Token | always |
Token ID | 1464 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
auc | +0.007 |
ays | +0.006 |
line | +0.006 |
bars | +0.006 |
ains | +0.006 |
Neg logit contributions | |
Amen | -0.006 |
lder | -0.005 |
defeat | -0.005 |
oc | -0.005 |
Frank | -0.005 |
Token | happy |
Token ID | 3772 |
Feature activation | -2.47e+00 |
Pos logit contributions | |
Amen | +0.029 |
lder | +0.023 |
defeat | +0.022 |
chat | +0.021 |
Frank | +0.021 |
Neg logit contributions | |
auc | -0.044 |
ays | -0.039 |
bars | -0.039 |
inter | -0.039 |
ains | -0.038 |
Token | , |
Token ID | 11 |
Feature activation | +6.44e-01 |
Pos logit contributions | |
Amen | +0.071 |
lder | +0.059 |
defeat | +0.057 |
chat | +0.055 |
Frank | +0.054 |
Neg logit contributions | |
auc | -0.079 |
bars | -0.069 |
ays | -0.068 |
ains | -0.068 |
inter | -0.068 |
Token | because |
Token ID | 780 |
Feature activation | +5.22e+00 |
Pos logit contributions | |
auc | +0.021 |
ays | +0.018 |
ains | +0.018 |
bars | +0.018 |
inter | +0.017 |
Neg logit contributions | |
Amen | -0.019 |
defeat | -0.015 |
lder | -0.015 |
chat | -0.015 |
Frank | -0.015 |
Token | when |
Token ID | 618 |
Feature activation | +4.62e+00 |
Pos logit contributions | |
auc | +0.134 |
inter | +0.124 |
ays | +0.120 |
ains | +0.113 |
ash | +0.111 |
Neg logit contributions | |
Amen | -0.190 |
defeat | -0.151 |
read | -0.149 |
Frank | -0.148 |
lder | -0.147 |
Token | everything |
Token ID | 2279 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
auc | +0.121 |
inter | +0.106 |
ays | +0.102 |
bars | +0.101 |
ash | +0.098 |
Neg logit contributions | |
Amen | -0.168 |
lder | -0.140 |
defeat | -0.140 |
Frank | -0.138 |
chat | -0.137 |
Token | was |
Token ID | 373 |
Feature activation | +2.77e+00 |
Pos logit contributions | |
auc | +0.286 |
ays | +0.272 |
inter | +0.269 |
line | +0.266 |
ains | +0.264 |
Neg logit contributions | |
Amen | -0.326 |
chat | -0.295 |
lder | -0.295 |
defeat | -0.279 |
read | -0.270 |
Token | sorted |
Token ID | 23243 |
Feature activation | +3.36e+00 |
Pos logit contributions | |
auc | +0.084 |
ays | +0.076 |
bars | +0.075 |
line | +0.072 |
ains | +0.072 |
Neg logit contributions | |
Amen | -0.087 |
lder | -0.069 |
defeat | -0.068 |
Frank | -0.067 |
chat | -0.066 |
Token | it |
Token ID | 340 |
Feature activation | +2.44e+00 |
Pos logit contributions | |
auc | +0.120 |
inter | +0.110 |
ays | +0.109 |
ours | +0.108 |
line | +0.108 |
Neg logit contributions | |
Amen | -0.064 |
defeat | -0.057 |
", | -0.055 |
lder | -0.053 |
chat | -0.052 |
Token | meant |
Token ID | 4001 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
auc | +0.086 |
ays | +0.078 |
inter | +0.077 |
idget | +0.077 |
bars | +0.075 |
Neg logit contributions | |
defeat | -0.047 |
", | -0.044 |
Amen | -0.043 |
read | -0.042 |
chat | -0.042 |
Token | everybody |
Token ID | 7288 |
Feature activation | +4.69e+00 |
Pos logit contributions | |
Amen | +0.033 |
defeat | +0.028 |
lder | +0.027 |
chat | +0.026 |
Frank | +0.025 |
Neg logit contributions | |
auc | -0.047 |
ays | -0.044 |
inter | -0.041 |
bars | -0.041 |
ains | -0.041 |
Token |
|
Token ID | 198 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
Amen | +0.062 |
lder | +0.053 |
defeat | +0.051 |
oc | +0.049 |
arrest | +0.048 |
Neg logit contributions | |
auc | -0.042 |
line | -0.035 |
ays | -0.035 |
ains | -0.034 |
ash | -0.033 |
Token | M |
Token ID | 44 |
Feature activation | +4.48e+00 |
Pos logit contributions | |
Amen | +0.048 |
lder | +0.040 |
defeat | +0.038 |
oc | +0.037 |
arrest | +0.035 |
Neg logit contributions | |
auc | -0.040 |
line | -0.034 |
ains | -0.033 |
bars | -0.032 |
ays | -0.032 |
Token | ama |
Token ID | 1689 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
auc | +0.123 |
ains | +0.103 |
ays | +0.103 |
bars | +0.101 |
line | +0.100 |
Neg logit contributions | |
Amen | -0.157 |
lder | -0.127 |
defeat | -0.125 |
Mic | -0.122 |
Frank | -0.121 |
Token | said |
Token ID | 531 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
auc | +0.044 |
line | +0.038 |
bars | +0.038 |
ays | +0.037 |
ains | +0.037 |
Neg logit contributions | |
Amen | -0.055 |
lder | -0.045 |
defeat | -0.042 |
Frank | -0.042 |
oc | -0.042 |
Token | that |
Token ID | 326 |
Feature activation | +3.81e+00 |
Pos logit contributions | |
Amen | +0.070 |
lder | +0.059 |
defeat | +0.055 |
Frank | +0.054 |
oc | +0.054 |
Neg logit contributions | |
auc | -0.051 |
ays | -0.044 |
ains | -0.044 |
bars | -0.043 |
line | -0.043 |
Token | those |
Token ID | 883 |
Feature activation | +1.08e+01 |
Pos logit contributions | |
auc | +0.108 |
ays | +0.103 |
inter | +0.097 |
idget | +0.095 |
ains | +0.094 |
Neg logit contributions | |
Amen | -0.118 |
read | -0.099 |
lder | -0.097 |
chat | -0.097 |
defeat | -0.093 |
Token | who |
Token ID | 508 |
Feature activation | +6.56e+00 |
Pos logit contributions | |
auc | +0.333 |
ays | +0.320 |
line | +0.306 |
ained | +0.282 |
inter | +0.281 |
Neg logit contributions | |
Amen | -0.289 |
lder | -0.271 |
read | -0.257 |
defeat | -0.240 |
novel | -0.227 |
Token | are |
Token ID | 389 |
Feature activation | +2.59e+00 |
Pos logit contributions | |
auc | +0.202 |
ays | +0.180 |
lights | +0.177 |
line | +0.176 |
unes | +0.172 |
Neg logit contributions | |
Amen | -0.208 |
lder | -0.176 |
defeat | -0.168 |
read | -0.167 |
oc | -0.164 |
Token | ignorant |
Token ID | 20109 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
auc | +0.093 |
ays | +0.085 |
bars | +0.083 |
inter | +0.083 |
line | +0.081 |
Neg logit contributions | |
Amen | -0.068 |
lder | -0.053 |
defeat | -0.051 |
read | -0.048 |
Frank | -0.046 |
Token | don |
Token ID | 836 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
auc | +0.048 |
line | +0.043 |
ays | +0.043 |
ains | +0.042 |
inter | +0.041 |
Neg logit contributions | |
Amen | -0.037 |
lder | -0.031 |
defeat | -0.029 |
", | -0.029 |
read | -0.029 |
Token | 't |
Token ID | 470 |
Feature activation | -3.12e-01 |
Pos logit contributions | |
auc | +0.073 |
line | +0.065 |
ays | +0.065 |
ains | +0.064 |
inter | +0.064 |
Neg logit contributions | |
Amen | -0.042 |
lder | -0.031 |
defeat | -0.028 |
Frank | -0.027 |
oc | -0.025 |
Token | , |
Token ID | 11 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
Amen | +0.055 |
lder | +0.045 |
defeat | +0.042 |
Frank | +0.041 |
oc | +0.041 |
Neg logit contributions | |
auc | -0.055 |
ays | -0.049 |
line | -0.049 |
bars | -0.048 |
ains | -0.048 |
Token | " |
Token ID | 366 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
auc | +0.032 |
ays | +0.028 |
bars | +0.027 |
ains | +0.026 |
line | +0.025 |
Neg logit contributions | |
Amen | -0.059 |
lder | -0.050 |
defeat | -0.047 |
Frank | -0.047 |
oc | -0.047 |
Token | M |
Token ID | 44 |
Feature activation | +5.05e+00 |
Pos logit contributions | |
auc | +0.041 |
ains | +0.038 |
lights | +0.038 |
ays | +0.037 |
tubes | +0.037 |
Neg logit contributions | |
Amen | -0.043 |
lder | -0.035 |
Frank | -0.035 |
Mic | -0.035 |
oc | -0.034 |
Token | mm |
Token ID | 3020 |
Feature activation | +5.65e+00 |
Pos logit contributions | |
auc | +0.143 |
line | +0.129 |
ays | +0.128 |
ains | +0.127 |
bars | +0.127 |
Neg logit contributions | |
Amen | -0.163 |
lder | -0.137 |
oc | -0.130 |
defeat | -0.125 |
Frank | -0.123 |
Token | , |
Token ID | 11 |
Feature activation | +4.38e+00 |
Pos logit contributions | |
auc | +0.185 |
ays | +0.163 |
line | +0.163 |
bars | +0.163 |
ains | +0.162 |
Neg logit contributions | |
Amen | -0.162 |
lder | -0.132 |
defeat | -0.121 |
oc | -0.120 |
Frank | -0.119 |
Token | those |
Token ID | 883 |
Feature activation | +1.06e+01 |
Pos logit contributions | |
auc | +0.138 |
ays | +0.133 |
ained | +0.129 |
inter | +0.126 |
ains | +0.126 |
Neg logit contributions | |
Amen | -0.125 |
oc | -0.100 |
lder | -0.098 |
defeat | -0.098 |
read | -0.094 |
Token | smell |
Token ID | 8508 |
Feature activation | +4.07e+00 |
Pos logit contributions | |
auc | +0.340 |
line | +0.307 |
ays | +0.299 |
lights | +0.295 |
ains | +0.287 |
Neg logit contributions | |
Amen | -0.303 |
lder | -0.268 |
oc | -0.248 |
", | -0.246 |
Frank | -0.235 |
Token | delicious |
Token ID | 12625 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
auc | +0.115 |
bars | +0.111 |
line | +0.110 |
ays | +0.108 |
idget | +0.107 |
Neg logit contributions | |
Amen | -0.117 |
lder | -0.106 |
defeat | -0.102 |
chat | -0.101 |
oc | -0.097 |
Token | !" |
Token ID | 2474 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
Amen | +0.032 |
lder | +0.026 |
oc | +0.024 |
defeat | +0.023 |
Frank | +0.022 |
Neg logit contributions | |
auc | -0.046 |
bars | -0.041 |
ains | -0.041 |
line | -0.041 |
ays | -0.040 |
Token | Lily |
Token ID | 20037 |
Feature activation | +6.71e-02 |
Pos logit contributions | |
Amen | +0.084 |
lder | +0.073 |
oc | +0.070 |
defeat | +0.069 |
chat | +0.068 |
Neg logit contributions | |
auc | -0.042 |
ains | -0.039 |
bars | -0.037 |
ays | -0.036 |
line | -0.034 |
Token | 's |
Token ID | 338 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
auc | +0.001 |
line | +0.001 |
bars | +0.001 |
ays | +0.001 |
tubes | +0.001 |
Neg logit contributions | |
Amen | -0.003 |
Frank | -0.003 |
lder | -0.002 |
Charles | -0.002 |
defeat | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
Amen | +0.060 |
lder | +0.048 |
defeat | +0.045 |
Frank | +0.045 |
oc | +0.045 |
Neg logit contributions | |
auc | -0.058 |
line | -0.051 |
ays | -0.050 |
bars | -0.050 |
ains | -0.049 |
Token | " |
Token ID | 366 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
auc | +0.032 |
ays | +0.027 |
bars | +0.026 |
ains | +0.026 |
ash | +0.026 |
Neg logit contributions | |
Amen | -0.053 |
lder | -0.045 |
Frank | -0.042 |
defeat | -0.042 |
oc | -0.042 |
Token | M |
Token ID | 44 |
Feature activation | +4.92e+00 |
Pos logit contributions | |
auc | +0.039 |
ays | +0.035 |
ains | +0.035 |
tubes | +0.035 |
bars | +0.034 |
Neg logit contributions | |
Amen | -0.038 |
lder | -0.030 |
Frank | -0.029 |
oc | -0.029 |
Mic | -0.028 |
Token | mm |
Token ID | 3020 |
Feature activation | +5.68e+00 |
Pos logit contributions | |
auc | +0.133 |
ays | +0.112 |
bars | +0.111 |
line | +0.111 |
ains | +0.110 |
Neg logit contributions | |
Amen | -0.174 |
lder | -0.144 |
defeat | -0.138 |
Frank | -0.136 |
oc | -0.133 |
Token | , |
Token ID | 11 |
Feature activation | +4.19e+00 |
Pos logit contributions | |
auc | +0.177 |
ays | +0.156 |
line | +0.155 |
ains | +0.155 |
bars | +0.154 |
Neg logit contributions | |
Amen | -0.174 |
lder | -0.138 |
defeat | -0.129 |
Frank | -0.129 |
oc | -0.127 |
Token | those |
Token ID | 883 |
Feature activation | +1.05e+01 |
Pos logit contributions | |
auc | +0.127 |
ays | +0.115 |
ains | +0.112 |
inter | +0.111 |
bars | +0.110 |
Neg logit contributions | |
Amen | -0.133 |
lder | -0.106 |
oc | -0.103 |
defeat | -0.100 |
chat | -0.097 |
Token | muff |
Token ID | 27563 |
Feature activation | +5.30e+00 |
Pos logit contributions | |
auc | +0.364 |
line | +0.347 |
ays | +0.333 |
lights | +0.331 |
ained | +0.331 |
Neg logit contributions | |
Amen | -0.241 |
lder | -0.231 |
oc | -0.200 |
", | -0.193 |
Raven | -0.186 |
Token | ins |
Token ID | 1040 |
Feature activation | +3.67e+00 |
Pos logit contributions | |
auc | +0.135 |
line | +0.127 |
inter | +0.126 |
tubes | +0.125 |
bars | +0.124 |
Neg logit contributions | |
Amen | -0.174 |
lder | -0.152 |
oc | -0.144 |
Frank | -0.133 |
defeat | -0.132 |
Token | smell |
Token ID | 8508 |
Feature activation | +2.97e+00 |
Pos logit contributions | |
auc | +0.084 |
bars | +0.078 |
lights | +0.075 |
line | +0.075 |
ays | +0.073 |
Neg logit contributions | |
Amen | -0.131 |
lder | -0.120 |
chat | -0.111 |
defeat | -0.110 |
oc | -0.108 |
Token | amazing |
Token ID | 4998 |
Feature activation | -8.16e-01 |
Pos logit contributions | |
auc | +0.103 |
bars | +0.099 |
ays | +0.095 |
ains | +0.095 |
line | +0.094 |
Neg logit contributions | |
Amen | -0.074 |
lder | -0.061 |
chat | -0.056 |
defeat | -0.055 |
", | -0.054 |
Token | !" |
Token ID | 2474 |
Feature activation | -2.59e+00 |
Pos logit contributions | |
Amen | +0.020 |
lder | +0.016 |
oc | +0.015 |
defeat | +0.014 |
Frank | +0.014 |
Neg logit contributions | |
auc | -0.030 |
bars | -0.027 |
line | -0.026 |
inter | -0.026 |
ains | -0.026 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
auc | +0.082 |
ays | +0.071 |
ains | +0.070 |
bars | +0.068 |
inter | +0.067 |
Neg logit contributions | |
Amen | -0.106 |
lder | -0.085 |
defeat | -0.084 |
oc | -0.083 |
Frank | -0.083 |
Token | didn |
Token ID | 1422 |
Feature activation | -4.77e-01 |
Pos logit contributions | |
auc | +0.028 |
line | +0.024 |
ays | +0.024 |
ains | +0.024 |
bars | +0.023 |
Neg logit contributions | |
Amen | -0.045 |
lder | -0.037 |
defeat | -0.035 |
Frank | -0.035 |
oc | -0.035 |
Token | 't |
Token ID | 470 |
Feature activation | -6.62e-01 |
Pos logit contributions | |
Amen | +0.014 |
lder | +0.011 |
oc | +0.010 |
defeat | +0.010 |
Frank | +0.010 |
Neg logit contributions | |
auc | -0.016 |
ays | -0.014 |
line | -0.014 |
bars | -0.014 |
ains | -0.013 |
Token | know |
Token ID | 760 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
Amen | +0.017 |
defeat | +0.013 |
lder | +0.013 |
chat | +0.012 |
read | +0.012 |
Neg logit contributions | |
auc | -0.024 |
ays | -0.023 |
bars | -0.022 |
ains | -0.022 |
lights | -0.021 |
Token | what |
Token ID | 644 |
Feature activation | +7.17e+00 |
Pos logit contributions | |
Amen | +0.043 |
Frank | +0.037 |
defeat | +0.036 |
lder | +0.035 |
oc | +0.034 |
Neg logit contributions | |
auc | -0.060 |
inter | -0.056 |
tub | -0.052 |
ays | -0.052 |
ash | -0.051 |
Token | those |
Token ID | 883 |
Feature activation | +1.05e+01 |
Pos logit contributions | |
auc | +0.226 |
ays | +0.219 |
bars | +0.211 |
line | +0.208 |
ains | +0.206 |
Neg logit contributions | |
Amen | -0.206 |
lder | -0.175 |
defeat | -0.168 |
Frank | -0.167 |
read | -0.165 |
Token | words |
Token ID | 2456 |
Feature activation | +5.23e+00 |
Pos logit contributions | |
auc | +0.327 |
line | +0.302 |
bars | +0.291 |
ays | +0.289 |
inter | +0.275 |
Neg logit contributions | |
lder | -0.274 |
Amen | -0.270 |
read | -0.264 |
chat | -0.242 |
Raven | -0.228 |
Token | meant |
Token ID | 4001 |
Feature activation | -9.25e-01 |
Pos logit contributions | |
auc | +0.154 |
ays | +0.133 |
ash | +0.130 |
inter | +0.129 |
ains | +0.129 |
Neg logit contributions | |
Amen | -0.146 |
read | -0.142 |
chat | -0.136 |
defeat | -0.134 |
lder | -0.133 |
Token | , |
Token ID | 11 |
Feature activation | +8.30e-01 |
Pos logit contributions | |
Amen | +0.019 |
lder | +0.015 |
defeat | +0.015 |
read | +0.015 |
chat | +0.014 |
Neg logit contributions | |
auc | -0.035 |
ays | -0.032 |
bars | -0.032 |
ains | -0.032 |
inter | -0.031 |
Token | so |
Token ID | 523 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
auc | +0.026 |
ays | +0.023 |
ains | +0.023 |
ash | +0.022 |
bars | +0.022 |
Neg logit contributions | |
Amen | -0.024 |
lder | -0.020 |
Frank | -0.019 |
defeat | -0.019 |
chat | -0.019 |
Token | she |
Token ID | 673 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
Amen | +0.084 |
lder | +0.071 |
defeat | +0.069 |
Frank | +0.068 |
oc | +0.068 |
Neg logit contributions | |
auc | -0.056 |
ains | -0.047 |
ays | -0.047 |
inter | -0.045 |
line | -0.045 |
Token | Lily |
Token ID | 20037 |
Feature activation | +7.24e-01 |
Pos logit contributions | |
Amen | +0.002 |
lder | +0.001 |
chat | +0.001 |
read | +0.001 |
oc | +0.001 |
Neg logit contributions | |
auc | -0.001 |
ains | -0.001 |
ays | -0.001 |
bars | -0.001 |
line | -0.001 |
Token | didn |
Token ID | 1422 |
Feature activation | -7.29e-01 |
Pos logit contributions | |
auc | +0.016 |
ains | +0.012 |
bars | +0.012 |
line | +0.012 |
tubes | +0.012 |
Neg logit contributions | |
Amen | -0.029 |
Frank | -0.023 |
defeat | -0.023 |
lder | -0.022 |
Charles | -0.022 |
Token | 't |
Token ID | 470 |
Feature activation | -6.84e-01 |
Pos logit contributions | |
Amen | +0.021 |
lder | +0.017 |
oc | +0.016 |
Frank | +0.016 |
defeat | +0.015 |
Neg logit contributions | |
auc | -0.024 |
ays | -0.021 |
bars | -0.021 |
line | -0.021 |
ains | -0.021 |
Token | know |
Token ID | 760 |
Feature activation | -2.14e+00 |
Pos logit contributions | |
Amen | +0.021 |
lder | +0.018 |
defeat | +0.016 |
Frank | +0.016 |
oc | +0.016 |
Neg logit contributions | |
auc | -0.021 |
ays | -0.018 |
line | -0.018 |
bars | -0.018 |
ains | -0.018 |
Token | what |
Token ID | 644 |
Feature activation | +6.94e+00 |
Pos logit contributions | |
Amen | +0.054 |
lder | +0.047 |
defeat | +0.045 |
Frank | +0.045 |
read | +0.044 |
Neg logit contributions | |
auc | -0.072 |
inter | -0.064 |
tub | -0.061 |
ays | -0.061 |
line | -0.060 |
Token | those |
Token ID | 883 |
Feature activation | +1.05e+01 |
Pos logit contributions | |
auc | +0.244 |
ays | +0.233 |
line | +0.226 |
bars | +0.222 |
lights | +0.220 |
Neg logit contributions | |
Amen | -0.181 |
lder | -0.159 |
read | -0.144 |
Frank | -0.138 |
defeat | -0.138 |
Token | words |
Token ID | 2456 |
Feature activation | +5.13e+00 |
Pos logit contributions | |
auc | +0.334 |
ays | +0.312 |
line | +0.308 |
bars | +0.295 |
lights | +0.282 |
Neg logit contributions | |
lder | -0.271 |
read | -0.248 |
Amen | -0.245 |
chat | -0.228 |
", | -0.210 |
Token | meant |
Token ID | 4001 |
Feature activation | -9.95e-01 |
Pos logit contributions | |
auc | +0.151 |
ays | +0.129 |
ash | +0.125 |
ains | +0.123 |
inter | +0.123 |
Neg logit contributions | |
Amen | -0.143 |
read | -0.140 |
lder | -0.135 |
chat | -0.135 |
defeat | -0.133 |
Token | , |
Token ID | 11 |
Feature activation | +6.71e-01 |
Pos logit contributions | |
Amen | +0.020 |
lder | +0.016 |
read | +0.015 |
defeat | +0.015 |
chat | +0.015 |
Neg logit contributions | |
auc | -0.039 |
ays | -0.036 |
bars | -0.035 |
ains | -0.035 |
lights | -0.034 |
Token | so |
Token ID | 523 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
auc | +0.021 |
ays | +0.019 |
ains | +0.018 |
bars | +0.018 |
ash | +0.018 |
Neg logit contributions | |
Amen | -0.020 |
lder | -0.016 |
Frank | -0.016 |
read | -0.016 |
chat | -0.015 |
Token | her |
Token ID | 607 |
Feature activation | -4.74e-02 |
Pos logit contributions | |
Amen | +0.091 |
lder | +0.078 |
defeat | +0.074 |
oc | +0.074 |
Frank | +0.074 |
Neg logit contributions | |
auc | -0.050 |
ays | -0.041 |
line | -0.041 |
bars | -0.041 |
ains | -0.041 |
Token | bugs |
Token ID | 11316 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
auc | +0.013 |
ays | +0.012 |
ains | +0.011 |
bars | +0.011 |
ash | +0.011 |
Neg logit contributions | |
Amen | -0.013 |
defeat | -0.011 |
read | -0.010 |
lder | -0.010 |
chat | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +9.09e-01 |
Pos logit contributions | |
auc | +0.054 |
ays | +0.047 |
bars | +0.047 |
ains | +0.047 |
inter | +0.045 |
Neg logit contributions | |
Amen | -0.073 |
lder | -0.061 |
defeat | -0.059 |
oc | -0.057 |
chat | -0.057 |
Token | They |
Token ID | 1119 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
auc | +0.024 |
bars | +0.021 |
ash | +0.020 |
ains | +0.020 |
line | +0.020 |
Neg logit contributions | |
Amen | -0.032 |
lder | -0.025 |
oc | -0.025 |
Frank | -0.024 |
defeat | -0.024 |
Token | knew |
Token ID | 2993 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
Amen | +0.010 |
lder | +0.008 |
defeat | +0.008 |
oc | +0.008 |
chat | +0.008 |
Neg logit contributions | |
auc | -0.008 |
ays | -0.007 |
line | -0.007 |
ains | -0.007 |
bars | -0.007 |
Token | that |
Token ID | 326 |
Feature activation | +3.49e+00 |
Pos logit contributions | |
Amen | +0.006 |
defeat | +0.005 |
Frank | +0.005 |
Raven | +0.005 |
read | +0.005 |
Neg logit contributions | |
auc | -0.008 |
inter | -0.007 |
ash | -0.007 |
ays | -0.007 |
line | -0.007 |
Token | those |
Token ID | 883 |
Feature activation | +1.04e+01 |
Pos logit contributions | |
auc | +0.093 |
ays | +0.088 |
bars | +0.085 |
inter | +0.084 |
line | +0.083 |
Neg logit contributions | |
Amen | -0.112 |
read | -0.095 |
chat | -0.095 |
defeat | -0.091 |
lder | -0.087 |
Token | were |
Token ID | 547 |
Feature activation | +2.28e+00 |
Pos logit contributions | |
ays | +0.318 |
line | +0.314 |
bars | +0.291 |
auc | +0.285 |
inter | +0.276 |
Neg logit contributions | |
read | -0.233 |
Amen | -0.231 |
lder | -0.222 |
chat | -0.213 |
defeat | -0.208 |
Token | more |
Token ID | 517 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
auc | +0.083 |
ays | +0.082 |
bars | +0.081 |
line | +0.078 |
inter | +0.076 |
Neg logit contributions | |
Amen | -0.052 |
defeat | -0.044 |
Frank | -0.040 |
read | -0.039 |
chat | -0.037 |
Token | fun |
Token ID | 1257 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
bars | +0.005 |
auc | +0.005 |
ays | +0.005 |
ains | +0.005 |
line | +0.005 |
Neg logit contributions | |
Amen | -0.004 |
defeat | -0.003 |
lder | -0.003 |
Frank | -0.003 |
chat | -0.003 |
Token | than |
Token ID | 621 |
Feature activation | +5.80e+00 |
Pos logit contributions | |
Amen | +0.059 |
defeat | +0.049 |
chat | +0.048 |
lder | +0.048 |
read | +0.046 |
Neg logit contributions | |
auc | -0.103 |
bars | -0.100 |
idget | -0.099 |
inter | -0.095 |
ays | -0.095 |
Token | any |
Token ID | 597 |
Feature activation | +5.40e+00 |
Pos logit contributions | |
auc | +0.171 |
ays | +0.161 |
bars | +0.151 |
line | +0.149 |
ains | +0.144 |
Neg logit contributions | |
Amen | -0.161 |
defeat | -0.149 |
chat | -0.143 |
read | -0.139 |
arrest | -0.137 |
Token | a |
Token ID | 257 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
auc | +0.066 |
ays | +0.057 |
bars | +0.057 |
line | +0.057 |
ains | +0.056 |
Neg logit contributions | |
Amen | -0.080 |
lder | -0.066 |
defeat | -0.063 |
Frank | -0.062 |
oc | -0.061 |
Token | voice |
Token ID | 3809 |
Feature activation | -6.55e-01 |
Pos logit contributions | |
Amen | +0.082 |
lder | +0.068 |
oc | +0.064 |
arrest | +0.063 |
Frank | +0.062 |
Neg logit contributions | |
auc | -0.065 |
tubes | -0.057 |
line | -0.056 |
inter | -0.054 |
tub | -0.054 |
Token | answered |
Token ID | 9373 |
Feature activation | -2.84e+00 |
Pos logit contributions | |
Amen | +0.023 |
lder | +0.019 |
defeat | +0.018 |
Frank | +0.018 |
oc | +0.018 |
Neg logit contributions | |
auc | -0.018 |
line | -0.015 |
ays | -0.015 |
bars | -0.014 |
ains | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | +8.12e-01 |
Pos logit contributions | |
Amen | +0.090 |
lder | +0.074 |
defeat | +0.068 |
oc | +0.068 |
Frank | +0.067 |
Neg logit contributions | |
auc | -0.086 |
line | -0.075 |
ays | -0.075 |
ains | -0.074 |
bars | -0.073 |
Token | " |
Token ID | 366 |
Feature activation | +6.96e-01 |
Pos logit contributions | |
auc | +0.016 |
ays | +0.012 |
line | +0.012 |
bars | +0.012 |
ains | +0.012 |
Neg logit contributions | |
Amen | -0.034 |
lder | -0.030 |
defeat | -0.028 |
Frank | -0.028 |
oc | -0.028 |
Token | Those |
Token ID | 9627 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
auc | +0.017 |
ays | +0.015 |
ains | +0.014 |
line | +0.014 |
bars | +0.014 |
Neg logit contributions | |
Amen | -0.026 |
lder | -0.022 |
Frank | -0.021 |
oc | -0.021 |
defeat | -0.020 |
Token | numbers |
Token ID | 3146 |
Feature activation | +4.23e+00 |
Pos logit contributions | |
auc | +0.238 |
line | +0.203 |
ays | +0.199 |
lights | +0.195 |
bars | +0.194 |
Neg logit contributions | |
Amen | -0.388 |
lder | -0.347 |
defeat | -0.329 |
oc | -0.324 |
Frank | -0.316 |
Token | are |
Token ID | 389 |
Feature activation | +2.64e+00 |
Pos logit contributions | |
auc | +0.098 |
inter | +0.083 |
line | +0.081 |
lights | +0.080 |
ays | +0.080 |
Neg logit contributions | |
Amen | -0.154 |
lder | -0.138 |
defeat | -0.133 |
oc | -0.128 |
chat | -0.126 |
Token | for |
Token ID | 329 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
auc | +0.078 |
ays | +0.069 |
line | +0.067 |
bars | +0.067 |
inter | +0.065 |
Neg logit contributions | |
Amen | -0.083 |
lder | -0.070 |
defeat | -0.069 |
Frank | -0.066 |
oc | -0.064 |
Token | a |
Token ID | 257 |
Feature activation | -3.38e+00 |
Pos logit contributions | |
auc | +0.044 |
ains | +0.038 |
ays | +0.038 |
line | +0.038 |
bars | +0.037 |
Neg logit contributions | |
Amen | -0.061 |
lder | -0.050 |
oc | -0.047 |
defeat | -0.047 |
Frank | -0.047 |
Token | treasure |
Token ID | 14068 |
Feature activation | -4.90e-01 |
Pos logit contributions | |
Amen | +0.108 |
lder | +0.086 |
Frank | +0.084 |
Charles | +0.079 |
oc | +0.077 |
Neg logit contributions | |
auc | -0.089 |
tubes | -0.081 |
inter | -0.079 |
tub | -0.076 |
line | -0.074 |
Token | dad |
Token ID | 9955 |
Feature activation | -3.45e+00 |
Pos logit contributions | |
Amen | +0.064 |
lder | +0.056 |
Frank | +0.054 |
Mic | +0.052 |
Charles | +0.051 |
Neg logit contributions | |
auc | -0.039 |
tubes | -0.035 |
lights | -0.033 |
ays | -0.033 |
ains | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | +8.91e-01 |
Pos logit contributions | |
Amen | +0.128 |
lder | +0.108 |
defeat | +0.103 |
oc | +0.102 |
Frank | +0.100 |
Neg logit contributions | |
auc | -0.084 |
ays | -0.071 |
ains | -0.070 |
bars | -0.070 |
line | -0.070 |
Token | " |
Token ID | 366 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
auc | +0.027 |
ains | +0.024 |
bars | +0.023 |
ays | +0.023 |
ash | +0.023 |
Neg logit contributions | |
Amen | -0.028 |
lder | -0.023 |
Frank | -0.021 |
Mic | -0.021 |
oc | -0.021 |
Token | Wow |
Token ID | 22017 |
Feature activation | +4.19e+00 |
Pos logit contributions | |
auc | +0.029 |
line | +0.025 |
ays | +0.025 |
bars | +0.025 |
ains | +0.024 |
Neg logit contributions | |
Amen | -0.038 |
lder | -0.033 |
defeat | -0.031 |
oc | -0.030 |
Frank | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | +4.42e+00 |
Pos logit contributions | |
auc | +0.128 |
ays | +0.114 |
ains | +0.113 |
line | +0.113 |
bars | +0.112 |
Neg logit contributions | |
Amen | -0.131 |
lder | -0.104 |
defeat | -0.097 |
Frank | -0.097 |
oc | -0.097 |
Token | those |
Token ID | 883 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
auc | +0.107 |
line | +0.091 |
bars | +0.089 |
ays | +0.088 |
ains | +0.087 |
Neg logit contributions | |
Amen | -0.163 |
lder | -0.138 |
oc | -0.133 |
defeat | -0.131 |
Frank | -0.131 |
Token | look |
Token ID | 804 |
Feature activation | +5.49e+00 |
Pos logit contributions | |
auc | +0.295 |
line | +0.273 |
inter | +0.262 |
lights | +0.260 |
ays | +0.259 |
Neg logit contributions | |
Amen | -0.305 |
lder | -0.278 |
", | -0.250 |
read | -0.244 |
oc | -0.243 |
Token | great |
Token ID | 1049 |
Feature activation | -4.66e-01 |
Pos logit contributions | |
auc | +0.176 |
bars | +0.158 |
ays | +0.157 |
line | +0.155 |
ains | +0.155 |
Neg logit contributions | |
Amen | -0.154 |
lder | -0.130 |
defeat | -0.120 |
oc | -0.117 |
chat | -0.115 |
Token | on |
Token ID | 319 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
Amen | +0.014 |
lder | +0.011 |
Mic | +0.010 |
Frank | +0.010 |
defeat | +0.010 |
Neg logit contributions | |
auc | -0.015 |
ays | -0.014 |
line | -0.014 |
bars | -0.014 |
lights | -0.013 |
Token | you |
Token ID | 345 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
auc | +0.075 |
ays | +0.066 |
bars | +0.065 |
line | +0.065 |
ains | +0.065 |
Neg logit contributions | |
Amen | -0.070 |
lder | -0.058 |
defeat | -0.053 |
Frank | -0.052 |
oc | -0.052 |
Token | !" |
Token ID | 2474 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
auc | +0.031 |
ains | +0.031 |
line | +0.030 |
ays | +0.029 |
ash | +0.028 |
Neg logit contributions | |
Amen | -0.042 |
Frank | -0.034 |
Mic | -0.034 |
lder | -0.032 |
Charles | -0.032 |
Token | flowers |
Token ID | 12734 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
Amen | +0.124 |
lder | +0.105 |
defeat | +0.099 |
Frank | +0.099 |
oc | +0.099 |
Neg logit contributions | |
auc | -0.091 |
ays | -0.078 |
bars | -0.078 |
ains | -0.078 |
line | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | -4.25e-01 |
Pos logit contributions | |
auc | +0.059 |
inter | +0.051 |
bars | +0.050 |
ains | +0.050 |
ash | +0.050 |
Neg logit contributions | |
Amen | -0.059 |
lder | -0.049 |
chat | -0.049 |
read | -0.048 |
defeat | -0.047 |
Token | " |
Token ID | 366 |
Feature activation | +5.81e-01 |
Pos logit contributions | |
Amen | +0.012 |
lder | +0.011 |
chat | +0.010 |
Frank | +0.010 |
read | +0.010 |
Neg logit contributions | |
auc | -0.012 |
iness | -0.011 |
lessness | -0.011 |
ying | -0.011 |
ains | -0.011 |
Token | Oh |
Token ID | 5812 |
Feature activation | +4.92e+00 |
Pos logit contributions | |
auc | +0.017 |
ays | +0.015 |
bars | +0.015 |
line | +0.015 |
ains | +0.015 |
Neg logit contributions | |
Amen | -0.019 |
lder | -0.015 |
Frank | -0.014 |
oc | -0.014 |
defeat | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | +3.64e+00 |
Pos logit contributions | |
auc | +0.137 |
ays | +0.118 |
line | +0.118 |
bars | +0.117 |
ains | +0.117 |
Neg logit contributions | |
Amen | -0.166 |
lder | -0.138 |
defeat | -0.130 |
Frank | -0.128 |
oc | -0.128 |
Token | those |
Token ID | 883 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
auc | +0.092 |
ays | +0.079 |
bars | +0.077 |
line | +0.077 |
ains | +0.076 |
Neg logit contributions | |
Amen | -0.133 |
lder | -0.113 |
defeat | -0.106 |
Frank | -0.104 |
oc | -0.104 |
Token | flowers |
Token ID | 12734 |
Feature activation | +4.18e+00 |
Pos logit contributions | |
auc | +0.305 |
line | +0.268 |
inter | +0.264 |
bars | +0.260 |
ays | +0.259 |
Neg logit contributions | |
lder | -0.290 |
Amen | -0.283 |
", | -0.262 |
chat | -0.248 |
read | -0.248 |
Token | are |
Token ID | 389 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
auc | +0.114 |
bars | +0.098 |
ays | +0.097 |
line | +0.094 |
ains | +0.094 |
Neg logit contributions | |
Amen | -0.140 |
lder | -0.122 |
defeat | -0.113 |
chat | -0.111 |
oc | -0.110 |
Token | so |
Token ID | 523 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
auc | +0.057 |
bars | +0.050 |
ays | +0.049 |
line | +0.049 |
ains | +0.048 |
Neg logit contributions | |
Amen | -0.060 |
lder | -0.050 |
Frank | -0.046 |
defeat | -0.046 |
chat | -0.045 |
Token | pretty |
Token ID | 2495 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
Amen | +0.063 |
lder | +0.053 |
chat | +0.047 |
Frank | +0.046 |
defeat | +0.042 |
Neg logit contributions | |
auc | -0.079 |
ays | -0.072 |
bars | -0.072 |
inter | -0.071 |
ains | -0.070 |
Token | !" |
Token ID | 2474 |
Feature activation | -3.20e+00 |
Pos logit contributions | |
Amen | +0.024 |
lder | +0.018 |
defeat | +0.017 |
Mic | +0.017 |
oc | +0.016 |
Neg logit contributions | |
auc | -0.038 |
ays | -0.035 |
line | -0.035 |
lights | -0.034 |
bars | -0.034 |
Token | sky |
Token ID | 6766 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
auc | +0.055 |
ays | +0.052 |
idget | +0.049 |
ours | +0.049 |
ains | +0.048 |
Neg logit contributions | |
Amen | -0.072 |
chat | -0.057 |
lder | -0.057 |
Frank | -0.056 |
defeat | -0.055 |
Token | . |
Token ID | 13 |
Feature activation | +5.26e-01 |
Pos logit contributions | |
auc | +0.038 |
ays | +0.033 |
bars | +0.033 |
ains | +0.033 |
inter | +0.032 |
Neg logit contributions | |
Amen | -0.043 |
lder | -0.036 |
chat | -0.035 |
defeat | -0.035 |
oc | -0.034 |
Token |
|
Token ID | 198 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
auc | +0.020 |
ains | +0.017 |
lights | +0.017 |
ash | +0.017 |
iness | +0.017 |
Neg logit contributions | |
Amen | -0.013 |
Frank | -0.009 |
lder | -0.009 |
Mic | -0.008 |
oc | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
Amen | +0.054 |
lder | +0.046 |
defeat | +0.042 |
oc | +0.042 |
Frank | +0.041 |
Neg logit contributions | |
auc | -0.042 |
line | -0.037 |
ays | -0.036 |
bars | -0.036 |
ains | -0.036 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.00e+00 |
Pos logit contributions | |
Amen | +0.034 |
lder | +0.026 |
defeat | +0.024 |
oc | +0.024 |
Frank | +0.024 |
Neg logit contributions | |
auc | -0.047 |
ays | -0.042 |
line | -0.042 |
bars | -0.042 |
ains | -0.041 |
Token | everything |
Token ID | 2279 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
auc | +0.110 |
ash | +0.096 |
vinegar | +0.092 |
sizes | +0.092 |
lights | +0.091 |
Neg logit contributions | |
Amen | -0.121 |
lder | -0.110 |
", | -0.100 |
defeat | -0.100 |
read | -0.098 |
Token | was |
Token ID | 373 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
auc | +0.271 |
line | +0.247 |
ains | +0.245 |
bars | +0.245 |
ays | +0.244 |
Neg logit contributions | |
Amen | -0.336 |
lder | -0.289 |
defeat | -0.279 |
chat | -0.275 |
Frank | -0.265 |
Token | unp |
Token ID | 8593 |
Feature activation | +4.45e+00 |
Pos logit contributions | |
auc | +0.053 |
ays | +0.045 |
bars | +0.045 |
line | +0.045 |
ains | +0.044 |
Neg logit contributions | |
Amen | -0.069 |
lder | -0.058 |
Frank | -0.055 |
defeat | -0.055 |
oc | -0.053 |
Token | acked |
Token ID | 6021 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
auc | +0.133 |
line | +0.121 |
bars | +0.118 |
ains | +0.116 |
tubes | +0.115 |
Neg logit contributions | |
Amen | -0.134 |
lder | -0.111 |
oc | -0.104 |
defeat | -0.100 |
Frank | -0.100 |
Token | , |
Token ID | 11 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
Amen | +0.027 |
lder | +0.023 |
defeat | +0.022 |
Frank | +0.022 |
chat | +0.021 |
Neg logit contributions | |
auc | -0.039 |
ays | -0.036 |
bars | -0.035 |
inter | -0.035 |
line | -0.035 |
Token | Tin |
Token ID | 22894 |
Feature activation | +3.84e+00 |
Pos logit contributions | |
auc | +0.062 |
ash | +0.053 |
unes | +0.050 |
ays | +0.050 |
oes | +0.050 |
Neg logit contributions | |
Amen | -0.042 |
Frank | -0.037 |
read | -0.037 |
chat | -0.035 |
defeat | -0.034 |
Token | sky |
Token ID | 6766 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
auc | +0.055 |
ays | +0.052 |
idget | +0.049 |
ours | +0.049 |
ains | +0.048 |
Neg logit contributions | |
Amen | -0.072 |
chat | -0.057 |
lder | -0.057 |
Frank | -0.056 |
defeat | -0.055 |
Token | . |
Token ID | 13 |
Feature activation | +5.26e-01 |
Pos logit contributions | |
auc | +0.038 |
ays | +0.033 |
bars | +0.033 |
ains | +0.033 |
inter | +0.032 |
Neg logit contributions | |
Amen | -0.043 |
lder | -0.036 |
chat | -0.035 |
defeat | -0.035 |
oc | -0.034 |
Token |
|
Token ID | 198 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
auc | +0.020 |
ains | +0.017 |
lights | +0.017 |
ash | +0.017 |
iness | +0.017 |
Neg logit contributions | |
Amen | -0.013 |
Frank | -0.009 |
lder | -0.009 |
Mic | -0.008 |
oc | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
Amen | +0.054 |
lder | +0.046 |
defeat | +0.042 |
oc | +0.042 |
Frank | +0.041 |
Neg logit contributions | |
auc | -0.042 |
line | -0.037 |
ays | -0.036 |
bars | -0.036 |
ains | -0.036 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.00e+00 |
Pos logit contributions | |
Amen | +0.034 |
lder | +0.026 |
defeat | +0.024 |
oc | +0.024 |
Frank | +0.024 |
Neg logit contributions | |
auc | -0.047 |
ays | -0.042 |
line | -0.042 |
bars | -0.042 |
ains | -0.041 |
Token | everything |
Token ID | 2279 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
auc | +0.110 |
ash | +0.096 |
vinegar | +0.092 |
sizes | +0.092 |
lights | +0.091 |
Neg logit contributions | |
Amen | -0.121 |
lder | -0.110 |
", | -0.100 |
defeat | -0.100 |
read | -0.098 |
Token | was |
Token ID | 373 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
auc | +0.271 |
line | +0.247 |
ains | +0.245 |
bars | +0.245 |
ays | +0.244 |
Neg logit contributions | |
Amen | -0.336 |
lder | -0.289 |
defeat | -0.279 |
chat | -0.275 |
Frank | -0.265 |
Token | unp |
Token ID | 8593 |
Feature activation | +4.45e+00 |
Pos logit contributions | |
auc | +0.053 |
ays | +0.045 |
bars | +0.045 |
line | +0.045 |
ains | +0.044 |
Neg logit contributions | |
Amen | -0.069 |
lder | -0.058 |
Frank | -0.055 |
defeat | -0.055 |
oc | -0.053 |
Token | acked |
Token ID | 6021 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
auc | +0.133 |
line | +0.121 |
bars | +0.118 |
ains | +0.116 |
tubes | +0.115 |
Neg logit contributions | |
Amen | -0.134 |
lder | -0.111 |
oc | -0.104 |
defeat | -0.100 |
Frank | -0.100 |
Token | , |
Token ID | 11 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
Amen | +0.027 |
lder | +0.023 |
defeat | +0.022 |
Frank | +0.022 |
chat | +0.021 |
Neg logit contributions | |
auc | -0.039 |
ays | -0.036 |
bars | -0.035 |
inter | -0.035 |
line | -0.035 |
Token | Tin |
Token ID | 22894 |
Feature activation | +3.84e+00 |
Pos logit contributions | |
auc | +0.062 |
ash | +0.053 |
unes | +0.050 |
ays | +0.050 |
oes | +0.050 |
Neg logit contributions | |
Amen | -0.042 |
Frank | -0.037 |
read | -0.037 |
chat | -0.035 |
defeat | -0.034 |
Token | taped |
Token ID | 36114 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
Amen | +0.008 |
read | +0.008 |
defeat | +0.008 |
chat | +0.007 |
oc | +0.007 |
Neg logit contributions | |
auc | -0.009 |
ays | -0.009 |
ains | -0.009 |
inter | -0.008 |
unes | -0.008 |
Token | all |
Token ID | 477 |
Feature activation | +5.30e+00 |
Pos logit contributions | |
auc | +0.004 |
line | +0.004 |
ays | +0.004 |
inter | +0.004 |
ains | +0.004 |
Neg logit contributions | |
Amen | -0.004 |
defeat | -0.004 |
chat | -0.003 |
Frank | -0.003 |
oc | -0.003 |
Token | her |
Token ID | 607 |
Feature activation | -8.76e-01 |
Pos logit contributions | |
ains | +0.181 |
der | +0.180 |
auc | +0.177 |
line | +0.177 |
ancer | +0.167 |
Neg logit contributions | |
Amen | -0.083 |
chat | -0.081 |
read | -0.076 |
defeat | -0.074 |
Jean | -0.072 |
Token | toys |
Token ID | 14958 |
Feature activation | -3.47e-01 |
Pos logit contributions | |
Amen | +0.025 |
defeat | +0.022 |
oc | +0.022 |
lder | +0.022 |
chat | +0.020 |
Neg logit contributions | |
auc | -0.026 |
line | -0.025 |
ays | -0.023 |
bars | -0.022 |
ains | -0.021 |
Token | until |
Token ID | 1566 |
Feature activation | +4.51e+00 |
Pos logit contributions | |
defeat | +0.007 |
chat | +0.006 |
read | +0.006 |
Amen | +0.006 |
", | +0.006 |
Neg logit contributions | |
auc | -0.012 |
ays | -0.011 |
ains | -0.011 |
ours | -0.011 |
inter | -0.010 |
Token | everything |
Token ID | 2279 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
auc | +0.145 |
ains | +0.133 |
inter | +0.132 |
ash | +0.129 |
ays | +0.128 |
Neg logit contributions | |
Amen | -0.117 |
chat | -0.108 |
read | -0.105 |
defeat | -0.099 |
lder | -0.097 |
Token | was |
Token ID | 373 |
Feature activation | +2.77e+00 |
Pos logit contributions | |
inter | +0.314 |
line | +0.308 |
ays | +0.306 |
auc | +0.304 |
ains | +0.298 |
Neg logit contributions | |
chat | -0.221 |
read | -0.202 |
lder | -0.201 |
Amen | -0.199 |
defeat | -0.198 |
Token | put |
Token ID | 1234 |
Feature activation | +3.06e+00 |
Pos logit contributions | |
auc | +0.083 |
ays | +0.073 |
bars | +0.072 |
inter | +0.070 |
line | +0.070 |
Neg logit contributions | |
Amen | -0.088 |
lder | -0.070 |
defeat | -0.069 |
Frank | -0.068 |
chat | -0.066 |
Token | away |
Token ID | 1497 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
auc | +0.091 |
inter | +0.085 |
ours | +0.084 |
bars | +0.084 |
line | +0.083 |
Neg logit contributions | |
Amen | -0.072 |
defeat | -0.066 |
chat | -0.064 |
lder | -0.061 |
reunited | -0.059 |
Token | . |
Token ID | 13 |
Feature activation | +9.72e-01 |
Pos logit contributions | |
Amen | +0.023 |
defeat | +0.022 |
chat | +0.019 |
", | +0.019 |
read | +0.019 |
Neg logit contributions | |
auc | -0.033 |
ays | -0.032 |
bars | -0.032 |
inter | -0.031 |
ours | -0.030 |
Token | Jen |
Token ID | 13364 |
Feature activation | +2.54e+00 |
Pos logit contributions | |
auc | +0.032 |
ains | +0.029 |
lights | +0.028 |
ays | +0.028 |
bars | +0.028 |
Neg logit contributions | |
Amen | -0.026 |
oc | -0.022 |
lder | -0.021 |
Frank | -0.020 |
chat | -0.019 |
Token | both |
Token ID | 1111 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
auc | +0.025 |
bars | +0.021 |
line | +0.021 |
ays | +0.021 |
ains | +0.020 |
Neg logit contributions | |
Amen | -0.054 |
lder | -0.048 |
Frank | -0.045 |
oc | -0.045 |
defeat | -0.044 |
Token | . |
Token ID | 13 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
auc | +0.027 |
line | +0.022 |
ays | +0.022 |
ains | +0.022 |
bars | +0.022 |
Neg logit contributions | |
Amen | -0.043 |
lder | -0.036 |
defeat | -0.034 |
oc | -0.034 |
Frank | -0.034 |
Token |
|
Token ID | 198 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
auc | +0.037 |
bars | +0.032 |
ash | +0.032 |
ains | +0.032 |
inter | +0.031 |
Neg logit contributions | |
Amen | -0.036 |
Mic | -0.027 |
Frank | -0.027 |
lder | -0.027 |
Offic | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
Amen | +0.052 |
lder | +0.045 |
defeat | +0.042 |
chat | +0.041 |
Frank | +0.040 |
Neg logit contributions | |
auc | -0.030 |
line | -0.026 |
ays | -0.025 |
ains | -0.025 |
tubes | -0.024 |
Token | " |
Token ID | 1 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
Amen | +0.033 |
lder | +0.027 |
defeat | +0.025 |
chat | +0.023 |
Charles | +0.023 |
Neg logit contributions | |
auc | -0.037 |
line | -0.034 |
bars | -0.032 |
tubes | -0.032 |
ains | -0.032 |
Token | Those |
Token ID | 9627 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
auc | +0.048 |
ains | +0.043 |
ays | +0.043 |
inter | +0.041 |
line | +0.041 |
Neg logit contributions | |
Amen | -0.061 |
lder | -0.049 |
Frank | -0.049 |
Mic | -0.047 |
Charles | -0.046 |
Token | are |
Token ID | 389 |
Feature activation | +2.95e+00 |
Pos logit contributions | |
auc | +0.238 |
line | +0.206 |
ays | +0.205 |
bars | +0.202 |
ash | +0.192 |
Neg logit contributions | |
Amen | -0.370 |
lder | -0.328 |
defeat | -0.316 |
chat | -0.315 |
read | -0.304 |
Token | wonderful |
Token ID | 7932 |
Feature activation | +5.22e-01 |
Pos logit contributions | |
auc | +0.095 |
line | +0.085 |
ays | +0.085 |
bars | +0.085 |
inter | +0.082 |
Neg logit contributions | |
Amen | -0.088 |
lder | -0.070 |
defeat | -0.068 |
Frank | -0.066 |
read | -0.064 |
Token | names |
Token ID | 3891 |
Feature activation | +3.69e+00 |
Pos logit contributions | |
auc | +0.015 |
ays | +0.013 |
line | +0.013 |
bars | +0.012 |
ains | +0.012 |
Neg logit contributions | |
Amen | -0.018 |
lder | -0.015 |
defeat | -0.014 |
Frank | -0.014 |
oc | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | +2.80e+00 |
Pos logit contributions | |
auc | +0.098 |
ays | +0.085 |
line | +0.084 |
bars | +0.084 |
ains | +0.083 |
Neg logit contributions | |
Amen | -0.128 |
lder | -0.107 |
defeat | -0.101 |
Frank | -0.100 |
oc | -0.100 |
Token | Lily |
Token ID | 20037 |
Feature activation | +7.83e-01 |
Pos logit contributions | |
auc | +0.058 |
ays | +0.048 |
line | +0.047 |
bars | +0.047 |
ains | +0.046 |
Neg logit contributions | |
Amen | -0.115 |
lder | -0.098 |
defeat | -0.094 |
Frank | -0.093 |
oc | -0.093 |
Token | said |
Token ID | 531 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
Amen | +0.013 |
lder | +0.011 |
chat | +0.010 |
defeat | +0.010 |
oc | +0.010 |
Neg logit contributions | |
auc | -0.016 |
ays | -0.015 |
ains | -0.014 |
bars | -0.014 |
inter | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
Amen | +0.076 |
lder | +0.061 |
defeat | +0.059 |
Frank | +0.059 |
oc | +0.058 |
Neg logit contributions | |
auc | -0.068 |
line | -0.058 |
bars | -0.057 |
ays | -0.056 |
ains | -0.056 |
Token | " |
Token ID | 366 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
auc | +0.026 |
ays | +0.023 |
ains | +0.022 |
bars | +0.022 |
ash | +0.021 |
Neg logit contributions | |
Amen | -0.045 |
lder | -0.038 |
defeat | -0.036 |
Frank | -0.036 |
oc | -0.036 |
Token | Wow |
Token ID | 22017 |
Feature activation | +3.92e+00 |
Pos logit contributions | |
auc | +0.032 |
ays | +0.029 |
ains | +0.029 |
lights | +0.028 |
bars | +0.028 |
Neg logit contributions | |
Amen | -0.035 |
Frank | -0.028 |
lder | -0.028 |
Mic | -0.028 |
oc | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | +3.99e+00 |
Pos logit contributions | |
auc | +0.135 |
bars | +0.120 |
line | +0.119 |
ays | +0.119 |
ains | +0.117 |
Neg logit contributions | |
Amen | -0.105 |
lder | -0.086 |
defeat | -0.078 |
Frank | -0.076 |
oc | -0.076 |
Token | those |
Token ID | 883 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
auc | +0.082 |
ays | +0.069 |
bars | +0.068 |
line | +0.067 |
ains | +0.067 |
Neg logit contributions | |
Amen | -0.165 |
lder | -0.143 |
defeat | -0.135 |
Frank | -0.133 |
oc | -0.133 |
Token | pancakes |
Token ID | 47008 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
auc | +0.309 |
line | +0.281 |
ained | +0.280 |
inter | +0.274 |
bars | +0.271 |
Neg logit contributions | |
lder | -0.283 |
Amen | -0.276 |
", | -0.247 |
read | -0.222 |
!", | -0.217 |
Token | look |
Token ID | 804 |
Feature activation | +4.31e+00 |
Pos logit contributions | |
auc | +0.048 |
bars | +0.044 |
inter | +0.043 |
lights | +0.042 |
ays | +0.041 |
Neg logit contributions | |
Amen | -0.059 |
lder | -0.055 |
chat | -0.051 |
read | -0.049 |
", | -0.048 |
Token | delicious |
Token ID | 12625 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
auc | +0.149 |
bars | +0.139 |
ays | +0.134 |
ains | +0.134 |
line | +0.134 |
Neg logit contributions | |
Amen | -0.105 |
lder | -0.088 |
oc | -0.081 |
defeat | -0.080 |
chat | -0.080 |
Token | ! |
Token ID | 0 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
Amen | +0.050 |
lder | +0.040 |
oc | +0.035 |
Frank | +0.033 |
defeat | +0.033 |
Neg logit contributions | |
auc | -0.089 |
bars | -0.081 |
ains | -0.080 |
line | -0.080 |
ays | -0.079 |
Token | Can |
Token ID | 1680 |
Feature activation | -3.10e+00 |
Pos logit contributions | |
auc | +0.065 |
bars | +0.058 |
line | +0.058 |
lights | +0.057 |
ays | +0.057 |
Neg logit contributions | |
Amen | -0.083 |
lder | -0.065 |
oc | -0.064 |
Frank | -0.058 |
", | -0.057 |
Token | bad |
Token ID | 2089 |
Feature activation | -7.55e-01 |
Pos logit contributions | |
ays | +0.092 |
auc | +0.091 |
ains | +0.088 |
bars | +0.088 |
line | +0.084 |
Neg logit contributions | |
Amen | -0.087 |
defeat | -0.077 |
lder | -0.070 |
read | -0.067 |
oc | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | +4.40e-01 |
Pos logit contributions | |
Amen | +0.025 |
lder | +0.021 |
defeat | +0.021 |
oc | +0.020 |
read | +0.019 |
Neg logit contributions | |
auc | -0.020 |
bars | -0.018 |
ays | -0.018 |
ains | -0.018 |
line | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
auc | +0.013 |
ays | +0.011 |
ains | +0.011 |
bars | +0.011 |
line | +0.011 |
Neg logit contributions | |
Amen | -0.015 |
lder | -0.012 |
oc | -0.012 |
Frank | -0.012 |
defeat | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
Amen | +0.062 |
lder | +0.054 |
defeat | +0.049 |
chat | +0.048 |
oc | +0.048 |
Neg logit contributions | |
auc | -0.041 |
line | -0.035 |
ays | -0.034 |
ains | -0.034 |
bars | -0.034 |
Token | " |
Token ID | 1 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
Amen | +0.041 |
lder | +0.034 |
defeat | +0.031 |
chat | +0.031 |
Charles | +0.029 |
Neg logit contributions | |
auc | -0.040 |
line | -0.035 |
bars | -0.034 |
tubes | -0.033 |
ains | -0.033 |
Token | Those |
Token ID | 9627 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
auc | +0.034 |
ays | +0.030 |
ains | +0.030 |
line | +0.030 |
bars | +0.029 |
Neg logit contributions | |
Amen | -0.044 |
lder | -0.036 |
Frank | -0.035 |
defeat | -0.034 |
oc | -0.034 |
Token | are |
Token ID | 389 |
Feature activation | +3.11e+00 |
Pos logit contributions | |
auc | +0.218 |
ays | +0.186 |
line | +0.184 |
ains | +0.179 |
bars | +0.178 |
Neg logit contributions | |
Amen | -0.399 |
lder | -0.349 |
defeat | -0.337 |
oc | -0.330 |
Frank | -0.324 |
Token | not |
Token ID | 407 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
auc | +0.097 |
line | +0.087 |
ays | +0.087 |
bars | +0.085 |
ains | +0.084 |
Neg logit contributions | |
Amen | -0.094 |
lder | -0.077 |
defeat | -0.073 |
Frank | -0.071 |
oc | -0.070 |
Token | your |
Token ID | 534 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
auc | +0.053 |
ays | +0.046 |
line | +0.046 |
bars | +0.045 |
ains | +0.045 |
Neg logit contributions | |
Amen | -0.062 |
lder | -0.052 |
defeat | -0.050 |
Frank | -0.049 |
oc | -0.048 |
Token | toys |
Token ID | 14958 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
auc | +0.057 |
ays | +0.050 |
line | +0.049 |
bars | +0.049 |
ains | +0.049 |
Neg logit contributions | |
Amen | -0.055 |
lder | -0.045 |
defeat | -0.042 |
oc | -0.041 |
Frank | -0.041 |
Token | , |
Token ID | 11 |
Feature activation | +3.26e+00 |
Pos logit contributions | |
auc | +0.005 |
ays | +0.005 |
ains | +0.005 |
line | +0.005 |
bars | +0.005 |
Neg logit contributions | |
Amen | -0.003 |
lder | -0.002 |
defeat | -0.002 |
oc | -0.002 |
Frank | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
auc | +0.038 |
bars | +0.033 |
ains | +0.032 |
ays | +0.031 |
ash | +0.031 |
Neg logit contributions | |
Amen | -0.055 |
lder | -0.046 |
Frank | -0.044 |
oc | -0.043 |
defeat | -0.043 |
Token |
|
Token ID | 198 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
Amen | +0.045 |
lder | +0.038 |
defeat | +0.036 |
oc | +0.035 |
Frank | +0.035 |
Neg logit contributions | |
auc | -0.034 |
line | -0.029 |
ays | -0.029 |
ains | -0.029 |
bars | -0.028 |
Token | " |
Token ID | 1 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
Amen | +0.032 |
lder | +0.025 |
defeat | +0.023 |
chat | +0.023 |
oc | +0.022 |
Neg logit contributions | |
auc | -0.031 |
line | -0.029 |
ains | -0.027 |
tubes | -0.027 |
ays | -0.027 |
Token | Wow |
Token ID | 22017 |
Feature activation | +4.01e+00 |
Pos logit contributions | |
auc | +0.043 |
ays | +0.037 |
ains | +0.036 |
line | +0.036 |
bars | +0.036 |
Neg logit contributions | |
Amen | -0.063 |
lder | -0.053 |
Frank | -0.050 |
defeat | -0.049 |
oc | -0.049 |
Token | , |
Token ID | 11 |
Feature activation | +4.34e+00 |
Pos logit contributions | |
auc | +0.123 |
ays | +0.110 |
ains | +0.110 |
line | +0.109 |
bars | +0.107 |
Neg logit contributions | |
Amen | -0.124 |
lder | -0.098 |
Frank | -0.092 |
oc | -0.092 |
defeat | -0.092 |
Token | those |
Token ID | 883 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
auc | +0.105 |
line | +0.088 |
ays | +0.088 |
bars | +0.087 |
ains | +0.087 |
Neg logit contributions | |
Amen | -0.163 |
lder | -0.138 |
defeat | -0.130 |
Frank | -0.129 |
oc | -0.129 |
Token | look |
Token ID | 804 |
Feature activation | +5.53e+00 |
Pos logit contributions | |
auc | +0.313 |
line | +0.285 |
lights | +0.282 |
bars | +0.280 |
inter | +0.279 |
Neg logit contributions | |
Amen | -0.275 |
lder | -0.270 |
", | -0.248 |
oc | -0.244 |
Frank | -0.237 |
Token | delicious |
Token ID | 12625 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
auc | +0.187 |
bars | +0.169 |
line | +0.167 |
ays | +0.166 |
ains | +0.164 |
Neg logit contributions | |
Amen | -0.148 |
lder | -0.123 |
defeat | -0.113 |
oc | -0.113 |
Frank | -0.107 |
Token | !" |
Token ID | 2474 |
Feature activation | -2.67e+00 |
Pos logit contributions | |
Amen | +0.026 |
lder | +0.019 |
defeat | +0.018 |
Frank | +0.017 |
oc | +0.017 |
Neg logit contributions | |
auc | -0.042 |
ays | -0.038 |
line | -0.038 |
ains | -0.038 |
bars | -0.037 |
Token | the |
Token ID | 262 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
Amen | +0.108 |
lder | +0.091 |
defeat | +0.090 |
chat | +0.088 |
Frank | +0.088 |
Neg logit contributions | |
auc | -0.055 |
ains | -0.048 |
ays | -0.046 |
bars | -0.046 |
line | -0.044 |
Token | man |
Token ID | 582 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
auc | +0.100 |
ays | +0.091 |
ains | +0.089 |
bars | +0.088 |
line | +0.087 |
Neg logit contributions | |
Amen | -0.072 |
lder | -0.057 |
defeat | -0.052 |
oc | -0.050 |
Frank | -0.050 |
Token | . |
Token ID | 13 |
Feature activation | +7.55e-01 |
Pos logit contributions | |
Amen | +0.048 |
lder | +0.040 |
defeat | +0.038 |
read | +0.037 |
chat | +0.036 |
Neg logit contributions | |
auc | -0.067 |
bars | -0.064 |
ays | -0.062 |
inter | -0.062 |
ains | -0.061 |
Token | She |
Token ID | 1375 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
auc | +0.024 |
ains | +0.022 |
ash | +0.022 |
bars | +0.021 |
lights | +0.021 |
Neg logit contributions | |
Amen | -0.021 |
lder | -0.017 |
Mic | -0.015 |
Frank | -0.015 |
oc | -0.015 |
Token | said |
Token ID | 531 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
Amen | +0.067 |
lder | +0.056 |
Frank | +0.053 |
defeat | +0.053 |
oc | +0.052 |
Neg logit contributions | |
auc | -0.041 |
bars | -0.034 |
ays | -0.034 |
line | -0.034 |
ains | -0.034 |
Token | , |
Token ID | 11 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
Amen | +0.053 |
lder | +0.044 |
defeat | +0.040 |
Frank | +0.039 |
oc | +0.038 |
Neg logit contributions | |
auc | -0.064 |
bars | -0.058 |
ays | -0.057 |
ains | -0.057 |
line | -0.057 |
Token | " |
Token ID | 366 |
Feature activation | +9.58e-01 |
Pos logit contributions | |
auc | +0.036 |
ays | +0.031 |
bars | +0.030 |
ains | +0.029 |
ash | +0.029 |
Neg logit contributions | |
Amen | -0.057 |
lder | -0.049 |
defeat | -0.045 |
Frank | -0.045 |
oc | -0.045 |
Token | Those |
Token ID | 9627 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
auc | +0.027 |
lights | +0.025 |
ays | +0.024 |
ains | +0.024 |
tubes | +0.024 |
Neg logit contributions | |
Amen | -0.031 |
Frank | -0.026 |
lder | -0.026 |
Mic | -0.025 |
Offic | -0.024 |
Token | are |
Token ID | 389 |
Feature activation | +3.04e+00 |
Pos logit contributions | |
auc | +0.285 |
line | +0.251 |
bars | +0.251 |
lights | +0.247 |
ays | +0.245 |
Neg logit contributions | |
Amen | -0.311 |
lder | -0.294 |
defeat | -0.271 |
chat | -0.268 |
oc | -0.260 |
Token | not |
Token ID | 407 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
auc | +0.101 |
line | +0.095 |
ays | +0.094 |
bars | +0.094 |
inter | +0.089 |
Neg logit contributions | |
Amen | -0.076 |
defeat | -0.066 |
lder | -0.066 |
Frank | -0.064 |
Chuck | -0.062 |
Token | grapes |
Token ID | 36656 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
auc | +0.056 |
ays | +0.051 |
bars | +0.050 |
line | +0.049 |
lights | +0.049 |
Neg logit contributions | |
Amen | -0.047 |
lder | -0.042 |
defeat | -0.040 |
Frank | -0.039 |
read | -0.037 |
Token | , |
Token ID | 11 |
Feature activation | +2.95e+00 |
Pos logit contributions | |
auc | +0.069 |
ays | +0.064 |
lights | +0.063 |
idget | +0.063 |
bars | +0.063 |
Neg logit contributions | |
Amen | -0.043 |
lder | -0.042 |
defeat | -0.040 |
read | -0.038 |
", | -0.038 |
Token | those |
Token ID | 883 |
Feature activation | +9.30e+00 |
Pos logit contributions | |
auc | +0.086 |
ays | +0.082 |
bars | +0.081 |
idget | +0.080 |
ash | +0.077 |
Neg logit contributions | |
Amen | -0.081 |
lder | -0.074 |
defeat | -0.070 |
oc | -0.067 |
Frank | -0.065 |
Token | stars |
Token ID | 5788 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
auc | +0.042 |
ays | +0.036 |
bars | +0.035 |
line | +0.035 |
inter | +0.034 |
Neg logit contributions | |
Amen | -0.060 |
lder | -0.050 |
defeat | -0.047 |
Frank | -0.046 |
chat | -0.046 |
Token | . |
Token ID | 13 |
Feature activation | -3.82e-01 |
Pos logit contributions | |
auc | +0.043 |
ains | +0.036 |
ays | +0.036 |
inter | +0.035 |
bars | +0.035 |
Neg logit contributions | |
Amen | -0.058 |
lder | -0.048 |
defeat | -0.047 |
Frank | -0.045 |
oc | -0.045 |
Token |
|
Token ID | 198 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
Amen | +0.013 |
lder | +0.011 |
Frank | +0.010 |
defeat | +0.010 |
chat | +0.010 |
Neg logit contributions | |
auc | -0.011 |
ains | -0.009 |
bars | -0.009 |
line | -0.009 |
ays | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
Amen | +0.077 |
lder | +0.064 |
defeat | +0.061 |
oc | +0.060 |
Frank | +0.059 |
Neg logit contributions | |
auc | -0.055 |
ays | -0.048 |
line | -0.047 |
ains | -0.047 |
bars | -0.046 |
Token | " |
Token ID | 1 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
Amen | +0.052 |
oc | +0.040 |
lder | +0.039 |
defeat | +0.039 |
Mic | +0.038 |
Neg logit contributions | |
auc | -0.054 |
line | -0.050 |
ays | -0.048 |
ains | -0.047 |
bars | -0.046 |
Token | Those |
Token ID | 9627 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
auc | +0.030 |
ays | +0.027 |
ains | +0.027 |
inter | +0.026 |
ash | +0.026 |
Neg logit contributions | |
Amen | -0.035 |
lder | -0.028 |
Frank | -0.028 |
Charles | -0.026 |
Mic | -0.026 |
Token | are |
Token ID | 389 |
Feature activation | +2.94e+00 |
Pos logit contributions | |
auc | +0.227 |
ays | +0.188 |
line | +0.186 |
bars | +0.185 |
inter | +0.179 |
Neg logit contributions | |
Amen | -0.401 |
lder | -0.348 |
defeat | -0.328 |
oc | -0.322 |
Frank | -0.319 |
Token | not |
Token ID | 407 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
auc | +0.086 |
ays | +0.076 |
bars | +0.075 |
line | +0.075 |
inter | +0.074 |
Neg logit contributions | |
Amen | -0.094 |
lder | -0.077 |
defeat | -0.074 |
Frank | -0.073 |
chat | -0.070 |
Token | fireworks |
Token ID | 26056 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
auc | +0.053 |
ays | +0.046 |
bars | +0.046 |
line | +0.045 |
inter | +0.044 |
Neg logit contributions | |
Amen | -0.059 |
lder | -0.049 |
defeat | -0.046 |
Frank | -0.046 |
Charles | -0.043 |
Token | , |
Token ID | 11 |
Feature activation | +2.80e+00 |
Pos logit contributions | |
auc | +0.057 |
ays | +0.051 |
line | +0.050 |
bars | +0.050 |
ains | +0.049 |
Neg logit contributions | |
Amen | -0.044 |
lder | -0.037 |
defeat | -0.035 |
oc | -0.033 |
", | -0.032 |
Token | Anna |
Token ID | 11735 |
Feature activation | -4.33e-01 |
Pos logit contributions | |
auc | +0.082 |
ays | +0.071 |
bars | +0.070 |
line | +0.069 |
inter | +0.068 |
Neg logit contributions | |
Amen | -0.092 |
lder | -0.076 |
defeat | -0.070 |
Frank | -0.069 |
oc | -0.068 |
Token | smiled |
Token ID | 13541 |
Feature activation | -8.27e-01 |
Pos logit contributions | |
auc | +0.044 |
ays | +0.038 |
line | +0.038 |
bars | +0.038 |
ains | +0.038 |
Neg logit contributions | |
Amen | -0.049 |
lder | -0.040 |
defeat | -0.037 |
oc | -0.037 |
Frank | -0.037 |
Token | and |
Token ID | 290 |
Feature activation | -8.28e-01 |
Pos logit contributions | |
Amen | +0.027 |
lder | +0.022 |
defeat | +0.020 |
Frank | +0.020 |
oc | +0.020 |
Neg logit contributions | |
auc | -0.024 |
ays | -0.021 |
line | -0.021 |
bars | -0.021 |
ains | -0.021 |
Token | said |
Token ID | 531 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
Amen | +0.020 |
lder | +0.016 |
chat | +0.015 |
Frank | +0.015 |
read | +0.014 |
Neg logit contributions | |
auc | -0.031 |
ains | -0.028 |
ays | -0.028 |
bars | -0.027 |
inter | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | +8.11e-01 |
Pos logit contributions | |
Amen | +0.058 |
lder | +0.047 |
defeat | +0.043 |
Frank | +0.042 |
oc | +0.042 |
Neg logit contributions | |
auc | -0.069 |
ays | -0.062 |
ains | -0.061 |
bars | -0.061 |
line | -0.061 |
Token | " |
Token ID | 366 |
Feature activation | +9.25e-01 |
Pos logit contributions | |
auc | +0.018 |
ays | +0.015 |
bars | +0.014 |
ains | +0.014 |
line | +0.014 |
Neg logit contributions | |
Amen | -0.033 |
lder | -0.028 |
Frank | -0.026 |
defeat | -0.026 |
oc | -0.026 |
Token | Those |
Token ID | 9627 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
auc | +0.026 |
ains | +0.023 |
inter | +0.023 |
ays | +0.023 |
lights | +0.023 |
Neg logit contributions | |
Amen | -0.030 |
Frank | -0.025 |
lder | -0.025 |
oc | -0.022 |
Charles | -0.022 |
Token | are |
Token ID | 389 |
Feature activation | +3.28e+00 |
Pos logit contributions | |
auc | +0.255 |
line | +0.224 |
ays | +0.221 |
idget | +0.221 |
bars | +0.211 |
Neg logit contributions | |
Amen | -0.346 |
lder | -0.313 |
defeat | -0.299 |
chat | -0.293 |
oc | -0.283 |
Token | stars |
Token ID | 5788 |
Feature activation | +2.27e+00 |
Pos logit contributions | |
auc | +0.102 |
inter | +0.095 |
ays | +0.093 |
line | +0.093 |
bars | +0.089 |
Neg logit contributions | |
Amen | -0.089 |
defeat | -0.073 |
lder | -0.073 |
read | -0.072 |
Frank | -0.072 |
Token | , |
Token ID | 11 |
Feature activation | +3.07e+00 |
Pos logit contributions | |
auc | +0.075 |
bars | +0.066 |
ays | +0.066 |
inter | +0.066 |
line | +0.064 |
Neg logit contributions | |
Amen | -0.062 |
lder | -0.051 |
defeat | -0.047 |
chat | -0.045 |
oc | -0.045 |
Token | sweet |
Token ID | 6029 |
Feature activation | +3.36e-01 |
Pos logit contributions | |
auc | +0.085 |
ays | +0.076 |
bars | +0.075 |
line | +0.074 |
ains | +0.072 |
Neg logit contributions | |
Amen | -0.103 |
lder | -0.084 |
Frank | -0.080 |
oc | -0.079 |
defeat | -0.077 |
Token | heart |
Token ID | 11499 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
auc | +0.009 |
ays | +0.008 |
line | +0.008 |
bars | +0.008 |
ains | +0.008 |
Neg logit contributions | |
Amen | -0.012 |
lder | -0.010 |
defeat | -0.009 |
Frank | -0.009 |
oc | -0.009 |
Token | a |
Token ID | 257 |
Feature activation | -3.25e+00 |
Pos logit contributions | |
Amen | +0.005 |
lder | +0.004 |
defeat | +0.004 |
chat | +0.004 |
oc | +0.004 |
Neg logit contributions | |
auc | -0.004 |
ays | -0.003 |
line | -0.003 |
ains | -0.003 |
bars | -0.003 |
Token | jar |
Token ID | 17379 |
Feature activation | +9.93e-01 |
Pos logit contributions | |
Amen | +0.147 |
lder | +0.124 |
Frank | +0.119 |
Charles | +0.118 |
oc | +0.115 |
Neg logit contributions | |
auc | -0.061 |
tubes | -0.050 |
inter | -0.047 |
tub | -0.046 |
line | -0.045 |
Token | and |
Token ID | 290 |
Feature activation | -6.84e-01 |
Pos logit contributions | |
ains | +0.026 |
auc | +0.026 |
bars | +0.025 |
ays | +0.025 |
inter | +0.024 |
Neg logit contributions | |
Amen | -0.028 |
chat | -0.024 |
defeat | -0.024 |
read | -0.023 |
lder | -0.023 |
Token | made |
Token ID | 925 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
Amen | +0.023 |
lder | +0.019 |
defeat | +0.018 |
chat | +0.018 |
oc | +0.018 |
Neg logit contributions | |
auc | -0.019 |
ays | -0.016 |
line | -0.016 |
ains | -0.016 |
bars | -0.016 |
Token | a |
Token ID | 257 |
Feature activation | -5.57e+00 |
Pos logit contributions | |
auc | +0.038 |
ays | +0.033 |
line | +0.033 |
bars | +0.033 |
ains | +0.033 |
Neg logit contributions | |
Amen | -0.045 |
lder | -0.038 |
defeat | -0.035 |
Frank | -0.035 |
oc | -0.035 |
Token | cozy |
Token ID | 37438 |
Feature activation | -9.13e+00 |
Pos logit contributions | |
Amen | +0.226 |
lder | +0.189 |
Frank | +0.189 |
Charles | +0.182 |
Mic | +0.178 |
Neg logit contributions | |
auc | -0.125 |
tubes | -0.111 |
tub | -0.106 |
line | -0.099 |
inter | -0.099 |
Token | home |
Token ID | 1363 |
Feature activation | -3.56e+00 |
Pos logit contributions | |
Amen | +0.334 |
lder | +0.296 |
Charles | +0.278 |
Frank | +0.277 |
Mic | +0.277 |
Neg logit contributions | |
auc | -0.218 |
tub | -0.195 |
tubes | -0.190 |
inter | -0.184 |
line | -0.184 |
Token | for |
Token ID | 329 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
Amen | +0.123 |
lder | +0.100 |
defeat | +0.099 |
chat | +0.096 |
oc | +0.096 |
Neg logit contributions | |
auc | -0.091 |
ays | -0.083 |
ains | -0.082 |
bars | -0.082 |
inter | -0.078 |
Token | the |
Token ID | 262 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
auc | +0.027 |
bars | +0.023 |
ays | +0.023 |
line | +0.022 |
ains | +0.022 |
Neg logit contributions | |
Amen | -0.036 |
lder | -0.031 |
defeat | -0.030 |
chat | -0.029 |
Frank | -0.029 |
Token | worm |
Token ID | 23849 |
Feature activation | -7.53e-01 |
Pos logit contributions | |
auc | +0.056 |
line | +0.049 |
tubes | +0.048 |
inter | +0.047 |
ains | +0.046 |
Neg logit contributions | |
Amen | -0.078 |
lder | -0.060 |
oc | -0.059 |
defeat | -0.058 |
Frank | -0.058 |
Token | . |
Token ID | 13 |
Feature activation | +6.50e-01 |
Pos logit contributions | |
Amen | +0.030 |
lder | +0.026 |
Frank | +0.024 |
defeat | +0.024 |
oc | +0.024 |
Neg logit contributions | |
auc | -0.017 |
line | -0.014 |
ays | -0.014 |
bars | -0.014 |
ains | -0.014 |
Token | eat |
Token ID | 4483 |
Feature activation | +7.39e-01 |
Pos logit contributions | |
Amen | +0.120 |
lder | +0.100 |
defeat | +0.093 |
Frank | +0.092 |
oc | +0.092 |
Neg logit contributions | |
auc | -0.100 |
ays | -0.087 |
line | -0.086 |
bars | -0.086 |
ains | -0.086 |
Token | , |
Token ID | 11 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
auc | +0.020 |
bars | +0.018 |
ays | +0.018 |
ains | +0.017 |
line | +0.017 |
Neg logit contributions | |
Amen | -0.025 |
lder | -0.021 |
defeat | -0.020 |
oc | -0.020 |
Frank | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | -6.70e-01 |
Pos logit contributions | |
auc | +0.031 |
ays | +0.027 |
ains | +0.026 |
bars | +0.026 |
ash | +0.026 |
Neg logit contributions | |
Amen | -0.037 |
lder | -0.031 |
defeat | -0.030 |
chat | -0.030 |
Frank | -0.030 |
Token | made |
Token ID | 925 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
Amen | +0.022 |
lder | +0.019 |
defeat | +0.019 |
chat | +0.019 |
Frank | +0.018 |
Neg logit contributions | |
auc | -0.019 |
ays | -0.016 |
ains | -0.016 |
bars | -0.016 |
ash | -0.015 |
Token | a |
Token ID | 257 |
Feature activation | -5.52e+00 |
Pos logit contributions | |
auc | +0.038 |
ays | +0.033 |
line | +0.032 |
ains | +0.032 |
bars | +0.032 |
Neg logit contributions | |
Amen | -0.051 |
lder | -0.043 |
defeat | -0.040 |
oc | -0.040 |
Frank | -0.040 |
Token | cozy |
Token ID | 37438 |
Feature activation | -8.87e+00 |
Pos logit contributions | |
Amen | +0.209 |
lder | +0.191 |
Frank | +0.180 |
Charles | +0.176 |
Mic | +0.172 |
Neg logit contributions | |
auc | -0.101 |
tub | -0.094 |
dripping | -0.094 |
shallow | -0.094 |
tubes | -0.094 |
Token | bed |
Token ID | 3996 |
Feature activation | -3.39e+00 |
Pos logit contributions | |
Amen | +0.281 |
lder | +0.261 |
Charles | +0.243 |
Mic | +0.238 |
Frank | +0.236 |
Neg logit contributions | |
auc | -0.210 |
tub | -0.206 |
lights | -0.198 |
line | -0.195 |
inter | -0.194 |
Token | for |
Token ID | 329 |
Feature activation | +8.19e-01 |
Pos logit contributions | |
Amen | +0.117 |
lder | +0.097 |
defeat | +0.095 |
chat | +0.093 |
oc | +0.091 |
Neg logit contributions | |
auc | -0.088 |
ays | -0.081 |
ains | -0.078 |
bars | -0.078 |
tubes | -0.075 |
Token | it |
Token ID | 340 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
auc | +0.023 |
bars | +0.020 |
ays | +0.020 |
ains | +0.019 |
inter | +0.019 |
Neg logit contributions | |
Amen | -0.027 |
lder | -0.023 |
defeat | -0.023 |
Frank | -0.023 |
chat | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | -4.55e+00 |
Pos logit contributions | |
auc | +0.050 |
ays | +0.044 |
bars | +0.043 |
ains | +0.043 |
line | +0.042 |
Neg logit contributions | |
Amen | -0.051 |
lder | -0.044 |
defeat | -0.044 |
chat | -0.043 |
read | -0.042 |
Token | rest |
Token ID | 1334 |
Feature activation | +2.21e+00 |
Pos logit contributions | |
Amen | +0.138 |
lder | +0.113 |
Frank | +0.103 |
oc | +0.103 |
defeat | +0.102 |
Neg logit contributions | |
auc | -0.140 |
line | -0.125 |
ains | -0.121 |
ays | -0.120 |
bars | -0.120 |
Token | Karen |
Token ID | 18678 |
Feature activation | +7.34e-01 |
Pos logit contributions | |
auc | +0.039 |
ays | +0.034 |
bars | +0.034 |
line | +0.033 |
ains | +0.032 |
Neg logit contributions | |
Amen | -0.072 |
lder | -0.064 |
read | -0.059 |
chat | -0.058 |
Frank | -0.058 |
Token | . |
Token ID | 13 |
Feature activation | +3.13e-01 |
Pos logit contributions | |
auc | +0.016 |
line | +0.014 |
bars | +0.014 |
ays | +0.014 |
ains | +0.013 |
Neg logit contributions | |
Amen | -0.029 |
lder | -0.025 |
defeat | -0.023 |
Frank | -0.023 |
oc | -0.023 |
Token | She |
Token ID | 1375 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
auc | +0.009 |
ays | +0.008 |
ains | +0.008 |
line | +0.008 |
bars | +0.008 |
Neg logit contributions | |
Amen | -0.010 |
lder | -0.009 |
Frank | -0.008 |
chat | -0.008 |
oc | -0.008 |
Token | wore |
Token ID | 12408 |
Feature activation | +3.51e-01 |
Pos logit contributions | |
Amen | +0.059 |
lder | +0.050 |
oc | +0.046 |
defeat | +0.046 |
Frank | +0.046 |
Neg logit contributions | |
auc | -0.046 |
ays | -0.040 |
line | -0.039 |
bars | -0.039 |
ains | -0.039 |
Token | a |
Token ID | 257 |
Feature activation | -5.27e+00 |
Pos logit contributions | |
auc | +0.010 |
ays | +0.009 |
line | +0.009 |
bars | +0.009 |
ains | +0.009 |
Neg logit contributions | |
Amen | -0.011 |
lder | -0.009 |
defeat | -0.009 |
oc | -0.009 |
Frank | -0.009 |
Token | cozy |
Token ID | 37438 |
Feature activation | -8.82e+00 |
Pos logit contributions | |
Amen | +0.204 |
lder | +0.174 |
defeat | +0.170 |
oc | +0.168 |
Frank | +0.164 |
Neg logit contributions | |
auc | -0.115 |
tubes | -0.096 |
line | -0.093 |
inter | -0.093 |
bars | -0.091 |
Token | scarf |
Token ID | 36953 |
Feature activation | -3.63e+00 |
Pos logit contributions | |
Amen | +0.300 |
lder | +0.260 |
defeat | +0.250 |
Frank | +0.245 |
oc | +0.238 |
Neg logit contributions | |
auc | -0.215 |
tub | -0.200 |
tubes | -0.197 |
lights | -0.196 |
inter | -0.189 |
Token | every |
Token ID | 790 |
Feature activation | +2.95e+00 |
Pos logit contributions | |
Amen | +0.132 |
lder | +0.112 |
defeat | +0.105 |
Frank | +0.104 |
oc | +0.104 |
Neg logit contributions | |
auc | -0.091 |
ays | -0.077 |
line | -0.077 |
bars | -0.077 |
ains | -0.076 |
Token | day |
Token ID | 1110 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
auc | +0.052 |
bars | +0.043 |
line | +0.042 |
inter | +0.041 |
ays | +0.040 |
Neg logit contributions | |
Amen | -0.126 |
lder | -0.110 |
defeat | -0.107 |
Frank | -0.106 |
oc | -0.105 |
Token | , |
Token ID | 11 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
Amen | +0.050 |
read | +0.042 |
lder | +0.042 |
chat | +0.040 |
oc | +0.038 |
Neg logit contributions | |
auc | -0.057 |
ains | -0.051 |
ays | -0.049 |
lights | -0.047 |
ours | -0.047 |
Token | no |
Token ID | 645 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
auc | +0.036 |
ays | +0.033 |
ains | +0.031 |
ash | +0.030 |
lights | +0.029 |
Neg logit contributions | |
Amen | -0.035 |
read | -0.030 |
chat | -0.030 |
lder | -0.029 |
Frank | -0.028 |
Token | the |
Token ID | 262 |
Feature activation | +2.05e+00 |
Pos logit contributions | |
auc | +0.082 |
inter | +0.071 |
line | +0.069 |
bars | +0.069 |
ains | +0.069 |
Neg logit contributions | |
Amen | -0.087 |
defeat | -0.074 |
chat | -0.073 |
Frank | -0.072 |
lder | -0.071 |
Token | books |
Token ID | 3835 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
auc | +0.054 |
line | +0.046 |
bars | +0.046 |
ays | +0.046 |
ains | +0.046 |
Neg logit contributions | |
Amen | -0.072 |
lder | -0.060 |
defeat | -0.056 |
Frank | -0.056 |
oc | -0.056 |
Token | and |
Token ID | 290 |
Feature activation | +2.87e-01 |
Pos logit contributions | |
auc | +0.046 |
ains | +0.041 |
inter | +0.040 |
tubes | +0.040 |
ays | +0.040 |
Neg logit contributions | |
Amen | -0.039 |
lder | -0.032 |
defeat | -0.030 |
oc | -0.029 |
chat | -0.029 |
Token | made |
Token ID | 925 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
auc | +0.009 |
ains | +0.008 |
inter | +0.007 |
ays | +0.007 |
ash | +0.007 |
Neg logit contributions | |
Amen | -0.009 |
read | -0.008 |
chat | -0.007 |
defeat | -0.007 |
lder | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | -5.82e+00 |
Pos logit contributions | |
auc | +0.031 |
line | +0.027 |
ays | +0.027 |
bars | +0.027 |
ains | +0.026 |
Neg logit contributions | |
Amen | -0.038 |
lder | -0.032 |
defeat | -0.031 |
Frank | -0.030 |
oc | -0.030 |
Token | cozy |
Token ID | 37438 |
Feature activation | -8.77e+00 |
Pos logit contributions | |
Amen | +0.212 |
lder | +0.186 |
Frank | +0.178 |
Mic | +0.172 |
Charles | +0.162 |
Neg logit contributions | |
auc | -0.137 |
tubes | -0.124 |
line | -0.122 |
tub | -0.116 |
lights | -0.116 |
Token | fort |
Token ID | 6285 |
Feature activation | -5.18e+00 |
Pos logit contributions | |
Amen | +0.284 |
lder | +0.256 |
Mic | +0.242 |
Frank | +0.240 |
Chuck | +0.230 |
Neg logit contributions | |
auc | -0.222 |
lights | -0.212 |
tub | -0.204 |
line | -0.202 |
tubes | -0.197 |
Token | . |
Token ID | 13 |
Feature activation | +4.70e-01 |
Pos logit contributions | |
Amen | +0.184 |
lder | +0.154 |
defeat | +0.150 |
oc | +0.145 |
chat | +0.144 |
Neg logit contributions | |
auc | -0.131 |
ays | -0.114 |
ains | -0.112 |
bars | -0.111 |
line | -0.109 |
Token | They |
Token ID | 1119 |
Feature activation | -7.15e-02 |
Pos logit contributions | |
auc | +0.013 |
bars | +0.011 |
ains | +0.011 |
line | +0.011 |
ays | +0.011 |
Neg logit contributions | |
Amen | -0.016 |
lder | -0.013 |
oc | -0.013 |
Frank | -0.013 |
Charles | -0.012 |
Token | crawled |
Token ID | 45668 |
Feature activation | +3.71e-03 |
Pos logit contributions | |
Amen | +0.003 |
lder | +0.002 |
defeat | +0.002 |
Frank | +0.002 |
oc | +0.002 |
Neg logit contributions | |
auc | -0.002 |
line | -0.002 |
bars | -0.002 |
ains | -0.002 |
ays | -0.002 |
Token | inside |
Token ID | 2641 |
Feature activation | +2.89e-02 |
Pos logit contributions | |
auc | +0.000 |
ays | +0.000 |
line | +0.000 |
bars | +0.000 |
ains | +0.000 |
Neg logit contributions | |
Amen | -0.000 |
lder | -0.000 |
defeat | -0.000 |
oc | -0.000 |
Frank | -0.000 |
Token | and |
Token ID | 290 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
auc | +0.019 |
ains | +0.018 |
ays | +0.018 |
idget | +0.018 |
line | +0.017 |
Neg logit contributions | |
Amen | -0.012 |
defeat | -0.011 |
oc | -0.010 |
lder | -0.010 |
Frank | -0.010 |
Token | nails |
Token ID | 23361 |
Feature activation | +2.99e+00 |
Pos logit contributions | |
Amen | +0.004 |
lder | +0.003 |
defeat | +0.003 |
oc | +0.003 |
Frank | +0.003 |
Neg logit contributions | |
auc | -0.003 |
ays | -0.002 |
line | -0.002 |
ains | -0.002 |
bars | -0.002 |
Token | to |
Token ID | 284 |
Feature activation | -3.86e+00 |
Pos logit contributions | |
auc | +0.107 |
ains | +0.099 |
ays | +0.097 |
idget | +0.095 |
ash | +0.095 |
Neg logit contributions | |
Amen | -0.056 |
defeat | -0.055 |
Frank | -0.046 |
oc | -0.046 |
chat | -0.046 |
Token | make |
Token ID | 787 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
Amen | +0.129 |
lder | +0.107 |
defeat | +0.102 |
oc | +0.100 |
Frank | +0.100 |
Neg logit contributions | |
auc | -0.108 |
ays | -0.095 |
bars | -0.093 |
line | -0.093 |
ains | -0.092 |
Token | a |
Token ID | 257 |
Feature activation | -5.65e+00 |
Pos logit contributions | |
auc | +0.051 |
ays | +0.043 |
line | +0.043 |
ains | +0.043 |
tubes | +0.042 |
Neg logit contributions | |
Amen | -0.083 |
lder | -0.070 |
Frank | -0.064 |
Mic | -0.063 |
oc | -0.063 |
Token | cozy |
Token ID | 37438 |
Feature activation | -8.71e+00 |
Pos logit contributions | |
Amen | +0.207 |
lder | +0.186 |
Frank | +0.173 |
Charles | +0.165 |
Mic | +0.163 |
Neg logit contributions | |
auc | -0.132 |
tubes | -0.116 |
line | -0.113 |
inter | -0.110 |
tub | -0.109 |
Token | place |
Token ID | 1295 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
Amen | +0.282 |
lder | +0.252 |
Frank | +0.228 |
Mic | +0.226 |
Charles | +0.224 |
Neg logit contributions | |
auc | -0.236 |
line | -0.217 |
tub | -0.211 |
inter | -0.210 |
lights | -0.209 |
Token | for |
Token ID | 329 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
Amen | +0.031 |
defeat | +0.026 |
chat | +0.024 |
oc | +0.024 |
lder | +0.024 |
Neg logit contributions | |
auc | -0.029 |
ains | -0.027 |
ays | -0.027 |
bars | -0.026 |
line | -0.026 |
Token | the |
Token ID | 262 |
Feature activation | +2.39e+00 |
Pos logit contributions | |
auc | +0.023 |
ays | +0.019 |
line | +0.019 |
bars | +0.019 |
ains | +0.019 |
Neg logit contributions | |
Amen | -0.041 |
lder | -0.035 |
defeat | -0.033 |
Frank | -0.033 |
oc | -0.033 |
Token | bird |
Token ID | 6512 |
Feature activation | -7.88e-01 |
Pos logit contributions | |
auc | +0.070 |
line | +0.065 |
tubes | +0.060 |
ains | +0.060 |
bars | +0.060 |
Neg logit contributions | |
Amen | -0.080 |
lder | -0.062 |
oc | -0.061 |
Frank | -0.059 |
Charles | -0.058 |
Token | to |
Token ID | 284 |
Feature activation | -4.22e+00 |
Pos logit contributions | |
Amen | +0.029 |
lder | +0.025 |
defeat | +0.023 |
Frank | +0.023 |
oc | +0.023 |
Neg logit contributions | |
auc | -0.020 |
line | -0.017 |
ays | -0.017 |
bars | -0.016 |
ains | -0.016 |
Token | wound |
Token ID | 11699 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
Amen | +0.076 |
lder | +0.064 |
Frank | +0.061 |
oc | +0.059 |
defeat | +0.058 |
Neg logit contributions | |
auc | -0.060 |
line | -0.053 |
bars | -0.052 |
tubes | -0.052 |
inter | -0.051 |
Token | and |
Token ID | 290 |
Feature activation | -3.40e-01 |
Pos logit contributions | |
auc | +0.034 |
ays | +0.030 |
ains | +0.030 |
bars | +0.029 |
line | +0.029 |
Neg logit contributions | |
Amen | -0.034 |
lder | -0.028 |
oc | -0.027 |
defeat | -0.026 |
Frank | -0.025 |
Token | makes |
Token ID | 1838 |
Feature activation | +2.99e+00 |
Pos logit contributions | |
Amen | +0.011 |
lder | +0.009 |
defeat | +0.009 |
oc | +0.008 |
Frank | +0.008 |
Neg logit contributions | |
auc | -0.010 |
line | -0.009 |
ays | -0.009 |
ains | -0.009 |
bars | -0.009 |
Token | him |
Token ID | 683 |
Feature activation | -3.56e+00 |
Pos logit contributions | |
auc | +0.074 |
ays | +0.063 |
line | +0.062 |
bars | +0.061 |
tubes | +0.060 |
Neg logit contributions | |
Amen | -0.112 |
lder | -0.095 |
oc | -0.089 |
Frank | -0.088 |
defeat | -0.086 |
Token | a |
Token ID | 257 |
Feature activation | -4.92e+00 |
Pos logit contributions | |
Amen | +0.126 |
lder | +0.114 |
Frank | +0.109 |
Mic | +0.109 |
Offic | +0.103 |
Neg logit contributions | |
auc | -0.089 |
line | -0.077 |
bars | -0.075 |
inter | -0.075 |
tubes | -0.074 |
Token | cozy |
Token ID | 37438 |
Feature activation | -8.62e+00 |
Pos logit contributions | |
Amen | +0.167 |
lder | +0.157 |
Frank | +0.148 |
Mic | +0.140 |
Charles | +0.137 |
Neg logit contributions | |
auc | -0.121 |
soap | -0.107 |
tub | -0.105 |
dripping | -0.103 |
inter | -0.100 |
Token | bed |
Token ID | 3996 |
Feature activation | -3.39e+00 |
Pos logit contributions | |
Amen | +0.277 |
lder | +0.254 |
Frank | +0.236 |
Mic | +0.233 |
Charles | +0.229 |
Neg logit contributions | |
auc | -0.225 |
tub | -0.202 |
line | -0.201 |
inter | -0.198 |
lights | -0.195 |
Token | . |
Token ID | 13 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
Amen | +0.120 |
lder | +0.098 |
defeat | +0.095 |
oc | +0.093 |
chat | +0.091 |
Neg logit contributions | |
auc | -0.088 |
ays | -0.078 |
ains | -0.077 |
bars | -0.075 |
inter | -0.073 |
Token | She |
Token ID | 1375 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
auc | +0.032 |
ains | +0.027 |
ays | +0.027 |
line | +0.026 |
bars | +0.026 |
Neg logit contributions | |
Amen | -0.044 |
lder | -0.037 |
Frank | -0.035 |
defeat | -0.035 |
oc | -0.035 |
Token | says |
Token ID | 1139 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
Amen | +0.055 |
lder | +0.046 |
defeat | +0.043 |
Frank | +0.043 |
oc | +0.042 |
Neg logit contributions | |
auc | -0.042 |
ays | -0.036 |
line | -0.036 |
bars | -0.036 |
ains | -0.036 |
Token | , |
Token ID | 11 |
Feature activation | +1.86e+00 |
Pos logit contributions | |
Amen | +0.051 |
lder | +0.042 |
oc | +0.041 |
Frank | +0.041 |
Mic | +0.040 |
Neg logit contributions | |
auc | -0.039 |
line | -0.033 |
bars | -0.032 |
ays | -0.032 |
ours | -0.031 |
Token | yes |
Token ID | 3763 |
Feature activation | -3.90e+00 |
Pos logit contributions | |
Amen | +0.058 |
lder | +0.046 |
defeat | +0.041 |
Frank | +0.041 |
oc | +0.039 |
Neg logit contributions | |
auc | -0.082 |
ays | -0.074 |
ains | -0.073 |
bars | -0.073 |
line | -0.072 |
Token | and |
Token ID | 290 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
Amen | +0.142 |
lder | +0.113 |
Frank | +0.109 |
Charles | +0.106 |
defeat | +0.106 |
Neg logit contributions | |
auc | -0.101 |
line | -0.091 |
bars | -0.087 |
inter | -0.084 |
ays | -0.084 |
Token | they |
Token ID | 484 |
Feature activation | +4.47e-01 |
Pos logit contributions | |
Amen | +0.065 |
lder | +0.056 |
defeat | +0.054 |
Frank | +0.053 |
oc | +0.053 |
Neg logit contributions | |
auc | -0.037 |
ains | -0.031 |
ays | -0.031 |
bars | -0.030 |
line | -0.030 |
Token | made |
Token ID | 925 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
auc | +0.011 |
ays | +0.009 |
bars | +0.009 |
line | +0.009 |
ains | +0.009 |
Neg logit contributions | |
Amen | -0.017 |
lder | -0.014 |
defeat | -0.014 |
Frank | -0.013 |
oc | -0.013 |
Token | a |
Token ID | 257 |
Feature activation | -5.71e+00 |
Pos logit contributions | |
auc | +0.033 |
ays | +0.029 |
ours | +0.027 |
tubes | +0.027 |
ains | +0.027 |
Neg logit contributions | |
Amen | -0.059 |
lder | -0.049 |
Mic | -0.047 |
Frank | -0.047 |
Charles | -0.046 |
Token | cozy |
Token ID | 37438 |
Feature activation | -8.58e+00 |
Pos logit contributions | |
Amen | +0.233 |
lder | +0.199 |
Frank | +0.191 |
Charles | +0.186 |
Mic | +0.179 |
Neg logit contributions | |
auc | -0.091 |
tubes | -0.091 |
shallow | -0.088 |
prett | -0.085 |
dripping | -0.085 |
Token | nest |
Token ID | 16343 |
Feature activation | -2.16e+00 |
Pos logit contributions | |
Amen | +0.274 |
lder | +0.264 |
Mic | +0.254 |
Charles | +0.243 |
Offic | +0.238 |
Neg logit contributions | |
auc | -0.190 |
tub | -0.188 |
lights | -0.179 |
inter | -0.176 |
tubes | -0.167 |
Token | for |
Token ID | 329 |
Feature activation | +8.14e-01 |
Pos logit contributions | |
Amen | +0.076 |
lder | +0.063 |
defeat | +0.061 |
oc | +0.060 |
chat | +0.059 |
Neg logit contributions | |
auc | -0.055 |
bars | -0.048 |
ays | -0.048 |
ains | -0.047 |
line | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
auc | +0.016 |
line | +0.014 |
ains | +0.013 |
ays | +0.013 |
bars | +0.013 |
Neg logit contributions | |
Amen | -0.034 |
lder | -0.029 |
oc | -0.027 |
defeat | -0.027 |
Frank | -0.027 |
Token | bird |
Token ID | 6512 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
auc | +0.051 |
line | +0.044 |
tubes | +0.044 |
bars | +0.044 |
ains | +0.044 |
Neg logit contributions | |
Amen | -0.062 |
lder | -0.049 |
oc | -0.048 |
defeat | -0.047 |
Frank | -0.046 |
Token | to |
Token ID | 284 |
Feature activation | -4.30e+00 |
Pos logit contributions | |
Amen | +0.041 |
lder | +0.035 |
Frank | +0.032 |
defeat | +0.032 |
Charles | +0.032 |
Neg logit contributions | |
auc | -0.023 |
line | -0.019 |
ays | -0.019 |
ains | -0.019 |
bars | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | +9.97e-01 |
Pos logit contributions | |
auc | +0.023 |
line | +0.019 |
ays | +0.019 |
bars | +0.019 |
ains | +0.019 |
Neg logit contributions | |
Amen | -0.045 |
lder | -0.038 |
Frank | -0.036 |
oc | -0.036 |
defeat | -0.036 |
Token | floor |
Token ID | 4314 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
auc | +0.020 |
ays | +0.017 |
line | +0.016 |
bars | +0.016 |
ains | +0.016 |
Neg logit contributions | |
Amen | -0.041 |
lder | -0.036 |
defeat | -0.034 |
Frank | -0.034 |
oc | -0.034 |
Token | to |
Token ID | 284 |
Feature activation | -4.46e+00 |
Pos logit contributions | |
auc | +0.033 |
ays | +0.029 |
ains | +0.029 |
bars | +0.028 |
inter | +0.028 |
Neg logit contributions | |
Amen | -0.043 |
lder | -0.037 |
defeat | -0.035 |
oc | -0.034 |
chat | -0.034 |
Token | make |
Token ID | 787 |
Feature activation | +2.21e+00 |
Pos logit contributions | |
Amen | +0.152 |
lder | +0.124 |
defeat | +0.122 |
oc | +0.118 |
chat | +0.114 |
Neg logit contributions | |
auc | -0.124 |
ays | -0.112 |
bars | -0.109 |
ains | -0.107 |
line | -0.104 |
Token | a |
Token ID | 257 |
Feature activation | -5.63e+00 |
Pos logit contributions | |
auc | +0.053 |
ains | +0.044 |
ays | +0.044 |
line | +0.044 |
tubes | +0.044 |
Neg logit contributions | |
Amen | -0.085 |
lder | -0.072 |
Frank | -0.068 |
oc | -0.066 |
Mic | -0.066 |
Token | cozy |
Token ID | 37438 |
Feature activation | -8.56e+00 |
Pos logit contributions | |
Amen | +0.209 |
lder | +0.206 |
Frank | +0.199 |
Charles | +0.181 |
Chuck | +0.181 |
Neg logit contributions | |
auc | -0.114 |
tubes | -0.104 |
tub | -0.103 |
lights | -0.098 |
ains | -0.090 |
Token | space |
Token ID | 2272 |
Feature activation | -3.74e+00 |
Pos logit contributions | |
Amen | +0.293 |
lder | +0.271 |
Frank | +0.256 |
Chuck | +0.246 |
Charles | +0.243 |
Neg logit contributions | |
auc | -0.207 |
lights | -0.195 |
tub | -0.193 |
tubes | -0.183 |
inter | -0.177 |
Token | . |
Token ID | 13 |
Feature activation | -3.34e-01 |
Pos logit contributions | |
Amen | +0.131 |
lder | +0.113 |
Frank | +0.108 |
Mic | +0.105 |
Charles | +0.102 |
Neg logit contributions | |
auc | -0.100 |
line | -0.085 |
lights | -0.084 |
ains | -0.083 |
tubes | -0.082 |
Token | |
Token ID | 220 |
Feature activation | +3.77e+00 |
Pos logit contributions | |
Amen | +0.010 |
lder | +0.008 |
oc | +0.007 |
Frank | +0.007 |
chat | +0.007 |
Neg logit contributions | |
auc | -0.011 |
ains | -0.009 |
bars | -0.009 |
ays | -0.009 |
line | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -2.50e+00 |
Pos logit contributions | |
auc | +0.131 |
bars | +0.116 |
line | +0.114 |
ays | +0.113 |
tubes | +0.113 |
Neg logit contributions | |
Amen | -0.099 |
lder | -0.076 |
Frank | -0.073 |
oc | -0.072 |
defeat | -0.068 |
Token |
|
Token ID | 198 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
Amen | +0.100 |
lder | +0.086 |
defeat | +0.083 |
arrest | +0.078 |
oc | +0.078 |
Neg logit contributions | |
auc | -0.050 |
line | -0.045 |
ays | -0.042 |
ains | -0.041 |
bars | -0.041 |
Token | a |
Token ID | 257 |
Feature activation | -3.35e+00 |
Pos logit contributions | |
auc | +0.178 |
bars | +0.158 |
line | +0.153 |
ays | +0.152 |
ains | +0.151 |
Neg logit contributions | |
Amen | -0.224 |
lder | -0.179 |
Frank | -0.178 |
chat | -0.174 |
defeat | -0.174 |
Token | blanket |
Token ID | 18447 |
Feature activation | -2.78e+00 |
Pos logit contributions | |
Amen | +0.111 |
lder | +0.098 |
oc | +0.086 |
Frank | +0.086 |
defeat | +0.086 |
Neg logit contributions | |
auc | -0.089 |
line | -0.080 |
inter | -0.077 |
tubes | -0.077 |
tub | -0.076 |
Token | , |
Token ID | 11 |
Feature activation | +2.05e+00 |
Pos logit contributions | |
Amen | +0.087 |
lder | +0.072 |
defeat | +0.069 |
Frank | +0.068 |
chat | +0.067 |
Neg logit contributions | |
auc | -0.081 |
ays | -0.071 |
ains | -0.071 |
bars | -0.070 |
line | -0.068 |
Token | enjoying |
Token ID | 13226 |
Feature activation | -8.00e-02 |
Pos logit contributions | |
auc | +0.068 |
ays | +0.060 |
ains | +0.059 |
bars | +0.058 |
ash | +0.058 |
Neg logit contributions | |
Amen | -0.059 |
lder | -0.046 |
chat | -0.045 |
Frank | -0.044 |
read | -0.044 |
Token | a |
Token ID | 257 |
Feature activation | -4.79e+00 |
Pos logit contributions | |
Amen | +0.003 |
lder | +0.003 |
defeat | +0.002 |
Frank | +0.002 |
oc | +0.002 |
Neg logit contributions | |
auc | -0.002 |
bars | -0.002 |
ays | -0.002 |
inter | -0.002 |
line | -0.002 |
Token | cozy |
Token ID | 37438 |
Feature activation | -8.54e+00 |
Pos logit contributions | |
Amen | +0.168 |
lder | +0.155 |
Mic | +0.147 |
Frank | +0.144 |
arrest | +0.142 |
Neg logit contributions | |
auc | -0.096 |
tub | -0.091 |
lights | -0.091 |
line | -0.089 |
shallow | -0.086 |
Token | cup |
Token ID | 6508 |
Feature activation | +6.56e-01 |
Pos logit contributions | |
Amen | +0.276 |
lder | +0.251 |
defeat | +0.232 |
Mic | +0.230 |
Frank | +0.229 |
Neg logit contributions | |
auc | -0.206 |
lights | -0.197 |
tub | -0.194 |
line | -0.189 |
tubes | -0.185 |
Token | of |
Token ID | 286 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
auc | +0.019 |
ains | +0.019 |
ays | +0.018 |
ours | +0.018 |
tubes | +0.017 |
Neg logit contributions | |
Amen | -0.018 |
lder | -0.015 |
chat | -0.015 |
oc | -0.015 |
defeat | -0.015 |
Token | hot |
Token ID | 3024 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
auc | +0.035 |
ays | +0.032 |
ains | +0.031 |
line | +0.031 |
bars | +0.031 |
Neg logit contributions | |
Amen | -0.036 |
lder | -0.028 |
chat | -0.027 |
defeat | -0.027 |
oc | -0.027 |
Token | chocolate |
Token ID | 11311 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
Amen | +0.041 |
lder | +0.032 |
oc | +0.031 |
chat | +0.031 |
Frank | +0.030 |
Neg logit contributions | |
auc | -0.038 |
ains | -0.035 |
bars | -0.034 |
line | -0.034 |
ays | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | +5.99e-01 |
Pos logit contributions | |
Amen | +0.034 |
lder | +0.029 |
oc | +0.027 |
defeat | +0.027 |
Frank | +0.027 |
Neg logit contributions | |
auc | -0.028 |
ays | -0.024 |
ains | -0.024 |
bars | -0.023 |
line | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
auc | +0.016 |
bars | +0.013 |
line | +0.013 |
ays | +0.013 |
ains | +0.013 |
Neg logit contributions | |
Amen | -0.026 |
lder | -0.022 |
defeat | -0.021 |
Frank | -0.021 |
oc | -0.021 |
Token | floor |
Token ID | 4314 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
auc | +0.050 |
ays | +0.044 |
ains | +0.043 |
line | +0.043 |
bars | +0.042 |
Neg logit contributions | |
Amen | -0.060 |
lder | -0.051 |
defeat | -0.047 |
oc | -0.047 |
Frank | -0.046 |
Token | and |
Token ID | 290 |
Feature activation | +9.07e-02 |
Pos logit contributions | |
auc | +0.056 |
ays | +0.050 |
ains | +0.049 |
bars | +0.049 |
inter | +0.049 |
Neg logit contributions | |
Amen | -0.043 |
lder | -0.036 |
defeat | -0.035 |
chat | -0.034 |
oc | -0.033 |
Token | made |
Token ID | 925 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
auc | +0.003 |
ays | +0.002 |
ains | +0.002 |
bars | +0.002 |
line | +0.002 |
Neg logit contributions | |
Amen | -0.003 |
chat | -0.002 |
lder | -0.002 |
read | -0.002 |
defeat | -0.002 |
Token | a |
Token ID | 257 |
Feature activation | -5.39e+00 |
Pos logit contributions | |
auc | +0.039 |
ays | +0.033 |
line | +0.033 |
bars | +0.033 |
ains | +0.033 |
Neg logit contributions | |
Amen | -0.048 |
lder | -0.040 |
defeat | -0.038 |
Frank | -0.038 |
oc | -0.037 |
Token | cozy |
Token ID | 37438 |
Feature activation | -8.51e+00 |
Pos logit contributions | |
Amen | +0.195 |
lder | +0.175 |
Frank | +0.166 |
Mic | +0.161 |
Charles | +0.151 |
Neg logit contributions | |
auc | -0.123 |
tubes | -0.110 |
tub | -0.107 |
lights | -0.105 |
line | -0.105 |
Token | spot |
Token ID | 4136 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
Amen | +0.270 |
lder | +0.239 |
Mic | +0.223 |
Frank | +0.221 |
defeat | +0.211 |
Neg logit contributions | |
auc | -0.228 |
lights | -0.211 |
tub | -0.206 |
line | -0.206 |
inter | -0.202 |
Token | on |
Token ID | 319 |
Feature activation | +4.47e-01 |
Pos logit contributions | |
Amen | +0.069 |
defeat | +0.057 |
lder | +0.056 |
chat | +0.056 |
Frank | +0.054 |
Neg logit contributions | |
auc | -0.049 |
ains | -0.043 |
ays | -0.043 |
bars | -0.043 |
line | -0.040 |
Token | the |
Token ID | 262 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
auc | +0.012 |
inter | +0.010 |
bars | +0.010 |
line | +0.010 |
ays | +0.010 |
Neg logit contributions | |
Amen | -0.015 |
lder | -0.013 |
defeat | -0.013 |
Frank | -0.012 |
oc | -0.012 |
Token | carpet |
Token ID | 20710 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
auc | +0.044 |
ays | +0.038 |
ains | +0.037 |
line | +0.037 |
bars | +0.037 |
Neg logit contributions | |
Amen | -0.058 |
lder | -0.049 |
defeat | -0.046 |
oc | -0.046 |
Frank | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | +5.17e-01 |
Pos logit contributions | |
Amen | +0.007 |
lder | +0.006 |
defeat | +0.006 |
oc | +0.006 |
chat | +0.006 |
Neg logit contributions | |
auc | -0.005 |
ays | -0.005 |
ains | -0.005 |
bars | -0.004 |
inter | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | +2.44e+00 |
Pos logit contributions | |
auc | +0.005 |
line | +0.004 |
ays | +0.004 |
ains | +0.004 |
bars | +0.004 |
Neg logit contributions | |
Amen | -0.007 |
lder | -0.006 |
defeat | -0.005 |
Frank | -0.005 |
oc | -0.005 |
Token | yarn |
Token ID | 21181 |
Feature activation | -8.70e-01 |
Pos logit contributions | |
auc | +0.063 |
bars | +0.058 |
tubes | +0.058 |
inter | +0.056 |
ays | +0.056 |
Neg logit contributions | |
Amen | -0.085 |
lder | -0.072 |
oc | -0.066 |
Frank | -0.066 |
Charles | -0.065 |
Token | to |
Token ID | 284 |
Feature activation | -3.91e+00 |
Pos logit contributions | |
Amen | +0.025 |
lder | +0.020 |
defeat | +0.020 |
oc | +0.019 |
chat | +0.019 |
Neg logit contributions | |
auc | -0.028 |
ains | -0.024 |
ays | -0.024 |
line | -0.024 |
bars | -0.024 |
Token | make |
Token ID | 787 |
Feature activation | +2.55e+00 |
Pos logit contributions | |
Amen | +0.137 |
lder | +0.113 |
Frank | +0.104 |
Charles | +0.102 |
oc | +0.102 |
Neg logit contributions | |
auc | -0.105 |
line | -0.096 |
tubes | -0.092 |
inter | -0.089 |
bars | -0.089 |
Token | a |
Token ID | 257 |
Feature activation | -5.27e+00 |
Pos logit contributions | |
auc | +0.047 |
tubes | +0.042 |
ays | +0.041 |
ains | +0.039 |
line | +0.039 |
Neg logit contributions | |
Amen | -0.110 |
lder | -0.094 |
Frank | -0.087 |
oc | -0.087 |
Charles | -0.087 |
Token | cozy |
Token ID | 37438 |
Feature activation | -8.49e+00 |
Pos logit contributions | |
Amen | +0.191 |
lder | +0.179 |
Frank | +0.172 |
Charles | +0.170 |
Robert | +0.152 |
Neg logit contributions | |
auc | -0.104 |
tubes | -0.098 |
dripping | -0.097 |
inter | -0.093 |
soap | -0.092 |
Token | nest |
Token ID | 16343 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
Amen | +0.278 |
lder | +0.258 |
Charles | +0.236 |
Frank | +0.235 |
Offic | +0.216 |
Neg logit contributions | |
auc | -0.209 |
inter | -0.196 |
tub | -0.193 |
lights | -0.190 |
line | -0.189 |
Token | for |
Token ID | 329 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
Amen | +0.058 |
defeat | +0.048 |
lder | +0.048 |
oc | +0.046 |
chat | +0.046 |
Neg logit contributions | |
auc | -0.045 |
ays | -0.040 |
ains | -0.040 |
bars | -0.040 |
ours | -0.038 |
Token | the |
Token ID | 262 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
auc | +0.021 |
line | +0.018 |
ains | +0.017 |
ays | +0.017 |
bars | +0.017 |
Neg logit contributions | |
Amen | -0.044 |
lder | -0.038 |
oc | -0.036 |
Frank | -0.036 |
defeat | -0.035 |
Token | bird |
Token ID | 6512 |
Feature activation | -8.31e-01 |
Pos logit contributions | |
auc | +0.056 |
tubes | +0.050 |
bars | +0.049 |
line | +0.048 |
ains | +0.048 |
Neg logit contributions | |
Amen | -0.070 |
oc | -0.057 |
lder | -0.057 |
Charles | -0.055 |
Frank | -0.054 |
Token | to |
Token ID | 284 |
Feature activation | -4.26e+00 |
Pos logit contributions | |
Amen | +0.032 |
lder | +0.027 |
Frank | +0.025 |
defeat | +0.025 |
Charles | +0.024 |
Neg logit contributions | |
auc | -0.020 |
line | -0.017 |
bars | -0.016 |
ays | -0.016 |
ains | -0.016 |
Token | was |
Token ID | 373 |
Feature activation | -6.73e-01 |
Pos logit contributions | |
Amen | +0.094 |
Frank | +0.081 |
defeat | +0.079 |
Mic | +0.079 |
lder | +0.078 |
Neg logit contributions | |
auc | -0.030 |
tubes | -0.023 |
ains | -0.022 |
tub | -0.021 |
flowed | -0.021 |
Token | tall |
Token ID | 7331 |
Feature activation | -3.93e+00 |
Pos logit contributions | |
Amen | +0.023 |
lder | +0.021 |
oc | +0.020 |
Frank | +0.019 |
defeat | +0.019 |
Neg logit contributions | |
auc | -0.016 |
line | -0.014 |
tubes | -0.014 |
ains | -0.013 |
inter | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
Amen | +0.138 |
lder | +0.115 |
Frank | +0.109 |
defeat | +0.105 |
Mic | +0.104 |
Neg logit contributions | |
auc | -0.103 |
line | -0.088 |
bars | -0.087 |
ays | -0.086 |
ains | -0.086 |
Token | wore |
Token ID | 12408 |
Feature activation | +3.36e-02 |
Pos logit contributions | |
Amen | +0.029 |
lder | +0.022 |
oc | +0.020 |
defeat | +0.019 |
Frank | +0.019 |
Neg logit contributions | |
auc | -0.040 |
ays | -0.038 |
ains | -0.036 |
bars | -0.036 |
line | -0.035 |
Token | a |
Token ID | 257 |
Feature activation | -5.46e+00 |
Pos logit contributions | |
auc | +0.001 |
tubes | +0.001 |
line | +0.001 |
lights | +0.001 |
ains | +0.001 |
Neg logit contributions | |
Amen | -0.001 |
lder | -0.001 |
oc | -0.001 |
defeat | -0.001 |
Frank | -0.001 |
Token | wide |
Token ID | 3094 |
Feature activation | -8.44e+00 |
Pos logit contributions | |
Amen | +0.218 |
lder | +0.199 |
Frank | +0.182 |
defeat | +0.181 |
arrest | +0.179 |
Neg logit contributions | |
auc | -0.092 |
tubes | -0.088 |
tub | -0.080 |
inter | -0.079 |
soap | -0.076 |
Token | hat |
Token ID | 6877 |
Feature activation | -3.09e+00 |
Pos logit contributions | |
Amen | +0.294 |
lder | +0.251 |
Frank | +0.243 |
Mic | +0.238 |
Chuck | +0.237 |
Neg logit contributions | |
auc | -0.201 |
tubes | -0.180 |
lights | -0.180 |
inter | -0.179 |
line | -0.173 |
Token | . |
Token ID | 13 |
Feature activation | +9.03e-02 |
Pos logit contributions | |
Amen | +0.129 |
lder | +0.113 |
Frank | +0.109 |
Mic | +0.106 |
Charles | +0.105 |
Neg logit contributions | |
auc | -0.058 |
line | -0.053 |
bars | -0.049 |
ays | -0.048 |
ains | -0.045 |
Token | One |
Token ID | 1881 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
auc | +0.003 |
ays | +0.002 |
ains | +0.002 |
line | +0.002 |
bars | +0.002 |
Neg logit contributions | |
Amen | -0.003 |
lder | -0.002 |
oc | -0.002 |
Frank | -0.002 |
defeat | -0.002 |
Token | day |
Token ID | 1110 |
Feature activation | -7.14e-01 |
Pos logit contributions | |
auc | +0.040 |
ays | +0.034 |
line | +0.033 |
bars | +0.033 |
ains | +0.033 |
Neg logit contributions | |
Amen | -0.074 |
lder | -0.063 |
oc | -0.060 |
defeat | -0.060 |
Frank | -0.059 |
Token | , |
Token ID | 11 |
Feature activation | +2.39e+00 |
Pos logit contributions | |
Amen | +0.023 |
lder | +0.019 |
defeat | +0.017 |
Frank | +0.017 |
oc | +0.017 |
Neg logit contributions | |
auc | -0.021 |
line | -0.018 |
ays | -0.018 |
bars | -0.018 |
ains | -0.018 |
Token | bird |
Token ID | 6512 |
Feature activation | +8.13e-01 |
Pos logit contributions | |
auc | +0.106 |
ains | +0.094 |
line | +0.094 |
bars | +0.092 |
tubes | +0.091 |
Neg logit contributions | |
Amen | -0.139 |
lder | -0.113 |
oc | -0.112 |
defeat | -0.111 |
Frank | -0.107 |
Token | wanted |
Token ID | 2227 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
auc | +0.017 |
ays | +0.014 |
line | +0.014 |
bars | +0.014 |
ains | +0.014 |
Neg logit contributions | |
Amen | -0.033 |
lder | -0.028 |
defeat | -0.027 |
oc | -0.026 |
Frank | -0.026 |
Token | to |
Token ID | 284 |
Feature activation | -5.60e+00 |
Pos logit contributions | |
Amen | +0.051 |
lder | +0.042 |
defeat | +0.041 |
Frank | +0.040 |
chat | +0.038 |
Neg logit contributions | |
auc | -0.058 |
ays | -0.052 |
line | -0.051 |
ains | -0.051 |
bars | -0.051 |
Token | make |
Token ID | 787 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
Amen | +0.225 |
lder | +0.191 |
Frank | +0.186 |
Charles | +0.184 |
Mic | +0.180 |
Neg logit contributions | |
auc | -0.119 |
tubes | -0.101 |
line | -0.100 |
ains | -0.088 |
inter | -0.088 |
Token | a |
Token ID | 257 |
Feature activation | -5.79e+00 |
Pos logit contributions | |
auc | +0.037 |
tubes | +0.032 |
ays | +0.032 |
ains | +0.031 |
line | +0.031 |
Neg logit contributions | |
Amen | -0.079 |
lder | -0.068 |
Mic | -0.064 |
oc | -0.063 |
Frank | -0.063 |
Token | cozy |
Token ID | 37438 |
Feature activation | -8.42e+00 |
Pos logit contributions | |
Amen | +0.221 |
lder | +0.198 |
Frank | +0.188 |
Charles | +0.181 |
Mic | +0.176 |
Neg logit contributions | |
auc | -0.122 |
tubes | -0.109 |
tub | -0.108 |
inter | -0.105 |
prett | -0.102 |
Token | nest |
Token ID | 16343 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
Amen | +0.282 |
lder | +0.267 |
Mic | +0.250 |
Frank | +0.249 |
Charles | +0.248 |
Neg logit contributions | |
tub | -0.190 |
auc | -0.189 |
lights | -0.177 |
inter | -0.173 |
line | -0.169 |
Token | in |
Token ID | 287 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
Amen | +0.075 |
lder | +0.063 |
defeat | +0.062 |
oc | +0.060 |
Frank | +0.059 |
Neg logit contributions | |
auc | -0.050 |
ays | -0.044 |
ains | -0.043 |
bars | -0.043 |
line | -0.041 |
Token | a |
Token ID | 257 |
Feature activation | -3.38e+00 |
Pos logit contributions | |
auc | +0.047 |
ays | +0.038 |
bars | +0.038 |
line | +0.038 |
ains | +0.038 |
Neg logit contributions | |
Amen | -0.085 |
lder | -0.072 |
defeat | -0.069 |
Frank | -0.069 |
oc | -0.069 |
Token | tree |
Token ID | 5509 |
Feature activation | -2.31e+00 |
Pos logit contributions | |
Amen | +0.114 |
lder | +0.098 |
Frank | +0.088 |
Charles | +0.087 |
defeat | +0.087 |
Neg logit contributions | |
auc | -0.087 |
line | -0.080 |
tub | -0.074 |
tubes | -0.073 |
ains | -0.072 |
Token | . |
Token ID | 13 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
Amen | +0.092 |
lder | +0.079 |
Frank | +0.074 |
defeat | +0.073 |
oc | +0.073 |
Neg logit contributions | |
auc | -0.050 |
line | -0.043 |
bars | -0.042 |
ays | -0.041 |
ains | -0.041 |
Token | home |
Token ID | 1363 |
Feature activation | -2.46e+00 |
Pos logit contributions | |
auc | +0.019 |
line | +0.015 |
tubes | +0.015 |
inter | +0.015 |
ays | +0.015 |
Neg logit contributions | |
Amen | -0.038 |
lder | -0.034 |
oc | -0.031 |
Charles | -0.030 |
Frank | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | +3.33e-01 |
Pos logit contributions | |
Amen | +0.096 |
lder | +0.082 |
oc | +0.077 |
Frank | +0.077 |
defeat | +0.076 |
Neg logit contributions | |
auc | -0.056 |
line | -0.048 |
bars | -0.046 |
ays | -0.046 |
ains | -0.045 |
Token | Mama |
Token ID | 40084 |
Feature activation | +2.47e+00 |
Pos logit contributions | |
auc | +0.009 |
line | +0.007 |
ays | +0.007 |
bars | +0.007 |
ains | +0.007 |
Neg logit contributions | |
Amen | -0.012 |
lder | -0.010 |
defeat | -0.010 |
Frank | -0.009 |
oc | -0.009 |
Token | made |
Token ID | 925 |
Feature activation | +9.23e-01 |
Pos logit contributions | |
auc | +0.055 |
ains | +0.046 |
bars | +0.045 |
line | +0.044 |
ays | +0.044 |
Neg logit contributions | |
Amen | -0.097 |
lder | -0.081 |
defeat | -0.078 |
oc | -0.076 |
Frank | -0.076 |
Token | a |
Token ID | 257 |
Feature activation | -5.73e+00 |
Pos logit contributions | |
auc | +0.022 |
ays | +0.019 |
line | +0.018 |
ains | +0.017 |
inter | +0.017 |
Neg logit contributions | |
Amen | -0.036 |
lder | -0.030 |
oc | -0.029 |
Mic | -0.028 |
defeat | -0.028 |
Token | cozy |
Token ID | 37438 |
Feature activation | -8.40e+00 |
Pos logit contributions | |
Amen | +0.221 |
lder | +0.193 |
Frank | +0.184 |
Charles | +0.178 |
Mic | +0.174 |
Neg logit contributions | |
auc | -0.107 |
inter | -0.099 |
dripping | -0.099 |
shallow | -0.096 |
tubes | -0.095 |
Token | nest |
Token ID | 16343 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
Amen | +0.274 |
lder | +0.252 |
Mic | +0.240 |
Charles | +0.235 |
Frank | +0.232 |
Neg logit contributions | |
auc | -0.187 |
tub | -0.183 |
inter | -0.178 |
line | -0.169 |
lights | -0.161 |
Token | for |
Token ID | 329 |
Feature activation | +7.88e-01 |
Pos logit contributions | |
Amen | +0.053 |
lder | +0.045 |
defeat | +0.043 |
oc | +0.042 |
Frank | +0.042 |
Neg logit contributions | |
auc | -0.038 |
ays | -0.033 |
ains | -0.033 |
bars | -0.033 |
tubes | -0.031 |
Token | the |
Token ID | 262 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
auc | +0.015 |
line | +0.013 |
ains | +0.013 |
ays | +0.012 |
bars | +0.012 |
Neg logit contributions | |
Amen | -0.033 |
lder | -0.028 |
oc | -0.027 |
defeat | -0.026 |
Frank | -0.026 |
Token | bird |
Token ID | 6512 |
Feature activation | +6.86e-01 |
Pos logit contributions | |
auc | +0.048 |
line | +0.043 |
ains | +0.042 |
inter | +0.042 |
bars | +0.040 |
Neg logit contributions | |
Amen | -0.066 |
oc | -0.052 |
defeat | -0.051 |
Mic | -0.051 |
lder | -0.051 |
Token | ie |
Token ID | 494 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
auc | +0.017 |
line | +0.014 |
ays | +0.014 |
ains | +0.014 |
bars | +0.014 |
Neg logit contributions | |
Amen | -0.025 |
lder | -0.022 |
defeat | -0.020 |
Frank | -0.020 |
oc | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | +5.90e-01 |
Pos logit contributions | |
auc | +0.051 |
inter | +0.046 |
ains | +0.046 |
ays | +0.043 |
bars | +0.043 |
Neg logit contributions | |
Amen | -0.044 |
defeat | -0.040 |
chat | -0.037 |
read | -0.036 |
Frank | -0.036 |
Token | He |
Token ID | 679 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
auc | +0.018 |
line | +0.016 |
ains | +0.016 |
ays | +0.016 |
lights | +0.015 |
Neg logit contributions | |
Amen | -0.018 |
oc | -0.014 |
lder | -0.014 |
Frank | -0.014 |
defeat | -0.013 |
Token | had |
Token ID | 550 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
Amen | +0.057 |
lder | +0.049 |
defeat | +0.047 |
oc | +0.046 |
chat | +0.045 |
Neg logit contributions | |
auc | -0.044 |
line | -0.038 |
ays | -0.038 |
ains | -0.037 |
bars | -0.037 |
Token | made |
Token ID | 925 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
Amen | +0.003 |
lder | +0.003 |
chat | +0.003 |
defeat | +0.003 |
read | +0.003 |
Neg logit contributions | |
auc | -0.003 |
ays | -0.003 |
ains | -0.003 |
line | -0.002 |
bars | -0.002 |
Token | a |
Token ID | 257 |
Feature activation | -5.34e+00 |
Pos logit contributions | |
auc | +0.049 |
ays | +0.043 |
line | +0.043 |
bars | +0.043 |
inter | +0.042 |
Neg logit contributions | |
Amen | -0.041 |
defeat | -0.035 |
lder | -0.033 |
oc | -0.032 |
chat | -0.031 |
Token | cozy |
Token ID | 37438 |
Feature activation | -8.39e+00 |
Pos logit contributions | |
Amen | +0.199 |
lder | +0.169 |
Frank | +0.163 |
Mic | +0.154 |
Charles | +0.152 |
Neg logit contributions | |
auc | -0.127 |
tubes | -0.113 |
line | -0.109 |
ains | -0.109 |
tub | -0.105 |
Token | little |
Token ID | 1310 |
Feature activation | -4.58e+00 |
Pos logit contributions | |
Amen | +0.306 |
lder | +0.261 |
Frank | +0.248 |
Mic | +0.239 |
Charles | +0.237 |
Neg logit contributions | |
auc | -0.203 |
line | -0.178 |
tubes | -0.177 |
lights | -0.174 |
tub | -0.173 |
Token | house |
Token ID | 2156 |
Feature activation | -3.64e+00 |
Pos logit contributions | |
Amen | +0.159 |
lder | +0.133 |
Frank | +0.125 |
defeat | +0.122 |
Mic | +0.120 |
Neg logit contributions | |
auc | -0.124 |
line | -0.106 |
bars | -0.106 |
ays | -0.106 |
ains | -0.104 |
Token | for |
Token ID | 329 |
Feature activation | +8.55e-01 |
Pos logit contributions | |
Amen | +0.107 |
defeat | +0.087 |
lder | +0.085 |
oc | +0.084 |
chat | +0.082 |
Neg logit contributions | |
auc | -0.110 |
ains | -0.101 |
ays | -0.100 |
inter | -0.097 |
bars | -0.096 |
Token | himself |
Token ID | 2241 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
auc | +0.025 |
ays | +0.023 |
bars | +0.022 |
inter | +0.021 |
line | +0.021 |
Neg logit contributions | |
Amen | -0.027 |
defeat | -0.023 |
lder | -0.022 |
oc | -0.021 |
Frank | -0.021 |
Token | out |
Token ID | 503 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
Amen | +0.005 |
lder | +0.004 |
defeat | +0.004 |
oc | +0.004 |
Frank | +0.004 |
Neg logit contributions | |
auc | -0.004 |
ays | -0.003 |
bars | -0.003 |
line | -0.003 |
ains | -0.003 |
Token | Ben |
Token ID | 3932 |
Feature activation | -6.88e-01 |
Pos logit contributions | |
Amen | +0.149 |
lder | +0.127 |
defeat | +0.123 |
Frank | +0.122 |
oc | +0.121 |
Neg logit contributions | |
auc | -0.060 |
bars | -0.049 |
ays | -0.049 |
ains | -0.049 |
line | -0.047 |
Token | asked |
Token ID | 1965 |
Feature activation | -3.83e+00 |
Pos logit contributions | |
Amen | +0.025 |
lder | +0.020 |
defeat | +0.019 |
Frank | +0.019 |
prise | +0.018 |
Neg logit contributions | |
auc | -0.016 |
line | -0.016 |
bars | -0.015 |
tubes | -0.015 |
ays | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | -3.37e-01 |
Pos logit contributions | |
Amen | +0.118 |
defeat | +0.095 |
lder | +0.091 |
Frank | +0.090 |
oc | +0.088 |
Neg logit contributions | |
auc | -0.113 |
bars | -0.103 |
ays | -0.103 |
ains | -0.102 |
line | -0.101 |
Token | holding |
Token ID | 4769 |
Feature activation | +3.52e+00 |
Pos logit contributions | |
Amen | +0.012 |
lder | +0.010 |
oc | +0.010 |
Frank | +0.009 |
Mic | +0.009 |
Neg logit contributions | |
auc | -0.008 |
line | -0.008 |
bars | -0.007 |
ains | -0.007 |
ays | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | -4.53e+00 |
Pos logit contributions | |
auc | +0.106 |
bars | +0.094 |
line | +0.093 |
ains | +0.089 |
lessness | +0.087 |
Neg logit contributions | |
Amen | -0.110 |
defeat | -0.092 |
oc | -0.087 |
lder | -0.085 |
Charles | -0.083 |
Token | wooden |
Token ID | 13510 |
Feature activation | -8.36e+00 |
Pos logit contributions | |
Amen | +0.169 |
lder | +0.151 |
Frank | +0.136 |
Mic | +0.130 |
Charles | +0.128 |
Neg logit contributions | |
auc | -0.105 |
tubes | -0.095 |
line | -0.092 |
tub | -0.087 |
inter | -0.086 |
Token | spoon |
Token ID | 24556 |
Feature activation | -7.14e-01 |
Pos logit contributions | |
Amen | +0.306 |
lder | +0.266 |
Frank | +0.257 |
Charles | +0.247 |
Mic | +0.241 |
Neg logit contributions | |
auc | -0.211 |
tubes | -0.181 |
line | -0.170 |
tub | -0.164 |
bars | -0.163 |
Token | as |
Token ID | 355 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
Amen | +0.027 |
lder | +0.023 |
defeat | +0.022 |
oc | +0.021 |
Frank | +0.021 |
Neg logit contributions | |
auc | -0.017 |
ays | -0.014 |
line | -0.014 |
bars | -0.014 |
ains | -0.014 |
Token | his |
Token ID | 465 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
auc | +0.038 |
ash | +0.035 |
ays | +0.035 |
inter | +0.035 |
idget | +0.034 |
Neg logit contributions | |
Amen | -0.035 |
defeat | -0.029 |
Charles | -0.025 |
lder | -0.025 |
Frank | -0.025 |
Token | sword |
Token ID | 8429 |
Feature activation | +6.46e-01 |
Pos logit contributions | |
auc | +0.005 |
tubes | +0.004 |
bars | +0.004 |
ains | +0.004 |
inter | +0.004 |
Neg logit contributions | |
Amen | -0.006 |
lder | -0.005 |
Frank | -0.005 |
oc | -0.004 |
Mic | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +7.08e-01 |
Pos logit contributions | |
auc | +0.014 |
line | +0.011 |
ays | +0.011 |
bars | +0.011 |
ains | +0.011 |
Neg logit contributions | |
Amen | -0.027 |
lder | -0.023 |
Frank | -0.021 |
Mic | -0.021 |
oc | -0.021 |
Token | and |
Token ID | 290 |
Feature activation | +2.38e-01 |
Pos logit contributions | |
auc | +0.108 |
ays | +0.096 |
bars | +0.095 |
ains | +0.094 |
ash | +0.093 |
Neg logit contributions | |
Amen | -0.061 |
defeat | -0.056 |
chat | -0.051 |
lder | -0.051 |
Frank | -0.050 |
Token | feathers |
Token ID | 27400 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
auc | +0.006 |
ays | +0.006 |
line | +0.005 |
ains | +0.005 |
bars | +0.005 |
Neg logit contributions | |
Amen | -0.008 |
lder | -0.007 |
defeat | -0.007 |
oc | -0.007 |
Frank | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | -3.45e+00 |
Pos logit contributions | |
auc | +0.037 |
ays | +0.033 |
ains | +0.032 |
bars | +0.032 |
ours | +0.032 |
Neg logit contributions | |
Amen | -0.026 |
defeat | -0.023 |
chat | -0.022 |
oc | -0.021 |
Frank | -0.021 |
Token | build |
Token ID | 1382 |
Feature activation | +6.07e-01 |
Pos logit contributions | |
Amen | +0.114 |
lder | +0.094 |
defeat | +0.094 |
oc | +0.090 |
Frank | +0.088 |
Neg logit contributions | |
auc | -0.097 |
ays | -0.088 |
bars | -0.084 |
ains | -0.083 |
line | -0.082 |
Token | a |
Token ID | 257 |
Feature activation | -5.35e+00 |
Pos logit contributions | |
auc | +0.017 |
bars | +0.015 |
ays | +0.015 |
line | +0.015 |
inter | +0.014 |
Neg logit contributions | |
Amen | -0.019 |
defeat | -0.017 |
oc | -0.016 |
lder | -0.016 |
chat | -0.015 |
Token | cozy |
Token ID | 37438 |
Feature activation | -8.32e+00 |
Pos logit contributions | |
Amen | +0.174 |
lder | +0.149 |
Frank | +0.133 |
Mic | +0.130 |
Charles | +0.129 |
Neg logit contributions | |
auc | -0.152 |
line | -0.136 |
tubes | -0.132 |
ains | -0.131 |
tub | -0.130 |
Token | shelter |
Token ID | 11772 |
Feature activation | -2.96e+00 |
Pos logit contributions | |
Amen | +0.248 |
lder | +0.208 |
Frank | +0.184 |
Mic | +0.183 |
defeat | +0.182 |
Neg logit contributions | |
auc | -0.259 |
line | -0.236 |
lights | -0.229 |
ays | -0.227 |
tubes | -0.226 |
Token | . |
Token ID | 13 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
Amen | +0.091 |
defeat | +0.079 |
oc | +0.076 |
lder | +0.073 |
chat | +0.073 |
Neg logit contributions | |
auc | -0.083 |
ays | -0.078 |
bars | -0.075 |
ains | -0.074 |
ours | -0.074 |
Token | |
Token ID | 220 |
Feature activation | +3.94e+00 |
Pos logit contributions | |
auc | +0.006 |
bars | +0.005 |
ays | +0.005 |
line | +0.005 |
ains | +0.005 |
Neg logit contributions | |
Amen | -0.008 |
lder | -0.006 |
oc | -0.006 |
Frank | -0.006 |
defeat | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
auc | +0.134 |
bars | +0.120 |
line | +0.118 |
tubes | +0.116 |
ays | +0.116 |
Neg logit contributions | |
Amen | -0.105 |
lder | -0.082 |
Frank | -0.081 |
oc | -0.080 |
defeat | -0.078 |
Token |
|
Token ID | 198 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
Amen | +0.079 |
lder | +0.067 |
defeat | +0.066 |
tail | +0.062 |
arrest | +0.062 |
Neg logit contributions | |
auc | -0.036 |
line | -0.033 |
ains | -0.032 |
tub | -0.030 |
ays | -0.028 |
Token | chair |
Token ID | 5118 |
Feature activation | +2.80e-01 |
Pos logit contributions | |
Amen | +0.021 |
lder | +0.018 |
Frank | +0.017 |
defeat | +0.017 |
oc | +0.017 |
Neg logit contributions | |
auc | -0.016 |
ays | -0.014 |
bars | -0.014 |
line | -0.014 |
ains | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | +2.88e-01 |
Pos logit contributions | |
auc | +0.007 |
ays | +0.006 |
ains | +0.006 |
line | +0.006 |
bars | +0.006 |
Neg logit contributions | |
Amen | -0.010 |
lder | -0.008 |
defeat | -0.008 |
oc | -0.008 |
Frank | -0.008 |
Token | They |
Token ID | 1119 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
auc | +0.009 |
ains | +0.008 |
line | +0.008 |
lights | +0.008 |
ays | +0.008 |
Neg logit contributions | |
Amen | -0.009 |
lder | -0.007 |
oc | -0.006 |
read | -0.006 |
Mic | -0.006 |
Token | made |
Token ID | 925 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
Amen | +0.005 |
lder | +0.004 |
Frank | +0.004 |
oc | +0.004 |
defeat | +0.004 |
Neg logit contributions | |
auc | -0.002 |
bars | -0.002 |
ains | -0.002 |
line | -0.002 |
tubes | -0.002 |
Token | a |
Token ID | 257 |
Feature activation | -5.23e+00 |
Pos logit contributions | |
auc | +0.037 |
ays | +0.032 |
ains | +0.031 |
tubes | +0.031 |
bars | +0.030 |
Neg logit contributions | |
Amen | -0.063 |
lder | -0.054 |
Frank | -0.051 |
Mic | -0.050 |
oc | -0.050 |
Token | cozy |
Token ID | 37438 |
Feature activation | -8.32e+00 |
Pos logit contributions | |
Amen | +0.204 |
lder | +0.194 |
Frank | +0.187 |
Mic | +0.172 |
Charles | +0.168 |
Neg logit contributions | |
auc | -0.094 |
tubes | -0.094 |
tub | -0.084 |
lights | -0.084 |
shallow | -0.079 |
Token | fort |
Token ID | 6285 |
Feature activation | -3.79e+00 |
Pos logit contributions | |
Amen | +0.300 |
lder | +0.283 |
Mic | +0.265 |
Frank | +0.264 |
Chuck | +0.254 |
Neg logit contributions | |
auc | -0.179 |
lights | -0.175 |
tub | -0.168 |
tubes | -0.164 |
tiles | -0.154 |
Token | and |
Token ID | 290 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
Amen | +0.130 |
lder | +0.108 |
Frank | +0.101 |
defeat | +0.100 |
oc | +0.100 |
Neg logit contributions | |
auc | -0.104 |
line | -0.090 |
ays | -0.089 |
bars | -0.089 |
ains | -0.088 |
Token | had |
Token ID | 550 |
Feature activation | -5.32e-01 |
Pos logit contributions | |
Amen | +0.044 |
lder | +0.038 |
Frank | +0.036 |
defeat | +0.036 |
oc | +0.035 |
Neg logit contributions | |
auc | -0.027 |
line | -0.022 |
bars | -0.022 |
ays | -0.022 |
ains | -0.022 |
Token | so |
Token ID | 523 |
Feature activation | -2.97e+00 |
Pos logit contributions | |
Amen | +0.020 |
lder | +0.017 |
oc | +0.016 |
defeat | +0.016 |
Frank | +0.015 |
Neg logit contributions | |
auc | -0.013 |
tubes | -0.011 |
ains | -0.011 |
bars | -0.011 |
ays | -0.011 |
Token | much |
Token ID | 881 |
Feature activation | -6.68e-01 |
Pos logit contributions | |
Amen | +0.110 |
lder | +0.094 |
Frank | +0.089 |
defeat | +0.088 |
Mic | +0.087 |
Neg logit contributions | |
auc | -0.072 |
bars | -0.061 |
ays | -0.061 |
ains | -0.059 |
lights | -0.059 |
Token | used |
Token ID | 973 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
auc | +0.010 |
ays | +0.009 |
ains | +0.009 |
line | +0.009 |
bars | +0.008 |
Neg logit contributions | |
Amen | -0.013 |
lder | -0.011 |
defeat | -0.011 |
oc | -0.011 |
Frank | -0.011 |
Token | them |
Token ID | 606 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
auc | +0.004 |
line | +0.003 |
ains | +0.003 |
ays | +0.003 |
bars | +0.003 |
Neg logit contributions | |
Amen | -0.004 |
defeat | -0.003 |
Frank | -0.003 |
chat | -0.003 |
lder | -0.003 |
Token | to |
Token ID | 284 |
Feature activation | -3.82e+00 |
Pos logit contributions | |
auc | +0.064 |
ays | +0.060 |
line | +0.058 |
ains | +0.058 |
bars | +0.056 |
Neg logit contributions | |
Amen | -0.036 |
defeat | -0.035 |
chat | -0.031 |
lder | -0.029 |
read | -0.029 |
Token | make |
Token ID | 787 |
Feature activation | +2.66e+00 |
Pos logit contributions | |
Amen | +0.125 |
lder | +0.104 |
defeat | +0.099 |
Frank | +0.096 |
oc | +0.096 |
Neg logit contributions | |
auc | -0.109 |
ays | -0.096 |
bars | -0.094 |
ains | -0.094 |
line | -0.093 |
Token | a |
Token ID | 257 |
Feature activation | -5.25e+00 |
Pos logit contributions | |
auc | +0.069 |
ays | +0.059 |
line | +0.059 |
bars | +0.058 |
ains | +0.058 |
Neg logit contributions | |
Amen | -0.095 |
lder | -0.080 |
defeat | -0.075 |
Frank | -0.074 |
oc | -0.074 |
Token | cozy |
Token ID | 37438 |
Feature activation | -8.29e+00 |
Pos logit contributions | |
Amen | +0.196 |
lder | +0.175 |
Frank | +0.160 |
Mic | +0.152 |
Charles | +0.151 |
Neg logit contributions | |
auc | -0.117 |
tubes | -0.108 |
line | -0.099 |
tub | -0.099 |
ains | -0.097 |
Token | home |
Token ID | 1363 |
Feature activation | -2.95e+00 |
Pos logit contributions | |
Amen | +0.273 |
lder | +0.244 |
Mic | +0.219 |
Frank | +0.219 |
Charles | +0.211 |
Neg logit contributions | |
auc | -0.218 |
lights | -0.198 |
tub | -0.197 |
line | -0.197 |
tubes | -0.194 |
Token | . |
Token ID | 13 |
Feature activation | +6.21e-01 |
Pos logit contributions | |
Amen | +0.100 |
defeat | +0.083 |
lder | +0.081 |
oc | +0.080 |
chat | +0.077 |
Neg logit contributions | |
auc | -0.077 |
ays | -0.071 |
ains | -0.068 |
bars | -0.067 |
inter | -0.066 |
Token | He |
Token ID | 679 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
auc | +0.019 |
ays | +0.017 |
ains | +0.016 |
bars | +0.016 |
line | +0.016 |
Neg logit contributions | |
Amen | -0.019 |
lder | -0.015 |
oc | -0.015 |
Frank | -0.015 |
defeat | -0.014 |
Token | thought |
Token ID | 1807 |
Feature activation | +8.27e-01 |
Pos logit contributions | |
Amen | +0.062 |
lder | +0.052 |
defeat | +0.049 |
Frank | +0.049 |
oc | +0.048 |
Neg logit contributions | |
auc | -0.044 |
ays | -0.038 |
line | -0.037 |
bars | -0.037 |
ains | -0.037 |
Token | his |
Token ID | 465 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
auc | +0.024 |
ays | +0.022 |
ains | +0.021 |
ained | +0.021 |
line | +0.021 |
Neg logit contributions | |
Amen | -0.024 |
defeat | -0.022 |
lder | -0.020 |
chat | -0.020 |
Frank | -0.020 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
Amen | +0.172 |
lder | +0.149 |
defeat | +0.141 |
Frank | +0.140 |
oc | +0.140 |
Neg logit contributions | |
auc | -0.082 |
line | -0.067 |
ays | -0.067 |
bars | -0.066 |
ains | -0.065 |
Token | said |
Token ID | 531 |
Feature activation | -4.59e+00 |
Pos logit contributions | |
Amen | +0.043 |
lder | +0.035 |
Frank | +0.034 |
Mic | +0.034 |
defeat | +0.032 |
Neg logit contributions | |
auc | -0.026 |
tubes | -0.023 |
bars | -0.022 |
line | -0.022 |
ays | -0.021 |
Token | , |
Token ID | 11 |
Feature activation | -5.88e-01 |
Pos logit contributions | |
Amen | +0.174 |
lder | +0.150 |
oc | +0.141 |
Frank | +0.141 |
Mic | +0.139 |
Neg logit contributions | |
auc | -0.106 |
line | -0.093 |
bars | -0.090 |
ays | -0.088 |
ains | -0.085 |
Token | holding |
Token ID | 4769 |
Feature activation | +3.33e+00 |
Pos logit contributions | |
Amen | +0.020 |
lder | +0.017 |
oc | +0.016 |
Mic | +0.016 |
Frank | +0.016 |
Neg logit contributions | |
line | -0.014 |
auc | -0.014 |
bars | -0.012 |
dripping | -0.012 |
tubes | -0.012 |
Token | a |
Token ID | 257 |
Feature activation | -4.59e+00 |
Pos logit contributions | |
auc | +0.098 |
line | +0.087 |
bars | +0.086 |
ains | +0.082 |
lights | +0.080 |
Neg logit contributions | |
Amen | -0.105 |
defeat | -0.088 |
lder | -0.085 |
oc | -0.083 |
Frank | -0.080 |
Token | wooden |
Token ID | 13510 |
Feature activation | -8.26e+00 |
Pos logit contributions | |
Amen | +0.166 |
lder | +0.148 |
Frank | +0.132 |
Mic | +0.127 |
Charles | +0.122 |
Neg logit contributions | |
auc | -0.113 |
tubes | -0.102 |
line | -0.097 |
tub | -0.095 |
inter | -0.091 |
Token | spoon |
Token ID | 24556 |
Feature activation | -5.83e-01 |
Pos logit contributions | |
Amen | +0.278 |
lder | +0.243 |
Frank | +0.230 |
Mic | +0.217 |
Charles | +0.217 |
Neg logit contributions | |
auc | -0.229 |
tubes | -0.198 |
line | -0.188 |
bars | -0.188 |
lights | -0.186 |
Token | as |
Token ID | 355 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
Amen | +0.022 |
lder | +0.018 |
defeat | +0.017 |
oc | +0.017 |
Frank | +0.017 |
Neg logit contributions | |
auc | -0.014 |
ays | -0.012 |
ains | -0.012 |
line | -0.012 |
bars | -0.012 |
Token | her |
Token ID | 607 |
Feature activation | -2.18e-01 |
Pos logit contributions | |
auc | +0.029 |
ays | +0.026 |
inter | +0.026 |
ash | +0.025 |
line | +0.025 |
Neg logit contributions | |
Amen | -0.035 |
defeat | -0.029 |
lder | -0.028 |
Frank | -0.028 |
Charles | -0.027 |
Token | sword |
Token ID | 8429 |
Feature activation | +7.28e-01 |
Pos logit contributions | |
Amen | +0.007 |
lder | +0.006 |
Frank | +0.005 |
oc | +0.005 |
Mic | +0.005 |
Neg logit contributions | |
auc | -0.006 |
bars | -0.006 |
tubes | -0.006 |
ains | -0.006 |
line | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +7.27e-03 |
Pos logit contributions | |
auc | +0.016 |
line | +0.013 |
ays | +0.013 |
bars | +0.013 |
ains | +0.013 |
Neg logit contributions | |
Amen | -0.029 |
lder | -0.025 |
Frank | -0.024 |
oc | -0.023 |
defeat | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | -7.61e-01 |
Pos logit contributions | |
Amen | +0.078 |
lder | +0.065 |
Frank | +0.062 |
Mic | +0.061 |
Charles | +0.061 |
Neg logit contributions | |
auc | -0.048 |
line | -0.045 |
bars | -0.040 |
tubes | -0.039 |
tub | -0.039 |
Token | her |
Token ID | 607 |
Feature activation | -6.51e-01 |
Pos logit contributions | |
Amen | +0.023 |
lder | +0.019 |
defeat | +0.018 |
Frank | +0.018 |
oc | +0.018 |
Neg logit contributions | |
auc | -0.023 |
line | -0.020 |
ays | -0.020 |
bars | -0.020 |
ains | -0.020 |
Token | mom |
Token ID | 1995 |
Feature activation | -9.56e-01 |
Pos logit contributions | |
Amen | +0.025 |
lder | +0.021 |
oc | +0.021 |
Frank | +0.021 |
Charles | +0.020 |
Neg logit contributions | |
auc | -0.014 |
tubes | -0.013 |
lights | -0.013 |
ains | -0.012 |
tub | -0.012 |
Token | asked |
Token ID | 1965 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
Amen | +0.035 |
lder | +0.030 |
Frank | +0.028 |
Charles | +0.028 |
defeat | +0.028 |
Neg logit contributions | |
auc | -0.023 |
line | -0.020 |
bars | -0.020 |
ains | -0.019 |
tubes | -0.018 |
Token | her |
Token ID | 607 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
Amen | +0.081 |
lder | +0.068 |
defeat | +0.064 |
Frank | +0.063 |
oc | +0.063 |
Neg logit contributions | |
auc | -0.061 |
line | -0.052 |
ays | -0.052 |
bars | -0.051 |
ains | -0.051 |
Token | why |
Token ID | 1521 |
Feature activation | +8.49e+00 |
Pos logit contributions | |
Amen | +0.039 |
lder | +0.033 |
Frank | +0.031 |
defeat | +0.031 |
oc | +0.031 |
Neg logit contributions | |
auc | -0.026 |
line | -0.022 |
ays | -0.022 |
bars | -0.022 |
ains | -0.022 |
Token | she |
Token ID | 673 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
auc | +0.266 |
ays | +0.263 |
bars | +0.257 |
ash | +0.253 |
inter | +0.252 |
Neg logit contributions | |
read | -0.193 |
Amen | -0.188 |
lder | -0.182 |
chat | -0.177 |
defeat | -0.170 |
Token | was |
Token ID | 373 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
auc | +0.029 |
line | +0.025 |
ays | +0.025 |
bars | +0.024 |
ains | +0.024 |
Neg logit contributions | |
Amen | -0.044 |
lder | -0.038 |
defeat | -0.035 |
oc | -0.035 |
chat | -0.035 |
Token | sad |
Token ID | 6507 |
Feature activation | -7.19e-01 |
Pos logit contributions | |
auc | +0.042 |
line | +0.037 |
ays | +0.037 |
ains | +0.036 |
bars | +0.036 |
Neg logit contributions | |
Amen | -0.048 |
lder | -0.040 |
oc | -0.037 |
defeat | -0.037 |
Frank | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | +2.79e-01 |
Pos logit contributions | |
Amen | +0.018 |
read | +0.015 |
chat | +0.014 |
lder | +0.014 |
defeat | +0.014 |
Neg logit contributions | |
auc | -0.023 |
ains | -0.022 |
ays | -0.022 |
bars | -0.021 |
ash | -0.021 |
Token | Lily |
Token ID | 20037 |
Feature activation | +8.82e-01 |
Pos logit contributions | |
auc | +0.007 |
ays | +0.006 |
bars | +0.006 |
line | +0.006 |
ains | +0.006 |
Neg logit contributions | |
Amen | -0.010 |
lder | -0.008 |
defeat | -0.008 |
Frank | -0.008 |
oc | -0.008 |
Token | our |
Token ID | 674 |
Feature activation | +7.87e-01 |
Pos logit contributions | |
auc | +0.110 |
bars | +0.095 |
line | +0.094 |
ays | +0.092 |
inter | +0.091 |
Neg logit contributions | |
Amen | -0.121 |
defeat | -0.102 |
Frank | -0.096 |
arrest | -0.095 |
lder | -0.094 |
Token | comfort |
Token ID | 4467 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
auc | +0.021 |
ains | +0.019 |
bars | +0.019 |
tubes | +0.019 |
inter | +0.019 |
Neg logit contributions | |
Amen | -0.027 |
lder | -0.022 |
Frank | -0.020 |
oc | -0.020 |
Charles | -0.020 |
Token | zone |
Token ID | 6516 |
Feature activation | -8.80e-01 |
Pos logit contributions | |
auc | +0.035 |
ays | +0.030 |
ours | +0.029 |
line | +0.029 |
ains | +0.029 |
Neg logit contributions | |
Amen | -0.043 |
lder | -0.037 |
oc | -0.036 |
defeat | -0.035 |
chat | -0.034 |
Token | to |
Token ID | 284 |
Feature activation | -4.41e+00 |
Pos logit contributions | |
Amen | +0.027 |
lder | +0.023 |
defeat | +0.022 |
oc | +0.021 |
chat | +0.021 |
Neg logit contributions | |
auc | -0.027 |
ays | -0.024 |
ains | -0.023 |
line | -0.022 |
bars | -0.022 |
Token | experience |
Token ID | 1998 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
Amen | +0.135 |
defeat | +0.111 |
lder | +0.107 |
oc | +0.102 |
Frank | +0.101 |
Neg logit contributions | |
auc | -0.143 |
ays | -0.128 |
bars | -0.125 |
line | -0.119 |
ains | -0.118 |
Token | all |
Token ID | 477 |
Feature activation | +5.90e+00 |
Pos logit contributions | |
auc | +0.045 |
ays | +0.040 |
bars | +0.039 |
line | +0.038 |
ains | +0.037 |
Neg logit contributions | |
Amen | -0.046 |
defeat | -0.041 |
oc | -0.038 |
lder | -0.037 |
Frank | -0.037 |
Token | the |
Token ID | 262 |
Feature activation | +3.51e+00 |
Pos logit contributions | |
auc | +0.181 |
bars | +0.164 |
ays | +0.164 |
idget | +0.158 |
line | +0.158 |
Neg logit contributions | |
Amen | -0.181 |
defeat | -0.157 |
oc | -0.143 |
Frank | -0.137 |
", | -0.136 |
Token | wonderful |
Token ID | 7932 |
Feature activation | +4.53e-01 |
Pos logit contributions | |
auc | +0.081 |
ains | +0.069 |
line | +0.067 |
bars | +0.067 |
tubes | +0.066 |
Neg logit contributions | |
Amen | -0.134 |
lder | -0.114 |
oc | -0.106 |
Frank | -0.105 |
defeat | -0.105 |
Token | things |
Token ID | 1243 |
Feature activation | +6.48e+00 |
Pos logit contributions | |
auc | +0.012 |
ays | +0.011 |
line | +0.010 |
bars | +0.010 |
ains | +0.010 |
Neg logit contributions | |
Amen | -0.016 |
lder | -0.013 |
defeat | -0.013 |
oc | -0.013 |
chat | -0.013 |
Token | life |
Token ID | 1204 |
Feature activation | +3.53e+00 |
Pos logit contributions | |
auc | +0.191 |
ays | +0.173 |
inter | +0.168 |
idget | +0.168 |
bars | +0.163 |
Neg logit contributions | |
Amen | -0.188 |
defeat | -0.170 |
chat | -0.160 |
lder | -0.159 |
", | -0.155 |
Token | has |
Token ID | 468 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
auc | +0.104 |
idget | +0.091 |
ays | +0.089 |
inter | +0.088 |
bars | +0.085 |
Neg logit contributions | |
Amen | -0.102 |
defeat | -0.093 |
lder | -0.090 |
", | -0.087 |
chat | -0.086 |
Token | go |
Token ID | 467 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
Amen | +0.096 |
lder | +0.078 |
oc | +0.075 |
Mic | +0.071 |
Frank | +0.071 |
Neg logit contributions | |
auc | -0.091 |
line | -0.091 |
ains | -0.088 |
bars | -0.087 |
ays | -0.085 |
Token | anywhere |
Token ID | 6609 |
Feature activation | +5.08e-01 |
Pos logit contributions | |
Amen | +0.050 |
lder | +0.043 |
oc | +0.039 |
Frank | +0.039 |
Mic | +0.039 |
Neg logit contributions | |
auc | -0.033 |
line | -0.031 |
ains | -0.030 |
ays | -0.030 |
bars | -0.029 |
Token | you |
Token ID | 345 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
auc | +0.014 |
line | +0.012 |
ays | +0.012 |
bars | +0.012 |
ains | +0.012 |
Neg logit contributions | |
Amen | -0.017 |
lder | -0.014 |
defeat | -0.013 |
Frank | -0.013 |
oc | -0.013 |
Token | want |
Token ID | 765 |
Feature activation | -4.14e-01 |
Pos logit contributions | |
auc | +0.042 |
line | +0.038 |
ains | +0.038 |
ays | +0.036 |
ash | +0.035 |
Neg logit contributions | |
Amen | -0.062 |
lder | -0.051 |
Frank | -0.050 |
Mic | -0.049 |
oc | -0.048 |
Token | with |
Token ID | 351 |
Feature activation | +5.71e+00 |
Pos logit contributions | |
Amen | +0.013 |
lder | +0.011 |
oc | +0.010 |
Mic | +0.010 |
defeat | +0.009 |
Neg logit contributions | |
auc | -0.012 |
line | -0.011 |
ains | -0.011 |
ays | -0.011 |
bars | -0.010 |
Token | those |
Token ID | 883 |
Feature activation | +8.35e+00 |
Pos logit contributions | |
auc | +0.145 |
ays | +0.130 |
ains | +0.129 |
line | +0.128 |
bars | +0.127 |
Neg logit contributions | |
Amen | -0.202 |
lder | -0.167 |
oc | -0.159 |
defeat | -0.153 |
Frank | -0.152 |
Token | wings |
Token ID | 12098 |
Feature activation | +5.67e-01 |
Pos logit contributions | |
auc | +0.239 |
line | +0.204 |
ays | +0.204 |
bars | +0.200 |
ains | +0.197 |
Neg logit contributions | |
Amen | -0.275 |
lder | -0.230 |
defeat | -0.216 |
Frank | -0.213 |
oc | -0.211 |
Token | of |
Token ID | 286 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
auc | +0.019 |
ays | +0.016 |
bars | +0.016 |
ains | +0.016 |
line | +0.016 |
Neg logit contributions | |
Amen | -0.016 |
lder | -0.013 |
defeat | -0.012 |
Frank | -0.012 |
oc | -0.012 |
Token | yours |
Token ID | 12431 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
auc | +0.068 |
line | +0.062 |
ains | +0.062 |
ays | +0.061 |
bars | +0.061 |
Neg logit contributions | |
Amen | -0.086 |
lder | -0.072 |
oc | -0.066 |
Frank | -0.065 |
Mic | -0.064 |
Token | ." |
Token ID | 526 |
Feature activation | -2.26e+00 |
Pos logit contributions | |
Amen | +0.058 |
lder | +0.048 |
Frank | +0.044 |
oc | +0.043 |
defeat | +0.043 |
Neg logit contributions | |
auc | -0.051 |
line | -0.046 |
ays | -0.046 |
ains | -0.046 |
bars | -0.045 |
Token |
|
Token ID | 198 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
Amen | +0.086 |
lder | +0.071 |
defeat | +0.070 |
Frank | +0.068 |
Charles | +0.066 |
Neg logit contributions | |
auc | -0.049 |
line | -0.045 |
ays | -0.043 |
ains | -0.043 |
inter | -0.041 |
Token | He |
Token ID | 679 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
auc | +0.029 |
ains | +0.025 |
ays | +0.025 |
line | +0.024 |
bars | +0.024 |
Neg logit contributions | |
Amen | -0.037 |
lder | -0.031 |
Frank | -0.030 |
oc | -0.029 |
defeat | -0.029 |
Token | decided |
Token ID | 3066 |
Feature activation | -4.45e+00 |
Pos logit contributions | |
Amen | +0.065 |
lder | +0.055 |
defeat | +0.052 |
oc | +0.051 |
Frank | +0.051 |
Neg logit contributions | |
auc | -0.053 |
ays | -0.046 |
line | -0.046 |
ains | -0.045 |
bars | -0.045 |
Token | to |
Token ID | 284 |
Feature activation | -5.11e+00 |
Pos logit contributions | |
Amen | +0.140 |
lder | +0.116 |
defeat | +0.109 |
Frank | +0.107 |
oc | +0.105 |
Neg logit contributions | |
auc | -0.134 |
ays | -0.118 |
ains | -0.117 |
bars | -0.115 |
line | -0.115 |
Token | take |
Token ID | 1011 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
Amen | +0.167 |
lder | +0.137 |
Frank | +0.127 |
oc | +0.127 |
defeat | +0.126 |
Neg logit contributions | |
auc | -0.148 |
line | -0.128 |
bars | -0.127 |
ays | -0.126 |
ains | -0.126 |
Token | it |
Token ID | 340 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
auc | +0.061 |
bars | +0.053 |
ains | +0.053 |
line | +0.053 |
ays | +0.052 |
Neg logit contributions | |
Amen | -0.060 |
defeat | -0.050 |
lder | -0.049 |
Frank | -0.047 |
oc | -0.047 |
Token | with |
Token ID | 351 |
Feature activation | +6.06e+00 |
Pos logit contributions | |
auc | +0.033 |
line | +0.029 |
ays | +0.028 |
bars | +0.028 |
ains | +0.028 |
Neg logit contributions | |
Amen | -0.048 |
lder | -0.040 |
oc | -0.037 |
Frank | -0.037 |
defeat | -0.037 |
Token | him |
Token ID | 683 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
auc | +0.196 |
bars | +0.173 |
ays | +0.173 |
line | +0.172 |
ains | +0.171 |
Neg logit contributions | |
Amen | -0.178 |
lder | -0.143 |
defeat | -0.138 |
Frank | -0.134 |
oc | -0.132 |
Token | and |
Token ID | 290 |
Feature activation | -8.29e-01 |
Pos logit contributions | |
Amen | +0.065 |
lder | +0.056 |
defeat | +0.052 |
Frank | +0.052 |
oc | +0.052 |
Neg logit contributions | |
auc | -0.042 |
ays | -0.036 |
line | -0.035 |
bars | -0.035 |
ains | -0.035 |
Token | he |
Token ID | 339 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
Amen | +0.026 |
lder | +0.021 |
defeat | +0.020 |
Frank | +0.020 |
oc | +0.019 |
Neg logit contributions | |
auc | -0.025 |
ays | -0.022 |
line | -0.022 |
bars | -0.022 |
ains | -0.022 |
Token | put |
Token ID | 1234 |
Feature activation | +2.48e+00 |
Pos logit contributions | |
Amen | +0.035 |
lder | +0.030 |
defeat | +0.028 |
Frank | +0.027 |
oc | +0.027 |
Neg logit contributions | |
auc | -0.026 |
ays | -0.023 |
line | -0.023 |
bars | -0.022 |
ains | -0.022 |
Token | it |
Token ID | 340 |
Feature activation | +1.86e+00 |
Pos logit contributions | |
auc | +0.091 |
ains | +0.079 |
bars | +0.078 |
inter | +0.077 |
line | +0.076 |
Neg logit contributions | |
Amen | -0.061 |
defeat | -0.049 |
Frank | -0.044 |
chat | -0.044 |
oc | -0.044 |
Token | it |
Token ID | 340 |
Feature activation | +5.10e+00 |
Pos logit contributions | |
auc | +0.103 |
ays | +0.094 |
ains | +0.091 |
inter | +0.091 |
line | +0.091 |
Neg logit contributions | |
Amen | -0.069 |
lder | -0.059 |
defeat | -0.056 |
chat | -0.054 |
Frank | -0.052 |
Token | flew |
Token ID | 13112 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
auc | +0.144 |
ays | +0.125 |
line | +0.124 |
bars | +0.124 |
ains | +0.122 |
Neg logit contributions | |
Amen | -0.169 |
lder | -0.144 |
defeat | -0.135 |
oc | -0.133 |
Frank | -0.132 |
Token | straight |
Token ID | 3892 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
auc | +0.004 |
ays | +0.003 |
bars | +0.003 |
line | +0.003 |
ains | +0.003 |
Neg logit contributions | |
Amen | -0.004 |
lder | -0.003 |
defeat | -0.003 |
Frank | -0.003 |
oc | -0.003 |
Token | towards |
Token ID | 3371 |
Feature activation | -5.56e-01 |
Pos logit contributions | |
Amen | +0.071 |
lder | +0.059 |
oc | +0.056 |
Frank | +0.056 |
defeat | +0.055 |
Neg logit contributions | |
auc | -0.047 |
ays | -0.041 |
line | -0.040 |
bars | -0.040 |
ains | -0.040 |
Token | Daisy |
Token ID | 40355 |
Feature activation | -2.39e+00 |
Pos logit contributions | |
Amen | +0.020 |
lder | +0.017 |
defeat | +0.017 |
Frank | +0.016 |
chat | +0.016 |
Neg logit contributions | |
auc | -0.014 |
bars | -0.012 |
ays | -0.012 |
line | -0.012 |
ains | -0.011 |
Token | with |
Token ID | 351 |
Feature activation | +6.05e+00 |
Pos logit contributions | |
Amen | +0.097 |
lder | +0.081 |
Frank | +0.077 |
Mic | +0.075 |
Offic | +0.075 |
Neg logit contributions | |
auc | -0.047 |
line | -0.046 |
bars | -0.044 |
ays | -0.043 |
ains | -0.042 |
Token | something |
Token ID | 1223 |
Feature activation | +3.72e+00 |
Pos logit contributions | |
auc | +0.160 |
ays | +0.142 |
line | +0.140 |
bars | +0.140 |
tubes | +0.138 |
Neg logit contributions | |
Amen | -0.205 |
lder | -0.174 |
oc | -0.163 |
defeat | -0.154 |
Frank | -0.154 |
Token | in |
Token ID | 287 |
Feature activation | +3.41e+00 |
Pos logit contributions | |
auc | +0.120 |
ays | +0.111 |
ains | +0.110 |
bars | +0.106 |
inter | +0.104 |
Neg logit contributions | |
Amen | -0.096 |
defeat | -0.088 |
chat | -0.084 |
lder | -0.082 |
Frank | -0.080 |
Token | its |
Token ID | 663 |
Feature activation | -8.37e-01 |
Pos logit contributions | |
auc | +0.096 |
ays | +0.084 |
line | +0.084 |
bars | +0.083 |
ains | +0.083 |
Neg logit contributions | |
Amen | -0.113 |
lder | -0.094 |
defeat | -0.087 |
oc | -0.087 |
Frank | -0.086 |
Token | mouth |
Token ID | 5422 |
Feature activation | +2.89e+00 |
Pos logit contributions | |
Amen | +0.026 |
lder | +0.022 |
Frank | +0.019 |
oc | +0.019 |
defeat | +0.019 |
Neg logit contributions | |
auc | -0.024 |
line | -0.022 |
ays | -0.022 |
bars | -0.022 |
ains | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | +6.57e-01 |
Pos logit contributions | |
auc | +0.086 |
ains | +0.077 |
ays | +0.076 |
ours | +0.074 |
bars | +0.073 |
Neg logit contributions | |
Amen | -0.079 |
lder | -0.071 |
defeat | -0.070 |
chat | -0.069 |
Frank | -0.067 |
Token | ages |
Token ID | 9337 |
Feature activation | +8.47e-01 |
Pos logit contributions | |
auc | +0.045 |
ays | +0.041 |
ains | +0.040 |
bars | +0.038 |
inter | +0.038 |
Neg logit contributions | |
Amen | -0.051 |
lder | -0.043 |
Frank | -0.043 |
defeat | -0.042 |
chat | -0.041 |
Token | , |
Token ID | 11 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
auc | +0.032 |
ays | +0.030 |
ains | +0.029 |
inter | +0.029 |
ash | +0.029 |
Neg logit contributions | |
Amen | -0.015 |
lder | -0.014 |
defeat | -0.013 |
chat | -0.013 |
", | -0.012 |
Token | finally |
Token ID | 3443 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
auc | +0.045 |
ays | +0.039 |
ains | +0.037 |
bars | +0.037 |
ash | +0.037 |
Neg logit contributions | |
Amen | -0.058 |
defeat | -0.049 |
Frank | -0.048 |
chat | -0.048 |
read | -0.046 |
Token | Randy |
Token ID | 21248 |
Feature activation | +8.06e-01 |
Pos logit contributions | |
Amen | +0.037 |
defeat | +0.035 |
lder | +0.034 |
Frank | +0.033 |
chat | +0.032 |
Neg logit contributions | |
auc | -0.055 |
bars | -0.050 |
ays | -0.049 |
lights | -0.049 |
inter | -0.048 |
Token | got |
Token ID | 1392 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
auc | +0.023 |
ays | +0.020 |
ains | +0.019 |
bars | +0.019 |
line | +0.019 |
Neg logit contributions | |
Amen | -0.027 |
lder | -0.023 |
defeat | -0.021 |
oc | -0.021 |
Frank | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
auc | +0.035 |
ains | +0.031 |
line | +0.030 |
ays | +0.030 |
lights | +0.030 |
Neg logit contributions | |
Amen | -0.037 |
lder | -0.030 |
defeat | -0.030 |
Frank | -0.029 |
chat | -0.028 |
Token | number |
Token ID | 1271 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
auc | +0.075 |
ays | +0.066 |
ains | +0.065 |
line | +0.065 |
bars | +0.064 |
Neg logit contributions | |
Amen | -0.073 |
lder | -0.061 |
defeat | -0.057 |
oc | -0.055 |
Frank | -0.055 |
Token | first |
Token ID | 717 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
auc | +0.064 |
ains | +0.059 |
ays | +0.055 |
ained | +0.055 |
inter | +0.055 |
Neg logit contributions | |
Amen | -0.065 |
defeat | -0.056 |
lder | -0.055 |
oc | -0.054 |
chat | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
Amen | +0.069 |
lder | +0.059 |
defeat | +0.058 |
oc | +0.056 |
chat | +0.056 |
Neg logit contributions | |
auc | -0.067 |
ains | -0.059 |
ays | -0.058 |
line | -0.056 |
lights | -0.055 |
Token | Max |
Token ID | 5436 |
Feature activation | -3.18e-01 |
Pos logit contributions | |
auc | +0.033 |
ains | +0.029 |
ays | +0.028 |
lights | +0.028 |
line | +0.027 |
Neg logit contributions | |
Amen | -0.032 |
oc | -0.025 |
Frank | -0.025 |
lder | -0.024 |
Charles | -0.024 |
Token | was |
Token ID | 373 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
Amen | +0.012 |
lder | +0.010 |
defeat | +0.010 |
Frank | +0.010 |
oc | +0.010 |
Neg logit contributions | |
auc | -0.007 |
bars | -0.006 |
line | -0.006 |
ays | -0.006 |
ains | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
auc | +0.053 |
ays | +0.049 |
ains | +0.048 |
ours | +0.047 |
lights | +0.045 |
Neg logit contributions | |
Amen | -0.037 |
defeat | -0.031 |
lder | -0.030 |
chat | -0.029 |
read | -0.029 |
Token | stepped |
Token ID | 10764 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
Amen | +0.034 |
lder | +0.029 |
defeat | +0.028 |
chat | +0.027 |
Frank | +0.026 |
Neg logit contributions | |
auc | -0.031 |
ays | -0.027 |
ains | -0.026 |
bars | -0.026 |
ash | -0.025 |
Token | into |
Token ID | 656 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
auc | +0.056 |
ays | +0.048 |
line | +0.048 |
bars | +0.047 |
ains | +0.046 |
Neg logit contributions | |
Amen | -0.083 |
lder | -0.072 |
oc | -0.066 |
defeat | -0.066 |
Frank | -0.066 |
Token | the |
Token ID | 262 |
Feature activation | +2.18e+00 |
Pos logit contributions | |
Amen | +0.005 |
lder | +0.004 |
defeat | +0.004 |
Frank | +0.004 |
oc | +0.004 |
Neg logit contributions | |
auc | -0.003 |
ays | -0.002 |
line | -0.002 |
bars | -0.002 |
ains | -0.002 |
Token | fountain |
Token ID | 29420 |
Feature activation | +2.84e+00 |
Pos logit contributions | |
auc | +0.048 |
bars | +0.044 |
ains | +0.043 |
ays | +0.042 |
line | +0.041 |
Neg logit contributions | |
Amen | -0.082 |
lder | -0.071 |
chat | -0.067 |
defeat | -0.067 |
oc | -0.065 |
Token | anyway |
Token ID | 6949 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
auc | +0.084 |
ains | +0.076 |
ays | +0.074 |
bars | +0.070 |
ours | +0.070 |
Neg logit contributions | |
Amen | -0.080 |
defeat | -0.070 |
lder | -0.069 |
chat | -0.068 |
oc | -0.066 |
Token | . |
Token ID | 13 |
Feature activation | +4.09e-01 |
Pos logit contributions | |
auc | +0.033 |
ays | +0.030 |
ains | +0.029 |
ours | +0.028 |
inter | +0.027 |
Neg logit contributions | |
Amen | -0.034 |
defeat | -0.028 |
lder | -0.027 |
Frank | -0.027 |
oc | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
auc | +0.012 |
bars | +0.010 |
ains | +0.010 |
ays | +0.010 |
line | +0.010 |
Neg logit contributions | |
Amen | -0.013 |
lder | -0.011 |
Frank | -0.010 |
defeat | -0.010 |
oc | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
Amen | +0.073 |
lder | +0.062 |
defeat | +0.059 |
oc | +0.057 |
arrest | +0.056 |
Neg logit contributions | |
auc | -0.035 |
line | -0.030 |
ays | -0.029 |
ains | -0.028 |
bars | -0.028 |
Token | The |
Token ID | 464 |
Feature activation | +3.98e+00 |
Pos logit contributions | |
Amen | +0.054 |
lder | +0.039 |
defeat | +0.038 |
oc | +0.038 |
Charles | +0.036 |
Neg logit contributions | |
auc | -0.040 |
line | -0.037 |
ained | -0.034 |
tubes | -0.034 |
inter | -0.034 |
Token | stranger |
Token ID | 16195 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
auc | +0.115 |
line | +0.099 |
ains | +0.096 |
bars | +0.095 |
tubes | +0.094 |
Neg logit contributions | |
Amen | -0.129 |
oc | -0.108 |
lder | -0.104 |
defeat | -0.101 |
Mic | -0.099 |
Token | 's |
Token ID | 338 |
Feature activation | +3.85e-01 |
Pos logit contributions | |
Amen | +0.014 |
defeat | +0.011 |
Frank | +0.011 |
lder | +0.011 |
Mic | +0.011 |
Neg logit contributions | |
auc | -0.005 |
tub | -0.004 |
ays | -0.004 |
inter | -0.004 |
line | -0.004 |
Token | mom |
Token ID | 1995 |
Feature activation | -7.87e-01 |
Pos logit contributions | |
auc | +0.008 |
bars | +0.008 |
ays | +0.007 |
tub | +0.007 |
ains | +0.007 |
Neg logit contributions | |
Amen | -0.014 |
oc | -0.012 |
defeat | -0.012 |
lder | -0.012 |
Frank | -0.012 |
Token | my |
Token ID | 1820 |
Feature activation | -4.70e-01 |
Pos logit contributions | |
Amen | +0.029 |
lder | +0.024 |
defeat | +0.024 |
Frank | +0.023 |
Charles | +0.023 |
Neg logit contributions | |
auc | -0.020 |
line | -0.016 |
ays | -0.016 |
bars | -0.016 |
ains | -0.016 |
Token | said |
Token ID | 531 |
Feature activation | -2.58e+00 |
Pos logit contributions | |
Amen | +0.018 |
defeat | +0.015 |
Frank | +0.014 |
Charles | +0.014 |
lder | +0.014 |
Neg logit contributions | |
auc | -0.011 |
line | -0.009 |
ains | -0.009 |
ays | -0.008 |
inter | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
Amen | +0.093 |
oc | +0.077 |
lder | +0.076 |
defeat | +0.076 |
Mic | +0.075 |
Neg logit contributions | |
auc | -0.064 |
line | -0.055 |
bars | -0.052 |
ays | -0.051 |
tubes | -0.050 |
Token | " |
Token ID | 366 |
Feature activation | +9.38e-01 |
Pos logit contributions | |
auc | +0.023 |
line | +0.018 |
bars | +0.018 |
ays | +0.018 |
ains | +0.017 |
Neg logit contributions | |
Amen | -0.053 |
lder | -0.046 |
defeat | -0.044 |
oc | -0.044 |
Frank | -0.044 |
Token | Tim |
Token ID | 14967 |
Feature activation | +2.50e+00 |
Pos logit contributions | |
auc | +0.028 |
ays | +0.025 |
line | +0.025 |
ains | +0.025 |
bars | +0.025 |
Neg logit contributions | |
Amen | -0.030 |
lder | -0.024 |
Frank | -0.023 |
oc | -0.023 |
defeat | -0.022 |
Token | my |
Token ID | 1820 |
Feature activation | +4.61e-01 |
Pos logit contributions | |
auc | +0.048 |
ays | +0.038 |
ains | +0.038 |
bars | +0.038 |
line | +0.036 |
Neg logit contributions | |
Amen | -0.106 |
lder | -0.091 |
defeat | -0.088 |
Frank | -0.087 |
Charles | -0.086 |
Token | , |
Token ID | 11 |
Feature activation | +3.46e+00 |
Pos logit contributions | |
auc | +0.013 |
ays | +0.012 |
ains | +0.012 |
line | +0.012 |
bars | +0.012 |
Neg logit contributions | |
Amen | -0.015 |
lder | -0.012 |
Frank | -0.011 |
defeat | -0.011 |
oc | -0.011 |
Token | you |
Token ID | 345 |
Feature activation | +2.50e+00 |
Pos logit contributions | |
auc | +0.088 |
ays | +0.078 |
bars | +0.075 |
ains | +0.075 |
line | +0.074 |
Neg logit contributions | |
Amen | -0.126 |
lder | -0.106 |
defeat | -0.100 |
oc | -0.098 |
Frank | -0.098 |
Token | 're |
Token ID | 821 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
auc | +0.043 |
ains | +0.035 |
inter | +0.033 |
tubes | +0.032 |
lessness | +0.032 |
Neg logit contributions | |
Amen | -0.105 |
lder | -0.093 |
Frank | -0.092 |
Charles | -0.089 |
Mic | -0.089 |
Token | fabric |
Token ID | 9664 |
Feature activation | -3.76e-01 |
Pos logit contributions | |
Amen | +0.077 |
lder | +0.061 |
oc | +0.057 |
defeat | +0.057 |
chat | +0.055 |
Neg logit contributions | |
auc | -0.097 |
ays | -0.086 |
ains | -0.086 |
line | -0.086 |
bars | -0.085 |
Token | with |
Token ID | 351 |
Feature activation | +6.35e+00 |
Pos logit contributions | |
Amen | +0.009 |
read | +0.007 |
lder | +0.007 |
defeat | +0.007 |
chat | +0.007 |
Neg logit contributions | |
auc | -0.013 |
ains | -0.011 |
ash | -0.011 |
ays | -0.011 |
inter | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | -3.34e+00 |
Pos logit contributions | |
auc | +0.207 |
ays | +0.182 |
ains | +0.181 |
inter | +0.180 |
line | +0.179 |
Neg logit contributions | |
Amen | -0.169 |
defeat | -0.138 |
chat | -0.138 |
Frank | -0.138 |
read | -0.137 |
Token | pattern |
Token ID | 3912 |
Feature activation | +8.86e-01 |
Pos logit contributions | |
Amen | +0.117 |
lder | +0.100 |
defeat | +0.092 |
Frank | +0.092 |
oc | +0.091 |
Neg logit contributions | |
auc | -0.087 |
line | -0.076 |
bars | -0.075 |
ays | -0.074 |
ains | -0.073 |
Token | made |
Token ID | 925 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
auc | +0.027 |
ains | +0.025 |
inter | +0.025 |
ash | +0.024 |
idget | +0.024 |
Neg logit contributions | |
Amen | -0.021 |
read | -0.019 |
defeat | -0.019 |
chat | -0.018 |
lder | -0.018 |
Token | of |
Token ID | 286 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
auc | +0.041 |
ains | +0.038 |
ays | +0.038 |
bars | +0.037 |
ours | +0.037 |
Neg logit contributions | |
Amen | -0.038 |
defeat | -0.036 |
chat | -0.034 |
read | -0.033 |
lder | -0.032 |
Token | tiny |
Token ID | 7009 |
Feature activation | -4.57e+00 |
Pos logit contributions | |
auc | +0.033 |
ash | +0.029 |
inter | +0.029 |
ays | +0.029 |
line | +0.028 |
Neg logit contributions | |
Amen | -0.029 |
defeat | -0.024 |
chat | -0.023 |
read | -0.023 |
Frank | -0.022 |
Token | silver |
Token ID | 8465 |
Feature activation | -2.31e+00 |
Pos logit contributions | |
Amen | +0.157 |
oc | +0.132 |
defeat | +0.129 |
lder | +0.129 |
chat | +0.122 |
Neg logit contributions | |
auc | -0.118 |
line | -0.101 |
ays | -0.099 |
ains | -0.099 |
bars | -0.097 |
Token | needles |
Token ID | 25209 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
Amen | +0.060 |
defeat | +0.054 |
oc | +0.053 |
read | +0.052 |
tail | +0.049 |
Neg logit contributions | |
auc | -0.070 |
line | -0.061 |
ains | -0.060 |
ays | -0.060 |
ash | -0.060 |
Token | . |
Token ID | 13 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
auc | +0.057 |
ains | +0.050 |
ours | +0.050 |
lights | +0.049 |
ays | +0.048 |
Neg logit contributions | |
Amen | -0.034 |
read | -0.031 |
", | -0.029 |
defeat | -0.029 |
tail | -0.028 |
Token | She |
Token ID | 1375 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
auc | +0.003 |
ains | +0.003 |
line | +0.003 |
ays | +0.003 |
bars | +0.003 |
Neg logit contributions | |
Amen | -0.003 |
lder | -0.002 |
chat | -0.002 |
oc | -0.002 |
read | -0.002 |
Token | while |
Token ID | 981 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
auc | +0.057 |
ays | +0.050 |
line | +0.050 |
bars | +0.050 |
ains | +0.050 |
Neg logit contributions | |
Amen | -0.050 |
lder | -0.040 |
defeat | -0.037 |
oc | -0.037 |
Frank | -0.036 |
Token | you |
Token ID | 345 |
Feature activation | +2.40e+00 |
Pos logit contributions | |
auc | +0.030 |
ays | +0.026 |
bars | +0.026 |
ains | +0.025 |
line | +0.025 |
Neg logit contributions | |
Amen | -0.037 |
lder | -0.031 |
defeat | -0.029 |
Frank | -0.029 |
oc | -0.028 |
Token | swim |
Token ID | 9422 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
auc | +0.064 |
ains | +0.056 |
line | +0.056 |
ays | +0.055 |
bars | +0.055 |
Neg logit contributions | |
Amen | -0.083 |
lder | -0.068 |
Frank | -0.065 |
oc | -0.063 |
Mic | -0.063 |
Token | up |
Token ID | 510 |
Feature activation | +9.12e-01 |
Pos logit contributions | |
auc | +0.032 |
bars | +0.027 |
ays | +0.027 |
line | +0.027 |
ains | +0.027 |
Neg logit contributions | |
Amen | -0.038 |
lder | -0.032 |
defeat | -0.030 |
oc | -0.030 |
Frank | -0.029 |
Token | to |
Token ID | 284 |
Feature activation | -3.16e+00 |
Pos logit contributions | |
auc | +0.026 |
ays | +0.023 |
bars | +0.022 |
line | +0.022 |
ains | +0.022 |
Neg logit contributions | |
Amen | -0.030 |
lder | -0.025 |
defeat | -0.023 |
Frank | -0.023 |
oc | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | +2.39e+00 |
Pos logit contributions | |
Amen | +0.103 |
lder | +0.085 |
defeat | +0.081 |
Frank | +0.079 |
oc | +0.079 |
Neg logit contributions | |
auc | -0.093 |
bars | -0.080 |
ays | -0.080 |
line | -0.078 |
ains | -0.078 |
Token | top |
Token ID | 1353 |
Feature activation | +4.91e+00 |
Pos logit contributions | |
auc | +0.061 |
line | +0.052 |
ains | +0.051 |
bars | +0.051 |
ays | +0.051 |
Neg logit contributions | |
Amen | -0.086 |
lder | -0.073 |
Frank | -0.068 |
defeat | -0.068 |
oc | -0.068 |
Token | of |
Token ID | 286 |
Feature activation | +3.70e+00 |
Pos logit contributions | |
auc | +0.150 |
ays | +0.131 |
ains | +0.130 |
line | +0.129 |
bars | +0.129 |
Neg logit contributions | |
Amen | -0.148 |
lder | -0.121 |
oc | -0.115 |
defeat | -0.114 |
Frank | -0.111 |
Token | the |
Token ID | 262 |
Feature activation | +3.39e+00 |
Pos logit contributions | |
auc | +0.100 |
bars | +0.086 |
ays | +0.084 |
inter | +0.082 |
ains | +0.081 |
Neg logit contributions | |
Amen | -0.131 |
lder | -0.109 |
defeat | -0.107 |
Frank | -0.105 |
chat | -0.101 |
Token | water |
Token ID | 1660 |
Feature activation | +2.64e+00 |
Pos logit contributions | |
auc | +0.094 |
line | +0.081 |
ays | +0.081 |
bars | +0.080 |
ains | +0.080 |
Neg logit contributions | |
Amen | -0.115 |
lder | -0.096 |
defeat | -0.090 |
Frank | -0.089 |
oc | -0.089 |
Token | , |
Token ID | 11 |
Feature activation | +2.55e+00 |
Pos logit contributions | |
auc | +0.084 |
ays | +0.074 |
bars | +0.074 |
ains | +0.073 |
inter | +0.072 |
Neg logit contributions | |
Amen | -0.076 |
lder | -0.062 |
defeat | -0.058 |
oc | -0.057 |
Frank | -0.056 |
Token | , |
Token ID | 11 |
Feature activation | +7.43e-01 |
Pos logit contributions | |
Amen | +0.023 |
defeat | +0.019 |
oc | +0.018 |
lder | +0.018 |
Frank | +0.017 |
Neg logit contributions | |
auc | -0.026 |
ains | -0.023 |
ays | -0.022 |
bars | -0.022 |
line | -0.021 |
Token | so |
Token ID | 523 |
Feature activation | -2.74e+00 |
Pos logit contributions | |
auc | +0.024 |
ays | +0.022 |
ains | +0.022 |
ours | +0.020 |
bars | +0.020 |
Neg logit contributions | |
Amen | -0.021 |
defeat | -0.017 |
Frank | -0.016 |
chat | -0.016 |
read | -0.016 |
Token | he |
Token ID | 339 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
Amen | +0.105 |
lder | +0.089 |
defeat | +0.085 |
Frank | +0.084 |
oc | +0.084 |
Neg logit contributions | |
auc | -0.065 |
ays | -0.054 |
ains | -0.054 |
line | -0.054 |
bars | -0.053 |
Token | decided |
Token ID | 3066 |
Feature activation | -4.66e+00 |
Pos logit contributions | |
Amen | +0.051 |
lder | +0.042 |
defeat | +0.040 |
oc | +0.039 |
Frank | +0.039 |
Neg logit contributions | |
auc | -0.039 |
line | -0.033 |
bars | -0.033 |
ays | -0.033 |
ains | -0.033 |
Token | to |
Token ID | 284 |
Feature activation | -5.25e+00 |
Pos logit contributions | |
Amen | +0.125 |
lder | +0.103 |
defeat | +0.100 |
Frank | +0.097 |
chat | +0.092 |
Neg logit contributions | |
auc | -0.161 |
ays | -0.146 |
ains | -0.144 |
lights | -0.137 |
inter | -0.136 |
Token | manage |
Token ID | 6687 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
Amen | +0.179 |
lder | +0.145 |
oc | +0.142 |
Frank | +0.138 |
Mic | +0.133 |
Neg logit contributions | |
auc | -0.139 |
line | -0.126 |
bars | -0.124 |
tubes | -0.119 |
inter | -0.118 |
Token | a |
Token ID | 257 |
Feature activation | -4.37e+00 |
Pos logit contributions | |
Amen | +0.004 |
defeat | +0.004 |
Frank | +0.003 |
chat | +0.003 |
read | +0.003 |
Neg logit contributions | |
auc | -0.005 |
ains | -0.004 |
ays | -0.004 |
line | -0.004 |
inter | -0.004 |
Token | thing |
Token ID | 1517 |
Feature activation | +3.27e+00 |
Pos logit contributions | |
Amen | +0.157 |
lder | +0.130 |
Frank | +0.122 |
oc | +0.121 |
Charles | +0.116 |
Neg logit contributions | |
auc | -0.111 |
bars | -0.097 |
line | -0.096 |
inter | -0.096 |
tubes | -0.096 |
Token | . |
Token ID | 13 |
Feature activation | -4.09e-01 |
Pos logit contributions | |
auc | +0.096 |
ains | +0.088 |
ays | +0.084 |
line | +0.083 |
lights | +0.082 |
Neg logit contributions | |
Amen | -0.100 |
defeat | -0.082 |
lder | -0.079 |
Frank | -0.076 |
read | -0.075 |
Token | |
Token ID | 220 |
Feature activation | +3.92e+00 |
Pos logit contributions | |
Amen | +0.011 |
lder | +0.008 |
Frank | +0.008 |
oc | +0.008 |
Mic | +0.008 |
Neg logit contributions | |
auc | -0.014 |
ains | -0.013 |
ays | -0.012 |
bars | -0.012 |
line | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -2.55e+00 |
Pos logit contributions | |
auc | +0.153 |
ays | +0.134 |
tubes | +0.134 |
bars | +0.132 |
lights | +0.132 |
Neg logit contributions | |
Amen | -0.082 |
Frank | -0.062 |
oc | -0.057 |
defeat | -0.054 |
Charles | -0.053 |
Token | to |
Token ID | 284 |
Feature activation | -4.43e+00 |
Pos logit contributions | |
Amen | +0.079 |
lder | +0.066 |
oc | +0.063 |
defeat | +0.062 |
Frank | +0.060 |
Neg logit contributions | |
auc | -0.087 |
ains | -0.077 |
ays | -0.076 |
bars | -0.076 |
line | -0.075 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
Amen | +0.151 |
lder | +0.125 |
oc | +0.117 |
defeat | +0.116 |
Frank | +0.116 |
Neg logit contributions | |
auc | -0.120 |
line | -0.105 |
ays | -0.103 |
bars | -0.103 |
ains | -0.102 |
Token | , |
Token ID | 11 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
Amen | +0.043 |
lder | +0.037 |
Frank | +0.035 |
Charles | +0.034 |
defeat | +0.034 |
Neg logit contributions | |
auc | -0.021 |
line | -0.019 |
bars | -0.018 |
ays | -0.018 |
ains | -0.017 |
Token | but |
Token ID | 475 |
Feature activation | +2.18e+00 |
Pos logit contributions | |
auc | +0.043 |
ays | +0.037 |
ains | +0.037 |
bars | +0.037 |
line | +0.037 |
Neg logit contributions | |
Amen | -0.054 |
lder | -0.045 |
defeat | -0.042 |
Frank | -0.042 |
oc | -0.042 |
Token | Lily |
Token ID | 20037 |
Feature activation | +4.63e-01 |
Pos logit contributions | |
auc | +0.052 |
ays | +0.044 |
ains | +0.043 |
bars | +0.042 |
line | +0.042 |
Neg logit contributions | |
Amen | -0.085 |
lder | -0.071 |
Frank | -0.067 |
defeat | -0.067 |
oc | -0.067 |
Token | noticed |
Token ID | 6810 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
auc | +0.011 |
line | +0.009 |
ays | +0.009 |
bars | +0.009 |
ains | +0.009 |
Neg logit contributions | |
Amen | -0.018 |
lder | -0.015 |
defeat | -0.014 |
Frank | -0.014 |
oc | -0.014 |
Token | that |
Token ID | 326 |
Feature activation | +2.16e+00 |
Pos logit contributions | |
Amen | +0.007 |
lder | +0.006 |
Frank | +0.006 |
read | +0.006 |
defeat | +0.005 |
Neg logit contributions | |
auc | -0.008 |
inter | -0.007 |
line | -0.007 |
ains | -0.007 |
ays | -0.007 |
Token | she |
Token ID | 673 |
Feature activation | -9.79e-01 |
Pos logit contributions | |
auc | +0.068 |
ains | +0.060 |
ays | +0.060 |
inter | +0.059 |
ash | +0.058 |
Neg logit contributions | |
Amen | -0.070 |
lder | -0.055 |
Frank | -0.055 |
read | -0.053 |
chat | -0.052 |
Token | never |
Token ID | 1239 |
Feature activation | +6.90e-01 |
Pos logit contributions | |
Amen | +0.038 |
lder | +0.033 |
defeat | +0.031 |
oc | +0.030 |
Frank | +0.030 |
Neg logit contributions | |
auc | -0.023 |
line | -0.019 |
ays | -0.019 |
bars | -0.019 |
ains | -0.019 |
Token | said |
Token ID | 531 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
auc | +0.021 |
ays | +0.019 |
ains | +0.019 |
lights | +0.019 |
bars | +0.018 |
Neg logit contributions | |
Amen | -0.020 |
read | -0.018 |
chat | -0.017 |
lder | -0.017 |
defeat | -0.016 |
Token | hello |
Token ID | 23748 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
Amen | +0.046 |
lder | +0.040 |
chat | +0.039 |
defeat | +0.038 |
read | +0.037 |
Neg logit contributions | |
auc | -0.057 |
ays | -0.052 |
ains | -0.050 |
ash | -0.050 |
inter | -0.048 |
Token | all |
Token ID | 477 |
Feature activation | +4.17e+00 |
Pos logit contributions | |
Amen | +0.019 |
defeat | +0.017 |
read | +0.017 |
chat | +0.016 |
bargain | +0.016 |
Neg logit contributions | |
auc | -0.025 |
line | -0.023 |
bars | -0.022 |
inter | -0.022 |
oes | -0.022 |
Token | of |
Token ID | 286 |
Feature activation | +2.47e+00 |
Pos logit contributions | |
auc | +0.132 |
unes | +0.130 |
line | +0.128 |
ains | +0.127 |
inter | +0.125 |
Neg logit contributions | |
Amen | -0.093 |
read | -0.072 |
oc | -0.068 |
chat | -0.067 |
defeat | -0.066 |
Token | her |
Token ID | 607 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
auc | +0.072 |
bars | +0.067 |
line | +0.065 |
ays | +0.064 |
inter | +0.064 |
Neg logit contributions | |
Amen | -0.071 |
defeat | -0.063 |
Mic | -0.061 |
read | -0.061 |
Frank | -0.060 |
Token | toys |
Token ID | 14958 |
Feature activation | -7.02e-01 |
Pos logit contributions | |
Amen | +0.042 |
lder | +0.033 |
oc | +0.033 |
defeat | +0.032 |
read | +0.031 |
Neg logit contributions | |
auc | -0.057 |
line | -0.054 |
bars | -0.051 |
ays | -0.051 |
inter | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | +8.67e-02 |
Pos logit contributions | |
Amen | +0.014 |
read | +0.013 |
defeat | +0.013 |
lder | +0.012 |
chat | +0.012 |
Neg logit contributions | |
auc | -0.024 |
ays | -0.023 |
inter | -0.021 |
ash | -0.021 |
ains | -0.021 |
Token | They |
Token ID | 1119 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
auc | +0.002 |
bars | +0.002 |
iness | +0.002 |
line | +0.002 |
ains | +0.002 |
Neg logit contributions | |
Amen | -0.003 |
lder | -0.002 |
read | -0.002 |
chat | -0.002 |
Mic | -0.002 |
Token | put |
Token ID | 1234 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
Amen | +0.007 |
lder | +0.006 |
read | +0.006 |
chat | +0.006 |
defeat | +0.006 |
Neg logit contributions | |
auc | -0.006 |
ays | -0.006 |
ains | -0.005 |
line | -0.005 |
bars | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
auc | +0.063 |
line | +0.059 |
bars | +0.056 |
inter | +0.055 |
ays | +0.053 |
Neg logit contributions | |
Amen | -0.060 |
defeat | -0.048 |
chat | -0.046 |
Mic | -0.046 |
read | -0.045 |
Token | dolls |
Token ID | 36062 |
Feature activation | -3.46e-01 |
Pos logit contributions | |
auc | +0.038 |
line | +0.034 |
ays | +0.034 |
bars | +0.033 |
ains | +0.033 |
Neg logit contributions | |
Amen | -0.042 |
lder | -0.034 |
defeat | -0.033 |
oc | -0.031 |
Mic | -0.031 |
Token | in |
Token ID | 287 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
Amen | +0.010 |
defeat | +0.009 |
read | +0.009 |
lder | +0.009 |
chat | +0.008 |
Neg logit contributions | |
auc | -0.010 |
ays | -0.009 |
inter | -0.008 |
ains | -0.008 |
ash | -0.008 |
Token | one |
Token ID | 530 |
Feature activation | +3.30e+00 |
Pos logit contributions | |
auc | +0.068 |
ays | +0.058 |
line | +0.058 |
bars | +0.057 |
ains | +0.057 |
Neg logit contributions | |
Amen | -0.091 |
lder | -0.076 |
defeat | -0.073 |
oc | -0.072 |
Frank | -0.072 |
Token | , |
Token ID | 11 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
Amen | +0.011 |
lder | +0.009 |
defeat | +0.008 |
oc | +0.008 |
Frank | +0.008 |
Neg logit contributions | |
auc | -0.014 |
bars | -0.013 |
ays | -0.013 |
line | -0.013 |
ains | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | +3.78e+00 |
Pos logit contributions | |
auc | +0.072 |
bars | +0.065 |
ains | +0.063 |
ays | +0.062 |
ash | +0.061 |
Neg logit contributions | |
Amen | -0.071 |
lder | -0.058 |
defeat | -0.058 |
Frank | -0.058 |
chat | -0.057 |
Token | raft |
Token ID | 31812 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
auc | +0.118 |
ays | +0.108 |
bars | +0.106 |
ash | +0.102 |
ains | +0.102 |
Neg logit contributions | |
Amen | -0.115 |
lder | -0.093 |
Frank | -0.091 |
defeat | -0.087 |
chat | -0.086 |
Token | started |
Token ID | 2067 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
auc | +0.038 |
ays | +0.032 |
bars | +0.032 |
ains | +0.032 |
line | +0.031 |
Neg logit contributions | |
Amen | -0.047 |
lder | -0.041 |
defeat | -0.040 |
oc | -0.039 |
Frank | -0.038 |
Token | to |
Token ID | 284 |
Feature activation | -5.24e+00 |
Pos logit contributions | |
Amen | +0.049 |
lder | +0.040 |
defeat | +0.038 |
Frank | +0.037 |
oc | +0.037 |
Neg logit contributions | |
auc | -0.050 |
ains | -0.044 |
ays | -0.044 |
line | -0.044 |
bars | -0.043 |
Token | move |
Token ID | 1445 |
Feature activation | +7.28e-01 |
Pos logit contributions | |
Amen | +0.182 |
lder | +0.152 |
oc | +0.142 |
Frank | +0.141 |
defeat | +0.140 |
Neg logit contributions | |
auc | -0.142 |
line | -0.123 |
bars | -0.121 |
ays | -0.121 |
ains | -0.121 |
Token | ! |
Token ID | 0 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
auc | +0.023 |
line | +0.021 |
ays | +0.021 |
bars | +0.021 |
ains | +0.020 |
Neg logit contributions | |
Amen | -0.020 |
defeat | -0.017 |
oc | -0.017 |
lder | -0.016 |
Frank | -0.016 |
Token | D |
Token ID | 360 |
Feature activation | +3.96e+00 |
Pos logit contributions | |
auc | +0.028 |
bars | +0.027 |
ains | +0.026 |
lights | +0.026 |
line | +0.025 |
Neg logit contributions | |
Amen | -0.036 |
lder | -0.033 |
oc | -0.031 |
defeat | -0.030 |
chat | -0.030 |
Token | ucky |
Token ID | 5309 |
Feature activation | +3.42e-02 |
Pos logit contributions | |
auc | +0.084 |
bars | +0.080 |
lights | +0.074 |
line | +0.074 |
ains | +0.072 |
Neg logit contributions | |
Amen | -0.146 |
lder | -0.130 |
oc | -0.125 |
Mic | -0.122 |
defeat | -0.121 |
Token | tried |
Token ID | 3088 |
Feature activation | -3.40e+00 |
Pos logit contributions | |
auc | +0.001 |
tubes | +0.001 |
line | +0.001 |
ains | +0.001 |
bars | +0.000 |
Neg logit contributions | |
Amen | -0.002 |
lder | -0.001 |
Frank | -0.001 |
Mic | -0.001 |
Charles | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | -5.15e+00 |
Pos logit contributions | |
Amen | +0.094 |
defeat | +0.077 |
lder | +0.076 |
Frank | +0.074 |
chat | +0.072 |
Neg logit contributions | |
auc | -0.115 |
ays | -0.100 |
ains | -0.099 |
bars | -0.098 |
line | -0.097 |
Token | ancient |
Token ID | 6156 |
Feature activation | -3.64e+00 |
Pos logit contributions | |
Amen | +0.038 |
lder | +0.033 |
defeat | +0.030 |
oc | +0.030 |
Frank | +0.030 |
Neg logit contributions | |
auc | -0.024 |
line | -0.020 |
bars | -0.020 |
tubes | -0.020 |
ains | -0.019 |
Token | cave |
Token ID | 11527 |
Feature activation | -2.56e+00 |
Pos logit contributions | |
Amen | +0.122 |
lder | +0.101 |
Frank | +0.094 |
defeat | +0.093 |
Mic | +0.093 |
Neg logit contributions | |
auc | -0.102 |
line | -0.087 |
tubes | -0.086 |
ains | -0.086 |
ays | -0.084 |
Token | . |
Token ID | 13 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
Amen | +0.078 |
lder | +0.064 |
oc | +0.062 |
defeat | +0.062 |
chat | +0.060 |
Neg logit contributions | |
auc | -0.074 |
ays | -0.069 |
idget | -0.068 |
inter | -0.067 |
bars | -0.067 |
Token | As |
Token ID | 1081 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
Amen | +0.004 |
lder | +0.003 |
Frank | +0.003 |
oc | +0.003 |
chat | +0.003 |
Neg logit contributions | |
auc | -0.003 |
ays | -0.002 |
bars | -0.002 |
ains | -0.002 |
line | -0.002 |
Token | they |
Token ID | 484 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
auc | +0.022 |
ays | +0.017 |
inter | +0.017 |
bars | +0.017 |
line | +0.016 |
Neg logit contributions | |
Amen | -0.054 |
lder | -0.047 |
Frank | -0.046 |
defeat | -0.045 |
Charles | -0.045 |
Token | approached |
Token ID | 10448 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
auc | +0.056 |
ays | +0.049 |
line | +0.049 |
bars | +0.049 |
ains | +0.048 |
Neg logit contributions | |
Amen | -0.064 |
lder | -0.053 |
defeat | -0.050 |
oc | -0.050 |
Frank | -0.049 |
Token | , |
Token ID | 11 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
Amen | +0.084 |
lder | +0.071 |
defeat | +0.068 |
Frank | +0.067 |
oc | +0.066 |
Neg logit contributions | |
auc | -0.057 |
ays | -0.048 |
bars | -0.048 |
line | -0.048 |
ains | -0.047 |
Token | they |
Token ID | 484 |
Feature activation | +8.29e-02 |
Pos logit contributions | |
auc | +0.023 |
ays | +0.019 |
bars | +0.019 |
line | +0.018 |
ains | +0.018 |
Neg logit contributions | |
Amen | -0.057 |
lder | -0.049 |
defeat | -0.047 |
Frank | -0.047 |
chat | -0.046 |
Token | noticed |
Token ID | 6810 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
auc | +0.003 |
ays | +0.002 |
bars | +0.002 |
line | +0.002 |
ains | +0.002 |
Neg logit contributions | |
Amen | -0.003 |
lder | -0.002 |
defeat | -0.002 |
Frank | -0.002 |
oc | -0.002 |
Token | a |
Token ID | 257 |
Feature activation | -3.73e+00 |
Pos logit contributions | |
Amen | +0.006 |
lder | +0.005 |
defeat | +0.005 |
Frank | +0.005 |
oc | +0.004 |
Neg logit contributions | |
auc | -0.006 |
ays | -0.005 |
line | -0.005 |
bars | -0.005 |
ains | -0.005 |
Token | long |
Token ID | 890 |
Feature activation | -3.99e+00 |
Pos logit contributions | |
Amen | +0.138 |
lder | +0.116 |
Frank | +0.106 |
arrest | +0.106 |
defeat | +0.104 |
Neg logit contributions | |
auc | -0.086 |
tubes | -0.076 |
tub | -0.075 |
line | -0.074 |
ains | -0.069 |
Token | on |
Token ID | 319 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
auc | +0.029 |
line | +0.025 |
bars | +0.025 |
ays | +0.025 |
ains | +0.024 |
Neg logit contributions | |
Amen | -0.040 |
defeat | -0.033 |
lder | -0.033 |
Frank | -0.032 |
chat | -0.032 |
Token | right |
Token ID | 826 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
auc | +0.028 |
line | +0.024 |
ays | +0.024 |
ains | +0.023 |
tubes | +0.023 |
Neg logit contributions | |
Amen | -0.051 |
lder | -0.043 |
Frank | -0.040 |
oc | -0.040 |
defeat | -0.039 |
Token | away |
Token ID | 1497 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
auc | +0.076 |
bars | +0.069 |
line | +0.067 |
ains | +0.066 |
inter | +0.066 |
Neg logit contributions | |
Amen | -0.032 |
defeat | -0.026 |
lder | -0.022 |
oc | -0.021 |
read | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | +2.69e-01 |
Pos logit contributions | |
Amen | +0.029 |
defeat | +0.026 |
oc | +0.024 |
lder | +0.024 |
Frank | +0.023 |
Neg logit contributions | |
auc | -0.030 |
bars | -0.027 |
ays | -0.027 |
ains | -0.026 |
inter | -0.026 |
Token | She |
Token ID | 1375 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
auc | +0.008 |
ains | +0.006 |
ays | +0.006 |
bars | +0.006 |
line | +0.006 |
Neg logit contributions | |
Amen | -0.009 |
lder | -0.008 |
oc | -0.007 |
Mic | -0.007 |
chat | -0.007 |
Token | could |
Token ID | 714 |
Feature activation | -4.60e+00 |
Pos logit contributions | |
Amen | +0.055 |
lder | +0.046 |
defeat | +0.043 |
Frank | +0.043 |
oc | +0.042 |
Neg logit contributions | |
auc | -0.042 |
ays | -0.036 |
line | -0.036 |
ains | -0.036 |
bars | -0.036 |
Token | see |
Token ID | 766 |
Feature activation | +6.20e-01 |
Pos logit contributions | |
Amen | +0.151 |
lder | +0.127 |
oc | +0.118 |
Frank | +0.116 |
Mic | +0.113 |
Neg logit contributions | |
auc | -0.131 |
line | -0.115 |
ays | -0.113 |
bars | -0.113 |
ains | -0.111 |
Token | everything |
Token ID | 2279 |
Feature activation | +6.97e+00 |
Pos logit contributions | |
auc | +0.018 |
inter | +0.017 |
line | +0.016 |
bars | +0.016 |
ays | +0.015 |
Neg logit contributions | |
Amen | -0.017 |
defeat | -0.015 |
lder | -0.015 |
Frank | -0.015 |
oc | -0.015 |
Token | clearly |
Token ID | 4084 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
auc | +0.229 |
inter | +0.228 |
ains | +0.207 |
ained | +0.206 |
ash | +0.202 |
Neg logit contributions | |
read | -0.152 |
defeat | -0.150 |
Amen | -0.148 |
chat | -0.147 |
Raven | -0.126 |
Token | now |
Token ID | 783 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
auc | +0.068 |
bars | +0.058 |
ays | +0.056 |
ains | +0.056 |
ours | +0.056 |
Neg logit contributions | |
Amen | -0.035 |
defeat | -0.033 |
chat | -0.032 |
read | -0.031 |
oc | -0.030 |
Token | ! |
Token ID | 0 |
Feature activation | +5.27e-01 |
Pos logit contributions | |
Amen | +0.028 |
read | +0.026 |
defeat | +0.026 |
lder | +0.025 |
chat | +0.024 |
Neg logit contributions | |
auc | -0.061 |
bars | -0.052 |
ash | -0.051 |
ays | -0.051 |
inter | -0.050 |
Token | a |
Token ID | 257 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
auc | +0.018 |
line | +0.016 |
tubes | +0.015 |
tub | +0.015 |
ains | +0.015 |
Neg logit contributions | |
Amen | -0.050 |
lder | -0.046 |
defeat | -0.043 |
oc | -0.043 |
Charles | -0.042 |
Token | time |
Token ID | 640 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
Amen | +0.069 |
lder | +0.058 |
defeat | +0.058 |
Frank | +0.056 |
Offic | +0.055 |
Neg logit contributions | |
auc | -0.028 |
inter | -0.023 |
ash | -0.023 |
ains | -0.023 |
ays | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
auc | +0.051 |
ays | +0.045 |
bars | +0.045 |
line | +0.045 |
ains | +0.044 |
Neg logit contributions | |
Amen | -0.041 |
lder | -0.033 |
defeat | -0.030 |
Frank | -0.030 |
oc | -0.030 |
Token | there |
Token ID | 612 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
auc | +0.055 |
line | +0.047 |
bars | +0.047 |
ays | +0.046 |
ains | +0.045 |
Neg logit contributions | |
Amen | -0.081 |
lder | -0.068 |
Frank | -0.064 |
defeat | -0.064 |
oc | -0.064 |
Token | was |
Token ID | 373 |
Feature activation | -4.55e-02 |
Pos logit contributions | |
auc | +0.058 |
bars | +0.052 |
line | +0.052 |
ays | +0.050 |
lights | +0.049 |
Neg logit contributions | |
Amen | -0.061 |
lder | -0.054 |
oc | -0.052 |
chat | -0.049 |
defeat | -0.049 |
Token | a |
Token ID | 257 |
Feature activation | -4.36e+00 |
Pos logit contributions | |
Amen | +0.002 |
lder | +0.002 |
defeat | +0.002 |
oc | +0.002 |
Charles | +0.001 |
Neg logit contributions | |
auc | -0.001 |
tubes | -0.001 |
ays | -0.001 |
ained | -0.001 |
ains | -0.001 |
Token | little |
Token ID | 1310 |
Feature activation | -3.26e+00 |
Pos logit contributions | |
Amen | +0.157 |
Frank | +0.128 |
defeat | +0.128 |
lder | +0.125 |
Charles | +0.123 |
Neg logit contributions | |
auc | -0.101 |
line | -0.088 |
tubes | -0.084 |
tub | -0.084 |
ains | -0.083 |
Token | girl |
Token ID | 2576 |
Feature activation | -3.90e+00 |
Pos logit contributions | |
Amen | +0.115 |
lder | +0.096 |
defeat | +0.094 |
Frank | +0.091 |
Charles | +0.089 |
Neg logit contributions | |
auc | -0.086 |
line | -0.072 |
tubes | -0.070 |
ays | -0.070 |
bars | -0.070 |
Token | named |
Token ID | 3706 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
Amen | +0.100 |
lder | +0.079 |
defeat | +0.074 |
Frank | +0.074 |
Charles | +0.071 |
Neg logit contributions | |
auc | -0.138 |
line | -0.121 |
ays | -0.120 |
ains | -0.119 |
bars | -0.119 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
auc | +0.033 |
ays | +0.026 |
bars | +0.025 |
line | +0.025 |
ains | +0.025 |
Neg logit contributions | |
Amen | -0.090 |
lder | -0.080 |
defeat | -0.075 |
Frank | -0.075 |
oc | -0.074 |
Token | . |
Token ID | 13 |
Feature activation | +7.42e-01 |
Pos logit contributions | |
auc | +0.004 |
ays | +0.004 |
line | +0.004 |
bars | +0.003 |
ains | +0.003 |
Neg logit contributions | |
Amen | -0.007 |
lder | -0.006 |
oc | -0.005 |
defeat | -0.005 |
Frank | -0.005 |
Token | They |
Token ID | 1119 |
Feature activation | -9.38e-01 |
Pos logit contributions | |
Amen | +0.095 |
lder | +0.081 |
oc | +0.076 |
defeat | +0.076 |
Frank | +0.075 |
Neg logit contributions | |
auc | -0.048 |
bars | -0.039 |
ains | -0.039 |
line | -0.039 |
ays | -0.038 |
Token | take |
Token ID | 1011 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
Amen | +0.031 |
lder | +0.026 |
Frank | +0.025 |
oc | +0.024 |
Mic | +0.024 |
Neg logit contributions | |
auc | -0.026 |
line | -0.023 |
bars | -0.023 |
ains | -0.022 |
tubes | -0.022 |
Token | Anna |
Token ID | 11735 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
auc | +0.047 |
line | +0.040 |
bars | +0.040 |
ays | +0.039 |
ains | +0.039 |
Neg logit contributions | |
Amen | -0.076 |
lder | -0.065 |
defeat | -0.061 |
oc | -0.061 |
Frank | -0.060 |
Token | and |
Token ID | 290 |
Feature activation | +5.89e-01 |
Pos logit contributions | |
Amen | +0.085 |
lder | +0.071 |
Frank | +0.066 |
Mic | +0.066 |
defeat | +0.066 |
Neg logit contributions | |
auc | -0.047 |
line | -0.042 |
ays | -0.042 |
bars | -0.040 |
inter | -0.040 |
Token | Ben |
Token ID | 3932 |
Feature activation | -9.61e-01 |
Pos logit contributions | |
auc | +0.009 |
ays | +0.007 |
ains | +0.007 |
line | +0.007 |
bars | +0.007 |
Neg logit contributions | |
Amen | -0.027 |
lder | -0.023 |
defeat | -0.023 |
oc | -0.022 |
Frank | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | -4.57e+00 |
Pos logit contributions | |
Amen | +0.036 |
lder | +0.031 |
Frank | +0.029 |
Mic | +0.029 |
oc | +0.028 |
Neg logit contributions | |
auc | -0.023 |
line | -0.020 |
bars | -0.020 |
ays | -0.020 |
ains | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | +9.04e-01 |
Pos logit contributions | |
Amen | +0.202 |
lder | +0.180 |
oc | +0.173 |
Frank | +0.168 |
Mic | +0.165 |
Neg logit contributions | |
auc | -0.075 |
line | -0.063 |
tubes | -0.059 |
ains | -0.058 |
ays | -0.056 |
Token | car |
Token ID | 1097 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
auc | +0.022 |
line | +0.018 |
ays | +0.018 |
ains | +0.018 |
bars | +0.018 |
Neg logit contributions | |
Amen | -0.034 |
lder | -0.029 |
oc | -0.027 |
Frank | -0.027 |
defeat | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
Amen | +0.013 |
lder | +0.012 |
Frank | +0.011 |
Mic | +0.011 |
defeat | +0.011 |
Neg logit contributions | |
auc | -0.006 |
line | -0.006 |
bars | -0.005 |
ays | -0.005 |
ains | -0.005 |
Token | Anna |
Token ID | 11735 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
auc | +0.027 |
bars | +0.022 |
ains | +0.022 |
line | +0.022 |
ays | +0.022 |
Neg logit contributions | |
Amen | -0.053 |
lder | -0.044 |
oc | -0.042 |
defeat | -0.042 |
Frank | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
Amen | +0.011 |
lder | +0.009 |
Frank | +0.009 |
defeat | +0.008 |
Mic | +0.008 |
Neg logit contributions | |
auc | -0.007 |
ays | -0.006 |
bars | -0.006 |
line | -0.006 |
ains | -0.006 |
Token | they |
Token ID | 484 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
auc | +0.065 |
bars | +0.058 |
ays | +0.058 |
inter | +0.056 |
ains | +0.056 |
Neg logit contributions | |
Amen | -0.067 |
Frank | -0.059 |
defeat | -0.057 |
chat | -0.056 |
Charles | -0.055 |
Token | were |
Token ID | 547 |
Feature activation | +3.28e-01 |
Pos logit contributions | |
auc | +0.004 |
line | +0.003 |
bars | +0.003 |
ains | +0.003 |
ays | +0.003 |
Neg logit contributions | |
Amen | -0.006 |
lder | -0.005 |
Frank | -0.005 |
oc | -0.005 |
defeat | -0.005 |
Token | welcomed |
Token ID | 15846 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
auc | +0.009 |
line | +0.008 |
bars | +0.008 |
ains | +0.008 |
ays | +0.008 |
Neg logit contributions | |
Amen | -0.011 |
lder | -0.009 |
oc | -0.009 |
Mic | -0.009 |
defeat | -0.009 |
Token | by |
Token ID | 416 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
Amen | +0.073 |
lder | +0.062 |
Frank | +0.059 |
defeat | +0.059 |
oc | +0.058 |
Neg logit contributions | |
auc | -0.058 |
ays | -0.050 |
bars | -0.050 |
ains | -0.050 |
line | -0.050 |
Token | a |
Token ID | 257 |
Feature activation | -3.68e+00 |
Pos logit contributions | |
auc | +0.040 |
ays | +0.034 |
bars | +0.034 |
line | +0.034 |
ains | +0.033 |
Neg logit contributions | |
Amen | -0.062 |
lder | -0.053 |
defeat | -0.051 |
oc | -0.050 |
Frank | -0.050 |
Token | friendly |
Token ID | 8030 |
Feature activation | -5.21e+00 |
Pos logit contributions | |
Amen | +0.121 |
lder | +0.103 |
Frank | +0.095 |
Mic | +0.093 |
Charles | +0.092 |
Neg logit contributions | |
auc | -0.099 |
line | -0.086 |
tubes | -0.081 |
bars | -0.081 |
inter | -0.081 |
Token | nurse |
Token ID | 15849 |
Feature activation | -3.98e+00 |
Pos logit contributions | |
Amen | +0.153 |
lder | +0.122 |
Frank | +0.114 |
defeat | +0.111 |
Mic | +0.111 |
Neg logit contributions | |
auc | -0.168 |
line | -0.147 |
bars | -0.145 |
inter | -0.142 |
ains | -0.142 |
Token | . |
Token ID | 13 |
Feature activation | +2.95e-01 |
Pos logit contributions | |
Amen | +0.160 |
lder | +0.141 |
Frank | +0.133 |
Mic | +0.131 |
defeat | +0.131 |
Neg logit contributions | |
auc | -0.082 |
line | -0.069 |
bars | -0.066 |
ays | -0.066 |
ains | -0.065 |
Token | She |
Token ID | 1375 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
auc | +0.008 |
bars | +0.007 |
ains | +0.007 |
ays | +0.007 |
line | +0.007 |
Neg logit contributions | |
Amen | -0.010 |
lder | -0.009 |
Frank | -0.008 |
oc | -0.008 |
defeat | -0.008 |
Token | said |
Token ID | 531 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
Amen | +0.058 |
lder | +0.047 |
defeat | +0.044 |
Frank | +0.044 |
oc | +0.044 |
Neg logit contributions | |
auc | -0.048 |
bars | -0.041 |
line | -0.041 |
ays | -0.041 |
ains | -0.041 |
Token | , |
Token ID | 11 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
Amen | +0.066 |
lder | +0.054 |
oc | +0.051 |
defeat | +0.050 |
Frank | +0.050 |
Neg logit contributions | |
auc | -0.058 |
line | -0.051 |
bars | -0.050 |
ays | -0.050 |
ains | -0.049 |
Token | , |
Token ID | 11 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
auc | +0.014 |
ays | +0.013 |
bars | +0.013 |
ains | +0.013 |
ours | +0.012 |
Neg logit contributions | |
Amen | -0.011 |
defeat | -0.009 |
lder | -0.009 |
oc | -0.009 |
read | -0.009 |
Token | beef |
Token ID | 12023 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
auc | +0.046 |
ays | +0.042 |
ains | +0.040 |
bars | +0.039 |
line | +0.039 |
Neg logit contributions | |
Amen | -0.055 |
lder | -0.046 |
defeat | -0.045 |
Frank | -0.045 |
Mic | -0.044 |
Token | , |
Token ID | 11 |
Feature activation | +1.98e+00 |
Pos logit contributions | |
Amen | +0.044 |
lder | +0.037 |
defeat | +0.036 |
oc | +0.036 |
chat | +0.034 |
Neg logit contributions | |
auc | -0.043 |
bars | -0.038 |
ays | -0.038 |
ains | -0.037 |
line | -0.037 |
Token | and |
Token ID | 290 |
Feature activation | -3.87e-01 |
Pos logit contributions | |
auc | +0.059 |
ays | +0.053 |
ains | +0.050 |
bars | +0.050 |
ash | +0.050 |
Neg logit contributions | |
Amen | -0.061 |
defeat | -0.051 |
lder | -0.050 |
oc | -0.050 |
chat | -0.049 |
Token | decided |
Token ID | 3066 |
Feature activation | -4.28e+00 |
Pos logit contributions | |
Amen | +0.014 |
lder | +0.012 |
defeat | +0.012 |
Frank | +0.011 |
oc | +0.011 |
Neg logit contributions | |
auc | -0.010 |
ays | -0.008 |
line | -0.008 |
bars | -0.008 |
ains | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | -5.02e+00 |
Pos logit contributions | |
Amen | +0.137 |
lder | +0.113 |
defeat | +0.106 |
Frank | +0.105 |
oc | +0.104 |
Neg logit contributions | |
auc | -0.127 |
ays | -0.111 |
ains | -0.110 |
line | -0.110 |
bars | -0.110 |
Token | collect |
Token ID | 2824 |
Feature activation | +3.56e+00 |
Pos logit contributions | |
Amen | +0.160 |
lder | +0.132 |
defeat | +0.122 |
Frank | +0.122 |
oc | +0.121 |
Neg logit contributions | |
auc | -0.148 |
line | -0.129 |
ays | -0.129 |
bars | -0.128 |
ains | -0.127 |
Token | beef |
Token ID | 12023 |
Feature activation | -2.72e+00 |
Pos logit contributions | |
auc | +0.111 |
line | +0.102 |
bars | +0.100 |
ains | +0.099 |
idget | +0.093 |
Neg logit contributions | |
Amen | -0.101 |
defeat | -0.090 |
oc | -0.090 |
Frank | -0.082 |
", | -0.081 |
Token | recipes |
Token ID | 14296 |
Feature activation | +7.14e-01 |
Pos logit contributions | |
Amen | +0.096 |
lder | +0.081 |
defeat | +0.080 |
oc | +0.078 |
Frank | +0.076 |
Neg logit contributions | |
auc | -0.069 |
bars | -0.060 |
ays | -0.059 |
ains | -0.059 |
line | -0.057 |
Token | . |
Token ID | 13 |
Feature activation | -4.34e-01 |
Pos logit contributions | |
auc | +0.021 |
ains | +0.019 |
inter | +0.019 |
ays | +0.019 |
bars | +0.018 |
Neg logit contributions | |
Amen | -0.021 |
defeat | -0.017 |
", | -0.017 |
oc | -0.017 |
Frank | -0.017 |
Token | Tim |
Token ID | 5045 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
Amen | +0.015 |
lder | +0.013 |
oc | +0.012 |
Frank | +0.012 |
defeat | +0.012 |
Neg logit contributions | |
auc | -0.012 |
ays | -0.010 |
ains | -0.010 |
line | -0.010 |
bars | -0.010 |