Token | has |
Token ID | 468 |
Feature activation | -2.02e-02 |
Pos logit contributions | |
Instruct | +0.157 |
Anon | +0.148 |
Teg | +0.147 |
Luxem | +0.146 |
Tuls | +0.142 |
Neg logit contributions | |
awaru | -0.149 |
�� | -0.142 |
ulia | -0.140 |
uese | -0.135 |
rogen | -0.135 |
Token | also |
Token ID | 635 |
Feature activation | +6.56e-02 |
Pos logit contributions | |
awaru | +0.039 |
�� | +0.037 |
ulia | +0.036 |
rogen | +0.036 |
Combat | +0.035 |
Neg logit contributions | |
Instruct | -0.036 |
Anon | -0.033 |
Teg | -0.033 |
Luxem | -0.033 |
theless | -0.031 |
Token | upset |
Token ID | 9247 |
Feature activation | -3.51e-01 |
Pos logit contributions | |
Instruct | +0.112 |
Anon | +0.104 |
Teg | +0.104 |
Luxem | +0.102 |
Tuls | +0.098 |
Neg logit contributions | |
awaru | -0.132 |
�� | -0.126 |
ulia | -0.122 |
rogen | -0.121 |
incial | -0.119 |
Token | defenders |
Token ID | 16355 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
awaru | +0.623 |
�� | +0.591 |
ulia | +0.586 |
rogen | +0.580 |
uese | +0.566 |
Neg logit contributions | |
Instruct | -0.665 |
Anon | -0.615 |
Teg | -0.611 |
theless | -0.611 |
Luxem | -0.603 |
Token | of |
Token ID | 286 |
Feature activation | +4.64e-02 |
Pos logit contributions | |
awaru | +0.183 |
�� | +0.175 |
ulia | +0.170 |
uese | +0.165 |
vantage | +0.164 |
Neg logit contributions | |
Instruct | -0.214 |
Anon | -0.202 |
Teg | -0.201 |
Luxem | -0.198 |
theless | -0.195 |
Token | wom |
Token ID | 1116 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
Instruct | +0.089 |
Anon | +0.083 |
Teg | +0.082 |
theless | +0.082 |
Luxem | +0.082 |
Neg logit contributions | |
awaru | -0.083 |
�� | -0.079 |
ulia | -0.077 |
rogen | -0.076 |
uese | -0.075 |
Token | ens |
Token ID | 641 |
Feature activation | -1.14e-02 |
Pos logit contributions | |
Instruct | +2.986 |
Anon | +2.804 |
Teg | +2.783 |
Luxem | +2.761 |
theless | +2.694 |
Neg logit contributions | |
awaru | -2.852 |
�� | -2.746 |
ulia | -2.644 |
vantage | -2.579 |
uese | -2.555 |
Token | � |
Token ID | 447 |
Feature activation | +5.98e-02 |
Pos logit contributions | |
awaru | +0.024 |
�� | +0.024 |
ulia | +0.023 |
rogen | +0.022 |
Combat | +0.022 |
Neg logit contributions | |
Instruct | -0.018 |
Anon | -0.017 |
Teg | -0.017 |
Luxem | -0.017 |
Tuls | -0.016 |
Token | � |
Token ID | 247 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
Instruct | +0.163 |
Anon | +0.156 |
Teg | +0.156 |
Luxem | +0.155 |
theless | +0.154 |
Neg logit contributions | |
awaru | -0.060 |
�� | -0.059 |
ulia | -0.053 |
vantage | -0.050 |
uese | -0.050 |
Token | rights |
Token ID | 2489 |
Feature activation | +3.76e-01 |
Pos logit contributions | |
Instruct | +0.208 |
Anon | +0.195 |
Teg | +0.193 |
Luxem | +0.191 |
Tuls | +0.189 |
Neg logit contributions | |
awaru | -0.199 |
�� | -0.190 |
ulia | -0.186 |
vantage | -0.182 |
uese | -0.182 |
Token | in |
Token ID | 287 |
Feature activation | +5.60e-02 |
Pos logit contributions | |
Instruct | +0.564 |
Anon | +0.519 |
Teg | +0.513 |
theless | +0.509 |
Luxem | +0.505 |
Neg logit contributions | |
awaru | -0.837 |
�� | -0.797 |
ulia | -0.791 |
uese | -0.769 |
vantage | -0.766 |
Token | string |
Token ID | 8841 |
Feature activation | -3.03e-01 |
Pos logit contributions | |
awaru | +0.065 |
�� | +0.061 |
ulia | +0.059 |
vantage | +0.057 |
uese | +0.057 |
Neg logit contributions | |
Instruct | -0.109 |
Anon | -0.104 |
Teg | -0.103 |
Luxem | -0.102 |
Tuls | -0.101 |
Token | format |
Token ID | 18982 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
awaru | +0.600 |
�� | +0.575 |
ulia | +0.564 |
uese | +0.547 |
vantage | +0.547 |
Neg logit contributions | |
Instruct | -0.528 |
Anon | -0.494 |
Teg | -0.485 |
Luxem | -0.483 |
theless | -0.476 |
Token | Switch |
Token ID | 14645 |
Feature activation | -2.88e-01 |
Pos logit contributions | |
awaru | +0.324 |
�� | +0.311 |
ulia | +0.307 |
vantage | +0.297 |
uese | +0.297 |
Neg logit contributions | |
Instruct | -0.242 |
Anon | -0.226 |
Teg | -0.222 |
Luxem | -0.222 |
Tuls | -0.217 |
Token | ($ |
Token ID | 7198 |
Feature activation | +5.00e-02 |
Pos logit contributions | |
awaru | +0.564 |
�� | +0.545 |
ulia | +0.536 |
uese | +0.517 |
vantage | +0.516 |
Neg logit contributions | |
Instruct | -0.503 |
Anon | -0.474 |
Teg | -0.467 |
Luxem | -0.464 |
Tuls | -0.455 |
Token | Al |
Token ID | 2348 |
Feature activation | +3.70e-01 |
Pos logit contributions | |
Instruct | +0.096 |
Anon | +0.091 |
Teg | +0.090 |
Luxem | +0.089 |
Tuls | +0.088 |
Neg logit contributions | |
awaru | -0.090 |
�� | -0.087 |
ulia | -0.085 |
vantage | -0.082 |
uese | -0.082 |
Token | ign |
Token ID | 570 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
Instruct | +0.847 |
Anon | +0.802 |
Teg | +0.799 |
Luxem | +0.793 |
Tuls | +0.780 |
Neg logit contributions | |
awaru | -0.536 |
�� | -0.510 |
ulia | -0.498 |
vantage | -0.483 |
uese | -0.476 |
Token | ) |
Token ID | 8 |
Feature activation | -7.30e-02 |
Pos logit contributions | |
Instruct | +3.052 |
Anon | +2.898 |
Teg | +2.888 |
Luxem | +2.848 |
theless | +2.796 |
Neg logit contributions | |
awaru | -1.742 |
�� | -1.645 |
ulia | -1.582 |
uese | -1.511 |
vantage | -1.507 |
Token | { |
Token ID | 1391 |
Feature activation | +3.71e-02 |
Pos logit contributions | |
awaru | +0.126 |
�� | +0.120 |
ulia | +0.118 |
uese | +0.113 |
vantage | +0.113 |
Neg logit contributions | |
Instruct | -0.144 |
Anon | -0.137 |
Teg | -0.136 |
Luxem | -0.135 |
theless | -0.134 |
Token | Center |
Token ID | 3337 |
Feature activation | +4.15e-01 |
Pos logit contributions | |
Instruct | +0.065 |
Anon | +0.061 |
Luxem | +0.061 |
Teg | +0.061 |
Tuls | +0.059 |
Neg logit contributions | |
awaru | -0.072 |
�� | -0.069 |
ulia | -0.068 |
vantage | -0.066 |
uese | -0.066 |
Token | { |
Token ID | 1391 |
Feature activation | -2.19e-02 |
Pos logit contributions | |
Instruct | +0.709 |
Anon | +0.661 |
Teg | +0.657 |
Luxem | +0.650 |
Tuls | +0.638 |
Neg logit contributions | |
awaru | -0.839 |
�� | -0.811 |
ulia | -0.792 |
vantage | -0.771 |
uese | -0.767 |
Token | $ |
Token ID | 720 |
Feature activation | -5.68e-02 |
Pos logit contributions | |
awaru | +0.042 |
�� | +0.041 |
ulia | +0.040 |
vantage | +0.039 |
uese | +0.038 |
Neg logit contributions | |
Instruct | -0.039 |
Anon | -0.037 |
Luxem | -0.036 |
Teg | -0.036 |
theless | -0.036 |
Token | of |
Token ID | 286 |
Feature activation | +2.57e-02 |
Pos logit contributions | |
awaru | +0.069 |
�� | +0.066 |
ulia | +0.065 |
uese | +0.064 |
vantage | +0.064 |
Neg logit contributions | |
Instruct | -0.068 |
Anon | -0.064 |
Luxem | -0.063 |
Teg | -0.063 |
theless | -0.061 |
Token | giant |
Token ID | 6175 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
Instruct | +0.045 |
Anon | +0.042 |
Teg | +0.042 |
Luxem | +0.041 |
theless | +0.041 |
Neg logit contributions | |
awaru | -0.050 |
�� | -0.048 |
ulia | -0.047 |
vantage | -0.046 |
uese | -0.045 |
Token | flying |
Token ID | 7348 |
Feature activation | -8.65e-03 |
Pos logit contributions | |
awaru | +0.025 |
�� | +0.024 |
ulia | +0.023 |
uese | +0.022 |
rogen | +0.022 |
Neg logit contributions | |
Instruct | -0.023 |
Anon | -0.021 |
Teg | -0.021 |
Luxem | -0.021 |
theless | -0.021 |
Token | men |
Token ID | 1450 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
awaru | +0.016 |
�� | +0.016 |
ulia | +0.015 |
uese | +0.015 |
rogen | +0.015 |
Neg logit contributions | |
Instruct | -0.016 |
Anon | -0.015 |
Teg | -0.015 |
Luxem | -0.015 |
theless | -0.014 |
Token | with |
Token ID | 351 |
Feature activation | +8.52e-02 |
Pos logit contributions | |
awaru | +0.040 |
�� | +0.039 |
ulia | +0.038 |
uese | +0.037 |
vantage | +0.037 |
Neg logit contributions | |
Instruct | -0.031 |
Anon | -0.029 |
Teg | -0.029 |
Luxem | -0.029 |
theless | -0.029 |
Token | wom |
Token ID | 1116 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
Instruct | +0.151 |
Anon | +0.141 |
Teg | +0.139 |
Luxem | +0.138 |
theless | +0.137 |
Neg logit contributions | |
awaru | -0.167 |
�� | -0.160 |
ulia | -0.155 |
uese | -0.151 |
rogen | -0.150 |
Token | bat |
Token ID | 8664 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
Instruct | +2.387 |
Anon | +2.247 |
Teg | +2.235 |
Luxem | +2.209 |
theless | +2.164 |
Neg logit contributions | |
awaru | -2.075 |
�� | -1.988 |
ulia | -1.928 |
vantage | -1.867 |
uese | -1.865 |
Token | be |
Token ID | 307 |
Feature activation | -5.93e-03 |
Pos logit contributions | |
Instruct | +0.365 |
Anon | +0.342 |
Teg | +0.335 |
Luxem | +0.334 |
theless | +0.328 |
Neg logit contributions | |
awaru | -0.365 |
�� | -0.349 |
ulia | -0.344 |
uese | -0.336 |
vantage | -0.333 |
Token | ards |
Token ID | 1371 |
Feature activation | +2.93e-01 |
Pos logit contributions | |
awaru | +0.011 |
�� | +0.011 |
ulia | +0.011 |
vantage | +0.010 |
uese | +0.010 |
Neg logit contributions | |
Instruct | -0.011 |
Anon | -0.010 |
Teg | -0.010 |
Luxem | -0.010 |
Tuls | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +3.00e-02 |
Pos logit contributions | |
Instruct | +0.474 |
Anon | +0.439 |
Teg | +0.435 |
Luxem | +0.431 |
theless | +0.421 |
Neg logit contributions | |
awaru | -0.619 |
�� | -0.597 |
ulia | -0.582 |
uese | -0.567 |
vantage | -0.567 |
Token | They |
Token ID | 1119 |
Feature activation | +2.57e-03 |
Pos logit contributions | |
Instruct | +0.054 |
Anon | +0.051 |
Teg | +0.051 |
theless | +0.051 |
Luxem | +0.050 |
Neg logit contributions | |
awaru | -0.055 |
�� | -0.053 |
ulia | -0.052 |
Combat | -0.051 |
vantage | -0.051 |
Token | during |
Token ID | 1141 |
Feature activation | +1.06e-03 |
Pos logit contributions | |
Instruct | +0.167 |
Teg | +0.154 |
Anon | +0.153 |
Luxem | +0.151 |
theless | +0.149 |
Neg logit contributions | |
awaru | -0.206 |
�� | -0.196 |
ulia | -0.193 |
incial | -0.188 |
rogen | -0.187 |
Token | my |
Token ID | 616 |
Feature activation | +9.64e-02 |
Pos logit contributions | |
Instruct | +0.002 |
Anon | +0.002 |
Teg | +0.002 |
Luxem | +0.002 |
theless | +0.002 |
Neg logit contributions | |
awaru | -0.002 |
�� | -0.002 |
ulia | -0.002 |
uese | -0.002 |
vantage | -0.002 |
Token | years |
Token ID | 812 |
Feature activation | -6.17e-02 |
Pos logit contributions | |
Instruct | +0.179 |
Teg | +0.168 |
Anon | +0.166 |
Luxem | +0.165 |
theless | +0.162 |
Neg logit contributions | |
awaru | -0.177 |
�� | -0.171 |
ulia | -0.169 |
uese | -0.163 |
incial | -0.162 |
Token | at |
Token ID | 379 |
Feature activation | -7.41e-03 |
Pos logit contributions | |
awaru | +0.116 |
�� | +0.110 |
ulia | +0.109 |
uese | +0.105 |
rogen | +0.104 |
Neg logit contributions | |
Instruct | -0.114 |
Teg | -0.106 |
Anon | -0.106 |
Luxem | -0.105 |
theless | -0.103 |
Token | C |
Token ID | 327 |
Feature activation | +2.56e-02 |
Pos logit contributions | |
awaru | +0.014 |
�� | +0.014 |
ulia | +0.014 |
vantage | +0.013 |
uese | +0.013 |
Neg logit contributions | |
Instruct | -0.013 |
Teg | -0.012 |
Anon | -0.012 |
Luxem | -0.012 |
theless | -0.012 |
Token | ign |
Token ID | 570 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
Instruct | +0.055 |
Anon | +0.051 |
Teg | +0.051 |
Luxem | +0.051 |
Tuls | +0.050 |
Neg logit contributions | |
awaru | -0.041 |
�� | -0.039 |
ulia | -0.038 |
vantage | -0.037 |
uese | -0.037 |
Token | a |
Token ID | 64 |
Feature activation | +3.20e-01 |
Pos logit contributions | |
Instruct | +2.221 |
Anon | +2.089 |
Teg | +2.079 |
Luxem | +2.068 |
theless | +2.030 |
Neg logit contributions | |
awaru | -1.861 |
�� | -1.784 |
ulia | -1.749 |
uese | -1.682 |
vantage | -1.679 |
Token | . |
Token ID | 13 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
Instruct | +0.521 |
Anon | +0.479 |
Teg | +0.477 |
Luxem | +0.468 |
theless | +0.467 |
Neg logit contributions | |
awaru | -0.668 |
�� | -0.650 |
ulia | -0.640 |
vantage | -0.624 |
uese | -0.618 |
Token |
|
Token ID | 198 |
Feature activation | +7.21e-02 |
Pos logit contributions | |
Instruct | +0.197 |
Anon | +0.183 |
Teg | +0.183 |
Luxem | +0.182 |
theless | +0.180 |
Neg logit contributions | |
awaru | -0.174 |
�� | -0.166 |
ulia | -0.165 |
vantage | -0.159 |
Combat | -0.157 |
Token |
|
Token ID | 198 |
Feature activation | +8.45e-02 |
Pos logit contributions | |
Instruct | +0.125 |
Anon | +0.116 |
Teg | +0.115 |
Luxem | +0.114 |
theless | +0.113 |
Neg logit contributions | |
awaru | -0.143 |
ulia | -0.136 |
�� | -0.135 |
Combat | -0.131 |
uese | -0.131 |
Token | The |
Token ID | 464 |
Feature activation | -6.27e-03 |
Pos logit contributions | |
Instruct | +0.176 |
Teg | +0.164 |
Anon | +0.164 |
Luxem | +0.163 |
theless | +0.162 |
Neg logit contributions | |
awaru | -0.139 |
�� | -0.133 |
ulia | -0.131 |
Combat | -0.129 |
vantage | -0.126 |
Token | the |
Token ID | 262 |
Feature activation | -6.37e-02 |
Pos logit contributions | |
awaru | +0.302 |
�� | +0.292 |
ulia | +0.285 |
vantage | +0.276 |
uese | +0.275 |
Neg logit contributions | |
Instruct | -0.290 |
Anon | -0.275 |
Luxem | -0.270 |
Teg | -0.270 |
theless | -0.264 |
Token | reign |
Token ID | 13580 |
Feature activation | -3.03e-02 |
Pos logit contributions | |
awaru | +0.116 |
�� | +0.113 |
ulia | +0.110 |
incial | +0.105 |
uese | +0.105 |
Neg logit contributions | |
Instruct | -0.119 |
Anon | -0.112 |
Luxem | -0.111 |
Teg | -0.111 |
theless | -0.109 |
Token | of |
Token ID | 286 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
awaru | +0.052 |
�� | +0.049 |
ulia | +0.048 |
uese | +0.046 |
vantage | +0.046 |
Neg logit contributions | |
Instruct | -0.061 |
Anon | -0.058 |
Teg | -0.057 |
Luxem | -0.057 |
theless | -0.056 |
Token | social |
Token ID | 1919 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
Instruct | +0.024 |
Anon | +0.023 |
Luxem | +0.022 |
Teg | +0.022 |
theless | +0.022 |
Neg logit contributions | |
awaru | -0.033 |
�� | -0.031 |
ulia | -0.031 |
rogen | -0.030 |
incial | -0.030 |
Token | justice |
Token ID | 5316 |
Feature activation | -8.83e-02 |
Pos logit contributions | |
Instruct | +0.387 |
Anon | +0.366 |
Teg | +0.363 |
Luxem | +0.359 |
theless | +0.354 |
Neg logit contributions | |
awaru | -0.378 |
�� | -0.361 |
ulia | -0.351 |
uese | -0.345 |
rogen | -0.342 |
Token | indo |
Token ID | 23046 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
awaru | +0.165 |
�� | +0.154 |
ulia | +0.153 |
rogen | +0.149 |
uese | +0.147 |
Neg logit contributions | |
Instruct | -0.163 |
Anon | -0.152 |
Teg | -0.149 |
theless | -0.149 |
Luxem | -0.148 |
Token | ct |
Token ID | 310 |
Feature activation | -3.82e-01 |
Pos logit contributions | |
Instruct | +1.334 |
Anon | +1.205 |
Teg | +1.190 |
Luxem | +1.163 |
Tuls | +1.128 |
Neg logit contributions | |
awaru | -2.521 |
�� | -2.444 |
ulia | -2.392 |
uese | -2.344 |
vantage | -2.335 |
Token | rin |
Token ID | 12769 |
Feature activation | +4.10e-02 |
Pos logit contributions | |
awaru | +0.684 |
�� | +0.653 |
ulia | +0.644 |
rogen | +0.623 |
vantage | +0.623 |
Neg logit contributions | |
Instruct | -0.739 |
Anon | -0.699 |
Teg | -0.694 |
Luxem | -0.691 |
theless | -0.671 |
Token | ators |
Token ID | 2024 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
Instruct | +0.083 |
Anon | +0.078 |
Teg | +0.077 |
Luxem | +0.077 |
theless | +0.075 |
Neg logit contributions | |
awaru | -0.071 |
�� | -0.067 |
ulia | -0.065 |
uese | -0.063 |
vantage | -0.063 |
Token | , |
Token ID | 11 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
awaru | +0.260 |
�� | +0.250 |
ulia | +0.247 |
uese | +0.241 |
rogen | +0.239 |
Neg logit contributions | |
Instruct | -0.195 |
Anon | -0.182 |
Teg | -0.179 |
Luxem | -0.178 |
theless | -0.177 |
Token | academics |
Token ID | 22321 |
Feature activation | -7.22e-02 |
Pos logit contributions | |
awaru | +0.020 |
�� | +0.019 |
ulia | +0.019 |
rogen | +0.018 |
uese | +0.018 |
Neg logit contributions | |
Instruct | -0.017 |
Anon | -0.016 |
Luxem | -0.016 |
Teg | -0.016 |
theless | -0.016 |
Token | , |
Token ID | 11 |
Feature activation | +7.88e-02 |
Pos logit contributions | |
Instruct | +0.157 |
Anon | +0.145 |
Teg | +0.144 |
Luxem | +0.141 |
theless | +0.139 |
Neg logit contributions | |
awaru | -0.228 |
�� | -0.219 |
ulia | -0.216 |
rogen | -0.209 |
uese | -0.209 |
Token | part |
Token ID | 636 |
Feature activation | -3.26e-02 |
Pos logit contributions | |
Instruct | +0.133 |
Anon | +0.122 |
Teg | +0.122 |
Luxem | +0.120 |
theless | +0.119 |
Neg logit contributions | |
awaru | -0.159 |
�� | -0.152 |
ulia | -0.150 |
rogen | -0.146 |
uese | -0.145 |
Token | of |
Token ID | 286 |
Feature activation | +3.07e-02 |
Pos logit contributions | |
awaru | +0.056 |
�� | +0.053 |
ulia | +0.052 |
uese | +0.050 |
vantage | +0.050 |
Neg logit contributions | |
Instruct | -0.066 |
Anon | -0.062 |
Teg | -0.062 |
Luxem | -0.061 |
Tuls | -0.060 |
Token | her |
Token ID | 607 |
Feature activation | +1.89e-03 |
Pos logit contributions | |
Instruct | +0.050 |
Anon | +0.045 |
Teg | +0.045 |
theless | +0.045 |
Luxem | +0.045 |
Neg logit contributions | |
awaru | -0.064 |
�� | -0.061 |
ulia | -0.060 |
uese | -0.058 |
vantage | -0.058 |
Token | cac |
Token ID | 40428 |
Feature activation | -6.75e-02 |
Pos logit contributions | |
Instruct | +0.003 |
Anon | +0.003 |
Luxem | +0.003 |
Teg | +0.003 |
Tuls | +0.003 |
Neg logit contributions | |
awaru | -0.004 |
�� | -0.003 |
ulia | -0.003 |
vantage | -0.003 |
rogen | -0.003 |
Token | het |
Token ID | 3202 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
awaru | +0.113 |
�� | +0.106 |
ulia | +0.106 |
uese | +0.101 |
vantage | +0.101 |
Neg logit contributions | |
Instruct | -0.140 |
Anon | -0.132 |
Teg | -0.131 |
Luxem | -0.129 |
Tuls | -0.128 |
Token | is |
Token ID | 318 |
Feature activation | -7.26e-03 |
Pos logit contributions | |
Instruct | +1.993 |
Anon | +1.875 |
Teg | +1.864 |
Luxem | +1.843 |
Tuls | +1.807 |
Neg logit contributions | |
awaru | -1.742 |
�� | -1.669 |
ulia | -1.620 |
uese | -1.570 |
vantage | -1.568 |
Token | that |
Token ID | 326 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
awaru | +0.015 |
�� | +0.014 |
ulia | +0.014 |
rogen | +0.014 |
incial | +0.013 |
Neg logit contributions | |
Instruct | -0.012 |
Anon | -0.011 |
Teg | -0.011 |
Luxem | -0.011 |
theless | -0.011 |
Token | her |
Token ID | 607 |
Feature activation | +7.64e-02 |
Pos logit contributions | |
awaru | +0.026 |
�� | +0.025 |
ulia | +0.024 |
rogen | +0.024 |
uese | +0.024 |
Neg logit contributions | |
Instruct | -0.017 |
Anon | -0.016 |
Teg | -0.015 |
theless | -0.015 |
Luxem | -0.015 |
Token | poems |
Token ID | 31888 |
Feature activation | +7.47e-02 |
Pos logit contributions | |
Instruct | +0.140 |
Anon | +0.131 |
Teg | +0.129 |
Luxem | +0.129 |
Tuls | +0.126 |
Neg logit contributions | |
awaru | -0.144 |
�� | -0.137 |
ulia | -0.134 |
rogen | -0.130 |
uese | -0.130 |
Token | are |
Token ID | 389 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
Instruct | +0.147 |
Anon | +0.140 |
Teg | +0.138 |
Luxem | +0.137 |
Tuls | +0.135 |
Neg logit contributions | |
awaru | -0.129 |
�� | -0.123 |
ulia | -0.121 |
rogen | -0.117 |
vantage | -0.117 |
Token | is |
Token ID | 318 |
Feature activation | +4.17e-03 |
Pos logit contributions | |
awaru | +0.242 |
�� | +0.234 |
ulia | +0.230 |
vantage | +0.225 |
uese | +0.223 |
Neg logit contributions | |
Instruct | -0.194 |
Anon | -0.182 |
Teg | -0.179 |
Luxem | -0.177 |
Tuls | -0.174 |
Token | for |
Token ID | 329 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
Instruct | +0.007 |
Anon | +0.007 |
Luxem | +0.007 |
Teg | +0.007 |
theless | +0.007 |
Neg logit contributions | |
awaru | -0.008 |
�� | -0.008 |
ulia | -0.007 |
rogen | -0.007 |
Combat | -0.007 |
Token | surgery |
Token ID | 8185 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
Instruct | +0.030 |
Anon | +0.028 |
Teg | +0.028 |
Luxem | +0.027 |
theless | +0.027 |
Neg logit contributions | |
awaru | -0.042 |
�� | -0.041 |
ulia | -0.040 |
vantage | -0.039 |
Combat | -0.039 |
Token | to |
Token ID | 284 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
awaru | +0.475 |
ulia | +0.446 |
�� | +0.446 |
vantage | +0.432 |
uese | +0.432 |
Neg logit contributions | |
Instruct | -0.418 |
Anon | -0.388 |
Teg | -0.387 |
Luxem | -0.386 |
theless | -0.383 |
Token | real |
Token ID | 1103 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
awaru | +0.033 |
�� | +0.032 |
ulia | +0.031 |
rogen | +0.031 |
uese | +0.031 |
Neg logit contributions | |
Instruct | -0.027 |
Anon | -0.026 |
theless | -0.025 |
Teg | -0.025 |
Luxem | -0.025 |
Token | ign |
Token ID | 570 |
Feature activation | +9.98e-01 |
Pos logit contributions | |
awaru | +0.279 |
�� | +0.263 |
vantage | +0.248 |
ulia | +0.248 |
uese | +0.241 |
Neg logit contributions | |
Instruct | -0.522 |
Anon | -0.498 |
Teg | -0.495 |
Luxem | -0.495 |
theless | -0.490 |
Token | and |
Token ID | 290 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
Instruct | +1.656 |
Anon | +1.544 |
Teg | +1.520 |
Luxem | +1.507 |
theless | +1.485 |
Neg logit contributions | |
awaru | -2.067 |
�� | -1.991 |
ulia | -1.947 |
vantage | -1.910 |
uese | -1.893 |
Token | fix |
Token ID | 4259 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
awaru | +0.032 |
�� | +0.030 |
rogen | +0.030 |
ulia | +0.029 |
Combat | +0.029 |
Neg logit contributions | |
Instruct | -0.027 |
Anon | -0.026 |
theless | -0.025 |
Teg | -0.025 |
Luxem | -0.025 |
Token | ate |
Token ID | 378 |
Feature activation | +4.22e-01 |
Pos logit contributions | |
Instruct | +0.370 |
Anon | +0.347 |
Teg | +0.341 |
Luxem | +0.339 |
theless | +0.338 |
Neg logit contributions | |
awaru | -0.450 |
�� | -0.434 |
ulia | -0.422 |
Combat | -0.412 |
vantage | -0.412 |
Token | the |
Token ID | 262 |
Feature activation | -2.37e-02 |
Pos logit contributions | |
Instruct | +0.705 |
Anon | +0.658 |
Teg | +0.647 |
Luxem | +0.642 |
theless | +0.638 |
Neg logit contributions | |
awaru | -0.865 |
�� | -0.833 |
ulia | -0.814 |
vantage | -0.796 |
rogen | -0.792 |
Token | fracture |
Token ID | 31846 |
Feature activation | -8.77e-02 |
Pos logit contributions | |
awaru | +0.042 |
�� | +0.040 |
ulia | +0.039 |
rogen | +0.038 |
incial | +0.037 |
Neg logit contributions | |
Instruct | -0.046 |
Anon | -0.044 |
Teg | -0.043 |
theless | -0.043 |
Luxem | -0.043 |
Token | � |
Token ID | 251 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
awaru | +0.082 |
�� | +0.078 |
ulia | +0.074 |
vantage | +0.069 |
uese | +0.069 |
Neg logit contributions | |
Instruct | -0.195 |
Anon | -0.186 |
Teg | -0.186 |
Luxem | -0.184 |
theless | -0.181 |
Token |
|
Token ID | 198 |
Feature activation | -7.90e-03 |
Pos logit contributions | |
Instruct | +0.227 |
Anon | +0.213 |
Teg | +0.211 |
Luxem | +0.210 |
Tuls | +0.205 |
Neg logit contributions | |
awaru | -0.233 |
�� | -0.224 |
ulia | -0.221 |
vantage | -0.213 |
uese | -0.212 |
Token |
|
Token ID | 198 |
Feature activation | +3.12e-02 |
Pos logit contributions | |
awaru | +0.016 |
�� | +0.015 |
ulia | +0.015 |
vantage | +0.015 |
Combat | +0.015 |
Neg logit contributions | |
Instruct | -0.014 |
Anon | -0.013 |
Teg | -0.012 |
Luxem | -0.012 |
theless | -0.012 |
Token | Facebook |
Token ID | 12025 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
Instruct | +0.063 |
Anon | +0.059 |
Teg | +0.058 |
Luxem | +0.058 |
theless | +0.057 |
Neg logit contributions | |
awaru | -0.054 |
�� | -0.052 |
ulia | -0.050 |
Combat | -0.049 |
vantage | -0.049 |
Token | Twitter |
Token ID | 3009 |
Feature activation | +6.13e-01 |
Pos logit contributions | |
Instruct | +0.185 |
Teg | +0.162 |
Anon | +0.161 |
Luxem | +0.156 |
Tuls | +0.151 |
Neg logit contributions | |
awaru | -0.486 |
�� | -0.475 |
ulia | -0.469 |
vantage | -0.460 |
uese | -0.458 |
Token | Pinterest |
Token ID | 17334 |
Feature activation | +9.84e-01 |
Pos logit contributions | |
Instruct | +0.716 |
Anon | +0.641 |
Teg | +0.639 |
Luxem | +0.632 |
theless | +0.616 |
Neg logit contributions | |
awaru | -1.559 |
ulia | -1.499 |
�� | -1.498 |
uese | -1.458 |
vantage | -1.448 |
Token | Liverpool |
Token ID | 11761 |
Feature activation | +3.75e-01 |
Pos logit contributions | |
Instruct | +1.669 |
Anon | +1.555 |
Teg | +1.543 |
Luxem | +1.523 |
theless | +1.489 |
Neg logit contributions | |
awaru | -1.998 |
�� | -1.926 |
ulia | -1.882 |
uese | -1.831 |
vantage | -1.830 |
Token | � |
Token ID | 447 |
Feature activation | +2.26e-01 |
Pos logit contributions | |
Instruct | +0.678 |
Anon | +0.631 |
Teg | +0.625 |
Luxem | +0.619 |
theless | +0.614 |
Neg logit contributions | |
awaru | -0.711 |
�� | -0.686 |
ulia | -0.672 |
uese | -0.658 |
vantage | -0.650 |
Token | � |
Token ID | 247 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
Instruct | +0.553 |
Anon | +0.528 |
Teg | +0.526 |
Luxem | +0.523 |
theless | +0.517 |
Neg logit contributions | |
awaru | -0.290 |
�� | -0.283 |
ulia | -0.263 |
vantage | -0.254 |
uese | -0.251 |
Token | s |
Token ID | 82 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
Instruct | +0.327 |
Anon | +0.306 |
Teg | +0.302 |
Luxem | +0.302 |
Tuls | +0.296 |
Neg logit contributions | |
awaru | -0.352 |
�� | -0.341 |
ulia | -0.334 |
vantage | -0.327 |
uese | -0.327 |
Token | Adam |
Token ID | 7244 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
Instruct | +0.292 |
Anon | +0.275 |
Teg | +0.267 |
Luxem | +0.265 |
theless | +0.264 |
Neg logit contributions | |
awaru | -0.333 |
�� | -0.325 |
ulia | -0.315 |
uese | -0.309 |
rogen | -0.305 |
Token | Flynn |
Token ID | 14654 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
awaru | +0.032 |
�� | +0.031 |
ulia | +0.030 |
uese | +0.029 |
Combat | +0.029 |
Neg logit contributions | |
Instruct | -0.021 |
Anon | -0.019 |
Luxem | -0.019 |
Teg | -0.019 |
theless | -0.018 |
Token | announced |
Token ID | 3414 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
awaru | +0.384 |
�� | +0.373 |
ulia | +0.359 |
rogen | +0.354 |
uese | +0.353 |
Neg logit contributions | |
Instruct | -0.309 |
Teg | -0.291 |
Anon | -0.290 |
Luxem | -0.286 |
theless | -0.280 |
Token | that |
Token ID | 326 |
Feature activation | +5.91e-02 |
Pos logit contributions | |
Instruct | +0.040 |
Anon | +0.037 |
Teg | +0.037 |
Luxem | +0.037 |
theless | +0.036 |
Neg logit contributions | |
awaru | -0.049 |
�� | -0.048 |
ulia | -0.046 |
uese | -0.045 |
rogen | -0.045 |
Token | Jen |
Token ID | 13364 |
Feature activation | +7.90e-02 |
Pos logit contributions | |
Instruct | +0.101 |
Anon | +0.094 |
Teg | +0.093 |
Luxem | +0.092 |
theless | +0.092 |
Neg logit contributions | |
awaru | -0.117 |
�� | -0.114 |
ulia | -0.111 |
rogen | -0.108 |
uese | -0.108 |
Token | was |
Token ID | 373 |
Feature activation | +3.91e-02 |
Pos logit contributions | |
Instruct | +0.136 |
Anon | +0.126 |
Teg | +0.125 |
Luxem | +0.124 |
theless | +0.122 |
Neg logit contributions | |
awaru | -0.157 |
�� | -0.153 |
ulia | -0.150 |
uese | -0.145 |
rogen | -0.144 |
Token | pregnant |
Token ID | 10423 |
Feature activation | +9.83e-01 |
Pos logit contributions | |
Instruct | +0.064 |
theless | +0.060 |
Anon | +0.059 |
Teg | +0.059 |
Luxem | +0.058 |
Neg logit contributions | |
awaru | -0.080 |
�� | -0.078 |
ulia | -0.076 |
rogen | -0.074 |
Combat | -0.073 |
Token | . |
Token ID | 13 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
Instruct | +1.580 |
Anon | +1.465 |
Teg | +1.455 |
Luxem | +1.433 |
Tuls | +1.398 |
Neg logit contributions | |
awaru | -2.092 |
�� | -2.018 |
ulia | -1.972 |
uese | -1.921 |
vantage | -1.920 |
Token | Flynn |
Token ID | 14654 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
Instruct | +0.243 |
behav | +0.228 |
theless | +0.228 |
Anon | +0.228 |
Teg | +0.226 |
Neg logit contributions | |
awaru | -0.248 |
ulia | -0.242 |
�� | -0.241 |
Combat | -0.233 |
incial | -0.231 |
Token | and |
Token ID | 290 |
Feature activation | +4.98e-02 |
Pos logit contributions | |
awaru | +0.215 |
�� | +0.208 |
ulia | +0.202 |
rogen | +0.200 |
uese | +0.196 |
Neg logit contributions | |
Instruct | -0.197 |
Teg | -0.186 |
Anon | -0.185 |
Luxem | -0.182 |
Tuls | -0.179 |
Token | his |
Token ID | 465 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
Instruct | +0.071 |
Anon | +0.065 |
Teg | +0.064 |
Luxem | +0.064 |
theless | +0.063 |
Neg logit contributions | |
awaru | -0.112 |
�� | -0.109 |
ulia | -0.107 |
rogen | -0.105 |
uese | -0.104 |
Token | wife |
Token ID | 3656 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
Instruct | +0.237 |
Anon | +0.220 |
Teg | +0.219 |
Luxem | +0.216 |
theless | +0.215 |
Neg logit contributions | |
awaru | -0.271 |
�� | -0.262 |
ulia | -0.256 |
rogen | -0.249 |
uese | -0.247 |
Token | the |
Token ID | 262 |
Feature activation | -3.82e-02 |
Pos logit contributions | |
Instruct | +0.102 |
Anon | +0.094 |
Teg | +0.093 |
Luxem | +0.093 |
theless | +0.092 |
Neg logit contributions | |
awaru | -0.120 |
�� | -0.116 |
ulia | -0.114 |
vantage | -0.112 |
uese | -0.111 |
Token | polls |
Token ID | 9231 |
Feature activation | +5.63e-02 |
Pos logit contributions | |
awaru | +0.059 |
�� | +0.058 |
ulia | +0.055 |
vantage | +0.054 |
rogen | +0.054 |
Neg logit contributions | |
Instruct | -0.081 |
Anon | -0.077 |
Teg | -0.076 |
Luxem | -0.075 |
theless | -0.075 |
Token | , |
Token ID | 11 |
Feature activation | -8.15e-02 |
Pos logit contributions | |
Instruct | +0.099 |
Anon | +0.093 |
Luxem | +0.092 |
Teg | +0.092 |
behav | +0.090 |
Neg logit contributions | |
awaru | -0.110 |
ulia | -0.104 |
�� | -0.103 |
rogen | -0.102 |
vantage | -0.101 |
Token | will |
Token ID | 481 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
awaru | +0.137 |
ulia | +0.128 |
�� | +0.127 |
vantage | +0.125 |
rogen | +0.124 |
Neg logit contributions | |
Instruct | -0.165 |
Anon | -0.156 |
Luxem | -0.154 |
Teg | -0.153 |
behav | -0.151 |
Token | cons |
Token ID | 762 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
awaru | +0.258 |
�� | +0.246 |
ulia | +0.244 |
rogen | +0.241 |
vantage | +0.235 |
Neg logit contributions | |
Instruct | -0.228 |
Anon | -0.218 |
Teg | -0.212 |
Luxem | -0.211 |
behav | -0.210 |
Token | ign |
Token ID | 570 |
Feature activation | +9.78e-01 |
Pos logit contributions | |
Instruct | +0.412 |
Anon | +0.397 |
Teg | +0.390 |
Luxem | +0.389 |
behav | +0.384 |
Neg logit contributions | |
awaru | -0.338 |
ulia | -0.318 |
�� | -0.314 |
vantage | -0.310 |
uese | -0.307 |
Token | their |
Token ID | 511 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
Instruct | +1.552 |
Anon | +1.428 |
Teg | +1.420 |
Luxem | +1.397 |
theless | +1.372 |
Neg logit contributions | |
awaru | -2.112 |
�� | -2.030 |
ulia | -1.998 |
vantage | -1.942 |
uese | -1.941 |
Token | votes |
Token ID | 5690 |
Feature activation | -6.79e-02 |
Pos logit contributions | |
awaru | +0.433 |
�� | +0.416 |
ulia | +0.410 |
rogen | +0.404 |
uese | +0.398 |
Neg logit contributions | |
Instruct | -0.339 |
Anon | -0.313 |
Teg | -0.309 |
theless | -0.307 |
Luxem | -0.302 |
Token | into |
Token ID | 656 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
awaru | +0.130 |
�� | +0.122 |
ulia | +0.121 |
rogen | +0.119 |
vantage | +0.118 |
Neg logit contributions | |
Instruct | -0.123 |
Anon | -0.115 |
Teg | -0.114 |
Luxem | -0.112 |
theless | -0.111 |
Token | computers |
Token ID | 9061 |
Feature activation | -6.78e-03 |
Pos logit contributions | |
Instruct | +0.041 |
Anon | +0.038 |
Teg | +0.038 |
Luxem | +0.037 |
theless | +0.037 |
Neg logit contributions | |
awaru | -0.047 |
�� | -0.045 |
ulia | -0.044 |
vantage | -0.043 |
rogen | -0.043 |
Token | that |
Token ID | 326 |
Feature activation | +7.46e-02 |
Pos logit contributions | |
awaru | +0.014 |
ulia | +0.013 |
�� | +0.013 |
rogen | +0.013 |
vantage | +0.012 |
Neg logit contributions | |
Instruct | -0.011 |
Anon | -0.011 |
Luxem | -0.011 |
Teg | -0.010 |
theless | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
Instruct | +0.055 |
Anon | +0.051 |
Teg | +0.050 |
Luxem | +0.050 |
theless | +0.049 |
Neg logit contributions | |
awaru | -0.069 |
�� | -0.066 |
ulia | -0.065 |
vantage | -0.064 |
rogen | -0.063 |
Token |
|
Token ID | 198 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
Instruct | +0.034 |
Anon | +0.033 |
Teg | +0.032 |
theless | +0.032 |
behav | +0.032 |
Neg logit contributions | |
awaru | -0.034 |
ulia | -0.033 |
�� | -0.033 |
Combat | -0.032 |
vantage | -0.032 |
Token |
|
Token ID | 198 |
Feature activation | +2.77e-02 |
Pos logit contributions | |
awaru | +0.053 |
ulia | +0.050 |
�� | +0.050 |
Combat | +0.049 |
rogen | +0.048 |
Neg logit contributions | |
Instruct | -0.047 |
Anon | -0.044 |
Teg | -0.043 |
Luxem | -0.043 |
theless | -0.043 |
Token | Facebook |
Token ID | 12025 |
Feature activation | +2.39e-02 |
Pos logit contributions | |
Instruct | +0.055 |
Anon | +0.052 |
Teg | +0.052 |
Luxem | +0.052 |
theless | +0.051 |
Neg logit contributions | |
awaru | -0.048 |
�� | -0.046 |
ulia | -0.045 |
Combat | -0.044 |
vantage | -0.043 |
Token | Twitter |
Token ID | 3009 |
Feature activation | +5.90e-01 |
Pos logit contributions | |
Instruct | +0.025 |
Teg | +0.022 |
Anon | +0.022 |
Luxem | +0.021 |
Tuls | +0.021 |
Neg logit contributions | |
awaru | -0.063 |
�� | -0.061 |
ulia | -0.061 |
vantage | -0.060 |
rogen | -0.060 |
Token | Pinterest |
Token ID | 17334 |
Feature activation | +9.67e-01 |
Pos logit contributions | |
Instruct | +0.451 |
Teg | +0.379 |
Luxem | +0.377 |
Anon | +0.377 |
theless | +0.361 |
Neg logit contributions | |
awaru | -1.735 |
ulia | -1.681 |
�� | -1.676 |
uese | -1.633 |
rogen | -1.630 |
Token | Lake |
Token ID | 6233 |
Feature activation | +3.05e-01 |
Pos logit contributions | |
Instruct | +1.680 |
Anon | +1.570 |
Teg | +1.559 |
Luxem | +1.540 |
theless | +1.506 |
Neg logit contributions | |
awaru | -1.924 |
�� | -1.851 |
ulia | -1.811 |
vantage | -1.758 |
uese | -1.758 |
Token | Ur |
Token ID | 8799 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
Instruct | +0.401 |
Anon | +0.365 |
Teg | +0.360 |
Luxem | +0.358 |
theless | +0.348 |
Neg logit contributions | |
awaru | -0.736 |
�� | -0.715 |
ulia | -0.702 |
vantage | -0.684 |
uese | -0.681 |
Token | mia |
Token ID | 20730 |
Feature activation | +8.19e-02 |
Pos logit contributions | |
Instruct | +0.325 |
Anon | +0.309 |
Teg | +0.307 |
Luxem | +0.305 |
Tuls | +0.300 |
Neg logit contributions | |
awaru | -0.189 |
�� | -0.177 |
ulia | -0.175 |
uese | -0.166 |
vantage | -0.166 |
Token | , |
Token ID | 11 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
Instruct | +0.143 |
Anon | +0.133 |
Teg | +0.132 |
Luxem | +0.131 |
theless | +0.131 |
Neg logit contributions | |
awaru | -0.161 |
�� | -0.155 |
ulia | -0.153 |
rogen | -0.147 |
vantage | -0.147 |
Token | Iran |
Token ID | 4068 |
Feature activation | +5.70e-02 |
Pos logit contributions | |
Instruct | +0.024 |
Anon | +0.022 |
Teg | +0.021 |
Luxem | +0.021 |
theless | +0.021 |
Neg logit contributions | |
awaru | -0.029 |
ulia | -0.027 |
�� | -0.027 |
rogen | -0.026 |
uese | -0.026 |
Token | R |
Token ID | 49 |
Feature activation | +2.10e-02 |
Pos logit contributions | |
Instruct | +0.091 |
Anon | +0.086 |
Teg | +0.086 |
Luxem | +0.085 |
theless | +0.085 |
Neg logit contributions | |
awaru | -0.060 |
�� | -0.057 |
ulia | -0.056 |
Combat | -0.054 |
vantage | -0.054 |
Token | . |
Token ID | 764 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
Instruct | +0.035 |
Anon | +0.033 |
Teg | +0.033 |
Luxem | +0.032 |
theless | +0.032 |
Neg logit contributions | |
awaru | -0.043 |
�� | -0.042 |
ulia | -0.041 |
vantage | -0.039 |
uese | -0.039 |
Token | Des |
Token ID | 5960 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
awaru | +0.337 |
�� | +0.323 |
ulia | +0.316 |
Combat | +0.310 |
vantage | +0.305 |
Neg logit contributions | |
Instruct | -0.404 |
Teg | -0.386 |
Anon | -0.383 |
theless | -0.380 |
Luxem | -0.379 |
Token | ren |
Token ID | 8851 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
Instruct | +0.336 |
Anon | +0.316 |
Teg | +0.313 |
Luxem | +0.311 |
Tuls | +0.304 |
Neg logit contributions | |
awaru | -0.358 |
�� | -0.345 |
ulia | -0.336 |
vantage | -0.326 |
uese | -0.326 |
Token | se |
Token ID | 325 |
Feature activation | +2.94e-01 |
Pos logit contributions | |
awaru | +0.275 |
�� | +0.260 |
ulia | +0.255 |
vantage | +0.245 |
incial | +0.244 |
Neg logit contributions | |
Instruct | -0.384 |
Anon | -0.367 |
Teg | -0.363 |
Luxem | -0.358 |
theless | -0.355 |
Token | ign |
Token ID | 570 |
Feature activation | +9.53e-01 |
Pos logit contributions | |
Instruct | +0.732 |
Anon | +0.701 |
Teg | +0.699 |
Luxem | +0.691 |
Tuls | +0.683 |
Neg logit contributions | |
awaru | -0.360 |
�� | -0.338 |
ulia | -0.331 |
uese | -0.310 |
incial | -0.308 |
Token | ements |
Token ID | 3196 |
Feature activation | +2.63e-01 |
Pos logit contributions | |
Instruct | +2.277 |
Anon | +2.169 |
Teg | +2.157 |
Luxem | +2.138 |
Tuls | +2.104 |
Neg logit contributions | |
awaru | -1.285 |
�� | -1.211 |
ulia | -1.166 |
uese | -1.121 |
vantage | -1.115 |
Token | ré |
Token ID | 40560 |
Feature activation | +2.75e-01 |
Pos logit contributions | |
Instruct | +0.435 |
Teg | +0.406 |
Anon | +0.405 |
Luxem | +0.397 |
Tuls | +0.392 |
Neg logit contributions | |
awaru | -0.541 |
�� | -0.521 |
ulia | -0.506 |
uese | -0.494 |
vantage | -0.493 |
Token | c |
Token ID | 66 |
Feature activation | +5.64e-02 |
Pos logit contributions | |
Instruct | +0.596 |
Anon | +0.566 |
Teg | +0.562 |
Luxem | +0.559 |
Tuls | +0.547 |
Neg logit contributions | |
awaru | -0.433 |
�� | -0.406 |
ulia | -0.399 |
uese | -0.385 |
vantage | -0.385 |
Token | ents |
Token ID | 658 |
Feature activation | +3.03e-01 |
Pos logit contributions | |
Instruct | +0.110 |
Anon | +0.106 |
Teg | +0.104 |
Luxem | +0.103 |
behav | +0.102 |
Neg logit contributions | |
awaru | -0.101 |
ulia | -0.094 |
�� | -0.094 |
incial | -0.092 |
uese | -0.091 |
Token | ind |
Token ID | 773 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
Instruct | +0.505 |
Anon | +0.470 |
Teg | +0.469 |
Luxem | +0.462 |
Tuls | +0.452 |
Neg logit contributions | |
awaru | -0.625 |
�� | -0.602 |
ulia | -0.586 |
uese | -0.571 |
vantage | -0.569 |
Token | serious |
Token ID | 2726 |
Feature activation | -5.64e-03 |
Pos logit contributions | |
awaru | +0.203 |
�� | +0.194 |
ulia | +0.188 |
uese | +0.184 |
rogen | +0.183 |
Neg logit contributions | |
Instruct | -0.191 |
Anon | -0.178 |
Teg | -0.176 |
Luxem | -0.174 |
theless | -0.172 |
Token | fiction |
Token ID | 10165 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
awaru | +0.011 |
�� | +0.010 |
ulia | +0.010 |
Combat | +0.010 |
uese | +0.010 |
Neg logit contributions | |
Instruct | -0.010 |
Anon | -0.009 |
Teg | -0.009 |
Luxem | -0.009 |
theless | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
awaru | +0.463 |
�� | +0.438 |
ulia | +0.434 |
rogen | +0.422 |
vantage | +0.421 |
Neg logit contributions | |
Instruct | -0.382 |
Anon | -0.357 |
Teg | -0.348 |
Luxem | -0.345 |
Tuls | -0.343 |
Token | The |
Token ID | 383 |
Feature activation | +3.31e-02 |
Pos logit contributions | |
awaru | +0.215 |
ulia | +0.208 |
�� | +0.206 |
Combat | +0.204 |
vantage | +0.202 |
Neg logit contributions | |
Instruct | -0.208 |
behav | -0.198 |
theless | -0.197 |
Anon | -0.196 |
Teg | -0.194 |
Token | expected |
Token ID | 2938 |
Feature activation | -2.60e-01 |
Pos logit contributions | |
Instruct | +0.060 |
Anon | +0.057 |
Teg | +0.056 |
Luxem | +0.055 |
Tuls | +0.055 |
Neg logit contributions | |
awaru | -0.062 |
�� | -0.060 |
ulia | -0.058 |
vantage | -0.057 |
uese | -0.057 |
Token | duel |
Token ID | 24662 |
Feature activation | +9.26e-01 |
Pos logit contributions | |
awaru | +0.511 |
�� | +0.485 |
ulia | +0.475 |
rogen | +0.468 |
uese | +0.461 |
Neg logit contributions | |
Instruct | -0.456 |
Anon | -0.427 |
Teg | -0.416 |
Luxem | -0.415 |
Tuls | -0.408 |
Token | never |
Token ID | 1239 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
Instruct | +1.585 |
Anon | +1.476 |
Teg | +1.467 |
Luxem | +1.447 |
Tuls | +1.411 |
Neg logit contributions | |
awaru | -1.876 |
�� | -1.806 |
ulia | -1.761 |
vantage | -1.713 |
uese | -1.711 |
Token | takes |
Token ID | 2753 |
Feature activation | +9.56e-02 |
Pos logit contributions | |
Instruct | +0.334 |
Anon | +0.319 |
Teg | +0.313 |
Luxem | +0.311 |
Tuls | +0.303 |
Neg logit contributions | |
awaru | -0.275 |
�� | -0.257 |
ulia | -0.251 |
rogen | -0.243 |
uese | -0.242 |
Token | place |
Token ID | 1295 |
Feature activation | -9.70e-02 |
Pos logit contributions | |
Instruct | +0.167 |
Anon | +0.157 |
Teg | +0.155 |
Luxem | +0.154 |
theless | +0.152 |
Neg logit contributions | |
awaru | -0.186 |
�� | -0.177 |
ulia | -0.176 |
vantage | -0.171 |
uese | -0.171 |
Token | . |
Token ID | 13 |
Feature activation | -9.24e-02 |
Pos logit contributions | |
awaru | +0.207 |
�� | +0.197 |
ulia | +0.195 |
vantage | +0.190 |
rogen | +0.189 |
Neg logit contributions | |
Instruct | -0.155 |
Anon | -0.144 |
Teg | -0.142 |
Luxem | -0.140 |
Tuls | -0.137 |
Token | The |
Token ID | 383 |
Feature activation | +6.15e-02 |
Pos logit contributions | |
awaru | +0.170 |
ulia | +0.165 |
�� | +0.163 |
Combat | +0.162 |
vantage | +0.160 |
Neg logit contributions | |
Instruct | -0.162 |
behav | -0.154 |
theless | -0.153 |
Anon | -0.152 |
Teg | -0.152 |
Token | of |
Token ID | 286 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
awaru | +0.000 |
�� | +0.000 |
ulia | +0.000 |
uese | +0.000 |
rogen | +0.000 |
Neg logit contributions | |
Instruct | -0.000 |
Anon | -0.000 |
Teg | -0.000 |
Luxem | -0.000 |
theless | -0.000 |
Token | the |
Token ID | 262 |
Feature activation | +4.54e-02 |
Pos logit contributions | |
awaru | +0.037 |
�� | +0.036 |
ulia | +0.035 |
uese | +0.034 |
vantage | +0.034 |
Neg logit contributions | |
Instruct | -0.032 |
Anon | -0.029 |
Teg | -0.029 |
Luxem | -0.029 |
theless | -0.028 |
Token | moral |
Token ID | 6573 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
Instruct | +0.079 |
Anon | +0.074 |
Teg | +0.073 |
Luxem | +0.072 |
Tuls | +0.071 |
Neg logit contributions | |
awaru | -0.090 |
�� | -0.087 |
ulia | -0.085 |
uese | -0.082 |
vantage | -0.082 |
Token | and |
Token ID | 290 |
Feature activation | -9.21e-03 |
Pos logit contributions | |
awaru | +0.287 |
�� | +0.278 |
ulia | +0.270 |
vantage | +0.264 |
uese | +0.264 |
Neg logit contributions | |
Instruct | -0.216 |
Anon | -0.202 |
Teg | -0.198 |
Luxem | -0.195 |
theless | -0.193 |
Token | intellectual |
Token ID | 9028 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
awaru | +0.018 |
�� | +0.018 |
ulia | +0.017 |
rogen | +0.017 |
vantage | +0.017 |
Neg logit contributions | |
Instruct | -0.016 |
Anon | -0.015 |
Teg | -0.015 |
Luxem | -0.015 |
Tuls | -0.014 |
Token | standard |
Token ID | 3210 |
Feature activation | +9.20e-01 |
Pos logit contributions | |
Instruct | +0.248 |
Anon | +0.231 |
Teg | +0.228 |
Luxem | +0.225 |
Tuls | +0.222 |
Neg logit contributions | |
awaru | -0.310 |
�� | -0.298 |
ulia | -0.291 |
uese | -0.284 |
vantage | -0.283 |
Token | the |
Token ID | 262 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
Instruct | +1.645 |
Anon | +1.537 |
Teg | +1.524 |
Luxem | +1.502 |
theless | +1.478 |
Neg logit contributions | |
awaru | -1.797 |
�� | -1.728 |
ulia | -1.681 |
uese | -1.636 |
vantage | -1.632 |
Token | West |
Token ID | 2688 |
Feature activation | +2.01e-01 |
Pos logit contributions | |
Instruct | +0.032 |
Anon | +0.030 |
Teg | +0.030 |
Luxem | +0.030 |
theless | +0.029 |
Neg logit contributions | |
awaru | -0.032 |
�� | -0.030 |
ulia | -0.030 |
uese | -0.029 |
vantage | -0.029 |
Token | identified |
Token ID | 5174 |
Feature activation | -5.28e-02 |
Pos logit contributions | |
Instruct | +0.376 |
Anon | +0.354 |
Teg | +0.348 |
Luxem | +0.341 |
Tuls | +0.341 |
Neg logit contributions | |
awaru | -0.372 |
�� | -0.359 |
ulia | -0.349 |
vantage | -0.341 |
uese | -0.339 |
Token | itself |
Token ID | 2346 |
Feature activation | +3.41e-01 |
Pos logit contributions | |
awaru | +0.111 |
�� | +0.107 |
ulia | +0.104 |
uese | +0.102 |
vantage | +0.102 |
Neg logit contributions | |
Instruct | -0.086 |
Anon | -0.080 |
Teg | -0.079 |
Luxem | -0.078 |
theless | -0.077 |
Token | with |
Token ID | 351 |
Feature activation | -4.22e-02 |
Pos logit contributions | |
Instruct | +0.517 |
Anon | +0.478 |
Teg | +0.475 |
Luxem | +0.469 |
Tuls | +0.454 |
Neg logit contributions | |
awaru | -0.752 |
�� | -0.731 |
ulia | -0.715 |
uese | -0.701 |
vantage | -0.695 |
Token | Kh |
Token ID | 5311 |
Feature activation | -7.22e-02 |
Pos logit contributions | |
Instruct | +0.067 |
Teg | +0.062 |
Luxem | +0.061 |
Anon | +0.061 |
theless | +0.061 |
Neg logit contributions | |
awaru | -0.072 |
�� | -0.069 |
ulia | -0.068 |
uese | -0.066 |
vantage | -0.065 |
Token | ul |
Token ID | 377 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
awaru | +0.115 |
ulia | +0.108 |
�� | +0.107 |
uese | +0.102 |
vantage | +0.102 |
Neg logit contributions | |
Instruct | -0.156 |
Anon | -0.148 |
Teg | -0.146 |
Luxem | -0.146 |
theless | -0.143 |
Token | ood |
Token ID | 702 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
awaru | +0.326 |
ulia | +0.305 |
�� | +0.305 |
uese | +0.289 |
vantage | +0.287 |
Neg logit contributions | |
Instruct | -0.452 |
Anon | -0.427 |
Teg | -0.426 |
Luxem | -0.422 |
theless | -0.419 |
Token | Bad |
Token ID | 7772 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
Instruct | +0.281 |
Anon | +0.261 |
Teg | +0.260 |
Luxem | +0.260 |
theless | +0.256 |
Neg logit contributions | |
awaru | -0.244 |
�� | -0.234 |
ulia | -0.230 |
uese | -0.221 |
vantage | -0.220 |
Token | awi |
Token ID | 23368 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
awaru | +0.343 |
ulia | +0.323 |
�� | +0.320 |
uese | +0.305 |
vantage | +0.305 |
Neg logit contributions | |
Instruct | -0.490 |
Luxem | -0.460 |
Anon | -0.458 |
Teg | -0.457 |
theless | -0.451 |
Token | (@ |
Token ID | 4275 |
Feature activation | +9.01e-01 |
Pos logit contributions | |
awaru | +0.526 |
�� | +0.506 |
ulia | +0.504 |
uese | +0.486 |
incial | +0.484 |
Neg logit contributions | |
Instruct | -0.410 |
Luxem | -0.374 |
Teg | -0.373 |
Anon | -0.372 |
theless | -0.371 |
Token | Kh |
Token ID | 33155 |
Feature activation | -5.57e-02 |
Pos logit contributions | |
Instruct | +2.454 |
Anon | +2.352 |
Teg | +2.337 |
Luxem | +2.316 |
Tuls | +2.286 |
Neg logit contributions | |
awaru | -0.904 |
�� | -0.839 |
ulia | -0.787 |
uese | -0.748 |
vantage | -0.741 |
Token | ul |
Token ID | 377 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
awaru | +0.099 |
�� | +0.094 |
ulia | +0.093 |
uese | +0.090 |
vantage | +0.089 |
Neg logit contributions | |
Instruct | -0.110 |
Anon | -0.103 |
Teg | -0.103 |
Luxem | -0.102 |
Tuls | -0.099 |
Token | ood |
Token ID | 702 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
awaru | +0.373 |
�� | +0.356 |
ulia | +0.350 |
uese | +0.339 |
vantage | +0.339 |
Neg logit contributions | |
Instruct | -0.354 |
Anon | -0.331 |
Teg | -0.331 |
Luxem | -0.327 |
Tuls | -0.319 |
Token | Bad |
Token ID | 22069 |
Feature activation | -2.26e-01 |
Pos logit contributions | |
awaru | +0.191 |
�� | +0.182 |
ulia | +0.177 |
uese | +0.171 |
vantage | +0.171 |
Neg logit contributions | |
Instruct | -0.257 |
Teg | -0.242 |
Anon | -0.241 |
Luxem | -0.239 |
theless | -0.235 |
Token | awi |
Token ID | 23368 |
Feature activation | -2.76e-01 |
Pos logit contributions | |
awaru | +0.266 |
�� | +0.239 |
ulia | +0.238 |
vantage | +0.224 |
uese | +0.222 |
Neg logit contributions | |
Instruct | -0.587 |
Teg | -0.562 |
Anon | -0.557 |
Luxem | -0.556 |
theless | -0.548 |
Token | … |
Token ID | 1399 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
Instruct | +0.032 |
Luxem | +0.030 |
Teg | +0.030 |
Anon | +0.030 |
theless | +0.030 |
Neg logit contributions | |
awaru | -0.037 |
�� | -0.035 |
ulia | -0.034 |
rogen | -0.033 |
uese | -0.033 |
Token | — |
Token ID | 851 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
Instruct | +0.216 |
Anon | +0.200 |
Teg | +0.200 |
Luxem | +0.199 |
theless | +0.195 |
Neg logit contributions | |
awaru | -0.236 |
�� | -0.225 |
ulia | -0.222 |
vantage | -0.214 |
uese | -0.213 |
Token | ACLU |
Token ID | 24381 |
Feature activation | -3.25e-02 |
Pos logit contributions | |
awaru | +0.416 |
�� | +0.399 |
ulia | +0.392 |
vantage | +0.376 |
uese | +0.375 |
Neg logit contributions | |
Instruct | -0.410 |
Teg | -0.381 |
Anon | -0.379 |
Luxem | -0.378 |
theless | -0.373 |
Token | of |
Token ID | 286 |
Feature activation | +7.07e-02 |
Pos logit contributions | |
awaru | +0.060 |
�� | +0.057 |
ulia | +0.056 |
uese | +0.054 |
vantage | +0.053 |
Neg logit contributions | |
Instruct | -0.063 |
Anon | -0.059 |
Teg | -0.058 |
Luxem | -0.058 |
Tuls | -0.057 |
Token | Oregon |
Token ID | 8819 |
Feature activation | +7.68e-02 |
Pos logit contributions | |
Instruct | +0.105 |
Anon | +0.096 |
Teg | +0.095 |
theless | +0.094 |
Luxem | +0.093 |
Neg logit contributions | |
awaru | -0.156 |
�� | -0.151 |
ulia | -0.149 |
vantage | -0.146 |
uese | -0.145 |
Token | (@ |
Token ID | 4275 |
Feature activation | +9.00e-01 |
Pos logit contributions | |
Instruct | +0.055 |
theless | +0.046 |
Teg | +0.045 |
Anon | +0.045 |
Luxem | +0.044 |
Neg logit contributions | |
awaru | -0.228 |
ulia | -0.221 |
�� | -0.221 |
Combat | -0.216 |
uese | -0.216 |
Token | AC |
Token ID | 2246 |
Feature activation | +3.51e-02 |
Pos logit contributions | |
Instruct | +2.122 |
Anon | +2.017 |
Teg | +2.003 |
Luxem | +1.981 |
Tuls | +1.951 |
Neg logit contributions | |
awaru | -1.237 |
�� | -1.179 |
ulia | -1.124 |
uese | -1.081 |
vantage | -1.077 |
Token | LU |
Token ID | 41596 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
Instruct | +0.061 |
Anon | +0.057 |
Teg | +0.057 |
Luxem | +0.056 |
theless | +0.055 |
Neg logit contributions | |
awaru | -0.070 |
�� | -0.067 |
ulia | -0.066 |
vantage | -0.064 |
uese | -0.064 |
Token | _ |
Token ID | 62 |
Feature activation | +9.30e-02 |
Pos logit contributions | |
Instruct | +0.479 |
Anon | +0.447 |
Teg | +0.445 |
Luxem | +0.442 |
theless | +0.436 |
Neg logit contributions | |
awaru | -0.456 |
�� | -0.433 |
ulia | -0.429 |
vantage | -0.415 |
uese | -0.415 |
Token | OR |
Token ID | 1581 |
Feature activation | +7.89e-02 |
Pos logit contributions | |
Instruct | +0.158 |
Teg | +0.144 |
theless | +0.144 |
Luxem | +0.143 |
Anon | +0.143 |
Neg logit contributions | |
awaru | -0.190 |
ulia | -0.180 |
�� | -0.179 |
Combat | -0.177 |
UFC | -0.176 |
Token | ) |
Token ID | 8 |
Feature activation | -7.43e-02 |
Pos logit contributions | |
Instruct | +0.145 |
Anon | +0.135 |
Teg | +0.134 |
Luxem | +0.133 |
theless | +0.132 |
Neg logit contributions | |
awaru | -0.149 |
�� | -0.141 |
ulia | -0.141 |
uese | -0.136 |
vantage | -0.136 |
Token | All |
Token ID | 1439 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
Instruct | +0.335 |
Anon | +0.313 |
Teg | +0.311 |
Luxem | +0.308 |
theless | +0.305 |
Neg logit contributions | |
awaru | -0.304 |
�� | -0.292 |
ulia | -0.287 |
vantage | -0.278 |
uese | -0.278 |
Token | Rights |
Token ID | 6923 |
Feature activation | +3.75e-01 |
Pos logit contributions | |
Instruct | +0.391 |
Anon | +0.363 |
Teg | +0.356 |
Luxem | +0.355 |
Tuls | +0.348 |
Neg logit contributions | |
awaru | -0.516 |
�� | -0.501 |
ulia | -0.491 |
uese | -0.480 |
vantage | -0.480 |
Token | Reserved |
Token ID | 33876 |
Feature activation | -2.45e-01 |
Pos logit contributions | |
Instruct | +0.596 |
Teg | +0.559 |
theless | +0.557 |
Anon | +0.557 |
Tuls | +0.548 |
Neg logit contributions | |
awaru | -0.782 |
�� | -0.743 |
ulia | -0.734 |
uese | -0.717 |
rogen | -0.713 |
Token | . |
Token ID | 13 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
awaru | +0.489 |
�� | +0.469 |
ulia | +0.461 |
vantage | +0.448 |
uese | +0.443 |
Neg logit contributions | |
Instruct | -0.422 |
Teg | -0.393 |
Anon | -0.392 |
Luxem | -0.386 |
theless | -0.383 |
Token | Rep |
Token ID | 1432 |
Feature activation | +2.97e-01 |
Pos logit contributions | |
Instruct | +0.273 |
Anon | +0.256 |
Teg | +0.255 |
Luxem | +0.252 |
Tuls | +0.247 |
Neg logit contributions | |
awaru | -0.250 |
�� | -0.239 |
ulia | -0.233 |
uese | -0.226 |
vantage | -0.226 |
Token | rint |
Token ID | 22272 |
Feature activation | +8.99e-01 |
Pos logit contributions | |
Instruct | +0.438 |
Anon | +0.407 |
Teg | +0.400 |
Luxem | +0.397 |
theless | +0.391 |
Neg logit contributions | |
awaru | -0.669 |
�� | -0.645 |
ulia | -0.635 |
uese | -0.622 |
rogen | -0.619 |
Token | ed |
Token ID | 276 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
Instruct | +1.578 |
Anon | +1.472 |
Teg | +1.459 |
Luxem | +1.440 |
theless | +1.435 |
Neg logit contributions | |
awaru | -1.763 |
�� | -1.684 |
ulia | -1.665 |
uese | -1.629 |
vantage | -1.624 |
Token | with |
Token ID | 351 |
Feature activation | +5.65e-02 |
Pos logit contributions | |
Instruct | +0.243 |
theless | +0.228 |
Anon | +0.225 |
Teg | +0.224 |
Luxem | +0.221 |
Neg logit contributions | |
awaru | -0.269 |
�� | -0.261 |
ulia | -0.261 |
uese | -0.252 |
rogen | -0.251 |
Token | permission |
Token ID | 7170 |
Feature activation | -1.49e-02 |
Pos logit contributions | |
Instruct | +0.106 |
theless | +0.100 |
Anon | +0.099 |
Luxem | +0.099 |
Teg | +0.099 |
Neg logit contributions | |
awaru | -0.103 |
�� | -0.097 |
ulia | -0.096 |
uese | -0.094 |
vantage | -0.094 |
Token | . |
Token ID | 13 |
Feature activation | +5.74e-02 |
Pos logit contributions | |
awaru | +0.030 |
�� | +0.028 |
ulia | +0.028 |
vantage | +0.027 |
uese | +0.027 |
Neg logit contributions | |
Instruct | -0.026 |
Teg | -0.024 |
Anon | -0.024 |
theless | -0.023 |
Luxem | -0.023 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
Instruct | +0.116 |
Anon | +0.108 |
Teg | +0.107 |
Luxem | +0.107 |
theless | +0.106 |
Neg logit contributions | |
awaru | -0.097 |
�� | -0.093 |
ulia | -0.092 |
vantage | -0.088 |
uese | -0.088 |
Token | process |
Token ID | 1429 |
Feature activation | -9.27e-02 |
Pos logit contributions | |
awaru | +0.018 |
�� | +0.018 |
ulia | +0.017 |
rogen | +0.017 |
uese | +0.017 |
Neg logit contributions | |
Instruct | -0.014 |
Anon | -0.013 |
Teg | -0.013 |
Luxem | -0.012 |
theless | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -6.11e-02 |
Pos logit contributions | |
awaru | +0.198 |
�� | +0.191 |
ulia | +0.186 |
uese | +0.181 |
vantage | +0.181 |
Neg logit contributions | |
Instruct | -0.149 |
Anon | -0.138 |
Teg | -0.137 |
Luxem | -0.135 |
Tuls | -0.132 |
Token | She |
Token ID | 1375 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
awaru | +0.112 |
ulia | +0.110 |
�� | +0.106 |
Combat | +0.105 |
vantage | +0.105 |
Neg logit contributions | |
Instruct | -0.108 |
behav | -0.102 |
Teg | -0.101 |
theless | -0.101 |
Anon | -0.100 |
Token | was |
Token ID | 373 |
Feature activation | -9.83e-03 |
Pos logit contributions | |
Instruct | +0.272 |
Anon | +0.254 |
Teg | +0.253 |
Luxem | +0.249 |
Tuls | +0.244 |
Neg logit contributions | |
awaru | -0.282 |
�� | -0.266 |
ulia | -0.263 |
rogen | -0.260 |
uese | -0.253 |
Token | also |
Token ID | 635 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
awaru | +0.020 |
�� | +0.019 |
rogen | +0.018 |
ulia | +0.018 |
uese | +0.018 |
Neg logit contributions | |
Instruct | -0.017 |
Anon | -0.015 |
Teg | -0.015 |
Luxem | -0.015 |
theless | -0.015 |
Token | pregnant |
Token ID | 10423 |
Feature activation | +8.93e-01 |
Pos logit contributions | |
Instruct | +0.282 |
Anon | +0.263 |
Teg | +0.262 |
Luxem | +0.257 |
theless | +0.253 |
Neg logit contributions | |
awaru | -0.340 |
�� | -0.325 |
ulia | -0.314 |
rogen | -0.313 |
uese | -0.305 |
Token | . |
Token ID | 13 |
Feature activation | +2.97e-02 |
Pos logit contributions | |
Instruct | +1.443 |
Anon | +1.334 |
Teg | +1.327 |
Luxem | +1.304 |
Tuls | +1.277 |
Neg logit contributions | |
awaru | -1.895 |
�� | -1.826 |
ulia | -1.789 |
uese | -1.738 |
vantage | -1.737 |
Token |
|
Token ID | 198 |
Feature activation | -4.58e-02 |
Pos logit contributions | |
Instruct | +0.052 |
Teg | +0.049 |
behav | +0.049 |
Anon | +0.049 |
theless | +0.048 |
Neg logit contributions | |
awaru | -0.055 |
ulia | -0.054 |
�� | -0.053 |
vantage | -0.051 |
Combat | -0.051 |
Token |
|
Token ID | 198 |
Feature activation | +2.99e-02 |
Pos logit contributions | |
awaru | +0.088 |
ulia | +0.085 |
�� | +0.083 |
Combat | +0.082 |
rogen | +0.082 |
Neg logit contributions | |
Instruct | -0.081 |
Anon | -0.075 |
Teg | -0.074 |
theless | -0.074 |
Luxem | -0.074 |
Token | return |
Token ID | 7783 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
Instruct | +0.061 |
Teg | +0.057 |
Anon | +0.057 |
Luxem | +0.057 |
Tuls | +0.056 |
Neg logit contributions | |
awaru | -0.051 |
�� | -0.048 |
ulia | -0.048 |
Combat | -0.047 |
vantage | -0.046 |
Token | ing |
Token ID | 278 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
Instruct | +0.188 |
Anon | +0.176 |
Teg | +0.175 |
Luxem | +0.173 |
Tuls | +0.169 |
Neg logit contributions | |
awaru | -0.205 |
�� | -0.196 |
ulia | -0.193 |
uese | -0.187 |
vantage | -0.187 |
Token | to |
Token ID | 284 |
Feature activation | +1.02e-03 |
Pos logit contributions | |
awaru | +0.082 |
�� | +0.078 |
ulia | +0.076 |
uese | +0.074 |
vantage | +0.074 |
Neg logit contributions | |
Instruct | -0.069 |
Anon | -0.064 |
Teg | -0.063 |
Luxem | -0.062 |
theless | -0.062 |
Token | compare |
Token ID | 8996 |
Feature activation | +4.72e-02 |
Pos logit contributions | |
Instruct | +0.001 |
Anon | +0.001 |
theless | +0.001 |
Luxem | +0.001 |
Teg | +0.001 |
Neg logit contributions | |
awaru | -0.002 |
�� | -0.002 |
ulia | -0.002 |
rogen | -0.002 |
uese | -0.002 |
Token | this |
Token ID | 428 |
Feature activation | -7.58e-02 |
Pos logit contributions | |
Instruct | +0.080 |
theless | +0.075 |
Anon | +0.074 |
Luxem | +0.074 |
Teg | +0.074 |
Neg logit contributions | |
awaru | -0.093 |
�� | -0.089 |
ulia | -0.089 |
uese | -0.085 |
rogen | -0.085 |
Token | to |
Token ID | 284 |
Feature activation | -7.85e-02 |
Pos logit contributions | |
awaru | +0.149 |
ulia | +0.141 |
�� | +0.140 |
vantage | +0.137 |
uese | +0.137 |
Neg logit contributions | |
Instruct | -0.127 |
theless | -0.121 |
Anon | -0.119 |
Luxem | -0.119 |
Teg | -0.118 |
Token | the |
Token ID | 262 |
Feature activation | +5.46e-02 |
Pos logit contributions | |
awaru | +0.156 |
�� | +0.151 |
ulia | +0.148 |
rogen | +0.144 |
vantage | +0.143 |
Neg logit contributions | |
Instruct | -0.130 |
theless | -0.123 |
Anon | -0.121 |
Teg | -0.120 |
Luxem | -0.120 |
Token | standard |
Token ID | 3210 |
Feature activation | +8.79e-01 |
Pos logit contributions | |
Instruct | +0.089 |
theless | +0.085 |
Luxem | +0.083 |
Teg | +0.083 |
Anon | +0.082 |
Neg logit contributions | |
awaru | -0.110 |
�� | -0.106 |
ulia | -0.105 |
rogen | -0.101 |
vantage | -0.101 |
Token | library |
Token ID | 5888 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
Instruct | +1.410 |
Anon | +1.301 |
Teg | +1.295 |
theless | +1.294 |
Luxem | +1.294 |
Neg logit contributions | |
awaru | -1.846 |
�� | -1.779 |
ulia | -1.754 |
uese | -1.694 |
rogen | -1.691 |
Token | � |
Token ID | 447 |
Feature activation | +5.79e-02 |
Pos logit contributions | |
awaru | +0.355 |
�� | +0.341 |
ulia | +0.339 |
rogen | +0.330 |
Combat | +0.329 |
Neg logit contributions | |
Instruct | -0.266 |
theless | -0.253 |
Luxem | -0.244 |
Anon | -0.243 |
Teg | -0.242 |
Token | � |
Token ID | 247 |
Feature activation | +7.40e-02 |
Pos logit contributions | |
Instruct | +0.150 |
Anon | +0.143 |
Teg | +0.142 |
Luxem | +0.141 |
theless | +0.139 |
Neg logit contributions | |
awaru | -0.067 |
�� | -0.064 |
ulia | -0.060 |
vantage | -0.056 |
uese | -0.056 |
Token | s |
Token ID | 82 |
Feature activation | +4.23e-02 |
Pos logit contributions | |
Instruct | +0.139 |
Anon | +0.130 |
Luxem | +0.130 |
Teg | +0.129 |
Tuls | +0.128 |
Neg logit contributions | |
awaru | -0.135 |
�� | -0.128 |
ulia | -0.127 |
vantage | -0.124 |
uese | -0.123 |
Token | Atomic |
Token ID | 28976 |
Feature activation | +5.51e-02 |
Pos logit contributions | |
Instruct | +0.068 |
theless | +0.064 |
Luxem | +0.063 |
Teg | +0.063 |
Anon | +0.063 |
Neg logit contributions | |
awaru | -0.087 |
�� | -0.083 |
ulia | -0.083 |
rogen | -0.080 |
Combat | -0.079 |
Token | bank |
Token ID | 3331 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
Instruct | +0.271 |
Anon | +0.250 |
Teg | +0.248 |
Luxem | +0.244 |
Tuls | +0.238 |
Neg logit contributions | |
awaru | -0.410 |
�� | -0.396 |
ulia | -0.387 |
uese | -0.378 |
vantage | -0.377 |
Token | loans |
Token ID | 10021 |
Feature activation | +1.35e-02 |
Pos logit contributions | |
awaru | +0.548 |
�� | +0.520 |
ulia | +0.516 |
rogen | +0.498 |
uese | +0.496 |
Neg logit contributions | |
Instruct | -0.513 |
Teg | -0.480 |
Anon | -0.475 |
Luxem | -0.469 |
Tuls | -0.465 |
Token | and |
Token ID | 290 |
Feature activation | +3.90e-02 |
Pos logit contributions | |
Instruct | +0.022 |
Anon | +0.020 |
Teg | +0.020 |
Luxem | +0.020 |
theless | +0.020 |
Neg logit contributions | |
awaru | -0.028 |
�� | -0.027 |
ulia | -0.027 |
vantage | -0.026 |
incial | -0.026 |
Token | the |
Token ID | 262 |
Feature activation | +2.74e-02 |
Pos logit contributions | |
Instruct | +0.069 |
Anon | +0.064 |
Teg | +0.063 |
Luxem | +0.063 |
theless | +0.062 |
Neg logit contributions | |
awaru | -0.076 |
�� | -0.073 |
ulia | -0.072 |
rogen | -0.070 |
incial | -0.069 |
Token | cac |
Token ID | 40428 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
Instruct | +0.048 |
Anon | +0.045 |
Teg | +0.044 |
theless | +0.044 |
Luxem | +0.044 |
Neg logit contributions | |
awaru | -0.053 |
�� | -0.051 |
ulia | -0.050 |
rogen | -0.049 |
incial | -0.048 |
Token | het |
Token ID | 3202 |
Feature activation | +8.79e-01 |
Pos logit contributions | |
awaru | +0.243 |
ulia | +0.231 |
�� | +0.226 |
uese | +0.220 |
rogen | +0.217 |
Neg logit contributions | |
Instruct | -0.371 |
Anon | -0.350 |
Teg | -0.348 |
Luxem | -0.347 |
Tuls | -0.343 |
Token | of |
Token ID | 286 |
Feature activation | -3.33e-03 |
Pos logit contributions | |
Instruct | +1.690 |
Anon | +1.592 |
Teg | +1.584 |
Luxem | +1.563 |
Tuls | +1.533 |
Neg logit contributions | |
awaru | -1.589 |
�� | -1.527 |
ulia | -1.482 |
vantage | -1.436 |
uese | -1.435 |
Token | having |
Token ID | 1719 |
Feature activation | +5.73e-02 |
Pos logit contributions | |
awaru | +0.006 |
�� | +0.006 |
ulia | +0.006 |
rogen | +0.006 |
uese | +0.006 |
Neg logit contributions | |
Instruct | -0.006 |
Anon | -0.005 |
theless | -0.005 |
Teg | -0.005 |
Luxem | -0.005 |
Token | planes |
Token ID | 13016 |
Feature activation | -9.21e-02 |
Pos logit contributions | |
Instruct | +0.100 |
Anon | +0.093 |
Teg | +0.091 |
theless | +0.091 |
Luxem | +0.090 |
Neg logit contributions | |
awaru | -0.114 |
�� | -0.108 |
ulia | -0.107 |
rogen | -0.104 |
uese | -0.103 |
Token | land |
Token ID | 1956 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
awaru | +0.182 |
�� | +0.177 |
ulia | +0.175 |
rogen | +0.170 |
inate | +0.169 |
Neg logit contributions | |
Instruct | -0.154 |
Anon | -0.145 |
Teg | -0.143 |
Luxem | -0.142 |
theless | -0.141 |
Token | in |
Token ID | 287 |
Feature activation | +1.82e-03 |
Pos logit contributions | |
Instruct | +0.332 |
Anon | +0.311 |
Teg | +0.307 |
Luxem | +0.306 |
theless | +0.301 |
Neg logit contributions | |
awaru | -0.379 |
�� | -0.367 |
ulia | -0.360 |
vantage | -0.353 |
rogen | -0.349 |
Token | Irvine |
Token ID | 37277 |
Feature activation | +3.68e-03 |
Pos logit contributions | |
Instruct | +0.020 |
Anon | +0.017 |
theless | +0.017 |
behav | +0.017 |
Luxem | +0.016 |
Neg logit contributions | |
awaru | -0.057 |
ulia | -0.055 |
�� | -0.055 |
rogen | -0.054 |
vantage | -0.053 |
Token | , |
Token ID | 11 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
Instruct | +0.006 |
theless | +0.005 |
Anon | +0.005 |
Luxem | +0.005 |
Teg | +0.005 |
Neg logit contributions | |
awaru | -0.008 |
ulia | -0.008 |
�� | -0.007 |
rogen | -0.007 |
vantage | -0.007 |
Token | California |
Token ID | 3442 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
Instruct | +0.289 |
Anon | +0.270 |
Teg | +0.269 |
Luxem | +0.266 |
theless | +0.263 |
Neg logit contributions | |
awaru | -0.322 |
�� | -0.310 |
ulia | -0.303 |
uese | -0.295 |
vantage | -0.294 |
Token | . |
Token ID | 13 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
Instruct | +0.249 |
Anon | +0.230 |
Teg | +0.229 |
Luxem | +0.222 |
Tuls | +0.217 |
Neg logit contributions | |
awaru | -0.374 |
�� | -0.363 |
ulia | -0.350 |
uese | -0.344 |
vantage | -0.342 |
Token | Hide |
Token ID | 10415 |
Feature activation | -3.89e-01 |
Pos logit contributions | |
Instruct | +0.392 |
Anon | +0.367 |
theless | +0.363 |
Teg | +0.362 |
behav | +0.361 |
Neg logit contributions | |
awaru | -0.448 |
�� | -0.432 |
ulia | -0.430 |
vantage | -0.417 |
uese | -0.414 |
Token | Caption |
Token ID | 11260 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
awaru | +0.468 |
ulia | +0.441 |
�� | +0.433 |
incial | +0.414 |
vantage | +0.413 |
Neg logit contributions | |
Instruct | -0.953 |
theless | -0.915 |
Anon | -0.915 |
Luxem | -0.908 |
Teg | -0.908 |
Token | 10 |
Token ID | 838 |
Feature activation | -1.94e-01 |
Pos logit contributions | |
ulia | +2.435 |
awaru | +2.415 |
Combat | +2.324 |
rogen | +2.300 |
�� | +2.291 |
Neg logit contributions | |
Instruct | -3.447 |
behav | -3.377 |
theless | -3.342 |
Anon | -3.307 |
Luxem | -3.247 |
Token | of |
Token ID | 286 |
Feature activation | +4.68e-02 |
Pos logit contributions | |
awaru | +0.309 |
ulia | +0.290 |
�� | +0.290 |
uese | +0.279 |
rogen | +0.278 |
Neg logit contributions | |
Instruct | -0.410 |
Anon | -0.388 |
Teg | -0.384 |
Luxem | -0.381 |
theless | -0.381 |
Token | 15 |
Token ID | 1315 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
Instruct | +0.095 |
Anon | +0.090 |
Teg | +0.088 |
Luxem | +0.088 |
theless | +0.088 |
Neg logit contributions | |
awaru | -0.079 |
�� | -0.075 |
ulia | -0.074 |
vantage | -0.071 |
uese | -0.071 |
Token | Photos |
Token ID | 9434 |
Feature activation | -9.23e-01 |
Pos logit contributions | |
Instruct | +0.253 |
Anon | +0.239 |
Teg | +0.238 |
Luxem | +0.235 |
theless | +0.231 |
Neg logit contributions | |
awaru | -0.185 |
�� | -0.176 |
ulia | -0.171 |
uese | -0.165 |
vantage | -0.165 |
Token | : |
Token ID | 25 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
awaru | +1.831 |
ulia | +1.809 |
Combat | +1.772 |
rogen | +1.761 |
incial | +1.742 |
Neg logit contributions | |
Instruct | -1.489 |
theless | -1.434 |
behav | -1.426 |
Anon | -1.385 |
princ | -1.372 |
Token | Pasadena |
Token ID | 41010 |
Feature activation | -9.83e-03 |
Pos logit contributions | |
awaru | +0.008 |
ulia | +0.008 |
�� | +0.008 |
rogen | +0.007 |
vantage | +0.007 |
Neg logit contributions | |
Instruct | -0.006 |
theless | -0.005 |
Anon | -0.005 |
behav | -0.005 |
toile | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
awaru | +0.023 |
�� | +0.022 |
ulia | +0.022 |
vantage | +0.021 |
uese | +0.021 |
Neg logit contributions | |
Instruct | -0.014 |
Anon | -0.012 |
Teg | -0.012 |
Luxem | -0.012 |
Tuls | -0.012 |
Token | California |
Token ID | 3442 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
Instruct | +0.242 |
Anon | +0.227 |
Teg | +0.224 |
Luxem | +0.222 |
theless | +0.221 |
Neg logit contributions | |
awaru | -0.259 |
�� | -0.248 |
ulia | -0.242 |
uese | -0.237 |
rogen | -0.236 |
Token | . |
Token ID | 13 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
Instruct | +0.204 |
Anon | +0.189 |
Teg | +0.188 |
Luxem | +0.184 |
Tuls | +0.178 |
Neg logit contributions | |
awaru | -0.308 |
�� | -0.300 |
ulia | -0.289 |
uese | -0.283 |
vantage | -0.283 |
Token | Hide |
Token ID | 10415 |
Feature activation | -3.80e-01 |
Pos logit contributions | |
Instruct | +0.305 |
Anon | +0.285 |
Teg | +0.282 |
theless | +0.281 |
behav | +0.279 |
Neg logit contributions | |
awaru | -0.351 |
�� | -0.338 |
ulia | -0.336 |
vantage | -0.326 |
uese | -0.324 |
Token | Caption |
Token ID | 11260 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
awaru | +0.459 |
ulia | +0.432 |
�� | +0.425 |
incial | +0.405 |
vantage | +0.405 |
Neg logit contributions | |
Instruct | -0.932 |
Anon | -0.894 |
theless | -0.892 |
Teg | -0.888 |
Luxem | -0.887 |
Token | 11 |
Token ID | 1367 |
Feature activation | -6.78e-02 |
Pos logit contributions | |
awaru | +2.539 |
ulia | +2.526 |
rogen | +2.508 |
Combat | +2.464 |
uese | +2.420 |
Neg logit contributions | |
theless | -3.244 |
behav | -3.238 |
Instruct | -3.178 |
Luxem | -3.122 |
princ | -3.106 |
Token | of |
Token ID | 286 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
awaru | +0.109 |
�� | +0.103 |
ulia | +0.103 |
uese | +0.099 |
rogen | +0.098 |
Neg logit contributions | |
Instruct | -0.142 |
Anon | -0.134 |
Teg | -0.133 |
Luxem | -0.132 |
theless | -0.132 |
Token | 15 |
Token ID | 1315 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
Instruct | +0.034 |
Anon | +0.032 |
Teg | +0.031 |
Luxem | +0.031 |
theless | +0.031 |
Neg logit contributions | |
awaru | -0.028 |
�� | -0.026 |
ulia | -0.026 |
vantage | -0.025 |
rogen | -0.025 |
Token | Photos |
Token ID | 9434 |
Feature activation | -9.51e-01 |
Pos logit contributions | |
Instruct | +0.058 |
Anon | +0.055 |
Teg | +0.055 |
Luxem | +0.054 |
theless | +0.054 |
Neg logit contributions | |
awaru | -0.042 |
�� | -0.040 |
ulia | -0.039 |
uese | -0.037 |
vantage | -0.037 |
Token | : |
Token ID | 25 |
Feature activation | +9.63e-02 |
Pos logit contributions | |
awaru | +1.874 |
ulia | +1.854 |
Combat | +1.816 |
rogen | +1.806 |
incial | +1.785 |
Neg logit contributions | |
Instruct | -1.542 |
theless | -1.490 |
behav | -1.481 |
Anon | -1.433 |
princ | -1.427 |
Token | 20 |
Token ID | 1160 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
awaru | +0.030 |
ulia | +0.029 |
�� | +0.028 |
Combat | +0.028 |
rogen | +0.028 |
Neg logit contributions | |
Instruct | -0.034 |
Anon | -0.032 |
behav | -0.032 |
Teg | -0.032 |
theless | -0.032 |
Token | , |
Token ID | 11 |
Feature activation | +2.05e-03 |
Pos logit contributions | |
awaru | +0.022 |
ulia | +0.021 |
�� | +0.021 |
uese | +0.021 |
vantage | +0.020 |
Neg logit contributions | |
Instruct | -0.015 |
Anon | -0.014 |
theless | -0.014 |
Teg | -0.014 |
Luxem | -0.014 |
Token | 2015 |
Token ID | 1853 |
Feature activation | +8.84e-02 |
Pos logit contributions | |
Instruct | +0.004 |
Anon | +0.003 |
Teg | +0.003 |
Luxem | +0.003 |
theless | +0.003 |
Neg logit contributions | |
awaru | -0.004 |
�� | -0.004 |
ulia | -0.004 |
vantage | -0.004 |
uese | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +4.30e-02 |
Pos logit contributions | |
Instruct | +0.132 |
Anon | +0.122 |
Teg | +0.121 |
Luxem | +0.119 |
Tuls | +0.116 |
Neg logit contributions | |
awaru | -0.200 |
�� | -0.194 |
ulia | -0.187 |
uese | -0.184 |
vantage | -0.183 |
Token | Hide |
Token ID | 10415 |
Feature activation | -4.19e-01 |
Pos logit contributions | |
Instruct | +0.075 |
Anon | +0.070 |
theless | +0.070 |
Teg | +0.070 |
behav | +0.069 |
Neg logit contributions | |
awaru | -0.082 |
ulia | -0.079 |
�� | -0.079 |
vantage | -0.077 |
uese | -0.076 |
Token | Caption |
Token ID | 11260 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
awaru | +0.504 |
ulia | +0.474 |
�� | +0.466 |
incial | +0.445 |
vantage | +0.445 |
Neg logit contributions | |
Instruct | -1.027 |
Anon | -0.986 |
theless | -0.983 |
Teg | -0.979 |
Luxem | -0.978 |
Token | 8 |
Token ID | 807 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
ulia | +2.858 |
awaru | +2.809 |
Combat | +2.677 |
incial | +2.665 |
uese | +2.644 |
Neg logit contributions | |
theless | -3.007 |
Instruct | -2.993 |
behav | -2.991 |
princ | -2.915 |
warr | -2.914 |
Token | of |
Token ID | 286 |
Feature activation | +3.74e-02 |
Pos logit contributions | |
awaru | +0.219 |
�� | +0.207 |
ulia | +0.205 |
uese | +0.198 |
vantage | +0.197 |
Neg logit contributions | |
Instruct | -0.284 |
Anon | -0.269 |
Teg | -0.267 |
Luxem | -0.264 |
theless | -0.263 |
Token | 15 |
Token ID | 1315 |
Feature activation | +8.09e-02 |
Pos logit contributions | |
Instruct | +0.076 |
Anon | +0.072 |
Teg | +0.071 |
Luxem | +0.070 |
theless | +0.070 |
Neg logit contributions | |
awaru | -0.063 |
�� | -0.059 |
ulia | -0.059 |
vantage | -0.057 |
uese | -0.057 |
Token | Photos |
Token ID | 9434 |
Feature activation | -8.81e-01 |
Pos logit contributions | |
Instruct | +0.174 |
Anon | +0.165 |
Teg | +0.164 |
Luxem | +0.162 |
theless | +0.160 |
Neg logit contributions | |
awaru | -0.128 |
�� | -0.121 |
ulia | -0.118 |
uese | -0.114 |
vantage | -0.113 |
Token | : |
Token ID | 25 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
awaru | +1.756 |
ulia | +1.732 |
Combat | +1.697 |
rogen | +1.685 |
incial | +1.668 |
Neg logit contributions | |
Instruct | -1.418 |
theless | -1.363 |
behav | -1.355 |
Anon | -1.318 |
Luxem | -1.304 |
Token | aga |
Token ID | 8126 |
Feature activation | -5.61e-02 |
Pos logit contributions | |
awaru | +0.693 |
ulia | +0.648 |
�� | +0.640 |
uese | +0.631 |
vantage | +0.621 |
Neg logit contributions | |
Instruct | -0.785 |
Anon | -0.735 |
theless | -0.733 |
behav | -0.725 |
Teg | -0.725 |
Token | , |
Token ID | 11 |
Feature activation | +7.59e-02 |
Pos logit contributions | |
awaru | +0.125 |
�� | +0.119 |
ulia | +0.119 |
uese | +0.115 |
vantage | +0.115 |
Neg logit contributions | |
Instruct | -0.083 |
Anon | -0.078 |
theless | -0.077 |
Luxem | -0.076 |
Teg | -0.076 |
Token | California |
Token ID | 3442 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
Instruct | +0.138 |
Anon | +0.130 |
theless | +0.129 |
behav | +0.129 |
Teg | +0.126 |
Neg logit contributions | |
awaru | -0.141 |
�� | -0.134 |
ulia | -0.132 |
rogen | -0.130 |
uese | -0.128 |
Token | . |
Token ID | 13 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
Instruct | +0.212 |
Anon | +0.196 |
Teg | +0.195 |
Luxem | +0.191 |
Tuls | +0.185 |
Neg logit contributions | |
awaru | -0.325 |
�� | -0.315 |
ulia | -0.305 |
vantage | -0.298 |
uese | -0.298 |
Token | Hide |
Token ID | 10415 |
Feature activation | -4.12e-01 |
Pos logit contributions | |
Instruct | +0.313 |
Anon | +0.294 |
Teg | +0.290 |
theless | +0.290 |
behav | +0.288 |
Neg logit contributions | |
awaru | -0.354 |
�� | -0.340 |
ulia | -0.338 |
vantage | -0.329 |
uese | -0.326 |
Token | Caption |
Token ID | 11260 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
awaru | +0.494 |
ulia | +0.466 |
�� | +0.456 |
incial | +0.436 |
vantage | +0.436 |
Neg logit contributions | |
Instruct | -1.011 |
theless | -0.972 |
Anon | -0.971 |
Luxem | -0.963 |
Teg | -0.963 |
Token | 13 |
Token ID | 1511 |
Feature activation | -4.65e-02 |
Pos logit contributions | |
awaru | +2.648 |
ulia | +2.547 |
Combat | +2.544 |
rogen | +2.525 |
inate | +2.466 |
Neg logit contributions | |
theless | -3.183 |
Instruct | -3.153 |
behav | -3.126 |
nodd | -3.091 |
princ | -3.078 |
Token | of |
Token ID | 286 |
Feature activation | -8.64e-03 |
Pos logit contributions | |
awaru | +0.076 |
�� | +0.071 |
ulia | +0.071 |
uese | +0.068 |
vantage | +0.068 |
Neg logit contributions | |
Instruct | -0.097 |
Anon | -0.092 |
Teg | -0.091 |
Luxem | -0.090 |
theless | -0.090 |
Token | 15 |
Token ID | 1315 |
Feature activation | +7.46e-03 |
Pos logit contributions | |
awaru | +0.015 |
�� | +0.014 |
ulia | +0.014 |
vantage | +0.013 |
rogen | +0.013 |
Neg logit contributions | |
Instruct | -0.018 |
Anon | -0.017 |
Teg | -0.016 |
Luxem | -0.016 |
theless | -0.016 |
Token | Photos |
Token ID | 9434 |
Feature activation | -9.37e-01 |
Pos logit contributions | |
Instruct | +0.016 |
Anon | +0.015 |
Teg | +0.015 |
Luxem | +0.015 |
theless | +0.015 |
Neg logit contributions | |
awaru | -0.012 |
�� | -0.011 |
ulia | -0.011 |
uese | -0.010 |
vantage | -0.010 |
Token | : |
Token ID | 25 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
awaru | +1.849 |
ulia | +1.830 |
Combat | +1.793 |
rogen | +1.783 |
incial | +1.761 |
Neg logit contributions | |
Instruct | -1.515 |
theless | -1.467 |
behav | -1.456 |
Anon | -1.409 |
princ | -1.402 |
Token | Hills |
Token ID | 14379 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
Instruct | +0.174 |
Anon | +0.161 |
theless | +0.158 |
Luxem | +0.157 |
Teg | +0.157 |
Neg logit contributions | |
awaru | -0.261 |
�� | -0.250 |
ulia | -0.249 |
uese | -0.242 |
vantage | -0.240 |
Token | , |
Token ID | 11 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
Instruct | +0.323 |
Anon | +0.299 |
Teg | +0.294 |
Luxem | +0.291 |
theless | +0.288 |
Neg logit contributions | |
awaru | -0.508 |
�� | -0.489 |
ulia | -0.482 |
uese | -0.468 |
vantage | -0.468 |
Token | California |
Token ID | 3442 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
Instruct | +0.198 |
Anon | +0.185 |
theless | +0.183 |
behav | +0.181 |
Teg | +0.181 |
Neg logit contributions | |
awaru | -0.199 |
�� | -0.189 |
ulia | -0.187 |
rogen | -0.182 |
uese | -0.181 |
Token | . |
Token ID | 13 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
Instruct | +0.250 |
Anon | +0.231 |
Teg | +0.230 |
Luxem | +0.225 |
Tuls | +0.218 |
Neg logit contributions | |
awaru | -0.382 |
�� | -0.372 |
ulia | -0.358 |
uese | -0.352 |
vantage | -0.351 |
Token | Hide |
Token ID | 10415 |
Feature activation | -4.61e-01 |
Pos logit contributions | |
Instruct | +0.298 |
Anon | +0.279 |
Teg | +0.276 |
theless | +0.275 |
behav | +0.273 |
Neg logit contributions | |
awaru | -0.343 |
�� | -0.330 |
ulia | -0.329 |
vantage | -0.319 |
uese | -0.317 |
Token | Caption |
Token ID | 11260 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
awaru | +0.550 |
ulia | +0.520 |
�� | +0.508 |
incial | +0.486 |
vantage | +0.485 |
Neg logit contributions | |
Instruct | -1.132 |
theless | -1.091 |
Anon | -1.088 |
Teg | -1.079 |
Luxem | -1.079 |
Token | 14 |
Token ID | 1478 |
Feature activation | -8.65e-02 |
Pos logit contributions | |
ulia | +2.733 |
awaru | +2.725 |
uese | +2.575 |
Combat | +2.560 |
incial | +2.551 |
Neg logit contributions | |
theless | -2.983 |
Instruct | -2.962 |
behav | -2.922 |
princ | -2.869 |
warr | -2.869 |
Token | of |
Token ID | 286 |
Feature activation | +3.28e-03 |
Pos logit contributions | |
awaru | +0.141 |
�� | +0.132 |
ulia | +0.131 |
uese | +0.127 |
vantage | +0.126 |
Neg logit contributions | |
Instruct | -0.181 |
Anon | -0.171 |
Teg | -0.170 |
Luxem | -0.168 |
theless | -0.167 |
Token | 15 |
Token ID | 1315 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
Instruct | +0.007 |
Anon | +0.006 |
Teg | +0.006 |
Luxem | +0.006 |
theless | +0.006 |
Neg logit contributions | |
awaru | -0.006 |
�� | -0.005 |
ulia | -0.005 |
vantage | -0.005 |
uese | -0.005 |
Token | Photos |
Token ID | 9434 |
Feature activation | -9.30e-01 |
Pos logit contributions | |
awaru | +0.034 |
�� | +0.032 |
ulia | +0.032 |
uese | +0.030 |
vantage | +0.030 |
Neg logit contributions | |
Instruct | -0.048 |
Anon | -0.045 |
Teg | -0.045 |
Luxem | -0.045 |
theless | -0.044 |
Token | : |
Token ID | 25 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
awaru | +1.839 |
ulia | +1.819 |
Combat | +1.783 |
rogen | +1.773 |
incial | +1.751 |
Neg logit contributions | |
Instruct | -1.500 |
theless | -1.452 |
behav | -1.439 |
Anon | -1.395 |
princ | -1.387 |
Token | , |
Token ID | 11 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
Instruct | +0.642 |
Anon | +0.587 |
Teg | +0.584 |
Luxem | +0.573 |
Tuls | +0.558 |
Neg logit contributions | |
awaru | -1.079 |
�� | -1.044 |
ulia | -1.020 |
uese | -0.997 |
vantage | -0.997 |
Token | New |
Token ID | 968 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
Instruct | +0.213 |
Anon | +0.202 |
Teg | +0.199 |
Luxem | +0.197 |
behav | +0.197 |
Neg logit contributions | |
awaru | -0.170 |
�� | -0.162 |
ulia | -0.158 |
uese | -0.154 |
rogen | -0.154 |
Token | Hampshire |
Token ID | 13910 |
Feature activation | -9.26e-02 |
Pos logit contributions | |
awaru | +0.328 |
ulia | +0.314 |
�� | +0.313 |
Combat | +0.309 |
incial | +0.306 |
Neg logit contributions | |
Instruct | -0.253 |
behav | -0.237 |
theless | -0.236 |
Anon | -0.234 |
Luxem | -0.232 |
Token | . |
Token ID | 13 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
awaru | +0.209 |
�� | +0.202 |
ulia | +0.196 |
uese | +0.192 |
vantage | +0.192 |
Neg logit contributions | |
Instruct | -0.140 |
Anon | -0.129 |
Teg | -0.129 |
Luxem | -0.126 |
Tuls | -0.123 |
Token | Hide |
Token ID | 10415 |
Feature activation | -4.97e-01 |
Pos logit contributions | |
Instruct | +0.343 |
Anon | +0.322 |
Teg | +0.319 |
theless | +0.317 |
Luxem | +0.315 |
Neg logit contributions | |
awaru | -0.391 |
�� | -0.376 |
ulia | -0.374 |
vantage | -0.362 |
uese | -0.360 |
Token | Caption |
Token ID | 11260 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
awaru | +0.592 |
ulia | +0.559 |
�� | +0.545 |
incial | +0.524 |
uese | +0.522 |
Neg logit contributions | |
Instruct | -1.225 |
Anon | -1.177 |
theless | -1.175 |
Luxem | -1.167 |
Teg | -1.167 |
Token | 6 |
Token ID | 718 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
awaru | +2.269 |
ulia | +2.264 |
Combat | +2.122 |
incial | +2.115 |
uese | +2.113 |
Neg logit contributions | |
Instruct | -3.457 |
behav | -3.393 |
theless | -3.389 |
Teg | -3.297 |
Luxem | -3.293 |
Token | of |
Token ID | 286 |
Feature activation | +2.19e-02 |
Pos logit contributions | |
awaru | +0.321 |
�� | +0.301 |
ulia | +0.301 |
uese | +0.289 |
rogen | +0.288 |
Neg logit contributions | |
Instruct | -0.424 |
Anon | -0.401 |
Teg | -0.397 |
Luxem | -0.394 |
theless | -0.393 |
Token | 15 |
Token ID | 1315 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
Instruct | +0.044 |
Anon | +0.042 |
Teg | +0.042 |
Luxem | +0.041 |
theless | +0.041 |
Neg logit contributions | |
awaru | -0.037 |
�� | -0.035 |
ulia | -0.034 |
vantage | -0.033 |
uese | -0.033 |
Token | Photos |
Token ID | 9434 |
Feature activation | -9.40e-01 |
Pos logit contributions | |
Instruct | +0.073 |
Anon | +0.069 |
Teg | +0.068 |
Luxem | +0.068 |
Tuls | +0.067 |
Neg logit contributions | |
awaru | -0.053 |
�� | -0.051 |
ulia | -0.049 |
uese | -0.048 |
vantage | -0.048 |
Token | : |
Token ID | 25 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
awaru | +1.858 |
ulia | +1.833 |
Combat | +1.799 |
rogen | +1.783 |
incial | +1.768 |
Neg logit contributions | |
Instruct | -1.519 |
theless | -1.468 |
behav | -1.460 |
Anon | -1.413 |
princ | -1.404 |
Token | in |
Token ID | 259 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
Instruct | +1.297 |
Anon | +1.229 |
Teg | +1.225 |
Luxem | +1.211 |
Tuls | +1.191 |
Neg logit contributions | |
awaru | -0.796 |
�� | -0.759 |
ulia | -0.721 |
uese | -0.699 |
vantage | -0.696 |
Token | , |
Token ID | 11 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
Instruct | +0.256 |
Anon | +0.237 |
Teg | +0.234 |
Luxem | +0.231 |
theless | +0.229 |
Neg logit contributions | |
awaru | -0.382 |
�� | -0.369 |
ulia | -0.364 |
uese | -0.353 |
vantage | -0.353 |
Token | Iowa |
Token ID | 9406 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
Instruct | +0.213 |
Teg | +0.195 |
Anon | +0.195 |
theless | +0.193 |
Luxem | +0.192 |
Neg logit contributions | |
awaru | -0.300 |
�� | -0.291 |
ulia | -0.285 |
uese | -0.277 |
vantage | -0.277 |
Token | . |
Token ID | 13 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
awaru | +0.286 |
�� | +0.275 |
ulia | +0.269 |
uese | +0.262 |
vantage | +0.262 |
Neg logit contributions | |
Instruct | -0.218 |
Anon | -0.202 |
Teg | -0.200 |
Luxem | -0.198 |
theless | -0.193 |
Token | Hide |
Token ID | 10415 |
Feature activation | -4.22e-01 |
Pos logit contributions | |
Instruct | +0.440 |
Anon | +0.413 |
Teg | +0.409 |
theless | +0.408 |
Luxem | +0.405 |
Neg logit contributions | |
awaru | -0.498 |
�� | -0.480 |
ulia | -0.476 |
vantage | -0.461 |
uese | -0.459 |
Token | Caption |
Token ID | 11260 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
awaru | +0.505 |
ulia | +0.476 |
�� | +0.465 |
incial | +0.449 |
uese | +0.447 |
Neg logit contributions | |
Instruct | -1.039 |
Anon | -0.997 |
theless | -0.995 |
Luxem | -0.991 |
Teg | -0.989 |
Token | 5 |
Token ID | 642 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
awaru | +2.386 |
ulia | +2.353 |
Combat | +2.309 |
�� | +2.289 |
rogen | +2.249 |
Neg logit contributions | |
Instruct | -3.333 |
behav | -3.287 |
theless | -3.268 |
Luxem | -3.167 |
Anon | -3.153 |
Token | of |
Token ID | 286 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
awaru | +0.405 |
�� | +0.381 |
ulia | +0.379 |
uese | +0.365 |
rogen | +0.363 |
Neg logit contributions | |
Instruct | -0.531 |
Anon | -0.502 |
Teg | -0.497 |
Luxem | -0.494 |
theless | -0.491 |
Token | 15 |
Token ID | 1315 |
Feature activation | +4.97e-02 |
Pos logit contributions | |
Instruct | +0.040 |
Anon | +0.038 |
Teg | +0.037 |
Luxem | +0.037 |
theless | +0.037 |
Neg logit contributions | |
awaru | -0.034 |
�� | -0.032 |
ulia | -0.031 |
vantage | -0.030 |
uese | -0.030 |
Token | Photos |
Token ID | 9434 |
Feature activation | -9.73e-01 |
Pos logit contributions | |
Instruct | +0.107 |
Anon | +0.101 |
Teg | +0.101 |
Luxem | +0.100 |
Tuls | +0.098 |
Neg logit contributions | |
awaru | -0.079 |
�� | -0.075 |
ulia | -0.073 |
uese | -0.070 |
vantage | -0.070 |
Token | : |
Token ID | 25 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
awaru | +1.941 |
ulia | +1.911 |
Combat | +1.878 |
rogen | +1.856 |
incial | +1.849 |
Neg logit contributions | |
Instruct | -1.556 |
theless | -1.510 |
behav | -1.500 |
Anon | -1.450 |
Luxem | -1.445 |
Token | Pasadena |
Token ID | 41010 |
Feature activation | +6.79e-02 |
Pos logit contributions | |
Instruct | +0.053 |
Anon | +0.049 |
theless | +0.048 |
behav | +0.048 |
Teg | +0.047 |
Neg logit contributions | |
awaru | -0.068 |
ulia | -0.065 |
�� | -0.065 |
rogen | -0.063 |
uese | -0.063 |
Token | , |
Token ID | 11 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
Instruct | +0.096 |
Anon | +0.088 |
Teg | +0.088 |
Luxem | +0.086 |
Tuls | +0.083 |
Neg logit contributions | |
awaru | -0.158 |
�� | -0.153 |
ulia | -0.149 |
vantage | -0.146 |
uese | -0.146 |
Token | California |
Token ID | 3442 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
Instruct | +0.198 |
Anon | +0.185 |
Teg | +0.183 |
Luxem | +0.181 |
Tuls | +0.177 |
Neg logit contributions | |
awaru | -0.233 |
�� | -0.224 |
ulia | -0.218 |
uese | -0.212 |
vantage | -0.212 |
Token | . |
Token ID | 13 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
Instruct | +0.335 |
Anon | +0.311 |
Teg | +0.309 |
Luxem | +0.303 |
Tuls | +0.294 |
Neg logit contributions | |
awaru | -0.502 |
�� | -0.487 |
ulia | -0.470 |
uese | -0.461 |
vantage | -0.460 |
Token | Hide |
Token ID | 10415 |
Feature activation | -3.40e-01 |
Pos logit contributions | |
Instruct | +0.360 |
Anon | +0.336 |
Teg | +0.333 |
theless | +0.331 |
Luxem | +0.328 |
Neg logit contributions | |
awaru | -0.411 |
�� | -0.395 |
ulia | -0.393 |
vantage | -0.381 |
uese | -0.379 |
Token | Caption |
Token ID | 11260 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
awaru | +0.411 |
ulia | +0.387 |
�� | +0.381 |
vantage | +0.363 |
incial | +0.363 |
Neg logit contributions | |
Instruct | -0.831 |
Anon | -0.797 |
theless | -0.796 |
Teg | -0.792 |
Luxem | -0.791 |
Token | 12 |
Token ID | 1105 |
Feature activation | -7.47e-02 |
Pos logit contributions | |
ulia | +2.306 |
awaru | +2.293 |
Combat | +2.224 |
vantage | +2.170 |
rogen | +2.157 |
Neg logit contributions | |
theless | -3.325 |
Instruct | -3.288 |
behav | -3.259 |
nodd | -3.229 |
princ | -3.176 |
Token | of |
Token ID | 286 |
Feature activation | +5.43e-02 |
Pos logit contributions | |
awaru | +0.120 |
�� | +0.113 |
ulia | +0.113 |
uese | +0.109 |
vantage | +0.108 |
Neg logit contributions | |
Instruct | -0.157 |
Anon | -0.148 |
Teg | -0.147 |
Luxem | -0.146 |
theless | -0.145 |
Token | 15 |
Token ID | 1315 |
Feature activation | +4.42e-02 |
Pos logit contributions | |
Instruct | +0.110 |
Anon | +0.104 |
Teg | +0.102 |
Luxem | +0.102 |
theless | +0.101 |
Neg logit contributions | |
awaru | -0.092 |
�� | -0.087 |
ulia | -0.086 |
vantage | -0.083 |
uese | -0.083 |
Token | Photos |
Token ID | 9434 |
Feature activation | -9.12e-01 |
Pos logit contributions | |
Instruct | +0.096 |
Anon | +0.091 |
Teg | +0.090 |
Luxem | +0.089 |
theless | +0.088 |
Neg logit contributions | |
awaru | -0.069 |
�� | -0.066 |
ulia | -0.064 |
uese | -0.061 |
vantage | -0.061 |
Token | : |
Token ID | 25 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
awaru | +1.803 |
ulia | +1.784 |
Combat | +1.747 |
rogen | +1.737 |
incial | +1.716 |
Neg logit contributions | |
Instruct | -1.474 |
theless | -1.423 |
behav | -1.414 |
Anon | -1.370 |
princ | -1.363 |
Token | Irvine |
Token ID | 37277 |
Feature activation | +4.66e-02 |
Pos logit contributions | |
awaru | +0.022 |
ulia | +0.022 |
�� | +0.022 |
rogen | +0.021 |
Combat | +0.021 |
Neg logit contributions | |
Instruct | -0.011 |
Anon | -0.010 |
theless | -0.010 |
behav | -0.010 |
toile | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
Instruct | +0.069 |
Anon | +0.064 |
Teg | +0.063 |
theless | +0.063 |
Luxem | +0.063 |
Neg logit contributions | |
awaru | -0.104 |
�� | -0.100 |
ulia | -0.099 |
vantage | -0.096 |
uese | -0.096 |
Token | California |
Token ID | 3442 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
Instruct | +0.220 |
theless | +0.214 |
behav | +0.210 |
Anon | +0.208 |
Luxem | +0.206 |
Neg logit contributions | |
awaru | -0.199 |
ulia | -0.189 |
�� | -0.189 |
rogen | -0.186 |
Combat | -0.182 |
Token | . |
Token ID | 13 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
Instruct | +0.233 |
Anon | +0.215 |
Teg | +0.214 |
Luxem | +0.208 |
Tuls | +0.203 |
Neg logit contributions | |
awaru | -0.353 |
�� | -0.342 |
ulia | -0.331 |
uese | -0.325 |
vantage | -0.323 |
Token | Hide |
Token ID | 10415 |
Feature activation | -3.22e-01 |
Pos logit contributions | |
Instruct | +0.367 |
theless | +0.345 |
Anon | +0.345 |
behav | +0.344 |
Teg | +0.340 |
Neg logit contributions | |
awaru | -0.420 |
ulia | -0.409 |
�� | -0.406 |
vantage | -0.397 |
rogen | -0.391 |
Token | Caption |
Token ID | 11260 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
awaru | +0.392 |
ulia | +0.368 |
�� | +0.363 |
vantage | +0.345 |
incial | +0.345 |
Neg logit contributions | |
Instruct | -0.788 |
Anon | -0.756 |
theless | -0.753 |
Teg | -0.750 |
Luxem | -0.750 |
Token | 9 |
Token ID | 860 |
Feature activation | +2.28e-02 |
Pos logit contributions | |
awaru | +2.354 |
ulia | +2.294 |
Combat | +2.221 |
�� | +2.211 |
rogen | +2.192 |
Neg logit contributions | |
Instruct | -3.327 |
theless | -3.246 |
behav | -3.232 |
Anon | -3.146 |
Luxem | -3.138 |
Token | of |
Token ID | 286 |
Feature activation | +3.44e-02 |
Pos logit contributions | |
Instruct | +0.048 |
Anon | +0.045 |
Teg | +0.045 |
Luxem | +0.044 |
theless | +0.044 |
Neg logit contributions | |
awaru | -0.037 |
�� | -0.035 |
ulia | -0.035 |
uese | -0.033 |
vantage | -0.033 |
Token | 15 |
Token ID | 1315 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
Instruct | +0.070 |
Anon | +0.066 |
Teg | +0.065 |
Luxem | +0.064 |
theless | +0.064 |
Neg logit contributions | |
awaru | -0.058 |
�� | -0.055 |
ulia | -0.054 |
vantage | -0.052 |
uese | -0.052 |
Token | Photos |
Token ID | 9434 |
Feature activation | -8.75e-01 |
Pos logit contributions | |
Instruct | +0.220 |
Anon | +0.209 |
Teg | +0.207 |
Luxem | +0.205 |
theless | +0.202 |
Neg logit contributions | |
awaru | -0.160 |
�� | -0.153 |
ulia | -0.148 |
uese | -0.143 |
vantage | -0.143 |
Token | : |
Token ID | 25 |
Feature activation | +1.97e-01 |
Pos logit contributions | |
awaru | +1.744 |
ulia | +1.721 |
Combat | +1.686 |
rogen | +1.675 |
incial | +1.659 |
Neg logit contributions | |
Instruct | -1.410 |
theless | -1.354 |
behav | -1.346 |
Anon | -1.308 |
Luxem | -1.293 |
Token | Grove |
Token ID | 24144 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
Instruct | +0.147 |
Anon | +0.138 |
Teg | +0.136 |
Luxem | +0.135 |
theless | +0.134 |
Neg logit contributions | |
awaru | -0.140 |
�� | -0.133 |
ulia | -0.132 |
uese | -0.128 |
vantage | -0.127 |
Token | , |
Token ID | 11 |
Feature activation | +7.38e-02 |
Pos logit contributions | |
Instruct | +0.359 |
Anon | +0.335 |
theless | +0.327 |
Teg | +0.326 |
Luxem | +0.325 |
Neg logit contributions | |
awaru | -0.540 |
�� | -0.521 |
ulia | -0.517 |
uese | -0.500 |
vantage | -0.497 |
Token | California |
Token ID | 3442 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
Instruct | +0.136 |
Anon | +0.128 |
theless | +0.127 |
behav | +0.126 |
Teg | +0.125 |
Neg logit contributions | |
awaru | -0.135 |
�� | -0.130 |
ulia | -0.128 |
rogen | -0.125 |
vantage | -0.123 |
Token | . |
Token ID | 13 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
Instruct | +0.193 |
Anon | +0.179 |
Teg | +0.178 |
Luxem | +0.174 |
Tuls | +0.169 |
Neg logit contributions | |
awaru | -0.292 |
�� | -0.283 |
ulia | -0.273 |
uese | -0.268 |
vantage | -0.267 |
Token | Hide |
Token ID | 10415 |
Feature activation | -4.44e-01 |
Pos logit contributions | |
Instruct | +0.277 |
Anon | +0.260 |
theless | +0.258 |
Teg | +0.257 |
behav | +0.256 |
Neg logit contributions | |
awaru | -0.319 |
ulia | -0.307 |
�� | -0.306 |
vantage | -0.297 |
uese | -0.295 |
Token | Caption |
Token ID | 11260 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
awaru | +0.528 |
ulia | +0.500 |
�� | +0.488 |
incial | +0.468 |
vantage | +0.467 |
Neg logit contributions | |
Instruct | -1.090 |
theless | -1.051 |
Anon | -1.049 |
Teg | -1.040 |
Luxem | -1.039 |
Token | 15 |
Token ID | 1315 |
Feature activation | -8.47e-02 |
Pos logit contributions | |
awaru | +2.261 |
ulia | +2.194 |
rogen | +2.141 |
Combat | +2.128 |
incial | +2.110 |
Neg logit contributions | |
Instruct | -3.396 |
theless | -3.305 |
behav | -3.260 |
Luxem | -3.209 |
Anon | -3.208 |
Token | of |
Token ID | 286 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
awaru | +0.141 |
�� | +0.134 |
ulia | +0.132 |
uese | +0.127 |
vantage | +0.127 |
Neg logit contributions | |
Instruct | -0.174 |
Anon | -0.164 |
Teg | -0.163 |
Luxem | -0.162 |
theless | -0.159 |
Token | 15 |
Token ID | 1315 |
Feature activation | -8.10e-02 |
Pos logit contributions | |
Instruct | +0.024 |
Anon | +0.022 |
Teg | +0.022 |
Luxem | +0.022 |
theless | +0.022 |
Neg logit contributions | |
awaru | -0.020 |
�� | -0.019 |
ulia | -0.019 |
uese | -0.018 |
vantage | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | +2.57e-01 |
Pos logit contributions | |
awaru | +0.163 |
�� | +0.156 |
ulia | +0.155 |
vantage | +0.150 |
uese | +0.150 |
Neg logit contributions | |
Instruct | -0.137 |
Anon | -0.127 |
Teg | -0.127 |
Luxem | -0.125 |
theless | -0.125 |
Token |
|
Token ID | 198 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
Instruct | +0.434 |
Anon | +0.404 |
Teg | +0.400 |
Luxem | +0.396 |
theless | +0.390 |
Neg logit contributions | |
awaru | -0.522 |
�� | -0.500 |
ulia | -0.495 |
uese | -0.479 |
vantage | -0.479 |
Token | 2002 |
Token ID | 6244 |
Feature activation | -3.85e-02 |
Pos logit contributions | |
Instruct | +0.290 |
Anon | +0.269 |
behav | +0.268 |
Teg | +0.267 |
theless | +0.264 |
Neg logit contributions | |
awaru | -0.252 |
ulia | -0.247 |
�� | -0.245 |
Combat | -0.238 |
rogen | -0.237 |
Token | United |
Token ID | 1578 |
Feature activation | +7.22e-02 |
Pos logit contributions | |
awaru | +0.081 |
�� | +0.078 |
ulia | +0.078 |
Combat | +0.075 |
uese | +0.075 |
Neg logit contributions | |
Instruct | -0.060 |
Anon | -0.055 |
Teg | -0.055 |
behav | -0.055 |
theless | -0.054 |
Token | States |
Token ID | 1829 |
Feature activation | -2.29e-01 |
Pos logit contributions | |
Instruct | +0.146 |
Anon | +0.138 |
Teg | +0.136 |
Luxem | +0.135 |
Tuls | +0.132 |
Neg logit contributions | |
awaru | -0.128 |
�� | -0.120 |
ulia | -0.117 |
vantage | -0.113 |
uese | -0.112 |
Token | 3 |
Token ID | 513 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
awaru | +0.437 |
ulia | +0.422 |
�� | +0.422 |
Combat | +0.408 |
incial | +0.405 |
Neg logit contributions | |
Instruct | -0.395 |
Anon | -0.367 |
Teg | -0.364 |
behav | -0.362 |
theless | -0.360 |
Token | 472 |
Token ID | 37856 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
awaru | +0.259 |
�� | +0.254 |
ulia | +0.250 |
incial | +0.241 |
vantage | +0.240 |
Neg logit contributions | |
Instruct | -0.261 |
Anon | -0.244 |
Teg | -0.243 |
Luxem | -0.239 |
behav | -0.239 |
Token | Posts |
Token ID | 12043 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
awaru | +0.041 |
�� | +0.039 |
ulia | +0.038 |
uese | +0.037 |
vantage | +0.037 |
Neg logit contributions | |
Instruct | -0.039 |
Anon | -0.037 |
Teg | -0.037 |
Luxem | -0.036 |
theless | -0.035 |
Token | # |
Token ID | 1303 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
awaru | +3.095 |
�� | +2.980 |
ulia | +2.919 |
vantage | +2.816 |
uese | +2.816 |
Neg logit contributions | |
Instruct | -2.716 |
Anon | -2.511 |
Teg | -2.499 |
Luxem | -2.478 |
Tuls | -2.442 |
Token | 11 |
Token ID | 1157 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
awaru | +2.130 |
vantage | +2.062 |
Combat | +2.057 |
rogen | +2.050 |
ulia | +2.038 |
Neg logit contributions | |
Instruct | -1.949 |
behav | -1.893 |
Luxem | -1.881 |
theless | -1.858 |
Tuls | -1.847 |
Token | On |
Token ID | 1550 |
Feature activation | -4.41e-01 |
Pos logit contributions | |
awaru | +0.212 |
�� | +0.202 |
ulia | +0.200 |
uese | +0.193 |
vantage | +0.193 |
Neg logit contributions | |
Instruct | -0.205 |
Anon | -0.191 |
Teg | -0.190 |
Luxem | -0.190 |
Tuls | -0.185 |
Token | September |
Token ID | 2693 |
Feature activation | +7.41e-02 |
Pos logit contributions | |
awaru | +0.847 |
�� | +0.816 |
ulia | +0.811 |
rogen | +0.779 |
uese | +0.775 |
Neg logit contributions | |
Instruct | -0.780 |
Teg | -0.725 |
Anon | -0.724 |
theless | -0.718 |
Luxem | -0.717 |
Token | 20 |
Token ID | 1160 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
Instruct | +0.132 |
behav | +0.124 |
theless | +0.124 |
Teg | +0.123 |
Anon | +0.123 |
Neg logit contributions | |
awaru | -0.135 |
ulia | -0.132 |
�� | -0.129 |
Combat | -0.128 |
rogen | -0.126 |
Token | June |
Token ID | 2795 |
Feature activation | +8.26e-02 |
Pos logit contributions | |
ulia | +2.063 |
Combat | +2.026 |
awaru | +2.025 |
rogen | +2.003 |
uese | +1.968 |
Neg logit contributions | |
Instruct | -2.331 |
behav | -2.253 |
theless | -2.215 |
Tuls | -2.145 |
opsis | -2.140 |
Token | 2012 |
Token ID | 2321 |
Feature activation | +3.52e-02 |
Pos logit contributions | |
Instruct | +0.157 |
behav | +0.147 |
Anon | +0.147 |
Teg | +0.145 |
theless | +0.145 |
Neg logit contributions | |
awaru | -0.141 |
ulia | -0.138 |
�� | -0.136 |
Combat | -0.134 |
rogen | -0.132 |
Token | Canada |
Token ID | 3340 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
Instruct | +0.054 |
Anon | +0.050 |
Teg | +0.049 |
behav | +0.049 |
theless | +0.049 |
Neg logit contributions | |
awaru | -0.075 |
�� | -0.072 |
ulia | -0.071 |
uese | -0.069 |
incial | -0.069 |
Token | 16 |
Token ID | 1467 |
Feature activation | +9.81e-02 |
Pos logit contributions | |
Instruct | +0.034 |
Anon | +0.032 |
behav | +0.032 |
theless | +0.032 |
Teg | +0.031 |
Neg logit contributions | |
awaru | -0.037 |
ulia | -0.036 |
�� | -0.035 |
Combat | -0.034 |
incial | -0.034 |
Token | 10 |
Token ID | 940 |
Feature activation | +6.99e-02 |
Pos logit contributions | |
Instruct | +0.160 |
Anon | +0.148 |
Teg | +0.147 |
behav | +0.147 |
Luxem | +0.145 |
Neg logit contributions | |
awaru | -0.197 |
�� | -0.192 |
ulia | -0.192 |
Combat | -0.186 |
incial | -0.186 |
Token | Posts |
Token ID | 12043 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
Instruct | +0.130 |
Anon | +0.122 |
Teg | +0.121 |
Luxem | +0.121 |
theless | +0.120 |
Neg logit contributions | |
awaru | -0.125 |
�� | -0.121 |
ulia | -0.120 |
Combat | -0.116 |
vantage | -0.116 |
Token | # |
Token ID | 1303 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
awaru | +3.194 |
�� | +3.076 |
ulia | +3.005 |
vantage | +2.920 |
uese | +2.920 |
Neg logit contributions | |
Instruct | -2.612 |
Anon | -2.425 |
Teg | -2.409 |
Luxem | -2.377 |
Tuls | -2.328 |
Token | 14 |
Token ID | 1415 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
awaru | +2.088 |
ulia | +2.040 |
Combat | +1.994 |
vantage | +1.990 |
UFC | +1.985 |
Neg logit contributions | |
Instruct | -2.071 |
theless | -1.948 |
behav | -1.937 |
Anon | -1.931 |
Luxem | -1.920 |
Token | On |
Token ID | 1550 |
Feature activation | -4.08e-01 |
Pos logit contributions | |
awaru | +0.422 |
�� | +0.400 |
ulia | +0.397 |
uese | +0.385 |
vantage | +0.384 |
Neg logit contributions | |
Instruct | -0.369 |
Anon | -0.341 |
Teg | -0.340 |
Luxem | -0.340 |
Tuls | -0.331 |
Token | September |
Token ID | 2693 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
awaru | +0.781 |
ulia | +0.742 |
rogen | +0.733 |
�� | +0.730 |
Combat | +0.716 |
Neg logit contributions | |
Instruct | -0.715 |
theless | -0.669 |
Anon | -0.659 |
behav | -0.656 |
Teg | -0.649 |
Token | 20 |
Token ID | 1160 |
Feature activation | +3.99e-02 |
Pos logit contributions | |
Instruct | +0.274 |
Anon | +0.257 |
Teg | +0.255 |
Luxem | +0.251 |
Tuls | +0.245 |
Neg logit contributions | |
awaru | -0.279 |
�� | -0.270 |
ulia | -0.260 |
vantage | -0.252 |
uese | -0.251 |
Token | Women |
Token ID | 6926 |
Feature activation | +9.44e-02 |
Pos logit contributions | |
Instruct | +0.061 |
theless | +0.058 |
Teg | +0.058 |
Anon | +0.058 |
Luxem | +0.058 |
Neg logit contributions | |
awaru | -0.036 |
�� | -0.034 |
ulia | -0.034 |
rogen | -0.032 |
vantage | -0.032 |
Token | in |
Token ID | 287 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
Instruct | +0.158 |
Anon | +0.147 |
theless | +0.147 |
Teg | +0.145 |
Luxem | +0.144 |
Neg logit contributions | |
awaru | -0.189 |
�� | -0.182 |
ulia | -0.180 |
rogen | -0.176 |
vantage | -0.175 |
Token | politics |
Token ID | 4819 |
Feature activation | +5.25e-02 |
Pos logit contributions | |
awaru | +0.033 |
�� | +0.033 |
ulia | +0.032 |
Combat | +0.031 |
uese | +0.031 |
Neg logit contributions | |
Instruct | -0.027 |
Anon | -0.025 |
theless | -0.024 |
Teg | -0.024 |
Luxem | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
Instruct | +0.089 |
Anon | +0.083 |
theless | +0.083 |
Teg | +0.082 |
Luxem | +0.081 |
Neg logit contributions | |
awaru | -0.104 |
ulia | -0.100 |
�� | -0.100 |
uese | -0.096 |
rogen | -0.096 |
Token | Hide |
Token ID | 10415 |
Feature activation | -3.10e-01 |
Pos logit contributions | |
Instruct | +0.377 |
Anon | +0.353 |
Teg | +0.349 |
theless | +0.347 |
Luxem | +0.346 |
Neg logit contributions | |
awaru | -0.408 |
�� | -0.392 |
ulia | -0.389 |
vantage | -0.376 |
uese | -0.374 |
Token | Caption |
Token ID | 11260 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
awaru | +0.373 |
ulia | +0.350 |
�� | +0.344 |
vantage | +0.328 |
incial | +0.328 |
Neg logit contributions | |
Instruct | -0.761 |
Anon | -0.731 |
theless | -0.728 |
Luxem | -0.725 |
Teg | -0.725 |
Token | 7 |
Token ID | 767 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
awaru | +2.320 |
ulia | +2.227 |
Combat | +2.214 |
rogen | +2.190 |
uese | +2.155 |
Neg logit contributions | |
Instruct | -3.088 |
theless | -3.026 |
behav | -2.993 |
Teg | -2.933 |
nodd | -2.921 |
Token | of |
Token ID | 286 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
awaru | +0.250 |
�� | +0.234 |
ulia | +0.234 |
uese | +0.225 |
rogen | +0.224 |
Neg logit contributions | |
Instruct | -0.328 |
Anon | -0.310 |
Teg | -0.307 |
Luxem | -0.305 |
theless | -0.304 |
Token | 15 |
Token ID | 1315 |
Feature activation | +7.10e-02 |
Pos logit contributions | |
Instruct | +0.032 |
Anon | +0.030 |
Teg | +0.030 |
Luxem | +0.030 |
theless | +0.030 |
Neg logit contributions | |
awaru | -0.027 |
�� | -0.025 |
ulia | -0.025 |
vantage | -0.024 |
uese | -0.024 |
Token | Photos |
Token ID | 9434 |
Feature activation | -8.78e-01 |
Pos logit contributions | |
Instruct | +0.153 |
Anon | +0.145 |
Teg | +0.144 |
Luxem | +0.142 |
theless | +0.140 |
Neg logit contributions | |
awaru | -0.112 |
�� | -0.107 |
ulia | -0.104 |
uese | -0.100 |
vantage | -0.100 |
Token | : |
Token ID | 25 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
awaru | +1.746 |
ulia | +1.722 |
Combat | +1.689 |
rogen | +1.675 |
incial | +1.660 |
Neg logit contributions | |
Instruct | -1.414 |
theless | -1.362 |
behav | -1.353 |
Anon | -1.313 |
Luxem | -1.301 |
Token | 2012 |
Token ID | 2321 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
Instruct | +0.260 |
behav | +0.245 |
Anon | +0.243 |
theless | +0.241 |
Teg | +0.240 |
Neg logit contributions | |
awaru | -0.227 |
ulia | -0.223 |
�� | -0.220 |
Combat | -0.215 |
uese | -0.212 |
Token | United |
Token ID | 1578 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
Instruct | +0.209 |
Anon | +0.192 |
Teg | +0.191 |
behav | +0.191 |
theless | +0.189 |
Neg logit contributions | |
awaru | -0.267 |
�� | -0.258 |
ulia | -0.256 |
Combat | -0.248 |
uese | -0.248 |
Token | States |
Token ID | 1829 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
Instruct | +0.251 |
Anon | +0.238 |
Teg | +0.235 |
Luxem | +0.233 |
Tuls | +0.229 |
Neg logit contributions | |
awaru | -0.223 |
�� | -0.212 |
ulia | -0.206 |
vantage | -0.199 |
uese | -0.197 |
Token | 5 |
Token ID | 642 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
awaru | +0.259 |
ulia | +0.250 |
�� | +0.249 |
Combat | +0.241 |
incial | +0.240 |
Neg logit contributions | |
Instruct | -0.237 |
Anon | -0.221 |
behav | -0.218 |
Teg | -0.218 |
theless | -0.217 |
Token | 667 |
Token ID | 28933 |
Feature activation | -7.58e-02 |
Pos logit contributions | |
awaru | +0.289 |
�� | +0.281 |
ulia | +0.279 |
incial | +0.270 |
uese | +0.270 |
Neg logit contributions | |
Instruct | -0.243 |
Anon | -0.227 |
Teg | -0.225 |
behav | -0.223 |
Luxem | -0.222 |
Token | Posts |
Token ID | 12043 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
awaru | +0.136 |
�� | +0.131 |
ulia | +0.130 |
uese | +0.125 |
Combat | +0.125 |
Neg logit contributions | |
Instruct | -0.142 |
Anon | -0.133 |
Teg | -0.132 |
Luxem | -0.131 |
theless | -0.130 |
Token | # |
Token ID | 1303 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
awaru | +3.064 |
�� | +2.949 |
ulia | +2.884 |
uese | +2.797 |
vantage | +2.796 |
Neg logit contributions | |
Instruct | -2.576 |
Anon | -2.389 |
Teg | -2.374 |
Luxem | -2.345 |
theless | -2.303 |
Token | 12 |
Token ID | 1065 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
awaru | +1.807 |
ulia | +1.763 |
Combat | +1.736 |
vantage | +1.728 |
UFC | +1.704 |
Neg logit contributions | |
Instruct | -2.243 |
theless | -2.142 |
behav | -2.119 |
Anon | -2.114 |
Luxem | -2.107 |
Token | I |
Token ID | 314 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
awaru | +0.315 |
�� | +0.300 |
ulia | +0.297 |
uese | +0.287 |
vantage | +0.287 |
Neg logit contributions | |
Instruct | -0.278 |
Luxem | -0.257 |
Anon | -0.257 |
Teg | -0.257 |
Tuls | -0.249 |
Token | 'm |
Token ID | 1101 |
Feature activation | +7.74e-02 |
Pos logit contributions | |
awaru | +0.453 |
ulia | +0.436 |
�� | +0.429 |
vantage | +0.420 |
rogen | +0.415 |
Neg logit contributions | |
Instruct | -0.521 |
Luxem | -0.490 |
Anon | -0.487 |
Teg | -0.487 |
Tuls | -0.480 |
Token | glad |
Token ID | 9675 |
Feature activation | +3.16e-02 |
Pos logit contributions | |
Instruct | +0.130 |
Luxem | +0.120 |
Teg | +0.120 |
Anon | +0.119 |
theless | +0.118 |
Neg logit contributions | |
awaru | -0.155 |
�� | -0.149 |
ulia | -0.146 |
uese | -0.143 |
rogen | -0.142 |
Token | 2011 |
Token ID | 2813 |
Feature activation | +2.81e-01 |
Pos logit contributions | |
Instruct | +0.477 |
behav | +0.446 |
Anon | +0.445 |
theless | +0.441 |
Teg | +0.441 |
Neg logit contributions | |
awaru | -0.425 |
ulia | -0.415 |
�� | -0.410 |
Combat | -0.400 |
rogen | -0.397 |
Token | United |
Token ID | 1578 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
Instruct | +0.427 |
Anon | +0.392 |
Teg | +0.388 |
Luxem | +0.380 |
theless | +0.379 |
Neg logit contributions | |
awaru | -0.608 |
�� | -0.589 |
ulia | -0.582 |
uese | -0.565 |
vantage | -0.564 |
Token | States |
Token ID | 1829 |
Feature activation | -9.72e-02 |
Pos logit contributions | |
Instruct | +0.244 |
Anon | +0.231 |
Teg | +0.228 |
Luxem | +0.226 |
Tuls | +0.222 |
Neg logit contributions | |
awaru | -0.219 |
�� | -0.207 |
ulia | -0.201 |
vantage | -0.195 |
uese | -0.194 |
Token | 228 |
Token ID | 29041 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
awaru | +0.184 |
ulia | +0.178 |
�� | +0.177 |
Combat | +0.172 |
rogen | +0.171 |
Neg logit contributions | |
Instruct | -0.169 |
Anon | -0.157 |
Teg | -0.155 |
behav | -0.155 |
theless | -0.154 |
Token | 65 |
Token ID | 2996 |
Feature activation | +4.66e-02 |
Pos logit contributions | |
Instruct | +0.242 |
Anon | +0.228 |
Teg | +0.226 |
Luxem | +0.224 |
behav | +0.223 |
Neg logit contributions | |
awaru | -0.263 |
�� | -0.254 |
ulia | -0.252 |
vantage | -0.245 |
Combat | -0.245 |
Token | Posts |
Token ID | 12043 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
Instruct | +0.086 |
Anon | +0.081 |
Teg | +0.080 |
Luxem | +0.079 |
theless | +0.079 |
Neg logit contributions | |
awaru | -0.086 |
�� | -0.083 |
ulia | -0.082 |
uese | -0.079 |
vantage | -0.079 |
Token | # |
Token ID | 1303 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
awaru | +3.060 |
�� | +2.946 |
ulia | +2.880 |
uese | +2.796 |
vantage | +2.796 |
Neg logit contributions | |
Instruct | -2.515 |
Anon | -2.334 |
Teg | -2.319 |
Luxem | -2.290 |
Tuls | -2.243 |
Token | 13 |
Token ID | 1485 |
Feature activation | -9.26e-02 |
Pos logit contributions | |
awaru | +2.122 |
ulia | +2.038 |
Combat | +2.025 |
vantage | +2.008 |
UFC | +2.000 |
Neg logit contributions | |
Instruct | -1.932 |
theless | -1.826 |
behav | -1.817 |
Teg | -1.815 |
Anon | -1.809 |
Token | On |
Token ID | 1550 |
Feature activation | -3.84e-01 |
Pos logit contributions | |
awaru | +0.180 |
�� | +0.171 |
ulia | +0.170 |
uese | +0.165 |
vantage | +0.164 |
Neg logit contributions | |
Instruct | -0.163 |
Anon | -0.151 |
Teg | -0.150 |
Luxem | -0.150 |
Tuls | -0.146 |
Token | September |
Token ID | 2693 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
awaru | +0.731 |
ulia | +0.694 |
rogen | +0.689 |
�� | +0.685 |
Combat | +0.671 |
Neg logit contributions | |
Instruct | -0.676 |
theless | -0.634 |
behav | -0.622 |
Anon | -0.622 |
Teg | -0.616 |
Token | 20 |
Token ID | 1160 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
Instruct | +0.315 |
Anon | +0.296 |
Teg | +0.294 |
Luxem | +0.291 |
Tuls | +0.285 |
Neg logit contributions | |
awaru | -0.294 |
�� | -0.281 |
ulia | -0.274 |
uese | -0.266 |
vantage | -0.266 |
Token | 2010 |
Token ID | 3050 |
Feature activation | -6.28e-02 |
Pos logit contributions | |
Instruct | +0.368 |
Anon | +0.343 |
behav | +0.342 |
theless | +0.340 |
Teg | +0.339 |
Neg logit contributions | |
awaru | -0.334 |
ulia | -0.325 |
�� | -0.321 |
uese | -0.312 |
Combat | -0.312 |
Token | United |
Token ID | 1578 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
awaru | +0.136 |
�� | +0.132 |
ulia | +0.131 |
uese | +0.128 |
incial | +0.127 |
Neg logit contributions | |
Instruct | -0.094 |
Anon | -0.085 |
Teg | -0.085 |
behav | -0.084 |
theless | -0.084 |
Token | States |
Token ID | 1829 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
Instruct | +0.211 |
Anon | +0.199 |
Teg | +0.197 |
Luxem | +0.195 |
Tuls | +0.192 |
Neg logit contributions | |
awaru | -0.190 |
�� | -0.180 |
ulia | -0.174 |
vantage | -0.169 |
uese | -0.167 |
Token | 10 |
Token ID | 838 |
Feature activation | -3.70e-02 |
Pos logit contributions | |
awaru | +0.287 |
ulia | +0.277 |
�� | +0.276 |
Combat | +0.267 |
incial | +0.266 |
Neg logit contributions | |
Instruct | -0.264 |
Anon | -0.245 |
behav | -0.243 |
Teg | -0.242 |
theless | -0.241 |
Token | 53 |
Token ID | 4310 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
awaru | +0.075 |
�� | +0.073 |
ulia | +0.072 |
vantage | +0.070 |
uese | +0.070 |
Neg logit contributions | |
Instruct | -0.061 |
Anon | -0.056 |
Teg | -0.056 |
Luxem | -0.055 |
behav | -0.055 |
Token | Posts |
Token ID | 12043 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
awaru | +0.202 |
�� | +0.195 |
ulia | +0.192 |
vantage | +0.186 |
uese | +0.185 |
Neg logit contributions | |
Instruct | -0.203 |
Anon | -0.191 |
Teg | -0.189 |
Luxem | -0.188 |
theless | -0.186 |
Token | # |
Token ID | 1303 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
awaru | +3.061 |
�� | +2.947 |
ulia | +2.878 |
uese | +2.799 |
vantage | +2.798 |
Neg logit contributions | |
Instruct | -2.501 |
Anon | -2.324 |
Teg | -2.308 |
Luxem | -2.278 |
Tuls | -2.227 |
Token | 15 |
Token ID | 1314 |
Feature activation | -2.54e-01 |
Pos logit contributions | |
awaru | +1.883 |
ulia | +1.827 |
rogen | +1.783 |
Combat | +1.779 |
vantage | +1.762 |
Neg logit contributions | |
Instruct | -2.463 |
Luxem | -2.306 |
Anon | -2.303 |
theless | -2.287 |
behav | -2.280 |
Token | honestly |
Token ID | 12698 |
Feature activation | +2.51e-02 |
Pos logit contributions | |
awaru | +0.503 |
�� | +0.479 |
ulia | +0.474 |
uese | +0.460 |
vantage | +0.459 |
Neg logit contributions | |
Instruct | -0.438 |
Anon | -0.405 |
Luxem | -0.404 |
Teg | -0.404 |
Tuls | -0.393 |
Token | if |
Token ID | 611 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
Instruct | +0.045 |
Anon | +0.042 |
Luxem | +0.042 |
Teg | +0.041 |
Tuls | +0.040 |
Neg logit contributions | |
awaru | -0.048 |
�� | -0.046 |
ulia | -0.045 |
vantage | -0.044 |
uese | -0.044 |
Token | we |
Token ID | 356 |
Feature activation | -8.88e-02 |
Pos logit contributions | |
Instruct | +0.268 |
Anon | +0.249 |
Teg | +0.248 |
Luxem | +0.247 |
Tuls | +0.241 |
Neg logit contributions | |
awaru | -0.292 |
�� | -0.280 |
ulia | -0.274 |
uese | -0.266 |
vantage | -0.266 |
Token | 2012 |
Token ID | 2321 |
Feature activation | +6.01e-02 |
Pos logit contributions | |
Instruct | +0.395 |
Anon | +0.367 |
Teg | +0.367 |
behav | +0.367 |
theless | +0.364 |
Neg logit contributions | |
awaru | -0.371 |
ulia | -0.360 |
�� | -0.357 |
Combat | -0.346 |
uese | -0.345 |
Token | United |
Token ID | 1578 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
Instruct | +0.091 |
Anon | +0.084 |
behav | +0.083 |
Teg | +0.083 |
theless | +0.082 |
Neg logit contributions | |
awaru | -0.128 |
�� | -0.124 |
ulia | -0.124 |
uese | -0.120 |
Combat | -0.120 |
Token | States |
Token ID | 1829 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
Instruct | +0.378 |
Anon | +0.357 |
Teg | +0.355 |
Luxem | +0.351 |
Tuls | +0.345 |
Neg logit contributions | |
awaru | -0.298 |
�� | -0.283 |
ulia | -0.275 |
vantage | -0.265 |
uese | -0.265 |
Token | 3 |
Token ID | 513 |
Feature activation | -8.58e-02 |
Pos logit contributions | |
awaru | +0.360 |
�� | +0.349 |
ulia | +0.348 |
Combat | +0.337 |
incial | +0.335 |
Neg logit contributions | |
Instruct | -0.330 |
Anon | -0.308 |
behav | -0.306 |
Teg | -0.305 |
theless | -0.304 |
Token | 240 |
Token ID | 16102 |
Feature activation | -1.94e-01 |
Pos logit contributions | |
awaru | +0.158 |
�� | +0.154 |
ulia | +0.150 |
uese | +0.145 |
vantage | +0.145 |
Neg logit contributions | |
Instruct | -0.158 |
Anon | -0.149 |
Teg | -0.147 |
Luxem | -0.146 |
Tuls | -0.143 |
Token | Posts |
Token ID | 12043 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
awaru | +0.338 |
�� | +0.324 |
ulia | +0.320 |
vantage | +0.311 |
rogen | +0.309 |
Neg logit contributions | |
Instruct | -0.372 |
Anon | -0.353 |
Teg | -0.348 |
Luxem | -0.347 |
behav | -0.340 |
Token | # |
Token ID | 1303 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
awaru | +2.727 |
�� | +2.621 |
ulia | +2.587 |
rogen | +2.488 |
vantage | +2.486 |
Neg logit contributions | |
Instruct | -2.710 |
Anon | -2.504 |
Teg | -2.490 |
Luxem | -2.478 |
theless | -2.460 |
Token | 10 |
Token ID | 940 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
awaru | +2.132 |
ulia | +2.092 |
Combat | +2.062 |
�� | +2.036 |
vantage | +2.032 |
Neg logit contributions | |
Instruct | -2.340 |
behav | -2.227 |
theless | -2.191 |
Teg | -2.184 |
Luxem | -2.183 |
Token | Happy |
Token ID | 14628 |
Feature activation | -3.27e-02 |
Pos logit contributions | |
awaru | +0.412 |
�� | +0.393 |
ulia | +0.388 |
vantage | +0.376 |
uese | +0.376 |
Neg logit contributions | |
Instruct | -0.380 |
Anon | -0.354 |
Teg | -0.353 |
Luxem | -0.350 |
theless | -0.341 |
Token | it |
Token ID | 340 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
awaru | +0.063 |
�� | +0.061 |
ulia | +0.059 |
uese | +0.058 |
rogen | +0.057 |
Neg logit contributions | |
Instruct | -0.058 |
Anon | -0.054 |
Teg | -0.054 |
Luxem | -0.053 |
theless | -0.052 |
Token | worked |
Token ID | 3111 |
Feature activation | -3.63e-01 |
Pos logit contributions | |
awaru | +0.293 |
�� | +0.279 |
ulia | +0.276 |
uese | +0.274 |
rogen | +0.269 |
Neg logit contributions | |
Instruct | -0.243 |
Anon | -0.223 |
Teg | -0.223 |
Luxem | -0.222 |
Tuls | -0.215 |
Token | 2010 |
Token ID | 3050 |
Feature activation | +7.16e-02 |
Pos logit contributions | |
Instruct | +0.559 |
Anon | +0.522 |
behav | +0.520 |
theless | +0.516 |
Teg | +0.515 |
Neg logit contributions | |
awaru | -0.501 |
ulia | -0.486 |
�� | -0.482 |
Combat | -0.468 |
rogen | -0.467 |
Token | United |
Token ID | 1578 |
Feature activation | -6.48e-03 |
Pos logit contributions | |
Instruct | +0.107 |
Anon | +0.098 |
Teg | +0.097 |
Luxem | +0.095 |
theless | +0.095 |
Neg logit contributions | |
awaru | -0.157 |
�� | -0.152 |
ulia | -0.150 |
uese | -0.146 |
vantage | -0.146 |
Token | States |
Token ID | 1829 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
awaru | +0.010 |
�� | +0.009 |
ulia | +0.009 |
uese | +0.009 |
vantage | +0.009 |
Neg logit contributions | |
Instruct | -0.014 |
Anon | -0.013 |
Teg | -0.013 |
Luxem | -0.013 |
theless | -0.013 |
Token | 6 |
Token ID | 718 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
awaru | +0.390 |
ulia | +0.376 |
�� | +0.375 |
Combat | +0.364 |
incial | +0.361 |
Neg logit contributions | |
Instruct | -0.359 |
Anon | -0.333 |
behav | -0.330 |
Teg | -0.329 |
theless | -0.327 |
Token | 83 |
Token ID | 5999 |
Feature activation | -2.04e-02 |
Pos logit contributions | |
awaru | +0.239 |
�� | +0.233 |
ulia | +0.231 |
Combat | +0.224 |
incial | +0.224 |
Neg logit contributions | |
Instruct | -0.191 |
Anon | -0.177 |
Teg | -0.176 |
Luxem | -0.174 |
behav | -0.174 |
Token | Posts |
Token ID | 12043 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
awaru | +0.036 |
�� | +0.035 |
ulia | +0.035 |
Combat | +0.034 |
vantage | +0.033 |
Neg logit contributions | |
Instruct | -0.038 |
Anon | -0.036 |
Teg | -0.036 |
Luxem | -0.035 |
theless | -0.035 |
Token | Last |
Token ID | 4586 |
Feature activation | -4.46e-01 |
Pos logit contributions | |
awaru | +3.075 |
�� | +2.963 |
ulia | +2.893 |
uese | +2.816 |
vantage | +2.815 |
Neg logit contributions | |
Instruct | -2.439 |
Anon | -2.266 |
Teg | -2.249 |
Luxem | -2.218 |
Tuls | -2.166 |
Token | Edited |
Token ID | 34212 |
Feature activation | -9.66e-01 |
Pos logit contributions | |
awaru | +0.547 |
ulia | +0.505 |
�� | +0.501 |
vantage | +0.483 |
uese | +0.480 |
Neg logit contributions | |
Instruct | -1.094 |
Teg | -1.041 |
Luxem | -1.039 |
theless | -1.034 |
Anon | -1.031 |
Token | : |
Token ID | 25 |
Feature activation | -9.66e-02 |
Pos logit contributions | |
awaru | +1.874 |
ulia | +1.861 |
rogen | +1.841 |
Combat | +1.816 |
incial | +1.784 |
Neg logit contributions | |
Instruct | -1.564 |
theless | -1.522 |
behav | -1.504 |
Luxem | -1.455 |
Anon | -1.439 |
Token | 2013 |
Token ID | 2211 |
Feature activation | +3.34e-02 |
Pos logit contributions | |
awaru | +0.160 |
ulia | +0.156 |
rogen | +0.154 |
Combat | +0.152 |
uese | +0.150 |
Neg logit contributions | |
Instruct | -0.188 |
behav | -0.179 |
theless | -0.177 |
Anon | -0.176 |
Tuls | -0.171 |
Token | - |
Token ID | 12 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
Instruct | +0.060 |
Anon | +0.056 |
theless | +0.056 |
Teg | +0.056 |
Luxem | +0.055 |
Neg logit contributions | |
awaru | -0.062 |
�� | -0.060 |
ulia | -0.060 |
Combat | -0.058 |
uese | -0.058 |
Token | 2011 |
Token ID | 2813 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
Instruct | +0.579 |
Anon | +0.542 |
Teg | +0.536 |
behav | +0.534 |
theless | +0.533 |
Neg logit contributions | |
awaru | -0.531 |
ulia | -0.515 |
�� | -0.512 |
rogen | -0.495 |
Combat | -0.494 |
Token | United |
Token ID | 1578 |
Feature activation | -2.00e-02 |
Pos logit contributions | |
Instruct | +0.251 |
Anon | +0.230 |
Teg | +0.227 |
behav | +0.223 |
theless | +0.222 |
Neg logit contributions | |
awaru | -0.367 |
�� | -0.356 |
ulia | -0.351 |
uese | -0.342 |
vantage | -0.341 |
Token | States |
Token ID | 1829 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
awaru | +0.036 |
�� | +0.034 |
ulia | +0.033 |
vantage | +0.032 |
uese | +0.032 |
Neg logit contributions | |
Instruct | -0.040 |
Anon | -0.038 |
Teg | -0.037 |
Luxem | -0.037 |
Tuls | -0.036 |
Token | 6 |
Token ID | 718 |
Feature activation | -1.94e-01 |
Pos logit contributions | |
awaru | +0.404 |
ulia | +0.391 |
�� | +0.389 |
Combat | +0.377 |
incial | +0.374 |
Neg logit contributions | |
Instruct | -0.377 |
Anon | -0.351 |
behav | -0.347 |
Teg | -0.346 |
theless | -0.344 |
Token | 12 |
Token ID | 1065 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
awaru | +0.379 |
�� | +0.371 |
ulia | +0.368 |
Combat | +0.357 |
vantage | +0.356 |
Neg logit contributions | |
Instruct | -0.328 |
Anon | -0.306 |
Teg | -0.303 |
behav | -0.300 |
Luxem | -0.300 |
Token | Posts |
Token ID | 12043 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
Instruct | +0.023 |
Anon | +0.021 |
Teg | +0.021 |
Luxem | +0.021 |
theless | +0.021 |
Neg logit contributions | |
awaru | -0.022 |
�� | -0.022 |
ulia | -0.021 |
Combat | -0.021 |
vantage | -0.021 |
Token | # |
Token ID | 1303 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
awaru | +3.055 |
�� | +2.944 |
ulia | +2.875 |
uese | +2.798 |
vantage | +2.797 |
Neg logit contributions | |
Instruct | -2.422 |
Anon | -2.249 |
Teg | -2.232 |
Luxem | -2.202 |
Tuls | -2.152 |
Token | 17 |
Token ID | 1558 |
Feature activation | +4.29e-02 |
Pos logit contributions | |
awaru | +2.254 |
ulia | +2.202 |
Combat | +2.150 |
rogen | +2.122 |
UFC | +2.114 |
Neg logit contributions | |
Instruct | -2.106 |
Anon | -1.998 |
theless | -1.985 |
Teg | -1.977 |
behav | -1.961 |
Token | Good |
Token ID | 4599 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
Instruct | +0.076 |
Anon | +0.071 |
Teg | +0.070 |
Luxem | +0.070 |
Tuls | +0.068 |
Neg logit contributions | |
awaru | -0.083 |
�� | -0.079 |
ulia | -0.078 |
uese | -0.076 |
vantage | -0.076 |
Token | to |
Token ID | 284 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
Instruct | +0.255 |
Teg | +0.236 |
Anon | +0.236 |
Luxem | +0.235 |
theless | +0.229 |
Neg logit contributions | |
awaru | -0.292 |
�� | -0.280 |
ulia | -0.272 |
uese | -0.267 |
vantage | -0.265 |
Token | hear |
Token ID | 3285 |
Feature activation | -7.61e-02 |
Pos logit contributions | |
awaru | +0.307 |
�� | +0.292 |
ulia | +0.283 |
uese | +0.281 |
rogen | +0.279 |
Neg logit contributions | |
Instruct | -0.306 |
Anon | -0.284 |
Luxem | -0.281 |
Teg | -0.279 |
theless | -0.278 |
Token | 2010 |
Token ID | 3050 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
Instruct | +0.520 |
Anon | +0.485 |
behav | +0.483 |
Teg | +0.479 |
theless | +0.478 |
Neg logit contributions | |
awaru | -0.450 |
ulia | -0.438 |
�� | -0.435 |
Combat | -0.421 |
rogen | -0.421 |
Token | United |
Token ID | 1578 |
Feature activation | +7.53e-02 |
Pos logit contributions | |
Instruct | +0.199 |
Anon | +0.183 |
Teg | +0.182 |
Luxem | +0.178 |
theless | +0.176 |
Neg logit contributions | |
awaru | -0.273 |
�� | -0.264 |
ulia | -0.260 |
uese | -0.253 |
vantage | -0.253 |
Token | States |
Token ID | 1829 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
Instruct | +0.150 |
Anon | +0.142 |
Teg | +0.140 |
Luxem | +0.139 |
Tuls | +0.137 |
Neg logit contributions | |
awaru | -0.135 |
�� | -0.127 |
ulia | -0.124 |
vantage | -0.119 |
uese | -0.119 |
Token | 21 |
Token ID | 2310 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
awaru | +0.256 |
ulia | +0.247 |
�� | +0.246 |
Combat | +0.239 |
incial | +0.237 |
Neg logit contributions | |
Instruct | -0.239 |
Anon | -0.222 |
behav | -0.220 |
Teg | -0.219 |
theless | -0.217 |
Token | 50 |
Token ID | 1120 |
Feature activation | +1.56e-03 |
Pos logit contributions | |
Instruct | +0.263 |
Anon | +0.245 |
Teg | +0.244 |
behav | +0.241 |
Luxem | +0.239 |
Neg logit contributions | |
awaru | -0.288 |
�� | -0.282 |
ulia | -0.277 |
vantage | -0.271 |
Combat | -0.270 |
Token | Posts |
Token ID | 12043 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
Instruct | +0.003 |
Anon | +0.003 |
Teg | +0.003 |
Luxem | +0.003 |
theless | +0.003 |
Neg logit contributions | |
awaru | -0.003 |
�� | -0.003 |
ulia | -0.003 |
vantage | -0.003 |
uese | -0.003 |
Token | # |
Token ID | 1303 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
awaru | +3.042 |
�� | +2.930 |
ulia | +2.861 |
uese | +2.785 |
vantage | +2.784 |
Neg logit contributions | |
Instruct | -2.429 |
Anon | -2.256 |
Teg | -2.240 |
Luxem | -2.210 |
Tuls | -2.159 |
Token | 16 |
Token ID | 1433 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
awaru | +2.139 |
ulia | +2.110 |
vantage | +2.050 |
Combat | +2.049 |
rogen | +2.017 |
Neg logit contributions | |
Instruct | -1.798 |
theless | -1.697 |
Teg | -1.684 |
Luxem | -1.683 |
Anon | -1.677 |
Token | Glad |
Token ID | 28442 |
Feature activation | +4.40e-01 |
Pos logit contributions | |
awaru | +0.300 |
�� | +0.285 |
ulia | +0.284 |
uese | +0.275 |
vantage | +0.274 |
Neg logit contributions | |
Instruct | -0.258 |
Anon | -0.238 |
Teg | -0.238 |
Luxem | -0.238 |
Tuls | -0.232 |
Token | to |
Token ID | 284 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
Instruct | +0.795 |
Anon | +0.741 |
Teg | +0.735 |
Luxem | +0.730 |
Tuls | +0.713 |
Neg logit contributions | |
awaru | -0.849 |
�� | -0.813 |
ulia | -0.795 |
uese | -0.773 |
vantage | -0.772 |
Token | hear |
Token ID | 3285 |
Feature activation | -1.94e-04 |
Pos logit contributions | |
awaru | +0.195 |
�� | +0.186 |
ulia | +0.181 |
uese | +0.179 |
inate | +0.177 |
Neg logit contributions | |
Instruct | -0.182 |
Anon | -0.168 |
Teg | -0.167 |
Luxem | -0.167 |
theless | -0.165 |
Token | drugs |
Token ID | 5010 |
Feature activation | +2.68e-01 |
Pos logit contributions | |
awaru | +0.166 |
�� | +0.162 |
ulia | +0.156 |
uese | +0.154 |
vantage | +0.153 |
Neg logit contributions | |
Instruct | -0.168 |
Anon | -0.161 |
Teg | -0.158 |
Luxem | -0.158 |
theless | -0.154 |
Token | come |
Token ID | 1282 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
Instruct | +0.438 |
Anon | +0.411 |
Teg | +0.402 |
Luxem | +0.400 |
theless | +0.398 |
Neg logit contributions | |
awaru | -0.557 |
�� | -0.526 |
ulia | -0.522 |
uese | -0.509 |
rogen | -0.509 |
Token | off |
Token ID | 572 |
Feature activation | +3.80e-02 |
Pos logit contributions | |
Instruct | +0.353 |
Anon | +0.328 |
Teg | +0.326 |
Luxem | +0.324 |
theless | +0.317 |
Neg logit contributions | |
awaru | -0.427 |
�� | -0.410 |
ulia | -0.403 |
uese | -0.393 |
vantage | -0.390 |
Token | their |
Token ID | 511 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
Instruct | +0.065 |
Anon | +0.062 |
Luxem | +0.060 |
Teg | +0.060 |
theless | +0.059 |
Neg logit contributions | |
awaru | -0.076 |
�� | -0.072 |
ulia | -0.071 |
vantage | -0.069 |
rogen | -0.069 |
Token | cr |
Token ID | 1067 |
Feature activation | -9.56e-02 |
Pos logit contributions | |
awaru | +0.186 |
�� | +0.177 |
ulia | +0.172 |
rogen | +0.169 |
Combat | +0.166 |
Neg logit contributions | |
Instruct | -0.188 |
Anon | -0.178 |
Luxem | -0.175 |
Teg | -0.174 |
theless | -0.172 |
Token | utch |
Token ID | 7140 |
Feature activation | +7.63e-01 |
Pos logit contributions | |
awaru | +0.162 |
ulia | +0.154 |
incial | +0.147 |
uese | +0.147 |
inate | +0.146 |
Neg logit contributions | |
Instruct | -0.197 |
Anon | -0.184 |
Teg | -0.181 |
Luxem | -0.181 |
Tuls | -0.181 |
Token | . |
Token ID | 13 |
Feature activation | -9.84e-02 |
Pos logit contributions | |
Instruct | +1.253 |
Anon | +1.167 |
Teg | +1.153 |
Luxem | +1.143 |
theless | +1.112 |
Neg logit contributions | |
awaru | -1.600 |
�� | -1.536 |
ulia | -1.506 |
vantage | -1.468 |
uese | -1.468 |
Token |
|
Token ID | 198 |
Feature activation | -7.36e-02 |
Pos logit contributions | |
awaru | +0.180 |
ulia | +0.173 |
�� | +0.172 |
Combat | +0.167 |
vantage | +0.166 |
Neg logit contributions | |
Instruct | -0.179 |
Anon | -0.168 |
Teg | -0.168 |
Luxem | -0.167 |
theless | -0.166 |
Token |
|
Token ID | 198 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
awaru | +0.143 |
ulia | +0.137 |
�� | +0.136 |
Combat | +0.133 |
rogen | +0.132 |
Neg logit contributions | |
Instruct | -0.128 |
Anon | -0.120 |
Teg | -0.119 |
Luxem | -0.118 |
theless | -0.118 |
Token | There |
Token ID | 1858 |
Feature activation | -6.20e-02 |
Pos logit contributions | |
Instruct | +0.022 |
Anon | +0.020 |
Teg | +0.020 |
Luxem | +0.020 |
theless | +0.020 |
Neg logit contributions | |
awaru | -0.017 |
�� | -0.016 |
ulia | -0.016 |
Combat | -0.016 |
vantage | -0.016 |
Token | are |
Token ID | 389 |
Feature activation | -2.72e-02 |
Pos logit contributions | |
awaru | +0.098 |
�� | +0.093 |
ulia | +0.091 |
vantage | +0.088 |
uese | +0.087 |
Neg logit contributions | |
Instruct | -0.133 |
Teg | -0.126 |
Anon | -0.125 |
Luxem | -0.125 |
Tuls | -0.122 |
Token | String |
Token ID | 10100 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
Instruct | +0.114 |
Anon | +0.107 |
Luxem | +0.106 |
Teg | +0.106 |
theless | +0.105 |
Neg logit contributions | |
awaru | -0.095 |
�� | -0.092 |
ulia | -0.090 |
Combat | -0.088 |
vantage | -0.087 |
Token | Al |
Token ID | 2348 |
Feature activation | +2.01e-01 |
Pos logit contributions | |
awaru | +0.235 |
�� | +0.219 |
ulia | +0.216 |
Combat | +0.212 |
uese | +0.204 |
Neg logit contributions | |
Instruct | -0.586 |
Anon | -0.557 |
Teg | -0.556 |
theless | -0.555 |
Luxem | -0.553 |
Token | ignment |
Token ID | 16747 |
Feature activation | +4.58e-02 |
Pos logit contributions | |
Instruct | +0.442 |
Anon | +0.418 |
Teg | +0.418 |
Luxem | +0.412 |
Tuls | +0.409 |
Neg logit contributions | |
awaru | -0.310 |
�� | -0.290 |
ulia | -0.288 |
vantage | -0.278 |
uese | -0.277 |
Token | ] |
Token ID | 60 |
Feature activation | +3.11e-01 |
Pos logit contributions | |
Instruct | +0.059 |
Anon | +0.054 |
Teg | +0.053 |
theless | +0.053 |
Luxem | +0.053 |
Neg logit contributions | |
awaru | -0.110 |
ulia | -0.106 |
�� | -0.105 |
Combat | -0.104 |
uese | -0.103 |
Token | :: |
Token ID | 3712 |
Feature activation | -2.76e-01 |
Pos logit contributions | |
Instruct | +0.512 |
Anon | +0.476 |
Teg | +0.470 |
Luxem | +0.469 |
theless | +0.468 |
Neg logit contributions | |
awaru | -0.640 |
ulia | -0.611 |
�� | -0.610 |
uese | -0.592 |
vantage | -0.591 |
Token | Near |
Token ID | 40640 |
Feature activation | +7.68e-01 |
Pos logit contributions | |
awaru | +0.465 |
�� | +0.449 |
ulia | +0.436 |
Combat | +0.430 |
vantage | +0.427 |
Neg logit contributions | |
Instruct | -0.554 |
Anon | -0.525 |
Teg | -0.525 |
Luxem | -0.516 |
theless | -0.515 |
Token | } |
Token ID | 1782 |
Feature activation | -2.64e-01 |
Pos logit contributions | |
Instruct | +1.498 |
Anon | +1.409 |
Teg | +1.398 |
Luxem | +1.382 |
Tuls | +1.353 |
Neg logit contributions | |
awaru | -1.372 |
�� | -1.312 |
ulia | -1.276 |
uese | -1.238 |
vantage | -1.234 |
Token | } |
Token ID | 1782 |
Feature activation | -2.92e-01 |
Pos logit contributions | |
awaru | +0.413 |
�� | +0.394 |
ulia | +0.386 |
vantage | +0.372 |
uese | +0.370 |
Neg logit contributions | |
Instruct | -0.558 |
Anon | -0.531 |
Teg | -0.528 |
Luxem | -0.523 |
theless | -0.521 |
Token | if |
Token ID | 611 |
Feature activation | +3.61e-02 |
Pos logit contributions | |
awaru | +0.539 |
�� | +0.513 |
ulia | +0.510 |
vantage | +0.488 |
Combat | +0.485 |
Neg logit contributions | |
Instruct | -0.536 |
Anon | -0.510 |
Luxem | -0.503 |
Teg | -0.503 |
Tuls | -0.494 |
Token | ($ |
Token ID | 7198 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
Instruct | +0.054 |
Anon | +0.050 |
Teg | +0.050 |
Luxem | +0.049 |
theless | +0.048 |
Neg logit contributions | |
awaru | -0.080 |
�� | -0.077 |
ulia | -0.076 |
vantage | -0.074 |
uese | -0.074 |
Token | Use |
Token ID | 11041 |
Feature activation | +7.68e-02 |
Pos logit contributions | |
Instruct | +0.422 |
Anon | +0.399 |
Teg | +0.397 |
Luxem | +0.393 |
Tuls | +0.386 |
Neg logit contributions | |
awaru | -0.326 |
�� | -0.311 |
ulia | -0.302 |
vantage | -0.291 |
uese | -0.291 |
Token | ized |
Token ID | 1143 |
Feature activation | -2.62e-01 |
Pos logit contributions | |
awaru | +0.169 |
�� | +0.161 |
ulia | +0.160 |
incial | +0.156 |
vantage | +0.156 |
Neg logit contributions | |
Instruct | -0.125 |
Anon | -0.118 |
Teg | -0.114 |
Luxem | -0.114 |
theless | -0.112 |
Token | . |
Token ID | 13 |
Feature activation | +5.88e-02 |
Pos logit contributions | |
awaru | +0.564 |
�� | +0.544 |
ulia | +0.535 |
vantage | +0.524 |
rogen | +0.512 |
Neg logit contributions | |
Instruct | -0.413 |
Anon | -0.382 |
Teg | -0.379 |
Luxem | -0.375 |
Tuls | -0.369 |
Token | � |
Token ID | 447 |
Feature activation | +2.94e-02 |
Pos logit contributions | |
Instruct | +0.105 |
Anon | +0.098 |
Teg | +0.098 |
theless | +0.097 |
Luxem | +0.097 |
Neg logit contributions | |
awaru | -0.111 |
�� | -0.107 |
ulia | -0.106 |
vantage | -0.103 |
Combat | -0.102 |
Token | � |
Token ID | 251 |
Feature activation | +9.69e-02 |
Pos logit contributions | |
Instruct | +0.077 |
Anon | +0.074 |
Teg | +0.074 |
Luxem | +0.073 |
theless | +0.072 |
Neg logit contributions | |
awaru | -0.032 |
�� | -0.031 |
ulia | -0.029 |
vantage | -0.027 |
uese | -0.027 |
Token | Ev |
Token ID | 4319 |
Feature activation | +4.94e-02 |
Pos logit contributions | |
Instruct | +0.175 |
Teg | +0.163 |
Anon | +0.163 |
Luxem | +0.162 |
theless | +0.160 |
Neg logit contributions | |
awaru | -0.182 |
�� | -0.175 |
ulia | -0.175 |
vantage | -0.168 |
incial | -0.166 |
Token | idently |
Token ID | 46046 |
Feature activation | +8.62e-01 |
Pos logit contributions | |
Instruct | +0.075 |
Anon | +0.070 |
Teg | +0.068 |
Luxem | +0.068 |
Tuls | +0.066 |
Neg logit contributions | |
awaru | -0.110 |
ulia | -0.105 |
�� | -0.105 |
vantage | -0.101 |
uese | -0.101 |
Token | , |
Token ID | 11 |
Feature activation | +8.15e-02 |
Pos logit contributions | |
Instruct | +1.335 |
Anon | +1.229 |
Teg | +1.219 |
Luxem | +1.205 |
theless | +1.177 |
Neg logit contributions | |
awaru | -1.891 |
�� | -1.820 |
ulia | -1.789 |
vantage | -1.735 |
uese | -1.733 |
Token | we |
Token ID | 356 |
Feature activation | +5.87e-02 |
Pos logit contributions | |
Instruct | +0.142 |
Anon | +0.131 |
Teg | +0.130 |
Luxem | +0.130 |
theless | +0.128 |
Neg logit contributions | |
awaru | -0.162 |
�� | -0.155 |
ulia | -0.153 |
rogen | -0.147 |
vantage | -0.147 |
Token | need |
Token ID | 761 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
Instruct | +0.122 |
Anon | +0.114 |
Teg | +0.114 |
Luxem | +0.113 |
theless | +0.111 |
Neg logit contributions | |
awaru | -0.097 |
�� | -0.092 |
ulia | -0.091 |
rogen | -0.087 |
vantage | -0.087 |
Token | to |
Token ID | 284 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
Instruct | +0.195 |
Anon | +0.181 |
Teg | +0.180 |
Luxem | +0.179 |
theless | +0.175 |
Neg logit contributions | |
awaru | -0.219 |
�� | -0.211 |
ulia | -0.206 |
uese | -0.199 |
vantage | -0.199 |
Token | conduct |
Token ID | 3189 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
awaru | +0.042 |
�� | +0.040 |
ulia | +0.040 |
rogen | +0.039 |
Combat | +0.038 |
Neg logit contributions | |
Instruct | -0.036 |
Anon | -0.034 |
theless | -0.034 |
Teg | -0.034 |
Luxem | -0.033 |
Token | Tennessee |
Token ID | 11287 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
Instruct | +0.026 |
Anon | +0.024 |
Teg | +0.024 |
Luxem | +0.024 |
theless | +0.023 |
Neg logit contributions | |
awaru | -0.034 |
�� | -0.032 |
ulia | -0.032 |
rogen | -0.032 |
uese | -0.031 |
Token | .) |
Token ID | 2014 |
Feature activation | -4.17e-02 |
Pos logit contributions | |
Instruct | +0.019 |
Anon | +0.018 |
Teg | +0.018 |
Luxem | +0.018 |
theless | +0.018 |
Neg logit contributions | |
awaru | -0.018 |
�� | -0.017 |
ulia | -0.017 |
uese | -0.016 |
vantage | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
awaru | +0.078 |
�� | +0.075 |
ulia | +0.073 |
uese | +0.071 |
vantage | +0.071 |
Neg logit contributions | |
Instruct | -0.076 |
Anon | -0.072 |
Teg | -0.071 |
Luxem | -0.071 |
theless | -0.070 |
Token |
|
Token ID | 198 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
Instruct | +0.212 |
Anon | +0.197 |
Teg | +0.196 |
Luxem | +0.194 |
theless | +0.193 |
Neg logit contributions | |
awaru | -0.243 |
ulia | -0.232 |
�� | -0.231 |
Combat | -0.224 |
uese | -0.223 |
Token | We |
Token ID | 1135 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
Instruct | +0.234 |
Anon | +0.220 |
Teg | +0.219 |
Luxem | +0.219 |
theless | +0.215 |
Neg logit contributions | |
awaru | -0.201 |
�� | -0.193 |
ulia | -0.188 |
Combat | -0.185 |
rogen | -0.182 |
Token | rated |
Token ID | 13178 |
Feature activation | +7.83e-01 |
Pos logit contributions | |
Instruct | +0.080 |
Teg | +0.075 |
Luxem | +0.075 |
Anon | +0.075 |
Tuls | +0.073 |
Neg logit contributions | |
awaru | -0.070 |
�� | -0.066 |
ulia | -0.065 |
rogen | -0.063 |
vantage | -0.063 |
Token | the |
Token ID | 262 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
Instruct | +1.283 |
Anon | +1.192 |
Teg | +1.182 |
Luxem | +1.165 |
Tuls | +1.138 |
Neg logit contributions | |
awaru | -1.643 |
�� | -1.583 |
ulia | -1.546 |
uese | -1.506 |
vantage | -1.505 |
Token | coalition |
Token ID | 9906 |
Feature activation | -4.68e-01 |
Pos logit contributions | |
awaru | +0.031 |
�� | +0.030 |
ulia | +0.029 |
uese | +0.028 |
rogen | +0.028 |
Neg logit contributions | |
Instruct | -0.023 |
Anon | -0.022 |
Luxem | -0.021 |
Teg | -0.021 |
theless | -0.021 |
Token | � |
Token ID | 447 |
Feature activation | -9.05e-02 |
Pos logit contributions | |
awaru | +0.989 |
�� | +0.951 |
ulia | +0.922 |
rogen | +0.911 |
uese | +0.903 |
Neg logit contributions | |
Instruct | -0.746 |
Anon | -0.686 |
theless | -0.686 |
Teg | -0.683 |
Luxem | -0.680 |
Token | � |
Token ID | 247 |
Feature activation | -4.41e-03 |
Pos logit contributions | |
awaru | +0.129 |
�� | +0.128 |
ulia | +0.119 |
vantage | +0.115 |
incial | +0.113 |
Neg logit contributions | |
Instruct | -0.207 |
Teg | -0.197 |
Anon | -0.196 |
Luxem | -0.196 |
theless | -0.194 |
Token | s |
Token ID | 82 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
awaru | +0.009 |
�� | +0.008 |
ulia | +0.008 |
vantage | +0.008 |
uese | +0.008 |
Neg logit contributions | |
Instruct | -0.008 |
Teg | -0.007 |
Anon | -0.007 |
Luxem | -0.007 |
theless | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +4.97e-02 |
Pos logit contributions | |
Instruct | +0.047 |
Anon | +0.044 |
Teg | +0.044 |
theless | +0.043 |
Luxem | +0.043 |
Neg logit contributions | |
awaru | -0.052 |
�� | -0.049 |
ulia | -0.048 |
uese | -0.047 |
vantage | -0.047 |
Token | — |
Token ID | 851 |
Feature activation | -8.59e-02 |
Pos logit contributions | |
Instruct | +0.086 |
Teg | +0.080 |
theless | +0.080 |
Anon | +0.080 |
Luxem | +0.079 |
Neg logit contributions | |
awaru | -0.096 |
�� | -0.092 |
ulia | -0.091 |
Combat | -0.088 |
vantage | -0.088 |
Token | ACLU |
Token ID | 24381 |
Feature activation | -5.59e-02 |
Pos logit contributions | |
awaru | +0.135 |
�� | +0.128 |
ulia | +0.124 |
uese | +0.120 |
vantage | +0.120 |
Neg logit contributions | |
Instruct | -0.186 |
Anon | -0.176 |
Teg | -0.175 |
Luxem | -0.173 |
Tuls | -0.170 |
Token | of |
Token ID | 286 |
Feature activation | +2.63e-03 |
Pos logit contributions | |
awaru | +0.086 |
ulia | +0.081 |
�� | +0.080 |
Combat | +0.078 |
uese | +0.078 |
Neg logit contributions | |
Instruct | -0.120 |
Anon | -0.113 |
theless | -0.112 |
Teg | -0.112 |
Luxem | -0.112 |
Token | Oregon |
Token ID | 8819 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
Instruct | +0.004 |
theless | +0.004 |
behav | +0.004 |
Anon | +0.004 |
Teg | +0.004 |
Neg logit contributions | |
awaru | -0.006 |
ulia | -0.005 |
�� | -0.005 |
rogen | -0.005 |
vantage | -0.005 |
Token | (@ |
Token ID | 4275 |
Feature activation | +8.46e-01 |
Pos logit contributions | |
Instruct | +0.017 |
theless | +0.015 |
Teg | +0.014 |
Anon | +0.014 |
behav | +0.014 |
Neg logit contributions | |
awaru | -0.067 |
ulia | -0.066 |
�� | -0.065 |
Combat | -0.064 |
uese | -0.064 |
Token | AC |
Token ID | 2246 |
Feature activation | -6.09e-02 |
Pos logit contributions | |
Instruct | +1.977 |
Anon | +1.878 |
Teg | +1.866 |
Luxem | +1.841 |
Tuls | +1.815 |
Neg logit contributions | |
awaru | -1.184 |
�� | -1.132 |
ulia | -1.073 |
uese | -1.036 |
vantage | -1.032 |
Token | LU |
Token ID | 41596 |
Feature activation | +1.98e-01 |
Pos logit contributions | |
awaru | +0.120 |
�� | +0.116 |
ulia | +0.114 |
vantage | +0.110 |
uese | +0.110 |
Neg logit contributions | |
Instruct | -0.107 |
Anon | -0.100 |
Teg | -0.099 |
Luxem | -0.098 |
theless | -0.097 |
Token | _ |
Token ID | 62 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
Instruct | +0.379 |
Anon | +0.354 |
Teg | +0.352 |
Luxem | +0.350 |
theless | +0.346 |
Neg logit contributions | |
awaru | -0.357 |
�� | -0.339 |
ulia | -0.337 |
uese | -0.325 |
vantage | -0.325 |
Token | OR |
Token ID | 1581 |
Feature activation | +5.42e-02 |
Pos logit contributions | |
Instruct | +0.029 |
theless | +0.027 |
Teg | +0.027 |
Luxem | +0.027 |
Anon | +0.027 |
Neg logit contributions | |
awaru | -0.036 |
�� | -0.034 |
ulia | -0.034 |
Combat | -0.033 |
UFC | -0.033 |
Token | ) |
Token ID | 8 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
Instruct | +0.100 |
Anon | +0.094 |
Teg | +0.093 |
Luxem | +0.092 |
theless | +0.092 |
Neg logit contributions | |
awaru | -0.101 |
ulia | -0.096 |
�� | -0.096 |
uese | -0.092 |
vantage | -0.092 |
Token | 21 |
Token ID | 2310 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
awaru | +0.212 |
�� | +0.201 |
ulia | +0.201 |
rogen | +0.197 |
vantage | +0.193 |
Neg logit contributions | |
Instruct | -0.196 |
Anon | -0.184 |
Teg | -0.180 |
Luxem | -0.177 |
Tuls | -0.176 |
Token | . |
Token ID | 13 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
Instruct | +0.039 |
Anon | +0.037 |
Teg | +0.036 |
Luxem | +0.036 |
theless | +0.035 |
Neg logit contributions | |
awaru | -0.042 |
�� | -0.040 |
ulia | -0.039 |
uese | -0.038 |
vantage | -0.038 |
Token | 0 |
Token ID | 15 |
Feature activation | +8.62e-02 |
Pos logit contributions | |
awaru | +0.032 |
�� | +0.031 |
ulia | +0.030 |
vantage | +0.029 |
Combat | +0.029 |
Neg logit contributions | |
Instruct | -0.045 |
Anon | -0.043 |
Teg | -0.043 |
Luxem | -0.043 |
behav | -0.042 |
Token | degrees |
Token ID | 7370 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
Instruct | +0.157 |
Anon | +0.147 |
Teg | +0.146 |
Luxem | +0.145 |
Tuls | +0.141 |
Neg logit contributions | |
awaru | -0.164 |
�� | -0.158 |
ulia | -0.154 |
uese | -0.149 |
vantage | -0.149 |
Token | east |
Token ID | 7627 |
Feature activation | -3.43e-02 |
Pos logit contributions | |
Instruct | +0.235 |
Anon | +0.220 |
Teg | +0.216 |
Luxem | +0.212 |
Tuls | +0.209 |
Neg logit contributions | |
awaru | -0.303 |
�� | -0.289 |
ulia | -0.287 |
rogen | -0.279 |
vantage | -0.277 |
Token | in |
Token ID | 287 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
awaru | +0.064 |
�� | +0.061 |
ulia | +0.060 |
vantage | +0.058 |
rogen | +0.058 |
Neg logit contributions | |
Instruct | -0.064 |
Anon | -0.060 |
Teg | -0.060 |
Luxem | -0.059 |
theless | -0.058 |
Token | Mare |
Token ID | 36989 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
awaru | +0.024 |
�� | +0.023 |
ulia | +0.023 |
vantage | +0.022 |
rogen | +0.022 |
Neg logit contributions | |
Instruct | -0.023 |
Anon | -0.022 |
Teg | -0.021 |
Luxem | -0.021 |
theless | -0.021 |
Token | Tr |
Token ID | 833 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
Instruct | +0.176 |
Anon | +0.164 |
Teg | +0.162 |
theless | +0.160 |
Luxem | +0.160 |
Neg logit contributions | |
awaru | -0.205 |
�� | -0.198 |
ulia | -0.198 |
vantage | -0.189 |
rogen | -0.188 |
Token | anqu |
Token ID | 26184 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
awaru | +0.295 |
ulia | +0.278 |
�� | +0.273 |
uese | +0.262 |
vantage | +0.259 |
Neg logit contributions | |
Instruct | -0.482 |
Anon | -0.458 |
Teg | -0.454 |
Luxem | -0.451 |
theless | -0.447 |
Token | ill |
Token ID | 359 |
Feature activation | -4.93e-02 |
Pos logit contributions | |
awaru | +0.354 |
�� | +0.338 |
ulia | +0.336 |
vantage | +0.323 |
uese | +0.323 |
Neg logit contributions | |
Instruct | -0.378 |
Anon | -0.357 |
Teg | -0.357 |
Luxem | -0.350 |
Tuls | -0.347 |
Token | it |
Token ID | 270 |
Feature activation | -3.74e-01 |
Pos logit contributions | |
awaru | +0.116 |
�� | +0.113 |
ulia | +0.110 |
uese | +0.108 |
vantage | +0.108 |
Neg logit contributions | |
Instruct | -0.068 |
Anon | -0.062 |
theless | -0.061 |
Teg | -0.061 |
Luxem | -0.061 |
Token | years |
Token ID | 812 |
Feature activation | +6.29e-03 |
Pos logit contributions | |
Instruct | +0.154 |
Luxem | +0.145 |
Anon | +0.145 |
Teg | +0.145 |
Tuls | +0.142 |
Neg logit contributions | |
awaru | -0.170 |
�� | -0.166 |
ulia | -0.163 |
uese | -0.157 |
incial | -0.156 |
Token | programming |
Token ID | 8300 |
Feature activation | +2.97e-01 |
Pos logit contributions | |
Instruct | +0.012 |
Anon | +0.011 |
Luxem | +0.011 |
Teg | +0.011 |
theless | +0.011 |
Neg logit contributions | |
awaru | -0.011 |
�� | -0.011 |
ulia | -0.010 |
uese | -0.010 |
vantage | -0.010 |
Token | 8 |
Token ID | 807 |
Feature activation | +3.31e-02 |
Pos logit contributions | |
Instruct | +0.492 |
Luxem | +0.460 |
Anon | +0.458 |
Teg | +0.454 |
Tuls | +0.448 |
Neg logit contributions | |
awaru | -0.604 |
�� | -0.580 |
ulia | -0.571 |
uese | -0.555 |
rogen | -0.550 |
Token | bit |
Token ID | 1643 |
Feature activation | +7.41e-05 |
Pos logit contributions | |
Instruct | +0.059 |
Anon | +0.055 |
Luxem | +0.055 |
Teg | +0.055 |
Tuls | +0.054 |
Neg logit contributions | |
awaru | -0.063 |
�� | -0.062 |
ulia | -0.060 |
uese | -0.058 |
vantage | -0.058 |
Token | MC |
Token ID | 13122 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
Instruct | +0.000 |
Luxem | +0.000 |
Anon | +0.000 |
Tuls | +0.000 |
theless | +0.000 |
Neg logit contributions | |
awaru | -0.000 |
�� | -0.000 |
ulia | -0.000 |
uese | -0.000 |
rogen | -0.000 |
Token | Us |
Token ID | 5842 |
Feature activation | +4.15e-01 |
Pos logit contributions | |
Instruct | +0.266 |
Anon | +0.251 |
Luxem | +0.250 |
Teg | +0.249 |
theless | +0.247 |
Neg logit contributions | |
awaru | -0.252 |
�� | -0.245 |
ulia | -0.241 |
vantage | -0.232 |
rogen | -0.231 |
Token | with |
Token ID | 351 |
Feature activation | +9.30e-02 |
Pos logit contributions | |
Instruct | +0.663 |
Anon | +0.613 |
Teg | +0.608 |
Luxem | +0.604 |
theless | +0.589 |
Neg logit contributions | |
awaru | -0.884 |
�� | -0.852 |
ulia | -0.838 |
uese | -0.816 |
vantage | -0.812 |
Token | fairly |
Token ID | 6547 |
Feature activation | +2.45e-01 |
Pos logit contributions | |
Instruct | +0.147 |
Luxem | +0.136 |
Anon | +0.135 |
theless | +0.133 |
Tuls | +0.132 |
Neg logit contributions | |
awaru | -0.195 |
�� | -0.187 |
ulia | -0.184 |
rogen | -0.179 |
vantage | -0.179 |
Token | simple |
Token ID | 2829 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
Instruct | +0.434 |
Anon | +0.402 |
Luxem | +0.397 |
Teg | +0.397 |
theless | +0.391 |
Neg logit contributions | |
awaru | -0.478 |
�� | -0.457 |
ulia | -0.446 |
rogen | -0.437 |
uese | -0.435 |
Token | peripher |
Token ID | 18375 |
Feature activation | +6.57e-02 |
Pos logit contributions | |
Instruct | +0.398 |
Anon | +0.370 |
Luxem | +0.369 |
theless | +0.363 |
Tuls | +0.360 |
Neg logit contributions | |
awaru | -0.509 |
�� | -0.490 |
ulia | -0.482 |
rogen | -0.472 |
uese | -0.466 |
Token | als |
Token ID | 874 |
Feature activation | +3.99e-02 |
Pos logit contributions | |
Instruct | +0.195 |
Anon | +0.189 |
Luxem | +0.188 |
Teg | +0.187 |
theless | +0.186 |
Neg logit contributions | |
awaru | -0.049 |
ulia | -0.042 |
�� | -0.042 |
uese | -0.038 |
vantage | -0.038 |
Token | not |
Token ID | 407 |
Feature activation | +6.72e-02 |
Pos logit contributions | |
Instruct | +0.194 |
Anon | +0.182 |
Luxem | +0.181 |
theless | +0.181 |
Teg | +0.181 |
Neg logit contributions | |
awaru | -0.232 |
�� | -0.221 |
ulia | -0.216 |
rogen | -0.216 |
incial | -0.212 |
Token | qualified |
Token ID | 10617 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
Instruct | +0.110 |
Anon | +0.104 |
Luxem | +0.103 |
Teg | +0.102 |
theless | +0.102 |
Neg logit contributions | |
awaru | -0.136 |
�� | -0.129 |
rogen | -0.126 |
ulia | -0.126 |
uese | -0.124 |
Token | to |
Token ID | 284 |
Feature activation | +3.87e-02 |
Pos logit contributions | |
Instruct | +0.281 |
Anon | +0.263 |
Teg | +0.262 |
Luxem | +0.257 |
theless | +0.253 |
Neg logit contributions | |
awaru | -0.332 |
�� | -0.317 |
ulia | -0.311 |
uese | -0.302 |
vantage | -0.301 |
Token | make |
Token ID | 787 |
Feature activation | +8.49e-03 |
Pos logit contributions | |
Instruct | +0.065 |
Anon | +0.062 |
Tuls | +0.061 |
theless | +0.061 |
Teg | +0.060 |
Neg logit contributions | |
awaru | -0.076 |
�� | -0.073 |
ulia | -0.072 |
rogen | -0.071 |
inate | -0.071 |
Token | clinical |
Token ID | 8668 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
Instruct | +0.015 |
Anon | +0.014 |
Teg | +0.014 |
Luxem | +0.014 |
theless | +0.014 |
Neg logit contributions | |
awaru | -0.017 |
�� | -0.016 |
ulia | -0.016 |
uese | -0.015 |
incial | -0.015 |
Token | decisions |
Token ID | 5370 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
Instruct | +0.332 |
Anon | +0.313 |
Teg | +0.309 |
Luxem | +0.308 |
Tuls | +0.302 |
Neg logit contributions | |
awaru | -0.283 |
�� | -0.271 |
ulia | -0.264 |
uese | -0.255 |
incial | -0.253 |
Token | , |
Token ID | 11 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
Instruct | +0.276 |
Anon | +0.257 |
Teg | +0.256 |
Luxem | +0.252 |
Tuls | +0.249 |
Neg logit contributions | |
awaru | -0.332 |
�� | -0.316 |
ulia | -0.310 |
vantage | -0.302 |
uese | -0.301 |
Token | on |
Token ID | 319 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
awaru | +0.045 |
�� | +0.043 |
ulia | +0.042 |
uese | +0.041 |
incial | +0.041 |
Neg logit contributions | |
Instruct | -0.035 |
Anon | -0.033 |
Teg | -0.033 |
Luxem | -0.033 |
theless | -0.032 |
Token | November |
Token ID | 3389 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
Instruct | +0.019 |
Anon | +0.018 |
Teg | +0.018 |
Luxem | +0.018 |
theless | +0.017 |
Neg logit contributions | |
awaru | -0.021 |
ulia | -0.020 |
�� | -0.020 |
incial | -0.019 |
rogen | -0.019 |
Token | 17 |
Token ID | 1596 |
Feature activation | +2.63e-01 |
Pos logit contributions | |
Instruct | +0.056 |
Anon | +0.053 |
Teg | +0.052 |
behav | +0.052 |
theless | +0.052 |
Neg logit contributions | |
awaru | -0.050 |
ulia | -0.049 |
�� | -0.048 |
Combat | -0.047 |
rogen | -0.047 |
Token | . |
Token ID | 13 |
Feature activation | +7.75e-02 |
Pos logit contributions | |
Instruct | +0.420 |
Anon | +0.392 |
Teg | +0.386 |
Luxem | +0.383 |
Tuls | +0.374 |
Neg logit contributions | |
awaru | -0.561 |
�� | -0.541 |
ulia | -0.531 |
uese | -0.517 |
vantage | -0.516 |
Token | � |
Token ID | 247 |
Feature activation | -1.61e-02 |
Pos logit contributions | |
awaru | +0.070 |
�� | +0.069 |
ulia | +0.062 |
vantage | +0.058 |
uese | +0.058 |
Neg logit contributions | |
Instruct | -0.186 |
Anon | -0.179 |
Teg | -0.178 |
Luxem | -0.177 |
theless | -0.175 |
Token | d |
Token ID | 67 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
awaru | +0.029 |
�� | +0.028 |
ulia | +0.027 |
uese | +0.026 |
vantage | +0.026 |
Neg logit contributions | |
Instruct | -0.031 |
Anon | -0.029 |
Teg | -0.029 |
Luxem | -0.029 |
Tuls | -0.028 |
Token | been |
Token ID | 587 |
Feature activation | +3.10e-02 |
Pos logit contributions | |
Instruct | +0.214 |
Anon | +0.198 |
Teg | +0.196 |
Luxem | +0.194 |
Tuls | +0.191 |
Neg logit contributions | |
awaru | -0.266 |
�� | -0.258 |
ulia | -0.253 |
vantage | -0.246 |
uese | -0.245 |
Token | no |
Token ID | 645 |
Feature activation | +5.26e-02 |
Pos logit contributions | |
Instruct | +0.058 |
Anon | +0.054 |
Luxem | +0.054 |
Teg | +0.054 |
Tuls | +0.053 |
Neg logit contributions | |
awaru | -0.057 |
�� | -0.055 |
ulia | -0.053 |
rogen | -0.052 |
uese | -0.051 |
Token | indication |
Token ID | 12955 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
Instruct | +0.093 |
Anon | +0.087 |
Teg | +0.086 |
Luxem | +0.086 |
theless | +0.084 |
Neg logit contributions | |
awaru | -0.101 |
�� | -0.098 |
ulia | -0.096 |
rogen | -0.093 |
uese | -0.093 |
Token | to |
Token ID | 284 |
Feature activation | +3.58e-02 |
Pos logit contributions | |
Instruct | +0.220 |
Anon | +0.205 |
Teg | +0.203 |
Luxem | +0.200 |
theless | +0.196 |
Neg logit contributions | |
awaru | -0.253 |
�� | -0.243 |
ulia | -0.238 |
uese | -0.231 |
rogen | -0.231 |
Token | this |
Token ID | 428 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
Instruct | +0.065 |
Anon | +0.061 |
Teg | +0.061 |
Luxem | +0.061 |
Tuls | +0.059 |
Neg logit contributions | |
awaru | -0.067 |
�� | -0.064 |
ulia | -0.063 |
uese | -0.061 |
incial | -0.061 |
Token | point |
Token ID | 966 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
awaru | +0.454 |
�� | +0.438 |
ulia | +0.422 |
uese | +0.415 |
vantage | +0.415 |
Neg logit contributions | |
Instruct | -0.389 |
Anon | -0.363 |
Teg | -0.362 |
Luxem | -0.359 |
theless | -0.353 |
Token | that |
Token ID | 326 |
Feature activation | -6.53e-02 |
Pos logit contributions | |
Instruct | +0.217 |
Anon | +0.202 |
Teg | +0.200 |
Luxem | +0.197 |
theless | +0.193 |
Neg logit contributions | |
awaru | -0.274 |
�� | -0.265 |
ulia | -0.258 |
uese | -0.252 |
vantage | -0.251 |
Token | Vol |
Token ID | 4709 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
awaru | +0.133 |
�� | +0.130 |
ulia | +0.127 |
uese | +0.123 |
rogen | +0.122 |
Neg logit contributions | |
Instruct | -0.108 |
Anon | -0.101 |
Teg | -0.099 |
Luxem | -0.099 |
Tuls | -0.096 |
Token | quez |
Token ID | 22281 |
Feature activation | -8.48e-02 |
Pos logit contributions | |
awaru | +0.107 |
ulia | +0.093 |
�� | +0.092 |
uese | +0.083 |
vantage | +0.083 |
Neg logit contributions | |
Instruct | -0.427 |
Anon | -0.411 |
Teg | -0.410 |
Luxem | -0.406 |
theless | -0.403 |
Token | f |
Token ID | 277 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
awaru | +0.050 |
�� | +0.048 |
ulia | +0.047 |
vantage | +0.046 |
uese | +0.045 |
Neg logit contributions | |
Instruct | -0.044 |
Anon | -0.041 |
Teg | -0.041 |
Luxem | -0.040 |
theless | -0.040 |
Token | / |
Token ID | 14 |
Feature activation | +4.22e-02 |
Pos logit contributions | |
awaru | +0.249 |
ulia | +0.240 |
�� | +0.236 |
vantage | +0.235 |
rogen | +0.231 |
Neg logit contributions | |
Instruct | -0.240 |
Anon | -0.221 |
Teg | -0.221 |
theless | -0.220 |
Luxem | -0.219 |
Token | 2 |
Token ID | 17 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
Instruct | +0.078 |
Anon | +0.071 |
Teg | +0.071 |
Luxem | +0.071 |
theless | +0.070 |
Neg logit contributions | |
awaru | -0.080 |
�� | -0.076 |
ulia | -0.076 |
vantage | -0.075 |
incial | -0.073 |
Token | . |
Token ID | 13 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
awaru | +0.245 |
�� | +0.237 |
ulia | +0.233 |
vantage | +0.229 |
incial | +0.224 |
Neg logit contributions | |
Instruct | -0.204 |
Teg | -0.189 |
Anon | -0.188 |
Luxem | -0.186 |
theless | -0.185 |
Token | 8 |
Token ID | 23 |
Feature activation | -4.39e-02 |
Pos logit contributions | |
Instruct | +0.199 |
theless | +0.188 |
Anon | +0.188 |
behav | +0.187 |
Teg | +0.186 |
Neg logit contributions | |
awaru | -0.168 |
�� | -0.157 |
ulia | -0.156 |
incial | -0.154 |
vantage | -0.153 |
Token | , |
Token ID | 11 |
Feature activation | +8.36e-02 |
Pos logit contributions | |
awaru | +0.091 |
�� | +0.088 |
ulia | +0.086 |
vantage | +0.085 |
rogen | +0.083 |
Neg logit contributions | |
Instruct | -0.072 |
Anon | -0.066 |
Teg | -0.066 |
Luxem | -0.065 |
theless | -0.064 |
Token |
|
Token ID | 198 |
Feature activation | -1.68e-03 |
Pos logit contributions | |
Instruct | +0.135 |
Anon | +0.125 |
Teg | +0.124 |
Luxem | +0.122 |
theless | +0.119 |
Neg logit contributions | |
awaru | -0.178 |
�� | -0.172 |
ulia | -0.167 |
uese | -0.163 |
vantage | -0.162 |
Token |
|
Token ID | 198 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
awaru | +0.003 |
ulia | +0.003 |
�� | +0.003 |
Combat | +0.003 |
rogen | +0.003 |
Neg logit contributions | |
Instruct | -0.003 |
Anon | -0.003 |
Teg | -0.003 |
Luxem | -0.003 |
theless | -0.003 |
Token | Four |
Token ID | 15137 |
Feature activation | +1.88e-01 |
Pos logit contributions | |
awaru | +0.028 |
�� | +0.026 |
ulia | +0.026 |
vantage | +0.025 |
Combat | +0.025 |
Neg logit contributions | |
Instruct | -0.036 |
Anon | -0.034 |
Teg | -0.033 |
Luxem | -0.033 |
Tuls | -0.033 |
Token | corner |
Token ID | 5228 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
Instruct | +0.260 |
Teg | +0.234 |
Anon | +0.233 |
Luxem | +0.229 |
Tuls | +0.226 |
Neg logit contributions | |
awaru | -0.439 |
ulia | -0.418 |
�� | -0.417 |
vantage | -0.409 |
uese | -0.407 |
Token | pin |
Token ID | 6757 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
Instruct | +0.025 |
Anon | +0.023 |
Teg | +0.023 |
Luxem | +0.023 |
Tuls | +0.022 |
Neg logit contributions | |
awaru | -0.043 |
�� | -0.042 |
ulia | -0.041 |
uese | -0.040 |
vantage | -0.040 |
Token | � |
Token ID | 250 |
Feature activation | +7.42e-02 |
Pos logit contributions | |
awaru | +0.055 |
�� | +0.054 |
ulia | +0.050 |
vantage | +0.047 |
Combat | +0.046 |
Neg logit contributions | |
Instruct | -0.119 |
Anon | -0.114 |
Teg | -0.114 |
Luxem | -0.114 |
theless | -0.112 |
Token | Mr |
Token ID | 5246 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
Instruct | +0.150 |
Anon | +0.142 |
Teg | +0.142 |
Luxem | +0.140 |
Tuls | +0.137 |
Neg logit contributions | |
awaru | -0.126 |
�� | -0.120 |
ulia | -0.118 |
vantage | -0.114 |
Combat | -0.113 |
Token | . |
Token ID | 13 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
awaru | +0.401 |
�� | +0.383 |
ulia | +0.382 |
Combat | +0.370 |
vantage | +0.370 |
Neg logit contributions | |
Instruct | -0.344 |
Anon | -0.320 |
theless | -0.319 |
Luxem | -0.318 |
Teg | -0.318 |
Token | Pruitt |
Token ID | 39637 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
awaru | +0.473 |
�� | +0.456 |
ulia | +0.446 |
uese | +0.436 |
vantage | +0.435 |
Neg logit contributions | |
Instruct | -0.321 |
Anon | -0.296 |
Teg | -0.294 |
Luxem | -0.288 |
Tuls | -0.283 |
Token | � |
Token ID | 447 |
Feature activation | -7.55e-02 |
Pos logit contributions | |
awaru | +0.367 |
�� | +0.352 |
ulia | +0.343 |
vantage | +0.333 |
rogen | +0.332 |
Neg logit contributions | |
Instruct | -0.312 |
Anon | -0.293 |
Teg | -0.293 |
Luxem | -0.290 |
theless | -0.281 |
Token | � |
Token ID | 247 |
Feature activation | -5.38e-02 |
Pos logit contributions | |
awaru | +0.072 |
�� | +0.068 |
ulia | +0.063 |
uese | +0.059 |
vantage | +0.059 |
Neg logit contributions | |
Instruct | -0.210 |
Anon | -0.201 |
Teg | -0.201 |
Luxem | -0.200 |
theless | -0.197 |
Token | s |
Token ID | 82 |
Feature activation | -5.79e-02 |
Pos logit contributions | |
awaru | +0.096 |
�� | +0.092 |
ulia | +0.090 |
vantage | +0.087 |
uese | +0.087 |
Neg logit contributions | |
Instruct | -0.104 |
Anon | -0.099 |
Teg | -0.098 |
Luxem | -0.097 |
Tuls | -0.095 |
Token | record |
Token ID | 1700 |
Feature activation | +5.88e-02 |
Pos logit contributions | |
awaru | +0.104 |
�� | +0.100 |
ulia | +0.097 |
vantage | +0.094 |
rogen | +0.094 |
Neg logit contributions | |
Instruct | -0.111 |
Anon | -0.105 |
Teg | -0.104 |
Luxem | -0.104 |
theless | -0.100 |
Token | and |
Token ID | 290 |
Feature activation | -2.17e-02 |
Pos logit contributions | |
Instruct | +0.113 |
Anon | +0.107 |
Teg | +0.106 |
Luxem | +0.105 |
Tuls | +0.102 |
Neg logit contributions | |
awaru | -0.106 |
�� | -0.101 |
ulia | -0.099 |
vantage | -0.096 |
uese | -0.095 |
Token | public |
Token ID | 1171 |
Feature activation | -2.37e-02 |
Pos logit contributions | |
awaru | +0.044 |
�� | +0.043 |
ulia | +0.042 |
rogen | +0.041 |
vantage | +0.040 |
Neg logit contributions | |
Instruct | -0.036 |
Anon | -0.034 |
Teg | -0.034 |
Luxem | -0.034 |
theless | -0.032 |
Token | statements |
Token ID | 6299 |
Feature activation | +8.17e-02 |
Pos logit contributions | |
awaru | +0.049 |
�� | +0.048 |
ulia | +0.046 |
incial | +0.045 |
rogen | +0.045 |
Neg logit contributions | |
Instruct | -0.039 |
Anon | -0.036 |
Teg | -0.036 |
Luxem | -0.036 |
Tuls | -0.034 |
Token | He |
Token ID | 679 |
Feature activation | -5.25e-02 |
Pos logit contributions | |
awaru | +0.139 |
�� | +0.134 |
ulia | +0.134 |
Combat | +0.130 |
vantage | +0.129 |
Neg logit contributions | |
Instruct | -0.169 |
theless | -0.161 |
Anon | -0.160 |
behav | -0.158 |
Luxem | -0.158 |
Token | just |
Token ID | 655 |
Feature activation | -8.94e-02 |
Pos logit contributions | |
awaru | +0.092 |
�� | +0.088 |
ulia | +0.087 |
rogen | +0.085 |
uese | +0.084 |
Neg logit contributions | |
Instruct | -0.102 |
Anon | -0.097 |
Teg | -0.094 |
Luxem | -0.094 |
Tuls | -0.092 |
Token | couldn |
Token ID | 3521 |
Feature activation | -2.34e-02 |
Pos logit contributions | |
awaru | +0.149 |
�� | +0.142 |
ulia | +0.140 |
rogen | +0.137 |
inate | +0.134 |
Neg logit contributions | |
Instruct | -0.181 |
Anon | -0.171 |
Teg | -0.168 |
Luxem | -0.168 |
Tuls | -0.163 |
Token | 't |
Token ID | 470 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
awaru | +0.043 |
ulia | +0.041 |
�� | +0.041 |
Combat | +0.040 |
rogen | +0.040 |
Neg logit contributions | |
Instruct | -0.043 |
Anon | -0.040 |
theless | -0.040 |
Teg | -0.040 |
Luxem | -0.040 |
Token | handle |
Token ID | 5412 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
awaru | +0.246 |
ulia | +0.232 |
�� | +0.232 |
inate | +0.226 |
rogen | +0.225 |
Neg logit contributions | |
Instruct | -0.228 |
Anon | -0.221 |
Teg | -0.216 |
Luxem | -0.215 |
theless | -0.215 |
Token | the |
Token ID | 262 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
awaru | +0.292 |
�� | +0.283 |
ulia | +0.279 |
uese | +0.272 |
rogen | +0.271 |
Neg logit contributions | |
Instruct | -0.228 |
Anon | -0.213 |
theless | -0.210 |
Luxem | -0.207 |
Teg | -0.206 |
Token | truth |
Token ID | 3872 |
Feature activation | +5.48e-02 |
Pos logit contributions | |
awaru | +0.215 |
�� | +0.209 |
ulia | +0.205 |
uese | +0.200 |
rogen | +0.199 |
Neg logit contributions | |
Instruct | -0.203 |
Anon | -0.192 |
theless | -0.187 |
Luxem | -0.187 |
Teg | -0.184 |
Token | . |
Token ID | 13 |
Feature activation | -7.53e-02 |
Pos logit contributions | |
Instruct | +0.096 |
Anon | +0.090 |
Luxem | +0.088 |
Teg | +0.088 |
theless | +0.086 |
Neg logit contributions | |
awaru | -0.108 |
�� | -0.103 |
ulia | -0.101 |
vantage | -0.099 |
rogen | -0.098 |
Token | Those |
Token ID | 5845 |
Feature activation | +7.96e-02 |
Pos logit contributions | |
awaru | +0.127 |
ulia | +0.122 |
�� | +0.122 |
Combat | +0.118 |
vantage | +0.117 |
Neg logit contributions | |
Instruct | -0.147 |
theless | -0.139 |
Anon | -0.139 |
behav | -0.137 |
Luxem | -0.137 |
Token | were |
Token ID | 547 |
Feature activation | -2.30e-01 |
Pos logit contributions | |
Instruct | +0.171 |
Anon | +0.160 |
Teg | +0.158 |
Luxem | +0.158 |
theless | +0.156 |
Neg logit contributions | |
awaru | -0.124 |
�� | -0.119 |
ulia | -0.117 |
uese | -0.113 |
rogen | -0.112 |
Token | facts |
Token ID | 6419 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
awaru | +0.445 |
�� | +0.432 |
ulia | +0.427 |
rogen | +0.416 |
incial | +0.408 |
Neg logit contributions | |
Instruct | -0.400 |
Anon | -0.373 |
Luxem | -0.368 |
theless | -0.367 |
Teg | -0.359 |
Token | one |
Token ID | 530 |
Feature activation | -4.22e-02 |
Pos logit contributions | |
Instruct | +0.176 |
Anon | +0.166 |
Luxem | +0.166 |
Teg | +0.165 |
Tuls | +0.160 |
Neg logit contributions | |
awaru | -0.195 |
�� | -0.186 |
ulia | -0.184 |
rogen | -0.179 |
uese | -0.177 |
Token | 's |
Token ID | 338 |
Feature activation | -6.10e-02 |
Pos logit contributions | |
awaru | +0.098 |
�� | +0.094 |
ulia | +0.093 |
uese | +0.091 |
vantage | +0.090 |
Neg logit contributions | |
Instruct | -0.059 |
Anon | -0.054 |
Teg | -0.054 |
Luxem | -0.053 |
Tuls | -0.052 |
Token | accuser |
Token ID | 43738 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
awaru | +0.122 |
�� | +0.116 |
ulia | +0.115 |
rogen | +0.113 |
Combat | +0.111 |
Neg logit contributions | |
Instruct | -0.103 |
Anon | -0.097 |
Luxem | -0.097 |
Teg | -0.096 |
theless | -0.094 |
Token | is |
Token ID | 318 |
Feature activation | -9.07e-02 |
Pos logit contributions | |
awaru | +0.398 |
�� | +0.372 |
ulia | +0.371 |
rogen | +0.362 |
vantage | +0.353 |
Neg logit contributions | |
Instruct | -0.427 |
Anon | -0.410 |
Luxem | -0.402 |
Teg | -0.401 |
Tuls | -0.388 |
Token | important |
Token ID | 1593 |
Feature activation | -1.50e-02 |
Pos logit contributions | |
awaru | +0.183 |
�� | +0.175 |
rogen | +0.170 |
ulia | +0.169 |
incial | +0.165 |
Neg logit contributions | |
Instruct | -0.153 |
Anon | -0.144 |
Luxem | -0.140 |
Teg | -0.139 |
theless | -0.139 |
Token | not |
Token ID | 407 |
Feature activation | -5.43e-02 |
Pos logit contributions | |
awaru | +0.029 |
�� | +0.028 |
ulia | +0.027 |
rogen | +0.027 |
vantage | +0.026 |
Neg logit contributions | |
Instruct | -0.026 |
Anon | -0.025 |
Teg | -0.024 |
Luxem | -0.024 |
theless | -0.024 |
Token | merely |
Token ID | 6974 |
Feature activation | +8.15e-02 |
Pos logit contributions | |
awaru | +0.124 |
�� | +0.119 |
ulia | +0.116 |
rogen | +0.114 |
vantage | +0.113 |
Neg logit contributions | |
Instruct | -0.080 |
Anon | -0.074 |
Luxem | -0.072 |
Teg | -0.072 |
Tuls | -0.069 |
Token | because |
Token ID | 780 |
Feature activation | -6.83e-02 |
Pos logit contributions | |
Instruct | +0.150 |
Anon | +0.140 |
Teg | +0.138 |
Luxem | +0.137 |
Tuls | +0.134 |
Neg logit contributions | |
awaru | -0.156 |
�� | -0.148 |
ulia | -0.145 |
rogen | -0.141 |
vantage | -0.140 |
Token | it |
Token ID | 340 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
awaru | +0.131 |
�� | +0.126 |
ulia | +0.124 |
rogen | +0.122 |
uese | +0.119 |
Neg logit contributions | |
Instruct | -0.120 |
Anon | -0.114 |
Luxem | -0.112 |
Teg | -0.111 |
theless | -0.108 |
Token | aids |
Token ID | 31378 |
Feature activation | -4.01e-02 |
Pos logit contributions | |
awaru | +0.180 |
�� | +0.168 |
rogen | +0.163 |
ulia | +0.162 |
vantage | +0.157 |
Neg logit contributions | |
Instruct | -0.237 |
Anon | -0.227 |
Teg | -0.224 |
Luxem | -0.221 |
Tuls | -0.217 |
Token | in |
Token ID | 287 |
Feature activation | -9.70e-02 |
Pos logit contributions | |
awaru | +0.075 |
�� | +0.071 |
ulia | +0.070 |
rogen | +0.068 |
vantage | +0.067 |
Neg logit contributions | |
Instruct | -0.073 |
Anon | -0.070 |
Luxem | -0.069 |
Teg | -0.068 |
theless | -0.067 |
Token | in |
Token ID | 287 |
Feature activation | +5.19e-03 |
Pos logit contributions | |
Instruct | +0.409 |
Anon | +0.388 |
Teg | +0.383 |
Luxem | +0.382 |
theless | +0.381 |
Neg logit contributions | |
awaru | -0.351 |
�� | -0.330 |
ulia | -0.329 |
rogen | -0.319 |
incial | -0.316 |
Token | Maryland |
Token ID | 10769 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
Instruct | +0.008 |
Anon | +0.008 |
theless | +0.008 |
Teg | +0.008 |
Luxem | +0.008 |
Neg logit contributions | |
awaru | -0.011 |
�� | -0.010 |
ulia | -0.010 |
rogen | -0.010 |
Combat | -0.010 |
Token | do |
Token ID | 466 |
Feature activation | +8.72e-02 |
Pos logit contributions | |
Instruct | +0.360 |
Anon | +0.342 |
Teg | +0.340 |
Luxem | +0.335 |
theless | +0.334 |
Neg logit contributions | |
awaru | -0.301 |
�� | -0.282 |
ulia | -0.281 |
rogen | -0.272 |
incial | -0.271 |
Token | not |
Token ID | 407 |
Feature activation | +1.35e-02 |
Pos logit contributions | |
Instruct | +0.130 |
Anon | +0.119 |
Teg | +0.118 |
Luxem | +0.117 |
theless | +0.114 |
Neg logit contributions | |
awaru | -0.196 |
�� | -0.189 |
ulia | -0.185 |
vantage | -0.181 |
uese | -0.180 |
Token | need |
Token ID | 761 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
Instruct | +0.028 |
Anon | +0.027 |
Teg | +0.027 |
Luxem | +0.026 |
theless | +0.026 |
Neg logit contributions | |
awaru | -0.021 |
�� | -0.020 |
ulia | -0.020 |
rogen | -0.020 |
inate | -0.019 |
Token | their |
Token ID | 511 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
Instruct | +0.223 |
Anon | +0.207 |
Teg | +0.206 |
Luxem | +0.202 |
Tuls | +0.197 |
Neg logit contributions | |
awaru | -0.260 |
�� | -0.250 |
ulia | -0.245 |
vantage | -0.238 |
uese | -0.238 |
Token | parents |
Token ID | 3397 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
awaru | +0.235 |
�� | +0.225 |
ulia | +0.219 |
rogen | +0.214 |
uese | +0.210 |
Neg logit contributions | |
Instruct | -0.245 |
theless | -0.231 |
Anon | -0.231 |
Teg | -0.231 |
Luxem | -0.230 |
Token | ' |
Token ID | 6 |
Feature activation | -9.11e-02 |
Pos logit contributions | |
Instruct | +0.312 |
Anon | +0.281 |
Teg | +0.278 |
Luxem | +0.272 |
Tuls | +0.263 |
Neg logit contributions | |
awaru | -0.678 |
�� | -0.658 |
ulia | -0.644 |
uese | -0.632 |
vantage | -0.632 |
Token | permission |
Token ID | 7170 |
Feature activation | -9.18e-02 |
Pos logit contributions | |
awaru | +0.159 |
�� | +0.152 |
ulia | +0.149 |
rogen | +0.143 |
uese | +0.142 |
Neg logit contributions | |
Instruct | -0.177 |
Anon | -0.167 |
Teg | -0.166 |
Luxem | -0.165 |
theless | -0.165 |
Token | to |
Token ID | 284 |
Feature activation | +3.89e-02 |
Pos logit contributions | |
awaru | +0.183 |
�� | +0.175 |
ulia | +0.171 |
vantage | +0.165 |
uese | +0.165 |
Neg logit contributions | |
Instruct | -0.160 |
Anon | -0.150 |
Teg | -0.149 |
Luxem | -0.147 |
theless | -0.144 |
Token | receive |
Token ID | 3328 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
Instruct | +0.064 |
theless | +0.061 |
Anon | +0.061 |
Luxem | +0.061 |
Teg | +0.061 |
Neg logit contributions | |
awaru | -0.078 |
ulia | -0.073 |
�� | -0.073 |
rogen | -0.073 |
inate | -0.072 |
Token | short |
Token ID | 1790 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
awaru | +0.380 |
�� | +0.368 |
ulia | +0.359 |
vantage | +0.353 |
Combat | +0.348 |
Neg logit contributions | |
Instruct | -0.355 |
Luxem | -0.333 |
Teg | -0.331 |
Anon | -0.330 |
theless | -0.325 |
Token | limit |
Token ID | 4179 |
Feature activation | +2.91e-02 |
Pos logit contributions | |
Instruct | +0.288 |
Luxem | +0.271 |
Teg | +0.269 |
Anon | +0.268 |
theless | +0.266 |
Neg logit contributions | |
awaru | -0.351 |
�� | -0.341 |
ulia | -0.332 |
vantage | -0.326 |
rogen | -0.324 |
Token | : |
Token ID | 25 |
Feature activation | -4.53e-02 |
Pos logit contributions | |
Instruct | +0.047 |
Luxem | +0.043 |
Teg | +0.043 |
Anon | +0.043 |
theless | +0.042 |
Neg logit contributions | |
awaru | -0.061 |
�� | -0.058 |
ulia | -0.057 |
vantage | -0.056 |
uese | -0.056 |
Token | 7 |
Token ID | 22 |
Feature activation | -3.46e-01 |
Pos logit contributions | |
awaru | +0.086 |
�� | +0.083 |
ulia | +0.082 |
vantage | +0.080 |
Combat | +0.079 |
Neg logit contributions | |
Instruct | -0.081 |
Teg | -0.075 |
Luxem | -0.075 |
Anon | -0.075 |
theless | -0.074 |
Token | R |
Token ID | 371 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
awaru | +0.659 |
�� | +0.647 |
ulia | +0.627 |
vantage | +0.611 |
uese | +0.606 |
Neg logit contributions | |
Instruct | -0.610 |
Teg | -0.571 |
Luxem | -0.570 |
Anon | -0.567 |
theless | -0.556 |
Token | TS |
Token ID | 4694 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
awaru | +0.453 |
�� | +0.430 |
ulia | +0.422 |
vantage | +0.412 |
rogen | +0.407 |
Neg logit contributions | |
Instruct | -0.531 |
Luxem | -0.500 |
Teg | -0.497 |
Anon | -0.496 |
Tuls | -0.489 |
Token | thr |
Token ID | 5636 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
awaru | +0.259 |
�� | +0.252 |
ulia | +0.244 |
vantage | +0.241 |
rogen | +0.239 |
Neg logit contributions | |
Instruct | -0.220 |
Luxem | -0.205 |
Teg | -0.205 |
Anon | -0.205 |
theless | -0.202 |
Token | : |
Token ID | 25 |
Feature activation | -7.45e-02 |
Pos logit contributions | |
awaru | +0.317 |
�� | +0.310 |
ulia | +0.301 |
vantage | +0.292 |
rogen | +0.291 |
Neg logit contributions | |
Instruct | -0.331 |
Luxem | -0.309 |
Anon | -0.308 |
Teg | -0.307 |
theless | -0.307 |
Token | off |
Token ID | 2364 |
Feature activation | -6.55e-02 |
Pos logit contributions | |
awaru | +0.136 |
�� | +0.131 |
ulia | +0.128 |
vantage | +0.126 |
rogen | +0.124 |
Neg logit contributions | |
Instruct | -0.140 |
Luxem | -0.130 |
Teg | -0.130 |
Anon | -0.130 |
theless | -0.127 |
Token | Frag |
Token ID | 24229 |
Feature activation | +2.79e-01 |
Pos logit contributions | |
awaru | +0.127 |
�� | +0.123 |
ulia | +0.120 |
vantage | +0.117 |
rogen | +0.117 |
Neg logit contributions | |
Instruct | -0.114 |
Luxem | -0.107 |
Anon | -0.106 |
Teg | -0.106 |
theless | -0.104 |
Token | ment |
Token ID | 434 |
Feature activation | -6.32e-01 |
Pos logit contributions | |
Instruct | +0.549 |
Anon | +0.516 |
Teg | +0.512 |
Luxem | +0.510 |
theless | +0.503 |
Neg logit contributions | |
awaru | -0.489 |
�� | -0.471 |
ulia | -0.457 |
vantage | -0.445 |
uese | -0.443 |
Token | States |
Token ID | 1829 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
awaru | +0.010 |
�� | +0.009 |
ulia | +0.009 |
uese | +0.009 |
vantage | +0.009 |
Neg logit contributions | |
Instruct | -0.014 |
Anon | -0.013 |
Teg | -0.013 |
Luxem | -0.013 |
theless | -0.013 |
Token | 6 |
Token ID | 718 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
awaru | +0.390 |
ulia | +0.376 |
�� | +0.375 |
Combat | +0.364 |
incial | +0.361 |
Neg logit contributions | |
Instruct | -0.359 |
Anon | -0.333 |
behav | -0.330 |
Teg | -0.329 |
theless | -0.327 |
Token | 83 |
Token ID | 5999 |
Feature activation | -2.04e-02 |
Pos logit contributions | |
awaru | +0.239 |
�� | +0.233 |
ulia | +0.231 |
Combat | +0.224 |
incial | +0.224 |
Neg logit contributions | |
Instruct | -0.191 |
Anon | -0.177 |
Teg | -0.176 |
Luxem | -0.174 |
behav | -0.174 |
Token | Posts |
Token ID | 12043 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
awaru | +0.036 |
�� | +0.035 |
ulia | +0.035 |
Combat | +0.034 |
vantage | +0.033 |
Neg logit contributions | |
Instruct | -0.038 |
Anon | -0.036 |
Teg | -0.036 |
Luxem | -0.035 |
theless | -0.035 |
Token | Last |
Token ID | 4586 |
Feature activation | -4.46e-01 |
Pos logit contributions | |
awaru | +3.075 |
�� | +2.963 |
ulia | +2.893 |
uese | +2.816 |
vantage | +2.815 |
Neg logit contributions | |
Instruct | -2.439 |
Anon | -2.266 |
Teg | -2.249 |
Luxem | -2.218 |
Tuls | -2.166 |
Token | Edited |
Token ID | 34212 |
Feature activation | -9.66e-01 |
Pos logit contributions | |
awaru | +0.547 |
ulia | +0.505 |
�� | +0.501 |
vantage | +0.483 |
uese | +0.480 |
Neg logit contributions | |
Instruct | -1.094 |
Teg | -1.041 |
Luxem | -1.039 |
theless | -1.034 |
Anon | -1.031 |
Token | : |
Token ID | 25 |
Feature activation | -9.66e-02 |
Pos logit contributions | |
awaru | +1.874 |
ulia | +1.861 |
rogen | +1.841 |
Combat | +1.816 |
incial | +1.784 |
Neg logit contributions | |
Instruct | -1.564 |
theless | -1.522 |
behav | -1.504 |
Luxem | -1.455 |
Anon | -1.439 |
Token | 2013 |
Token ID | 2211 |
Feature activation | +3.34e-02 |
Pos logit contributions | |
awaru | +0.160 |
ulia | +0.156 |
rogen | +0.154 |
Combat | +0.152 |
uese | +0.150 |
Neg logit contributions | |
Instruct | -0.188 |
behav | -0.179 |
theless | -0.177 |
Anon | -0.176 |
Tuls | -0.171 |
Token | - |
Token ID | 12 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
Instruct | +0.060 |
Anon | +0.056 |
theless | +0.056 |
Teg | +0.056 |
Luxem | +0.055 |
Neg logit contributions | |
awaru | -0.062 |
�� | -0.060 |
ulia | -0.060 |
Combat | -0.058 |
uese | -0.058 |
Token | 09 |
Token ID | 2931 |
Feature activation | -2.47e-01 |
Pos logit contributions | |
awaru | +0.024 |
ulia | +0.023 |
Combat | +0.023 |
�� | +0.023 |
UFC | +0.022 |
Neg logit contributions | |
Instruct | -0.020 |
behav | -0.019 |
Anon | -0.019 |
theless | -0.019 |
Teg | -0.019 |
Token | - |
Token ID | 12 |
Feature activation | +5.67e-02 |
Pos logit contributions | |
awaru | +0.455 |
ulia | +0.437 |
�� | +0.431 |
Combat | +0.423 |
uese | +0.422 |
Neg logit contributions | |
Instruct | -0.453 |
Anon | -0.425 |
theless | -0.421 |
Luxem | -0.419 |
Teg | -0.419 |
Token | Hills |
Token ID | 14379 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
Instruct | +0.174 |
Anon | +0.161 |
theless | +0.158 |
Luxem | +0.157 |
Teg | +0.157 |
Neg logit contributions | |
awaru | -0.261 |
�� | -0.250 |
ulia | -0.249 |
uese | -0.242 |
vantage | -0.240 |
Token | , |
Token ID | 11 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
Instruct | +0.323 |
Anon | +0.299 |
Teg | +0.294 |
Luxem | +0.291 |
theless | +0.288 |
Neg logit contributions | |
awaru | -0.508 |
�� | -0.489 |
ulia | -0.482 |
uese | -0.468 |
vantage | -0.468 |
Token | California |
Token ID | 3442 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
Instruct | +0.198 |
Anon | +0.185 |
theless | +0.183 |
behav | +0.181 |
Teg | +0.181 |
Neg logit contributions | |
awaru | -0.199 |
�� | -0.189 |
ulia | -0.187 |
rogen | -0.182 |
uese | -0.181 |
Token | . |
Token ID | 13 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
Instruct | +0.250 |
Anon | +0.231 |
Teg | +0.230 |
Luxem | +0.225 |
Tuls | +0.218 |
Neg logit contributions | |
awaru | -0.382 |
�� | -0.372 |
ulia | -0.358 |
uese | -0.352 |
vantage | -0.351 |
Token | Hide |
Token ID | 10415 |
Feature activation | -4.61e-01 |
Pos logit contributions | |
Instruct | +0.298 |
Anon | +0.279 |
Teg | +0.276 |
theless | +0.275 |
behav | +0.273 |
Neg logit contributions | |
awaru | -0.343 |
�� | -0.330 |
ulia | -0.329 |
vantage | -0.319 |
uese | -0.317 |
Token | Caption |
Token ID | 11260 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
awaru | +0.550 |
ulia | +0.520 |
�� | +0.508 |
incial | +0.486 |
vantage | +0.485 |
Neg logit contributions | |
Instruct | -1.132 |
theless | -1.091 |
Anon | -1.088 |
Teg | -1.079 |
Luxem | -1.079 |
Token | 14 |
Token ID | 1478 |
Feature activation | -8.65e-02 |
Pos logit contributions | |
ulia | +2.733 |
awaru | +2.725 |
uese | +2.575 |
Combat | +2.560 |
incial | +2.551 |
Neg logit contributions | |
theless | -2.983 |
Instruct | -2.962 |
behav | -2.922 |
princ | -2.869 |
warr | -2.869 |
Token | of |
Token ID | 286 |
Feature activation | +3.28e-03 |
Pos logit contributions | |
awaru | +0.141 |
�� | +0.132 |
ulia | +0.131 |
uese | +0.127 |
vantage | +0.126 |
Neg logit contributions | |
Instruct | -0.181 |
Anon | -0.171 |
Teg | -0.170 |
Luxem | -0.168 |
theless | -0.167 |
Token | 15 |
Token ID | 1315 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
Instruct | +0.007 |
Anon | +0.006 |
Teg | +0.006 |
Luxem | +0.006 |
theless | +0.006 |
Neg logit contributions | |
awaru | -0.006 |
�� | -0.005 |
ulia | -0.005 |
vantage | -0.005 |
uese | -0.005 |
Token | Photos |
Token ID | 9434 |
Feature activation | -9.30e-01 |
Pos logit contributions | |
awaru | +0.034 |
�� | +0.032 |
ulia | +0.032 |
uese | +0.030 |
vantage | +0.030 |
Neg logit contributions | |
Instruct | -0.048 |
Anon | -0.045 |
Teg | -0.045 |
Luxem | -0.045 |
theless | -0.044 |
Token | : |
Token ID | 25 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
awaru | +1.839 |
ulia | +1.819 |
Combat | +1.783 |
rogen | +1.773 |
incial | +1.751 |
Neg logit contributions | |
Instruct | -1.500 |
theless | -1.452 |
behav | -1.439 |
Anon | -1.395 |
princ | -1.387 |
Token | was |
Token ID | 373 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
awaru | +0.153 |
�� | +0.144 |
ulia | +0.142 |
vantage | +0.139 |
rogen | +0.139 |
Neg logit contributions | |
Instruct | -0.145 |
Anon | -0.136 |
Teg | -0.136 |
Luxem | -0.135 |
Tuls | -0.131 |
Token | about |
Token ID | 546 |
Feature activation | -5.64e-02 |
Pos logit contributions | |
awaru | +0.368 |
�� | +0.343 |
ulia | +0.340 |
rogen | +0.339 |
uese | +0.331 |
Neg logit contributions | |
Instruct | -0.319 |
Teg | -0.298 |
Luxem | -0.297 |
Anon | -0.296 |
theless | -0.287 |
Token | homosexuality |
Token ID | 19822 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
awaru | +0.105 |
�� | +0.100 |
ulia | +0.098 |
rogen | +0.097 |
uese | +0.097 |
Neg logit contributions | |
Instruct | -0.103 |
Anon | -0.096 |
Teg | -0.096 |
Luxem | -0.095 |
theless | -0.094 |
Token |
|
Token ID | 198 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
Instruct | +0.293 |
Teg | +0.273 |
Anon | +0.271 |
Luxem | +0.269 |
theless | +0.262 |
Neg logit contributions | |
awaru | -0.357 |
�� | -0.340 |
ulia | -0.335 |
uese | -0.331 |
rogen | -0.330 |
Token |
|
Token ID | 198 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
Instruct | +0.308 |
Anon | +0.286 |
Teg | +0.284 |
Luxem | +0.281 |
Tuls | +0.274 |
Neg logit contributions | |
awaru | -0.384 |
�� | -0.370 |
ulia | -0.361 |
vantage | -0.352 |
uese | -0.352 |
Token | Reviewer |
Token ID | 35407 |
Feature activation | -9.18e-01 |
Pos logit contributions | |
Instruct | +0.259 |
Teg | +0.244 |
Anon | +0.242 |
Luxem | +0.241 |
Tuls | +0.236 |
Neg logit contributions | |
awaru | -0.233 |
�� | -0.223 |
ulia | -0.217 |
Combat | -0.212 |
uese | -0.211 |
Token | : |
Token ID | 25 |
Feature activation | +3.29e-01 |
Pos logit contributions | |
awaru | +1.888 |
ulia | +1.823 |
�� | +1.768 |
rogen | +1.766 |
uese | +1.754 |
Neg logit contributions | |
Instruct | -1.439 |
theless | -1.346 |
Luxem | -1.344 |
Anon | -1.343 |
behav | -1.333 |
Token | k |
Token ID | 479 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
Instruct | +0.582 |
Anon | +0.541 |
Teg | +0.540 |
Luxem | +0.537 |
theless | +0.525 |
Neg logit contributions | |
awaru | -0.639 |
�� | -0.614 |
ulia | -0.601 |
uese | -0.582 |
vantage | -0.581 |
Token | imp |
Token ID | 11011 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
Instruct | +0.318 |
Anon | +0.300 |
Teg | +0.299 |
Luxem | +0.296 |
Tuls | +0.290 |
Neg logit contributions | |
awaru | -0.261 |
�� | -0.246 |
ulia | -0.243 |
uese | -0.234 |
vantage | -0.233 |
Token | unk |
Token ID | 2954 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
Instruct | +0.453 |
Anon | +0.424 |
Teg | +0.422 |
Luxem | +0.418 |
Tuls | +0.411 |
Neg logit contributions | |
awaru | -0.414 |
�� | -0.394 |
ulia | -0.388 |
uese | -0.374 |
vantage | -0.374 |
Token | rock |
Token ID | 10823 |
Feature activation | -2.96e-01 |
Pos logit contributions | |
Instruct | +0.235 |
Teg | +0.218 |
Anon | +0.218 |
Luxem | +0.215 |
theless | +0.211 |
Neg logit contributions | |
awaru | -0.264 |
�� | -0.253 |
ulia | -0.248 |
uese | -0.241 |
vantage | -0.241 |
Token | June |
Token ID | 2795 |
Feature activation | +7.32e-02 |
Pos logit contributions | |
ulia | +2.099 |
awaru | +2.067 |
Combat | +2.052 |
rogen | +2.030 |
�� | +2.001 |
Neg logit contributions | |
Instruct | -2.337 |
behav | -2.240 |
theless | -2.190 |
opsis | -2.139 |
Tuls | -2.131 |
Token | 2009 |
Token ID | 3717 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
Instruct | +0.139 |
behav | +0.130 |
Anon | +0.130 |
Teg | +0.129 |
theless | +0.128 |
Neg logit contributions | |
awaru | -0.125 |
ulia | -0.122 |
�� | -0.121 |
Combat | -0.118 |
rogen | -0.117 |
Token | Germany |
Token ID | 4486 |
Feature activation | +8.60e-02 |
Pos logit contributions | |
awaru | +0.210 |
�� | +0.202 |
ulia | +0.200 |
uese | +0.195 |
vantage | +0.194 |
Neg logit contributions | |
Instruct | -0.167 |
Anon | -0.155 |
Teg | -0.153 |
behav | -0.151 |
Luxem | -0.150 |
Token | 247 |
Token ID | 30179 |
Feature activation | -2.60e-01 |
Pos logit contributions | |
Instruct | +0.151 |
Anon | +0.139 |
behav | +0.139 |
Teg | +0.138 |
theless | +0.137 |
Neg logit contributions | |
awaru | -0.162 |
ulia | -0.157 |
�� | -0.157 |
Combat | -0.152 |
uese | -0.150 |
Token | 34 |
Token ID | 2682 |
Feature activation | -4.95e-02 |
Pos logit contributions | |
awaru | +0.474 |
ulia | +0.465 |
�� | +0.462 |
Combat | +0.452 |
vantage | +0.447 |
Neg logit contributions | |
Instruct | -0.459 |
behav | -0.432 |
Anon | -0.431 |
Teg | -0.430 |
theless | -0.424 |
Token | Posts |
Token ID | 12043 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
awaru | +0.088 |
�� | +0.085 |
ulia | +0.085 |
Combat | +0.082 |
uese | +0.082 |
Neg logit contributions | |
Instruct | -0.093 |
Anon | -0.087 |
Teg | -0.086 |
theless | -0.086 |
Luxem | -0.086 |
Token | # |
Token ID | 1303 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
awaru | +2.986 |
�� | +2.874 |
ulia | +2.810 |
vantage | +2.730 |
uese | +2.729 |
Neg logit contributions | |
Instruct | -2.441 |
Anon | -2.264 |
Teg | -2.251 |
Luxem | -2.222 |
Tuls | -2.177 |
Token | 20 |
Token ID | 1238 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
awaru | +2.211 |
ulia | +2.141 |
Combat | +2.084 |
vantage | +2.066 |
�� | +2.062 |
Neg logit contributions | |
Instruct | -2.421 |
Anon | -2.283 |
Teg | -2.253 |
Luxem | -2.246 |
theless | -2.241 |
Token | Nice |
Token ID | 18460 |
Feature activation | +2.65e-02 |
Pos logit contributions | |
awaru | +0.231 |
�� | +0.220 |
ulia | +0.217 |
uese | +0.210 |
vantage | +0.210 |
Neg logit contributions | |
Instruct | -0.228 |
Anon | -0.213 |
Teg | -0.212 |
Luxem | -0.212 |
Tuls | -0.206 |
Token | , |
Token ID | 11 |
Feature activation | +5.66e-02 |
Pos logit contributions | |
Instruct | +0.045 |
Anon | +0.041 |
Luxem | +0.041 |
Teg | +0.041 |
Tuls | +0.040 |
Neg logit contributions | |
awaru | -0.053 |
�� | -0.051 |
ulia | -0.050 |
uese | -0.049 |
rogen | -0.048 |
Token | happy |
Token ID | 3772 |
Feature activation | +8.20e-02 |
Pos logit contributions | |
Instruct | +0.108 |
Anon | +0.101 |
Teg | +0.100 |
Luxem | +0.100 |
Tuls | +0.098 |
Neg logit contributions | |
awaru | -0.102 |
�� | -0.098 |
ulia | -0.096 |
uese | -0.093 |
vantage | -0.093 |
Token |
|
Token ID | 198 |
Feature activation | -7.80e-02 |
Pos logit contributions | |
Instruct | +0.399 |
Anon | +0.372 |
Teg | +0.371 |
Luxem | +0.366 |
Tuls | +0.358 |
Neg logit contributions | |
awaru | -0.478 |
�� | -0.457 |
ulia | -0.447 |
vantage | -0.434 |
uese | -0.434 |
Token |
|
Token ID | 198 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
awaru | +0.152 |
ulia | +0.145 |
�� | +0.144 |
Combat | +0.141 |
rogen | +0.140 |
Neg logit contributions | |
Instruct | -0.136 |
Anon | -0.127 |
Teg | -0.125 |
Luxem | -0.125 |
theless | -0.125 |
Token | Ko |
Token ID | 48735 |
Feature activation | -6.46e-01 |
Pos logit contributions | |
awaru | +0.192 |
�� | +0.183 |
ulia | +0.178 |
vantage | +0.172 |
uese | +0.171 |
Neg logit contributions | |
Instruct | -0.242 |
Anon | -0.227 |
Teg | -0.226 |
Luxem | -0.225 |
Tuls | -0.221 |
Token | Do |
Token ID | 5211 |
Feature activation | -7.09e-02 |
Pos logit contributions | |
awaru | +1.113 |
�� | +1.054 |
ulia | +1.026 |
uese | +0.991 |
rogen | +0.991 |
Neg logit contributions | |
Instruct | -1.312 |
Anon | -1.223 |
Teg | -1.221 |
Luxem | -1.212 |
theless | -1.197 |
Token | Profile |
Token ID | 13118 |
Feature activation | -6.57e-01 |
Pos logit contributions | |
awaru | +0.121 |
�� | +0.116 |
ulia | +0.113 |
uese | +0.109 |
vantage | +0.109 |
Neg logit contributions | |
Instruct | -0.144 |
Anon | -0.135 |
Teg | -0.135 |
Luxem | -0.133 |
theless | -0.131 |
Token | Joined |
Token ID | 16709 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
awaru | +1.547 |
�� | +1.501 |
ulia | +1.475 |
vantage | +1.443 |
uese | +1.442 |
Neg logit contributions | |
Instruct | -0.885 |
Teg | -0.800 |
Anon | -0.799 |
Luxem | -0.786 |
theless | -0.776 |
Token | December |
Token ID | 3426 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
ulia | +2.135 |
Combat | +2.098 |
awaru | +2.087 |
rogen | +2.073 |
uese | +2.050 |
Neg logit contributions | |
Instruct | -2.415 |
behav | -2.335 |
theless | -2.293 |
Teg | -2.234 |
Tuls | -2.228 |
Token | 2010 |
Token ID | 3050 |
Feature activation | +7.16e-02 |
Pos logit contributions | |
Instruct | +0.559 |
Anon | +0.522 |
behav | +0.520 |
theless | +0.516 |
Teg | +0.515 |
Neg logit contributions | |
awaru | -0.501 |
ulia | -0.486 |
�� | -0.482 |
Combat | -0.468 |
rogen | -0.467 |
Token | United |
Token ID | 1578 |
Feature activation | -6.48e-03 |
Pos logit contributions | |
Instruct | +0.107 |
Anon | +0.098 |
Teg | +0.097 |
Luxem | +0.095 |
theless | +0.095 |
Neg logit contributions | |
awaru | -0.157 |
�� | -0.152 |
ulia | -0.150 |
uese | -0.146 |
vantage | -0.146 |
Token | States |
Token ID | 1829 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
awaru | +0.010 |
�� | +0.009 |
ulia | +0.009 |
uese | +0.009 |
vantage | +0.009 |
Neg logit contributions | |
Instruct | -0.014 |
Anon | -0.013 |
Teg | -0.013 |
Luxem | -0.013 |
theless | -0.013 |
Token | 6 |
Token ID | 718 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
awaru | +0.390 |
ulia | +0.376 |
�� | +0.375 |
Combat | +0.364 |
incial | +0.361 |
Neg logit contributions | |
Instruct | -0.359 |
Anon | -0.333 |
behav | -0.330 |
Teg | -0.329 |
theless | -0.327 |