Token | is |
Token ID | 318 |
Feature activation | +3.49e-02 |
Pos logit contributions | |
ine | +0.022 |
birthday | +0.021 |
hip | +0.020 |
Office | +0.019 |
Jupiter | +0.019 |
Neg logit contributions | |
ls | -0.025 |
Run | -0.025 |
mun | -0.024 |
P | -0.022 |
Hey | -0.022 |
Token | not |
Token ID | 407 |
Feature activation | +3.41e-02 |
Pos logit contributions | |
hip | +0.061 |
ine | +0.061 |
Office | +0.055 |
rap | +0.055 |
birthday | +0.054 |
Neg logit contributions | |
ls | -0.053 |
Run | -0.051 |
bitten | -0.047 |
P | -0.047 |
mun | -0.045 |
Token | the |
Token ID | 262 |
Feature activation | +2.53e-03 |
Pos logit contributions | |
ine | +0.057 |
hip | +0.053 |
Office | +0.051 |
birthday | +0.051 |
rap | +0.051 |
Neg logit contributions | |
ls | -0.054 |
Run | -0.052 |
bitten | -0.049 |
mun | -0.047 |
P | -0.047 |
Token | only |
Token ID | 691 |
Feature activation | +2.19e-02 |
Pos logit contributions | |
ine | +0.004 |
birthday | +0.003 |
hip | +0.003 |
Jupiter | +0.003 |
Office | +0.003 |
Neg logit contributions | |
Run | -0.005 |
ls | -0.005 |
mun | -0.004 |
P | -0.004 |
Hey | -0.004 |
Token | red |
Token ID | 2266 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
ine | +0.031 |
hip | +0.029 |
rap | +0.028 |
birthday | +0.028 |
Jupiter | +0.028 |
Neg logit contributions | |
ls | -0.040 |
Run | -0.039 |
mun | -0.037 |
bitten | -0.035 |
Hey | -0.034 |
Token | cr |
Token ID | 1067 |
Feature activation | +5.92e-02 |
Pos logit contributions | |
ine | +0.030 |
Jupiter | +0.029 |
Office | +0.028 |
birthday | +0.028 |
hip | +0.027 |
Neg logit contributions | |
ls | -0.033 |
Run | -0.032 |
mun | -0.030 |
Hey | -0.028 |
bitten | -0.027 |
Token | ay |
Token ID | 323 |
Feature activation | +2.54e-02 |
Pos logit contributions | |
birthday | +0.074 |
Office | +0.074 |
sing | +0.071 |
experiment | +0.069 |
Jupiter | +0.068 |
Neg logit contributions | |
ls | -0.111 |
P | -0.100 |
awed | -0.092 |
led | -0.091 |
Hey | -0.090 |
Token | on |
Token ID | 261 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
ine | +0.054 |
birthday | +0.054 |
sing | +0.052 |
Office | +0.052 |
Jupiter | +0.051 |
Neg logit contributions | |
ls | -0.029 |
Run | -0.024 |
mun | -0.023 |
led | -0.021 |
P | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
ine | +0.029 |
hip | +0.028 |
Jupiter | +0.027 |
Office | +0.027 |
birthday | +0.027 |
Neg logit contributions | |
ls | -0.020 |
Run | -0.015 |
led | -0.015 |
mun | -0.015 |
bitten | -0.014 |
Token | There |
Token ID | 1318 |
Feature activation | +9.96e-03 |
Pos logit contributions | |
ine | +0.036 |
birthday | +0.031 |
rap | +0.031 |
desk | +0.030 |
end | +0.030 |
Neg logit contributions | |
Run | -0.039 |
ls | -0.034 |
Hey | -0.034 |
P | -0.033 |
Sorry | -0.032 |
Token | are |
Token ID | 389 |
Feature activation | +2.21e-02 |
Pos logit contributions | |
ine | +0.017 |
birthday | +0.016 |
hip | +0.015 |
Office | +0.015 |
rap | +0.015 |
Neg logit contributions | |
ls | -0.017 |
Run | -0.016 |
mun | -0.016 |
Hey | -0.014 |
P | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | +1.74e-02 |
Pos logit contributions | |
hip | +0.035 |
ine | +0.035 |
Office | +0.034 |
birthday | +0.033 |
uries | +0.033 |
Neg logit contributions | |
ls | -0.035 |
led | -0.026 |
bitten | -0.025 |
Run | -0.024 |
mun | -0.024 |
Token | scissors |
Token ID | 41786 |
Feature activation | +2.47e-02 |
Pos logit contributions | |
ine | +0.027 |
hip | +0.027 |
Office | +0.026 |
rap | +0.026 |
birthday | +0.025 |
Neg logit contributions | |
ls | -0.030 |
Run | -0.029 |
mun | -0.027 |
bitten | -0.025 |
P | -0.025 |
Token | , |
Token ID | 11 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
birthday | +0.044 |
Office | +0.043 |
hip | +0.041 |
Jupiter | +0.041 |
ine | +0.040 |
Neg logit contributions | |
ls | -0.039 |
bitten | -0.032 |
led | -0.031 |
mun | -0.031 |
Run | -0.031 |
Token | glue |
Token ID | 22749 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
ine | +0.036 |
hip | +0.035 |
Office | +0.035 |
birthday | +0.034 |
rap | +0.033 |
Neg logit contributions | |
ls | -0.038 |
Run | -0.036 |
mun | -0.033 |
P | -0.033 |
bitten | -0.032 |
Token | and |
Token ID | 290 |
Feature activation | +3.02e-02 |
Pos logit contributions | |
birthday | +0.045 |
hip | +0.044 |
Office | +0.044 |
ine | +0.043 |
Jupiter | +0.043 |
Neg logit contributions | |
ls | -0.045 |
mun | -0.036 |
Run | -0.035 |
led | -0.035 |
bitten | -0.035 |
Token | cr |
Token ID | 1067 |
Feature activation | +5.84e-02 |
Pos logit contributions | |
ine | +0.049 |
hip | +0.046 |
birthday | +0.045 |
Office | +0.045 |
rap | +0.044 |
Neg logit contributions | |
ls | -0.051 |
Run | -0.050 |
mun | -0.047 |
P | -0.045 |
bitten | -0.045 |
Token | ay |
Token ID | 323 |
Feature activation | +2.83e-02 |
Pos logit contributions | |
Office | +0.077 |
birthday | +0.074 |
experiment | +0.072 |
sing | +0.072 |
reminder | +0.071 |
Neg logit contributions | |
ls | -0.117 |
P | -0.100 |
led | -0.099 |
awed | -0.097 |
bitten | -0.095 |
Token | ons |
Token ID | 684 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
birthday | +0.073 |
ine | +0.071 |
Office | +0.071 |
hip | +0.070 |
Jupiter | +0.069 |
Neg logit contributions | |
ls | -0.022 |
mun | -0.016 |
Run | -0.016 |
P | -0.015 |
led | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
ine | +0.041 |
birthday | +0.040 |
Office | +0.040 |
Jupiter | +0.039 |
hip | +0.039 |
Neg logit contributions | |
ls | -0.029 |
Run | -0.023 |
mun | -0.022 |
led | -0.022 |
bitten | -0.022 |
Token | They |
Token ID | 1119 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
ine | +0.025 |
birthday | +0.022 |
hip | +0.022 |
urger | +0.022 |
Office | +0.022 |
Neg logit contributions | |
Run | -0.020 |
ls | -0.019 |
mun | -0.018 |
Hey | -0.017 |
P | -0.017 |
Token | make |
Token ID | 787 |
Feature activation | +3.33e-02 |
Pos logit contributions | |
ine | +0.034 |
birthday | +0.032 |
hip | +0.032 |
Office | +0.031 |
rap | +0.030 |
Neg logit contributions | |
ls | -0.045 |
Run | -0.043 |
mun | -0.043 |
distracted | -0.040 |
P | -0.040 |
Token | to |
Token ID | 284 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
ls | +0.001 |
Run | +0.001 |
mun | +0.001 |
P | +0.001 |
bitten | +0.001 |
Neg logit contributions | |
ine | -0.001 |
birthday | -0.001 |
hip | -0.001 |
Office | -0.001 |
X | -0.001 |
Token | draw |
Token ID | 3197 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
ine | +0.014 |
Office | +0.014 |
birthday | +0.013 |
Jupiter | +0.013 |
hip | +0.013 |
Neg logit contributions | |
Run | -0.019 |
ls | -0.018 |
mun | -0.018 |
P | -0.017 |
bitten | -0.017 |
Token | pictures |
Token ID | 5986 |
Feature activation | +3.83e-02 |
Pos logit contributions | |
ine | +0.038 |
sing | +0.037 |
hip | +0.036 |
Office | +0.036 |
uries | +0.035 |
Neg logit contributions | |
ls | -0.032 |
mun | -0.026 |
led | -0.025 |
P | -0.024 |
Run | -0.024 |
Token | with |
Token ID | 351 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
ine | +0.080 |
sing | +0.077 |
Office | +0.077 |
urger | +0.076 |
artments | +0.074 |
Neg logit contributions | |
ls | -0.049 |
mun | -0.039 |
Run | -0.035 |
P | -0.033 |
led | -0.032 |
Token | her |
Token ID | 607 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
ine | +0.034 |
Office | +0.033 |
hip | +0.033 |
birthday | +0.032 |
sing | +0.031 |
Neg logit contributions | |
ls | -0.033 |
mun | -0.029 |
Run | -0.028 |
P | -0.027 |
bitten | -0.026 |
Token | cr |
Token ID | 1067 |
Feature activation | +5.64e-02 |
Pos logit contributions | |
ine | +0.034 |
hip | +0.033 |
Office | +0.031 |
rap | +0.031 |
birthday | +0.030 |
Neg logit contributions | |
ls | -0.038 |
mun | -0.034 |
Run | -0.033 |
P | -0.031 |
bitten | -0.030 |
Token | ay |
Token ID | 323 |
Feature activation | +2.87e-02 |
Pos logit contributions | |
Office | +0.073 |
birthday | +0.067 |
sing | +0.065 |
experiment | +0.065 |
believe | +0.064 |
Neg logit contributions | |
ls | -0.110 |
P | -0.099 |
awed | -0.095 |
led | -0.091 |
bitten | -0.088 |
Token | ons |
Token ID | 684 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
birthday | +0.076 |
Office | +0.074 |
ine | +0.074 |
sing | +0.073 |
hip | +0.072 |
Neg logit contributions | |
ls | -0.019 |
mun | -0.012 |
P | -0.011 |
led | -0.010 |
Run | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
ine | +0.023 |
Office | +0.022 |
birthday | +0.021 |
X | +0.021 |
Jupiter | +0.021 |
Neg logit contributions | |
ls | -0.018 |
mun | -0.015 |
Run | -0.014 |
P | -0.014 |
led | -0.014 |
Token | but |
Token ID | 475 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
ine | +0.024 |
Office | +0.023 |
hip | +0.022 |
rap | +0.022 |
birthday | +0.021 |
Neg logit contributions | |
ls | -0.025 |
Run | -0.025 |
mun | -0.025 |
Hey | -0.021 |
P | -0.021 |
Token | she |
Token ID | 673 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
ine | +0.038 |
hip | +0.037 |
Office | +0.036 |
birthday | +0.036 |
rap | +0.035 |
Neg logit contributions | |
ls | -0.021 |
Run | -0.020 |
mun | -0.020 |
distracted | -0.017 |
P | -0.016 |
Token | draw |
Token ID | 3197 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
ine | +0.022 |
Office | +0.022 |
uries | +0.021 |
rap | +0.021 |
hip | +0.021 |
Neg logit contributions | |
Run | -0.021 |
ls | -0.020 |
P | -0.020 |
mun | -0.020 |
bitten | -0.019 |
Token | and |
Token ID | 290 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
ine | +0.047 |
sing | +0.046 |
Office | +0.044 |
uries | +0.044 |
hip | +0.044 |
Neg logit contributions | |
ls | -0.039 |
mun | -0.031 |
P | -0.031 |
led | -0.030 |
distracted | -0.027 |
Token | color |
Token ID | 3124 |
Feature activation | +3.03e-02 |
Pos logit contributions | |
ine | +0.030 |
Office | +0.027 |
hip | +0.026 |
urger | +0.025 |
uries | +0.025 |
Neg logit contributions | |
mun | -0.044 |
ls | -0.044 |
Run | -0.042 |
av | -0.040 |
P | -0.039 |
Token | with |
Token ID | 351 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
sing | +0.068 |
uries | +0.065 |
Office | +0.065 |
ine | +0.064 |
hip | +0.063 |
Neg logit contributions | |
ls | -0.042 |
mun | -0.028 |
ries | -0.027 |
av | -0.026 |
P | -0.025 |
Token | her |
Token ID | 607 |
Feature activation | +1.74e-02 |
Pos logit contributions | |
Office | +0.044 |
ine | +0.042 |
sing | +0.042 |
hip | +0.042 |
uries | +0.041 |
Neg logit contributions | |
ls | -0.027 |
P | -0.023 |
mun | -0.022 |
av | -0.021 |
bitten | -0.021 |
Token | cr |
Token ID | 1067 |
Feature activation | +5.62e-02 |
Pos logit contributions | |
ine | +0.030 |
hip | +0.029 |
Office | +0.028 |
rap | +0.027 |
uries | +0.027 |
Neg logit contributions | |
ls | -0.029 |
mun | -0.025 |
P | -0.024 |
Run | -0.024 |
bitten | -0.023 |
Token | ay |
Token ID | 323 |
Feature activation | +2.81e-02 |
Pos logit contributions | |
Office | +0.072 |
sing | +0.065 |
birthday | +0.064 |
experiment | +0.062 |
believe | +0.062 |
Neg logit contributions | |
ls | -0.112 |
P | -0.103 |
awed | -0.097 |
led | -0.094 |
bitten | -0.091 |
Token | ons |
Token ID | 684 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
birthday | +0.075 |
Office | +0.074 |
ine | +0.073 |
hip | +0.073 |
sing | +0.072 |
Neg logit contributions | |
ls | -0.019 |
P | -0.013 |
mun | -0.012 |
led | -0.010 |
Run | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +2.12e-02 |
Pos logit contributions | |
ine | +0.023 |
Office | +0.021 |
birthday | +0.021 |
X | +0.021 |
hip | +0.021 |
Neg logit contributions | |
ls | -0.015 |
mun | -0.013 |
Run | -0.012 |
P | -0.012 |
led | -0.012 |
Token | One |
Token ID | 1881 |
Feature activation | +3.28e-02 |
Pos logit contributions | |
ine | +0.035 |
urger | +0.032 |
rap | +0.031 |
Office | +0.031 |
believe | +0.031 |
Neg logit contributions | |
mun | -0.034 |
Run | -0.032 |
ls | -0.032 |
Hey | -0.030 |
P | -0.030 |
Token | day |
Token ID | 1110 |
Feature activation | -6.20e-05 |
Pos logit contributions | |
uries | +0.062 |
artments | +0.061 |
hip | +0.059 |
urger | +0.059 |
rap | +0.057 |
Neg logit contributions | |
P | -0.044 |
ls | -0.043 |
Run | -0.038 |
mun | -0.037 |
led | -0.035 |
Token | draw |
Token ID | 3197 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
ine | +0.022 |
Office | +0.022 |
uries | +0.021 |
rap | +0.021 |
hip | +0.021 |
Neg logit contributions | |
Run | -0.021 |
ls | -0.020 |
P | -0.020 |
mun | -0.020 |
bitten | -0.019 |
Token | and |
Token ID | 290 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
ine | +0.047 |
sing | +0.046 |
Office | +0.044 |
uries | +0.044 |
hip | +0.044 |
Neg logit contributions | |
ls | -0.039 |
mun | -0.031 |
P | -0.031 |
led | -0.030 |
distracted | -0.027 |
Token | color |
Token ID | 3124 |
Feature activation | +3.03e-02 |
Pos logit contributions | |
ine | +0.030 |
Office | +0.027 |
hip | +0.026 |
urger | +0.025 |
uries | +0.025 |
Neg logit contributions | |
mun | -0.044 |
ls | -0.044 |
Run | -0.042 |
av | -0.040 |
P | -0.039 |
Token | with |
Token ID | 351 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
sing | +0.068 |
uries | +0.065 |
Office | +0.065 |
ine | +0.064 |
hip | +0.063 |
Neg logit contributions | |
ls | -0.042 |
mun | -0.028 |
ries | -0.027 |
av | -0.026 |
P | -0.025 |
Token | her |
Token ID | 607 |
Feature activation | +1.74e-02 |
Pos logit contributions | |
Office | +0.044 |
ine | +0.042 |
sing | +0.042 |
hip | +0.042 |
uries | +0.041 |
Neg logit contributions | |
ls | -0.027 |
P | -0.023 |
mun | -0.022 |
av | -0.021 |
bitten | -0.021 |
Token | cr |
Token ID | 1067 |
Feature activation | +5.62e-02 |
Pos logit contributions | |
ine | +0.030 |
hip | +0.029 |
Office | +0.028 |
rap | +0.027 |
uries | +0.027 |
Neg logit contributions | |
ls | -0.029 |
mun | -0.025 |
P | -0.024 |
Run | -0.024 |
bitten | -0.023 |
Token | ay |
Token ID | 323 |
Feature activation | +2.81e-02 |
Pos logit contributions | |
Office | +0.072 |
sing | +0.065 |
birthday | +0.064 |
experiment | +0.062 |
believe | +0.062 |
Neg logit contributions | |
ls | -0.112 |
P | -0.103 |
awed | -0.097 |
led | -0.094 |
bitten | -0.091 |
Token | ons |
Token ID | 684 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
birthday | +0.075 |
Office | +0.074 |
ine | +0.073 |
hip | +0.073 |
sing | +0.072 |
Neg logit contributions | |
ls | -0.019 |
P | -0.013 |
mun | -0.012 |
led | -0.010 |
Run | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +2.12e-02 |
Pos logit contributions | |
ine | +0.023 |
Office | +0.021 |
birthday | +0.021 |
X | +0.021 |
hip | +0.021 |
Neg logit contributions | |
ls | -0.015 |
mun | -0.013 |
Run | -0.012 |
P | -0.012 |
led | -0.012 |
Token | One |
Token ID | 1881 |
Feature activation | +3.28e-02 |
Pos logit contributions | |
ine | +0.035 |
urger | +0.032 |
rap | +0.031 |
Office | +0.031 |
believe | +0.031 |
Neg logit contributions | |
mun | -0.034 |
Run | -0.032 |
ls | -0.032 |
Hey | -0.030 |
P | -0.030 |
Token | day |
Token ID | 1110 |
Feature activation | -6.20e-05 |
Pos logit contributions | |
uries | +0.062 |
artments | +0.061 |
hip | +0.059 |
urger | +0.059 |
rap | +0.057 |
Neg logit contributions | |
P | -0.044 |
ls | -0.043 |
Run | -0.038 |
mun | -0.037 |
led | -0.035 |
Token | wrote |
Token ID | 2630 |
Feature activation | +2.78e-02 |
Pos logit contributions | |
ine | +0.026 |
birthday | +0.025 |
hip | +0.024 |
Office | +0.023 |
Jupiter | +0.023 |
Neg logit contributions | |
ls | -0.018 |
Run | -0.018 |
mun | -0.017 |
P | -0.015 |
bitten | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | +7.16e-03 |
Pos logit contributions | |
ine | +0.054 |
sing | +0.050 |
birthday | +0.050 |
uries | +0.049 |
Office | +0.049 |
Neg logit contributions | |
ls | -0.040 |
led | -0.035 |
distracted | -0.033 |
mun | -0.031 |
Run | -0.031 |
Token | wrote |
Token ID | 2630 |
Feature activation | +3.76e-02 |
Pos logit contributions | |
ine | +0.012 |
birthday | +0.011 |
hip | +0.010 |
Office | +0.010 |
Jupiter | +0.010 |
Neg logit contributions | |
Run | -0.012 |
ls | -0.012 |
mun | -0.012 |
P | -0.011 |
bitten | -0.011 |
Token | on |
Token ID | 319 |
Feature activation | +2.43e-02 |
Pos logit contributions | |
ine | +0.068 |
sing | +0.067 |
birthday | +0.063 |
uries | +0.062 |
Office | +0.062 |
Neg logit contributions | |
ls | -0.055 |
led | -0.049 |
mun | -0.045 |
distracted | -0.044 |
Hey | -0.042 |
Token | the |
Token ID | 262 |
Feature activation | +3.33e-02 |
Pos logit contributions | |
birthday | +0.048 |
ine | +0.048 |
hip | +0.046 |
sing | +0.046 |
Office | +0.046 |
Neg logit contributions | |
ls | -0.030 |
mun | -0.026 |
Run | -0.025 |
Hey | -0.022 |
distracted | -0.022 |
Token | soft |
Token ID | 2705 |
Feature activation | +5.59e-02 |
Pos logit contributions | |
ine | +0.056 |
uries | +0.048 |
birthday | +0.048 |
hip | +0.048 |
rap | +0.047 |
Neg logit contributions | |
av | -0.047 |
ls | -0.047 |
Run | -0.046 |
bitten | -0.045 |
P | -0.044 |
Token | wood |
Token ID | 4898 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
uries | +0.089 |
birthday | +0.082 |
anniversary | +0.082 |
icity | +0.080 |
sing | +0.079 |
Neg logit contributions | |
av | -0.072 |
ls | -0.070 |
st | -0.062 |
led | -0.060 |
bitten | -0.060 |
Token | . |
Token ID | 13 |
Feature activation | +1.74e-02 |
Pos logit contributions | |
ine | +0.041 |
birthday | +0.039 |
rap | +0.037 |
Office | +0.037 |
hip | +0.037 |
Neg logit contributions | |
ls | -0.036 |
mun | -0.032 |
Run | -0.032 |
led | -0.029 |
bitten | -0.028 |
Token | Now |
Token ID | 2735 |
Feature activation | +1.45e-02 |
Pos logit contributions | |
ine | +0.033 |
urger | +0.030 |
birthday | +0.030 |
Office | +0.029 |
hip | +0.029 |
Neg logit contributions | |
Run | -0.023 |
ls | -0.022 |
mun | -0.022 |
Hey | -0.020 |
P | -0.019 |
Token | she |
Token ID | 673 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
ine | +0.031 |
sing | +0.030 |
hip | +0.030 |
Office | +0.029 |
birthday | +0.029 |
Neg logit contributions | |
ls | -0.017 |
Run | -0.015 |
mun | -0.014 |
distracted | -0.013 |
Hey | -0.012 |
Token | could |
Token ID | 714 |
Feature activation | +5.60e-03 |
Pos logit contributions | |
ine | +0.034 |
birthday | +0.033 |
hip | +0.032 |
Office | +0.031 |
Jupiter | +0.030 |
Neg logit contributions | |
ls | -0.027 |
mun | -0.025 |
Run | -0.025 |
led | -0.022 |
bitten | -0.022 |
Token | a |
Token ID | 257 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
hip | +0.101 |
basis | +0.097 |
Office | +0.091 |
sing | +0.090 |
jo | +0.090 |
Neg logit contributions | |
ls | -0.067 |
bitten | -0.061 |
st | -0.056 |
P | -0.056 |
distracted | -0.056 |
Token | special |
Token ID | 2041 |
Feature activation | +2.64e-02 |
Pos logit contributions | |
hip | +0.034 |
ine | +0.033 |
rap | +0.032 |
urg | +0.029 |
ika | +0.029 |
Neg logit contributions | |
P | -0.033 |
ls | -0.032 |
Run | -0.031 |
mun | -0.029 |
st | -0.029 |
Token | color |
Token ID | 3124 |
Feature activation | +2.50e-02 |
Pos logit contributions | |
hip | +0.037 |
ine | +0.035 |
rap | +0.035 |
Jupiter | +0.035 |
Him | +0.033 |
Neg logit contributions | |
ls | -0.043 |
Run | -0.040 |
mun | -0.039 |
P | -0.039 |
bitten | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
sing | +0.054 |
ine | +0.053 |
Office | +0.053 |
Jupiter | +0.053 |
artments | +0.052 |
Neg logit contributions | |
ls | -0.025 |
P | -0.016 |
guard | -0.014 |
mun | -0.013 |
Run | -0.013 |
Token | Purple |
Token ID | 17265 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
ine | +0.047 |
birthday | +0.043 |
desk | +0.043 |
sing | +0.042 |
urger | +0.042 |
Neg logit contributions | |
Run | -0.037 |
Hey | -0.034 |
mun | -0.033 |
P | -0.032 |
ls | -0.032 |
Token | is |
Token ID | 318 |
Feature activation | +5.57e-02 |
Pos logit contributions | |
ine | +0.036 |
sing | +0.036 |
birthday | +0.035 |
hip | +0.035 |
X | +0.035 |
Neg logit contributions | |
ls | -0.037 |
mun | -0.032 |
Run | -0.031 |
P | -0.030 |
Hey | -0.029 |
Token | a |
Token ID | 257 |
Feature activation | +2.10e-02 |
Pos logit contributions | |
hip | +0.102 |
basis | +0.098 |
Office | +0.092 |
otive | +0.091 |
sing | +0.091 |
Neg logit contributions | |
ls | -0.069 |
bitten | -0.062 |
P | -0.058 |
st | -0.058 |
distracted | -0.057 |
Token | fun |
Token ID | 1257 |
Feature activation | +2.08e-02 |
Pos logit contributions | |
hip | +0.033 |
ine | +0.033 |
rap | +0.031 |
artments | +0.029 |
urg | +0.029 |
Neg logit contributions | |
P | -0.033 |
ls | -0.031 |
Run | -0.030 |
mun | -0.029 |
st | -0.028 |
Token | color |
Token ID | 3124 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
hip | +0.027 |
Jupiter | +0.027 |
ine | +0.027 |
sing | +0.026 |
Him | +0.026 |
Neg logit contributions | |
ls | -0.033 |
P | -0.032 |
Run | -0.032 |
mun | -0.030 |
Hey | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | +2.77e-02 |
Pos logit contributions | |
ine | +0.049 |
sing | +0.049 |
Office | +0.049 |
Jupiter | +0.048 |
artments | +0.048 |
Neg logit contributions | |
ls | -0.024 |
P | -0.016 |
mun | -0.014 |
guard | -0.014 |
Run | -0.014 |
Token | I |
Token ID | 314 |
Feature activation | +2.81e-02 |
Pos logit contributions | |
ine | +0.050 |
birthday | +0.047 |
desk | +0.047 |
urger | +0.045 |
rap | +0.045 |
Neg logit contributions | |
Run | -0.040 |
Hey | -0.036 |
mun | -0.036 |
P | -0.035 |
ls | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -3.54e-03 |
Pos logit contributions | |
ine | +0.052 |
sing | +0.051 |
urger | +0.050 |
hip | +0.049 |
Office | +0.049 |
Neg logit contributions | |
ls | -0.029 |
Run | -0.021 |
Hey | -0.019 |
av | -0.018 |
mun | -0.017 |
Token | She |
Token ID | 1375 |
Feature activation | +1.43e-02 |
Pos logit contributions | |
Run | +0.006 |
ls | +0.006 |
mun | +0.006 |
P | +0.005 |
Hey | +0.005 |
Neg logit contributions | |
ine | -0.006 |
birthday | -0.006 |
hip | -0.005 |
Office | -0.005 |
X | -0.005 |
Token | picks |
Token ID | 11103 |
Feature activation | +9.16e-03 |
Pos logit contributions | |
ine | +0.023 |
birthday | +0.022 |
hip | +0.021 |
Office | +0.020 |
Jupiter | +0.020 |
Neg logit contributions | |
ls | -0.025 |
Run | -0.024 |
mun | -0.023 |
P | -0.021 |
Hey | -0.021 |
Token | a |
Token ID | 257 |
Feature activation | +2.74e-02 |
Pos logit contributions | |
ine | +0.020 |
birthday | +0.019 |
hip | +0.019 |
sing | +0.019 |
uries | +0.019 |
Neg logit contributions | |
ls | -0.012 |
Run | -0.009 |
distracted | -0.009 |
led | -0.008 |
Hey | -0.008 |
Token | red |
Token ID | 2266 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
ine | +0.040 |
hip | +0.040 |
uries | +0.039 |
Office | +0.036 |
artments | +0.035 |
Neg logit contributions | |
ls | -0.052 |
Run | -0.043 |
P | -0.041 |
mun | -0.039 |
bitten | -0.039 |
Token | cr |
Token ID | 1067 |
Feature activation | +5.55e-02 |
Pos logit contributions | |
uries | +0.043 |
ine | +0.040 |
Office | +0.037 |
artments | +0.036 |
hip | +0.036 |
Neg logit contributions | |
ls | -0.045 |
Run | -0.041 |
Hey | -0.039 |
P | -0.038 |
av | -0.035 |
Token | ay |
Token ID | 323 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
birthday | +0.084 |
Office | +0.081 |
sing | +0.080 |
appearance | +0.080 |
experiment | +0.077 |
Neg logit contributions | |
ls | -0.090 |
P | -0.085 |
awed | -0.077 |
Hey | -0.074 |
umbled | -0.073 |
Token | on |
Token ID | 261 |
Feature activation | +1.27e-02 |
Pos logit contributions | |
birthday | +0.055 |
uries | +0.054 |
sing | +0.053 |
appearance | +0.052 |
ine | +0.052 |
Neg logit contributions | |
ls | -0.024 |
Run | -0.018 |
P | -0.017 |
Hey | -0.015 |
led | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | +1.53e-02 |
Pos logit contributions | |
ine | +0.024 |
Office | +0.024 |
uries | +0.023 |
perfection | +0.023 |
hip | +0.022 |
Neg logit contributions | |
ls | -0.018 |
Run | -0.013 |
led | -0.013 |
bitten | -0.013 |
distracted | -0.012 |
Token | starts |
Token ID | 4940 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
ine | +0.024 |
birthday | +0.023 |
hip | +0.022 |
Office | +0.022 |
uries | +0.021 |
Neg logit contributions | |
ls | -0.027 |
Run | -0.027 |
mun | -0.025 |
P | -0.023 |
bitten | -0.023 |
Token | to |
Token ID | 284 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
ine | +0.033 |
hip | +0.032 |
Office | +0.032 |
uries | +0.032 |
sing | +0.031 |
Neg logit contributions | |
ls | -0.017 |
bitten | -0.015 |
distracted | -0.014 |
Run | -0.014 |
mun | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
ine | +0.034 |
birthday | +0.033 |
Office | +0.031 |
X | +0.031 |
hip | +0.031 |
Neg logit contributions | |
ls | -0.024 |
Run | -0.024 |
mun | -0.023 |
P | -0.021 |
Hey | -0.020 |
Token | M |
Token ID | 44 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
ine | +0.023 |
birthday | +0.022 |
X | +0.020 |
Office | +0.020 |
Jupiter | +0.020 |
Neg logit contributions | |
Run | -0.019 |
ls | -0.019 |
mun | -0.018 |
P | -0.017 |
Hey | -0.017 |
Token | oral |
Token ID | 6864 |
Feature activation | +2.12e-02 |
Pos logit contributions | |
ine | +0.026 |
birthday | +0.024 |
Office | +0.023 |
hip | +0.023 |
X | +0.023 |
Neg logit contributions | |
ls | -0.029 |
Run | -0.028 |
mun | -0.027 |
P | -0.025 |
Hey | -0.024 |
Token | of |
Token ID | 286 |
Feature activation | +4.37e-02 |
Pos logit contributions | |
ine | +0.037 |
birthday | +0.035 |
X | +0.032 |
hip | +0.032 |
Office | +0.032 |
Neg logit contributions | |
ls | -0.034 |
Run | -0.033 |
mun | -0.031 |
P | -0.030 |
Hey | -0.029 |
Token | the |
Token ID | 262 |
Feature activation | +1.22e-02 |
Pos logit contributions | |
uries | +0.099 |
ine | +0.099 |
artments | +0.098 |
hip | +0.092 |
sing | +0.092 |
Neg logit contributions | |
mun | -0.038 |
Run | -0.036 |
ls | -0.035 |
P | -0.034 |
Hey | -0.030 |
Token | story |
Token ID | 1621 |
Feature activation | +5.54e-02 |
Pos logit contributions | |
ine | +0.014 |
hip | +0.012 |
artments | +0.012 |
birthday | +0.012 |
Office | +0.011 |
Neg logit contributions | |
ls | -0.027 |
Run | -0.027 |
mun | -0.026 |
P | -0.025 |
Hey | -0.024 |
Token | : |
Token ID | 25 |
Feature activation | +3.49e-02 |
Pos logit contributions | |
ine | +0.069 |
birthday | +0.062 |
X | +0.060 |
urger | +0.060 |
sing | +0.060 |
Neg logit contributions | |
ls | -0.112 |
Run | -0.101 |
mun | -0.099 |
Hey | -0.094 |
led | -0.093 |
Token | Regular |
Token ID | 23603 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
ine | +0.057 |
birthday | +0.052 |
X | +0.050 |
eye | +0.050 |
pattern | +0.050 |
Neg logit contributions | |
Run | -0.055 |
mun | -0.053 |
ls | -0.050 |
Hey | -0.046 |
distracted | -0.045 |
Token | practice |
Token ID | 3357 |
Feature activation | +1.45e-02 |
Pos logit contributions | |
ine | +0.023 |
birthday | +0.021 |
X | +0.021 |
Office | +0.020 |
Jupiter | +0.020 |
Neg logit contributions | |
ls | -0.029 |
Run | -0.028 |
mun | -0.028 |
led | -0.026 |
Hey | -0.025 |
Token | makes |
Token ID | 1838 |
Feature activation | +1.72e-02 |
Pos logit contributions | |
ine | +0.021 |
birthday | +0.021 |
X | +0.020 |
Office | +0.019 |
Jupiter | +0.019 |
Neg logit contributions | |
ls | -0.026 |
mun | -0.025 |
Run | -0.025 |
P | -0.022 |
led | -0.022 |
Token | perfect |
Token ID | 2818 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
ine | +0.028 |
X | +0.027 |
birthday | +0.026 |
Office | +0.026 |
hip | +0.026 |
Neg logit contributions | |
ls | -0.028 |
mun | -0.027 |
Run | -0.027 |
P | -0.025 |
bitten | -0.023 |
Token | can |
Token ID | 460 |
Feature activation | +1.77e-04 |
Pos logit contributions | |
ine | +0.025 |
birthday | +0.023 |
hip | +0.022 |
Office | +0.022 |
rap | +0.022 |
Neg logit contributions | |
Run | -0.029 |
ls | -0.029 |
mun | -0.028 |
P | -0.025 |
Hey | -0.025 |
Token | draw |
Token ID | 3197 |
Feature activation | +3.40e-02 |
Pos logit contributions | |
ine | +0.000 |
birthday | +0.000 |
Office | +0.000 |
hip | +0.000 |
X | +0.000 |
Neg logit contributions | |
ls | -0.000 |
Run | -0.000 |
mun | -0.000 |
P | -0.000 |
Hey | -0.000 |
Token | on |
Token ID | 319 |
Feature activation | +2.76e-02 |
Pos logit contributions | |
ine | +0.063 |
hip | +0.059 |
sing | +0.057 |
Office | +0.057 |
birthday | +0.056 |
Neg logit contributions | |
ls | -0.057 |
Run | -0.046 |
mun | -0.046 |
led | -0.043 |
P | -0.041 |
Token | paper |
Token ID | 3348 |
Feature activation | +3.32e-02 |
Pos logit contributions | |
ine | +0.053 |
hip | +0.051 |
birthday | +0.049 |
Office | +0.049 |
sing | +0.049 |
Neg logit contributions | |
ls | -0.039 |
Run | -0.035 |
mun | -0.034 |
P | -0.030 |
Hey | -0.030 |
Token | with |
Token ID | 351 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
Office | +0.058 |
hip | +0.058 |
ine | +0.057 |
sing | +0.056 |
uries | +0.055 |
Neg logit contributions | |
ls | -0.052 |
mun | -0.038 |
led | -0.038 |
Run | -0.034 |
P | -0.034 |
Token | cr |
Token ID | 1067 |
Feature activation | +5.47e-02 |
Pos logit contributions | |
ine | +0.033 |
Office | +0.032 |
hip | +0.031 |
rap | +0.030 |
uries | +0.029 |
Neg logit contributions | |
ls | -0.033 |
Run | -0.030 |
mun | -0.028 |
P | -0.028 |
bitten | -0.028 |
Token | ay |
Token ID | 323 |
Feature activation | +2.91e-02 |
Pos logit contributions | |
Office | +0.067 |
birthday | +0.064 |
sing | +0.063 |
reminder | +0.060 |
experiment | +0.059 |
Neg logit contributions | |
ls | -0.112 |
P | -0.103 |
awed | -0.095 |
led | -0.094 |
mun | -0.091 |
Token | ons |
Token ID | 684 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
birthday | +0.076 |
ine | +0.075 |
Office | +0.074 |
hip | +0.074 |
sing | +0.073 |
Neg logit contributions | |
ls | -0.021 |
mun | -0.015 |
Run | -0.014 |
P | -0.014 |
led | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | +2.62e-02 |
Pos logit contributions | |
ine | +0.035 |
Office | +0.034 |
birthday | +0.033 |
hip | +0.033 |
rap | +0.033 |
Neg logit contributions | |
ls | -0.031 |
mun | -0.025 |
Run | -0.025 |
P | -0.024 |
led | -0.024 |
Token | stickers |
Token ID | 28568 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
ine | +0.039 |
hip | +0.037 |
rap | +0.036 |
Office | +0.036 |
birthday | +0.034 |
Neg logit contributions | |
ls | -0.047 |
Run | -0.047 |
mun | -0.044 |
P | -0.042 |
bitten | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | +7.85e-03 |
Pos logit contributions | |
ine | +0.076 |
Office | +0.074 |
birthday | +0.073 |
sing | +0.072 |
Jupiter | +0.070 |
Neg logit contributions | |
ls | -0.048 |
led | -0.037 |
mun | -0.033 |
Run | -0.032 |
P | -0.032 |
Token | a |
Token ID | 257 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
hip | +0.095 |
basis | +0.090 |
sing | +0.085 |
jo | +0.085 |
ine | +0.084 |
Neg logit contributions | |
ls | -0.058 |
bitten | -0.053 |
distracted | -0.048 |
st | -0.047 |
P | -0.047 |
Token | different |
Token ID | 1180 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
hip | +0.034 |
ine | +0.033 |
rap | +0.031 |
Him | +0.029 |
urg | +0.029 |
Neg logit contributions | |
P | -0.031 |
ls | -0.030 |
Run | -0.029 |
st | -0.027 |
bitten | -0.027 |
Token | color |
Token ID | 3124 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
hip | +0.017 |
ine | +0.017 |
rap | +0.017 |
Jupiter | +0.016 |
Office | +0.016 |
Neg logit contributions | |
ls | -0.022 |
Run | -0.021 |
mun | -0.020 |
bitten | -0.019 |
Hey | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +2.27e-02 |
Pos logit contributions | |
ine | +0.049 |
sing | +0.048 |
Office | +0.048 |
Jupiter | +0.048 |
hip | +0.047 |
Neg logit contributions | |
ls | -0.023 |
P | -0.013 |
guard | -0.013 |
Run | -0.012 |
bitten | -0.011 |
Token | Purple |
Token ID | 17265 |
Feature activation | +1.98e-02 |
Pos logit contributions | |
ine | +0.041 |
birthday | +0.038 |
desk | +0.037 |
rap | +0.037 |
sing | +0.037 |
Neg logit contributions | |
Run | -0.033 |
Hey | -0.029 |
ls | -0.028 |
mun | -0.028 |
P | -0.028 |
Token | is |
Token ID | 318 |
Feature activation | +5.47e-02 |
Pos logit contributions | |
ine | +0.032 |
birthday | +0.031 |
hip | +0.031 |
sing | +0.031 |
X | +0.031 |
Neg logit contributions | |
ls | -0.033 |
mun | -0.028 |
Run | -0.028 |
P | -0.026 |
Hey | -0.026 |
Token | a |
Token ID | 257 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
hip | +0.101 |
basis | +0.097 |
Office | +0.091 |
sing | +0.090 |
jo | +0.090 |
Neg logit contributions | |
ls | -0.067 |
bitten | -0.061 |
st | -0.056 |
P | -0.056 |
distracted | -0.056 |
Token | special |
Token ID | 2041 |
Feature activation | +2.64e-02 |
Pos logit contributions | |
hip | +0.034 |
ine | +0.033 |
rap | +0.032 |
urg | +0.029 |
ika | +0.029 |
Neg logit contributions | |
P | -0.033 |
ls | -0.032 |
Run | -0.031 |
mun | -0.029 |
st | -0.029 |
Token | color |
Token ID | 3124 |
Feature activation | +2.50e-02 |
Pos logit contributions | |
hip | +0.037 |
ine | +0.035 |
rap | +0.035 |
Jupiter | +0.035 |
Him | +0.033 |
Neg logit contributions | |
ls | -0.043 |
Run | -0.040 |
mun | -0.039 |
P | -0.039 |
bitten | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
sing | +0.054 |
ine | +0.053 |
Office | +0.053 |
Jupiter | +0.053 |
artments | +0.052 |
Neg logit contributions | |
ls | -0.025 |
P | -0.016 |
guard | -0.014 |
mun | -0.013 |
Run | -0.013 |
Token | Purple |
Token ID | 17265 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
ine | +0.047 |
birthday | +0.043 |
desk | +0.043 |
sing | +0.042 |
urger | +0.042 |
Neg logit contributions | |
Run | -0.037 |
Hey | -0.034 |
mun | -0.033 |
P | -0.032 |
ls | -0.032 |
Token |
|
Token ID | 198 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
ine | +0.033 |
birthday | +0.031 |
X | +0.031 |
Office | +0.030 |
hip | +0.030 |
Neg logit contributions | |
ls | -0.021 |
Run | -0.020 |
mun | -0.020 |
P | -0.019 |
Hey | -0.017 |
Token | M |
Token ID | 44 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
ine | +0.021 |
birthday | +0.020 |
X | +0.020 |
Office | +0.019 |
hip | +0.019 |
Neg logit contributions | |
Run | -0.016 |
ls | -0.016 |
mun | -0.016 |
P | -0.015 |
Hey | -0.014 |
Token | oral |
Token ID | 6864 |
Feature activation | +1.52e-02 |
Pos logit contributions | |
ine | +0.032 |
birthday | +0.029 |
Office | +0.029 |
X | +0.029 |
hip | +0.029 |
Neg logit contributions | |
ls | -0.035 |
Run | -0.033 |
mun | -0.031 |
P | -0.029 |
led | -0.028 |
Token | of |
Token ID | 286 |
Feature activation | +4.73e-02 |
Pos logit contributions | |
ine | +0.024 |
birthday | +0.023 |
hip | +0.021 |
X | +0.021 |
Office | +0.021 |
Neg logit contributions | |
ls | -0.027 |
Run | -0.026 |
mun | -0.025 |
P | -0.024 |
Hey | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
ine | +0.108 |
artments | +0.106 |
uries | +0.106 |
sing | +0.099 |
urger | +0.099 |
Neg logit contributions | |
mun | -0.041 |
P | -0.041 |
Run | -0.040 |
ls | -0.037 |
Hey | -0.036 |
Token | story |
Token ID | 1621 |
Feature activation | +5.42e-02 |
Pos logit contributions | |
ine | +0.013 |
birthday | +0.011 |
hip | +0.011 |
artments | +0.011 |
Office | +0.011 |
Neg logit contributions | |
ls | -0.025 |
Run | -0.024 |
mun | -0.024 |
P | -0.023 |
Hey | -0.022 |
Token | : |
Token ID | 25 |
Feature activation | +3.14e-02 |
Pos logit contributions | |
ine | +0.075 |
urger | +0.073 |
X | +0.071 |
uries | +0.071 |
artments | +0.070 |
Neg logit contributions | |
ls | -0.101 |
mun | -0.084 |
Run | -0.081 |
led | -0.081 |
Hey | -0.080 |
Token | If |
Token ID | 1002 |
Feature activation | +7.54e-03 |
Pos logit contributions | |
ine | +0.055 |
X | +0.049 |
birthday | +0.049 |
urger | +0.048 |
Office | +0.047 |
Neg logit contributions | |
Run | -0.046 |
mun | -0.045 |
ls | -0.044 |
distracted | -0.039 |
bitten | -0.039 |
Token | you |
Token ID | 345 |
Feature activation | -9.88e-03 |
Pos logit contributions | |
ine | +0.016 |
birthday | +0.015 |
hip | +0.014 |
Office | +0.014 |
X | +0.014 |
Neg logit contributions | |
ls | -0.009 |
Run | -0.009 |
mun | -0.009 |
P | -0.008 |
Hey | -0.008 |
Token | are |
Token ID | 389 |
Feature activation | -9.32e-03 |
Pos logit contributions | |
ls | +0.023 |
Run | +0.022 |
mun | +0.022 |
P | +0.020 |
awed | +0.020 |
Neg logit contributions | |
ine | -0.011 |
birthday | -0.010 |
hip | -0.010 |
X | -0.009 |
eye | -0.009 |
Token | sorry |
Token ID | 7926 |
Feature activation | +9.00e-03 |
Pos logit contributions | |
Run | +0.017 |
ls | +0.017 |
mun | +0.016 |
P | +0.015 |
Hey | +0.015 |
Neg logit contributions | |
ine | -0.014 |
birthday | -0.013 |
hip | -0.012 |
Office | -0.012 |
X | -0.012 |
Token | She |
Token ID | 1375 |
Feature activation | -4.96e-03 |
Pos logit contributions | |
ls | +0.009 |
Run | +0.009 |
mun | +0.009 |
P | +0.008 |
led | +0.008 |
Neg logit contributions | |
ine | -0.008 |
birthday | -0.008 |
Office | -0.007 |
X | -0.007 |
hip | -0.007 |
Token | is |
Token ID | 318 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
Run | +0.008 |
ls | +0.008 |
mun | +0.008 |
P | +0.008 |
bitten | +0.007 |
Neg logit contributions | |
ine | -0.008 |
birthday | -0.007 |
X | -0.007 |
Office | -0.007 |
Jupiter | -0.007 |
Token | using |
Token ID | 1262 |
Feature activation | +2.30e-03 |
Pos logit contributions | |
ine | +0.026 |
birthday | +0.025 |
hip | +0.024 |
Office | +0.024 |
X | +0.023 |
Neg logit contributions | |
ls | -0.019 |
Run | -0.019 |
mun | -0.018 |
bitten | -0.017 |
P | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | +2.04e-02 |
Pos logit contributions | |
birthday | +0.005 |
hip | +0.005 |
Office | +0.005 |
ine | +0.005 |
Birthday | +0.004 |
Neg logit contributions | |
ls | -0.003 |
bitten | -0.002 |
Run | -0.002 |
mun | -0.002 |
distracted | -0.002 |
Token | red |
Token ID | 2266 |
Feature activation | +2.19e-02 |
Pos logit contributions | |
ine | +0.035 |
hip | +0.033 |
Office | +0.032 |
rap | +0.031 |
uries | +0.031 |
Neg logit contributions | |
ls | -0.032 |
Run | -0.029 |
P | -0.029 |
mun | -0.028 |
bitten | -0.027 |
Token | cr |
Token ID | 1067 |
Feature activation | +5.39e-02 |
Pos logit contributions | |
ine | +0.040 |
uries | +0.039 |
hip | +0.039 |
Jupiter | +0.039 |
Office | +0.038 |
Neg logit contributions | |
ls | -0.030 |
P | -0.026 |
Run | -0.026 |
Hey | -0.024 |
mun | -0.024 |
Token | ay |
Token ID | 323 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
Office | +0.074 |
birthday | +0.074 |
sing | +0.070 |
appearance | +0.069 |
experiment | +0.068 |
Neg logit contributions | |
ls | -0.098 |
P | -0.093 |
awed | -0.083 |
led | -0.081 |
itt | -0.080 |
Token | on |
Token ID | 261 |
Feature activation | +7.79e-03 |
Pos logit contributions | |
birthday | +0.048 |
Office | +0.047 |
sing | +0.047 |
ine | +0.047 |
uries | +0.047 |
Neg logit contributions | |
ls | -0.024 |
P | -0.019 |
Run | -0.017 |
led | -0.017 |
bitten | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +1.44e-03 |
Pos logit contributions | |
ine | +0.015 |
birthday | +0.014 |
Office | +0.014 |
hip | +0.014 |
Jupiter | +0.014 |
Neg logit contributions | |
ls | -0.011 |
Run | -0.010 |
mun | -0.009 |
led | -0.009 |
P | -0.009 |
Token | He |
Token ID | 679 |
Feature activation | +3.28e-03 |
Pos logit contributions | |
ine | +0.002 |
birthday | +0.002 |
Office | +0.002 |
hip | +0.002 |
X | +0.002 |
Neg logit contributions | |
ls | -0.003 |
Run | -0.003 |
mun | -0.002 |
P | -0.002 |
bitten | -0.002 |
Token | says |
Token ID | 1139 |
Feature activation | +8.47e-03 |
Pos logit contributions | |
ine | +0.005 |
birthday | +0.005 |
Office | +0.004 |
hip | +0.004 |
X | +0.004 |
Neg logit contributions | |
Run | -0.006 |
ls | -0.006 |
mun | -0.006 |
P | -0.005 |
Hey | -0.005 |
Token | liked |
Token ID | 8288 |
Feature activation | -7.01e-03 |
Pos logit contributions | |
ine | +0.009 |
birthday | +0.008 |
Office | +0.008 |
hip | +0.008 |
X | +0.008 |
Neg logit contributions | |
Run | -0.008 |
ls | -0.008 |
mun | -0.008 |
P | -0.007 |
Hey | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | +9.54e-03 |
Pos logit contributions | |
Run | +0.012 |
ls | +0.012 |
mun | +0.012 |
P | +0.011 |
led | +0.010 |
Neg logit contributions | |
ine | -0.011 |
birthday | -0.011 |
X | -0.010 |
Office | -0.010 |
Jupiter | -0.010 |
Token | draw |
Token ID | 3197 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
ine | +0.013 |
Office | +0.012 |
birthday | +0.012 |
rap | +0.012 |
Jupiter | +0.012 |
Neg logit contributions | |
Run | -0.017 |
ls | -0.017 |
mun | -0.016 |
P | -0.016 |
bitten | -0.015 |
Token | with |
Token ID | 351 |
Feature activation | +6.35e-03 |
Pos logit contributions | |
ine | +0.031 |
uries | +0.029 |
hip | +0.029 |
sing | +0.029 |
Office | +0.029 |
Neg logit contributions | |
ls | -0.025 |
mun | -0.020 |
led | -0.020 |
P | -0.020 |
Run | -0.020 |
Token | his |
Token ID | 465 |
Feature activation | +1.53e-02 |
Pos logit contributions | |
ine | +0.012 |
birthday | +0.011 |
Office | +0.011 |
hip | +0.011 |
uries | +0.010 |
Neg logit contributions | |
ls | -0.009 |
Run | -0.009 |
mun | -0.009 |
P | -0.008 |
bitten | -0.008 |
Token | cr |
Token ID | 1067 |
Feature activation | +5.38e-02 |
Pos logit contributions | |
ine | +0.025 |
hip | +0.024 |
Office | +0.023 |
birthday | +0.022 |
rap | +0.022 |
Neg logit contributions | |
ls | -0.027 |
Run | -0.024 |
mun | -0.023 |
P | -0.023 |
bitten | -0.022 |
Token | ay |
Token ID | 323 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
Office | +0.068 |
birthday | +0.062 |
sing | +0.059 |
ine | +0.059 |
experiment | +0.058 |
Neg logit contributions | |
ls | -0.110 |
P | -0.100 |
awed | -0.094 |
led | -0.092 |
bitten | -0.089 |
Token | ons |
Token ID | 684 |
Feature activation | +8.13e-03 |
Pos logit contributions | |
birthday | +0.070 |
ine | +0.069 |
Office | +0.069 |
uries | +0.067 |
sing | +0.067 |
Neg logit contributions | |
ls | -0.019 |
mun | -0.013 |
P | -0.012 |
led | -0.011 |
Run | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
ine | +0.015 |
Office | +0.014 |
birthday | +0.014 |
Jupiter | +0.014 |
rap | +0.014 |
Neg logit contributions | |
ls | -0.012 |
mun | -0.010 |
P | -0.010 |
led | -0.010 |
Run | -0.010 |
Token | markers |
Token ID | 19736 |
Feature activation | +2.60e-02 |
Pos logit contributions | |
ine | +0.032 |
hip | +0.028 |
Office | +0.028 |
birthday | +0.027 |
rap | +0.027 |
Neg logit contributions | |
ls | -0.041 |
mun | -0.039 |
Run | -0.039 |
bitten | -0.035 |
P | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
ine | +0.052 |
Office | +0.051 |
birthday | +0.048 |
sing | +0.048 |
ika | +0.048 |
Neg logit contributions | |
ls | -0.034 |
led | -0.028 |
mun | -0.026 |
bitten | -0.024 |
P | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +8.38e-03 |
Pos logit contributions | |
ine | +0.073 |
Office | +0.069 |
sing | +0.068 |
artments | +0.066 |
urger | +0.066 |
Neg logit contributions | |
ls | -0.039 |
mun | -0.028 |
Run | -0.026 |
led | -0.025 |
P | -0.025 |
Token | Tom |
Token ID | 4186 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
ine | +0.015 |
birthday | +0.014 |
hip | +0.014 |
rap | +0.013 |
Office | +0.013 |
Neg logit contributions | |
Run | -0.013 |
ls | -0.012 |
mun | -0.012 |
Hey | -0.011 |
P | -0.011 |
Token | has |
Token ID | 468 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
birthday | +0.026 |
hip | +0.024 |
ine | +0.023 |
perfection | +0.022 |
sing | +0.022 |
Neg logit contributions | |
mun | -0.024 |
ls | -0.024 |
Run | -0.021 |
Hey | -0.021 |
teased | -0.021 |
Token | a |
Token ID | 257 |
Feature activation | +6.48e-03 |
Pos logit contributions | |
ine | +0.024 |
hip | +0.023 |
rap | +0.022 |
Office | +0.022 |
birthday | +0.022 |
Neg logit contributions | |
Run | -0.019 |
ls | -0.018 |
P | -0.018 |
bitten | -0.018 |
mun | -0.018 |
Token | red |
Token ID | 2266 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
hip | +0.011 |
ine | +0.011 |
rap | +0.010 |
uries | +0.010 |
Office | +0.010 |
Neg logit contributions | |
ls | -0.010 |
P | -0.010 |
mun | -0.009 |
Run | -0.009 |
bitten | -0.008 |
Token | cr |
Token ID | 1067 |
Feature activation | +5.36e-02 |
Pos logit contributions | |
uries | +0.030 |
ine | +0.028 |
hip | +0.028 |
Office | +0.028 |
Jupiter | +0.028 |
Neg logit contributions | |
P | -0.027 |
ls | -0.027 |
Run | -0.026 |
Hey | -0.024 |
mun | -0.024 |
Token | ay |
Token ID | 323 |
Feature activation | +2.50e-02 |
Pos logit contributions | |
Office | +0.072 |
birthday | +0.068 |
experiment | +0.066 |
sing | +0.064 |
Jupiter | +0.063 |
Neg logit contributions | |
ls | -0.097 |
P | -0.095 |
awed | -0.086 |
bitten | -0.083 |
led | -0.083 |
Token | on |
Token ID | 261 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
birthday | +0.052 |
Office | +0.051 |
ine | +0.051 |
uries | +0.051 |
sing | +0.049 |
Neg logit contributions | |
ls | -0.029 |
P | -0.026 |
Run | -0.024 |
bitten | -0.023 |
mun | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
ine | +0.030 |
Office | +0.029 |
Jupiter | +0.028 |
hip | +0.028 |
rap | +0.027 |
Neg logit contributions | |
ls | -0.022 |
P | -0.018 |
led | -0.017 |
bitten | -0.017 |
Run | -0.017 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.40e-02 |
Pos logit contributions | |
ine | +0.024 |
hip | +0.023 |
birthday | +0.022 |
Office | +0.022 |
rap | +0.021 |
Neg logit contributions | |
ls | -0.022 |
Run | -0.021 |
mun | -0.020 |
P | -0.019 |
bitten | -0.019 |
Token | has |
Token ID | 468 |
Feature activation | +2.14e-02 |
Pos logit contributions | |
birthday | +0.025 |
hip | +0.025 |
Jupiter | +0.024 |
rap | +0.024 |
ine | +0.024 |
Neg logit contributions | |
ls | -0.019 |
mun | -0.018 |
led | -0.017 |
cried | -0.017 |
bitten | -0.017 |
Token | with |
Token ID | 351 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
ine | +0.005 |
birthday | +0.005 |
Office | +0.004 |
hip | +0.004 |
X | +0.004 |
Neg logit contributions | |
Run | -0.006 |
ls | -0.006 |
mun | -0.005 |
P | -0.005 |
Hey | -0.005 |
Token | his |
Token ID | 465 |
Feature activation | +9.31e-03 |
Pos logit contributions | |
ine | +0.017 |
birthday | +0.016 |
X | +0.015 |
moon | +0.015 |
Jupiter | +0.015 |
Neg logit contributions | |
Run | -0.018 |
ls | -0.018 |
mun | -0.018 |
P | -0.016 |
Hey | -0.016 |
Token | drum |
Token ID | 13026 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
ine | +0.013 |
birthday | +0.013 |
X | +0.012 |
Office | +0.012 |
Jupiter | +0.011 |
Neg logit contributions | |
Run | -0.018 |
ls | -0.017 |
mun | -0.017 |
Hey | -0.016 |
P | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +2.83e-02 |
Pos logit contributions | |
ine | +0.029 |
birthday | +0.026 |
Office | +0.026 |
X | +0.026 |
rap | +0.025 |
Neg logit contributions | |
ls | -0.028 |
mun | -0.026 |
Run | -0.025 |
P | -0.024 |
led | -0.023 |
Token | It |
Token ID | 632 |
Feature activation | +2.28e-02 |
Pos logit contributions | |
ine | +0.056 |
believe | +0.053 |
birthday | +0.052 |
uries | +0.051 |
urger | +0.051 |
Neg logit contributions | |
Run | -0.030 |
mun | -0.028 |
ls | -0.027 |
Hey | -0.027 |
� | -0.022 |
Token | was |
Token ID | 373 |
Feature activation | +5.34e-02 |
Pos logit contributions | |
ine | +0.037 |
birthday | +0.035 |
Office | +0.032 |
hip | +0.032 |
Jupiter | +0.032 |
Neg logit contributions | |
Run | -0.039 |
ls | -0.039 |
mun | -0.037 |
P | -0.034 |
Hey | -0.033 |
Token | big |
Token ID | 1263 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
Office | +0.098 |
hip | +0.098 |
ine | +0.095 |
uries | +0.094 |
urger | +0.092 |
Neg logit contributions | |
Run | -0.067 |
P | -0.066 |
ls | -0.066 |
bitten | -0.066 |
distracted | -0.064 |
Token | and |
Token ID | 290 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
ine | +0.022 |
hip | +0.020 |
birthday | +0.020 |
Office | +0.020 |
uries | +0.019 |
Neg logit contributions | |
ls | -0.022 |
Run | -0.021 |
mun | -0.021 |
P | -0.019 |
bitten | -0.019 |
Token | heavy |
Token ID | 4334 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
ine | +0.032 |
birthday | +0.030 |
hip | +0.029 |
Office | +0.029 |
rap | +0.029 |
Neg logit contributions | |
Run | -0.032 |
ls | -0.031 |
mun | -0.029 |
P | -0.028 |
Hey | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | +7.01e-03 |
Pos logit contributions | |
ine | +0.022 |
rap | +0.021 |
Office | +0.021 |
uries | +0.020 |
urger | +0.020 |
Neg logit contributions | |
ls | -0.015 |
Run | -0.014 |
mun | -0.013 |
bitten | -0.012 |
led | -0.012 |
Token | but |
Token ID | 475 |
Feature activation | +1.91e-02 |
Pos logit contributions | |
ine | +0.011 |
birthday | +0.010 |
Office | +0.009 |
hip | +0.009 |
X | +0.009 |
Neg logit contributions | |
Run | -0.013 |
ls | -0.013 |
mun | -0.012 |
P | -0.011 |
Hey | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | +6.21e-03 |
Pos logit contributions | |
ine | +0.028 |
birthday | +0.027 |
Office | +0.025 |
X | +0.025 |
hip | +0.025 |
Neg logit contributions | |
ls | -0.019 |
Run | -0.019 |
mun | -0.018 |
P | -0.017 |
Hey | -0.016 |
Token | M |
Token ID | 44 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
ine | +0.011 |
birthday | +0.010 |
hip | +0.010 |
Office | +0.010 |
X | +0.009 |
Neg logit contributions | |
ls | -0.010 |
Run | -0.010 |
mun | -0.010 |
P | -0.008 |
bitten | -0.008 |
Token | oral |
Token ID | 6864 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
ine | +0.016 |
birthday | +0.015 |
Office | +0.014 |
hip | +0.014 |
Jupiter | +0.014 |
Neg logit contributions | |
ls | -0.018 |
Run | -0.017 |
mun | -0.017 |
P | -0.016 |
Hey | -0.015 |
Token | of |
Token ID | 286 |
Feature activation | +4.11e-02 |
Pos logit contributions | |
ine | +0.022 |
birthday | +0.021 |
X | +0.020 |
sing | +0.020 |
hip | +0.020 |
Neg logit contributions | |
ls | -0.027 |
Run | -0.026 |
mun | -0.025 |
P | -0.024 |
Hey | -0.024 |
Token | the |
Token ID | 262 |
Feature activation | +7.40e-03 |
Pos logit contributions | |
ine | +0.093 |
artments | +0.093 |
uries | +0.092 |
sing | +0.089 |
urger | +0.088 |
Neg logit contributions | |
mun | -0.036 |
Run | -0.034 |
P | -0.034 |
ls | -0.034 |
Hey | -0.030 |
Token | story |
Token ID | 1621 |
Feature activation | +5.33e-02 |
Pos logit contributions | |
ine | +0.009 |
artments | +0.008 |
sing | +0.008 |
birthday | +0.008 |
hip | +0.008 |
Neg logit contributions | |
ls | -0.015 |
Run | -0.015 |
mun | -0.015 |
P | -0.014 |
Hey | -0.014 |
Token | : |
Token ID | 25 |
Feature activation | +3.34e-02 |
Pos logit contributions | |
ine | +0.070 |
urger | +0.067 |
birthday | +0.066 |
sing | +0.065 |
uries | +0.063 |
Neg logit contributions | |
ls | -0.103 |
Run | -0.087 |
mun | -0.086 |
Hey | -0.084 |
led | -0.081 |
Token | Sharing |
Token ID | 36644 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
ine | +0.056 |
birthday | +0.051 |
urger | +0.051 |
X | +0.050 |
eye | +0.050 |
Neg logit contributions | |
Run | -0.049 |
mun | -0.047 |
ls | -0.046 |
Hey | -0.042 |
distracted | -0.041 |
Token | is |
Token ID | 318 |
Feature activation | +2.12e-02 |
Pos logit contributions | |
ine | +0.019 |
birthday | +0.018 |
X | +0.017 |
Office | +0.017 |
hip | +0.017 |
Neg logit contributions | |
ls | -0.022 |
Run | -0.021 |
mun | -0.021 |
P | -0.019 |
Hey | -0.019 |
Token | caring |
Token ID | 18088 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
ine | +0.030 |
birthday | +0.027 |
X | +0.027 |
hip | +0.026 |
Office | +0.026 |
Neg logit contributions | |
ls | -0.039 |
Run | -0.039 |
mun | -0.038 |
P | -0.036 |
distracted | -0.035 |
Token | ! |
Token ID | 0 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
ine | +0.026 |
X | +0.024 |
Jupiter | +0.024 |
Office | +0.024 |
birthday | +0.024 |
Neg logit contributions | |
ls | -0.029 |
mun | -0.026 |
Run | -0.026 |
led | -0.024 |
Hey | -0.023 |
Token | his |
Token ID | 465 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
ine | +0.008 |
birthday | +0.007 |
hip | +0.007 |
Office | +0.007 |
X | +0.007 |
Neg logit contributions | |
ls | -0.005 |
Run | -0.004 |
mun | -0.004 |
P | -0.004 |
Hey | -0.004 |
Token | sc |
Token ID | 629 |
Feature activation | +3.99e-03 |
Pos logit contributions | |
ine | +0.019 |
birthday | +0.017 |
hip | +0.016 |
Office | +0.016 |
Jupiter | +0.016 |
Neg logit contributions | |
ls | -0.026 |
Run | -0.025 |
mun | -0.025 |
P | -0.023 |
Hey | -0.022 |
Token | ooter |
Token ID | 25141 |
Feature activation | +8.00e-03 |
Pos logit contributions | |
ine | +0.007 |
birthday | +0.006 |
Office | +0.006 |
Jupiter | +0.006 |
moon | +0.006 |
Neg logit contributions | |
ls | -0.006 |
Run | -0.006 |
mun | -0.006 |
P | -0.005 |
distracted | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
ine | +0.013 |
birthday | +0.012 |
Office | +0.012 |
Jupiter | +0.012 |
X | +0.012 |
Neg logit contributions | |
ls | -0.013 |
Run | -0.013 |
mun | -0.013 |
P | -0.012 |
led | -0.011 |
Token | It |
Token ID | 632 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
ine | +0.051 |
birthday | +0.047 |
believe | +0.047 |
urger | +0.046 |
Birthday | +0.045 |
Neg logit contributions | |
Run | -0.025 |
mun | -0.024 |
Hey | -0.022 |
ls | -0.022 |
P | -0.019 |
Token | was |
Token ID | 373 |
Feature activation | +5.31e-02 |
Pos logit contributions | |
ine | +0.031 |
birthday | +0.029 |
hip | +0.027 |
Office | +0.027 |
X | +0.027 |
Neg logit contributions | |
ls | -0.028 |
Run | -0.028 |
mun | -0.027 |
P | -0.024 |
Hey | -0.024 |
Token | red |
Token ID | 2266 |
Feature activation | +6.35e-03 |
Pos logit contributions | |
hip | +0.101 |
Office | +0.100 |
ine | +0.099 |
uries | +0.096 |
urger | +0.096 |
Neg logit contributions | |
Run | -0.061 |
distracted | -0.060 |
P | -0.060 |
bitten | -0.060 |
tangled | -0.059 |
Token | and |
Token ID | 290 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
ine | +0.010 |
Office | +0.009 |
birthday | +0.009 |
Jupiter | +0.009 |
hip | +0.009 |
Neg logit contributions | |
ls | -0.011 |
Run | -0.011 |
mun | -0.010 |
P | -0.009 |
Hey | -0.009 |
Token | had |
Token ID | 550 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
ine | +0.044 |
birthday | +0.043 |
hip | +0.042 |
rap | +0.042 |
Office | +0.041 |
Neg logit contributions | |
ls | -0.033 |
Run | -0.032 |
mun | -0.029 |
P | -0.028 |
Hey | -0.027 |
Token | two |
Token ID | 734 |
Feature activation | +5.35e-03 |
Pos logit contributions | |
ine | +0.065 |
birthday | +0.060 |
Office | +0.057 |
hip | +0.057 |
rap | +0.057 |
Neg logit contributions | |
Run | -0.063 |
ls | -0.062 |
mun | -0.059 |
bitten | -0.056 |
P | -0.056 |
Token | wheels |
Token ID | 13666 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
ine | +0.008 |
birthday | +0.007 |
hip | +0.007 |
Office | +0.007 |
Jupiter | +0.007 |
Neg logit contributions | |
ls | -0.010 |
Run | -0.010 |
mun | -0.010 |
P | -0.009 |
bitten | -0.009 |
Token | ay |
Token ID | 323 |
Feature activation | +9.78e-03 |
Pos logit contributions | |
Office | +0.041 |
birthday | +0.038 |
ine | +0.038 |
X | +0.037 |
Jupiter | +0.037 |
Neg logit contributions | |
ls | -0.078 |
P | -0.073 |
awed | -0.069 |
mun | -0.068 |
led | -0.068 |
Token | ons |
Token ID | 684 |
Feature activation | +1.86e-03 |
Pos logit contributions | |
ine | +0.024 |
birthday | +0.024 |
Office | +0.023 |
X | +0.023 |
hip | +0.023 |
Neg logit contributions | |
ls | -0.008 |
mun | -0.007 |
Run | -0.007 |
P | -0.006 |
led | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +6.03e-03 |
Pos logit contributions | |
ine | +0.003 |
birthday | +0.003 |
Office | +0.003 |
hip | +0.003 |
X | +0.003 |
Neg logit contributions | |
ls | -0.003 |
Run | -0.003 |
mun | -0.003 |
P | -0.002 |
led | -0.002 |
Token | She |
Token ID | 1375 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
ine | +0.011 |
Jupiter | +0.009 |
Office | +0.009 |
birthday | +0.009 |
hip | +0.009 |
Neg logit contributions | |
Run | -0.009 |
mun | -0.009 |
ls | -0.009 |
Hey | -0.008 |
P | -0.008 |
Token | writes |
Token ID | 6797 |
Feature activation | +3.28e-02 |
Pos logit contributions | |
hip | +0.029 |
birthday | +0.029 |
ine | +0.028 |
Office | +0.027 |
rap | +0.026 |
Neg logit contributions | |
ls | -0.031 |
mun | -0.030 |
Run | -0.028 |
P | -0.027 |
distracted | -0.026 |
Token | words |
Token ID | 2456 |
Feature activation | +5.31e-02 |
Pos logit contributions | |
ine | +0.061 |
sing | +0.059 |
eye | +0.058 |
basis | +0.057 |
hip | +0.057 |
Neg logit contributions | |
ls | -0.051 |
led | -0.042 |
mun | -0.041 |
Hey | -0.037 |
distracted | -0.036 |
Token | and |
Token ID | 290 |
Feature activation | +2.73e-02 |
Pos logit contributions | |
ine | +0.100 |
sing | +0.100 |
Office | +0.098 |
birthday | +0.098 |
urger | +0.096 |
Neg logit contributions | |
ls | -0.070 |
mun | -0.057 |
led | -0.056 |
Hey | -0.053 |
teased | -0.052 |
Token | draws |
Token ID | 14293 |
Feature activation | +3.11e-02 |
Pos logit contributions | |
ine | +0.037 |
Office | +0.035 |
rap | +0.035 |
hip | +0.035 |
birthday | +0.033 |
Neg logit contributions | |
ls | -0.049 |
mun | -0.049 |
Run | -0.049 |
teased | -0.044 |
bitten | -0.044 |
Token | pictures |
Token ID | 5986 |
Feature activation | +3.97e-02 |
Pos logit contributions | |
sing | +0.057 |
ine | +0.055 |
hip | +0.055 |
Office | +0.054 |
otive | +0.053 |
Neg logit contributions | |
ls | -0.047 |
mun | -0.040 |
led | -0.038 |
P | -0.037 |
Run | -0.033 |
Token | on |
Token ID | 319 |
Feature activation | +2.71e-02 |
Pos logit contributions | |
ine | +0.084 |
sing | +0.081 |
Office | +0.080 |
urger | +0.079 |
birthday | +0.077 |
Neg logit contributions | |
ls | -0.049 |
mun | -0.039 |
Run | -0.034 |
led | -0.033 |
Hey | -0.032 |
Token | paper |
Token ID | 3348 |
Feature activation | +3.05e-02 |
Pos logit contributions | |
sing | +0.054 |
birthday | +0.053 |
Office | +0.052 |
ine | +0.052 |
hip | +0.052 |
Neg logit contributions | |
ls | -0.034 |
mun | -0.027 |
Run | -0.026 |
P | -0.026 |
Hey | -0.026 |
Token | in |
Token ID | 287 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
ls | +0.007 |
Run | +0.007 |
mun | +0.007 |
led | +0.006 |
bitten | +0.006 |
Neg logit contributions | |
birthday | -0.005 |
ine | -0.004 |
pattern | -0.004 |
X | -0.004 |
Office | -0.004 |
Token | her |
Token ID | 607 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
ine | +0.033 |
birthday | +0.031 |
hip | +0.030 |
Office | +0.030 |
X | +0.029 |
Neg logit contributions | |
Run | -0.029 |
ls | -0.028 |
mun | -0.027 |
P | -0.025 |
Hey | -0.025 |
Token | bedroom |
Token ID | 14043 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
ine | +0.018 |
hip | +0.016 |
Office | +0.015 |
birthday | +0.015 |
Jupiter | +0.015 |
Neg logit contributions | |
ls | -0.030 |
Run | -0.028 |
mun | -0.028 |
P | -0.027 |
bitten | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | +2.47e-02 |
Pos logit contributions | |
ine | +0.037 |
birthday | +0.035 |
Office | +0.032 |
hip | +0.032 |
X | +0.031 |
Neg logit contributions | |
Run | -0.041 |
ls | -0.041 |
mun | -0.040 |
P | -0.036 |
Hey | -0.035 |
Token | It |
Token ID | 632 |
Feature activation | +3.08e-02 |
Pos logit contributions | |
ine | +0.051 |
urger | +0.046 |
rap | +0.045 |
hip | +0.045 |
birthday | +0.044 |
Neg logit contributions | |
Run | -0.030 |
ls | -0.027 |
mun | -0.026 |
Hey | -0.026 |
bitten | -0.024 |
Token | was |
Token ID | 373 |
Feature activation | +5.31e-02 |
Pos logit contributions | |
ine | +0.061 |
birthday | +0.059 |
hip | +0.057 |
Office | +0.055 |
X | +0.055 |
Neg logit contributions | |
ls | -0.042 |
Run | -0.041 |
mun | -0.040 |
P | -0.036 |
Hey | -0.035 |
Token | a |
Token ID | 257 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
hip | +0.104 |
ine | +0.102 |
urger | +0.098 |
Office | +0.098 |
uries | +0.097 |
Neg logit contributions | |
Run | -0.065 |
bitten | -0.063 |
P | -0.063 |
ls | -0.061 |
distracted | -0.060 |
Token | very |
Token ID | 845 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
ine | +0.036 |
hip | +0.035 |
uries | +0.033 |
rap | +0.033 |
artments | +0.031 |
Neg logit contributions | |
Run | -0.026 |
ls | -0.025 |
P | -0.025 |
mun | -0.022 |
Hey | -0.021 |
Token | messy |
Token ID | 24097 |
Feature activation | +1.88e-02 |
Pos logit contributions | |
ine | +0.024 |
hip | +0.022 |
Office | +0.021 |
birthday | +0.021 |
Jupiter | +0.020 |
Neg logit contributions | |
ls | -0.026 |
Run | -0.024 |
mun | -0.023 |
P | -0.022 |
bitten | -0.021 |
Token | place |
Token ID | 1295 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
ine | +0.027 |
hip | +0.026 |
uries | +0.026 |
rap | +0.025 |
Jupiter | +0.025 |
Neg logit contributions | |
ls | -0.032 |
Run | -0.032 |
mun | -0.030 |
P | -0.030 |
Hey | -0.029 |
Token | , |
Token ID | 11 |
Feature activation | +8.06e-03 |
Pos logit contributions | |
ine | +0.023 |
hip | +0.022 |
Office | +0.021 |
urger | +0.021 |
Jupiter | +0.021 |
Neg logit contributions | |
ls | -0.020 |
Run | -0.019 |
mun | -0.018 |
bitten | -0.017 |
P | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | -9.61e-03 |
Pos logit contributions | |
mun | +0.026 |
ls | +0.026 |
Run | +0.025 |
P | +0.024 |
led | +0.024 |
Neg logit contributions | |
ine | -0.018 |
birthday | -0.018 |
Grand | -0.017 |
desk | -0.017 |
Today | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
Run | +0.017 |
ls | +0.017 |
mun | +0.016 |
led | +0.015 |
P | +0.015 |
Neg logit contributions | |
ine | -0.015 |
birthday | -0.014 |
hip | -0.013 |
X | -0.013 |
Office | -0.013 |
Token | Tom |
Token ID | 13787 |
Feature activation | -9.87e-03 |
Pos logit contributions | |
ls | +0.024 |
Run | +0.023 |
mun | +0.023 |
led | +0.020 |
bitten | +0.020 |
Neg logit contributions | |
ine | -0.025 |
birthday | -0.023 |
hip | -0.022 |
Office | -0.022 |
X | -0.022 |
Token | felt |
Token ID | 2936 |
Feature activation | -3.36e-02 |
Pos logit contributions | |
ls | +0.013 |
Run | +0.013 |
mun | +0.012 |
P | +0.011 |
Hey | +0.011 |
Neg logit contributions | |
ine | -0.020 |
birthday | -0.018 |
Office | -0.018 |
hip | -0.018 |
X | -0.018 |
Token | angry |
Token ID | 7954 |
Feature activation | -2.26e-02 |
Pos logit contributions | |
Run | +0.050 |
mun | +0.049 |
ls | +0.049 |
P | +0.047 |
Hey | +0.042 |
Neg logit contributions | |
ine | -0.058 |
birthday | -0.055 |
X | -0.053 |
eye | -0.053 |
end | -0.052 |
Token | and |
Token ID | 290 |
Feature activation | -5.78e-02 |
Pos logit contributions | |
ls | +0.036 |
Run | +0.036 |
mun | +0.034 |
P | +0.030 |
bitten | +0.030 |
Neg logit contributions | |
ine | -0.040 |
birthday | -0.037 |
Office | -0.036 |
hip | -0.035 |
Jupiter | -0.035 |
Token | sad |
Token ID | 6507 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
P | +0.108 |
mun | +0.105 |
ries | +0.101 |
ls | +0.096 |
Run | +0.095 |
Neg logit contributions | |
for | -0.072 |
be | -0.068 |
eye | -0.066 |
sing | -0.065 |
end | -0.065 |
Token | . |
Token ID | 13 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
ls | +0.023 |
Run | +0.023 |
mun | +0.022 |
P | +0.019 |
led | +0.019 |
Neg logit contributions | |
ine | -0.029 |
birthday | -0.026 |
Office | -0.026 |
hip | -0.025 |
Jupiter | -0.025 |
Token | He |
Token ID | 679 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
Run | +0.037 |
ls | +0.036 |
mun | +0.035 |
P | +0.031 |
Hey | +0.031 |
Neg logit contributions | |
ine | -0.045 |
birthday | -0.041 |
hip | -0.039 |
Office | -0.039 |
Jupiter | -0.039 |
Token | tried |
Token ID | 3088 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
ls | +0.037 |
mun | +0.036 |
Run | +0.036 |
P | +0.032 |
aro | +0.031 |
Neg logit contributions | |
ine | -0.043 |
X | -0.036 |
Office | -0.035 |
pattern | -0.035 |
birthday | -0.034 |
Token | to |
Token ID | 284 |
Feature activation | -3.20e-02 |
Pos logit contributions | |
Run | +0.021 |
ls | +0.021 |
mun | +0.020 |
P | +0.019 |
led | +0.018 |
Neg logit contributions | |
ine | -0.020 |
birthday | -0.019 |
hip | -0.017 |
Office | -0.017 |
X | -0.017 |
Token | burned |
Token ID | 11544 |
Feature activation | -1.75e-03 |
Pos logit contributions | |
ine | +0.003 |
birthday | +0.003 |
Office | +0.003 |
X | +0.003 |
eye | +0.003 |
Neg logit contributions | |
Run | -0.002 |
mun | -0.002 |
ls | -0.002 |
P | -0.002 |
Hey | -0.002 |
Token | his |
Token ID | 465 |
Feature activation | +3.36e-04 |
Pos logit contributions | |
ls | +0.003 |
Run | +0.003 |
mun | +0.002 |
Hey | +0.002 |
led | +0.002 |
Neg logit contributions | |
ine | -0.003 |
birthday | -0.003 |
hip | -0.003 |
Office | -0.003 |
Jupiter | -0.003 |
Token | hand |
Token ID | 1021 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
ine | +0.000 |
hip | +0.000 |
Jupiter | +0.000 |
rap | +0.000 |
Office | +0.000 |
Neg logit contributions | |
ls | -0.001 |
Run | -0.001 |
mun | -0.001 |
bitten | -0.001 |
led | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | -2.71e-02 |
Pos logit contributions | |
ls | +0.021 |
Run | +0.020 |
mun | +0.019 |
bitten | +0.018 |
led | +0.018 |
Neg logit contributions | |
ine | -0.019 |
birthday | -0.019 |
Jupiter | -0.017 |
Office | -0.017 |
hip | -0.017 |
Token | made |
Token ID | 925 |
Feature activation | +2.86e-03 |
Pos logit contributions | |
ls | +0.041 |
Run | +0.041 |
mun | +0.041 |
P | +0.037 |
Hey | +0.035 |
Neg logit contributions | |
ine | -0.049 |
birthday | -0.046 |
X | -0.043 |
Office | -0.043 |
Jupiter | -0.042 |
Token | him |
Token ID | 683 |
Feature activation | -5.64e-02 |
Pos logit contributions | |
ine | +0.005 |
hip | +0.005 |
Office | +0.005 |
birthday | +0.005 |
uries | +0.005 |
Neg logit contributions | |
ls | -0.004 |
Run | -0.004 |
bitten | -0.004 |
distracted | -0.003 |
led | -0.003 |
Token | cry |
Token ID | 3960 |
Feature activation | -4.00e-03 |
Pos logit contributions | |
Run | +0.095 |
ls | +0.093 |
mun | +0.093 |
P | +0.087 |
Hey | +0.086 |
Neg logit contributions | |
ine | -0.096 |
birthday | -0.093 |
eye | -0.085 |
desk | -0.082 |
hip | -0.081 |
Token | . |
Token ID | 13 |
Feature activation | -2.02e-02 |
Pos logit contributions | |
ls | +0.004 |
Run | +0.003 |
mun | +0.003 |
led | +0.003 |
distracted | +0.003 |
Neg logit contributions | |
ine | -0.009 |
X | -0.009 |
Office | -0.009 |
rap | -0.009 |
artments | -0.009 |
Token | His |
Token ID | 2399 |
Feature activation | -2.68e-03 |
Pos logit contributions | |
ls | +0.036 |
Run | +0.036 |
mun | +0.034 |
P | +0.031 |
Hey | +0.031 |
Neg logit contributions | |
ine | -0.032 |
birthday | -0.030 |
Office | -0.028 |
hip | -0.028 |
X | -0.028 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
ls | +0.005 |
Run | +0.005 |
mun | +0.005 |
bitten | +0.005 |
distracted | +0.005 |
Neg logit contributions | |
ine | -0.003 |
hip | -0.003 |
Office | -0.003 |
rap | -0.003 |
Jupiter | -0.003 |
Token | heard |
Token ID | 2982 |
Feature activation | +5.03e-04 |
Pos logit contributions | |
ls | +0.018 |
Run | +0.018 |
mun | +0.017 |
P | +0.015 |
Hey | +0.015 |
Neg logit contributions | |
ine | -0.023 |
birthday | -0.022 |
Office | -0.021 |
hip | -0.020 |
X | -0.020 |
Token | so |
Token ID | 523 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
ls | +0.013 |
Run | +0.013 |
mun | +0.013 |
P | +0.012 |
Hey | +0.011 |
Neg logit contributions | |
ine | -0.011 |
birthday | -0.010 |
Office | -0.009 |
hip | -0.009 |
Jupiter | -0.009 |
Token | disappointed |
Token ID | 11679 |
Feature activation | -9.16e-03 |
Pos logit contributions | |
ls | +0.063 |
Run | +0.062 |
mun | +0.062 |
P | +0.057 |
Hey | +0.055 |
Neg logit contributions | |
ine | -0.045 |
birthday | -0.044 |
believe | -0.039 |
eye | -0.039 |
end | -0.038 |
Token | and |
Token ID | 290 |
Feature activation | -4.09e-02 |
Pos logit contributions | |
ls | +0.017 |
Run | +0.016 |
mun | +0.016 |
P | +0.014 |
Hey | +0.014 |
Neg logit contributions | |
ine | -0.014 |
birthday | -0.013 |
Office | -0.012 |
hip | -0.012 |
X | -0.012 |
Token | he |
Token ID | 339 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
mun | +0.062 |
ls | +0.062 |
P | +0.059 |
Run | +0.059 |
awed | +0.052 |
Neg logit contributions | |
ine | -0.073 |
birthday | -0.065 |
Office | -0.063 |
sleep | -0.063 |
X | -0.063 |
Token | never |
Token ID | 1239 |
Feature activation | -3.73e-03 |
Pos logit contributions | |
ls | +0.021 |
Run | +0.020 |
mun | +0.020 |
P | +0.018 |
Hey | +0.017 |
Neg logit contributions | |
ine | -0.028 |
birthday | -0.024 |
Office | -0.024 |
X | -0.024 |
Jupiter | -0.023 |
Token | got |
Token ID | 1392 |
Feature activation | -5.56e-02 |
Pos logit contributions | |
Run | +0.005 |
ls | +0.005 |
mun | +0.005 |
bitten | +0.004 |
P | +0.004 |
Neg logit contributions | |
ine | -0.008 |
birthday | -0.007 |
hip | -0.007 |
Office | -0.007 |
Jupiter | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
Run | +0.138 |
mun | +0.130 |
ls | +0.124 |
Sorry | +0.118 |
Hey | +0.118 |
Neg logit contributions | |
ine | -0.050 |
for | -0.050 |
birthday | -0.048 |
sleep | -0.041 |
three | -0.040 |
Token | special |
Token ID | 2041 |
Feature activation | +1.27e-02 |
Pos logit contributions | |
Run | +0.024 |
ls | +0.024 |
mun | +0.024 |
P | +0.022 |
Hey | +0.022 |
Neg logit contributions | |
ine | -0.010 |
birthday | -0.010 |
Office | -0.008 |
X | -0.008 |
hip | -0.008 |
Token | gift |
Token ID | 6979 |
Feature activation | -6.30e-03 |
Pos logit contributions | |
ine | +0.013 |
birthday | +0.011 |
hip | +0.010 |
Office | +0.010 |
Jupiter | +0.010 |
Neg logit contributions | |
ls | -0.030 |
Run | -0.030 |
mun | -0.029 |
P | -0.027 |
Hey | -0.027 |
Token | from |
Token ID | 422 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
ls | +0.010 |
Run | +0.010 |
mun | +0.009 |
bitten | +0.008 |
P | +0.008 |
Neg logit contributions | |
ine | -0.011 |
birthday | -0.010 |
Office | -0.010 |
Jupiter | -0.010 |
hip | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | -2.77e-03 |
Pos logit contributions | |
ls | +0.023 |
Run | +0.022 |
mun | +0.021 |
bitten | +0.018 |
Hey | +0.018 |
Neg logit contributions | |
ine | -0.032 |
hip | -0.029 |
birthday | -0.029 |
Office | -0.028 |
X | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
mun | +0.048 |
Run | +0.046 |
ls | +0.044 |
P | +0.044 |
Hey | +0.044 |
Neg logit contributions | |
ine | -0.027 |
birthday | -0.024 |
sleep | -0.024 |
X | -0.023 |
end | -0.023 |
Token | Tim |
Token ID | 5045 |
Feature activation | -3.09e-02 |
Pos logit contributions | |
ls | +0.045 |
mun | +0.045 |
Run | +0.044 |
led | +0.040 |
P | +0.040 |
Neg logit contributions | |
ine | -0.028 |
birthday | -0.026 |
Jupiter | -0.025 |
X | -0.025 |
hip | -0.024 |
Token | and |
Token ID | 290 |
Feature activation | +3.25e-03 |
Pos logit contributions | |
ls | +0.050 |
mun | +0.049 |
Run | +0.049 |
P | +0.046 |
av | +0.045 |
Neg logit contributions | |
ine | -0.052 |
eye | -0.045 |
end | -0.043 |
X | -0.041 |
aunt | -0.040 |
Token | Sam |
Token ID | 3409 |
Feature activation | -3.71e-02 |
Pos logit contributions | |
ine | +0.005 |
birthday | +0.004 |
Office | +0.004 |
hip | +0.004 |
X | +0.004 |
Neg logit contributions | |
ls | -0.006 |
Run | -0.006 |
mun | -0.006 |
P | -0.006 |
Hey | -0.006 |
Token | were |
Token ID | 547 |
Feature activation | -2.61e-02 |
Pos logit contributions | |
ls | +0.058 |
mun | +0.055 |
P | +0.055 |
Run | +0.054 |
av | +0.053 |
Neg logit contributions | |
ine | -0.058 |
end | -0.057 |
eye | -0.055 |
sleep | -0.051 |
X | -0.050 |
Token | both |
Token ID | 1111 |
Feature activation | -5.51e-02 |
Pos logit contributions | |
mun | +0.043 |
Run | +0.043 |
ls | +0.041 |
P | +0.039 |
Hey | +0.038 |
Neg logit contributions | |
ine | -0.042 |
birthday | -0.040 |
eye | -0.038 |
end | -0.038 |
sing | -0.037 |
Token | hurt |
Token ID | 5938 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
Run | +0.092 |
ls | +0.090 |
P | +0.087 |
mun | +0.085 |
Hey | +0.085 |
Neg logit contributions | |
birthday | -0.091 |
eye | -0.087 |
sleep | -0.085 |
ine | -0.084 |
sing | -0.082 |
Token | and |
Token ID | 290 |
Feature activation | -4.15e-02 |
Pos logit contributions | |
mun | +0.054 |
ls | +0.054 |
Run | +0.054 |
P | +0.051 |
Hey | +0.049 |
Neg logit contributions | |
ine | -0.026 |
birthday | -0.023 |
eye | -0.021 |
end | -0.021 |
hip | -0.021 |
Token | scared |
Token ID | 12008 |
Feature activation | -2.02e-02 |
Pos logit contributions | |
mun | +0.061 |
P | +0.060 |
Run | +0.059 |
ls | +0.059 |
Hey | +0.055 |
Neg logit contributions | |
ine | -0.072 |
birthday | -0.071 |
eye | -0.068 |
sleep | -0.067 |
end | -0.066 |
Token | . |
Token ID | 13 |
Feature activation | -2.66e-02 |
Pos logit contributions | |
ls | +0.037 |
Run | +0.036 |
mun | +0.036 |
P | +0.034 |
bitten | +0.032 |
Neg logit contributions | |
ine | -0.030 |
birthday | -0.029 |
hip | -0.026 |
X | -0.026 |
Jupiter | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | +1.83e-03 |
Pos logit contributions | |
ls | +0.056 |
mun | +0.055 |
Run | +0.054 |
led | +0.050 |
P | +0.050 |
Neg logit contributions | |
ine | -0.033 |
birthday | -0.030 |
Jupiter | -0.029 |
Office | -0.028 |
Today | -0.028 |
Token | � |
Token ID | 129 |
Feature activation | -9.10e-03 |
Pos logit contributions | |
Run | +0.036 |
ls | +0.035 |
mun | +0.034 |
P | +0.031 |
Hey | +0.030 |
Neg logit contributions | |
ine | -0.039 |
birthday | -0.037 |
Office | -0.034 |
hip | -0.034 |
X | -0.033 |
Token | � |
Token ID | 241 |
Feature activation | -3.67e-02 |
Pos logit contributions | |
Run | +0.015 |
ls | +0.015 |
mun | +0.014 |
distracted | +0.013 |
P | +0.013 |
Neg logit contributions | |
birthday | -0.015 |
ine | -0.013 |
moon | -0.013 |
X | -0.013 |
astronaut | -0.013 |
Token | What |
Token ID | 2061 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
ls | +0.056 |
mun | +0.052 |
Run | +0.049 |
led | +0.046 |
P | +0.044 |
Neg logit contributions | |
ine | -0.070 |
birthday | -0.066 |
Office | -0.060 |
Today | -0.060 |
hip | -0.060 |
Token | â |
Token ID | 22940 |
Feature activation | -2.12e-02 |
Pos logit contributions | |
Run | +0.038 |
ls | +0.037 |
mun | +0.035 |
awed | +0.034 |
P | +0.034 |
Neg logit contributions | |
birthday | -0.023 |
ine | -0.022 |
desk | -0.019 |
Office | -0.019 |
hip | -0.018 |
Token | € |
Token ID | 26391 |
Feature activation | -2.60e-02 |
Pos logit contributions | |
ls | +0.041 |
Run | +0.039 |
mun | +0.038 |
bitten | +0.035 |
distracted | +0.034 |
Neg logit contributions | |
ine | -0.031 |
birthday | -0.029 |
Office | -0.028 |
hip | -0.027 |
X | -0.026 |
Token | ™ |
Token ID | 8151 |
Feature activation | -5.47e-02 |
Pos logit contributions | |
Run | +0.045 |
ls | +0.044 |
mun | +0.043 |
P | +0.038 |
Hey | +0.038 |
Neg logit contributions | |
ine | -0.043 |
birthday | -0.040 |
Office | -0.037 |
hip | -0.036 |
Jupiter | -0.036 |
Token | s |
Token ID | 82 |
Feature activation | -3.21e-03 |
Pos logit contributions | |
Run | +0.096 |
mun | +0.084 |
bitten | +0.079 |
Help | +0.078 |
Sorry | +0.078 |
Neg logit contributions | |
ine | -0.086 |
Today | -0.078 |
birthday | -0.075 |
for | -0.071 |
desk | -0.070 |
Token | wrong |
Token ID | 2642 |
Feature activation | -2.54e-03 |
Pos logit contributions | |
Run | +0.005 |
mun | +0.004 |
ls | +0.004 |
Hey | +0.004 |
P | +0.004 |
Neg logit contributions | |
ine | -0.006 |
birthday | -0.006 |
Office | -0.006 |
hip | -0.006 |
Jupiter | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
ls | +0.004 |
Run | +0.004 |
mun | +0.004 |
P | +0.003 |
Hey | +0.003 |
Neg logit contributions | |
ine | -0.005 |
birthday | -0.004 |
hip | -0.004 |
X | -0.004 |
Jupiter | -0.004 |
Token | Jane |
Token ID | 12091 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
ls | +0.023 |
Run | +0.023 |
mun | +0.023 |
P | +0.021 |
bitten | +0.020 |
Neg logit contributions | |
ine | -0.018 |
birthday | -0.017 |
Office | -0.015 |
X | -0.015 |
hip | -0.015 |
Token | ? |
Token ID | 30 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
Run | +0.025 |
ls | +0.024 |
mun | +0.024 |
P | +0.022 |
Hey | +0.021 |
Neg logit contributions | |
ine | -0.023 |
birthday | -0.021 |
Office | -0.020 |
hip | -0.019 |
Jupiter | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
ine | +0.003 |
birthday | +0.002 |
Office | +0.002 |
hip | +0.002 |
Jupiter | +0.002 |
Neg logit contributions | |
ls | -0.003 |
Run | -0.003 |
mun | -0.003 |
P | -0.003 |
led | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +4.50e-03 |
Pos logit contributions | |
ine | +0.027 |
birthday | +0.025 |
hip | +0.024 |
Office | +0.024 |
X | +0.024 |
Neg logit contributions | |
ls | -0.022 |
Run | -0.022 |
mun | -0.021 |
P | -0.019 |
Hey | -0.018 |
Token | The |
Token ID | 464 |
Feature activation | -2.38e-03 |
Pos logit contributions | |
ine | +0.007 |
birthday | +0.007 |
X | +0.006 |
eye | +0.006 |
hip | +0.006 |
Neg logit contributions | |
Run | -0.008 |
ls | -0.008 |
mun | -0.008 |
P | -0.007 |
Hey | -0.007 |
Token | two |
Token ID | 734 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
Run | +0.003 |
ls | +0.003 |
mun | +0.003 |
Hey | +0.003 |
P | +0.003 |
Neg logit contributions | |
ine | -0.005 |
hip | -0.005 |
artments | -0.004 |
uries | -0.004 |
rap | -0.004 |
Token | became |
Token ID | 2627 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
Run | +0.031 |
ls | +0.030 |
mun | +0.028 |
P | +0.026 |
bitten | +0.026 |
Neg logit contributions | |
ine | -0.033 |
birthday | -0.032 |
X | -0.030 |
hip | -0.029 |
Office | -0.029 |
Token | insepar |
Token ID | 50140 |
Feature activation | -5.46e-02 |
Pos logit contributions | |
Run | +0.023 |
ls | +0.022 |
mun | +0.021 |
awed | +0.021 |
led | +0.020 |
Neg logit contributions | |
ine | -0.014 |
birthday | -0.014 |
Jupiter | -0.012 |
Office | -0.012 |
X | -0.012 |
Token | able |
Token ID | 540 |
Feature activation | -2.08e-02 |
Pos logit contributions | |
Run | +0.070 |
bitten | +0.058 |
ls | +0.055 |
mun | +0.055 |
awed | +0.053 |
Neg logit contributions | |
ine | -0.109 |
birthday | -0.101 |
end | -0.101 |
eye | -0.097 |
hip | -0.097 |
Token | and |
Token ID | 290 |
Feature activation | -6.79e-03 |
Pos logit contributions | |
Run | +0.038 |
ls | +0.037 |
mun | +0.036 |
P | +0.034 |
Hey | +0.033 |
Neg logit contributions | |
ine | -0.032 |
birthday | -0.031 |
X | -0.028 |
Office | -0.028 |
hip | -0.027 |
Token | each |
Token ID | 1123 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
Run | +0.011 |
ls | +0.011 |
mun | +0.010 |
P | +0.009 |
led | +0.009 |
Neg logit contributions | |
ine | -0.012 |
birthday | -0.011 |
hip | -0.011 |
Office | -0.011 |
rap | -0.011 |
Token | day |
Token ID | 1110 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
ls | +0.030 |
Run | +0.030 |
mun | +0.029 |
P | +0.026 |
awed | +0.025 |
Neg logit contributions | |
ine | -0.030 |
birthday | -0.027 |
Office | -0.026 |
hip | -0.026 |
X | -0.026 |
Token | they |
Token ID | 484 |
Feature activation | -2.53e-03 |
Pos logit contributions | |
Run | +0.018 |
ls | +0.018 |
mun | +0.017 |
P | +0.016 |
Hey | +0.016 |
Neg logit contributions | |
ine | -0.019 |
birthday | -0.018 |
Office | -0.017 |
Jupiter | -0.017 |
hip | -0.016 |
Token | scissors |
Token ID | 41786 |
Feature activation | -2.77e-02 |
Pos logit contributions | |
Run | +0.005 |
ls | +0.005 |
mun | +0.004 |
P | +0.004 |
Hey | +0.004 |
Neg logit contributions | |
ine | -0.004 |
birthday | -0.004 |
hip | -0.004 |
Office | -0.004 |
Jupiter | -0.004 |
Token | slipped |
Token ID | 18859 |
Feature activation | -8.58e-03 |
Pos logit contributions | |
Run | +0.044 |
ls | +0.042 |
mun | +0.041 |
P | +0.040 |
Hey | +0.037 |
Neg logit contributions | |
ine | -0.050 |
birthday | -0.044 |
hip | -0.044 |
desk | -0.042 |
pattern | -0.042 |
Token | , |
Token ID | 11 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
ls | +0.014 |
Run | +0.014 |
mun | +0.014 |
P | +0.013 |
Hey | +0.012 |
Neg logit contributions | |
ine | -0.015 |
birthday | -0.013 |
hip | -0.013 |
X | -0.013 |
Office | -0.012 |
Token | giving |
Token ID | 3501 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
ls | +0.052 |
Run | +0.050 |
mun | +0.050 |
P | +0.048 |
Hey | +0.045 |
Neg logit contributions | |
ine | -0.031 |
birthday | -0.026 |
X | -0.025 |
hip | -0.025 |
desk | -0.024 |
Token | the |
Token ID | 262 |
Feature activation | -7.05e-03 |
Pos logit contributions | |
Run | +0.024 |
ls | +0.024 |
mun | +0.023 |
P | +0.021 |
Hey | +0.020 |
Neg logit contributions | |
ine | -0.020 |
birthday | -0.018 |
X | -0.017 |
hip | -0.017 |
Office | -0.017 |
Token | boy |
Token ID | 2933 |
Feature activation | -5.46e-02 |
Pos logit contributions | |
Run | +0.012 |
ls | +0.012 |
mun | +0.011 |
P | +0.010 |
Hey | +0.010 |
Neg logit contributions | |
ine | -0.012 |
birthday | -0.011 |
hip | -0.011 |
Office | -0.011 |
Jupiter | -0.010 |
Token | a |
Token ID | 257 |
Feature activation | -2.49e-02 |
Pos logit contributions | |
Run | +0.109 |
ls | +0.108 |
mun | +0.104 |
Hey | +0.095 |
P | +0.095 |
Neg logit contributions | |
ine | -0.077 |
birthday | -0.070 |
Office | -0.064 |
hip | -0.063 |
X | -0.062 |
Token | bad |
Token ID | 2089 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
Run | +0.048 |
ls | +0.048 |
mun | +0.046 |
P | +0.042 |
led | +0.041 |
Neg logit contributions | |
ine | -0.036 |
birthday | -0.034 |
X | -0.030 |
Office | -0.030 |
hip | -0.030 |
Token | cut |
Token ID | 2005 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
Run | +0.019 |
ls | +0.019 |
mun | +0.018 |
bitten | +0.017 |
P | +0.017 |
Neg logit contributions | |
ine | -0.014 |
Jupiter | -0.013 |
Office | -0.013 |
birthday | -0.013 |
hip | -0.013 |
Token | on |
Token ID | 319 |
Feature activation | -8.34e-03 |
Pos logit contributions | |
ls | +0.021 |
Run | +0.021 |
mun | +0.020 |
P | +0.018 |
bitten | +0.018 |
Neg logit contributions | |
ine | -0.025 |
birthday | -0.024 |
Office | -0.023 |
Jupiter | -0.023 |
hip | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | +3.81e-03 |
Pos logit contributions | |
Run | +0.011 |
ls | +0.010 |
mun | +0.010 |
bitten | +0.009 |
Hey | +0.009 |
Neg logit contributions | |
ine | -0.016 |
birthday | -0.016 |
Jupiter | -0.015 |
Office | -0.015 |
hip | -0.015 |
Token | got |
Token ID | 1392 |
Feature activation | -4.12e-02 |
Pos logit contributions | |
ls | +0.033 |
mun | +0.031 |
Run | +0.031 |
P | +0.027 |
aro | +0.026 |
Neg logit contributions | |
ine | -0.033 |
sleep | -0.030 |
X | -0.030 |
birthday | -0.030 |
eye | -0.030 |
Token | hurt |
Token ID | 5938 |
Feature activation | -7.16e-03 |
Pos logit contributions | |
mun | +0.081 |
Run | +0.076 |
ls | +0.070 |
� | +0.069 |
awed | +0.068 |
Neg logit contributions | |
for | -0.054 |
birthday | -0.052 |
ine | -0.051 |
sleep | -0.049 |
Grand | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | -5.13e-03 |
Pos logit contributions | |
mun | +0.014 |
ls | +0.014 |
Run | +0.014 |
P | +0.012 |
awed | +0.012 |
Neg logit contributions | |
ine | -0.009 |
birthday | -0.009 |
hip | -0.008 |
end | -0.008 |
X | -0.008 |
Token | You |
Token ID | 921 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
ls | +0.009 |
mun | +0.009 |
Run | +0.009 |
led | +0.008 |
P | +0.008 |
Neg logit contributions | |
ine | -0.008 |
birthday | -0.008 |
Office | -0.007 |
hip | -0.007 |
X | -0.007 |
Token | made |
Token ID | 925 |
Feature activation | +3.86e-03 |
Pos logit contributions | |
ls | +0.028 |
Run | +0.027 |
mun | +0.027 |
awed | +0.026 |
P | +0.026 |
Neg logit contributions | |
ine | -0.015 |
eye | -0.013 |
believe | -0.013 |
sleep | -0.013 |
imagine | -0.013 |
Token | us |
Token ID | 514 |
Feature activation | -5.45e-02 |
Pos logit contributions | |
ine | +0.006 |
birthday | +0.006 |
Office | +0.006 |
hip | +0.006 |
X | +0.006 |
Neg logit contributions | |
Run | -0.007 |
ls | -0.007 |
mun | -0.006 |
Hey | -0.006 |
bitten | -0.006 |
Token | worry |
Token ID | 5490 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
Run | +0.097 |
ls | +0.094 |
mun | +0.093 |
P | +0.088 |
Hey | +0.087 |
Neg logit contributions | |
birthday | -0.087 |
ine | -0.085 |
hip | -0.076 |
eye | -0.075 |
end | -0.073 |
Token | ." |
Token ID | 526 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
ls | +0.017 |
Run | +0.016 |
mun | +0.015 |
P | +0.014 |
Hey | +0.014 |
Neg logit contributions | |
ine | -0.018 |
birthday | -0.017 |
X | -0.016 |
Office | -0.016 |
hip | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | +4.18e-04 |
Pos logit contributions | |
mun | +0.028 |
ls | +0.028 |
Run | +0.028 |
led | +0.027 |
bitten | +0.026 |
Neg logit contributions | |
ine | -0.019 |
birthday | -0.018 |
Today | -0.017 |
X | -0.016 |
Jupiter | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | -7.81e-03 |
Pos logit contributions | |
ine | +0.001 |
birthday | +0.001 |
hip | +0.001 |
Office | +0.001 |
X | +0.001 |
Neg logit contributions | |
Run | -0.001 |
ls | -0.001 |
mun | -0.001 |
P | -0.001 |
Hey | -0.001 |
Token | " |
Token ID | 1 |
Feature activation | -3.43e-02 |
Pos logit contributions | |
ls | +0.013 |
mun | +0.012 |
Run | +0.012 |
led | +0.011 |
bitten | +0.011 |
Neg logit contributions | |
ine | -0.013 |
hip | -0.012 |
birthday | -0.012 |
Office | -0.011 |
X | -0.011 |
Token | 't |
Token ID | 470 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
Run | +0.013 |
ls | +0.013 |
mun | +0.013 |
led | +0.011 |
P | +0.011 |
Neg logit contributions | |
ine | -0.020 |
birthday | -0.018 |
hip | -0.018 |
Office | -0.017 |
rap | -0.017 |
Token | leave |
Token ID | 2666 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
ls | +0.024 |
Run | +0.023 |
mun | +0.023 |
led | +0.021 |
P | +0.021 |
Neg logit contributions | |
ine | -0.012 |
birthday | -0.011 |
hip | -0.011 |
X | -0.010 |
Office | -0.010 |
Token | our |
Token ID | 674 |
Feature activation | -2.60e-03 |
Pos logit contributions | |
Run | +0.029 |
ls | +0.029 |
mun | +0.028 |
P | +0.025 |
Hey | +0.025 |
Neg logit contributions | |
ine | -0.028 |
birthday | -0.027 |
hip | -0.025 |
Office | -0.025 |
X | -0.025 |
Token | homes |
Token ID | 5682 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
ls | +0.005 |
Run | +0.005 |
mun | +0.005 |
P | +0.004 |
bitten | +0.004 |
Neg logit contributions | |
ine | -0.004 |
birthday | -0.004 |
hip | -0.004 |
Office | -0.003 |
Jupiter | -0.003 |
Token | for |
Token ID | 329 |
Feature activation | -6.86e-03 |
Pos logit contributions | |
ls | +0.031 |
Run | +0.030 |
mun | +0.029 |
P | +0.027 |
bitten | +0.026 |
Neg logit contributions | |
ine | -0.028 |
birthday | -0.026 |
Office | -0.025 |
hip | -0.025 |
X | -0.025 |
Token | anyone |
Token ID | 2687 |
Feature activation | -5.43e-02 |
Pos logit contributions | |
Run | +0.011 |
ls | +0.011 |
mun | +0.010 |
P | +0.009 |
bitten | +0.009 |
Neg logit contributions | |
ine | -0.012 |
birthday | -0.011 |
Office | -0.011 |
hip | -0.011 |
X | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
Run | +0.108 |
mun | +0.101 |
P | +0.098 |
ls | +0.098 |
Hey | +0.094 |
Neg logit contributions | |
birthday | -0.077 |
ine | -0.073 |
hip | -0.064 |
sleep | -0.063 |
believe | -0.063 |
Token | not |
Token ID | 407 |
Feature activation | +6.88e-03 |
Pos logit contributions | |
ls | +0.023 |
Run | +0.023 |
mun | +0.022 |
P | +0.020 |
Hey | +0.020 |
Neg logit contributions | |
ine | -0.019 |
birthday | -0.017 |
hip | -0.016 |
Office | -0.016 |
X | -0.016 |
Token | even |
Token ID | 772 |
Feature activation | -2.28e-02 |
Pos logit contributions | |
ine | +0.010 |
birthday | +0.009 |
hip | +0.008 |
Office | +0.008 |
X | +0.008 |
Neg logit contributions | |
Run | -0.013 |
ls | -0.013 |
mun | -0.013 |
P | -0.012 |
Hey | -0.012 |
Token | a |
Token ID | 257 |
Feature activation | -2.32e-03 |
Pos logit contributions | |
Run | +0.042 |
ls | +0.041 |
mun | +0.040 |
P | +0.038 |
awed | +0.037 |
Neg logit contributions | |
ine | -0.033 |
birthday | -0.033 |
X | -0.029 |
hip | -0.029 |
Office | -0.029 |
Token | mighty |
Token ID | 18680 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
Run | +0.004 |
ls | +0.004 |
mun | +0.004 |
P | +0.004 |
Hey | +0.003 |
Neg logit contributions | |
ine | -0.004 |
birthday | -0.003 |
hip | -0.003 |
Office | -0.003 |
X | -0.003 |
Token | it |
Token ID | 340 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
ls | +0.027 |
Run | +0.026 |
mun | +0.025 |
bitten | +0.023 |
P | +0.023 |
Neg logit contributions | |
ine | -0.030 |
birthday | -0.027 |
Office | -0.027 |
hip | -0.026 |
Jupiter | -0.026 |
Token | , |
Token ID | 11 |
Feature activation | -2.24e-02 |
Pos logit contributions | |
Run | +0.027 |
ls | +0.027 |
mun | +0.026 |
P | +0.024 |
Hey | +0.024 |
Neg logit contributions | |
ine | -0.023 |
birthday | -0.022 |
hip | -0.020 |
Office | -0.020 |
X | -0.020 |
Token | so |
Token ID | 523 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
ls | +0.041 |
Run | +0.041 |
mun | +0.040 |
P | +0.036 |
Hey | +0.035 |
Neg logit contributions | |
ine | -0.034 |
birthday | -0.031 |
hip | -0.029 |
Office | -0.029 |
X | -0.029 |
Token | Jill |
Token ID | 22772 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
Run | +0.020 |
ls | +0.020 |
mun | +0.019 |
P | +0.017 |
Hey | +0.016 |
Neg logit contributions | |
ine | -0.036 |
birthday | -0.035 |
hip | -0.033 |
X | -0.033 |
Office | -0.032 |
Token | got |
Token ID | 1392 |
Feature activation | -4.89e-02 |
Pos logit contributions | |
Run | +0.034 |
ls | +0.032 |
mun | +0.032 |
P | +0.028 |
Hey | +0.028 |
Neg logit contributions | |
ine | -0.029 |
birthday | -0.022 |
Office | -0.022 |
desk | -0.022 |
eye | -0.022 |
Token | very |
Token ID | 845 |
Feature activation | -5.37e-02 |
Pos logit contributions | |
Run | +0.103 |
mun | +0.098 |
ls | +0.092 |
awed | +0.087 |
Hey | +0.087 |
Neg logit contributions | |
for | -0.062 |
birthday | -0.058 |
ine | -0.058 |
end | -0.055 |
eye | -0.053 |
Token | mad |
Token ID | 8805 |
Feature activation | -2.71e-02 |
Pos logit contributions | |
Run | +0.098 |
mun | +0.094 |
Hey | +0.092 |
P | +0.081 |
rob | +0.080 |
Neg logit contributions | |
birthday | -0.081 |
eye | -0.074 |
ine | -0.074 |
for | -0.073 |
end | -0.070 |
Token | . |
Token ID | 13 |
Feature activation | -2.17e-02 |
Pos logit contributions | |
Run | +0.050 |
ls | +0.050 |
mun | +0.048 |
P | +0.044 |
Hey | +0.043 |
Neg logit contributions | |
ine | -0.041 |
birthday | -0.038 |
hip | -0.035 |
Office | -0.035 |
X | -0.035 |
Token | |
Token ID | 220 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
ls | +0.040 |
Run | +0.040 |
mun | +0.038 |
P | +0.035 |
bitten | +0.034 |
Neg logit contributions | |
ine | -0.033 |
birthday | -0.031 |
hip | -0.029 |
Office | -0.029 |
Jupiter | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
Run | +0.041 |
ls | +0.040 |
mun | +0.038 |
P | +0.035 |
led | +0.035 |
Neg logit contributions | |
ine | -0.032 |
birthday | -0.029 |
hip | -0.028 |
Office | -0.027 |
rap | -0.027 |
Token | Her |
Token ID | 9360 |
Feature activation | -4.30e-03 |
Pos logit contributions | |
ls | +0.025 |
Run | +0.024 |
mun | +0.023 |
bitten | +0.021 |
led | +0.021 |
Neg logit contributions | |
ine | -0.024 |
birthday | -0.022 |
hip | -0.021 |
Office | -0.020 |
Jupiter | -0.020 |
Token | then |
Token ID | 788 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
ine | +0.024 |
hip | +0.022 |
X | +0.022 |
Office | +0.022 |
birthday | +0.022 |
Neg logit contributions | |
Run | -0.021 |
ls | -0.021 |
mun | -0.020 |
distracted | -0.018 |
Hey | -0.018 |
Token | Lily |
Token ID | 20037 |
Feature activation | +3.05e-03 |
Pos logit contributions | |
ls | +0.016 |
Run | +0.015 |
mun | +0.015 |
distracted | +0.013 |
P | +0.013 |
Neg logit contributions | |
ine | -0.018 |
hip | -0.018 |
birthday | -0.017 |
Office | -0.017 |
X | -0.017 |
Token | 's |
Token ID | 338 |
Feature activation | +2.20e-02 |
Pos logit contributions | |
ine | +0.005 |
birthday | +0.005 |
hip | +0.005 |
Office | +0.005 |
X | +0.005 |
Neg logit contributions | |
ls | -0.005 |
Run | -0.005 |
mun | -0.004 |
P | -0.004 |
bitten | -0.004 |
Token | grandma |
Token ID | 49890 |
Feature activation | +9.26e-04 |
Pos logit contributions | |
ine | +0.028 |
Jupiter | +0.024 |
rap | +0.024 |
hip | +0.023 |
Office | +0.022 |
Neg logit contributions | |
ls | -0.043 |
Run | -0.043 |
mun | -0.042 |
bitten | -0.040 |
distracted | -0.040 |
Token | got |
Token ID | 1392 |
Feature activation | -3.84e-02 |
Pos logit contributions | |
ine | +0.002 |
birthday | +0.002 |
hip | +0.002 |
X | +0.002 |
Office | +0.002 |
Neg logit contributions | |
ls | -0.001 |
Run | -0.001 |
mun | -0.001 |
P | -0.001 |
bitten | -0.001 |
Token | very |
Token ID | 845 |
Feature activation | -5.37e-02 |
Pos logit contributions | |
Run | +0.077 |
ls | +0.072 |
mun | +0.067 |
Hey | +0.065 |
awed | +0.064 |
Neg logit contributions | |
ine | -0.056 |
birthday | -0.051 |
for | -0.047 |
end | -0.045 |
eye | -0.044 |
Token | tired |
Token ID | 10032 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
Run | +0.095 |
mun | +0.087 |
Hey | +0.087 |
aro | +0.080 |
awed | +0.078 |
Neg logit contributions | |
birthday | -0.090 |
ine | -0.085 |
eye | -0.078 |
sleep | -0.074 |
for | -0.072 |
Token | and |
Token ID | 290 |
Feature activation | -3.58e-02 |
Pos logit contributions | |
ls | +0.018 |
Run | +0.018 |
mun | +0.017 |
P | +0.015 |
bitten | +0.015 |
Neg logit contributions | |
ine | -0.019 |
birthday | -0.018 |
Office | -0.018 |
Jupiter | -0.017 |
X | -0.017 |
Token | fell |
Token ID | 3214 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
Run | +0.056 |
ls | +0.055 |
mun | +0.054 |
P | +0.052 |
Hey | +0.049 |
Neg logit contributions | |
ine | -0.062 |
birthday | -0.061 |
sleep | -0.055 |
desk | -0.054 |
Jupiter | -0.053 |
Token | asleep |
Token ID | 16039 |
Feature activation | -7.10e-03 |
Pos logit contributions | |
Run | +0.036 |
mun | +0.035 |
ls | +0.034 |
P | +0.032 |
Hey | +0.031 |
Neg logit contributions | |
ine | -0.028 |
birthday | -0.026 |
Jupiter | -0.024 |
hip | -0.023 |
moon | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | +3.65e-03 |
Pos logit contributions | |
Run | +0.012 |
ls | +0.012 |
mun | +0.011 |
P | +0.010 |
Hey | +0.010 |
Neg logit contributions | |
ine | -0.012 |
birthday | -0.011 |
Office | -0.010 |
hip | -0.010 |
X | -0.010 |
Token | so |
Token ID | 523 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
ls | +0.016 |
Run | +0.016 |
mun | +0.015 |
P | +0.014 |
Hey | +0.013 |
Neg logit contributions | |
ine | -0.017 |
birthday | -0.016 |
Office | -0.015 |
hip | -0.015 |
X | -0.015 |
Token | sad |
Token ID | 6507 |
Feature activation | -6.55e-03 |
Pos logit contributions | |
Run | +0.043 |
ls | +0.042 |
mun | +0.042 |
P | +0.037 |
Hey | +0.036 |
Neg logit contributions | |
ine | -0.049 |
birthday | -0.048 |
eye | -0.042 |
hip | -0.042 |
end | -0.041 |
Token | that |
Token ID | 326 |
Feature activation | +1.23e-03 |
Pos logit contributions | |
ls | +0.012 |
Run | +0.012 |
mun | +0.011 |
P | +0.010 |
Hey | +0.010 |
Neg logit contributions | |
ine | -0.010 |
birthday | -0.009 |
Office | -0.009 |
hip | -0.009 |
X | -0.009 |
Token | she |
Token ID | 673 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
ine | +0.003 |
hip | +0.003 |
X | +0.003 |
Office | +0.003 |
rap | +0.003 |
Neg logit contributions | |
ls | -0.001 |
Run | -0.001 |
mun | -0.001 |
distracted | -0.001 |
Hey | -0.001 |
Token | never |
Token ID | 1239 |
Feature activation | -3.26e-03 |
Pos logit contributions | |
Run | +0.027 |
ls | +0.027 |
mun | +0.027 |
P | +0.024 |
Hey | +0.023 |
Neg logit contributions | |
ine | -0.028 |
birthday | -0.023 |
Office | -0.023 |
desk | -0.022 |
eye | -0.022 |
Token | got |
Token ID | 1392 |
Feature activation | -5.36e-02 |
Pos logit contributions | |
ls | +0.005 |
Run | +0.005 |
mun | +0.004 |
P | +0.004 |
Hey | +0.004 |
Neg logit contributions | |
ine | -0.006 |
birthday | -0.006 |
hip | -0.006 |
Office | -0.006 |
X | -0.005 |
Token | her |
Token ID | 607 |
Feature activation | -1.41e-03 |
Pos logit contributions | |
Run | +0.124 |
mun | +0.115 |
ls | +0.105 |
Sorry | +0.103 |
Hey | +0.103 |
Neg logit contributions | |
ine | -0.064 |
birthday | -0.057 |
for | -0.054 |
sleep | -0.049 |
aunt | -0.049 |
Token | skirt |
Token ID | 23967 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
Run | +0.003 |
ls | +0.003 |
mun | +0.003 |
P | +0.003 |
Hey | +0.003 |
Neg logit contributions | |
ine | -0.001 |
birthday | -0.001 |
hip | -0.001 |
Office | -0.001 |
X | -0.001 |
Token | back |
Token ID | 736 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
Run | +0.042 |
ls | +0.042 |
mun | +0.041 |
P | +0.038 |
Hey | +0.037 |
Neg logit contributions | |
ine | -0.031 |
birthday | -0.029 |
hip | -0.027 |
Office | -0.027 |
X | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -2.21e-03 |
Pos logit contributions | |
Run | +0.035 |
mun | +0.033 |
ls | +0.033 |
P | +0.031 |
Hey | +0.030 |
Neg logit contributions | |
ine | -0.025 |
birthday | -0.023 |
hip | -0.021 |
eye | -0.020 |
desk | -0.020 |
Token | Lily |
Token ID | 20037 |
Feature activation | -9.38e-03 |
Pos logit contributions | |
Run | +0.003 |
ls | +0.003 |
mun | +0.003 |
Hey | +0.003 |
bitten | +0.003 |
Neg logit contributions | |
ine | -0.004 |
X | -0.003 |
hip | -0.003 |
birthday | -0.003 |
rap | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | -6.77e-03 |
Pos logit contributions | |
Run | +0.022 |
ls | +0.021 |
mun | +0.021 |
P | +0.019 |
awed | +0.018 |
Neg logit contributions | |
ine | -0.017 |
birthday | -0.016 |
X | -0.016 |
hip | -0.015 |
Office | -0.015 |
Token | toy |
Token ID | 13373 |
Feature activation | -7.93e-03 |
Pos logit contributions | |
Run | +0.016 |
mun | +0.016 |
ls | +0.016 |
P | +0.015 |
awed | +0.014 |
Neg logit contributions | |
birthday | -0.007 |
ine | -0.006 |
desk | -0.006 |
eye | -0.006 |
X | -0.006 |
Token | away |
Token ID | 1497 |
Feature activation | -2.74e-03 |
Pos logit contributions | |
Run | +0.016 |
mun | +0.016 |
ls | +0.015 |
P | +0.014 |
Hey | +0.014 |
Neg logit contributions | |
ine | -0.010 |
birthday | -0.010 |
X | -0.009 |
hip | -0.009 |
Office | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
Run | +0.005 |
ls | +0.005 |
mun | +0.005 |
P | +0.005 |
Hey | +0.005 |
Neg logit contributions | |
ine | -0.004 |
birthday | -0.003 |
hip | -0.003 |
Office | -0.003 |
X | -0.003 |
Token | made |
Token ID | 925 |
Feature activation | -4.41e-04 |
Pos logit contributions | |
ls | +0.045 |
Run | +0.044 |
mun | +0.043 |
P | +0.041 |
Hey | +0.038 |
Neg logit contributions | |
ine | -0.040 |
birthday | -0.037 |
desk | -0.036 |
X | -0.036 |
Office | -0.035 |
Token | him |
Token ID | 683 |
Feature activation | -5.36e-02 |
Pos logit contributions | |
Run | +0.001 |
ls | +0.001 |
mun | +0.001 |
P | +0.001 |
Hey | +0.001 |
Neg logit contributions | |
ine | -0.001 |
birthday | -0.001 |
hip | -0.001 |
Office | -0.001 |
X | -0.001 |
Token | go |
Token ID | 467 |
Feature activation | -1.29e-02 |
Pos logit contributions | |
Run | +0.126 |
ls | +0.121 |
mun | +0.120 |
Hey | +0.117 |
P | +0.117 |
Neg logit contributions | |
birthday | -0.062 |
ine | -0.056 |
sleep | -0.053 |
eye | -0.053 |
desk | -0.050 |
Token | back |
Token ID | 736 |
Feature activation | -8.89e-03 |
Pos logit contributions | |
Run | +0.031 |
mun | +0.029 |
ls | +0.029 |
Hey | +0.028 |
P | +0.027 |
Neg logit contributions | |
ine | -0.014 |
birthday | -0.013 |
desk | -0.011 |
eye | -0.011 |
hip | -0.011 |
Token | to |
Token ID | 284 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
Run | +0.020 |
ls | +0.020 |
mun | +0.019 |
P | +0.018 |
Hey | +0.018 |
Neg logit contributions | |
ine | -0.010 |
birthday | -0.009 |
Office | -0.008 |
hip | -0.008 |
Jupiter | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | -9.07e-03 |
Pos logit contributions | |
Run | +0.027 |
ls | +0.027 |
mun | +0.026 |
bitten | +0.024 |
P | +0.023 |
Neg logit contributions | |
ine | -0.032 |
birthday | -0.029 |
hip | -0.028 |
Office | -0.028 |
X | -0.027 |
Token | store |
Token ID | 3650 |
Feature activation | -8.18e-03 |
Pos logit contributions | |
ls | +0.017 |
Run | +0.017 |
mun | +0.016 |
P | +0.015 |
Hey | +0.014 |
Neg logit contributions | |
ine | -0.014 |
birthday | -0.013 |
hip | -0.012 |
Office | -0.012 |
X | -0.012 |
Token | day |
Token ID | 1110 |
Feature activation | -9.36e-03 |
Pos logit contributions | |
ine | +0.017 |
hip | +0.015 |
birthday | +0.015 |
Jupiter | +0.014 |
Office | +0.014 |
Neg logit contributions | |
Run | -0.029 |
ls | -0.028 |
mun | -0.027 |
P | -0.026 |
Hey | -0.025 |
Token | , |
Token ID | 11 |
Feature activation | -7.00e-03 |
Pos logit contributions | |
Run | +0.019 |
ls | +0.019 |
mun | +0.018 |
P | +0.017 |
Hey | +0.017 |
Neg logit contributions | |
ine | -0.012 |
birthday | -0.011 |
Office | -0.010 |
hip | -0.010 |
X | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | -4.21e-03 |
Pos logit contributions | |
ls | +0.014 |
Run | +0.013 |
mun | +0.013 |
P | +0.012 |
bitten | +0.012 |
Neg logit contributions | |
birthday | -0.010 |
ine | -0.009 |
desk | -0.008 |
moon | -0.008 |
Jupiter | -0.008 |
Token | puppy |
Token ID | 26188 |
Feature activation | -1.95e-02 |
Pos logit contributions | |
ls | +0.008 |
Run | +0.008 |
mun | +0.008 |
P | +0.007 |
bitten | +0.007 |
Neg logit contributions | |
birthday | -0.006 |
ine | -0.006 |
hip | -0.006 |
X | -0.005 |
Office | -0.005 |
Token | got |
Token ID | 1392 |
Feature activation | -4.63e-02 |
Pos logit contributions | |
Run | +0.032 |
mun | +0.031 |
ls | +0.031 |
aro | +0.029 |
P | +0.028 |
Neg logit contributions | |
ine | -0.033 |
eye | -0.027 |
Office | -0.027 |
desk | -0.027 |
end | -0.027 |
Token | very |
Token ID | 845 |
Feature activation | -5.36e-02 |
Pos logit contributions | |
Run | +0.087 |
mun | +0.086 |
ls | +0.075 |
Hey | +0.075 |
Sorry | +0.072 |
Neg logit contributions | |
ine | -0.068 |
for | -0.062 |
end | -0.058 |
eye | -0.056 |
even | -0.055 |
Token | sick |
Token ID | 6639 |
Feature activation | -2.07e-02 |
Pos logit contributions | |
mun | +0.104 |
Run | +0.099 |
Hey | +0.095 |
� | +0.083 |
rob | +0.081 |
Neg logit contributions | |
birthday | -0.075 |
ine | -0.073 |
eye | -0.070 |
sleep | -0.066 |
for | -0.064 |
Token | and |
Token ID | 290 |
Feature activation | -4.09e-02 |
Pos logit contributions | |
Run | +0.038 |
ls | +0.038 |
mun | +0.037 |
P | +0.034 |
Hey | +0.033 |
Neg logit contributions | |
ine | -0.031 |
birthday | -0.029 |
hip | -0.027 |
Office | -0.027 |
X | -0.027 |
Token | had |
Token ID | 550 |
Feature activation | -3.61e-02 |
Pos logit contributions | |
mun | +0.071 |
Run | +0.070 |
P | +0.066 |
ls | +0.064 |
Hey | +0.060 |
Neg logit contributions | |
ine | -0.064 |
birthday | -0.058 |
sleep | -0.055 |
desk | -0.055 |
pattern | -0.054 |
Token | to |
Token ID | 284 |
Feature activation | -2.91e-02 |
Pos logit contributions | |
Run | +0.079 |
mun | +0.078 |
ls | +0.077 |
led | +0.069 |
P | +0.068 |
Neg logit contributions | |
ine | -0.044 |
birthday | -0.044 |
moon | -0.035 |
pattern | -0.035 |
desk | -0.034 |
Token | go |
Token ID | 467 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
Run | +0.062 |
ls | +0.061 |
mun | +0.059 |
P | +0.056 |
bitten | +0.055 |
Neg logit contributions | |
ine | -0.034 |
birthday | -0.030 |
Office | -0.029 |
hip | -0.028 |
Jupiter | -0.028 |
Token | Unfortunately |
Token ID | 13898 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
ls | +0.005 |
Run | +0.005 |
mun | +0.004 |
P | +0.004 |
bitten | +0.004 |
Neg logit contributions | |
ine | -0.005 |
birthday | -0.005 |
hip | -0.005 |
Office | -0.005 |
X | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | -5.17e-03 |
Pos logit contributions | |
ine | +0.025 |
birthday | +0.024 |
hip | +0.022 |
Office | +0.022 |
X | +0.021 |
Neg logit contributions | |
ls | -0.030 |
Run | -0.030 |
mun | -0.028 |
led | -0.026 |
P | -0.025 |
Token | the |
Token ID | 262 |
Feature activation | -2.46e-04 |
Pos logit contributions | |
ls | +0.009 |
Run | +0.009 |
mun | +0.009 |
P | +0.008 |
Hey | +0.008 |
Neg logit contributions | |
ine | -0.008 |
birthday | -0.008 |
Office | -0.007 |
Jupiter | -0.007 |
X | -0.007 |
Token | infant |
Token ID | 11212 |
Feature activation | -3.13e-03 |
Pos logit contributions | |
ls | +0.000 |
Run | +0.000 |
mun | +0.000 |
P | +0.000 |
awed | +0.000 |
Neg logit contributions | |
birthday | -0.000 |
moon | -0.000 |
ine | -0.000 |
end | -0.000 |
pattern | -0.000 |
Token | never |
Token ID | 1239 |
Feature activation | -7.96e-03 |
Pos logit contributions | |
ls | +0.005 |
Run | +0.005 |
mun | +0.005 |
P | +0.004 |
Hey | +0.004 |
Neg logit contributions | |
ine | -0.005 |
birthday | -0.005 |
Office | -0.005 |
hip | -0.005 |
X | -0.005 |
Token | got |
Token ID | 1392 |
Feature activation | -5.35e-02 |
Pos logit contributions | |
ls | +0.012 |
Run | +0.012 |
mun | +0.012 |
P | +0.011 |
Hey | +0.010 |
Neg logit contributions | |
ine | -0.014 |
birthday | -0.013 |
eye | -0.012 |
hip | -0.012 |
Office | -0.012 |
Token | to |
Token ID | 284 |
Feature activation | -3.30e-02 |
Pos logit contributions | |
Run | +0.125 |
mun | +0.122 |
Hey | +0.110 |
P | +0.108 |
Sorry | +0.106 |
Neg logit contributions | |
for | -0.056 |
ine | -0.056 |
birthday | -0.053 |
sleep | -0.048 |
end | -0.044 |
Token | see |
Token ID | 766 |
Feature activation | -9.99e-03 |
Pos logit contributions | |
ls | +0.068 |
Run | +0.067 |
mun | +0.065 |
P | +0.061 |
led | +0.060 |
Neg logit contributions | |
ine | -0.044 |
birthday | -0.042 |
X | -0.037 |
hip | -0.037 |
Jupiter | -0.037 |
Token | his |
Token ID | 465 |
Feature activation | -4.07e-03 |
Pos logit contributions | |
Run | +0.018 |
ls | +0.018 |
mun | +0.017 |
P | +0.016 |
Hey | +0.015 |
Neg logit contributions | |
ine | -0.016 |
birthday | -0.015 |
Office | -0.014 |
hip | -0.014 |
Jupiter | -0.014 |
Token | friend |
Token ID | 1545 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
Run | +0.009 |
ls | +0.009 |
mun | +0.008 |
P | +0.008 |
Hey | +0.008 |
Neg logit contributions | |
ine | -0.005 |
birthday | -0.005 |
Office | -0.004 |
hip | -0.004 |
X | -0.004 |
Token | again |
Token ID | 757 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
ls | +0.026 |
Run | +0.026 |
mun | +0.025 |
bitten | +0.022 |
Hey | +0.022 |
Neg logit contributions | |
ine | -0.027 |
birthday | -0.025 |
Office | -0.024 |
hip | -0.024 |
X | -0.024 |
Token | and |
Token ID | 290 |
Feature activation | -2.50e-02 |
Pos logit contributions | |
ine | +0.004 |
X | +0.004 |
Office | +0.004 |
hip | +0.004 |
Jupiter | +0.004 |
Neg logit contributions | |
ls | -0.003 |
Run | -0.002 |
mun | -0.002 |
led | -0.002 |
bitten | -0.002 |
Token | gave |
Token ID | 2921 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
ls | +0.048 |
mun | +0.047 |
Run | +0.047 |
P | +0.044 |
awed | +0.043 |
Neg logit contributions | |
birthday | -0.034 |
for | -0.033 |
eye | -0.031 |
be | -0.030 |
end | -0.030 |
Token | back |
Token ID | 736 |
Feature activation | -2.24e-02 |
Pos logit contributions | |
Run | +0.026 |
ls | +0.025 |
mun | +0.025 |
P | +0.022 |
awed | +0.022 |
Neg logit contributions | |
ine | -0.015 |
birthday | -0.015 |
end | -0.012 |
hip | -0.012 |
Office | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | -4.21e-04 |
Pos logit contributions | |
Run | +0.047 |
mun | +0.044 |
ls | +0.044 |
awed | +0.040 |
led | +0.040 |
Neg logit contributions | |
ine | -0.029 |
birthday | -0.028 |
Office | -0.025 |
Jupiter | -0.025 |
X | -0.024 |
Token | shiny |
Token ID | 22441 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
ls | +0.001 |
Run | +0.001 |
mun | +0.001 |
P | +0.001 |
Hey | +0.001 |
Neg logit contributions | |
ine | -0.001 |
birthday | -0.001 |
hip | -0.001 |
Office | -0.001 |
X | -0.001 |
Token | things |
Token ID | 1243 |
Feature activation | -5.35e-02 |
Pos logit contributions | |
ine | +0.019 |
hip | +0.017 |
birthday | +0.017 |
Office | +0.016 |
Jupiter | +0.016 |
Neg logit contributions | |
ls | -0.021 |
Run | -0.019 |
mun | -0.019 |
led | -0.018 |
P | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
mun | +0.126 |
Run | +0.120 |
ls | +0.110 |
awed | +0.109 |
P | +0.109 |
Neg logit contributions | |
birthday | -0.060 |
for | -0.055 |
end | -0.052 |
hip | -0.047 |
eye | -0.046 |
Token |
|
Token ID | 198 |
Feature activation | +8.11e-03 |
Pos logit contributions | |
ls | +0.024 |
Run | +0.023 |
mun | +0.023 |
P | +0.021 |
led | +0.021 |
Neg logit contributions | |
ine | -0.017 |
birthday | -0.016 |
Office | -0.014 |
Jupiter | -0.014 |
hip | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | -5.32e-03 |
Pos logit contributions | |
ine | +0.017 |
birthday | +0.016 |
Office | +0.015 |
hip | +0.015 |
X | +0.015 |
Neg logit contributions | |
ls | -0.010 |
Run | -0.010 |
mun | -0.009 |
P | -0.009 |
bitten | -0.008 |
Token | They |
Token ID | 2990 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
ls | +0.009 |
Run | +0.009 |
mun | +0.008 |
P | +0.007 |
bitten | +0.007 |
Neg logit contributions | |
ine | -0.009 |
birthday | -0.009 |
Office | -0.008 |
hip | -0.008 |
X | -0.008 |
Token | hugged |
Token ID | 41130 |
Feature activation | -6.96e-03 |
Pos logit contributions | |
ls | +0.019 |
Run | +0.019 |
mun | +0.018 |
aro | +0.016 |
bitten | +0.016 |
Neg logit contributions | |
ine | -0.017 |
eye | -0.015 |
birthday | -0.015 |
end | -0.015 |
X | -0.015 |
Token | me |
Token ID | 502 |
Feature activation | -7.49e-03 |
Pos logit contributions | |
ine | +0.003 |
birthday | +0.002 |
hip | +0.002 |
Office | +0.002 |
X | +0.002 |
Neg logit contributions | |
Run | -0.002 |
ls | -0.002 |
mun | -0.002 |
P | -0.002 |
Hey | -0.002 |
Token | ! |
Token ID | 0 |
Feature activation | +1.23e-03 |
Pos logit contributions | |
ls | +0.011 |
Run | +0.010 |
mun | +0.010 |
P | +0.009 |
bitten | +0.009 |
Neg logit contributions | |
ine | -0.013 |
birthday | -0.013 |
X | -0.013 |
Jupiter | -0.012 |
Office | -0.012 |
Token | Aren |
Token ID | 9843 |
Feature activation | +6.95e-03 |
Pos logit contributions | |
ine | +0.002 |
birthday | +0.002 |
hip | +0.002 |
Office | +0.002 |
X | +0.002 |
Neg logit contributions | |
Run | -0.002 |
ls | -0.002 |
mun | -0.002 |
P | -0.002 |
Hey | -0.002 |
Token | â |
Token ID | 22940 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
ine | +0.013 |
birthday | +0.013 |
hip | +0.012 |
Office | +0.012 |
rap | +0.012 |
Neg logit contributions | |
ls | -0.010 |
Run | -0.010 |
mun | -0.009 |
led | -0.008 |
bitten | -0.008 |
Token | € |
Token ID | 26391 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
ls | +0.039 |
Run | +0.037 |
mun | +0.036 |
bitten | +0.033 |
awed | +0.032 |
Neg logit contributions | |
ine | -0.035 |
birthday | -0.032 |
hip | -0.031 |
Office | -0.030 |
X | -0.029 |
Token | ™ |
Token ID | 8151 |
Feature activation | -5.32e-02 |
Pos logit contributions | |
ls | +0.036 |
Run | +0.036 |
mun | +0.034 |
P | +0.031 |
Hey | +0.030 |
Neg logit contributions | |
ine | -0.039 |
birthday | -0.036 |
hip | -0.034 |
Office | -0.034 |
X | -0.034 |
Token | t |
Token ID | 83 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
Run | +0.082 |
ls | +0.072 |
mun | +0.072 |
P | +0.067 |
bitten | +0.065 |
Neg logit contributions | |
ine | -0.096 |
birthday | -0.089 |
hip | -0.084 |
Today | -0.081 |
ary | -0.077 |
Token | we |
Token ID | 356 |
Feature activation | -2.76e-02 |
Pos logit contributions | |
Run | +0.027 |
ls | +0.027 |
mun | +0.026 |
P | +0.024 |
awed | +0.023 |
Neg logit contributions | |
ine | -0.024 |
birthday | -0.023 |
hip | -0.021 |
Office | -0.021 |
Jupiter | -0.020 |
Token | both |
Token ID | 1111 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
Run | +0.057 |
ls | +0.056 |
mun | +0.054 |
awed | +0.050 |
P | +0.049 |
Neg logit contributions | |
ine | -0.037 |
birthday | -0.036 |
hip | -0.032 |
Office | -0.030 |
X | -0.030 |
Token | so |
Token ID | 523 |
Feature activation | -3.93e-02 |
Pos logit contributions | |
ls | +0.052 |
Run | +0.052 |
mun | +0.050 |
P | +0.046 |
Hey | +0.045 |
Neg logit contributions | |
ine | -0.035 |
birthday | -0.032 |
Office | -0.029 |
hip | -0.029 |
X | -0.029 |
Token | excited |
Token ID | 6568 |
Feature activation | -9.96e-03 |
Pos logit contributions | |
Run | +0.075 |
ls | +0.073 |
mun | +0.073 |
P | +0.066 |
bitten | +0.066 |
Neg logit contributions | |
birthday | -0.057 |
ine | -0.055 |
hip | -0.047 |
Office | -0.046 |
X | -0.046 |
Token | . |
Token ID | 13 |
Feature activation | -3.32e-03 |
Pos logit contributions | |
Run | +0.014 |
mun | +0.014 |
ls | +0.013 |
P | +0.013 |
Hey | +0.013 |
Neg logit contributions | |
ine | -0.011 |
birthday | -0.010 |
eye | -0.010 |
X | -0.009 |
pattern | -0.009 |
Token | It |
Token ID | 632 |
Feature activation | +3.65e-03 |
Pos logit contributions | |
Run | +0.006 |
ls | +0.006 |
mun | +0.006 |
P | +0.005 |
Hey | +0.005 |
Neg logit contributions | |
ine | -0.005 |
birthday | -0.005 |
Office | -0.004 |
hip | -0.004 |
X | -0.004 |
Token | hurt |
Token ID | 5938 |
Feature activation | -3.58e-03 |
Pos logit contributions | |
ine | +0.006 |
pattern | +0.005 |
birthday | +0.005 |
Jupiter | +0.005 |
moon | +0.005 |
Neg logit contributions | |
Run | -0.007 |
ls | -0.006 |
mun | -0.006 |
P | -0.006 |
Hey | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | -2.89e-02 |
Pos logit contributions | |
ls | +0.006 |
Run | +0.006 |
mun | +0.006 |
P | +0.006 |
Hey | +0.005 |
Neg logit contributions | |
ine | -0.006 |
birthday | -0.005 |
Office | -0.005 |
hip | -0.005 |
X | -0.005 |
Token | made |
Token ID | 925 |
Feature activation | -3.68e-04 |
Pos logit contributions | |
mun | +0.046 |
Run | +0.045 |
ls | +0.044 |
P | +0.042 |
Hey | +0.040 |
Neg logit contributions | |
ine | -0.050 |
birthday | -0.045 |
X | -0.043 |
pattern | -0.043 |
Jupiter | -0.042 |
Token | him |
Token ID | 683 |
Feature activation | -5.31e-02 |
Pos logit contributions | |
ls | +0.001 |
Run | +0.001 |
mun | +0.001 |
P | +0.001 |
bitten | +0.001 |
Neg logit contributions | |
ine | -0.001 |
birthday | -0.001 |
Office | -0.001 |
hip | -0.001 |
X | -0.001 |
Token | sad |
Token ID | 6507 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
Run | +0.106 |
mun | +0.104 |
ls | +0.102 |
Hey | +0.100 |
P | +0.099 |
Neg logit contributions | |
birthday | -0.074 |
ine | -0.073 |
eye | -0.069 |
sleep | -0.066 |
pattern | -0.063 |
Token | . |
Token ID | 13 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
ls | +0.027 |
Run | +0.026 |
mun | +0.024 |
bitten | +0.022 |
led | +0.021 |
Neg logit contributions | |
ine | -0.032 |
birthday | -0.030 |
Office | -0.029 |
hip | -0.029 |
X | -0.029 |
Token | D |
Token ID | 360 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
Run | +0.021 |
ls | +0.021 |
mun | +0.021 |
Hey | +0.018 |
led | +0.018 |
Neg logit contributions | |
ine | -0.016 |
birthday | -0.015 |
Today | -0.014 |
Jupiter | -0.013 |
Office | -0.013 |
Token | ucky |
Token ID | 5309 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
Run | +0.032 |
ls | +0.030 |
mun | +0.030 |
aro | +0.027 |
bitten | +0.027 |
Neg logit contributions | |
ine | -0.024 |
birthday | -0.022 |
eye | -0.021 |
hip | -0.021 |
pattern | -0.020 |
Token | started |
Token ID | 2067 |
Feature activation | -2.92e-02 |
Pos logit contributions | |
Run | +0.025 |
ls | +0.024 |
mun | +0.023 |
P | +0.022 |
Hey | +0.022 |
Neg logit contributions | |
ine | -0.022 |
end | -0.018 |
eye | -0.018 |
birthday | -0.017 |
pattern | -0.017 |
Token | but |
Token ID | 475 |
Feature activation | +3.47e-03 |
Pos logit contributions | |
ls | +0.049 |
Run | +0.049 |
mun | +0.047 |
P | +0.043 |
Hey | +0.042 |
Neg logit contributions | |
ine | -0.042 |
birthday | -0.039 |
hip | -0.036 |
Office | -0.036 |
X | -0.036 |
Token | Buddy |
Token ID | 36896 |
Feature activation | -2.62e-02 |
Pos logit contributions | |
ine | +0.006 |
birthday | +0.006 |
Jupiter | +0.006 |
X | +0.006 |
hip | +0.006 |
Neg logit contributions | |
ls | -0.005 |
Run | -0.005 |
mun | -0.005 |
P | -0.005 |
awed | -0.004 |
Token | just |
Token ID | 655 |
Feature activation | -3.02e-03 |
Pos logit contributions | |
Run | +0.042 |
ls | +0.042 |
mun | +0.040 |
P | +0.037 |
Hey | +0.036 |
Neg logit contributions | |
ine | -0.047 |
birthday | -0.041 |
Office | -0.040 |
hip | -0.039 |
X | -0.038 |
Token | loved |
Token ID | 6151 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
Run | +0.004 |
ls | +0.004 |
mun | +0.004 |
led | +0.003 |
bitten | +0.003 |
Neg logit contributions | |
ine | -0.006 |
birthday | -0.006 |
hip | -0.005 |
Office | -0.005 |
Jupiter | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | -2.38e-02 |
Pos logit contributions | |
ls | +0.017 |
Run | +0.017 |
mun | +0.016 |
P | +0.015 |
bitten | +0.014 |
Neg logit contributions | |
ine | -0.018 |
birthday | -0.017 |
hip | -0.016 |
Office | -0.016 |
X | -0.016 |
Token | mud |
Token ID | 17492 |
Feature activation | -5.29e-02 |
Pos logit contributions | |
ls | +0.050 |
Run | +0.048 |
mun | +0.046 |
P | +0.044 |
led | +0.043 |
Neg logit contributions | |
birthday | -0.032 |
ine | -0.031 |
pattern | -0.029 |
eye | -0.028 |
moon | -0.027 |
Token | too |
Token ID | 1165 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
Run | +0.101 |
mun | +0.098 |
ls | +0.096 |
P | +0.090 |
Hey | +0.088 |
Neg logit contributions | |
ine | -0.076 |
birthday | -0.074 |
pattern | -0.067 |
hip | -0.065 |
X | -0.065 |
Token | much |
Token ID | 881 |
Feature activation | -2.86e-03 |
Pos logit contributions | |
mun | +0.028 |
Run | +0.028 |
ls | +0.027 |
P | +0.025 |
Hey | +0.025 |
Neg logit contributions | |
ine | -0.019 |
birthday | -0.018 |
hip | -0.016 |
Office | -0.015 |
X | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +1.15e-03 |
Pos logit contributions | |
ls | +0.004 |
Run | +0.004 |
mun | +0.004 |
bitten | +0.003 |
led | +0.003 |
Neg logit contributions | |
ine | -0.006 |
birthday | -0.005 |
Jupiter | -0.005 |
Office | -0.005 |
hip | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
ine | +0.002 |
birthday | +0.002 |
Office | +0.002 |
hip | +0.002 |
Jupiter | +0.002 |
Neg logit contributions | |
Run | -0.002 |
ls | -0.002 |
mun | -0.002 |
P | -0.002 |
bitten | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | +8.24e-03 |
Pos logit contributions | |
ine | +0.036 |
birthday | +0.034 |
Office | +0.033 |
X | +0.032 |
hip | +0.032 |
Neg logit contributions | |
ls | -0.026 |
Run | -0.025 |
mun | -0.024 |
P | -0.022 |
Hey | -0.021 |
Token | she |
Token ID | 673 |
Feature activation | -4.46e-03 |
Pos logit contributions | |
Run | +0.013 |
ls | +0.013 |
mun | +0.013 |
P | +0.011 |
Hey | +0.011 |
Neg logit contributions | |
ine | -0.026 |
birthday | -0.024 |
hip | -0.023 |
Office | -0.023 |
X | -0.023 |
Token | decided |
Token ID | 3066 |
Feature activation | -5.15e-03 |
Pos logit contributions | |
Run | +0.007 |
mun | +0.007 |
ls | +0.007 |
Hey | +0.006 |
P | +0.006 |
Neg logit contributions | |
ine | -0.008 |
Office | -0.007 |
birthday | -0.007 |
moon | -0.007 |
eye | -0.006 |
Token | to |
Token ID | 284 |
Feature activation | -1.75e-02 |
Pos logit contributions | |
Run | +0.012 |
mun | +0.011 |
ls | +0.011 |
P | +0.010 |
Hey | +0.010 |
Neg logit contributions | |
birthday | -0.006 |
ine | -0.006 |
hip | -0.005 |
desk | -0.005 |
for | -0.005 |
Token | give |
Token ID | 1577 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
ls | +0.039 |
Run | +0.039 |
mun | +0.037 |
P | +0.035 |
led | +0.034 |
Neg logit contributions | |
ine | -0.020 |
birthday | -0.019 |
hip | -0.017 |
X | -0.017 |
Office | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | -1.29e-02 |
Pos logit contributions | |
Run | +0.031 |
ls | +0.029 |
mun | +0.028 |
P | +0.026 |
awed | +0.026 |
Neg logit contributions | |
ine | -0.016 |
birthday | -0.015 |
hip | -0.014 |
end | -0.014 |
rap | -0.013 |
Token | boy |
Token ID | 2933 |
Feature activation | -5.29e-02 |
Pos logit contributions | |
ls | +0.028 |
aro | +0.027 |
Run | +0.027 |
P | +0.026 |
mun | +0.026 |
Neg logit contributions | |
birthday | -0.017 |
end | -0.016 |
moon | -0.015 |
eye | -0.015 |
X | -0.014 |
Token | one |
Token ID | 530 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
Run | +0.105 |
ls | +0.103 |
mun | +0.100 |
P | +0.090 |
Hey | +0.090 |
Neg logit contributions | |
ine | -0.075 |
birthday | -0.070 |
hip | -0.063 |
Office | -0.063 |
X | -0.062 |
Token | of |
Token ID | 286 |
Feature activation | -1.75e-02 |
Pos logit contributions | |
Run | +0.037 |
mun | +0.036 |
ls | +0.035 |
P | +0.032 |
Hey | +0.032 |
Neg logit contributions | |
birthday | -0.026 |
ine | -0.025 |
eye | -0.023 |
end | -0.022 |
X | -0.022 |
Token | her |
Token ID | 607 |
Feature activation | -4.17e-03 |
Pos logit contributions | |
Run | +0.035 |
ls | +0.035 |
mun | +0.033 |
led | +0.031 |
P | +0.030 |
Neg logit contributions | |
ine | -0.025 |
birthday | -0.023 |
hip | -0.021 |
X | -0.020 |
rap | -0.020 |
Token | own |
Token ID | 898 |
Feature activation | -1.59e-02 |
Pos logit contributions | |
Run | +0.008 |
ls | +0.008 |
mun | +0.008 |
Hey | +0.007 |
P | +0.007 |
Neg logit contributions | |
ine | -0.006 |
birthday | -0.006 |
X | -0.005 |
hip | -0.005 |
Office | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | -2.38e-03 |
Pos logit contributions | |
Run | +0.032 |
ls | +0.030 |
mun | +0.030 |
Hey | +0.028 |
P | +0.028 |
Neg logit contributions | |
ine | -0.023 |
birthday | -0.022 |
eye | -0.019 |
hip | -0.019 |
X | -0.019 |
Token | !" |
Token ID | 2474 |
Feature activation | -8.60e-03 |
Pos logit contributions | |
ine | +0.026 |
birthday | +0.024 |
hip | +0.023 |
Office | +0.023 |
uries | +0.023 |
Neg logit contributions | |
ls | -0.021 |
Run | -0.021 |
mun | -0.018 |
P | -0.016 |
Hey | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | +1.33e-03 |
Pos logit contributions | |
Run | +0.017 |
mun | +0.016 |
ls | +0.016 |
P | +0.015 |
led | +0.015 |
Neg logit contributions | |
ine | -0.012 |
birthday | -0.012 |
X | -0.011 |
Jupiter | -0.011 |
Today | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | -8.37e-03 |
Pos logit contributions | |
ine | +0.003 |
birthday | +0.002 |
Office | +0.002 |
hip | +0.002 |
X | +0.002 |
Neg logit contributions | |
ls | -0.002 |
Run | -0.002 |
mun | -0.002 |
P | -0.002 |
Hey | -0.002 |
Token | " |
Token ID | 1 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
Run | +0.014 |
ls | +0.014 |
mun | +0.014 |
P | +0.013 |
Hey | +0.012 |
Neg logit contributions | |
ine | -0.014 |
birthday | -0.013 |
hip | -0.012 |
Office | -0.012 |
X | -0.012 |
Token | This |
Token ID | 1212 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
ls | +0.013 |
Run | +0.012 |
mun | +0.012 |
bitten | +0.010 |
led | +0.010 |
Neg logit contributions | |
ine | -0.025 |
birthday | -0.024 |
hip | -0.023 |
Office | -0.023 |
X | -0.023 |
Token | is |
Token ID | 318 |
Feature activation | +3.01e-02 |
Pos logit contributions | |
ine | +0.025 |
birthday | +0.023 |
hip | +0.023 |
rap | +0.022 |
sing | +0.022 |
Neg logit contributions | |
Run | -0.026 |
ls | -0.026 |
mun | -0.025 |
Hey | -0.023 |
P | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | -1.23e-03 |
Pos logit contributions | |
ine | +0.056 |
hip | +0.054 |
sing | +0.051 |
desk | +0.051 |
eye | +0.051 |
Neg logit contributions | |
Run | -0.037 |
ls | -0.037 |
P | -0.033 |
Hey | -0.033 |
distracted | -0.031 |
Token | best |
Token ID | 1266 |
Feature activation | +7.34e-03 |
Pos logit contributions | |
Run | +0.002 |
mun | +0.002 |
ls | +0.002 |
P | +0.002 |
Hey | +0.002 |
Neg logit contributions | |
ine | -0.002 |
rap | -0.002 |
hip | -0.002 |
Office | -0.002 |
artments | -0.002 |
Token | gift |
Token ID | 6979 |
Feature activation | -4.25e-03 |
Pos logit contributions | |
ine | +0.010 |
hip | +0.009 |
Office | +0.008 |
uries | +0.008 |
ika | +0.008 |
Neg logit contributions | |
ls | -0.014 |
Run | -0.013 |
mun | -0.012 |
P | -0.011 |
Hey | -0.011 |
Token | ever |
Token ID | 1683 |
Feature activation | -8.18e-03 |
Pos logit contributions | |
ls | +0.007 |
Run | +0.007 |
mun | +0.006 |
bitten | +0.006 |
P | +0.006 |
Neg logit contributions | |
ine | -0.008 |
birthday | -0.007 |
hip | -0.007 |
Office | -0.007 |
Jupiter | -0.006 |
Token | !" |
Token ID | 2474 |
Feature activation | -6.27e-03 |
Pos logit contributions | |
ls | +0.013 |
Run | +0.012 |
mun | +0.012 |
led | +0.011 |
P | +0.011 |
Neg logit contributions | |
ine | -0.015 |
birthday | -0.014 |
Office | -0.013 |
hip | -0.013 |
Jupiter | -0.013 |
Token | From |
Token ID | 3574 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
ine | +0.015 |
birthday | +0.014 |
hip | +0.013 |
Office | +0.013 |
X | +0.013 |
Neg logit contributions | |
Run | -0.017 |
ls | -0.017 |
mun | -0.016 |
P | -0.015 |
Hey | -0.015 |
Token | that |
Token ID | 326 |
Feature activation | +2.75e-02 |
Pos logit contributions | |
ine | +0.015 |
birthday | +0.014 |
hip | +0.014 |
Office | +0.014 |
X | +0.013 |
Neg logit contributions | |
ls | -0.021 |
Run | -0.021 |
mun | -0.019 |
P | -0.018 |
Hey | -0.018 |
Token | day |
Token ID | 1110 |
Feature activation | +3.29e-04 |
Pos logit contributions | |
ine | +0.047 |
artments | +0.047 |
urger | +0.046 |
uries | +0.046 |
rap | +0.044 |
Neg logit contributions | |
mun | -0.040 |
Run | -0.040 |
ls | -0.039 |
P | -0.037 |
Hey | -0.035 |
Token | on |
Token ID | 319 |
Feature activation | +3.28e-03 |
Pos logit contributions | |
ine | +0.001 |
birthday | +0.001 |
hip | +0.001 |
Office | +0.001 |
X | +0.001 |
Neg logit contributions | |
Run | -0.000 |
ls | -0.000 |
mun | -0.000 |
P | -0.000 |
Hey | -0.000 |
Token | , |
Token ID | 11 |
Feature activation | +7.88e-03 |
Pos logit contributions | |
ine | +0.004 |
birthday | +0.004 |
Office | +0.003 |
hip | +0.003 |
Jupiter | +0.003 |
Neg logit contributions | |
Run | -0.007 |
ls | -0.007 |
mun | -0.007 |
P | -0.006 |
bitten | -0.006 |
Token | Lily |
Token ID | 20037 |
Feature activation | +3.94e-02 |
Pos logit contributions | |
ine | +0.012 |
birthday | +0.012 |
Office | +0.011 |
hip | +0.011 |
Jupiter | +0.011 |
Neg logit contributions | |
Run | -0.014 |
ls | -0.014 |
mun | -0.013 |
P | -0.012 |
bitten | -0.012 |
Token | loved |
Token ID | 6151 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
birthday | +0.075 |
sing | +0.070 |
X | +0.069 |
ine | +0.068 |
Today | +0.067 |
Neg logit contributions | |
mun | -0.053 |
ls | -0.051 |
distracted | -0.048 |
led | -0.048 |
P | -0.047 |
Token | playing |
Token ID | 2712 |
Feature activation | +4.04e-03 |
Pos logit contributions | |
ine | +0.022 |
hip | +0.020 |
eye | +0.019 |
uries | +0.019 |
Today | +0.019 |
Neg logit contributions | |
ls | -0.016 |
Run | -0.015 |
mun | -0.015 |
bitten | -0.014 |
distracted | -0.014 |
Token | in |
Token ID | 287 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
ine | +0.007 |
birthday | +0.007 |
hip | +0.007 |
Office | +0.006 |
X | +0.006 |
Neg logit contributions | |
ls | -0.006 |
Run | -0.006 |
mun | -0.006 |
P | -0.005 |
Hey | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | +1.67e-03 |
Pos logit contributions | |
ine | +0.046 |
urger | +0.046 |
hip | +0.044 |
uries | +0.043 |
Office | +0.042 |
Neg logit contributions | |
Run | -0.017 |
ls | -0.016 |
mun | -0.015 |
bitten | -0.015 |
P | -0.014 |
Token | empty |
Token ID | 6565 |
Feature activation | +2.71e-02 |
Pos logit contributions | |
ine | +0.002 |
hip | +0.002 |
birthday | +0.002 |
Office | +0.002 |
Jupiter | +0.002 |
Neg logit contributions | |
ls | -0.003 |
Run | -0.003 |
mun | -0.003 |
P | -0.003 |
Hey | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | -3.10e-04 |
Pos logit contributions | |
ine | +0.024 |
hip | +0.022 |
birthday | +0.022 |
X | +0.022 |
Jupiter | +0.022 |
Neg logit contributions | |
ls | -0.027 |
mun | -0.026 |
Run | -0.026 |
P | -0.023 |
led | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | +4.31e-03 |
Pos logit contributions | |
Run | +0.001 |
ls | +0.001 |
mun | +0.001 |
P | +0.001 |
bitten | +0.000 |
Neg logit contributions | |
ine | -0.000 |
birthday | -0.000 |
Jupiter | -0.000 |
Office | -0.000 |
hip | -0.000 |
Token | teacher |
Token ID | 4701 |
Feature activation | +1.29e-02 |
Pos logit contributions | |
ine | +0.006 |
birthday | +0.005 |
hip | +0.005 |
Office | +0.005 |
Jupiter | +0.005 |
Neg logit contributions | |
ls | -0.008 |
Run | -0.008 |
mun | -0.008 |
P | -0.007 |
Hey | -0.007 |
Token | asked |
Token ID | 1965 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
ine | +0.025 |
birthday | +0.024 |
hip | +0.023 |
Office | +0.023 |
X | +0.022 |
Neg logit contributions | |
ls | -0.018 |
Run | -0.018 |
mun | -0.017 |
P | -0.015 |
Hey | -0.015 |
Token | Lily |
Token ID | 20037 |
Feature activation | +2.44e-02 |
Pos logit contributions | |
ine | +0.035 |
X | +0.032 |
birthday | +0.031 |
Office | +0.031 |
sing | +0.031 |
Neg logit contributions | |
ls | -0.029 |
Run | -0.028 |
mun | -0.028 |
Hey | -0.025 |
led | -0.024 |
Token | 's |
Token ID | 338 |
Feature activation | +2.83e-02 |
Pos logit contributions | |
ine | +0.046 |
X | +0.045 |
sing | +0.043 |
oval | +0.042 |
Office | +0.042 |
Neg logit contributions | |
ls | -0.032 |
mun | -0.031 |
Run | -0.026 |
distracted | -0.026 |
bitten | -0.024 |
Token | mom |
Token ID | 1995 |
Feature activation | +2.81e-02 |
Pos logit contributions | |
ine | +0.044 |
hip | +0.039 |
rap | +0.035 |
uries | +0.034 |
artments | +0.034 |
Neg logit contributions | |
ls | -0.047 |
mun | -0.045 |
Run | -0.045 |
bitten | -0.042 |
distracted | -0.041 |
Token | to |
Token ID | 284 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
ine | +0.044 |
X | +0.043 |
sing | +0.042 |
Jupiter | +0.042 |
Office | +0.041 |
Neg logit contributions | |
mun | -0.044 |
ls | -0.042 |
Run | -0.035 |
distracted | -0.035 |
led | -0.035 |
Token | send |
Token ID | 3758 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
ine | +0.016 |
Office | +0.014 |
Jupiter | +0.014 |
birthday | +0.013 |
hip | +0.013 |
Neg logit contributions | |
Run | -0.018 |
mun | -0.017 |
P | -0.016 |
Hey | -0.016 |
ls | -0.016 |
Token | a |
Token ID | 257 |
Feature activation | +7.76e-03 |
Pos logit contributions | |
ine | +0.020 |
birthday | +0.019 |
Office | +0.018 |
hip | +0.018 |
X | +0.018 |
Neg logit contributions | |
ls | -0.018 |
Run | -0.018 |
mun | -0.017 |
P | -0.016 |
Hey | -0.016 |
Token | note |
Token ID | 3465 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
ine | +0.010 |
hip | +0.009 |
Office | +0.009 |
rap | +0.009 |
Jupiter | +0.008 |
Neg logit contributions | |
ls | -0.015 |
Run | -0.015 |
mun | -0.015 |
P | -0.014 |
bitten | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
ls | +0.003 |
Run | +0.003 |
mun | +0.003 |
bitten | +0.002 |
P | +0.002 |
Neg logit contributions | |
ine | -0.003 |
birthday | -0.003 |
Office | -0.003 |
hip | -0.003 |
X | -0.003 |
Token | From |
Token ID | 3574 |
Feature activation | +8.48e-03 |
Pos logit contributions | |
ine | +0.021 |
birthday | +0.019 |
Office | +0.018 |
hip | +0.018 |
X | +0.018 |
Neg logit contributions | |
ls | -0.019 |
Run | -0.019 |
mun | -0.018 |
P | -0.017 |
Hey | -0.016 |
Token | then |
Token ID | 788 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
ine | +0.012 |
birthday | +0.011 |
hip | +0.010 |
Office | +0.010 |
X | +0.010 |
Neg logit contributions | |
ls | -0.017 |
Run | -0.016 |
mun | -0.016 |
P | -0.015 |
Hey | -0.014 |
Token | on |
Token ID | 319 |
Feature activation | +1.01e-03 |
Pos logit contributions | |
ine | +0.039 |
birthday | +0.037 |
Office | +0.036 |
Jupiter | +0.036 |
hip | +0.036 |
Neg logit contributions | |
ls | -0.016 |
Run | -0.016 |
mun | -0.014 |
P | -0.014 |
Hey | -0.012 |
Token | , |
Token ID | 11 |
Feature activation | +5.08e-03 |
Pos logit contributions | |
ine | +0.001 |
birthday | +0.001 |
hip | +0.001 |
end | +0.001 |
Office | +0.001 |
Neg logit contributions | |
Run | -0.002 |
ls | -0.002 |
mun | -0.002 |
bitten | -0.002 |
P | -0.002 |
Token | Lily |
Token ID | 20037 |
Feature activation | +3.63e-02 |
Pos logit contributions | |
ine | +0.008 |
birthday | +0.008 |
hip | +0.007 |
Jupiter | +0.007 |
Office | +0.007 |
Neg logit contributions | |
Run | -0.009 |
ls | -0.009 |
mun | -0.009 |
P | -0.008 |
bitten | -0.008 |
Token | always |
Token ID | 1464 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
birthday | +0.061 |
ine | +0.061 |
hip | +0.057 |
X | +0.057 |
Jupiter | +0.056 |
Neg logit contributions | |
ls | -0.057 |
mun | -0.056 |
Run | -0.055 |
P | -0.052 |
distracted | -0.049 |
Token | brushed |
Token ID | 31782 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
birthday | +0.037 |
ine | +0.037 |
hip | +0.036 |
Jupiter | +0.036 |
X | +0.035 |
Neg logit contributions | |
mun | -0.020 |
Run | -0.020 |
ls | -0.018 |
bitten | -0.018 |
distracted | -0.018 |
Token | Max |
Token ID | 5436 |
Feature activation | +7.09e-03 |
Pos logit contributions | |
birthday | +0.028 |
ine | +0.027 |
X | +0.026 |
Office | +0.026 |
Jupiter | +0.026 |
Neg logit contributions | |
ls | -0.024 |
mun | -0.022 |
led | -0.021 |
Run | -0.021 |
Hey | -0.019 |
Token | softly |
Token ID | 26625 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
X | +0.013 |
ine | +0.013 |
birthday | +0.013 |
Jupiter | +0.013 |
Office | +0.012 |
Neg logit contributions | |
ls | -0.010 |
led | -0.009 |
mun | -0.009 |
Run | -0.009 |
distracted | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | +8.68e-03 |
Pos logit contributions | |
Jupiter | +0.051 |
rap | +0.050 |
ine | +0.049 |
astronaut | +0.049 |
X | +0.049 |
Neg logit contributions | |
ls | -0.030 |
led | -0.029 |
Run | -0.028 |
mun | -0.027 |
bitten | -0.027 |
Token | in |
Token ID | 287 |
Feature activation | +9.49e-03 |
Pos logit contributions | |
Run | +0.021 |
ls | +0.020 |
mun | +0.020 |
Hey | +0.018 |
Sorry | +0.017 |
Neg logit contributions | |
birthday | -0.012 |
for | -0.011 |
ine | -0.011 |
rap | -0.011 |
Jupiter | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | -1.74e-03 |
Pos logit contributions | |
ine | +0.018 |
birthday | +0.016 |
hip | +0.016 |
Office | +0.015 |
rap | +0.015 |
Neg logit contributions | |
Run | -0.014 |
ls | -0.013 |
mun | -0.013 |
bitten | -0.012 |
P | -0.012 |
Token | jungle |
Token ID | 20712 |
Feature activation | +9.91e-04 |
Pos logit contributions | |
P | +0.003 |
mun | +0.003 |
Run | +0.003 |
Hey | +0.002 |
ls | +0.002 |
Neg logit contributions | |
ine | -0.003 |
hip | -0.003 |
rap | -0.003 |
artments | -0.003 |
uries | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
ine | +0.001 |
birthday | +0.001 |
hip | +0.001 |
Office | +0.001 |
Jupiter | +0.001 |
Neg logit contributions | |
Run | -0.002 |
ls | -0.002 |
mun | -0.002 |
Hey | -0.002 |
P | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
ine | +0.028 |
birthday | +0.026 |
hip | +0.025 |
rap | +0.024 |
Office | +0.024 |
Neg logit contributions | |
Run | -0.026 |
ls | -0.025 |
mun | -0.025 |
Hey | -0.023 |
P | -0.022 |
Token | One |
Token ID | 3198 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
ine | +0.030 |
birthday | +0.028 |
hip | +0.026 |
Office | +0.026 |
X | +0.026 |
Neg logit contributions | |
ls | -0.025 |
Run | -0.025 |
mun | -0.024 |
P | -0.022 |
Hey | -0.021 |
Token | day |
Token ID | 1110 |
Feature activation | -3.84e-03 |
Pos logit contributions | |
ine | +0.043 |
artments | +0.040 |
uries | +0.040 |
hip | +0.040 |
rap | +0.039 |
Neg logit contributions | |
Run | -0.055 |
ls | -0.055 |
P | -0.051 |
mun | -0.049 |
Hey | -0.047 |
Token | , |
Token ID | 11 |
Feature activation | +3.00e-03 |
Pos logit contributions | |
mun | +0.009 |
ls | +0.009 |
Run | +0.009 |
awed | +0.009 |
P | +0.009 |
Neg logit contributions | |
birthday | -0.003 |
ine | -0.003 |
X | -0.003 |
moon | -0.002 |
Jupiter | -0.002 |
Token | he |
Token ID | 339 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
birthday | +0.004 |
ine | +0.004 |
X | +0.004 |
moon | +0.004 |
Jupiter | +0.004 |
Neg logit contributions | |
ls | -0.006 |
mun | -0.006 |
Run | -0.005 |
P | -0.005 |
awed | -0.005 |
Token | heard |
Token ID | 2982 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
ine | +0.016 |
birthday | +0.014 |
Office | +0.014 |
X | +0.013 |
Jupiter | +0.013 |
Neg logit contributions | |
ls | -0.019 |
Run | -0.018 |
mun | -0.018 |
aro | -0.016 |
Sorry | -0.015 |
Token | a |
Token ID | 257 |
Feature activation | -7.33e-03 |
Pos logit contributions | |
ine | +0.027 |
birthday | +0.025 |
hip | +0.024 |
Office | +0.024 |
rap | +0.023 |
Neg logit contributions | |
Run | -0.028 |
ls | -0.028 |
mun | -0.026 |
Hey | -0.024 |
bitten | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
ls | +0.007 |
Run | +0.007 |
mun | +0.007 |
P | +0.006 |
Hey | +0.006 |
Neg logit contributions | |
ine | -0.009 |
birthday | -0.009 |
hip | -0.008 |
X | -0.008 |
Office | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | +9.44e-03 |
Pos logit contributions | |
birthday | +0.032 |
ine | +0.031 |
X | +0.030 |
Office | +0.029 |
eye | +0.029 |
Neg logit contributions | |
ls | -0.018 |
Run | -0.018 |
mun | -0.017 |
P | -0.017 |
distracted | -0.014 |
Token | When |
Token ID | 2215 |
Feature activation | -9.84e-04 |
Pos logit contributions | |
birthday | +0.018 |
ine | +0.018 |
X | +0.017 |
Office | +0.017 |
hip | +0.016 |
Neg logit contributions | |
Run | -0.014 |
ls | -0.013 |
P | -0.013 |
mun | -0.012 |
Hey | -0.011 |
Token | it |
Token ID | 340 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
ls | +0.001 |
Run | +0.001 |
mun | +0.001 |
P | +0.001 |
Hey | +0.001 |
Neg logit contributions | |
ine | -0.002 |
birthday | -0.002 |
hip | -0.002 |
Office | -0.002 |
uries | -0.002 |
Token | started |
Token ID | 2067 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
ine | +0.017 |
birthday | +0.016 |
Office | +0.015 |
hip | +0.015 |
X | +0.015 |
Neg logit contributions | |
Run | -0.017 |
ls | -0.017 |
mun | -0.017 |
P | -0.015 |
bitten | -0.015 |
Token | getting |
Token ID | 1972 |
Feature activation | +2.49e-03 |
Pos logit contributions | |
Run | +0.026 |
ls | +0.025 |
mun | +0.025 |
P | +0.024 |
Hey | +0.023 |
Neg logit contributions | |
ine | -0.020 |
birthday | -0.019 |
Jupiter | -0.017 |
pattern | -0.017 |
hip | -0.016 |
Token | dark |
Token ID | 3223 |
Feature activation | +7.86e-03 |
Pos logit contributions | |
ine | +0.003 |
birthday | +0.003 |
Jupiter | +0.003 |
moon | +0.003 |
end | +0.003 |
Neg logit contributions | |
Run | -0.005 |
mun | -0.005 |
ls | -0.005 |
awed | -0.004 |
P | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | -6.76e-03 |
Pos logit contributions | |
ine | +0.011 |
birthday | +0.010 |
Office | +0.009 |
hip | +0.009 |
X | +0.009 |
Neg logit contributions | |
ls | -0.015 |
Run | -0.015 |
mun | -0.015 |
P | -0.013 |
Hey | -0.013 |
Token | Mild |
Token ID | 29087 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
ls | +0.012 |
Run | +0.012 |
mun | +0.011 |
P | +0.010 |
Hey | +0.010 |
Neg logit contributions | |
ine | -0.011 |
birthday | -0.010 |
hip | -0.010 |
Office | -0.010 |
X | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | +1.72e-03 |
Pos logit contributions | |
birthday | +0.028 |
ine | +0.027 |
X | +0.025 |
hip | +0.025 |
Jupiter | +0.025 |
Neg logit contributions | |
ls | -0.026 |
Run | -0.026 |
mun | -0.025 |
led | -0.024 |
Hey | -0.023 |
Token | her |
Token ID | 607 |
Feature activation | +5.12e-03 |
Pos logit contributions | |
ine | +0.003 |
birthday | +0.003 |
hip | +0.003 |
Office | +0.003 |
uries | +0.002 |
Neg logit contributions | |
ls | -0.003 |
Run | -0.003 |
mun | -0.003 |
bitten | -0.003 |
distracted | -0.002 |
Token | with |
Token ID | 351 |
Feature activation | -7.99e-03 |
Pos logit contributions | |
ls | +0.009 |
Run | +0.009 |
mun | +0.009 |
P | +0.008 |
Hey | +0.008 |
Neg logit contributions | |
ine | -0.010 |
birthday | -0.009 |
hip | -0.009 |
Office | -0.009 |
X | -0.009 |
Token | hair |
Token ID | 4190 |
Feature activation | +6.43e-03 |
Pos logit contributions | |
Run | +0.014 |
ls | +0.014 |
mun | +0.013 |
P | +0.012 |
Hey | +0.012 |
Neg logit contributions | |
ine | -0.013 |
birthday | -0.012 |
Office | -0.011 |
hip | -0.011 |
X | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -4.12e-03 |
Pos logit contributions | |
ine | +0.010 |
birthday | +0.009 |
Office | +0.009 |
hip | +0.009 |
X | +0.009 |
Neg logit contributions | |
ls | -0.012 |
Run | -0.011 |
mun | -0.011 |
P | -0.010 |
led | -0.010 |
Token | |
Token ID | 220 |
Feature activation | -1.97e-03 |
Pos logit contributions | |
ls | +0.008 |
Run | +0.007 |
mun | +0.007 |
P | +0.007 |
led | +0.007 |
Neg logit contributions | |
ine | -0.006 |
birthday | -0.006 |
Office | -0.006 |
hip | -0.005 |
Jupiter | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +5.88e-03 |
Pos logit contributions | |
Run | +0.003 |
ls | +0.003 |
mun | +0.003 |
P | +0.003 |
Hey | +0.003 |
Neg logit contributions | |
ine | -0.003 |
birthday | -0.003 |
hip | -0.003 |
Office | -0.003 |
X | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -2.82e-03 |
Pos logit contributions | |
ine | +0.010 |
birthday | +0.009 |
hip | +0.009 |
Office | +0.009 |
Jupiter | +0.009 |
Neg logit contributions | |
ls | -0.010 |
Run | -0.009 |
mun | -0.009 |
P | -0.008 |
bitten | -0.008 |
Token | The |
Token ID | 464 |
Feature activation | +9.39e-03 |
Pos logit contributions | |
ls | +0.005 |
Run | +0.005 |
mun | +0.004 |
led | +0.004 |
bitten | +0.004 |
Neg logit contributions | |
ine | -0.005 |
birthday | -0.004 |
hip | -0.004 |
Office | -0.004 |
Jupiter | -0.004 |
Token | bunny |
Token ID | 44915 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
ine | +0.016 |
birthday | +0.014 |
hip | +0.014 |
Office | +0.014 |
Jupiter | +0.014 |
Neg logit contributions | |
Run | -0.016 |
ls | -0.016 |
mun | -0.015 |
P | -0.014 |
Hey | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | +1.65e-03 |
Pos logit contributions | |
ine | +0.029 |
birthday | +0.028 |
hip | +0.026 |
Office | +0.026 |
X | +0.026 |
Neg logit contributions | |
ls | -0.024 |
Run | -0.024 |
mun | -0.023 |
P | -0.021 |
Hey | -0.020 |
Token | the |
Token ID | 262 |
Feature activation | -1.98e-04 |
Pos logit contributions | |
ine | +0.003 |
birthday | +0.003 |
hip | +0.002 |
Office | +0.002 |
X | +0.002 |
Neg logit contributions | |
ls | -0.003 |
Run | -0.003 |
mun | -0.003 |
P | -0.002 |
Hey | -0.002 |
Token | k |
Token ID | 479 |
Feature activation | -7.66e-03 |
Pos logit contributions | |
Run | +0.000 |
ls | +0.000 |
mun | +0.000 |
P | +0.000 |
Hey | +0.000 |
Neg logit contributions | |
ine | -0.000 |
birthday | -0.000 |
hip | -0.000 |
Office | -0.000 |
Jupiter | -0.000 |
Token | but |
Token ID | 475 |
Feature activation | +8.51e-03 |
Pos logit contributions | |
Run | +0.020 |
ls | +0.019 |
mun | +0.018 |
P | +0.016 |
distracted | +0.016 |
Neg logit contributions | |
ine | -0.020 |
birthday | -0.018 |
Office | -0.018 |
hip | -0.018 |
rap | -0.018 |
Token | warned |
Token ID | 7728 |
Feature activation | +2.35e-02 |
Pos logit contributions | |
ine | +0.016 |
birthday | +0.015 |
Office | +0.014 |
hip | +0.014 |
Jupiter | +0.014 |
Neg logit contributions | |
ls | -0.013 |
Run | -0.013 |
mun | -0.012 |
P | -0.011 |
bitten | -0.011 |
Token | her |
Token ID | 607 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
X | +0.045 |
Office | +0.044 |
ine | +0.043 |
Jupiter | +0.042 |
birthday | +0.041 |
Neg logit contributions | |
ls | -0.032 |
Run | -0.030 |
mun | -0.030 |
Hey | -0.028 |
led | -0.027 |
Token | that |
Token ID | 326 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
ine | +0.031 |
X | +0.029 |
birthday | +0.028 |
Office | +0.028 |
hip | +0.028 |
Neg logit contributions | |
ls | -0.028 |
Run | -0.026 |
mun | -0.025 |
bitten | -0.023 |
distracted | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
ine | +0.027 |
birthday | +0.025 |
hip | +0.024 |
Office | +0.024 |
X | +0.024 |
Neg logit contributions | |
ls | -0.017 |
Run | -0.017 |
mun | -0.017 |
P | -0.014 |
Hey | -0.014 |
Token | toy |
Token ID | 13373 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
ine | +0.014 |
birthday | +0.013 |
hip | +0.012 |
Office | +0.012 |
Jupiter | +0.012 |
Neg logit contributions | |
Run | -0.020 |
ls | -0.020 |
mun | -0.019 |
P | -0.018 |
Hey | -0.018 |
Token | was |
Token ID | 373 |
Feature activation | +7.99e-03 |
Pos logit contributions | |
ine | +0.019 |
birthday | +0.018 |
hip | +0.017 |
X | +0.017 |
Office | +0.017 |
Neg logit contributions | |
ls | -0.017 |
Run | -0.016 |
mun | -0.016 |
Hey | -0.014 |
bitten | -0.014 |
Token | very |
Token ID | 845 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
ine | +0.012 |
birthday | +0.011 |
Office | +0.010 |
hip | +0.010 |
Jupiter | +0.010 |
Neg logit contributions | |
Run | -0.015 |
ls | -0.015 |
mun | -0.015 |
P | -0.014 |
Hey | -0.013 |
Token | special |
Token ID | 2041 |
Feature activation | -3.35e-03 |
Pos logit contributions | |
Run | +0.045 |
mun | +0.044 |
ls | +0.043 |
Hey | +0.040 |
P | +0.039 |
Neg logit contributions | |
ine | -0.029 |
birthday | -0.026 |
Office | -0.023 |
hip | -0.023 |
X | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | -5.93e-03 |
Pos logit contributions | |
ls | +0.005 |
Run | +0.005 |
mun | +0.005 |
bitten | +0.004 |
P | +0.004 |
Neg logit contributions | |
ine | -0.006 |
hip | -0.005 |
Jupiter | -0.005 |
rap | -0.005 |
Office | -0.005 |
Token | she |
Token ID | 673 |
Feature activation | -4.96e-03 |
Pos logit contributions | |
ls | +0.009 |
Run | +0.009 |
mun | +0.008 |
bitten | +0.007 |
P | +0.007 |
Neg logit contributions | |
ine | -0.011 |
birthday | -0.010 |
hip | -0.010 |
Office | -0.010 |
X | -0.010 |
Token | back |
Token ID | 736 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
mun | +0.010 |
Run | +0.010 |
ls | +0.010 |
P | +0.009 |
led | +0.009 |
Neg logit contributions | |
ine | -0.007 |
birthday | -0.007 |
hip | -0.006 |
Jupiter | -0.006 |
X | -0.006 |
Token | to |
Token ID | 284 |
Feature activation | -9.35e-03 |
Pos logit contributions | |
Run | +0.025 |
mun | +0.024 |
ls | +0.023 |
P | +0.023 |
Hey | +0.023 |
Neg logit contributions | |
ine | -0.014 |
birthday | -0.013 |
Jupiter | -0.012 |
sleep | -0.012 |
for | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | -7.18e-03 |
Pos logit contributions | |
Run | +0.015 |
ls | +0.015 |
mun | +0.014 |
bitten | +0.013 |
P | +0.013 |
Neg logit contributions | |
ine | -0.016 |
birthday | -0.015 |
Office | -0.014 |
hip | -0.014 |
X | -0.014 |
Token | house |
Token ID | 2156 |
Feature activation | +1.29e-02 |
Pos logit contributions | |
Run | +0.013 |
ls | +0.013 |
mun | +0.013 |
P | +0.012 |
Hey | +0.012 |
Neg logit contributions | |
ine | -0.011 |
birthday | -0.010 |
hip | -0.009 |
Office | -0.009 |
X | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | -6.51e-03 |
Pos logit contributions | |
ine | +0.020 |
birthday | +0.018 |
Office | +0.017 |
hip | +0.017 |
X | +0.017 |
Neg logit contributions | |
ls | -0.023 |
Run | -0.022 |
mun | -0.022 |
bitten | -0.019 |
P | -0.019 |
Token | but |
Token ID | 475 |
Feature activation | +4.39e-03 |
Pos logit contributions | |
ls | +0.012 |
Run | +0.012 |
mun | +0.011 |
P | +0.011 |
Hey | +0.010 |
Neg logit contributions | |
ine | -0.010 |
birthday | -0.009 |
Office | -0.009 |
hip | -0.008 |
X | -0.008 |
Token | she |
Token ID | 673 |
Feature activation | -4.24e-03 |
Pos logit contributions | |
ine | +0.008 |
birthday | +0.008 |
Office | +0.007 |
hip | +0.007 |
X | +0.007 |
Neg logit contributions | |
ls | -0.006 |
Run | -0.006 |
mun | -0.006 |
P | -0.005 |
Hey | -0.005 |
Token | slipped |
Token ID | 18859 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
Run | +0.007 |
ls | +0.007 |
mun | +0.007 |
P | +0.006 |
Hey | +0.006 |
Neg logit contributions | |
ine | -0.007 |
birthday | -0.007 |
Office | -0.006 |
hip | -0.006 |
X | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | -3.55e-02 |
Pos logit contributions | |
ls | +0.024 |
mun | +0.024 |
Run | +0.024 |
P | +0.022 |
led | +0.021 |
Neg logit contributions | |
ine | -0.016 |
birthday | -0.014 |
hip | -0.013 |
X | -0.013 |
Office | -0.013 |
Token | fell |
Token ID | 3214 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
mun | +0.063 |
ls | +0.056 |
Run | +0.055 |
ins | +0.054 |
awed | +0.053 |
Neg logit contributions | |
ine | -0.048 |
sleep | -0.047 |
end | -0.044 |
for | -0.042 |
desk | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
mun | +0.025 |
Run | +0.024 |
ls | +0.023 |
P | +0.022 |
Hey | +0.022 |
Neg logit contributions | |
ine | -0.017 |
birthday | -0.015 |
for | -0.015 |
moon | -0.015 |
pattern | -0.015 |
Token | magic |
Token ID | 5536 |
Feature activation | +2.83e-02 |
Pos logit contributions | |
ine | +0.007 |
birthday | +0.006 |
Office | +0.006 |
hip | +0.006 |
X | +0.006 |
Neg logit contributions | |
ls | -0.010 |
Run | -0.010 |
mun | -0.009 |
P | -0.008 |
Hey | -0.008 |
Token | wand |
Token ID | 11569 |
Feature activation | -8.12e-03 |
Pos logit contributions | |
ine | +0.042 |
birthday | +0.037 |
hip | +0.036 |
Office | +0.035 |
X | +0.034 |
Neg logit contributions | |
ls | -0.055 |
Run | -0.053 |
mun | -0.050 |
Hey | -0.047 |
P | -0.047 |
Token | again |
Token ID | 757 |
Feature activation | -2.55e-03 |
Pos logit contributions | |
Run | +0.014 |
ls | +0.014 |
mun | +0.013 |
P | +0.012 |
Hey | +0.012 |
Neg logit contributions | |
ine | -0.013 |
birthday | -0.013 |
hip | -0.012 |
Office | -0.012 |
X | -0.012 |
Token | !" |
Token ID | 2474 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
ls | +0.004 |
Run | +0.004 |
mun | +0.004 |
P | +0.004 |
Hey | +0.003 |
Neg logit contributions | |
ine | -0.004 |
birthday | -0.004 |
hip | -0.004 |
Office | -0.004 |
X | -0.004 |
Token | They |
Token ID | 1119 |
Feature activation | -5.20e-03 |
Pos logit contributions | |
Run | +0.029 |
ls | +0.029 |
mun | +0.029 |
P | +0.027 |
bitten | +0.027 |
Neg logit contributions | |
ine | -0.017 |
birthday | -0.017 |
Today | -0.015 |
desk | -0.014 |
Grand | -0.014 |
Token | went |
Token ID | 1816 |
Feature activation | -8.14e-04 |
Pos logit contributions | |
ls | +0.008 |
Run | +0.008 |
mun | +0.008 |
Hey | +0.007 |
P | +0.007 |
Neg logit contributions | |
ine | -0.010 |
birthday | -0.009 |
X | -0.009 |
eye | -0.008 |
hip | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | -4.91e-03 |
Pos logit contributions | |
Run | +0.002 |
ls | +0.002 |
mun | +0.002 |
P | +0.001 |
Hey | +0.001 |
Neg logit contributions | |
ine | -0.001 |
birthday | -0.001 |
X | -0.001 |
Office | -0.001 |
Jupiter | -0.001 |
Token | dry |
Token ID | 5894 |
Feature activation | -9.04e-03 |
Pos logit contributions | |
Run | +0.008 |
ls | +0.007 |
mun | +0.007 |
bitten | +0.006 |
distracted | +0.006 |
Neg logit contributions | |
ine | -0.009 |
hip | -0.008 |
birthday | -0.008 |
Office | -0.008 |
urger | -0.008 |
Token | themselves |
Token ID | 2405 |
Feature activation | -2.95e-04 |
Pos logit contributions | |
Run | +0.015 |
ls | +0.015 |
mun | +0.014 |
P | +0.013 |
Hey | +0.013 |
Neg logit contributions | |
ine | -0.015 |
birthday | -0.014 |
Office | -0.014 |
hip | -0.014 |
X | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
ls | +0.000 |
Run | +0.000 |
mun | +0.000 |
bitten | +0.000 |
P | +0.000 |
Neg logit contributions | |
ine | -0.001 |
birthday | -0.000 |
Office | -0.000 |
hip | -0.000 |
X | -0.000 |
Token | find |
Token ID | 1064 |
Feature activation | +4.97e-03 |
Pos logit contributions | |
ls | +0.026 |
Run | +0.026 |
mun | +0.025 |
P | +0.023 |
Hey | +0.022 |
Neg logit contributions | |
ine | -0.027 |
birthday | -0.025 |
Office | -0.023 |
X | -0.023 |
hip | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
ine | +0.002 |
birthday | +0.002 |
Office | +0.002 |
Jupiter | +0.002 |
hip | +0.002 |
Neg logit contributions | |
ls | -0.004 |
Run | -0.004 |
mun | -0.004 |
P | -0.003 |
led | -0.003 |
Token | But |
Token ID | 1537 |
Feature activation | +1.27e-02 |
Pos logit contributions | |
ine | +0.023 |
birthday | +0.022 |
hip | +0.021 |
Office | +0.021 |
X | +0.021 |
Neg logit contributions | |
ls | -0.017 |
Run | -0.017 |
mun | -0.016 |
P | -0.015 |
Hey | -0.014 |
Token | then |
Token ID | 788 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
ine | +0.018 |
birthday | +0.017 |
hip | +0.016 |
Office | +0.016 |
X | +0.015 |
Neg logit contributions | |
Run | -0.025 |
ls | -0.024 |
mun | -0.023 |
distracted | -0.021 |
Hey | -0.021 |
Token | something |
Token ID | 1223 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
ls | +0.034 |
Run | +0.034 |
mun | +0.032 |
P | +0.030 |
Hey | +0.029 |
Neg logit contributions | |
ine | -0.024 |
birthday | -0.022 |
hip | -0.020 |
Office | -0.020 |
X | -0.020 |
Token | happened |
Token ID | 3022 |
Feature activation | -1.61e-02 |
Pos logit contributions | |
mun | +0.023 |
Run | +0.021 |
ls | +0.021 |
P | +0.019 |
awed | +0.019 |
Neg logit contributions | |
ine | -0.017 |
end | -0.016 |
birthday | -0.016 |
pattern | -0.016 |
believe | -0.015 |
Token | that |
Token ID | 326 |
Feature activation | -6.35e-03 |
Pos logit contributions | |
ls | +0.030 |
Run | +0.030 |
mun | +0.030 |
P | +0.027 |
Hey | +0.026 |
Neg logit contributions | |
ine | -0.024 |
birthday | -0.021 |
hip | -0.020 |
Office | -0.020 |
X | -0.020 |
Token | changed |
Token ID | 3421 |
Feature activation | -9.92e-03 |
Pos logit contributions | |
mun | +0.011 |
Run | +0.011 |
ls | +0.011 |
P | +0.011 |
awed | +0.010 |
Neg logit contributions | |
ine | -0.009 |
moon | -0.009 |
Jupiter | -0.009 |
birthday | -0.009 |
X | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | -8.78e-03 |
Pos logit contributions | |
Run | +0.017 |
ls | +0.017 |
mun | +0.017 |
P | +0.015 |
Hey | +0.015 |
Neg logit contributions | |
ine | -0.016 |
birthday | -0.015 |
hip | -0.014 |
Office | -0.014 |
Jupiter | -0.014 |
Token | race |
Token ID | 3234 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
ls | +0.017 |
Run | +0.017 |
mun | +0.017 |
led | +0.015 |
P | +0.015 |
Neg logit contributions | |
birthday | -0.012 |
ine | -0.012 |
pattern | -0.011 |
X | -0.011 |
moon | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -6.20e-03 |
Pos logit contributions | |
ls | +0.027 |
Run | +0.027 |
mun | +0.027 |
P | +0.025 |
bitten | +0.023 |
Neg logit contributions | |
ine | -0.023 |
birthday | -0.022 |
hip | -0.021 |
Office | -0.020 |
X | -0.020 |
Token | The |
Token ID | 383 |
Feature activation | -2.11e-03 |
Pos logit contributions | |
ls | +0.013 |
mun | +0.013 |
Run | +0.013 |
P | +0.012 |
led | +0.012 |
Neg logit contributions | |
ine | -0.008 |
birthday | -0.007 |
Office | -0.007 |
Jupiter | -0.007 |
hip | -0.007 |
Token | not |
Token ID | 407 |
Feature activation | -1.85e-03 |
Pos logit contributions | |
Run | +0.003 |
ls | +0.003 |
mun | +0.003 |
P | +0.003 |
Hey | +0.003 |
Neg logit contributions | |
ine | -0.003 |
birthday | -0.002 |
Office | -0.002 |
hip | -0.002 |
X | -0.002 |
Token | want |
Token ID | 765 |
Feature activation | +5.02e-03 |
Pos logit contributions | |
Run | +0.004 |
ls | +0.003 |
mun | +0.003 |
bitten | +0.003 |
bothered | +0.003 |
Neg logit contributions | |
ine | -0.003 |
birthday | -0.002 |
Office | -0.002 |
Jupiter | -0.002 |
rap | -0.002 |
Token | to |
Token ID | 284 |
Feature activation | -2.40e-02 |
Pos logit contributions | |
ine | +0.009 |
birthday | +0.008 |
hip | +0.008 |
Office | +0.008 |
X | +0.008 |
Neg logit contributions | |
ls | -0.008 |
Run | -0.008 |
mun | -0.008 |
P | -0.007 |
Hey | -0.007 |
Token | play |
Token ID | 711 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
Run | +0.050 |
ls | +0.049 |
mun | +0.048 |
P | +0.044 |
bitten | +0.044 |
Neg logit contributions | |
ine | -0.030 |
birthday | -0.027 |
Office | -0.026 |
hip | -0.025 |
Jupiter | -0.025 |
Token | with |
Token ID | 351 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
Run | +0.021 |
ls | +0.021 |
mun | +0.020 |
P | +0.018 |
Hey | +0.017 |
Neg logit contributions | |
ine | -0.022 |
birthday | -0.021 |
Office | -0.019 |
hip | -0.019 |
X | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | -5.30e-03 |
Pos logit contributions | |
Run | +0.037 |
ls | +0.037 |
mun | +0.036 |
P | +0.034 |
Hey | +0.032 |
Neg logit contributions | |
ine | -0.034 |
birthday | -0.032 |
Office | -0.029 |
hip | -0.029 |
X | -0.029 |
Token | baby |
Token ID | 5156 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
ls | +0.010 |
Run | +0.010 |
mun | +0.009 |
P | +0.008 |
Hey | +0.008 |
Neg logit contributions | |
ine | -0.008 |
birthday | -0.008 |
hip | -0.007 |
Office | -0.007 |
X | -0.007 |
Token | sheep |
Token ID | 15900 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
Run | +0.025 |
ls | +0.023 |
P | +0.023 |
mun | +0.023 |
Hey | +0.022 |
Neg logit contributions | |
ine | -0.016 |
birthday | -0.014 |
moon | -0.013 |
Office | -0.013 |
Jupiter | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -9.68e-03 |
Pos logit contributions | |
Run | +0.048 |
ls | +0.047 |
mun | +0.046 |
P | +0.043 |
Hey | +0.042 |
Neg logit contributions | |
ine | -0.039 |
birthday | -0.037 |
hip | -0.034 |
Office | -0.034 |
Jupiter | -0.033 |
Token | They |
Token ID | 1119 |
Feature activation | -2.06e-02 |
Pos logit contributions | |
ls | +0.019 |
Run | +0.019 |
mun | +0.018 |
P | +0.017 |
bitten | +0.017 |
Neg logit contributions | |
ine | -0.014 |
birthday | -0.013 |
Office | -0.012 |
hip | -0.012 |
Jupiter | -0.011 |
Token | wanted |
Token ID | 2227 |
Feature activation | +1.03e-03 |
Pos logit contributions | |
ls | +0.032 |
Run | +0.031 |
mun | +0.030 |
P | +0.027 |
Hey | +0.026 |
Neg logit contributions | |
ine | -0.038 |
birthday | -0.034 |
X | -0.034 |
Office | -0.033 |
eye | -0.033 |
Token | to |
Token ID | 284 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
ls | +0.013 |
Run | +0.013 |
mun | +0.013 |
P | +0.011 |
Hey | +0.011 |
Neg logit contributions | |
ine | -0.011 |
birthday | -0.010 |
Office | -0.009 |
hip | -0.009 |
X | -0.009 |
Token | explore |
Token ID | 7301 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
ine | +0.023 |
Birthday | +0.021 |
Office | +0.021 |
urger | +0.021 |
birthday | +0.019 |
Neg logit contributions | |
Run | -0.021 |
ls | -0.020 |
mun | -0.019 |
Hey | -0.019 |
hurt | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -2.49e-03 |
Pos logit contributions | |
ine | +0.020 |
birthday | +0.019 |
Office | +0.018 |
hip | +0.018 |
rap | +0.017 |
Neg logit contributions | |
ls | -0.019 |
Run | -0.019 |
mun | -0.018 |
Hey | -0.016 |
led | -0.016 |
Token | Tom |
Token ID | 4186 |
Feature activation | +5.08e-03 |
Pos logit contributions | |
ls | +0.004 |
Run | +0.004 |
mun | +0.004 |
P | +0.004 |
led | +0.004 |
Neg logit contributions | |
ine | -0.004 |
birthday | -0.004 |
Office | -0.003 |
hip | -0.003 |
X | -0.003 |
Token | started |
Token ID | 2067 |
Feature activation | -8.50e-03 |
Pos logit contributions | |
birthday | +0.010 |
ine | +0.010 |
hip | +0.009 |
Office | +0.009 |
X | +0.009 |
Neg logit contributions | |
ls | -0.007 |
Run | -0.007 |
mun | -0.007 |
led | -0.006 |
distracted | -0.006 |
Token | to |
Token ID | 284 |
Feature activation | -4.68e-03 |
Pos logit contributions | |
ls | +0.013 |
Run | +0.013 |
mun | +0.013 |
P | +0.012 |
bitten | +0.011 |
Neg logit contributions | |
ine | -0.015 |
birthday | -0.014 |
hip | -0.013 |
Office | -0.013 |
X | -0.013 |
Token | explore |
Token ID | 7301 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
Run | +0.009 |
mun | +0.008 |
ls | +0.008 |
Hey | +0.007 |
bitten | +0.007 |
Neg logit contributions | |
ine | -0.007 |
birthday | -0.006 |
Office | -0.006 |
rap | -0.006 |
hip | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | +1.70e-03 |
Pos logit contributions | |
ine | +0.030 |
birthday | +0.029 |
hip | +0.027 |
Office | +0.027 |
Birthday | +0.027 |
Neg logit contributions | |
ls | -0.025 |
Run | -0.024 |
mun | -0.023 |
led | -0.020 |
Hey | -0.020 |
Token | camp |
Token ID | 1413 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
ine | +0.003 |
hip | +0.002 |
birthday | +0.002 |
rap | +0.002 |
X | +0.002 |
Neg logit contributions | |
ls | -0.003 |
mun | -0.003 |
Run | -0.003 |
Hey | -0.003 |
P | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | +1.02e-03 |
Pos logit contributions | |
ine | +0.039 |
birthday | +0.037 |
rap | +0.034 |
hip | +0.034 |
X | +0.034 |
Neg logit contributions | |
ls | -0.039 |
Run | -0.037 |
mun | -0.037 |
led | -0.033 |
P | -0.031 |
Token | discovering |
Token ID | 21611 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
ine | +0.002 |
birthday | +0.002 |
hip | +0.001 |
Office | +0.001 |
rap | +0.001 |
Neg logit contributions | |
Run | -0.002 |
mun | -0.002 |
ls | -0.002 |
bitten | -0.002 |
distracted | -0.001 |
Token | sour |
Token ID | 11348 |
Feature activation | -2.19e-02 |
Pos logit contributions | |
Run | +0.023 |
ls | +0.022 |
mun | +0.022 |
Hey | +0.020 |
P | +0.020 |
Neg logit contributions | |
birthday | -0.020 |
ine | -0.019 |
X | -0.018 |
eye | -0.017 |
pattern | -0.017 |
Token | candy |
Token ID | 18550 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
Run | +0.035 |
ls | +0.035 |
mun | +0.034 |
P | +0.030 |
Hey | +0.029 |
Neg logit contributions | |
ine | -0.038 |
birthday | -0.036 |
X | -0.033 |
hip | -0.033 |
Office | -0.033 |
Token | to |
Token ID | 284 |
Feature activation | +9.15e-03 |
Pos logit contributions | |
Run | +0.023 |
mun | +0.022 |
ls | +0.021 |
P | +0.019 |
Hey | +0.019 |
Neg logit contributions | |
ine | -0.018 |
birthday | -0.017 |
hip | -0.016 |
X | -0.015 |
Office | -0.015 |
Token | share |
Token ID | 2648 |
Feature activation | -5.31e-03 |
Pos logit contributions | |
ine | +0.014 |
Jupiter | +0.014 |
Office | +0.013 |
birthday | +0.013 |
astronaut | +0.013 |
Neg logit contributions | |
mun | -0.015 |
ls | -0.014 |
Run | -0.014 |
bitten | -0.014 |
P | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | +5.54e-03 |
Pos logit contributions | |
ls | +0.008 |
Run | +0.008 |
mun | +0.008 |
P | +0.007 |
Hey | +0.007 |
Neg logit contributions | |
ine | -0.009 |
birthday | -0.009 |
Office | -0.008 |
hip | -0.008 |
Jupiter | -0.008 |
Token | Tim |
Token ID | 5045 |
Feature activation | -5.71e-03 |
Pos logit contributions | |
ine | +0.010 |
hip | +0.009 |
birthday | +0.009 |
rap | +0.009 |
Office | +0.009 |
Neg logit contributions | |
Run | -0.008 |
ls | -0.008 |
mun | -0.007 |
Hey | -0.007 |
distracted | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | +9.96e-03 |
Pos logit contributions | |
Run | +0.009 |
ls | +0.009 |
mun | +0.009 |
P | +0.008 |
Hey | +0.008 |
Neg logit contributions | |
ine | -0.010 |
birthday | -0.009 |
Office | -0.009 |
hip | -0.009 |
X | -0.009 |
Token | Jake |
Token ID | 14757 |
Feature activation | -3.94e-03 |
Pos logit contributions | |
ine | +0.012 |
birthday | +0.011 |
uries | +0.010 |
hip | +0.010 |
Office | +0.010 |
Neg logit contributions | |
Run | -0.021 |
ls | -0.020 |
mun | -0.020 |
Hey | -0.019 |
led | -0.018 |
Token | both |
Token ID | 1111 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
Run | +0.006 |
ls | +0.006 |
mun | +0.005 |
P | +0.005 |
Hey | +0.005 |
Neg logit contributions | |
ine | -0.007 |
birthday | -0.007 |
hip | -0.007 |
Office | -0.007 |
Jupiter | -0.007 |
Token | loved |
Token ID | 6151 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
ls | +0.019 |
Run | +0.019 |
mun | +0.018 |
P | +0.016 |
Hey | +0.016 |
Neg logit contributions | |
ine | -0.025 |
birthday | -0.024 |
Office | -0.022 |
hip | -0.022 |
X | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
ls | +0.019 |
Run | +0.018 |
mun | +0.018 |
bitten | +0.016 |
P | +0.016 |
Neg logit contributions | |
ine | -0.021 |
birthday | -0.019 |
hip | -0.018 |
Office | -0.018 |
Jupiter | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -4.43e-03 |
Pos logit contributions | |
Run | +0.017 |
mun | +0.016 |
ls | +0.016 |
P | +0.016 |
led | +0.015 |
Neg logit contributions | |
birthday | -0.015 |
ine | -0.014 |
Jupiter | -0.013 |
moon | -0.013 |
X | -0.013 |
Token | They |
Token ID | 1119 |
Feature activation | +1.11e-03 |
Pos logit contributions | |
ls | +0.008 |
Run | +0.008 |
mun | +0.008 |
P | +0.007 |
bitten | +0.007 |
Neg logit contributions | |
ine | -0.007 |
birthday | -0.006 |
Office | -0.006 |
hip | -0.006 |
X | -0.006 |
Token | are |
Token ID | 389 |
Feature activation | +7.91e-03 |
Pos logit contributions | |
ine | +0.001 |
birthday | +0.001 |
X | +0.001 |
hip | +0.001 |
Office | +0.001 |
Neg logit contributions | |
ls | -0.002 |
Run | -0.002 |
mun | -0.002 |
P | -0.002 |
bitten | -0.002 |
Token | glad |
Token ID | 9675 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
ine | +0.012 |
birthday | +0.011 |
Jupiter | +0.010 |
Office | +0.010 |
X | +0.010 |
Neg logit contributions | |
Run | -0.015 |
ls | -0.015 |
mun | -0.015 |
Hey | -0.013 |
P | -0.013 |
Token | they |
Token ID | 484 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
Run | +0.020 |
mun | +0.020 |
ls | +0.019 |
P | +0.019 |
led | +0.018 |
Neg logit contributions | |
birthday | -0.023 |
ine | -0.022 |
Jupiter | -0.020 |
moon | -0.020 |
X | -0.020 |
Token | have |
Token ID | 423 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
ls | +0.025 |
Run | +0.025 |
mun | +0.024 |
P | +0.022 |
awed | +0.021 |
Neg logit contributions | |
ine | -0.018 |
birthday | -0.017 |
hip | -0.017 |
Office | -0.017 |
X | -0.016 |
Token | float |
Token ID | 12178 |
Feature activation | -3.54e-03 |
Pos logit contributions | |
Run | +0.036 |
ls | +0.036 |
mun | +0.035 |
Hey | +0.031 |
P | +0.031 |
Neg logit contributions | |
ine | -0.023 |
birthday | -0.022 |
Office | -0.019 |
hip | -0.019 |
X | -0.019 |
Token | ies |
Token ID | 444 |
Feature activation | -2.39e-03 |
Pos logit contributions | |
Run | +0.006 |
mun | +0.006 |
ls | +0.006 |
P | +0.005 |
Hey | +0.005 |
Neg logit contributions | |
ine | -0.006 |
birthday | -0.005 |
hip | -0.005 |
X | -0.005 |
Office | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | -4.85e-04 |
Pos logit contributions | |
Run | +0.004 |
ls | +0.004 |
mun | +0.004 |
P | +0.004 |
Hey | +0.004 |
Neg logit contributions | |
ine | -0.004 |
birthday | -0.004 |
hip | -0.003 |
Office | -0.003 |
X | -0.003 |
Token | They |
Token ID | 1119 |
Feature activation | +3.86e-03 |
Pos logit contributions | |
Run | +0.001 |
ls | +0.001 |
mun | +0.001 |
P | +0.001 |
Hey | +0.001 |
Neg logit contributions | |
ine | -0.001 |
birthday | -0.001 |
hip | -0.001 |
Office | -0.001 |
X | -0.001 |
Token | are |
Token ID | 389 |
Feature activation | +1.28e-02 |
Pos logit contributions | |
ine | +0.005 |
birthday | +0.004 |
X | +0.004 |
hip | +0.004 |
Office | +0.004 |
Neg logit contributions | |
ls | -0.008 |
Run | -0.008 |
mun | -0.008 |
P | -0.007 |
led | -0.007 |
Token | hairy |
Token ID | 43873 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
Run | +0.042 |
ls | +0.041 |
mun | +0.041 |
P | +0.038 |
Hey | +0.036 |
Neg logit contributions | |
ine | -0.038 |
birthday | -0.036 |
hip | -0.033 |
X | -0.032 |
pattern | -0.032 |
Token | and |
Token ID | 290 |
Feature activation | -2.09e-03 |
Pos logit contributions | |
ls | +0.031 |
Run | +0.030 |
mun | +0.029 |
bitten | +0.026 |
P | +0.026 |
Neg logit contributions | |
ine | -0.029 |
birthday | -0.027 |
Office | -0.026 |
hip | -0.026 |
Jupiter | -0.026 |
Token | bent |
Token ID | 17157 |
Feature activation | +1.22e-03 |
Pos logit contributions | |
Run | +0.003 |
ls | +0.003 |
mun | +0.003 |
P | +0.003 |
Hey | +0.003 |
Neg logit contributions | |
ine | -0.004 |
birthday | -0.003 |
hip | -0.003 |
Office | -0.003 |
X | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +1.30e-04 |
Pos logit contributions | |
ine | +0.002 |
birthday | +0.002 |
Office | +0.002 |
hip | +0.002 |
X | +0.002 |
Neg logit contributions | |
ls | -0.002 |
Run | -0.002 |
mun | -0.002 |
P | -0.001 |
Hey | -0.001 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
ine | +0.000 |
birthday | +0.000 |
Office | +0.000 |
Jupiter | +0.000 |
hip | +0.000 |
Neg logit contributions | |
ls | -0.000 |
Run | -0.000 |
mun | -0.000 |
P | -0.000 |
led | -0.000 |
Token | got |
Token ID | 1392 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
Run | +0.017 |
ls | +0.017 |
mun | +0.016 |
P | +0.015 |
Hey | +0.015 |
Neg logit contributions | |
ine | -0.017 |
pattern | -0.014 |
desk | -0.014 |
eye | -0.013 |
Office | -0.013 |
Token | upset |
Token ID | 9247 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
mun | +0.063 |
Run | +0.060 |
ls | +0.058 |
P | +0.054 |
awed | +0.054 |
Neg logit contributions | |
for | -0.053 |
ine | -0.053 |
eye | -0.052 |
end | -0.052 |
birthday | -0.050 |
Token | and |
Token ID | 290 |
Feature activation | -3.53e-02 |
Pos logit contributions | |
ls | +0.028 |
Run | +0.028 |
mun | +0.027 |
P | +0.025 |
Hey | +0.024 |
Neg logit contributions | |
ine | -0.020 |
birthday | -0.019 |
Office | -0.018 |
hip | -0.017 |
X | -0.017 |
Token | told |
Token ID | 1297 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
P | +0.059 |
mun | +0.057 |
Run | +0.057 |
ls | +0.056 |
� | +0.054 |
Neg logit contributions | |
ine | -0.052 |
for | -0.049 |
end | -0.044 |
desk | -0.044 |
be | -0.044 |
Token | Max |
Token ID | 5436 |
Feature activation | -8.87e-03 |
Pos logit contributions | |
ine | +0.047 |
birthday | +0.044 |
Office | +0.044 |
uries | +0.042 |
eye | +0.042 |
Neg logit contributions | |
ls | -0.034 |
Run | -0.033 |
mun | -0.031 |
Hey | -0.029 |
bitten | -0.028 |
Token | he |
Token ID | 339 |
Feature activation | -1.65e-02 |
Pos logit contributions | |
Run | +0.017 |
ls | +0.016 |
mun | +0.016 |
P | +0.015 |
Hey | +0.015 |
Neg logit contributions | |
ine | -0.013 |
birthday | -0.012 |
hip | -0.011 |
Office | -0.011 |
Jupiter | -0.011 |
Token | for |
Token ID | 329 |
Feature activation | +8.03e-03 |
Pos logit contributions | |
Run | +0.021 |
ls | +0.021 |
mun | +0.020 |
P | +0.019 |
Hey | +0.018 |
Neg logit contributions | |
ine | -0.014 |
birthday | -0.013 |
hip | -0.012 |
Office | -0.012 |
X | -0.012 |
Token | a |
Token ID | 257 |
Feature activation | -1.25e-02 |
Pos logit contributions | |
ine | +0.016 |
hip | +0.016 |
Office | +0.015 |
Birthday | +0.015 |
rap | +0.015 |
Neg logit contributions | |
Run | -0.010 |
ls | -0.009 |
bitten | -0.009 |
killed | -0.009 |
distracted | -0.008 |
Token | while |
Token ID | 981 |
Feature activation | -1.42e-03 |
Pos logit contributions | |
Run | +0.022 |
ls | +0.022 |
mun | +0.021 |
P | +0.019 |
bitten | +0.019 |
Neg logit contributions | |
ine | -0.020 |
birthday | -0.018 |
Office | -0.018 |
hip | -0.018 |
X | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | -2.35e-02 |
Pos logit contributions | |
ls | +0.003 |
Run | +0.003 |
mun | +0.003 |
P | +0.003 |
Hey | +0.003 |
Neg logit contributions | |
ine | -0.002 |
birthday | -0.002 |
X | -0.002 |
hip | -0.002 |
Jupiter | -0.002 |
Token | he |
Token ID | 339 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
ls | +0.035 |
mun | +0.034 |
Run | +0.033 |
P | +0.031 |
awed | +0.030 |
Neg logit contributions | |
ine | -0.042 |
birthday | -0.041 |
X | -0.039 |
Jupiter | -0.039 |
moon | -0.038 |
Token | started |
Token ID | 2067 |
Feature activation | -3.84e-02 |
Pos logit contributions | |
ls | +0.018 |
Run | +0.018 |
mun | +0.018 |
P | +0.016 |
Hey | +0.015 |
Neg logit contributions | |
ine | -0.022 |
birthday | -0.019 |
Office | -0.018 |
X | -0.018 |
eye | -0.018 |
Token | to |
Token ID | 284 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
Run | +0.081 |
ls | +0.078 |
mun | +0.076 |
Hey | +0.075 |
P | +0.074 |
Neg logit contributions | |
ine | -0.052 |
birthday | -0.045 |
desk | -0.039 |
pattern | -0.039 |
Jupiter | -0.039 |
Token | get |
Token ID | 651 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
ls | +0.071 |
Run | +0.070 |
mun | +0.068 |
P | +0.064 |
Hey | +0.062 |
Neg logit contributions | |
ine | -0.046 |
birthday | -0.042 |
hip | -0.039 |
Office | -0.039 |
X | -0.038 |
Token | tired |
Token ID | 10032 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
Run | +0.067 |
mun | +0.066 |
ls | +0.062 |
awed | +0.058 |
Hey | +0.057 |
Neg logit contributions | |
ine | -0.061 |
birthday | -0.055 |
eye | -0.055 |
end | -0.055 |
sleep | -0.054 |
Token | . |
Token ID | 13 |
Feature activation | -3.83e-03 |
Pos logit contributions | |
ls | +0.019 |
Run | +0.019 |
mun | +0.018 |
P | +0.016 |
Hey | +0.016 |
Neg logit contributions | |
ine | -0.021 |
birthday | -0.019 |
Office | -0.019 |
hip | -0.018 |
Jupiter | -0.018 |
Token | He |
Token ID | 679 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
Run | +0.005 |
ls | +0.005 |
mun | +0.005 |
Hey | +0.005 |
bitten | +0.005 |
Neg logit contributions | |
ine | -0.007 |
birthday | -0.007 |
hip | -0.006 |
rap | -0.006 |
Office | -0.006 |
Token | a |
Token ID | 257 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
Run | +0.008 |
ls | +0.008 |
mun | +0.007 |
P | +0.007 |
Hey | +0.007 |
Neg logit contributions | |
ine | -0.008 |
birthday | -0.007 |
hip | -0.007 |
Office | -0.007 |
X | -0.007 |
Token | big |
Token ID | 1263 |
Feature activation | -8.03e-03 |
Pos logit contributions | |
ls | +0.021 |
Run | +0.020 |
P | +0.019 |
mun | +0.019 |
bitten | +0.017 |
Neg logit contributions | |
ine | -0.029 |
hip | -0.028 |
Office | -0.028 |
birthday | -0.027 |
uries | -0.027 |
Token | rock |
Token ID | 3881 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
ls | +0.009 |
Run | +0.009 |
P | +0.007 |
mun | +0.007 |
slapped | +0.007 |
Neg logit contributions | |
uries | -0.017 |
ine | -0.017 |
hip | -0.017 |
Office | -0.016 |
birthday | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -2.79e-04 |
Pos logit contributions | |
mun | +0.037 |
ls | +0.036 |
Run | +0.035 |
P | +0.032 |
Sorry | +0.032 |
Neg logit contributions | |
ine | -0.021 |
eye | -0.018 |
X | -0.018 |
hip | -0.018 |
birthday | -0.017 |
Token | Jane |
Token ID | 12091 |
Feature activation | -9.45e-03 |
Pos logit contributions | |
Run | +0.000 |
ls | +0.000 |
mun | +0.000 |
Hey | +0.000 |
bitten | +0.000 |
Neg logit contributions | |
ine | -0.000 |
birthday | -0.000 |
hip | -0.000 |
Office | -0.000 |
Jupiter | -0.000 |
Token | got |
Token ID | 1392 |
Feature activation | -3.62e-02 |
Pos logit contributions | |
Run | +0.015 |
ls | +0.015 |
mun | +0.015 |
P | +0.014 |
Hey | +0.013 |
Neg logit contributions | |
ine | -0.017 |
birthday | -0.014 |
Office | -0.014 |
X | -0.014 |
desk | -0.014 |
Token | scared |
Token ID | 12008 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
mun | +0.063 |
Run | +0.060 |
ls | +0.060 |
awed | +0.057 |
P | +0.054 |
Neg logit contributions | |
ine | -0.050 |
for | -0.049 |
birthday | -0.048 |
eye | -0.048 |
name | -0.047 |
Token | and |
Token ID | 290 |
Feature activation | -3.69e-02 |
Pos logit contributions | |
ls | +0.025 |
Run | +0.025 |
mun | +0.024 |
P | +0.022 |
Hey | +0.022 |
Neg logit contributions | |
ine | -0.020 |
birthday | -0.018 |
Office | -0.018 |
hip | -0.017 |
X | -0.017 |
Token | started |
Token ID | 2067 |
Feature activation | -3.62e-02 |
Pos logit contributions | |
mun | +0.058 |
P | +0.054 |
ls | +0.054 |
� | +0.050 |
Run | +0.049 |
Neg logit contributions | |
ine | -0.062 |
sleep | -0.056 |
for | -0.055 |
imagine | -0.054 |
desk | -0.053 |
Token | to |
Token ID | 284 |
Feature activation | -4.19e-02 |
Pos logit contributions | |
Run | +0.081 |
ls | +0.079 |
Hey | +0.077 |
mun | +0.077 |
P | +0.075 |
Neg logit contributions | |
ine | -0.045 |
birthday | -0.034 |
sleep | -0.033 |
desk | -0.033 |
pattern | -0.033 |
Token | cry |
Token ID | 3960 |
Feature activation | -3.71e-03 |
Pos logit contributions | |
ls | +0.079 |
Run | +0.073 |
mun | +0.073 |
P | +0.071 |
led | +0.069 |
Neg logit contributions | |
ine | -0.065 |
birthday | -0.058 |
sleep | -0.056 |
X | -0.055 |
eye | -0.055 |
Token | were |
Token ID | 547 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
ine | +0.004 |
birthday | +0.004 |
hip | +0.004 |
Office | +0.004 |
X | +0.004 |
Neg logit contributions | |
Run | -0.004 |
ls | -0.003 |
mun | -0.003 |
P | -0.003 |
Hey | -0.003 |
Token | not |
Token ID | 407 |
Feature activation | -3.78e-03 |
Pos logit contributions | |
ine | +0.042 |
birthday | +0.039 |
hip | +0.037 |
Office | +0.037 |
X | +0.037 |
Neg logit contributions | |
ls | -0.048 |
Run | -0.048 |
mun | -0.047 |
P | -0.043 |
bitten | -0.042 |
Token | ashamed |
Token ID | 22461 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
Run | +0.006 |
ls | +0.005 |
mun | +0.005 |
P | +0.005 |
Hey | +0.005 |
Neg logit contributions | |
ine | -0.007 |
birthday | -0.007 |
Jupiter | -0.006 |
Office | -0.006 |
hip | -0.006 |
Token | anymore |
Token ID | 7471 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
Run | +0.030 |
mun | +0.030 |
P | +0.029 |
ls | +0.029 |
awed | +0.027 |
Neg logit contributions | |
birthday | -0.018 |
ine | -0.016 |
end | -0.015 |
for | -0.015 |
about | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +6.23e-03 |
Pos logit contributions | |
ls | +0.021 |
Run | +0.021 |
mun | +0.020 |
P | +0.018 |
bitten | +0.018 |
Neg logit contributions | |
ine | -0.024 |
birthday | -0.022 |
hip | -0.021 |
Office | -0.021 |
X | -0.021 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -3.70e-02 |
Pos logit contributions | |
ine | +0.012 |
birthday | +0.011 |
hip | +0.010 |
X | +0.010 |
Office | +0.010 |
Neg logit contributions | |
Run | -0.009 |
ls | -0.009 |
mun | -0.009 |
P | -0.008 |
Hey | -0.008 |
Token | L |
Token ID | 43 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
Run | +0.055 |
ls | +0.054 |
mun | +0.050 |
slapped | +0.047 |
led | +0.047 |
Neg logit contributions | |
birthday | -0.064 |
Today | -0.062 |
ine | -0.062 |
Office | -0.059 |
moon | -0.059 |
Token | ily |
Token ID | 813 |
Feature activation | +6.76e-03 |
Pos logit contributions | |
ls | +0.031 |
Run | +0.030 |
mun | +0.029 |
P | +0.027 |
Hey | +0.026 |
Neg logit contributions | |
ine | -0.029 |
birthday | -0.028 |
hip | -0.027 |
Office | -0.027 |
Jupiter | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
ine | +0.011 |
birthday | +0.010 |
hip | +0.010 |
Office | +0.010 |
Jupiter | +0.010 |
Neg logit contributions | |
ls | -0.011 |
Run | -0.011 |
mun | -0.011 |
P | -0.010 |
Hey | -0.010 |
Token | Tom |
Token ID | 4186 |
Feature activation | -8.07e-05 |
Pos logit contributions | |
ls | +0.032 |
mun | +0.031 |
Run | +0.030 |
led | +0.028 |
awed | +0.028 |
Neg logit contributions | |
birthday | -0.020 |
ine | -0.019 |
X | -0.017 |
hip | -0.017 |
eye | -0.016 |
Token | were |
Token ID | 547 |
Feature activation | -2.39e-02 |
Pos logit contributions | |
ls | +0.000 |
Run | +0.000 |
mun | +0.000 |
P | +0.000 |
Hey | +0.000 |
Neg logit contributions | |
ine | -0.000 |
birthday | -0.000 |
hip | -0.000 |
Office | -0.000 |
X | -0.000 |
Token | girl |
Token ID | 2576 |
Feature activation | -3.89e-03 |
Pos logit contributions | |
ine | +0.009 |
birthday | +0.008 |
hip | +0.008 |
Office | +0.008 |
X | +0.008 |
Neg logit contributions | |
Run | -0.010 |
ls | -0.010 |
mun | -0.009 |
P | -0.009 |
Hey | -0.008 |
Token | was |
Token ID | 373 |
Feature activation | -3.08e-03 |
Pos logit contributions | |
ls | +0.005 |
Run | +0.005 |
mun | +0.005 |
P | +0.005 |
Hey | +0.004 |
Neg logit contributions | |
ine | -0.008 |
birthday | -0.007 |
Office | -0.007 |
hip | -0.007 |
X | -0.007 |
Token | scared |
Token ID | 12008 |
Feature activation | -1.50e-02 |
Pos logit contributions | |
Run | +0.003 |
ls | +0.003 |
mun | +0.003 |
bitten | +0.003 |
P | +0.003 |
Neg logit contributions | |
ine | -0.007 |
birthday | -0.007 |
hip | -0.006 |
Office | -0.006 |
X | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | -3.25e-02 |
Pos logit contributions | |
ls | +0.028 |
Run | +0.027 |
mun | +0.026 |
Hey | +0.023 |
bitten | +0.023 |
Neg logit contributions | |
ine | -0.023 |
birthday | -0.021 |
Office | -0.021 |
hip | -0.020 |
X | -0.020 |
Token | began |
Token ID | 2540 |
Feature activation | -2.78e-02 |
Pos logit contributions | |
ls | +0.049 |
mun | +0.048 |
P | +0.048 |
Run | +0.046 |
Hey | +0.042 |
Neg logit contributions | |
ine | -0.058 |
Jupiter | -0.050 |
birthday | -0.049 |
for | -0.049 |
end | -0.048 |
Token | to |
Token ID | 284 |
Feature activation | -3.52e-02 |
Pos logit contributions | |
Run | +0.054 |
ls | +0.053 |
mun | +0.051 |
Hey | +0.049 |
P | +0.048 |
Neg logit contributions | |
ine | -0.042 |
birthday | -0.037 |
Jupiter | -0.033 |
Office | -0.033 |
hip | -0.033 |
Token | back |
Token ID | 736 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
ls | +0.061 |
Run | +0.060 |
mun | +0.058 |
P | +0.054 |
led | +0.051 |
Neg logit contributions | |
ine | -0.058 |
birthday | -0.054 |
hip | -0.050 |
Office | -0.050 |
X | -0.050 |
Token | away |
Token ID | 1497 |
Feature activation | -9.22e-03 |
Pos logit contributions | |
ls | +0.020 |
Run | +0.020 |
mun | +0.019 |
P | +0.017 |
bitten | +0.017 |
Neg logit contributions | |
ine | -0.017 |
birthday | -0.016 |
hip | -0.015 |
Office | -0.015 |
X | -0.014 |
Token | slowly |
Token ID | 6364 |
Feature activation | +7.37e-03 |
Pos logit contributions | |
Run | +0.016 |
ls | +0.016 |
mun | +0.015 |
P | +0.014 |
Hey | +0.013 |
Neg logit contributions | |
ine | -0.015 |
birthday | -0.014 |
hip | -0.013 |
Office | -0.013 |
X | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -2.84e-04 |
Pos logit contributions | |
ine | +0.013 |
birthday | +0.012 |
Office | +0.011 |
hip | +0.011 |
Jupiter | +0.011 |
Neg logit contributions | |
ls | -0.012 |
Run | -0.012 |
mun | -0.011 |
bitten | -0.010 |
P | -0.010 |
Token | She |
Token ID | 1375 |
Feature activation | -6.75e-03 |
Pos logit contributions | |
Run | +0.000 |
ls | +0.000 |
mun | +0.000 |
Hey | +0.000 |
distracted | +0.000 |
Neg logit contributions | |
ine | -0.000 |
birthday | -0.000 |
X | -0.000 |
hip | -0.000 |
urger | -0.000 |