Token | , |
Token ID | 11 |
Feature activation | -8.17e-04 |
Pos logit contributions | |
committee | +0.226 |
contestant | +0.219 |
brawl | +0.208 |
rookies | +0.200 |
duel | +0.194 |
Neg logit contributions | |
His | -0.291 |
HIS | -0.273 |
his | -0.266 |
his | -0.265 |
him | -0.262 |
Token | run |
Token ID | 1057 |
Feature activation | +5.32e-02 |
Pos logit contributions | |
His | +0.001 |
HIS | +0.001 |
his | +0.001 |
his | +0.001 |
him | +0.001 |
Neg logit contributions | |
committee | -0.001 |
contestant | -0.001 |
brawl | -0.001 |
rookies | -0.001 |
duel | -0.001 |
Token | - |
Token ID | 12 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
committee | +0.048 |
contestant | +0.047 |
brawl | +0.045 |
rookies | +0.043 |
duel | +0.042 |
Neg logit contributions | |
His | -0.052 |
HIS | -0.048 |
his | -0.047 |
his | -0.047 |
He | -0.046 |
Token | by |
Token ID | 1525 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
contestant | +0.114 |
committee | +0.113 |
brawl | +0.109 |
rookies | +0.105 |
duel | +0.103 |
Neg logit contributions | |
His | -0.115 |
HIS | -0.106 |
his | -0.105 |
his | -0.103 |
He | -0.102 |
Token | - |
Token ID | 12 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
committee | +0.165 |
contestant | +0.162 |
brawl | +0.153 |
rookies | +0.148 |
duel | +0.142 |
Neg logit contributions | |
His | -0.208 |
HIS | -0.195 |
his | -0.190 |
his | -0.189 |
him | -0.188 |
Token | committee |
Token ID | 26799 |
Feature activation | +7.82e-01 |
Pos logit contributions | |
contestant | +0.097 |
committee | +0.094 |
brawl | +0.093 |
rookies | +0.088 |
duel | +0.086 |
Neg logit contributions | |
His | -0.135 |
his | -0.125 |
HIS | -0.124 |
his | -0.122 |
He | -0.121 |
Token | affair |
Token ID | 14669 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
committee | +0.647 |
contestant | +0.639 |
brawl | +0.605 |
rookies | +0.581 |
duel | +0.562 |
Neg logit contributions | |
His | -0.820 |
HIS | -0.770 |
his | -0.747 |
his | -0.747 |
He | -0.738 |
Token | that |
Token ID | 326 |
Feature activation | -7.13e-02 |
Pos logit contributions | |
committee | +0.166 |
contestant | +0.164 |
brawl | +0.158 |
rookies | +0.151 |
duel | +0.148 |
Neg logit contributions | |
His | -0.155 |
HIS | -0.143 |
his | -0.139 |
his | -0.138 |
He | -0.137 |
Token | has |
Token ID | 468 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
His | +0.064 |
HIS | +0.059 |
his | +0.057 |
his | +0.057 |
He | +0.056 |
Neg logit contributions | |
committee | -0.070 |
contestant | -0.070 |
brawl | -0.067 |
rookies | -0.064 |
duel | -0.063 |
Token | mostly |
Token ID | 4632 |
Feature activation | +9.11e-02 |
Pos logit contributions | |
His | +0.122 |
HIS | +0.114 |
his | +0.111 |
his | +0.110 |
He | +0.108 |
Neg logit contributions | |
committee | -0.114 |
contestant | -0.113 |
brawl | -0.109 |
rookies | -0.103 |
duel | -0.101 |
Token | served |
Token ID | 4983 |
Feature activation | +3.57e-02 |
Pos logit contributions | |
committee | +0.083 |
contestant | +0.083 |
brawl | +0.079 |
rookies | +0.075 |
duel | +0.074 |
Neg logit contributions | |
His | -0.088 |
HIS | -0.082 |
his | -0.079 |
his | -0.079 |
him | -0.078 |
Token | � |
Token ID | 447 |
Feature activation | -1.14e-03 |
Pos logit contributions | |
His | +0.461 |
HIS | +0.427 |
his | +0.417 |
his | +0.413 |
He | +0.409 |
Neg logit contributions | |
committee | -0.448 |
contestant | -0.444 |
brawl | -0.425 |
rookies | -0.407 |
duel | -0.397 |
Token | � |
Token ID | 251 |
Feature activation | -7.69e-03 |
Pos logit contributions | |
His | +0.001 |
HIS | +0.001 |
his | +0.001 |
his | +0.001 |
He | +0.001 |
Neg logit contributions | |
committee | -0.001 |
contestant | -0.001 |
brawl | -0.001 |
rookies | -0.001 |
duel | -0.001 |
Token | Mayweather |
Token ID | 38718 |
Feature activation | +3.67e-02 |
Pos logit contributions | |
His | +0.008 |
HIS | +0.007 |
his | +0.007 |
his | +0.007 |
He | +0.007 |
Neg logit contributions | |
committee | -0.007 |
contestant | -0.007 |
brawl | -0.007 |
rookies | -0.006 |
duel | -0.006 |
Token | to |
Token ID | 284 |
Feature activation | +9.92e-02 |
Pos logit contributions | |
committee | +0.034 |
contestant | +0.033 |
brawl | +0.032 |
rookies | +0.030 |
duel | +0.030 |
Neg logit contributions | |
His | -0.035 |
HIS | -0.033 |
his | -0.032 |
his | -0.032 |
He | -0.031 |
Token | a |
Token ID | 257 |
Feature activation | +1.88e-01 |
Pos logit contributions | |
committee | +0.076 |
contestant | +0.075 |
brawl | +0.071 |
rookies | +0.067 |
duel | +0.066 |
Neg logit contributions | |
His | -0.110 |
HIS | -0.104 |
his | -0.101 |
his | -0.101 |
He | -0.099 |
Token | duel |
Token ID | 24662 |
Feature activation | +6.65e-01 |
Pos logit contributions | |
committee | +0.146 |
contestant | +0.143 |
brawl | +0.135 |
rookies | +0.129 |
duel | +0.124 |
Neg logit contributions | |
His | -0.207 |
HIS | -0.195 |
his | -0.190 |
his | -0.190 |
He | -0.187 |
Token | ! |
Token ID | 0 |
Feature activation | -4.39e-02 |
Pos logit contributions | |
committee | +0.636 |
contestant | +0.625 |
brawl | +0.600 |
rookies | +0.575 |
duel | +0.561 |
Neg logit contributions | |
His | -0.616 |
HIS | -0.570 |
his | -0.555 |
his | -0.552 |
He | -0.545 |
Token | Okay |
Token ID | 16805 |
Feature activation | +7.77e-03 |
Pos logit contributions | |
His | +0.043 |
HIS | +0.040 |
his | +0.038 |
his | +0.038 |
He | +0.038 |
Neg logit contributions | |
committee | -0.040 |
contestant | -0.040 |
brawl | -0.038 |
rookies | -0.036 |
duel | -0.036 |
Token | it |
Token ID | 340 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
committee | +0.008 |
contestant | +0.008 |
brawl | +0.007 |
rookies | +0.007 |
duel | +0.007 |
Neg logit contributions | |
His | -0.007 |
HIS | -0.006 |
his | -0.006 |
his | -0.006 |
He | -0.006 |
Token | wasn |
Token ID | 2492 |
Feature activation | -4.73e-02 |
Pos logit contributions | |
committee | +0.102 |
contestant | +0.101 |
brawl | +0.097 |
rookies | +0.093 |
duel | +0.091 |
Neg logit contributions | |
His | -0.104 |
HIS | -0.097 |
his | -0.094 |
his | -0.093 |
He | -0.092 |
Token | � |
Token ID | 447 |
Feature activation | -5.79e-03 |
Pos logit contributions | |
His | +0.040 |
HIS | +0.037 |
his | +0.036 |
his | +0.035 |
He | +0.035 |
Neg logit contributions | |
committee | -0.049 |
contestant | -0.048 |
brawl | -0.047 |
rookies | -0.044 |
duel | -0.044 |
Token | wasn |
Token ID | 2492 |
Feature activation | -4.73e-02 |
Pos logit contributions | |
committee | +0.102 |
contestant | +0.101 |
brawl | +0.097 |
rookies | +0.093 |
duel | +0.091 |
Neg logit contributions | |
His | -0.104 |
HIS | -0.097 |
his | -0.094 |
his | -0.093 |
He | -0.092 |
Token | � |
Token ID | 447 |
Feature activation | -5.79e-03 |
Pos logit contributions | |
His | +0.040 |
HIS | +0.037 |
his | +0.036 |
his | +0.035 |
He | +0.035 |
Neg logit contributions | |
committee | -0.049 |
contestant | -0.048 |
brawl | -0.047 |
rookies | -0.044 |
duel | -0.044 |
Token | � |
Token ID | 247 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
His | +0.007 |
HIS | +0.006 |
his | +0.006 |
his | +0.006 |
He | +0.006 |
Neg logit contributions | |
committee | -0.004 |
contestant | -0.004 |
brawl | -0.004 |
rookies | -0.004 |
duel | -0.004 |
Token | t |
Token ID | 83 |
Feature activation | -6.48e-03 |
Pos logit contributions | |
His | +0.021 |
HIS | +0.020 |
his | +0.019 |
his | +0.019 |
He | +0.019 |
Neg logit contributions | |
committee | -0.022 |
contestant | -0.022 |
brawl | -0.021 |
rookies | -0.020 |
duel | -0.019 |
Token | a |
Token ID | 257 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
His | +0.006 |
HIS | +0.006 |
his | +0.006 |
his | +0.006 |
him | +0.006 |
Neg logit contributions | |
committee | -0.006 |
contestant | -0.006 |
brawl | -0.006 |
rookies | -0.005 |
duel | -0.005 |
Token | duel |
Token ID | 24662 |
Feature activation | +6.62e-01 |
Pos logit contributions | |
committee | +0.130 |
contestant | +0.125 |
brawl | +0.117 |
rookies | +0.112 |
players | +0.106 |
Neg logit contributions | |
His | -0.228 |
HIS | -0.216 |
his | -0.211 |
his | -0.210 |
him | -0.209 |
Token | , |
Token ID | 11 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
committee | +0.651 |
contestant | +0.643 |
brawl | +0.618 |
rookies | +0.592 |
duel | +0.579 |
Neg logit contributions | |
His | -0.594 |
HIS | -0.548 |
his | -0.532 |
his | -0.529 |
He | -0.523 |
Token | but |
Token ID | 475 |
Feature activation | -6.96e-02 |
Pos logit contributions | |
His | +0.021 |
HIS | +0.019 |
his | +0.019 |
his | +0.019 |
He | +0.018 |
Neg logit contributions | |
committee | -0.021 |
contestant | -0.021 |
brawl | -0.020 |
rookies | -0.019 |
duel | -0.019 |
Token | a |
Token ID | 257 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
His | +0.069 |
HIS | +0.065 |
his | +0.063 |
his | +0.063 |
him | +0.062 |
Neg logit contributions | |
committee | -0.062 |
contestant | -0.061 |
brawl | -0.058 |
rookies | -0.055 |
duel | -0.054 |
Token | girl |
Token ID | 2576 |
Feature activation | +9.38e-02 |
Pos logit contributions | |
committee | +0.140 |
contestant | +0.136 |
brawl | +0.126 |
rookies | +0.123 |
players | +0.116 |
Neg logit contributions | |
His | -0.225 |
HIS | -0.214 |
his | -0.208 |
his | -0.207 |
He | -0.205 |
Token | can |
Token ID | 460 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
committee | +0.083 |
contestant | +0.080 |
brawl | +0.077 |
rookies | +0.074 |
duel | +0.071 |
Neg logit contributions | |
His | -0.094 |
HIS | -0.088 |
his | -0.086 |
his | -0.085 |
He | -0.084 |
Token | on |
Token ID | 319 |
Feature activation | -7.32e-03 |
Pos logit contributions | |
His | +0.034 |
HIS | +0.032 |
his | +0.031 |
his | +0.031 |
him | +0.030 |
Neg logit contributions | |
committee | -0.034 |
contestant | -0.034 |
brawl | -0.032 |
rookies | -0.031 |
duel | -0.030 |
Token | the |
Token ID | 262 |
Feature activation | +1.91e-02 |
Pos logit contributions | |
His | +0.007 |
HIS | +0.006 |
his | +0.006 |
his | +0.006 |
him | +0.006 |
Neg logit contributions | |
committee | -0.007 |
contestant | -0.007 |
brawl | -0.007 |
rookies | -0.006 |
duel | -0.006 |
Token | rooft |
Token ID | 27291 |
Feature activation | +5.96e-03 |
Pos logit contributions | |
committee | +0.019 |
contestant | +0.019 |
brawl | +0.018 |
rookies | +0.017 |
duel | +0.017 |
Neg logit contributions | |
His | -0.017 |
HIS | -0.016 |
his | -0.015 |
his | -0.015 |
He | -0.015 |
Token | ops |
Token ID | 2840 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
committee | +0.005 |
contestant | +0.005 |
brawl | +0.005 |
rookies | +0.005 |
duel | +0.004 |
Neg logit contributions | |
His | -0.006 |
HIS | -0.006 |
his | -0.006 |
his | -0.006 |
He | -0.006 |
Token | to |
Token ID | 284 |
Feature activation | +9.96e-02 |
Pos logit contributions | |
committee | +0.099 |
contestant | +0.097 |
brawl | +0.093 |
rookies | +0.089 |
duel | +0.087 |
Neg logit contributions | |
His | -0.096 |
HIS | -0.089 |
his | -0.086 |
his | -0.086 |
He | -0.085 |
Token | duel |
Token ID | 24662 |
Feature activation | +6.61e-01 |
Pos logit contributions | |
committee | +0.092 |
contestant | +0.091 |
brawl | +0.087 |
rookies | +0.083 |
duel | +0.081 |
Neg logit contributions | |
His | -0.095 |
HIS | -0.089 |
his | -0.086 |
his | -0.086 |
him | -0.085 |
Token | to |
Token ID | 284 |
Feature activation | +9.18e-02 |
Pos logit contributions | |
committee | +0.766 |
contestant | +0.759 |
brawl | +0.734 |
rookies | +0.705 |
duel | +0.696 |
Neg logit contributions | |
His | -0.478 |
HIS | -0.434 |
his | -0.418 |
his | -0.414 |
him | -0.408 |
Token | the |
Token ID | 262 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
committee | +0.088 |
contestant | +0.087 |
brawl | +0.083 |
rookies | +0.079 |
duel | +0.078 |
Neg logit contributions | |
His | -0.085 |
HIS | -0.079 |
his | -0.077 |
his | -0.076 |
He | -0.075 |
Token | death |
Token ID | 1918 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
committee | +0.020 |
contestant | +0.019 |
brawl | +0.019 |
rookies | +0.018 |
duel | +0.018 |
Neg logit contributions | |
His | -0.019 |
HIS | -0.018 |
his | -0.017 |
his | -0.017 |
He | -0.017 |
Token | in |
Token ID | 287 |
Feature activation | +1.79e-02 |
Pos logit contributions | |
His | +0.024 |
HIS | +0.022 |
his | +0.022 |
his | +0.021 |
him | +0.021 |
Neg logit contributions | |
committee | -0.025 |
contestant | -0.025 |
brawl | -0.024 |
rookies | -0.023 |
duel | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | +1.86e-02 |
Pos logit contributions | |
committee | +0.016 |
contestant | +0.016 |
brawl | +0.015 |
rookies | +0.015 |
duel | +0.014 |
Neg logit contributions | |
His | -0.017 |
HIS | -0.016 |
his | -0.016 |
his | -0.016 |
He | -0.016 |
Token | serious |
Token ID | 2726 |
Feature activation | +2.71e-01 |
Pos logit contributions | |
His | +0.005 |
HIS | +0.004 |
his | +0.004 |
his | +0.004 |
him | +0.004 |
Neg logit contributions | |
committee | -0.004 |
contestant | -0.004 |
brawl | -0.004 |
rookies | -0.004 |
duel | -0.004 |
Token | fiction |
Token ID | 10165 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
committee | +0.245 |
contestant | +0.241 |
brawl | +0.231 |
rookies | +0.219 |
duel | +0.214 |
Neg logit contributions | |
His | -0.265 |
HIS | -0.247 |
his | -0.240 |
his | -0.239 |
He | -0.236 |
Token | . |
Token ID | 13 |
Feature activation | +3.61e-02 |
Pos logit contributions | |
committee | +0.015 |
contestant | +0.015 |
brawl | +0.014 |
rookies | +0.013 |
duel | +0.013 |
Neg logit contributions | |
His | -0.014 |
HIS | -0.013 |
his | -0.013 |
his | -0.013 |
him | -0.013 |
Token | The |
Token ID | 383 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
committee | +0.034 |
contestant | +0.034 |
brawl | +0.032 |
rookies | +0.031 |
duel | +0.030 |
Neg logit contributions | |
His | -0.034 |
HIS | -0.032 |
his | -0.031 |
his | -0.031 |
He | -0.030 |
Token | expected |
Token ID | 2938 |
Feature activation | +6.62e-02 |
Pos logit contributions | |
committee | +0.108 |
contestant | +0.106 |
brawl | +0.102 |
rookies | +0.096 |
duel | +0.094 |
Neg logit contributions | |
His | -0.133 |
HIS | -0.124 |
his | -0.121 |
his | -0.121 |
He | -0.119 |
Token | duel |
Token ID | 24662 |
Feature activation | +6.57e-01 |
Pos logit contributions | |
committee | +0.060 |
contestant | +0.059 |
brawl | +0.056 |
rookies | +0.053 |
duel | +0.053 |
Neg logit contributions | |
His | -0.065 |
HIS | -0.060 |
his | -0.059 |
his | -0.058 |
He | -0.058 |
Token | never |
Token ID | 1239 |
Feature activation | +1.53e-02 |
Pos logit contributions | |
committee | +0.634 |
contestant | +0.622 |
brawl | +0.596 |
rookies | +0.571 |
duel | +0.556 |
Neg logit contributions | |
His | -0.603 |
HIS | -0.556 |
his | -0.542 |
his | -0.542 |
him | -0.535 |
Token | takes |
Token ID | 2753 |
Feature activation | -5.70e-02 |
Pos logit contributions | |
committee | +0.015 |
contestant | +0.015 |
brawl | +0.014 |
rookies | +0.014 |
duel | +0.013 |
Neg logit contributions | |
His | -0.014 |
HIS | -0.013 |
his | -0.012 |
his | -0.012 |
He | -0.012 |
Token | place |
Token ID | 1295 |
Feature activation | -5.04e-02 |
Pos logit contributions | |
His | +0.050 |
HIS | +0.047 |
his | +0.045 |
his | +0.045 |
him | +0.044 |
Neg logit contributions | |
committee | -0.057 |
contestant | -0.056 |
brawl | -0.054 |
rookies | -0.052 |
duel | -0.051 |
Token | . |
Token ID | 13 |
Feature activation | +3.67e-02 |
Pos logit contributions | |
His | +0.046 |
HIS | +0.042 |
his | +0.041 |
his | +0.041 |
him | +0.040 |
Neg logit contributions | |
committee | -0.049 |
contestant | -0.048 |
brawl | -0.047 |
rookies | -0.044 |
duel | -0.043 |
Token | The |
Token ID | 383 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
committee | +0.034 |
contestant | +0.033 |
brawl | +0.032 |
rookies | +0.030 |
duel | +0.030 |
Neg logit contributions | |
His | -0.036 |
HIS | -0.033 |
his | -0.032 |
his | -0.032 |
He | -0.032 |
Token | to |
Token ID | 284 |
Feature activation | +9.31e-02 |
Pos logit contributions | |
His | +0.227 |
HIS | +0.212 |
his | +0.207 |
his | +0.205 |
He | +0.204 |
Neg logit contributions | |
committee | -0.201 |
contestant | -0.199 |
brawl | -0.190 |
rookies | -0.180 |
duel | -0.178 |
Token | some |
Token ID | 617 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
committee | +0.093 |
contestant | +0.092 |
brawl | +0.089 |
rookies | +0.085 |
duel | +0.084 |
Neg logit contributions | |
His | -0.082 |
HIS | -0.075 |
his | -0.073 |
his | -0.072 |
He | -0.071 |
Token | of |
Token ID | 286 |
Feature activation | -4.02e-03 |
Pos logit contributions | |
committee | +0.120 |
contestant | +0.119 |
brawl | +0.115 |
rookies | +0.109 |
duel | +0.108 |
Neg logit contributions | |
His | -0.112 |
HIS | -0.104 |
his | -0.101 |
his | -0.100 |
He | -0.099 |
Token | the |
Token ID | 262 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
His | +0.004 |
HIS | +0.003 |
his | +0.003 |
his | +0.003 |
He | +0.003 |
Neg logit contributions | |
committee | -0.004 |
contestant | -0.004 |
brawl | -0.004 |
rookies | -0.004 |
duel | -0.004 |
Token | cos |
Token ID | 8615 |
Feature activation | -1.91e-01 |
Pos logit contributions | |
committee | +0.014 |
contestant | +0.014 |
brawl | +0.014 |
rookies | +0.013 |
duel | +0.013 |
Neg logit contributions | |
His | -0.012 |
HIS | -0.011 |
his | -0.010 |
his | -0.010 |
He | -0.010 |
Token | players |
Token ID | 32399 |
Feature activation | +6.39e-01 |
Pos logit contributions | |
His | +0.245 |
HIS | +0.231 |
his | +0.228 |
his | +0.225 |
He | +0.224 |
Neg logit contributions | |
committee | -0.113 |
contestant | -0.113 |
brawl | -0.106 |
rookies | -0.097 |
duel | -0.095 |
Token | featured |
Token ID | 8096 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
committee | +0.546 |
contestant | +0.539 |
brawl | +0.518 |
rookies | +0.489 |
duel | +0.481 |
Neg logit contributions | |
His | -0.654 |
HIS | -0.611 |
his | -0.594 |
his | -0.591 |
He | -0.585 |
Token | above |
Token ID | 2029 |
Feature activation | -8.01e-02 |
Pos logit contributions | |
committee | +0.199 |
contestant | +0.197 |
brawl | +0.190 |
rookies | +0.180 |
duel | +0.178 |
Neg logit contributions | |
His | -0.197 |
HIS | -0.183 |
his | -0.177 |
his | -0.176 |
He | -0.174 |
Token | : |
Token ID | 25 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
His | +0.076 |
HIS | +0.071 |
his | +0.069 |
his | +0.069 |
He | +0.068 |
Neg logit contributions | |
committee | -0.074 |
contestant | -0.073 |
brawl | -0.071 |
rookies | -0.067 |
duel | -0.066 |
Token |
|
Token ID | 198 |
Feature activation | +8.92e-03 |
Pos logit contributions | |
committee | +0.095 |
contestant | +0.094 |
brawl | +0.091 |
rookies | +0.086 |
duel | +0.085 |
Neg logit contributions | |
His | -0.096 |
HIS | -0.089 |
his | -0.087 |
his | -0.086 |
He | -0.085 |
Token |
|
Token ID | 198 |
Feature activation | +5.75e-03 |
Pos logit contributions | |
committee | +0.008 |
contestant | +0.008 |
brawl | +0.008 |
rookies | +0.008 |
duel | +0.007 |
Neg logit contributions | |
His | -0.008 |
HIS | -0.008 |
his | -0.008 |
his | -0.007 |
He | -0.007 |
Token | � |
Token ID | 247 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
His | +0.003 |
HIS | +0.003 |
his | +0.003 |
his | +0.003 |
He | +0.003 |
Neg logit contributions | |
committee | -0.003 |
contestant | -0.003 |
brawl | -0.003 |
rookies | -0.003 |
duel | -0.003 |
Token | ve |
Token ID | 303 |
Feature activation | +1.25e-03 |
Pos logit contributions | |
His | +0.018 |
HIS | +0.016 |
his | +0.016 |
his | +0.016 |
He | +0.015 |
Neg logit contributions | |
committee | -0.022 |
contestant | -0.022 |
brawl | -0.021 |
rookies | -0.021 |
duel | -0.020 |
Token | seen |
Token ID | 1775 |
Feature activation | +3.35e-02 |
Pos logit contributions | |
committee | +0.001 |
contestant | +0.001 |
brawl | +0.001 |
rookies | +0.001 |
duel | +0.001 |
Neg logit contributions | |
His | -0.001 |
HIS | -0.001 |
his | -0.001 |
his | -0.001 |
He | -0.001 |
Token | some |
Token ID | 617 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
committee | +0.031 |
contestant | +0.030 |
brawl | +0.029 |
rookies | +0.027 |
duel | +0.027 |
Neg logit contributions | |
His | -0.032 |
HIS | -0.030 |
his | -0.029 |
his | -0.029 |
him | -0.029 |
Token | cos |
Token ID | 8615 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
committee | +0.118 |
contestant | +0.116 |
brawl | +0.111 |
rookies | +0.105 |
duel | +0.104 |
Neg logit contributions | |
His | -0.128 |
HIS | -0.119 |
his | -0.116 |
his | -0.115 |
He | -0.114 |
Token | players |
Token ID | 32399 |
Feature activation | +6.37e-01 |
Pos logit contributions | |
His | +0.305 |
HIS | +0.291 |
his | +0.288 |
his | +0.286 |
He | +0.284 |
Neg logit contributions | |
committee | -0.056 |
contestant | -0.055 |
brawl | -0.048 |
rookies | -0.039 |
duel | -0.037 |
Token | really |
Token ID | 1107 |
Feature activation | -9.54e-03 |
Pos logit contributions | |
committee | +0.567 |
contestant | +0.558 |
brawl | +0.530 |
rookies | +0.507 |
duel | +0.495 |
Neg logit contributions | |
His | -0.630 |
HIS | -0.589 |
his | -0.571 |
his | -0.571 |
him | -0.563 |
Token | nail |
Token ID | 17864 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
His | +0.009 |
HIS | +0.009 |
his | +0.009 |
his | +0.009 |
him | +0.008 |
Neg logit contributions | |
committee | -0.008 |
contestant | -0.008 |
brawl | -0.008 |
rookies | -0.008 |
duel | -0.007 |
Token | some |
Token ID | 617 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
committee | +0.122 |
contestant | +0.121 |
brawl | +0.116 |
rookies | +0.110 |
duel | +0.109 |
Neg logit contributions | |
His | -0.122 |
HIS | -0.113 |
his | -0.110 |
his | -0.110 |
him | -0.108 |
Token | amazing |
Token ID | 4998 |
Feature activation | -9.58e-02 |
Pos logit contributions | |
committee | +0.124 |
contestant | +0.123 |
brawl | +0.118 |
rookies | +0.113 |
duel | +0.111 |
Neg logit contributions | |
His | -0.116 |
HIS | -0.107 |
his | -0.104 |
his | -0.103 |
He | -0.102 |
Token | set |
Token ID | 900 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
His | +0.084 |
HIS | +0.077 |
his | +0.075 |
his | +0.074 |
He | +0.073 |
Neg logit contributions | |
committee | -0.096 |
contestant | -0.095 |
brawl | -0.092 |
rookies | -0.087 |
duel | -0.086 |
Token | selection |
Token ID | 6356 |
Feature activation | +3.04e-01 |
Pos logit contributions | |
His | +0.065 |
HIS | +0.061 |
his | +0.060 |
his | +0.059 |
him | +0.059 |
Neg logit contributions | |
committee | -0.040 |
contestant | -0.038 |
brawl | -0.037 |
rookies | -0.034 |
duel | -0.034 |
Token | and |
Token ID | 290 |
Feature activation | +9.46e-02 |
Pos logit contributions | |
committee | +0.304 |
contestant | +0.297 |
brawl | +0.289 |
rookies | +0.273 |
duel | +0.270 |
Neg logit contributions | |
His | -0.271 |
HIS | -0.249 |
his | -0.245 |
his | -0.242 |
him | -0.240 |
Token | among |
Token ID | 1871 |
Feature activation | +9.20e-02 |
Pos logit contributions | |
committee | +0.083 |
contestant | +0.081 |
brawl | +0.078 |
rookies | +0.073 |
duel | +0.072 |
Neg logit contributions | |
His | -0.095 |
HIS | -0.089 |
his | -0.088 |
his | -0.087 |
him | -0.086 |
Token | the |
Token ID | 262 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
committee | +0.087 |
contestant | +0.087 |
brawl | +0.084 |
rookies | +0.078 |
duel | +0.078 |
Neg logit contributions | |
His | -0.086 |
HIS | -0.080 |
his | -0.078 |
his | -0.077 |
him | -0.076 |
Token | first |
Token ID | 717 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
committee | +0.027 |
contestant | +0.026 |
brawl | +0.025 |
rookies | +0.024 |
duel | +0.024 |
Neg logit contributions | |
His | -0.028 |
HIS | -0.026 |
his | -0.026 |
his | -0.025 |
him | -0.025 |
Token | trio |
Token ID | 19886 |
Feature activation | +6.31e-01 |
Pos logit contributions | |
committee | +0.078 |
contestant | +0.076 |
brawl | +0.074 |
duel | +0.067 |
skirm | +0.066 |
Neg logit contributions | |
His | -0.116 |
HIS | -0.109 |
his | -0.108 |
his | -0.106 |
him | -0.106 |
Token | to |
Token ID | 284 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
committee | +0.570 |
contestant | +0.561 |
brawl | +0.543 |
duel | +0.505 |
rookies | +0.504 |
Neg logit contributions | |
His | -0.617 |
HIS | -0.573 |
his | -0.558 |
his | -0.556 |
He | -0.549 |
Token | be |
Token ID | 307 |
Feature activation | -3.68e-01 |
Pos logit contributions | |
committee | +0.100 |
contestant | +0.098 |
brawl | +0.095 |
rookies | +0.090 |
duel | +0.088 |
Neg logit contributions | |
His | -0.092 |
HIS | -0.086 |
his | -0.083 |
his | -0.083 |
He | -0.081 |
Token | selected |
Token ID | 6163 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
His | +0.365 |
HIS | +0.343 |
his | +0.336 |
his | +0.329 |
him | +0.329 |
Neg logit contributions | |
committee | -0.325 |
contestant | -0.320 |
brawl | -0.311 |
rookies | -0.286 |
duel | -0.285 |
Token | to |
Token ID | 284 |
Feature activation | +9.87e-02 |
Pos logit contributions | |
committee | +0.201 |
contestant | +0.198 |
brawl | +0.192 |
rookies | +0.179 |
duel | +0.179 |
Neg logit contributions | |
His | -0.187 |
HIS | -0.173 |
his | -0.169 |
his | -0.167 |
him | -0.166 |
Token | the |
Token ID | 262 |
Feature activation | +2.21e-02 |
Pos logit contributions | |
committee | +0.093 |
contestant | +0.092 |
brawl | +0.089 |
rookies | +0.084 |
duel | +0.083 |
Neg logit contributions | |
His | -0.092 |
HIS | -0.086 |
his | -0.083 |
his | -0.083 |
He | -0.082 |
Token | the |
Token ID | 262 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
committee | +0.319 |
contestant | +0.315 |
brawl | +0.303 |
rookies | +0.289 |
duel | +0.285 |
Neg logit contributions | |
His | -0.286 |
HIS | -0.265 |
his | -0.256 |
his | -0.256 |
him | -0.252 |
Token | futuristic |
Token ID | 36701 |
Feature activation | +6.08e-02 |
Pos logit contributions | |
committee | +0.019 |
contestant | +0.019 |
brawl | +0.018 |
rookies | +0.017 |
duel | +0.017 |
Neg logit contributions | |
His | -0.015 |
HIS | -0.014 |
his | -0.013 |
his | -0.013 |
He | -0.013 |
Token | fort |
Token ID | 6285 |
Feature activation | +6.93e-02 |
Pos logit contributions | |
committee | +0.064 |
contestant | +0.064 |
brawl | +0.061 |
rookies | +0.059 |
duel | +0.058 |
Neg logit contributions | |
His | -0.050 |
HIS | -0.046 |
his | -0.044 |
his | -0.044 |
He | -0.043 |
Token | ifications |
Token ID | 6637 |
Feature activation | +3.81e-02 |
Pos logit contributions | |
committee | +0.067 |
contestant | +0.067 |
brawl | +0.064 |
rookies | +0.061 |
duel | +0.060 |
Neg logit contributions | |
His | -0.063 |
HIS | -0.058 |
his | -0.057 |
his | -0.056 |
He | -0.055 |
Token | that |
Token ID | 326 |
Feature activation | -9.83e-02 |
Pos logit contributions | |
committee | +0.037 |
contestant | +0.036 |
brawl | +0.035 |
rookies | +0.033 |
duel | +0.032 |
Neg logit contributions | |
His | -0.035 |
HIS | -0.033 |
his | -0.032 |
his | -0.032 |
him | -0.031 |
Token | antagonist |
Token ID | 34667 |
Feature activation | +6.29e-01 |
Pos logit contributions | |
His | +0.104 |
HIS | +0.097 |
his | +0.096 |
his | +0.094 |
him | +0.094 |
Neg logit contributions | |
committee | -0.081 |
contestant | -0.079 |
brawl | -0.075 |
rookies | -0.071 |
duel | -0.069 |
Token | Dr |
Token ID | 1583 |
Feature activation | -9.80e-04 |
Pos logit contributions | |
committee | +0.568 |
contestant | +0.558 |
brawl | +0.537 |
rookies | +0.509 |
duel | +0.501 |
Neg logit contributions | |
His | -0.614 |
HIS | -0.572 |
his | -0.557 |
his | -0.556 |
him | -0.548 |
Token | . |
Token ID | 13 |
Feature activation | +4.14e-02 |
Pos logit contributions | |
His | +0.001 |
HIS | +0.001 |
his | +0.001 |
his | +0.001 |
He | +0.001 |
Neg logit contributions | |
committee | -0.001 |
contestant | -0.001 |
brawl | -0.001 |
rookies | -0.001 |
duel | -0.001 |
Token | M |
Token ID | 337 |
Feature activation | -3.82e-02 |
Pos logit contributions | |
committee | +0.038 |
contestant | +0.037 |
brawl | +0.036 |
rookies | +0.034 |
duel | +0.034 |
Neg logit contributions | |
His | -0.040 |
HIS | -0.037 |
his | -0.036 |
his | -0.036 |
He | -0.036 |
Token | has |
Token ID | 468 |
Feature activation | -1.58e-01 |
Pos logit contributions | |
His | +0.036 |
HIS | +0.033 |
his | +0.032 |
his | +0.032 |
He | +0.032 |
Neg logit contributions | |
committee | -0.036 |
contestant | -0.036 |
brawl | -0.035 |
rookies | -0.033 |
duel | -0.032 |
Token | constructed |
Token ID | 12006 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
His | +0.145 |
HIS | +0.134 |
his | +0.130 |
his | +0.130 |
him | +0.128 |
Neg logit contributions | |
committee | -0.153 |
contestant | -0.151 |
brawl | -0.145 |
rookies | -0.138 |
duel | -0.135 |
Token | … |
Token ID | 3926 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
committee | +0.150 |
contestant | +0.149 |
brawl | +0.143 |
rookies | +0.137 |
duel | +0.134 |
Neg logit contributions | |
His | -0.133 |
HIS | -0.123 |
his | -0.119 |
his | -0.119 |
He | -0.117 |
Token | it |
Token ID | 340 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
His | +0.012 |
HIS | +0.011 |
his | +0.011 |
his | +0.011 |
He | +0.011 |
Neg logit contributions | |
committee | -0.011 |
contestant | -0.011 |
brawl | -0.010 |
rookies | -0.010 |
duel | -0.010 |
Token | could |
Token ID | 714 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
committee | +0.084 |
contestant | +0.083 |
brawl | +0.079 |
rookies | +0.074 |
duel | +0.073 |
Neg logit contributions | |
His | -0.122 |
HIS | -0.114 |
his | -0.112 |
his | -0.111 |
He | -0.110 |
Token | lead |
Token ID | 1085 |
Feature activation | +2.39e-01 |
Pos logit contributions | |
committee | +0.187 |
contestant | +0.187 |
brawl | +0.180 |
rookies | +0.171 |
duel | +0.169 |
Neg logit contributions | |
His | -0.176 |
HIS | -0.162 |
his | -0.158 |
his | -0.155 |
He | -0.154 |
Token | to |
Token ID | 284 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
committee | +0.201 |
contestant | +0.200 |
brawl | +0.191 |
rookies | +0.180 |
duel | +0.177 |
Neg logit contributions | |
His | -0.247 |
HIS | -0.230 |
his | -0.225 |
his | -0.223 |
He | -0.221 |
Token | skirm |
Token ID | 37238 |
Feature activation | +6.26e-01 |
Pos logit contributions | |
committee | +0.088 |
contestant | +0.086 |
brawl | +0.082 |
rookies | +0.078 |
duel | +0.076 |
Neg logit contributions | |
His | -0.104 |
HIS | -0.098 |
his | -0.095 |
his | -0.095 |
him | -0.094 |
Token | ishes |
Token ID | 5614 |
Feature activation | -2.79e-01 |
Pos logit contributions | |
committee | +0.813 |
contestant | +0.812 |
brawl | +0.786 |
rookies | +0.762 |
duel | +0.751 |
Neg logit contributions | |
His | -0.362 |
HIS | -0.317 |
his | -0.305 |
his | -0.298 |
He | -0.294 |
Token | in |
Token ID | 287 |
Feature activation | +2.60e-02 |
Pos logit contributions | |
His | +0.251 |
HIS | +0.231 |
his | +0.227 |
his | +0.224 |
him | +0.223 |
Neg logit contributions | |
committee | -0.275 |
contestant | -0.271 |
brawl | -0.259 |
rookies | -0.248 |
duel | -0.241 |
Token | various |
Token ID | 2972 |
Feature activation | +8.92e-02 |
Pos logit contributions | |
committee | +0.025 |
contestant | +0.025 |
brawl | +0.024 |
rookies | +0.023 |
duel | +0.023 |
Neg logit contributions | |
His | -0.023 |
HIS | -0.022 |
his | -0.021 |
his | -0.021 |
him | -0.021 |
Token | places |
Token ID | 4113 |
Feature activation | -3.71e-01 |
Pos logit contributions | |
committee | +0.095 |
contestant | +0.094 |
brawl | +0.091 |
rookies | +0.087 |
duel | +0.086 |
Neg logit contributions | |
His | -0.072 |
HIS | -0.066 |
his | -0.064 |
his | -0.063 |
He | -0.063 |
Token | around |
Token ID | 1088 |
Feature activation | -9.01e-02 |
Pos logit contributions | |
His | +0.323 |
HIS | +0.297 |
his | +0.289 |
his | +0.289 |
He | +0.284 |
Neg logit contributions | |
committee | -0.373 |
contestant | -0.370 |
brawl | -0.355 |
rookies | -0.340 |
duel | -0.333 |
Token | of |
Token ID | 286 |
Feature activation | -6.76e-03 |
Pos logit contributions | |
His | +0.029 |
HIS | +0.027 |
his | +0.026 |
his | +0.026 |
He | +0.026 |
Neg logit contributions | |
committee | -0.027 |
contestant | -0.027 |
brawl | -0.026 |
rookies | -0.024 |
duel | -0.024 |
Token | the |
Token ID | 262 |
Feature activation | +1.43e-02 |
Pos logit contributions | |
His | +0.006 |
HIS | +0.006 |
his | +0.006 |
his | +0.006 |
He | +0.005 |
Neg logit contributions | |
committee | -0.006 |
contestant | -0.006 |
brawl | -0.006 |
rookies | -0.006 |
duel | -0.006 |
Token | hip |
Token ID | 10359 |
Feature activation | -4.29e-03 |
Pos logit contributions | |
committee | +0.013 |
contestant | +0.013 |
brawl | +0.013 |
rookies | +0.012 |
duel | +0.012 |
Neg logit contributions | |
His | -0.014 |
HIS | -0.013 |
his | -0.012 |
his | -0.012 |
He | -0.012 |
Token | - |
Token ID | 12 |
Feature activation | +9.08e-02 |
Pos logit contributions | |
His | +0.004 |
HIS | +0.004 |
his | +0.004 |
his | +0.004 |
He | +0.004 |
Neg logit contributions | |
committee | -0.004 |
contestant | -0.004 |
brawl | -0.003 |
rookies | -0.003 |
duel | -0.003 |
Token | hop |
Token ID | 8548 |
Feature activation | -3.22e-02 |
Pos logit contributions | |
committee | +0.086 |
contestant | +0.086 |
brawl | +0.082 |
rookies | +0.078 |
duel | +0.077 |
Neg logit contributions | |
His | -0.085 |
HIS | -0.078 |
his | -0.077 |
his | -0.075 |
He | -0.075 |
Token | trio |
Token ID | 19886 |
Feature activation | +6.26e-01 |
Pos logit contributions | |
His | +0.037 |
HIS | +0.035 |
his | +0.034 |
his | +0.034 |
He | +0.034 |
Neg logit contributions | |
committee | -0.023 |
contestant | -0.023 |
brawl | -0.022 |
rookies | -0.020 |
duel | -0.020 |
Token | Mig |
Token ID | 24488 |
Feature activation | -5.75e-03 |
Pos logit contributions | |
committee | +0.583 |
contestant | +0.579 |
brawl | +0.557 |
rookies | +0.529 |
duel | +0.522 |
Neg logit contributions | |
His | -0.590 |
HIS | -0.548 |
his | -0.532 |
his | -0.525 |
He | -0.522 |
Token | os |
Token ID | 418 |
Feature activation | -5.26e-02 |
Pos logit contributions | |
His | +0.005 |
HIS | +0.005 |
his | +0.005 |
his | +0.005 |
He | +0.005 |
Neg logit contributions | |
committee | -0.006 |
contestant | -0.006 |
brawl | -0.005 |
rookies | -0.005 |
duel | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | -3.14e-02 |
Pos logit contributions | |
His | +0.046 |
HIS | +0.043 |
his | +0.041 |
his | +0.041 |
He | +0.041 |
Neg logit contributions | |
committee | -0.052 |
contestant | -0.052 |
brawl | -0.050 |
rookies | -0.048 |
duel | -0.047 |
Token | this |
Token ID | 428 |
Feature activation | -6.43e-02 |
Pos logit contributions | |
His | +0.030 |
HIS | +0.028 |
his | +0.028 |
his | +0.028 |
He | +0.027 |
Neg logit contributions | |
committee | -0.029 |
contestant | -0.028 |
brawl | -0.027 |
rookies | -0.026 |
duel | -0.025 |
Token | is |
Token ID | 318 |
Feature activation | -6.96e-02 |
Pos logit contributions | |
His | +0.061 |
HIS | +0.057 |
his | +0.055 |
his | +0.055 |
He | +0.054 |
Neg logit contributions | |
committee | -0.060 |
contestant | -0.059 |
brawl | -0.057 |
rookies | -0.054 |
duel | -0.053 |
Token | like |
Token ID | 588 |
Feature activation | -2.50e-01 |
Pos logit contributions | |
His | +0.034 |
HIS | +0.031 |
his | +0.030 |
his | +0.030 |
He | +0.030 |
Neg logit contributions | |
committee | -0.034 |
contestant | -0.033 |
brawl | -0.032 |
rookies | -0.030 |
duel | -0.030 |
Token | Dani |
Token ID | 22254 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
His | +0.210 |
HIS | +0.194 |
his | +0.189 |
his | +0.188 |
him | +0.186 |
Neg logit contributions | |
committee | -0.259 |
contestant | -0.254 |
brawl | -0.246 |
rookies | -0.233 |
duel | -0.232 |
Token | , |
Token ID | 11 |
Feature activation | -4.31e-02 |
Pos logit contributions | |
His | +0.011 |
HIS | +0.011 |
his | +0.010 |
his | +0.010 |
He | +0.010 |
Neg logit contributions | |
committee | -0.012 |
contestant | -0.012 |
brawl | -0.012 |
rookies | -0.011 |
duel | -0.011 |
Token | will |
Token ID | 481 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
His | +0.045 |
HIS | +0.042 |
his | +0.041 |
his | +0.041 |
him | +0.041 |
Neg logit contributions | |
committee | -0.036 |
contestant | -0.035 |
brawl | -0.034 |
rookies | -0.032 |
duel | -0.031 |
Token | prefer |
Token ID | 4702 |
Feature activation | -9.78e-03 |
Pos logit contributions | |
committee | +0.124 |
contestant | +0.123 |
brawl | +0.118 |
rookies | +0.113 |
duel | +0.111 |
Neg logit contributions | |
His | -0.118 |
HIS | -0.109 |
his | -0.106 |
his | -0.105 |
He | -0.104 |
Token | spectator |
Token ID | 43893 |
Feature activation | +6.25e-01 |
Pos logit contributions | |
His | +0.010 |
HIS | +0.009 |
his | +0.009 |
his | +0.009 |
him | +0.009 |
Neg logit contributions | |
committee | -0.009 |
contestant | -0.009 |
brawl | -0.008 |
rookies | -0.008 |
duel | -0.008 |
Token | activities |
Token ID | 4568 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
committee | +0.630 |
contestant | +0.622 |
brawl | +0.599 |
rookies | +0.573 |
duel | +0.564 |
Neg logit contributions | |
His | -0.545 |
HIS | -0.502 |
his | -0.487 |
his | -0.483 |
He | -0.478 |
Token | that |
Token ID | 326 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
His | +0.119 |
HIS | +0.110 |
his | +0.107 |
his | +0.107 |
him | +0.105 |
Neg logit contributions | |
committee | -0.134 |
contestant | -0.133 |
brawl | -0.128 |
rookies | -0.122 |
duel | -0.120 |
Token | don |
Token ID | 836 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
His | +0.136 |
HIS | +0.128 |
his | +0.126 |
his | +0.124 |
him | +0.124 |
Neg logit contributions | |
committee | -0.105 |
contestant | -0.104 |
brawl | -0.098 |
rookies | -0.093 |
duel | -0.091 |
Token | � |
Token ID | 447 |
Feature activation | -2.17e-02 |
Pos logit contributions | |
committee | +0.179 |
contestant | +0.178 |
brawl | +0.172 |
rookies | +0.164 |
duel | +0.161 |
Neg logit contributions | |
His | -0.159 |
HIS | -0.146 |
his | -0.142 |
his | -0.140 |
He | -0.139 |
Token | � |
Token ID | 247 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
His | +0.021 |
HIS | +0.019 |
his | +0.019 |
his | +0.018 |
He | +0.018 |
Neg logit contributions | |
committee | -0.020 |
contestant | -0.020 |
brawl | -0.019 |
rookies | -0.018 |
duel | -0.018 |
Token | play |
Token ID | 711 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
committee | +0.112 |
contestant | +0.109 |
brawl | +0.104 |
rookies | +0.099 |
duel | +0.096 |
Neg logit contributions | |
His | -0.131 |
HIS | -0.122 |
his | -0.121 |
his | -0.119 |
him | -0.118 |
Token | a |
Token ID | 257 |
Feature activation | +2.20e-01 |
Pos logit contributions | |
committee | +0.196 |
contestant | +0.190 |
brawl | +0.185 |
rookies | +0.171 |
duel | +0.169 |
Neg logit contributions | |
His | -0.232 |
HIS | -0.218 |
his | -0.216 |
him | -0.213 |
his | -0.211 |
Token | mother |
Token ID | 2802 |
Feature activation | -5.09e-02 |
Pos logit contributions | |
committee | +0.165 |
contestant | +0.158 |
brawl | +0.152 |
rookies | +0.144 |
players | +0.138 |
Neg logit contributions | |
His | -0.247 |
HIS | -0.234 |
his | -0.230 |
him | -0.227 |
his | -0.226 |
Token | and |
Token ID | 290 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
His | +0.051 |
HIS | +0.047 |
his | +0.047 |
his | +0.046 |
him | +0.046 |
Neg logit contributions | |
committee | -0.045 |
contestant | -0.044 |
brawl | -0.042 |
rookies | -0.040 |
duel | -0.039 |
Token | daughter |
Token ID | 4957 |
Feature activation | -4.83e-02 |
Pos logit contributions | |
committee | +0.130 |
contestant | +0.126 |
brawl | +0.123 |
rookies | +0.117 |
duel | +0.115 |
Neg logit contributions | |
His | -0.095 |
HIS | -0.088 |
his | -0.088 |
his | -0.085 |
him | -0.085 |
Token | duo |
Token ID | 18545 |
Feature activation | +6.23e-01 |
Pos logit contributions | |
His | +0.049 |
his | +0.046 |
HIS | +0.046 |
him | +0.045 |
his | +0.045 |
Neg logit contributions | |
committee | -0.041 |
contestant | -0.039 |
brawl | -0.037 |
rookies | -0.036 |
�� | -0.035 |
Token | who |
Token ID | 508 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
committee | +0.557 |
contestant | +0.541 |
brawl | +0.520 |
rookies | +0.489 |
duel | +0.480 |
Neg logit contributions | |
His | -0.617 |
HIS | -0.577 |
his | -0.566 |
his | -0.558 |
him | -0.555 |
Token | find |
Token ID | 1064 |
Feature activation | +3.57e-02 |
Pos logit contributions | |
committee | +0.129 |
contestant | +0.126 |
brawl | +0.119 |
rookies | +0.113 |
duel | +0.109 |
Neg logit contributions | |
His | -0.167 |
HIS | -0.156 |
his | -0.154 |
his | -0.152 |
him | -0.151 |
Token | themselves |
Token ID | 2405 |
Feature activation | +2.68e-01 |
Pos logit contributions | |
committee | +0.028 |
contestant | +0.027 |
brawl | +0.026 |
rookies | +0.024 |
duel | +0.023 |
Neg logit contributions | |
His | -0.039 |
HIS | -0.037 |
his | -0.036 |
him | -0.036 |
his | -0.036 |
Token | the |
Token ID | 262 |
Feature activation | +7.30e-02 |
Pos logit contributions | |
committee | +0.229 |
contestant | +0.221 |
brawl | +0.211 |
rookies | +0.197 |
duel | +0.192 |
Neg logit contributions | |
His | -0.277 |
HIS | -0.260 |
his | -0.258 |
him | -0.253 |
his | -0.251 |
Token | targets |
Token ID | 6670 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
committee | +0.051 |
contestant | +0.050 |
brawl | +0.047 |
rookies | +0.044 |
duel | +0.043 |
Neg logit contributions | |
His | -0.086 |
HIS | -0.082 |
his | -0.080 |
his | -0.079 |
He | -0.079 |
Token | , |
Token ID | 11 |
Feature activation | -3.51e-02 |
Pos logit contributions | |
committee | +0.166 |
contestant | +0.163 |
brawl | +0.157 |
rookies | +0.150 |
duel | +0.147 |
Neg logit contributions | |
His | -0.165 |
HIS | -0.153 |
his | -0.148 |
his | -0.147 |
He | -0.146 |
Token | a |
Token ID | 257 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
His | +0.034 |
HIS | +0.032 |
his | +0.031 |
his | +0.031 |
He | +0.030 |
Neg logit contributions | |
committee | -0.032 |
contestant | -0.032 |
brawl | -0.031 |
rookies | -0.029 |
duel | -0.028 |
Token | Boston |
Token ID | 6182 |
Feature activation | -2.04e-02 |
Pos logit contributions | |
committee | +0.156 |
contestant | +0.153 |
brawl | +0.147 |
rookies | +0.140 |
duel | +0.137 |
Neg logit contributions | |
His | -0.178 |
HIS | -0.166 |
his | -0.161 |
his | -0.160 |
He | -0.159 |
Token | - |
Token ID | 12 |
Feature activation | +8.16e-02 |
Pos logit contributions | |
His | +0.020 |
HIS | +0.019 |
his | +0.019 |
his | +0.018 |
He | +0.018 |
Neg logit contributions | |
committee | -0.018 |
contestant | -0.018 |
brawl | -0.017 |
rookies | -0.016 |
duel | -0.016 |
Token | metal |
Token ID | 28469 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
committee | +0.071 |
contestant | +0.071 |
brawl | +0.068 |
rookies | +0.064 |
duel | +0.063 |
Neg logit contributions | |
His | -0.082 |
HIS | -0.076 |
his | -0.074 |
his | -0.073 |
He | -0.073 |
Token | trio |
Token ID | 19886 |
Feature activation | +6.23e-01 |
Pos logit contributions | |
His | +0.144 |
HIS | +0.135 |
his | +0.132 |
his | +0.132 |
He | +0.131 |
Neg logit contributions | |
committee | -0.093 |
contestant | -0.091 |
brawl | -0.086 |
rookies | -0.081 |
duel | -0.079 |
Token | with |
Token ID | 351 |
Feature activation | +3.21e-02 |
Pos logit contributions | |
committee | +0.589 |
contestant | +0.582 |
brawl | +0.561 |
rookies | +0.532 |
duel | +0.525 |
Neg logit contributions | |
His | -0.581 |
HIS | -0.539 |
his | -0.524 |
his | -0.519 |
He | -0.514 |
Token | an |
Token ID | 281 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
committee | +0.028 |
contestant | +0.028 |
brawl | +0.026 |
rookies | +0.025 |
duel | +0.025 |
Neg logit contributions | |
His | -0.032 |
HIS | -0.030 |
his | -0.029 |
his | -0.029 |
him | -0.029 |
Token | EP |
Token ID | 14724 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
committee | +0.125 |
contestant | +0.124 |
brawl | +0.119 |
rookies | +0.114 |
duel | +0.112 |
Neg logit contributions | |
His | -0.128 |
HIS | -0.118 |
his | -0.115 |
his | -0.113 |
He | -0.113 |
Token | called |
Token ID | 1444 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
committee | +0.201 |
contestant | +0.200 |
brawl | +0.192 |
rookies | +0.182 |
duel | +0.179 |
Neg logit contributions | |
His | -0.202 |
HIS | -0.187 |
his | -0.182 |
his | -0.180 |
He | -0.179 |
Token | You |
Token ID | 921 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
His | +0.028 |
HIS | +0.026 |
his | +0.026 |
his | +0.025 |
He | +0.025 |
Neg logit contributions | |
committee | -0.027 |
contestant | -0.027 |
brawl | -0.026 |
rookies | -0.024 |
duel | -0.024 |
Token | active |
Token ID | 4075 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
committee | +0.007 |
contestant | +0.007 |
brawl | +0.007 |
rookies | +0.006 |
duel | +0.006 |
Neg logit contributions | |
His | -0.008 |
HIS | -0.007 |
his | -0.007 |
his | -0.007 |
He | -0.007 |
Token | CR |
Token ID | 8740 |
Feature activation | +8.08e-03 |
Pos logit contributions | |
committee | +0.132 |
contestant | +0.130 |
brawl | +0.125 |
rookies | +0.119 |
duel | +0.117 |
Neg logit contributions | |
His | -0.133 |
HIS | -0.123 |
his | -0.120 |
his | -0.119 |
He | -0.118 |
Token | F |
Token ID | 37 |
Feature activation | -8.12e-02 |
Pos logit contributions | |
committee | +0.009 |
contestant | +0.009 |
brawl | +0.009 |
rookies | +0.008 |
duel | +0.008 |
Neg logit contributions | |
His | -0.006 |
HIS | -0.006 |
his | -0.005 |
his | -0.005 |
He | -0.005 |
Token | 1 |
Token ID | 16 |
Feature activation | -3.58e-02 |
Pos logit contributions | |
His | +0.115 |
HIS | +0.110 |
his | +0.107 |
his | +0.107 |
He | +0.107 |
Neg logit contributions | |
committee | -0.038 |
contestant | -0.037 |
brawl | -0.034 |
rookies | -0.030 |
duel | -0.029 |
Token | receptor |
Token ID | 17408 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
His | +0.049 |
HIS | +0.046 |
his | +0.045 |
his | +0.045 |
He | +0.045 |
Neg logit contributions | |
committee | -0.019 |
contestant | -0.018 |
brawl | -0.017 |
rookies | -0.015 |
duel | -0.015 |
Token | antagonist |
Token ID | 34667 |
Feature activation | +6.22e-01 |
Pos logit contributions | |
committee | +0.067 |
contestant | +0.065 |
brawl | +0.058 |
rookies | +0.050 |
duel | +0.048 |
Neg logit contributions | |
His | -0.280 |
HIS | -0.268 |
his | -0.262 |
his | -0.262 |
He | -0.260 |
Token | . |
Token ID | 13 |
Feature activation | +3.95e-02 |
Pos logit contributions | |
committee | +0.602 |
contestant | +0.597 |
brawl | +0.575 |
rookies | +0.548 |
duel | +0.540 |
Neg logit contributions | |
His | -0.567 |
HIS | -0.524 |
his | -0.508 |
his | -0.504 |
He | -0.499 |
Token | Oral |
Token ID | 42222 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
committee | +0.036 |
contestant | +0.036 |
brawl | +0.034 |
rookies | +0.033 |
duel | +0.032 |
Neg logit contributions | |
His | -0.039 |
HIS | -0.036 |
his | -0.035 |
his | -0.035 |
He | -0.034 |
Token | administration |
Token ID | 3662 |
Feature activation | +8.40e-02 |
Pos logit contributions | |
committee | +0.224 |
contestant | +0.222 |
brawl | +0.213 |
rookies | +0.203 |
duel | +0.200 |
Neg logit contributions | |
His | -0.213 |
HIS | -0.198 |
his | -0.191 |
his | -0.190 |
He | -0.188 |
Token | of |
Token ID | 286 |
Feature activation | +8.83e-03 |
Pos logit contributions | |
committee | +0.079 |
contestant | +0.078 |
brawl | +0.075 |
rookies | +0.071 |
duel | +0.070 |
Neg logit contributions | |
His | -0.079 |
HIS | -0.073 |
his | -0.071 |
his | -0.071 |
him | -0.070 |
Token | a |
Token ID | 257 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
committee | +0.009 |
contestant | +0.009 |
brawl | +0.009 |
rookies | +0.008 |
duel | +0.008 |
Neg logit contributions | |
His | -0.007 |
HIS | -0.007 |
his | -0.007 |
his | -0.007 |
him | -0.006 |
Token | Shir |
Token ID | 21972 |
Feature activation | +2.73e-02 |
Pos logit contributions | |
committee | +0.084 |
contestant | +0.083 |
brawl | +0.080 |
rookies | +0.076 |
duel | +0.075 |
Neg logit contributions | |
His | -0.076 |
HIS | -0.070 |
his | -0.068 |
his | -0.068 |
He | -0.067 |
Token | az |
Token ID | 1031 |
Feature activation | -9.74e-02 |
Pos logit contributions | |
committee | +0.029 |
contestant | +0.029 |
brawl | +0.028 |
rookies | +0.027 |
duel | +0.027 |
Neg logit contributions | |
His | -0.022 |
HIS | -0.020 |
his | -0.019 |
his | -0.019 |
He | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | -3.84e-02 |
Pos logit contributions | |
His | +0.088 |
HIS | +0.081 |
his | +0.080 |
his | +0.079 |
He | +0.078 |
Neg logit contributions | |
committee | -0.095 |
contestant | -0.094 |
brawl | -0.090 |
rookies | -0.086 |
duel | -0.085 |
Token | where |
Token ID | 810 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
His | +0.034 |
HIS | +0.031 |
his | +0.031 |
his | +0.031 |
him | +0.030 |
Neg logit contributions | |
committee | -0.038 |
contestant | -0.038 |
brawl | -0.036 |
rookies | -0.034 |
duel | -0.034 |
Token | protesters |
Token ID | 9584 |
Feature activation | +3.82e-01 |
Pos logit contributions | |
His | +0.204 |
HIS | +0.192 |
his | +0.190 |
him | +0.186 |
his | +0.186 |
Neg logit contributions | |
committee | -0.157 |
contestant | -0.153 |
brawl | -0.144 |
rookies | -0.137 |
duel | -0.132 |
Token | skirm |
Token ID | 37238 |
Feature activation | +6.20e-01 |
Pos logit contributions | |
committee | +0.298 |
contestant | +0.293 |
brawl | +0.273 |
rookies | +0.260 |
duel | +0.250 |
Neg logit contributions | |
His | -0.422 |
HIS | -0.397 |
his | -0.390 |
his | -0.387 |
him | -0.384 |
Token | ished |
Token ID | 1348 |
Feature activation | -1.58e-01 |
Pos logit contributions | |
contestant | +0.707 |
committee | +0.706 |
brawl | +0.678 |
rookies | +0.658 |
duel | +0.642 |
Neg logit contributions | |
His | -0.458 |
HIS | -0.412 |
his | -0.403 |
his | -0.397 |
He | -0.391 |
Token | with |
Token ID | 351 |
Feature activation | +3.87e-02 |
Pos logit contributions | |
His | +0.153 |
HIS | +0.142 |
his | +0.140 |
him | +0.138 |
his | +0.138 |
Neg logit contributions | |
committee | -0.147 |
contestant | -0.144 |
brawl | -0.135 |
rookies | -0.130 |
duel | -0.125 |
Token | Bas |
Token ID | 6455 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
committee | +0.033 |
contestant | +0.032 |
brawl | +0.031 |
rookies | +0.029 |
duel | +0.028 |
Neg logit contributions | |
His | -0.040 |
HIS | -0.037 |
his | -0.037 |
him | -0.036 |
his | -0.036 |
Token | ij |
Token ID | 2926 |
Feature activation | -4.77e-02 |
Pos logit contributions | |
committee | +0.125 |
contestant | +0.125 |
brawl | +0.120 |
rookies | +0.114 |
duel | +0.113 |
Neg logit contributions | |
His | -0.110 |
HIS | -0.101 |
his | -0.099 |
his | -0.097 |
He | -0.096 |
Token | milit |
Token ID | 2036 |
Feature activation | +6.86e-02 |
Pos logit contributions | |
His | +0.056 |
HIS | +0.052 |
his | +0.051 |
his | +0.051 |
He | +0.051 |
Neg logit contributions | |
committee | -0.034 |
contestant | -0.033 |
brawl | -0.032 |
rookies | -0.029 |
duel | -0.029 |
Token | Kendall |
Token ID | 36779 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
committee | +0.059 |
contestant | +0.058 |
brawl | +0.056 |
rookies | +0.052 |
duel | +0.052 |
Neg logit contributions | |
His | -0.073 |
HIS | -0.069 |
his | -0.067 |
his | -0.067 |
him | -0.066 |
Token | Reyes |
Token ID | 40599 |
Feature activation | +4.25e-02 |
Pos logit contributions | |
committee | +0.178 |
contestant | +0.176 |
brawl | +0.169 |
rookies | +0.161 |
duel | +0.158 |
Neg logit contributions | |
His | -0.191 |
HIS | -0.177 |
his | -0.172 |
his | -0.170 |
He | -0.169 |
Token | as |
Token ID | 355 |
Feature activation | -6.80e-02 |
Pos logit contributions | |
committee | +0.039 |
contestant | +0.039 |
brawl | +0.037 |
rookies | +0.035 |
duel | +0.034 |
Neg logit contributions | |
His | -0.040 |
HIS | -0.038 |
his | -0.037 |
his | -0.037 |
him | -0.036 |
Token | the |
Token ID | 262 |
Feature activation | +2.35e-03 |
Pos logit contributions | |
His | +0.067 |
his | +0.063 |
HIS | +0.062 |
his | +0.061 |
him | +0.060 |
Neg logit contributions | |
committee | -0.061 |
contestant | -0.059 |
brawl | -0.057 |
duel | -0.052 |
rookies | -0.052 |
Token | starting |
Token ID | 3599 |
Feature activation | +2.53e-01 |
Pos logit contributions | |
committee | +0.002 |
contestant | +0.002 |
brawl | +0.002 |
duel | +0.002 |
rookies | +0.002 |
Neg logit contributions | |
His | -0.003 |
HIS | -0.002 |
his | -0.002 |
him | -0.002 |
his | -0.002 |
Token | trio |
Token ID | 19886 |
Feature activation | +6.17e-01 |
Pos logit contributions | |
committee | +0.192 |
contestant | +0.188 |
brawl | +0.177 |
rookies | +0.162 |
duel | +0.161 |
Neg logit contributions | |
His | -0.284 |
HIS | -0.267 |
his | -0.262 |
his | -0.259 |
him | -0.259 |
Token | . |
Token ID | 13 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
committee | +0.596 |
contestant | +0.589 |
brawl | +0.566 |
rookies | +0.531 |
duel | +0.526 |
Neg logit contributions | |
His | -0.564 |
HIS | -0.523 |
his | -0.511 |
his | -0.507 |
him | -0.502 |
Token | There |
Token ID | 1318 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
committee | +0.011 |
contestant | +0.010 |
brawl | +0.010 |
rookies | +0.009 |
duel | +0.009 |
Neg logit contributions | |
His | -0.012 |
HIS | -0.011 |
his | -0.011 |
his | -0.011 |
He | -0.011 |
Token | was |
Token ID | 373 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
contestant | +0.166 |
committee | +0.165 |
brawl | +0.158 |
rookies | +0.147 |
duel | +0.145 |
Neg logit contributions | |
His | -0.227 |
HIS | -0.211 |
his | -0.208 |
his | -0.207 |
He | -0.205 |
Token | a |
Token ID | 257 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
His | +0.122 |
HIS | +0.114 |
his | +0.113 |
him | +0.111 |
his | +0.111 |
Neg logit contributions | |
committee | -0.104 |
contestant | -0.103 |
brawl | -0.098 |
rookies | -0.092 |
duel | -0.090 |
Token | lot |
Token ID | 1256 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
committee | +0.147 |
contestant | +0.144 |
brawl | +0.135 |
rookies | +0.129 |
duel | +0.126 |
Neg logit contributions | |
His | -0.184 |
HIS | -0.173 |
his | -0.170 |
him | -0.167 |
his | -0.167 |
Token | has |
Token ID | 468 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
committee | +0.067 |
contestant | +0.066 |
brawl | +0.064 |
rookies | +0.061 |
duel | +0.060 |
Neg logit contributions | |
His | -0.056 |
HIS | -0.052 |
his | -0.050 |
his | -0.050 |
him | -0.049 |
Token | already |
Token ID | 1541 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
His | +0.171 |
HIS | +0.158 |
his | +0.153 |
his | +0.152 |
He | +0.150 |
Neg logit contributions | |
committee | -0.184 |
contestant | -0.182 |
brawl | -0.175 |
rookies | -0.167 |
duel | -0.164 |
Token | broken |
Token ID | 5445 |
Feature activation | +7.56e-02 |
Pos logit contributions | |
His | +0.150 |
HIS | +0.139 |
his | +0.135 |
his | +0.134 |
He | +0.133 |
Neg logit contributions | |
committee | -0.148 |
contestant | -0.147 |
brawl | -0.141 |
rookies | -0.133 |
duel | -0.132 |
Token | up |
Token ID | 510 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
committee | +0.084 |
contestant | +0.084 |
brawl | +0.081 |
rookies | +0.078 |
duel | +0.077 |
Neg logit contributions | |
His | -0.057 |
HIS | -0.052 |
his | -0.050 |
his | -0.049 |
He | -0.049 |
Token | that |
Token ID | 326 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
His | +0.213 |
HIS | +0.197 |
his | +0.193 |
his | +0.191 |
him | +0.188 |
Neg logit contributions | |
committee | -0.242 |
contestant | -0.238 |
brawl | -0.229 |
rookies | -0.217 |
duel | -0.214 |
Token | trio |
Token ID | 19886 |
Feature activation | +6.15e-01 |
Pos logit contributions | |
His | +0.163 |
HIS | +0.155 |
his | +0.152 |
him | +0.151 |
his | +0.150 |
Neg logit contributions | |
committee | -0.085 |
contestant | -0.082 |
brawl | -0.076 |
rookies | -0.071 |
duel | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
committee | +0.593 |
contestant | +0.584 |
brawl | +0.560 |
rookies | +0.530 |
duel | +0.523 |
Neg logit contributions | |
His | -0.567 |
HIS | -0.526 |
his | -0.513 |
his | -0.510 |
him | -0.504 |
Token | L |
Token ID | 406 |
Feature activation | -6.98e-02 |
Pos logit contributions | |
committee | +0.015 |
contestant | +0.015 |
brawl | +0.015 |
rookies | +0.014 |
duel | +0.014 |
Neg logit contributions | |
His | -0.014 |
HIS | -0.013 |
his | -0.013 |
his | -0.013 |
He | -0.013 |
Token | add |
Token ID | 2860 |
Feature activation | -5.05e-02 |
Pos logit contributions | |
His | +0.062 |
HIS | +0.057 |
his | +0.056 |
his | +0.055 |
He | +0.055 |
Neg logit contributions | |
committee | -0.069 |
contestant | -0.069 |
brawl | -0.066 |
rookies | -0.063 |
duel | -0.062 |
Token | played |
Token ID | 2826 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
His | +0.046 |
HIS | +0.043 |
his | +0.042 |
his | +0.042 |
He | +0.041 |
Neg logit contributions | |
committee | -0.049 |
contestant | -0.048 |
brawl | -0.046 |
rookies | -0.044 |
duel | -0.043 |
Token | left |
Token ID | 1364 |
Feature activation | +9.58e-02 |
Pos logit contributions | |
committee | +0.289 |
contestant | +0.285 |
brawl | +0.273 |
rookies | +0.258 |
duel | +0.255 |
Neg logit contributions | |
His | -0.306 |
HIS | -0.285 |
his | -0.277 |
his | -0.277 |
him | -0.272 |
Token | Tai |
Token ID | 47976 |
Feature activation | -7.74e-02 |
Pos logit contributions | |
committee | +0.003 |
contestant | +0.003 |
brawl | +0.003 |
rookies | +0.003 |
duel | +0.003 |
Neg logit contributions | |
His | -0.003 |
HIS | -0.002 |
his | -0.002 |
He | -0.002 |
his | -0.002 |
Token | Lung |
Token ID | 39337 |
Feature activation | -2.17e-02 |
Pos logit contributions | |
His | +0.068 |
HIS | +0.062 |
his | +0.061 |
his | +0.060 |
He | +0.059 |
Neg logit contributions | |
committee | -0.078 |
contestant | -0.077 |
brawl | -0.074 |
rookies | -0.071 |
duel | -0.070 |
Token | , |
Token ID | 11 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
His | +0.020 |
HIS | +0.019 |
his | +0.018 |
his | +0.018 |
He | +0.018 |
Neg logit contributions | |
committee | -0.020 |
contestant | -0.020 |
brawl | -0.019 |
rookies | -0.018 |
duel | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
His | +0.030 |
HIS | +0.028 |
his | +0.028 |
his | +0.027 |
him | +0.027 |
Neg logit contributions | |
committee | -0.025 |
contestant | -0.025 |
brawl | -0.024 |
rookies | -0.023 |
duel | -0.022 |
Token | main |
Token ID | 1388 |
Feature activation | +9.30e-02 |
Pos logit contributions | |
committee | +0.020 |
contestant | +0.020 |
brawl | +0.019 |
rookies | +0.018 |
duel | +0.018 |
Neg logit contributions | |
His | -0.023 |
HIS | -0.021 |
his | -0.021 |
his | -0.021 |
He | -0.020 |
Token | antagonist |
Token ID | 34667 |
Feature activation | +6.12e-01 |
Pos logit contributions | |
committee | +0.059 |
contestant | +0.057 |
brawl | +0.054 |
rookies | +0.051 |
duel | +0.049 |
Neg logit contributions | |
His | -0.116 |
HIS | -0.109 |
his | -0.107 |
his | -0.107 |
He | -0.106 |
Token | of |
Token ID | 286 |
Feature activation | -5.80e-03 |
Pos logit contributions | |
committee | +0.581 |
contestant | +0.571 |
brawl | +0.551 |
rookies | +0.526 |
duel | +0.514 |
Neg logit contributions | |
His | -0.570 |
HIS | -0.529 |
his | -0.514 |
his | -0.513 |
He | -0.506 |
Token | the |
Token ID | 262 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
His | +0.006 |
HIS | +0.005 |
his | +0.005 |
his | +0.005 |
He | +0.005 |
Neg logit contributions | |
committee | -0.005 |
contestant | -0.005 |
brawl | -0.005 |
rookies | -0.005 |
duel | -0.005 |
Token | 2008 |
Token ID | 3648 |
Feature activation | -9.95e-02 |
Pos logit contributions | |
committee | +0.017 |
contestant | +0.017 |
brawl | +0.016 |
rookies | +0.016 |
duel | +0.015 |
Neg logit contributions | |
His | -0.018 |
HIS | -0.017 |
his | -0.017 |
his | -0.017 |
He | -0.016 |
Token | film |
Token ID | 2646 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
His | +0.103 |
HIS | +0.096 |
his | +0.093 |
his | +0.093 |
him | +0.092 |
Neg logit contributions | |
committee | -0.085 |
contestant | -0.083 |
brawl | -0.079 |
rookies | -0.075 |
duel | -0.073 |
Token | Kung |
Token ID | 44753 |
Feature activation | -1.63e-03 |
Pos logit contributions | |
committee | +0.125 |
contestant | +0.123 |
brawl | +0.118 |
rookies | +0.112 |
duel | +0.110 |
Neg logit contributions | |
His | -0.132 |
HIS | -0.123 |
his | -0.120 |
his | -0.119 |
He | -0.118 |
Token | after |
Token ID | 706 |
Feature activation | +5.15e-04 |
Pos logit contributions | |
His | +0.021 |
HIS | +0.020 |
his | +0.019 |
his | +0.019 |
him | +0.019 |
Neg logit contributions | |
committee | -0.020 |
contestant | -0.019 |
brawl | -0.018 |
rookies | -0.017 |
duel | -0.017 |
Token | a |
Token ID | 257 |
Feature activation | +1.98e-01 |
Pos logit contributions | |
committee | +0.000 |
contestant | +0.000 |
brawl | +0.000 |
rookies | +0.000 |
duel | +0.000 |
Neg logit contributions | |
His | -0.000 |
HIS | -0.000 |
his | -0.000 |
his | -0.000 |
him | -0.000 |
Token | try |
Token ID | 1949 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
committee | +0.159 |
contestant | +0.154 |
brawl | +0.145 |
rookies | +0.138 |
duel | +0.134 |
Neg logit contributions | |
His | -0.214 |
HIS | -0.201 |
his | -0.198 |
his | -0.196 |
him | -0.194 |
Token | out |
Token ID | 448 |
Feature activation | -7.57e-03 |
Pos logit contributions | |
committee | +0.214 |
contestant | +0.212 |
brawl | +0.204 |
rookies | +0.194 |
duel | +0.191 |
Neg logit contributions | |
His | -0.197 |
HIS | -0.180 |
his | -0.176 |
his | -0.176 |
He | -0.174 |
Token | in |
Token ID | 287 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
His | +0.007 |
HIS | +0.007 |
his | +0.007 |
his | +0.007 |
him | +0.006 |
Neg logit contributions | |
committee | -0.007 |
contestant | -0.007 |
brawl | -0.007 |
rookies | -0.006 |
duel | -0.006 |
Token | rookie |
Token ID | 12302 |
Feature activation | +6.09e-01 |
Pos logit contributions | |
committee | +0.023 |
contestant | +0.023 |
brawl | +0.022 |
rookies | +0.021 |
duel | +0.021 |
Neg logit contributions | |
His | -0.023 |
HIS | -0.021 |
his | -0.021 |
his | -0.021 |
him | -0.020 |
Token | mini |
Token ID | 9927 |
Feature activation | +2.62e-01 |
Pos logit contributions | |
committee | +0.518 |
contestant | +0.509 |
brawl | +0.486 |
rookies | +0.457 |
duel | +0.453 |
Neg logit contributions | |
His | -0.627 |
HIS | -0.584 |
his | -0.572 |
his | -0.570 |
him | -0.562 |
Token | - |
Token ID | 12 |
Feature activation | +9.95e-02 |
Pos logit contributions | |
committee | +0.206 |
contestant | +0.204 |
brawl | +0.193 |
rookies | +0.181 |
duel | +0.179 |
Neg logit contributions | |
His | -0.287 |
HIS | -0.269 |
his | -0.263 |
his | -0.261 |
He | -0.259 |
Token | camp |
Token ID | 16544 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
contestant | +0.073 |
committee | +0.072 |
brawl | +0.068 |
rookies | +0.063 |
duel | +0.063 |
Neg logit contributions | |
His | -0.115 |
HIS | -0.107 |
his | -0.106 |
He | -0.104 |
his | -0.104 |
Token | . |
Token ID | 13 |
Feature activation | +3.67e-02 |
Pos logit contributions | |
committee | +0.183 |
contestant | +0.181 |
brawl | +0.173 |
rookies | +0.164 |
duel | +0.163 |
Neg logit contributions | |
His | -0.179 |
HIS | -0.165 |
his | -0.161 |
his | -0.161 |
He | -0.158 |
Token | Chris |
Token ID | 5180 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
committee | +0.034 |
contestant | +0.034 |
brawl | +0.032 |
rookies | +0.031 |
duel | +0.030 |
Neg logit contributions | |
His | -0.035 |
HIS | -0.033 |
his | -0.032 |
his | -0.031 |
He | -0.031 |
Token | at |
Token ID | 379 |
Feature activation | -6.94e-03 |
Pos logit contributions | |
His | +0.027 |
HIS | +0.025 |
his | +0.024 |
his | +0.024 |
He | +0.024 |
Neg logit contributions | |
committee | -0.024 |
contestant | -0.023 |
brawl | -0.022 |
rookies | -0.021 |
duel | -0.021 |
Token | May |
Token ID | 1737 |
Feature activation | +5.77e-02 |
Pos logit contributions | |
His | +0.006 |
HIS | +0.006 |
his | +0.006 |
his | +0.006 |
He | +0.006 |
Neg logit contributions | |
committee | -0.007 |
contestant | -0.007 |
brawl | -0.006 |
rookies | -0.006 |
duel | -0.006 |
Token | y |
Token ID | 88 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
committee | +0.053 |
contestant | +0.053 |
brawl | +0.050 |
rookies | +0.048 |
duel | +0.047 |
Neg logit contributions | |
His | -0.055 |
HIS | -0.051 |
his | -0.050 |
his | -0.050 |
He | -0.049 |
Token | ar |
Token ID | 283 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
His | +0.156 |
HIS | +0.143 |
his | +0.141 |
his | +0.139 |
He | +0.138 |
Neg logit contributions | |
committee | -0.159 |
contestant | -0.158 |
brawl | -0.152 |
rookies | -0.145 |
duel | -0.143 |
Token | ( |
Token ID | 357 |
Feature activation | -1.33e-03 |
Pos logit contributions | |
committee | +0.099 |
contestant | +0.098 |
brawl | +0.095 |
rookies | +0.090 |
duel | +0.089 |
Neg logit contributions | |
His | -0.097 |
HIS | -0.090 |
his | -0.087 |
his | -0.086 |
He | -0.086 |
Token | His |
Token ID | 6653 |
Feature activation | -7.60e-01 |
Pos logit contributions | |
His | +0.001 |
HIS | +0.001 |
his | +0.001 |
his | +0.001 |
He | +0.001 |
Neg logit contributions | |
committee | -0.001 |
contestant | -0.001 |
brawl | -0.001 |
rookies | -0.001 |
duel | -0.001 |
Token | ar |
Token ID | 283 |
Feature activation | +9.41e-02 |
Pos logit contributions | |
His | +0.695 |
HIS | +0.641 |
his | +0.627 |
his | +0.618 |
He | +0.614 |
Neg logit contributions | |
committee | -0.730 |
contestant | -0.727 |
brawl | -0.698 |
rookies | -0.666 |
duel | -0.655 |
Token | ) |
Token ID | 8 |
Feature activation | -9.54e-02 |
Pos logit contributions | |
committee | +0.093 |
contestant | +0.092 |
brawl | +0.089 |
rookies | +0.085 |
duel | +0.084 |
Neg logit contributions | |
His | -0.084 |
HIS | -0.077 |
his | -0.075 |
his | -0.074 |
He | -0.074 |
Token | � |
Token ID | 447 |
Feature activation | -2.34e-02 |
Pos logit contributions | |
His | +0.090 |
HIS | +0.083 |
his | +0.081 |
his | +0.080 |
He | +0.080 |
Neg logit contributions | |
committee | -0.089 |
contestant | -0.089 |
brawl | -0.085 |
rookies | -0.081 |
duel | -0.080 |
Token | � |
Token ID | 251 |
Feature activation | -2.97e-02 |
Pos logit contributions | |
His | +0.025 |
HIS | +0.023 |
his | +0.022 |
his | +0.022 |
He | +0.022 |
Neg logit contributions | |
committee | -0.019 |
contestant | -0.019 |
brawl | -0.019 |
rookies | -0.017 |
duel | -0.017 |
Token | by |
Token ID | 416 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
His | +0.028 |
HIS | +0.026 |
his | +0.025 |
his | +0.025 |
He | +0.025 |
Neg logit contributions | |
committee | -0.028 |
contestant | -0.028 |
brawl | -0.027 |
rookies | -0.025 |
duel | -0.025 |
Token | central |
Token ID | 4318 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
committee | +0.007 |
contestant | +0.007 |
brawl | +0.007 |
rookies | +0.007 |
duel | +0.007 |
Neg logit contributions | |
His | -0.006 |
HIS | -0.005 |
his | -0.005 |
his | -0.005 |
He | -0.005 |
Token | Alberta |
Token ID | 15555 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
committee | +0.192 |
contestant | +0.190 |
brawl | +0.185 |
rookies | +0.178 |
duel | +0.176 |
Neg logit contributions | |
His | -0.108 |
HIS | -0.097 |
his | -0.093 |
his | -0.092 |
He | -0.091 |
Token | . |
Token ID | 13 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
His | +0.216 |
HIS | +0.200 |
his | +0.195 |
his | +0.193 |
He | +0.191 |
Neg logit contributions | |
committee | -0.220 |
contestant | -0.218 |
brawl | -0.210 |
rookies | -0.200 |
duel | -0.197 |
Token |
|
Token ID | 198 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
committee | +0.013 |
contestant | +0.013 |
brawl | +0.012 |
rookies | +0.012 |
duel | +0.012 |
Neg logit contributions | |
His | -0.013 |
HIS | -0.012 |
his | -0.011 |
his | -0.011 |
He | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | -1.61e-02 |
Pos logit contributions | |
His | +0.016 |
HIS | +0.015 |
his | +0.015 |
his | +0.014 |
He | +0.014 |
Neg logit contributions | |
committee | -0.016 |
contestant | -0.016 |
brawl | -0.016 |
rookies | -0.015 |
duel | -0.015 |
Token | His |
Token ID | 6653 |
Feature activation | -7.55e-01 |
Pos logit contributions | |
His | +0.011 |
HIS | +0.010 |
his | +0.010 |
his | +0.010 |
He | +0.010 |
Neg logit contributions | |
committee | -0.019 |
contestant | -0.018 |
brawl | -0.018 |
rookies | -0.017 |
duel | -0.017 |
Token | sentence |
Token ID | 6827 |
Feature activation | +4.11e-02 |
Pos logit contributions | |
His | +0.791 |
HIS | +0.741 |
his | +0.720 |
his | +0.716 |
He | +0.709 |
Neg logit contributions | |
committee | -0.627 |
contestant | -0.620 |
brawl | -0.592 |
rookies | -0.560 |
duel | -0.550 |
Token | is |
Token ID | 318 |
Feature activation | -8.37e-02 |
Pos logit contributions | |
committee | +0.040 |
contestant | +0.040 |
brawl | +0.039 |
rookies | +0.037 |
duel | +0.036 |
Neg logit contributions | |
His | -0.037 |
HIS | -0.034 |
his | -0.033 |
his | -0.033 |
He | -0.032 |
Token | to |
Token ID | 284 |
Feature activation | +7.96e-02 |
Pos logit contributions | |
His | +0.081 |
HIS | +0.075 |
his | +0.073 |
his | +0.072 |
He | +0.072 |
Neg logit contributions | |
committee | -0.076 |
contestant | -0.076 |
brawl | -0.073 |
rookies | -0.069 |
duel | -0.068 |
Token | expire |
Token ID | 24264 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
committee | +0.078 |
contestant | +0.078 |
brawl | +0.075 |
rookies | +0.072 |
duel | +0.071 |
Neg logit contributions | |
His | -0.071 |
HIS | -0.065 |
his | -0.063 |
He | -0.062 |
his | -0.061 |
Token | in |
Token ID | 287 |
Feature activation | +2.90e-03 |
Pos logit contributions | |
His | +0.135 |
HIS | +0.126 |
his | +0.122 |
his | +0.120 |
He | +0.120 |
Neg logit contributions | |
committee | -0.130 |
contestant | -0.130 |
brawl | -0.124 |
rookies | -0.118 |
duel | -0.117 |
Token | Art |
Token ID | 3683 |
Feature activation | -8.35e-02 |
Pos logit contributions | |
committee | +0.060 |
contestant | +0.060 |
brawl | +0.057 |
rookies | +0.055 |
duel | +0.054 |
Neg logit contributions | |
His | -0.059 |
HIS | -0.055 |
his | -0.053 |
his | -0.053 |
He | -0.052 |
Token | z |
Token ID | 89 |
Feature activation | +8.08e-04 |
Pos logit contributions | |
His | +0.077 |
HIS | +0.071 |
his | +0.069 |
his | +0.068 |
He | +0.068 |
Neg logit contributions | |
committee | -0.080 |
contestant | -0.079 |
brawl | -0.076 |
rookies | -0.073 |
duel | -0.072 |
Token | . |
Token ID | 13 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
committee | +0.001 |
contestant | +0.001 |
brawl | +0.001 |
rookies | +0.001 |
duel | +0.001 |
Neg logit contributions | |
His | -0.001 |
HIS | -0.001 |
his | -0.001 |
his | -0.001 |
He | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -5.44e-03 |
Pos logit contributions | |
committee | +0.016 |
contestant | +0.015 |
brawl | +0.015 |
rookies | +0.014 |
duel | +0.014 |
Neg logit contributions | |
His | -0.015 |
HIS | -0.014 |
his | -0.013 |
his | -0.013 |
He | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
His | +0.005 |
HIS | +0.005 |
his | +0.005 |
his | +0.005 |
He | +0.005 |
Neg logit contributions | |
committee | -0.005 |
contestant | -0.005 |
brawl | -0.005 |
rookies | -0.005 |
duel | -0.005 |
Token | His |
Token ID | 6653 |
Feature activation | -7.52e-01 |
Pos logit contributions | |
His | +0.010 |
HIS | +0.009 |
his | +0.009 |
his | +0.009 |
He | +0.009 |
Neg logit contributions | |
committee | -0.013 |
contestant | -0.013 |
brawl | -0.012 |
rookies | -0.012 |
duel | -0.012 |
Token | friends |
Token ID | 2460 |
Feature activation | -2.93e-02 |
Pos logit contributions | |
His | +0.692 |
HIS | +0.641 |
his | +0.622 |
his | +0.614 |
He | +0.610 |
Neg logit contributions | |
committee | -0.720 |
contestant | -0.714 |
brawl | -0.687 |
rookies | -0.653 |
duel | -0.644 |
Token | and |
Token ID | 290 |
Feature activation | +7.01e-02 |
Pos logit contributions | |
His | +0.029 |
HIS | +0.027 |
his | +0.027 |
his | +0.027 |
He | +0.026 |
Neg logit contributions | |
committee | -0.026 |
contestant | -0.026 |
brawl | -0.024 |
rookies | -0.023 |
duel | -0.023 |
Token | Oak |
Token ID | 9121 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
committee | +0.069 |
contestant | +0.069 |
brawl | +0.066 |
rookies | +0.063 |
duel | +0.062 |
Neg logit contributions | |
His | -0.063 |
HIS | -0.058 |
his | -0.056 |
He | -0.055 |
his | -0.055 |
Token | ton |
Token ID | 1122 |
Feature activation | +5.20e-02 |
Pos logit contributions | |
His | +0.174 |
HIS | +0.161 |
his | +0.157 |
his | +0.155 |
He | +0.154 |
Neg logit contributions | |
committee | -0.175 |
contestant | -0.174 |
brawl | -0.167 |
rookies | -0.159 |
duel | -0.157 |
Token | neighbors |
Token ID | 12020 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
committee | +0.049 |
contestant | +0.048 |
brawl | +0.047 |
rookies | +0.044 |
duel | +0.044 |
Neg logit contributions | |
His | -0.049 |
HIS | -0.045 |
his | -0.044 |
his | -0.043 |
He | -0.043 |
Token | NYPD |
Token ID | 27615 |
Feature activation | +3.80e-02 |
Pos logit contributions | |
committee | +0.115 |
contestant | +0.114 |
brawl | +0.109 |
rookies | +0.104 |
duel | +0.102 |
Neg logit contributions | |
His | -0.136 |
HIS | -0.127 |
his | -0.124 |
his | -0.122 |
He | -0.121 |
Token | flyer |
Token ID | 42464 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
committee | +0.030 |
contestant | +0.030 |
brawl | +0.028 |
rookies | +0.027 |
duel | +0.026 |
Neg logit contributions | |
His | -0.041 |
HIS | -0.039 |
his | -0.038 |
his | -0.038 |
He | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
committee | +0.199 |
contestant | +0.198 |
brawl | +0.190 |
rookies | +0.181 |
duel | +0.179 |
Neg logit contributions | |
His | -0.190 |
HIS | -0.177 |
his | -0.171 |
his | -0.170 |
He | -0.168 |
Token |
|
Token ID | 198 |
Feature activation | -7.58e-03 |
Pos logit contributions | |
committee | +0.023 |
contestant | +0.023 |
brawl | +0.022 |
rookies | +0.021 |
duel | +0.021 |
Neg logit contributions | |
His | -0.021 |
HIS | -0.019 |
his | -0.019 |
his | -0.019 |
He | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
His | +0.007 |
HIS | +0.007 |
his | +0.006 |
his | +0.006 |
He | +0.006 |
Neg logit contributions | |
committee | -0.007 |
contestant | -0.007 |
brawl | -0.007 |
rookies | -0.006 |
duel | -0.006 |
Token | His |
Token ID | 6653 |
Feature activation | -7.50e-01 |
Pos logit contributions | |
His | +0.012 |
HIS | +0.011 |
his | +0.010 |
his | +0.010 |
He | +0.010 |
Neg logit contributions | |
committee | -0.016 |
contestant | -0.016 |
brawl | -0.015 |
rookies | -0.015 |
duel | -0.014 |
Token | show |
Token ID | 905 |
Feature activation | +6.43e-02 |
Pos logit contributions | |
His | +0.651 |
HIS | +0.600 |
his | +0.580 |
his | +0.577 |
He | +0.570 |
Neg logit contributions | |
committee | -0.758 |
contestant | -0.750 |
brawl | -0.722 |
rookies | -0.690 |
duel | -0.680 |
Token | tells |
Token ID | 4952 |
Feature activation | -9.67e-02 |
Pos logit contributions | |
committee | +0.062 |
contestant | +0.061 |
brawl | +0.059 |
rookies | +0.056 |
duel | +0.055 |
Neg logit contributions | |
His | -0.059 |
HIS | -0.055 |
his | -0.053 |
his | -0.053 |
He | -0.052 |
Token | the |
Token ID | 262 |
Feature activation | +3.40e-03 |
Pos logit contributions | |
His | +0.090 |
HIS | +0.084 |
his | +0.082 |
his | +0.082 |
him | +0.081 |
Neg logit contributions | |
committee | -0.092 |
contestant | -0.090 |
brawl | -0.087 |
rookies | -0.082 |
duel | -0.081 |
Token | back |
Token ID | 736 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
committee | +0.003 |
contestant | +0.003 |
brawl | +0.003 |
rookies | +0.002 |
duel | +0.002 |
Neg logit contributions | |
His | -0.004 |
HIS | -0.003 |
his | -0.003 |
his | -0.003 |
He | -0.003 |
Token | story |
Token ID | 1621 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
His | +0.255 |
HIS | +0.237 |
his | +0.234 |
his | +0.233 |
He | +0.232 |
Neg logit contributions | |
committee | -0.202 |
contestant | -0.199 |
brawl | -0.190 |
rookies | -0.178 |
duel | -0.176 |
Token | typically |
Token ID | 6032 |
Feature activation | +1.97e-01 |
Pos logit contributions | |
His | +0.119 |
HIS | +0.111 |
his | +0.108 |
his | +0.107 |
He | +0.106 |
Neg logit contributions | |
committee | -0.107 |
contestant | -0.107 |
brawl | -0.102 |
rookies | -0.097 |
duel | -0.095 |
Token | fatal |
Token ID | 10800 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
committee | +0.171 |
contestant | +0.170 |
brawl | +0.163 |
rookies | +0.154 |
duel | +0.152 |
Neg logit contributions | |
His | -0.199 |
HIS | -0.186 |
his | -0.181 |
his | -0.179 |
He | -0.177 |
Token | . |
Token ID | 13 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
committee | +0.197 |
contestant | +0.196 |
brawl | +0.188 |
rookies | +0.179 |
duel | +0.176 |
Neg logit contributions | |
His | -0.196 |
HIS | -0.182 |
his | -0.176 |
his | -0.175 |
He | -0.173 |
Token |
|
Token ID | 198 |
Feature activation | -8.62e-03 |
Pos logit contributions | |
committee | +0.017 |
contestant | +0.017 |
brawl | +0.017 |
rookies | +0.016 |
duel | +0.016 |
Neg logit contributions | |
His | -0.017 |
HIS | -0.016 |
his | -0.015 |
his | -0.015 |
He | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | -7.20e-03 |
Pos logit contributions | |
His | +0.008 |
HIS | +0.007 |
his | +0.007 |
his | +0.007 |
He | +0.007 |
Neg logit contributions | |
committee | -0.008 |
contestant | -0.008 |
brawl | -0.008 |
rookies | -0.007 |
duel | -0.007 |
Token | His |
Token ID | 6653 |
Feature activation | -7.47e-01 |
Pos logit contributions | |
His | +0.006 |
HIS | +0.006 |
his | +0.005 |
his | +0.005 |
He | +0.005 |
Neg logit contributions | |
committee | -0.008 |
contestant | -0.007 |
brawl | -0.007 |
rookies | -0.007 |
duel | -0.007 |
Token | skin |
Token ID | 4168 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
His | +0.704 |
HIS | +0.652 |
his | +0.633 |
his | +0.630 |
He | +0.622 |
Neg logit contributions | |
committee | -0.699 |
contestant | -0.692 |
brawl | -0.665 |
rookies | -0.633 |
duel | -0.623 |
Token | sl |
Token ID | 1017 |
Feature activation | -6.59e-03 |
Pos logit contributions | |
His | +0.167 |
HIS | +0.154 |
his | +0.150 |
his | +0.149 |
He | +0.147 |
Neg logit contributions | |
committee | -0.168 |
contestant | -0.167 |
brawl | -0.160 |
rookies | -0.153 |
duel | -0.151 |
Token | oug |
Token ID | 20805 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
His | +0.006 |
HIS | +0.006 |
his | +0.006 |
his | +0.006 |
He | +0.006 |
Neg logit contributions | |
committee | -0.006 |
contestant | -0.006 |
brawl | -0.006 |
rookies | -0.006 |
duel | -0.005 |
Token | hed |
Token ID | 704 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
His | +0.016 |
HIS | +0.015 |
his | +0.014 |
his | +0.014 |
He | +0.014 |
Neg logit contributions | |
committee | -0.021 |
contestant | -0.021 |
brawl | -0.020 |
rookies | -0.019 |
duel | -0.019 |
Token | off |
Token ID | 572 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
His | +0.099 |
HIS | +0.090 |
his | +0.087 |
He | +0.085 |
his | +0.085 |
Neg logit contributions | |
committee | -0.142 |
contestant | -0.142 |
brawl | -0.137 |
rookies | -0.132 |
duel | -0.130 |
Token | oct |
Token ID | 19318 |
Feature activation | -8.85e-02 |
Pos logit contributions | |
His | +0.016 |
HIS | +0.014 |
his | +0.014 |
his | +0.014 |
He | +0.014 |
Neg logit contributions | |
committee | -0.017 |
contestant | -0.017 |
brawl | -0.017 |
rookies | -0.016 |
duel | -0.016 |
Token | opus |
Token ID | 25790 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
His | +0.101 |
HIS | +0.095 |
his | +0.093 |
his | +0.092 |
He | +0.091 |
Neg logit contributions | |
committee | -0.065 |
contestant | -0.064 |
brawl | -0.061 |
rookies | -0.058 |
duel | -0.056 |
Token | ." |
Token ID | 526 |
Feature activation | -7.76e-02 |
Pos logit contributions | |
committee | +0.133 |
contestant | +0.131 |
brawl | +0.126 |
rookies | +0.120 |
duel | +0.119 |
Neg logit contributions | |
His | -0.124 |
HIS | -0.114 |
his | -0.111 |
his | -0.110 |
He | -0.109 |
Token |
|
Token ID | 198 |
Feature activation | -4.68e-03 |
Pos logit contributions | |
His | +0.072 |
HIS | +0.067 |
his | +0.065 |
his | +0.064 |
He | +0.064 |
Neg logit contributions | |
committee | -0.074 |
contestant | -0.073 |
brawl | -0.071 |
rookies | -0.067 |
duel | -0.066 |
Token |
|
Token ID | 198 |
Feature activation | -5.82e-03 |
Pos logit contributions | |
His | +0.004 |
HIS | +0.004 |
his | +0.004 |
his | +0.004 |
He | +0.004 |
Neg logit contributions | |
committee | -0.004 |
contestant | -0.004 |
brawl | -0.004 |
rookies | -0.004 |
duel | -0.004 |
Token | His |
Token ID | 6653 |
Feature activation | -7.46e-01 |
Pos logit contributions | |
His | +0.005 |
HIS | +0.004 |
his | +0.004 |
He | +0.004 |
his | +0.004 |
Neg logit contributions | |
committee | -0.006 |
contestant | -0.006 |
brawl | -0.006 |
rookies | -0.006 |
duel | -0.006 |
Token | comments |
Token ID | 3651 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
His | +0.745 |
HIS | +0.695 |
his | +0.678 |
his | +0.676 |
He | +0.666 |
Neg logit contributions | |
committee | -0.656 |
contestant | -0.639 |
brawl | -0.616 |
rookies | -0.581 |
duel | -0.573 |
Token | caused |
Token ID | 4073 |
Feature activation | +9.78e-02 |
Pos logit contributions | |
His | +0.156 |
HIS | +0.146 |
his | +0.145 |
his | +0.143 |
him | +0.143 |
Neg logit contributions | |
committee | -0.096 |
contestant | -0.094 |
brawl | -0.088 |
rookies | -0.083 |
duel | -0.080 |
Token | confusion |
Token ID | 10802 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
committee | +0.082 |
contestant | +0.081 |
brawl | +0.077 |
rookies | +0.073 |
duel | +0.072 |
Neg logit contributions | |
His | -0.102 |
HIS | -0.095 |
his | -0.093 |
his | -0.093 |
him | -0.092 |
Token | among |
Token ID | 1871 |
Feature activation | +8.19e-02 |
Pos logit contributions | |
committee | +0.137 |
contestant | +0.136 |
brawl | +0.131 |
rookies | +0.124 |
duel | +0.122 |
Neg logit contributions | |
His | -0.135 |
HIS | -0.126 |
his | -0.122 |
his | -0.121 |
He | -0.120 |
Token | fans |
Token ID | 3296 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
committee | +0.075 |
contestant | +0.074 |
brawl | +0.071 |
rookies | +0.067 |
duel | +0.066 |
Neg logit contributions | |
His | -0.079 |
HIS | -0.074 |
his | -0.072 |
his | -0.072 |
him | -0.071 |
Token | May |
Token ID | 1737 |
Feature activation | +6.50e-02 |
Pos logit contributions | |
committee | +0.008 |
contestant | +0.008 |
brawl | +0.008 |
rookies | +0.008 |
duel | +0.008 |
Neg logit contributions | |
His | -0.008 |
HIS | -0.008 |
his | -0.008 |
his | -0.008 |
He | -0.007 |
Token | 2010 |
Token ID | 3050 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
committee | +0.066 |
contestant | +0.065 |
brawl | +0.063 |
rookies | +0.060 |
duel | +0.059 |
Neg logit contributions | |
His | -0.056 |
HIS | -0.052 |
his | -0.050 |
his | -0.050 |
He | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
His | +0.161 |
HIS | +0.149 |
his | +0.145 |
his | +0.145 |
He | +0.143 |
Neg logit contributions | |
committee | -0.159 |
contestant | -0.156 |
brawl | -0.150 |
rookies | -0.143 |
duel | -0.141 |
Token |
|
Token ID | 198 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
committee | +0.019 |
contestant | +0.019 |
brawl | +0.018 |
rookies | +0.017 |
duel | +0.017 |
Neg logit contributions | |
His | -0.018 |
HIS | -0.016 |
his | -0.016 |
his | -0.016 |
He | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | -4.79e-03 |
Pos logit contributions | |
His | +0.014 |
HIS | +0.013 |
his | +0.012 |
his | +0.012 |
He | +0.012 |
Neg logit contributions | |
committee | -0.014 |
contestant | -0.013 |
brawl | -0.013 |
rookies | -0.012 |
duel | -0.012 |
Token | His |
Token ID | 6653 |
Feature activation | -7.45e-01 |
Pos logit contributions | |
His | +0.003 |
HIS | +0.003 |
his | +0.003 |
his | +0.003 |
him | +0.003 |
Neg logit contributions | |
committee | -0.006 |
contestant | -0.006 |
brawl | -0.005 |
rookies | -0.005 |
duel | -0.005 |
Token | pen |
Token ID | 3112 |
Feature activation | -2.58e-02 |
Pos logit contributions | |
His | +0.750 |
HIS | +0.701 |
his | +0.681 |
his | +0.679 |
He | +0.669 |
Neg logit contributions | |
committee | -0.649 |
contestant | -0.640 |
brawl | -0.611 |
rookies | -0.577 |
duel | -0.567 |
Token | ultimate |
Token ID | 44818 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
His | +0.029 |
HIS | +0.027 |
his | +0.027 |
his | +0.026 |
He | +0.026 |
Neg logit contributions | |
committee | -0.019 |
contestant | -0.019 |
brawl | -0.018 |
rookies | -0.017 |
duel | -0.017 |
Token | game |
Token ID | 983 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
committee | +0.118 |
contestant | +0.115 |
brawl | +0.110 |
rookies | +0.104 |
duel | +0.102 |
Neg logit contributions | |
His | -0.146 |
HIS | -0.137 |
his | -0.133 |
his | -0.133 |
He | -0.131 |
Token | was |
Token ID | 373 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
committee | +0.157 |
contestant | +0.155 |
brawl | +0.149 |
rookies | +0.142 |
duel | +0.140 |
Neg logit contributions | |
His | -0.150 |
HIS | -0.138 |
his | -0.134 |
his | -0.134 |
He | -0.132 |
Token | a |
Token ID | 257 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
His | +0.117 |
HIS | +0.110 |
his | +0.107 |
his | +0.107 |
him | +0.105 |
Neg logit contributions | |
committee | -0.104 |
contestant | -0.102 |
brawl | -0.097 |
rookies | -0.092 |
duel | -0.091 |
Token | Opp |
Token ID | 9385 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
committee | +0.015 |
contestant | +0.015 |
brawl | +0.014 |
rookies | +0.013 |
duel | +0.013 |
Neg logit contributions | |
His | -0.016 |
HIS | -0.015 |
his | -0.015 |
his | -0.015 |
He | -0.014 |
Token | ressed |
Token ID | 2790 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
committee | +0.182 |
contestant | +0.182 |
brawl | +0.172 |
rookies | +0.162 |
duel | +0.157 |
Neg logit contributions | |
His | -0.316 |
HIS | -0.295 |
his | -0.292 |
his | -0.289 |
He | -0.288 |
Token | .") |
Token ID | 19570 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
committee | +0.217 |
contestant | +0.216 |
brawl | +0.209 |
rookies | +0.199 |
duel | +0.197 |
Neg logit contributions | |
His | -0.159 |
HIS | -0.146 |
his | -0.141 |
his | -0.139 |
He | -0.138 |
Token |
|
Token ID | 198 |
Feature activation | -5.56e-03 |
Pos logit contributions | |
His | +0.191 |
HIS | +0.180 |
his | +0.173 |
his | +0.172 |
He | +0.171 |
Neg logit contributions | |
committee | -0.171 |
contestant | -0.170 |
brawl | -0.162 |
rookies | -0.153 |
duel | -0.151 |
Token |
|
Token ID | 198 |
Feature activation | -8.51e-03 |
Pos logit contributions | |
His | +0.005 |
HIS | +0.005 |
his | +0.005 |
his | +0.005 |
He | +0.005 |
Neg logit contributions | |
committee | -0.005 |
contestant | -0.005 |
brawl | -0.005 |
rookies | -0.005 |
duel | -0.005 |
Token | His |
Token ID | 6653 |
Feature activation | -7.43e-01 |
Pos logit contributions | |
His | +0.007 |
HIS | +0.006 |
his | +0.006 |
He | +0.006 |
his | +0.006 |
Neg logit contributions | |
contestant | -0.009 |
committee | -0.009 |
brawl | -0.009 |
rookies | -0.008 |
duel | -0.008 |
Token | ac |
Token ID | 936 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
His | +0.760 |
HIS | +0.713 |
his | +0.693 |
his | +0.690 |
He | +0.681 |
Neg logit contributions | |
committee | -0.633 |
contestant | -0.628 |
brawl | -0.599 |
rookies | -0.567 |
duel | -0.556 |
Token | oly |
Token ID | 3366 |
Feature activation | +4.67e-02 |
Pos logit contributions | |
His | +0.137 |
HIS | +0.127 |
his | +0.125 |
his | +0.123 |
He | +0.122 |
Neg logit contributions | |
committee | -0.109 |
contestant | -0.109 |
brawl | -0.104 |
rookies | -0.098 |
duel | -0.096 |
Token | tes |
Token ID | 4879 |
Feature activation | +7.55e-02 |
Pos logit contributions | |
committee | +0.039 |
contestant | +0.039 |
brawl | +0.037 |
rookies | +0.035 |
duel | +0.035 |
Neg logit contributions | |
His | -0.049 |
HIS | -0.045 |
his | -0.044 |
his | -0.044 |
He | -0.044 |
Token | explain |
Token ID | 4727 |
Feature activation | +9.91e-03 |
Pos logit contributions | |
committee | +0.065 |
contestant | +0.064 |
brawl | +0.061 |
rookies | +0.058 |
duel | +0.057 |
Neg logit contributions | |
His | -0.077 |
HIS | -0.072 |
his | -0.070 |
his | -0.070 |
him | -0.069 |
Token | the |
Token ID | 262 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
committee | +0.010 |
contestant | +0.010 |
brawl | +0.010 |
rookies | +0.009 |
duel | +0.009 |
Neg logit contributions | |
His | -0.009 |
HIS | -0.008 |
his | -0.008 |
his | -0.008 |
He | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +3.69e-02 |
Pos logit contributions | |
committee | +0.062 |
contestant | +0.061 |
brawl | +0.059 |
rookies | +0.056 |
duel | +0.055 |
Neg logit contributions | |
His | -0.063 |
HIS | -0.058 |
his | -0.056 |
his | -0.056 |
He | -0.055 |
Token | � |
Token ID | 447 |
Feature activation | +2.64e-03 |
Pos logit contributions | |
committee | +0.035 |
contestant | +0.035 |
brawl | +0.034 |
rookies | +0.032 |
duel | +0.032 |
Neg logit contributions | |
His | -0.034 |
HIS | -0.032 |
his | -0.031 |
his | -0.030 |
He | -0.030 |
Token | � |
Token ID | 251 |
Feature activation | -6.38e-04 |
Pos logit contributions | |
committee | +0.002 |
contestant | +0.002 |
brawl | +0.002 |
rookies | +0.002 |
duel | +0.002 |
Neg logit contributions | |
His | -0.003 |
HIS | -0.003 |
his | -0.002 |
his | -0.002 |
He | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | +4.69e-03 |
Pos logit contributions | |
His | +0.001 |
HIS | +0.001 |
his | +0.001 |
his | +0.001 |
He | +0.001 |
Neg logit contributions | |
committee | -0.001 |
contestant | -0.001 |
brawl | -0.001 |
rookies | -0.001 |
duel | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +3.67e-03 |
Pos logit contributions | |
committee | +0.004 |
contestant | +0.004 |
brawl | +0.004 |
rookies | +0.004 |
duel | +0.004 |
Neg logit contributions | |
His | -0.004 |
HIS | -0.004 |
his | -0.004 |
his | -0.004 |
He | -0.004 |
Token | His |
Token ID | 6653 |
Feature activation | -7.40e-01 |
Pos logit contributions | |
committee | +0.004 |
contestant | +0.004 |
brawl | +0.004 |
rookies | +0.004 |
duel | +0.004 |
Neg logit contributions | |
His | -0.003 |
HIS | -0.003 |
his | -0.003 |
his | -0.003 |
He | -0.003 |
Token | comments |
Token ID | 3651 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
His | +0.764 |
HIS | +0.715 |
his | +0.697 |
his | +0.695 |
He | +0.686 |
Neg logit contributions | |
committee | -0.623 |
contestant | -0.617 |
brawl | -0.589 |
rookies | -0.555 |
duel | -0.543 |
Token | seemed |
Token ID | 3947 |
Feature activation | +1.96e-03 |
Pos logit contributions | |
His | +0.128 |
HIS | +0.119 |
his | +0.118 |
his | +0.117 |
him | +0.116 |
Neg logit contributions | |
committee | -0.092 |
contestant | -0.090 |
brawl | -0.085 |
rookies | -0.080 |
duel | -0.078 |
Token | to |
Token ID | 284 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
committee | +0.002 |
contestant | +0.002 |
brawl | +0.002 |
rookies | +0.002 |
duel | +0.001 |
Neg logit contributions | |
His | -0.002 |
HIS | -0.002 |
his | -0.002 |
his | -0.002 |
him | -0.002 |
Token | put |
Token ID | 1234 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
committee | +0.094 |
contestant | +0.094 |
brawl | +0.090 |
rookies | +0.085 |
duel | +0.084 |
Neg logit contributions | |
His | -0.096 |
HIS | -0.089 |
his | -0.086 |
his | -0.086 |
He | -0.085 |
Token | McDonnell |
Token ID | 39441 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
His | +0.114 |
HIS | +0.105 |
his | +0.102 |
his | +0.101 |
He | +0.100 |
Neg logit contributions | |
committee | -0.127 |
contestant | -0.126 |
brawl | -0.121 |
rookies | -0.115 |
duel | -0.114 |
Token | British |
Token ID | 3517 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
committee | +0.021 |
contestant | +0.021 |
brawl | +0.020 |
rookies | +0.019 |
duel | +0.019 |
Neg logit contributions | |
His | -0.024 |
HIS | -0.022 |
his | -0.021 |
his | -0.021 |
He | -0.021 |
Token | establishment |
Token ID | 9323 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
His | +0.114 |
HIS | +0.105 |
his | +0.101 |
his | +0.101 |
He | +0.100 |
Neg logit contributions | |
committee | -0.127 |
contestant | -0.125 |
brawl | -0.121 |
rookies | -0.116 |
duel | -0.114 |
Token | . |
Token ID | 13 |
Feature activation | +3.83e-02 |
Pos logit contributions | |
committee | +0.100 |
contestant | +0.099 |
brawl | +0.095 |
rookies | +0.091 |
duel | +0.090 |
Neg logit contributions | |
His | -0.097 |
HIS | -0.089 |
his | -0.087 |
his | -0.086 |
He | -0.085 |
Token |
|
Token ID | 198 |
Feature activation | +3.25e-03 |
Pos logit contributions | |
committee | +0.036 |
contestant | +0.036 |
brawl | +0.035 |
rookies | +0.033 |
duel | +0.033 |
Neg logit contributions | |
His | -0.035 |
HIS | -0.033 |
his | -0.032 |
his | -0.031 |
He | -0.031 |
Token |
|
Token ID | 198 |
Feature activation | +5.09e-03 |
Pos logit contributions | |
committee | +0.003 |
contestant | +0.003 |
brawl | +0.003 |
rookies | +0.003 |
duel | +0.003 |
Neg logit contributions | |
His | -0.003 |
HIS | -0.003 |
his | -0.003 |
his | -0.003 |
He | -0.003 |
Token | His |
Token ID | 6653 |
Feature activation | -7.35e-01 |
Pos logit contributions | |
committee | +0.006 |
contestant | +0.006 |
brawl | +0.006 |
rookies | +0.005 |
duel | +0.005 |
Neg logit contributions | |
His | -0.004 |
HIS | -0.003 |
his | -0.003 |
his | -0.003 |
him | -0.003 |
Token | family |
Token ID | 1641 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
His | +0.663 |
HIS | +0.614 |
his | +0.595 |
his | +0.590 |
He | +0.584 |
Neg logit contributions | |
committee | -0.717 |
contestant | -0.711 |
brawl | -0.684 |
rookies | -0.652 |
duel | -0.642 |
Token | 's |
Token ID | 338 |
Feature activation | +8.24e-02 |
Pos logit contributions | |
His | +0.169 |
HIS | +0.156 |
his | +0.152 |
his | +0.149 |
He | +0.148 |
Neg logit contributions | |
committee | -0.175 |
contestant | -0.174 |
brawl | -0.168 |
rookies | -0.160 |
duel | -0.158 |
Token | Stellar |
Token ID | 39336 |
Feature activation | -3.89e-02 |
Pos logit contributions | |
committee | +0.089 |
contestant | +0.088 |
brawl | +0.085 |
rookies | +0.082 |
duel | +0.081 |
Neg logit contributions | |
His | -0.066 |
HIS | -0.060 |
his | -0.058 |
his | -0.057 |
He | -0.057 |
Token | International |
Token ID | 4037 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
His | +0.033 |
HIS | +0.030 |
his | +0.029 |
his | +0.029 |
He | +0.029 |
Neg logit contributions | |
committee | -0.040 |
contestant | -0.040 |
brawl | -0.038 |
rookies | -0.037 |
duel | -0.036 |
Token | Art |
Token ID | 3683 |
Feature activation | -7.08e-02 |
Pos logit contributions | |
His | +0.156 |
HIS | +0.143 |
his | +0.139 |
his | +0.136 |
He | +0.135 |
Neg logit contributions | |
committee | -0.192 |
contestant | -0.191 |
brawl | -0.185 |
rookies | -0.176 |
duel | -0.174 |
Token | felony |
Token ID | 14544 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
committee | +0.085 |
contestant | +0.085 |
brawl | +0.081 |
rookies | +0.077 |
duel | +0.077 |
Neg logit contributions | |
His | -0.080 |
HIS | -0.074 |
his | -0.072 |
his | -0.071 |
He | -0.071 |
Token | charge |
Token ID | 3877 |
Feature activation | +3.47e-02 |
Pos logit contributions | |
committee | +0.161 |
contestant | +0.160 |
brawl | +0.154 |
rookies | +0.146 |
duel | +0.144 |
Neg logit contributions | |
His | -0.169 |
HIS | -0.157 |
his | -0.153 |
his | -0.151 |
He | -0.150 |
Token | . |
Token ID | 13 |
Feature activation | +3.39e-02 |
Pos logit contributions | |
committee | +0.034 |
contestant | +0.033 |
brawl | +0.032 |
rookies | +0.031 |
duel | +0.030 |
Neg logit contributions | |
His | -0.032 |
HIS | -0.029 |
his | -0.028 |
his | -0.028 |
He | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | +3.89e-03 |
Pos logit contributions | |
committee | +0.035 |
contestant | +0.034 |
brawl | +0.033 |
rookies | +0.032 |
duel | +0.031 |
Neg logit contributions | |
His | -0.029 |
HIS | -0.026 |
his | -0.026 |
his | -0.025 |
He | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | +6.42e-03 |
Pos logit contributions | |
committee | +0.004 |
contestant | +0.004 |
brawl | +0.003 |
rookies | +0.003 |
duel | +0.003 |
Neg logit contributions | |
His | -0.004 |
HIS | -0.003 |
his | -0.003 |
his | -0.003 |
He | -0.003 |
Token | His |
Token ID | 6653 |
Feature activation | -7.33e-01 |
Pos logit contributions | |
committee | +0.007 |
contestant | +0.007 |
brawl | +0.007 |
rookies | +0.007 |
duel | +0.006 |
Neg logit contributions | |
His | -0.005 |
HIS | -0.004 |
his | -0.004 |
his | -0.004 |
him | -0.004 |
Token | company |
Token ID | 1664 |
Feature activation | -3.07e-01 |
Pos logit contributions | |
His | +0.665 |
HIS | +0.616 |
his | +0.597 |
his | +0.590 |
He | +0.585 |
Neg logit contributions | |
committee | -0.711 |
contestant | -0.705 |
brawl | -0.678 |
rookies | -0.646 |
duel | -0.637 |
Token | , |
Token ID | 11 |
Feature activation | -2.79e-02 |
Pos logit contributions | |
His | +0.282 |
HIS | +0.261 |
his | +0.254 |
his | +0.249 |
He | +0.248 |
Neg logit contributions | |
committee | -0.295 |
contestant | -0.293 |
brawl | -0.282 |
rookies | -0.268 |
duel | -0.265 |
Token | Fog |
Token ID | 33381 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
His | +0.024 |
HIS | +0.022 |
his | +0.021 |
his | +0.021 |
He | +0.021 |
Neg logit contributions | |
committee | -0.029 |
contestant | -0.028 |
brawl | -0.027 |
rookies | -0.026 |
duel | -0.026 |
Token | Cutter |
Token ID | 49160 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
His | +0.126 |
HIS | +0.116 |
his | +0.112 |
his | +0.110 |
He | +0.109 |
Neg logit contributions | |
committee | -0.153 |
contestant | -0.152 |
brawl | -0.147 |
rookies | -0.140 |
duel | -0.139 |
Token | Capitol |
Token ID | 13241 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
committee | +0.136 |
contestant | +0.135 |
brawl | +0.130 |
rookies | +0.125 |
duel | +0.123 |
Neg logit contributions | |
His | -0.110 |
HIS | -0.101 |
his | -0.098 |
his | -0.096 |
He | -0.095 |
Token | the |
Token ID | 262 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
committee | +0.251 |
contestant | +0.251 |
brawl | +0.241 |
rookies | +0.230 |
duel | +0.227 |
Neg logit contributions | |
His | -0.228 |
HIS | -0.210 |
his | -0.205 |
He | -0.200 |
his | -0.200 |
Token | mortgage |
Token ID | 13682 |
Feature activation | -2.90e-01 |
Pos logit contributions | |
committee | +0.023 |
contestant | +0.023 |
brawl | +0.022 |
rookies | +0.021 |
duel | +0.021 |
Neg logit contributions | |
His | -0.017 |
HIS | -0.015 |
his | -0.015 |
his | -0.014 |
He | -0.014 |
Token | market |
Token ID | 1910 |
Feature activation | -3.66e-01 |
Pos logit contributions | |
His | +0.225 |
HIS | +0.205 |
his | +0.198 |
his | +0.195 |
He | +0.194 |
Neg logit contributions | |
committee | -0.320 |
contestant | -0.318 |
brawl | -0.307 |
rookies | -0.294 |
duel | -0.291 |
Token | . |
Token ID | 13 |
Feature activation | +3.32e-02 |
Pos logit contributions | |
His | +0.343 |
HIS | +0.318 |
his | +0.310 |
his | +0.309 |
He | +0.304 |
Neg logit contributions | |
committee | -0.345 |
contestant | -0.341 |
brawl | -0.327 |
rookies | -0.311 |
duel | -0.307 |
Token | ( |
Token ID | 357 |
Feature activation | +1.98e-02 |
Pos logit contributions | |
committee | +0.031 |
contestant | +0.031 |
brawl | +0.029 |
rookies | +0.028 |
duel | +0.028 |
Neg logit contributions | |
His | -0.032 |
HIS | -0.029 |
his | -0.028 |
his | -0.028 |
He | -0.028 |
Token | His |
Token ID | 6653 |
Feature activation | -7.32e-01 |
Pos logit contributions | |
committee | +0.021 |
contestant | +0.021 |
brawl | +0.020 |
rookies | +0.020 |
duel | +0.019 |
Neg logit contributions | |
His | -0.016 |
HIS | -0.014 |
his | -0.014 |
his | -0.014 |
He | -0.014 |
Token | personal |
Token ID | 2614 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
His | +0.646 |
HIS | +0.597 |
his | +0.578 |
his | +0.573 |
He | +0.567 |
Neg logit contributions | |
committee | -0.729 |
contestant | -0.722 |
brawl | -0.695 |
rookies | -0.664 |
duel | -0.654 |
Token | take |
Token ID | 1011 |
Feature activation | +4.34e-02 |
Pos logit contributions | |
His | +0.088 |
HIS | +0.079 |
his | +0.077 |
his | +0.075 |
He | +0.075 |
Neg logit contributions | |
committee | -0.135 |
contestant | -0.134 |
brawl | -0.129 |
rookies | -0.124 |
duel | -0.123 |
Token | was |
Token ID | 373 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
committee | +0.041 |
contestant | +0.041 |
brawl | +0.039 |
rookies | +0.037 |
duel | +0.037 |
Neg logit contributions | |
His | -0.040 |
HIS | -0.037 |
his | -0.036 |
his | -0.036 |
He | -0.035 |
Token | nearly |
Token ID | 3016 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
His | +0.086 |
HIS | +0.079 |
his | +0.076 |
his | +0.075 |
He | +0.074 |
Neg logit contributions | |
committee | -0.115 |
contestant | -0.115 |
brawl | -0.111 |
rookies | -0.106 |
duel | -0.105 |
Token | $ |
Token ID | 720 |
Feature activation | -4.98e-01 |
Pos logit contributions | |
His | +0.017 |
HIS | +0.015 |
his | +0.014 |
his | +0.013 |
He | +0.013 |
Neg logit contributions | |
committee | -0.038 |
contestant | -0.038 |
brawl | -0.037 |
rookies | -0.036 |
duel | -0.036 |
Token | , |
Token ID | 11 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
committee | +0.162 |
contestant | +0.160 |
brawl | +0.154 |
rookies | +0.147 |
duel | +0.144 |
Neg logit contributions | |
His | -0.161 |
HIS | -0.150 |
his | -0.145 |
his | -0.144 |
He | -0.143 |
Token | in |
Token ID | 287 |
Feature activation | +3.45e-02 |
Pos logit contributions | |
His | +0.015 |
HIS | +0.014 |
his | +0.014 |
his | +0.014 |
He | +0.014 |
Neg logit contributions | |
committee | -0.014 |
contestant | -0.013 |
brawl | -0.013 |
rookies | -0.012 |
duel | -0.012 |
Token | person |
Token ID | 1048 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
committee | +0.033 |
contestant | +0.033 |
brawl | +0.031 |
rookies | +0.030 |
duel | +0.029 |
Neg logit contributions | |
His | -0.032 |
HIS | -0.030 |
his | -0.029 |
his | -0.028 |
He | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | +4.59e-02 |
Pos logit contributions | |
committee | +0.137 |
contestant | +0.136 |
brawl | +0.130 |
rookies | +0.123 |
duel | +0.121 |
Neg logit contributions | |
His | -0.143 |
HIS | -0.133 |
his | -0.129 |
his | -0.129 |
He | -0.127 |
Token | ( |
Token ID | 357 |
Feature activation | +3.88e-02 |
Pos logit contributions | |
committee | +0.045 |
contestant | +0.045 |
brawl | +0.043 |
rookies | +0.041 |
duel | +0.041 |
Neg logit contributions | |
His | -0.041 |
HIS | -0.038 |
his | -0.037 |
his | -0.036 |
He | -0.036 |
Token | His |
Token ID | 6653 |
Feature activation | -7.24e-01 |
Pos logit contributions | |
committee | +0.043 |
contestant | +0.042 |
brawl | +0.041 |
rookies | +0.039 |
duel | +0.039 |
Neg logit contributions | |
His | -0.030 |
HIS | -0.028 |
his | -0.027 |
his | -0.026 |
He | -0.026 |
Token | created |
Token ID | 2727 |
Feature activation | -4.46e-02 |
Pos logit contributions | |
His | +0.665 |
HIS | +0.616 |
his | +0.597 |
his | +0.592 |
He | +0.586 |
Neg logit contributions | |
committee | -0.695 |
contestant | -0.688 |
brawl | -0.662 |
rookies | -0.630 |
duel | -0.621 |
Token | , |
Token ID | 11 |
Feature activation | -9.43e-03 |
Pos logit contributions | |
His | +0.043 |
HIS | +0.040 |
his | +0.039 |
his | +0.039 |
He | +0.039 |
Neg logit contributions | |
committee | -0.040 |
contestant | -0.040 |
brawl | -0.038 |
rookies | -0.036 |
duel | -0.036 |
Token | over |
Token ID | 625 |
Feature activation | -2.18e-01 |
Pos logit contributions | |
His | +0.009 |
HIS | +0.008 |
his | +0.008 |
his | +0.008 |
He | +0.008 |
Neg logit contributions | |
committee | -0.009 |
contestant | -0.009 |
brawl | -0.008 |
rookies | -0.008 |
duel | -0.008 |
Token | - |
Token ID | 12 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
His | +0.204 |
HIS | +0.189 |
his | +0.184 |
his | +0.183 |
He | +0.181 |
Neg logit contributions | |
committee | -0.204 |
contestant | -0.203 |
brawl | -0.195 |
rookies | -0.186 |
duel | -0.183 |
Token | the |
Token ID | 1169 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
committee | +0.084 |
contestant | +0.084 |
brawl | +0.080 |
rookies | +0.076 |
duel | +0.075 |
Neg logit contributions | |
His | -0.104 |
HIS | -0.097 |
his | -0.095 |
his | -0.093 |
He | -0.093 |
Token | time |
Token ID | 640 |
Feature activation | -5.81e-02 |
Pos logit contributions | |
His | +0.008 |
HIS | +0.007 |
his | +0.007 |
his | +0.007 |
He | +0.007 |
Neg logit contributions | |
committee | -0.009 |
contestant | -0.009 |
brawl | -0.009 |
rookies | -0.008 |
duel | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | +8.48e-02 |
Pos logit contributions | |
His | +0.055 |
HIS | +0.051 |
his | +0.049 |
his | +0.049 |
He | +0.049 |
Neg logit contributions | |
committee | -0.054 |
contestant | -0.054 |
brawl | -0.052 |
rookies | -0.049 |
duel | -0.048 |
Token | some |
Token ID | 617 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
committee | +0.083 |
contestant | +0.083 |
brawl | +0.080 |
rookies | +0.076 |
duel | +0.075 |
Neg logit contributions | |
His | -0.076 |
HIS | -0.070 |
his | -0.068 |
his | -0.067 |
He | -0.067 |
Token | degree |
Token ID | 4922 |
Feature activation | +7.74e-02 |
Pos logit contributions | |
committee | +0.123 |
contestant | +0.122 |
brawl | +0.117 |
rookies | +0.111 |
duel | +0.110 |
Neg logit contributions | |
His | -0.113 |
HIS | -0.104 |
his | -0.102 |
his | -0.101 |
He | -0.100 |
Token | ( |
Token ID | 357 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
committee | +0.075 |
contestant | +0.075 |
brawl | +0.072 |
rookies | +0.068 |
duel | +0.067 |
Neg logit contributions | |
His | -0.070 |
HIS | -0.065 |
his | -0.063 |
his | -0.062 |
He | -0.062 |
Token | his |
Token ID | 14363 |
Feature activation | -7.20e-01 |
Pos logit contributions | |
committee | +0.012 |
contestant | +0.012 |
brawl | +0.011 |
rookies | +0.011 |
duel | +0.010 |
Neg logit contributions | |
His | -0.011 |
HIS | -0.011 |
his | -0.010 |
his | -0.010 |
He | -0.010 |
Token | abhor |
Token ID | 42934 |
Feature activation | +9.74e-02 |
Pos logit contributions | |
His | +0.675 |
HIS | +0.626 |
his | +0.608 |
his | +0.604 |
He | +0.597 |
Neg logit contributions | |
committee | -0.678 |
contestant | -0.670 |
brawl | -0.644 |
rookies | -0.612 |
duel | -0.601 |
Token | rence |
Token ID | 6784 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
committee | +0.081 |
contestant | +0.081 |
brawl | +0.077 |
rookies | +0.072 |
duel | +0.071 |
Neg logit contributions | |
His | -0.102 |
HIS | -0.095 |
his | -0.093 |
his | -0.093 |
He | -0.092 |
Token | of |
Token ID | 286 |
Feature activation | -1.44e-03 |
Pos logit contributions | |
committee | +0.113 |
contestant | +0.112 |
brawl | +0.108 |
rookies | +0.102 |
duel | +0.101 |
Neg logit contributions | |
His | -0.107 |
HIS | -0.099 |
his | -0.096 |
his | -0.095 |
He | -0.094 |
Token | modern |
Token ID | 3660 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
His | +0.001 |
HIS | +0.001 |
his | +0.001 |
his | +0.001 |
He | +0.001 |
Neg logit contributions | |
committee | -0.001 |
contestant | -0.001 |
brawl | -0.001 |
rookies | -0.001 |
duel | -0.001 |
Token | literature |
Token ID | 9285 |
Feature activation | -8.09e-02 |
Pos logit contributions | |
His | +0.092 |
HIS | +0.085 |
his | +0.082 |
his | +0.081 |
He | +0.080 |
Neg logit contributions | |
committee | -0.102 |
contestant | -0.101 |
brawl | -0.097 |
rookies | -0.093 |
duel | -0.091 |
Token | " |
Token ID | 366 |
Feature activation | -5.04e-02 |
Pos logit contributions | |
His | +0.218 |
HIS | +0.201 |
his | +0.195 |
his | +0.193 |
He | +0.191 |
Neg logit contributions | |
committee | -0.243 |
contestant | -0.241 |
brawl | -0.232 |
rookies | -0.221 |
duel | -0.218 |
Token | he |
Token ID | 258 |
Feature activation | -3.70e-01 |
Pos logit contributions | |
His | +0.041 |
HIS | +0.038 |
his | +0.036 |
his | +0.036 |
He | +0.036 |
Neg logit contributions | |
committee | -0.053 |
contestant | -0.053 |
brawl | -0.051 |
rookies | -0.049 |
duel | -0.048 |
Token | " |
Token ID | 1 |
Feature activation | -8.83e-02 |
Pos logit contributions | |
His | +0.302 |
HIS | +0.279 |
his | +0.268 |
his | +0.262 |
He | +0.261 |
Neg logit contributions | |
committee | -0.389 |
contestant | -0.387 |
brawl | -0.373 |
rookies | -0.356 |
duel | -0.353 |
Token | and |
Token ID | 290 |
Feature activation | +8.32e-02 |
Pos logit contributions | |
His | +0.089 |
HIS | +0.083 |
his | +0.081 |
his | +0.080 |
He | +0.080 |
Neg logit contributions | |
committee | -0.077 |
contestant | -0.076 |
brawl | -0.073 |
rookies | -0.069 |
duel | -0.068 |
Token | " |
Token ID | 366 |
Feature activation | -4.64e-02 |
Pos logit contributions | |
committee | +0.082 |
contestant | +0.081 |
brawl | +0.078 |
rookies | +0.075 |
duel | +0.074 |
Neg logit contributions | |
His | -0.074 |
HIS | -0.069 |
his | -0.067 |
his | -0.065 |
He | -0.065 |
Token | his |
Token ID | 14363 |
Feature activation | -7.10e-01 |
Pos logit contributions | |
His | +0.031 |
HIS | +0.029 |
his | +0.027 |
his | +0.027 |
He | +0.026 |
Neg logit contributions | |
committee | -0.055 |
contestant | -0.055 |
brawl | -0.053 |
rookies | -0.051 |
duel | -0.050 |
Token | "). |
Token ID | 11074 |
Feature activation | -1.91e-01 |
Pos logit contributions | |
His | +0.531 |
HIS | +0.484 |
his | +0.465 |
his | +0.456 |
He | +0.453 |
Neg logit contributions | |
committee | -0.799 |
contestant | -0.795 |
brawl | -0.769 |
rookies | -0.737 |
duel | -0.730 |
Token | Male |
Token ID | 12674 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
His | +0.202 |
HIS | +0.189 |
his | +0.183 |
his | +0.183 |
He | +0.181 |
Neg logit contributions | |
committee | -0.161 |
contestant | -0.158 |
brawl | -0.151 |
rookies | -0.143 |
duel | -0.141 |
Token | - |
Token ID | 12 |
Feature activation | +8.64e-02 |
Pos logit contributions | |
committee | +0.116 |
contestant | +0.114 |
brawl | +0.109 |
rookies | +0.103 |
duel | +0.102 |
Neg logit contributions | |
His | -0.128 |
HIS | -0.119 |
his | -0.116 |
his | -0.115 |
He | -0.114 |
Token | to |
Token ID | 1462 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
contestant | +0.067 |
committee | +0.067 |
brawl | +0.065 |
rookies | +0.060 |
duel | +0.059 |
Neg logit contributions | |
His | -0.096 |
HIS | -0.089 |
his | -0.088 |
He | -0.087 |
his | -0.087 |
Token | - |
Token ID | 12 |
Feature activation | +8.32e-02 |
Pos logit contributions | |
committee | +0.165 |
contestant | +0.164 |
brawl | +0.157 |
rookies | +0.149 |
duel | +0.147 |
Neg logit contributions | |
His | -0.187 |
HIS | -0.173 |
his | -0.170 |
his | -0.168 |
He | -0.167 |
Token | - |
Token ID | 12 |
Feature activation | +8.82e-02 |
Pos logit contributions | |
His | +0.296 |
HIS | +0.272 |
his | +0.267 |
his | +0.266 |
He | +0.264 |
Neg logit contributions | |
committee | -0.266 |
contestant | -0.264 |
brawl | -0.253 |
rookies | -0.240 |
duel | -0.236 |
Token | mail |
Token ID | 4529 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
committee | +0.100 |
contestant | +0.100 |
brawl | +0.097 |
rookies | +0.093 |
duel | +0.092 |
Neg logit contributions | |
His | -0.065 |
HIS | -0.059 |
his | -0.057 |
his | -0.056 |
He | -0.056 |
Token | to |
Token ID | 284 |
Feature activation | +9.07e-02 |
Pos logit contributions | |
His | +0.157 |
HIS | +0.147 |
his | +0.144 |
his | +0.143 |
him | +0.142 |
Neg logit contributions | |
committee | -0.133 |
contestant | -0.129 |
brawl | -0.123 |
rookies | -0.118 |
duel | -0.114 |
Token | � |
Token ID | 564 |
Feature activation | +3.60e-02 |
Pos logit contributions | |
committee | +0.087 |
contestant | +0.086 |
brawl | +0.083 |
rookies | +0.079 |
duel | +0.078 |
Neg logit contributions | |
His | -0.083 |
HIS | -0.077 |
his | -0.075 |
his | -0.074 |
He | -0.073 |
Token | � |
Token ID | 250 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
committee | +0.028 |
contestant | +0.027 |
brawl | +0.026 |
rookies | +0.024 |
duel | +0.024 |
Neg logit contributions | |
His | -0.040 |
HIS | -0.037 |
his | -0.037 |
his | -0.036 |
He | -0.036 |
Token | his |
Token ID | 14363 |
Feature activation | -7.10e-01 |
Pos logit contributions | |
committee | +0.162 |
contestant | +0.162 |
brawl | +0.157 |
rookies | +0.149 |
duel | +0.147 |
Neg logit contributions | |
His | -0.144 |
HIS | -0.132 |
his | -0.129 |
his | -0.127 |
He | -0.126 |
Token | fans |
Token ID | 3296 |
Feature activation | +2.53e-01 |
Pos logit contributions | |
His | +0.696 |
HIS | +0.649 |
his | +0.631 |
his | +0.630 |
He | +0.621 |
Neg logit contributions | |
committee | -0.635 |
contestant | -0.624 |
brawl | -0.603 |
rookies | -0.568 |
duel | -0.562 |
Token | � |
Token ID | 447 |
Feature activation | -9.57e-03 |
Pos logit contributions | |
committee | +0.230 |
contestant | +0.227 |
brawl | +0.218 |
rookies | +0.207 |
duel | +0.203 |
Neg logit contributions | |
His | -0.246 |
HIS | -0.228 |
his | -0.223 |
his | -0.223 |
He | -0.220 |
Token | � |
Token ID | 251 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
His | +0.009 |
HIS | +0.009 |
his | +0.009 |
his | +0.008 |
He | +0.008 |
Neg logit contributions | |
committee | -0.009 |
contestant | -0.008 |
brawl | -0.008 |
rookies | -0.008 |
duel | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | +8.77e-02 |
Pos logit contributions | |
His | +0.019 |
HIS | +0.017 |
his | +0.017 |
his | +0.017 |
He | +0.017 |
Neg logit contributions | |
committee | -0.016 |
contestant | -0.016 |
brawl | -0.015 |
rookies | -0.014 |
duel | -0.014 |
Token | let |
Token ID | 1309 |
Feature activation | +4.41e-02 |
Pos logit contributions | |
committee | +0.084 |
contestant | +0.083 |
brawl | +0.079 |
rookies | +0.076 |
duel | +0.074 |
Neg logit contributions | |
His | -0.081 |
HIS | -0.075 |
his | -0.073 |
his | -0.072 |
He | -0.071 |
Token | that |
Token ID | 326 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
committee | +0.160 |
contestant | +0.160 |
brawl | +0.154 |
rookies | +0.148 |
duel | +0.146 |
Neg logit contributions | |
His | -0.110 |
HIS | -0.100 |
his | -0.096 |
his | -0.094 |
He | -0.094 |
Token | the |
Token ID | 262 |
Feature activation | +5.34e-03 |
Pos logit contributions | |
His | +0.107 |
HIS | +0.098 |
his | +0.094 |
his | +0.094 |
He | +0.092 |
Neg logit contributions | |
committee | -0.148 |
contestant | -0.147 |
brawl | -0.142 |
rookies | -0.136 |
duel | -0.134 |
Token | tattoo |
Token ID | 18785 |
Feature activation | -4.67e-02 |
Pos logit contributions | |
committee | +0.005 |
contestant | +0.005 |
brawl | +0.004 |
rookies | +0.004 |
duel | +0.004 |
Neg logit contributions | |
His | -0.005 |
HIS | -0.005 |
his | -0.005 |
his | -0.005 |
He | -0.005 |
Token | sign |
Token ID | 1051 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
His | +0.044 |
HIS | +0.040 |
his | +0.039 |
his | +0.039 |
He | +0.039 |
Neg logit contributions | |
committee | -0.044 |
contestant | -0.044 |
brawl | -0.042 |
rookies | -0.040 |
duel | -0.039 |
Token | ified |
Token ID | 1431 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
committee | +0.026 |
contestant | +0.026 |
brawl | +0.025 |
rookies | +0.024 |
duel | +0.024 |
Neg logit contributions | |
His | -0.018 |
HIS | -0.017 |
his | -0.016 |
his | -0.016 |
He | -0.016 |
Token | his |
Token ID | 465 |
Feature activation | -7.05e-01 |
Pos logit contributions | |
His | +0.168 |
HIS | +0.154 |
his | +0.149 |
his | +0.147 |
He | +0.146 |
Neg logit contributions | |
committee | -0.218 |
contestant | -0.216 |
brawl | -0.209 |
rookies | -0.199 |
duel | -0.197 |
Token | place |
Token ID | 1295 |
Feature activation | -6.59e-02 |
Pos logit contributions | |
His | +0.666 |
HIS | +0.620 |
his | +0.601 |
his | +0.599 |
He | +0.590 |
Neg logit contributions | |
committee | -0.660 |
contestant | -0.649 |
brawl | -0.622 |
rookies | -0.593 |
duel | -0.582 |
Token | of |
Token ID | 286 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
His | +0.062 |
HIS | +0.058 |
his | +0.056 |
his | +0.056 |
He | +0.055 |
Neg logit contributions | |
committee | -0.062 |
contestant | -0.061 |
brawl | -0.059 |
rookies | -0.056 |
duel | -0.055 |
Token | birth |
Token ID | 4082 |
Feature activation | -2.73e-01 |
Pos logit contributions | |
His | +0.014 |
HIS | +0.013 |
his | +0.012 |
He | +0.012 |
his | +0.012 |
Neg logit contributions | |
contestant | -0.022 |
committee | -0.022 |
brawl | -0.021 |
rookies | -0.021 |
duel | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
His | +0.258 |
HIS | +0.240 |
his | +0.233 |
his | +0.231 |
He | +0.229 |
Neg logit contributions | |
committee | -0.254 |
contestant | -0.251 |
brawl | -0.242 |
rookies | -0.230 |
duel | -0.226 |
Token |
|
Token ID | 198 |
Feature activation | -9.74e-03 |
Pos logit contributions | |
committee | +0.016 |
contestant | +0.016 |
brawl | +0.015 |
rookies | +0.015 |
duel | +0.014 |
Neg logit contributions | |
His | -0.015 |
HIS | -0.014 |
his | -0.013 |
his | -0.013 |
He | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | -6.87e-05 |
Pos logit contributions | |
committee | +0.044 |
contestant | +0.044 |
brawl | +0.042 |
rookies | +0.040 |
duel | +0.039 |
Neg logit contributions | |
His | -0.047 |
HIS | -0.044 |
his | -0.043 |
his | -0.042 |
He | -0.042 |
Token | analyst |
Token ID | 12499 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
His | +0.000 |
HIS | +0.000 |
his | +0.000 |
his | +0.000 |
He | +0.000 |
Neg logit contributions | |
committee | -0.000 |
contestant | -0.000 |
brawl | -0.000 |
rookies | -0.000 |
duel | -0.000 |
Token | to |
Token ID | 284 |
Feature activation | +7.34e-02 |
Pos logit contributions | |
committee | +0.223 |
contestant | +0.220 |
brawl | +0.213 |
rookies | +0.201 |
duel | +0.196 |
Neg logit contributions | |
His | -0.256 |
HIS | -0.239 |
his | -0.233 |
his | -0.233 |
him | -0.229 |
Token | indicate |
Token ID | 7603 |
Feature activation | +8.10e-02 |
Pos logit contributions | |
committee | +0.072 |
contestant | +0.071 |
brawl | +0.069 |
rookies | +0.065 |
duel | +0.064 |
Neg logit contributions | |
His | -0.066 |
HIS | -0.061 |
his | -0.059 |
his | -0.058 |
He | -0.058 |
Token | to |
Token ID | 284 |
Feature activation | +7.14e-02 |
Pos logit contributions | |
committee | +0.083 |
contestant | +0.082 |
brawl | +0.079 |
rookies | +0.076 |
duel | +0.075 |
Neg logit contributions | |
His | -0.069 |
HIS | -0.064 |
his | -0.062 |
his | -0.061 |
He | -0.060 |
Token | his |
Token ID | 465 |
Feature activation | -7.05e-01 |
Pos logit contributions | |
committee | +0.081 |
contestant | +0.081 |
brawl | +0.078 |
rookies | +0.075 |
duel | +0.074 |
Neg logit contributions | |
His | -0.053 |
HIS | -0.048 |
his | -0.046 |
his | -0.046 |
He | -0.045 |
Token | patient |
Token ID | 5827 |
Feature activation | +9.98e-02 |
Pos logit contributions | |
His | +0.780 |
HIS | +0.734 |
his | +0.719 |
his | +0.715 |
him | +0.706 |
Neg logit contributions | |
committee | -0.544 |
contestant | -0.528 |
brawl | -0.511 |
rookies | -0.472 |
duel | -0.465 |
Token | that |
Token ID | 326 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
committee | +0.105 |
contestant | +0.104 |
brawl | +0.101 |
rookies | +0.096 |
duel | +0.095 |
Neg logit contributions | |
His | -0.083 |
HIS | -0.076 |
his | -0.073 |
his | -0.073 |
him | -0.072 |
Token | what |
Token ID | 644 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
His | +0.115 |
HIS | +0.106 |
his | +0.102 |
his | +0.102 |
He | +0.100 |
Neg logit contributions | |
committee | -0.141 |
contestant | -0.139 |
brawl | -0.134 |
rookies | -0.128 |
duel | -0.126 |
Token | she |
Token ID | 673 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
His | +0.190 |
HIS | +0.174 |
his | +0.168 |
his | +0.168 |
him | +0.164 |
Neg logit contributions | |
committee | -0.264 |
contestant | -0.261 |
brawl | -0.253 |
rookies | -0.241 |
duel | -0.238 |
Token | was |
Token ID | 373 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
His | +0.204 |
HIS | +0.190 |
his | +0.184 |
his | +0.184 |
him | +0.181 |
Neg logit contributions | |
committee | -0.210 |
contestant | -0.208 |
brawl | -0.200 |
rookies | -0.190 |
duel | -0.187 |
Token | put |
Token ID | 1234 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
His | +0.335 |
HIS | +0.313 |
his | +0.305 |
his | +0.303 |
him | +0.299 |
Neg logit contributions | |
committee | -0.290 |
contestant | -0.287 |
brawl | -0.274 |
rookies | -0.260 |
duel | -0.254 |
Token | what |
Token ID | 644 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
His | +0.089 |
HIS | +0.081 |
his | +0.078 |
his | +0.077 |
He | +0.076 |
Neg logit contributions | |
committee | -0.130 |
contestant | -0.129 |
brawl | -0.125 |
rookies | -0.120 |
duel | -0.119 |
Token | he |
Token ID | 339 |
Feature activation | -3.34e-01 |
Pos logit contributions | |
His | +0.179 |
HIS | +0.164 |
his | +0.159 |
his | +0.159 |
He | +0.156 |
Neg logit contributions | |
committee | -0.233 |
contestant | -0.231 |
brawl | -0.223 |
rookies | -0.213 |
duel | -0.210 |
Token | considered |
Token ID | 3177 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
His | +0.341 |
HIS | +0.319 |
his | +0.311 |
his | +0.308 |
He | +0.305 |
Neg logit contributions | |
committee | -0.287 |
contestant | -0.284 |
brawl | -0.272 |
rookies | -0.257 |
duel | -0.253 |
Token | " |
Token ID | 366 |
Feature activation | -3.41e-02 |
Pos logit contributions | |
committee | +0.126 |
contestant | +0.125 |
brawl | +0.120 |
rookies | +0.114 |
duel | +0.112 |
Neg logit contributions | |
His | -0.135 |
HIS | -0.126 |
his | -0.122 |
his | -0.121 |
He | -0.120 |
Token | his |
Token ID | 14363 |
Feature activation | -7.04e-01 |
Pos logit contributions | |
His | +0.035 |
HIS | +0.032 |
his | +0.032 |
his | +0.031 |
He | +0.031 |
Neg logit contributions | |
committee | -0.029 |
contestant | -0.029 |
brawl | -0.028 |
rookies | -0.026 |
duel | -0.026 |
Token | duty |
Token ID | 7077 |
Feature activation | -6.14e-03 |
Pos logit contributions | |
His | +0.619 |
HIS | +0.571 |
his | +0.554 |
his | +0.547 |
He | +0.543 |
Neg logit contributions | |
committee | -0.702 |
contestant | -0.697 |
brawl | -0.671 |
rookies | -0.641 |
duel | -0.631 |
Token | to |
Token ID | 284 |
Feature activation | +9.78e-02 |
Pos logit contributions | |
His | +0.006 |
HIS | +0.005 |
his | +0.005 |
his | +0.005 |
He | +0.005 |
Neg logit contributions | |
committee | -0.006 |
contestant | -0.006 |
brawl | -0.006 |
rookies | -0.005 |
duel | -0.005 |
Token | God |
Token ID | 1793 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
committee | +0.104 |
contestant | +0.104 |
brawl | +0.100 |
rookies | +0.096 |
duel | +0.095 |
Neg logit contributions | |
His | -0.079 |
HIS | -0.072 |
his | -0.070 |
He | -0.068 |
his | -0.068 |
Token | and |
Token ID | 290 |
Feature activation | +9.69e-02 |
Pos logit contributions | |
His | +0.164 |
HIS | +0.152 |
his | +0.147 |
his | +0.146 |
He | +0.144 |
Neg logit contributions | |
committee | -0.178 |
contestant | -0.176 |
brawl | -0.170 |
rookies | -0.162 |
duel | -0.160 |
Token | the |
Token ID | 262 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
committee | +0.112 |
contestant | +0.112 |
brawl | +0.108 |
rookies | +0.104 |
duel | +0.103 |
Neg logit contributions | |
His | -0.069 |
HIS | -0.062 |
his | -0.060 |
He | -0.058 |
his | -0.058 |
Token | nd |
Token ID | 358 |
Feature activation | +9.09e-02 |
Pos logit contributions | |
His | +0.024 |
HIS | +0.022 |
his | +0.022 |
his | +0.022 |
He | +0.022 |
Neg logit contributions | |
committee | -0.022 |
contestant | -0.022 |
brawl | -0.021 |
rookies | -0.020 |
duel | -0.020 |
Token | Street |
Token ID | 3530 |
Feature activation | -3.68e-03 |
Pos logit contributions | |
committee | +0.088 |
contestant | +0.087 |
brawl | +0.084 |
rookies | +0.080 |
duel | +0.079 |
Neg logit contributions | |
His | -0.083 |
HIS | -0.077 |
his | -0.074 |
his | -0.074 |
He | -0.073 |
Token | , |
Token ID | 11 |
Feature activation | -4.78e-02 |
Pos logit contributions | |
His | +0.004 |
HIS | +0.003 |
his | +0.003 |
his | +0.003 |
He | +0.003 |
Neg logit contributions | |
committee | -0.003 |
contestant | -0.003 |
brawl | -0.003 |
rookies | -0.003 |
duel | -0.003 |
Token | according |
Token ID | 1864 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
His | +0.046 |
HIS | +0.042 |
his | +0.041 |
his | +0.041 |
He | +0.040 |
Neg logit contributions | |
committee | -0.044 |
contestant | -0.044 |
brawl | -0.042 |
rookies | -0.040 |
duel | -0.039 |
Token | to |
Token ID | 284 |
Feature activation | +7.57e-02 |
Pos logit contributions | |
His | +0.119 |
HIS | +0.111 |
his | +0.108 |
his | +0.107 |
He | +0.106 |
Neg logit contributions | |
committee | -0.096 |
contestant | -0.095 |
brawl | -0.091 |
rookies | -0.086 |
duel | -0.084 |
Token | his |
Token ID | 465 |
Feature activation | -7.02e-01 |
Pos logit contributions | |
committee | +0.081 |
contestant | +0.081 |
brawl | +0.078 |
rookies | +0.075 |
duel | +0.074 |
Neg logit contributions | |
His | -0.061 |
HIS | -0.056 |
his | -0.054 |
his | -0.053 |
He | -0.052 |
Token | page |
Token ID | 2443 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
His | +0.548 |
HIS | +0.500 |
his | +0.482 |
his | +0.478 |
He | +0.472 |
Neg logit contributions | |
committee | -0.773 |
contestant | -0.764 |
brawl | -0.740 |
rookies | -0.709 |
duel | -0.700 |
Token | on |
Token ID | 319 |
Feature activation | -2.86e-02 |
Pos logit contributions | |
His | +0.035 |
HIS | +0.033 |
his | +0.032 |
his | +0.031 |
He | +0.031 |
Neg logit contributions | |
committee | -0.035 |
contestant | -0.035 |
brawl | -0.033 |
rookies | -0.032 |
duel | -0.031 |
Token | Sh |
Token ID | 911 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
His | +0.020 |
HIS | +0.018 |
his | +0.018 |
He | +0.017 |
his | +0.017 |
Neg logit contributions | |
committee | -0.033 |
contestant | -0.033 |
brawl | -0.032 |
rookies | -0.031 |
duel | -0.030 |
Token | uffle |
Token ID | 18137 |
Feature activation | +3.38e-01 |
Pos logit contributions | |
His | +0.142 |
HIS | +0.135 |
his | +0.133 |
his | +0.131 |
He | +0.131 |
Neg logit contributions | |
committee | -0.051 |
contestant | -0.050 |
brawl | -0.046 |
rookies | -0.042 |
duel | -0.041 |
Token | Along |
Token ID | 17159 |
Feature activation | +5.90e-02 |
Pos logit contributions | |
committee | +0.287 |
contestant | +0.285 |
brawl | +0.273 |
rookies | +0.258 |
duel | +0.254 |
Neg logit contributions | |
His | -0.347 |
HIS | -0.324 |
his | -0.315 |
his | -0.312 |
He | -0.310 |
Token | that |
Token ID | 326 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
committee | +0.030 |
contestant | +0.030 |
brawl | +0.028 |
rookies | +0.027 |
duel | +0.026 |
Neg logit contributions | |
His | -0.033 |
HIS | -0.031 |
his | -0.030 |
his | -0.030 |
him | -0.030 |
Token | conversation |
Token ID | 5273 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
His | +0.109 |
HIS | +0.103 |
his | +0.100 |
his | +0.100 |
him | +0.099 |
Neg logit contributions | |
committee | -0.089 |
contestant | -0.087 |
brawl | -0.082 |
rookies | -0.079 |
duel | -0.076 |
Token | . |
Token ID | 13 |
Feature activation | +4.16e-02 |
Pos logit contributions | |
committee | +0.186 |
contestant | +0.184 |
brawl | +0.176 |
rookies | +0.168 |
duel | +0.165 |
Neg logit contributions | |
His | -0.173 |
HIS | -0.161 |
his | -0.157 |
his | -0.155 |
him | -0.154 |
Token | I |
Token ID | 314 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
committee | +0.037 |
contestant | +0.037 |
brawl | +0.035 |
rookies | +0.033 |
duel | +0.033 |
Neg logit contributions | |
His | -0.042 |
HIS | -0.039 |
his | -0.038 |
his | -0.038 |
He | -0.038 |
Token | think |
Token ID | 892 |
Feature activation | -4.47e-04 |
Pos logit contributions | |
committee | +0.115 |
contestant | +0.114 |
brawl | +0.109 |
rookies | +0.104 |
duel | +0.102 |
Neg logit contributions | |
His | -0.125 |
HIS | -0.116 |
his | -0.113 |
his | -0.113 |
He | -0.111 |
Token | they |
Token ID | 484 |
Feature activation | +5.02e-01 |
Pos logit contributions | |
His | +0.000 |
HIS | +0.000 |
his | +0.000 |
his | +0.000 |
him | +0.000 |
Neg logit contributions | |
committee | -0.000 |
contestant | -0.000 |
brawl | -0.000 |
rookies | -0.000 |
duel | -0.000 |
Token | would |
Token ID | 561 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
committee | +0.365 |
contestant | +0.357 |
brawl | +0.334 |
rookies | +0.313 |
duel | +0.307 |
Neg logit contributions | |
His | -0.582 |
HIS | -0.548 |
his | -0.539 |
his | -0.535 |
him | -0.532 |
Token | have |
Token ID | 423 |
Feature activation | +7.42e-02 |
Pos logit contributions | |
committee | +0.127 |
contestant | +0.126 |
brawl | +0.121 |
rookies | +0.116 |
duel | +0.115 |
Neg logit contributions | |
His | -0.092 |
HIS | -0.084 |
his | -0.082 |
his | -0.081 |
He | -0.080 |
Token | to |
Token ID | 284 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
committee | +0.066 |
contestant | +0.065 |
brawl | +0.062 |
rookies | +0.058 |
duel | +0.057 |
Neg logit contributions | |
His | -0.074 |
HIS | -0.069 |
his | -0.068 |
him | -0.067 |
his | -0.067 |
Token | be |
Token ID | 307 |
Feature activation | -3.58e-01 |
Pos logit contributions | |
committee | +0.103 |
contestant | +0.102 |
brawl | +0.098 |
rookies | +0.094 |
duel | +0.092 |
Neg logit contributions | |
His | -0.093 |
HIS | -0.086 |
his | -0.083 |
his | -0.083 |
He | -0.082 |
Token | considering |
Token ID | 6402 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
His | +0.354 |
HIS | +0.333 |
his | +0.326 |
him | +0.321 |
his | +0.321 |
Neg logit contributions | |
committee | -0.316 |
contestant | -0.309 |
brawl | -0.297 |
rookies | -0.277 |
duel | -0.275 |
Token | , |
Token ID | 11 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
committee | +0.168 |
contestant | +0.167 |
brawl | +0.160 |
rookies | +0.153 |
duel | +0.150 |
Neg logit contributions | |
His | -0.157 |
HIS | -0.145 |
his | -0.141 |
his | -0.140 |
He | -0.138 |
Token | and |
Token ID | 290 |
Feature activation | +9.02e-02 |
Pos logit contributions | |
His | +0.021 |
HIS | +0.020 |
his | +0.020 |
his | +0.020 |
him | +0.019 |
Neg logit contributions | |
committee | -0.019 |
contestant | -0.019 |
brawl | -0.018 |
rookies | -0.017 |
duel | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | +2.77e-02 |
Pos logit contributions | |
committee | +0.078 |
contestant | +0.077 |
brawl | +0.073 |
rookies | +0.069 |
duel | +0.068 |
Neg logit contributions | |
His | -0.092 |
HIS | -0.086 |
his | -0.084 |
his | -0.084 |
him | -0.082 |
Token | only |
Token ID | 691 |
Feature activation | +6.53e-02 |
Pos logit contributions | |
committee | +0.024 |
contestant | +0.024 |
brawl | +0.023 |
rookies | +0.022 |
duel | +0.021 |
Neg logit contributions | |
His | -0.028 |
HIS | -0.026 |
his | -0.025 |
his | -0.025 |
He | -0.025 |
Token | one |
Token ID | 530 |
Feature activation | +2.59e-01 |
Pos logit contributions | |
committee | +0.061 |
contestant | +0.059 |
brawl | +0.057 |
rookies | +0.054 |
duel | +0.053 |
Neg logit contributions | |
His | -0.062 |
HIS | -0.058 |
his | -0.056 |
his | -0.056 |
him | -0.056 |
Token | they |
Token ID | 484 |
Feature activation | +4.82e-01 |
Pos logit contributions | |
committee | +0.234 |
contestant | +0.228 |
brawl | +0.220 |
rookies | +0.207 |
duel | +0.204 |
Neg logit contributions | |
His | -0.256 |
HIS | -0.239 |
his | -0.233 |
his | -0.232 |
him | -0.229 |
Token | know |
Token ID | 760 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
committee | +0.443 |
contestant | +0.437 |
brawl | +0.416 |
rookies | +0.396 |
duel | +0.387 |
Neg logit contributions | |
His | -0.467 |
HIS | -0.433 |
his | -0.423 |
his | -0.422 |
He | -0.416 |
Token | who |
Token ID | 508 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
His | +0.214 |
HIS | +0.198 |
his | +0.192 |
his | +0.190 |
He | +0.188 |
Neg logit contributions | |
committee | -0.239 |
contestant | -0.237 |
brawl | -0.228 |
rookies | -0.217 |
duel | -0.214 |
Token | understands |
Token ID | 14759 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
committee | +0.117 |
contestant | +0.116 |
brawl | +0.112 |
rookies | +0.106 |
duel | +0.105 |
Neg logit contributions | |
His | -0.112 |
HIS | -0.104 |
his | -0.101 |
his | -0.100 |
He | -0.099 |
Token | the |
Token ID | 262 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
His | +0.204 |
HIS | +0.187 |
his | +0.181 |
his | +0.178 |
He | +0.177 |
Neg logit contributions | |
committee | -0.268 |
contestant | -0.266 |
brawl | -0.257 |
rookies | -0.246 |
duel | -0.243 |
Token | heart |
Token ID | 2612 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
committee | +0.020 |
contestant | +0.020 |
brawl | +0.019 |
rookies | +0.018 |
duel | +0.018 |
Neg logit contributions | |
His | -0.022 |
HIS | -0.021 |
his | -0.020 |
his | -0.020 |
He | -0.020 |
Token | many |
Token ID | 867 |
Feature activation | +7.33e-02 |
Pos logit contributions | |
committee | +0.035 |
contestant | +0.034 |
brawl | +0.033 |
rookies | +0.031 |
duel | +0.031 |
Neg logit contributions | |
His | -0.036 |
HIS | -0.034 |
his | -0.033 |
his | -0.033 |
him | -0.032 |
Token | ways |
Token ID | 2842 |
Feature activation | -1.91e-01 |
Pos logit contributions | |
committee | +0.076 |
contestant | +0.075 |
brawl | +0.072 |
rookies | +0.069 |
duel | +0.068 |
Neg logit contributions | |
His | -0.062 |
HIS | -0.057 |
his | -0.055 |
his | -0.055 |
He | -0.054 |
Token | by |
Token ID | 416 |
Feature activation | -5.75e-02 |
Pos logit contributions | |
His | +0.173 |
HIS | +0.160 |
his | +0.156 |
his | +0.156 |
He | +0.153 |
Neg logit contributions | |
committee | -0.186 |
contestant | -0.184 |
brawl | -0.177 |
rookies | -0.168 |
duel | -0.166 |
Token | pop |
Token ID | 1461 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
His | +0.057 |
HIS | +0.054 |
his | +0.053 |
his | +0.052 |
him | +0.052 |
Neg logit contributions | |
committee | -0.050 |
contestant | -0.049 |
brawl | -0.048 |
rookies | -0.045 |
duel | -0.044 |
Token | culture |
Token ID | 3968 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
committee | +0.027 |
contestant | +0.027 |
brawl | +0.026 |
rookies | +0.025 |
duel | +0.025 |
Neg logit contributions | |
His | -0.022 |
HIS | -0.020 |
his | -0.019 |
his | -0.019 |
He | -0.019 |
Token | pundits |
Token ID | 30855 |
Feature activation | +4.26e-01 |
Pos logit contributions | |
His | +0.159 |
HIS | +0.148 |
his | +0.145 |
his | +0.144 |
him | +0.142 |
Neg logit contributions | |
committee | -0.140 |
contestant | -0.137 |
brawl | -0.131 |
rookies | -0.123 |
duel | -0.122 |
Token | . |
Token ID | 13 |
Feature activation | +6.96e-02 |
Pos logit contributions | |
committee | +0.403 |
contestant | +0.397 |
brawl | +0.383 |
rookies | +0.363 |
duel | +0.356 |
Neg logit contributions | |
His | -0.398 |
HIS | -0.369 |
his | -0.362 |
his | -0.359 |
him | -0.356 |
Token | There |
Token ID | 1318 |
Feature activation | +2.71e-01 |
Pos logit contributions | |
committee | +0.065 |
contestant | +0.063 |
brawl | +0.061 |
rookies | +0.058 |
duel | +0.057 |
Neg logit contributions | |
His | -0.068 |
HIS | -0.063 |
his | -0.061 |
his | -0.061 |
He | -0.060 |
Token | are |
Token ID | 389 |
Feature activation | +4.76e-01 |
Pos logit contributions | |
committee | +0.201 |
contestant | +0.199 |
brawl | +0.189 |
rookies | +0.176 |
duel | +0.173 |
Neg logit contributions | |
His | -0.309 |
HIS | -0.289 |
his | -0.285 |
his | -0.283 |
He | -0.280 |
Token | albums |
Token ID | 16788 |
Feature activation | -5.15e-02 |
Pos logit contributions | |
committee | +0.437 |
contestant | +0.428 |
brawl | +0.412 |
rookies | +0.389 |
duel | +0.383 |
Neg logit contributions | |
His | -0.459 |
HIS | -0.428 |
his | -0.416 |
his | -0.415 |
him | -0.409 |
Token | you |
Token ID | 345 |
Feature activation | +1.97e-01 |
Pos logit contributions | |
His | +0.045 |
HIS | +0.042 |
his | +0.040 |
his | +0.040 |
him | +0.040 |
Neg logit contributions | |
committee | -0.052 |
contestant | -0.051 |
brawl | -0.049 |
rookies | -0.047 |
duel | -0.046 |
Token | least |
Token ID | 1551 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
His | +0.006 |
HIS | +0.005 |
his | +0.005 |
his | +0.005 |
him | +0.005 |
Neg logit contributions | |
committee | -0.006 |
contestant | -0.006 |
brawl | -0.005 |
rookies | -0.005 |
duel | -0.005 |
Token | one |
Token ID | 530 |
Feature activation | +2.37e-01 |
Pos logit contributions | |
His | +0.014 |
HIS | +0.013 |
his | +0.013 |
his | +0.013 |
him | +0.013 |
Neg logit contributions | |
committee | -0.009 |
contestant | -0.009 |
brawl | -0.008 |
rookies | -0.007 |
duel | -0.007 |
Token | very |
Token ID | 845 |
Feature activation | +2.11e-04 |
Pos logit contributions | |
committee | +0.218 |
contestant | +0.212 |
brawl | +0.204 |
rookies | +0.192 |
duel | +0.190 |
Neg logit contributions | |
His | -0.229 |
HIS | -0.213 |
his | -0.208 |
his | -0.206 |
him | -0.205 |
Token | good |
Token ID | 922 |
Feature activation | -9.37e-02 |
Pos logit contributions | |
committee | +0.000 |
contestant | +0.000 |
brawl | +0.000 |
rookies | +0.000 |
duel | +0.000 |
Neg logit contributions | |
His | -0.000 |
HIS | -0.000 |
his | -0.000 |
his | -0.000 |
him | -0.000 |
Token | role |
Token ID | 2597 |
Feature activation | +2.50e-01 |
Pos logit contributions | |
His | +0.109 |
HIS | +0.103 |
his | +0.102 |
his | +0.101 |
him | +0.101 |
Neg logit contributions | |
committee | -0.067 |
contestant | -0.064 |
brawl | -0.061 |
rookies | -0.056 |
duel | -0.055 |
Token | player |
Token ID | 2137 |
Feature activation | +4.31e-01 |
Pos logit contributions | |
committee | +0.108 |
contestant | +0.102 |
brawl | +0.096 |
rookies | +0.082 |
duel | +0.080 |
Neg logit contributions | |
His | -0.360 |
HIS | -0.342 |
his | -0.339 |
his | -0.336 |
him | -0.336 |
Token | . |
Token ID | 13 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
committee | +0.417 |
contestant | +0.411 |
brawl | +0.397 |
rookies | +0.374 |
duel | +0.370 |
Neg logit contributions | |
His | -0.391 |
HIS | -0.363 |
his | -0.355 |
his | -0.352 |
him | -0.349 |
Token | Getting |
Token ID | 18067 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
committee | +0.019 |
contestant | +0.019 |
brawl | +0.018 |
rookies | +0.017 |
duel | +0.017 |
Neg logit contributions | |
His | -0.019 |
HIS | -0.018 |
his | -0.017 |
his | -0.017 |
He | -0.017 |
Token | rid |
Token ID | 5755 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
committee | +0.147 |
contestant | +0.146 |
brawl | +0.141 |
rookies | +0.134 |
duel | +0.133 |
Neg logit contributions | |
His | -0.131 |
HIS | -0.121 |
his | -0.118 |
his | -0.116 |
He | -0.115 |
Token | of |
Token ID | 286 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
His | +0.211 |
HIS | +0.196 |
his | +0.191 |
his | +0.189 |
He | +0.187 |
Neg logit contributions | |
committee | -0.206 |
contestant | -0.204 |
brawl | -0.196 |
rookies | -0.186 |
duel | -0.184 |
Token | the |
Token ID | 262 |
Feature activation | +2.97e-03 |
Pos logit contributions | |
His | +0.015 |
HIS | +0.014 |
his | +0.013 |
his | +0.013 |
He | +0.013 |
Neg logit contributions | |
committee | -0.016 |
contestant | -0.015 |
brawl | -0.015 |
rookies | -0.014 |
duel | -0.014 |
Token | and |
Token ID | 290 |
Feature activation | +8.68e-02 |
Pos logit contributions | |
committee | +0.168 |
contestant | +0.166 |
brawl | +0.159 |
rookies | +0.152 |
duel | +0.149 |
Neg logit contributions | |
His | -0.162 |
HIS | -0.150 |
his | -0.145 |
his | -0.145 |
He | -0.143 |
Token | wholly |
Token ID | 18174 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
committee | +0.072 |
contestant | +0.071 |
brawl | +0.067 |
rookies | +0.064 |
duel | +0.062 |
Neg logit contributions | |
His | -0.091 |
HIS | -0.086 |
his | -0.083 |
his | -0.083 |
him | -0.082 |
Token | believable |
Token ID | 41632 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
His | +0.130 |
HIS | +0.122 |
his | +0.118 |
his | +0.118 |
him | +0.117 |
Neg logit contributions | |
committee | -0.097 |
contestant | -0.095 |
brawl | -0.091 |
rookies | -0.085 |
duel | -0.083 |
Token | friendship |
Token ID | 14738 |
Feature activation | +9.03e-02 |
Pos logit contributions | |
committee | +0.128 |
contestant | +0.126 |
brawl | +0.121 |
rookies | +0.115 |
duel | +0.113 |
Neg logit contributions | |
His | -0.132 |
HIS | -0.123 |
his | -0.119 |
his | -0.118 |
He | -0.117 |
Token | -- |
Token ID | 1377 |
Feature activation | -3.56e-02 |
Pos logit contributions | |
committee | +0.084 |
contestant | +0.082 |
brawl | +0.079 |
rookies | +0.074 |
duel | +0.073 |
Neg logit contributions | |
His | -0.086 |
HIS | -0.080 |
his | -0.079 |
his | -0.078 |
him | -0.077 |
Token | they |
Token ID | 484 |
Feature activation | +4.84e-01 |
Pos logit contributions | |
His | +0.036 |
HIS | +0.033 |
his | +0.032 |
his | +0.032 |
him | +0.032 |
Neg logit contributions | |
committee | -0.032 |
contestant | -0.031 |
brawl | -0.030 |
rookies | -0.028 |
duel | -0.027 |
Token | have |
Token ID | 423 |
Feature activation | +5.91e-02 |
Pos logit contributions | |
committee | +0.412 |
contestant | +0.404 |
brawl | +0.381 |
rookies | +0.363 |
duel | +0.352 |
Neg logit contributions | |
His | -0.501 |
HIS | -0.468 |
his | -0.458 |
his | -0.457 |
him | -0.451 |
Token | little |
Token ID | 1310 |
Feature activation | +7.50e-02 |
Pos logit contributions | |
committee | +0.054 |
contestant | +0.053 |
brawl | +0.051 |
rookies | +0.049 |
duel | +0.048 |
Neg logit contributions | |
His | -0.057 |
HIS | -0.053 |
his | -0.052 |
his | -0.051 |
him | -0.051 |
Token | in |
Token ID | 287 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
committee | +0.070 |
contestant | +0.068 |
brawl | +0.065 |
rookies | +0.062 |
duel | +0.061 |
Neg logit contributions | |
His | -0.071 |
HIS | -0.067 |
his | -0.064 |
his | -0.064 |
him | -0.064 |
Token | common |
Token ID | 2219 |
Feature activation | -4.05e-02 |
Pos logit contributions | |
committee | +0.023 |
contestant | +0.023 |
brawl | +0.022 |
rookies | +0.021 |
duel | +0.020 |
Neg logit contributions | |
His | -0.022 |
HIS | -0.020 |
his | -0.019 |
his | -0.019 |
He | -0.019 |
Token | at |
Token ID | 379 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
His | +0.038 |
HIS | +0.036 |
his | +0.035 |
his | +0.034 |
He | +0.034 |
Neg logit contributions | |
committee | -0.038 |
contestant | -0.037 |
brawl | -0.036 |
rookies | -0.034 |
duel | -0.034 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +5.23e-02 |
Pos logit contributions | |
committee | +0.035 |
contestant | +0.035 |
brawl | +0.033 |
rookies | +0.032 |
duel | +0.031 |
Neg logit contributions | |
His | -0.039 |
HIS | -0.037 |
his | -0.036 |
his | -0.035 |
He | -0.035 |
Token | Their |
Token ID | 14574 |
Feature activation | +4.81e-02 |
Pos logit contributions | |
committee | +0.054 |
contestant | +0.054 |
brawl | +0.052 |
rookies | +0.050 |
duel | +0.049 |
Neg logit contributions | |
His | -0.044 |
HIS | -0.040 |
his | -0.039 |
He | -0.038 |
his | -0.038 |
Token | solution |
Token ID | 4610 |
Feature activation | +2.70e-02 |
Pos logit contributions | |
committee | +0.046 |
contestant | +0.045 |
brawl | +0.043 |
rookies | +0.041 |
duel | +0.041 |
Neg logit contributions | |
His | -0.045 |
HIS | -0.041 |
his | -0.040 |
his | -0.040 |
He | -0.039 |
Token | involves |
Token ID | 9018 |
Feature activation | +2.68e-02 |
Pos logit contributions | |
committee | +0.025 |
contestant | +0.025 |
brawl | +0.024 |
rookies | +0.023 |
duel | +0.023 |
Neg logit contributions | |
His | -0.025 |
HIS | -0.023 |
his | -0.023 |
his | -0.023 |
He | -0.022 |
Token | encaps |
Token ID | 32652 |
Feature activation | +4.41e-02 |
Pos logit contributions | |
committee | +0.022 |
contestant | +0.021 |
brawl | +0.021 |
rookies | +0.019 |
duel | +0.019 |
Neg logit contributions | |
His | -0.028 |
HIS | -0.027 |
his | -0.026 |
his | -0.026 |
him | -0.026 |
Token | ulating |
Token ID | 8306 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
committee | +0.040 |
contestant | +0.040 |
brawl | +0.038 |
rookies | +0.037 |
duel | +0.036 |
Neg logit contributions | |
His | -0.042 |
HIS | -0.039 |
his | -0.038 |
his | -0.038 |
He | -0.038 |
Token | per |
Token ID | 583 |
Feature activation | -6.50e-02 |
Pos logit contributions | |
committee | +0.022 |
contestant | +0.022 |
brawl | +0.021 |
rookies | +0.020 |
duel | +0.020 |
Neg logit contributions | |
His | -0.022 |
HIS | -0.021 |
his | -0.020 |
his | -0.020 |
He | -0.020 |
Token | ov |
Token ID | 709 |
Feature activation | +3.04e-02 |
Pos logit contributions | |
His | +0.064 |
HIS | +0.059 |
his | +0.058 |
his | +0.057 |
He | +0.057 |
Neg logit contributions | |
committee | -0.058 |
contestant | -0.058 |
brawl | -0.056 |
rookies | -0.053 |
duel | -0.052 |
Token | sk |
Token ID | 8135 |
Feature activation | +9.50e-03 |
Pos logit contributions | |
committee | +0.029 |
contestant | +0.029 |
brawl | +0.028 |
rookies | +0.027 |
duel | +0.026 |
Neg logit contributions | |
His | -0.028 |
HIS | -0.026 |
his | -0.025 |
his | -0.025 |
He | -0.025 |
Token | ite |
Token ID | 578 |
Feature activation | -5.76e-03 |
Pos logit contributions | |
committee | +0.009 |
contestant | +0.009 |
brawl | +0.009 |
rookies | +0.008 |
duel | +0.008 |
Neg logit contributions | |
His | -0.009 |
HIS | -0.008 |
his | -0.008 |
his | -0.008 |
He | -0.008 |
Token | micro |
Token ID | 4580 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
His | +0.005 |
HIS | +0.005 |
his | +0.004 |
his | +0.004 |
He | +0.004 |
Neg logit contributions | |
committee | -0.006 |
contestant | -0.006 |
brawl | -0.006 |
rookies | -0.005 |
duel | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
committee | +0.067 |
contestant | +0.066 |
brawl | +0.063 |
rookies | +0.060 |
duel | +0.058 |
Neg logit contributions | |
His | -0.067 |
HIS | -0.062 |
his | -0.061 |
his | -0.060 |
him | -0.060 |
Token |
|
Token ID | 198 |
Feature activation | -2.57e-03 |
Pos logit contributions | |
committee | +0.022 |
contestant | +0.022 |
brawl | +0.021 |
rookies | +0.020 |
duel | +0.020 |
Neg logit contributions | |
His | -0.022 |
HIS | -0.021 |
his | -0.020 |
his | -0.020 |
He | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -3.82e-03 |
Pos logit contributions | |
His | +0.002 |
HIS | +0.002 |
his | +0.002 |
his | +0.002 |
He | +0.002 |
Neg logit contributions | |
committee | -0.002 |
contestant | -0.002 |
brawl | -0.002 |
rookies | -0.002 |
duel | -0.002 |
Token | F |
Token ID | 37 |
Feature activation | -9.46e-02 |
Pos logit contributions | |
His | +0.003 |
HIS | +0.003 |
his | +0.003 |
his | +0.003 |
He | +0.003 |
Neg logit contributions | |
committee | -0.004 |
contestant | -0.004 |
brawl | -0.004 |
rookies | -0.004 |
duel | -0.004 |
Token | oster |
Token ID | 6197 |
Feature activation | +9.32e-02 |
Pos logit contributions | |
His | +0.100 |
HIS | +0.093 |
his | +0.091 |
his | +0.090 |
He | +0.090 |
Neg logit contributions | |
committee | -0.078 |
contestant | -0.077 |
brawl | -0.074 |
rookies | -0.070 |
duel | -0.069 |
Token | 's |
Token ID | 338 |
Feature activation | +7.48e-02 |
Pos logit contributions | |
committee | +0.093 |
contestant | +0.092 |
brawl | +0.089 |
rookies | +0.085 |
duel | +0.083 |
Neg logit contributions | |
His | -0.082 |
HIS | -0.076 |
his | -0.073 |
his | -0.073 |
He | -0.072 |
Token | condition |
Token ID | 4006 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
committee | +0.074 |
contestant | +0.073 |
brawl | +0.070 |
rookies | +0.067 |
duel | +0.066 |
Neg logit contributions | |
His | -0.067 |
HIS | -0.062 |
his | -0.060 |
his | -0.059 |
He | -0.059 |
Token | wasn |
Token ID | 2492 |
Feature activation | -4.45e-02 |
Pos logit contributions | |
His | +0.012 |
HIS | +0.011 |
his | +0.010 |
his | +0.010 |
He | +0.010 |
Neg logit contributions | |
committee | -0.014 |
contestant | -0.014 |
brawl | -0.013 |
rookies | -0.012 |
duel | -0.012 |
Token | 't |
Token ID | 470 |
Feature activation | -3.17e-02 |
Pos logit contributions | |
His | +0.038 |
HIS | +0.035 |
his | +0.034 |
his | +0.033 |
He | +0.033 |
Neg logit contributions | |
committee | -0.046 |
contestant | -0.045 |
brawl | -0.044 |
rookies | -0.042 |
duel | -0.041 |
Token | immediately |
Token ID | 3393 |
Feature activation | -2.45e-02 |
Pos logit contributions | |
His | +0.029 |
HIS | +0.027 |
his | +0.027 |
He | +0.026 |
his | +0.026 |
Neg logit contributions | |
committee | -0.030 |
contestant | -0.030 |
brawl | -0.029 |
rookies | -0.027 |
duel | -0.027 |
Token | available |
Token ID | 1695 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
His | +0.023 |
HIS | +0.021 |
his | +0.020 |
He | +0.020 |
his | +0.020 |
Neg logit contributions | |
committee | -0.023 |
contestant | -0.023 |
brawl | -0.022 |
rookies | -0.021 |
duel | -0.021 |
Token | later |
Token ID | 1568 |
Feature activation | +5.04e-02 |
Pos logit contributions | |
His | +0.283 |
HIS | +0.263 |
his | +0.255 |
his | +0.254 |
He | +0.251 |
Neg logit contributions | |
committee | -0.275 |
contestant | -0.272 |
brawl | -0.261 |
rookies | -0.247 |
duel | -0.244 |
Token | , |
Token ID | 11 |
Feature activation | -2.11e-02 |
Pos logit contributions | |
committee | +0.049 |
contestant | +0.048 |
brawl | +0.047 |
rookies | +0.044 |
duel | +0.044 |
Neg logit contributions | |
His | -0.046 |
HIS | -0.042 |
his | -0.041 |
his | -0.041 |
He | -0.040 |
Token | we |
Token ID | 356 |
Feature activation | +3.79e-01 |
Pos logit contributions | |
His | +0.026 |
HIS | +0.025 |
his | +0.024 |
his | +0.024 |
He | +0.024 |
Neg logit contributions | |
committee | -0.013 |
contestant | -0.013 |
brawl | -0.012 |
rookies | -0.011 |
duel | -0.011 |
Token | � |
Token ID | 447 |
Feature activation | -6.79e-03 |
Pos logit contributions | |
committee | +0.296 |
contestant | +0.291 |
brawl | +0.275 |
rookies | +0.258 |
duel | +0.253 |
Neg logit contributions | |
His | -0.419 |
HIS | -0.393 |
his | -0.384 |
his | -0.383 |
He | -0.379 |
Token | � |
Token ID | 247 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
His | +0.006 |
HIS | +0.006 |
his | +0.006 |
his | +0.005 |
He | +0.005 |
Neg logit contributions | |
contestant | -0.007 |
committee | -0.007 |
brawl | -0.006 |
rookies | -0.006 |
duel | -0.006 |
Token | re |
Token ID | 260 |
Feature activation | +2.39e-01 |
Pos logit contributions | |
His | +0.017 |
HIS | +0.016 |
his | +0.016 |
his | +0.016 |
He | +0.015 |
Neg logit contributions | |
contestant | -0.020 |
committee | -0.020 |
brawl | -0.019 |
rookies | -0.018 |
duel | -0.018 |
Token | excited |
Token ID | 6568 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
committee | +0.200 |
contestant | +0.198 |
brawl | +0.190 |
rookies | +0.179 |
duel | +0.176 |
Neg logit contributions | |
His | -0.248 |
HIS | -0.232 |
his | -0.226 |
his | -0.225 |
He | -0.222 |
Token | to |
Token ID | 284 |
Feature activation | +9.49e-02 |
Pos logit contributions | |
committee | +0.146 |
contestant | +0.144 |
brawl | +0.138 |
rookies | +0.131 |
duel | +0.129 |
Neg logit contributions | |
His | -0.166 |
HIS | -0.155 |
his | -0.151 |
his | -0.150 |
He | -0.149 |
Token | announce |
Token ID | 5453 |
Feature activation | +9.93e-02 |
Pos logit contributions | |
committee | +0.086 |
contestant | +0.085 |
brawl | +0.082 |
rookies | +0.078 |
duel | +0.077 |
Neg logit contributions | |
His | -0.092 |
HIS | -0.086 |
his | -0.083 |
his | -0.082 |
He | -0.082 |
Token | that |
Token ID | 326 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
committee | +0.097 |
contestant | +0.096 |
brawl | +0.092 |
rookies | +0.087 |
duel | +0.086 |
Neg logit contributions | |
His | -0.090 |
HIS | -0.083 |
his | -0.081 |
his | -0.081 |
He | -0.079 |
Token | we |
Token ID | 356 |
Feature activation | +3.76e-01 |
Pos logit contributions | |
His | +0.135 |
HIS | +0.127 |
his | +0.124 |
his | +0.124 |
He | +0.122 |
Neg logit contributions | |
committee | -0.088 |
contestant | -0.086 |
brawl | -0.082 |
rookies | -0.076 |
duel | -0.075 |
Token | able |
Token ID | 1498 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
His | +0.117 |
HIS | +0.110 |
his | +0.107 |
his | +0.107 |
him | +0.105 |
Neg logit contributions | |
committee | -0.100 |
contestant | -0.097 |
brawl | -0.094 |
rookies | -0.088 |
duel | -0.087 |
Token | to |
Token ID | 284 |
Feature activation | +8.81e-02 |
Pos logit contributions | |
committee | +0.092 |
contestant | +0.090 |
brawl | +0.087 |
rookies | +0.082 |
duel | +0.080 |
Neg logit contributions | |
His | -0.108 |
HIS | -0.101 |
his | -0.098 |
his | -0.098 |
him | -0.097 |
Token | capitalize |
Token ID | 35160 |
Feature activation | -8.65e-03 |
Pos logit contributions | |
committee | +0.083 |
contestant | +0.082 |
brawl | +0.078 |
rookies | +0.074 |
duel | +0.073 |
Neg logit contributions | |
His | -0.083 |
HIS | -0.077 |
his | -0.075 |
his | -0.074 |
He | -0.073 |
Token | on |
Token ID | 319 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
His | +0.008 |
HIS | +0.007 |
his | +0.007 |
his | +0.007 |
He | +0.007 |
Neg logit contributions | |
committee | -0.008 |
contestant | -0.008 |
brawl | -0.008 |
rookies | -0.007 |
duel | -0.007 |
Token | its |
Token ID | 663 |
Feature activation | -1.88e-01 |
Pos logit contributions | |
His | +0.016 |
HIS | +0.015 |
his | +0.015 |
his | +0.014 |
him | +0.014 |
Neg logit contributions | |
committee | -0.017 |
contestant | -0.016 |
brawl | -0.016 |
rookies | -0.015 |
duel | -0.015 |
Token | extra |
Token ID | 3131 |
Feature activation | +3.06e-02 |
Pos logit contributions | |
His | +0.184 |
HIS | +0.172 |
his | +0.167 |
his | +0.167 |
him | +0.164 |
Neg logit contributions | |
committee | -0.169 |
contestant | -0.165 |
brawl | -0.159 |
rookies | -0.150 |
duel | -0.147 |
Token | placement |
Token ID | 13127 |
Feature activation | +7.17e-02 |
Pos logit contributions | |
committee | +0.027 |
contestant | +0.027 |
brawl | +0.026 |
rookies | +0.025 |
duel | +0.024 |
Neg logit contributions | |
His | -0.030 |
HIS | -0.028 |
his | -0.027 |
his | -0.027 |
He | -0.027 |
Token | with |
Token ID | 351 |
Feature activation | +4.58e-02 |
Pos logit contributions | |
committee | +0.069 |
contestant | +0.068 |
brawl | +0.065 |
rookies | +0.062 |
duel | +0.061 |
Neg logit contributions | |
His | -0.066 |
HIS | -0.061 |
his | -0.059 |
his | -0.059 |
him | -0.059 |
Token | a |
Token ID | 257 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
committee | +0.041 |
contestant | +0.040 |
brawl | +0.038 |
rookies | +0.036 |
duel | +0.036 |
Neg logit contributions | |
His | -0.045 |
HIS | -0.042 |
his | -0.041 |
his | -0.041 |
him | -0.041 |
Token | win |
Token ID | 1592 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
committee | +0.149 |
contestant | +0.145 |
brawl | +0.139 |
rookies | +0.132 |
duel | +0.128 |
Neg logit contributions | |
His | -0.187 |
HIS | -0.176 |
his | -0.171 |
his | -0.171 |
him | -0.169 |
Token | for |
Token ID | 329 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
His | +0.102 |
HIS | +0.093 |
his | +0.091 |
his | +0.091 |
him | +0.089 |
Neg logit contributions | |
committee | -0.133 |
contestant | -0.131 |
brawl | -0.127 |
rookies | -0.121 |
duel | -0.119 |
Token | . |
Token ID | 13 |
Feature activation | +3.20e-02 |
Pos logit contributions | |
His | +0.100 |
HIS | +0.093 |
his | +0.090 |
his | +0.089 |
He | +0.088 |
Neg logit contributions | |
committee | -0.107 |
contestant | -0.107 |
brawl | -0.102 |
rookies | -0.098 |
duel | -0.096 |
Token |
|
Token ID | 198 |
Feature activation | +2.85e-03 |
Pos logit contributions | |
committee | +0.029 |
contestant | +0.029 |
brawl | +0.028 |
rookies | +0.026 |
duel | +0.026 |
Neg logit contributions | |
His | -0.031 |
HIS | -0.029 |
his | -0.028 |
his | -0.028 |
He | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | +4.23e-03 |
Pos logit contributions | |
committee | +0.003 |
contestant | +0.003 |
brawl | +0.003 |
rookies | +0.002 |
duel | +0.002 |
Neg logit contributions | |
His | -0.003 |
HIS | -0.002 |
his | -0.002 |
his | -0.002 |
He | -0.002 |
Token | More |
Token ID | 5167 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
committee | +0.004 |
contestant | +0.004 |
brawl | +0.004 |
rookies | +0.004 |
duel | +0.004 |
Neg logit contributions | |
His | -0.004 |
HIS | -0.003 |
his | -0.003 |
He | -0.003 |
his | -0.003 |
Token | subtle |
Token ID | 11800 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
His | +0.157 |
HIS | +0.145 |
his | +0.141 |
his | +0.141 |
He | +0.139 |
Neg logit contributions | |
committee | -0.148 |
contestant | -0.146 |
brawl | -0.140 |
rookies | -0.132 |
duel | -0.130 |
Token | and |
Token ID | 290 |
Feature activation | +8.31e-02 |
Pos logit contributions | |
committee | +0.162 |
contestant | +0.161 |
brawl | +0.155 |
rookies | +0.147 |
duel | +0.144 |
Neg logit contributions | |
His | -0.173 |
HIS | -0.161 |
his | -0.157 |
his | -0.155 |
He | -0.154 |
Token | more |
Token ID | 517 |
Feature activation | -4.31e-02 |
Pos logit contributions | |
committee | +0.069 |
contestant | +0.068 |
brawl | +0.065 |
rookies | +0.062 |
duel | +0.060 |
Neg logit contributions | |
His | -0.087 |
HIS | -0.081 |
his | -0.079 |
his | -0.079 |
He | -0.078 |
Token | insidious |
Token ID | 43732 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
His | +0.047 |
HIS | +0.044 |
his | +0.043 |
his | +0.043 |
him | +0.043 |
Neg logit contributions | |
committee | -0.034 |
contestant | -0.033 |
brawl | -0.032 |
rookies | -0.030 |
duel | -0.029 |
Token | was |
Token ID | 373 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
committee | +0.194 |
contestant | +0.192 |
brawl | +0.183 |
rookies | +0.172 |
duel | +0.169 |
Neg logit contributions | |
His | -0.267 |
HIS | -0.251 |
his | -0.245 |
his | -0.242 |
He | -0.241 |
Token | the |
Token ID | 262 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
His | +0.101 |
HIS | +0.093 |
his | +0.091 |
his | +0.091 |
him | +0.089 |
Neg logit contributions | |
committee | -0.103 |
contestant | -0.102 |
brawl | -0.098 |
rookies | -0.093 |
duel | -0.091 |
Token | fear |
Token ID | 3252 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
committee | +0.017 |
contestant | +0.017 |
brawl | +0.016 |
rookies | +0.015 |
duel | +0.015 |
Neg logit contributions | |
His | -0.020 |
HIS | -0.019 |
his | -0.018 |
his | -0.018 |
He | -0.018 |
Token | claim |
Token ID | 1624 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
His | +0.727 |
HIS | +0.679 |
his | +0.663 |
his | +0.662 |
He | +0.654 |
Neg logit contributions | |
committee | -0.578 |
contestant | -0.570 |
brawl | -0.546 |
rookies | -0.515 |
duel | -0.502 |
Token | is |
Token ID | 318 |
Feature activation | -8.37e-02 |
Pos logit contributions | |
His | +0.022 |
HIS | +0.021 |
his | +0.020 |
his | +0.020 |
He | +0.019 |
Neg logit contributions | |
committee | -0.026 |
contestant | -0.026 |
brawl | -0.025 |
rookies | -0.024 |
duel | -0.023 |
Token | this |
Token ID | 428 |
Feature activation | -7.09e-02 |
Pos logit contributions | |
His | +0.070 |
HIS | +0.065 |
his | +0.063 |
his | +0.062 |
He | +0.061 |
Neg logit contributions | |
committee | -0.087 |
contestant | -0.086 |
brawl | -0.083 |
rookies | -0.080 |
duel | -0.078 |
Token | : |
Token ID | 25 |
Feature activation | +7.94e-02 |
Pos logit contributions | |
His | +0.070 |
HIS | +0.066 |
his | +0.064 |
his | +0.063 |
He | +0.063 |
Neg logit contributions | |
committee | -0.063 |
contestant | -0.062 |
brawl | -0.060 |
rookies | -0.057 |
duel | -0.056 |
Token | modern |
Token ID | 3660 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
committee | +0.075 |
contestant | +0.074 |
brawl | +0.071 |
rookies | +0.068 |
duel | +0.067 |
Neg logit contributions | |
His | -0.075 |
HIS | -0.069 |
his | -0.067 |
his | -0.067 |
He | -0.066 |
Token | notions |
Token ID | 26226 |
Feature activation | -4.77e-02 |
Pos logit contributions | |
His | +0.095 |
HIS | +0.087 |
his | +0.085 |
his | +0.084 |
He | +0.083 |
Neg logit contributions | |
committee | -0.114 |
contestant | -0.114 |
brawl | -0.110 |
rookies | -0.105 |
duel | -0.103 |
Token | of |
Token ID | 286 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
His | +0.045 |
HIS | +0.042 |
his | +0.041 |
his | +0.040 |
He | +0.040 |
Neg logit contributions | |
committee | -0.044 |
contestant | -0.044 |
brawl | -0.042 |
rookies | -0.040 |
duel | -0.040 |
Token | goodness |
Token ID | 20437 |
Feature activation | -4.72e-01 |
Pos logit contributions | |
His | +0.016 |
HIS | +0.015 |
his | +0.015 |
his | +0.014 |
He | +0.014 |
Neg logit contributions | |
committee | -0.019 |
contestant | -0.019 |
brawl | -0.018 |
rookies | -0.017 |
duel | -0.017 |
Token | as |
Token ID | 355 |
Feature activation | -7.00e-02 |
Pos logit contributions | |
His | +0.481 |
HIS | +0.448 |
his | +0.438 |
his | +0.438 |
He | +0.432 |
Neg logit contributions | |
committee | -0.407 |
contestant | -0.399 |
brawl | -0.383 |
rookies | -0.363 |
duel | -0.354 |
Token | altru |
Token ID | 37677 |
Feature activation | -9.31e-02 |
Pos logit contributions | |
His | +0.069 |
HIS | +0.064 |
his | +0.063 |
his | +0.062 |
him | +0.062 |
Neg logit contributions | |
committee | -0.063 |
contestant | -0.062 |
brawl | -0.060 |
rookies | -0.056 |
duel | -0.055 |
Token | istic |
Token ID | 2569 |
Feature activation | +6.27e-02 |
Pos logit contributions | |
His | +0.086 |
HIS | +0.080 |
his | +0.078 |
his | +0.077 |
He | +0.076 |
Neg logit contributions | |
committee | -0.089 |
contestant | -0.088 |
brawl | -0.085 |
rookies | -0.080 |
duel | -0.079 |
Token | pensions |
Token ID | 27581 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
committee | +0.064 |
contestant | +0.063 |
brawl | +0.061 |
rookies | +0.057 |
duel | +0.056 |
Neg logit contributions | |
His | -0.068 |
HIS | -0.064 |
his | -0.062 |
his | -0.062 |
him | -0.061 |
Token | — |
Token ID | 851 |
Feature activation | -3.04e-02 |
Pos logit contributions | |
His | +0.308 |
HIS | +0.288 |
his | +0.281 |
his | +0.281 |
He | +0.277 |
Neg logit contributions | |
committee | -0.228 |
contestant | -0.226 |
brawl | -0.216 |
rookies | -0.203 |
duel | -0.199 |
Token | previously |
Token ID | 4271 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
His | +0.029 |
HIS | +0.027 |
his | +0.026 |
his | +0.026 |
He | +0.025 |
Neg logit contributions | |
committee | -0.028 |
contestant | -0.028 |
brawl | -0.027 |
rookies | -0.026 |
duel | -0.025 |
Token | the |
Token ID | 262 |
Feature activation | +2.34e-03 |
Pos logit contributions | |
committee | +0.017 |
contestant | +0.017 |
brawl | +0.016 |
rookies | +0.016 |
duel | +0.015 |
Neg logit contributions | |
His | -0.019 |
HIS | -0.018 |
his | -0.018 |
his | -0.017 |
He | -0.017 |
Token | one |
Token ID | 530 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
committee | +0.002 |
contestant | +0.002 |
brawl | +0.002 |
rookies | +0.002 |
duel | +0.002 |
Neg logit contributions | |
His | -0.002 |
HIS | -0.002 |
his | -0.002 |
his | -0.002 |
He | -0.002 |
Token | area |
Token ID | 1989 |
Feature activation | -5.33e-02 |
Pos logit contributions | |
committee | +0.215 |
contestant | +0.211 |
brawl | +0.204 |
rookies | +0.193 |
duel | +0.190 |
Neg logit contributions | |
His | -0.236 |
HIS | -0.219 |
his | -0.213 |
his | -0.213 |
He | -0.210 |
Token | in |
Token ID | 287 |
Feature activation | +3.69e-03 |
Pos logit contributions | |
His | +0.045 |
HIS | +0.041 |
his | +0.040 |
his | +0.040 |
him | +0.039 |
Neg logit contributions | |
committee | -0.055 |
contestant | -0.054 |
brawl | -0.053 |
rookies | -0.050 |
duel | -0.049 |
Token | which |
Token ID | 543 |
Feature activation | -9.59e-02 |
Pos logit contributions | |
committee | +0.004 |
contestant | +0.004 |
brawl | +0.004 |
rookies | +0.003 |
duel | +0.003 |
Neg logit contributions | |
His | -0.003 |
HIS | -0.003 |
his | -0.003 |
his | -0.003 |
He | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | +6.40e-04 |
Pos logit contributions | |
His | +0.089 |
HIS | +0.083 |
his | +0.081 |
his | +0.080 |
him | +0.079 |
Neg logit contributions | |
committee | -0.091 |
contestant | -0.089 |
brawl | -0.086 |
rookies | -0.081 |
duel | -0.080 |
Token | government |
Token ID | 1230 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
committee | +0.001 |
contestant | +0.001 |
brawl | +0.001 |
rookies | +0.001 |
duel | +0.001 |
Neg logit contributions | |
His | -0.001 |
HIS | -0.001 |
his | -0.000 |
his | -0.000 |
him | -0.000 |
Token | wasn |
Token ID | 2492 |
Feature activation | -6.20e-02 |
Pos logit contributions | |
His | +0.142 |
HIS | +0.131 |
his | +0.127 |
his | +0.127 |
him | +0.125 |
Neg logit contributions | |
committee | -0.150 |
contestant | -0.149 |
brawl | -0.143 |
rookies | -0.136 |
duel | -0.134 |
Token | � |
Token ID | 247 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
committee | +0.003 |
contestant | +0.003 |
brawl | +0.002 |
rookies | +0.002 |
duel | +0.002 |
Neg logit contributions | |
His | -0.002 |
HIS | -0.002 |
his | -0.002 |
his | -0.002 |
He | -0.002 |
Token | s |
Token ID | 82 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
His | +0.025 |
HIS | +0.023 |
his | +0.023 |
his | +0.023 |
He | +0.022 |
Neg logit contributions | |
committee | -0.015 |
contestant | -0.015 |
brawl | -0.014 |
rookies | -0.013 |
duel | -0.013 |
Token | going |
Token ID | 1016 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
committee | +0.152 |
contestant | +0.150 |
brawl | +0.143 |
rookies | +0.135 |
duel | +0.133 |
Neg logit contributions | |
His | -0.203 |
HIS | -0.191 |
his | -0.186 |
his | -0.185 |
He | -0.183 |
Token | to |
Token ID | 284 |
Feature activation | +9.55e-02 |
Pos logit contributions | |
committee | +0.088 |
contestant | +0.087 |
brawl | +0.083 |
rookies | +0.079 |
duel | +0.078 |
Neg logit contributions | |
His | -0.107 |
HIS | -0.100 |
his | -0.097 |
his | -0.097 |
He | -0.096 |
Token | be |
Token ID | 307 |
Feature activation | -3.70e-01 |
Pos logit contributions | |
committee | +0.108 |
contestant | +0.108 |
brawl | +0.104 |
rookies | +0.100 |
duel | +0.099 |
Neg logit contributions | |
His | -0.071 |
HIS | -0.065 |
his | -0.062 |
his | -0.061 |
He | -0.061 |
Token | good |
Token ID | 922 |
Feature activation | -7.09e-02 |
Pos logit contributions | |
His | +0.370 |
HIS | +0.346 |
his | +0.336 |
his | +0.336 |
him | +0.331 |
Neg logit contributions | |
committee | -0.326 |
contestant | -0.321 |
brawl | -0.308 |
rookies | -0.291 |
duel | -0.286 |
Token | for |
Token ID | 329 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
His | +0.055 |
HIS | +0.050 |
his | +0.049 |
his | +0.048 |
He | +0.047 |
Neg logit contributions | |
committee | -0.078 |
contestant | -0.077 |
brawl | -0.075 |
rookies | -0.072 |
duel | -0.071 |
Token | the |
Token ID | 262 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
His | +0.161 |
HIS | +0.148 |
his | +0.144 |
his | +0.141 |
He | +0.141 |
Neg logit contributions | |
committee | -0.184 |
contestant | -0.183 |
brawl | -0.176 |
rookies | -0.168 |
duel | -0.166 |
Token | overall |
Token ID | 4045 |
Feature activation | +2.65e-02 |
Pos logit contributions | |
contestant | +0.028 |
committee | +0.028 |
brawl | +0.028 |
rookies | +0.027 |
duel | +0.026 |
Neg logit contributions | |
His | -0.015 |
HIS | -0.013 |
his | -0.013 |
He | -0.012 |
his | -0.012 |
Token | health |
Token ID | 1535 |
Feature activation | -3.76e-01 |
Pos logit contributions | |
committee | +0.032 |
contestant | +0.032 |
brawl | +0.031 |
rookies | +0.030 |
duel | +0.030 |
Neg logit contributions | |
His | -0.017 |
HIS | -0.015 |
his | -0.015 |
his | -0.014 |
He | -0.014 |
Token | of |
Token ID | 286 |
Feature activation | +2.59e-03 |
Pos logit contributions | |
His | +0.355 |
HIS | +0.329 |
his | +0.320 |
his | +0.318 |
He | +0.315 |
Neg logit contributions | |
committee | -0.349 |
contestant | -0.346 |
brawl | -0.333 |
rookies | -0.316 |
duel | -0.311 |
Token | , |
Token ID | 11 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
His | +0.011 |
HIS | +0.010 |
his | +0.010 |
his | +0.010 |
He | +0.010 |
Neg logit contributions | |
committee | -0.012 |
contestant | -0.012 |
brawl | -0.011 |
rookies | -0.011 |
duel | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | +9.26e-02 |
Pos logit contributions | |
His | +0.019 |
HIS | +0.018 |
his | +0.018 |
his | +0.018 |
He | +0.017 |
Neg logit contributions | |
committee | -0.016 |
contestant | -0.016 |
brawl | -0.015 |
rookies | -0.015 |
duel | -0.014 |
Token | not |
Token ID | 407 |
Feature activation | -8.83e-02 |
Pos logit contributions | |
committee | +0.084 |
contestant | +0.083 |
brawl | +0.080 |
rookies | +0.076 |
duel | +0.075 |
Neg logit contributions | |
His | -0.090 |
HIS | -0.084 |
his | -0.082 |
his | -0.081 |
He | -0.080 |
Token | in |
Token ID | 287 |
Feature activation | +2.19e-02 |
Pos logit contributions | |
His | +0.087 |
HIS | +0.081 |
his | +0.078 |
his | +0.078 |
He | +0.077 |
Neg logit contributions | |
committee | -0.079 |
contestant | -0.079 |
brawl | -0.075 |
rookies | -0.071 |
duel | -0.070 |
Token | the |
Token ID | 262 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
committee | +0.020 |
contestant | +0.020 |
brawl | +0.019 |
rookies | +0.018 |
duel | +0.018 |
Neg logit contributions | |
His | -0.021 |
HIS | -0.019 |
his | -0.019 |
his | -0.019 |
He | -0.019 |
Token | much |
Token ID | 881 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
committee | +0.020 |
contestant | +0.020 |
brawl | +0.019 |
rookies | +0.018 |
duel | +0.018 |
Neg logit contributions | |
His | -0.019 |
HIS | -0.018 |
his | -0.018 |
his | -0.017 |
He | -0.017 |
Token | bigger |
Token ID | 5749 |
Feature activation | -3.64e-02 |
Pos logit contributions | |
His | +0.014 |
HIS | +0.013 |
his | +0.012 |
his | +0.012 |
He | +0.012 |
Neg logit contributions | |
committee | -0.014 |
contestant | -0.014 |
brawl | -0.013 |
rookies | -0.012 |
duel | -0.012 |
Token | sphere |
Token ID | 16558 |
Feature activation | -7.49e-04 |
Pos logit contributions | |
His | +0.032 |
HIS | +0.030 |
his | +0.029 |
his | +0.028 |
He | +0.028 |
Neg logit contributions | |
committee | -0.036 |
contestant | -0.036 |
brawl | -0.035 |
rookies | -0.033 |
duel | -0.033 |
Token | of |
Token ID | 286 |
Feature activation | -5.33e-04 |
Pos logit contributions | |
His | +0.001 |
HIS | +0.001 |
his | +0.001 |
his | +0.001 |
He | +0.001 |
Neg logit contributions | |
committee | -0.001 |
contestant | -0.001 |
brawl | -0.001 |
rookies | -0.001 |
duel | -0.001 |
Token | programming |
Token ID | 8300 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
His | +0.000 |
HIS | +0.000 |
his | +0.000 |
He | +0.000 |
his | +0.000 |
Neg logit contributions | |
committee | -0.001 |
contestant | -0.001 |
brawl | -0.000 |
rookies | -0.000 |
duel | -0.000 |
Token | , |
Token ID | 11 |
Feature activation | -2.46e-02 |
Pos logit contributions | |
His | +0.018 |
HIS | +0.017 |
his | +0.016 |
He | +0.016 |
his | +0.016 |
Neg logit contributions | |
committee | -0.021 |
contestant | -0.021 |
brawl | -0.020 |
rookies | -0.019 |
duel | -0.019 |
Token | is |
Token ID | 318 |
Feature activation | -5.67e-02 |
Pos logit contributions | |
committee | +0.223 |
contestant | +0.220 |
brawl | +0.211 |
rookies | +0.199 |
duel | +0.196 |
Neg logit contributions | |
His | -0.250 |
HIS | -0.232 |
his | -0.227 |
his | -0.227 |
He | -0.223 |
Token | Media |
Token ID | 6343 |
Feature activation | -7.60e-02 |
Pos logit contributions | |
His | +0.058 |
HIS | +0.055 |
his | +0.053 |
his | +0.053 |
him | +0.052 |
Neg logit contributions | |
committee | -0.048 |
contestant | -0.048 |
brawl | -0.045 |
rookies | -0.043 |
duel | -0.042 |
Token | Day |
Token ID | 3596 |
Feature activation | +5.27e-02 |
Pos logit contributions | |
His | +0.075 |
HIS | +0.070 |
his | +0.068 |
his | +0.067 |
He | +0.067 |
Neg logit contributions | |
committee | -0.068 |
contestant | -0.067 |
brawl | -0.065 |
rookies | -0.061 |
duel | -0.060 |
Token | . |
Token ID | 13 |
Feature activation | +3.84e-02 |
Pos logit contributions | |
committee | +0.051 |
contestant | +0.050 |
brawl | +0.048 |
rookies | +0.045 |
duel | +0.045 |
Neg logit contributions | |
His | -0.049 |
HIS | -0.045 |
his | -0.044 |
his | -0.044 |
him | -0.043 |
Token | This |
Token ID | 770 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
committee | +0.036 |
contestant | +0.035 |
brawl | +0.034 |
rookies | +0.032 |
duel | +0.032 |
Neg logit contributions | |
His | -0.037 |
HIS | -0.034 |
his | -0.033 |
his | -0.033 |
He | -0.033 |
Token | year |
Token ID | 614 |
Feature activation | -5.39e-02 |
Pos logit contributions | |
committee | +0.023 |
contestant | +0.023 |
brawl | +0.022 |
rookies | +0.021 |
duel | +0.021 |
Neg logit contributions | |
His | -0.022 |
HIS | -0.020 |
his | -0.020 |
his | -0.020 |
He | -0.019 |
Token | � |
Token ID | 447 |
Feature activation | +4.19e-03 |
Pos logit contributions | |
His | +0.050 |
HIS | +0.046 |
his | +0.045 |
his | +0.045 |
He | +0.044 |
Neg logit contributions | |
committee | -0.052 |
contestant | -0.051 |
brawl | -0.049 |
rookies | -0.046 |
duel | -0.046 |
Token | � |
Token ID | 247 |
Feature activation | -8.41e-03 |
Pos logit contributions | |
committee | +0.004 |
contestant | +0.004 |
brawl | +0.004 |
rookies | +0.003 |
duel | +0.003 |
Neg logit contributions | |
His | -0.004 |
HIS | -0.004 |
his | -0.004 |
his | -0.004 |
He | -0.004 |
Token | s |
Token ID | 82 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
His | +0.008 |
HIS | +0.008 |
his | +0.007 |
his | +0.007 |
He | +0.007 |
Neg logit contributions | |
committee | -0.008 |
contestant | -0.008 |
brawl | -0.007 |
rookies | -0.007 |
duel | -0.007 |
Token | event |
Token ID | 1785 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
committee | +0.163 |
contestant | +0.161 |
brawl | +0.153 |
rookies | +0.144 |
duel | +0.143 |
Neg logit contributions | |
His | -0.199 |
HIS | -0.186 |
his | -0.181 |
his | -0.180 |
He | -0.178 |
Token | , |
Token ID | 11 |
Feature activation | -1.49e-02 |
Pos logit contributions | |
committee | +0.151 |
contestant | +0.150 |
brawl | +0.144 |
rookies | +0.137 |
duel | +0.136 |
Neg logit contributions | |
His | -0.139 |
HIS | -0.129 |
his | -0.125 |
his | -0.124 |
He | -0.123 |
Token | get |
Token ID | 651 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
committee | +0.381 |
contestant | +0.375 |
brawl | +0.355 |
rookies | +0.336 |
duel | +0.328 |
Neg logit contributions | |
His | -0.500 |
HIS | -0.469 |
his | -0.458 |
his | -0.456 |
He | -0.450 |
Token | sick |
Token ID | 6639 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
His | +0.097 |
HIS | +0.090 |
his | +0.087 |
his | +0.086 |
He | +0.085 |
Neg logit contributions | |
committee | -0.106 |
contestant | -0.105 |
brawl | -0.101 |
rookies | -0.096 |
duel | -0.095 |
Token | or |
Token ID | 393 |
Feature activation | +1.47e-02 |
Pos logit contributions | |
committee | +0.024 |
contestant | +0.024 |
brawl | +0.023 |
rookies | +0.022 |
duel | +0.021 |
Neg logit contributions | |
His | -0.021 |
HIS | -0.019 |
his | -0.019 |
his | -0.018 |
He | -0.018 |
Token | otherwise |
Token ID | 4306 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
committee | +0.013 |
contestant | +0.013 |
brawl | +0.013 |
rookies | +0.012 |
duel | +0.012 |
Neg logit contributions | |
His | -0.014 |
HIS | -0.013 |
his | -0.013 |
his | -0.013 |
him | -0.013 |
Token | need |
Token ID | 761 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
His | +0.010 |
HIS | +0.009 |
his | +0.009 |
his | +0.009 |
him | +0.009 |
Neg logit contributions | |
committee | -0.009 |
contestant | -0.009 |
brawl | -0.009 |
rookies | -0.009 |
duel | -0.008 |
Token | health |
Token ID | 1535 |
Feature activation | -3.80e-01 |
Pos logit contributions | |
committee | +0.119 |
contestant | +0.118 |
brawl | +0.113 |
rookies | +0.108 |
duel | +0.106 |
Neg logit contributions | |
His | -0.107 |
HIS | -0.099 |
his | -0.096 |
his | -0.096 |
him | -0.095 |
Token | services |
Token ID | 2594 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
His | +0.285 |
HIS | +0.259 |
his | +0.250 |
his | +0.246 |
He | +0.244 |
Neg logit contributions | |
committee | -0.427 |
contestant | -0.425 |
brawl | -0.411 |
rookies | -0.394 |
duel | -0.390 |
Token | ," |
Token ID | 553 |
Feature activation | -7.46e-02 |
Pos logit contributions | |
His | +0.363 |
HIS | +0.334 |
his | +0.323 |
his | +0.322 |
He | +0.317 |
Neg logit contributions | |
committee | -0.433 |
contestant | -0.430 |
brawl | -0.416 |
rookies | -0.395 |
duel | -0.390 |
Token | CNBC |
Token ID | 34881 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
His | +0.068 |
HIS | +0.063 |
his | +0.061 |
his | +0.061 |
He | +0.060 |
Neg logit contributions | |
committee | -0.072 |
contestant | -0.071 |
brawl | -0.069 |
rookies | -0.065 |
duel | -0.064 |
Token | reported |
Token ID | 2098 |
Feature activation | +3.44e-02 |
Pos logit contributions | |
His | +0.010 |
HIS | +0.009 |
his | +0.009 |
his | +0.009 |
He | +0.009 |
Neg logit contributions | |
committee | -0.009 |
contestant | -0.009 |
brawl | -0.009 |
rookies | -0.009 |
duel | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +2.47e-02 |
Pos logit contributions | |
committee | +0.032 |
contestant | +0.031 |
brawl | +0.030 |
rookies | +0.029 |
duel | +0.028 |
Neg logit contributions | |
His | -0.033 |
HIS | -0.031 |
his | -0.030 |
his | -0.029 |
He | -0.029 |
Token | go |
Token ID | 467 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
committee | +0.089 |
contestant | +0.088 |
brawl | +0.085 |
rookies | +0.081 |
duel | +0.080 |
Neg logit contributions | |
His | -0.077 |
HIS | -0.071 |
his | -0.068 |
his | -0.068 |
He | -0.067 |
Token | upstairs |
Token ID | 26148 |
Feature activation | -3.09e-02 |
Pos logit contributions | |
His | +0.128 |
HIS | +0.118 |
his | +0.115 |
his | +0.114 |
him | +0.112 |
Neg logit contributions | |
committee | -0.151 |
contestant | -0.150 |
brawl | -0.144 |
rookies | -0.138 |
duel | -0.136 |
Token | to |
Token ID | 284 |
Feature activation | +8.45e-02 |
Pos logit contributions | |
His | +0.029 |
HIS | +0.027 |
his | +0.026 |
his | +0.026 |
him | +0.026 |
Neg logit contributions | |
committee | -0.029 |
contestant | -0.029 |
brawl | -0.028 |
rookies | -0.027 |
duel | -0.026 |
Token | check |
Token ID | 2198 |
Feature activation | +2.47e-02 |
Pos logit contributions | |
committee | +0.086 |
contestant | +0.085 |
brawl | +0.082 |
rookies | +0.078 |
duel | +0.077 |
Neg logit contributions | |
His | -0.073 |
HIS | -0.067 |
his | -0.065 |
his | -0.064 |
He | -0.064 |
Token | his |
Token ID | 465 |
Feature activation | -6.85e-01 |
Pos logit contributions | |
contestant | +0.024 |
committee | +0.024 |
brawl | +0.024 |
rookies | +0.023 |
duel | +0.022 |
Neg logit contributions | |
His | -0.022 |
HIS | -0.020 |
his | -0.019 |
He | -0.019 |
his | -0.019 |
Token | health |
Token ID | 1535 |
Feature activation | -3.89e-01 |
Pos logit contributions | |
His | +0.628 |
HIS | +0.582 |
his | +0.565 |
his | +0.558 |
He | +0.553 |
Neg logit contributions | |
committee | -0.658 |
contestant | -0.653 |
brawl | -0.627 |
rookies | -0.597 |
duel | -0.589 |
Token | . |
Token ID | 13 |
Feature activation | +2.57e-02 |
Pos logit contributions | |
His | +0.349 |
HIS | +0.321 |
his | +0.311 |
his | +0.311 |
He | +0.307 |
Neg logit contributions | |
committee | -0.383 |
contestant | -0.378 |
brawl | -0.365 |
rookies | -0.348 |
duel | -0.343 |
Token | However |
Token ID | 2102 |
Feature activation | +5.52e-02 |
Pos logit contributions | |
committee | +0.023 |
contestant | +0.022 |
brawl | +0.021 |
rookies | +0.020 |
duel | +0.020 |
Neg logit contributions | |
His | -0.026 |
HIS | -0.024 |
his | -0.024 |
his | -0.023 |
He | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | -3.66e-02 |
Pos logit contributions | |
committee | +0.054 |
contestant | +0.054 |
brawl | +0.052 |
rookies | +0.049 |
duel | +0.049 |
Neg logit contributions | |
His | -0.050 |
HIS | -0.046 |
his | -0.044 |
his | -0.044 |
He | -0.044 |
Token | Man |
Token ID | 1869 |
Feature activation | -6.72e-02 |
Pos logit contributions | |
His | +0.037 |
HIS | +0.034 |
his | +0.033 |
his | +0.033 |
him | +0.033 |
Neg logit contributions | |
committee | -0.032 |
contestant | -0.032 |
brawl | -0.030 |
rookies | -0.029 |
duel | -0.028 |
Token | j |
Token ID | 73 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
His | +0.064 |
HIS | +0.059 |
his | +0.058 |
his | +0.057 |
He | +0.057 |
Neg logit contributions | |
committee | -0.063 |
contestant | -0.062 |
brawl | -0.060 |
rookies | -0.057 |
duel | -0.056 |
Token | purchased |
Token ID | 8155 |
Feature activation | -3.03e-01 |
Pos logit contributions | |
committee | +0.005 |
contestant | +0.005 |
brawl | +0.005 |
rookies | +0.005 |
duel | +0.005 |
Neg logit contributions | |
His | -0.005 |
HIS | -0.005 |
his | -0.005 |
his | -0.005 |
He | -0.005 |
Token | before |
Token ID | 878 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
His | +0.254 |
HIS | +0.234 |
his | +0.226 |
He | +0.221 |
his | +0.221 |
Neg logit contributions | |
committee | -0.313 |
contestant | -0.312 |
brawl | -0.301 |
rookies | -0.288 |
duel | -0.285 |
Token | the |
Token ID | 262 |
Feature activation | +2.20e-02 |
Pos logit contributions | |
His | +0.026 |
HIS | +0.024 |
his | +0.023 |
his | +0.023 |
him | +0.023 |
Neg logit contributions | |
committee | -0.033 |
contestant | -0.033 |
brawl | -0.032 |
rookies | -0.030 |
duel | -0.030 |
Token | bombings |
Token ID | 26579 |
Feature activation | +3.13e-01 |
Pos logit contributions | |
committee | +0.013 |
contestant | +0.012 |
brawl | +0.011 |
rookies | +0.011 |
duel | +0.010 |
Neg logit contributions | |
His | -0.029 |
HIS | -0.027 |
his | -0.027 |
his | -0.027 |
him | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | +2.74e-02 |
Pos logit contributions | |
committee | +0.297 |
contestant | +0.295 |
brawl | +0.282 |
rookies | +0.269 |
duel | +0.264 |
Neg logit contributions | |
His | -0.293 |
HIS | -0.272 |
his | -0.264 |
his | -0.264 |
He | -0.260 |
Token | He |
Token ID | 679 |
Feature activation | -4.71e-01 |
Pos logit contributions | |
committee | +0.029 |
contestant | +0.029 |
brawl | +0.028 |
rookies | +0.026 |
duel | +0.026 |
Neg logit contributions | |
His | -0.023 |
HIS | -0.021 |
his | -0.020 |
his | -0.020 |
He | -0.020 |
Token | said |
Token ID | 531 |
Feature activation | -2.96e-01 |
Pos logit contributions | |
His | +0.428 |
HIS | +0.395 |
his | +0.385 |
his | +0.381 |
He | +0.376 |
Neg logit contributions | |
committee | -0.458 |
contestant | -0.453 |
brawl | -0.436 |
rookies | -0.416 |
duel | -0.409 |
Token | an |
Token ID | 281 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
His | +0.199 |
HIS | +0.180 |
his | +0.172 |
his | +0.171 |
He | +0.168 |
Neg logit contributions | |
committee | -0.358 |
contestant | -0.355 |
brawl | -0.344 |
rookies | -0.331 |
duel | -0.327 |
Token | FBI |
Token ID | 5349 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
committee | +0.112 |
contestant | +0.110 |
brawl | +0.105 |
rookies | +0.099 |
duel | +0.098 |
Neg logit contributions | |
His | -0.142 |
HIS | -0.133 |
his | -0.129 |
his | -0.129 |
He | -0.128 |
Token | agent |
Token ID | 5797 |
Feature activation | +4.75e-02 |
Pos logit contributions | |
committee | +0.105 |
contestant | +0.103 |
brawl | +0.099 |
rookies | +0.094 |
duel | +0.092 |
Neg logit contributions | |
His | -0.129 |
HIS | -0.120 |
his | -0.117 |
his | -0.117 |
He | -0.116 |
Token | strongly |
Token ID | 7634 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
committee | +0.041 |
contestant | +0.041 |
brawl | +0.039 |
rookies | +0.037 |
duel | +0.036 |
Neg logit contributions | |
His | -0.048 |
HIS | -0.045 |
his | -0.044 |
his | -0.044 |
him | -0.043 |
Token | she |
Token ID | 673 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
His | +0.008 |
HIS | +0.007 |
his | +0.007 |
his | +0.007 |
him | +0.007 |
Neg logit contributions | |
committee | -0.012 |
contestant | -0.011 |
brawl | -0.011 |
rookies | -0.011 |
duel | -0.010 |
Token | was |
Token ID | 373 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
His | +0.185 |
HIS | +0.172 |
his | +0.167 |
his | +0.166 |
him | +0.163 |
Neg logit contributions | |
committee | -0.192 |
contestant | -0.189 |
brawl | -0.181 |
rookies | -0.173 |
duel | -0.169 |
Token | just |
Token ID | 655 |
Feature activation | -3.10e-02 |
Pos logit contributions | |
His | +0.106 |
HIS | +0.099 |
his | +0.096 |
his | +0.096 |
him | +0.095 |
Neg logit contributions | |
committee | -0.095 |
contestant | -0.094 |
brawl | -0.089 |
rookies | -0.085 |
duel | -0.083 |
Token | 12 |
Token ID | 1105 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
His | +0.031 |
HIS | +0.029 |
his | +0.028 |
his | +0.028 |
him | +0.028 |
Neg logit contributions | |
committee | -0.028 |
contestant | -0.027 |
brawl | -0.026 |
rookies | -0.025 |
duel | -0.024 |
Token | - |
Token ID | 12 |
Feature activation | +9.16e-02 |
Pos logit contributions | |
committee | +0.117 |
contestant | +0.116 |
brawl | +0.112 |
rookies | +0.108 |
duel | +0.106 |
Neg logit contributions | |
His | -0.089 |
HIS | -0.081 |
his | -0.078 |
his | -0.077 |
He | -0.077 |
Token | years |
Token ID | 19002 |
Feature activation | -3.92e-01 |
Pos logit contributions | |
committee | +0.127 |
contestant | +0.126 |
brawl | +0.123 |
rookies | +0.119 |
duel | +0.118 |
Neg logit contributions | |
His | -0.045 |
HIS | -0.039 |
his | -0.037 |
his | -0.036 |
He | -0.035 |
Token | - |
Token ID | 12 |
Feature activation | +9.46e-02 |
Pos logit contributions | |
His | +0.384 |
HIS | +0.357 |
his | +0.348 |
his | +0.347 |
him | +0.342 |
Neg logit contributions | |
committee | -0.356 |
contestant | -0.350 |
brawl | -0.335 |
rookies | -0.317 |
duel | -0.313 |
Token | old |
Token ID | 727 |
Feature activation | -7.01e-03 |
Pos logit contributions | |
committee | +0.093 |
contestant | +0.093 |
brawl | +0.089 |
rookies | +0.085 |
duel | +0.084 |
Neg logit contributions | |
His | -0.085 |
HIS | -0.078 |
his | -0.076 |
his | -0.075 |
He | -0.075 |
Token | has |
Token ID | 468 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
His | +0.007 |
HIS | +0.006 |
his | +0.006 |
his | +0.006 |
He | +0.006 |
Neg logit contributions | |
committee | -0.006 |
contestant | -0.006 |
brawl | -0.006 |
rookies | -0.006 |
duel | -0.006 |
Token | raised |
Token ID | 4376 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
His | +0.194 |
HIS | +0.182 |
his | +0.178 |
his | +0.177 |
He | +0.175 |
Neg logit contributions | |
committee | -0.139 |
contestant | -0.138 |
brawl | -0.131 |
rookies | -0.123 |
duel | -0.121 |
Token | questions |
Token ID | 2683 |
Feature activation | -6.04e-03 |
Pos logit contributions | |
His | +0.015 |
HIS | +0.014 |
his | +0.014 |
his | +0.014 |
him | +0.014 |
Neg logit contributions | |
committee | -0.014 |
contestant | -0.014 |
brawl | -0.014 |
rookies | -0.013 |
duel | -0.013 |
Token | proposals |
Token ID | 11628 |
Feature activation | +3.06e-02 |
Pos logit contributions | |
committee | +0.024 |
contestant | +0.024 |
brawl | +0.023 |
rookies | +0.022 |
duel | +0.021 |
Neg logit contributions | |
His | -0.025 |
HIS | -0.024 |
his | -0.023 |
his | -0.023 |
He | -0.022 |
Token | included |
Token ID | 3017 |
Feature activation | -5.30e-02 |
Pos logit contributions | |
contestant | +0.027 |
committee | +0.027 |
brawl | +0.026 |
rookies | +0.024 |
duel | +0.024 |
Neg logit contributions | |
His | -0.030 |
HIS | -0.028 |
his | -0.028 |
his | -0.028 |
He | -0.027 |
Token | selling |
Token ID | 6301 |
Feature activation | -7.13e-02 |
Pos logit contributions | |
His | +0.052 |
HIS | +0.049 |
his | +0.048 |
his | +0.047 |
him | +0.047 |
Neg logit contributions | |
contestant | -0.047 |
committee | -0.047 |
brawl | -0.045 |
rookies | -0.042 |
duel | -0.041 |
Token | bonds |
Token ID | 13100 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
His | +0.060 |
HIS | +0.055 |
his | +0.053 |
his | +0.052 |
He | +0.052 |
Neg logit contributions | |
committee | -0.074 |
contestant | -0.074 |
brawl | -0.071 |
rookies | -0.068 |
duel | -0.067 |
Token | to |
Token ID | 284 |
Feature activation | +8.90e-02 |
Pos logit contributions | |
His | +0.244 |
HIS | +0.226 |
his | +0.220 |
his | +0.219 |
He | +0.216 |
Neg logit contributions | |
committee | -0.242 |
contestant | -0.241 |
brawl | -0.232 |
rookies | -0.220 |
duel | -0.216 |
Token | pay |
Token ID | 1414 |
Feature activation | -3.95e-01 |
Pos logit contributions | |
committee | +0.102 |
contestant | +0.101 |
brawl | +0.098 |
rookies | +0.094 |
duel | +0.093 |
Neg logit contributions | |
His | -0.065 |
HIS | -0.059 |
his | -0.057 |
his | -0.055 |
He | -0.055 |
Token | for |
Token ID | 329 |
Feature activation | -1.94e-01 |
Pos logit contributions | |
His | +0.293 |
HIS | +0.265 |
his | +0.256 |
his | +0.250 |
He | +0.250 |
Neg logit contributions | |
committee | -0.448 |
contestant | -0.445 |
brawl | -0.431 |
rookies | -0.414 |
duel | -0.409 |
Token | some |
Token ID | 617 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
His | +0.170 |
HIS | +0.157 |
his | +0.152 |
his | +0.151 |
He | +0.149 |
Neg logit contributions | |
committee | -0.194 |
contestant | -0.193 |
brawl | -0.185 |
rookies | -0.177 |
duel | -0.174 |
Token | fixes |
Token ID | 13040 |
Feature activation | -2.88e-01 |
Pos logit contributions | |
committee | +0.113 |
contestant | +0.113 |
brawl | +0.108 |
rookies | +0.103 |
duel | +0.102 |
Neg logit contributions | |
His | -0.108 |
HIS | -0.100 |
his | -0.097 |
his | -0.095 |
He | -0.095 |
Token | , |
Token ID | 11 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
His | +0.266 |
HIS | +0.246 |
his | +0.239 |
his | +0.238 |
He | +0.235 |
Neg logit contributions | |
committee | -0.274 |
contestant | -0.273 |
brawl | -0.262 |
rookies | -0.249 |
duel | -0.245 |
Token | requiring |
Token ID | 10616 |
Feature activation | +2.67e-01 |
Pos logit contributions | |
His | +0.031 |
HIS | +0.029 |
his | +0.028 |
his | +0.028 |
He | +0.028 |
Neg logit contributions | |
committee | -0.028 |
contestant | -0.028 |
brawl | -0.026 |
rookies | -0.025 |
duel | -0.024 |