Token | and |
Token ID | 290 |
Feature activation | +3.46e-02 |
Pos logit contributions | |
lor | +0.039 |
actus | +0.038 |
� | +0.038 |
sky | +0.035 |
course | +0.032 |
Neg logit contributions | |
Jill | -0.038 |
type | -0.035 |
kil | -0.035 |
ooks | -0.034 |
Jen | -0.033 |
Token | make |
Token ID | 787 |
Feature activation | -2.88e-02 |
Pos logit contributions | |
lor | +0.063 |
actus | +0.062 |
� | +0.062 |
sky | +0.059 |
course | +0.057 |
Neg logit contributions | |
Jill | -0.044 |
kil | -0.041 |
ooks | -0.039 |
tim | -0.038 |
ots | -0.038 |
Token | better |
Token ID | 1365 |
Feature activation | +2.66e-02 |
Pos logit contributions | |
Jill | +0.042 |
kil | +0.039 |
type | +0.038 |
Jen | +0.037 |
ooks | +0.036 |
Neg logit contributions | |
lor | -0.049 |
actus | -0.048 |
� | -0.046 |
sky | -0.044 |
course | -0.042 |
Token | choices |
Token ID | 7747 |
Feature activation | -3.14e-02 |
Pos logit contributions | |
actus | +0.037 |
lor | +0.037 |
� | +0.035 |
sky | +0.035 |
skies | +0.031 |
Neg logit contributions | |
Jill | -0.044 |
kil | -0.041 |
ots | -0.041 |
type | -0.040 |
Matt | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
Jill | +0.052 |
Jen | +0.048 |
ooks | +0.048 |
type | +0.047 |
kil | +0.046 |
Neg logit contributions | |
lor | -0.049 |
� | -0.047 |
actus | -0.046 |
sky | -0.043 |
course | -0.041 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
Jill | +0.024 |
Jen | +0.022 |
like | +0.021 |
theirs | +0.021 |
ooks | +0.021 |
Neg logit contributions | |
actus | -0.025 |
lor | -0.024 |
� | -0.022 |
poster | -0.021 |
sky | -0.020 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.33e-02 |
Pos logit contributions | |
actus | +0.200 |
lor | +0.196 |
ican | +0.178 |
� | +0.178 |
poster | +0.169 |
Neg logit contributions | |
Jill | -0.176 |
like | -0.155 |
Jen | -0.154 |
Mia | -0.150 |
Jane | -0.149 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.14e-02 |
Pos logit contributions | |
lor | +0.099 |
� | +0.094 |
actus | +0.093 |
ican | +0.085 |
sky | +0.084 |
Neg logit contributions | |
Jill | -0.050 |
Jen | -0.040 |
theirs | -0.038 |
type | -0.037 |
Liz | -0.037 |
Token | a |
Token ID | 257 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
Jill | +0.011 |
like | +0.011 |
type | +0.009 |
theirs | +0.009 |
Jen | +0.009 |
Neg logit contributions | |
lor | -0.027 |
actus | -0.027 |
� | -0.026 |
ican | -0.024 |
eros | -0.024 |
Token | time |
Token ID | 640 |
Feature activation | +5.23e-02 |
Pos logit contributions | |
lor | +0.025 |
actus | +0.025 |
� | +0.024 |
sky | +0.022 |
ican | +0.022 |
Neg logit contributions | |
Jill | -0.015 |
type | -0.014 |
like | -0.013 |
kil | -0.013 |
tim | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | -9.40e-03 |
Pos logit contributions | |
lor | +0.088 |
actus | +0.088 |
� | +0.081 |
sky | +0.076 |
ican | +0.071 |
Neg logit contributions | |
Jill | -0.085 |
Jen | -0.075 |
type | -0.075 |
like | -0.074 |
kil | -0.073 |
Token | the |
Token ID | 262 |
Feature activation | +5.10e-02 |
Pos logit contributions | |
� | +0.026 |
lor | +0.026 |
actus | +0.025 |
sky | +0.025 |
course | +0.024 |
Neg logit contributions | |
Jill | -0.024 |
kil | -0.023 |
type | -0.022 |
rated | -0.021 |
ots | -0.021 |
Token | best |
Token ID | 1266 |
Feature activation | -1.66e-02 |
Pos logit contributions | |
lor | +0.106 |
actus | +0.102 |
� | +0.100 |
eros | +0.089 |
sky | +0.088 |
Neg logit contributions | |
Jill | -0.061 |
type | -0.056 |
like | -0.056 |
neat | -0.054 |
kil | -0.052 |
Token | of |
Token ID | 286 |
Feature activation | -6.97e-02 |
Pos logit contributions | |
Jill | +0.021 |
type | +0.018 |
kil | +0.018 |
Jen | +0.017 |
theirs | +0.017 |
Neg logit contributions | |
lor | -0.033 |
actus | -0.033 |
� | -0.032 |
sky | -0.030 |
course | -0.028 |
Token | friends |
Token ID | 2460 |
Feature activation | -7.69e-03 |
Pos logit contributions | |
Jill | +0.107 |
type | +0.095 |
kil | +0.093 |
Jen | +0.093 |
theirs | +0.093 |
Neg logit contributions | |
lor | -0.124 |
actus | -0.120 |
� | -0.117 |
sky | -0.107 |
poster | -0.098 |
Token | . |
Token ID | 13 |
Feature activation | -6.98e-03 |
Pos logit contributions | |
Jill | +0.013 |
type | +0.012 |
kil | +0.012 |
ooks | +0.012 |
Jen | +0.012 |
Neg logit contributions | |
lor | -0.011 |
actus | -0.011 |
� | -0.011 |
sky | -0.010 |
course | -0.009 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
Jill | +0.013 |
type | +0.012 |
Jen | +0.012 |
kil | +0.011 |
like | +0.011 |
Neg logit contributions | |
lor | -0.010 |
actus | -0.010 |
� | -0.009 |
sky | -0.008 |
poster | -0.008 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.23e-02 |
Pos logit contributions | |
lor | +0.190 |
actus | +0.188 |
� | +0.174 |
ican | +0.167 |
poster | +0.160 |
Neg logit contributions | |
Jill | -0.183 |
like | -0.164 |
Jen | -0.159 |
tim | -0.156 |
Jane | -0.155 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
lor | +0.091 |
� | +0.086 |
actus | +0.086 |
sky | +0.078 |
ican | +0.077 |
Neg logit contributions | |
Jill | -0.052 |
Jen | -0.043 |
type | -0.042 |
theirs | -0.041 |
ooks | -0.041 |
Token | a |
Token ID | 257 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
Jill | +0.016 |
like | +0.014 |
type | +0.014 |
Jen | +0.013 |
theirs | +0.013 |
Neg logit contributions | |
lor | -0.030 |
actus | -0.030 |
� | -0.029 |
sky | -0.026 |
ican | -0.026 |
Token | time |
Token ID | 640 |
Feature activation | +5.34e-02 |
Pos logit contributions | |
lor | +0.024 |
actus | +0.023 |
� | +0.023 |
sky | +0.021 |
eros | +0.020 |
Neg logit contributions | |
Jill | -0.015 |
type | -0.014 |
like | -0.013 |
kil | -0.013 |
tim | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | -8.57e-03 |
Pos logit contributions | |
lor | +0.101 |
actus | +0.100 |
� | +0.089 |
sky | +0.086 |
ican | +0.083 |
Neg logit contributions | |
Jill | -0.080 |
Jen | -0.070 |
like | -0.069 |
type | -0.067 |
tim | -0.067 |
Token | know |
Token ID | 760 |
Feature activation | +2.15e-02 |
Pos logit contributions | |
ooks | +0.012 |
kil | +0.011 |
Jill | +0.011 |
Matt | +0.010 |
ots | +0.010 |
Neg logit contributions | |
actus | -0.016 |
sky | -0.015 |
lor | -0.015 |
� | -0.014 |
course | -0.014 |
Token | what |
Token ID | 644 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
lor | +0.041 |
� | +0.040 |
course | +0.040 |
actus | +0.039 |
sky | +0.038 |
Neg logit contributions | |
Jill | -0.023 |
kil | -0.023 |
type | -0.023 |
ooks | -0.022 |
tim | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | +9.28e-03 |
Pos logit contributions | |
ooks | +0.022 |
kil | +0.022 |
Jill | +0.022 |
rated | +0.021 |
ots | +0.021 |
Neg logit contributions | |
lor | -0.015 |
� | -0.015 |
actus | -0.015 |
sky | -0.015 |
course | -0.014 |
Token | do |
Token ID | 466 |
Feature activation | +5.97e-02 |
Pos logit contributions | |
actus | +0.017 |
lor | +0.017 |
� | +0.016 |
sky | +0.016 |
course | +0.015 |
Neg logit contributions | |
Jill | -0.012 |
kil | -0.011 |
ooks | -0.011 |
type | -0.010 |
tim | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -9.76e-03 |
Pos logit contributions | |
lor | +0.103 |
actus | +0.100 |
� | +0.100 |
sky | +0.093 |
course | +0.088 |
Neg logit contributions | |
Jill | -0.085 |
kil | -0.079 |
type | -0.078 |
ooks | -0.077 |
Jen | -0.075 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
Jill | +0.017 |
Jen | +0.014 |
Jane | +0.014 |
Liz | +0.014 |
Matt | +0.014 |
Neg logit contributions | |
lor | -0.017 |
actus | -0.017 |
� | -0.015 |
eros | -0.015 |
rot | -0.015 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.31e-02 |
Pos logit contributions | |
actus | +0.182 |
lor | +0.181 |
� | +0.163 |
ican | +0.159 |
poster | +0.152 |
Neg logit contributions | |
Jill | -0.192 |
like | -0.168 |
Jen | -0.166 |
Jane | -0.164 |
Mia | -0.160 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.25e-02 |
Pos logit contributions | |
lor | +0.089 |
actus | +0.084 |
� | +0.083 |
ican | +0.075 |
sky | +0.075 |
Neg logit contributions | |
Jill | -0.058 |
Jen | -0.048 |
type | -0.046 |
theirs | -0.046 |
Liz | -0.046 |
Token | a |
Token ID | 257 |
Feature activation | +1.40e-02 |
Pos logit contributions | |
Jill | +0.013 |
like | +0.012 |
type | +0.011 |
neat | +0.010 |
Jen | +0.010 |
Neg logit contributions | |
lor | -0.029 |
actus | -0.029 |
� | -0.028 |
ican | -0.026 |
eros | -0.025 |
Token | time |
Token ID | 640 |
Feature activation | +5.42e-02 |
Pos logit contributions | |
lor | +0.028 |
actus | +0.028 |
� | +0.026 |
sky | +0.025 |
eros | +0.024 |
Neg logit contributions | |
Jill | -0.017 |
type | -0.016 |
like | -0.015 |
kil | -0.015 |
neat | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | -4.25e-03 |
Pos logit contributions | |
actus | +0.098 |
lor | +0.097 |
� | +0.087 |
sky | +0.084 |
ican | +0.079 |
Neg logit contributions | |
Jill | -0.086 |
Jen | -0.075 |
like | -0.072 |
Jane | -0.072 |
type | -0.072 |
Token | like |
Token ID | 588 |
Feature activation | -3.20e-02 |
Pos logit contributions | |
lor | +0.120 |
actus | +0.119 |
� | +0.118 |
sky | +0.109 |
course | +0.104 |
Neg logit contributions | |
Jill | -0.067 |
kil | -0.059 |
type | -0.058 |
Jen | -0.056 |
ooks | -0.055 |
Token | to |
Token ID | 284 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
Jill | +0.051 |
kil | +0.048 |
tim | +0.048 |
type | +0.046 |
ooks | +0.046 |
Neg logit contributions | |
actus | -0.048 |
lor | -0.048 |
� | -0.047 |
sky | -0.045 |
course | -0.043 |
Token | be |
Token ID | 307 |
Feature activation | -3.41e-02 |
Pos logit contributions | |
� | +0.022 |
actus | +0.021 |
lor | +0.021 |
sky | +0.020 |
course | +0.019 |
Neg logit contributions | |
Jill | -0.014 |
kil | -0.012 |
ooks | -0.012 |
theirs | -0.011 |
Jen | -0.011 |
Token | treated |
Token ID | 5716 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
kil | +0.055 |
Jamie | +0.053 |
Matt | +0.052 |
rated | +0.052 |
tim | +0.052 |
Neg logit contributions | |
sky | -0.043 |
actus | -0.042 |
handed | -0.040 |
bird | -0.040 |
course | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | -6.07e-03 |
Pos logit contributions | |
Jill | +0.022 |
kil | +0.021 |
type | +0.021 |
Jen | +0.020 |
tim | +0.020 |
Neg logit contributions | |
lor | -0.024 |
actus | -0.024 |
� | -0.023 |
sky | -0.022 |
course | -0.021 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
Jill | +0.008 |
Jen | +0.007 |
type | +0.007 |
ooks | +0.007 |
like | +0.007 |
Neg logit contributions | |
lor | -0.012 |
actus | -0.012 |
� | -0.011 |
sky | -0.010 |
poster | -0.010 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.29e-02 |
Pos logit contributions | |
actus | +0.191 |
lor | +0.190 |
� | +0.170 |
ican | +0.167 |
poster | +0.162 |
Neg logit contributions | |
Jill | -0.184 |
like | -0.161 |
Jen | -0.159 |
Jane | -0.156 |
Mia | -0.153 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
lor | +0.093 |
actus | +0.088 |
� | +0.088 |
ican | +0.079 |
sky | +0.079 |
Neg logit contributions | |
Jill | -0.053 |
Jen | -0.044 |
theirs | -0.042 |
type | -0.041 |
Liz | -0.041 |
Token | a |
Token ID | 257 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
Jill | +0.013 |
like | +0.012 |
type | +0.011 |
theirs | +0.011 |
neat | +0.011 |
Neg logit contributions | |
lor | -0.026 |
actus | -0.025 |
� | -0.025 |
ican | -0.023 |
eros | -0.022 |
Token | time |
Token ID | 640 |
Feature activation | +5.42e-02 |
Pos logit contributions | |
lor | +0.026 |
actus | +0.025 |
� | +0.024 |
sky | +0.023 |
ican | +0.022 |
Neg logit contributions | |
Jill | -0.017 |
type | -0.015 |
like | -0.015 |
kil | -0.015 |
neat | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | -6.49e-03 |
Pos logit contributions | |
actus | +0.102 |
lor | +0.101 |
� | +0.090 |
sky | +0.087 |
ican | +0.083 |
Neg logit contributions | |
Jill | -0.083 |
Jen | -0.071 |
like | -0.070 |
Jane | -0.069 |
type | -0.069 |
Token | fun |
Token ID | 1257 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
lor | +0.043 |
actus | +0.041 |
� | +0.041 |
sky | +0.039 |
course | +0.037 |
Neg logit contributions | |
Jill | -0.030 |
kil | -0.027 |
type | -0.027 |
ooks | -0.026 |
Jen | -0.025 |
Token | on |
Token ID | 319 |
Feature activation | +7.21e-03 |
Pos logit contributions | |
lor | +0.023 |
actus | +0.022 |
� | +0.022 |
sky | +0.020 |
course | +0.019 |
Neg logit contributions | |
Jill | -0.017 |
type | -0.015 |
kil | -0.015 |
ooks | -0.014 |
Jen | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | +3.26e-02 |
Pos logit contributions | |
lor | +0.017 |
actus | +0.017 |
� | +0.016 |
sky | +0.015 |
rot | +0.014 |
Neg logit contributions | |
Jill | -0.007 |
Jen | -0.006 |
type | -0.006 |
kil | -0.006 |
theirs | -0.006 |
Token | ride |
Token ID | 6594 |
Feature activation | +2.99e-02 |
Pos logit contributions | |
lor | +0.060 |
� | +0.058 |
actus | +0.053 |
rot | +0.049 |
pillar | +0.046 |
Neg logit contributions | |
type | -0.044 |
sunny | -0.043 |
neat | -0.043 |
kil | -0.042 |
Jill | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | -8.56e-03 |
Pos logit contributions | |
lor | +0.052 |
actus | +0.051 |
� | +0.050 |
sky | +0.048 |
course | +0.045 |
Neg logit contributions | |
Jill | -0.043 |
kil | -0.039 |
type | -0.039 |
ooks | -0.038 |
Jen | -0.037 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
Jill | +0.012 |
Jen | +0.011 |
like | +0.011 |
Joe | +0.010 |
Jane | +0.010 |
Neg logit contributions | |
lor | -0.016 |
actus | -0.015 |
� | -0.014 |
poster | -0.014 |
rot | -0.014 |
Token | Once |
Token ID | 7454 |
Feature activation | +3.81e-02 |
Pos logit contributions | |
lor | +0.203 |
actus | +0.203 |
� | +0.188 |
ican | +0.187 |
eros | +0.177 |
Neg logit contributions | |
Jill | -0.170 |
like | -0.148 |
Jen | -0.146 |
Mia | -0.145 |
Jane | -0.143 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
lor | +0.089 |
� | +0.084 |
actus | +0.082 |
ican | +0.076 |
sky | +0.075 |
Neg logit contributions | |
Jill | -0.043 |
Jen | -0.035 |
theirs | -0.032 |
Liz | -0.031 |
like | -0.031 |
Token | a |
Token ID | 257 |
Feature activation | +8.73e-03 |
Pos logit contributions | |
Jill | +0.013 |
like | +0.013 |
type | +0.011 |
sunny | +0.010 |
Jen | +0.010 |
Neg logit contributions | |
lor | -0.034 |
actus | -0.034 |
� | -0.033 |
ican | -0.031 |
eros | -0.031 |
Token | time |
Token ID | 640 |
Feature activation | +4.97e-02 |
Pos logit contributions | |
lor | +0.019 |
actus | +0.018 |
� | +0.017 |
sky | +0.016 |
eros | +0.016 |
Neg logit contributions | |
Jill | -0.010 |
type | -0.009 |
like | -0.009 |
tim | -0.009 |
kil | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
actus | +0.094 |
lor | +0.094 |
� | +0.084 |
sky | +0.081 |
ican | +0.078 |
Neg logit contributions | |
Jill | -0.074 |
Jen | -0.065 |
like | -0.063 |
type | -0.061 |
Jane | -0.061 |
Token | take |
Token ID | 1011 |
Feature activation | -3.88e-02 |
Pos logit contributions | |
lor | +0.022 |
actus | +0.022 |
� | +0.022 |
sky | +0.020 |
course | +0.019 |
Neg logit contributions | |
Jill | -0.015 |
type | -0.014 |
kil | -0.014 |
Jen | -0.013 |
theirs | -0.013 |
Token | care |
Token ID | 1337 |
Feature activation | -2.79e-02 |
Pos logit contributions | |
Jill | +0.051 |
type | +0.044 |
kil | +0.043 |
Jen | +0.043 |
like | +0.043 |
Neg logit contributions | |
lor | -0.075 |
actus | -0.075 |
� | -0.073 |
sky | -0.067 |
poster | -0.063 |
Token | of |
Token ID | 286 |
Feature activation | -5.73e-02 |
Pos logit contributions | |
Jill | +0.040 |
kil | +0.037 |
type | +0.037 |
Jen | +0.035 |
Matt | +0.035 |
Neg logit contributions | |
lor | -0.049 |
actus | -0.047 |
� | -0.045 |
sky | -0.044 |
course | -0.040 |
Token | her |
Token ID | 607 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
Jill | +0.074 |
kil | +0.070 |
type | +0.066 |
ooks | +0.065 |
tim | +0.062 |
Neg logit contributions | |
� | -0.106 |
lor | -0.106 |
actus | -0.104 |
sky | -0.097 |
course | -0.095 |
Token | . |
Token ID | 13 |
Feature activation | -4.32e-03 |
Pos logit contributions | |
lor | +0.022 |
actus | +0.021 |
� | +0.021 |
sky | +0.019 |
course | +0.018 |
Neg logit contributions | |
Jill | -0.017 |
type | -0.015 |
kil | -0.015 |
Jen | -0.015 |
theirs | -0.015 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
Jill | +0.007 |
Jen | +0.006 |
Liz | +0.006 |
ooks | +0.006 |
like | +0.006 |
Neg logit contributions | |
lor | -0.007 |
actus | -0.007 |
� | -0.007 |
ican | -0.007 |
rot | -0.006 |
Token | Once |
Token ID | 7454 |
Feature activation | +3.93e-02 |
Pos logit contributions | |
lor | +0.183 |
actus | +0.182 |
� | +0.168 |
ican | +0.163 |
eros | +0.155 |
Neg logit contributions | |
Jill | -0.186 |
like | -0.167 |
Jen | -0.163 |
tim | -0.159 |
Jane | -0.158 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
lor | +0.089 |
actus | +0.084 |
� | +0.084 |
ican | +0.076 |
sky | +0.076 |
Neg logit contributions | |
Jill | -0.045 |
Jen | -0.036 |
type | -0.035 |
like | -0.034 |
theirs | -0.034 |
Token | a |
Token ID | 257 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
Jill | +0.014 |
like | +0.013 |
type | +0.012 |
neat | +0.011 |
Jen | +0.011 |
Neg logit contributions | |
lor | -0.030 |
actus | -0.030 |
� | -0.029 |
ican | -0.027 |
eros | -0.026 |
Token | time |
Token ID | 640 |
Feature activation | +5.22e-02 |
Pos logit contributions | |
lor | +0.022 |
actus | +0.021 |
� | +0.021 |
sky | +0.019 |
ican | +0.018 |
Neg logit contributions | |
Jill | -0.015 |
type | -0.013 |
kil | -0.013 |
like | -0.013 |
Jen | -0.012 |
Token | , |
Token ID | 11 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
lor | +0.095 |
actus | +0.095 |
� | +0.085 |
sky | +0.081 |
ican | +0.078 |
Neg logit contributions | |
Jill | -0.081 |
Jen | -0.070 |
like | -0.070 |
type | -0.068 |
Jane | -0.067 |
Token | 't |
Token ID | 470 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
actus | +0.104 |
lor | +0.100 |
sky | +0.075 |
rot | +0.074 |
� | +0.073 |
Neg logit contributions | |
Jill | -0.199 |
Jen | -0.187 |
organize | -0.186 |
like | -0.185 |
Liz | -0.183 |
Token | get |
Token ID | 651 |
Feature activation | -1.00e-02 |
Pos logit contributions | |
Jill | +0.017 |
kil | +0.015 |
type | +0.015 |
ooks | +0.015 |
theirs | +0.014 |
Neg logit contributions | |
� | -0.026 |
lor | -0.026 |
actus | -0.026 |
sky | -0.024 |
course | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
Jill | +0.015 |
type | +0.015 |
kil | +0.015 |
ooks | +0.014 |
ots | +0.014 |
Neg logit contributions | |
actus | -0.016 |
lor | -0.015 |
� | -0.015 |
sky | -0.014 |
course | -0.013 |
Token | them |
Token ID | 606 |
Feature activation | -2.25e-02 |
Pos logit contributions | |
actus | +0.030 |
� | +0.030 |
lor | +0.029 |
sky | +0.028 |
course | +0.027 |
Neg logit contributions | |
kil | -0.026 |
Jill | -0.025 |
ooks | -0.024 |
ots | -0.023 |
type | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -8.92e-03 |
Pos logit contributions | |
Jill | +0.034 |
kil | +0.032 |
type | +0.031 |
ooks | +0.031 |
Jen | +0.030 |
Neg logit contributions | |
lor | -0.036 |
actus | -0.036 |
� | -0.036 |
sky | -0.033 |
course | -0.031 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
Jill | +0.015 |
Jen | +0.014 |
Liz | +0.013 |
like | +0.013 |
Mia | +0.013 |
Neg logit contributions | |
lor | -0.014 |
actus | -0.013 |
� | -0.013 |
poster | -0.012 |
ican | -0.012 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.12e-02 |
Pos logit contributions | |
lor | +0.190 |
actus | +0.190 |
� | +0.173 |
ican | +0.171 |
eros | +0.163 |
Neg logit contributions | |
Jill | -0.181 |
like | -0.160 |
Jen | -0.156 |
Jane | -0.153 |
Mia | -0.152 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
lor | +0.091 |
� | +0.086 |
actus | +0.086 |
ican | +0.078 |
sky | +0.077 |
Neg logit contributions | |
Jill | -0.049 |
Jen | -0.040 |
type | -0.038 |
theirs | -0.038 |
ooks | -0.037 |
Token | a |
Token ID | 257 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
Jill | +0.012 |
like | +0.012 |
type | +0.010 |
neat | +0.010 |
Jen | +0.010 |
Neg logit contributions | |
lor | -0.028 |
actus | -0.027 |
� | -0.026 |
ican | -0.024 |
eros | -0.024 |
Token | time |
Token ID | 640 |
Feature activation | +5.35e-02 |
Pos logit contributions | |
lor | +0.021 |
actus | +0.020 |
� | +0.019 |
sky | +0.018 |
eros | +0.018 |
Neg logit contributions | |
Jill | -0.013 |
type | -0.012 |
like | -0.011 |
kil | -0.011 |
tim | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | -5.19e-03 |
Pos logit contributions | |
actus | +0.097 |
lor | +0.097 |
� | +0.086 |
sky | +0.083 |
ican | +0.080 |
Neg logit contributions | |
Jill | -0.083 |
Jen | -0.073 |
like | -0.072 |
type | -0.071 |
Jane | -0.069 |
Token | so |
Token ID | 523 |
Feature activation | +1.39e-02 |
Pos logit contributions | |
Jill | +0.010 |
type | +0.009 |
kil | +0.009 |
Jen | +0.008 |
ooks | +0.008 |
Neg logit contributions | |
lor | -0.015 |
actus | -0.015 |
� | -0.015 |
sky | -0.014 |
course | -0.013 |
Token | happy |
Token ID | 3772 |
Feature activation | +6.28e-02 |
Pos logit contributions | |
lor | +0.031 |
actus | +0.030 |
� | +0.030 |
sky | +0.027 |
ican | +0.027 |
Neg logit contributions | |
Jill | -0.014 |
type | -0.012 |
like | -0.012 |
neat | -0.012 |
kil | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | +4.56e-02 |
Pos logit contributions | |
lor | +0.095 |
actus | +0.092 |
� | +0.092 |
sky | +0.078 |
rot | +0.073 |
Neg logit contributions | |
Jill | -0.113 |
like | -0.102 |
type | -0.099 |
Jen | -0.099 |
theirs | -0.096 |
Token | loved |
Token ID | 6151 |
Feature activation | +3.94e-02 |
Pos logit contributions | |
lor | +0.079 |
actus | +0.079 |
� | +0.076 |
sky | +0.071 |
course | +0.065 |
Neg logit contributions | |
Jill | -0.066 |
kil | -0.062 |
type | -0.061 |
ooks | -0.058 |
theirs | -0.058 |
Token | . |
Token ID | 13 |
Feature activation | -5.56e-04 |
Pos logit contributions | |
lor | +0.073 |
actus | +0.072 |
� | +0.072 |
sky | +0.067 |
course | +0.063 |
Neg logit contributions | |
Jill | -0.052 |
kil | -0.047 |
type | -0.047 |
ooks | -0.046 |
Jen | -0.044 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
Jill | +0.001 |
Jen | +0.001 |
ooks | +0.001 |
like | +0.001 |
Liz | +0.001 |
Neg logit contributions | |
lor | -0.001 |
actus | -0.001 |
� | -0.001 |
course | -0.001 |
ican | -0.001 |
Token | Once |
Token ID | 7454 |
Feature activation | +3.89e-02 |
Pos logit contributions | |
lor | +0.186 |
actus | +0.185 |
� | +0.171 |
ican | +0.165 |
eros | +0.158 |
Neg logit contributions | |
Jill | -0.184 |
like | -0.163 |
Jen | -0.160 |
Jane | -0.155 |
tim | -0.154 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
lor | +0.086 |
� | +0.081 |
actus | +0.081 |
ican | +0.074 |
sky | +0.073 |
Neg logit contributions | |
Jill | -0.046 |
Jen | -0.038 |
type | -0.036 |
theirs | -0.036 |
Liz | -0.035 |
Token | a |
Token ID | 257 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
Jill | +0.014 |
like | +0.013 |
type | +0.012 |
Jen | +0.012 |
theirs | +0.012 |
Neg logit contributions | |
lor | -0.029 |
actus | -0.029 |
� | -0.028 |
ican | -0.025 |
eros | -0.025 |
Token | time |
Token ID | 640 |
Feature activation | +5.27e-02 |
Pos logit contributions | |
lor | +0.022 |
actus | +0.022 |
� | +0.021 |
sky | +0.020 |
eros | +0.019 |
Neg logit contributions | |
Jill | -0.013 |
type | -0.012 |
like | -0.011 |
tim | -0.011 |
kil | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | -9.97e-03 |
Pos logit contributions | |
lor | +0.098 |
actus | +0.098 |
� | +0.088 |
sky | +0.084 |
ican | +0.080 |
Neg logit contributions | |
Jill | -0.081 |
Jen | -0.070 |
like | -0.068 |
Jane | -0.068 |
type | -0.067 |
Token | to |
Token ID | 284 |
Feature activation | +4.69e-02 |
Pos logit contributions | |
actus | +0.163 |
lor | +0.158 |
� | +0.157 |
poster | +0.145 |
pillar | +0.143 |
Neg logit contributions | |
Jill | -0.064 |
Jen | -0.054 |
like | -0.051 |
theirs | -0.048 |
Joe | -0.048 |
Token | help |
Token ID | 1037 |
Feature activation | -3.13e-03 |
Pos logit contributions | |
lor | +0.086 |
actus | +0.085 |
� | +0.084 |
sky | +0.077 |
course | +0.073 |
Neg logit contributions | |
Jill | -0.065 |
kil | -0.058 |
type | -0.057 |
Jen | -0.055 |
theirs | -0.055 |
Token | others |
Token ID | 1854 |
Feature activation | -4.18e-03 |
Pos logit contributions | |
Jill | +0.005 |
kil | +0.005 |
tim | +0.005 |
Jen | +0.005 |
Matt | +0.005 |
Neg logit contributions | |
lor | -0.004 |
actus | -0.004 |
sky | -0.004 |
course | -0.004 |
� | -0.004 |
Token | too |
Token ID | 1165 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
Jill | +0.006 |
Jen | +0.006 |
kil | +0.006 |
type | +0.005 |
ooks | +0.005 |
Neg logit contributions | |
lor | -0.007 |
actus | -0.007 |
� | -0.007 |
sky | -0.006 |
course | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -1.80e-03 |
Pos logit contributions | |
lor | +0.033 |
� | +0.033 |
actus | +0.033 |
sky | +0.031 |
skies | +0.027 |
Neg logit contributions | |
Jill | -0.030 |
type | -0.028 |
kil | -0.027 |
Jen | -0.027 |
plugged | -0.026 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
Jill | +0.002 |
Jen | +0.002 |
ooks | +0.002 |
like | +0.002 |
theirs | +0.002 |
Neg logit contributions | |
lor | -0.003 |
� | -0.003 |
poster | -0.003 |
actus | -0.003 |
ican | -0.003 |
Token | Once |
Token ID | 7454 |
Feature activation | +3.98e-02 |
Pos logit contributions | |
lor | +0.188 |
actus | +0.188 |
� | +0.173 |
ican | +0.167 |
poster | +0.160 |
Neg logit contributions | |
Jill | -0.184 |
like | -0.161 |
Jen | -0.159 |
Jane | -0.155 |
Mia | -0.153 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
lor | +0.093 |
� | +0.088 |
actus | +0.087 |
ican | +0.080 |
sky | +0.078 |
Neg logit contributions | |
Jill | -0.044 |
Jen | -0.035 |
theirs | -0.033 |
Liz | -0.033 |
type | -0.032 |
Token | a |
Token ID | 257 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
Jill | +0.014 |
like | +0.013 |
type | +0.011 |
theirs | +0.011 |
neat | +0.011 |
Neg logit contributions | |
lor | -0.035 |
actus | -0.034 |
� | -0.034 |
ican | -0.031 |
eros | -0.031 |
Token | time |
Token ID | 640 |
Feature activation | +5.06e-02 |
Pos logit contributions | |
lor | +0.023 |
actus | +0.022 |
� | +0.021 |
sky | +0.020 |
ican | +0.019 |
Neg logit contributions | |
Jill | -0.013 |
type | -0.012 |
like | -0.012 |
kil | -0.011 |
tim | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
actus | +0.093 |
lor | +0.092 |
� | +0.083 |
sky | +0.079 |
ican | +0.075 |
Neg logit contributions | |
Jill | -0.078 |
Jen | -0.068 |
like | -0.066 |
type | -0.066 |
Jane | -0.065 |
Token | could |
Token ID | 714 |
Feature activation | +7.20e-02 |
Pos logit contributions | |
Jill | +0.024 |
kil | +0.022 |
theirs | +0.022 |
ooks | +0.022 |
type | +0.022 |
Neg logit contributions | |
actus | -0.032 |
lor | -0.032 |
� | -0.032 |
sky | -0.029 |
course | -0.027 |
Token | help |
Token ID | 1037 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
� | +0.138 |
lor | +0.132 |
actus | +0.128 |
sky | +0.120 |
course | +0.115 |
Neg logit contributions | |
Jill | -0.094 |
kil | -0.088 |
theirs | -0.083 |
ooks | -0.082 |
Jen | -0.081 |
Token | her |
Token ID | 607 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
kil | +0.030 |
tim | +0.030 |
rated | +0.029 |
theirs | +0.028 |
ooks | +0.028 |
Neg logit contributions | |
� | -0.025 |
sky | -0.023 |
actus | -0.023 |
lor | -0.022 |
course | -0.021 |
Token | daddy |
Token ID | 35822 |
Feature activation | +4.45e-02 |
Pos logit contributions | |
actus | +0.017 |
� | +0.017 |
lor | +0.017 |
sky | +0.016 |
poster | +0.015 |
Neg logit contributions | |
Jill | -0.017 |
kil | -0.016 |
ooks | -0.016 |
Jen | -0.016 |
theirs | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
lor | +0.078 |
� | +0.077 |
actus | +0.076 |
sky | +0.072 |
poster | +0.068 |
Neg logit contributions | |
Jill | -0.061 |
ooks | -0.057 |
Jen | -0.055 |
kil | -0.054 |
theirs | -0.053 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
Jill | +0.018 |
Jen | +0.015 |
Liz | +0.015 |
ooks | +0.015 |
neat | +0.015 |
Neg logit contributions | |
lor | -0.020 |
� | -0.018 |
actus | -0.018 |
rot | -0.017 |
ican | -0.017 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.37e-02 |
Pos logit contributions | |
lor | +0.182 |
actus | +0.182 |
� | +0.161 |
ican | +0.159 |
poster | +0.153 |
Neg logit contributions | |
Jill | -0.188 |
like | -0.165 |
Jen | -0.162 |
Jane | -0.159 |
tim | -0.157 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
lor | +0.092 |
actus | +0.088 |
� | +0.088 |
sky | +0.080 |
ican | +0.078 |
Neg logit contributions | |
Jill | -0.054 |
Jen | -0.044 |
type | -0.044 |
theirs | -0.043 |
ooks | -0.042 |
Token | a |
Token ID | 257 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
Jill | +0.014 |
like | +0.012 |
type | +0.012 |
Jen | +0.011 |
theirs | +0.011 |
Neg logit contributions | |
lor | -0.027 |
actus | -0.027 |
� | -0.026 |
ican | -0.024 |
sky | -0.024 |
Token | time |
Token ID | 640 |
Feature activation | +5.26e-02 |
Pos logit contributions | |
lor | +0.024 |
actus | +0.024 |
� | +0.023 |
sky | +0.022 |
ican | +0.020 |
Neg logit contributions | |
Jill | -0.016 |
type | -0.015 |
kil | -0.014 |
like | -0.014 |
Jen | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | -8.64e-03 |
Pos logit contributions | |
lor | +0.089 |
actus | +0.088 |
� | +0.082 |
sky | +0.076 |
ican | +0.071 |
Neg logit contributions | |
Jill | -0.086 |
Jen | -0.076 |
type | -0.075 |
like | -0.075 |
kil | -0.074 |
Token | not |
Token ID | 407 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
lor | +0.065 |
actus | +0.065 |
� | +0.064 |
sky | +0.060 |
course | +0.057 |
Neg logit contributions | |
Jill | -0.043 |
kil | -0.040 |
type | -0.038 |
ooks | -0.038 |
ots | -0.036 |
Token | break |
Token ID | 2270 |
Feature activation | -3.97e-02 |
Pos logit contributions | |
Jill | +0.026 |
kil | +0.024 |
ooks | +0.024 |
type | +0.023 |
theirs | +0.023 |
Neg logit contributions | |
� | -0.033 |
actus | -0.033 |
lor | -0.032 |
sky | -0.031 |
poster | -0.029 |
Token | the |
Token ID | 262 |
Feature activation | +3.93e-02 |
Pos logit contributions | |
ooks | +0.060 |
type | +0.059 |
kil | +0.059 |
tim | +0.057 |
ierce | +0.056 |
Neg logit contributions | |
actus | -0.057 |
sky | -0.055 |
course | -0.053 |
� | -0.052 |
lor | -0.052 |
Token | toy |
Token ID | 13373 |
Feature activation | +5.25e-02 |
Pos logit contributions | |
lor | +0.069 |
� | +0.067 |
actus | +0.065 |
sky | +0.057 |
course | +0.054 |
Neg logit contributions | |
Jill | -0.061 |
type | -0.056 |
kil | -0.054 |
Jen | -0.053 |
like | -0.053 |
Token | . |
Token ID | 13 |
Feature activation | -9.59e-03 |
Pos logit contributions | |
lor | +0.088 |
actus | +0.087 |
� | +0.085 |
sky | +0.078 |
course | +0.073 |
Neg logit contributions | |
Jill | -0.081 |
type | -0.073 |
kil | -0.073 |
Jen | -0.070 |
ooks | -0.070 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
Jill | +0.015 |
Jen | +0.012 |
Liz | +0.012 |
like | +0.012 |
Jane | +0.012 |
Neg logit contributions | |
lor | -0.018 |
actus | -0.017 |
� | -0.016 |
rot | -0.016 |
ican | -0.015 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.08e-02 |
Pos logit contributions | |
actus | +0.183 |
lor | +0.182 |
� | +0.168 |
ican | +0.162 |
poster | +0.154 |
Neg logit contributions | |
Jill | -0.185 |
like | -0.164 |
Jen | -0.161 |
Jane | -0.157 |
tim | -0.156 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
lor | +0.091 |
actus | +0.086 |
� | +0.085 |
ican | +0.078 |
sky | +0.077 |
Neg logit contributions | |
Jill | -0.049 |
Jen | -0.040 |
theirs | -0.037 |
type | -0.037 |
Liz | -0.037 |
Token | a |
Token ID | 257 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
Jill | +0.013 |
like | +0.012 |
type | +0.011 |
neat | +0.011 |
Jen | +0.011 |
Neg logit contributions | |
lor | -0.030 |
actus | -0.030 |
� | -0.029 |
ican | -0.027 |
eros | -0.027 |
Token | time |
Token ID | 640 |
Feature activation | +5.29e-02 |
Pos logit contributions | |
lor | +0.027 |
actus | +0.026 |
� | +0.025 |
sky | +0.024 |
ican | +0.023 |
Neg logit contributions | |
Jill | -0.015 |
type | -0.014 |
like | -0.014 |
kil | -0.013 |
tim | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | -9.24e-03 |
Pos logit contributions | |
actus | +0.096 |
lor | +0.095 |
� | +0.085 |
sky | +0.082 |
ican | +0.078 |
Neg logit contributions | |
Jill | -0.083 |
Jen | -0.072 |
like | -0.071 |
type | -0.070 |
Jane | -0.069 |
Token | he |
Token ID | 339 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
Jill | +0.038 |
kil | +0.037 |
type | +0.037 |
ooks | +0.036 |
ots | +0.036 |
Neg logit contributions | |
� | -0.032 |
actus | -0.031 |
lor | -0.030 |
sky | -0.030 |
skies | -0.027 |
Token | 's |
Token ID | 338 |
Feature activation | +6.90e-04 |
Pos logit contributions | |
Jill | +0.015 |
type | +0.014 |
like | +0.014 |
kil | +0.014 |
Jen | +0.013 |
Neg logit contributions | |
lor | -0.019 |
� | -0.018 |
actus | -0.018 |
sky | -0.016 |
poster | -0.015 |
Token | feeling |
Token ID | 4203 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
lor | +0.001 |
actus | +0.001 |
� | +0.001 |
sky | +0.001 |
course | +0.001 |
Neg logit contributions | |
Jill | -0.001 |
type | -0.001 |
kil | -0.001 |
Jen | -0.001 |
theirs | -0.001 |
Token | sore |
Token ID | 19597 |
Feature activation | +6.44e-03 |
Pos logit contributions | |
Jill | +0.023 |
Jen | +0.021 |
ooks | +0.021 |
type | +0.021 |
kil | +0.021 |
Neg logit contributions | |
actus | -0.019 |
� | -0.019 |
lor | -0.019 |
sky | -0.018 |
course | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -9.58e-03 |
Pos logit contributions | |
lor | +0.011 |
actus | +0.011 |
� | +0.011 |
sky | +0.010 |
course | +0.009 |
Neg logit contributions | |
Jill | -0.009 |
type | -0.009 |
kil | -0.008 |
Jen | -0.008 |
ooks | -0.008 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
Jill | +0.018 |
Jen | +0.015 |
Joe | +0.015 |
like | +0.015 |
ooks | +0.015 |
Neg logit contributions | |
lor | -0.014 |
actus | -0.014 |
� | -0.013 |
sky | -0.012 |
poster | -0.012 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.11e-02 |
Pos logit contributions | |
actus | +0.181 |
lor | +0.179 |
� | +0.162 |
ican | +0.158 |
poster | +0.155 |
Neg logit contributions | |
Jill | -0.189 |
Jen | -0.162 |
like | -0.162 |
Jane | -0.160 |
Mia | -0.158 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
lor | +0.092 |
� | +0.087 |
actus | +0.087 |
sky | +0.079 |
ican | +0.079 |
Neg logit contributions | |
Jill | -0.048 |
Jen | -0.038 |
type | -0.037 |
theirs | -0.036 |
Liz | -0.036 |
Token | a |
Token ID | 257 |
Feature activation | +8.95e-03 |
Pos logit contributions | |
Jill | +0.012 |
like | +0.011 |
type | +0.010 |
theirs | +0.010 |
Jen | +0.009 |
Neg logit contributions | |
lor | -0.025 |
actus | -0.024 |
� | -0.024 |
ican | -0.022 |
eros | -0.022 |
Token | time |
Token ID | 640 |
Feature activation | +5.20e-02 |
Pos logit contributions | |
lor | +0.018 |
actus | +0.017 |
� | +0.017 |
sky | +0.016 |
ican | +0.015 |
Neg logit contributions | |
Jill | -0.011 |
type | -0.010 |
like | -0.010 |
kil | -0.010 |
neat | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | -7.57e-03 |
Pos logit contributions | |
lor | +0.087 |
actus | +0.087 |
� | +0.081 |
sky | +0.076 |
ican | +0.070 |
Neg logit contributions | |
Jill | -0.086 |
Jen | -0.075 |
type | -0.074 |
like | -0.073 |
kil | -0.072 |
Token | want |
Token ID | 765 |
Feature activation | +3.15e-02 |
Pos logit contributions | |
Jill | +0.014 |
ooks | +0.012 |
kil | +0.012 |
Matt | +0.012 |
Jamie | +0.012 |
Neg logit contributions | |
actus | -0.019 |
sky | -0.018 |
� | -0.018 |
lor | -0.018 |
course | -0.017 |
Token | to |
Token ID | 284 |
Feature activation | +2.57e-02 |
Pos logit contributions | |
actus | +0.046 |
lor | +0.046 |
� | +0.045 |
sky | +0.044 |
course | +0.041 |
Neg logit contributions | |
Jill | -0.050 |
ooks | -0.048 |
type | -0.048 |
tim | -0.047 |
kil | -0.047 |
Token | complain |
Token ID | 13121 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
actus | +0.047 |
� | +0.047 |
lor | +0.046 |
sky | +0.044 |
course | +0.041 |
Neg logit contributions | |
Jill | -0.034 |
kil | -0.031 |
ooks | -0.030 |
theirs | -0.029 |
type | -0.029 |
Token | anymore |
Token ID | 7471 |
Feature activation | +5.33e-02 |
Pos logit contributions | |
sky | +0.017 |
actus | +0.016 |
lor | +0.015 |
� | +0.014 |
course | +0.013 |
Neg logit contributions | |
kil | -0.040 |
type | -0.040 |
akers | -0.040 |
Jill | -0.040 |
ooks | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | -5.62e-03 |
Pos logit contributions | |
lor | +0.072 |
actus | +0.070 |
� | +0.067 |
sky | +0.062 |
course | +0.058 |
Neg logit contributions | |
Jill | -0.091 |
ooks | -0.086 |
type | -0.085 |
kil | -0.085 |
ierce | -0.083 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
Jill | +0.009 |
type | +0.008 |
Jen | +0.008 |
kil | +0.008 |
like | +0.007 |
Neg logit contributions | |
lor | -0.010 |
actus | -0.010 |
� | -0.009 |
sky | -0.008 |
poster | -0.008 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.12e-02 |
Pos logit contributions | |
lor | +0.180 |
actus | +0.180 |
� | +0.165 |
ican | +0.158 |
eros | +0.150 |
Neg logit contributions | |
Jill | -0.189 |
like | -0.167 |
Jen | -0.164 |
Jane | -0.160 |
type | -0.159 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
lor | +0.095 |
� | +0.090 |
actus | +0.089 |
ican | +0.081 |
sky | +0.081 |
Neg logit contributions | |
Jill | -0.046 |
Jen | -0.037 |
theirs | -0.035 |
type | -0.035 |
Liz | -0.035 |
Token | a |
Token ID | 257 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
Jill | +0.012 |
like | +0.011 |
type | +0.010 |
Jen | +0.009 |
theirs | +0.009 |
Neg logit contributions | |
lor | -0.025 |
actus | -0.024 |
� | -0.024 |
ican | -0.022 |
eros | -0.022 |
Token | time |
Token ID | 640 |
Feature activation | +5.33e-02 |
Pos logit contributions | |
lor | +0.024 |
actus | +0.023 |
� | +0.022 |
sky | +0.021 |
eros | +0.020 |
Neg logit contributions | |
Jill | -0.015 |
type | -0.014 |
kil | -0.013 |
like | -0.013 |
Jen | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | -7.30e-03 |
Pos logit contributions | |
lor | +0.091 |
actus | +0.091 |
� | +0.083 |
sky | +0.078 |
ican | +0.073 |
Neg logit contributions | |
Jill | -0.087 |
Jen | -0.076 |
type | -0.076 |
like | -0.075 |
kil | -0.074 |
Token | didn |
Token ID | 1422 |
Feature activation | +5.29e-02 |
Pos logit contributions | |
Jill | +0.029 |
type | +0.026 |
kil | +0.026 |
Jen | +0.025 |
like | +0.025 |
Neg logit contributions | |
lor | -0.034 |
actus | -0.033 |
� | -0.033 |
sky | -0.030 |
course | -0.028 |
Token | 't |
Token ID | 470 |
Feature activation | -9.36e-03 |
Pos logit contributions | |
actus | +0.051 |
lor | +0.050 |
� | +0.036 |
sky | +0.035 |
eros | +0.035 |
Neg logit contributions | |
Jill | -0.121 |
organize | -0.114 |
like | -0.113 |
type | -0.112 |
Jen | -0.109 |
Token | feel |
Token ID | 1254 |
Feature activation | -1.78e-02 |
Pos logit contributions | |
kil | +0.013 |
Jill | +0.012 |
ooks | +0.012 |
theirs | +0.011 |
Matt | +0.011 |
Neg logit contributions | |
� | -0.016 |
actus | -0.015 |
sky | -0.015 |
lor | -0.014 |
't | -0.014 |
Token | well |
Token ID | 880 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
Jill | +0.030 |
kil | +0.028 |
type | +0.028 |
ooks | +0.028 |
Jen | +0.028 |
Neg logit contributions | |
actus | -0.026 |
lor | -0.025 |
� | -0.024 |
sky | -0.024 |
course | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -8.09e-03 |
Pos logit contributions | |
lor | +0.021 |
actus | +0.021 |
� | +0.020 |
sky | +0.019 |
course | +0.017 |
Neg logit contributions | |
Jill | -0.018 |
type | -0.017 |
kil | -0.017 |
Jen | -0.016 |
ooks | -0.016 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
Jill | +0.014 |
Jen | +0.012 |
like | +0.012 |
Liz | +0.012 |
Mia | +0.011 |
Neg logit contributions | |
lor | -0.014 |
actus | -0.013 |
� | -0.012 |
ican | -0.012 |
rot | -0.012 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.11e-02 |
Pos logit contributions | |
actus | +0.189 |
lor | +0.189 |
� | +0.174 |
ican | +0.172 |
eros | +0.163 |
Neg logit contributions | |
Jill | -0.179 |
like | -0.158 |
Jen | -0.154 |
Jane | -0.152 |
Mia | -0.150 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.25e-02 |
Pos logit contributions | |
lor | +0.094 |
� | +0.088 |
actus | +0.088 |
ican | +0.081 |
sky | +0.079 |
Neg logit contributions | |
Jill | -0.047 |
Jen | -0.038 |
theirs | -0.036 |
type | -0.036 |
Liz | -0.035 |
Token | a |
Token ID | 257 |
Feature activation | +9.30e-03 |
Pos logit contributions | |
Jill | +0.013 |
like | +0.012 |
type | +0.011 |
neat | +0.010 |
theirs | +0.010 |
Neg logit contributions | |
lor | -0.029 |
actus | -0.029 |
� | -0.028 |
ican | -0.026 |
eros | -0.026 |
Token | time |
Token ID | 640 |
Feature activation | +5.00e-02 |
Pos logit contributions | |
lor | +0.019 |
actus | +0.018 |
� | +0.018 |
sky | +0.017 |
ican | +0.016 |
Neg logit contributions | |
Jill | -0.011 |
type | -0.010 |
like | -0.010 |
kil | -0.010 |
tim | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | -9.71e-03 |
Pos logit contributions | |
actus | +0.089 |
lor | +0.089 |
� | +0.081 |
sky | +0.077 |
ican | +0.073 |
Neg logit contributions | |
Jill | -0.079 |
Jen | -0.069 |
like | -0.068 |
type | -0.067 |
Jane | -0.065 |
Token | help |
Token ID | 1037 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
lor | +0.140 |
� | +0.138 |
actus | +0.136 |
sky | +0.124 |
course | +0.118 |
Neg logit contributions | |
Jill | -0.094 |
kil | -0.084 |
type | -0.083 |
ooks | -0.078 |
Jen | -0.078 |
Token | her |
Token ID | 607 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
kil | +0.026 |
rated | +0.025 |
tim | +0.025 |
ooks | +0.024 |
ash | +0.023 |
Neg logit contributions | |
� | -0.020 |
sky | -0.018 |
lor | -0.018 |
shop | -0.018 |
actus | -0.017 |
Token | mom |
Token ID | 1995 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
lor | +0.020 |
actus | +0.020 |
� | +0.020 |
sky | +0.018 |
poster | +0.016 |
Neg logit contributions | |
Jill | -0.025 |
kil | -0.024 |
ooks | -0.023 |
type | -0.023 |
ots | -0.023 |
Token | my |
Token ID | 1820 |
Feature activation | +2.07e-02 |
Pos logit contributions | |
lor | +0.028 |
actus | +0.028 |
� | +0.028 |
sky | +0.026 |
course | +0.023 |
Neg logit contributions | |
Jill | -0.035 |
kil | -0.033 |
type | -0.032 |
ooks | -0.032 |
Jen | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | -6.57e-03 |
Pos logit contributions | |
lor | +0.037 |
actus | +0.036 |
� | +0.036 |
sky | +0.033 |
course | +0.030 |
Neg logit contributions | |
Jill | -0.030 |
kil | -0.026 |
type | -0.026 |
ooks | -0.026 |
Jen | -0.026 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
Jill | +0.010 |
Jen | +0.008 |
Liz | +0.008 |
Jane | +0.008 |
Mia | +0.008 |
Neg logit contributions | |
lor | -0.013 |
actus | -0.012 |
� | -0.011 |
ican | -0.011 |
rot | -0.011 |
Token | Once |
Token ID | 7454 |
Feature activation | +3.95e-02 |
Pos logit contributions | |
actus | +0.188 |
lor | +0.186 |
� | +0.170 |
ican | +0.169 |
eros | +0.160 |
Neg logit contributions | |
Jill | -0.181 |
like | -0.158 |
Jen | -0.155 |
Jane | -0.153 |
Mia | -0.151 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
lor | +0.086 |
actus | +0.082 |
� | +0.082 |
sky | +0.074 |
ican | +0.074 |
Neg logit contributions | |
Jill | -0.048 |
Jen | -0.039 |
type | -0.038 |
theirs | -0.037 |
like | -0.036 |
Token | a |
Token ID | 257 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
Jill | +0.013 |
like | +0.012 |
type | +0.011 |
neat | +0.011 |
Jen | +0.011 |
Neg logit contributions | |
lor | -0.028 |
actus | -0.027 |
� | -0.027 |
ican | -0.025 |
eros | -0.024 |
Token | time |
Token ID | 640 |
Feature activation | +5.18e-02 |
Pos logit contributions | |
lor | +0.021 |
actus | +0.020 |
� | +0.019 |
sky | +0.018 |
ican | +0.018 |
Neg logit contributions | |
Jill | -0.013 |
type | -0.012 |
like | -0.011 |
kil | -0.011 |
tim | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
actus | +0.097 |
lor | +0.097 |
� | +0.086 |
sky | +0.083 |
ican | +0.080 |
Neg logit contributions | |
Jill | -0.079 |
Jen | -0.068 |
like | -0.067 |
Jane | -0.066 |
type | -0.065 |
Token | help |
Token ID | 1037 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
Jill | +0.011 |
kil | +0.009 |
ooks | +0.009 |
type | +0.009 |
Jen | +0.009 |
Neg logit contributions | |
� | -0.016 |
actus | -0.016 |
lor | -0.015 |
sky | -0.014 |
course | -0.014 |
Token | her |
Token ID | 607 |
Feature activation | +8.39e-03 |
Pos logit contributions | |
Jill | +0.026 |
kil | +0.026 |
ash | +0.025 |
rated | +0.025 |
tim | +0.025 |
Neg logit contributions | |
� | -0.024 |
actus | -0.024 |
sky | -0.024 |
lor | -0.023 |
course | -0.022 |
Token | mom |
Token ID | 1995 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
actus | +0.012 |
lor | +0.012 |
� | +0.012 |
sky | +0.011 |
poster | +0.009 |
Neg logit contributions | |
Jill | -0.015 |
kil | -0.014 |
type | -0.013 |
ooks | -0.013 |
ots | -0.013 |
Token | my |
Token ID | 1820 |
Feature activation | +1.74e-02 |
Pos logit contributions | |
actus | +0.022 |
lor | +0.022 |
� | +0.022 |
sky | +0.020 |
course | +0.018 |
Neg logit contributions | |
Jill | -0.025 |
kil | -0.023 |
type | -0.023 |
ooks | -0.023 |
Jen | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
lor | +0.028 |
actus | +0.028 |
� | +0.028 |
sky | +0.025 |
course | +0.024 |
Neg logit contributions | |
Jill | -0.027 |
kil | -0.025 |
type | -0.024 |
Jen | -0.024 |
ooks | -0.024 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
Jill | +0.017 |
Jen | +0.014 |
Jane | +0.014 |
Liz | +0.014 |
Mia | +0.014 |
Neg logit contributions | |
lor | -0.019 |
actus | -0.018 |
� | -0.017 |
ican | -0.016 |
poster | -0.016 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.04e-02 |
Pos logit contributions | |
actus | +0.187 |
lor | +0.183 |
ican | +0.166 |
� | +0.164 |
poster | +0.160 |
Neg logit contributions | |
Jill | -0.182 |
Jane | -0.156 |
Jen | -0.155 |
like | -0.155 |
Mia | -0.152 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
lor | +0.090 |
� | +0.085 |
actus | +0.085 |
ican | +0.077 |
sky | +0.077 |
Neg logit contributions | |
Jill | -0.048 |
Jen | -0.039 |
type | -0.037 |
theirs | -0.036 |
Liz | -0.036 |
Token | a |
Token ID | 257 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
Jill | +0.014 |
like | +0.013 |
type | +0.012 |
theirs | +0.011 |
Jen | +0.011 |
Neg logit contributions | |
lor | -0.029 |
actus | -0.029 |
� | -0.028 |
ican | -0.026 |
eros | -0.026 |
Token | time |
Token ID | 640 |
Feature activation | +5.35e-02 |
Pos logit contributions | |
lor | +0.023 |
actus | +0.022 |
� | +0.021 |
sky | +0.020 |
ican | +0.019 |
Neg logit contributions | |
Jill | -0.014 |
type | -0.013 |
like | -0.012 |
kil | -0.012 |
tim | -0.012 |
Token | , |
Token ID | 11 |
Feature activation | -8.41e-03 |
Pos logit contributions | |
actus | +0.098 |
lor | +0.097 |
� | +0.088 |
sky | +0.084 |
ican | +0.080 |
Neg logit contributions | |
Jill | -0.084 |
Jen | -0.073 |
Jane | -0.070 |
like | -0.070 |
Joe | -0.070 |
Token | Spot |
Token ID | 15899 |
Feature activation | +1.53e-02 |
Pos logit contributions | |
actus | +0.061 |
lor | +0.060 |
� | +0.059 |
sky | +0.059 |
course | +0.054 |
Neg logit contributions | |
kil | -0.058 |
Jill | -0.054 |
ots | -0.054 |
ooks | -0.053 |
type | -0.053 |
Token | all |
Token ID | 477 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
actus | +0.024 |
lor | +0.024 |
� | +0.024 |
sky | +0.022 |
course | +0.020 |
Neg logit contributions | |
Jill | -0.024 |
kil | -0.022 |
type | -0.022 |
ooks | -0.022 |
Jen | -0.022 |
Token | day |
Token ID | 1110 |
Feature activation | +8.35e-02 |
Pos logit contributions | |
kil | +0.062 |
ooks | +0.060 |
ots | +0.056 |
type | +0.056 |
ierce | +0.056 |
Neg logit contributions | |
sky | -0.052 |
actus | -0.051 |
course | -0.047 |
� | -0.047 |
lor | -0.045 |
Token | long |
Token ID | 890 |
Feature activation | -7.76e-03 |
Pos logit contributions | |
lor | +0.169 |
actus | +0.167 |
� | +0.158 |
sky | +0.149 |
course | +0.144 |
Neg logit contributions | |
Jill | -0.103 |
type | -0.089 |
kil | -0.088 |
like | -0.086 |
neat | -0.086 |
Token | . |
Token ID | 13 |
Feature activation | -3.72e-03 |
Pos logit contributions | |
Jill | +0.010 |
type | +0.009 |
kil | +0.009 |
Jen | +0.009 |
theirs | +0.009 |
Neg logit contributions | |
lor | -0.015 |
actus | -0.015 |
� | -0.014 |
sky | -0.013 |
course | -0.012 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
Jill | +0.006 |
Jen | +0.006 |
like | +0.006 |
theirs | +0.005 |
type | +0.005 |
Neg logit contributions | |
lor | -0.006 |
actus | -0.006 |
� | -0.005 |
poster | -0.005 |
sky | -0.005 |
Token | Once |
Token ID | 7454 |
Feature activation | +3.88e-02 |
Pos logit contributions | |
lor | +0.189 |
actus | +0.187 |
� | +0.173 |
ican | +0.168 |
eros | +0.160 |
Neg logit contributions | |
Jill | -0.179 |
like | -0.159 |
Jen | -0.155 |
Jane | -0.152 |
Mia | -0.150 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
lor | +0.082 |
actus | +0.078 |
� | +0.078 |
sky | +0.070 |
ican | +0.069 |
Neg logit contributions | |
Jill | -0.048 |
Jen | -0.040 |
type | -0.039 |
theirs | -0.039 |
like | -0.038 |
Token | a |
Token ID | 257 |
Feature activation | +1.22e-02 |
Pos logit contributions | |
Jill | +0.016 |
like | +0.014 |
type | +0.014 |
Jen | +0.013 |
theirs | +0.013 |
Neg logit contributions | |
lor | -0.032 |
actus | -0.031 |
� | -0.030 |
ican | -0.027 |
sky | -0.027 |
Token | time |
Token ID | 640 |
Feature activation | +5.19e-02 |
Pos logit contributions | |
lor | +0.025 |
actus | +0.024 |
� | +0.023 |
sky | +0.022 |
ican | +0.022 |
Neg logit contributions | |
Jill | -0.014 |
type | -0.013 |
like | -0.013 |
tim | -0.012 |
kil | -0.012 |
Token | , |
Token ID | 11 |
Feature activation | -8.54e-03 |
Pos logit contributions | |
lor | +0.095 |
actus | +0.094 |
� | +0.084 |
sky | +0.081 |
ican | +0.077 |
Neg logit contributions | |
Jill | -0.080 |
Jen | -0.070 |
like | -0.069 |
type | -0.068 |
Joe | -0.067 |
Token | was |
Token ID | 373 |
Feature activation | -1.36e-03 |
Pos logit contributions | |
Jill | +0.016 |
type | +0.014 |
kil | +0.014 |
Jen | +0.014 |
ooks | +0.014 |
Neg logit contributions | |
lor | -0.017 |
actus | -0.017 |
� | -0.017 |
sky | -0.015 |
course | -0.014 |
Token | for |
Token ID | 329 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
kil | +0.002 |
tim | +0.002 |
ooks | +0.002 |
Jill | +0.002 |
type | +0.002 |
Neg logit contributions | |
lor | -0.002 |
sky | -0.002 |
� | -0.002 |
actus | -0.002 |
course | -0.002 |
Token | her |
Token ID | 607 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
rated | +0.015 |
kil | +0.014 |
tim | +0.014 |
ooks | +0.014 |
ash | +0.014 |
Neg logit contributions | |
� | -0.018 |
course | -0.018 |
lor | -0.018 |
sky | -0.016 |
actus | -0.016 |
Token | love |
Token ID | 1842 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
actus | +0.027 |
lor | +0.027 |
� | +0.026 |
sky | +0.024 |
course | +0.023 |
Neg logit contributions | |
Jill | -0.022 |
kil | -0.020 |
type | -0.020 |
ooks | -0.019 |
Jen | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
actus | +0.019 |
lor | +0.019 |
� | +0.019 |
sky | +0.018 |
course | +0.017 |
Neg logit contributions | |
Jill | -0.018 |
kil | -0.018 |
ooks | -0.017 |
type | -0.017 |
tim | -0.017 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
Jill | +0.020 |
Jen | +0.018 |
Liz | +0.017 |
Joe | +0.017 |
Jane | +0.017 |
Neg logit contributions | |
lor | -0.017 |
actus | -0.016 |
� | -0.016 |
rot | -0.015 |
poster | -0.015 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.11e-02 |
Pos logit contributions | |
actus | +0.193 |
lor | +0.190 |
ican | +0.174 |
� | +0.169 |
poster | +0.168 |
Neg logit contributions | |
Jill | -0.173 |
like | -0.149 |
Jane | -0.148 |
Jen | -0.147 |
Mia | -0.144 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
lor | +0.094 |
� | +0.089 |
actus | +0.089 |
ican | +0.081 |
sky | +0.080 |
Neg logit contributions | |
Jill | -0.047 |
Jen | -0.038 |
theirs | -0.036 |
Liz | -0.035 |
type | -0.035 |
Token | a |
Token ID | 257 |
Feature activation | +1.29e-02 |
Pos logit contributions | |
Jill | +0.011 |
like | +0.011 |
type | +0.010 |
neat | +0.009 |
theirs | +0.009 |
Neg logit contributions | |
lor | -0.024 |
actus | -0.024 |
� | -0.023 |
ican | -0.021 |
eros | -0.021 |
Token | time |
Token ID | 640 |
Feature activation | +5.20e-02 |
Pos logit contributions | |
lor | +0.026 |
actus | +0.025 |
� | +0.024 |
sky | +0.023 |
eros | +0.022 |
Neg logit contributions | |
Jill | -0.016 |
type | -0.014 |
like | -0.014 |
kil | -0.014 |
tim | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | -6.65e-03 |
Pos logit contributions | |
actus | +0.094 |
lor | +0.094 |
� | +0.086 |
sky | +0.081 |
ican | +0.077 |
Neg logit contributions | |
Jill | -0.082 |
Jen | -0.071 |
like | -0.069 |
type | -0.069 |
Jane | -0.068 |
Token | around |
Token ID | 1088 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
lor | +0.056 |
actus | +0.055 |
� | +0.054 |
sky | +0.049 |
course | +0.046 |
Neg logit contributions | |
Jill | -0.039 |
type | -0.035 |
like | -0.035 |
kil | -0.034 |
neat | -0.033 |
Token | the |
Token ID | 262 |
Feature activation | +3.57e-02 |
Pos logit contributions | |
actus | +0.031 |
lor | +0.030 |
� | +0.030 |
sky | +0.029 |
course | +0.028 |
Neg logit contributions | |
Jill | -0.025 |
kil | -0.025 |
ooks | -0.023 |
type | -0.023 |
tim | -0.023 |
Token | c |
Token ID | 269 |
Feature activation | -4.84e-03 |
Pos logit contributions | |
� | +0.063 |
lor | +0.063 |
actus | +0.050 |
course | +0.050 |
ane | +0.050 |
Neg logit contributions | |
Jill | -0.055 |
type | -0.053 |
neat | -0.052 |
like | -0.051 |
sunny | -0.049 |
Token | actus |
Token ID | 34144 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
ooks | +0.011 |
Jill | +0.011 |
ores | +0.011 |
ook | +0.011 |
arter | +0.011 |
Neg logit contributions | |
course | -0.003 |
� | -0.003 |
lor | -0.003 |
helicopter | -0.002 |
poster | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -7.04e-03 |
Pos logit contributions | |
lor | +0.024 |
actus | +0.024 |
� | +0.023 |
sky | +0.021 |
poster | +0.020 |
Neg logit contributions | |
Jill | -0.020 |
like | -0.018 |
type | -0.018 |
kil | -0.017 |
Jen | -0.017 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
Jill | +0.013 |
Jen | +0.011 |
Liz | +0.011 |
ooks | +0.011 |
like | +0.011 |
Neg logit contributions | |
lor | -0.011 |
actus | -0.011 |
� | -0.010 |
ican | -0.009 |
poster | -0.009 |
Token | Once |
Token ID | 7454 |
Feature activation | +3.94e-02 |
Pos logit contributions | |
lor | +0.182 |
actus | +0.181 |
� | +0.168 |
ican | +0.162 |
eros | +0.154 |
Neg logit contributions | |
Jill | -0.184 |
like | -0.164 |
Jen | -0.159 |
Jane | -0.156 |
tim | -0.154 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
lor | +0.089 |
� | +0.085 |
actus | +0.084 |
ican | +0.077 |
sky | +0.076 |
Neg logit contributions | |
Jill | -0.045 |
Jen | -0.037 |
theirs | -0.035 |
type | -0.034 |
ooks | -0.034 |
Token | a |
Token ID | 257 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
Jill | +0.014 |
like | +0.013 |
type | +0.012 |
neat | +0.011 |
theirs | +0.011 |
Neg logit contributions | |
lor | -0.030 |
actus | -0.030 |
� | -0.029 |
ican | -0.027 |
eros | -0.026 |
Token | time |
Token ID | 640 |
Feature activation | +5.23e-02 |
Pos logit contributions | |
lor | +0.021 |
actus | +0.021 |
� | +0.020 |
sky | +0.019 |
eros | +0.018 |
Neg logit contributions | |
Jill | -0.012 |
type | -0.011 |
like | -0.011 |
kil | -0.011 |
tim | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | -9.09e-03 |
Pos logit contributions | |
actus | +0.098 |
lor | +0.098 |
� | +0.088 |
sky | +0.084 |
ican | +0.081 |
Neg logit contributions | |
Jill | -0.080 |
Jen | -0.069 |
like | -0.068 |
Jane | -0.066 |
type | -0.066 |
Token | friend |
Token ID | 1545 |
Feature activation | +6.22e-02 |
Pos logit contributions | |
lor | +0.050 |
actus | +0.045 |
� | +0.044 |
sky | +0.040 |
ican | +0.038 |
Neg logit contributions | |
Jill | -0.033 |
type | -0.033 |
theirs | -0.031 |
neat | -0.031 |
kil | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | -4.22e-03 |
Pos logit contributions | |
actus | +0.098 |
lor | +0.098 |
� | +0.098 |
sky | +0.091 |
course | +0.086 |
Neg logit contributions | |
Jill | -0.096 |
ooks | -0.090 |
type | -0.088 |
kil | -0.086 |
Jen | -0.084 |
Token | The |
Token ID | 383 |
Feature activation | +2.55e-02 |
Pos logit contributions | |
Jill | +0.007 |
Jen | +0.006 |
Joe | +0.006 |
Liz | +0.006 |
Matt | +0.006 |
Neg logit contributions | |
lor | -0.007 |
actus | -0.007 |
� | -0.006 |
poster | -0.006 |
rot | -0.006 |
Token | end |
Token ID | 886 |
Feature activation | +8.85e-03 |
Pos logit contributions | |
lor | +0.058 |
actus | +0.053 |
� | +0.050 |
ican | +0.045 |
eros | +0.045 |
Neg logit contributions | |
sunny | -0.031 |
type | -0.030 |
neat | -0.030 |
matter | -0.029 |
like | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | -8.79e-03 |
Pos logit contributions | |
actus | +0.013 |
� | +0.013 |
lor | +0.013 |
sky | +0.012 |
course | +0.011 |
Neg logit contributions | |
Jill | -0.015 |
type | -0.013 |
ooks | -0.013 |
theirs | -0.013 |
kil | -0.013 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
Jill | +0.016 |
Jen | +0.014 |
like | +0.014 |
theirs | +0.014 |
type | +0.014 |
Neg logit contributions | |
lor | -0.013 |
actus | -0.013 |
� | -0.012 |
poster | -0.010 |
sky | -0.010 |
Token | Once |
Token ID | 7454 |
Feature activation | +3.74e-02 |
Pos logit contributions | |
actus | +0.192 |
lor | +0.190 |
� | +0.172 |
ican | +0.171 |
poster | +0.163 |
Neg logit contributions | |
Jill | -0.177 |
like | -0.155 |
Jen | -0.152 |
Jane | -0.150 |
tim | -0.149 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
lor | +0.084 |
� | +0.079 |
actus | +0.079 |
ican | +0.071 |
sky | +0.071 |
Neg logit contributions | |
Jill | -0.044 |
Jen | -0.036 |
theirs | -0.034 |
type | -0.034 |
Liz | -0.033 |
Token | a |
Token ID | 257 |
Feature activation | +8.93e-03 |
Pos logit contributions | |
Jill | +0.015 |
like | +0.013 |
type | +0.013 |
Jen | +0.012 |
theirs | +0.012 |
Neg logit contributions | |
lor | -0.030 |
actus | -0.030 |
� | -0.029 |
ican | -0.027 |
eros | -0.026 |
Token | time |
Token ID | 640 |
Feature activation | +5.15e-02 |
Pos logit contributions | |
lor | +0.019 |
actus | +0.019 |
� | +0.018 |
sky | +0.017 |
eros | +0.017 |
Neg logit contributions | |
Jill | -0.010 |
type | -0.009 |
like | -0.009 |
tim | -0.009 |
sunny | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | -8.55e-03 |
Pos logit contributions | |
actus | +0.096 |
lor | +0.096 |
� | +0.086 |
sky | +0.082 |
ican | +0.079 |
Neg logit contributions | |
Jill | -0.078 |
Jen | -0.068 |
like | -0.066 |
Joe | -0.066 |
type | -0.065 |
Token | chose |
Token ID | 7690 |
Feature activation | -2.44e-03 |
Pos logit contributions | |
Jill | +0.031 |
type | +0.028 |
kil | +0.027 |
like | +0.026 |
Jen | +0.026 |
Neg logit contributions | |
lor | -0.045 |
actus | -0.045 |
� | -0.044 |
sky | -0.040 |
course | -0.038 |
Token | both |
Token ID | 1111 |
Feature activation | -3.45e-02 |
Pos logit contributions | |
Jill | +0.003 |
kil | +0.003 |
type | +0.003 |
ooks | +0.003 |
tim | +0.003 |
Neg logit contributions | |
lor | -0.004 |
actus | -0.004 |
� | -0.004 |
sky | -0.004 |
course | -0.004 |
Token | vendors |
Token ID | 17192 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
Jill | +0.051 |
kil | +0.048 |
ooks | +0.046 |
type | +0.045 |
tim | +0.045 |
Neg logit contributions | |
lor | -0.056 |
actus | -0.056 |
� | -0.055 |
sky | -0.053 |
course | -0.050 |
Token | . |
Token ID | 13 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
Jill | +0.017 |
kil | +0.015 |
ooks | +0.015 |
type | +0.015 |
Jen | +0.015 |
Neg logit contributions | |
lor | -0.017 |
actus | -0.017 |
� | -0.017 |
sky | -0.015 |
course | -0.015 |
Token | The |
Token ID | 383 |
Feature activation | +1.28e-02 |
Pos logit contributions | |
Jill | +0.030 |
Jen | +0.026 |
type | +0.026 |
kil | +0.026 |
ooks | +0.026 |
Neg logit contributions | |
actus | -0.026 |
lor | -0.026 |
� | -0.024 |
sky | -0.022 |
poster | -0.021 |
Token | frag |
Token ID | 7956 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
lor | +0.023 |
actus | +0.021 |
� | +0.021 |
sky | +0.018 |
rot | +0.018 |
Neg logit contributions | |
Jill | -0.020 |
type | -0.018 |
neat | -0.018 |
kil | -0.018 |
like | -0.017 |
Token | rant |
Token ID | 5250 |
Feature activation | -3.80e-02 |
Pos logit contributions | |
Jill | +0.170 |
kil | +0.149 |
type | +0.147 |
Jen | +0.145 |
neat | +0.138 |
Neg logit contributions | |
lor | -0.301 |
actus | -0.292 |
� | -0.289 |
sky | -0.272 |
course | -0.272 |
Token | flowers |
Token ID | 12734 |
Feature activation | -4.75e-02 |
Pos logit contributions | |
Jill | +0.066 |
kil | +0.062 |
ooks | +0.060 |
type | +0.059 |
Jen | +0.058 |
Neg logit contributions | |
actus | -0.054 |
lor | -0.054 |
� | -0.053 |
sky | -0.051 |
course | -0.046 |
Token | for |
Token ID | 329 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
Jill | +0.071 |
type | +0.064 |
kil | +0.064 |
Jen | +0.061 |
theirs | +0.061 |
Neg logit contributions | |
lor | -0.082 |
actus | -0.080 |
� | -0.079 |
sky | -0.072 |
course | -0.067 |
Token | the |
Token ID | 262 |
Feature activation | +2.62e-02 |
Pos logit contributions | |
Jill | +0.037 |
kil | +0.034 |
type | +0.034 |
ooks | +0.032 |
Jen | +0.031 |
Neg logit contributions | |
lor | -0.057 |
� | -0.057 |
actus | -0.056 |
sky | -0.052 |
course | -0.050 |
Token | sweet |
Token ID | 6029 |
Feature activation | -2.51e-02 |
Pos logit contributions | |
lor | +0.042 |
actus | +0.041 |
� | +0.040 |
sky | +0.036 |
course | +0.033 |
Neg logit contributions | |
Jill | -0.043 |
type | -0.040 |
kil | -0.039 |
Jen | -0.038 |
like | -0.038 |
Token | er |
Token ID | 263 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
Jill | +0.041 |
kil | +0.039 |
type | +0.038 |
like | +0.036 |
remember | +0.036 |
Neg logit contributions | |
actus | -0.045 |
� | -0.044 |
lor | -0.044 |
sky | -0.041 |
poster | -0.036 |
Token | ful |
Token ID | 913 |
Feature activation | -2.95e-02 |
Pos logit contributions | |
Jill | +0.051 |
type | +0.046 |
kil | +0.045 |
Jen | +0.044 |
like | +0.044 |
Neg logit contributions | |
lor | -0.061 |
actus | -0.061 |
� | -0.059 |
sky | -0.054 |
course | -0.050 |
Token | " |
Token ID | 1 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
Jill | +0.045 |
type | +0.042 |
kil | +0.040 |
like | +0.039 |
Jen | +0.039 |
Neg logit contributions | |
actus | -0.049 |
� | -0.049 |
lor | -0.049 |
sky | -0.043 |
ican | -0.040 |
Token | and |
Token ID | 290 |
Feature activation | +2.97e-02 |
Pos logit contributions | |
lor | +0.029 |
actus | +0.029 |
� | +0.028 |
sky | +0.026 |
course | +0.024 |
Neg logit contributions | |
Jill | -0.029 |
type | -0.027 |
kil | -0.027 |
Jen | -0.026 |
ooks | -0.026 |
Token | " |
Token ID | 366 |
Feature activation | +6.22e-03 |
Pos logit contributions | |
lor | +0.050 |
actus | +0.048 |
� | +0.048 |
sky | +0.047 |
course | +0.044 |
Neg logit contributions | |
Jill | -0.041 |
kil | -0.040 |
ooks | -0.040 |
type | -0.037 |
theirs | -0.037 |
Token | sm |
Token ID | 5796 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
lor | +0.011 |
actus | +0.011 |
� | +0.010 |
sky | +0.009 |
eros | +0.009 |
Neg logit contributions | |
Jill | -0.009 |
type | -0.009 |
kil | -0.008 |
like | -0.008 |
Jen | -0.008 |
Token | iley |
Token ID | 9618 |
Feature activation | -9.83e-03 |
Pos logit contributions | |
Jill | +0.223 |
type | +0.204 |
kil | +0.203 |
Jen | +0.196 |
theirs | +0.195 |
Neg logit contributions | |
lor | -0.217 |
actus | -0.216 |
� | -0.211 |
sky | -0.192 |
course | -0.179 |
Token | ". |
Token ID | 1911 |
Feature activation | +8.06e-03 |
Pos logit contributions | |
Jill | +0.017 |
kil | +0.015 |
type | +0.015 |
theirs | +0.015 |
ooks | +0.015 |
Neg logit contributions | |
lor | -0.015 |
actus | -0.014 |
� | -0.014 |
sky | -0.013 |
course | -0.012 |
Token | But |
Token ID | 887 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
actus | +0.016 |
lor | +0.015 |
poster | +0.013 |
rot | +0.013 |
� | +0.013 |
Neg logit contributions | |
Jill | -0.012 |
Jen | -0.011 |
Mia | -0.010 |
Joe | -0.010 |
Jane | -0.010 |
Token | it |
Token ID | 340 |
Feature activation | +2.78e-03 |
Pos logit contributions | |
Jill | +0.031 |
kil | +0.029 |
type | +0.028 |
ooks | +0.028 |
Jen | +0.027 |
Neg logit contributions | |
lor | -0.036 |
� | -0.036 |
actus | -0.036 |
sky | -0.034 |
course | -0.031 |
Token | was |
Token ID | 373 |
Feature activation | +3.85e-03 |
Pos logit contributions | |
lor | +0.005 |
actus | +0.005 |
� | +0.005 |
sky | +0.004 |
course | +0.004 |
Neg logit contributions | |
Jill | -0.004 |
type | -0.004 |
kil | -0.004 |
Jen | -0.004 |
like | -0.004 |
Token | didn |
Token ID | 1422 |
Feature activation | +5.23e-02 |
Pos logit contributions | |
Jill | +0.007 |
type | +0.006 |
plugged | +0.006 |
like | +0.006 |
added | +0.006 |
Neg logit contributions | |
lor | -0.011 |
� | -0.010 |
actus | -0.010 |
eros | -0.009 |
sky | -0.009 |
Token | 't |
Token ID | 470 |
Feature activation | -6.24e-04 |
Pos logit contributions | |
actus | +0.059 |
lor | +0.057 |
sky | +0.045 |
eros | +0.043 |
rot | +0.042 |
Neg logit contributions | |
Jill | -0.113 |
like | -0.106 |
organize | -0.106 |
neat | -0.102 |
type | -0.102 |
Token | know |
Token ID | 760 |
Feature activation | +2.66e-02 |
Pos logit contributions | |
kil | +0.001 |
ooks | +0.001 |
Jill | +0.001 |
Matt | +0.001 |
tim | +0.001 |
Neg logit contributions | |
actus | -0.001 |
sky | -0.001 |
� | -0.001 |
lor | -0.001 |
course | -0.001 |
Token | what |
Token ID | 644 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
lor | +0.060 |
actus | +0.059 |
� | +0.059 |
sky | +0.055 |
course | +0.053 |
Neg logit contributions | |
Jill | -0.025 |
type | -0.021 |
kil | -0.021 |
ooks | -0.020 |
Jen | -0.019 |
Token | " |
Token ID | 366 |
Feature activation | +8.51e-03 |
Pos logit contributions | |
rated | +0.025 |
kil | +0.024 |
tim | +0.024 |
ooks | +0.024 |
cest | +0.022 |
Neg logit contributions | |
� | -0.021 |
course | -0.019 |
actus | -0.019 |
sky | -0.018 |
lor | -0.018 |
Token | sm |
Token ID | 5796 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
lor | +0.015 |
actus | +0.015 |
� | +0.014 |
sky | +0.013 |
course | +0.012 |
Neg logit contributions | |
Jill | -0.013 |
type | -0.011 |
kil | -0.011 |
Jen | -0.011 |
theirs | -0.011 |
Token | oother |
Token ID | 31724 |
Feature activation | -3.22e-02 |
Pos logit contributions | |
Jill | +0.227 |
type | +0.203 |
kil | +0.197 |
like | +0.197 |
Jen | +0.197 |
Neg logit contributions | |
lor | -0.225 |
actus | -0.211 |
� | -0.205 |
sky | -0.185 |
rot | -0.171 |
Token | " |
Token ID | 1 |
Feature activation | +2.08e-02 |
Pos logit contributions | |
Jill | +0.054 |
kil | +0.049 |
type | +0.049 |
Jen | +0.048 |
theirs | +0.048 |
Neg logit contributions | |
lor | -0.050 |
actus | -0.048 |
� | -0.047 |
sky | -0.043 |
course | -0.040 |
Token | meant |
Token ID | 4001 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
actus | +0.042 |
lor | +0.041 |
� | +0.041 |
sky | +0.039 |
skies | +0.036 |
Neg logit contributions | |
Jill | -0.024 |
kil | -0.022 |
type | -0.021 |
ooks | -0.021 |
theirs | -0.020 |
Token | , |
Token ID | 11 |
Feature activation | -6.41e-02 |
Pos logit contributions | |
� | +0.023 |
lor | +0.020 |
actus | +0.019 |
sky | +0.019 |
poster | +0.017 |
Neg logit contributions | |
kil | -0.038 |
Jill | -0.037 |
tim | -0.037 |
ooks | -0.037 |
theirs | -0.036 |
Token | but |
Token ID | 475 |
Feature activation | -1.76e-02 |
Pos logit contributions | |
Jill | +0.075 |
kil | +0.069 |
type | +0.068 |
ooks | +0.066 |
theirs | +0.064 |
Neg logit contributions | |
lor | -0.126 |
� | -0.126 |
actus | -0.125 |
sky | -0.118 |
course | -0.111 |
Token | them |
Token ID | 606 |
Feature activation | -3.46e-02 |
Pos logit contributions | |
Jill | +0.095 |
kil | +0.090 |
type | +0.088 |
tim | +0.086 |
ierce | +0.085 |
Neg logit contributions | |
actus | -0.083 |
lor | -0.082 |
sky | -0.080 |
� | -0.080 |
course | -0.074 |
Token | in |
Token ID | 287 |
Feature activation | -4.32e-02 |
Pos logit contributions | |
Jill | +0.057 |
kil | +0.053 |
Jen | +0.051 |
ooks | +0.051 |
type | +0.050 |
Neg logit contributions | |
� | -0.052 |
actus | -0.051 |
lor | -0.051 |
sky | -0.049 |
skies | -0.044 |
Token | the |
Token ID | 262 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
kil | +0.063 |
ierce | +0.061 |
Jill | +0.061 |
tim | +0.060 |
ooks | +0.058 |
Neg logit contributions | |
� | -0.070 |
sky | -0.069 |
course | -0.066 |
lor | -0.065 |
actus | -0.061 |
Token | ground |
Token ID | 2323 |
Feature activation | -8.89e-03 |
Pos logit contributions | |
actus | +0.027 |
lor | +0.026 |
� | +0.026 |
sky | +0.025 |
course | +0.022 |
Neg logit contributions | |
Jill | -0.036 |
kil | -0.034 |
theirs | -0.034 |
type | -0.033 |
Jen | -0.032 |
Token | and |
Token ID | 290 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
Jill | +0.016 |
kil | +0.015 |
type | +0.015 |
theirs | +0.015 |
Jen | +0.015 |
Neg logit contributions | |
lor | -0.012 |
actus | -0.012 |
� | -0.012 |
sky | -0.011 |
course | -0.010 |
Token | watering |
Token ID | 41737 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
sky | +0.016 |
� | +0.015 |
actus | +0.015 |
lor | +0.015 |
skies | +0.014 |
Neg logit contributions | |
kil | -0.015 |
ooks | -0.014 |
ierce | -0.014 |
tim | -0.013 |
Jill | -0.013 |
Token | them |
Token ID | 606 |
Feature activation | -3.94e-02 |
Pos logit contributions | |
Jill | +0.210 |
kil | +0.208 |
type | +0.202 |
ooks | +0.199 |
tim | +0.198 |
Neg logit contributions | |
actus | -0.204 |
lor | -0.201 |
sky | -0.198 |
� | -0.196 |
course | -0.185 |
Token | , |
Token ID | 11 |
Feature activation | -4.24e-02 |
Pos logit contributions | |
Jill | +0.065 |
kil | +0.061 |
Jen | +0.059 |
Matt | +0.059 |
tim | +0.059 |
Neg logit contributions | |
� | -0.056 |
sky | -0.055 |
actus | -0.055 |
lor | -0.053 |
skies | -0.048 |
Token | so |
Token ID | 523 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
kil | +0.058 |
Jill | +0.057 |
tim | +0.053 |
type | +0.053 |
ooks | +0.053 |
Neg logit contributions | |
sky | -0.071 |
� | -0.070 |
actus | -0.068 |
lor | -0.068 |
course | -0.063 |
Token | they |
Token ID | 484 |
Feature activation | -2.78e-02 |
Pos logit contributions | |
kil | +0.025 |
Jill | +0.025 |
type | +0.024 |
tim | +0.023 |
ooks | +0.023 |
Neg logit contributions | |
� | -0.019 |
lor | -0.018 |
actus | -0.018 |
sky | -0.018 |
skies | -0.015 |
Token | could |
Token ID | 714 |
Feature activation | +6.52e-02 |
Pos logit contributions | |
Jill | +0.041 |
kil | +0.040 |
ooks | +0.039 |
type | +0.037 |
Jen | +0.037 |
Neg logit contributions | |
� | -0.044 |
actus | -0.044 |
lor | -0.043 |
sky | -0.043 |
course | -0.039 |
Token | the |
Token ID | 262 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
lor | +0.020 |
actus | +0.020 |
� | +0.019 |
sky | +0.018 |
rot | +0.017 |
Neg logit contributions | |
Jill | -0.011 |
type | -0.009 |
Jen | -0.009 |
like | -0.009 |
theirs | -0.009 |
Token | garden |
Token ID | 11376 |
Feature activation | +5.84e-03 |
Pos logit contributions | |
lor | +0.057 |
� | +0.054 |
actus | +0.053 |
sky | +0.044 |
course | +0.043 |
Neg logit contributions | |
Jill | -0.075 |
type | -0.070 |
like | -0.067 |
neat | -0.066 |
Jen | -0.066 |
Token | and |
Token ID | 290 |
Feature activation | +3.48e-02 |
Pos logit contributions | |
lor | +0.011 |
actus | +0.011 |
� | +0.011 |
sky | +0.010 |
course | +0.009 |
Neg logit contributions | |
Jill | -0.008 |
type | -0.007 |
kil | -0.007 |
Jen | -0.006 |
like | -0.006 |
Token | smelled |
Token ID | 37609 |
Feature activation | -6.61e-02 |
Pos logit contributions | |
actus | +0.059 |
lor | +0.058 |
� | +0.057 |
sky | +0.054 |
course | +0.050 |
Neg logit contributions | |
Jill | -0.050 |
kil | -0.050 |
type | -0.046 |
ooks | -0.046 |
theirs | -0.045 |
Token | the |
Token ID | 262 |
Feature activation | +3.40e-02 |
Pos logit contributions | |
Jill | +0.094 |
kil | +0.091 |
ooks | +0.086 |
tim | +0.085 |
type | +0.085 |
Neg logit contributions | |
actus | -0.114 |
sky | -0.108 |
lor | -0.106 |
� | -0.105 |
course | -0.098 |
Token | frag |
Token ID | 7956 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
actus | +0.050 |
lor | +0.049 |
� | +0.047 |
sky | +0.046 |
skies | +0.040 |
Neg logit contributions | |
Jill | -0.059 |
kil | -0.055 |
type | -0.054 |
theirs | -0.053 |
Jen | -0.053 |
Token | rant |
Token ID | 5250 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
Jill | +0.139 |
type | +0.120 |
kil | +0.118 |
Jen | +0.113 |
theirs | +0.110 |
Neg logit contributions | |
lor | -0.296 |
actus | -0.293 |
� | -0.288 |
sky | -0.270 |
course | -0.257 |
Token | flowers |
Token ID | 12734 |
Feature activation | -3.27e-02 |
Pos logit contributions | |
Jill | +0.062 |
kil | +0.058 |
theirs | +0.056 |
ooks | +0.056 |
Matt | +0.056 |
Neg logit contributions | |
actus | -0.039 |
sky | -0.037 |
lor | -0.036 |
� | -0.036 |
skies | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | -6.00e-03 |
Pos logit contributions | |
Jill | +0.048 |
kil | +0.043 |
type | +0.043 |
ooks | +0.042 |
Jen | +0.042 |
Neg logit contributions | |
actus | -0.056 |
lor | -0.056 |
� | -0.055 |
sky | -0.052 |
course | -0.047 |
Token | She |
Token ID | 1375 |
Feature activation | -9.11e-03 |
Pos logit contributions | |
Jill | +0.010 |
type | +0.009 |
Jen | +0.009 |
kil | +0.009 |
like | +0.009 |
Neg logit contributions | |
lor | -0.009 |
actus | -0.009 |
� | -0.009 |
sky | -0.008 |
course | -0.007 |
Token | loved |
Token ID | 6151 |
Feature activation | +3.27e-02 |
Pos logit contributions | |
kil | +0.013 |
Jill | +0.013 |
theirs | +0.012 |
ooks | +0.012 |
type | +0.011 |
Neg logit contributions | |
actus | -0.016 |
� | -0.015 |
lor | -0.015 |
sky | -0.015 |
course | -0.013 |
Token | there |
Token ID | 612 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
lor | +0.059 |
actus | +0.058 |
� | +0.056 |
sky | +0.054 |
course | +0.050 |
Neg logit contributions | |
Jill | -0.055 |
kil | -0.055 |
tim | -0.053 |
ots | -0.053 |
type | -0.053 |
Token | were |
Token ID | 547 |
Feature activation | -3.80e-03 |
Pos logit contributions | |
lor | +0.027 |
actus | +0.027 |
� | +0.026 |
sky | +0.025 |
course | +0.023 |
Neg logit contributions | |
Jill | -0.040 |
ooks | -0.038 |
kil | -0.037 |
akers | -0.036 |
tim | -0.036 |
Token | other |
Token ID | 584 |
Feature activation | +8.84e-03 |
Pos logit contributions | |
Jill | +0.006 |
kil | +0.005 |
tim | +0.005 |
ooks | +0.005 |
type | +0.005 |
Neg logit contributions | |
actus | -0.006 |
lor | -0.006 |
sky | -0.006 |
� | -0.005 |
skies | -0.005 |
Token | long |
Token ID | 890 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
actus | +0.012 |
sky | +0.011 |
lor | +0.010 |
skies | +0.010 |
� | +0.010 |
Neg logit contributions | |
tim | -0.016 |
Jill | -0.015 |
ooks | -0.015 |
kil | -0.015 |
Liz | -0.015 |
Token | sh |
Token ID | 427 |
Feature activation | -3.70e-02 |
Pos logit contributions | |
Jill | +0.029 |
type | +0.027 |
kil | +0.027 |
Jen | +0.026 |
ooks | +0.026 |
Neg logit contributions | |
actus | -0.031 |
lor | -0.031 |
� | -0.030 |
sky | -0.028 |
course | -0.026 |
Token | rim |
Token ID | 3036 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
Jill | +0.039 |
type | +0.034 |
kil | +0.034 |
ooks | +0.033 |
like | +0.032 |
Neg logit contributions | |
lor | -0.080 |
actus | -0.077 |
� | -0.077 |
sky | -0.073 |
poster | -0.070 |
Token | ps |
Token ID | 862 |
Feature activation | -2.05e-02 |
Pos logit contributions | |
Jill | +0.242 |
type | +0.228 |
kil | +0.225 |
neat | +0.219 |
Jen | +0.218 |
Neg logit contributions | |
lor | -0.191 |
actus | -0.178 |
� | -0.165 |
sky | -0.151 |
rot | -0.149 |
Token | in |
Token ID | 287 |
Feature activation | -3.91e-02 |
Pos logit contributions | |
ooks | +0.032 |
Jill | +0.031 |
kil | +0.030 |
type | +0.030 |
Jen | +0.029 |
Neg logit contributions | |
sky | -0.030 |
actus | -0.030 |
lor | -0.029 |
� | -0.029 |
course | -0.027 |
Token | the |
Token ID | 262 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
Jill | +0.048 |
kil | +0.045 |
ooks | +0.043 |
type | +0.043 |
tim | +0.042 |
Neg logit contributions | |
lor | -0.074 |
� | -0.073 |
actus | -0.072 |
sky | -0.069 |
course | -0.066 |
Token | ocean |
Token ID | 9151 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
lor | +0.035 |
� | +0.034 |
actus | +0.032 |
sky | +0.027 |
rot | +0.027 |
Neg logit contributions | |
Jill | -0.039 |
type | -0.035 |
kil | -0.034 |
like | -0.034 |
neat | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
actus | +0.019 |
lor | +0.019 |
� | +0.018 |
sky | +0.017 |
course | +0.016 |
Neg logit contributions | |
Jill | -0.018 |
ooks | -0.016 |
type | -0.016 |
kil | -0.016 |
Jen | -0.016 |
Token | looked |
Token ID | 3114 |
Feature activation | -5.55e-02 |
Pos logit contributions | |
Jill | +0.015 |
kil | +0.014 |
type | +0.013 |
ooks | +0.013 |
theirs | +0.013 |
Neg logit contributions | |
lor | -0.018 |
actus | -0.018 |
� | -0.017 |
sky | -0.016 |
course | -0.015 |
Token | around |
Token ID | 1088 |
Feature activation | +5.18e-03 |
Pos logit contributions | |
Jill | +0.072 |
kil | +0.066 |
type | +0.066 |
ooks | +0.063 |
Jen | +0.063 |
Neg logit contributions | |
actus | -0.104 |
lor | -0.100 |
� | -0.099 |
sky | -0.096 |
course | -0.089 |
Token | and |
Token ID | 290 |
Feature activation | +3.24e-02 |
Pos logit contributions | |
� | +0.007 |
actus | +0.007 |
lor | +0.007 |
sky | +0.007 |
course | +0.006 |
Neg logit contributions | |
Jill | -0.009 |
kil | -0.008 |
type | -0.008 |
ooks | -0.008 |
theirs | -0.008 |
Token | saw |
Token ID | 2497 |
Feature activation | -1.97e-02 |
Pos logit contributions | |
actus | +0.052 |
lor | +0.051 |
� | +0.050 |
sky | +0.048 |
course | +0.043 |
Neg logit contributions | |
Jill | -0.049 |
kil | -0.048 |
type | -0.046 |
theirs | -0.045 |
ooks | -0.045 |
Token | a |
Token ID | 257 |
Feature activation | -5.42e-03 |
Pos logit contributions | |
Jill | +0.024 |
type | +0.020 |
Jen | +0.020 |
like | +0.019 |
neat | +0.019 |
Neg logit contributions | |
lor | -0.043 |
� | -0.043 |
actus | -0.042 |
rot | -0.036 |
sky | -0.036 |
Token | frag |
Token ID | 7956 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
Jill | +0.009 |
kil | +0.008 |
type | +0.008 |
Jen | +0.008 |
theirs | +0.008 |
Neg logit contributions | |
lor | -0.009 |
actus | -0.009 |
� | -0.008 |
sky | -0.008 |
course | -0.007 |
Token | lie |
Token ID | 14485 |
Feature activation | -3.01e-02 |
Pos logit contributions | |
Jill | +0.121 |
type | +0.104 |
kil | +0.101 |
like | +0.096 |
ooks | +0.095 |
Neg logit contributions | |
lor | -0.311 |
actus | -0.308 |
� | -0.306 |
sky | -0.282 |
poster | -0.266 |
Token | thing |
Token ID | 1517 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
Jill | +0.045 |
type | +0.042 |
like | +0.041 |
kil | +0.040 |
Jen | +0.039 |
Neg logit contributions | |
lor | -0.053 |
� | -0.050 |
actus | -0.049 |
sky | -0.042 |
course | -0.041 |
Token | in |
Token ID | 287 |
Feature activation | -5.06e-02 |
Pos logit contributions | |
sky | +0.022 |
actus | +0.022 |
� | +0.021 |
lor | +0.021 |
poster | +0.019 |
Neg logit contributions | |
ooks | -0.026 |
Jill | -0.025 |
kil | -0.024 |
rated | -0.024 |
type | -0.024 |
Token | the |
Token ID | 262 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
Jill | +0.061 |
kil | +0.055 |
type | +0.054 |
ooks | +0.052 |
Jen | +0.051 |
Neg logit contributions | |
lor | -0.101 |
actus | -0.099 |
� | -0.098 |
sky | -0.092 |
course | -0.086 |
Token | sky |
Token ID | 6766 |
Feature activation | -7.76e-03 |
Pos logit contributions | |
lor | +0.018 |
� | +0.017 |
actus | +0.017 |
sky | +0.014 |
course | +0.013 |
Neg logit contributions | |
Jill | -0.034 |
type | -0.032 |
kil | -0.031 |
like | -0.031 |
Jen | -0.031 |
Token | stove |
Token ID | 27635 |
Feature activation | -7.90e-03 |
Pos logit contributions | |
actus | +0.032 |
lor | +0.031 |
� | +0.030 |
sky | +0.029 |
course | +0.025 |
Neg logit contributions | |
Jill | -0.040 |
type | -0.036 |
kil | -0.036 |
Jen | -0.036 |
theirs | -0.036 |
Token | - |
Token ID | 532 |
Feature activation | -4.48e-02 |
Pos logit contributions | |
Jill | +0.013 |
ooks | +0.012 |
kil | +0.012 |
type | +0.011 |
theirs | +0.011 |
Neg logit contributions | |
lor | -0.012 |
� | -0.012 |
actus | -0.012 |
sky | -0.012 |
course | -0.011 |
Token | two |
Token ID | 734 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
rated | +0.072 |
Jill | +0.071 |
ooks | +0.071 |
ash | +0.070 |
ierce | +0.069 |
Neg logit contributions | |
sky | -0.063 |
� | -0.062 |
lor | -0.060 |
actus | -0.059 |
skies | -0.053 |
Token | more |
Token ID | 517 |
Feature activation | -2.87e-02 |
Pos logit contributions | |
Jill | +0.018 |
tim | +0.016 |
theirs | +0.016 |
ierce | +0.016 |
Jen | +0.016 |
Neg logit contributions | |
actus | -0.018 |
sky | -0.017 |
� | -0.017 |
lor | -0.017 |
skies | -0.016 |
Token | sh |
Token ID | 427 |
Feature activation | -4.68e-02 |
Pos logit contributions | |
Jill | +0.041 |
ooks | +0.038 |
tim | +0.038 |
Jen | +0.037 |
ierce | +0.037 |
Neg logit contributions | |
actus | -0.047 |
sky | -0.046 |
� | -0.044 |
lor | -0.042 |
skies | -0.041 |
Token | rim |
Token ID | 3036 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
Jill | +0.052 |
ooks | +0.046 |
type | +0.045 |
kil | +0.045 |
ots | +0.044 |
Neg logit contributions | |
lor | -0.097 |
actus | -0.097 |
� | -0.096 |
sky | -0.091 |
poster | -0.086 |
Token | ps |
Token ID | 862 |
Feature activation | -2.06e-02 |
Pos logit contributions | |
Jill | +0.241 |
type | +0.229 |
kil | +0.224 |
Jen | +0.218 |
neat | +0.216 |
Neg logit contributions | |
lor | -0.188 |
actus | -0.173 |
� | -0.162 |
sky | -0.151 |
eros | -0.143 |
Token | ! |
Token ID | 0 |
Feature activation | -9.86e-03 |
Pos logit contributions | |
ooks | +0.032 |
Jill | +0.031 |
kil | +0.030 |
type | +0.030 |
ierce | +0.029 |
Neg logit contributions | |
sky | -0.032 |
actus | -0.031 |
lor | -0.030 |
� | -0.030 |
skies | -0.028 |
Token | His |
Token ID | 2399 |
Feature activation | -4.65e-03 |
Pos logit contributions | |
Jill | +0.014 |
type | +0.013 |
kil | +0.012 |
Jen | +0.012 |
theirs | +0.012 |
Neg logit contributions | |
lor | -0.018 |
actus | -0.018 |
� | -0.018 |
sky | -0.016 |
course | -0.015 |
Token | dad |
Token ID | 9955 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
Jill | +0.006 |
type | +0.005 |
kil | +0.005 |
Jen | +0.005 |
neat | +0.005 |
Neg logit contributions | |
lor | -0.009 |
� | -0.009 |
actus | -0.009 |
sky | -0.008 |
course | -0.008 |
Token | had |
Token ID | 550 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
lor | +0.020 |
� | +0.019 |
actus | +0.019 |
sky | +0.018 |
course | +0.017 |
Neg logit contributions | |
Jill | -0.020 |
kil | -0.019 |
theirs | -0.019 |
ooks | -0.018 |
type | -0.018 |
Token | another |
Token ID | 1194 |
Feature activation | +9.94e-03 |
Pos logit contributions | |
ooks | +0.065 |
type | +0.065 |
Jill | +0.065 |
kil | +0.064 |
rated | +0.063 |
Neg logit contributions | |
lor | -0.069 |
sky | -0.064 |
actus | -0.063 |
� | -0.063 |
course | -0.061 |
Token | wand |
Token ID | 11569 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
lor | +0.016 |
� | +0.015 |
actus | +0.015 |
sky | +0.013 |
ican | +0.012 |
Neg logit contributions | |
Jill | -0.017 |
type | -0.016 |
kil | -0.015 |
like | -0.015 |
Jen | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -2.55e-02 |
Pos logit contributions | |
Jill | +0.033 |
type | +0.030 |
kil | +0.029 |
Jen | +0.028 |
theirs | +0.028 |
Neg logit contributions | |
lor | -0.045 |
actus | -0.044 |
� | -0.043 |
sky | -0.040 |
course | -0.037 |
Token | " |
Token ID | 366 |
Feature activation | -1.29e-02 |
Pos logit contributions | |
Jill | +0.048 |
type | +0.045 |
kil | +0.044 |
Jen | +0.043 |
theirs | +0.043 |
Neg logit contributions | |
lor | -0.034 |
actus | -0.034 |
� | -0.033 |
sky | -0.029 |
course | -0.026 |
Token | A |
Token ID | 32 |
Feature activation | -2.65e-02 |
Pos logit contributions | |
Jill | +0.022 |
type | +0.020 |
kil | +0.020 |
Jen | +0.019 |
like | +0.019 |
Neg logit contributions | |
lor | -0.019 |
actus | -0.019 |
� | -0.019 |
sky | -0.017 |
course | -0.015 |
Token | br |
Token ID | 1671 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
Jill | +0.043 |
type | +0.039 |
kil | +0.039 |
Jen | +0.038 |
like | +0.037 |
Neg logit contributions | |
lor | -0.042 |
actus | -0.042 |
� | -0.041 |
sky | -0.037 |
course | -0.034 |
Token | ac |
Token ID | 330 |
Feature activation | -4.79e-02 |
Pos logit contributions | |
Jill | +0.077 |
type | +0.057 |
kil | +0.055 |
Jen | +0.049 |
theirs | +0.048 |
Neg logit contributions | |
lor | -0.354 |
actus | -0.348 |
� | -0.343 |
sky | -0.324 |
course | -0.312 |
Token | ad |
Token ID | 324 |
Feature activation | -8.15e-02 |
Pos logit contributions | |
Jill | +0.067 |
Jen | +0.063 |
ac | +0.060 |
type | +0.060 |
like | +0.057 |
Neg logit contributions | |
� | -0.098 |
actus | -0.089 |
lor | -0.083 |
sky | -0.079 |
poster | -0.074 |
Token | ab |
Token ID | 397 |
Feature activation | -5.16e-02 |
Pos logit contributions | |
Jill | +0.116 |
type | +0.104 |
kil | +0.104 |
like | +0.100 |
Jen | +0.100 |
Neg logit contributions | |
lor | -0.146 |
actus | -0.145 |
� | -0.144 |
sky | -0.130 |
skies | -0.121 |
Token | ra |
Token ID | 430 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
Jill | +0.088 |
type | +0.080 |
kil | +0.080 |
Jen | +0.078 |
like | +0.077 |
Neg logit contributions | |
lor | -0.078 |
actus | -0.077 |
� | -0.075 |
sky | -0.068 |
course | -0.063 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
Jill | +0.016 |
type | +0.014 |
kil | +0.014 |
ooks | +0.014 |
theirs | +0.014 |
Neg logit contributions | |
lor | -0.022 |
actus | -0.021 |
� | -0.021 |
sky | -0.019 |
course | -0.018 |
Token | around |
Token ID | 1088 |
Feature activation | -6.88e-04 |
Pos logit contributions | |
Jill | +0.030 |
type | +0.027 |
kil | +0.027 |
ooks | +0.026 |
Jen | +0.026 |
Neg logit contributions | |
lor | -0.037 |
� | -0.037 |
actus | -0.037 |
sky | -0.035 |
course | -0.032 |
Token | the |
Token ID | 262 |
Feature activation | +2.58e-02 |
Pos logit contributions | |
ooks | +0.001 |
kil | +0.001 |
type | +0.001 |
ierce | +0.001 |
Jill | +0.001 |
Neg logit contributions | |
sky | -0.001 |
� | -0.001 |
course | -0.001 |
lor | -0.001 |
skies | -0.001 |
Token | garden |
Token ID | 11376 |
Feature activation | -1.39e-03 |
Pos logit contributions | |
actus | +0.033 |
lor | +0.032 |
� | +0.031 |
sky | +0.030 |
skies | +0.025 |
Neg logit contributions | |
Jill | -0.050 |
kil | -0.047 |
type | -0.046 |
theirs | -0.046 |
Jen | -0.045 |
Token | , |
Token ID | 11 |
Feature activation | -4.25e-02 |
Pos logit contributions | |
Jill | +0.002 |
kil | +0.002 |
type | +0.002 |
Jen | +0.002 |
theirs | +0.002 |
Neg logit contributions | |
lor | -0.002 |
actus | -0.002 |
� | -0.002 |
sky | -0.002 |
course | -0.002 |
Token | carefully |
Token ID | 7773 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
Jill | +0.063 |
kil | +0.061 |
type | +0.059 |
ooks | +0.058 |
theirs | +0.057 |
Neg logit contributions | |
� | -0.067 |
lor | -0.066 |
actus | -0.066 |
sky | -0.066 |
course | -0.059 |
Token | watering |
Token ID | 41737 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
� | +0.019 |
lor | +0.019 |
actus | +0.019 |
sky | +0.019 |
skies | +0.017 |
Neg logit contributions | |
Jill | -0.016 |
ooks | -0.015 |
kil | -0.015 |
type | -0.014 |
theirs | -0.014 |
Token | each |
Token ID | 1123 |
Feature activation | -2.71e-02 |
Pos logit contributions | |
Jill | +0.213 |
ooks | +0.213 |
type | +0.208 |
tim | +0.203 |
supposed | +0.202 |
Neg logit contributions | |
sky | -0.196 |
lor | -0.187 |
actus | -0.186 |
course | -0.180 |
� | -0.178 |
Token | plant |
Token ID | 4618 |
Feature activation | -2.90e-02 |
Pos logit contributions | |
ooks | +0.051 |
tim | +0.051 |
Jill | +0.049 |
uggets | +0.049 |
kil | +0.048 |
Neg logit contributions | |
sky | -0.033 |
course | -0.031 |
actus | -0.030 |
� | -0.027 |
tree | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -2.00e-02 |
Pos logit contributions | |
Jill | +0.046 |
kil | +0.042 |
type | +0.041 |
ooks | +0.041 |
Jen | +0.041 |
Neg logit contributions | |
lor | -0.045 |
actus | -0.044 |
� | -0.043 |
sky | -0.043 |
course | -0.039 |
Token | He |
Token ID | 679 |
Feature activation | +4.84e-03 |
Pos logit contributions | |
type | +0.038 |
kil | +0.037 |
cest | +0.037 |
akers | +0.036 |
matter | +0.036 |
Neg logit contributions | |
� | -0.027 |
sky | -0.023 |
lor | -0.023 |
actus | -0.021 |
skies | -0.019 |
Token | was |
Token ID | 373 |
Feature activation | -6.14e-03 |
Pos logit contributions | |
actus | +0.007 |
lor | +0.007 |
� | +0.007 |
sky | +0.007 |
course | +0.006 |
Neg logit contributions | |
Jill | -0.008 |
theirs | -0.008 |
ooks | -0.007 |
kil | -0.007 |
tim | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | +2.54e-02 |
Pos logit contributions | |
kil | +0.054 |
akers | +0.053 |
ooks | +0.053 |
ierce | +0.053 |
supposed | +0.052 |
Neg logit contributions | |
sky | -0.075 |
lor | -0.072 |
� | -0.072 |
course | -0.070 |
actus | -0.067 |
Token | air |
Token ID | 1633 |
Feature activation | -9.15e-03 |
Pos logit contributions | |
lor | +0.030 |
actus | +0.029 |
� | +0.029 |
ican | +0.022 |
rot | +0.021 |
Neg logit contributions | |
Jill | -0.053 |
neat | -0.049 |
type | -0.049 |
like | -0.048 |
kil | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
Jill | +0.014 |
kil | +0.013 |
type | +0.013 |
ooks | +0.012 |
ierce | +0.012 |
Neg logit contributions | |
lor | -0.015 |
actus | -0.015 |
� | -0.014 |
sky | -0.014 |
poster | -0.012 |
Token | " |
Token ID | 366 |
Feature activation | -8.41e-03 |
Pos logit contributions | |
type | +0.028 |
kil | +0.028 |
matter | +0.028 |
supposed | +0.027 |
akers | +0.027 |
Neg logit contributions | |
� | -0.031 |
lor | -0.029 |
sky | -0.029 |
actus | -0.028 |
skies | -0.024 |
Token | A |
Token ID | 32 |
Feature activation | -2.58e-02 |
Pos logit contributions | |
Jill | +0.014 |
type | +0.013 |
kil | +0.013 |
ooks | +0.013 |
theirs | +0.012 |
Neg logit contributions | |
lor | -0.013 |
actus | -0.013 |
� | -0.013 |
sky | -0.011 |
course | -0.011 |
Token | br |
Token ID | 1671 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
Jill | +0.044 |
kil | +0.039 |
type | +0.039 |
Jen | +0.039 |
theirs | +0.039 |
Neg logit contributions | |
lor | -0.040 |
actus | -0.039 |
� | -0.038 |
sky | -0.035 |
course | -0.032 |
Token | ac |
Token ID | 330 |
Feature activation | -4.43e-02 |
Pos logit contributions | |
Jill | +0.080 |
type | +0.060 |
kil | +0.058 |
Jen | +0.053 |
theirs | +0.052 |
Neg logit contributions | |
lor | -0.348 |
actus | -0.342 |
� | -0.337 |
sky | -0.319 |
course | -0.306 |
Token | ad |
Token ID | 324 |
Feature activation | -7.97e-02 |
Pos logit contributions | |
Jill | +0.061 |
type | +0.055 |
Jen | +0.055 |
ac | +0.055 |
like | +0.052 |
Neg logit contributions | |
� | -0.091 |
actus | -0.086 |
lor | -0.081 |
sky | -0.073 |
poster | -0.069 |
Token | ab |
Token ID | 397 |
Feature activation | -4.87e-02 |
Pos logit contributions | |
Jill | +0.112 |
kil | +0.101 |
type | +0.101 |
like | +0.098 |
Jen | +0.096 |
Neg logit contributions | |
lor | -0.144 |
actus | -0.143 |
� | -0.142 |
sky | -0.128 |
skies | -0.119 |
Token | ra |
Token ID | 430 |
Feature activation | -9.81e-03 |
Pos logit contributions | |
Jill | +0.080 |
type | +0.073 |
kil | +0.072 |
Jen | +0.072 |
like | +0.069 |
Neg logit contributions | |
actus | -0.077 |
lor | -0.076 |
� | -0.074 |
sky | -0.068 |
course | -0.063 |
Token | !" |
Token ID | 2474 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
Jill | +0.014 |
type | +0.012 |
kil | +0.012 |
ooks | +0.012 |
theirs | +0.012 |
Neg logit contributions | |
lor | -0.018 |
actus | -0.018 |
� | -0.017 |
sky | -0.016 |
course | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +5.64e-03 |
Pos logit contributions | |
actus | +0.012 |
lor | +0.012 |
sky | +0.012 |
� | +0.011 |
course | +0.011 |
Neg logit contributions | |
Jill | -0.013 |
kil | -0.012 |
Jen | -0.012 |
ooks | -0.012 |
type | -0.011 |
Token | One |
Token ID | 1881 |
Feature activation | -3.07e-02 |
Pos logit contributions | |
lor | +0.009 |
actus | +0.009 |
� | +0.009 |
sky | +0.008 |
course | +0.007 |
Neg logit contributions | |
Jill | -0.009 |
type | -0.008 |
kil | -0.008 |
Jen | -0.008 |
theirs | -0.008 |
Token | day |
Token ID | 1110 |
Feature activation | +5.59e-02 |
Pos logit contributions | |
Jill | +0.057 |
kil | +0.053 |
ooks | +0.052 |
type | +0.052 |
theirs | +0.051 |
Neg logit contributions | |
actus | -0.041 |
lor | -0.040 |
� | -0.039 |
sky | -0.037 |
course | -0.032 |
Token | , |
Token ID | 11 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
lor | +0.075 |
actus | +0.074 |
� | +0.074 |
sky | +0.065 |
course | +0.059 |
Neg logit contributions | |
Jill | -0.105 |
type | -0.096 |
kil | -0.095 |
theirs | -0.093 |
ooks | -0.093 |
Token | E |
Token ID | 412 |
Feature activation | -2.61e-02 |
Pos logit contributions | |
Jill | +0.054 |
type | +0.048 |
kil | +0.047 |
Jen | +0.046 |
like | +0.046 |
Neg logit contributions | |
lor | -0.073 |
actus | -0.071 |
� | -0.069 |
sky | -0.064 |
course | -0.060 |
Token | sm |
Token ID | 5796 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
Jill | +0.044 |
type | +0.040 |
kil | +0.039 |
like | +0.038 |
Jen | +0.038 |
Neg logit contributions | |
lor | -0.041 |
� | -0.039 |
actus | -0.039 |
sky | -0.036 |
course | -0.032 |
Token | ie |
Token ID | 494 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
Jill | +0.188 |
type | +0.168 |
kil | +0.166 |
Jen | +0.161 |
theirs | +0.160 |
Neg logit contributions | |
lor | -0.243 |
actus | -0.238 |
� | -0.235 |
sky | -0.215 |
course | -0.201 |
Token | decided |
Token ID | 3066 |
Feature activation | +5.21e-02 |
Pos logit contributions | |
Jill | +0.023 |
like | +0.022 |
plugged | +0.020 |
type | +0.020 |
Joe | +0.020 |
Neg logit contributions | |
� | -0.051 |
lor | -0.046 |
actus | -0.044 |
course | -0.041 |
poster | -0.040 |
Token | to |
Token ID | 284 |
Feature activation | +4.63e-02 |
Pos logit contributions | |
lor | +0.102 |
� | +0.102 |
actus | +0.101 |
sky | +0.090 |
course | +0.086 |
Neg logit contributions | |
Jill | -0.069 |
type | -0.059 |
like | -0.059 |
Jen | -0.058 |
kil | -0.057 |
Token | travel |
Token ID | 3067 |
Feature activation | -2.53e-03 |
Pos logit contributions | |
lor | +0.099 |
actus | +0.098 |
� | +0.096 |
sky | +0.089 |
course | +0.084 |
Neg logit contributions | |
Jill | -0.050 |
type | -0.044 |
kil | -0.042 |
Jen | -0.041 |
theirs | -0.041 |
Token | to |
Token ID | 284 |
Feature activation | +2.31e-02 |
Pos logit contributions | |
Jill | +0.004 |
kil | +0.004 |
ooks | +0.004 |
type | +0.004 |
Jen | +0.004 |
Neg logit contributions | |
lor | -0.004 |
actus | -0.004 |
sky | -0.004 |
� | -0.003 |
course | -0.003 |
Token | soft |
Token ID | 2705 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
Jill | +0.004 |
type | +0.004 |
kil | +0.004 |
Jen | +0.003 |
theirs | +0.003 |
Neg logit contributions | |
lor | -0.004 |
actus | -0.004 |
� | -0.004 |
sky | -0.004 |
course | -0.004 |
Token | copper |
Token ID | 15317 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
Jill | +0.041 |
type | +0.040 |
like | +0.038 |
neat | +0.037 |
kil | +0.036 |
Neg logit contributions | |
lor | -0.038 |
� | -0.037 |
actus | -0.036 |
sky | -0.030 |
course | -0.029 |
Token | coin |
Token ID | 10752 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
lor | +0.023 |
� | +0.022 |
actus | +0.021 |
sky | +0.018 |
ican | +0.017 |
Neg logit contributions | |
type | -0.030 |
Jill | -0.030 |
like | -0.030 |
kil | -0.030 |
neat | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
lor | +0.021 |
actus | +0.021 |
� | +0.021 |
sky | +0.019 |
course | +0.018 |
Neg logit contributions | |
Jill | -0.016 |
type | -0.014 |
kil | -0.014 |
Jen | -0.013 |
like | -0.013 |
Token | E |
Token ID | 412 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
Jill | +0.032 |
Jen | +0.028 |
like | +0.028 |
type | +0.027 |
neat | +0.026 |
Neg logit contributions | |
lor | -0.031 |
actus | -0.030 |
� | -0.029 |
course | -0.027 |
rot | -0.027 |
Token | sm |
Token ID | 5796 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
Jill | +0.035 |
type | +0.032 |
kil | +0.032 |
like | +0.031 |
ots | +0.031 |
Neg logit contributions | |
lor | -0.031 |
actus | -0.029 |
� | -0.029 |
sky | -0.027 |
poster | -0.024 |
Token | ie |
Token ID | 494 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
Jill | +0.210 |
type | +0.190 |
kil | +0.188 |
Jen | +0.184 |
like | +0.182 |
Neg logit contributions | |
lor | -0.220 |
actus | -0.213 |
� | -0.211 |
sky | -0.190 |
course | -0.176 |
Token | wanted |
Token ID | 2227 |
Feature activation | +4.82e-02 |
Pos logit contributions | |
Jill | +0.033 |
type | +0.032 |
plugged | +0.031 |
added | +0.029 |
like | +0.029 |
Neg logit contributions | |
� | -0.064 |
lor | -0.062 |
actus | -0.060 |
sky | -0.054 |
ican | -0.054 |
Token | to |
Token ID | 284 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
lor | +0.092 |
actus | +0.091 |
� | +0.090 |
sky | +0.080 |
poster | +0.076 |
Neg logit contributions | |
Jill | -0.065 |
type | -0.058 |
kil | -0.057 |
like | -0.056 |
Jen | -0.056 |
Token | pick |
Token ID | 2298 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
lor | +0.078 |
actus | +0.077 |
� | +0.076 |
sky | +0.071 |
course | +0.067 |
Neg logit contributions | |
Jill | -0.043 |
kil | -0.037 |
type | -0.037 |
ooks | -0.035 |
Jen | -0.035 |
Token | up |
Token ID | 510 |
Feature activation | -3.75e-03 |
Pos logit contributions | |
Jill | +0.026 |
like | +0.023 |
type | +0.022 |
kil | +0.022 |
Jen | +0.021 |
Neg logit contributions | |
lor | -0.038 |
� | -0.038 |
actus | -0.037 |
sky | -0.033 |
poster | -0.032 |
Token | wine |
Token ID | 8237 |
Feature activation | -2.50e-02 |
Pos logit contributions | |
Jill | +0.057 |
kil | +0.054 |
ooks | +0.053 |
type | +0.052 |
Jen | +0.052 |
Neg logit contributions | |
� | -0.052 |
actus | -0.051 |
lor | -0.049 |
sky | -0.048 |
course | -0.044 |
Token | that |
Token ID | 326 |
Feature activation | -2.68e-02 |
Pos logit contributions | |
ooks | +0.041 |
Jill | +0.040 |
Jen | +0.039 |
kil | +0.038 |
Matt | +0.037 |
Neg logit contributions | |
� | -0.039 |
lor | -0.037 |
actus | -0.036 |
sky | -0.036 |
course | -0.034 |
Token | was |
Token ID | 373 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
ooks | +0.047 |
kil | +0.044 |
tim | +0.042 |
theirs | +0.042 |
ierce | +0.041 |
Neg logit contributions | |
� | -0.038 |
sky | -0.038 |
lor | -0.037 |
actus | -0.036 |
course | -0.034 |
Token | sweet |
Token ID | 6029 |
Feature activation | -2.76e-02 |
Pos logit contributions | |
kil | +0.026 |
ooks | +0.026 |
ierce | +0.025 |
tim | +0.025 |
Jill | +0.025 |
Neg logit contributions | |
� | -0.027 |
sky | -0.026 |
actus | -0.024 |
lor | -0.024 |
course | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | -4.72e-03 |
Pos logit contributions | |
Jill | +0.050 |
ooks | +0.050 |
tim | +0.048 |
ierce | +0.048 |
kil | +0.048 |
Neg logit contributions | |
actus | -0.036 |
� | -0.035 |
sky | -0.034 |
lor | -0.034 |
course | -0.031 |
Token | frag |
Token ID | 7956 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
tim | +0.007 |
ooks | +0.007 |
ierce | +0.007 |
kil | +0.007 |
iding | +0.007 |
Neg logit contributions | |
� | -0.007 |
sky | -0.007 |
actus | -0.007 |
course | -0.007 |
lor | -0.006 |
Token | rant |
Token ID | 5250 |
Feature activation | -3.72e-02 |
Pos logit contributions | |
Jill | +0.163 |
type | +0.143 |
Jen | +0.142 |
kil | +0.140 |
neat | +0.135 |
Neg logit contributions | |
lor | -0.267 |
� | -0.258 |
actus | -0.258 |
sky | -0.236 |
course | -0.233 |
Token | . |
Token ID | 13 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
Jill | +0.057 |
kil | +0.056 |
tim | +0.055 |
ooks | +0.054 |
ierce | +0.052 |
Neg logit contributions | |
� | -0.058 |
actus | -0.057 |
lor | -0.056 |
sky | -0.055 |
course | -0.049 |
Token |
|
Token ID | 198 |
Feature activation | -1.09e-03 |
Pos logit contributions | |
Jill | +0.026 |
type | +0.024 |
kil | +0.024 |
theirs | +0.023 |
Jen | +0.023 |
Neg logit contributions | |
lor | -0.024 |
actus | -0.024 |
� | -0.023 |
sky | -0.021 |
course | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | -6.48e-03 |
Pos logit contributions | |
Jill | +0.002 |
Jen | +0.002 |
type | +0.002 |
kil | +0.002 |
ooks | +0.002 |
Neg logit contributions | |
lor | -0.001 |
actus | -0.001 |
� | -0.001 |
sky | -0.001 |
poster | -0.001 |
Token | She |
Token ID | 3347 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
Jill | +0.011 |
type | +0.010 |
kil | +0.010 |
Jen | +0.009 |
like | +0.009 |
Neg logit contributions | |
lor | -0.010 |
actus | -0.010 |
� | -0.010 |
sky | -0.009 |
poster | -0.008 |
Token | They |
Token ID | 2990 |
Feature activation | -4.42e-03 |
Pos logit contributions | |
Jill | +0.013 |
type | +0.011 |
kil | +0.011 |
Jen | +0.011 |
theirs | +0.011 |
Neg logit contributions | |
lor | -0.012 |
actus | -0.011 |
� | -0.011 |
sky | -0.010 |
course | -0.009 |
Token | all |
Token ID | 477 |
Feature activation | -3.94e-02 |
Pos logit contributions | |
Jill | +0.006 |
kil | +0.006 |
ooks | +0.006 |
type | +0.006 |
Jen | +0.006 |
Neg logit contributions | |
lor | -0.007 |
� | -0.007 |
actus | -0.007 |
sky | -0.007 |
course | -0.006 |
Token | worked |
Token ID | 3111 |
Feature activation | +1.34e-03 |
Pos logit contributions | |
kil | +0.064 |
Jill | +0.063 |
ooks | +0.062 |
type | +0.059 |
theirs | +0.059 |
Neg logit contributions | |
� | -0.052 |
actus | -0.052 |
sky | -0.052 |
lor | -0.051 |
course | -0.051 |
Token | together |
Token ID | 1978 |
Feature activation | -6.67e-03 |
Pos logit contributions | |
lor | +0.003 |
actus | +0.003 |
� | +0.003 |
sky | +0.002 |
course | +0.002 |
Neg logit contributions | |
Jill | -0.002 |
type | -0.002 |
kil | -0.002 |
Jen | -0.001 |
ooks | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | -3.06e-02 |
Pos logit contributions | |
Jill | +0.012 |
kil | +0.011 |
ooks | +0.011 |
type | +0.011 |
Jen | +0.011 |
Neg logit contributions | |
� | -0.009 |
lor | -0.008 |
sky | -0.008 |
actus | -0.008 |
course | -0.007 |
Token | watering |
Token ID | 41737 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
Jill | +0.049 |
kil | +0.048 |
type | +0.047 |
ooks | +0.045 |
theirs | +0.044 |
Neg logit contributions | |
� | -0.045 |
lor | -0.043 |
sky | -0.042 |
actus | -0.042 |
course | -0.038 |
Token | and |
Token ID | 290 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
Jill | +0.199 |
kil | +0.193 |
ooks | +0.190 |
type | +0.189 |
supposed | +0.184 |
Neg logit contributions | |
sky | -0.199 |
� | -0.197 |
lor | -0.195 |
actus | -0.194 |
course | -0.187 |
Token | tending |
Token ID | 44681 |
Feature activation | -5.59e-02 |
Pos logit contributions | |
sky | +0.021 |
� | +0.020 |
skies | +0.019 |
actus | +0.019 |
course | +0.019 |
Neg logit contributions | |
kil | -0.016 |
Jill | -0.015 |
tim | -0.015 |
ooks | -0.015 |
theirs | -0.014 |
Token | to |
Token ID | 284 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
kil | +0.098 |
tim | +0.096 |
ooks | +0.094 |
ierce | +0.093 |
Jill | +0.093 |
Neg logit contributions | |
sky | -0.072 |
course | -0.066 |
� | -0.062 |
actus | -0.059 |
skies | -0.058 |
Token | the |
Token ID | 262 |
Feature activation | +3.82e-02 |
Pos logit contributions | |
rated | +0.028 |
kil | +0.027 |
ash | +0.026 |
ooks | +0.026 |
iding | +0.026 |
Neg logit contributions | |
sky | -0.023 |
course | -0.023 |
� | -0.022 |
tree | -0.020 |
skies | -0.020 |
Token | tree |
Token ID | 5509 |
Feature activation | +2.44e-02 |
Pos logit contributions | |
lor | +0.051 |
actus | +0.049 |
� | +0.048 |
sky | +0.041 |
course | +0.038 |
Neg logit contributions | |
Jill | -0.073 |
type | -0.068 |
kil | -0.067 |
Jen | -0.065 |
like | -0.065 |
Token | care |
Token ID | 1337 |
Feature activation | -2.84e-02 |
Pos logit contributions | |
Jill | +0.065 |
type | +0.058 |
kil | +0.057 |
like | +0.056 |
Jen | +0.056 |
Neg logit contributions | |
lor | -0.083 |
actus | -0.082 |
� | -0.080 |
sky | -0.072 |
poster | -0.068 |
Token | of |
Token ID | 286 |
Feature activation | -5.83e-02 |
Pos logit contributions | |
Jill | +0.043 |
kil | +0.039 |
type | +0.039 |
Jen | +0.038 |
tim | +0.038 |
Neg logit contributions | |
lor | -0.046 |
actus | -0.045 |
� | -0.044 |
sky | -0.043 |
course | -0.040 |
Token | the |
Token ID | 262 |
Feature activation | +3.31e-02 |
Pos logit contributions | |
Jill | +0.078 |
kil | +0.072 |
ooks | +0.070 |
type | +0.069 |
tim | +0.068 |
Neg logit contributions | |
lor | -0.104 |
� | -0.103 |
actus | -0.103 |
sky | -0.096 |
course | -0.095 |
Token | grass |
Token ID | 8701 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
actus | +0.043 |
lor | +0.040 |
� | +0.040 |
sky | +0.040 |
skies | +0.034 |
Neg logit contributions | |
Jill | -0.065 |
kil | -0.059 |
ooks | -0.057 |
type | -0.057 |
Jen | -0.057 |
Token | , |
Token ID | 11 |
Feature activation | -4.08e-02 |
Pos logit contributions | |
lor | +0.022 |
actus | +0.022 |
� | +0.022 |
sky | +0.020 |
course | +0.018 |
Neg logit contributions | |
Jill | -0.020 |
kil | -0.018 |
type | -0.018 |
ooks | -0.018 |
Jen | -0.018 |
Token | watering |
Token ID | 41737 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
Jill | +0.057 |
type | +0.051 |
kil | +0.051 |
Jen | +0.049 |
ooks | +0.048 |
Neg logit contributions | |
lor | -0.074 |
actus | -0.073 |
� | -0.072 |
sky | -0.066 |
course | -0.062 |
Token | it |
Token ID | 340 |
Feature activation | -3.19e-03 |
Pos logit contributions | |
Jill | +0.205 |
kil | +0.192 |
type | +0.189 |
ooks | +0.189 |
tim | +0.182 |
Neg logit contributions | |
actus | -0.209 |
lor | -0.208 |
� | -0.202 |
sky | -0.196 |
course | -0.187 |
Token | and |
Token ID | 290 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
Jill | +0.005 |
kil | +0.005 |
ooks | +0.005 |
Jen | +0.004 |
type | +0.004 |
Neg logit contributions | |
lor | -0.005 |
� | -0.005 |
actus | -0.005 |
sky | -0.005 |
course | -0.004 |
Token | trim |
Token ID | 15797 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
lor | +0.023 |
� | +0.023 |
actus | +0.023 |
sky | +0.022 |
course | +0.021 |
Neg logit contributions | |
Jill | -0.017 |
kil | -0.016 |
ooks | -0.015 |
tim | -0.015 |
type | -0.015 |
Token | ming |
Token ID | 2229 |
Feature activation | -2.76e-02 |
Pos logit contributions | |
Jill | +0.042 |
type | +0.038 |
kil | +0.038 |
ooks | +0.037 |
Jen | +0.036 |
Neg logit contributions | |
lor | -0.044 |
actus | -0.043 |
� | -0.042 |
sky | -0.038 |
course | -0.036 |
Token | it |
Token ID | 340 |
Feature activation | -7.80e-03 |
Pos logit contributions | |
Jill | +0.040 |
ooks | +0.037 |
kil | +0.037 |
type | +0.037 |
tim | +0.035 |
Neg logit contributions | |
lor | -0.045 |
actus | -0.044 |
� | -0.044 |
sky | -0.043 |
course | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
actus | +0.017 |
lor | +0.016 |
� | +0.016 |
sky | +0.015 |
poster | +0.014 |
Neg logit contributions | |
Jill | -0.014 |
kil | -0.012 |
type | -0.012 |
ooks | -0.012 |
Jen | -0.012 |
Token | They |
Token ID | 1119 |
Feature activation | -3.77e-03 |
Pos logit contributions | |
Jill | +0.021 |
kil | +0.019 |
type | +0.019 |
theirs | +0.018 |
neat | +0.018 |
Neg logit contributions | |
� | -0.027 |
lor | -0.027 |
actus | -0.027 |
sky | -0.026 |
skies | -0.023 |
Token | see |
Token ID | 766 |
Feature activation | -3.80e-02 |
Pos logit contributions | |
Jill | +0.004 |
type | +0.004 |
like | +0.004 |
kil | +0.004 |
Jen | +0.004 |
Neg logit contributions | |
lor | -0.008 |
actus | -0.008 |
� | -0.008 |
sky | -0.007 |
rot | -0.007 |
Token | many |
Token ID | 867 |
Feature activation | -7.01e-03 |
Pos logit contributions | |
Jill | +0.046 |
type | +0.040 |
kil | +0.040 |
Jen | +0.039 |
theirs | +0.038 |
Neg logit contributions | |
lor | -0.077 |
actus | -0.076 |
� | -0.075 |
sky | -0.069 |
course | -0.065 |
Token | sh |
Token ID | 427 |
Feature activation | -3.94e-02 |
Pos logit contributions | |
Jill | +0.012 |
tim | +0.012 |
Jen | +0.011 |
Liz | +0.011 |
ierce | +0.011 |
Neg logit contributions | |
actus | -0.011 |
sky | -0.010 |
course | -0.009 |
skies | -0.009 |
� | -0.008 |
Token | rim |
Token ID | 3036 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
Jill | +0.036 |
ooks | +0.032 |
type | +0.030 |
kil | +0.030 |
ots | +0.029 |
Neg logit contributions | |
lor | -0.090 |
� | -0.088 |
actus | -0.087 |
sky | -0.083 |
poster | -0.079 |
Token | ps |
Token ID | 862 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
Jill | +0.238 |
type | +0.224 |
kil | +0.222 |
neat | +0.211 |
Jen | +0.211 |
Neg logit contributions | |
lor | -0.182 |
actus | -0.176 |
� | -0.162 |
sky | -0.151 |
course | -0.142 |
Token | and |
Token ID | 290 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
Jill | +0.028 |
ooks | +0.028 |
kil | +0.027 |
type | +0.027 |
Jen | +0.026 |
Neg logit contributions | |
sky | -0.022 |
actus | -0.021 |
� | -0.021 |
lor | -0.020 |
course | -0.019 |
Token | fish |
Token ID | 5916 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
sky | +0.033 |
actus | +0.033 |
� | +0.033 |
lor | +0.032 |
poster | +0.030 |
Neg logit contributions | |
Jill | -0.025 |
ooks | -0.025 |
kil | -0.025 |
ots | -0.023 |
ash | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | +3.00e-02 |
Pos logit contributions | |
Jill | +0.025 |
type | +0.023 |
kil | +0.023 |
Jen | +0.022 |
theirs | +0.022 |
Neg logit contributions | |
lor | -0.032 |
actus | -0.032 |
� | -0.031 |
sky | -0.029 |
course | -0.027 |
Token | crabs |
Token ID | 49172 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
actus | +0.047 |
sky | +0.047 |
� | +0.044 |
lor | +0.044 |
course | +0.042 |
Neg logit contributions | |
Jill | -0.043 |
ooks | -0.043 |
kil | -0.042 |
ots | -0.041 |
type | -0.041 |
Token | covering |
Token ID | 9505 |
Feature activation | -3.93e-02 |
Pos logit contributions | |
Jill | +0.003 |
kil | +0.002 |
type | +0.002 |
ooks | +0.002 |
tim | +0.002 |
Neg logit contributions | |
lor | -0.004 |
actus | -0.004 |
� | -0.004 |
sky | -0.004 |
course | -0.004 |
Token | them |
Token ID | 606 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
Jill | +0.051 |
kil | +0.051 |
ooks | +0.047 |
tim | +0.047 |
type | +0.047 |
Neg logit contributions | |
actus | -0.069 |
lor | -0.069 |
� | -0.066 |
sky | -0.066 |
course | -0.062 |
Token | with |
Token ID | 351 |
Feature activation | -5.98e-02 |
Pos logit contributions | |
Jill | +0.040 |
kil | +0.036 |
Jen | +0.035 |
ooks | +0.035 |
type | +0.035 |
Neg logit contributions | |
lor | -0.043 |
actus | -0.043 |
� | -0.042 |
sky | -0.039 |
course | -0.036 |
Token | soil |
Token ID | 9260 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
Jill | +0.085 |
kil | +0.079 |
ooks | +0.077 |
tim | +0.075 |
type | +0.074 |
Neg logit contributions | |
� | -0.100 |
lor | -0.100 |
actus | -0.100 |
sky | -0.096 |
course | -0.091 |
Token | and |
Token ID | 290 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
Jill | +0.022 |
type | +0.020 |
kil | +0.020 |
Jen | +0.020 |
ooks | +0.020 |
Neg logit contributions | |
lor | -0.020 |
actus | -0.020 |
� | -0.020 |
sky | -0.018 |
course | -0.016 |
Token | watering |
Token ID | 41737 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
actus | +0.036 |
lor | +0.036 |
� | +0.036 |
sky | +0.035 |
course | +0.032 |
Neg logit contributions | |
Jill | -0.029 |
kil | -0.028 |
tim | -0.027 |
ooks | -0.026 |
type | -0.026 |
Token | them |
Token ID | 606 |
Feature activation | -3.88e-02 |
Pos logit contributions | |
Jill | +0.212 |
kil | +0.208 |
type | +0.201 |
tim | +0.200 |
ooks | +0.199 |
Neg logit contributions | |
actus | -0.192 |
lor | -0.186 |
sky | -0.186 |
� | -0.184 |
course | -0.169 |
Token | every |
Token ID | 790 |
Feature activation | -6.22e-04 |
Pos logit contributions | |
Jill | +0.064 |
kil | +0.060 |
ooks | +0.059 |
Jen | +0.059 |
Matt | +0.059 |
Neg logit contributions | |
actus | -0.055 |
lor | -0.054 |
� | -0.054 |
sky | -0.054 |
skies | -0.046 |
Token | day |
Token ID | 1110 |
Feature activation | +4.23e-02 |
Pos logit contributions | |
Jill | +0.001 |
kil | +0.001 |
ooks | +0.001 |
type | +0.001 |
Jen | +0.001 |
Neg logit contributions | |
actus | -0.001 |
lor | -0.001 |
� | -0.001 |
sky | -0.001 |
course | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
� | +0.061 |
lor | +0.061 |
sky | +0.061 |
actus | +0.059 |
skies | +0.052 |
Neg logit contributions | |
Jill | -0.067 |
kil | -0.064 |
ooks | -0.063 |
type | -0.063 |
tim | -0.063 |
Token | Her |
Token ID | 2332 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
Jill | +0.020 |
Jen | +0.018 |
Mia | +0.017 |
Liz | +0.017 |
Jane | +0.017 |
Neg logit contributions | |
lor | -0.022 |
actus | -0.021 |
� | -0.018 |
rot | -0.018 |
poster | -0.018 |
Token | car |
Token ID | 1097 |
Feature activation | +1.79e-02 |
Pos logit contributions | |
lor | +0.037 |
actus | +0.034 |
� | +0.034 |
course | +0.027 |
pillar | +0.027 |
Neg logit contributions | |
Jill | -0.040 |
type | -0.037 |
kil | -0.036 |
Jen | -0.036 |
like | -0.036 |
Token | , |
Token ID | 11 |
Feature activation | -5.16e-02 |
Pos logit contributions | |
lor | +0.031 |
� | +0.030 |
actus | +0.030 |
ican | +0.025 |
sky | +0.024 |
Neg logit contributions | |
Jill | -0.027 |
like | -0.026 |
Jen | -0.025 |
type | -0.025 |
kil | -0.024 |
Token | and |
Token ID | 290 |
Feature activation | +8.11e-03 |
Pos logit contributions | |
Jill | +0.074 |
type | +0.071 |
kil | +0.071 |
ooks | +0.071 |
ierce | +0.067 |
Neg logit contributions | |
sky | -0.085 |
� | -0.081 |
actus | -0.080 |
lor | -0.077 |
course | -0.076 |
Token | the |
Token ID | 262 |
Feature activation | +2.51e-02 |
Pos logit contributions | |
sky | +0.012 |
actus | +0.011 |
course | +0.011 |
helicopter | +0.011 |
� | +0.011 |
Neg logit contributions | |
ooks | -0.013 |
kil | -0.012 |
ierce | -0.012 |
ots | -0.012 |
Jill | -0.012 |
Token | fire |
Token ID | 2046 |
Feature activation | -3.57e-02 |
Pos logit contributions | |
lor | +0.039 |
actus | +0.037 |
� | +0.036 |
sky | +0.029 |
course | +0.029 |
Neg logit contributions | |
Jill | -0.044 |
type | -0.041 |
kil | -0.040 |
Jen | -0.039 |
like | -0.039 |
Token | hyd |
Token ID | 7409 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
Jill | +0.057 |
type | +0.052 |
kil | +0.051 |
Jen | +0.051 |
like | +0.050 |
Neg logit contributions | |
lor | -0.059 |
actus | -0.057 |
� | -0.056 |
sky | -0.050 |
course | -0.047 |
Token | rant |
Token ID | 5250 |
Feature activation | -4.24e-02 |
Pos logit contributions | |
Jill | +0.161 |
type | +0.146 |
kil | +0.137 |
Jen | +0.129 |
whether | +0.129 |
Neg logit contributions | |
actus | -0.262 |
lor | -0.259 |
� | -0.244 |
sky | -0.244 |
skies | -0.225 |
Token | are |
Token ID | 389 |
Feature activation | -1.65e-02 |
Pos logit contributions | |
Jill | +0.067 |
type | +0.061 |
kil | +0.060 |
ooks | +0.059 |
Jen | +0.057 |
Neg logit contributions | |
actus | -0.067 |
lor | -0.067 |
� | -0.066 |
sky | -0.063 |
course | -0.057 |
Token | wet |
Token ID | 9583 |
Feature activation | -3.06e-02 |
Pos logit contributions | |
Jill | +0.027 |
ierce | +0.027 |
ash | +0.026 |
kil | +0.025 |
type | +0.025 |
Neg logit contributions | |
sky | -0.023 |
� | -0.020 |
skies | -0.019 |
cage | -0.019 |
actus | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -2.61e-02 |
Pos logit contributions | |
Jill | +0.046 |
type | +0.042 |
kil | +0.042 |
Jen | +0.040 |
ooks | +0.040 |
Neg logit contributions | |
lor | -0.051 |
actus | -0.051 |
� | -0.050 |
sky | -0.047 |
course | -0.043 |
Token | Lily |
Token ID | 20037 |
Feature activation | -5.39e-03 |
Pos logit contributions | |
Jill | +0.046 |
type | +0.044 |
kil | +0.044 |
theirs | +0.042 |
neat | +0.042 |
Neg logit contributions | |
lor | -0.036 |
� | -0.036 |
actus | -0.035 |
sky | -0.032 |
course | -0.029 |
Token | river |
Token ID | 7850 |
Feature activation | +5.02e-03 |
Pos logit contributions | |
lor | +0.042 |
� | +0.039 |
actus | +0.038 |
rot | +0.032 |
course | +0.031 |
Neg logit contributions | |
Jill | -0.049 |
kil | -0.045 |
type | -0.045 |
like | -0.045 |
neat | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | -2.04e-02 |
Pos logit contributions | |
lor | +0.009 |
actus | +0.009 |
� | +0.009 |
sky | +0.008 |
course | +0.008 |
Neg logit contributions | |
Jill | -0.007 |
type | -0.006 |
kil | -0.006 |
Jen | -0.006 |
like | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +6.35e-03 |
Pos logit contributions | |
Jill | +0.035 |
Jen | +0.033 |
Liz | +0.030 |
Matt | +0.030 |
like | +0.030 |
Neg logit contributions | |
actus | -0.033 |
lor | -0.032 |
� | -0.029 |
course | -0.028 |
poster | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | -1.50e-04 |
Pos logit contributions | |
actus | +0.011 |
� | +0.010 |
lor | +0.010 |
ican | +0.009 |
eros | +0.009 |
Neg logit contributions | |
Jill | -0.010 |
Jen | -0.009 |
Matt | -0.009 |
like | -0.009 |
Jane | -0.009 |
Token | When |
Token ID | 2215 |
Feature activation | -1.17e-03 |
Pos logit contributions | |
Jill | +0.000 |
type | +0.000 |
like | +0.000 |
Jen | +0.000 |
kil | +0.000 |
Neg logit contributions | |
actus | -0.000 |
� | -0.000 |
lor | -0.000 |
sky | -0.000 |
eros | -0.000 |
Token | Sil |
Token ID | 4243 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
Jill | +0.002 |
Jen | +0.002 |
type | +0.002 |
kil | +0.001 |
like | +0.001 |
Neg logit contributions | |
lor | -0.002 |
actus | -0.002 |
� | -0.002 |
sky | -0.002 |
course | -0.002 |
Token | via |
Token ID | 8869 |
Feature activation | -3.38e-02 |
Pos logit contributions | |
Jill | +0.132 |
type | +0.121 |
ots | +0.116 |
kil | +0.113 |
ierce | +0.111 |
Neg logit contributions | |
� | -0.296 |
actus | -0.285 |
lor | -0.285 |
sky | -0.268 |
skies | -0.253 |
Token | arrived |
Token ID | 5284 |
Feature activation | +4.03e-03 |
Pos logit contributions | |
Jill | +0.045 |
like | +0.040 |
type | +0.038 |
Jen | +0.038 |
plugged | +0.037 |
Neg logit contributions | |
lor | -0.066 |
� | -0.065 |
actus | -0.063 |
sky | -0.057 |
poster | -0.056 |
Token | at |
Token ID | 379 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
lor | +0.009 |
actus | +0.009 |
� | +0.009 |
sky | +0.008 |
course | +0.008 |
Neg logit contributions | |
Jill | -0.004 |
type | -0.003 |
kil | -0.003 |
Jen | -0.003 |
like | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | +2.89e-02 |
Pos logit contributions | |
Jill | +0.020 |
Jen | +0.016 |
type | +0.015 |
theirs | +0.014 |
like | +0.014 |
Neg logit contributions | |
lor | -0.047 |
actus | -0.045 |
� | -0.044 |
rot | -0.040 |
sky | -0.039 |
Token | river |
Token ID | 7850 |
Feature activation | +7.67e-03 |
Pos logit contributions | |
lor | +0.049 |
actus | +0.045 |
� | +0.045 |
eros | +0.038 |
rot | +0.038 |
Neg logit contributions | |
Jill | -0.046 |
type | -0.044 |
like | -0.044 |
neat | -0.042 |
organize | -0.041 |
Token | for |
Token ID | 329 |
Feature activation | -3.67e-02 |
Pos logit contributions | |
ooks | +0.049 |
ierce | +0.047 |
kil | +0.047 |
rated | +0.046 |
tim | +0.046 |
Neg logit contributions | |
lor | -0.043 |
sky | -0.041 |
course | -0.041 |
actus | -0.039 |
� | -0.039 |
Token | you |
Token ID | 345 |
Feature activation | -3.35e-02 |
Pos logit contributions | |
Jill | +0.068 |
kil | +0.067 |
ooks | +0.065 |
type | +0.064 |
rated | +0.064 |
Neg logit contributions | |
� | -0.044 |
lor | -0.043 |
actus | -0.041 |
course | -0.039 |
sky | -0.038 |
Token | Jane |
Token ID | 12091 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
Jill | +0.052 |
kil | +0.050 |
type | +0.049 |
ooks | +0.047 |
tim | +0.047 |
Neg logit contributions | |
lor | -0.050 |
actus | -0.049 |
� | -0.049 |
sky | -0.046 |
course | -0.043 |
Token | , |
Token ID | 11 |
Feature activation | -6.70e-02 |
Pos logit contributions | |
Jill | +0.038 |
kil | +0.035 |
type | +0.035 |
ooks | +0.034 |
theirs | +0.034 |
Neg logit contributions | |
lor | -0.036 |
actus | -0.035 |
� | -0.034 |
sky | -0.031 |
course | -0.029 |
Token | you |
Token ID | 345 |
Feature activation | -3.79e-02 |
Pos logit contributions | |
kil | +0.089 |
ooks | +0.088 |
tim | +0.087 |
Jill | +0.083 |
akers | +0.082 |
Neg logit contributions | |
lor | -0.110 |
course | -0.108 |
� | -0.108 |
sky | -0.107 |
actus | -0.104 |
Token | can |
Token ID | 460 |
Feature activation | +6.87e-02 |
Pos logit contributions | |
Jill | +0.056 |
kil | +0.053 |
type | +0.052 |
ooks | +0.051 |
theirs | +0.050 |
Neg logit contributions | |
lor | -0.062 |
actus | -0.061 |
� | -0.061 |
sky | -0.056 |
course | -0.054 |
Token | chew |
Token ID | 34722 |
Feature activation | -2.02e-02 |
Pos logit contributions | |
lor | +0.144 |
actus | +0.142 |
� | +0.138 |
sky | +0.130 |
course | +0.121 |
Neg logit contributions | |
Jill | -0.078 |
type | -0.068 |
kil | -0.067 |
like | -0.065 |
Jen | -0.064 |
Token | it |
Token ID | 340 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
Jill | +0.031 |
type | +0.028 |
kil | +0.028 |
ooks | +0.027 |
Jen | +0.027 |
Neg logit contributions | |
lor | -0.034 |
actus | -0.034 |
� | -0.033 |
sky | -0.030 |
course | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -2.70e-02 |
Pos logit contributions | |
lor | +0.020 |
actus | +0.020 |
� | +0.020 |
sky | +0.019 |
course | +0.018 |
Neg logit contributions | |
Jill | -0.021 |
ooks | -0.020 |
kil | -0.020 |
type | -0.020 |
Jen | -0.019 |
Token | But |
Token ID | 887 |
Feature activation | -3.46e-02 |
Pos logit contributions | |
Jill | +0.043 |
type | +0.040 |
kil | +0.040 |
theirs | +0.039 |
neat | +0.039 |
Neg logit contributions | |
lor | -0.043 |
� | -0.042 |
actus | -0.040 |
sky | -0.037 |
course | -0.035 |
Token | no |
Token ID | 645 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
Jill | +0.047 |
ooks | +0.047 |
kil | +0.047 |
tim | +0.046 |
ots | +0.044 |
Neg logit contributions | |
� | -0.056 |
lor | -0.056 |
course | -0.053 |
actus | -0.052 |
sky | -0.052 |
Token | to |
Token ID | 284 |
Feature activation | +3.69e-02 |
Pos logit contributions | |
lor | +0.075 |
actus | +0.073 |
sky | +0.070 |
� | +0.069 |
skies | +0.063 |
Neg logit contributions | |
Jill | -0.073 |
kil | -0.067 |
type | -0.066 |
ooks | -0.065 |
Jen | -0.064 |
Token | tell |
Token ID | 1560 |
Feature activation | -2.75e-03 |
Pos logit contributions | |
lor | +0.075 |
actus | +0.074 |
� | +0.073 |
sky | +0.067 |
course | +0.063 |
Neg logit contributions | |
Jill | -0.045 |
type | -0.039 |
kil | -0.038 |
Jen | -0.037 |
like | -0.037 |
Token | everyone |
Token ID | 2506 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
Jill | +0.003 |
ooks | +0.003 |
kil | +0.003 |
akers | +0.003 |
type | +0.003 |
Neg logit contributions | |
lor | -0.005 |
actus | -0.005 |
sky | -0.005 |
course | -0.004 |
skies | -0.004 |
Token | about |
Token ID | 546 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
ooks | +0.021 |
Jill | +0.020 |
akers | +0.019 |
ots | +0.019 |
blend | +0.019 |
Neg logit contributions | |
actus | -0.023 |
lor | -0.023 |
helicopter | -0.022 |
sky | -0.022 |
course | -0.021 |
Token | his |
Token ID | 465 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
Jill | +0.014 |
ooks | +0.013 |
kil | +0.013 |
type | +0.012 |
ierce | +0.012 |
Neg logit contributions | |
lor | -0.018 |
actus | -0.018 |
� | -0.018 |
sky | -0.017 |
course | -0.017 |
Token | dream |
Token ID | 4320 |
Feature activation | +5.12e-02 |
Pos logit contributions | |
lor | +0.034 |
� | +0.033 |
actus | +0.030 |
ican | +0.027 |
course | +0.027 |
Neg logit contributions | |
type | -0.026 |
Jill | -0.025 |
like | -0.025 |
neat | -0.024 |
remember | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
lor | +0.095 |
actus | +0.094 |
� | +0.092 |
sky | +0.084 |
course | +0.079 |
Neg logit contributions | |
Jill | -0.070 |
type | -0.063 |
kil | -0.062 |
Jen | -0.059 |
like | -0.059 |
Token |
|
Token ID | 198 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
Jill | +0.019 |
Jen | +0.017 |
type | +0.017 |
kil | +0.017 |
like | +0.017 |
Neg logit contributions | |
lor | -0.014 |
actus | -0.014 |
� | -0.013 |
sky | -0.011 |
rot | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | +2.18e-03 |
Pos logit contributions | |
actus | +0.018 |
lor | +0.018 |
� | +0.017 |
ican | +0.015 |
sky | +0.015 |
Neg logit contributions | |
Jill | -0.020 |
Jen | -0.018 |
type | -0.018 |
like | -0.017 |
ooks | -0.017 |
Token | So |
Token ID | 2396 |
Feature activation | +1.32e-04 |
Pos logit contributions | |
lor | +0.003 |
actus | +0.003 |
� | +0.003 |
sky | +0.003 |
course | +0.003 |
Neg logit contributions | |
Jill | -0.004 |
type | -0.003 |
kil | -0.003 |
Jen | -0.003 |
theirs | -0.003 |
Token | one |
Token ID | 530 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
lor | +0.000 |
actus | +0.000 |
� | +0.000 |
sky | +0.000 |
course | +0.000 |
Neg logit contributions | |
Jill | -0.000 |
kil | -0.000 |
type | -0.000 |
ooks | -0.000 |
ots | -0.000 |
Token | , |
Token ID | 11 |
Feature activation | -3.23e-02 |
Pos logit contributions | |
lor | +0.103 |
actus | +0.100 |
� | +0.097 |
sky | +0.093 |
course | +0.084 |
Neg logit contributions | |
Jill | -0.088 |
kil | -0.080 |
type | -0.080 |
ooks | -0.078 |
Jen | -0.077 |
Token | so |
Token ID | 523 |
Feature activation | -2.18e-03 |
Pos logit contributions | |
Jill | +0.047 |
kil | +0.044 |
type | +0.044 |
ooks | +0.044 |
tim | +0.041 |
Neg logit contributions | |
lor | -0.053 |
� | -0.052 |
actus | -0.052 |
sky | -0.051 |
course | -0.046 |
Token | one |
Token ID | 530 |
Feature activation | -2.79e-02 |
Pos logit contributions | |
kil | +0.003 |
Jill | +0.003 |
type | +0.003 |
tim | +0.003 |
organize | +0.003 |
Neg logit contributions | |
� | -0.003 |
sky | -0.003 |
lor | -0.003 |
actus | -0.003 |
skies | -0.003 |
Token | day |
Token ID | 1110 |
Feature activation | +4.38e-02 |
Pos logit contributions | |
Jill | +0.054 |
ooks | +0.051 |
kil | +0.050 |
theirs | +0.050 |
type | +0.049 |
Neg logit contributions | |
actus | -0.034 |
lor | -0.033 |
sky | -0.032 |
� | -0.031 |
course | -0.027 |
Token | the |
Token ID | 262 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
sky | +0.054 |
� | +0.047 |
lor | +0.046 |
skies | +0.045 |
actus | +0.041 |
Neg logit contributions | |
kil | -0.080 |
rated | -0.078 |
ash | -0.078 |
ooks | -0.077 |
type | -0.076 |
Token | wish |
Token ID | 4601 |
Feature activation | +5.25e-02 |
Pos logit contributions | |
� | +0.035 |
lor | +0.033 |
actus | +0.032 |
ican | +0.028 |
course | +0.026 |
Neg logit contributions | |
Jill | -0.029 |
type | -0.027 |
like | -0.026 |
neat | -0.026 |
kil | -0.026 |
Token | took |
Token ID | 1718 |
Feature activation | -2.79e-02 |
Pos logit contributions | |
lor | +0.093 |
actus | +0.092 |
� | +0.089 |
sky | +0.084 |
course | +0.077 |
Neg logit contributions | |
Jill | -0.075 |
type | -0.068 |
kil | -0.068 |
ooks | -0.065 |
Jen | -0.065 |
Token | a |
Token ID | 257 |
Feature activation | +2.95e-03 |
Pos logit contributions | |
Jill | +0.038 |
kil | +0.036 |
type | +0.035 |
ooks | +0.034 |
Jen | +0.033 |
Neg logit contributions | |
lor | -0.050 |
actus | -0.047 |
� | -0.046 |
sky | -0.045 |
course | -0.043 |
Token | stand |
Token ID | 1302 |
Feature activation | +2.11e-06 |
Pos logit contributions | |
� | +0.006 |
lor | +0.006 |
actus | +0.006 |
ican | +0.005 |
eros | +0.005 |
Neg logit contributions | |
Jill | -0.004 |
type | -0.004 |
neat | -0.004 |
like | -0.004 |
kil | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -1.39e-02 |
Token |
|
Token ID | 198 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
Jill | +0.025 |
Jen | +0.022 |
type | +0.022 |
Joe | +0.021 |
Jane | +0.021 |
Neg logit contributions | |
actus | -0.022 |
lor | -0.022 |
� | -0.019 |
rot | -0.017 |
poster | -0.017 |
Token | they |
Token ID | 484 |
Feature activation | -7.56e-03 |
Pos logit contributions | |
Jill | +0.037 |
kil | +0.036 |
type | +0.035 |
ooks | +0.035 |
tim | +0.034 |
Neg logit contributions | |
lor | -0.042 |
� | -0.041 |
actus | -0.041 |
sky | -0.040 |
course | -0.037 |
Token | always |
Token ID | 1464 |
Feature activation | -4.80e-03 |
Pos logit contributions | |
Jill | +0.010 |
type | +0.009 |
like | +0.009 |
kil | +0.009 |
Jen | +0.009 |
Neg logit contributions | |
lor | -0.015 |
actus | -0.014 |
� | -0.014 |
sky | -0.013 |
eros | -0.012 |
Token | asked |
Token ID | 1965 |
Feature activation | +2.66e-02 |
Pos logit contributions | |
kil | +0.008 |
ooks | +0.008 |
Jill | +0.008 |
ierce | +0.007 |
theirs | +0.007 |
Neg logit contributions | |
actus | -0.007 |
� | -0.007 |
sky | -0.006 |
course | -0.006 |
lor | -0.006 |
Token | their |
Token ID | 511 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
sky | +0.038 |
actus | +0.038 |
course | +0.036 |
shop | +0.036 |
skies | +0.036 |
Neg logit contributions | |
Jill | -0.035 |
kil | -0.035 |
cest | -0.035 |
rated | -0.034 |
supposed | -0.034 |
Token | mom |
Token ID | 1995 |
Feature activation | +3.05e-02 |
Pos logit contributions | |
actus | +0.017 |
lor | +0.017 |
� | +0.016 |
sky | +0.016 |
poster | +0.014 |
Neg logit contributions | |
Jill | -0.022 |
kil | -0.020 |
type | -0.019 |
ooks | -0.019 |
theirs | -0.019 |
Token | first |
Token ID | 717 |
Feature activation | +4.93e-02 |
Pos logit contributions | |
lor | +0.057 |
actus | +0.057 |
� | +0.054 |
sky | +0.054 |
course | +0.050 |
Neg logit contributions | |
Jill | -0.038 |
kil | -0.034 |
type | -0.033 |
ooks | -0.032 |
Matt | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
lor | +0.062 |
actus | +0.061 |
� | +0.061 |
sky | +0.061 |
course | +0.053 |
Neg logit contributions | |
Jill | -0.087 |
kil | -0.081 |
ooks | -0.081 |
Matt | -0.078 |
tim | -0.078 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +9.77e-02 |
Pos logit contributions | |
Jill | +0.035 |
Jen | +0.032 |
type | +0.031 |
ooks | +0.031 |
theirs | +0.030 |
Neg logit contributions | |
lor | -0.030 |
actus | -0.030 |
� | -0.028 |
sky | -0.025 |
course | -0.025 |
Token | Anna |
Token ID | 31160 |
Feature activation | +1.39e-03 |
Pos logit contributions | |
lor | +0.138 |
actus | +0.133 |
� | +0.121 |
ican | +0.115 |
chief | +0.111 |
Neg logit contributions | |
Jill | -0.172 |
like | -0.156 |
kil | -0.155 |
tim | -0.153 |
Jen | -0.152 |
Token | liked |
Token ID | 8288 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
lor | +0.002 |
actus | +0.002 |
� | +0.002 |
sky | +0.002 |
course | +0.002 |
Neg logit contributions | |
Jill | -0.002 |
kil | -0.002 |
type | -0.002 |
theirs | -0.002 |
ooks | -0.002 |
Token | to |
Token ID | 284 |
Feature activation | +3.21e-02 |
Pos logit contributions | |
actus | +0.040 |
lor | +0.039 |
� | +0.039 |
sky | +0.036 |
course | +0.033 |
Neg logit contributions | |
Jill | -0.036 |
kil | -0.033 |
type | -0.032 |
ooks | -0.032 |
Jen | -0.031 |
Token | to |
Token ID | 284 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
lor | +0.036 |
sky | +0.036 |
actus | +0.036 |
� | +0.035 |
skies | +0.030 |
Neg logit contributions | |
Jill | -0.067 |
ots | -0.065 |
ooks | -0.064 |
type | -0.063 |
kil | -0.063 |
Token | race |
Token ID | 3234 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
� | +0.026 |
lor | +0.025 |
actus | +0.025 |
sky | +0.023 |
course | +0.023 |
Neg logit contributions | |
Jill | -0.020 |
ooks | -0.018 |
kil | -0.018 |
type | -0.018 |
theirs | -0.017 |
Token | again |
Token ID | 757 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
lor | +0.017 |
� | +0.017 |
actus | +0.017 |
sky | +0.017 |
course | +0.016 |
Neg logit contributions | |
ooks | -0.023 |
type | -0.023 |
Jill | -0.023 |
kil | -0.022 |
ots | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
lor | +0.026 |
� | +0.025 |
sky | +0.024 |
actus | +0.024 |
course | +0.022 |
Neg logit contributions | |
Jill | -0.030 |
ooks | -0.029 |
kil | -0.029 |
type | -0.029 |
ots | -0.028 |
Token | The |
Token ID | 383 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
Jill | +0.022 |
kil | +0.020 |
Jen | +0.020 |
type | +0.019 |
ooks | +0.019 |
Neg logit contributions | |
actus | -0.021 |
lor | -0.021 |
� | -0.019 |
sky | -0.018 |
poster | -0.017 |
Token | moral |
Token ID | 6573 |
Feature activation | +5.37e-02 |
Pos logit contributions | |
lor | +0.034 |
� | +0.028 |
actus | +0.028 |
ane | +0.027 |
salon | +0.024 |
Neg logit contributions | |
type | -0.022 |
Jill | -0.022 |
like | -0.021 |
matter | -0.021 |
sunny | -0.021 |
Token | of |
Token ID | 286 |
Feature activation | -5.77e-02 |
Pos logit contributions | |
� | +0.133 |
actus | +0.132 |
lor | +0.130 |
ican | +0.116 |
poster | +0.114 |
Neg logit contributions | |
Jill | -0.044 |
like | -0.037 |
type | -0.037 |
Jen | -0.036 |
kil | -0.032 |
Token | this |
Token ID | 428 |
Feature activation | +1.05e-03 |
Pos logit contributions | |
Jill | +0.062 |
type | +0.050 |
Jen | +0.049 |
like | +0.048 |
theirs | +0.047 |
Neg logit contributions | |
actus | -0.130 |
lor | -0.130 |
� | -0.128 |
sky | -0.116 |
rot | -0.111 |
Token | story |
Token ID | 1621 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
� | +0.001 |
lor | +0.001 |
actus | +0.001 |
handed | +0.001 |
ican | +0.001 |
Neg logit contributions | |
Jill | -0.002 |
Joe | -0.002 |
type | -0.002 |
like | -0.002 |
remember | -0.002 |
Token | is |
Token ID | 318 |
Feature activation | -2.65e-03 |
Pos logit contributions | |
actus | +0.026 |
� | +0.025 |
lor | +0.025 |
ican | +0.022 |
sky | +0.021 |
Neg logit contributions | |
Jill | -0.028 |
type | -0.026 |
like | -0.025 |
Jen | -0.024 |
theirs | -0.024 |
Token | : |
Token ID | 25 |
Feature activation | +4.72e-03 |
Pos logit contributions | |
Jill | +0.006 |
kil | +0.006 |
type | +0.006 |
theirs | +0.005 |
Jen | +0.005 |
Neg logit contributions | |
lor | -0.003 |
actus | -0.002 |
� | -0.002 |
sky | -0.002 |
course | -0.002 |
Token | But |
Token ID | 1537 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
actus | +0.012 |
lor | +0.012 |
� | +0.012 |
sky | +0.011 |
course | +0.010 |
Neg logit contributions | |
Jill | -0.016 |
kil | -0.015 |
theirs | -0.015 |
type | -0.014 |
Jen | -0.014 |
Token | then |
Token ID | 788 |
Feature activation | +3.78e-02 |
Pos logit contributions | |
Jill | +0.030 |
kil | +0.030 |
type | +0.029 |
ooks | +0.028 |
ots | +0.028 |
Neg logit contributions | |
� | -0.026 |
lor | -0.026 |
actus | -0.026 |
sky | -0.024 |
course | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | -3.80e-02 |
Pos logit contributions | |
� | +0.041 |
lor | +0.039 |
actus | +0.039 |
sky | +0.037 |
course | +0.036 |
Neg logit contributions | |
kil | -0.073 |
Jill | -0.072 |
type | -0.072 |
ots | -0.071 |
ooks | -0.070 |
Token | when |
Token ID | 618 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
kil | +0.048 |
ooks | +0.045 |
type | +0.044 |
ots | +0.043 |
Jill | +0.042 |
Neg logit contributions | |
� | -0.073 |
sky | -0.066 |
course | -0.065 |
actus | -0.065 |
lor | -0.063 |
Token | they |
Token ID | 484 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
kil | +0.024 |
Jill | +0.024 |
type | +0.024 |
ots | +0.023 |
ooks | +0.023 |
Neg logit contributions | |
� | -0.033 |
actus | -0.030 |
sky | -0.030 |
lor | -0.030 |
course | -0.028 |
Token | got |
Token ID | 1392 |
Feature activation | -8.40e-03 |
Pos logit contributions | |
Jill | +0.027 |
type | +0.024 |
kil | +0.023 |
Jen | +0.023 |
like | +0.023 |
Neg logit contributions | |
lor | -0.032 |
actus | -0.031 |
� | -0.030 |
sky | -0.028 |
course | -0.026 |
Token | home |
Token ID | 1363 |
Feature activation | +2.73e-02 |
Pos logit contributions | |
Jill | +0.011 |
Jen | +0.010 |
theirs | +0.010 |
like | +0.009 |
neat | +0.009 |
Neg logit contributions | |
lor | -0.018 |
actus | -0.017 |
� | -0.016 |
pillar | -0.015 |
course | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | -3.85e-02 |
Pos logit contributions | |
� | +0.026 |
actus | +0.026 |
lor | +0.023 |
sky | +0.022 |
car | +0.020 |
Neg logit contributions | |
kil | -0.059 |
type | -0.057 |
sizes | -0.057 |
ots | -0.056 |
Jill | -0.055 |
Token | Tim |
Token ID | 5045 |
Feature activation | -1.97e-02 |
Pos logit contributions | |
kil | +0.045 |
ooks | +0.041 |
type | +0.041 |
ash | +0.039 |
ots | +0.039 |
Neg logit contributions | |
� | -0.078 |
sky | -0.072 |
actus | -0.069 |
course | -0.068 |
lor | -0.067 |
Token | my |
Token ID | 1820 |
Feature activation | +3.71e-03 |
Pos logit contributions | |
Jill | +0.025 |
type | +0.022 |
Jen | +0.021 |
like | +0.021 |
kil | +0.021 |
Neg logit contributions | |
lor | -0.040 |
� | -0.039 |
actus | -0.038 |
sky | -0.034 |
ican | -0.033 |
Token | realized |
Token ID | 6939 |
Feature activation | +9.92e-03 |
Pos logit contributions | |
lor | +0.007 |
� | +0.007 |
actus | +0.007 |
course | +0.006 |
ican | +0.006 |
Neg logit contributions | |
Jill | -0.005 |
type | -0.004 |
plugged | -0.004 |
like | -0.004 |
neat | -0.004 |
Token | that |
Token ID | 326 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
lor | +0.016 |
� | +0.015 |
actus | +0.013 |
sky | +0.012 |
rot | +0.011 |
Neg logit contributions | |
Jill | -0.016 |
like | -0.015 |
why | -0.014 |
remember | -0.014 |
Jen | -0.014 |
Token | day |
Token ID | 1110 |
Feature activation | +7.46e-02 |
Pos logit contributions | |
lor | +0.036 |
actus | +0.035 |
� | +0.034 |
sky | +0.030 |
ican | +0.029 |
Neg logit contributions | |
Jill | -0.034 |
type | -0.032 |
like | -0.030 |
kil | -0.030 |
neat | -0.030 |
Token | on |
Token ID | 319 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
lor | +0.132 |
� | +0.129 |
actus | +0.129 |
sky | +0.118 |
course | +0.111 |
Neg logit contributions | |
Jill | -0.108 |
type | -0.098 |
kil | -0.096 |
theirs | -0.093 |
Jen | -0.092 |
Token | , |
Token ID | 11 |
Feature activation | -2.37e-02 |
Pos logit contributions | |
lor | +0.019 |
actus | +0.018 |
� | +0.017 |
sky | +0.016 |
rot | +0.015 |
Neg logit contributions | |
Jill | -0.018 |
type | -0.015 |
Jen | -0.015 |
neat | -0.015 |
Liz | -0.015 |
Token | she |
Token ID | 673 |
Feature activation | -1.01e-03 |
Pos logit contributions | |
Jill | +0.028 |
kil | +0.026 |
type | +0.025 |
Jen | +0.024 |
ooks | +0.024 |
Neg logit contributions | |
lor | -0.047 |
actus | -0.047 |
� | -0.047 |
sky | -0.043 |
course | -0.041 |
Token | asked |
Token ID | 1965 |
Feature activation | +3.83e-02 |
Pos logit contributions | |
Jill | +0.001 |
kil | +0.001 |
type | +0.001 |
theirs | +0.001 |
Jen | +0.001 |
Neg logit contributions | |
actus | -0.002 |
lor | -0.002 |
� | -0.002 |
sky | -0.002 |
course | -0.002 |
Token | her |
Token ID | 607 |
Feature activation | +6.48e-03 |
Pos logit contributions | |
actus | +0.057 |
sky | +0.054 |
� | +0.054 |
course | +0.053 |
lor | +0.052 |
Neg logit contributions | |
kil | -0.059 |
sizes | -0.054 |
Jill | -0.054 |
Matt | -0.050 |
Jen | -0.050 |
Token | mom |
Token ID | 1995 |
Feature activation | +2.85e-02 |
Pos logit contributions | |
actus | +0.008 |
� | +0.008 |
sky | +0.008 |
lor | +0.007 |
poster | +0.007 |
Neg logit contributions | |
Jill | -0.012 |
kil | -0.011 |
sizes | -0.011 |
ots | -0.011 |
type | -0.011 |
Token | my |
Token ID | 1820 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
actus | +0.040 |
lor | +0.040 |
� | +0.039 |
sky | +0.037 |
course | +0.034 |
Neg logit contributions | |
Jill | -0.049 |
kil | -0.047 |
type | -0.045 |
Jen | -0.044 |
ooks | -0.044 |
Token | to |
Token ID | 284 |
Feature activation | +4.13e-02 |
Pos logit contributions | |
lor | +0.049 |
actus | +0.049 |
� | +0.048 |
sky | +0.045 |
course | +0.043 |
Neg logit contributions | |
Jill | -0.029 |
kil | -0.026 |
type | -0.025 |
Jen | -0.024 |
ooks | -0.023 |
Token | help |
Token ID | 1037 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
� | +0.080 |
lor | +0.074 |
sky | +0.073 |
course | +0.072 |
actus | +0.071 |
Neg logit contributions | |
Jill | -0.050 |
kil | -0.049 |
theirs | -0.047 |
sizes | -0.045 |
tim | -0.045 |
Token | was |
Token ID | 373 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
lor | +0.009 |
actus | +0.008 |
� | +0.008 |
sky | +0.007 |
course | +0.007 |
Neg logit contributions | |
Jill | -0.009 |
type | -0.008 |
kil | -0.008 |
Jen | -0.008 |
theirs | -0.007 |
Token | starting |
Token ID | 3599 |
Feature activation | +9.82e-03 |
Pos logit contributions | |
Jill | +0.026 |
kil | +0.026 |
ierce | +0.024 |
Jen | +0.024 |
ooks | +0.024 |
Neg logit contributions | |
sky | -0.025 |
� | -0.025 |
lor | -0.024 |
actus | -0.023 |
skies | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | +8.02e-03 |
Pos logit contributions | |
lor | +0.016 |
actus | +0.016 |
� | +0.016 |
sky | +0.015 |
skies | +0.013 |
Neg logit contributions | |
Jill | -0.015 |
kil | -0.014 |
type | -0.013 |
Jen | -0.013 |
theirs | -0.013 |
Token | look |
Token ID | 804 |
Feature activation | -5.91e-02 |
Pos logit contributions | |
� | +0.014 |
lor | +0.013 |
sky | +0.013 |
actus | +0.013 |
course | +0.012 |
Neg logit contributions | |
Jill | -0.011 |
kil | -0.011 |
ooks | -0.010 |
Jen | -0.010 |
ierce | -0.010 |
Token | old |
Token ID | 1468 |
Feature activation | +5.13e-02 |
Pos logit contributions | |
Jen | +0.096 |
ierce | +0.093 |
ooks | +0.092 |
Matt | +0.092 |
Liz | +0.088 |
Neg logit contributions | |
sky | -0.084 |
actus | -0.078 |
l | -0.068 |
car | -0.068 |
helicopter | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
actus | +0.068 |
lor | +0.067 |
� | +0.066 |
sky | +0.063 |
course | +0.055 |
Neg logit contributions | |
Jill | -0.094 |
ooks | -0.086 |
kil | -0.086 |
Jen | -0.085 |
type | -0.084 |
Token | She |
Token ID | 1375 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
Jill | +0.027 |
type | +0.025 |
Jen | +0.024 |
kil | +0.024 |
like | +0.023 |
Neg logit contributions | |
lor | -0.024 |
actus | -0.023 |
� | -0.022 |
sky | -0.020 |
course | -0.019 |
Token | started |
Token ID | 2067 |
Feature activation | +8.45e-03 |
Pos logit contributions | |
Jill | +0.019 |
type | +0.017 |
kil | +0.017 |
Jen | +0.016 |
like | +0.016 |
Neg logit contributions | |
lor | -0.026 |
actus | -0.026 |
� | -0.026 |
sky | -0.023 |
course | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | +2.27e-02 |
Pos logit contributions | |
lor | +0.016 |
actus | +0.016 |
� | +0.016 |
sky | +0.014 |
course | +0.013 |
Neg logit contributions | |
Jill | -0.011 |
type | -0.010 |
kil | -0.010 |
Jen | -0.010 |
like | -0.010 |
Token | worry |
Token ID | 5490 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
� | +0.033 |
sky | +0.032 |
actus | +0.032 |
lor | +0.031 |
poster | +0.030 |
Neg logit contributions | |
kil | -0.035 |
Jill | -0.035 |
ooks | -0.035 |
theirs | -0.033 |
akers | -0.032 |
Token | that |
Token ID | 326 |
Feature activation | -7.12e-03 |
Pos logit contributions | |
lor | +0.028 |
� | +0.027 |
sky | +0.026 |
actus | +0.026 |
course | +0.023 |
Neg logit contributions | |
Jill | -0.041 |
kil | -0.040 |
ooks | -0.039 |
type | -0.039 |
tim | -0.038 |
Token | trees |
Token ID | 7150 |
Feature activation | -2.00e-02 |
Pos logit contributions | |
Jill | +0.054 |
type | +0.048 |
kil | +0.048 |
Jen | +0.047 |
like | +0.046 |
Neg logit contributions | |
lor | -0.065 |
actus | -0.063 |
� | -0.062 |
sky | -0.056 |
course | -0.053 |
Token | , |
Token ID | 11 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
Jill | +0.033 |
like | +0.030 |
type | +0.030 |
Jen | +0.029 |
kil | +0.029 |
Neg logit contributions | |
lor | -0.033 |
actus | -0.032 |
� | -0.030 |
sky | -0.026 |
rot | -0.025 |
Token | and |
Token ID | 290 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
Jill | +0.066 |
like | +0.061 |
type | +0.060 |
Jen | +0.059 |
kil | +0.058 |
Neg logit contributions | |
lor | -0.063 |
actus | -0.061 |
� | -0.060 |
rot | -0.050 |
eros | -0.048 |
Token | even |
Token ID | 772 |
Feature activation | -5.12e-02 |
Pos logit contributions | |
lor | +0.018 |
actus | +0.017 |
sky | +0.017 |
� | +0.017 |
course | +0.016 |
Neg logit contributions | |
Jill | -0.016 |
kil | -0.014 |
ooks | -0.014 |
type | -0.014 |
theirs | -0.014 |
Token | her |
Token ID | 607 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
Jill | +0.076 |
ooks | +0.075 |
kil | +0.071 |
ots | +0.069 |
theirs | +0.067 |
Neg logit contributions | |
sky | -0.080 |
actus | -0.072 |
skies | -0.070 |
course | -0.070 |
lor | -0.070 |
Token | luggage |
Token ID | 31488 |
Feature activation | +4.27e-02 |
Pos logit contributions | |
Jill | +0.016 |
kil | +0.015 |
type | +0.015 |
theirs | +0.014 |
ooks | +0.014 |
Neg logit contributions | |
lor | -0.016 |
actus | -0.016 |
� | -0.016 |
sky | -0.015 |
course | -0.014 |
Token | ! |
Token ID | 0 |
Feature activation | -2.08e-02 |
Pos logit contributions | |
lor | +0.074 |
actus | +0.072 |
� | +0.071 |
sky | +0.068 |
course | +0.062 |
Neg logit contributions | |
Jill | -0.062 |
kil | -0.058 |
type | -0.055 |
ooks | -0.055 |
Jen | -0.054 |
Token |
|
Token ID | 198 |
Feature activation | -1.57e-03 |
Pos logit contributions | |
Jill | +0.039 |
type | +0.035 |
Jen | +0.035 |
kil | +0.034 |
like | +0.034 |
Neg logit contributions | |
lor | -0.029 |
actus | -0.029 |
� | -0.027 |
sky | -0.024 |
poster | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | -7.64e-03 |
Pos logit contributions | |
Jill | +0.003 |
type | +0.002 |
Jen | +0.002 |
ooks | +0.002 |
Jane | +0.002 |
Neg logit contributions | |
actus | -0.002 |
lor | -0.002 |
� | -0.002 |
sky | -0.002 |
eros | -0.002 |
Token | One |
Token ID | 3198 |
Feature activation | -4.65e-02 |
Pos logit contributions | |
Jill | +0.013 |
type | +0.012 |
kil | +0.012 |
Jen | +0.011 |
ooks | +0.011 |
Neg logit contributions | |
lor | -0.012 |
actus | -0.012 |
� | -0.011 |
sky | -0.010 |
course | -0.009 |
Token | day |
Token ID | 1110 |
Feature activation | +4.73e-02 |
Pos logit contributions | |
Jill | +0.097 |
kil | +0.093 |
theirs | +0.093 |
ooks | +0.093 |
uggets | +0.088 |
Neg logit contributions | |
sky | -0.047 |
actus | -0.047 |
� | -0.042 |
lor | -0.041 |
course | -0.038 |
Token | Mom |
Token ID | 11254 |
Feature activation | -2.42e-03 |
Pos logit contributions | |
ooks | +0.068 |
Jill | +0.067 |
kil | +0.067 |
type | +0.065 |
ots | +0.063 |
Neg logit contributions | |
actus | -0.100 |
� | -0.099 |
lor | -0.099 |
sky | -0.095 |
course | -0.090 |
Token | my |
Token ID | 1820 |
Feature activation | -8.88e-04 |
Pos logit contributions | |
Jill | +0.004 |
kil | +0.004 |
type | +0.004 |
ooks | +0.004 |
theirs | +0.003 |
Neg logit contributions | |
lor | -0.004 |
actus | -0.004 |
� | -0.004 |
sky | -0.003 |
course | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -2.05e-02 |
Pos logit contributions | |
Jill | +0.001 |
kil | +0.001 |
ooks | +0.001 |
type | +0.001 |
theirs | +0.001 |
Neg logit contributions | |
lor | -0.001 |
actus | -0.001 |
� | -0.001 |
sky | -0.001 |
course | -0.001 |
Token | I |
Token ID | 314 |
Feature activation | -5.02e-03 |
Pos logit contributions | |
Jill | +0.026 |
type | +0.024 |
kil | +0.024 |
theirs | +0.023 |
ooks | +0.022 |
Neg logit contributions | |
lor | -0.039 |
actus | -0.039 |
� | -0.039 |
sky | -0.036 |
course | -0.033 |
Token | didn |
Token ID | 1422 |
Feature activation | +4.85e-02 |
Pos logit contributions | |
Jill | +0.007 |
type | +0.006 |
kil | +0.006 |
like | +0.006 |
Jen | +0.006 |
Neg logit contributions | |
lor | -0.010 |
actus | -0.010 |
� | -0.009 |
sky | -0.009 |
course | -0.008 |
Token | 't |
Token ID | 470 |
Feature activation | -4.53e-03 |
Pos logit contributions | |
actus | +0.066 |
lor | +0.062 |
sky | +0.048 |
� | +0.046 |
rot | +0.046 |
Neg logit contributions | |
Jill | -0.097 |
organize | -0.090 |
Liz | -0.089 |
like | -0.088 |
Jen | -0.088 |
Token | mean |
Token ID | 1612 |
Feature activation | -6.96e-03 |
Pos logit contributions | |
kil | +0.006 |
ooks | +0.006 |
Jamie | +0.006 |
Jill | +0.006 |
ots | +0.006 |
Neg logit contributions | |
actus | -0.007 |
sky | -0.007 |
course | -0.007 |
't | -0.006 |
lor | -0.006 |
Token | to |
Token ID | 284 |
Feature activation | +9.55e-03 |
Pos logit contributions | |
Jill | +0.011 |
type | +0.010 |
kil | +0.010 |
ooks | +0.009 |
Jen | +0.009 |
Neg logit contributions | |
lor | -0.012 |
actus | -0.011 |
� | -0.011 |
sky | -0.010 |
course | -0.010 |
Token | make |
Token ID | 787 |
Feature activation | -2.65e-02 |
Pos logit contributions | |
actus | +0.016 |
� | +0.016 |
lor | +0.015 |
sky | +0.014 |
course | +0.014 |
Neg logit contributions | |
Jill | -0.014 |
kil | -0.014 |
ooks | -0.013 |
theirs | -0.013 |
type | -0.013 |
Token | a |
Token ID | 257 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
Jill | +0.043 |
type | +0.039 |
kil | +0.038 |
Jen | +0.037 |
theirs | +0.037 |
Neg logit contributions | |
lor | -0.044 |
actus | -0.043 |
� | -0.042 |
sky | -0.038 |
course | -0.035 |
Token | mess |
Token ID | 2085 |
Feature activation | -3.18e-02 |
Pos logit contributions | |
lor | +0.017 |
actus | +0.017 |
� | +0.017 |
sky | +0.015 |
course | +0.014 |
Neg logit contributions | |
Jill | -0.017 |
type | -0.015 |
kil | -0.015 |
Jen | -0.015 |
theirs | -0.014 |
Token | is |
Token ID | 318 |
Feature activation | -2.25e-02 |
Pos logit contributions | |
Jill | +0.004 |
type | +0.003 |
kil | +0.003 |
Jen | +0.003 |
like | +0.003 |
Neg logit contributions | |
lor | -0.003 |
actus | -0.003 |
� | -0.002 |
sky | -0.002 |
course | -0.002 |
Token | the |
Token ID | 262 |
Feature activation | +2.85e-02 |
Pos logit contributions | |
ooks | +0.039 |
rated | +0.038 |
kil | +0.037 |
iding | +0.037 |
cest | +0.037 |
Neg logit contributions | |
sky | -0.029 |
actus | -0.022 |
cage | -0.021 |
skies | -0.021 |
� | -0.020 |
Token | best |
Token ID | 1266 |
Feature activation | -4.04e-02 |
Pos logit contributions | |
actus | +0.044 |
lor | +0.043 |
� | +0.043 |
sky | +0.042 |
course | +0.038 |
Neg logit contributions | |
Jill | -0.047 |
ooks | -0.042 |
kil | -0.042 |
Jen | -0.041 |
Liz | -0.041 |
Token | surprise |
Token ID | 5975 |
Feature activation | +6.25e-03 |
Pos logit contributions | |
ierce | +0.069 |
Jill | +0.068 |
kil | +0.068 |
tim | +0.067 |
ooks | +0.066 |
Neg logit contributions | |
actus | -0.055 |
sky | -0.055 |
� | -0.050 |
lor | -0.049 |
course | -0.045 |
Token | ever |
Token ID | 1683 |
Feature activation | -4.71e-02 |
Pos logit contributions | |
lor | +0.010 |
sky | +0.010 |
actus | +0.010 |
� | +0.009 |
course | +0.009 |
Neg logit contributions | |
ooks | -0.009 |
kil | -0.009 |
tim | -0.008 |
ierce | -0.008 |
rated | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -3.02e-02 |
Pos logit contributions | |
Jill | +0.075 |
ooks | +0.074 |
kil | +0.072 |
type | +0.072 |
organize | +0.070 |
Neg logit contributions | |
actus | -0.070 |
lor | -0.069 |
sky | -0.063 |
course | -0.060 |
� | -0.059 |
Token | Mom |
Token ID | 11254 |
Feature activation | +2.36e-03 |
Pos logit contributions | |
type | +0.056 |
Jill | +0.055 |
kil | +0.055 |
theirs | +0.054 |
neat | +0.053 |
Neg logit contributions | |
� | -0.038 |
lor | -0.036 |
actus | -0.035 |
sky | -0.034 |
course | -0.028 |
Token | says |
Token ID | 1139 |
Feature activation | +3.65e-02 |
Pos logit contributions | |
lor | +0.004 |
� | +0.004 |
actus | +0.004 |
sky | +0.003 |
ican | +0.003 |
Neg logit contributions | |
Jill | -0.004 |
like | -0.004 |
Jen | -0.004 |
type | -0.004 |
kil | -0.004 |
Token | she |
Token ID | 673 |
Feature activation | -2.02e-02 |
Pos logit contributions | |
lor | +0.059 |
actus | +0.058 |
� | +0.057 |
sky | +0.053 |
course | +0.049 |
Neg logit contributions | |
Jill | -0.057 |
type | -0.053 |
kil | -0.052 |
ooks | -0.051 |
theirs | -0.050 |
Token | is |
Token ID | 318 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
Jill | +0.032 |
type | +0.029 |
kil | +0.029 |
like | +0.028 |
Jen | +0.028 |
Neg logit contributions | |
lor | -0.033 |
actus | -0.032 |
� | -0.032 |
sky | -0.029 |
course | -0.027 |
Token | glad |
Token ID | 9675 |
Feature activation | +4.75e-02 |
Pos logit contributions | |
rated | +0.020 |
kil | +0.020 |
ooks | +0.020 |
ash | +0.020 |
ots | +0.020 |
Neg logit contributions | |
sky | -0.017 |
� | -0.016 |
lor | -0.015 |
actus | -0.015 |
poster | -0.014 |
Token | more |
Token ID | 517 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
Jill | +0.003 |
ooks | +0.003 |
type | +0.003 |
kil | +0.003 |
Jen | +0.002 |
Neg logit contributions | |
lor | -0.003 |
� | -0.003 |
actus | -0.003 |
sky | -0.003 |
course | -0.003 |
Token | than |
Token ID | 621 |
Feature activation | -7.55e-02 |
Pos logit contributions | |
Jill | +0.040 |
Jen | +0.034 |
theirs | +0.033 |
ooks | +0.033 |
type | +0.033 |
Neg logit contributions | |
lor | -0.052 |
actus | -0.051 |
� | -0.051 |
sky | -0.050 |
course | -0.047 |
Token | anything |
Token ID | 1997 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
Jill | +0.096 |
kil | +0.092 |
ooks | +0.092 |
tim | +0.090 |
type | +0.089 |
Neg logit contributions | |
� | -0.129 |
lor | -0.128 |
actus | -0.126 |
sky | -0.125 |
course | -0.119 |
Token | you |
Token ID | 345 |
Feature activation | -5.24e-02 |
Pos logit contributions | |
Jill | +0.030 |
ooks | +0.029 |
Jen | +0.027 |
tim | +0.027 |
Matt | +0.027 |
Neg logit contributions | |
lor | -0.026 |
actus | -0.026 |
� | -0.026 |
sky | -0.025 |
course | -0.024 |
Token | can |
Token ID | 460 |
Feature activation | +6.53e-02 |
Pos logit contributions | |
Jill | +0.058 |
ooks | +0.055 |
kil | +0.051 |
type | +0.050 |
theirs | +0.050 |
Neg logit contributions | |
� | -0.105 |
actus | -0.105 |
lor | -0.104 |
sky | -0.100 |
course | -0.095 |
Token | find |
Token ID | 1064 |
Feature activation | -3.44e-02 |
Pos logit contributions | |
lor | +0.129 |
� | +0.127 |
actus | +0.126 |
sky | +0.117 |
course | +0.111 |
Neg logit contributions | |
Jill | -0.080 |
kil | -0.071 |
type | -0.070 |
Jen | -0.068 |
ooks | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | -2.08e-02 |
Pos logit contributions | |
ooks | +0.054 |
Jill | +0.054 |
type | +0.053 |
ierce | +0.053 |
tim | +0.052 |
Neg logit contributions | |
lor | -0.049 |
course | -0.047 |
sky | -0.045 |
actus | -0.044 |
� | -0.041 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +9.15e-02 |
Pos logit contributions | |
Jill | +0.039 |
type | +0.035 |
kil | +0.035 |
Jen | +0.035 |
ooks | +0.034 |
Neg logit contributions | |
actus | -0.028 |
lor | -0.028 |
� | -0.027 |
sky | -0.024 |
poster | -0.022 |
Token | Once |
Token ID | 7454 |
Feature activation | +2.82e-02 |
Pos logit contributions | |
lor | +0.116 |
actus | +0.114 |
� | +0.106 |
ican | +0.092 |
poster | +0.091 |
Neg logit contributions | |
Jill | -0.177 |
Jen | -0.159 |
kil | -0.159 |
like | -0.158 |
tim | -0.155 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
lor | +0.060 |
actus | +0.058 |
� | +0.058 |
sky | +0.052 |
course | +0.051 |
Neg logit contributions | |
Jill | -0.034 |
Jen | -0.028 |
type | -0.028 |
kil | -0.027 |
theirs | -0.027 |
Token | a |
Token ID | 257 |
Feature activation | +3.16e-04 |
Pos logit contributions | |
Jill | +0.020 |
like | +0.017 |
type | +0.017 |
Jen | +0.017 |
kil | +0.016 |
Neg logit contributions | |
lor | -0.036 |
actus | -0.036 |
� | -0.035 |
sky | -0.031 |
ican | -0.031 |
Token | my |
Token ID | 1820 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
Jill | +0.009 |
type | +0.008 |
like | +0.008 |
Jen | +0.008 |
kil | +0.008 |
Neg logit contributions | |
lor | -0.012 |
actus | -0.012 |
� | -0.012 |
sky | -0.011 |
skies | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | +4.45e-02 |
Pos logit contributions | |
lor | +0.029 |
actus | +0.028 |
� | +0.027 |
sky | +0.024 |
course | +0.022 |
Neg logit contributions | |
Jill | -0.032 |
type | -0.029 |
kil | -0.029 |
like | -0.029 |
Jen | -0.028 |
Token | said |
Token ID | 531 |
Feature activation | +5.11e-02 |
Pos logit contributions | |
� | +0.060 |
course | +0.058 |
lor | +0.058 |
actus | +0.058 |
sky | +0.057 |
Neg logit contributions | |
kil | -0.076 |
ooks | -0.072 |
tim | -0.069 |
ots | -0.067 |
type | -0.067 |
Token | , |
Token ID | 11 |
Feature activation | -1.78e-02 |
Pos logit contributions | |
lor | +0.062 |
� | +0.060 |
actus | +0.059 |
sky | +0.052 |
course | +0.049 |
Neg logit contributions | |
Jill | -0.098 |
kil | -0.093 |
type | -0.093 |
ooks | -0.090 |
theirs | -0.089 |
Token | " |
Token ID | 366 |
Feature activation | +2.78e-03 |
Pos logit contributions | |
Jill | +0.023 |
type | +0.021 |
kil | +0.021 |
ooks | +0.020 |
theirs | +0.020 |
Neg logit contributions | |
� | -0.034 |
lor | -0.033 |
actus | -0.032 |
sky | -0.031 |
course | -0.028 |
Token | Tim |
Token ID | 14967 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
lor | +0.005 |
actus | +0.005 |
� | +0.005 |
sky | +0.004 |
course | +0.004 |
Neg logit contributions | |
Jill | -0.004 |
type | -0.004 |
kil | -0.004 |
Jen | -0.004 |
theirs | -0.004 |
Token | my |
Token ID | 1820 |
Feature activation | -2.30e-03 |
Pos logit contributions | |
Jill | +0.037 |
type | +0.033 |
kil | +0.033 |
Jen | +0.032 |
theirs | +0.032 |
Neg logit contributions | |
lor | -0.039 |
actus | -0.039 |
� | -0.038 |
sky | -0.035 |
course | -0.032 |
Token | , |
Token ID | 11 |
Feature activation | -5.57e-02 |
Pos logit contributions | |
Jill | +0.005 |
kil | +0.004 |
theirs | +0.004 |
ooks | +0.004 |
type | +0.004 |
Neg logit contributions | |
lor | -0.003 |
� | -0.003 |
actus | -0.003 |
sky | -0.002 |
course | -0.002 |
Token | it |
Token ID | 340 |
Feature activation | -2.53e-04 |
Pos logit contributions | |
Jill | +0.084 |
ooks | +0.080 |
kil | +0.079 |
type | +0.077 |
tim | +0.077 |
Neg logit contributions | |
lor | -0.087 |
� | -0.083 |
actus | -0.083 |
sky | -0.080 |
course | -0.080 |
Token | 's |
Token ID | 338 |
Feature activation | -3.25e-02 |
Pos logit contributions | |
Jill | +0.000 |
type | +0.000 |
kil | +0.000 |
Jen | +0.000 |
theirs | +0.000 |
Neg logit contributions | |
lor | -0.000 |
actus | -0.000 |
� | -0.000 |
sky | -0.000 |
course | -0.000 |
Token | important |
Token ID | 1593 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
Jill | +0.049 |
kil | +0.047 |
ooks | +0.044 |
type | +0.044 |
cest | +0.043 |
Neg logit contributions | |
sky | -0.049 |
� | -0.048 |
lor | -0.046 |
actus | -0.045 |
course | -0.043 |
Token | her |
Token ID | 607 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
Jill | +0.017 |
Jen | +0.013 |
type | +0.013 |
Jane | +0.012 |
neat | +0.012 |
Neg logit contributions | |
actus | -0.044 |
lor | -0.041 |
� | -0.040 |
rot | -0.038 |
pillar | -0.038 |
Token | way |
Token ID | 835 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
Jill | +0.019 |
type | +0.018 |
neat | +0.018 |
organize | +0.018 |
Jen | +0.017 |
Neg logit contributions | |
lor | -0.030 |
actus | -0.029 |
� | -0.027 |
ican | -0.026 |
rot | -0.026 |
Token | there |
Token ID | 612 |
Feature activation | +1.79e-02 |
Pos logit contributions | |
lor | +0.019 |
actus | +0.019 |
� | +0.018 |
sky | +0.016 |
course | +0.015 |
Neg logit contributions | |
Jill | -0.020 |
Jen | -0.018 |
type | -0.018 |
like | -0.017 |
kil | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | -4.37e-02 |
Pos logit contributions | |
lor | +0.025 |
actus | +0.025 |
� | +0.024 |
sky | +0.022 |
course | +0.020 |
Neg logit contributions | |
Jill | -0.032 |
kil | -0.030 |
type | -0.029 |
theirs | -0.029 |
ooks | -0.028 |
Token | she |
Token ID | 673 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
kil | +0.096 |
ooks | +0.091 |
theirs | +0.090 |
ots | +0.090 |
type | +0.090 |
Neg logit contributions | |
� | -0.038 |
sky | -0.037 |
actus | -0.035 |
lor | -0.035 |
course | -0.029 |
Token | saw |
Token ID | 2497 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
Jill | +0.032 |
type | +0.029 |
like | +0.027 |
Jen | +0.027 |
neat | +0.026 |
Neg logit contributions | |
lor | -0.055 |
actus | -0.054 |
� | -0.053 |
sky | -0.048 |
course | -0.047 |
Token | her |
Token ID | 607 |
Feature activation | -2.06e-03 |
Pos logit contributions | |
Jill | +0.020 |
Jen | +0.016 |
type | +0.016 |
like | +0.016 |
neat | +0.015 |
Neg logit contributions | |
lor | -0.042 |
� | -0.041 |
actus | -0.041 |
poster | -0.036 |
eros | -0.035 |
Token | friend |
Token ID | 1545 |
Feature activation | +4.44e-02 |
Pos logit contributions | |
Jill | +0.003 |
type | +0.003 |
like | +0.003 |
Jen | +0.002 |
neat | +0.002 |
Neg logit contributions | |
lor | -0.004 |
actus | -0.004 |
� | -0.004 |
course | -0.003 |
sky | -0.003 |
Token | Sam |
Token ID | 3409 |
Feature activation | +3.37e-03 |
Pos logit contributions | |
� | +0.062 |
actus | +0.060 |
sky | +0.059 |
lor | +0.058 |
course | +0.053 |
Neg logit contributions | |
kil | -0.076 |
ooks | -0.071 |
type | -0.070 |
rated | -0.069 |
sizes | -0.069 |
Token | j |
Token ID | 474 |
Feature activation | -8.24e-03 |
Pos logit contributions | |
lor | +0.007 |
actus | +0.006 |
� | +0.006 |
sky | +0.006 |
rot | +0.005 |
Neg logit contributions | |
Jill | -0.004 |
type | -0.004 |
like | -0.004 |
kil | -0.004 |
theirs | -0.004 |
Token | ogging |
Token ID | 30853 |
Feature activation | +1.40e-02 |
Pos logit contributions | |
Jill | +0.015 |
ash | +0.013 |
ooks | +0.013 |
ook | +0.012 |
Jen | +0.012 |
Neg logit contributions | |
lor | -0.013 |
� | -0.011 |
sky | -0.011 |
actus | -0.011 |
skies | -0.010 |
Token | bird |
Token ID | 6512 |
Feature activation | +6.65e-04 |
Pos logit contributions | |
lor | +0.033 |
� | +0.029 |
ane | +0.027 |
rob | +0.027 |
orted | +0.026 |
Neg logit contributions | |
sunny | -0.022 |
type | -0.020 |
neat | -0.019 |
let | -0.018 |
active | -0.018 |
Token | was |
Token ID | 373 |
Feature activation | -1.67e-03 |
Pos logit contributions | |
lor | +0.001 |
� | +0.001 |
actus | +0.001 |
sky | +0.001 |
poster | +0.001 |
Neg logit contributions | |
Jill | -0.001 |
type | -0.001 |
like | -0.001 |
Jen | -0.001 |
plugged | -0.001 |
Token | happy |
Token ID | 3772 |
Feature activation | +5.66e-02 |
Pos logit contributions | |
Jill | +0.002 |
type | +0.002 |
kil | +0.002 |
Jen | +0.002 |
theirs | +0.002 |
Neg logit contributions | |
lor | -0.003 |
actus | -0.003 |
� | -0.003 |
sky | -0.003 |
course | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | +3.96e-02 |
Pos logit contributions | |
lor | +0.089 |
actus | +0.088 |
� | +0.086 |
sky | +0.078 |
course | +0.072 |
Neg logit contributions | |
Jill | -0.094 |
type | -0.086 |
kil | -0.085 |
Jen | -0.082 |
theirs | -0.082 |
Token | ch |
Token ID | 442 |
Feature activation | -6.82e-02 |
Pos logit contributions | |
actus | +0.060 |
lor | +0.059 |
� | +0.058 |
sky | +0.054 |
course | +0.049 |
Neg logit contributions | |
Jill | -0.065 |
kil | -0.063 |
type | -0.060 |
ooks | -0.059 |
theirs | -0.058 |
Token | ir |
Token ID | 343 |
Feature activation | -4.79e-02 |
Pos logit contributions | |
Jill | +0.129 |
kil | +0.116 |
theirs | +0.116 |
type | +0.116 |
Jen | +0.116 |
Neg logit contributions | |
lor | -0.096 |
actus | -0.091 |
� | -0.091 |
sky | -0.081 |
poster | -0.075 |
Token | ped |
Token ID | 9124 |
Feature activation | +2.89e-02 |
Pos logit contributions | |
Jill | +0.071 |
like | +0.063 |
ots | +0.062 |
Jen | +0.062 |
type | +0.061 |
Neg logit contributions | |
lor | -0.084 |
� | -0.083 |
actus | -0.081 |
poster | -0.072 |
sky | -0.072 |
Token | a |
Token ID | 257 |
Feature activation | -5.93e-03 |
Pos logit contributions | |
lor | +0.046 |
� | +0.046 |
actus | +0.045 |
sky | +0.043 |
course | +0.039 |
Neg logit contributions | |
Jill | -0.044 |
kil | -0.042 |
type | -0.041 |
ooks | -0.040 |
Jen | -0.040 |
Token | sweet |
Token ID | 6029 |
Feature activation | -1.75e-02 |
Pos logit contributions | |
Jill | +0.008 |
type | +0.007 |
kil | +0.007 |
like | +0.007 |
neat | +0.007 |
Neg logit contributions | |
lor | -0.011 |
actus | -0.011 |
� | -0.011 |
sky | -0.010 |
ican | -0.009 |
Token | song |
Token ID | 3496 |
Feature activation | +7.38e-03 |
Pos logit contributions | |
Jill | +0.023 |
type | +0.022 |
like | +0.022 |
kil | +0.020 |
remember | +0.020 |
Neg logit contributions | |
lor | -0.034 |
� | -0.032 |
actus | -0.032 |
sky | -0.028 |
skies | -0.027 |
Token | to |
Token ID | 284 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
lor | +0.013 |
actus | +0.013 |
� | +0.013 |
sky | +0.012 |
course | +0.011 |
Neg logit contributions | |
Jill | -0.010 |
kil | -0.009 |
type | -0.009 |
ooks | -0.009 |
Jen | -0.009 |
Token | deer |
Token ID | 20096 |
Feature activation | -6.89e-03 |
Pos logit contributions | |
lor | +0.027 |
� | +0.025 |
actus | +0.024 |
ane | +0.021 |
rob | +0.020 |
Neg logit contributions | |
type | -0.019 |
Jill | -0.018 |
like | -0.018 |
sunny | -0.018 |
neat | -0.017 |
Token | would |
Token ID | 561 |
Feature activation | +5.48e-02 |
Pos logit contributions | |
Jill | +0.008 |
like | +0.008 |
type | +0.008 |
kil | +0.007 |
remember | +0.007 |
Neg logit contributions | |
lor | -0.013 |
� | -0.013 |
actus | -0.013 |
sky | -0.012 |
ican | -0.011 |
Token | peek |
Token ID | 27185 |
Feature activation | -7.51e-02 |
Pos logit contributions | |
lor | +0.110 |
actus | +0.107 |
� | +0.107 |
sky | +0.099 |
course | +0.095 |
Neg logit contributions | |
Jill | -0.067 |
kil | -0.058 |
type | -0.057 |
ooks | -0.056 |
Jen | -0.055 |
Token | out |
Token ID | 503 |
Feature activation | -7.67e-02 |
Pos logit contributions | |
Jill | +0.102 |
type | +0.099 |
kil | +0.095 |
Jen | +0.093 |
ooks | +0.092 |
Neg logit contributions | |
lor | -0.133 |
actus | -0.130 |
sky | -0.125 |
� | -0.118 |
skies | -0.114 |
Token | from |
Token ID | 422 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
type | +0.109 |
Jill | +0.107 |
akers | +0.106 |
kil | +0.106 |
ooks | +0.105 |
Neg logit contributions | |
sky | -0.124 |
lor | -0.112 |
skies | -0.109 |
actus | -0.109 |
� | -0.109 |
Token | between |
Token ID | 1022 |
Feature activation | -8.25e-02 |
Pos logit contributions | |
kil | +0.017 |
ooks | +0.017 |
Jill | +0.017 |
type | +0.016 |
ots | +0.016 |
Neg logit contributions | |
sky | -0.024 |
actus | -0.023 |
� | -0.023 |
course | -0.022 |
lor | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | +2.99e-02 |
Pos logit contributions | |
Jill | +0.102 |
type | +0.094 |
kil | +0.093 |
ooks | +0.089 |
Jen | +0.087 |
Neg logit contributions | |
lor | -0.157 |
� | -0.154 |
actus | -0.153 |
sky | -0.145 |
course | -0.138 |
Token | branches |
Token ID | 13737 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
lor | +0.043 |
actus | +0.041 |
� | +0.041 |
sky | +0.036 |
course | +0.033 |
Neg logit contributions | |
Jill | -0.054 |
type | -0.050 |
kil | -0.050 |
Jen | -0.048 |
like | -0.048 |
Token | and |
Token ID | 290 |
Feature activation | +7.34e-03 |
Pos logit contributions | |
Jill | +0.044 |
type | +0.039 |
like | +0.039 |
kil | +0.039 |
Jen | +0.037 |
Neg logit contributions | |
lor | -0.055 |
actus | -0.055 |
� | -0.052 |
sky | -0.046 |
eros | -0.045 |
Token | watch |
Token ID | 2342 |
Feature activation | -4.82e-02 |
Pos logit contributions | |
sky | +0.012 |
lor | +0.012 |
actus | +0.012 |
� | +0.011 |
course | +0.011 |
Neg logit contributions | |
ooks | -0.010 |
Jill | -0.010 |
ots | -0.010 |
kil | -0.010 |
theirs | -0.009 |
Token | as |
Token ID | 355 |
Feature activation | -1.14e-02 |
Pos logit contributions | |
ooks | +0.067 |
Jill | +0.066 |
kil | +0.065 |
type | +0.063 |
akers | +0.062 |
Neg logit contributions | |
sky | -0.078 |
lor | -0.075 |
actus | -0.072 |
� | -0.072 |
skies | -0.069 |
Token | tree |
Token ID | 5509 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
lor | +0.012 |
actus | +0.011 |
� | +0.011 |
sky | +0.008 |
course | +0.007 |
Neg logit contributions | |
Jill | -0.033 |
type | -0.031 |
kil | -0.030 |
Jen | -0.030 |
like | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
lor | +0.053 |
actus | +0.052 |
� | +0.051 |
sky | +0.046 |
course | +0.044 |
Neg logit contributions | |
Jill | -0.034 |
type | -0.030 |
kil | -0.029 |
like | -0.029 |
Jen | -0.028 |
Token | |
Token ID | 220 |
Feature activation | +5.65e-03 |
Pos logit contributions | |
Jill | +0.020 |
Jen | +0.017 |
Jane | +0.016 |
Liz | +0.016 |
Mia | +0.016 |
Neg logit contributions | |
lor | -0.021 |
actus | -0.020 |
� | -0.019 |
rot | -0.018 |
course | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | +8.09e-03 |
Pos logit contributions | |
lor | +0.007 |
actus | +0.007 |
� | +0.007 |
sky | +0.006 |
course | +0.006 |
Neg logit contributions | |
Jill | -0.011 |
Jen | -0.010 |
type | -0.010 |
kil | -0.010 |
ooks | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | +1.02e-03 |
Pos logit contributions | |
actus | +0.013 |
lor | +0.013 |
� | +0.012 |
ican | +0.011 |
eros | +0.011 |
Neg logit contributions | |
Jill | -0.014 |
Jen | -0.012 |
like | -0.012 |
Jane | -0.012 |
Mia | -0.012 |
Token | L |
Token ID | 43 |
Feature activation | -8.57e-02 |
Pos logit contributions | |
lor | +0.002 |
actus | +0.002 |
� | +0.002 |
sky | +0.002 |
eros | +0.002 |
Neg logit contributions | |
Jill | -0.001 |
type | -0.001 |
neat | -0.001 |
like | -0.001 |
organize | -0.001 |
Token | ily |
Token ID | 813 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
Jill | +0.082 |
like | +0.062 |
Jen | +0.060 |
theirs | +0.059 |
type | +0.058 |
Neg logit contributions | |
� | -0.205 |
lor | -0.203 |
actus | -0.197 |
sky | -0.189 |
skies | -0.180 |
Token | finished |
Token ID | 5201 |
Feature activation | +3.50e-02 |
Pos logit contributions | |
Jill | +0.015 |
type | +0.013 |
plugged | +0.013 |
like | +0.013 |
Joe | +0.013 |
Neg logit contributions | |
lor | -0.029 |
� | -0.029 |
actus | -0.028 |
ane | -0.025 |
sky | -0.024 |
Token | cleaning |
Token ID | 12724 |
Feature activation | -4.21e-02 |
Pos logit contributions | |
� | +0.069 |
lor | +0.068 |
actus | +0.067 |
sky | +0.062 |
course | +0.059 |
Neg logit contributions | |
Jill | -0.043 |
kil | -0.039 |
type | -0.039 |
ooks | -0.037 |
theirs | -0.036 |
Token | the |
Token ID | 262 |
Feature activation | +3.72e-02 |
Pos logit contributions | |
Jill | +0.059 |
kil | +0.056 |
type | +0.055 |
ooks | +0.052 |
theirs | +0.052 |
Neg logit contributions | |
actus | -0.073 |
lor | -0.073 |
� | -0.072 |
sky | -0.068 |
course | -0.064 |
Token | patch |
Token ID | 8529 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
lor | +0.070 |
� | +0.064 |
ane | +0.056 |
rot | +0.054 |
actus | +0.054 |
Neg logit contributions | |
like | -0.051 |
neat | -0.051 |
type | -0.049 |
sunny | -0.048 |
Jill | -0.048 |
Token | on |
Token ID | 319 |
Feature activation | -2.86e-02 |
Pos logit contributions | |
lor | +0.021 |
actus | +0.021 |
� | +0.021 |
sky | +0.019 |
course | +0.018 |
Neg logit contributions | |
Jill | -0.013 |
type | -0.012 |
kil | -0.012 |
like | -0.012 |
Jen | -0.011 |
Token | it |
Token ID | 340 |
Feature activation | -6.89e-03 |
Pos logit contributions | |
Jill | +0.042 |
type | +0.038 |
kil | +0.037 |
Jen | +0.036 |
theirs | +0.036 |
Neg logit contributions | |
lor | -0.051 |
actus | -0.050 |
� | -0.049 |
sky | -0.044 |
course | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
Jill | +0.010 |
kil | +0.009 |
type | +0.009 |
ooks | +0.008 |
Jen | +0.008 |
Neg logit contributions | |
lor | -0.013 |
actus | -0.012 |
� | -0.012 |
sky | -0.011 |
course | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | +8.72e-03 |
Pos logit contributions | |
Jill | +0.028 |
Jen | +0.023 |
Jane | +0.023 |
Mia | +0.023 |
Liz | +0.023 |
Neg logit contributions | |
lor | -0.032 |
actus | -0.029 |
� | -0.028 |
rot | -0.027 |
eros | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | +2.10e-03 |
Pos logit contributions | |
lor | +0.014 |
� | +0.013 |
actus | +0.013 |
ican | +0.012 |
sky | +0.011 |
Neg logit contributions | |
Jill | -0.015 |
Jen | -0.013 |
type | -0.013 |
like | -0.013 |
Jane | -0.013 |
Token | L |
Token ID | 43 |
Feature activation | -8.51e-02 |
Pos logit contributions | |
lor | +0.003 |
actus | +0.003 |
� | +0.003 |
sky | +0.003 |
course | +0.003 |
Neg logit contributions | |
Jill | -0.003 |
kil | -0.003 |
type | -0.003 |
theirs | -0.003 |
Jen | -0.003 |
Token | ily |
Token ID | 813 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
Jill | +0.090 |
Jen | +0.070 |
type | +0.069 |
theirs | +0.069 |
like | +0.069 |
Neg logit contributions | |
lor | -0.193 |
� | -0.192 |
actus | -0.188 |
sky | -0.177 |
skies | -0.168 |
Token | loved |
Token ID | 6151 |
Feature activation | +3.79e-02 |
Pos logit contributions | |
Jill | +0.013 |
type | +0.012 |
plugged | +0.012 |
like | +0.012 |
organize | +0.011 |
Neg logit contributions | |
lor | -0.026 |
� | -0.026 |
actus | -0.024 |
ican | -0.022 |
eros | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | +3.82e-02 |
Pos logit contributions | |
lor | +0.088 |
� | +0.085 |
actus | +0.083 |
sky | +0.074 |
rot | +0.073 |
Neg logit contributions | |
Jill | -0.042 |
type | -0.035 |
Jen | -0.035 |
theirs | -0.034 |
like | -0.033 |
Token | necklace |
Token ID | 38987 |
Feature activation | +4.32e-02 |
Pos logit contributions | |
� | +0.092 |
lor | +0.083 |
ican | +0.082 |
eros | +0.078 |
itt | +0.077 |
Neg logit contributions | |
like | -0.046 |
neat | -0.045 |
Jill | -0.044 |
color | -0.044 |
type | -0.043 |
Token | so |
Token ID | 523 |
Feature activation | +2.89e-03 |
Pos logit contributions | |
lor | +0.083 |
actus | +0.081 |
� | +0.080 |
sky | +0.071 |
ican | +0.069 |
Neg logit contributions | |
Jill | -0.059 |
like | -0.053 |
type | -0.053 |
neat | -0.050 |
Jen | -0.049 |
Token | rest |
Token ID | 1334 |
Feature activation | -4.13e-02 |
Pos logit contributions | |
lor | +0.039 |
� | +0.038 |
actus | +0.037 |
sky | +0.031 |
ican | +0.030 |
Neg logit contributions | |
Jill | -0.046 |
type | -0.043 |
kil | -0.042 |
like | -0.042 |
neat | -0.041 |
Token | later |
Token ID | 1568 |
Feature activation | +3.10e-02 |
Pos logit contributions | |
Jill | +0.057 |
kil | +0.055 |
ooks | +0.053 |
type | +0.052 |
whether | +0.051 |
Neg logit contributions | |
actus | -0.071 |
lor | -0.070 |
sky | -0.066 |
� | -0.065 |
ane | -0.060 |
Token | ?" |
Token ID | 1701 |
Feature activation | +2.99e-02 |
Pos logit contributions | |
lor | +0.053 |
� | +0.052 |
actus | +0.052 |
sky | +0.048 |
course | +0.044 |
Neg logit contributions | |
Jill | -0.046 |
kil | -0.043 |
type | -0.042 |
ooks | -0.041 |
theirs | -0.040 |
Token |
|
Token ID | 198 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
lor | +0.043 |
actus | +0.043 |
� | +0.041 |
sky | +0.036 |
ican | +0.035 |
Neg logit contributions | |
Jill | -0.057 |
Jen | -0.050 |
like | -0.048 |
Jane | -0.048 |
kil | -0.047 |
Token |
|
Token ID | 198 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
Jill | +0.018 |
Jen | +0.016 |
ooks | +0.015 |
Mia | +0.015 |
Liz | +0.015 |
Neg logit contributions | |
lor | -0.016 |
� | -0.016 |
actus | -0.015 |
ican | -0.014 |
sky | -0.013 |
Token | L |
Token ID | 43 |
Feature activation | -8.31e-02 |
Pos logit contributions | |
Jill | +0.019 |
type | +0.017 |
kil | +0.017 |
Jen | +0.016 |
ooks | +0.016 |
Neg logit contributions | |
� | -0.027 |
actus | -0.027 |
lor | -0.027 |
sky | -0.025 |
ican | -0.023 |
Token | ily |
Token ID | 813 |
Feature activation | -7.63e-03 |
Pos logit contributions | |
Jill | +0.091 |
Jen | +0.073 |
kil | +0.071 |
theirs | +0.071 |
type | +0.070 |
Neg logit contributions | |
� | -0.180 |
actus | -0.178 |
lor | -0.178 |
sky | -0.168 |
poster | -0.158 |
Token | agreed |
Token ID | 4987 |
Feature activation | +3.90e-02 |
Pos logit contributions | |
Jill | +0.011 |
type | +0.009 |
plugged | +0.009 |
like | +0.009 |
Jane | +0.009 |
Neg logit contributions | |
lor | -0.015 |
� | -0.014 |
actus | -0.014 |
sky | -0.012 |
eros | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
actus | +0.073 |
lor | +0.073 |
� | +0.069 |
ican | +0.062 |
sky | +0.062 |
Neg logit contributions | |
Jill | -0.058 |
like | -0.052 |
type | -0.049 |
neat | -0.049 |
Jen | -0.048 |
Token | She |
Token ID | 1375 |
Feature activation | +3.41e-03 |
Pos logit contributions | |
Jill | +0.032 |
type | +0.029 |
kil | +0.029 |
Jen | +0.029 |
theirs | +0.028 |
Neg logit contributions | |
lor | -0.027 |
actus | -0.027 |
� | -0.026 |
sky | -0.024 |
course | -0.022 |
Token | was |
Token ID | 373 |
Feature activation | -3.16e-03 |
Pos logit contributions | |
lor | +0.006 |
actus | +0.006 |
� | +0.005 |
sky | +0.005 |
course | +0.005 |
Neg logit contributions | |
Jill | -0.005 |
kil | -0.005 |
type | -0.005 |
ooks | -0.005 |
theirs | -0.005 |
Token | deep |
Token ID | 2769 |
Feature activation | -3.69e-02 |
Pos logit contributions | |
lor | +0.027 |
� | +0.025 |
actus | +0.025 |
sky | +0.023 |
skies | +0.022 |
Neg logit contributions | |
Jill | -0.019 |
type | -0.018 |
neat | -0.018 |
kil | -0.017 |
like | -0.017 |
Token | breath |
Token ID | 8033 |
Feature activation | +4.38e-03 |
Pos logit contributions | |
Jill | +0.061 |
type | +0.056 |
kil | +0.054 |
like | +0.054 |
Jen | +0.054 |
Neg logit contributions | |
lor | -0.060 |
actus | -0.058 |
� | -0.057 |
sky | -0.050 |
rot | -0.047 |
Token | and |
Token ID | 290 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
lor | +0.007 |
actus | +0.007 |
� | +0.007 |
sky | +0.006 |
poster | +0.006 |
Neg logit contributions | |
Jill | -0.007 |
like | -0.007 |
type | -0.006 |
kil | -0.006 |
Jen | -0.006 |
Token | shouted |
Token ID | 17293 |
Feature activation | -3.45e-03 |
Pos logit contributions | |
lor | +0.026 |
actus | +0.026 |
� | +0.025 |
sky | +0.025 |
course | +0.022 |
Neg logit contributions | |
Jill | -0.024 |
kil | -0.023 |
type | -0.023 |
ooks | -0.023 |
theirs | -0.022 |
Token | " |
Token ID | 366 |
Feature activation | +8.81e-03 |
Pos logit contributions | |
matter | +0.007 |
sizes | +0.006 |
type | +0.006 |
sorts | +0.006 |
rated | +0.006 |
Neg logit contributions | |
lor | -0.004 |
� | -0.004 |
sky | -0.004 |
actus | -0.003 |
rot | -0.003 |
Token | ST |
Token ID | 2257 |
Feature activation | -8.88e-02 |
Pos logit contributions | |
lor | +0.015 |
actus | +0.014 |
� | +0.014 |
sky | +0.013 |
course | +0.012 |
Neg logit contributions | |
Jill | -0.014 |
type | -0.012 |
kil | -0.012 |
Jen | -0.012 |
like | -0.012 |
Token | OP |
Token ID | 3185 |
Feature activation | -4.15e-02 |
Pos logit contributions | |
Jill | +0.126 |
type | +0.116 |
kil | +0.111 |
neat | +0.110 |
theirs | +0.109 |
Neg logit contributions | |
lor | -0.162 |
� | -0.157 |
actus | -0.154 |
sky | -0.139 |
course | -0.134 |
Token | !" |
Token ID | 2474 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
Jill | +0.053 |
type | +0.047 |
theirs | +0.047 |
Jen | +0.045 |
ooks | +0.044 |
Neg logit contributions | |
lor | -0.083 |
actus | -0.081 |
� | -0.080 |
sky | -0.073 |
course | -0.069 |
Token | |
Token ID | 220 |
Feature activation | +2.10e-03 |
Pos logit contributions | |
lor | +0.034 |
actus | +0.034 |
� | +0.032 |
sky | +0.029 |
course | +0.027 |
Neg logit contributions | |
Jill | -0.039 |
type | -0.034 |
Jen | -0.034 |
kil | -0.034 |
like | -0.034 |
Token |
|
Token ID | 198 |
Feature activation | -3.80e-03 |
Pos logit contributions | |
actus | +0.003 |
lor | +0.003 |
� | +0.002 |
sky | +0.002 |
course | +0.002 |
Neg logit contributions | |
Jill | -0.004 |
kil | -0.004 |
type | -0.004 |
Jen | -0.004 |
ooks | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -8.77e-03 |
Pos logit contributions | |
Jill | +0.007 |
Jen | +0.006 |
type | +0.006 |
kil | +0.006 |
ooks | +0.006 |
Neg logit contributions | |
actus | -0.006 |
lor | -0.005 |
� | -0.005 |
sky | -0.005 |
poster | -0.004 |