Token | his |
Token ID | 465 |
Feature activation | +2.32e+00 |
Pos logit contributions | |
blonde | +0.030 |
Eva | +0.030 |
chard | +0.030 |
wrists | +0.030 |
Robert | +0.030 |
Neg logit contributions | |
of | -0.019 |
loo | -0.018 |
ick | -0.017 |
el | -0.016 |
iren | -0.015 |
Token | head |
Token ID | 1182 |
Feature activation | +9.87e-01 |
Pos logit contributions | |
loo | +0.083 |
of | +0.075 |
ick | +0.069 |
ig | +0.069 |
auc | +0.066 |
Neg logit contributions | |
Eva | -0.103 |
chard | -0.102 |
Robert | -0.102 |
wrists | -0.100 |
blonde | -0.097 |
Token | out |
Token ID | 503 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
loo | +0.027 |
of | +0.024 |
ick | +0.022 |
iren | +0.021 |
auc | +0.021 |
Neg logit contributions | |
Eva | -0.052 |
Robert | -0.052 |
chard | -0.052 |
wrists | -0.052 |
blonde | -0.050 |
Token | of |
Token ID | 286 |
Feature activation | +4.11e+00 |
Pos logit contributions | |
Eva | +0.061 |
Robert | +0.060 |
chard | +0.059 |
wrists | +0.059 |
blonde | +0.058 |
Neg logit contributions | |
loo | -0.028 |
iren | -0.021 |
auc | -0.021 |
ick | -0.021 |
arl | -0.020 |
Token | his |
Token ID | 465 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
loo | +0.157 |
auc | +0.135 |
ick | +0.134 |
iren | +0.132 |
of | +0.129 |
Neg logit contributions | |
Eva | -0.179 |
Robert | -0.179 |
chard | -0.176 |
wrists | -0.175 |
blonde | -0.170 |
Token | bur |
Token ID | 4356 |
Feature activation | +8.15e+00 |
Pos logit contributions | |
loo | +0.038 |
of | +0.033 |
ick | +0.032 |
iren | +0.031 |
auc | +0.031 |
Neg logit contributions | |
Eva | -0.050 |
chard | -0.050 |
Robert | -0.050 |
wrists | -0.049 |
blonde | -0.047 |
Token | row |
Token ID | 808 |
Feature activation | +5.51e+00 |
Pos logit contributions | |
loo | +0.164 |
auc | +0.124 |
pour | +0.121 |
iren | +0.120 |
of | +0.103 |
Neg logit contributions | |
chard | -0.466 |
wrists | -0.461 |
blonde | -0.450 |
Eva | -0.448 |
Robert | -0.438 |
Token | and |
Token ID | 290 |
Feature activation | -2.25e-01 |
Pos logit contributions | |
loo | +0.175 |
iren | +0.145 |
ick | +0.142 |
auc | +0.141 |
of | +0.139 |
Neg logit contributions | |
chard | -0.267 |
Robert | -0.265 |
Eva | -0.263 |
wrists | -0.260 |
blonde | -0.254 |
Token | saw |
Token ID | 2497 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
Eva | +0.009 |
Robert | +0.009 |
chard | +0.009 |
wrists | +0.009 |
blonde | +0.009 |
Neg logit contributions | |
loo | -0.009 |
of | -0.008 |
ick | -0.008 |
iren | -0.008 |
auc | -0.008 |
Token | something |
Token ID | 1223 |
Feature activation | -2.98e+00 |
Pos logit contributions | |
wrists | +0.011 |
chard | +0.011 |
Eva | +0.011 |
Robert | +0.011 |
blonde | +0.010 |
Neg logit contributions | |
loo | -0.008 |
of | -0.008 |
ick | -0.007 |
auc | -0.007 |
ig | -0.007 |
Token | new |
Token ID | 649 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
Eva | +0.125 |
wrists | +0.124 |
Robert | +0.123 |
chard | +0.123 |
blonde | +0.118 |
Neg logit contributions | |
loo | -0.117 |
of | -0.106 |
ick | -0.101 |
auc | -0.099 |
iren | -0.098 |
Token | who |
Token ID | 508 |
Feature activation | -4.55e+00 |
Pos logit contributions | |
chard | +0.103 |
Eva | +0.103 |
Robert | +0.103 |
wrists | +0.102 |
blonde | +0.098 |
Neg logit contributions | |
loo | -0.101 |
of | -0.087 |
ick | -0.087 |
iren | -0.086 |
auc | -0.086 |
Token | lived |
Token ID | 5615 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
wrists | +0.191 |
chard | +0.186 |
Eva | +0.184 |
Robert | +0.183 |
blonde | +0.169 |
Neg logit contributions | |
loo | -0.176 |
of | -0.172 |
ick | -0.159 |
auc | -0.147 |
iren | -0.147 |
Token | in |
Token ID | 287 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
loo | +0.041 |
of | +0.035 |
ick | +0.035 |
iren | +0.033 |
auc | +0.033 |
Neg logit contributions | |
Eva | -0.066 |
Robert | -0.065 |
wrists | -0.065 |
chard | -0.065 |
blonde | -0.063 |
Token | his |
Token ID | 465 |
Feature activation | +4.54e-01 |
Pos logit contributions | |
Eva | +0.083 |
Robert | +0.083 |
chard | +0.081 |
wrists | +0.079 |
blonde | +0.078 |
Neg logit contributions | |
loo | -0.097 |
auc | -0.085 |
iren | -0.084 |
ick | -0.083 |
of | -0.083 |
Token | cozy |
Token ID | 37438 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
loo | +0.015 |
iren | +0.013 |
auc | +0.013 |
ick | +0.013 |
of | +0.013 |
Neg logit contributions | |
Eva | -0.022 |
Robert | -0.021 |
chard | -0.021 |
wrists | -0.021 |
blonde | -0.021 |
Token | bur |
Token ID | 4356 |
Feature activation | +7.65e+00 |
Pos logit contributions | |
loo | +0.069 |
auc | +0.064 |
iren | +0.063 |
ick | +0.058 |
arl | +0.057 |
Neg logit contributions | |
wrists | -0.094 |
chard | -0.093 |
Robert | -0.093 |
blonde | -0.092 |
Eva | -0.092 |
Token | row |
Token ID | 808 |
Feature activation | +5.10e+00 |
Pos logit contributions | |
loo | +0.162 |
auc | +0.141 |
iren | +0.133 |
pour | +0.130 |
ick | +0.114 |
Neg logit contributions | |
chard | -0.435 |
wrists | -0.418 |
blonde | -0.410 |
Robert | -0.394 |
Eva | -0.394 |
Token | . |
Token ID | 13 |
Feature activation | +6.30e-01 |
Pos logit contributions | |
loo | +0.159 |
iren | +0.131 |
ick | +0.129 |
auc | +0.128 |
of | +0.125 |
Neg logit contributions | |
chard | -0.254 |
Robert | -0.250 |
Eva | -0.249 |
wrists | -0.249 |
blonde | -0.242 |
Token | He |
Token ID | 679 |
Feature activation | -7.73e-01 |
Pos logit contributions | |
of | +0.019 |
loo | +0.018 |
late | +0.015 |
ig | +0.015 |
ick | +0.015 |
Neg logit contributions | |
wrists | -0.031 |
chard | -0.030 |
cious | -0.029 |
yes | -0.028 |
rice | -0.028 |
Token | loved |
Token ID | 6151 |
Feature activation | -2.60e-01 |
Pos logit contributions | |
chard | +0.038 |
Robert | +0.038 |
Eva | +0.038 |
wrists | +0.037 |
Edward | +0.035 |
Neg logit contributions | |
of | -0.023 |
loo | -0.021 |
ig | -0.020 |
ick | -0.019 |
row | -0.018 |
Token | to |
Token ID | 284 |
Feature activation | -3.38e-02 |
Pos logit contributions | |
Eva | +0.012 |
Robert | +0.012 |
chard | +0.012 |
wrists | +0.012 |
blonde | +0.012 |
Neg logit contributions | |
loo | -0.009 |
of | -0.007 |
ick | -0.007 |
iren | -0.007 |
auc | -0.007 |
Token | mist |
Token ID | 4020 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
loo | +0.038 |
ick | +0.033 |
of | +0.032 |
iren | +0.032 |
auc | +0.032 |
Neg logit contributions | |
Eva | -0.047 |
chard | -0.047 |
Robert | -0.047 |
wrists | -0.047 |
blonde | -0.046 |
Token | and |
Token ID | 290 |
Feature activation | -3.41e-01 |
Pos logit contributions | |
loo | +0.066 |
iren | +0.055 |
ick | +0.055 |
auc | +0.054 |
of | +0.053 |
Neg logit contributions | |
Eva | -0.099 |
Robert | -0.098 |
chard | -0.098 |
wrists | -0.097 |
blonde | -0.095 |
Token | into |
Token ID | 656 |
Feature activation | -4.33e-01 |
Pos logit contributions | |
Robert | +0.014 |
wrists | +0.014 |
Eva | +0.014 |
blonde | +0.014 |
chard | +0.014 |
Neg logit contributions | |
loo | -0.013 |
of | -0.012 |
pour | -0.011 |
auc | -0.010 |
ick | -0.010 |
Token | their |
Token ID | 511 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
chard | +0.019 |
Eva | +0.019 |
Robert | +0.019 |
wrists | +0.019 |
blonde | +0.018 |
Neg logit contributions | |
loo | -0.016 |
of | -0.014 |
ick | -0.014 |
iren | -0.013 |
auc | -0.013 |
Token | cozy |
Token ID | 37438 |
Feature activation | +1.98e+00 |
Pos logit contributions | |
Eva | +0.005 |
chard | +0.005 |
Robert | +0.005 |
wrists | +0.005 |
blonde | +0.005 |
Neg logit contributions | |
loo | -0.003 |
of | -0.003 |
ick | -0.003 |
iren | -0.003 |
auc | -0.003 |
Token | bur |
Token ID | 4356 |
Feature activation | +7.47e+00 |
Pos logit contributions | |
loo | +0.071 |
auc | +0.066 |
iren | +0.065 |
ick | +0.059 |
arl | +0.058 |
Neg logit contributions | |
wrists | -0.090 |
blonde | -0.089 |
Eva | -0.085 |
chard | -0.085 |
Claire | -0.084 |
Token | rows |
Token ID | 8516 |
Feature activation | +9.52e-01 |
Pos logit contributions | |
loo | +0.162 |
auc | +0.133 |
iren | +0.130 |
pour | +0.121 |
ick | +0.108 |
Neg logit contributions | |
wrists | -0.398 |
chard | -0.396 |
blonde | -0.389 |
cious | -0.386 |
Eva | -0.378 |
Token | . |
Token ID | 13 |
Feature activation | +3.70e-01 |
Pos logit contributions | |
loo | +0.030 |
ick | +0.024 |
iren | +0.024 |
of | +0.024 |
auc | +0.024 |
Neg logit contributions | |
chard | -0.047 |
Eva | -0.047 |
Robert | -0.047 |
wrists | -0.047 |
blonde | -0.045 |
Token | The |
Token ID | 383 |
Feature activation | +2.44e-02 |
Pos logit contributions | |
loo | +0.014 |
of | +0.013 |
ick | +0.011 |
iren | +0.011 |
auc | +0.011 |
Neg logit contributions | |
wrists | -0.017 |
chard | -0.016 |
blonde | -0.015 |
cious | -0.015 |
Eva | -0.014 |
Token | end |
Token ID | 886 |
Feature activation | +4.00e+00 |
Pos logit contributions | |
loo | +0.001 |
of | +0.001 |
ick | +0.001 |
auc | +0.001 |
iren | +0.001 |
Neg logit contributions | |
Eva | -0.001 |
Robert | -0.001 |
chard | -0.001 |
wrists | -0.001 |
blonde | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
loo | +0.150 |
arl | +0.134 |
iren | +0.132 |
auc | +0.131 |
ick | +0.124 |
Neg logit contributions | |
Eva | -0.166 |
Robert | -0.157 |
blonde | -0.154 |
Joy | -0.153 |
aloud | -0.151 |
Token | able |
Token ID | 1498 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
loo | +0.093 |
of | +0.082 |
ick | +0.080 |
iren | +0.078 |
auc | +0.077 |
Neg logit contributions | |
Eva | -0.108 |
chard | -0.107 |
wrists | -0.107 |
Robert | -0.107 |
blonde | -0.103 |
Token | to |
Token ID | 284 |
Feature activation | -6.13e-03 |
Pos logit contributions | |
Robert | +0.103 |
Eva | +0.103 |
chard | +0.103 |
wrists | +0.102 |
blonde | +0.098 |
Neg logit contributions | |
loo | -0.075 |
ick | -0.063 |
iren | -0.062 |
auc | -0.062 |
of | -0.059 |
Token | escape |
Token ID | 6654 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
Eva | +0.000 |
Robert | +0.000 |
chard | +0.000 |
wrists | +0.000 |
blonde | +0.000 |
Neg logit contributions | |
loo | -0.000 |
ick | -0.000 |
of | -0.000 |
auc | -0.000 |
iren | -0.000 |
Token | from |
Token ID | 422 |
Feature activation | -5.19e-01 |
Pos logit contributions | |
loo | +0.033 |
ick | +0.027 |
iren | +0.027 |
auc | +0.026 |
of | +0.026 |
Neg logit contributions | |
Robert | -0.050 |
chard | -0.050 |
Eva | -0.050 |
wrists | -0.049 |
blonde | -0.047 |
Token | his |
Token ID | 465 |
Feature activation | +9.35e-02 |
Pos logit contributions | |
wrists | +0.024 |
Eva | +0.024 |
Robert | +0.024 |
chard | +0.024 |
blonde | +0.023 |
Neg logit contributions | |
loo | -0.018 |
of | -0.016 |
ick | -0.015 |
iren | -0.015 |
auc | -0.014 |
Token | bur |
Token ID | 4356 |
Feature activation | +7.46e+00 |
Pos logit contributions | |
loo | +0.003 |
of | +0.003 |
ick | +0.003 |
iren | +0.003 |
auc | +0.003 |
Neg logit contributions | |
Eva | -0.004 |
chard | -0.004 |
Robert | -0.004 |
wrists | -0.004 |
blonde | -0.004 |
Token | row |
Token ID | 808 |
Feature activation | +4.86e+00 |
Pos logit contributions | |
loo | +0.155 |
auc | +0.120 |
iren | +0.112 |
pour | +0.110 |
ick | +0.105 |
Neg logit contributions | |
chard | -0.414 |
wrists | -0.410 |
blonde | -0.393 |
rice | -0.385 |
Eva | -0.384 |
Token | and |
Token ID | 290 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
loo | +0.158 |
iren | +0.131 |
ick | +0.129 |
auc | +0.127 |
arl | +0.122 |
Neg logit contributions | |
chard | -0.232 |
Robert | -0.230 |
Eva | -0.227 |
wrists | -0.225 |
blonde | -0.219 |
Token | play |
Token ID | 711 |
Feature activation | +7.90e-02 |
Pos logit contributions | |
Eva | +0.007 |
Robert | +0.007 |
chard | +0.007 |
wrists | +0.007 |
blonde | +0.006 |
Neg logit contributions | |
loo | -0.007 |
ick | -0.006 |
of | -0.006 |
iren | -0.006 |
auc | -0.006 |
Token | in |
Token ID | 287 |
Feature activation | -2.48e+00 |
Pos logit contributions | |
loo | +0.002 |
of | +0.002 |
ick | +0.002 |
auc | +0.002 |
iren | +0.002 |
Neg logit contributions | |
Eva | -0.004 |
chard | -0.004 |
wrists | -0.004 |
Robert | -0.004 |
blonde | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | +5.17e-01 |
Pos logit contributions | |
Robert | +0.102 |
Eva | +0.101 |
chard | +0.098 |
wrists | +0.096 |
Claire | +0.095 |
Neg logit contributions | |
loo | -0.101 |
auc | -0.089 |
iren | -0.088 |
arl | -0.086 |
ick | -0.086 |
Token | and |
Token ID | 290 |
Feature activation | -3.64e-02 |
Pos logit contributions | |
Eva | +0.164 |
chard | +0.163 |
Robert | +0.163 |
wrists | +0.162 |
blonde | +0.157 |
Neg logit contributions | |
loo | -0.095 |
of | -0.079 |
ick | -0.078 |
iren | -0.076 |
auc | -0.075 |
Token | showed |
Token ID | 3751 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
chard | +0.002 |
Eva | +0.002 |
Robert | +0.002 |
wrists | +0.002 |
blonde | +0.001 |
Neg logit contributions | |
loo | -0.001 |
of | -0.001 |
ick | -0.001 |
iren | -0.001 |
auc | -0.001 |
Token | him |
Token ID | 683 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
chard | +0.061 |
wrists | +0.058 |
blonde | +0.058 |
Eva | +0.057 |
Robert | +0.056 |
Neg logit contributions | |
loo | -0.039 |
of | -0.039 |
ick | -0.035 |
auc | -0.034 |
iren | -0.032 |
Token | around |
Token ID | 1088 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
chard | +0.015 |
Robert | +0.015 |
Eva | +0.015 |
wrists | +0.015 |
blonde | +0.015 |
Neg logit contributions | |
loo | -0.011 |
of | -0.010 |
ick | -0.009 |
auc | -0.009 |
ig | -0.009 |
Token | their |
Token ID | 511 |
Feature activation | -2.61e-01 |
Pos logit contributions | |
Eva | +0.095 |
chard | +0.095 |
Robert | +0.094 |
wrists | +0.094 |
blonde | +0.091 |
Neg logit contributions | |
loo | -0.059 |
of | -0.051 |
ick | -0.049 |
auc | -0.047 |
iren | -0.047 |
Token | bur |
Token ID | 4356 |
Feature activation | +7.39e+00 |
Pos logit contributions | |
chard | +0.012 |
Eva | +0.012 |
Robert | +0.012 |
blonde | +0.011 |
wrists | +0.011 |
Neg logit contributions | |
loo | -0.009 |
of | -0.008 |
ig | -0.008 |
ick | -0.008 |
tunnel | -0.008 |
Token | row |
Token ID | 808 |
Feature activation | +4.66e+00 |
Pos logit contributions | |
loo | +0.128 |
auc | +0.091 |
pour | +0.090 |
iren | +0.090 |
ick | +0.088 |
Neg logit contributions | |
wrists | -0.452 |
chard | -0.441 |
blonde | -0.434 |
Eva | -0.426 |
Robert | -0.418 |
Token | . |
Token ID | 13 |
Feature activation | +5.21e-01 |
Pos logit contributions | |
loo | +0.140 |
ick | +0.114 |
of | +0.114 |
iren | +0.113 |
auc | +0.111 |
Neg logit contributions | |
chard | -0.235 |
Eva | -0.235 |
Robert | -0.234 |
wrists | -0.234 |
blonde | -0.225 |
Token | They |
Token ID | 1119 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
loo | +0.019 |
of | +0.019 |
auc | +0.017 |
ick | +0.017 |
iren | +0.017 |
Neg logit contributions | |
wrists | -0.022 |
chard | -0.022 |
cious | -0.020 |
Eva | -0.020 |
blonde | -0.020 |
Token | took |
Token ID | 1718 |
Feature activation | +4.48e-02 |
Pos logit contributions | |
Eva | +0.006 |
chard | +0.006 |
Robert | +0.006 |
wrists | +0.006 |
blonde | +0.006 |
Neg logit contributions | |
loo | -0.005 |
of | -0.005 |
ick | -0.005 |
ig | -0.005 |
auc | -0.005 |
Token | turns |
Token ID | 4962 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
loo | +0.002 |
of | +0.001 |
ick | +0.001 |
auc | +0.001 |
iren | +0.001 |
Neg logit contributions | |
chard | -0.002 |
wrists | -0.002 |
Eva | -0.002 |
Robert | -0.002 |
blonde | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | -3.22e-01 |
Pos logit contributions | |
of | +0.011 |
loo | +0.010 |
ick | +0.009 |
pour | +0.009 |
late | +0.009 |
Neg logit contributions | |
wrists | -0.016 |
Eva | -0.015 |
Robert | -0.015 |
chard | -0.015 |
blonde | -0.015 |
Token | soon |
Token ID | 2582 |
Feature activation | +8.67e-01 |
Pos logit contributions | |
wrists | +0.014 |
Robert | +0.014 |
Eva | +0.014 |
blonde | +0.014 |
chard | +0.014 |
Neg logit contributions | |
loo | -0.011 |
of | -0.011 |
ick | -0.009 |
pour | -0.009 |
auc | -0.009 |
Token | they |
Token ID | 484 |
Feature activation | +9.57e-01 |
Pos logit contributions | |
loo | +0.028 |
of | +0.027 |
ick | +0.023 |
pour | +0.022 |
auc | +0.021 |
Neg logit contributions | |
wrists | -0.042 |
chard | -0.041 |
blonde | -0.039 |
Robert | -0.039 |
Eva | -0.039 |
Token | found |
Token ID | 1043 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
loo | +0.034 |
of | +0.032 |
ick | +0.029 |
pour | +0.028 |
auc | +0.028 |
Neg logit contributions | |
wrists | -0.043 |
Eva | -0.042 |
chard | -0.042 |
Robert | -0.041 |
blonde | -0.041 |
Token | their |
Token ID | 511 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
loo | +0.040 |
of | +0.039 |
ick | +0.033 |
auc | +0.033 |
ig | +0.032 |
Neg logit contributions | |
wrists | -0.055 |
chard | -0.055 |
blonde | -0.054 |
Eva | -0.054 |
Robert | -0.054 |
Token | bur |
Token ID | 4356 |
Feature activation | +7.31e+00 |
Pos logit contributions | |
Eva | +0.011 |
Robert | +0.011 |
chard | +0.011 |
wrists | +0.010 |
blonde | +0.010 |
Neg logit contributions | |
loo | -0.008 |
of | -0.007 |
ig | -0.007 |
ick | -0.007 |
tunnel | -0.007 |
Token | row |
Token ID | 808 |
Feature activation | +4.70e+00 |
Pos logit contributions | |
loo | +0.159 |
iren | +0.133 |
auc | +0.128 |
pour | +0.119 |
ick | +0.113 |
Neg logit contributions | |
chard | -0.404 |
wrists | -0.393 |
blonde | -0.379 |
Robert | -0.375 |
Eva | -0.375 |
Token | . |
Token ID | 13 |
Feature activation | +3.62e-01 |
Pos logit contributions | |
loo | +0.150 |
iren | +0.125 |
ick | +0.124 |
of | +0.121 |
auc | +0.120 |
Neg logit contributions | |
chard | -0.228 |
Robert | -0.226 |
Eva | -0.226 |
wrists | -0.224 |
blonde | -0.216 |
Token | The |
Token ID | 383 |
Feature activation | +6.06e-02 |
Pos logit contributions | |
loo | +0.012 |
of | +0.012 |
iren | +0.011 |
auc | +0.010 |
ick | +0.010 |
Neg logit contributions | |
wrists | -0.016 |
chard | -0.016 |
blonde | -0.015 |
cious | -0.015 |
Eva | -0.014 |
Token | animals |
Token ID | 4695 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
loo | +0.002 |
of | +0.002 |
ick | +0.002 |
ig | +0.002 |
auc | +0.002 |
Neg logit contributions | |
Eva | -0.003 |
Robert | -0.003 |
wrists | -0.003 |
chard | -0.003 |
blonde | -0.003 |
Token | were |
Token ID | 547 |
Feature activation | +8.74e-01 |
Pos logit contributions | |
Eva | +0.087 |
wrists | +0.086 |
Robert | +0.086 |
chard | +0.085 |
blonde | +0.083 |
Neg logit contributions | |
loo | -0.073 |
of | -0.068 |
ick | -0.063 |
auc | -0.061 |
iren | -0.061 |
Token | enough |
Token ID | 1576 |
Feature activation | +4.38e-01 |
Pos logit contributions | |
loo | +0.020 |
of | +0.020 |
ick | +0.018 |
pour | +0.016 |
ig | +0.016 |
Neg logit contributions | |
wrists | -0.033 |
chard | -0.032 |
blonde | -0.030 |
Eva | -0.030 |
Robert | -0.029 |
Token | it |
Token ID | 340 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
loo | +0.015 |
of | +0.013 |
ick | +0.012 |
iren | +0.012 |
auc | +0.012 |
Neg logit contributions | |
chard | -0.021 |
Eva | -0.021 |
wrists | -0.021 |
Robert | -0.021 |
blonde | -0.020 |
Token | disappeared |
Token ID | 12120 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
loo | +0.007 |
of | +0.006 |
ick | +0.006 |
iren | +0.006 |
auc | +0.006 |
Neg logit contributions | |
Eva | -0.008 |
Robert | -0.008 |
chard | -0.008 |
wrists | -0.008 |
blonde | -0.007 |
Token | into |
Token ID | 656 |
Feature activation | -4.59e-01 |
Pos logit contributions | |
loo | +0.066 |
iren | +0.056 |
auc | +0.053 |
ick | +0.053 |
of | +0.051 |
Neg logit contributions | |
chard | -0.091 |
Robert | -0.091 |
Eva | -0.090 |
wrists | -0.088 |
blonde | -0.087 |
Token | a |
Token ID | 257 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
Eva | +0.019 |
Robert | +0.019 |
chard | +0.019 |
wrists | +0.019 |
blonde | +0.018 |
Neg logit contributions | |
loo | -0.018 |
auc | -0.016 |
iren | -0.016 |
ick | -0.016 |
of | -0.015 |
Token | bur |
Token ID | 4356 |
Feature activation | +7.26e+00 |
Pos logit contributions | |
chard | +0.048 |
Eva | +0.048 |
Robert | +0.048 |
wrists | +0.048 |
blonde | +0.046 |
Neg logit contributions | |
loo | -0.038 |
ick | -0.033 |
of | -0.032 |
iren | -0.032 |
auc | -0.032 |
Token | row |
Token ID | 808 |
Feature activation | +4.57e+00 |
Pos logit contributions | |
loo | +0.174 |
auc | +0.143 |
shower | +0.136 |
iren | +0.135 |
pour | +0.134 |
Neg logit contributions | |
atively | -0.324 |
chard | -0.318 |
blonde | -0.314 |
wrists | -0.314 |
ya | -0.313 |
Token | . |
Token ID | 13 |
Feature activation | +3.26e-01 |
Pos logit contributions | |
loo | +0.154 |
iren | +0.133 |
arl | +0.126 |
ick | +0.125 |
auc | +0.120 |
Neg logit contributions | |
chard | -0.206 |
Robert | -0.200 |
Eva | -0.196 |
Claire | -0.190 |
wrists | -0.190 |
Token |
|
Token ID | 198 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
loo | +0.011 |
of | +0.011 |
late | +0.010 |
iren | +0.010 |
ick | +0.010 |
Neg logit contributions | |
wrists | -0.015 |
chard | -0.014 |
cious | -0.013 |
blonde | -0.013 |
Eva | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | +4.52e-02 |
Pos logit contributions | |
loo | +0.004 |
of | +0.003 |
iren | +0.003 |
ick | +0.003 |
auc | +0.003 |
Neg logit contributions | |
wrists | -0.006 |
chard | -0.006 |
Robert | -0.005 |
Eva | -0.005 |
blonde | -0.005 |
Token | The |
Token ID | 464 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
loo | +0.002 |
of | +0.002 |
iren | +0.002 |
ick | +0.002 |
auc | +0.001 |
Neg logit contributions | |
wrists | -0.002 |
chard | -0.002 |
Robert | -0.002 |
Eva | -0.002 |
blonde | -0.002 |
Token | peek |
Token ID | 27185 |
Feature activation | -9.04e-01 |
Pos logit contributions | |
chard | +0.059 |
Robert | +0.058 |
Eva | +0.058 |
wrists | +0.057 |
blonde | +0.055 |
Neg logit contributions | |
loo | -0.052 |
of | -0.048 |
ick | -0.045 |
auc | -0.044 |
iren | -0.044 |
Token | ed |
Token ID | 276 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
chard | +0.037 |
Robert | +0.037 |
Eva | +0.037 |
wrists | +0.037 |
blonde | +0.035 |
Neg logit contributions | |
loo | -0.036 |
of | -0.033 |
ick | -0.032 |
auc | -0.032 |
iren | -0.031 |
Token | out |
Token ID | 503 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
wrists | +0.087 |
chard | +0.087 |
Eva | +0.086 |
Robert | +0.085 |
blonde | +0.083 |
Neg logit contributions | |
loo | -0.053 |
of | -0.048 |
ick | -0.045 |
auc | -0.043 |
iren | -0.042 |
Token | of |
Token ID | 286 |
Feature activation | +3.12e+00 |
Pos logit contributions | |
Eva | +0.125 |
Robert | +0.122 |
Claire | +0.121 |
yes | +0.121 |
blonde | +0.118 |
Neg logit contributions | |
loo | -0.057 |
iren | -0.043 |
auc | -0.041 |
ick | -0.039 |
arl | -0.037 |
Token | his |
Token ID | 465 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
loo | +0.119 |
ick | +0.102 |
auc | +0.102 |
of | +0.101 |
iren | +0.100 |
Neg logit contributions | |
Eva | -0.136 |
Robert | -0.135 |
chard | -0.135 |
wrists | -0.133 |
blonde | -0.128 |
Token | bur |
Token ID | 4356 |
Feature activation | +7.25e+00 |
Pos logit contributions | |
wrists | +0.005 |
chard | +0.005 |
Robert | +0.005 |
Eva | +0.005 |
blonde | +0.005 |
Neg logit contributions | |
loo | -0.005 |
iren | -0.005 |
auc | -0.005 |
ick | -0.005 |
of | -0.005 |
Token | row |
Token ID | 808 |
Feature activation | +4.60e+00 |
Pos logit contributions | |
loo | +0.166 |
auc | +0.131 |
pour | +0.130 |
iren | +0.124 |
of | +0.118 |
Neg logit contributions | |
wrists | -0.394 |
chard | -0.376 |
blonde | -0.367 |
cious | -0.364 |
ya | -0.361 |
Token | . |
Token ID | 13 |
Feature activation | +5.01e-01 |
Pos logit contributions | |
loo | +0.158 |
iren | +0.132 |
auc | +0.128 |
ick | +0.126 |
arl | +0.124 |
Neg logit contributions | |
chard | -0.213 |
Robert | -0.210 |
Eva | -0.210 |
wrists | -0.207 |
blonde | -0.201 |
Token |
|
Token ID | 198 |
Feature activation | +2.01e-01 |
Pos logit contributions | |
loo | +0.018 |
of | +0.017 |
iren | +0.015 |
pour | +0.015 |
ick | +0.015 |
Neg logit contributions | |
wrists | -0.023 |
chard | -0.022 |
cious | -0.021 |
Eva | -0.020 |
blonde | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
loo | +0.007 |
of | +0.007 |
ick | +0.006 |
iren | +0.006 |
auc | +0.006 |
Neg logit contributions | |
wrists | -0.009 |
chard | -0.009 |
Robert | -0.009 |
Eva | -0.009 |
blonde | -0.009 |
Token | He |
Token ID | 1544 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
loo | +0.006 |
ick | +0.005 |
of | +0.005 |
auc | +0.005 |
iren | +0.005 |
Neg logit contributions | |
Eva | -0.008 |
Robert | -0.008 |
chard | -0.008 |
wrists | -0.007 |
blonde | -0.007 |
Token | ies |
Token ID | 444 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
loo | +0.051 |
iren | +0.045 |
pour | +0.041 |
auc | +0.039 |
arl | +0.037 |
Neg logit contributions | |
wrists | -0.097 |
Eva | -0.097 |
Claire | -0.096 |
chard | -0.096 |
cious | -0.095 |
Token | lived |
Token ID | 5615 |
Feature activation | +8.36e-01 |
Pos logit contributions | |
chard | +0.004 |
Robert | +0.004 |
Eva | +0.004 |
wrists | +0.004 |
blonde | +0.004 |
Neg logit contributions | |
loo | -0.008 |
iren | -0.007 |
ick | -0.007 |
auc | -0.007 |
arl | -0.007 |
Token | in |
Token ID | 287 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
loo | +0.026 |
of | +0.022 |
ick | +0.022 |
iren | +0.021 |
auc | +0.021 |
Neg logit contributions | |
Eva | -0.041 |
Robert | -0.041 |
wrists | -0.041 |
chard | -0.041 |
blonde | -0.040 |
Token | a |
Token ID | 257 |
Feature activation | -7.79e-01 |
Pos logit contributions | |
Eva | +0.083 |
Robert | +0.082 |
chard | +0.080 |
wrists | +0.078 |
blonde | +0.078 |
Neg logit contributions | |
loo | -0.111 |
auc | -0.097 |
iren | -0.097 |
ick | -0.094 |
of | -0.093 |
Token | cozy |
Token ID | 37438 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
chard | +0.033 |
Eva | +0.033 |
Robert | +0.032 |
wrists | +0.032 |
blonde | +0.031 |
Neg logit contributions | |
loo | -0.030 |
iren | -0.026 |
of | -0.026 |
auc | -0.026 |
ick | -0.026 |
Token | bur |
Token ID | 4356 |
Feature activation | +7.24e+00 |
Pos logit contributions | |
loo | +0.069 |
auc | +0.065 |
iren | +0.063 |
ick | +0.056 |
arl | +0.055 |
Neg logit contributions | |
chard | -0.085 |
wrists | -0.084 |
Claire | -0.084 |
Eva | -0.083 |
Robert | -0.083 |
Token | row |
Token ID | 808 |
Feature activation | +4.76e+00 |
Pos logit contributions | |
loo | +0.154 |
pour | +0.138 |
auc | +0.134 |
iren | +0.125 |
accidents | +0.121 |
Neg logit contributions | |
chard | -0.379 |
wrists | -0.368 |
cious | -0.367 |
blonde | -0.358 |
ya | -0.352 |
Token | near |
Token ID | 1474 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
loo | +0.157 |
iren | +0.134 |
auc | +0.128 |
ick | +0.127 |
arl | +0.125 |
Neg logit contributions | |
chard | -0.229 |
Robert | -0.222 |
Eva | -0.221 |
wrists | -0.219 |
blonde | -0.212 |
Token | a |
Token ID | 257 |
Feature activation | -7.72e-01 |
Pos logit contributions | |
wrists | +0.064 |
chard | +0.064 |
Robert | +0.064 |
Eva | +0.064 |
blonde | +0.062 |
Neg logit contributions | |
loo | -0.044 |
of | -0.041 |
ick | -0.039 |
ig | -0.037 |
auc | -0.036 |
Token | small |
Token ID | 1402 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
chard | +0.036 |
Eva | +0.035 |
Robert | +0.035 |
wrists | +0.035 |
blonde | +0.034 |
Neg logit contributions | |
loo | -0.027 |
of | -0.023 |
ick | -0.023 |
iren | -0.023 |
auc | -0.022 |
Token | lake |
Token ID | 13546 |
Feature activation | -5.57e-01 |
Pos logit contributions | |
loo | +0.067 |
iren | +0.059 |
auc | +0.058 |
ick | +0.056 |
arl | +0.055 |
Neg logit contributions | |
chard | -0.085 |
Robert | -0.081 |
Eva | -0.081 |
wrists | -0.081 |
Claire | -0.079 |
Token | until |
Token ID | 1566 |
Feature activation | +4.06e-01 |
Pos logit contributions | |
Eva | +0.056 |
chard | +0.056 |
Robert | +0.056 |
wrists | +0.056 |
blonde | +0.054 |
Neg logit contributions | |
loo | -0.044 |
of | -0.038 |
ick | -0.037 |
iren | -0.037 |
auc | -0.036 |
Token | it |
Token ID | 340 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
loo | +0.015 |
of | +0.013 |
ick | +0.013 |
iren | +0.013 |
auc | +0.013 |
Neg logit contributions | |
Eva | -0.017 |
chard | -0.017 |
Robert | -0.017 |
wrists | -0.017 |
blonde | -0.017 |
Token | reached |
Token ID | 4251 |
Feature activation | -4.93e-01 |
Pos logit contributions | |
loo | +0.008 |
of | +0.007 |
ick | +0.007 |
iren | +0.007 |
auc | +0.007 |
Neg logit contributions | |
Eva | -0.008 |
chard | -0.008 |
Robert | -0.008 |
wrists | -0.008 |
blonde | -0.008 |
Token | its |
Token ID | 663 |
Feature activation | -7.77e-01 |
Pos logit contributions | |
chard | +0.023 |
wrists | +0.023 |
Eva | +0.022 |
Robert | +0.022 |
blonde | +0.022 |
Neg logit contributions | |
loo | -0.017 |
of | -0.015 |
ick | -0.015 |
iren | -0.014 |
auc | -0.014 |
Token | cozy |
Token ID | 37438 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
Robert | +0.037 |
Eva | +0.037 |
chard | +0.036 |
wrists | +0.036 |
blonde | +0.035 |
Neg logit contributions | |
loo | -0.026 |
of | -0.023 |
ick | -0.022 |
iren | -0.021 |
auc | -0.021 |
Token | bur |
Token ID | 4356 |
Feature activation | +7.24e+00 |
Pos logit contributions | |
loo | +0.057 |
iren | +0.053 |
auc | +0.053 |
ick | +0.047 |
arl | +0.047 |
Neg logit contributions | |
wrists | -0.079 |
blonde | -0.078 |
chard | -0.078 |
Eva | -0.077 |
Robert | -0.077 |
Token | row |
Token ID | 808 |
Feature activation | +4.63e+00 |
Pos logit contributions | |
loo | +0.158 |
iren | +0.126 |
auc | +0.125 |
pour | +0.114 |
ick | +0.104 |
Neg logit contributions | |
chard | -0.383 |
wrists | -0.368 |
atively | -0.362 |
blonde | -0.362 |
rice | -0.356 |
Token | . |
Token ID | 13 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
loo | +0.154 |
iren | +0.129 |
auc | +0.124 |
ick | +0.123 |
arl | +0.118 |
Neg logit contributions | |
chard | -0.220 |
Robert | -0.216 |
Eva | -0.215 |
wrists | -0.212 |
blonde | -0.208 |
Token | The |
Token ID | 383 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
loo | +0.006 |
of | +0.006 |
ick | +0.006 |
pour | +0.006 |
late | +0.006 |
Neg logit contributions | |
wrists | -0.007 |
chard | -0.007 |
blonde | -0.006 |
Eva | -0.006 |
cious | -0.006 |
Token | end |
Token ID | 886 |
Feature activation | +3.84e+00 |
Pos logit contributions | |
Eva | +0.001 |
wrists | +0.001 |
Robert | +0.001 |
chard | +0.001 |
blonde | +0.001 |
Neg logit contributions | |
loo | -0.001 |
of | -0.001 |
ick | -0.001 |
iren | -0.001 |
auc | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +2.20e-01 |
Pos logit contributions | |
loo | +0.136 |
arl | +0.119 |
auc | +0.119 |
iren | +0.117 |
ick | +0.114 |
Neg logit contributions | |
Eva | -0.166 |
Robert | -0.163 |
blonde | -0.160 |
aloud | -0.154 |
rice | -0.153 |
Token | he |
Token ID | 339 |
Feature activation | +5.64e-01 |
Pos logit contributions | |
loo | +0.007 |
of | +0.006 |
ick | +0.006 |
iren | +0.006 |
auc | +0.006 |
Neg logit contributions | |
wrists | -0.009 |
Eva | -0.009 |
chard | -0.009 |
Robert | -0.009 |
blonde | -0.009 |
Token | finally |
Token ID | 3443 |
Feature activation | +3.65e-01 |
Pos logit contributions | |
loo | +0.021 |
of | +0.020 |
ick | +0.018 |
pour | +0.017 |
auc | +0.017 |
Neg logit contributions | |
wrists | -0.025 |
chard | -0.024 |
Robert | -0.024 |
blonde | -0.024 |
Eva | -0.023 |
Token | reached |
Token ID | 4251 |
Feature activation | -6.28e-01 |
Pos logit contributions | |
loo | +0.015 |
of | +0.014 |
ick | +0.013 |
iren | +0.013 |
auc | +0.013 |
Neg logit contributions | |
wrists | -0.015 |
chard | -0.015 |
Robert | -0.014 |
Eva | -0.014 |
blonde | -0.014 |
Token | his |
Token ID | 465 |
Feature activation | -3.65e-01 |
Pos logit contributions | |
wrists | +0.030 |
chard | +0.030 |
Eva | +0.030 |
Robert | +0.030 |
blonde | +0.029 |
Neg logit contributions | |
loo | -0.021 |
of | -0.018 |
ick | -0.017 |
iren | -0.016 |
auc | -0.016 |
Token | cozy |
Token ID | 37438 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
chard | +0.017 |
Robert | +0.017 |
Eva | +0.017 |
wrists | +0.016 |
blonde | +0.016 |
Neg logit contributions | |
loo | -0.012 |
of | -0.011 |
ig | -0.010 |
ick | -0.010 |
tunnel | -0.010 |
Token | bur |
Token ID | 4356 |
Feature activation | +7.23e+00 |
Pos logit contributions | |
loo | +0.051 |
auc | +0.045 |
iren | +0.045 |
ick | +0.042 |
of | +0.041 |
Neg logit contributions | |
wrists | -0.078 |
Eva | -0.077 |
Robert | -0.077 |
chard | -0.076 |
blonde | -0.075 |
Token | row |
Token ID | 808 |
Feature activation | +4.58e+00 |
Pos logit contributions | |
loo | +0.154 |
auc | +0.120 |
iren | +0.118 |
pour | +0.111 |
ick | +0.103 |
Neg logit contributions | |
chard | -0.400 |
wrists | -0.392 |
Eva | -0.382 |
blonde | -0.376 |
Robert | -0.376 |
Token | . |
Token ID | 13 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
loo | +0.150 |
iren | +0.126 |
ick | +0.122 |
auc | +0.121 |
arl | +0.117 |
Neg logit contributions | |
chard | -0.219 |
Robert | -0.218 |
Eva | -0.216 |
wrists | -0.211 |
blonde | -0.205 |
Token | From |
Token ID | 3574 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
loo | +0.004 |
of | +0.004 |
iren | +0.004 |
ick | +0.004 |
late | +0.004 |
Neg logit contributions | |
wrists | -0.006 |
chard | -0.006 |
cious | -0.005 |
blonde | -0.005 |
nam | -0.005 |
Token | then |
Token ID | 788 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
Eva | +0.012 |
wrists | +0.012 |
chard | +0.012 |
Robert | +0.012 |
blonde | +0.011 |
Neg logit contributions | |
loo | -0.010 |
of | -0.009 |
ick | -0.008 |
iren | -0.008 |
auc | -0.008 |
Token | on |
Token ID | 319 |
Feature activation | -2.14e+00 |
Pos logit contributions | |
Robert | +0.060 |
Eva | +0.059 |
blonde | +0.057 |
yes | +0.056 |
wrists | +0.056 |
Neg logit contributions | |
loo | -0.045 |
auc | -0.041 |
rim | -0.038 |
outage | -0.037 |
iren | -0.036 |
Token | Benny |
Token ID | 44275 |
Feature activation | -9.46e-01 |
Pos logit contributions | |
loo | +0.022 |
of | +0.021 |
ick | +0.020 |
auc | +0.019 |
iren | +0.019 |
Neg logit contributions | |
wrists | -0.025 |
Eva | -0.025 |
chard | -0.024 |
Robert | -0.023 |
cious | -0.023 |
Token | lived |
Token ID | 5615 |
Feature activation | +8.93e-01 |
Pos logit contributions | |
chard | +0.038 |
Eva | +0.038 |
wrists | +0.038 |
Robert | +0.037 |
blonde | +0.035 |
Neg logit contributions | |
loo | -0.036 |
of | -0.035 |
ick | -0.033 |
ig | -0.032 |
tunnel | -0.031 |
Token | in |
Token ID | 287 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
loo | +0.030 |
of | +0.025 |
ick | +0.025 |
iren | +0.024 |
auc | +0.024 |
Neg logit contributions | |
chard | -0.043 |
Eva | -0.043 |
Robert | -0.042 |
wrists | -0.042 |
blonde | -0.041 |
Token | a |
Token ID | 257 |
Feature activation | -8.74e-01 |
Pos logit contributions | |
Robert | +0.083 |
Eva | +0.081 |
chard | +0.078 |
blonde | +0.076 |
Claire | +0.074 |
Neg logit contributions | |
loo | -0.122 |
auc | -0.113 |
iren | -0.109 |
arl | -0.107 |
row | -0.104 |
Token | cozy |
Token ID | 37438 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
chard | +0.038 |
Eva | +0.037 |
Robert | +0.037 |
wrists | +0.036 |
blonde | +0.036 |
Neg logit contributions | |
loo | -0.034 |
iren | -0.029 |
auc | -0.029 |
of | -0.029 |
ick | -0.029 |
Token | bur |
Token ID | 4356 |
Feature activation | +7.23e+00 |
Pos logit contributions | |
loo | +0.060 |
auc | +0.058 |
iren | +0.056 |
ick | +0.049 |
arl | +0.049 |
Neg logit contributions | |
chard | -0.079 |
Robert | -0.078 |
wrists | -0.077 |
Eva | -0.077 |
Claire | -0.076 |
Token | row |
Token ID | 808 |
Feature activation | +4.65e+00 |
Pos logit contributions | |
loo | +0.148 |
auc | +0.139 |
iren | +0.128 |
pour | +0.126 |
ick | +0.111 |
Neg logit contributions | |
chard | -0.405 |
wrists | -0.375 |
blonde | -0.370 |
cious | -0.359 |
ya | -0.355 |
Token | with |
Token ID | 351 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
loo | +0.161 |
iren | +0.139 |
auc | +0.134 |
ick | +0.131 |
arl | +0.129 |
Neg logit contributions | |
chard | -0.219 |
Robert | -0.210 |
Eva | -0.208 |
wrists | -0.205 |
blonde | -0.203 |
Token | his |
Token ID | 465 |
Feature activation | +5.86e-02 |
Pos logit contributions | |
wrists | +0.096 |
Eva | +0.095 |
Robert | +0.093 |
chard | +0.093 |
rice | +0.089 |
Neg logit contributions | |
loo | -0.075 |
of | -0.068 |
ick | -0.064 |
auc | -0.060 |
iren | -0.060 |
Token | family |
Token ID | 1641 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
loo | +0.002 |
iren | +0.002 |
auc | +0.002 |
of | +0.002 |
ick | +0.002 |
Neg logit contributions | |
chard | -0.003 |
wrists | -0.003 |
Robert | -0.003 |
Eva | -0.003 |
blonde | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +6.97e-01 |
Pos logit contributions | |
loo | +0.076 |
iren | +0.064 |
auc | +0.063 |
ick | +0.061 |
arl | +0.060 |
Neg logit contributions | |
chard | -0.113 |
Robert | -0.112 |
wrists | -0.111 |
Eva | -0.110 |
blonde | -0.107 |
Token | was |
Token ID | 373 |
Feature activation | +6.88e-01 |
Pos logit contributions | |
loo | +0.214 |
ick | +0.183 |
iren | +0.183 |
auc | +0.181 |
of | +0.179 |
Neg logit contributions | |
Eva | -0.200 |
wrists | -0.199 |
Robert | -0.197 |
chard | -0.197 |
blonde | -0.191 |
Token | too |
Token ID | 1165 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
loo | +0.029 |
auc | +0.024 |
iren | +0.023 |
ick | +0.023 |
arl | +0.022 |
Neg logit contributions | |
chard | -0.029 |
Eva | -0.028 |
Robert | -0.028 |
blonde | -0.027 |
wrists | -0.027 |
Token | strong |
Token ID | 1913 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
loo | +0.060 |
auc | +0.048 |
iren | +0.048 |
ig | +0.045 |
rim | +0.044 |
Neg logit contributions | |
chard | -0.061 |
Eva | -0.058 |
blonde | -0.058 |
Claire | -0.056 |
Robert | -0.056 |
Token | and |
Token ID | 290 |
Feature activation | -2.81e-01 |
Pos logit contributions | |
loo | +0.045 |
auc | +0.036 |
rim | +0.036 |
pour | +0.035 |
iren | +0.035 |
Neg logit contributions | |
Robert | -0.044 |
Eva | -0.041 |
Stella | -0.039 |
Claire | -0.038 |
blonde | -0.038 |
Token | the |
Token ID | 262 |
Feature activation | +3.28e-01 |
Pos logit contributions | |
chard | +0.012 |
Eva | +0.012 |
Robert | +0.012 |
wrists | +0.012 |
blonde | +0.011 |
Neg logit contributions | |
loo | -0.011 |
ick | -0.010 |
of | -0.009 |
iren | -0.009 |
auc | -0.009 |
Token | bur |
Token ID | 4356 |
Feature activation | +7.22e+00 |
Pos logit contributions | |
loo | +0.009 |
of | +0.008 |
ick | +0.007 |
iren | +0.007 |
auc | +0.007 |
Neg logit contributions | |
Eva | -0.017 |
Robert | -0.017 |
chard | -0.017 |
wrists | -0.017 |
blonde | -0.017 |
Token | row |
Token ID | 808 |
Feature activation | +4.66e+00 |
Pos logit contributions | |
loo | +0.202 |
iren | +0.160 |
ig | +0.158 |
auc | +0.157 |
pour | +0.150 |
Neg logit contributions | |
wrists | -0.333 |
atively | -0.332 |
chard | -0.327 |
rice | -0.321 |
cious | -0.319 |
Token | started |
Token ID | 2067 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
loo | +0.188 |
iren | +0.160 |
ick | +0.157 |
auc | +0.156 |
of | +0.155 |
Neg logit contributions | |
chard | -0.185 |
wrists | -0.185 |
Eva | -0.184 |
Robert | -0.183 |
blonde | -0.179 |
Token | to |
Token ID | 284 |
Feature activation | -1.73e-01 |
Pos logit contributions | |
loo | +0.060 |
iren | +0.048 |
auc | +0.048 |
ick | +0.047 |
arl | +0.046 |
Neg logit contributions | |
wrists | -0.066 |
chard | -0.065 |
Eva | -0.064 |
blonde | -0.064 |
Robert | -0.063 |
Token | fill |
Token ID | 6070 |
Feature activation | +3.58e-01 |
Pos logit contributions | |
chard | +0.008 |
Eva | +0.008 |
wrists | +0.008 |
Robert | +0.008 |
blonde | +0.008 |
Neg logit contributions | |
loo | -0.006 |
iren | -0.005 |
auc | -0.005 |
of | -0.005 |
ick | -0.005 |
Token | with |
Token ID | 351 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
loo | +0.012 |
of | +0.011 |
ick | +0.011 |
auc | +0.010 |
pour | +0.010 |
Neg logit contributions | |
chard | -0.016 |
Robert | -0.016 |
Eva | -0.016 |
wrists | -0.016 |
blonde | -0.016 |
Token | from |
Token ID | 422 |
Feature activation | -5.14e-01 |
Pos logit contributions | |
loo | +0.071 |
arl | +0.056 |
pour | +0.055 |
iren | +0.055 |
ick | +0.053 |
Neg logit contributions | |
yes | -0.079 |
wrists | -0.074 |
chard | -0.074 |
Robert | -0.073 |
Eva | -0.071 |
Token | the |
Token ID | 262 |
Feature activation | +4.23e-01 |
Pos logit contributions | |
chard | +0.023 |
Robert | +0.023 |
Eva | +0.023 |
wrists | +0.022 |
blonde | +0.021 |
Neg logit contributions | |
loo | -0.019 |
of | -0.016 |
ick | -0.016 |
auc | -0.016 |
iren | -0.016 |
Token | river |
Token ID | 7850 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
loo | +0.014 |
of | +0.012 |
ick | +0.012 |
iren | +0.011 |
auc | +0.011 |
Neg logit contributions | |
chard | -0.020 |
Robert | -0.020 |
Eva | -0.020 |
wrists | -0.020 |
blonde | -0.019 |
Token | to |
Token ID | 284 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
loo | +0.040 |
ick | +0.033 |
iren | +0.032 |
auc | +0.031 |
arl | +0.031 |
Neg logit contributions | |
chard | -0.058 |
Robert | -0.056 |
Eva | -0.056 |
wrists | -0.056 |
Claire | -0.053 |
Token | their |
Token ID | 511 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
chard | +0.006 |
Eva | +0.006 |
Robert | +0.006 |
Claire | +0.005 |
wrists | +0.005 |
Neg logit contributions | |
loo | -0.007 |
iren | -0.006 |
auc | -0.006 |
ick | -0.006 |
arl | -0.006 |
Token | bur |
Token ID | 4356 |
Feature activation | +7.22e+00 |
Pos logit contributions | |
chard | +0.002 |
wrists | +0.002 |
Robert | +0.002 |
Eva | +0.002 |
blonde | +0.002 |
Neg logit contributions | |
loo | -0.001 |
of | -0.001 |
ick | -0.001 |
iren | -0.001 |
auc | -0.001 |
Token | row |
Token ID | 808 |
Feature activation | +4.82e+00 |
Pos logit contributions | |
loo | +0.174 |
pour | +0.149 |
accidents | +0.148 |
auc | +0.141 |
iren | +0.135 |
Neg logit contributions | |
chard | -0.341 |
wrists | -0.320 |
rice | -0.300 |
atively | -0.299 |
ya | -0.299 |
Token | . |
Token ID | 13 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
loo | +0.163 |
iren | +0.137 |
ick | +0.132 |
arl | +0.129 |
auc | +0.129 |
Neg logit contributions | |
chard | -0.229 |
Robert | -0.219 |
Eva | -0.217 |
wrists | -0.214 |
Claire | -0.208 |
Token | |
Token ID | 220 |
Feature activation | -2.47e+00 |
Pos logit contributions | |
loo | +0.004 |
of | +0.004 |
iren | +0.004 |
ick | +0.004 |
pour | +0.004 |
Neg logit contributions | |
wrists | -0.005 |
chard | -0.005 |
cious | -0.005 |
Eva | -0.005 |
blonde | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
chard | +0.106 |
wrists | +0.106 |
Eva | +0.106 |
Robert | +0.106 |
blonde | +0.102 |
Neg logit contributions | |
loo | -0.094 |
of | -0.083 |
ick | -0.081 |
iren | -0.079 |
auc | -0.078 |
Token |
|
Token ID | 198 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
loo | +0.007 |
of | +0.006 |
iren | +0.006 |
ick | +0.006 |
auc | +0.005 |
Neg logit contributions | |
wrists | -0.009 |
chard | -0.009 |
Eva | -0.009 |
Robert | -0.009 |
blonde | -0.009 |
Token | he |
Token ID | 339 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
Robert | +0.072 |
Eva | +0.072 |
chard | +0.072 |
wrists | +0.071 |
blonde | +0.068 |
Neg logit contributions | |
loo | -0.063 |
of | -0.054 |
ick | -0.054 |
iren | -0.053 |
auc | -0.053 |
Token | could |
Token ID | 714 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
of | +0.032 |
loo | +0.031 |
ick | +0.029 |
row | +0.026 |
pour | +0.026 |
Neg logit contributions | |
Robert | -0.050 |
Eva | -0.050 |
wrists | -0.049 |
chard | -0.049 |
Claire | -0.047 |
Token | back |
Token ID | 736 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
wrists | +0.017 |
chard | +0.017 |
Robert | +0.017 |
Eva | +0.017 |
blonde | +0.016 |
Neg logit contributions | |
loo | -0.009 |
of | -0.008 |
ick | -0.007 |
iren | -0.007 |
auc | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
Eva | +0.048 |
chard | +0.048 |
Robert | +0.048 |
wrists | +0.046 |
Joy | +0.045 |
Neg logit contributions | |
loo | -0.049 |
auc | -0.046 |
rim | -0.045 |
iren | -0.045 |
ick | -0.044 |
Token | his |
Token ID | 465 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
chard | +0.008 |
wrists | +0.008 |
Robert | +0.008 |
Eva | +0.008 |
blonde | +0.008 |
Neg logit contributions | |
loo | -0.006 |
of | -0.005 |
ick | -0.005 |
iren | -0.005 |
ig | -0.005 |
Token | bur |
Token ID | 4356 |
Feature activation | +7.18e+00 |
Pos logit contributions | |
Eva | +0.001 |
Robert | +0.001 |
chard | +0.001 |
wrists | +0.001 |
blonde | +0.001 |
Neg logit contributions | |
loo | -0.000 |
of | -0.000 |
ick | -0.000 |
iren | -0.000 |
auc | -0.000 |
Token | row |
Token ID | 808 |
Feature activation | +4.56e+00 |
Pos logit contributions | |
loo | +0.167 |
auc | +0.135 |
iren | +0.125 |
pour | +0.122 |
ick | +0.119 |
Neg logit contributions | |
chard | -0.367 |
wrists | -0.359 |
blonde | -0.347 |
Eva | -0.342 |
rice | -0.341 |
Token | . |
Token ID | 13 |
Feature activation | +5.30e-01 |
Pos logit contributions | |
loo | +0.160 |
iren | +0.135 |
auc | +0.131 |
arl | +0.129 |
ick | +0.129 |
Neg logit contributions | |
chard | -0.208 |
Robert | -0.204 |
Eva | -0.202 |
wrists | -0.198 |
blonde | -0.195 |
Token |
|
Token ID | 198 |
Feature activation | +4.02e-01 |
Pos logit contributions | |
loo | +0.018 |
of | +0.017 |
ick | +0.016 |
iren | +0.016 |
late | +0.015 |
Neg logit contributions | |
wrists | -0.026 |
chard | -0.024 |
Eva | -0.023 |
blonde | -0.023 |
cious | -0.022 |
Token |
|
Token ID | 198 |
Feature activation | +3.66e-01 |
Pos logit contributions | |
loo | +0.014 |
of | +0.013 |
iren | +0.012 |
ick | +0.012 |
auc | +0.011 |
Neg logit contributions | |
wrists | -0.019 |
chard | -0.019 |
Eva | -0.018 |
Robert | -0.018 |
blonde | -0.018 |
Token | When |
Token ID | 2215 |
Feature activation | -3.20e-01 |
Pos logit contributions | |
loo | +0.016 |
of | +0.015 |
ick | +0.014 |
iren | +0.014 |
auc | +0.014 |
Neg logit contributions | |
Eva | -0.013 |
chard | -0.013 |
Robert | -0.013 |
wrists | -0.013 |
blonde | -0.013 |
Token | got |
Token ID | 1392 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
loo | +0.028 |
of | +0.026 |
ick | +0.025 |
auc | +0.024 |
iren | +0.024 |
Neg logit contributions | |
chard | -0.041 |
Robert | -0.040 |
Eva | -0.040 |
wrists | -0.039 |
blonde | -0.038 |
Token | m |
Token ID | 285 |
Feature activation | +6.71e-01 |
Pos logit contributions | |
loo | +0.076 |
of | +0.067 |
ick | +0.065 |
iren | +0.064 |
auc | +0.064 |
Neg logit contributions | |
Eva | -0.096 |
Robert | -0.095 |
chard | -0.095 |
wrists | -0.095 |
blonde | -0.091 |
Token | - |
Token ID | 12 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
loo | +0.021 |
tunnel | +0.017 |
pour | +0.016 |
auc | +0.016 |
of | +0.016 |
Neg logit contributions | |
wrists | -0.032 |
blonde | -0.032 |
Eva | -0.032 |
cious | -0.031 |
Robert | -0.031 |
Token | mad |
Token ID | 9937 |
Feature activation | -4.04e-01 |
Pos logit contributions | |
loo | +0.046 |
of | +0.041 |
iren | +0.040 |
ick | +0.039 |
auc | +0.039 |
Neg logit contributions | |
Robert | -0.065 |
wrists | -0.065 |
Eva | -0.065 |
chard | -0.064 |
blonde | -0.062 |
Token | and |
Token ID | 290 |
Feature activation | -6.76e-02 |
Pos logit contributions | |
Eva | +0.019 |
wrists | +0.019 |
Robert | +0.019 |
blonde | +0.019 |
chard | +0.018 |
Neg logit contributions | |
loo | -0.014 |
ick | -0.011 |
auc | -0.011 |
iren | -0.011 |
of | -0.011 |
Token | r |
Token ID | 374 |
Feature activation | +7.18e+00 |
Pos logit contributions | |
chard | +0.003 |
Robert | +0.003 |
Eva | +0.003 |
wrists | +0.003 |
blonde | +0.003 |
Neg logit contributions | |
loo | -0.002 |
of | -0.002 |
iren | -0.002 |
ick | -0.002 |
auc | -0.002 |
Token | - |
Token ID | 12 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
loo | +0.220 |
of | +0.201 |
tunnel | +0.197 |
storm | +0.188 |
outage | +0.185 |
Neg logit contributions | |
wrists | -0.334 |
blonde | -0.322 |
Eva | -0.314 |
chard | -0.307 |
blond | -0.302 |
Token | ran |
Token ID | 2596 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
loo | +0.042 |
of | +0.036 |
iren | +0.036 |
ick | +0.035 |
auc | +0.035 |
Neg logit contributions | |
Robert | -0.060 |
Eva | -0.060 |
wrists | -0.059 |
chard | -0.059 |
blonde | -0.057 |
Token | away |
Token ID | 1497 |
Feature activation | +4.62e-01 |
Pos logit contributions | |
chard | +0.056 |
Eva | +0.054 |
Robert | +0.054 |
wrists | +0.053 |
Claire | +0.051 |
Neg logit contributions | |
loo | -0.036 |
of | -0.033 |
ick | -0.033 |
iren | -0.032 |
auc | -0.030 |
Token | and |
Token ID | 290 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
loo | +0.015 |
iren | +0.012 |
auc | +0.012 |
ick | +0.012 |
of | +0.011 |
Neg logit contributions | |
Eva | -0.022 |
wrists | -0.022 |
Robert | -0.022 |
chard | -0.022 |
blonde | -0.021 |
Token | I |
Token ID | 314 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
chard | +0.008 |
Eva | +0.008 |
Robert | +0.008 |
wrists | +0.008 |
blonde | +0.008 |
Neg logit contributions | |
loo | -0.008 |
of | -0.007 |
ick | -0.007 |
iren | -0.007 |
auc | -0.007 |
Token | fluffy |
Token ID | 39145 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
loo | +0.001 |
of | +0.001 |
ick | +0.001 |
iren | +0.001 |
auc | +0.001 |
Neg logit contributions | |
Eva | -0.001 |
Robert | -0.001 |
wrists | -0.001 |
chard | -0.001 |
blonde | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
loo | +0.006 |
of | +0.005 |
ick | +0.005 |
ig | +0.004 |
iren | +0.004 |
Neg logit contributions | |
Eva | -0.014 |
chard | -0.014 |
wrists | -0.014 |
Robert | -0.014 |
blonde | -0.013 |
Token | lived |
Token ID | 5615 |
Feature activation | +8.71e-01 |
Pos logit contributions | |
Eva | +0.001 |
wrists | +0.001 |
Robert | +0.001 |
chard | +0.001 |
blonde | +0.001 |
Neg logit contributions | |
loo | -0.001 |
of | -0.001 |
ick | -0.001 |
ig | -0.001 |
iren | -0.001 |
Token | in |
Token ID | 287 |
Feature activation | -2.52e+00 |
Pos logit contributions | |
loo | +0.025 |
of | +0.022 |
ick | +0.021 |
iren | +0.020 |
auc | +0.019 |
Neg logit contributions | |
wrists | -0.045 |
Eva | -0.045 |
Robert | -0.045 |
chard | -0.044 |
blonde | -0.043 |
Token | a |
Token ID | 257 |
Feature activation | -9.00e-01 |
Pos logit contributions | |
Robert | +0.088 |
Eva | +0.088 |
chard | +0.086 |
blonde | +0.083 |
wrists | +0.081 |
Neg logit contributions | |
loo | -0.119 |
auc | -0.107 |
iren | -0.104 |
ick | -0.103 |
arl | -0.102 |
Token | bur |
Token ID | 4356 |
Feature activation | +7.18e+00 |
Pos logit contributions | |
chard | +0.039 |
Eva | +0.039 |
Robert | +0.038 |
wrists | +0.038 |
blonde | +0.037 |
Neg logit contributions | |
loo | -0.034 |
auc | -0.029 |
ick | -0.029 |
iren | -0.029 |
of | -0.029 |
Token | row |
Token ID | 808 |
Feature activation | +4.74e+00 |
Pos logit contributions | |
loo | +0.163 |
auc | +0.154 |
pour | +0.141 |
accidents | +0.139 |
iren | +0.134 |
Neg logit contributions | |
chard | -0.367 |
wrists | -0.341 |
blonde | -0.337 |
cious | -0.337 |
ya | -0.331 |
Token | in |
Token ID | 287 |
Feature activation | -2.46e+00 |
Pos logit contributions | |
loo | +0.159 |
iren | +0.136 |
auc | +0.130 |
ick | +0.128 |
arl | +0.128 |
Neg logit contributions | |
chard | -0.229 |
Robert | -0.216 |
Eva | -0.214 |
wrists | -0.213 |
blonde | -0.209 |
Token | the |
Token ID | 262 |
Feature activation | +6.70e-01 |
Pos logit contributions | |
Eva | +0.089 |
Robert | +0.089 |
chard | +0.088 |
Claire | +0.083 |
blonde | +0.082 |
Neg logit contributions | |
loo | -0.111 |
auc | -0.104 |
iren | -0.101 |
arl | -0.100 |
ick | -0.098 |
Token | forest |
Token ID | 8222 |
Feature activation | +9.07e-01 |
Pos logit contributions | |
loo | +0.025 |
ick | +0.022 |
iren | +0.021 |
of | +0.021 |
auc | +0.021 |
Neg logit contributions | |
chard | -0.030 |
Eva | -0.028 |
blonde | -0.028 |
Robert | -0.028 |
wrists | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | +8.07e-01 |
Pos logit contributions | |
loo | +0.027 |
ick | +0.022 |
iren | +0.022 |
of | +0.022 |
auc | +0.022 |
Neg logit contributions | |
chard | -0.046 |
Eva | -0.046 |
Robert | -0.046 |
wrists | -0.045 |
blonde | -0.044 |
Token | who |
Token ID | 508 |
Feature activation | -4.60e+00 |
Pos logit contributions | |
Eva | +0.108 |
chard | +0.108 |
Robert | +0.108 |
wrists | +0.107 |
blonde | +0.103 |
Neg logit contributions | |
loo | -0.101 |
of | -0.089 |
ick | -0.087 |
iren | -0.086 |
auc | -0.085 |
Token | lived |
Token ID | 5615 |
Feature activation | +7.54e-01 |
Pos logit contributions | |
wrists | +0.181 |
chard | +0.181 |
Eva | +0.178 |
Robert | +0.178 |
blonde | +0.163 |
Neg logit contributions | |
loo | -0.188 |
of | -0.180 |
ick | -0.174 |
auc | -0.162 |
iren | -0.158 |
Token | in |
Token ID | 287 |
Feature activation | -2.39e+00 |
Pos logit contributions | |
loo | +0.023 |
of | +0.020 |
ick | +0.019 |
iren | +0.018 |
auc | +0.018 |
Neg logit contributions | |
Eva | -0.038 |
Robert | -0.038 |
wrists | -0.038 |
chard | -0.038 |
blonde | -0.037 |
Token | the |
Token ID | 262 |
Feature activation | +7.18e-01 |
Pos logit contributions | |
wrists | +0.103 |
chard | +0.102 |
Robert | +0.100 |
Eva | +0.100 |
blonde | +0.096 |
Neg logit contributions | |
loo | -0.088 |
of | -0.081 |
ick | -0.079 |
iren | -0.074 |
auc | -0.073 |
Token | same |
Token ID | 976 |
Feature activation | +6.19e-01 |
Pos logit contributions | |
loo | +0.023 |
of | +0.020 |
ick | +0.019 |
iren | +0.018 |
auc | +0.018 |
Neg logit contributions | |
wrists | -0.035 |
Eva | -0.035 |
Robert | -0.035 |
chard | -0.035 |
blonde | -0.033 |
Token | bur |
Token ID | 4356 |
Feature activation | +7.17e+00 |
Pos logit contributions | |
loo | +0.023 |
ick | +0.019 |
of | +0.019 |
iren | +0.019 |
auc | +0.019 |
Neg logit contributions | |
Eva | -0.027 |
chard | -0.027 |
Robert | -0.027 |
wrists | -0.027 |
blonde | -0.026 |
Token | row |
Token ID | 808 |
Feature activation | +4.68e+00 |
Pos logit contributions | |
loo | +0.159 |
pour | +0.136 |
auc | +0.131 |
ick | +0.127 |
iren | +0.122 |
Neg logit contributions | |
chard | -0.382 |
wrists | -0.379 |
cious | -0.364 |
blonde | -0.352 |
ya | -0.351 |
Token | . |
Token ID | 13 |
Feature activation | +9.82e-01 |
Pos logit contributions | |
loo | +0.142 |
iren | +0.116 |
ick | +0.115 |
auc | +0.112 |
of | +0.111 |
Neg logit contributions | |
chard | -0.237 |
Robert | -0.233 |
Eva | -0.233 |
wrists | -0.233 |
blonde | -0.224 |
Token | They |
Token ID | 1119 |
Feature activation | +2.53e-01 |
Pos logit contributions | |
of | +0.032 |
loo | +0.032 |
ig | +0.028 |
ick | +0.028 |
late | +0.027 |
Neg logit contributions | |
wrists | -0.044 |
cious | -0.041 |
chard | -0.041 |
rice | -0.041 |
yes | -0.040 |
Token | were |
Token ID | 547 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
of | +0.009 |
loo | +0.008 |
ick | +0.008 |
ig | +0.007 |
ful | +0.007 |
Neg logit contributions | |
Eva | -0.011 |
wrists | -0.011 |
Robert | -0.011 |
chard | -0.011 |
blonde | -0.010 |
Token | very |
Token ID | 845 |
Feature activation | -2.34e+00 |
Pos logit contributions | |
loo | +0.037 |
of | +0.036 |
ick | +0.034 |
ig | +0.031 |
iren | +0.030 |
Neg logit contributions | |
wrists | -0.048 |
chard | -0.047 |
Eva | -0.047 |
Robert | -0.047 |
blonde | -0.045 |
Token | it |
Token ID | 340 |
Feature activation | +8.64e-02 |
Pos logit contributions | |
loo | +0.148 |
of | +0.130 |
ick | +0.128 |
iren | +0.126 |
auc | +0.125 |
Neg logit contributions | |
chard | -0.147 |
Eva | -0.146 |
Robert | -0.145 |
wrists | -0.145 |
blonde | -0.139 |
Token | to |
Token ID | 284 |
Feature activation | -4.00e-01 |
Pos logit contributions | |
loo | +0.003 |
iren | +0.003 |
ick | +0.003 |
auc | +0.003 |
arl | +0.003 |
Neg logit contributions | |
Robert | -0.004 |
Eva | -0.004 |
chard | -0.004 |
wrists | -0.004 |
blonde | -0.004 |
Token | his |
Token ID | 465 |
Feature activation | -4.50e-01 |
Pos logit contributions | |
wrists | +0.019 |
chard | +0.019 |
Eva | +0.018 |
Robert | +0.018 |
blonde | +0.018 |
Neg logit contributions | |
loo | -0.013 |
of | -0.012 |
ick | -0.011 |
iren | -0.011 |
auc | -0.011 |
Token | friends |
Token ID | 2460 |
Feature activation | -8.54e-01 |
Pos logit contributions | |
chard | +0.021 |
Eva | +0.020 |
Robert | +0.020 |
wrists | +0.020 |
blonde | +0.020 |
Neg logit contributions | |
loo | -0.015 |
of | -0.014 |
ig | -0.013 |
ick | -0.013 |
tunnel | -0.012 |
Token | ' |
Token ID | 6 |
Feature activation | -1.87e-01 |
Pos logit contributions | |
Eva | +0.047 |
wrists | +0.046 |
chard | +0.046 |
Robert | +0.046 |
blonde | +0.045 |
Neg logit contributions | |
loo | -0.021 |
of | -0.020 |
ick | -0.018 |
auc | -0.017 |
iren | -0.016 |
Token | bur |
Token ID | 4356 |
Feature activation | +7.16e+00 |
Pos logit contributions | |
Eva | +0.009 |
chard | +0.009 |
Robert | +0.009 |
wrists | +0.009 |
blonde | +0.009 |
Neg logit contributions | |
loo | -0.006 |
of | -0.005 |
ick | -0.005 |
iren | -0.005 |
auc | -0.005 |
Token | rows |
Token ID | 8516 |
Feature activation | +7.16e-01 |
Pos logit contributions | |
loo | +0.161 |
auc | +0.125 |
iren | +0.125 |
pour | +0.123 |
ick | +0.111 |
Neg logit contributions | |
chard | -0.389 |
wrists | -0.369 |
Robert | -0.364 |
Eva | -0.363 |
rice | -0.359 |
Token | . |
Token ID | 13 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
loo | +0.022 |
ick | +0.018 |
of | +0.018 |
iren | +0.018 |
auc | +0.018 |
Neg logit contributions | |
Robert | -0.036 |
Eva | -0.036 |
chard | -0.036 |
wrists | -0.035 |
blonde | -0.034 |
Token | They |
Token ID | 1119 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
loo | +0.006 |
of | +0.006 |
ick | +0.005 |
iren | +0.005 |
auc | +0.005 |
Neg logit contributions | |
wrists | -0.008 |
chard | -0.008 |
blonde | -0.007 |
cious | -0.007 |
Eva | -0.007 |
Token | all |
Token ID | 477 |
Feature activation | +4.56e+00 |
Pos logit contributions | |
Eva | +0.006 |
wrists | +0.006 |
blonde | +0.006 |
chard | +0.006 |
Robert | +0.006 |
Neg logit contributions | |
loo | -0.004 |
of | -0.004 |
ick | -0.004 |
pour | -0.003 |
ig | -0.003 |
Token | thought |
Token ID | 1807 |
Feature activation | +3.84e-01 |
Pos logit contributions | |
loo | +0.209 |
of | +0.193 |
ick | +0.185 |
auc | +0.181 |
iren | +0.180 |
Neg logit contributions | |
chard | -0.159 |
Eva | -0.158 |
wrists | -0.158 |
Robert | -0.157 |
blonde | -0.149 |
Token | ate |
Token ID | 15063 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
Robert | +0.083 |
Eva | +0.082 |
chard | +0.079 |
wrists | +0.078 |
Claire | +0.076 |
Neg logit contributions | |
loo | -0.067 |
auc | -0.058 |
iren | -0.057 |
ick | -0.057 |
arl | -0.056 |
Token | their |
Token ID | 511 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
loo | +0.038 |
of | +0.035 |
ick | +0.031 |
auc | +0.031 |
iren | +0.030 |
Neg logit contributions | |
wrists | -0.057 |
chard | -0.057 |
blonde | -0.057 |
Robert | -0.057 |
Eva | -0.057 |
Token | popcorn |
Token ID | 38914 |
Feature activation | -4.61e+00 |
Pos logit contributions | |
loo | +0.006 |
of | +0.005 |
ig | +0.005 |
ick | +0.005 |
auc | +0.005 |
Neg logit contributions | |
Robert | -0.007 |
Eva | -0.007 |
chard | -0.007 |
blonde | -0.007 |
wrists | -0.007 |
Token | while |
Token ID | 981 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
Eva | +0.227 |
chard | +0.227 |
Robert | +0.226 |
wrists | +0.225 |
blonde | +0.217 |
Neg logit contributions | |
loo | -0.146 |
of | -0.122 |
ick | -0.121 |
iren | -0.118 |
auc | -0.117 |
Token | it |
Token ID | 340 |
Feature activation | +6.01e-01 |
Pos logit contributions | |
Eva | +0.087 |
Robert | +0.087 |
chard | +0.087 |
wrists | +0.086 |
blonde | +0.082 |
Neg logit contributions | |
loo | -0.073 |
of | -0.062 |
ick | -0.062 |
iren | -0.062 |
auc | -0.062 |
Token | r |
Token ID | 374 |
Feature activation | +7.16e+00 |
Pos logit contributions | |
loo | +0.022 |
of | +0.019 |
ick | +0.018 |
auc | +0.018 |
iren | +0.018 |
Neg logit contributions | |
chard | -0.027 |
Robert | -0.027 |
Eva | -0.027 |
wrists | -0.027 |
blonde | -0.026 |
Token | ained |
Token ID | 1328 |
Feature activation | +4.13e+00 |
Pos logit contributions | |
loo | +0.151 |
of | +0.146 |
tunnel | +0.143 |
outage | +0.138 |
ig | +0.120 |
Neg logit contributions | |
chard | -0.378 |
wrists | -0.366 |
Eva | -0.351 |
blonde | -0.345 |
Robert | -0.343 |
Token | outside |
Token ID | 2354 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
loo | +0.150 |
auc | +0.123 |
iren | +0.123 |
ick | +0.121 |
rim | +0.121 |
Neg logit contributions | |
Robert | -0.176 |
chard | -0.176 |
Eva | -0.175 |
Claire | -0.170 |
wrists | -0.166 |
Token | . |
Token ID | 13 |
Feature activation | +5.15e-01 |
Pos logit contributions | |
Eva | +0.066 |
Robert | +0.065 |
wrists | +0.064 |
Claire | +0.064 |
chard | +0.064 |
Neg logit contributions | |
loo | -0.055 |
iren | -0.047 |
arl | -0.046 |
ick | -0.045 |
auc | -0.045 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
loo | +0.018 |
of | +0.018 |
auc | +0.016 |
iren | +0.016 |
ick | +0.016 |
Neg logit contributions | |
chard | -0.023 |
wrists | -0.023 |
cious | -0.021 |
Eva | -0.021 |
Robert | -0.021 |
Token | was |
Token ID | 373 |
Feature activation | +5.47e-01 |
Pos logit contributions | |
loo | +0.006 |
of | +0.006 |
ick | +0.005 |
auc | +0.005 |
iren | +0.005 |
Neg logit contributions | |
chard | -0.008 |
Robert | -0.007 |
wrists | -0.007 |
Eva | -0.007 |
blonde | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | -3.23e-01 |
Pos logit contributions | |
loo | +0.023 |
ick | +0.019 |
of | +0.019 |
iren | +0.018 |
auc | +0.018 |
Neg logit contributions | |
chard | -0.034 |
Eva | -0.034 |
Robert | -0.034 |
wrists | -0.034 |
blonde | -0.033 |
Token | share |
Token ID | 2648 |
Feature activation | -4.62e+00 |
Pos logit contributions | |
Eva | +0.013 |
wrists | +0.013 |
Robert | +0.013 |
chard | +0.013 |
blonde | +0.013 |
Neg logit contributions | |
loo | -0.013 |
auc | -0.011 |
ick | -0.011 |
iren | -0.011 |
of | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
Eva | +0.245 |
Robert | +0.244 |
chard | +0.244 |
wrists | +0.240 |
blonde | +0.238 |
Neg logit contributions | |
loo | -0.124 |
of | -0.113 |
ick | -0.100 |
iren | -0.098 |
auc | -0.098 |
Token | wait |
Token ID | 4043 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
wrists | +0.009 |
Eva | +0.009 |
chard | +0.009 |
Robert | +0.009 |
blonde | +0.008 |
Neg logit contributions | |
loo | -0.008 |
ick | -0.007 |
iren | -0.007 |
auc | -0.007 |
of | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +7.94e-01 |
Pos logit contributions | |
chard | +0.010 |
wrists | +0.010 |
Robert | +0.010 |
Eva | +0.010 |
blonde | +0.010 |
Neg logit contributions | |
loo | -0.004 |
of | -0.004 |
ick | -0.003 |
auc | -0.003 |
arl | -0.003 |
Token | How |
Token ID | 1374 |
Feature activation | -9.76e+00 |
Pos logit contributions | |
loo | +0.034 |
of | +0.030 |
ick | +0.030 |
iren | +0.030 |
auc | +0.029 |
Neg logit contributions | |
chard | -0.030 |
Eva | -0.030 |
Robert | -0.030 |
wrists | -0.030 |
blonde | -0.028 |
Token | about |
Token ID | 546 |
Feature activation | -2.54e+00 |
Pos logit contributions | |
Eva | +0.376 |
Robert | +0.374 |
chard | +0.374 |
wrists | +0.369 |
blonde | +0.355 |
Neg logit contributions | |
loo | -0.414 |
ick | -0.360 |
of | -0.359 |
iren | -0.359 |
auc | -0.352 |
Token | we |
Token ID | 356 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
Eva | +0.121 |
chard | +0.121 |
Robert | +0.121 |
wrists | +0.120 |
blonde | +0.116 |
Neg logit contributions | |
loo | -0.085 |
of | -0.073 |
ick | -0.071 |
iren | -0.070 |
auc | -0.070 |
Token | clean |
Token ID | 3424 |
Feature activation | +3.45e+00 |
Pos logit contributions | |
loo | +0.043 |
iren | +0.037 |
auc | +0.037 |
ick | +0.036 |
of | +0.035 |
Neg logit contributions | |
chard | -0.045 |
Robert | -0.044 |
Eva | -0.044 |
wrists | -0.043 |
Claire | -0.041 |
Token | up |
Token ID | 510 |
Feature activation | -5.84e+00 |
Pos logit contributions | |
loo | +0.139 |
of | +0.121 |
ick | +0.120 |
iren | +0.118 |
auc | +0.117 |
Neg logit contributions | |
chard | -0.141 |
Eva | -0.141 |
Robert | -0.141 |
wrists | -0.139 |
blonde | -0.135 |
Token | this |
Token ID | 428 |
Feature activation | -7.08e-02 |
Pos logit contributions | |
Eva | +0.275 |
chard | +0.273 |
Robert | +0.272 |
wrists | +0.272 |
Claire | +0.258 |
Neg logit contributions | |
loo | -0.200 |
ick | -0.166 |
iren | -0.165 |
auc | -0.160 |
of | -0.158 |
Token | play |
Token ID | 711 |
Feature activation | -3.78e-01 |
Pos logit contributions | |
Robert | +0.018 |
Eva | +0.018 |
wrists | +0.017 |
chard | +0.017 |
Claire | +0.017 |
Neg logit contributions | |
loo | -0.018 |
auc | -0.015 |
iren | -0.015 |
ick | -0.015 |
pour | -0.014 |
Token | with |
Token ID | 351 |
Feature activation | -2.59e+00 |
Pos logit contributions | |
chard | +0.020 |
Eva | +0.020 |
Robert | +0.020 |
wrists | +0.020 |
blonde | +0.020 |
Neg logit contributions | |
loo | -0.010 |
of | -0.009 |
ick | -0.008 |
auc | -0.008 |
iren | -0.007 |
Token | other |
Token ID | 584 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
chard | +0.139 |
Eva | +0.138 |
Robert | +0.138 |
wrists | +0.137 |
blonde | +0.135 |
Neg logit contributions | |
loo | -0.068 |
of | -0.060 |
ick | -0.055 |
auc | -0.053 |
iren | -0.052 |
Token | toys |
Token ID | 14958 |
Feature activation | -6.45e-01 |
Pos logit contributions | |
chard | +0.007 |
Eva | +0.007 |
Robert | +0.007 |
wrists | +0.007 |
blonde | +0.007 |
Neg logit contributions | |
loo | -0.004 |
of | -0.004 |
ick | -0.003 |
auc | -0.003 |
iren | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +8.37e-01 |
Pos logit contributions | |
chard | +0.031 |
Robert | +0.031 |
Eva | +0.031 |
wrists | +0.031 |
Claire | +0.029 |
Neg logit contributions | |
loo | -0.021 |
iren | -0.017 |
ick | -0.017 |
auc | -0.017 |
of | -0.017 |
Token | How |
Token ID | 1374 |
Feature activation | -9.75e+00 |
Pos logit contributions | |
loo | +0.038 |
of | +0.035 |
ick | +0.034 |
auc | +0.033 |
iren | +0.033 |
Neg logit contributions | |
chard | -0.029 |
wrists | -0.029 |
Eva | -0.028 |
Robert | -0.028 |
blonde | -0.027 |
Token | about |
Token ID | 546 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
chard | +0.450 |
wrists | +0.447 |
Robert | +0.439 |
Eva | +0.435 |
blonde | +0.420 |
Neg logit contributions | |
loo | -0.354 |
of | -0.340 |
ick | -0.303 |
auc | -0.298 |
arl | -0.275 |
Token | this |
Token ID | 428 |
Feature activation | +7.94e-02 |
Pos logit contributions | |
Eva | +0.132 |
chard | +0.132 |
wrists | +0.131 |
Robert | +0.130 |
blonde | +0.129 |
Neg logit contributions | |
loo | -0.078 |
of | -0.070 |
ick | -0.064 |
iren | -0.064 |
auc | -0.063 |
Token | puzzle |
Token ID | 15027 |
Feature activation | +9.48e-01 |
Pos logit contributions | |
loo | +0.002 |
of | +0.002 |
ick | +0.002 |
iren | +0.002 |
auc | +0.002 |
Neg logit contributions | |
Eva | -0.004 |
Robert | -0.004 |
chard | -0.004 |
wrists | -0.004 |
blonde | -0.004 |
Token | ?" |
Token ID | 1701 |
Feature activation | -3.95e+00 |
Pos logit contributions | |
loo | +0.027 |
ick | +0.022 |
iren | +0.022 |
auc | +0.021 |
arl | +0.020 |
Neg logit contributions | |
chard | -0.049 |
Robert | -0.049 |
Eva | -0.049 |
wrists | -0.048 |
Claire | -0.047 |
Token | She |
Token ID | 1375 |
Feature activation | -6.58e-01 |
Pos logit contributions | |
chard | +0.200 |
wrists | +0.199 |
cious | +0.193 |
blonde | +0.189 |
Robert | +0.186 |
Neg logit contributions | |
of | -0.117 |
loo | -0.114 |
iren | -0.097 |
ick | -0.093 |
auc | -0.091 |
Token | downstairs |
Token ID | 34624 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
loo | +0.024 |
auc | +0.021 |
ick | +0.021 |
rim | +0.021 |
iren | +0.020 |
Neg logit contributions | |
Robert | -0.025 |
chard | -0.025 |
wrists | -0.025 |
Eva | -0.024 |
Claire | -0.023 |
Token | for |
Token ID | 329 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
Eva | +0.087 |
Robert | +0.087 |
chard | +0.087 |
wrists | +0.086 |
blonde | +0.082 |
Neg logit contributions | |
loo | -0.057 |
iren | -0.046 |
auc | -0.045 |
ick | -0.045 |
of | -0.045 |
Token | a |
Token ID | 257 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
loo | +0.041 |
of | +0.036 |
ick | +0.035 |
iren | +0.035 |
auc | +0.034 |
Neg logit contributions | |
Eva | -0.051 |
Robert | -0.051 |
chard | -0.051 |
wrists | -0.050 |
blonde | -0.048 |
Token | snack |
Token ID | 26906 |
Feature activation | -8.42e-01 |
Pos logit contributions | |
wrists | +0.055 |
Eva | +0.054 |
chard | +0.054 |
Robert | +0.054 |
blonde | +0.052 |
Neg logit contributions | |
loo | -0.044 |
of | -0.039 |
ick | -0.038 |
iren | -0.036 |
auc | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | +9.01e-01 |
Pos logit contributions | |
Eva | +0.035 |
Robert | +0.035 |
chard | +0.035 |
wrists | +0.035 |
Claire | +0.034 |
Neg logit contributions | |
loo | -0.031 |
iren | -0.028 |
ick | -0.026 |
rim | -0.026 |
auc | -0.026 |
Token | How |
Token ID | 1374 |
Feature activation | -9.74e+00 |
Pos logit contributions | |
loo | +0.038 |
of | +0.034 |
ick | +0.033 |
iren | +0.033 |
auc | +0.033 |
Neg logit contributions | |
chard | -0.035 |
wrists | -0.034 |
Robert | -0.034 |
Eva | -0.034 |
blonde | -0.032 |
Token | does |
Token ID | 857 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
wrists | +0.427 |
chard | +0.426 |
Robert | +0.417 |
Eva | +0.415 |
blonde | +0.401 |
Neg logit contributions | |
loo | -0.374 |
of | -0.350 |
ick | -0.323 |
auc | -0.315 |
iren | -0.295 |
Token | that |
Token ID | 326 |
Feature activation | +3.03e+00 |
Pos logit contributions | |
Robert | +0.051 |
Eva | +0.050 |
chard | +0.048 |
wrists | +0.047 |
Claire | +0.046 |
Neg logit contributions | |
loo | -0.041 |
auc | -0.035 |
ick | -0.035 |
iren | -0.034 |
arl | -0.034 |
Token | sound |
Token ID | 2128 |
Feature activation | +8.88e-01 |
Pos logit contributions | |
loo | +0.107 |
ick | +0.092 |
iren | +0.090 |
auc | +0.089 |
arl | +0.087 |
Neg logit contributions | |
Eva | -0.137 |
Robert | -0.137 |
chard | -0.134 |
yes | -0.133 |
Claire | -0.129 |
Token | ?" |
Token ID | 1701 |
Feature activation | -3.74e+00 |
Pos logit contributions | |
loo | +0.028 |
ick | +0.024 |
iren | +0.022 |
auc | +0.022 |
of | +0.022 |
Neg logit contributions | |
Eva | -0.043 |
Robert | -0.043 |
chard | -0.043 |
wrists | -0.042 |
blonde | -0.041 |
Token |
|
Token ID | 198 |
Feature activation | -3.12e-01 |
Pos logit contributions | |
wrists | +0.190 |
chard | +0.189 |
Robert | +0.183 |
Eva | +0.179 |
blonde | +0.179 |
Neg logit contributions | |
of | -0.110 |
loo | -0.109 |
ick | -0.093 |
iren | -0.092 |
pour | -0.088 |
Token | that |
Token ID | 326 |
Feature activation | +3.06e+00 |
Pos logit contributions | |
loo | +0.005 |
of | +0.004 |
ick | +0.004 |
iren | +0.004 |
auc | +0.004 |
Neg logit contributions | |
Eva | -0.009 |
Robert | -0.009 |
chard | -0.009 |
wrists | -0.009 |
blonde | -0.008 |
Token | you |
Token ID | 345 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
loo | +0.104 |
ick | +0.090 |
iren | +0.088 |
auc | +0.085 |
arl | +0.084 |
Neg logit contributions | |
Robert | -0.143 |
Eva | -0.143 |
chard | -0.141 |
wrists | -0.140 |
Claire | -0.134 |
Token | both |
Token ID | 1111 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
loo | +0.004 |
of | +0.003 |
ick | +0.003 |
auc | +0.003 |
iren | +0.003 |
Neg logit contributions | |
chard | -0.006 |
Robert | -0.006 |
Eva | -0.006 |
wrists | -0.006 |
blonde | -0.006 |
Token | like |
Token ID | 588 |
Feature activation | -2.55e+00 |
Pos logit contributions | |
loo | +0.071 |
ick | +0.060 |
iren | +0.059 |
auc | +0.059 |
of | +0.057 |
Neg logit contributions | |
Eva | -0.086 |
Robert | -0.085 |
chard | -0.085 |
wrists | -0.085 |
blonde | -0.081 |
Token | . |
Token ID | 13 |
Feature activation | +9.34e-01 |
Pos logit contributions | |
Eva | +0.129 |
chard | +0.129 |
Robert | +0.128 |
wrists | +0.127 |
blonde | +0.124 |
Neg logit contributions | |
loo | -0.077 |
of | -0.067 |
ick | -0.063 |
iren | -0.062 |
auc | -0.061 |
Token | How |
Token ID | 1374 |
Feature activation | -9.74e+00 |
Pos logit contributions | |
loo | +0.040 |
of | +0.037 |
ick | +0.036 |
iren | +0.035 |
auc | +0.035 |
Neg logit contributions | |
chard | -0.035 |
wrists | -0.034 |
Eva | -0.033 |
Robert | -0.033 |
blonde | -0.032 |
Token | about |
Token ID | 546 |
Feature activation | -2.44e+00 |
Pos logit contributions | |
chard | +0.447 |
wrists | +0.441 |
Robert | +0.437 |
Eva | +0.433 |
blonde | +0.419 |
Neg logit contributions | |
loo | -0.357 |
of | -0.334 |
ick | -0.307 |
auc | -0.298 |
arl | -0.277 |
Token | a |
Token ID | 257 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
Eva | +0.114 |
Robert | +0.114 |
chard | +0.114 |
wrists | +0.113 |
blonde | +0.109 |
Neg logit contributions | |
loo | -0.084 |
of | -0.071 |
ick | -0.071 |
iren | -0.069 |
auc | -0.069 |
Token | cartoon |
Token ID | 16251 |
Feature activation | -5.36e+00 |
Pos logit contributions | |
wrists | +0.058 |
chard | +0.058 |
Eva | +0.058 |
Robert | +0.056 |
blonde | +0.055 |
Neg logit contributions | |
loo | -0.044 |
of | -0.040 |
ick | -0.037 |
iren | -0.036 |
auc | -0.035 |
Token | about |
Token ID | 546 |
Feature activation | -2.44e+00 |
Pos logit contributions | |
Eva | +0.271 |
Robert | +0.270 |
chard | +0.267 |
wrists | +0.263 |
Claire | +0.254 |
Neg logit contributions | |
loo | -0.159 |
ick | -0.134 |
auc | -0.127 |
iren | -0.127 |
arl | -0.123 |
Token | animals |
Token ID | 4695 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
Eva | +0.111 |
chard | +0.111 |
Robert | +0.111 |
wrists | +0.110 |
blonde | +0.109 |
Neg logit contributions | |
loo | -0.085 |
of | -0.076 |
ick | -0.072 |
iren | -0.069 |
auc | -0.069 |
Token | in |
Token ID | 287 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
loo | +0.136 |
iren | +0.120 |
ick | +0.119 |
auc | +0.117 |
rim | +0.115 |
Neg logit contributions | |
Robert | -0.157 |
Eva | -0.156 |
wrists | -0.152 |
chard | -0.151 |
Claire | -0.148 |
Token | the |
Token ID | 262 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
Robert | +0.093 |
chard | +0.093 |
Eva | +0.089 |
wrists | +0.088 |
Claire | +0.086 |
Neg logit contributions | |
loo | -0.123 |
auc | -0.113 |
iren | -0.111 |
arl | -0.109 |
rim | -0.108 |
Token | living |
Token ID | 2877 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
loo | +0.008 |
ick | +0.007 |
of | +0.007 |
iren | +0.007 |
auc | +0.007 |
Neg logit contributions | |
chard | -0.010 |
Robert | -0.010 |
Eva | -0.010 |
wrists | -0.010 |
blonde | -0.010 |
Token | room |
Token ID | 2119 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
loo | +0.047 |
iren | +0.044 |
ick | +0.042 |
of | +0.042 |
arl | +0.042 |
Neg logit contributions | |
wrists | -0.072 |
Eva | -0.069 |
chard | -0.069 |
Robert | -0.068 |
Claire | -0.065 |
Token | . |
Token ID | 13 |
Feature activation | +6.86e-01 |
Pos logit contributions | |
loo | +0.083 |
iren | +0.070 |
ick | +0.070 |
auc | +0.068 |
of | +0.067 |
Neg logit contributions | |
Eva | -0.113 |
Robert | -0.112 |
wrists | -0.112 |
chard | -0.112 |
blonde | -0.107 |
Token | How |
Token ID | 1374 |
Feature activation | -9.74e+00 |
Pos logit contributions | |
loo | +0.030 |
of | +0.026 |
ick | +0.026 |
iren | +0.025 |
auc | +0.025 |
Neg logit contributions | |
chard | -0.026 |
Eva | -0.026 |
wrists | -0.026 |
Robert | -0.026 |
blonde | -0.024 |
Token | does |
Token ID | 857 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
wrists | +0.442 |
chard | +0.441 |
Robert | +0.437 |
Eva | +0.435 |
blonde | +0.420 |
Neg logit contributions | |
loo | -0.358 |
of | -0.329 |
ick | -0.304 |
auc | -0.297 |
iren | -0.280 |
Token | that |
Token ID | 326 |
Feature activation | +3.01e+00 |
Pos logit contributions | |
Robert | +0.053 |
Eva | +0.053 |
chard | +0.053 |
wrists | +0.051 |
blonde | +0.049 |
Neg logit contributions | |
loo | -0.045 |
auc | -0.039 |
ick | -0.039 |
iren | -0.038 |
arl | -0.037 |
Token | sound |
Token ID | 2128 |
Feature activation | +8.54e-01 |
Pos logit contributions | |
loo | +0.105 |
ick | +0.091 |
arl | +0.090 |
iren | +0.089 |
auc | +0.088 |
Neg logit contributions | |
Robert | -0.135 |
chard | -0.135 |
Eva | -0.134 |
yes | -0.131 |
Claire | -0.126 |
Token | ?" |
Token ID | 1701 |
Feature activation | -4.36e+00 |
Pos logit contributions | |
loo | +0.025 |
ick | +0.021 |
iren | +0.020 |
of | +0.020 |
auc | +0.020 |
Neg logit contributions | |
chard | -0.044 |
Eva | -0.044 |
Robert | -0.044 |
wrists | -0.043 |
blonde | -0.041 |
Token |
|
Token ID | 198 |
Feature activation | -2.60e-01 |
Pos logit contributions | |
wrists | +0.223 |
chard | +0.214 |
blonde | +0.211 |
Eva | +0.210 |
Robert | +0.209 |
Neg logit contributions | |
loo | -0.134 |
of | -0.129 |
iren | -0.112 |
ick | -0.109 |
auc | -0.106 |
Token | are |
Token ID | 389 |
Feature activation | +6.33e-01 |
Pos logit contributions | |
Robert | +0.001 |
Eva | +0.001 |
wrists | +0.001 |
chard | +0.001 |
Claire | +0.001 |
Neg logit contributions | |
loo | -0.001 |
iren | -0.001 |
auc | -0.001 |
ick | -0.001 |
arl | -0.001 |
Token | plenty |
Token ID | 6088 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
loo | +0.024 |
of | +0.021 |
ick | +0.020 |
iren | +0.020 |
auc | +0.020 |
Neg logit contributions | |
Eva | -0.027 |
Robert | -0.027 |
chard | -0.027 |
wrists | -0.027 |
blonde | -0.026 |
Token | of |
Token ID | 286 |
Feature activation | +3.08e+00 |
Pos logit contributions | |
yes | +0.001 |
Claire | +0.001 |
Eva | +0.001 |
Robert | +0.001 |
chard | +0.001 |
Neg logit contributions | |
loo | -0.000 |
rim | -0.000 |
el | -0.000 |
ig | -0.000 |
ick | -0.000 |
Token | them |
Token ID | 606 |
Feature activation | -3.13e+00 |
Pos logit contributions | |
loo | +0.116 |
ick | +0.103 |
auc | +0.099 |
pour | +0.099 |
iren | +0.097 |
Neg logit contributions | |
wrists | -0.131 |
Robert | -0.129 |
blonde | -0.127 |
Eva | -0.126 |
chard | -0.123 |
Token | . |
Token ID | 13 |
Feature activation | +8.98e-01 |
Pos logit contributions | |
chard | +0.149 |
Robert | +0.148 |
Eva | +0.148 |
wrists | +0.147 |
blonde | +0.140 |
Neg logit contributions | |
loo | -0.104 |
iren | -0.086 |
ick | -0.086 |
auc | -0.085 |
of | -0.082 |
Token | How |
Token ID | 1374 |
Feature activation | -9.73e+00 |
Pos logit contributions | |
loo | +0.039 |
of | +0.035 |
ick | +0.034 |
iren | +0.033 |
auc | +0.033 |
Neg logit contributions | |
chard | -0.034 |
wrists | -0.034 |
Eva | -0.034 |
Robert | -0.033 |
blonde | -0.032 |
Token | about |
Token ID | 546 |
Feature activation | -2.48e+00 |
Pos logit contributions | |
chard | +0.439 |
wrists | +0.437 |
Robert | +0.434 |
Eva | +0.433 |
blonde | +0.416 |
Neg logit contributions | |
loo | -0.358 |
of | -0.329 |
ick | -0.305 |
auc | -0.300 |
iren | -0.285 |
Token | that |
Token ID | 326 |
Feature activation | +2.98e+00 |
Pos logit contributions | |
Eva | +0.123 |
chard | +0.122 |
Robert | +0.122 |
wrists | +0.121 |
blonde | +0.119 |
Neg logit contributions | |
loo | -0.077 |
of | -0.067 |
ick | -0.064 |
auc | -0.062 |
iren | -0.062 |
Token | ?" |
Token ID | 1701 |
Feature activation | -3.87e+00 |
Pos logit contributions | |
loo | +0.100 |
ick | +0.085 |
iren | +0.083 |
auc | +0.083 |
arl | +0.079 |
Neg logit contributions | |
Robert | -0.140 |
chard | -0.139 |
Eva | -0.139 |
wrists | -0.133 |
yes | -0.133 |
Token |
|
Token ID | 198 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
wrists | +0.186 |
chard | +0.186 |
Robert | +0.181 |
Eva | +0.180 |
blonde | +0.175 |
Neg logit contributions | |
loo | -0.126 |
of | -0.120 |
iren | -0.107 |
ick | -0.107 |
auc | -0.103 |
Token |
|
Token ID | 198 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
Eva | +0.011 |
Robert | +0.011 |
wrists | +0.011 |
chard | +0.011 |
blonde | +0.010 |
Neg logit contributions | |
loo | -0.012 |
ick | -0.010 |
auc | -0.010 |
iren | -0.010 |
of | -0.010 |
Token | is |
Token ID | 318 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
Eva | +0.015 |
Robert | +0.015 |
chard | +0.015 |
wrists | +0.015 |
blonde | +0.014 |
Neg logit contributions | |
loo | -0.010 |
ick | -0.009 |
iren | -0.009 |
of | -0.009 |
auc | -0.009 |
Token | safe |
Token ID | 3338 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
loo | +0.036 |
of | +0.031 |
ick | +0.031 |
iren | +0.029 |
auc | +0.029 |
Neg logit contributions | |
wrists | -0.051 |
Eva | -0.051 |
Robert | -0.051 |
chard | -0.051 |
blonde | -0.049 |
Token | and |
Token ID | 290 |
Feature activation | +8.38e-02 |
Pos logit contributions | |
loo | +0.060 |
iren | +0.053 |
auc | +0.051 |
ick | +0.050 |
rim | +0.048 |
Neg logit contributions | |
Robert | -0.071 |
chard | -0.071 |
Eva | -0.070 |
wrists | -0.068 |
Claire | -0.067 |
Token | clean |
Token ID | 3424 |
Feature activation | +3.64e+00 |
Pos logit contributions | |
loo | +0.004 |
ick | +0.003 |
iren | +0.003 |
auc | +0.003 |
of | +0.003 |
Neg logit contributions | |
chard | -0.003 |
Robert | -0.003 |
Eva | -0.003 |
wrists | -0.003 |
blonde | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +7.31e-01 |
Pos logit contributions | |
loo | +0.117 |
of | +0.098 |
ick | +0.097 |
iren | +0.096 |
auc | +0.095 |
Neg logit contributions | |
Eva | -0.177 |
Robert | -0.177 |
chard | -0.177 |
wrists | -0.176 |
blonde | -0.169 |
Token | How |
Token ID | 1374 |
Feature activation | -9.73e+00 |
Pos logit contributions | |
loo | +0.030 |
of | +0.027 |
ick | +0.026 |
iren | +0.025 |
auc | +0.025 |
Neg logit contributions | |
chard | -0.029 |
wrists | -0.029 |
Eva | -0.029 |
Robert | -0.028 |
blonde | -0.028 |
Token | does |
Token ID | 857 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
chard | +0.451 |
wrists | +0.451 |
Robert | +0.436 |
Eva | +0.434 |
blonde | +0.427 |
Neg logit contributions | |
loo | -0.350 |
of | -0.329 |
ick | -0.300 |
auc | -0.289 |
ig | -0.268 |
Token | that |
Token ID | 326 |
Feature activation | +3.05e+00 |
Pos logit contributions | |
Robert | +0.049 |
Eva | +0.049 |
chard | +0.048 |
wrists | +0.047 |
Claire | +0.045 |
Neg logit contributions | |
loo | -0.037 |
ick | -0.031 |
auc | -0.031 |
iren | -0.030 |
arl | -0.029 |
Token | sound |
Token ID | 2128 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
loo | +0.103 |
ick | +0.088 |
auc | +0.087 |
iren | +0.085 |
arl | +0.083 |
Neg logit contributions | |
Robert | -0.145 |
Eva | -0.144 |
chard | -0.140 |
yes | -0.138 |
Claire | -0.135 |
Token | ?" |
Token ID | 1701 |
Feature activation | -4.06e+00 |
Pos logit contributions | |
loo | +0.030 |
ick | +0.025 |
iren | +0.024 |
auc | +0.024 |
of | +0.023 |
Neg logit contributions | |
Eva | -0.051 |
Robert | -0.051 |
chard | -0.051 |
wrists | -0.050 |
Claire | -0.048 |
Token |
|
Token ID | 198 |
Feature activation | -2.50e-01 |
Pos logit contributions | |
wrists | +0.202 |
chard | +0.201 |
Robert | +0.191 |
blonde | +0.191 |
Eva | +0.191 |
Neg logit contributions | |
loo | -0.123 |
of | -0.117 |
ick | -0.106 |
iren | -0.104 |
pour | -0.098 |
Token | just |
Token ID | 655 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
loo | +0.104 |
of | +0.088 |
ick | +0.088 |
iren | +0.087 |
auc | +0.086 |
Neg logit contributions | |
chard | -0.129 |
Eva | -0.129 |
Robert | -0.129 |
wrists | -0.128 |
blonde | -0.123 |
Token | one |
Token ID | 530 |
Feature activation | +4.36e-01 |
Pos logit contributions | |
loo | +0.061 |
of | +0.055 |
ick | +0.052 |
iren | +0.051 |
auc | +0.051 |
Neg logit contributions | |
Eva | -0.070 |
chard | -0.070 |
wrists | -0.070 |
Robert | -0.070 |
blonde | -0.067 |
Token | of |
Token ID | 286 |
Feature activation | +3.12e+00 |
Pos logit contributions | |
loo | +0.013 |
of | +0.011 |
ick | +0.011 |
iren | +0.010 |
auc | +0.010 |
Neg logit contributions | |
Eva | -0.022 |
chard | -0.022 |
Robert | -0.022 |
wrists | -0.022 |
blonde | -0.021 |
Token | you |
Token ID | 345 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
loo | +0.119 |
ick | +0.101 |
auc | +0.100 |
iren | +0.099 |
of | +0.096 |
Neg logit contributions | |
Eva | -0.137 |
Robert | -0.136 |
wrists | -0.135 |
chard | -0.134 |
Claire | -0.128 |
Token | . |
Token ID | 13 |
Feature activation | +6.84e-01 |
Pos logit contributions | |
loo | +0.001 |
of | +0.000 |
ick | +0.000 |
iren | +0.000 |
auc | +0.000 |
Neg logit contributions | |
chard | -0.001 |
Eva | -0.001 |
Robert | -0.001 |
wrists | -0.001 |
blonde | -0.001 |
Token | How |
Token ID | 1374 |
Feature activation | -9.72e+00 |
Pos logit contributions | |
loo | +0.031 |
ick | +0.028 |
of | +0.027 |
iren | +0.027 |
auc | +0.027 |
Neg logit contributions | |
Robert | -0.024 |
Eva | -0.024 |
chard | -0.024 |
wrists | -0.023 |
blonde | -0.023 |
Token | would |
Token ID | 561 |
Feature activation | -2.84e+00 |
Pos logit contributions | |
Eva | +0.388 |
chard | +0.387 |
Robert | +0.387 |
wrists | +0.385 |
blonde | +0.369 |
Neg logit contributions | |
loo | -0.403 |
of | -0.355 |
ick | -0.349 |
auc | -0.342 |
iren | -0.342 |
Token | you |
Token ID | 345 |
Feature activation | +1.46e-03 |
Pos logit contributions | |
Robert | +0.151 |
Eva | +0.151 |
chard | +0.150 |
wrists | +0.150 |
blonde | +0.144 |
Neg logit contributions | |
loo | -0.081 |
ick | -0.065 |
iren | -0.064 |
of | -0.064 |
auc | -0.063 |
Token | feel |
Token ID | 1254 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
loo | +0.000 |
iren | +0.000 |
auc | +0.000 |
ick | +0.000 |
of | +0.000 |
Neg logit contributions | |
chard | -0.000 |
Robert | -0.000 |
Eva | -0.000 |
wrists | -0.000 |
blonde | -0.000 |
Token | if |
Token ID | 611 |
Feature activation | -3.36e+00 |
Pos logit contributions | |
chard | +0.051 |
Robert | +0.051 |
Eva | +0.051 |
wrists | +0.050 |
blonde | +0.048 |
Neg logit contributions | |
loo | -0.047 |
ick | -0.040 |
iren | -0.040 |
of | -0.040 |
auc | -0.039 |
Token | someone |
Token ID | 2130 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
Eva | +0.167 |
chard | +0.167 |
Robert | +0.166 |
wrists | +0.164 |
blonde | +0.158 |
Neg logit contributions | |
loo | -0.106 |
ick | -0.088 |
iren | -0.087 |
of | -0.087 |
auc | -0.086 |
Token | you |
Token ID | 345 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
loo | +0.075 |
of | +0.061 |
ick | +0.059 |
iren | +0.057 |
auc | +0.057 |
Neg logit contributions | |
chard | -0.162 |
Eva | -0.162 |
Robert | -0.161 |
wrists | -0.161 |
blonde | -0.156 |
Token | for |
Token ID | 329 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
loo | +0.005 |
of | +0.004 |
ick | +0.004 |
iren | +0.004 |
auc | +0.004 |
Neg logit contributions | |
chard | -0.009 |
Eva | -0.009 |
wrists | -0.009 |
Robert | -0.009 |
blonde | -0.009 |
Token | saying |
Token ID | 2282 |
Feature activation | -7.52e-01 |
Pos logit contributions | |
loo | +0.041 |
of | +0.036 |
ick | +0.035 |
iren | +0.034 |
auc | +0.033 |
Neg logit contributions | |
Eva | -0.055 |
chard | -0.055 |
Robert | -0.055 |
wrists | -0.054 |
blonde | -0.053 |
Token | sorry |
Token ID | 7926 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
chard | +0.033 |
Eva | +0.032 |
wrists | +0.031 |
Robert | +0.031 |
blonde | +0.031 |
Neg logit contributions | |
loo | -0.028 |
of | -0.026 |
iren | -0.025 |
ick | -0.024 |
arl | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
Eva | +0.053 |
Robert | +0.052 |
wrists | +0.052 |
chard | +0.050 |
Claire | +0.049 |
Neg logit contributions | |
loo | -0.038 |
ick | -0.032 |
iren | -0.032 |
auc | -0.031 |
of | -0.030 |
Token | How |
Token ID | 1374 |
Feature activation | -9.72e+00 |
Pos logit contributions | |
loo | +0.043 |
of | +0.038 |
ick | +0.037 |
iren | +0.037 |
auc | +0.037 |
Neg logit contributions | |
chard | -0.041 |
Eva | -0.040 |
wrists | -0.040 |
Robert | -0.040 |
blonde | -0.038 |
Token | about |
Token ID | 546 |
Feature activation | -2.53e+00 |
Pos logit contributions | |
wrists | +0.431 |
chard | +0.429 |
Eva | +0.419 |
Robert | +0.418 |
blonde | +0.403 |
Neg logit contributions | |
loo | -0.373 |
of | -0.347 |
ick | -0.320 |
auc | -0.315 |
iren | -0.295 |
Token | we |
Token ID | 356 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
Eva | +0.123 |
chard | +0.123 |
Robert | +0.122 |
wrists | +0.122 |
blonde | +0.119 |
Neg logit contributions | |
loo | -0.081 |
of | -0.070 |
ick | -0.067 |
iren | -0.066 |
auc | -0.066 |
Token | go |
Token ID | 467 |
Feature activation | +7.46e-01 |
Pos logit contributions | |
loo | +0.052 |
iren | +0.044 |
auc | +0.044 |
ick | +0.042 |
of | +0.040 |
Neg logit contributions | |
chard | -0.047 |
Robert | -0.046 |
wrists | -0.045 |
Eva | -0.044 |
yes | -0.044 |
Token | home |
Token ID | 1363 |
Feature activation | +3.11e+00 |
Pos logit contributions | |
loo | +0.029 |
ick | +0.025 |
auc | +0.025 |
iren | +0.025 |
arl | +0.024 |
Neg logit contributions | |
chard | -0.030 |
Robert | -0.030 |
wrists | -0.029 |
Eva | -0.029 |
yes | -0.028 |
Token | and |
Token ID | 290 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
loo | +0.115 |
iren | +0.099 |
auc | +0.098 |
ick | +0.097 |
arl | +0.090 |
Neg logit contributions | |
chard | -0.135 |
Eva | -0.134 |
Robert | -0.133 |
wrists | -0.130 |
Claire | -0.127 |
Token | will |
Token ID | 481 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
Robert | +0.020 |
Eva | +0.020 |
chard | +0.019 |
wrists | +0.019 |
blonde | +0.019 |
Neg logit contributions | |
loo | -0.017 |
of | -0.015 |
ick | -0.015 |
iren | -0.014 |
auc | -0.014 |
Token | be |
Token ID | 307 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
Robert | +0.049 |
wrists | +0.049 |
chard | +0.049 |
Eva | +0.049 |
blonde | +0.046 |
Neg logit contributions | |
loo | -0.053 |
auc | -0.045 |
ick | -0.044 |
iren | -0.043 |
of | -0.042 |
Token | sad |
Token ID | 6507 |
Feature activation | +7.82e-01 |
Pos logit contributions | |
loo | +0.052 |
of | +0.049 |
iren | +0.045 |
ick | +0.045 |
auc | +0.044 |
Neg logit contributions | |
Eva | -0.071 |
Robert | -0.071 |
chard | -0.070 |
wrists | -0.070 |
blonde | -0.068 |
Token | too |
Token ID | 1165 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
loo | +0.035 |
iren | +0.031 |
auc | +0.030 |
rim | +0.029 |
ick | +0.029 |
Neg logit contributions | |
wrists | -0.027 |
chard | -0.027 |
Robert | -0.027 |
Eva | -0.026 |
Claire | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | +8.73e-01 |
Pos logit contributions | |
loo | +0.042 |
auc | +0.035 |
ick | +0.035 |
iren | +0.035 |
of | +0.035 |
Neg logit contributions | |
Eva | -0.052 |
Robert | -0.052 |
chard | -0.052 |
wrists | -0.051 |
blonde | -0.049 |
Token | How |
Token ID | 1374 |
Feature activation | -9.72e+00 |
Pos logit contributions | |
loo | +0.035 |
of | +0.031 |
iren | +0.030 |
ick | +0.030 |
auc | +0.030 |
Neg logit contributions | |
chard | -0.036 |
wrists | -0.035 |
Robert | -0.035 |
Eva | -0.035 |
blonde | -0.033 |
Token | can |
Token ID | 460 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
chard | +0.399 |
Eva | +0.399 |
Robert | +0.398 |
wrists | +0.398 |
blonde | +0.381 |
Neg logit contributions | |
loo | -0.392 |
of | -0.348 |
ick | -0.338 |
auc | -0.332 |
iren | -0.329 |
Token | we |
Token ID | 356 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
Eva | +0.021 |
Robert | +0.021 |
chard | +0.021 |
wrists | +0.021 |
blonde | +0.020 |
Neg logit contributions | |
loo | -0.013 |
ick | -0.011 |
iren | -0.010 |
auc | -0.010 |
of | -0.010 |
Token | fix |
Token ID | 4259 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
loo | +0.048 |
auc | +0.041 |
iren | +0.040 |
ick | +0.040 |
arl | +0.037 |
Neg logit contributions | |
chard | -0.046 |
Robert | -0.045 |
wrists | -0.045 |
Eva | -0.043 |
yes | -0.042 |
Token | it |
Token ID | 340 |
Feature activation | -4.18e-01 |
Pos logit contributions | |
loo | +0.045 |
of | +0.041 |
ick | +0.039 |
iren | +0.038 |
auc | +0.037 |
Neg logit contributions | |
chard | -0.059 |
blonde | -0.057 |
Eva | -0.057 |
Robert | -0.057 |
wrists | -0.056 |
Token | ?" |
Token ID | 1701 |
Feature activation | -5.13e+00 |
Pos logit contributions | |
chard | +0.020 |
Robert | +0.019 |
Eva | +0.019 |
wrists | +0.019 |
yes | +0.018 |
Neg logit contributions | |
loo | -0.014 |
ick | -0.012 |
iren | -0.012 |
auc | -0.011 |
arl | -0.011 |
Token | share |
Token ID | 2648 |
Feature activation | -4.62e+00 |
Pos logit contributions | |
wrists | +0.012 |
Eva | +0.011 |
Robert | +0.011 |
chard | +0.011 |
blonde | +0.011 |
Neg logit contributions | |
loo | -0.014 |
auc | -0.012 |
ick | -0.011 |
iren | -0.011 |
of | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
chard | +0.246 |
Robert | +0.245 |
Eva | +0.245 |
wrists | +0.241 |
blonde | +0.239 |
Neg logit contributions | |
loo | -0.123 |
of | -0.110 |
ick | -0.099 |
iren | -0.098 |
auc | -0.096 |
Token | take |
Token ID | 1011 |
Feature activation | +8.52e-02 |
Pos logit contributions | |
wrists | +0.005 |
Eva | +0.005 |
Robert | +0.005 |
chard | +0.005 |
blonde | +0.005 |
Neg logit contributions | |
loo | -0.005 |
ick | -0.005 |
auc | -0.004 |
iren | -0.004 |
pour | -0.004 |
Token | turns |
Token ID | 4962 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
loo | +0.003 |
of | +0.003 |
ick | +0.003 |
iren | +0.003 |
auc | +0.002 |
Neg logit contributions | |
chard | -0.004 |
Eva | -0.004 |
Robert | -0.004 |
wrists | -0.004 |
blonde | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +9.45e-01 |
Pos logit contributions | |
wrists | +0.061 |
Robert | +0.060 |
Eva | +0.060 |
chard | +0.059 |
blonde | +0.057 |
Neg logit contributions | |
loo | -0.052 |
ick | -0.042 |
auc | -0.042 |
ig | -0.040 |
pour | -0.040 |
Token | How |
Token ID | 1374 |
Feature activation | -9.72e+00 |
Pos logit contributions | |
loo | +0.039 |
of | +0.034 |
ick | +0.034 |
iren | +0.033 |
auc | +0.033 |
Neg logit contributions | |
Eva | -0.038 |
Robert | -0.038 |
chard | -0.037 |
wrists | -0.037 |
blonde | -0.036 |
Token | about |
Token ID | 546 |
Feature activation | -2.54e+00 |
Pos logit contributions | |
Eva | +0.395 |
Robert | +0.392 |
chard | +0.390 |
wrists | +0.385 |
blonde | +0.373 |
Neg logit contributions | |
loo | -0.392 |
iren | -0.338 |
ick | -0.338 |
of | -0.331 |
auc | -0.328 |
Token | this |
Token ID | 428 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
chard | +0.124 |
Eva | +0.124 |
Robert | +0.124 |
wrists | +0.123 |
blonde | +0.120 |
Neg logit contributions | |
loo | -0.081 |
of | -0.070 |
ick | -0.067 |
iren | -0.066 |
auc | -0.066 |
Token | : |
Token ID | 25 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
Eva | +0.007 |
Robert | +0.007 |
chard | +0.007 |
wrists | +0.007 |
blonde | +0.007 |
Neg logit contributions | |
loo | -0.005 |
ick | -0.004 |
auc | -0.004 |
iren | -0.004 |
of | -0.004 |
Token | you |
Token ID | 345 |
Feature activation | +4.04e-03 |
Pos logit contributions | |
Eva | +0.084 |
Robert | +0.084 |
wrists | +0.083 |
chard | +0.083 |
blonde | +0.081 |
Neg logit contributions | |
loo | -0.069 |
of | -0.065 |
iren | -0.059 |
auc | -0.059 |
ick | -0.058 |
Token | can |
Token ID | 460 |
Feature activation | -4.21e-01 |
Pos logit contributions | |
loo | +0.000 |
iren | +0.000 |
auc | +0.000 |
ick | +0.000 |
of | +0.000 |
Neg logit contributions | |
chard | -0.000 |
Robert | -0.000 |
wrists | -0.000 |
Eva | -0.000 |
blonde | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | +9.50e-01 |
Pos logit contributions | |
Eva | +0.239 |
chard | +0.239 |
Robert | +0.238 |
wrists | +0.236 |
blonde | +0.232 |
Neg logit contributions | |
loo | -0.127 |
of | -0.110 |
ick | -0.102 |
iren | -0.100 |
auc | -0.100 |
Token | Sharing |
Token ID | 36644 |
Feature activation | -4.05e+00 |
Pos logit contributions | |
loo | +0.040 |
of | +0.035 |
ick | +0.034 |
iren | +0.034 |
auc | +0.034 |
Neg logit contributions | |
chard | -0.037 |
Eva | -0.037 |
Robert | -0.037 |
wrists | -0.037 |
blonde | -0.035 |
Token | is |
Token ID | 318 |
Feature activation | +9.36e-01 |
Pos logit contributions | |
wrists | +0.184 |
Robert | +0.182 |
Eva | +0.178 |
chard | +0.170 |
Claire | +0.166 |
Neg logit contributions | |
loo | -0.150 |
ick | -0.130 |
auc | -0.124 |
ig | -0.123 |
pour | -0.117 |
Token | caring |
Token ID | 18088 |
Feature activation | -3.08e+00 |
Pos logit contributions | |
loo | +0.031 |
of | +0.030 |
iren | +0.027 |
ick | +0.027 |
auc | +0.026 |
Neg logit contributions | |
Eva | -0.043 |
chard | -0.043 |
Robert | -0.042 |
wrists | -0.042 |
blonde | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
Eva | +0.132 |
Robert | +0.131 |
wrists | +0.131 |
Claire | +0.123 |
chard | +0.121 |
Neg logit contributions | |
loo | -0.113 |
iren | -0.098 |
pour | -0.096 |
ick | -0.095 |
auc | -0.095 |
Token | How |
Token ID | 1374 |
Feature activation | -9.72e+00 |
Pos logit contributions | |
loo | +0.042 |
of | +0.037 |
ick | +0.036 |
iren | +0.036 |
auc | +0.036 |
Neg logit contributions | |
chard | -0.040 |
Eva | -0.039 |
Robert | -0.039 |
wrists | -0.039 |
blonde | -0.038 |
Token | about |
Token ID | 546 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
Eva | +0.394 |
Robert | +0.392 |
chard | +0.391 |
wrists | +0.386 |
blonde | +0.372 |
Neg logit contributions | |
loo | -0.393 |
ick | -0.339 |
iren | -0.338 |
of | -0.336 |
auc | -0.331 |
Token | you |
Token ID | 345 |
Feature activation | +5.60e-02 |
Pos logit contributions | |
Eva | +0.128 |
blonde | +0.128 |
chard | +0.126 |
Robert | +0.125 |
wrists | +0.124 |
Neg logit contributions | |
loo | -0.067 |
of | -0.064 |
iren | -0.055 |
ick | -0.055 |
auc | -0.053 |
Token | take |
Token ID | 1011 |
Feature activation | +4.29e-02 |
Pos logit contributions | |
loo | +0.002 |
ick | +0.002 |
iren | +0.002 |
auc | +0.002 |
of | +0.002 |
Neg logit contributions | |
chard | -0.002 |
Robert | -0.002 |
Eva | -0.002 |
wrists | -0.002 |
blonde | -0.002 |
Token | turns |
Token ID | 4962 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
loo | +0.001 |
of | +0.001 |
iren | +0.001 |
ick | +0.001 |
auc | +0.001 |
Neg logit contributions | |
chard | -0.002 |
Eva | -0.002 |
wrists | -0.002 |
Robert | -0.002 |
blonde | -0.002 |
Token | playing |
Token ID | 2712 |
Feature activation | -2.75e+00 |
Pos logit contributions | |
wrists | +0.059 |
Robert | +0.058 |
chard | +0.057 |
Eva | +0.057 |
Claire | +0.054 |
Neg logit contributions | |
loo | -0.049 |
ig | -0.040 |
ick | -0.039 |
auc | -0.039 |
pour | -0.037 |
Token | back |
Token ID | 736 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
chard | +0.055 |
Robert | +0.055 |
Eva | +0.055 |
wrists | +0.054 |
blonde | +0.051 |
Neg logit contributions | |
loo | -0.052 |
iren | -0.045 |
ick | -0.045 |
auc | -0.044 |
of | -0.044 |
Token | and |
Token ID | 290 |
Feature activation | -7.70e-02 |
Pos logit contributions | |
Eva | +0.065 |
chard | +0.065 |
Robert | +0.065 |
wrists | +0.065 |
blonde | +0.063 |
Neg logit contributions | |
loo | -0.043 |
of | -0.037 |
ick | -0.036 |
iren | -0.035 |
auc | -0.035 |
Token | relax |
Token ID | 8960 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
chard | +0.003 |
Eva | +0.003 |
Robert | +0.003 |
wrists | +0.003 |
blonde | +0.003 |
Neg logit contributions | |
loo | -0.003 |
ick | -0.003 |
iren | -0.003 |
of | -0.003 |
auc | -0.003 |
Token | together |
Token ID | 1978 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
Eva | +0.079 |
Robert | +0.079 |
chard | +0.078 |
wrists | +0.078 |
blonde | +0.076 |
Neg logit contributions | |
loo | -0.047 |
of | -0.039 |
ick | -0.038 |
iren | -0.037 |
auc | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | +7.03e-01 |
Pos logit contributions | |
chard | +0.101 |
Eva | +0.101 |
Robert | +0.101 |
wrists | +0.100 |
blonde | +0.099 |
Neg logit contributions | |
loo | -0.061 |
of | -0.054 |
ick | -0.051 |
iren | -0.047 |
auc | -0.047 |
Token | How |
Token ID | 1374 |
Feature activation | -9.72e+00 |
Pos logit contributions | |
loo | +0.029 |
of | +0.026 |
ick | +0.025 |
iren | +0.025 |
auc | +0.025 |
Neg logit contributions | |
chard | -0.028 |
Eva | -0.027 |
wrists | -0.027 |
Robert | -0.027 |
blonde | -0.026 |
Token | does |
Token ID | 857 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
chard | +0.431 |
wrists | +0.429 |
Robert | +0.425 |
Eva | +0.424 |
blonde | +0.407 |
Neg logit contributions | |
loo | -0.366 |
of | -0.332 |
ick | -0.313 |
auc | -0.305 |
iren | -0.292 |
Token | that |
Token ID | 326 |
Feature activation | +3.03e+00 |
Pos logit contributions | |
Robert | +0.052 |
Eva | +0.052 |
chard | +0.051 |
wrists | +0.050 |
blonde | +0.048 |
Neg logit contributions | |
loo | -0.045 |
auc | -0.039 |
ick | -0.039 |
iren | -0.038 |
of | -0.037 |
Token | sound |
Token ID | 2128 |
Feature activation | +9.74e-01 |
Pos logit contributions | |
loo | +0.101 |
ick | +0.086 |
auc | +0.086 |
iren | +0.082 |
arl | +0.080 |
Neg logit contributions | |
Robert | -0.143 |
Eva | -0.143 |
chard | -0.140 |
wrists | -0.136 |
yes | -0.136 |
Token | ?" |
Token ID | 1701 |
Feature activation | -4.25e+00 |
Pos logit contributions | |
loo | +0.030 |
ick | +0.025 |
auc | +0.024 |
iren | +0.024 |
of | +0.023 |
Neg logit contributions | |
chard | -0.048 |
Robert | -0.048 |
Eva | -0.048 |
wrists | -0.046 |
Claire | -0.045 |
Token |
|
Token ID | 198 |
Feature activation | -3.30e-01 |
Pos logit contributions | |
chard | +0.204 |
wrists | +0.203 |
Robert | +0.199 |
Eva | +0.197 |
blonde | +0.191 |
Neg logit contributions | |
loo | -0.144 |
of | -0.132 |
iren | -0.123 |
ick | -0.120 |
auc | -0.114 |
Token | on |
Token ID | 319 |
Feature activation | -2.14e+00 |
Pos logit contributions | |
loo | +0.056 |
ick | +0.049 |
auc | +0.047 |
iren | +0.044 |
rim | +0.043 |
Neg logit contributions | |
chard | -0.076 |
wrists | -0.076 |
Robert | -0.075 |
Eva | -0.075 |
blonde | -0.071 |
Token | the |
Token ID | 262 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
chard | +0.101 |
Eva | +0.099 |
wrists | +0.099 |
Robert | +0.099 |
blonde | +0.097 |
Neg logit contributions | |
loo | -0.071 |
of | -0.063 |
ick | -0.060 |
iren | -0.058 |
auc | -0.057 |
Token | big |
Token ID | 1263 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
loo | +0.008 |
of | +0.007 |
iren | +0.006 |
auc | +0.006 |
ick | +0.006 |
Neg logit contributions | |
Eva | -0.010 |
chard | -0.010 |
Robert | -0.010 |
wrists | -0.010 |
blonde | -0.010 |
Token | slide |
Token ID | 10649 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
loo | +0.092 |
of | +0.080 |
ick | +0.080 |
iren | +0.078 |
auc | +0.078 |
Neg logit contributions | |
Eva | -0.095 |
Robert | -0.094 |
chard | -0.094 |
wrists | -0.094 |
blonde | -0.090 |
Token | . |
Token ID | 13 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
chard | +0.094 |
Eva | +0.094 |
Robert | +0.094 |
wrists | +0.093 |
blonde | +0.090 |
Neg logit contributions | |
loo | -0.059 |
ick | -0.048 |
of | -0.048 |
iren | -0.048 |
auc | -0.047 |
Token | How |
Token ID | 1374 |
Feature activation | -9.71e+00 |
Pos logit contributions | |
loo | +0.044 |
of | +0.039 |
ick | +0.038 |
iren | +0.037 |
auc | +0.037 |
Neg logit contributions | |
chard | -0.042 |
Eva | -0.042 |
wrists | -0.042 |
Robert | -0.042 |
blonde | -0.040 |
Token | was |
Token ID | 373 |
Feature activation | +3.70e-01 |
Pos logit contributions | |
chard | +0.453 |
wrists | +0.452 |
Robert | +0.442 |
Eva | +0.439 |
blonde | +0.427 |
Neg logit contributions | |
loo | -0.352 |
of | -0.325 |
ick | -0.297 |
auc | -0.292 |
iren | -0.273 |
Token | it |
Token ID | 340 |
Feature activation | -3.98e-01 |
Pos logit contributions | |
loo | +0.009 |
of | +0.008 |
iren | +0.006 |
ick | +0.006 |
auc | +0.006 |
Neg logit contributions | |
wrists | -0.021 |
blonde | -0.021 |
Eva | -0.020 |
chard | -0.020 |
Robert | -0.020 |
Token | ?" |
Token ID | 1701 |
Feature activation | -5.27e+00 |
Pos logit contributions | |
chard | +0.019 |
Eva | +0.019 |
Robert | +0.019 |
yes | +0.018 |
Claire | +0.018 |
Neg logit contributions | |
loo | -0.012 |
iren | -0.011 |
ick | -0.010 |
auc | -0.010 |
arl | -0.010 |
Token | Sara |
Token ID | 24799 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
wrists | +0.251 |
chard | +0.247 |
Robert | +0.247 |
Eva | +0.243 |
blonde | +0.238 |
Neg logit contributions | |
loo | -0.173 |
of | -0.168 |
iren | -0.149 |
ick | -0.144 |
auc | -0.143 |
Token | 's |
Token ID | 338 |
Feature activation | +9.81e-01 |
Pos logit contributions | |
chard | +0.086 |
Robert | +0.083 |
wrists | +0.083 |
Eva | +0.083 |
blonde | +0.082 |
Neg logit contributions | |
loo | -0.059 |
of | -0.054 |
iren | -0.051 |
ick | -0.051 |
auc | -0.050 |
Token | you |
Token ID | 345 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
loo | +0.103 |
iren | +0.088 |
ick | +0.088 |
arl | +0.084 |
of | +0.084 |
Neg logit contributions | |
Eva | -0.142 |
Robert | -0.141 |
chard | -0.139 |
wrists | -0.139 |
Claire | -0.132 |
Token | can |
Token ID | 460 |
Feature activation | -3.29e-01 |
Pos logit contributions | |
loo | +0.004 |
of | +0.004 |
ick | +0.003 |
auc | +0.003 |
iren | +0.003 |
Neg logit contributions | |
Eva | -0.007 |
Robert | -0.007 |
wrists | -0.007 |
chard | -0.007 |
blonde | -0.007 |
Token | use |
Token ID | 779 |
Feature activation | +2.90e+00 |
Pos logit contributions | |
chard | +0.013 |
Robert | +0.013 |
wrists | +0.013 |
yes | +0.012 |
blonde | +0.012 |
Neg logit contributions | |
loo | -0.014 |
auc | -0.012 |
iren | -0.011 |
pour | -0.011 |
rim | -0.011 |
Token | safely |
Token ID | 11512 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
loo | +0.094 |
ick | +0.077 |
of | +0.076 |
iren | +0.076 |
auc | +0.076 |
Neg logit contributions | |
Robert | -0.142 |
chard | -0.141 |
Eva | -0.141 |
wrists | -0.141 |
blonde | -0.135 |
Token | . |
Token ID | 13 |
Feature activation | +7.29e-01 |
Pos logit contributions | |
loo | +0.040 |
auc | +0.033 |
iren | +0.033 |
ick | +0.033 |
pour | +0.031 |
Neg logit contributions | |
chard | -0.052 |
Robert | -0.050 |
wrists | -0.050 |
Eva | -0.049 |
blonde | -0.047 |
Token | How |
Token ID | 1374 |
Feature activation | -9.71e+00 |
Pos logit contributions | |
loo | +0.031 |
of | +0.029 |
ick | +0.028 |
iren | +0.027 |
auc | +0.027 |
Neg logit contributions | |
chard | -0.027 |
wrists | -0.026 |
Eva | -0.026 |
Robert | -0.026 |
blonde | -0.025 |
Token | about |
Token ID | 546 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
chard | +0.460 |
wrists | +0.458 |
Robert | +0.449 |
Eva | +0.449 |
blonde | +0.432 |
Neg logit contributions | |
loo | -0.339 |
of | -0.323 |
ick | -0.289 |
auc | -0.281 |
ig | -0.258 |
Token | that |
Token ID | 326 |
Feature activation | +3.03e+00 |
Pos logit contributions | |
Robert | +0.116 |
Eva | +0.116 |
chard | +0.116 |
wrists | +0.115 |
blonde | +0.111 |
Neg logit contributions | |
loo | -0.088 |
of | -0.074 |
ick | -0.074 |
iren | -0.073 |
auc | -0.072 |
Token | ?" |
Token ID | 1701 |
Feature activation | -5.35e+00 |
Pos logit contributions | |
loo | +0.099 |
ick | +0.083 |
auc | +0.081 |
iren | +0.080 |
of | +0.079 |
Neg logit contributions | |
chard | -0.148 |
Robert | -0.148 |
Eva | -0.147 |
wrists | -0.144 |
blonde | -0.139 |
Token | he |
Token ID | 339 |
Feature activation | +6.54e-01 |
Pos logit contributions | |
wrists | +0.264 |
chard | +0.252 |
Eva | +0.250 |
Robert | +0.248 |
blonde | +0.245 |
Neg logit contributions | |
loo | -0.163 |
of | -0.157 |
ick | -0.142 |
iren | -0.139 |
auc | -0.134 |
Token | said |
Token ID | 531 |
Feature activation | +7.09e-01 |
Pos logit contributions | |
loo | +0.024 |
of | +0.023 |
iren | +0.022 |
ick | +0.022 |
auc | +0.021 |
Neg logit contributions | |
chard | -0.027 |
Eva | -0.026 |
wrists | -0.026 |
Robert | -0.026 |
blonde | -0.025 |
Token | play |
Token ID | 711 |
Feature activation | -3.89e-01 |
Pos logit contributions | |
Eva | +0.017 |
Robert | +0.017 |
wrists | +0.017 |
chard | +0.017 |
blonde | +0.016 |
Neg logit contributions | |
loo | -0.016 |
auc | -0.014 |
ick | -0.014 |
iren | -0.014 |
of | -0.013 |
Token | with |
Token ID | 351 |
Feature activation | -2.50e+00 |
Pos logit contributions | |
blonde | +0.022 |
Eva | +0.022 |
chard | +0.022 |
wrists | +0.022 |
Robert | +0.022 |
Neg logit contributions | |
loo | -0.009 |
of | -0.009 |
ick | -0.007 |
iren | -0.007 |
auc | -0.007 |
Token | something |
Token ID | 1223 |
Feature activation | -3.64e+00 |
Pos logit contributions | |
blonde | +0.129 |
Robert | +0.129 |
chard | +0.129 |
Eva | +0.129 |
wrists | +0.127 |
Neg logit contributions | |
loo | -0.062 |
of | -0.060 |
ick | -0.054 |
iren | -0.049 |
auc | -0.047 |
Token | else |
Token ID | 2073 |
Feature activation | -4.49e+00 |
Pos logit contributions | |
chard | +0.133 |
Eva | +0.133 |
Robert | +0.132 |
wrists | +0.132 |
blonde | +0.127 |
Neg logit contributions | |
loo | -0.162 |
of | -0.148 |
ick | -0.142 |
auc | -0.140 |
iren | -0.139 |
Token | . |
Token ID | 13 |
Feature activation | +8.91e-01 |
Pos logit contributions | |
Robert | +0.249 |
Eva | +0.248 |
chard | +0.248 |
blonde | +0.247 |
wrists | +0.247 |
Neg logit contributions | |
loo | -0.109 |
of | -0.100 |
ick | -0.087 |
auc | -0.081 |
iren | -0.076 |
Token | How |
Token ID | 1374 |
Feature activation | -9.71e+00 |
Pos logit contributions | |
loo | +0.039 |
of | +0.035 |
ick | +0.034 |
iren | +0.034 |
auc | +0.034 |
Neg logit contributions | |
chard | -0.033 |
wrists | -0.033 |
Eva | -0.032 |
Robert | -0.032 |
blonde | -0.031 |
Token | about |
Token ID | 546 |
Feature activation | -2.50e+00 |
Pos logit contributions | |
wrists | +0.434 |
chard | +0.432 |
Robert | +0.425 |
Eva | +0.423 |
blonde | +0.410 |
Neg logit contributions | |
loo | -0.366 |
of | -0.348 |
ick | -0.313 |
auc | -0.308 |
iren | -0.285 |
Token | the |
Token ID | 262 |
Feature activation | +9.90e-02 |
Pos logit contributions | |
Eva | +0.122 |
wrists | +0.121 |
chard | +0.121 |
blonde | +0.120 |
Robert | +0.120 |
Neg logit contributions | |
loo | -0.077 |
of | -0.070 |
ick | -0.063 |
iren | -0.063 |
auc | -0.062 |
Token | blocks |
Token ID | 7021 |
Feature activation | -4.38e-01 |
Pos logit contributions | |
loo | +0.004 |
of | +0.003 |
ick | +0.003 |
iren | +0.003 |
auc | +0.003 |
Neg logit contributions | |
Eva | -0.004 |
Robert | -0.004 |
wrists | -0.004 |
chard | -0.004 |
blonde | -0.004 |
Token | ?" |
Token ID | 1701 |
Feature activation | -3.88e+00 |
Pos logit contributions | |
chard | +0.021 |
Robert | +0.020 |
Eva | +0.020 |
wrists | +0.020 |
yes | +0.019 |
Neg logit contributions | |
loo | -0.014 |
ick | -0.012 |
iren | -0.012 |
auc | -0.011 |
ig | -0.011 |
Token | Tim |
Token ID | 5045 |
Feature activation | +9.35e-01 |
Pos logit contributions | |
chard | +0.192 |
wrists | +0.191 |
cious | +0.185 |
blonde | +0.181 |
Robert | +0.179 |
Neg logit contributions | |
of | -0.120 |
loo | -0.118 |
iren | -0.099 |
ick | -0.097 |
late | -0.095 |
Token | magic |
Token ID | 5536 |
Feature activation | -4.04e-01 |
Pos logit contributions | |
chard | +0.007 |
wrists | +0.007 |
Robert | +0.007 |
Eva | +0.007 |
blonde | +0.007 |
Neg logit contributions | |
loo | -0.005 |
of | -0.004 |
iren | -0.004 |
ick | -0.004 |
auc | -0.004 |
Token | book |
Token ID | 1492 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
chard | +0.019 |
Robert | +0.019 |
Eva | +0.019 |
wrists | +0.019 |
blonde | +0.018 |
Neg logit contributions | |
loo | -0.014 |
of | -0.012 |
ick | -0.012 |
iren | -0.011 |
auc | -0.011 |
Token | ? |
Token ID | 30 |
Feature activation | -6.85e-01 |
Pos logit contributions | |
Eva | +0.069 |
Robert | +0.069 |
chard | +0.067 |
wrists | +0.066 |
Claire | +0.064 |
Neg logit contributions | |
loo | -0.043 |
ick | -0.038 |
iren | -0.035 |
auc | -0.035 |
rim | -0.034 |
Token | Really |
Token ID | 16123 |
Feature activation | -4.97e+00 |
Pos logit contributions | |
chard | +0.024 |
wrists | +0.024 |
Robert | +0.024 |
Eva | +0.023 |
blonde | +0.022 |
Neg logit contributions | |
loo | -0.031 |
of | -0.028 |
auc | -0.028 |
ick | -0.027 |
iren | -0.027 |
Token | ? |
Token ID | 30 |
Feature activation | -7.07e-01 |
Pos logit contributions | |
chard | +0.285 |
Eva | +0.285 |
Robert | +0.283 |
wrists | +0.280 |
blonde | +0.272 |
Neg logit contributions | |
loo | -0.117 |
auc | -0.088 |
iren | -0.088 |
ick | -0.088 |
of | -0.086 |
Token | How |
Token ID | 1374 |
Feature activation | -9.71e+00 |
Pos logit contributions | |
chard | +0.026 |
wrists | +0.026 |
Robert | +0.026 |
Eva | +0.025 |
blonde | +0.024 |
Neg logit contributions | |
loo | -0.030 |
of | -0.028 |
auc | -0.027 |
ick | -0.027 |
iren | -0.026 |
Token | much |
Token ID | 881 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
chard | +0.496 |
wrists | +0.493 |
Robert | +0.483 |
Eva | +0.476 |
blonde | +0.459 |
Neg logit contributions | |
loo | -0.304 |
of | -0.292 |
ick | -0.256 |
auc | -0.251 |
arl | -0.227 |
Token | do |
Token ID | 466 |
Feature activation | -2.79e-01 |
Pos logit contributions | |
loo | +0.054 |
ick | +0.047 |
auc | +0.046 |
iren | +0.045 |
arl | +0.043 |
Neg logit contributions | |
Eva | -0.079 |
Robert | -0.077 |
chard | -0.075 |
wrists | -0.074 |
blonde | -0.073 |
Token | you |
Token ID | 345 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
chard | +0.015 |
Eva | +0.015 |
Robert | +0.014 |
wrists | +0.014 |
blonde | +0.014 |
Neg logit contributions | |
loo | -0.008 |
of | -0.007 |
ick | -0.006 |
iren | -0.006 |
auc | -0.006 |
Token | want |
Token ID | 765 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
loo | +0.001 |
of | +0.001 |
ick | +0.001 |
iren | +0.001 |
auc | +0.001 |
Neg logit contributions | |
chard | -0.001 |
Eva | -0.001 |
Robert | -0.001 |
wrists | -0.001 |
blonde | -0.001 |
Token | for |
Token ID | 329 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
Eva | +0.010 |
Robert | +0.010 |
chard | +0.010 |
wrists | +0.010 |
blonde | +0.009 |
Neg logit contributions | |
loo | -0.007 |
ick | -0.006 |
of | -0.005 |
iren | -0.005 |
auc | -0.005 |
Token | play |
Token ID | 711 |
Feature activation | -2.77e-01 |
Pos logit contributions | |
Eva | +0.011 |
Robert | +0.011 |
wrists | +0.011 |
chard | +0.011 |
blonde | +0.010 |
Neg logit contributions | |
loo | -0.011 |
auc | -0.010 |
iren | -0.010 |
of | -0.009 |
ick | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
Eva | +0.015 |
Robert | +0.015 |
chard | +0.015 |
wrists | +0.015 |
blonde | +0.014 |
Neg logit contributions | |
loo | -0.008 |
of | -0.007 |
ick | -0.006 |
auc | -0.006 |
iren | -0.006 |
Token | watch |
Token ID | 2342 |
Feature activation | -5.17e-01 |
Pos logit contributions | |
Eva | +0.011 |
Robert | +0.011 |
chard | +0.011 |
wrists | +0.011 |
blonde | +0.010 |
Neg logit contributions | |
loo | -0.010 |
of | -0.008 |
ick | -0.008 |
iren | -0.008 |
auc | -0.008 |
Token | cartoons |
Token ID | 30070 |
Feature activation | -5.58e+00 |
Pos logit contributions | |
chard | +0.026 |
Eva | +0.025 |
wrists | +0.025 |
Robert | +0.025 |
blonde | +0.025 |
Neg logit contributions | |
loo | -0.015 |
of | -0.014 |
ick | -0.013 |
auc | -0.012 |
iren | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +7.15e-01 |
Pos logit contributions | |
Eva | +0.286 |
chard | +0.286 |
Robert | +0.286 |
wrists | +0.285 |
blonde | +0.279 |
Neg logit contributions | |
loo | -0.162 |
of | -0.138 |
ick | -0.133 |
iren | -0.129 |
auc | -0.128 |
Token | How |
Token ID | 1374 |
Feature activation | -9.71e+00 |
Pos logit contributions | |
loo | +0.030 |
of | +0.028 |
ick | +0.027 |
iren | +0.027 |
auc | +0.026 |
Neg logit contributions | |
chard | -0.027 |
wrists | -0.027 |
Eva | -0.026 |
Robert | -0.026 |
blonde | -0.025 |
Token | about |
Token ID | 546 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
chard | +0.484 |
wrists | +0.482 |
Robert | +0.471 |
Eva | +0.461 |
blonde | +0.451 |
Neg logit contributions | |
of | -0.318 |
loo | -0.313 |
ick | -0.267 |
auc | -0.260 |
ig | -0.231 |
Token | that |
Token ID | 326 |
Feature activation | +2.96e+00 |
Pos logit contributions | |
chard | +0.127 |
Eva | +0.126 |
Robert | +0.126 |
wrists | +0.125 |
blonde | +0.123 |
Neg logit contributions | |
loo | -0.079 |
of | -0.070 |
ick | -0.066 |
iren | -0.064 |
auc | -0.064 |
Token | ?" |
Token ID | 1701 |
Feature activation | -4.09e+00 |
Pos logit contributions | |
loo | +0.098 |
ick | +0.081 |
auc | +0.080 |
iren | +0.080 |
arl | +0.076 |
Neg logit contributions | |
Eva | -0.143 |
Robert | -0.143 |
chard | -0.141 |
wrists | -0.137 |
Claire | -0.135 |
Token |
|
Token ID | 198 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
wrists | +0.212 |
chard | +0.206 |
blonde | +0.206 |
cious | +0.205 |
Eva | +0.195 |
Neg logit contributions | |
of | -0.110 |
loo | -0.108 |
iren | -0.095 |
late | -0.091 |
ick | -0.085 |
Token |
|
Token ID | 198 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
chard | +0.006 |
wrists | +0.006 |
blonde | +0.006 |
Robert | +0.006 |
Eva | +0.006 |
Neg logit contributions | |
loo | -0.004 |
of | -0.004 |
iren | -0.003 |
ick | -0.003 |
auc | -0.003 |
Token | can |
Token ID | 460 |
Feature activation | -2.79e-01 |
Pos logit contributions | |
loo | +0.044 |
iren | +0.038 |
auc | +0.037 |
ick | +0.036 |
rim | +0.035 |
Neg logit contributions | |
chard | -0.052 |
Robert | -0.052 |
wrists | -0.051 |
Eva | -0.051 |
Claire | -0.048 |
Token | play |
Token ID | 711 |
Feature activation | -4.10e-01 |
Pos logit contributions | |
wrists | +0.011 |
Robert | +0.011 |
Eva | +0.010 |
chard | +0.010 |
blonde | +0.010 |
Neg logit contributions | |
loo | -0.012 |
auc | -0.010 |
iren | -0.010 |
pour | -0.010 |
ick | -0.010 |
Token | pretend |
Token ID | 16614 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
Eva | +0.020 |
chard | +0.020 |
Robert | +0.020 |
wrists | +0.020 |
blonde | +0.020 |
Neg logit contributions | |
loo | -0.013 |
of | -0.011 |
ick | -0.011 |
auc | -0.010 |
iren | -0.010 |
Token | together |
Token ID | 1978 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
Robert | +0.096 |
Eva | +0.096 |
chard | +0.095 |
wrists | +0.095 |
Claire | +0.090 |
Neg logit contributions | |
loo | -0.069 |
ick | -0.059 |
iren | -0.058 |
auc | -0.056 |
arl | -0.056 |
Token | . |
Token ID | 13 |
Feature activation | +8.88e-01 |
Pos logit contributions | |
Eva | +0.095 |
chard | +0.095 |
Robert | +0.095 |
wrists | +0.094 |
blonde | +0.091 |
Neg logit contributions | |
loo | -0.057 |
of | -0.047 |
ick | -0.046 |
iren | -0.045 |
auc | -0.045 |
Token | How |
Token ID | 1374 |
Feature activation | -9.71e+00 |
Pos logit contributions | |
loo | +0.037 |
of | +0.034 |
iren | +0.033 |
ick | +0.033 |
auc | +0.032 |
Neg logit contributions | |
chard | -0.034 |
wrists | -0.034 |
Eva | -0.033 |
Robert | -0.033 |
blonde | -0.032 |
Token | about |
Token ID | 546 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
chard | +0.430 |
wrists | +0.428 |
Robert | +0.424 |
Eva | +0.424 |
blonde | +0.406 |
Neg logit contributions | |
loo | -0.366 |
of | -0.334 |
ick | -0.313 |
auc | -0.307 |
iren | -0.295 |
Token | we |
Token ID | 356 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
Eva | +0.117 |
chard | +0.117 |
Robert | +0.117 |
wrists | +0.116 |
blonde | +0.113 |
Neg logit contributions | |
loo | -0.085 |
of | -0.073 |
ick | -0.071 |
iren | -0.070 |
auc | -0.069 |
Token | pretend |
Token ID | 16614 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
loo | +0.056 |
iren | +0.048 |
auc | +0.047 |
ick | +0.045 |
rim | +0.044 |
Neg logit contributions | |
chard | -0.043 |
Robert | -0.043 |
wrists | -0.042 |
yes | -0.042 |
Eva | -0.041 |
Token | to |
Token ID | 284 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
Robert | +0.100 |
Eva | +0.100 |
chard | +0.099 |
wrists | +0.098 |
blonde | +0.093 |
Neg logit contributions | |
loo | -0.066 |
ick | -0.056 |
iren | -0.055 |
auc | -0.054 |
of | -0.054 |
Token | be |
Token ID | 307 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
wrists | +0.012 |
Robert | +0.012 |
Eva | +0.012 |
chard | +0.012 |
blonde | +0.011 |
Neg logit contributions | |
loo | -0.012 |
ick | -0.011 |
iren | -0.010 |
auc | -0.010 |
arl | -0.010 |
Token | go |
Token ID | 467 |
Feature activation | +4.62e-01 |
Pos logit contributions | |
Eva | +0.049 |
Robert | +0.049 |
chard | +0.049 |
wrists | +0.048 |
blonde | +0.047 |
Neg logit contributions | |
loo | -0.043 |
of | -0.037 |
ick | -0.036 |
auc | -0.036 |
iren | -0.036 |
Token | the |
Token ID | 262 |
Feature activation | -2.52e-02 |
Pos logit contributions | |
loo | +0.016 |
auc | +0.013 |
ick | +0.013 |
iren | +0.013 |
of | +0.013 |
Neg logit contributions | |
chard | -0.022 |
Robert | -0.022 |
Eva | -0.021 |
wrists | -0.021 |
blonde | -0.020 |
Token | other |
Token ID | 584 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
Eva | +0.001 |
wrists | +0.001 |
Robert | +0.001 |
blonde | +0.001 |
chard | +0.001 |
Neg logit contributions | |
loo | -0.001 |
of | -0.001 |
ick | -0.001 |
ig | -0.001 |
iren | -0.001 |
Token | way |
Token ID | 835 |
Feature activation | +9.12e-01 |
Pos logit contributions | |
loo | +0.000 |
of | +0.000 |
ick | +0.000 |
iren | +0.000 |
auc | +0.000 |
Neg logit contributions | |
chard | -0.001 |
Eva | -0.001 |
Robert | -0.001 |
wrists | -0.001 |
blonde | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +2.78e-01 |
Pos logit contributions | |
loo | +0.033 |
auc | +0.028 |
iren | +0.028 |
ick | +0.027 |
rim | +0.026 |
Neg logit contributions | |
Robert | -0.041 |
chard | -0.040 |
Eva | -0.040 |
wrists | -0.039 |
Claire | -0.038 |
Token | How |
Token ID | 1374 |
Feature activation | -9.70e+00 |
Pos logit contributions | |
loo | +0.010 |
of | +0.010 |
ick | +0.009 |
iren | +0.009 |
arl | +0.009 |
Neg logit contributions | |
chard | -0.012 |
wrists | -0.012 |
Eva | -0.011 |
Robert | -0.011 |
blonde | -0.011 |
Token | does |
Token ID | 857 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
chard | +0.458 |
wrists | +0.458 |
Eva | +0.450 |
Robert | +0.449 |
blonde | +0.433 |
Neg logit contributions | |
loo | -0.333 |
of | -0.316 |
ick | -0.287 |
auc | -0.277 |
pour | -0.258 |
Token | that |
Token ID | 326 |
Feature activation | +2.96e+00 |
Pos logit contributions | |
Robert | +0.059 |
Eva | +0.059 |
chard | +0.057 |
wrists | +0.057 |
blonde | +0.055 |
Neg logit contributions | |
loo | -0.052 |
auc | -0.045 |
ick | -0.044 |
iren | -0.043 |
of | -0.042 |
Token | sound |
Token ID | 2128 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
loo | +0.108 |
auc | +0.090 |
iren | +0.089 |
ick | +0.088 |
arl | +0.086 |
Neg logit contributions | |
Robert | -0.135 |
Eva | -0.132 |
yes | -0.126 |
chard | -0.126 |
wrists | -0.125 |
Token | ?" |
Token ID | 1701 |
Feature activation | -4.40e+00 |
Pos logit contributions | |
loo | +0.035 |
ick | +0.028 |
auc | +0.028 |
iren | +0.028 |
of | +0.027 |
Neg logit contributions | |
Robert | -0.050 |
Eva | -0.050 |
chard | -0.050 |
wrists | -0.049 |
blonde | -0.047 |
Token |
|
Token ID | 198 |
Feature activation | -5.85e-02 |
Pos logit contributions | |
chard | +0.204 |
wrists | +0.199 |
cious | +0.195 |
blonde | +0.192 |
Eva | +0.184 |
Neg logit contributions | |
of | -0.144 |
loo | -0.136 |
iren | -0.124 |
ick | -0.119 |
pour | -0.117 |
Token | ained |
Token ID | 1328 |
Feature activation | +3.62e+00 |
Pos logit contributions | |
loo | +0.150 |
of | +0.129 |
ig | +0.113 |
outage | +0.106 |
tunnel | +0.104 |
Neg logit contributions | |
wrists | -0.342 |
blonde | -0.340 |
chard | -0.336 |
Eva | -0.328 |
Robert | -0.321 |
Token | and |
Token ID | 290 |
Feature activation | -3.82e-01 |
Pos logit contributions | |
loo | +0.142 |
auc | +0.119 |
iren | +0.119 |
ick | +0.116 |
arl | +0.116 |
Neg logit contributions | |
Robert | -0.150 |
Eva | -0.147 |
chard | -0.146 |
Claire | -0.140 |
wrists | -0.140 |
Token | r |
Token ID | 374 |
Feature activation | +6.54e+00 |
Pos logit contributions | |
Eva | +0.016 |
chard | +0.016 |
wrists | +0.016 |
Robert | +0.016 |
blonde | +0.015 |
Neg logit contributions | |
loo | -0.015 |
ick | -0.013 |
iren | -0.013 |
auc | -0.012 |
arl | -0.012 |
Token | ained |
Token ID | 1328 |
Feature activation | +3.55e+00 |
Pos logit contributions | |
loo | +0.146 |
of | +0.125 |
ig | +0.103 |
iren | +0.096 |
tunnel | +0.096 |
Neg logit contributions | |
wrists | -0.356 |
blonde | -0.349 |
chard | -0.342 |
Eva | -0.336 |
Robert | -0.328 |
Token | and |
Token ID | 290 |
Feature activation | -3.67e-01 |
Pos logit contributions | |
loo | +0.137 |
auc | +0.115 |
iren | +0.114 |
ick | +0.112 |
arl | +0.110 |
Neg logit contributions | |
Robert | -0.149 |
Eva | -0.148 |
chard | -0.147 |
wrists | -0.142 |
blonde | -0.140 |
Token | r |
Token ID | 374 |
Feature activation | +6.51e+00 |
Pos logit contributions | |
chard | +0.016 |
wrists | +0.016 |
Eva | +0.016 |
Robert | +0.016 |
blonde | +0.015 |
Neg logit contributions | |
loo | -0.014 |
ick | -0.012 |
iren | -0.012 |
auc | -0.012 |
arl | -0.012 |
Token | ained |
Token ID | 1328 |
Feature activation | +3.51e+00 |
Pos logit contributions | |
loo | +0.151 |
of | +0.127 |
ig | +0.104 |
iren | +0.102 |
auc | +0.099 |
Neg logit contributions | |
wrists | -0.352 |
blonde | -0.343 |
chard | -0.335 |
Eva | -0.326 |
rice | -0.323 |
Token | . |
Token ID | 13 |
Feature activation | +3.34e-01 |
Pos logit contributions | |
loo | +0.131 |
auc | +0.109 |
iren | +0.107 |
ick | +0.106 |
rim | +0.104 |
Neg logit contributions | |
Robert | -0.151 |
Eva | -0.150 |
chard | -0.150 |
wrists | -0.145 |
blonde | -0.143 |
Token | The |
Token ID | 383 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
of | +0.013 |
loo | +0.012 |
ick | +0.012 |
pour | +0.011 |
late | +0.011 |
Neg logit contributions | |
chard | -0.014 |
wrists | -0.013 |
cious | -0.012 |
Eva | -0.012 |
Robert | -0.012 |
Token | rain |
Token ID | 6290 |
Feature activation | +4.86e+00 |
Pos logit contributions | |
Eva | +0.010 |
chard | +0.010 |
Robert | +0.010 |
wrists | +0.010 |
blonde | +0.010 |
Neg logit contributions | |
loo | -0.006 |
of | -0.005 |
ick | -0.005 |
iren | -0.005 |
auc | -0.005 |
Token | was |
Token ID | 373 |
Feature activation | +3.52e-01 |
Pos logit contributions | |
loo | +0.209 |
ick | +0.177 |
iren | +0.176 |
auc | +0.176 |
arl | +0.176 |
Neg logit contributions | |
wrists | -0.183 |
Eva | -0.179 |
blonde | -0.176 |
Robert | -0.175 |
chard | -0.175 |
Token | . |
Token ID | 13 |
Feature activation | +7.08e-01 |
Pos logit contributions | |
loo | +0.033 |
iren | +0.027 |
ick | +0.027 |
auc | +0.026 |
of | +0.026 |
Neg logit contributions | |
chard | -0.052 |
Eva | -0.052 |
Robert | -0.052 |
wrists | -0.051 |
blonde | -0.049 |
Token | But |
Token ID | 887 |
Feature activation | +8.98e-03 |
Pos logit contributions | |
of | +0.024 |
loo | +0.022 |
ick | +0.020 |
iren | +0.020 |
pour | +0.020 |
Neg logit contributions | |
wrists | -0.032 |
chard | -0.031 |
cious | -0.030 |
Eva | -0.029 |
rice | -0.028 |
Token | it |
Token ID | 340 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
loo | +0.000 |
of | +0.000 |
ick | +0.000 |
iren | +0.000 |
auc | +0.000 |
Neg logit contributions | |
Eva | -0.000 |
wrists | -0.000 |
Robert | -0.000 |
chard | -0.000 |
blonde | -0.000 |
Token | was |
Token ID | 373 |
Feature activation | +8.12e-01 |
Pos logit contributions | |
loo | +0.010 |
iren | +0.008 |
ick | +0.008 |
of | +0.008 |
auc | +0.008 |
Neg logit contributions | |
chard | -0.011 |
Eva | -0.011 |
Robert | -0.011 |
wrists | -0.010 |
blonde | -0.010 |
Token | too |
Token ID | 1165 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
loo | +0.032 |
auc | +0.028 |
iren | +0.027 |
ick | +0.026 |
arl | +0.026 |
Neg logit contributions | |
chard | -0.036 |
Eva | -0.034 |
Robert | -0.034 |
blonde | -0.034 |
wrists | -0.033 |
Token | wind |
Token ID | 2344 |
Feature activation | +6.04e+00 |
Pos logit contributions | |
loo | +0.065 |
auc | +0.058 |
iren | +0.057 |
rim | +0.056 |
ick | +0.052 |
Neg logit contributions | |
chard | -0.058 |
Eva | -0.052 |
blonde | -0.051 |
Robert | -0.050 |
Claire | -0.048 |
Token | y |
Token ID | 88 |
Feature activation | +5.54e-01 |
Pos logit contributions | |
loo | +0.198 |
auc | +0.159 |
iren | +0.156 |
pour | +0.155 |
ig | +0.148 |
Neg logit contributions | |
chard | -0.292 |
Robert | -0.286 |
wrists | -0.281 |
Eva | -0.279 |
rice | -0.275 |
Token | outside |
Token ID | 2354 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
loo | +0.020 |
iren | +0.017 |
auc | +0.017 |
ick | +0.016 |
pour | +0.016 |
Neg logit contributions | |
Robert | -0.025 |
chard | -0.025 |
Eva | -0.025 |
wrists | -0.024 |
blonde | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | +7.33e-01 |
Pos logit contributions | |
Robert | +0.063 |
chard | +0.063 |
Eva | +0.062 |
wrists | +0.061 |
blonde | +0.061 |
Neg logit contributions | |
loo | -0.058 |
iren | -0.051 |
arl | -0.049 |
rim | -0.048 |
auc | -0.048 |
Token | So |
Token ID | 1406 |
Feature activation | -2.26e+00 |
Pos logit contributions | |
of | +0.023 |
loo | +0.021 |
ick | +0.020 |
iren | +0.019 |
pour | +0.019 |
Neg logit contributions | |
chard | -0.035 |
wrists | -0.035 |
cious | -0.033 |
ya | -0.032 |
Eva | -0.032 |
Token | the |
Token ID | 262 |
Feature activation | +7.01e-01 |
Pos logit contributions | |
wrists | +0.134 |
chard | +0.133 |
Eva | +0.131 |
Robert | +0.129 |
blonde | +0.127 |
Neg logit contributions | |
loo | -0.047 |
of | -0.043 |
ick | -0.038 |
ig | -0.032 |
pour | -0.032 |
Token | home |
Token ID | 1363 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
loo | +0.024 |
auc | +0.022 |
ick | +0.021 |
iren | +0.021 |
arl | +0.020 |
Neg logit contributions | |
chard | -0.028 |
Robert | -0.026 |
Eva | -0.025 |
Claire | -0.024 |
wrists | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +8.74e-01 |
Pos logit contributions | |
loo | +0.090 |
iren | +0.075 |
auc | +0.075 |
ick | +0.074 |
of | +0.072 |
Neg logit contributions | |
Eva | -0.122 |
chard | -0.122 |
Robert | -0.122 |
wrists | -0.119 |
Claire | -0.115 |
Token | It |
Token ID | 632 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
loo | +0.034 |
of | +0.032 |
auc | +0.030 |
iren | +0.030 |
ick | +0.030 |
Neg logit contributions | |
chard | -0.036 |
Eva | -0.035 |
wrists | -0.035 |
Robert | -0.035 |
blonde | -0.033 |
Token | 's |
Token ID | 338 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
Eva | +0.051 |
wrists | +0.051 |
Robert | +0.051 |
chard | +0.050 |
Claire | +0.047 |
Neg logit contributions | |
loo | -0.051 |
iren | -0.045 |
ick | -0.044 |
arl | -0.043 |
of | -0.043 |
Token | too |
Token ID | 1165 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
loo | +0.042 |
of | +0.036 |
ick | +0.035 |
iren | +0.035 |
auc | +0.035 |
Neg logit contributions | |
chard | -0.051 |
Eva | -0.051 |
Robert | -0.051 |
wrists | -0.050 |
blonde | -0.048 |
Token | wind |
Token ID | 2344 |
Feature activation | +5.60e+00 |
Pos logit contributions | |
loo | +0.044 |
auc | +0.038 |
iren | +0.038 |
ick | +0.037 |
of | +0.037 |
Neg logit contributions | |
chard | -0.050 |
Eva | -0.049 |
Robert | -0.049 |
wrists | -0.047 |
blonde | -0.046 |
Token | y |
Token ID | 88 |
Feature activation | +2.69e-01 |
Pos logit contributions | |
loo | +0.179 |
ick | +0.147 |
auc | +0.146 |
of | +0.144 |
iren | +0.144 |
Neg logit contributions | |
wrists | -0.268 |
chard | -0.265 |
Eva | -0.265 |
Robert | -0.265 |
rice | -0.251 |
Token | here |
Token ID | 994 |
Feature activation | -3.32e+00 |
Pos logit contributions | |
loo | +0.009 |
iren | +0.008 |
ick | +0.008 |
auc | +0.008 |
of | +0.007 |
Neg logit contributions | |
chard | -0.012 |
Robert | -0.012 |
Eva | -0.012 |
wrists | -0.012 |
blonde | -0.011 |
Token | ." |
Token ID | 526 |
Feature activation | -2.99e+00 |
Pos logit contributions | |
Eva | +0.161 |
chard | +0.161 |
Robert | +0.161 |
wrists | +0.160 |
blonde | +0.155 |
Neg logit contributions | |
loo | -0.106 |
of | -0.092 |
ick | -0.089 |
iren | -0.086 |
auc | -0.085 |
Token |
|
Token ID | 198 |
Feature activation | +7.28e-03 |
Pos logit contributions | |
chard | +0.130 |
wrists | +0.130 |
Eva | +0.129 |
Robert | +0.129 |
blonde | +0.124 |
Neg logit contributions | |
loo | -0.114 |
of | -0.101 |
ick | -0.097 |
iren | -0.097 |
auc | -0.095 |
Token |
|
Token ID | 198 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
loo | +0.000 |
ick | +0.000 |
auc | +0.000 |
iren | +0.000 |
pour | +0.000 |
Neg logit contributions | |
Eva | -0.000 |
Robert | -0.000 |
chard | -0.000 |
wrists | -0.000 |
Claire | -0.000 |
Token | be |
Token ID | 307 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
Eva | +0.007 |
Robert | +0.007 |
chard | +0.007 |
wrists | +0.007 |
blonde | +0.007 |
Neg logit contributions | |
loo | -0.007 |
auc | -0.006 |
ick | -0.006 |
iren | -0.006 |
of | -0.006 |
Token | careful |
Token ID | 8161 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
loo | +0.072 |
of | +0.068 |
ick | +0.064 |
iren | +0.063 |
auc | +0.061 |
Neg logit contributions | |
chard | -0.085 |
wrists | -0.085 |
Robert | -0.085 |
Eva | -0.085 |
blonde | -0.079 |
Token | , |
Token ID | 11 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
loo | +0.060 |
auc | +0.048 |
iren | +0.048 |
ick | +0.047 |
pour | +0.047 |
Neg logit contributions | |
Robert | -0.078 |
Eva | -0.077 |
wrists | -0.076 |
chard | -0.075 |
blonde | -0.073 |
Token | because |
Token ID | 780 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
loo | +0.014 |
of | +0.012 |
ick | +0.012 |
iren | +0.012 |
auc | +0.012 |
Neg logit contributions | |
Eva | -0.010 |
Robert | -0.010 |
chard | -0.010 |
wrists | -0.010 |
blonde | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | +3.28e-01 |
Pos logit contributions | |
loo | +0.110 |
arl | +0.107 |
rim | +0.104 |
iren | +0.102 |
ick | +0.102 |
Neg logit contributions | |
chard | -0.075 |
Robert | -0.074 |
Eva | -0.069 |
yes | -0.067 |
blonde | -0.066 |
Token | tunnel |
Token ID | 13275 |
Feature activation | +5.36e+00 |
Pos logit contributions | |
loo | +0.007 |
rim | +0.006 |
iren | +0.006 |
ick | +0.006 |
arl | +0.006 |
Neg logit contributions | |
blonde | -0.018 |
wrists | -0.018 |
Robert | -0.017 |
Eva | -0.017 |
chard | -0.017 |
Token | was |
Token ID | 373 |
Feature activation | +6.99e-01 |
Pos logit contributions | |
loo | +0.216 |
ick | +0.187 |
iren | +0.187 |
arl | +0.186 |
auc | +0.181 |
Neg logit contributions | |
Robert | -0.212 |
wrists | -0.210 |
Eva | -0.210 |
chard | -0.207 |
blonde | -0.199 |
Token | fragile |
Token ID | 21049 |
Feature activation | +2.48e+00 |
Pos logit contributions | |
loo | +0.031 |
auc | +0.027 |
iren | +0.026 |
ick | +0.026 |
arl | +0.025 |
Neg logit contributions | |
chard | -0.027 |
Robert | -0.026 |
Eva | -0.026 |
blonde | -0.025 |
wrists | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | +3.88e-01 |
Pos logit contributions | |
loo | +0.087 |
iren | +0.073 |
ick | +0.070 |
auc | +0.069 |
arl | +0.069 |
Neg logit contributions | |
Robert | -0.115 |
wrists | -0.113 |
Eva | -0.113 |
chard | -0.112 |
blonde | -0.110 |
Token | |
Token ID | 220 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
loo | +0.016 |
of | +0.015 |
ick | +0.014 |
iren | +0.014 |
auc | +0.014 |
Neg logit contributions | |
chard | -0.015 |
wrists | -0.015 |
Eva | -0.015 |
Robert | -0.015 |
blonde | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | +3.07e-01 |
Pos logit contributions | |
chard | +0.102 |
wrists | +0.102 |
Eva | +0.101 |
Robert | +0.101 |
blonde | +0.097 |
Neg logit contributions | |
loo | -0.087 |
of | -0.076 |
ick | -0.075 |
iren | -0.073 |
auc | -0.073 |
Token | as |
Token ID | 355 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
loo | +0.012 |
of | +0.009 |
ick | +0.009 |
iren | +0.009 |
auc | +0.009 |
Neg logit contributions | |
Eva | -0.022 |
Robert | -0.022 |
chard | -0.022 |
wrists | -0.022 |
blonde | -0.021 |
Token | he |
Token ID | 339 |
Feature activation | +5.81e-01 |
Pos logit contributions | |
wrists | +0.093 |
chard | +0.091 |
Eva | +0.090 |
Robert | +0.089 |
blonde | +0.087 |
Neg logit contributions | |
loo | -0.067 |
of | -0.060 |
ick | -0.057 |
auc | -0.053 |
iren | -0.053 |
Token | walked |
Token ID | 6807 |
Feature activation | -6.59e-01 |
Pos logit contributions | |
loo | +0.024 |
of | +0.022 |
ick | +0.021 |
iren | +0.020 |
auc | +0.020 |
Neg logit contributions | |
chard | -0.024 |
Eva | -0.023 |
Robert | -0.023 |
wrists | -0.023 |
blonde | -0.022 |
Token | home |
Token ID | 1363 |
Feature activation | +3.06e+00 |
Pos logit contributions | |
Eva | +0.033 |
wrists | +0.032 |
chard | +0.032 |
Robert | +0.032 |
blonde | +0.031 |
Neg logit contributions | |
loo | -0.021 |
of | -0.018 |
ick | -0.017 |
iren | -0.017 |
auc | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
loo | +0.102 |
of | +0.085 |
ick | +0.085 |
auc | +0.083 |
iren | +0.083 |
Neg logit contributions | |
Eva | -0.146 |
chard | -0.146 |
Robert | -0.146 |
wrists | -0.145 |
blonde | -0.140 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +4.84e+00 |
Pos logit contributions | |
of | +0.006 |
loo | +0.006 |
late | +0.005 |
iren | +0.005 |
ick | +0.004 |
Neg logit contributions | |
wrists | -0.011 |
chard | -0.011 |
blonde | -0.010 |
cious | -0.010 |
Eva | -0.010 |
Token | Once |
Token ID | 7454 |
Feature activation | +7.43e-01 |
Pos logit contributions | |
loo | +0.160 |
of | +0.143 |
ick | +0.134 |
iren | +0.128 |
pour | +0.123 |
Neg logit contributions | |
wrists | -0.244 |
chard | -0.239 |
Eva | -0.231 |
Robert | -0.228 |
blonde | -0.228 |
Token | upon |
Token ID | 2402 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
loo | +0.025 |
iren | +0.021 |
auc | +0.021 |
ick | +0.021 |
of | +0.020 |
Neg logit contributions | |
Robert | -0.035 |
Eva | -0.035 |
chard | -0.035 |
wrists | -0.033 |
blonde | -0.033 |
Token | a |
Token ID | 257 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
of | +0.062 |
loo | +0.060 |
ick | +0.050 |
late | +0.049 |
ig | +0.048 |
Neg logit contributions | |
wrists | -0.106 |
chard | -0.100 |
Eva | -0.098 |
Robert | -0.097 |
rice | -0.094 |
Token | time |
Token ID | 640 |
Feature activation | +2.64e+00 |
Pos logit contributions | |
chard | +0.058 |
wrists | +0.057 |
Robert | +0.055 |
Eva | +0.054 |
Claire | +0.052 |
Neg logit contributions | |
loo | -0.025 |
of | -0.023 |
ick | -0.023 |
arl | -0.019 |
ig | -0.018 |
Token | there |
Token ID | 612 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
loo | +0.070 |
of | +0.058 |
ick | +0.056 |
auc | +0.053 |
iren | +0.053 |
Neg logit contributions | |
wrists | -0.145 |
chard | -0.144 |
Eva | -0.143 |
Robert | -0.143 |
blonde | -0.139 |
Token | prince |
Token ID | 19716 |
Feature activation | -2.46e+00 |
Pos logit contributions | |
loo | +0.040 |
of | +0.034 |
ick | +0.033 |
auc | +0.032 |
ig | +0.030 |
Neg logit contributions | |
Eva | -0.065 |
chard | -0.065 |
Robert | -0.065 |
wrists | -0.064 |
blonde | -0.062 |
Token | ?" |
Token ID | 1701 |
Feature activation | -3.90e+00 |
Pos logit contributions | |
Robert | +0.142 |
Eva | +0.141 |
chard | +0.139 |
wrists | +0.136 |
blonde | +0.133 |
Neg logit contributions | |
loo | -0.058 |
iren | -0.045 |
ick | -0.044 |
auc | -0.044 |
of | -0.043 |
Token |
|
Token ID | 198 |
Feature activation | -7.79e-02 |
Pos logit contributions | |
wrists | +0.197 |
chard | +0.191 |
Eva | +0.183 |
blonde | +0.180 |
cious | +0.179 |
Neg logit contributions | |
loo | -0.115 |
of | -0.112 |
iren | -0.106 |
ick | -0.097 |
auc | -0.094 |
Token |
|
Token ID | 198 |
Feature activation | -6.62e-02 |
Pos logit contributions | |
wrists | +0.004 |
chard | +0.004 |
Eva | +0.004 |
blonde | +0.004 |
Robert | +0.004 |
Neg logit contributions | |
loo | -0.003 |
of | -0.002 |
iren | -0.002 |
ick | -0.002 |
auc | -0.002 |
Token | The |
Token ID | 464 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
Eva | +0.003 |
chard | +0.003 |
Robert | +0.003 |
wrists | +0.003 |
blonde | +0.003 |
Neg logit contributions | |
loo | -0.003 |
of | -0.002 |
ick | -0.002 |
iren | -0.002 |
auc | -0.002 |
Token | little |
Token ID | 1310 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
wrists | +0.006 |
Eva | +0.005 |
chard | +0.005 |
Robert | +0.005 |
blonde | +0.005 |
Neg logit contributions | |
loo | -0.008 |
of | -0.007 |
ick | -0.007 |
ig | -0.007 |
auc | -0.007 |
Token | prince |
Token ID | 19716 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
loo | +0.061 |
of | +0.055 |
ick | +0.054 |
auc | +0.053 |
iren | +0.053 |
Neg logit contributions | |
Eva | -0.041 |
wrists | -0.041 |
chard | -0.041 |
Robert | -0.041 |
blonde | -0.039 |
Token | showed |
Token ID | 3751 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
chard | +0.108 |
wrists | +0.104 |
blonde | +0.102 |
Eva | +0.101 |
Robert | +0.100 |
Neg logit contributions | |
loo | -0.088 |
of | -0.087 |
ick | -0.080 |
ig | -0.075 |
iren | -0.074 |
Token | his |
Token ID | 465 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
chard | +0.063 |
wrists | +0.062 |
Eva | +0.062 |
Robert | +0.061 |
blonde | +0.060 |
Neg logit contributions | |
loo | -0.047 |
of | -0.042 |
ick | -0.040 |
iren | -0.039 |
auc | -0.039 |
Token | mum |
Token ID | 25682 |
Feature activation | +7.69e-01 |
Pos logit contributions | |
chard | +0.009 |
Eva | +0.009 |
Robert | +0.008 |
wrists | +0.008 |
blonde | +0.008 |
Neg logit contributions | |
loo | -0.005 |
of | -0.005 |
ig | -0.005 |
tunnel | -0.004 |
ick | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | +3.46e-01 |
Pos logit contributions | |
of | +0.022 |
loo | +0.021 |
ick | +0.018 |
iren | +0.018 |
ig | +0.018 |
Neg logit contributions | |
chard | -0.039 |
wrists | -0.039 |
blonde | -0.038 |
Eva | -0.038 |
Robert | -0.037 |
Token | a |
Token ID | 257 |
Feature activation | -7.67e-01 |
Pos logit contributions | |
loo | +0.028 |
of | +0.028 |
ick | +0.025 |
ig | +0.023 |
late | +0.023 |
Neg logit contributions | |
wrists | -0.042 |
chard | -0.040 |
Eva | -0.040 |
Robert | -0.039 |
rice | -0.039 |
Token | small |
Token ID | 1402 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
wrists | +0.035 |
Eva | +0.035 |
Robert | +0.035 |
chard | +0.033 |
blonde | +0.033 |
Neg logit contributions | |
loo | -0.027 |
of | -0.024 |
ick | -0.023 |
ig | -0.022 |
auc | -0.021 |
Token | girl |
Token ID | 2576 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
loo | +0.067 |
iren | +0.058 |
auc | +0.058 |
of | +0.058 |
ick | +0.058 |
Neg logit contributions | |
chard | -0.073 |
Robert | -0.072 |
Eva | -0.072 |
wrists | -0.071 |
blonde | -0.069 |
Token | named |
Token ID | 3706 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
chard | +0.040 |
Eva | +0.039 |
wrists | +0.039 |
Robert | +0.038 |
blonde | +0.038 |
Neg logit contributions | |
loo | -0.061 |
iren | -0.054 |
ick | -0.053 |
arl | -0.053 |
auc | -0.053 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
of | +0.004 |
loo | +0.004 |
ick | +0.004 |
arl | +0.003 |
auc | +0.003 |
Neg logit contributions | |
wrists | -0.007 |
Robert | -0.007 |
chard | -0.007 |
Edward | -0.007 |
Percy | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
loo | +0.006 |
iren | +0.005 |
auc | +0.005 |
ick | +0.005 |
arl | +0.005 |
Neg logit contributions | |
chard | -0.009 |
Robert | -0.009 |
Eva | -0.009 |
wrists | -0.008 |
blonde | -0.008 |
Token | She |
Token ID | 1375 |
Feature activation | +8.87e-02 |
Pos logit contributions | |
loo | +0.047 |
of | +0.042 |
ick | +0.040 |
auc | +0.039 |
iren | +0.039 |
Neg logit contributions | |
chard | -0.052 |
Robert | -0.052 |
wrists | -0.052 |
Eva | -0.052 |
blonde | -0.048 |
Token | lived |
Token ID | 5615 |
Feature activation | +6.94e-01 |
Pos logit contributions | |
loo | +0.005 |
of | +0.004 |
ick | +0.004 |
iren | +0.004 |
auc | +0.004 |
Neg logit contributions | |
Eva | -0.003 |
chard | -0.003 |
wrists | -0.003 |
Robert | -0.003 |
blonde | -0.002 |
Token | in |
Token ID | 287 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
loo | +0.027 |
iren | +0.024 |
auc | +0.024 |
ick | +0.023 |
outage | +0.022 |
Neg logit contributions | |
chard | -0.030 |
Eva | -0.028 |
Robert | -0.028 |
wrists | -0.027 |
Claire | -0.027 |
Token | a |
Token ID | 257 |
Feature activation | -7.92e-01 |
Pos logit contributions | |
Eva | +0.087 |
Robert | +0.086 |
chard | +0.086 |
wrists | +0.084 |
blonde | +0.082 |
Neg logit contributions | |
loo | -0.103 |
of | -0.090 |
ick | -0.089 |
auc | -0.089 |
iren | -0.089 |
Token | warm |
Token ID | 5814 |
Feature activation | +3.31e+00 |
Pos logit contributions | |
chard | +0.032 |
Eva | +0.031 |
Robert | +0.031 |
wrists | +0.031 |
blonde | +0.030 |
Neg logit contributions | |
loo | -0.033 |
iren | -0.029 |
auc | -0.028 |
of | -0.028 |
ick | -0.028 |
Token | She |
Token ID | 3347 |
Feature activation | +7.19e-01 |
Pos logit contributions | |
loo | +0.010 |
of | +0.010 |
iren | +0.009 |
ick | +0.008 |
auc | +0.008 |
Neg logit contributions | |
wrists | -0.018 |
chard | -0.018 |
blonde | -0.017 |
Robert | -0.017 |
Eva | -0.017 |
Token | said |
Token ID | 531 |
Feature activation | +7.13e-01 |
Pos logit contributions | |
of | +0.025 |
loo | +0.024 |
auc | +0.022 |
dri | +0.021 |
ick | +0.021 |
Neg logit contributions | |
chard | -0.032 |
Robert | -0.031 |
Eva | -0.030 |
blonde | -0.029 |
wrists | -0.029 |
Token | , |
Token ID | 11 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
of | +0.017 |
loo | +0.014 |
ful | +0.013 |
auc | +0.012 |
ick | +0.012 |
Neg logit contributions | |
chard | -0.041 |
wrists | -0.040 |
Eva | -0.038 |
blonde | -0.038 |
Robert | -0.038 |
Token | " |
Token ID | 366 |
Feature activation | -3.37e+00 |
Pos logit contributions | |
of | +0.008 |
loo | +0.007 |
iren | +0.007 |
ful | +0.007 |
arl | +0.007 |
Neg logit contributions | |
wrists | -0.009 |
chard | -0.008 |
blonde | -0.008 |
Eva | -0.008 |
Robert | -0.008 |
Token | Tom |
Token ID | 13787 |
Feature activation | +6.56e-01 |
Pos logit contributions | |
wrists | +0.145 |
chard | +0.144 |
Eva | +0.138 |
Robert | +0.138 |
blonde | +0.134 |
Neg logit contributions | |
loo | -0.115 |
of | -0.108 |
iren | -0.105 |
pour | -0.102 |
auc | -0.099 |
Token | , |
Token ID | 11 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
loo | +0.018 |
of | +0.015 |
ick | +0.014 |
iren | +0.014 |
auc | +0.014 |
Neg logit contributions | |
Eva | -0.035 |
Robert | -0.035 |
chard | -0.035 |
wrists | -0.035 |
blonde | -0.034 |
Token | why |
Token ID | 1521 |
Feature activation | -6.35e+00 |
Pos logit contributions | |
loo | +0.006 |
of | +0.006 |
ick | +0.006 |
iren | +0.005 |
auc | +0.005 |
Neg logit contributions | |
chard | -0.008 |
Eva | -0.008 |
Robert | -0.008 |
wrists | -0.008 |
blonde | -0.007 |
Token | don |
Token ID | 836 |
Feature activation | +6.64e-01 |
Pos logit contributions | |
Eva | +0.271 |
Robert | +0.269 |
chard | +0.262 |
wrists | +0.258 |
blonde | +0.251 |
Neg logit contributions | |
loo | -0.246 |
iren | -0.215 |
ick | -0.211 |
auc | -0.207 |
of | -0.203 |
Token | 't |
Token ID | 470 |
Feature activation | +3.53e+00 |
Pos logit contributions | |
loo | +0.031 |
auc | +0.027 |
ig | +0.027 |
row | +0.026 |
ick | +0.026 |
Neg logit contributions | |
Robert | -0.025 |
Eva | -0.023 |
Stella | -0.022 |
chard | -0.022 |
Edward | -0.021 |
Token | you |
Token ID | 345 |
Feature activation | +3.91e-01 |
Pos logit contributions | |
loo | +0.147 |
iren | +0.123 |
auc | +0.123 |
ick | +0.121 |
arl | +0.120 |
Neg logit contributions | |
Robert | -0.157 |
Eva | -0.152 |
chard | -0.148 |
wrists | -0.146 |
Claire | -0.145 |
Token | take |
Token ID | 1011 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
loo | +0.015 |
of | +0.012 |
ick | +0.012 |
iren | +0.012 |
auc | +0.012 |
Neg logit contributions | |
Eva | -0.017 |
chard | -0.017 |
Robert | -0.017 |
wrists | -0.017 |
blonde | -0.016 |
Token | tried |
Token ID | 3088 |
Feature activation | +2.20e+00 |
Pos logit contributions | |
Eva | +0.013 |
Robert | +0.013 |
chard | +0.013 |
wrists | +0.013 |
blonde | +0.013 |
Neg logit contributions | |
loo | -0.013 |
of | -0.011 |
ick | -0.011 |
iren | -0.011 |
auc | -0.011 |
Token | to |
Token ID | 284 |
Feature activation | +2.54e-01 |
Pos logit contributions | |
loo | +0.067 |
of | +0.058 |
ick | +0.055 |
iren | +0.054 |
auc | +0.053 |
Neg logit contributions | |
Eva | -0.110 |
chard | -0.110 |
Robert | -0.109 |
wrists | -0.109 |
blonde | -0.106 |
Token | dig |
Token ID | 3100 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
loo | +0.010 |
of | +0.008 |
ick | +0.008 |
iren | +0.008 |
auc | +0.008 |
Neg logit contributions | |
Eva | -0.011 |
Robert | -0.011 |
chard | -0.011 |
wrists | -0.011 |
blonde | -0.010 |
Token | a |
Token ID | 257 |
Feature activation | -6.21e-01 |
Pos logit contributions | |
loo | +0.050 |
of | +0.045 |
ick | +0.043 |
auc | +0.042 |
iren | +0.041 |
Neg logit contributions | |
chard | -0.064 |
wrists | -0.064 |
Eva | -0.064 |
Robert | -0.064 |
blonde | -0.062 |
Token | hole |
Token ID | 7604 |
Feature activation | +6.96e+00 |
Pos logit contributions | |
chard | +0.032 |
wrists | +0.031 |
Eva | +0.031 |
Robert | +0.031 |
blonde | +0.030 |
Neg logit contributions | |
loo | -0.018 |
iren | -0.015 |
ick | -0.015 |
auc | -0.014 |
arl | -0.014 |
Token | for |
Token ID | 329 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
loo | +0.237 |
iren | +0.202 |
ick | +0.195 |
auc | +0.191 |
arl | +0.183 |
Neg logit contributions | |
chard | -0.325 |
Eva | -0.324 |
Robert | -0.322 |
wrists | -0.313 |
Claire | -0.306 |
Token | more |
Token ID | 517 |
Feature activation | -3.46e-01 |
Pos logit contributions | |
loo | +0.061 |
auc | +0.053 |
iren | +0.052 |
ick | +0.052 |
of | +0.052 |
Neg logit contributions | |
Eva | -0.061 |
Robert | -0.060 |
chard | -0.060 |
wrists | -0.059 |
blonde | -0.057 |
Token | light |
Token ID | 1657 |
Feature activation | +3.76e+00 |
Pos logit contributions | |
wrists | +0.014 |
Robert | +0.014 |
blonde | +0.014 |
chard | +0.014 |
Eva | +0.014 |
Neg logit contributions | |
loo | -0.013 |
iren | -0.012 |
ick | -0.012 |
auc | -0.011 |
arl | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
loo | +0.126 |
iren | +0.107 |
ick | +0.106 |
auc | +0.102 |
of | +0.099 |
Neg logit contributions | |
Eva | -0.175 |
wrists | -0.175 |
Robert | -0.173 |
chard | -0.172 |
blonde | -0.166 |
Token | but |
Token ID | 475 |
Feature activation | -2.16e+00 |
Pos logit contributions | |
loo | +0.007 |
of | +0.006 |
ick | +0.006 |
auc | +0.006 |
iren | +0.006 |
Neg logit contributions | |
Eva | -0.005 |
Robert | -0.005 |
chard | -0.005 |
wrists | -0.005 |
blonde | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
wrists | +0.089 |
Eva | +0.088 |
Robert | +0.088 |
chard | +0.087 |
blonde | +0.086 |
Neg logit contributions | |
loo | -0.085 |
of | -0.076 |
ick | -0.073 |
iren | -0.070 |
auc | -0.070 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +4.89e+00 |
Pos logit contributions | |
loo | +0.022 |
of | +0.021 |
ick | +0.018 |
iren | +0.018 |
pour | +0.017 |
Neg logit contributions | |
chard | -0.032 |
wrists | -0.032 |
blonde | -0.029 |
Eva | -0.029 |
Robert | -0.029 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
loo | +0.175 |
of | +0.153 |
ick | +0.148 |
iren | +0.143 |
auc | +0.141 |
Neg logit contributions | |
wrists | -0.225 |
chard | -0.224 |
Eva | -0.221 |
Robert | -0.220 |
blonde | -0.214 |
Token | upon |
Token ID | 2402 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
loo | +0.038 |
iren | +0.033 |
auc | +0.033 |
ick | +0.031 |
pour | +0.031 |
Neg logit contributions | |
Eva | -0.050 |
Robert | -0.050 |
chard | -0.049 |
wrists | -0.047 |
Claire | -0.047 |
Token | a |
Token ID | 257 |
Feature activation | -6.76e-01 |
Pos logit contributions | |
of | +0.074 |
loo | +0.069 |
ick | +0.060 |
late | +0.059 |
ig | +0.055 |
Neg logit contributions | |
wrists | -0.129 |
chard | -0.121 |
Robert | -0.117 |
Eva | -0.116 |
rice | -0.113 |
Token | time |
Token ID | 640 |
Feature activation | +2.77e+00 |
Pos logit contributions | |
chard | +0.040 |
wrists | +0.039 |
Robert | +0.037 |
Eva | +0.037 |
nam | +0.036 |
Neg logit contributions | |
loo | -0.015 |
of | -0.014 |
ick | -0.014 |
arl | -0.011 |
ig | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
loo | +0.080 |
of | +0.065 |
ick | +0.064 |
iren | +0.064 |
auc | +0.063 |
Neg logit contributions | |
Eva | -0.145 |
Robert | -0.144 |
chard | -0.144 |
wrists | -0.142 |
blonde | -0.139 |
Token | there |
Token ID | 612 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
loo | +0.030 |
of | +0.026 |
ick | +0.024 |
auc | +0.022 |
iren | +0.022 |
Neg logit contributions | |
wrists | -0.062 |
Robert | -0.061 |
Eva | -0.061 |
chard | -0.060 |
blonde | -0.058 |
Token | was |
Token ID | 373 |
Feature activation | +9.65e-01 |
Pos logit contributions | |
Eva | +0.052 |
Robert | +0.052 |
chard | +0.052 |
wrists | +0.051 |
blonde | +0.050 |
Neg logit contributions | |
loo | -0.043 |
of | -0.037 |
ick | -0.037 |
iren | -0.037 |
auc | -0.036 |
Token | a |
Token ID | 257 |
Feature activation | -6.75e-01 |
Pos logit contributions | |
of | +0.029 |
loo | +0.028 |
ick | +0.025 |
late | +0.024 |
ig | +0.024 |
Neg logit contributions | |
wrists | -0.050 |
chard | -0.047 |
Eva | -0.046 |
rice | -0.046 |
Robert | -0.045 |
Token | little |
Token ID | 1310 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
wrists | +0.031 |
Eva | +0.031 |
Robert | +0.030 |
chard | +0.029 |
blonde | +0.029 |
Neg logit contributions | |
loo | -0.024 |
of | -0.021 |
ick | -0.021 |
ig | -0.020 |
auc | -0.019 |
Token | girl |
Token ID | 2576 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
loo | +0.071 |
iren | +0.062 |
auc | +0.061 |
ick | +0.061 |
of | +0.061 |
Neg logit contributions | |
chard | -0.067 |
Robert | -0.066 |
Eva | -0.066 |
wrists | -0.065 |
blonde | -0.063 |
Token | . |
Token ID | 13 |
Feature activation | +9.99e-01 |
Pos logit contributions | |
chard | +0.030 |
Eva | +0.029 |
Robert | +0.029 |
wrists | +0.029 |
blonde | +0.029 |
Neg logit contributions | |
loo | -0.012 |
of | -0.012 |
ick | -0.010 |
iren | -0.009 |
auc | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | +4.01e-01 |
Pos logit contributions | |
loo | +0.038 |
of | +0.034 |
iren | +0.033 |
ick | +0.033 |
auc | +0.032 |
Neg logit contributions | |
chard | -0.043 |
Eva | -0.043 |
wrists | -0.043 |
Robert | -0.042 |
blonde | -0.041 |
Token |
|
Token ID | 198 |
Feature activation | +4.20e-01 |
Pos logit contributions | |
loo | +0.016 |
of | +0.014 |
ick | +0.014 |
iren | +0.013 |
auc | +0.013 |
Neg logit contributions | |
Eva | -0.017 |
chard | -0.017 |
Robert | -0.017 |
wrists | -0.017 |
blonde | -0.016 |
Token | He |
Token ID | 1544 |
Feature activation | +3.24e-01 |
Pos logit contributions | |
loo | +0.019 |
ick | +0.017 |
auc | +0.017 |
of | +0.016 |
iren | +0.016 |
Neg logit contributions | |
Robert | -0.015 |
chard | -0.015 |
Eva | -0.015 |
wrists | -0.014 |
blonde | -0.014 |
Token | runs |
Token ID | 4539 |
Feature activation | -6.61e-01 |
Pos logit contributions | |
loo | +0.010 |
of | +0.010 |
auc | +0.009 |
iren | +0.009 |
ick | +0.009 |
Neg logit contributions | |
chard | -0.016 |
Robert | -0.015 |
Eva | -0.015 |
wrists | -0.015 |
blonde | -0.015 |
Token | to |
Token ID | 284 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
wrists | +0.034 |
chard | +0.034 |
Eva | +0.034 |
Robert | +0.034 |
blonde | +0.033 |
Neg logit contributions | |
loo | -0.019 |
of | -0.017 |
iren | -0.016 |
ick | -0.016 |
auc | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | +4.94e-01 |
Pos logit contributions | |
wrists | +0.010 |
blonde | +0.010 |
cious | +0.010 |
chard | +0.010 |
Eva | +0.010 |
Neg logit contributions | |
loo | -0.005 |
of | -0.005 |
iren | -0.005 |
ick | -0.004 |
pour | -0.004 |
Token | musician |
Token ID | 21623 |
Feature activation | -4.10e+00 |
Pos logit contributions | |
loo | +0.020 |
of | +0.017 |
ick | +0.017 |
iren | +0.017 |
auc | +0.017 |
Neg logit contributions | |
Eva | -0.020 |
chard | -0.020 |
Robert | -0.020 |
wrists | -0.020 |
blonde | -0.019 |
Token | and |
Token ID | 290 |
Feature activation | +5.79e-01 |
Pos logit contributions | |
chard | +0.212 |
Eva | +0.211 |
Robert | +0.210 |
wrists | +0.210 |
blonde | +0.209 |
Neg logit contributions | |
loo | -0.115 |
of | -0.105 |
iren | -0.094 |
ick | -0.093 |
auc | -0.090 |
Token | says |
Token ID | 1139 |
Feature activation | +3.79e-01 |
Pos logit contributions | |
loo | +0.018 |
of | +0.017 |
iren | +0.017 |
auc | +0.015 |
dri | +0.014 |
Neg logit contributions | |
chard | -0.027 |
blonde | -0.027 |
Robert | -0.026 |
Eva | -0.026 |
wrists | -0.026 |
Token | , |
Token ID | 11 |
Feature activation | +7.53e-01 |
Pos logit contributions | |
of | +0.011 |
loo | +0.011 |
iren | +0.009 |
auc | +0.009 |
ick | +0.009 |
Neg logit contributions | |
wrists | -0.020 |
chard | -0.020 |
Eva | -0.019 |
blonde | -0.019 |
rice | -0.018 |
Token | hon |
Token ID | 3032 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
loo | +0.049 |
ick | +0.042 |
ained | +0.041 |
auc | +0.041 |
pour | +0.039 |
Neg logit contributions | |
Eva | -0.050 |
Robert | -0.049 |
wrists | -0.047 |
Claire | -0.045 |
blonde | -0.045 |
Token | k |
Token ID | 74 |
Feature activation | -2.75e+00 |
Pos logit contributions | |
wrists | +0.051 |
Eva | +0.050 |
rice | +0.049 |
cious | +0.049 |
Robert | +0.048 |
Neg logit contributions | |
loo | -0.097 |
pour | -0.082 |
auc | -0.081 |
ick | -0.079 |
of | -0.079 |
Token | . |
Token ID | 13 |
Feature activation | +9.69e-01 |
Pos logit contributions | |
Eva | +0.100 |
Robert | +0.097 |
yes | +0.095 |
wrists | +0.093 |
Edward | +0.091 |
Neg logit contributions | |
loo | -0.111 |
ick | -0.092 |
auc | -0.091 |
rim | -0.089 |
iren | -0.087 |
Token | They |
Token ID | 1119 |
Feature activation | +5.26e-02 |
Pos logit contributions | |
loo | +0.041 |
of | +0.036 |
ick | +0.035 |
auc | +0.035 |
iren | +0.035 |
Neg logit contributions | |
chard | -0.038 |
Robert | -0.038 |
Eva | -0.038 |
wrists | -0.037 |
blonde | -0.036 |
Token | see |
Token ID | 766 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
loo | +0.002 |
of | +0.002 |
ick | +0.002 |
iren | +0.002 |
auc | +0.002 |
Neg logit contributions | |
Eva | -0.002 |
Robert | -0.002 |
chard | -0.002 |
wrists | -0.002 |
blonde | -0.002 |
Token | a |
Token ID | 257 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
Eva | +0.000 |
Robert | +0.000 |
wrists | +0.000 |
chard | +0.000 |
blonde | +0.000 |
Neg logit contributions | |
loo | -0.001 |
ick | -0.000 |
iren | -0.000 |
auc | -0.000 |
arl | -0.000 |
Token | big |
Token ID | 1263 |
Feature activation | +2.37e+00 |
Pos logit contributions | |
chard | +0.046 |
Robert | +0.045 |
Eva | +0.044 |
wrists | +0.044 |
blonde | +0.042 |
Neg logit contributions | |
loo | -0.043 |
auc | -0.039 |
ick | -0.039 |
of | -0.038 |
iren | -0.036 |
Token | truck |
Token ID | 7779 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
loo | +0.093 |
auc | +0.088 |
ick | +0.086 |
ained | +0.085 |
rim | +0.081 |
Neg logit contributions | |
Robert | -0.092 |
wrists | -0.090 |
Eva | -0.090 |
chard | -0.089 |
blonde | -0.085 |
Token | . |
Token ID | 13 |
Feature activation | +9.70e-01 |
Pos logit contributions | |
Robert | +0.004 |
chard | +0.004 |
Eva | +0.004 |
wrists | +0.004 |
yes | +0.004 |
Neg logit contributions | |
loo | -0.004 |
ick | -0.004 |
auc | -0.004 |
rim | -0.004 |
arl | -0.004 |
Token | The |
Token ID | 383 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
loo | +0.049 |
of | +0.044 |
ick | +0.044 |
iren | +0.043 |
auc | +0.043 |
Neg logit contributions | |
Eva | -0.029 |
Robert | -0.029 |
chard | -0.029 |
wrists | -0.029 |
blonde | -0.028 |
Token | truck |
Token ID | 7779 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
Robert | +0.005 |
chard | +0.005 |
wrists | +0.005 |
yes | +0.005 |
blonde | +0.005 |
Neg logit contributions | |
loo | -0.006 |
ful | -0.006 |
auc | -0.006 |
angle | -0.006 |
ick | -0.006 |
Token | ! |
Token ID | 0 |
Feature activation | -2.13e-03 |
Pos logit contributions | |
Robert | +0.119 |
Eva | +0.119 |
chard | +0.119 |
wrists | +0.117 |
blonde | +0.114 |
Neg logit contributions | |
loo | -0.075 |
ick | -0.061 |
iren | -0.061 |
auc | -0.060 |
of | -0.060 |
Token | He |
Token ID | 679 |
Feature activation | -7.88e-01 |
Pos logit contributions | |
wrists | +0.000 |
chard | +0.000 |
Eva | +0.000 |
Robert | +0.000 |
blonde | +0.000 |
Neg logit contributions | |
loo | -0.000 |
of | -0.000 |
iren | -0.000 |
ick | -0.000 |
auc | -0.000 |
Token | said |
Token ID | 531 |
Feature activation | +8.37e-01 |
Pos logit contributions | |
wrists | +0.037 |
Eva | +0.037 |
chard | +0.036 |
Robert | +0.036 |
blonde | +0.035 |
Neg logit contributions | |
of | -0.026 |
loo | -0.026 |
ick | -0.022 |
ig | -0.021 |
auc | -0.021 |
Token | he |
Token ID | 339 |
Feature activation | +9.04e-01 |
Pos logit contributions | |
of | +0.018 |
loo | +0.018 |
ick | +0.013 |
auc | +0.013 |
iren | +0.012 |
Neg logit contributions | |
wrists | -0.050 |
chard | -0.049 |
Eva | -0.048 |
blonde | -0.048 |
Robert | -0.047 |
Token | was |
Token ID | 373 |
Feature activation | +5.69e-01 |
Pos logit contributions | |
of | +0.028 |
loo | +0.026 |
ick | +0.024 |
auc | +0.021 |
ig | +0.021 |
Neg logit contributions | |
wrists | -0.046 |
chard | -0.044 |
Eva | -0.044 |
Robert | -0.043 |
Stella | -0.041 |
Token | here |
Token ID | 994 |
Feature activation | -3.04e+00 |
Pos logit contributions | |
of | +0.018 |
loo | +0.017 |
ick | +0.016 |
ig | +0.014 |
late | +0.014 |
Neg logit contributions | |
wrists | -0.029 |
Eva | -0.026 |
blonde | -0.026 |
chard | -0.026 |
cious | -0.026 |
Token | to |
Token ID | 284 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
wrists | +0.159 |
Eva | +0.159 |
chard | +0.158 |
Robert | +0.157 |
blonde | +0.153 |
Neg logit contributions | |
loo | -0.086 |
of | -0.077 |
ick | -0.070 |
auc | -0.066 |
iren | -0.066 |
Token | help |
Token ID | 1037 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
Eva | +0.008 |
chard | +0.008 |
Robert | +0.008 |
wrists | +0.008 |
blonde | +0.007 |
Neg logit contributions | |
loo | -0.007 |
of | -0.006 |
ick | -0.006 |
iren | -0.005 |
auc | -0.005 |
Token | her |
Token ID | 607 |
Feature activation | -2.57e-01 |
Pos logit contributions | |
loo | +0.083 |
of | +0.069 |
ick | +0.068 |
iren | +0.067 |
auc | +0.066 |
Neg logit contributions | |
Eva | -0.126 |
Robert | -0.126 |
chard | -0.125 |
wrists | -0.125 |
blonde | -0.120 |
Token | make |
Token ID | 787 |
Feature activation | +6.90e-01 |
Pos logit contributions | |
Eva | +0.012 |
chard | +0.012 |
Robert | +0.012 |
wrists | +0.012 |
blonde | +0.012 |
Neg logit contributions | |
loo | -0.009 |
of | -0.007 |
ick | -0.007 |
iren | -0.007 |
auc | -0.007 |
Token | friends |
Token ID | 2460 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
loo | +0.027 |
of | +0.024 |
ick | +0.024 |
iren | +0.023 |
auc | +0.023 |
Neg logit contributions | |
chard | -0.029 |
Eva | -0.028 |
Robert | -0.028 |
wrists | -0.028 |
blonde | -0.027 |
Token | she |
Token ID | 673 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
of | +0.022 |
loo | +0.022 |
ick | +0.019 |
iren | +0.017 |
pour | +0.017 |
Neg logit contributions | |
wrists | -0.036 |
Eva | -0.036 |
Robert | -0.036 |
cious | -0.035 |
chard | -0.034 |
Token | tried |
Token ID | 3088 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
loo | +0.073 |
arl | +0.066 |
iren | +0.066 |
ick | +0.064 |
pour | +0.062 |
Neg logit contributions | |
Eva | -0.048 |
Robert | -0.048 |
chard | -0.047 |
wrists | -0.046 |
blonde | -0.045 |
Token | to |
Token ID | 284 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
loo | +0.035 |
of | +0.029 |
ick | +0.029 |
iren | +0.028 |
auc | +0.028 |
Neg logit contributions | |
Eva | -0.049 |
Robert | -0.048 |
chard | -0.048 |
wrists | -0.048 |
blonde | -0.046 |
Token | climb |
Token ID | 12080 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
wrists | +0.004 |
chard | +0.004 |
Eva | +0.004 |
Robert | +0.004 |
blonde | +0.004 |
Neg logit contributions | |
loo | -0.005 |
iren | -0.004 |
auc | -0.004 |
arl | -0.004 |
ick | -0.004 |
Token | up |
Token ID | 510 |
Feature activation | -5.68e+00 |
Pos logit contributions | |
loo | +0.035 |
of | +0.031 |
ick | +0.030 |
iren | +0.029 |
auc | +0.028 |
Neg logit contributions | |
wrists | -0.050 |
Eva | -0.050 |
Robert | -0.049 |
chard | -0.049 |
blonde | -0.048 |
Token | on |
Token ID | 319 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
Eva | +0.239 |
Robert | +0.238 |
chard | +0.234 |
wrists | +0.231 |
blonde | +0.229 |
Neg logit contributions | |
loo | -0.227 |
auc | -0.194 |
iren | -0.189 |
ick | -0.188 |
arl | -0.184 |
Token | a |
Token ID | 257 |
Feature activation | -9.13e-01 |
Pos logit contributions | |
chard | +0.090 |
Eva | +0.088 |
Robert | +0.088 |
wrists | +0.088 |
blonde | +0.084 |
Neg logit contributions | |
loo | -0.061 |
of | -0.055 |
ick | -0.053 |
iren | -0.051 |
pour | -0.049 |
Token | chair |
Token ID | 5118 |
Feature activation | -3.68e-01 |
Pos logit contributions | |
Eva | +0.040 |
chard | +0.039 |
Robert | +0.039 |
wrists | +0.039 |
blonde | +0.038 |
Neg logit contributions | |
loo | -0.034 |
of | -0.029 |
ick | -0.029 |
iren | -0.029 |
auc | -0.029 |
Token | to |
Token ID | 284 |
Feature activation | -9.63e-02 |
Pos logit contributions | |
Robert | +0.015 |
chard | +0.015 |
Eva | +0.015 |
blonde | +0.015 |
wrists | +0.014 |
Neg logit contributions | |
loo | -0.014 |
arl | -0.012 |
rim | -0.012 |
auc | -0.012 |
iren | -0.012 |
Token | put |
Token ID | 1234 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
Robert | +0.005 |
wrists | +0.005 |
Eva | +0.005 |
chard | +0.004 |
blonde | +0.004 |
Neg logit contributions | |
loo | -0.003 |
of | -0.003 |
ick | -0.003 |
iren | -0.003 |
auc | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | -9.37e-01 |
Pos logit contributions | |
loo | +0.005 |
of | +0.004 |
ick | +0.004 |
iren | +0.004 |
auc | +0.004 |
Neg logit contributions | |
Eva | -0.007 |
chard | -0.007 |
Robert | -0.007 |
wrists | -0.007 |
blonde | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +7.87e-01 |
Pos logit contributions | |
loo | +0.058 |
iren | +0.047 |
auc | +0.047 |
ick | +0.047 |
arl | +0.045 |
Neg logit contributions | |
wrists | -0.068 |
chard | -0.067 |
Robert | -0.066 |
Eva | -0.066 |
blonde | -0.065 |
Token | The |
Token ID | 383 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
loo | +0.027 |
of | +0.027 |
iren | +0.025 |
ick | +0.024 |
late | +0.024 |
Neg logit contributions | |
chard | -0.035 |
wrists | -0.035 |
cious | -0.033 |
Eva | -0.033 |
blonde | -0.032 |
Token | frog |
Token ID | 21264 |
Feature activation | -3.44e+00 |
Pos logit contributions | |
loo | +0.006 |
of | +0.005 |
ick | +0.005 |
iren | +0.005 |
auc | +0.005 |
Neg logit contributions | |
Robert | -0.007 |
Eva | -0.007 |
wrists | -0.007 |
chard | -0.007 |
blonde | -0.007 |
Token | was |
Token ID | 373 |
Feature activation | +7.52e-01 |
Pos logit contributions | |
Eva | +0.142 |
Robert | +0.142 |
chard | +0.142 |
wrists | +0.141 |
blonde | +0.136 |
Neg logit contributions | |
loo | -0.136 |
of | -0.121 |
ick | -0.117 |
iren | -0.115 |
auc | -0.114 |
Token | so |
Token ID | 523 |
Feature activation | -2.80e+00 |
Pos logit contributions | |
loo | +0.033 |
auc | +0.029 |
ick | +0.028 |
iren | +0.028 |
of | +0.027 |
Neg logit contributions | |
chard | -0.029 |
Eva | -0.029 |
Robert | -0.028 |
wrists | -0.028 |
blonde | -0.027 |
Token | shocked |
Token ID | 11472 |
Feature activation | -6.97e-02 |
Pos logit contributions | |
chard | +0.094 |
Eva | +0.089 |
Robert | +0.088 |
blonde | +0.085 |
wrists | +0.085 |
Neg logit contributions | |
loo | -0.138 |
auc | -0.122 |
ick | -0.120 |
ig | -0.113 |
iren | -0.112 |
Token | he |
Token ID | 339 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
Robert | +0.003 |
Eva | +0.003 |
chard | +0.003 |
wrists | +0.003 |
Stella | +0.003 |
Neg logit contributions | |
loo | -0.003 |
auc | -0.002 |
ick | -0.002 |
iren | -0.002 |
arl | -0.002 |
Token | jumped |
Token ID | 11687 |
Feature activation | -3.60e-01 |
Pos logit contributions | |
loo | +0.040 |
of | +0.037 |
ick | +0.035 |
iren | +0.034 |
auc | +0.033 |
Neg logit contributions | |
Eva | -0.045 |
Robert | -0.044 |
chard | -0.044 |
wrists | -0.043 |
blonde | -0.043 |
Token | straight |
Token ID | 3892 |
Feature activation | -7.04e-01 |
Pos logit contributions | |
Robert | +0.014 |
chard | +0.014 |
Eva | +0.014 |
wrists | +0.013 |
Claire | +0.013 |
Neg logit contributions | |
loo | -0.015 |
ick | -0.012 |
iren | -0.012 |
auc | -0.012 |
ig | -0.012 |
Token | out |
Token ID | 503 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
Robert | +0.032 |
Eva | +0.031 |
wrists | +0.031 |
chard | +0.031 |
blonde | +0.030 |
Neg logit contributions | |
loo | -0.026 |
ick | -0.022 |
auc | -0.021 |
iren | -0.021 |
of | -0.021 |
Token | of |
Token ID | 286 |
Feature activation | +3.37e+00 |
Pos logit contributions | |
Eva | +0.105 |
Joy | +0.103 |
yes | +0.103 |
aloud | +0.102 |
Robert | +0.099 |
Neg logit contributions | |
loo | -0.064 |
ained | -0.053 |
auc | -0.047 |
ick | -0.046 |
w | -0.044 |
Token | boy |
Token ID | 2933 |
Feature activation | -9.14e-01 |
Pos logit contributions | |
loo | +0.022 |
of | +0.019 |
iren | +0.019 |
ick | +0.019 |
auc | +0.019 |
Neg logit contributions | |
Robert | -0.026 |
chard | -0.026 |
Eva | -0.026 |
wrists | -0.025 |
blonde | -0.024 |
Token | was |
Token ID | 373 |
Feature activation | +7.82e-01 |
Pos logit contributions | |
Eva | +0.042 |
wrists | +0.041 |
chard | +0.041 |
Robert | +0.040 |
blonde | +0.040 |
Neg logit contributions | |
loo | -0.032 |
of | -0.031 |
ick | -0.027 |
ig | -0.027 |
auc | -0.026 |
Token | gone |
Token ID | 3750 |
Feature activation | +4.51e+00 |
Pos logit contributions | |
loo | +0.028 |
of | +0.026 |
ick | +0.024 |
iren | +0.024 |
auc | +0.023 |
Neg logit contributions | |
wrists | -0.036 |
Eva | -0.035 |
Robert | -0.035 |
chard | -0.034 |
blonde | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | +8.72e-01 |
Pos logit contributions | |
loo | +0.165 |
auc | +0.144 |
iren | +0.143 |
ick | +0.134 |
rim | +0.131 |
Neg logit contributions | |
Robert | -0.203 |
chard | -0.202 |
Eva | -0.196 |
wrists | -0.193 |
blonde | -0.189 |
Token | " |
Token ID | 366 |
Feature activation | -3.14e+00 |
Pos logit contributions | |
loo | +0.031 |
of | +0.027 |
iren | +0.026 |
auc | +0.025 |
ick | +0.025 |
Neg logit contributions | |
wrists | -0.040 |
chard | -0.039 |
Eva | -0.038 |
Robert | -0.038 |
yes | -0.037 |
Token | Where |
Token ID | 8496 |
Feature activation | -5.05e+00 |
Pos logit contributions | |
Robert | +0.115 |
chard | +0.115 |
Eva | +0.114 |
wrists | +0.111 |
blonde | +0.108 |
Neg logit contributions | |
loo | -0.142 |
ick | -0.125 |
of | -0.124 |
auc | -0.122 |
iren | -0.122 |
Token | are |
Token ID | 389 |
Feature activation | +8.58e-01 |
Pos logit contributions | |
Robert | +0.199 |
Eva | +0.198 |
chard | +0.191 |
blonde | +0.182 |
yes | +0.181 |
Neg logit contributions | |
loo | -0.208 |
iren | -0.186 |
ick | -0.182 |
arl | -0.178 |
auc | -0.178 |
Token | you |
Token ID | 345 |
Feature activation | +3.38e-01 |
Pos logit contributions | |
loo | +0.024 |
of | +0.019 |
ick | +0.019 |
iren | +0.019 |
auc | +0.019 |
Neg logit contributions | |
Robert | -0.046 |
Eva | -0.046 |
chard | -0.046 |
wrists | -0.045 |
blonde | -0.044 |
Token | ?" |
Token ID | 1701 |
Feature activation | -3.21e+00 |
Pos logit contributions | |
loo | +0.013 |
iren | +0.012 |
auc | +0.012 |
ick | +0.011 |
of | +0.011 |
Neg logit contributions | |
Robert | -0.014 |
chard | -0.014 |
Eva | -0.014 |
Claire | -0.013 |
wrists | -0.013 |
Token | she |
Token ID | 673 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
wrists | +0.151 |
chard | +0.147 |
Eva | +0.145 |
Robert | +0.144 |
blonde | +0.141 |
Neg logit contributions | |
loo | -0.102 |
of | -0.099 |
iren | -0.090 |
ick | -0.085 |
auc | -0.082 |
Token | shouted |
Token ID | 17293 |
Feature activation | +4.44e-01 |
Pos logit contributions | |
of | +0.042 |
loo | +0.040 |
ick | +0.037 |
iren | +0.037 |
ig | +0.036 |
Neg logit contributions | |
chard | -0.061 |
wrists | -0.061 |
Eva | -0.060 |
Robert | -0.060 |
blonde | -0.058 |
Token | is |
Token ID | 318 |
Feature activation | +8.71e-01 |
Pos logit contributions | |
loo | +0.006 |
ick | +0.005 |
iren | +0.005 |
of | +0.005 |
auc | +0.005 |
Neg logit contributions | |
Robert | -0.011 |
Eva | -0.011 |
wrists | -0.011 |
chard | -0.011 |
blonde | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
loo | +0.029 |
of | +0.027 |
ick | +0.024 |
ig | +0.024 |
tunnel | +0.023 |
Neg logit contributions | |
wrists | -0.042 |
chard | -0.041 |
blonde | -0.040 |
Eva | -0.039 |
cious | -0.038 |
Token | best |
Token ID | 1266 |
Feature activation | -3.63e+00 |
Pos logit contributions | |
loo | +0.008 |
of | +0.007 |
ick | +0.007 |
iren | +0.007 |
auc | +0.007 |
Neg logit contributions | |
Eva | -0.010 |
chard | -0.010 |
Robert | -0.010 |
wrists | -0.010 |
blonde | -0.010 |
Token | summer |
Token ID | 3931 |
Feature activation | +2.85e+00 |
Pos logit contributions | |
Robert | +0.162 |
Eva | +0.153 |
wrists | +0.150 |
Claire | +0.147 |
Stella | +0.143 |
Neg logit contributions | |
loo | -0.122 |
iren | -0.120 |
ick | -0.116 |
auc | -0.115 |
arl | -0.114 |
Token | ever |
Token ID | 1683 |
Feature activation | +2.00e+00 |
Pos logit contributions | |
loo | +0.086 |
arl | +0.080 |
iren | +0.077 |
ick | +0.076 |
auc | +0.076 |
Neg logit contributions | |
Robert | -0.141 |
Eva | -0.139 |
wrists | -0.131 |
Claire | -0.131 |
chard | -0.129 |
Token | !" |
Token ID | 2474 |
Feature activation | -5.08e+00 |
Pos logit contributions | |
loo | +0.072 |
iren | +0.062 |
ick | +0.062 |
auc | +0.061 |
of | +0.059 |
Neg logit contributions | |
Eva | -0.090 |
Robert | -0.090 |
chard | -0.087 |
wrists | -0.087 |
Claire | -0.083 |
Token | Anna |
Token ID | 11735 |
Feature activation | -3.10e+00 |
Pos logit contributions | |
chard | +0.215 |
wrists | +0.212 |
Robert | +0.212 |
Eva | +0.212 |
blonde | +0.205 |
Neg logit contributions | |
loo | -0.196 |
of | -0.176 |
ick | -0.168 |
iren | -0.167 |
auc | -0.166 |
Token | says |
Token ID | 1139 |
Feature activation | +2.69e-01 |
Pos logit contributions | |
chard | +0.164 |
blonde | +0.154 |
Robert | +0.154 |
wrists | +0.154 |
Eva | +0.152 |
Neg logit contributions | |
loo | -0.088 |
of | -0.086 |
ick | -0.074 |
iren | -0.074 |
auc | -0.074 |
Token | . |
Token ID | 13 |
Feature activation | +8.19e-01 |
Pos logit contributions | |
loo | +0.006 |
of | +0.006 |
ick | +0.005 |
iren | +0.005 |
auc | +0.005 |
Neg logit contributions | |
chard | -0.015 |
Robert | -0.015 |
Eva | -0.015 |
wrists | -0.015 |
blonde | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
of | +0.027 |
loo | +0.025 |
iren | +0.024 |
arl | +0.023 |
ick | +0.022 |
Neg logit contributions | |
chard | -0.039 |
wrists | -0.039 |
Robert | -0.038 |
Eva | -0.037 |
cious | -0.037 |
Token |
|
Token ID | 198 |
Feature activation | -1.77e-03 |
Pos logit contributions | |
Eva | +0.001 |
Robert | +0.001 |
chard | +0.001 |
wrists | +0.001 |
blonde | +0.001 |
Neg logit contributions | |
loo | -0.001 |
of | -0.001 |
ick | -0.001 |
auc | -0.001 |
iren | -0.001 |
Token | ze |
Token ID | 2736 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
chard | +0.090 |
Robert | +0.089 |
Eva | +0.087 |
wrists | +0.085 |
Claire | +0.081 |
Neg logit contributions | |
loo | -0.100 |
of | -0.091 |
ick | -0.089 |
iren | -0.087 |
auc | -0.087 |
Token | makes |
Token ID | 1838 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
Eva | +0.122 |
wrists | +0.121 |
Robert | +0.120 |
blonde | +0.116 |
chard | +0.113 |
Neg logit contributions | |
loo | -0.111 |
ick | -0.093 |
arl | -0.090 |
auc | -0.090 |
ig | -0.088 |
Token | the |
Token ID | 262 |
Feature activation | +4.84e-01 |
Pos logit contributions | |
loo | +0.064 |
auc | +0.054 |
iren | +0.054 |
ick | +0.054 |
arl | +0.053 |
Neg logit contributions | |
Eva | -0.053 |
Robert | -0.052 |
wrists | -0.051 |
blonde | -0.050 |
chard | -0.048 |
Token | blocks |
Token ID | 7021 |
Feature activation | -4.77e-01 |
Pos logit contributions | |
loo | +0.018 |
ick | +0.016 |
of | +0.016 |
iren | +0.016 |
auc | +0.016 |
Neg logit contributions | |
wrists | -0.020 |
Eva | -0.020 |
chard | -0.020 |
Robert | -0.020 |
blonde | -0.020 |
Token | fall |
Token ID | 2121 |
Feature activation | +4.91e+00 |
Pos logit contributions | |
yes | +0.018 |
wrists | +0.016 |
blonde | +0.016 |
Eva | +0.015 |
aloud | +0.015 |
Neg logit contributions | |
loo | -0.021 |
enn | -0.018 |
iren | -0.018 |
rim | -0.017 |
arl | -0.017 |
Token | down |
Token ID | 866 |
Feature activation | -4.92e+00 |
Pos logit contributions | |
loo | +0.232 |
rim | +0.203 |
auc | +0.189 |
iren | +0.180 |
enn | +0.177 |
Neg logit contributions | |
wrists | -0.157 |
aloud | -0.152 |
blonde | -0.152 |
Robert | -0.150 |
yes | -0.145 |
Token | . |
Token ID | 13 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
Robert | +0.183 |
Eva | +0.176 |
chard | +0.174 |
aloud | +0.168 |
Claire | +0.163 |
Neg logit contributions | |
loo | -0.202 |
rim | -0.186 |
auc | -0.176 |
arl | -0.175 |
row | -0.168 |
Token | The |
Token ID | 383 |
Feature activation | +5.30e-02 |
Pos logit contributions | |
loo | +0.039 |
of | +0.038 |
ick | +0.034 |
auc | +0.033 |
iren | +0.033 |
Neg logit contributions | |
chard | -0.056 |
wrists | -0.055 |
cious | -0.052 |
Robert | -0.052 |
blonde | -0.052 |
Token | building |
Token ID | 2615 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
loo | +0.002 |
ick | +0.002 |
arl | +0.002 |
auc | +0.002 |
iren | +0.002 |
Neg logit contributions | |
wrists | -0.002 |
Eva | -0.002 |
chard | -0.002 |
blonde | -0.002 |
Robert | -0.002 |
Token | is |
Token ID | 318 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
wrists | +0.002 |
Eva | +0.002 |
blonde | +0.001 |
Robert | +0.001 |
yes | +0.001 |
Neg logit contributions | |
loo | -0.001 |
arl | -0.001 |
rim | -0.001 |
ick | -0.001 |
pour | -0.001 |
Token | gone |
Token ID | 3750 |
Feature activation | +4.36e+00 |
Pos logit contributions | |
loo | +0.050 |
auc | +0.042 |
ick | +0.041 |
iren | +0.040 |
arp | +0.039 |
Neg logit contributions | |
chard | -0.044 |
blonde | -0.043 |
wrists | -0.042 |
Eva | -0.042 |
Robert | -0.041 |
Token | , |
Token ID | 11 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
wrists | +0.313 |
chard | +0.311 |
Eva | +0.309 |
Robert | +0.307 |
blonde | +0.294 |
Neg logit contributions | |
loo | -0.121 |
of | -0.107 |
ick | -0.099 |
ig | -0.085 |
pour | -0.084 |
Token | little |
Token ID | 1310 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
loo | +0.011 |
of | +0.010 |
ick | +0.009 |
iren | +0.009 |
auc | +0.009 |
Neg logit contributions | |
Eva | -0.010 |
chard | -0.010 |
Robert | -0.010 |
wrists | -0.010 |
blonde | -0.009 |
Token | spirit |
Token ID | 4437 |
Feature activation | +4.95e-01 |
Pos logit contributions | |
loo | +0.055 |
of | +0.048 |
ick | +0.047 |
auc | +0.045 |
iren | +0.045 |
Neg logit contributions | |
Eva | -0.067 |
chard | -0.067 |
Robert | -0.067 |
wrists | -0.067 |
blonde | -0.063 |
Token | , |
Token ID | 11 |
Feature activation | +2.75e-01 |
Pos logit contributions | |
loo | +0.015 |
ick | +0.012 |
iren | +0.012 |
auc | +0.012 |
of | +0.012 |
Neg logit contributions | |
Robert | -0.025 |
Eva | -0.025 |
chard | -0.025 |
wrists | -0.024 |
blonde | -0.024 |
Token | â |
Token ID | 22940 |
Feature activation | +3.74e+00 |
Pos logit contributions | |
loo | +0.010 |
of | +0.009 |
ick | +0.009 |
auc | +0.008 |
iren | +0.008 |
Neg logit contributions | |
wrists | -0.012 |
chard | -0.012 |
Eva | -0.012 |
Robert | -0.012 |
blonde | -0.012 |
Token | € |
Token ID | 26391 |
Feature activation | -6.38e+00 |
Pos logit contributions | |
loo | +0.201 |
of | +0.181 |
ick | +0.180 |
iren | +0.178 |
auc | +0.177 |
Neg logit contributions | |
Eva | -0.102 |
chard | -0.102 |
Robert | -0.102 |
wrists | -0.102 |
blonde | -0.095 |
Token | said |
Token ID | 531 |
Feature activation | +8.94e-01 |
Pos logit contributions | |
chard | +0.260 |
wrists | +0.258 |
Eva | +0.257 |
Robert | +0.255 |
Claire | +0.242 |
Neg logit contributions | |
loo | -0.254 |
of | -0.223 |
ick | -0.223 |
iren | -0.214 |
auc | -0.209 |
Token | F |
Token ID | 376 |
Feature activation | +6.13e-02 |
Pos logit contributions | |
loo | +0.021 |
of | +0.020 |
ick | +0.017 |
auc | +0.015 |
iren | +0.015 |
Neg logit contributions | |
wrists | -0.052 |
chard | -0.051 |
Eva | -0.050 |
Robert | -0.049 |
blonde | -0.049 |
Token | awn |
Token ID | 3832 |
Feature activation | -2.93e+00 |
Pos logit contributions | |
loo | +0.002 |
ick | +0.002 |
auc | +0.002 |
of | +0.002 |
arl | +0.002 |
Neg logit contributions | |
wrists | -0.003 |
Eva | -0.003 |
Robert | -0.003 |
chard | -0.003 |
blonde | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +4.99e-01 |
Pos logit contributions | |
chard | +0.166 |
Eva | +0.166 |
wrists | +0.165 |
Robert | +0.165 |
blonde | +0.161 |
Neg logit contributions | |
loo | -0.073 |
of | -0.061 |
ick | -0.058 |
auc | -0.055 |
iren | -0.053 |
Token | � |
Token ID | 6184 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
loo | +0.020 |
of | +0.019 |
ick | +0.018 |
iren | +0.018 |
auc | +0.018 |
Neg logit contributions | |
chard | -0.020 |
Eva | -0.020 |
wrists | -0.020 |
Robert | -0.019 |
blonde | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | +3.83e-01 |
Pos logit contributions | |
of | +0.001 |
loo | +0.001 |
ick | +0.001 |
auc | +0.001 |
iren | +0.001 |
Neg logit contributions | |
wrists | -0.003 |
chard | -0.003 |
blonde | -0.003 |
Eva | -0.003 |
cious | -0.003 |
Token | old |
Token ID | 1468 |
Feature activation | +8.96e-01 |
Pos logit contributions | |
loo | +0.014 |
of | +0.012 |
ick | +0.012 |
auc | +0.012 |
iren | +0.012 |
Neg logit contributions | |
wrists | -0.017 |
Eva | -0.017 |
Robert | -0.017 |
chard | -0.017 |
blonde | -0.016 |
Token | man |
Token ID | 582 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
loo | +0.033 |
of | +0.029 |
iren | +0.028 |
ick | +0.027 |
auc | +0.027 |
Neg logit contributions | |
Eva | -0.040 |
chard | -0.040 |
wrists | -0.039 |
Robert | -0.039 |
blonde | -0.038 |
Token | , |
Token ID | 11 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
wrists | +0.074 |
blonde | +0.073 |
chard | +0.069 |
Eva | +0.069 |
Robert | +0.069 |
Neg logit contributions | |
of | -0.036 |
loo | -0.032 |
ick | -0.026 |
auc | -0.026 |
ful | -0.026 |
Token | " |
Token ID | 366 |
Feature activation | -3.56e+00 |
Pos logit contributions | |
of | +0.013 |
loo | +0.012 |
ick | +0.011 |
ful | +0.011 |
iren | +0.011 |
Neg logit contributions | |
wrists | -0.012 |
blonde | -0.011 |
Eva | -0.011 |
cious | -0.010 |
Robert | -0.010 |
Token | What |
Token ID | 2061 |
Feature activation | -7.31e+00 |
Pos logit contributions | |
wrists | +0.146 |
Eva | +0.142 |
blonde | +0.140 |
Robert | +0.139 |
chard | +0.136 |
Neg logit contributions | |
loo | -0.138 |
of | -0.127 |
iren | -0.123 |
auc | -0.120 |
pour | -0.120 |
Token | is |
Token ID | 318 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
wrists | +0.385 |
chard | +0.383 |
Eva | +0.371 |
Robert | +0.368 |
blonde | +0.366 |
Neg logit contributions | |
loo | -0.211 |
of | -0.198 |
ick | -0.171 |
auc | -0.164 |
iren | -0.152 |
Token | it |
Token ID | 340 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
loo | +0.034 |
of | +0.030 |
ick | +0.029 |
auc | +0.028 |
iren | +0.028 |
Neg logit contributions | |
wrists | -0.049 |
Eva | -0.049 |
chard | -0.048 |
Robert | -0.047 |
blonde | -0.047 |
Token | ?" |
Token ID | 1701 |
Feature activation | -3.86e+00 |
Pos logit contributions | |
loo | +0.006 |
of | +0.005 |
ick | +0.005 |
iren | +0.005 |
auc | +0.005 |
Neg logit contributions | |
Eva | -0.012 |
chard | -0.012 |
Robert | -0.012 |
wrists | -0.012 |
blonde | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -4.34e-02 |
Pos logit contributions | |
wrists | +0.194 |
chard | +0.185 |
blonde | +0.184 |
cious | +0.182 |
Eva | +0.176 |
Neg logit contributions | |
of | -0.121 |
loo | -0.112 |
iren | -0.109 |
pour | -0.098 |
late | -0.096 |
Token |
|
Token ID | 198 |
Feature activation | -5.39e-02 |
Pos logit contributions | |
wrists | +0.002 |
chard | +0.002 |
blonde | +0.002 |
Eva | +0.002 |
Robert | +0.002 |
Neg logit contributions | |
loo | -0.001 |
of | -0.001 |
iren | -0.001 |
ick | -0.001 |
auc | -0.001 |