Token | left |
Token ID | 1364 |
Feature activation | +3.32e-02 |
Pos logit contributions | |
ble | +0.001 |
bled | +0.000 |
ster | +0.000 |
balls | +0.000 |
fin | +0.000 |
Neg logit contributions | |
rell | -0.000 |
Karen | -0.000 |
aren | -0.000 |
ird | -0.000 |
Ell | -0.000 |
Token | this |
Token ID | 428 |
Feature activation | +6.12e-02 |
Pos logit contributions | |
ble | +0.070 |
bled | +0.064 |
balls | +0.061 |
ster | +0.058 |
fin | +0.057 |
Neg logit contributions | |
ird | -0.060 |
Karen | -0.055 |
aren | -0.055 |
rell | -0.054 |
Stella | -0.052 |
Token | shark |
Token ID | 21027 |
Feature activation | +6.21e-02 |
Pos logit contributions | |
bled | +0.091 |
ble | +0.085 |
balls | +0.077 |
ster | +0.069 |
Miles | +0.067 |
Neg logit contributions | |
ird | -0.130 |
hell | -0.121 |
rell | -0.119 |
he | -0.117 |
chair | -0.114 |
Token | in |
Token ID | 287 |
Feature activation | +2.84e-02 |
Pos logit contributions | |
ble | +0.143 |
bled | +0.143 |
balls | +0.131 |
fin | +0.119 |
ster | +0.117 |
Neg logit contributions | |
ird | -0.095 |
he | -0.075 |
old | -0.074 |
rell | -0.071 |
aren | -0.070 |
Token | the |
Token ID | 262 |
Feature activation | +4.22e-02 |
Pos logit contributions | |
ble | +0.063 |
bled | +0.059 |
balls | +0.056 |
ster | +0.053 |
fin | +0.050 |
Neg logit contributions | |
ird | -0.050 |
Karen | -0.045 |
rell | -0.045 |
aren | -0.044 |
Ellen | -0.043 |
Token | bath |
Token ID | 7837 |
Feature activation | +9.42e-02 |
Pos logit contributions | |
ble | +0.110 |
bled | +0.107 |
balls | +0.097 |
ster | +0.093 |
Miles | +0.088 |
Neg logit contributions | |
rell | -0.052 |
ird | -0.052 |
Eve | -0.046 |
chair | -0.040 |
he | -0.039 |
Token | tub |
Token ID | 37995 |
Feature activation | +5.23e-02 |
Pos logit contributions | |
ble | +0.135 |
bled | +0.130 |
balls | +0.115 |
fin | +0.106 |
ster | +0.105 |
Neg logit contributions | |
ird | -0.220 |
rell | -0.199 |
he | -0.191 |
Eve | -0.185 |
ives | -0.181 |
Token | ?" |
Token ID | 1701 |
Feature activation | +4.89e-02 |
Pos logit contributions | |
ble | +0.118 |
bled | +0.115 |
balls | +0.108 |
ster | +0.101 |
fin | +0.099 |
Neg logit contributions | |
ird | -0.086 |
rell | -0.072 |
aren | -0.070 |
Karen | -0.069 |
old | -0.066 |
Token | Sara |
Token ID | 24799 |
Feature activation | -1.90e-04 |
Pos logit contributions | |
bled | +0.139 |
ble | +0.137 |
balls | +0.127 |
ster | +0.121 |
bing | +0.114 |
Neg logit contributions | |
ird | -0.049 |
Karen | -0.047 |
Stella | -0.042 |
Eve | -0.041 |
aren | -0.040 |
Token | and |
Token ID | 290 |
Feature activation | +4.39e-02 |
Pos logit contributions | |
ird | +0.000 |
rell | +0.000 |
aren | +0.000 |
Karen | +0.000 |
Ellen | +0.000 |
Neg logit contributions | |
ble | -0.000 |
bled | -0.000 |
balls | -0.000 |
ster | -0.000 |
fin | -0.000 |
Token | Ben |
Token ID | 3932 |
Feature activation | -2.47e-02 |
Pos logit contributions | |
ble | +0.089 |
bled | +0.085 |
balls | +0.078 |
ster | +0.072 |
fin | +0.069 |
Neg logit contributions | |
ird | -0.079 |
rell | -0.076 |
Karen | -0.076 |
aren | -0.073 |
Kathy | -0.072 |
Token |
|
Token ID | 198 |
Feature activation | +5.34e-03 |
Pos logit contributions | |
ird | +0.027 |
rell | +0.025 |
aren | +0.024 |
Karen | +0.023 |
Ellen | +0.023 |
Neg logit contributions | |
ble | -0.028 |
bled | -0.024 |
balls | -0.023 |
emo | -0.022 |
fin | -0.021 |
Token | As |
Token ID | 1722 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
ble | +0.013 |
bled | +0.011 |
ster | +0.011 |
balls | +0.011 |
fin | +0.011 |
Neg logit contributions | |
ird | -0.009 |
Karen | -0.008 |
rell | -0.008 |
aren | -0.008 |
Ell | -0.007 |
Token | they |
Token ID | 484 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
ble | +0.042 |
bled | +0.038 |
ster | +0.037 |
balls | +0.036 |
fin | +0.035 |
Neg logit contributions | |
Karen | -0.033 |
ird | -0.032 |
aren | -0.031 |
rell | -0.030 |
Stella | -0.029 |
Token | buried |
Token ID | 11694 |
Feature activation | +2.96e-02 |
Pos logit contributions | |
ble | +0.029 |
bled | +0.026 |
balls | +0.025 |
ster | +0.025 |
fin | +0.024 |
Neg logit contributions | |
ird | -0.013 |
rell | -0.012 |
Karen | -0.011 |
aren | -0.011 |
hell | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | +4.41e-02 |
Pos logit contributions | |
ble | +0.072 |
bled | +0.065 |
ster | +0.065 |
balls | +0.063 |
fin | +0.060 |
Neg logit contributions | |
Karen | -0.041 |
Ellie | -0.037 |
rell | -0.037 |
Stella | -0.037 |
Ellen | -0.036 |
Token | ham |
Token ID | 8891 |
Feature activation | +9.08e-02 |
Pos logit contributions | |
bled | +0.092 |
ble | +0.091 |
balls | +0.078 |
Miles | +0.072 |
fin | +0.072 |
Neg logit contributions | |
hell | -0.070 |
ird | -0.065 |
rell | -0.065 |
aback | -0.061 |
Stella | -0.061 |
Token | ster |
Token ID | 1706 |
Feature activation | +4.18e-02 |
Pos logit contributions | |
bled | +0.082 |
ble | +0.078 |
surfing | +0.073 |
fin | +0.065 |
skating | +0.064 |
Neg logit contributions | |
hell | -0.213 |
ives | -0.203 |
ird | -0.200 |
bars | -0.195 |
hel | -0.186 |
Token | , |
Token ID | 11 |
Feature activation | +8.13e-03 |
Pos logit contributions | |
ble | +0.105 |
bled | +0.100 |
balls | +0.093 |
fin | +0.090 |
emo | +0.090 |
Neg logit contributions | |
ird | -0.051 |
rell | -0.047 |
hell | -0.043 |
Karen | -0.042 |
chair | -0.040 |
Token | Tim |
Token ID | 5045 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
ble | +0.020 |
bled | +0.018 |
ster | +0.018 |
balls | +0.017 |
fin | +0.017 |
Neg logit contributions | |
Karen | -0.012 |
ird | -0.012 |
aren | -0.011 |
rell | -0.011 |
Stella | -0.011 |
Token | my |
Token ID | 1820 |
Feature activation | +4.23e-02 |
Pos logit contributions | |
ble | +0.032 |
bled | +0.029 |
ster | +0.029 |
balls | +0.028 |
fin | +0.028 |
Neg logit contributions | |
ird | -0.012 |
Karen | -0.011 |
aren | -0.010 |
rell | -0.010 |
Stella | -0.009 |
Token | felt |
Token ID | 2936 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
ble | +0.108 |
bled | +0.101 |
balls | +0.096 |
ster | +0.088 |
fin | +0.087 |
Neg logit contributions | |
ird | -0.058 |
rell | -0.048 |
old | -0.046 |
Karen | -0.046 |
aren | -0.045 |
Token | One |
Token ID | 3198 |
Feature activation | +3.61e-02 |
Pos logit contributions | |
ble | +0.017 |
bled | +0.015 |
balls | +0.015 |
ster | +0.014 |
fin | +0.014 |
Neg logit contributions | |
ird | -0.009 |
rell | -0.008 |
aren | -0.008 |
Karen | -0.008 |
Ellen | -0.007 |
Token | sunny |
Token ID | 27737 |
Feature activation | +6.45e-02 |
Pos logit contributions | |
bled | +0.103 |
balls | +0.098 |
ble | +0.097 |
ster | +0.086 |
bing | +0.085 |
Neg logit contributions | |
old | -0.033 |
ird | -0.030 |
he | -0.029 |
chair | -0.029 |
Karen | -0.029 |
Token | day |
Token ID | 1110 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
bled | +0.180 |
balls | +0.179 |
ble | +0.173 |
ster | +0.155 |
fin | +0.149 |
Neg logit contributions | |
aren | -0.051 |
rell | -0.048 |
ird | -0.046 |
chair | -0.045 |
Karen | -0.045 |
Token | , |
Token ID | 11 |
Feature activation | +1.28e-02 |
Pos logit contributions | |
ble | +0.080 |
bled | +0.075 |
balls | +0.071 |
ster | +0.071 |
fin | +0.069 |
Neg logit contributions | |
ird | -0.026 |
Karen | -0.023 |
rell | -0.022 |
aren | -0.020 |
Stella | -0.020 |
Token | the |
Token ID | 262 |
Feature activation | +5.44e-02 |
Pos logit contributions | |
ble | +0.034 |
bled | +0.032 |
ster | +0.031 |
balls | +0.030 |
fin | +0.029 |
Neg logit contributions | |
Karen | -0.015 |
Stella | -0.014 |
ird | -0.014 |
Alice | -0.013 |
Ellie | -0.013 |
Token | oy |
Token ID | 35104 |
Feature activation | +9.01e-02 |
Pos logit contributions | |
bled | +0.121 |
balls | +0.118 |
ble | +0.112 |
bing | +0.100 |
zes | +0.097 |
Neg logit contributions | |
hell | -0.065 |
Eve | -0.061 |
aren | -0.059 |
Mad | -0.059 |
Stella | -0.059 |
Token | ster |
Token ID | 1706 |
Feature activation | +6.07e-02 |
Pos logit contributions | |
bled | +0.096 |
balls | +0.080 |
bing | +0.075 |
sing | +0.075 |
ble | +0.075 |
Neg logit contributions | |
ird | -0.198 |
hell | -0.188 |
old | -0.183 |
ives | -0.181 |
mie | -0.178 |
Token | saw |
Token ID | 2497 |
Feature activation | +3.40e-02 |
Pos logit contributions | |
ble | +0.166 |
bled | +0.159 |
balls | +0.150 |
fin | +0.137 |
ster | +0.135 |
Neg logit contributions | |
ird | -0.073 |
old | -0.059 |
rell | -0.059 |
aren | -0.056 |
Karen | -0.054 |
Token | a |
Token ID | 257 |
Feature activation | +2.94e-02 |
Pos logit contributions | |
ble | +0.087 |
balls | +0.083 |
bled | +0.081 |
ster | +0.079 |
fin | +0.075 |
Neg logit contributions | |
Karen | -0.034 |
ird | -0.034 |
rell | -0.032 |
old | -0.032 |
Stella | -0.030 |
Token | shiny |
Token ID | 22441 |
Feature activation | +3.87e-02 |
Pos logit contributions | |
bled | +0.079 |
ble | +0.074 |
balls | +0.071 |
bing | +0.065 |
ster | +0.062 |
Neg logit contributions | |
ird | -0.031 |
Karen | -0.029 |
hell | -0.028 |
aren | -0.028 |
ha | -0.027 |
Token | pearl |
Token ID | 43836 |
Feature activation | +4.69e-02 |
Pos logit contributions | |
bled | +0.077 |
balls | +0.076 |
ble | +0.068 |
Miles | +0.062 |
bing | +0.058 |
Neg logit contributions | |
hell | -0.063 |
ird | -0.063 |
aren | -0.059 |
rell | -0.059 |
Karen | -0.059 |
Token | little |
Token ID | 1310 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
bled | +0.074 |
ble | +0.069 |
balls | +0.066 |
bing | +0.060 |
urg | +0.057 |
Neg logit contributions | |
ird | -0.026 |
Karen | -0.024 |
Stella | -0.023 |
aren | -0.022 |
ha | -0.021 |
Token | oy |
Token ID | 35104 |
Feature activation | +8.05e-02 |
Pos logit contributions | |
bled | +0.065 |
ble | +0.064 |
balls | +0.057 |
ster | +0.048 |
bing | +0.047 |
Neg logit contributions | |
ird | -0.062 |
aren | -0.060 |
Karen | -0.060 |
Stella | -0.057 |
rell | -0.057 |
Token | ster |
Token ID | 1706 |
Feature activation | +3.33e-02 |
Pos logit contributions | |
bled | +0.091 |
ble | +0.065 |
bing | +0.059 |
balls | +0.057 |
zes | +0.049 |
Neg logit contributions | |
ird | -0.195 |
hell | -0.187 |
aren | -0.170 |
old | -0.167 |
ha | -0.165 |
Token | . |
Token ID | 13 |
Feature activation | -3.73e-03 |
Pos logit contributions | |
bled | +0.099 |
ble | +0.094 |
balls | +0.090 |
fin | +0.080 |
Miles | +0.080 |
Neg logit contributions | |
ird | -0.030 |
hell | -0.022 |
old | -0.022 |
chair | -0.021 |
Karen | -0.020 |
Token | This |
Token ID | 770 |
Feature activation | +7.36e-02 |
Pos logit contributions | |
ird | +0.006 |
Karen | +0.005 |
rell | +0.005 |
aren | +0.005 |
Ellen | +0.005 |
Neg logit contributions | |
ble | -0.009 |
bled | -0.008 |
balls | -0.008 |
ster | -0.008 |
fin | -0.007 |
Token | oy |
Token ID | 35104 |
Feature activation | +8.96e-02 |
Pos logit contributions | |
bled | +0.138 |
balls | +0.132 |
ble | +0.129 |
bing | +0.113 |
bling | +0.102 |
Neg logit contributions | |
hell | -0.126 |
ird | -0.123 |
ha | -0.123 |
aren | -0.119 |
rell | -0.118 |
Token | ster |
Token ID | 1706 |
Feature activation | +4.24e-02 |
Pos logit contributions | |
bled | +0.087 |
ble | +0.068 |
balls | +0.065 |
bing | +0.060 |
fin | +0.059 |
Neg logit contributions | |
ird | -0.217 |
hell | -0.216 |
aren | -0.203 |
old | -0.200 |
mie | -0.195 |
Token | was |
Token ID | 373 |
Feature activation | +2.49e-02 |
Pos logit contributions | |
ble | +0.105 |
bled | +0.104 |
balls | +0.096 |
fin | +0.089 |
diving | +0.088 |
Neg logit contributions | |
ird | -0.055 |
hell | -0.049 |
chair | -0.047 |
old | -0.047 |
Karen | -0.047 |
Token | very |
Token ID | 845 |
Feature activation | +4.64e-02 |
Pos logit contributions | |
ble | +0.064 |
balls | +0.063 |
ster | +0.063 |
bled | +0.062 |
bing | +0.059 |
Neg logit contributions | |
Ellie | -0.021 |
seen | -0.020 |
Karen | -0.020 |
Stella | -0.019 |
rell | -0.017 |
Token | happy |
Token ID | 3772 |
Feature activation | +3.44e-02 |
Pos logit contributions | |
bled | +0.121 |
ble | +0.120 |
balls | +0.115 |
bing | +0.110 |
ster | +0.109 |
Neg logit contributions | |
hell | -0.039 |
ird | -0.039 |
old | -0.037 |
know | -0.035 |
Karen | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -3.27e-03 |
Pos logit contributions | |
bled | +0.103 |
ble | +0.101 |
balls | +0.090 |
ster | +0.088 |
fin | +0.085 |
Neg logit contributions | |
ird | -0.024 |
Karen | -0.022 |
chair | -0.020 |
Ellie | -0.020 |
rell | -0.019 |
Token | they |
Token ID | 484 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
ble | +0.051 |
bled | +0.048 |
ster | +0.044 |
balls | +0.043 |
fin | +0.042 |
Neg logit contributions | |
ird | -0.034 |
Karen | -0.034 |
rell | -0.032 |
Ellie | -0.032 |
Ellen | -0.031 |
Token | needed |
Token ID | 2622 |
Feature activation | +3.50e-03 |
Pos logit contributions | |
ird | +0.026 |
rell | +0.024 |
aren | +0.024 |
Karen | +0.024 |
Ellen | +0.022 |
Neg logit contributions | |
ble | -0.032 |
bled | -0.029 |
balls | -0.027 |
ster | -0.026 |
fin | -0.026 |
Token | to |
Token ID | 284 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
ble | +0.008 |
bled | +0.007 |
ster | +0.007 |
balls | +0.007 |
fin | +0.006 |
Neg logit contributions | |
ird | -0.006 |
Karen | -0.006 |
rell | -0.005 |
aren | -0.005 |
Stella | -0.005 |
Token | put |
Token ID | 1234 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
ble | +0.036 |
bled | +0.034 |
balls | +0.033 |
ster | +0.031 |
fin | +0.029 |
Neg logit contributions | |
ird | -0.031 |
Karen | -0.027 |
rell | -0.027 |
aren | -0.026 |
Stella | -0.025 |
Token | the |
Token ID | 262 |
Feature activation | +5.05e-02 |
Pos logit contributions | |
ble | +0.024 |
bled | +0.022 |
balls | +0.021 |
ster | +0.020 |
fin | +0.020 |
Neg logit contributions | |
ird | -0.020 |
rell | -0.019 |
Karen | -0.019 |
aren | -0.018 |
Stella | -0.017 |
Token | ham |
Token ID | 8891 |
Feature activation | +8.89e-02 |
Pos logit contributions | |
ble | +0.103 |
bled | +0.102 |
balls | +0.087 |
bing | +0.079 |
diving | +0.077 |
Neg logit contributions | |
hell | -0.081 |
ird | -0.078 |
rell | -0.074 |
Stella | -0.073 |
aback | -0.073 |
Token | ster |
Token ID | 1706 |
Feature activation | +4.31e-02 |
Pos logit contributions | |
bled | +0.080 |
ble | +0.077 |
diving | +0.065 |
surfing | +0.059 |
Miles | +0.057 |
Neg logit contributions | |
hell | -0.224 |
ird | -0.221 |
ives | -0.211 |
bars | -0.195 |
he | -0.194 |
Token | in |
Token ID | 287 |
Feature activation | -2.31e-04 |
Pos logit contributions | |
ble | +0.084 |
bled | +0.077 |
balls | +0.072 |
fin | +0.068 |
ster | +0.067 |
Neg logit contributions | |
ird | -0.081 |
rell | -0.073 |
hell | -0.072 |
Karen | -0.068 |
aren | -0.068 |
Token | a |
Token ID | 257 |
Feature activation | -4.08e-03 |
Pos logit contributions | |
ird | +0.000 |
rell | +0.000 |
aren | +0.000 |
Karen | +0.000 |
Ellen | +0.000 |
Neg logit contributions | |
ble | -0.000 |
bled | -0.000 |
fin | -0.000 |
balls | -0.000 |
ster | -0.000 |
Token | special |
Token ID | 2041 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
ird | +0.008 |
rell | +0.007 |
Karen | +0.007 |
aren | +0.007 |
Ellen | +0.007 |
Neg logit contributions | |
ble | -0.008 |
bled | -0.008 |
balls | -0.007 |
ster | -0.007 |
fin | -0.007 |
Token | place |
Token ID | 1295 |
Feature activation | +4.17e-03 |
Pos logit contributions | |
ble | +0.032 |
bled | +0.029 |
balls | +0.027 |
ster | +0.026 |
fin | +0.026 |
Neg logit contributions | |
ird | -0.028 |
aren | -0.025 |
rell | -0.025 |
Karen | -0.025 |
Ellen | -0.024 |
Token | understood |
Token ID | 7247 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
ble | +0.008 |
bled | +0.007 |
balls | +0.007 |
ster | +0.007 |
fin | +0.007 |
Neg logit contributions | |
ird | -0.006 |
rell | -0.006 |
Karen | -0.006 |
aren | -0.006 |
Ellen | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +5.73e-03 |
Pos logit contributions | |
ble | +0.030 |
bled | +0.027 |
balls | +0.026 |
ster | +0.025 |
fin | +0.025 |
Neg logit contributions | |
ird | -0.024 |
Karen | -0.022 |
aren | -0.022 |
rell | -0.022 |
Ellen | -0.020 |
Token | He |
Token ID | 679 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
ble | +0.011 |
bled | +0.010 |
balls | +0.009 |
ster | +0.009 |
fin | +0.009 |
Neg logit contributions | |
ird | -0.012 |
rell | -0.011 |
Karen | -0.011 |
aren | -0.011 |
Ellen | -0.010 |
Token | opened |
Token ID | 4721 |
Feature activation | +7.12e-03 |
Pos logit contributions | |
ble | +0.028 |
bled | +0.025 |
balls | +0.024 |
ster | +0.023 |
fin | +0.023 |
Neg logit contributions | |
ird | -0.021 |
rell | -0.019 |
Karen | -0.019 |
aren | -0.019 |
Ellen | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | +3.56e-02 |
Pos logit contributions | |
ble | +0.013 |
bled | +0.012 |
balls | +0.011 |
ster | +0.011 |
fin | +0.010 |
Neg logit contributions | |
ird | -0.015 |
rell | -0.014 |
Karen | -0.014 |
aren | -0.014 |
Ellen | -0.013 |
Token | oy |
Token ID | 35104 |
Feature activation | +8.78e-02 |
Pos logit contributions | |
ble | +0.057 |
bled | +0.056 |
balls | +0.054 |
emo | +0.042 |
bing | +0.042 |
Neg logit contributions | |
ird | -0.073 |
rell | -0.072 |
aren | -0.071 |
hell | -0.070 |
Karen | -0.069 |
Token | ster |
Token ID | 1706 |
Feature activation | +3.44e-02 |
Pos logit contributions | |
bled | +0.083 |
ble | +0.071 |
balls | +0.071 |
spitting | +0.056 |
diving | +0.055 |
Neg logit contributions | |
ird | -0.206 |
hell | -0.199 |
land | -0.193 |
ives | -0.191 |
mie | -0.188 |
Token | carefully |
Token ID | 7773 |
Feature activation | +2.65e-03 |
Pos logit contributions | |
ble | +0.083 |
bled | +0.079 |
balls | +0.074 |
fin | +0.068 |
emo | +0.068 |
Neg logit contributions | |
ird | -0.051 |
rell | -0.044 |
aren | -0.043 |
hell | -0.043 |
Karen | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | +4.89e-03 |
Pos logit contributions | |
ble | +0.006 |
bled | +0.006 |
balls | +0.005 |
ster | +0.005 |
fin | +0.005 |
Neg logit contributions | |
ird | -0.004 |
rell | -0.004 |
aren | -0.004 |
Karen | -0.004 |
Ellen | -0.004 |
Token | looked |
Token ID | 3114 |
Feature activation | -3.48e-03 |
Pos logit contributions | |
ble | +0.012 |
bled | +0.010 |
balls | +0.010 |
fin | +0.010 |
ster | +0.010 |
Neg logit contributions | |
ird | -0.008 |
Karen | -0.007 |
aren | -0.007 |
rell | -0.007 |
Stella | -0.006 |
Token | inside |
Token ID | 2641 |
Feature activation | +2.93e-02 |
Pos logit contributions | |
ird | +0.007 |
rell | +0.007 |
Karen | +0.007 |
aren | +0.007 |
Ellen | +0.006 |
Neg logit contributions | |
ble | -0.007 |
bled | -0.006 |
balls | -0.006 |
ster | -0.006 |
fin | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +6.42e-03 |
Pos logit contributions | |
ird | +0.036 |
rell | +0.034 |
aren | +0.032 |
Karen | +0.031 |
Ellen | +0.030 |
Neg logit contributions | |
ble | -0.030 |
balls | -0.024 |
bled | -0.024 |
emo | -0.023 |
diving | -0.022 |
Token | One |
Token ID | 3198 |
Feature activation | +3.31e-02 |
Pos logit contributions | |
ble | +0.017 |
bled | +0.016 |
balls | +0.015 |
ster | +0.015 |
fin | +0.015 |
Neg logit contributions | |
ird | -0.008 |
rell | -0.008 |
Karen | -0.008 |
aren | -0.008 |
Ellen | -0.007 |
Token | day |
Token ID | 1110 |
Feature activation | +2.31e-02 |
Pos logit contributions | |
bled | +0.095 |
balls | +0.090 |
ble | +0.089 |
ster | +0.078 |
bing | +0.075 |
Neg logit contributions | |
old | -0.031 |
Karen | -0.029 |
aren | -0.029 |
ird | -0.028 |
chair | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | +4.40e-03 |
Pos logit contributions | |
ble | +0.062 |
bled | +0.057 |
balls | +0.055 |
ster | +0.054 |
fin | +0.053 |
Neg logit contributions | |
ird | -0.029 |
Karen | -0.027 |
rell | -0.026 |
aren | -0.025 |
Stella | -0.024 |
Token | the |
Token ID | 262 |
Feature activation | +5.07e-02 |
Pos logit contributions | |
ble | +0.011 |
bled | +0.010 |
ster | +0.010 |
balls | +0.010 |
fin | +0.009 |
Neg logit contributions | |
Karen | -0.006 |
ird | -0.006 |
Stella | -0.005 |
Ellen | -0.005 |
Ellie | -0.005 |
Token | oy |
Token ID | 35104 |
Feature activation | +8.70e-02 |
Pos logit contributions | |
bled | +0.099 |
balls | +0.099 |
ble | +0.091 |
bing | +0.080 |
zes | +0.077 |
Neg logit contributions | |
hell | -0.074 |
Stella | -0.072 |
Mad | -0.071 |
aren | -0.070 |
ird | -0.069 |
Token | ster |
Token ID | 1706 |
Feature activation | +5.46e-02 |
Pos logit contributions | |
bled | +0.086 |
balls | +0.077 |
ble | +0.069 |
diving | +0.067 |
sing | +0.065 |
Neg logit contributions | |
ird | -0.194 |
hell | -0.190 |
mie | -0.183 |
old | -0.183 |
aren | -0.182 |
Token | met |
Token ID | 1138 |
Feature activation | +2.20e-02 |
Pos logit contributions | |
ble | +0.145 |
bled | +0.139 |
balls | +0.133 |
fin | +0.120 |
ster | +0.119 |
Neg logit contributions | |
ird | -0.068 |
rell | -0.057 |
old | -0.056 |
aren | -0.055 |
Karen | -0.055 |
Token | a |
Token ID | 257 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
ble | +0.049 |
bled | +0.045 |
balls | +0.043 |
ster | +0.041 |
fin | +0.041 |
Neg logit contributions | |
ird | -0.038 |
Karen | -0.035 |
rell | -0.035 |
aren | -0.035 |
Ellen | -0.032 |
Token | little |
Token ID | 1310 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
ble | +0.026 |
bled | +0.026 |
balls | +0.023 |
ster | +0.022 |
fin | +0.021 |
Neg logit contributions | |
ird | -0.014 |
Karen | -0.014 |
aren | -0.013 |
Stella | -0.013 |
Ellen | -0.012 |
Token | fish |
Token ID | 5916 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
bled | +0.046 |
ble | +0.045 |
balls | +0.040 |
ster | +0.034 |
emo | +0.034 |
Neg logit contributions | |
ird | -0.028 |
Karen | -0.028 |
aren | -0.028 |
Stella | -0.027 |
old | -0.026 |
Token | interesting |
Token ID | 3499 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
Karen | +0.009 |
ird | +0.009 |
hell | +0.009 |
rell | +0.008 |
aren | +0.008 |
Neg logit contributions | |
ble | -0.018 |
bled | -0.018 |
ster | -0.017 |
balls | -0.016 |
fin | -0.015 |
Token | happening |
Token ID | 5836 |
Feature activation | +1.91e-02 |
Pos logit contributions | |
ble | +0.029 |
bled | +0.028 |
balls | +0.026 |
ster | +0.026 |
fin | +0.025 |
Neg logit contributions | |
ird | -0.016 |
Karen | -0.016 |
hell | -0.015 |
rell | -0.015 |
aren | -0.015 |
Token | on |
Token ID | 319 |
Feature activation | +4.32e-02 |
Pos logit contributions | |
ble | +0.042 |
bled | +0.040 |
balls | +0.040 |
ster | +0.038 |
fin | +0.036 |
Neg logit contributions | |
Karen | -0.027 |
aren | -0.026 |
Stella | -0.025 |
rell | -0.025 |
aback | -0.024 |
Token | the |
Token ID | 262 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
ble | +0.103 |
bled | +0.102 |
balls | +0.099 |
ster | +0.095 |
fin | +0.086 |
Neg logit contributions | |
Karen | -0.058 |
rell | -0.054 |
Stella | -0.053 |
ird | -0.052 |
aren | -0.052 |
Token | books |
Token ID | 3835 |
Feature activation | +6.03e-03 |
Pos logit contributions | |
bled | +0.044 |
ble | +0.044 |
balls | +0.038 |
ster | +0.037 |
emo | +0.034 |
Neg logit contributions | |
rell | -0.042 |
ird | -0.038 |
hell | -0.037 |
Eve | -0.036 |
Karen | -0.036 |
Token | he |
Token ID | 258 |
Feature activation | +8.62e-02 |
Pos logit contributions | |
ble | +0.020 |
bled | +0.018 |
balls | +0.017 |
ster | +0.017 |
fin | +0.017 |
Neg logit contributions | |
ird | -0.004 |
Karen | -0.004 |
aren | -0.004 |
rell | -0.004 |
Ellen | -0.003 |
Token | lf |
Token ID | 1652 |
Feature activation | +2.76e-02 |
Pos logit contributions | |
bled | +0.182 |
ble | +0.179 |
ster | +0.164 |
balls | +0.162 |
hyd | +0.158 |
Neg logit contributions | |
chair | -0.110 |
ird | -0.110 |
he | -0.102 |
hell | -0.102 |
rell | -0.095 |
Token | near |
Token ID | 1474 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
ble | +0.070 |
bled | +0.064 |
balls | +0.061 |
ster | +0.060 |
fin | +0.060 |
Neg logit contributions | |
ird | -0.039 |
rell | -0.035 |
Karen | -0.035 |
aren | -0.035 |
Ellen | -0.031 |
Token | the |
Token ID | 262 |
Feature activation | +6.46e-03 |
Pos logit contributions | |
ble | +0.044 |
bled | +0.041 |
balls | +0.040 |
ster | +0.038 |
fin | +0.035 |
Neg logit contributions | |
ird | -0.038 |
Karen | -0.036 |
aren | -0.035 |
rell | -0.035 |
Stella | -0.033 |
Token | river |
Token ID | 7850 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
ble | +0.013 |
bled | +0.012 |
balls | +0.011 |
ster | +0.011 |
fin | +0.010 |
Neg logit contributions | |
ird | -0.012 |
rell | -0.012 |
Karen | -0.011 |
aren | -0.011 |
hell | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | +3.18e-02 |
Pos logit contributions | |
ble | +0.060 |
bled | +0.055 |
balls | +0.052 |
ster | +0.051 |
fin | +0.050 |
Neg logit contributions | |
ird | -0.034 |
rell | -0.031 |
Karen | -0.031 |
aren | -0.031 |
Ellen | -0.028 |
Token | One |
Token ID | 3198 |
Feature activation | +4.05e-02 |
Pos logit contributions | |
ble | +0.071 |
bled | +0.065 |
balls | +0.061 |
ster | +0.060 |
fin | +0.059 |
Neg logit contributions | |
ird | -0.053 |
aren | -0.048 |
Karen | -0.048 |
rell | -0.046 |
Ell | -0.044 |
Token | day |
Token ID | 1110 |
Feature activation | +3.32e-02 |
Pos logit contributions | |
bled | +0.112 |
ble | +0.110 |
balls | +0.106 |
ster | +0.098 |
bing | +0.094 |
Neg logit contributions | |
old | -0.039 |
ird | -0.036 |
hell | -0.036 |
aren | -0.036 |
he | -0.034 |
Token | , |
Token ID | 11 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
ble | +0.088 |
bled | +0.081 |
balls | +0.076 |
ster | +0.076 |
fin | +0.074 |
Neg logit contributions | |
ird | -0.045 |
Karen | -0.041 |
aren | -0.039 |
rell | -0.039 |
Stella | -0.038 |
Token | the |
Token ID | 262 |
Feature activation | +4.81e-02 |
Pos logit contributions | |
ble | +0.055 |
bled | +0.051 |
ster | +0.046 |
balls | +0.046 |
fin | +0.045 |
Neg logit contributions | |
ird | -0.038 |
Karen | -0.038 |
aren | -0.035 |
Stella | -0.035 |
Ellen | -0.034 |
Token | graceful |
Token ID | 44363 |
Feature activation | +6.19e-02 |
Pos logit contributions | |
ble | +0.091 |
bled | +0.086 |
balls | +0.073 |
bing | +0.071 |
ster | +0.068 |
Neg logit contributions | |
ird | -0.082 |
Stella | -0.081 |
aren | -0.079 |
hell | -0.078 |
Karen | -0.077 |
Token | oy |
Token ID | 35104 |
Feature activation | +8.61e-02 |
Pos logit contributions | |
ble | +0.110 |
bled | +0.101 |
balls | +0.088 |
ster | +0.078 |
fin | +0.078 |
Neg logit contributions | |
Stella | -0.119 |
aren | -0.119 |
rell | -0.118 |
ird | -0.118 |
Karen | -0.115 |
Token | ster |
Token ID | 1706 |
Feature activation | +6.38e-02 |
Pos logit contributions | |
bled | +0.083 |
ble | +0.078 |
bing | +0.069 |
fin | +0.063 |
sing | +0.060 |
Neg logit contributions | |
ird | -0.198 |
ives | -0.190 |
hell | -0.188 |
old | -0.184 |
aren | -0.182 |
Token | had |
Token ID | 550 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
ble | +0.157 |
bled | +0.149 |
balls | +0.137 |
fin | +0.125 |
ster | +0.125 |
Neg logit contributions | |
ird | -0.088 |
old | -0.077 |
rell | -0.077 |
aren | -0.074 |
hell | -0.072 |
Token | a |
Token ID | 257 |
Feature activation | +7.57e-03 |
Pos logit contributions | |
ble | +0.026 |
bled | +0.024 |
balls | +0.022 |
ster | +0.022 |
fin | +0.021 |
Neg logit contributions | |
ird | -0.019 |
Karen | -0.018 |
aren | -0.017 |
rell | -0.017 |
Stella | -0.016 |
Token | special |
Token ID | 2041 |
Feature activation | +3.31e-02 |
Pos logit contributions | |
ble | +0.019 |
bled | +0.019 |
balls | +0.017 |
ster | +0.016 |
bing | +0.016 |
Neg logit contributions | |
ird | -0.010 |
Karen | -0.009 |
aren | -0.009 |
Stella | -0.009 |
hell | -0.008 |
Token | treat |
Token ID | 2190 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
ble | +0.070 |
bled | +0.068 |
balls | +0.060 |
ster | +0.057 |
bing | +0.054 |
Neg logit contributions | |
ird | -0.057 |
aren | -0.052 |
Karen | -0.052 |
rell | -0.050 |
Stella | -0.050 |
Token | he |
Token ID | 339 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
ble | +0.045 |
ster | +0.041 |
bled | +0.040 |
fin | +0.038 |
iflower | +0.035 |
Neg logit contributions | |
Karen | -0.034 |
Ellie | -0.032 |
ird | -0.031 |
Stella | -0.029 |
rell | -0.029 |
Token | decided |
Token ID | 3066 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
ble | +0.104 |
bled | +0.094 |
balls | +0.089 |
ster | +0.086 |
fin | +0.081 |
Neg logit contributions | |
ird | -0.055 |
ha | -0.046 |
Karen | -0.044 |
rell | -0.044 |
aren | -0.042 |
Token | to |
Token ID | 284 |
Feature activation | +2.76e-02 |
Pos logit contributions | |
ble | +0.028 |
bled | +0.026 |
balls | +0.024 |
fin | +0.023 |
ster | +0.023 |
Neg logit contributions | |
Karen | -0.016 |
ird | -0.015 |
Stella | -0.015 |
aren | -0.015 |
Ellen | -0.014 |
Token | remove |
Token ID | 4781 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
ble | +0.059 |
bled | +0.055 |
balls | +0.051 |
ster | +0.051 |
fin | +0.048 |
Neg logit contributions | |
ird | -0.051 |
Karen | -0.046 |
rell | -0.046 |
aren | -0.045 |
Ellen | -0.042 |
Token | the |
Token ID | 262 |
Feature activation | +5.40e-02 |
Pos logit contributions | |
ble | +0.042 |
bled | +0.038 |
balls | +0.036 |
ster | +0.035 |
fin | +0.034 |
Neg logit contributions | |
ird | -0.037 |
rell | -0.035 |
Karen | -0.035 |
aren | -0.035 |
Ellen | -0.032 |
Token | seaw |
Token ID | 32085 |
Feature activation | +8.59e-02 |
Pos logit contributions | |
bled | +0.103 |
ble | +0.100 |
balls | +0.091 |
ster | +0.080 |
bing | +0.078 |
Neg logit contributions | |
hell | -0.092 |
rell | -0.087 |
ird | -0.085 |
ha | -0.082 |
aren | -0.080 |
Token | eed |
Token ID | 2308 |
Feature activation | +6.36e-02 |
Pos logit contributions | |
bled | +0.127 |
ble | +0.122 |
balls | +0.102 |
shark | +0.095 |
digging | +0.092 |
Neg logit contributions | |
hell | -0.185 |
ird | -0.180 |
ives | -0.171 |
rell | -0.168 |
ha | -0.157 |
Token | from |
Token ID | 422 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
bled | +0.146 |
ble | +0.143 |
balls | +0.130 |
emo | +0.127 |
fin | +0.124 |
Neg logit contributions | |
hell | -0.079 |
ird | -0.076 |
rell | -0.073 |
aback | -0.068 |
he | -0.068 |
Token | the |
Token ID | 262 |
Feature activation | +5.16e-02 |
Pos logit contributions | |
ble | +0.040 |
ster | +0.037 |
bled | +0.036 |
balls | +0.035 |
fin | +0.032 |
Neg logit contributions | |
Karen | -0.032 |
ird | -0.031 |
rell | -0.030 |
aren | -0.029 |
Stella | -0.028 |
Token | water |
Token ID | 1660 |
Feature activation | +5.66e-02 |
Pos logit contributions | |
bled | +0.104 |
ble | +0.100 |
balls | +0.087 |
bing | +0.081 |
emo | +0.080 |
Neg logit contributions | |
hell | -0.074 |
ird | -0.072 |
rell | -0.072 |
bars | -0.068 |
hy | -0.068 |
Token | so |
Token ID | 523 |
Feature activation | +3.31e-02 |
Pos logit contributions | |
ble | +0.152 |
bled | +0.147 |
balls | +0.134 |
ster | +0.130 |
emo | +0.128 |
Neg logit contributions | |
ird | -0.064 |
rell | -0.058 |
aren | -0.050 |
hell | -0.050 |
Karen | -0.049 |
Token | , |
Token ID | 11 |
Feature activation | +2.93e-02 |
Pos logit contributions | |
ble | +0.168 |
bled | +0.155 |
ster | +0.154 |
fin | +0.145 |
ich | +0.145 |
Neg logit contributions | |
he | -0.072 |
rell | -0.072 |
ird | -0.071 |
Ellie | -0.069 |
hell | -0.068 |
Token | he |
Token ID | 339 |
Feature activation | +5.79e-02 |
Pos logit contributions | |
ble | +0.071 |
ster | +0.067 |
bled | +0.064 |
fin | +0.061 |
balls | +0.061 |
Neg logit contributions | |
Karen | -0.046 |
Ellie | -0.041 |
Stella | -0.038 |
ird | -0.038 |
Ellen | -0.038 |
Token | noticed |
Token ID | 6810 |
Feature activation | +3.09e-02 |
Pos logit contributions | |
ble | +0.158 |
bled | +0.146 |
balls | +0.142 |
ster | +0.130 |
fin | +0.128 |
Neg logit contributions | |
ird | -0.072 |
rell | -0.055 |
old | -0.054 |
Karen | -0.052 |
aren | -0.052 |
Token | that |
Token ID | 326 |
Feature activation | +4.56e-02 |
Pos logit contributions | |
ble | +0.067 |
bled | +0.062 |
balls | +0.060 |
ster | +0.058 |
fin | +0.056 |
Neg logit contributions | |
ird | -0.052 |
Karen | -0.050 |
aren | -0.047 |
rell | -0.047 |
old | -0.045 |
Token | his |
Token ID | 465 |
Feature activation | +8.14e-02 |
Pos logit contributions | |
ble | +0.089 |
bled | +0.083 |
ster | +0.078 |
balls | +0.076 |
fin | +0.068 |
Neg logit contributions | |
Karen | -0.088 |
ird | -0.086 |
Ellie | -0.082 |
Stella | -0.081 |
Alice | -0.081 |
Token | toy |
Token ID | 13373 |
Feature activation | +8.56e-02 |
Pos logit contributions | |
ble | +0.187 |
bled | +0.180 |
bling | +0.149 |
ster | +0.145 |
bing | +0.144 |
Neg logit contributions | |
ird | -0.092 |
hell | -0.092 |
keys | -0.092 |
rell | -0.087 |
chair | -0.086 |
Token | car |
Token ID | 1097 |
Feature activation | +2.57e-02 |
Pos logit contributions | |
ble | +0.194 |
bled | +0.182 |
diving | +0.170 |
emo | +0.164 |
surfing | +0.161 |
Neg logit contributions | |
chairs | -0.093 |
ird | -0.092 |
chair | -0.091 |
rell | -0.089 |
Ellie | -0.086 |
Token | was |
Token ID | 373 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
ble | +0.057 |
bled | +0.051 |
balls | +0.049 |
fin | +0.049 |
ster | +0.048 |
Neg logit contributions | |
ird | -0.044 |
rell | -0.041 |
Karen | -0.040 |
aren | -0.039 |
hell | -0.037 |
Token | missing |
Token ID | 4814 |
Feature activation | +2.20e-02 |
Pos logit contributions | |
ble | +0.046 |
ster | +0.043 |
balls | +0.042 |
bled | +0.041 |
fin | +0.040 |
Neg logit contributions | |
ird | -0.032 |
rell | -0.030 |
Karen | -0.030 |
Stella | -0.027 |
hell | -0.027 |
Token | a |
Token ID | 257 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
ble | +0.058 |
bled | +0.054 |
balls | +0.052 |
ster | +0.051 |
fin | +0.050 |
Neg logit contributions | |
ird | -0.028 |
Karen | -0.024 |
rell | -0.024 |
aren | -0.023 |
hell | -0.022 |
Token | wheel |
Token ID | 7825 |
Feature activation | +4.04e-04 |
Pos logit contributions | |
ble | +0.023 |
bled | +0.022 |
balls | +0.019 |
ster | +0.019 |
fin | +0.018 |
Neg logit contributions | |
ird | -0.018 |
Karen | -0.016 |
rell | -0.015 |
aren | -0.015 |
hell | -0.015 |
Token | them |
Token ID | 606 |
Feature activation | -9.09e-03 |
Pos logit contributions | |
ble | +0.045 |
bled | +0.041 |
balls | +0.038 |
ster | +0.037 |
fin | +0.037 |
Neg logit contributions | |
ird | -0.045 |
rell | -0.042 |
aren | -0.042 |
Karen | -0.042 |
Ellen | -0.039 |
Token | to |
Token ID | 284 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
ird | +0.019 |
rell | +0.018 |
aren | +0.017 |
Karen | +0.017 |
Ellen | +0.016 |
Neg logit contributions | |
ble | -0.017 |
bled | -0.015 |
balls | -0.015 |
ster | -0.014 |
fin | -0.014 |
Token | go |
Token ID | 467 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
ble | +0.035 |
bled | +0.031 |
balls | +0.029 |
ster | +0.029 |
fin | +0.028 |
Neg logit contributions | |
ird | -0.029 |
aren | -0.027 |
rell | -0.027 |
Karen | -0.027 |
Ellen | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -2.60e-03 |
Pos logit contributions | |
ble | +0.026 |
bled | +0.024 |
balls | +0.022 |
ster | +0.022 |
fin | +0.021 |
Neg logit contributions | |
ird | -0.019 |
Karen | -0.018 |
rell | -0.018 |
aren | -0.018 |
Ellen | -0.016 |
Token | The |
Token ID | 383 |
Feature activation | +5.67e-02 |
Pos logit contributions | |
ird | +0.004 |
rell | +0.004 |
Karen | +0.004 |
aren | +0.004 |
Ellen | +0.004 |
Neg logit contributions | |
ble | -0.006 |
bled | -0.006 |
balls | -0.005 |
ster | -0.005 |
fin | -0.005 |
Token | careless |
Token ID | 36138 |
Feature activation | +8.43e-02 |
Pos logit contributions | |
bled | +0.107 |
ble | +0.101 |
balls | +0.090 |
bing | +0.083 |
urg | +0.079 |
Neg logit contributions | |
hell | -0.088 |
ird | -0.088 |
Eve | -0.080 |
ha | -0.079 |
he | -0.078 |
Token | jelly |
Token ID | 27644 |
Feature activation | +5.51e-02 |
Pos logit contributions | |
bled | +0.144 |
ble | +0.133 |
balls | +0.126 |
bing | +0.101 |
fin | +0.097 |
Neg logit contributions | |
hell | -0.161 |
ird | -0.150 |
ha | -0.150 |
Eve | -0.142 |
rell | -0.140 |
Token | fish |
Token ID | 11084 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
ble | +0.060 |
bled | +0.054 |
balls | +0.043 |
fin | +0.039 |
ster | +0.034 |
Neg logit contributions | |
ird | -0.137 |
hell | -0.134 |
ha | -0.130 |
aren | -0.127 |
rell | -0.126 |
Token | said |
Token ID | 531 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
ble | +0.034 |
bled | +0.031 |
balls | +0.029 |
ster | +0.028 |
fin | +0.028 |
Neg logit contributions | |
ird | -0.028 |
Karen | -0.025 |
rell | -0.025 |
aren | -0.025 |
Ellen | -0.024 |
Token | goodbye |
Token ID | 24829 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
ble | +0.038 |
bled | +0.034 |
balls | +0.032 |
fin | +0.032 |
ster | +0.032 |
Neg logit contributions | |
ird | -0.017 |
Karen | -0.016 |
Stella | -0.015 |
Ellen | -0.015 |
Ellie | -0.014 |
Token | to |
Token ID | 284 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
ble | +0.044 |
bled | +0.040 |
balls | +0.037 |
ster | +0.036 |
fin | +0.036 |
Neg logit contributions | |
ird | -0.033 |
Karen | -0.031 |
rell | -0.031 |
aren | -0.030 |
Ellen | -0.028 |
Token | They |
Token ID | 2990 |
Feature activation | -8.85e-03 |
Pos logit contributions | |
ble | +0.038 |
bled | +0.035 |
balls | +0.033 |
ster | +0.033 |
fin | +0.032 |
Neg logit contributions | |
ird | -0.032 |
Karen | -0.031 |
aren | -0.031 |
rell | -0.029 |
Ell | -0.028 |
Token | decide |
Token ID | 5409 |
Feature activation | -7.97e-04 |
Pos logit contributions | |
ird | +0.017 |
rell | +0.017 |
aren | +0.016 |
Karen | +0.016 |
chairs | +0.015 |
Neg logit contributions | |
ble | -0.017 |
bled | -0.015 |
fin | -0.015 |
ster | -0.014 |
balls | -0.014 |
Token | to |
Token ID | 284 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
ird | +0.002 |
Karen | +0.002 |
aren | +0.001 |
rell | +0.001 |
Ellen | +0.001 |
Neg logit contributions | |
ble | -0.002 |
bled | -0.001 |
balls | -0.001 |
ster | -0.001 |
fin | -0.001 |
Token | leave |
Token ID | 2666 |
Feature activation | +4.05e-03 |
Pos logit contributions | |
ble | +0.022 |
bled | +0.020 |
balls | +0.019 |
ster | +0.018 |
fin | +0.018 |
Neg logit contributions | |
ird | -0.020 |
rell | -0.019 |
aren | -0.019 |
Karen | -0.019 |
Ellen | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | +4.71e-02 |
Pos logit contributions | |
ble | +0.007 |
bled | +0.007 |
balls | +0.006 |
fin | +0.006 |
ster | +0.006 |
Neg logit contributions | |
ird | -0.009 |
rell | -0.008 |
aren | -0.008 |
Karen | -0.008 |
Ellen | -0.008 |
Token | oy |
Token ID | 35104 |
Feature activation | +8.39e-02 |
Pos logit contributions | |
bled | +0.089 |
balls | +0.085 |
ble | +0.081 |
Miles | +0.067 |
bing | +0.066 |
Neg logit contributions | |
aren | -0.075 |
Mad | -0.071 |
hell | -0.071 |
ha | -0.070 |
Karen | -0.069 |
Token | ster |
Token ID | 1706 |
Feature activation | +3.51e-02 |
Pos logit contributions | |
bled | +0.085 |
balls | +0.076 |
ble | +0.060 |
Finn | +0.057 |
diving | +0.056 |
Neg logit contributions | |
ird | -0.190 |
hell | -0.179 |
mie | -0.178 |
seen | -0.170 |
land | -0.166 |
Token | alone |
Token ID | 3436 |
Feature activation | +1.98e-02 |
Pos logit contributions | |
ble | +0.076 |
bled | +0.070 |
balls | +0.066 |
ster | +0.062 |
fin | +0.061 |
Neg logit contributions | |
ird | -0.062 |
rell | -0.055 |
aren | -0.055 |
Karen | -0.054 |
hell | -0.052 |
Token | and |
Token ID | 290 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
ble | +0.045 |
bled | +0.041 |
balls | +0.039 |
ster | +0.038 |
fin | +0.038 |
Neg logit contributions | |
ird | -0.034 |
rell | -0.031 |
Karen | -0.031 |
aren | -0.031 |
Ellen | -0.029 |
Token | play |
Token ID | 711 |
Feature activation | +3.56e-02 |
Pos logit contributions | |
ble | +0.043 |
bled | +0.040 |
balls | +0.037 |
ster | +0.036 |
fin | +0.035 |
Neg logit contributions | |
ird | -0.028 |
Karen | -0.026 |
aren | -0.025 |
rell | -0.024 |
Stella | -0.024 |
Token | with |
Token ID | 351 |
Feature activation | +7.66e-03 |
Pos logit contributions | |
ble | +0.073 |
bled | +0.069 |
balls | +0.064 |
ster | +0.061 |
fin | +0.058 |
Neg logit contributions | |
ird | -0.059 |
hell | -0.054 |
Stella | -0.049 |
aback | -0.049 |
Karen | -0.049 |
Token | play |
Token ID | 711 |
Feature activation | +3.21e-02 |
Pos logit contributions | |
ird | +0.000 |
rell | +0.000 |
aren | +0.000 |
Karen | +0.000 |
Ellen | +0.000 |
Neg logit contributions | |
ble | -0.000 |
bled | -0.000 |
balls | -0.000 |
ster | -0.000 |
fin | -0.000 |
Token | and |
Token ID | 290 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
ble | +0.085 |
bled | +0.082 |
ster | +0.075 |
balls | +0.075 |
fin | +0.072 |
Neg logit contributions | |
ird | -0.040 |
hell | -0.032 |
rell | -0.031 |
he | -0.030 |
Karen | -0.029 |
Token | saw |
Token ID | 2497 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
ble | +0.027 |
bled | +0.025 |
balls | +0.023 |
ster | +0.023 |
fin | +0.023 |
Neg logit contributions | |
ird | -0.018 |
Karen | -0.016 |
rell | -0.016 |
aren | -0.016 |
Ellen | -0.015 |
Token | Tim |
Token ID | 5045 |
Feature activation | -4.10e-03 |
Pos logit contributions | |
ble | +0.026 |
bled | +0.023 |
balls | +0.022 |
ster | +0.021 |
fin | +0.021 |
Neg logit contributions | |
ird | -0.025 |
Karen | -0.023 |
rell | -0.023 |
aren | -0.023 |
Ellen | -0.021 |
Token | 's |
Token ID | 338 |
Feature activation | +6.78e-02 |
Pos logit contributions | |
ird | +0.007 |
rell | +0.006 |
Karen | +0.006 |
aren | +0.006 |
hell | +0.006 |
Neg logit contributions | |
ble | -0.009 |
bled | -0.009 |
balls | -0.008 |
ster | -0.008 |
fin | -0.008 |
Token | broken |
Token ID | 5445 |
Feature activation | +8.39e-02 |
Pos logit contributions | |
bled | +0.134 |
ble | +0.128 |
ster | +0.112 |
bing | +0.110 |
balls | +0.109 |
Neg logit contributions | |
hell | -0.101 |
ird | -0.097 |
rell | -0.096 |
umbrella | -0.092 |
Karen | -0.086 |
Token | sp |
Token ID | 599 |
Feature activation | +6.34e-02 |
Pos logit contributions | |
bled | +0.165 |
balls | +0.150 |
Shark | +0.148 |
ble | +0.148 |
ster | +0.142 |
Neg logit contributions | |
rell | -0.118 |
hell | -0.112 |
ird | -0.111 |
umbrella | -0.105 |
keys | -0.100 |
Token | ade |
Token ID | 671 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
ble | +0.163 |
bled | +0.159 |
fin | +0.158 |
ster | +0.155 |
Shark | +0.153 |
Neg logit contributions | |
hell | -0.068 |
ird | -0.067 |
rell | -0.062 |
he | -0.054 |
ives | -0.053 |
Token | . |
Token ID | 13 |
Feature activation | -8.85e-03 |
Pos logit contributions | |
ble | +0.056 |
bled | +0.052 |
balls | +0.049 |
ster | +0.049 |
fin | +0.049 |
Neg logit contributions | |
ird | -0.029 |
rell | -0.026 |
Karen | -0.025 |
aren | -0.025 |
hell | -0.023 |
Token | Tom |
Token ID | 4186 |
Feature activation | -4.20e-04 |
Pos logit contributions | |
ird | +0.014 |
Karen | +0.014 |
rell | +0.013 |
aren | +0.013 |
Ellen | +0.012 |
Neg logit contributions | |
ble | -0.022 |
bled | -0.019 |
ster | -0.018 |
fin | -0.018 |
balls | -0.018 |
Token | said |
Token ID | 531 |
Feature activation | +5.32e-03 |
Pos logit contributions | |
ird | +0.001 |
rell | +0.001 |
Karen | +0.001 |
aren | +0.001 |
Ellen | +0.000 |
Neg logit contributions | |
ble | -0.001 |
bled | -0.001 |
balls | -0.001 |
ster | -0.001 |
fin | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +9.24e-03 |
Pos logit contributions | |
ird | +0.027 |
rell | +0.026 |
aren | +0.024 |
Karen | +0.024 |
Ellen | +0.022 |
Neg logit contributions | |
ble | -0.027 |
bled | -0.022 |
balls | -0.022 |
fin | -0.021 |
emo | -0.021 |
Token | One |
Token ID | 3198 |
Feature activation | +3.84e-02 |
Pos logit contributions | |
ble | +0.024 |
bled | +0.022 |
balls | +0.021 |
ster | +0.021 |
fin | +0.021 |
Neg logit contributions | |
ird | -0.013 |
rell | -0.012 |
aren | -0.011 |
Karen | -0.011 |
Ellen | -0.010 |
Token | day |
Token ID | 1110 |
Feature activation | +2.73e-02 |
Pos logit contributions | |
bled | +0.108 |
balls | +0.104 |
ble | +0.103 |
ster | +0.091 |
bing | +0.088 |
Neg logit contributions | |
old | -0.035 |
Karen | -0.032 |
aren | -0.032 |
ird | -0.031 |
hell | -0.031 |
Token | , |
Token ID | 11 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
ble | +0.077 |
bled | +0.072 |
balls | +0.068 |
ster | +0.068 |
fin | +0.066 |
Neg logit contributions | |
ird | -0.030 |
Karen | -0.027 |
rell | -0.026 |
aren | -0.024 |
Stella | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | +5.01e-02 |
Pos logit contributions | |
ble | +0.031 |
bled | +0.029 |
ster | +0.028 |
balls | +0.027 |
fin | +0.026 |
Neg logit contributions | |
Karen | -0.016 |
ird | -0.015 |
Stella | -0.014 |
Ellie | -0.014 |
Ellen | -0.014 |
Token | oy |
Token ID | 35104 |
Feature activation | +8.34e-02 |
Pos logit contributions | |
bled | +0.097 |
balls | +0.096 |
ble | +0.088 |
bing | +0.079 |
zes | +0.075 |
Neg logit contributions | |
aren | -0.074 |
hell | -0.073 |
Stella | -0.073 |
Karen | -0.072 |
Mad | -0.072 |
Token | ster |
Token ID | 1706 |
Feature activation | +5.85e-02 |
Pos logit contributions | |
bled | +0.080 |
balls | +0.068 |
ble | +0.065 |
fin | +0.059 |
diving | +0.058 |
Neg logit contributions | |
ird | -0.198 |
hell | -0.190 |
aren | -0.186 |
old | -0.186 |
mie | -0.182 |
Token | went |
Token ID | 1816 |
Feature activation | +3.15e-02 |
Pos logit contributions | |
ble | +0.157 |
bled | +0.151 |
balls | +0.142 |
fin | +0.130 |
ster | +0.128 |
Neg logit contributions | |
ird | -0.073 |
old | -0.060 |
rell | -0.060 |
aren | -0.058 |
Karen | -0.057 |
Token | swimming |
Token ID | 14899 |
Feature activation | +4.90e-02 |
Pos logit contributions | |
ble | +0.082 |
bled | +0.079 |
ster | +0.070 |
balls | +0.067 |
fin | +0.067 |
Neg logit contributions | |
rell | -0.037 |
Ellie | -0.032 |
aback | -0.031 |
Ellen | -0.029 |
Ell | -0.028 |
Token | and |
Token ID | 290 |
Feature activation | +2.82e-02 |
Pos logit contributions | |
ble | +0.111 |
bled | +0.104 |
ster | +0.096 |
balls | +0.095 |
fin | +0.088 |
Neg logit contributions | |
rell | -0.068 |
ird | -0.067 |
hell | -0.064 |
aback | -0.064 |
Karen | -0.062 |
Token | it |
Token ID | 340 |
Feature activation | +4.14e-02 |
Pos logit contributions | |
ble | +0.077 |
bled | +0.069 |
ster | +0.066 |
iflower | +0.062 |
balls | +0.062 |
Neg logit contributions | |
rell | -0.028 |
Karen | -0.028 |
Stella | -0.027 |
Ellie | -0.027 |
seen | -0.025 |
Token | and |
Token ID | 290 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
ird | +0.025 |
rell | +0.024 |
Karen | +0.024 |
aren | +0.023 |
Ellen | +0.022 |
Neg logit contributions | |
ble | -0.029 |
bled | -0.027 |
ster | -0.025 |
balls | -0.025 |
fin | -0.024 |
Token | her |
Token ID | 607 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
ble | +0.026 |
bled | +0.023 |
balls | +0.022 |
ster | +0.022 |
fin | +0.021 |
Neg logit contributions | |
ird | -0.019 |
rell | -0.018 |
aren | -0.018 |
Karen | -0.018 |
Ellen | -0.016 |
Token | mom |
Token ID | 1995 |
Feature activation | -4.46e-02 |
Pos logit contributions | |
ble | +0.024 |
bled | +0.021 |
balls | +0.019 |
ster | +0.019 |
fin | +0.018 |
Neg logit contributions | |
ird | -0.036 |
rell | -0.034 |
Karen | -0.034 |
aren | -0.034 |
Ellen | -0.032 |
Token | left |
Token ID | 1364 |
Feature activation | -8.05e-03 |
Pos logit contributions | |
rell | +0.093 |
Ellen | +0.087 |
ird | +0.087 |
Karen | +0.085 |
aren | +0.085 |
Neg logit contributions | |
ble | -0.071 |
bled | -0.070 |
ster | -0.062 |
balls | -0.060 |
fin | -0.057 |
Token | the |
Token ID | 262 |
Feature activation | +8.79e-03 |
Pos logit contributions | |
ird | +0.018 |
rell | +0.017 |
aren | +0.016 |
hell | +0.016 |
Karen | +0.016 |
Neg logit contributions | |
ble | -0.014 |
bled | -0.013 |
balls | -0.012 |
fin | -0.011 |
ster | -0.011 |
Token | oy |
Token ID | 35104 |
Feature activation | +8.33e-02 |
Pos logit contributions | |
ble | +0.020 |
bled | +0.018 |
balls | +0.017 |
ster | +0.016 |
fin | +0.016 |
Neg logit contributions | |
ird | -0.015 |
Karen | -0.014 |
aren | -0.014 |
rell | -0.014 |
Stella | -0.013 |
Token | ster |
Token ID | 1706 |
Feature activation | +3.67e-02 |
Pos logit contributions | |
bled | +0.077 |
balls | +0.067 |
ble | +0.066 |
fin | +0.052 |
spitting | +0.050 |
Neg logit contributions | |
ird | -0.206 |
mie | -0.189 |
hell | -0.189 |
aren | -0.182 |
seen | -0.179 |
Token | at |
Token ID | 379 |
Feature activation | +7.99e-03 |
Pos logit contributions | |
ble | +0.078 |
bled | +0.072 |
balls | +0.067 |
fin | +0.062 |
ster | +0.062 |
Neg logit contributions | |
ird | -0.066 |
rell | -0.058 |
aren | -0.057 |
hell | -0.057 |
Karen | -0.056 |
Token | the |
Token ID | 262 |
Feature activation | +3.10e-02 |
Pos logit contributions | |
ble | +0.018 |
bled | +0.016 |
balls | +0.016 |
ster | +0.015 |
fin | +0.015 |
Neg logit contributions | |
ird | -0.013 |
Karen | -0.013 |
aren | -0.013 |
rell | -0.012 |
Stella | -0.012 |
Token | beach |
Token ID | 10481 |
Feature activation | +2.64e-02 |
Pos logit contributions | |
bled | +0.062 |
ble | +0.061 |
balls | +0.055 |
bing | +0.048 |
emo | +0.047 |
Neg logit contributions | |
Stella | -0.048 |
ird | -0.048 |
aren | -0.047 |
Karen | -0.047 |
rell | -0.046 |
Token | , |
Token ID | 11 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
ble | +0.066 |
bled | +0.061 |
balls | +0.057 |
ster | +0.056 |
fin | +0.056 |
Neg logit contributions | |
ird | -0.039 |
rell | -0.035 |
Karen | -0.035 |
aren | -0.035 |
Ellen | -0.032 |
Token | my |
Token ID | 1820 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
ird | +0.033 |
rell | +0.032 |
aren | +0.030 |
Karen | +0.029 |
Stella | +0.028 |
Neg logit contributions | |
ble | -0.069 |
bled | -0.064 |
ster | -0.062 |
balls | -0.061 |
fin | -0.058 |
Token | told |
Token ID | 1297 |
Feature activation | -2.78e-02 |
Pos logit contributions | |
ird | +0.025 |
rell | +0.024 |
aren | +0.023 |
Karen | +0.022 |
hell | +0.021 |
Neg logit contributions | |
ble | -0.032 |
bled | -0.029 |
ster | -0.028 |
balls | -0.028 |
fin | -0.027 |
Token | him |
Token ID | 683 |
Feature activation | +4.08e-03 |
Pos logit contributions | |
ird | +0.067 |
aren | +0.061 |
rell | +0.059 |
Aud | +0.057 |
Karen | +0.057 |
Neg logit contributions | |
ble | -0.045 |
bled | -0.040 |
balls | -0.035 |
fin | -0.032 |
ster | -0.032 |
Token | that |
Token ID | 326 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
ble | +0.009 |
bled | +0.008 |
ster | +0.007 |
fin | +0.007 |
balls | +0.007 |
Neg logit contributions | |
ird | -0.007 |
Karen | -0.007 |
Ellen | -0.006 |
rell | -0.006 |
aren | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
ble | +0.048 |
bled | +0.046 |
ster | +0.042 |
balls | +0.041 |
fin | +0.039 |
Neg logit contributions | |
Karen | -0.037 |
ird | -0.036 |
rell | -0.034 |
Ellen | -0.033 |
Ellie | -0.033 |
Token | ham |
Token ID | 8891 |
Feature activation | +8.28e-02 |
Pos logit contributions | |
bled | +0.090 |
ble | +0.087 |
balls | +0.074 |
bing | +0.071 |
fin | +0.068 |
Neg logit contributions | |
ird | -0.058 |
hell | -0.057 |
Stella | -0.055 |
rell | -0.055 |
aback | -0.052 |
Token | ster |
Token ID | 1706 |
Feature activation | +2.22e-02 |
Pos logit contributions | |
ble | +0.105 |
bled | +0.103 |
diving | +0.091 |
surfing | +0.090 |
bing | +0.088 |
Neg logit contributions | |
hell | -0.179 |
ird | -0.173 |
ives | -0.167 |
bars | -0.151 |
aren | -0.151 |
Token | would |
Token ID | 561 |
Feature activation | +9.62e-03 |
Pos logit contributions | |
ble | +0.049 |
bled | +0.045 |
balls | +0.041 |
ster | +0.040 |
fin | +0.039 |
Neg logit contributions | |
ird | -0.039 |
rell | -0.036 |
Karen | -0.036 |
aren | -0.035 |
old | -0.033 |
Token | always |
Token ID | 1464 |
Feature activation | -4.46e-03 |
Pos logit contributions | |
ble | +0.018 |
bled | +0.017 |
balls | +0.016 |
ster | +0.015 |
fin | +0.015 |
Neg logit contributions | |
ird | -0.020 |
rell | -0.018 |
Karen | -0.018 |
aren | -0.018 |
Ellen | -0.017 |
Token | be |
Token ID | 307 |
Feature activation | +9.72e-03 |
Pos logit contributions | |
ird | +0.009 |
rell | +0.008 |
Karen | +0.008 |
aren | +0.008 |
Ellen | +0.008 |
Neg logit contributions | |
ble | -0.009 |
bled | -0.008 |
balls | -0.008 |
ster | -0.007 |
fin | -0.007 |
Token | with |
Token ID | 351 |
Feature activation | -5.43e-03 |
Pos logit contributions | |
ble | +0.022 |
bled | +0.020 |
balls | +0.019 |
ster | +0.018 |
fin | +0.018 |
Neg logit contributions | |
ird | -0.017 |
rell | -0.016 |
aren | -0.016 |
Karen | -0.016 |
Ellen | -0.014 |
Token | he |
Token ID | 339 |
Feature activation | +4.45e-02 |
Pos logit contributions | |
ble | +0.068 |
bled | +0.061 |
balls | +0.058 |
ster | +0.056 |
fin | +0.055 |
Neg logit contributions | |
ird | -0.061 |
rell | -0.057 |
Karen | -0.057 |
aren | -0.056 |
Ellen | -0.052 |
Token | felt |
Token ID | 2936 |
Feature activation | +9.47e-03 |
Pos logit contributions | |
ble | +0.101 |
bled | +0.092 |
balls | +0.087 |
ster | +0.083 |
fin | +0.082 |
Neg logit contributions | |
ird | -0.072 |
rell | -0.067 |
Karen | -0.062 |
hell | -0.062 |
aren | -0.061 |
Token | better |
Token ID | 1365 |
Feature activation | +2.73e-02 |
Pos logit contributions | |
ble | +0.024 |
bled | +0.022 |
balls | +0.021 |
ster | +0.021 |
fin | +0.020 |
Neg logit contributions | |
ird | -0.014 |
rell | -0.012 |
Karen | -0.012 |
aren | -0.012 |
hell | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | +6.75e-03 |
Pos logit contributions | |
ble | +0.075 |
bled | +0.072 |
balls | +0.069 |
fin | +0.065 |
ster | +0.065 |
Neg logit contributions | |
ird | -0.031 |
aren | -0.024 |
Karen | -0.023 |
rell | -0.023 |
hell | -0.022 |
Token | The |
Token ID | 383 |
Feature activation | +5.13e-02 |
Pos logit contributions | |
ble | +0.015 |
bled | +0.014 |
balls | +0.013 |
ster | +0.013 |
fin | +0.013 |
Neg logit contributions | |
ird | -0.012 |
rell | -0.011 |
Karen | -0.011 |
aren | -0.011 |
Ellen | -0.010 |
Token | green |
Token ID | 4077 |
Feature activation | +8.26e-02 |
Pos logit contributions | |
bled | +0.131 |
ble | +0.128 |
balls | +0.114 |
ster | +0.105 |
bing | +0.104 |
Neg logit contributions | |
rell | -0.053 |
hell | -0.053 |
ird | -0.051 |
Many | -0.045 |
officer | -0.044 |
Token | oct |
Token ID | 19318 |
Feature activation | +5.91e-02 |
Pos logit contributions | |
bled | +0.167 |
ble | +0.141 |
bling | +0.133 |
balls | +0.133 |
ross | +0.119 |
Neg logit contributions | |
rell | -0.122 |
hell | -0.116 |
les | -0.106 |
est | -0.103 |
ha | -0.096 |
Token | opus |
Token ID | 25790 |
Feature activation | +2.62e-02 |
Pos logit contributions | |
bled | +0.100 |
ble | +0.095 |
balls | +0.078 |
fin | +0.078 |
hyd | +0.073 |
Neg logit contributions | |
ird | -0.118 |
hell | -0.118 |
ives | -0.110 |
aren | -0.105 |
he | -0.103 |
Token | and |
Token ID | 290 |
Feature activation | +5.37e-02 |
Pos logit contributions | |
ble | +0.062 |
bled | +0.057 |
balls | +0.054 |
ster | +0.051 |
fin | +0.050 |
Neg logit contributions | |
ird | -0.041 |
rell | -0.037 |
Karen | -0.035 |
aren | -0.035 |
hell | -0.033 |
Token | his |
Token ID | 465 |
Feature activation | +5.68e-02 |
Pos logit contributions | |
bled | +0.121 |
ble | +0.121 |
balls | +0.112 |
ster | +0.107 |
ider | +0.104 |
Neg logit contributions | |
Ellie | -0.072 |
rell | -0.070 |
Karen | -0.067 |
hell | -0.065 |
Ellen | -0.063 |
Token | friends |
Token ID | 2460 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
ble | +0.129 |
bled | +0.123 |
ster | +0.107 |
balls | +0.103 |
bing | +0.099 |
Neg logit contributions | |
rell | -0.080 |
ird | -0.080 |
Karen | -0.076 |
hell | -0.075 |
aren | -0.071 |
Token | to |
Token ID | 284 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
ble | +0.021 |
bled | +0.019 |
balls | +0.017 |
ster | +0.017 |
fin | +0.017 |
Neg logit contributions | |
ird | -0.019 |
Karen | -0.018 |
rell | -0.018 |
aren | -0.018 |
Stella | -0.016 |
Token | get |
Token ID | 651 |
Feature activation | +3.68e-02 |
Pos logit contributions | |
ble | +0.048 |
bled | +0.043 |
balls | +0.040 |
ster | +0.040 |
fin | +0.039 |
Neg logit contributions | |
ird | -0.044 |
rell | -0.040 |
Karen | -0.040 |
aren | -0.040 |
Ellen | -0.037 |
Token | out |
Token ID | 503 |
Feature activation | +3.08e-02 |
Pos logit contributions | |
ble | +0.083 |
ster | +0.077 |
bled | +0.073 |
balls | +0.071 |
fin | +0.070 |
Neg logit contributions | |
ird | -0.057 |
rell | -0.052 |
Karen | -0.050 |
Eve | -0.049 |
old | -0.047 |
Token | of |
Token ID | 286 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
ble | +0.078 |
bled | +0.072 |
balls | +0.068 |
ster | +0.067 |
fin | +0.066 |
Neg logit contributions | |
ird | -0.045 |
rell | -0.040 |
Karen | -0.039 |
aren | -0.039 |
old | -0.036 |
Token | the |
Token ID | 262 |
Feature activation | +4.53e-02 |
Pos logit contributions | |
ble | +0.035 |
bled | +0.031 |
balls | +0.030 |
ster | +0.030 |
fin | +0.028 |
Neg logit contributions | |
ird | -0.028 |
Karen | -0.027 |
rell | -0.026 |
aren | -0.026 |
Ellen | -0.025 |
Token | bath |
Token ID | 7837 |
Feature activation | +8.25e-02 |
Pos logit contributions | |
ble | +0.106 |
bled | +0.103 |
ster | +0.095 |
balls | +0.093 |
Miles | +0.087 |
Neg logit contributions | |
ird | -0.061 |
rell | -0.060 |
Eve | -0.056 |
Karen | -0.051 |
Mad | -0.050 |
Token | tub |
Token ID | 37995 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
ble | +0.121 |
bled | +0.108 |
fin | +0.095 |
ster | +0.094 |
balls | +0.090 |
Neg logit contributions | |
ird | -0.197 |
rell | -0.187 |
Eve | -0.172 |
aren | -0.169 |
ives | -0.168 |
Token | now |
Token ID | 783 |
Feature activation | +1.98e-02 |
Pos logit contributions | |
ble | +0.048 |
bled | +0.044 |
balls | +0.042 |
ster | +0.041 |
fin | +0.040 |
Neg logit contributions | |
ird | -0.030 |
rell | -0.028 |
Karen | -0.028 |
aren | -0.027 |
Ellen | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
ble | +0.050 |
bled | +0.046 |
balls | +0.043 |
ster | +0.043 |
fin | +0.042 |
Neg logit contributions | |
ird | -0.029 |
rell | -0.026 |
Karen | -0.026 |
aren | -0.026 |
Ellen | -0.024 |
Token | The |
Token ID | 383 |
Feature activation | +4.69e-02 |
Pos logit contributions | |
ble | +0.025 |
bled | +0.023 |
balls | +0.022 |
ster | +0.021 |
fin | +0.021 |
Neg logit contributions | |
ird | -0.019 |
Karen | -0.018 |
rell | -0.017 |
aren | -0.017 |
Ellen | -0.016 |
Token | water |
Token ID | 1660 |
Feature activation | +4.80e-02 |
Pos logit contributions | |
ble | +0.102 |
bled | +0.098 |
ster | +0.090 |
balls | +0.087 |
iflower | +0.081 |
Neg logit contributions | |
ird | -0.072 |
rell | -0.068 |
Karen | -0.064 |
Eve | -0.063 |
Stella | -0.061 |
Token | with |
Token ID | 351 |
Feature activation | +7.61e-03 |
Pos logit contributions | |
ble | +0.017 |
bled | +0.016 |
balls | +0.014 |
ster | +0.014 |
fin | +0.013 |
Neg logit contributions | |
ird | -0.020 |
rell | -0.019 |
Karen | -0.019 |
aren | -0.018 |
hell | -0.018 |
Token | ." |
Token ID | 526 |
Feature activation | +4.01e-03 |
Pos logit contributions | |
ble | +0.018 |
bled | +0.016 |
balls | +0.015 |
ster | +0.015 |
fin | +0.015 |
Neg logit contributions | |
ird | -0.012 |
aren | -0.011 |
rell | -0.011 |
Karen | -0.011 |
Ellen | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | +4.74e-03 |
Pos logit contributions | |
ble | +0.010 |
bled | +0.009 |
balls | +0.009 |
ster | +0.009 |
fin | +0.008 |
Neg logit contributions | |
ird | -0.006 |
rell | -0.005 |
Karen | -0.005 |
aren | -0.005 |
Ellen | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
ble | +0.011 |
bled | +0.010 |
balls | +0.010 |
ster | +0.009 |
fin | +0.009 |
Neg logit contributions | |
ird | -0.008 |
rell | -0.007 |
aren | -0.007 |
Karen | -0.007 |
Ellen | -0.006 |
Token | The |
Token ID | 464 |
Feature activation | +4.90e-02 |
Pos logit contributions | |
ble | +0.042 |
bled | +0.040 |
balls | +0.038 |
ster | +0.036 |
fin | +0.036 |
Neg logit contributions | |
ird | -0.024 |
Ell | -0.022 |
aren | -0.022 |
Karen | -0.022 |
Alice | -0.021 |
Token | white |
Token ID | 2330 |
Feature activation | +8.25e-02 |
Pos logit contributions | |
ble | +0.128 |
bled | +0.127 |
balls | +0.116 |
ster | +0.109 |
bing | +0.106 |
Neg logit contributions | |
hell | -0.044 |
rell | -0.041 |
ird | -0.040 |
Karen | -0.036 |
Stella | -0.036 |
Token | oct |
Token ID | 19318 |
Feature activation | +5.89e-02 |
Pos logit contributions | |
bled | +0.142 |
ble | +0.125 |
balls | +0.118 |
bling | +0.109 |
bing | +0.100 |
Neg logit contributions | |
hell | -0.137 |
rell | -0.136 |
ha | -0.124 |
aren | -0.120 |
aff | -0.120 |
Token | opus |
Token ID | 25790 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
bled | +0.096 |
ble | +0.092 |
balls | +0.076 |
fin | +0.074 |
bing | +0.072 |
Neg logit contributions | |
ird | -0.122 |
hell | -0.119 |
ives | -0.113 |
aren | -0.112 |
old | -0.110 |
Token | felt |
Token ID | 2936 |
Feature activation | -7.01e-03 |
Pos logit contributions | |
ble | +0.026 |
bled | +0.023 |
balls | +0.022 |
ster | +0.021 |
fin | +0.021 |
Neg logit contributions | |
ird | -0.016 |
rell | -0.014 |
Karen | -0.014 |
aren | -0.014 |
Ellen | -0.013 |
Token | bad |
Token ID | 2089 |
Feature activation | +1.65e-03 |
Pos logit contributions | |
ird | +0.012 |
rell | +0.011 |
aren | +0.011 |
Karen | +0.011 |
Ellen | +0.010 |
Neg logit contributions | |
ble | -0.016 |
bled | -0.014 |
balls | -0.013 |
ster | -0.013 |
fin | -0.013 |
Token | for |
Token ID | 329 |
Feature activation | +8.42e-03 |
Pos logit contributions | |
ble | +0.004 |
bled | +0.004 |
balls | +0.003 |
ster | +0.003 |
fin | +0.003 |
Neg logit contributions | |
ird | -0.002 |
rell | -0.002 |
Karen | -0.002 |
aren | -0.002 |
hell | -0.002 |
Token | it |
Token ID | 340 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
ird | +0.036 |
aren | +0.034 |
rell | +0.033 |
Karen | +0.032 |
Ell | +0.030 |
Neg logit contributions | |
ble | -0.024 |
bled | -0.019 |
fin | -0.018 |
ster | -0.018 |
balls | -0.018 |
Token | before |
Token ID | 878 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
ird | +0.030 |
aren | +0.029 |
rell | +0.028 |
Karen | +0.027 |
old | +0.026 |
Neg logit contributions | |
ble | -0.035 |
bled | -0.030 |
balls | -0.030 |
ster | -0.029 |
fin | -0.028 |
Token | it |
Token ID | 340 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
ird | +0.033 |
aren | +0.031 |
rell | +0.031 |
old | +0.029 |
Karen | +0.029 |
Neg logit contributions | |
ble | -0.035 |
bled | -0.030 |
balls | -0.029 |
fin | -0.027 |
ster | -0.027 |
Token | was |
Token ID | 373 |
Feature activation | +1.91e-03 |
Pos logit contributions | |
ble | +0.022 |
bled | +0.020 |
balls | +0.018 |
ster | +0.018 |
fin | +0.018 |
Neg logit contributions | |
ird | -0.019 |
rell | -0.017 |
Karen | -0.017 |
aren | -0.017 |
Ellen | -0.016 |
Token | gone |
Token ID | 3750 |
Feature activation | +4.71e-03 |
Pos logit contributions | |
ble | +0.004 |
bled | +0.004 |
balls | +0.003 |
fin | +0.003 |
ster | +0.003 |
Neg logit contributions | |
ird | -0.004 |
aren | -0.003 |
rell | -0.003 |
Karen | -0.003 |
Ellen | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -8.87e-02 |
Pos logit contributions | |
ble | +0.011 |
bled | +0.010 |
balls | +0.009 |
ster | +0.009 |
fin | +0.009 |
Neg logit contributions | |
ird | -0.008 |
aren | -0.007 |
rell | -0.007 |
Karen | -0.007 |
old | -0.007 |
Token | She |
Token ID | 1375 |
Feature activation | +3.73e-03 |
Pos logit contributions | |
hell | +0.170 |
ird | +0.159 |
ock | +0.155 |
ives | +0.153 |
Aud | +0.152 |
Neg logit contributions | |
ble | -0.145 |
bled | -0.135 |
next | -0.127 |
balls | -0.118 |
ster | -0.115 |
Token | chased |
Token ID | 26172 |
Feature activation | -5.08e-02 |
Pos logit contributions | |
ble | +0.009 |
bled | +0.008 |
balls | +0.007 |
ster | +0.007 |
fin | +0.007 |
Neg logit contributions | |
ird | -0.006 |
Karen | -0.006 |
aren | -0.006 |
rell | -0.006 |
Ellen | -0.005 |
Token | it |
Token ID | 340 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
ird | +0.138 |
rell | +0.127 |
chairs | +0.117 |
aren | +0.116 |
hell | +0.115 |
Neg logit contributions | |
ble | -0.064 |
bled | -0.051 |
next | -0.046 |
bing | -0.043 |
ster | -0.043 |
Token | and |
Token ID | 290 |
Feature activation | -2.87e-02 |
Pos logit contributions | |
ird | +0.031 |
aren | +0.028 |
Karen | +0.027 |
rell | +0.027 |
old | +0.025 |
Neg logit contributions | |
ble | -0.031 |
bled | -0.027 |
balls | -0.027 |
diving | -0.025 |
fin | -0.025 |
Token | chased |
Token ID | 26172 |
Feature activation | -4.82e-02 |
Pos logit contributions | |
ird | +0.056 |
rell | +0.053 |
aren | +0.052 |
Karen | +0.050 |
hell | +0.049 |
Neg logit contributions | |
ble | -0.058 |
bled | -0.053 |
balls | -0.050 |
ster | -0.048 |
fin | -0.047 |
Token | for |
Token ID | 329 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
ird | +0.012 |
aren | +0.012 |
rell | +0.012 |
Karen | +0.011 |
Ellen | +0.011 |
Neg logit contributions | |
ble | -0.009 |
bled | -0.008 |
balls | -0.007 |
ster | -0.007 |
fin | -0.007 |
Token | ways |
Token ID | 2842 |
Feature activation | -3.86e-02 |
Pos logit contributions | |
rell | +0.053 |
ird | +0.052 |
aren | +0.052 |
ives | +0.048 |
Ell | +0.048 |
Neg logit contributions | |
ble | -0.038 |
diving | -0.033 |
fin | -0.032 |
bled | -0.031 |
balls | -0.031 |
Token | to |
Token ID | 284 |
Feature activation | -9.80e-03 |
Pos logit contributions | |
ird | +0.094 |
rell | +0.089 |
aren | +0.087 |
Karen | +0.085 |
hell | +0.082 |
Neg logit contributions | |
ble | -0.059 |
bled | -0.049 |
balls | -0.049 |
ster | -0.046 |
diving | -0.044 |
Token | help |
Token ID | 1037 |
Feature activation | -3.34e-02 |
Pos logit contributions | |
ird | +0.020 |
aren | +0.020 |
rell | +0.019 |
Karen | +0.019 |
chairs | +0.018 |
Neg logit contributions | |
ble | -0.019 |
bled | -0.016 |
balls | -0.015 |
fin | -0.015 |
ster | -0.015 |
Token | others |
Token ID | 1854 |
Feature activation | -9.70e-03 |
Pos logit contributions | |
aren | +0.064 |
ird | +0.063 |
rell | +0.061 |
Karen | +0.057 |
Aud | +0.057 |
Neg logit contributions | |
ble | -0.068 |
bled | -0.054 |
balls | -0.052 |
ster | -0.052 |
fin | -0.050 |
Token | . |
Token ID | 13 |
Feature activation | -8.60e-02 |
Pos logit contributions | |
ird | +0.015 |
rell | +0.014 |
aren | +0.014 |
Karen | +0.013 |
Ellen | +0.012 |
Neg logit contributions | |
ble | -0.024 |
bled | -0.022 |
balls | -0.021 |
ster | -0.020 |
fin | -0.020 |
Token | |
Token ID | 220 |
Feature activation | -4.42e-02 |
Pos logit contributions | |
hell | +0.186 |
ives | +0.184 |
ird | +0.182 |
ited | +0.179 |
aney | +0.176 |
Neg logit contributions | |
ble | -0.109 |
next | -0.096 |
bled | -0.083 |
bing | -0.072 |
balls | -0.071 |
Token |
|
Token ID | 198 |
Feature activation | -6.82e-02 |
Pos logit contributions | |
rell | +0.102 |
ird | +0.099 |
ited | +0.095 |
aren | +0.094 |
ha | +0.092 |
Neg logit contributions | |
ble | -0.071 |
bled | -0.057 |
balls | -0.054 |
emo | -0.046 |
ster | -0.043 |
Token |
|
Token ID | 198 |
Feature activation | -6.30e-02 |
Pos logit contributions | |
ird | +0.166 |
rell | +0.162 |
ives | +0.159 |
vious | +0.157 |
hell | +0.151 |
Neg logit contributions | |
ble | -0.089 |
next | -0.067 |
bled | -0.062 |
balls | -0.057 |
diving | -0.050 |
Token | One |
Token ID | 3198 |
Feature activation | +5.12e-03 |
Pos logit contributions | |
ives | +0.125 |
ird | +0.123 |
aback | +0.119 |
aren | +0.118 |
aney | +0.117 |
Neg logit contributions | |
ble | -0.102 |
next | -0.090 |
bled | -0.085 |
bing | -0.079 |
ster | -0.068 |
Token | day |
Token ID | 1110 |
Feature activation | -4.13e-02 |
Pos logit contributions | |
ble | +0.012 |
bled | +0.012 |
balls | +0.011 |
ster | +0.010 |
fin | +0.010 |
Neg logit contributions | |
ird | -0.008 |
aren | -0.008 |
Karen | -0.008 |
old | -0.007 |
chair | -0.007 |
Token | was |
Token ID | 373 |
Feature activation | +8.94e-03 |
Pos logit contributions | |
ble | +0.024 |
bled | +0.021 |
balls | +0.020 |
ster | +0.020 |
fin | +0.019 |
Neg logit contributions | |
ird | -0.024 |
rell | -0.022 |
aren | -0.022 |
Karen | -0.022 |
Ellen | -0.021 |
Token | too |
Token ID | 1165 |
Feature activation | +7.54e-03 |
Pos logit contributions | |
ble | +0.019 |
bled | +0.017 |
balls | +0.016 |
ster | +0.016 |
fin | +0.016 |
Neg logit contributions | |
ird | -0.016 |
Karen | -0.015 |
rell | -0.015 |
aren | -0.015 |
Ellen | -0.014 |
Token | dangerous |
Token ID | 4923 |
Feature activation | -3.41e-02 |
Pos logit contributions | |
ble | +0.015 |
bled | +0.013 |
balls | +0.012 |
ster | +0.012 |
fin | +0.012 |
Neg logit contributions | |
ird | -0.015 |
rell | -0.014 |
aren | -0.014 |
Karen | -0.014 |
Ellen | -0.013 |
Token | for |
Token ID | 329 |
Feature activation | -4.52e-02 |
Pos logit contributions | |
ird | +0.073 |
aren | +0.068 |
rell | +0.065 |
Karen | +0.063 |
Ell | +0.062 |
Neg logit contributions | |
ble | -0.061 |
balls | -0.055 |
bled | -0.053 |
diving | -0.049 |
ster | -0.049 |
Token | her |
Token ID | 607 |
Feature activation | +8.40e-03 |
Pos logit contributions | |
ives | +0.089 |
ird | +0.088 |
hell | +0.086 |
aback | +0.082 |
old | +0.081 |
Neg logit contributions | |
ble | -0.085 |
bled | -0.075 |
balls | -0.069 |
diving | -0.066 |
next | -0.062 |
Token | . |
Token ID | 13 |
Feature activation | -8.48e-02 |
Pos logit contributions | |
ble | +0.019 |
bled | +0.017 |
balls | +0.016 |
ster | +0.016 |
fin | +0.016 |
Neg logit contributions | |
ird | -0.014 |
rell | -0.013 |
aren | -0.013 |
Karen | -0.013 |
Ellen | -0.012 |
Token | |
Token ID | 220 |
Feature activation | -4.35e-02 |
Pos logit contributions | |
hell | +0.136 |
ives | +0.131 |
ird | +0.130 |
ock | +0.127 |
ited | +0.121 |
Neg logit contributions | |
ble | -0.161 |
balls | -0.152 |
bled | -0.146 |
next | -0.141 |
bing | -0.129 |
Token |
|
Token ID | 198 |
Feature activation | -6.48e-02 |
Pos logit contributions | |
rell | +0.098 |
ird | +0.097 |
ited | +0.094 |
hy | +0.092 |
aren | +0.092 |
Neg logit contributions | |
ble | -0.069 |
bled | -0.055 |
balls | -0.053 |
emo | -0.043 |
Miles | -0.039 |
Token |
|
Token ID | 198 |
Feature activation | -5.94e-02 |
Pos logit contributions | |
ird | +0.161 |
rell | +0.161 |
vious | +0.159 |
ives | +0.150 |
hell | +0.148 |
Neg logit contributions | |
ble | -0.077 |
balls | -0.052 |
bled | -0.049 |
next | -0.049 |
emo | -0.042 |
Token | K |
Token ID | 42 |
Feature activation | -5.00e-02 |
Pos logit contributions | |
ird | +0.089 |
ives | +0.088 |
rell | +0.085 |
hell | +0.083 |
aback | +0.075 |
Neg logit contributions | |
ble | -0.129 |
bled | -0.112 |
next | -0.108 |
bing | -0.103 |
balls | -0.102 |
Token | aren |
Token ID | 5757 |
Feature activation | -6.23e-03 |
Pos logit contributions | |
ird | +0.042 |
hell | +0.040 |
rell | +0.039 |
Ell | +0.037 |
Ellen | +0.037 |
Neg logit contributions | |
ble | -0.127 |
ster | -0.117 |
bled | -0.113 |
next | -0.107 |
ich | -0.107 |
Token | a |
Token ID | 257 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
ird | +0.055 |
rell | +0.053 |
aren | +0.052 |
hell | +0.050 |
Karen | +0.049 |
Neg logit contributions | |
ble | -0.044 |
bled | -0.039 |
fin | -0.036 |
balls | -0.035 |
diving | -0.035 |
Token | big |
Token ID | 1263 |
Feature activation | -3.02e-02 |
Pos logit contributions | |
ird | +0.032 |
Karen | +0.029 |
aren | +0.029 |
rell | +0.029 |
Ellen | +0.027 |
Neg logit contributions | |
ble | -0.041 |
bled | -0.037 |
balls | -0.035 |
ster | -0.034 |
fin | -0.033 |
Token | and |
Token ID | 290 |
Feature activation | +5.60e-03 |
Pos logit contributions | |
ird | +0.061 |
rell | +0.056 |
Karen | +0.055 |
aren | +0.055 |
hell | +0.052 |
Neg logit contributions | |
ble | -0.057 |
bled | -0.049 |
ster | -0.049 |
fin | -0.049 |
balls | -0.047 |
Token | beautiful |
Token ID | 4950 |
Feature activation | -5.43e-03 |
Pos logit contributions | |
ble | +0.012 |
bled | +0.011 |
balls | +0.010 |
ster | +0.010 |
fin | +0.010 |
Neg logit contributions | |
ird | -0.010 |
rell | -0.010 |
aren | -0.010 |
Karen | -0.009 |
Ellen | -0.009 |
Token | nation |
Token ID | 3277 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
ird | +0.011 |
rell | +0.010 |
aren | +0.010 |
Karen | +0.010 |
Ellen | +0.009 |
Neg logit contributions | |
ble | -0.011 |
bled | -0.010 |
balls | -0.009 |
ster | -0.009 |
fin | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -8.44e-02 |
Pos logit contributions | |
ird | +0.026 |
rell | +0.025 |
Karen | +0.025 |
aren | +0.024 |
Ellen | +0.023 |
Neg logit contributions | |
ble | -0.033 |
bled | -0.029 |
balls | -0.028 |
ster | -0.028 |
fin | -0.027 |
Token | Every |
Token ID | 3887 |
Feature activation | +1.05e-03 |
Pos logit contributions | |
ird | +0.163 |
hell | +0.149 |
ha | +0.148 |
ives | +0.142 |
rell | +0.140 |
Neg logit contributions | |
next | -0.121 |
ble | -0.099 |
Then | -0.098 |
balls | -0.095 |
bled | -0.094 |
Token | day |
Token ID | 1110 |
Feature activation | -4.30e-02 |
Pos logit contributions | |
ble | +0.002 |
bled | +0.002 |
balls | +0.002 |
ster | +0.002 |
fin | +0.002 |
Neg logit contributions | |
ird | -0.002 |
rell | -0.002 |
Karen | -0.002 |
aren | -0.002 |
Ellen | -0.002 |
Token | she |
Token ID | 673 |
Feature activation | +2.79e-02 |
Pos logit contributions | |
rell | +0.094 |
ird | +0.090 |
aren | +0.086 |
hell | +0.086 |
Ell | +0.085 |
Neg logit contributions | |
ble | -0.057 |
bled | -0.055 |
balls | -0.052 |
next | -0.050 |
diving | -0.048 |
Token | would |
Token ID | 561 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
ble | +0.064 |
bled | +0.059 |
balls | +0.055 |
ster | +0.054 |
fin | +0.053 |
Neg logit contributions | |
ird | -0.047 |
rell | -0.043 |
aren | -0.042 |
Karen | -0.042 |
Ellen | -0.039 |
Token | go |
Token ID | 467 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
ird | +0.022 |
rell | +0.021 |
aren | +0.021 |
Karen | +0.020 |
hell | +0.019 |
Neg logit contributions | |
ble | -0.025 |
bled | -0.022 |
balls | -0.021 |
ster | -0.021 |
fin | -0.021 |
Token | little |
Token ID | 1310 |
Feature activation | -1.02e-03 |
Pos logit contributions | |
ird | +0.012 |
rell | +0.011 |
aren | +0.011 |
Karen | +0.011 |
Ell | +0.010 |
Neg logit contributions | |
ble | -0.013 |
ster | -0.011 |
bled | -0.011 |
fin | -0.011 |
shark | -0.011 |
Token | girl |
Token ID | 2576 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
ird | +0.002 |
Karen | +0.002 |
aren | +0.002 |
rell | +0.002 |
Ellen | +0.001 |
Neg logit contributions | |
ble | -0.002 |
bled | -0.002 |
balls | -0.002 |
ster | -0.002 |
fin | -0.002 |
Token | named |
Token ID | 3706 |
Feature activation | -2.58e-02 |
Pos logit contributions | |
ird | +0.017 |
rell | +0.016 |
aren | +0.016 |
Karen | +0.015 |
Ellen | +0.014 |
Neg logit contributions | |
ble | -0.023 |
bled | -0.020 |
ster | -0.020 |
balls | -0.019 |
fin | -0.019 |
Token | Mad |
Token ID | 4627 |
Feature activation | -5.98e-02 |
Pos logit contributions | |
ird | +0.057 |
rell | +0.057 |
hell | +0.055 |
stret | +0.055 |
chairs | +0.054 |
Neg logit contributions | |
ble | -0.030 |
balls | -0.029 |
next | -0.027 |
bled | -0.026 |
diving | -0.025 |
Token | eline |
Token ID | 4470 |
Feature activation | -9.59e-03 |
Pos logit contributions | |
ird | +0.075 |
rell | +0.072 |
Aud | +0.068 |
ives | +0.067 |
hell | +0.064 |
Neg logit contributions | |
ble | -0.145 |
ster | -0.135 |
bled | -0.132 |
bing | -0.117 |
next | -0.115 |
Token | . |
Token ID | 13 |
Feature activation | -8.42e-02 |
Pos logit contributions | |
ird | +0.017 |
rell | +0.017 |
aren | +0.016 |
Aud | +0.016 |
Karen | +0.016 |
Neg logit contributions | |
ble | -0.020 |
ster | -0.018 |
bled | -0.018 |
balls | -0.016 |
fin | -0.016 |
Token | One |
Token ID | 1881 |
Feature activation | +4.05e-03 |
Pos logit contributions | |
ird | +0.150 |
old | +0.148 |
ives | +0.142 |
hell | +0.138 |
anne | +0.129 |
Neg logit contributions | |
ble | -0.150 |
bled | -0.140 |
next | -0.134 |
bing | -0.116 |
balls | -0.109 |
Token | day |
Token ID | 1110 |
Feature activation | -4.57e-02 |
Pos logit contributions | |
ble | +0.009 |
bled | +0.008 |
balls | +0.008 |
ster | +0.007 |
fin | +0.007 |
Neg logit contributions | |
ird | -0.007 |
Karen | -0.007 |
aren | -0.007 |
rell | -0.007 |
chair | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | -6.51e-02 |
Pos logit contributions | |
ives | +0.097 |
rell | +0.095 |
Aud | +0.094 |
old | +0.093 |
hell | +0.093 |
Neg logit contributions | |
ble | -0.069 |
bled | -0.064 |
balls | -0.052 |
ster | -0.051 |
next | -0.047 |
Token | she |
Token ID | 673 |
Feature activation | +2.90e-02 |
Pos logit contributions | |
ives | +0.128 |
rell | +0.121 |
hell | +0.119 |
old | +0.117 |
ird | +0.112 |
Neg logit contributions | |
ble | -0.121 |
bled | -0.118 |
next | -0.102 |
ster | -0.095 |
balls | -0.094 |
Token | was |
Token ID | 373 |
Feature activation | -5.22e-04 |
Pos logit contributions | |
ble | +0.062 |
bled | +0.055 |
balls | +0.052 |
ster | +0.051 |
fin | +0.050 |
Neg logit contributions | |
ird | -0.054 |
rell | -0.050 |
aren | -0.050 |
Karen | -0.050 |
Ellen | -0.046 |
Token | catch |
Token ID | 4929 |
Feature activation | -6.55e-03 |
Pos logit contributions | |
ird | +0.030 |
aren | +0.029 |
rell | +0.028 |
Karen | +0.028 |
chairs | +0.027 |
Neg logit contributions | |
ble | -0.024 |
bled | -0.020 |
balls | -0.019 |
fin | -0.019 |
ster | -0.018 |
Token | it |
Token ID | 340 |
Feature activation | -9.70e-03 |
Pos logit contributions | |
ird | +0.016 |
aren | +0.015 |
rell | +0.015 |
Karen | +0.014 |
Ell | +0.013 |
Neg logit contributions | |
ble | -0.011 |
bled | -0.009 |
balls | -0.009 |
fin | -0.009 |
ster | -0.008 |
Token | in |
Token ID | 287 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
ird | +0.017 |
aren | +0.016 |
Karen | +0.015 |
rell | +0.015 |
Ellen | +0.014 |
Neg logit contributions | |
ble | -0.022 |
bled | -0.019 |
balls | -0.019 |
ster | -0.018 |
fin | -0.018 |
Token | her |
Token ID | 607 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
rell | +0.056 |
ird | +0.056 |
aren | +0.052 |
ives | +0.052 |
hell | +0.051 |
Neg logit contributions | |
ble | -0.034 |
fin | -0.029 |
diving | -0.028 |
bled | -0.027 |
digging | -0.026 |
Token | hands |
Token ID | 2832 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
ble | +0.029 |
bled | +0.027 |
balls | +0.025 |
ster | +0.023 |
emo | +0.022 |
Neg logit contributions | |
ird | -0.023 |
aren | -0.020 |
Karen | -0.020 |
rell | -0.020 |
Ellen | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -8.40e-02 |
Pos logit contributions | |
ird | +0.022 |
aren | +0.021 |
rell | +0.021 |
Karen | +0.021 |
Ellen | +0.019 |
Neg logit contributions | |
ble | -0.030 |
bled | -0.026 |
balls | -0.025 |
ster | -0.025 |
fin | -0.024 |
Token | |
Token ID | 220 |
Feature activation | -4.69e-02 |
Pos logit contributions | |
hell | +0.163 |
ird | +0.155 |
ock | +0.151 |
ited | +0.150 |
Aud | +0.149 |
Neg logit contributions | |
ble | -0.136 |
bled | -0.126 |
next | -0.113 |
balls | -0.108 |
bing | -0.101 |
Token |
|
Token ID | 198 |
Feature activation | -6.60e-02 |
Pos logit contributions | |
rell | +0.100 |
ird | +0.099 |
ited | +0.098 |
aren | +0.094 |
hell | +0.093 |
Neg logit contributions | |
ble | -0.082 |
bled | -0.070 |
balls | -0.065 |
emo | -0.053 |
ster | -0.050 |
Token |
|
Token ID | 198 |
Feature activation | -5.45e-02 |
Pos logit contributions | |
ird | +0.159 |
rell | +0.159 |
vious | +0.155 |
ited | +0.151 |
hell | +0.149 |
Neg logit contributions | |
ble | -0.088 |
bled | -0.066 |
next | -0.059 |
balls | -0.058 |
emo | -0.049 |
Token | When |
Token ID | 2215 |
Feature activation | -4.44e-02 |
Pos logit contributions | |
ird | +0.108 |
rell | +0.108 |
aren | +0.104 |
hell | +0.102 |
ives | +0.101 |
Neg logit contributions | |
ble | -0.097 |
bled | -0.085 |
bing | -0.072 |
next | -0.071 |
ster | -0.069 |
Token | she |
Token ID | 673 |
Feature activation | +7.21e-03 |
Pos logit contributions | |
ird | +0.072 |
rell | +0.070 |
hell | +0.070 |
ives | +0.069 |
old | +0.067 |
Neg logit contributions | |
ble | -0.096 |
bled | -0.089 |
ster | -0.078 |
balls | -0.076 |
next | -0.075 |
Token | lunch |
Token ID | 9965 |
Feature activation | -1.61e-02 |
Pos logit contributions | |
ird | +0.009 |
aren | +0.008 |
rell | +0.008 |
Karen | +0.008 |
hell | +0.008 |
Neg logit contributions | |
ble | -0.007 |
bled | -0.006 |
balls | -0.006 |
fin | -0.006 |
ster | -0.006 |
Token | ," |
Token ID | 553 |
Feature activation | -4.28e-02 |
Pos logit contributions | |
rell | +0.026 |
aren | +0.025 |
Karen | +0.025 |
ird | +0.024 |
Ellen | +0.023 |
Neg logit contributions | |
ble | -0.036 |
balls | -0.032 |
bled | -0.031 |
ster | -0.030 |
diving | -0.029 |
Token | her |
Token ID | 607 |
Feature activation | -8.16e-03 |
Pos logit contributions | |
rell | +0.083 |
ird | +0.077 |
ives | +0.075 |
hell | +0.074 |
aren | +0.073 |
Neg logit contributions | |
ble | -0.086 |
bled | -0.076 |
balls | -0.071 |
ster | -0.070 |
fin | -0.067 |
Token | mom |
Token ID | 1995 |
Feature activation | -3.95e-02 |
Pos logit contributions | |
rell | +0.022 |
ird | +0.022 |
aren | +0.021 |
hell | +0.021 |
ives | +0.021 |
Neg logit contributions | |
ble | -0.010 |
bled | -0.008 |
balls | -0.008 |
ster | -0.007 |
fin | -0.007 |
Token | shouted |
Token ID | 17293 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
rell | +0.069 |
aren | +0.065 |
hell | +0.064 |
Aud | +0.062 |
Karen | +0.061 |
Neg logit contributions | |
ble | -0.080 |
bled | -0.074 |
balls | -0.071 |
ster | -0.069 |
fin | -0.064 |
Token | . |
Token ID | 13 |
Feature activation | -8.38e-02 |
Pos logit contributions | |
ird | +0.045 |
rell | +0.043 |
aren | +0.043 |
Karen | +0.041 |
hell | +0.038 |
Neg logit contributions | |
ble | -0.057 |
bled | -0.052 |
balls | -0.050 |
ster | -0.048 |
fin | -0.047 |
Token | |
Token ID | 220 |
Feature activation | -5.97e-02 |
Pos logit contributions | |
hell | +0.164 |
ird | +0.159 |
rell | +0.155 |
ock | +0.154 |
vious | +0.142 |
Neg logit contributions | |
balls | -0.130 |
ble | -0.125 |
bled | -0.121 |
next | -0.119 |
Mitch | -0.100 |
Token |
|
Token ID | 198 |
Feature activation | -6.74e-02 |
Pos logit contributions | |
ird | +0.128 |
rell | +0.123 |
hell | +0.121 |
aren | +0.117 |
ited | +0.117 |
Neg logit contributions | |
ble | -0.103 |
bled | -0.089 |
balls | -0.087 |
emo | -0.066 |
next | -0.064 |
Token |
|
Token ID | 198 |
Feature activation | -5.59e-02 |
Pos logit contributions | |
rell | +0.153 |
ird | +0.140 |
vious | +0.139 |
hell | +0.133 |
chairs | +0.132 |
Neg logit contributions | |
ble | -0.095 |
next | -0.083 |
bled | -0.082 |
balls | -0.078 |
But | -0.063 |
Token | Alice |
Token ID | 44484 |
Feature activation | +8.57e-03 |
Pos logit contributions | |
rell | +0.081 |
ird | +0.065 |
hell | +0.063 |
chairs | +0.060 |
ives | +0.058 |
Neg logit contributions | |
ble | -0.126 |
bled | -0.122 |
next | -0.118 |
balls | -0.112 |
ster | -0.109 |
Token | took |
Token ID | 1718 |
Feature activation | -4.90e-02 |
Pos logit contributions | |
ble | +0.020 |
bled | +0.018 |
balls | +0.017 |
ster | +0.016 |
fin | +0.016 |
Neg logit contributions | |
ird | -0.015 |
Karen | -0.013 |
aren | -0.013 |
rell | -0.013 |
Ellen | -0.012 |
Token | she |
Token ID | 673 |
Feature activation | +5.51e-02 |
Pos logit contributions | |
ives | +0.018 |
aren | +0.018 |
ird | +0.017 |
Aud | +0.017 |
hell | +0.017 |
Neg logit contributions | |
ble | -0.014 |
bled | -0.012 |
ster | -0.011 |
balls | -0.011 |
fin | -0.010 |
Token | heard |
Token ID | 2982 |
Feature activation | -3.33e-02 |
Pos logit contributions | |
ble | +0.134 |
bled | +0.122 |
balls | +0.119 |
ster | +0.114 |
fin | +0.110 |
Neg logit contributions | |
ird | -0.082 |
rell | -0.068 |
aren | -0.064 |
Karen | -0.064 |
hell | -0.062 |
Token | a |
Token ID | 257 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
ird | +0.088 |
aren | +0.087 |
rell | +0.083 |
irst | +0.083 |
ives | +0.082 |
Neg logit contributions | |
ble | -0.036 |
bled | -0.023 |
next | -0.021 |
ster | -0.017 |
his | -0.016 |
Token | loud |
Token ID | 7812 |
Feature activation | -9.40e-03 |
Pos logit contributions | |
ird | +0.022 |
Karen | +0.021 |
rell | +0.020 |
aren | +0.020 |
Ellen | +0.019 |
Neg logit contributions | |
ble | -0.033 |
bled | -0.031 |
balls | -0.029 |
ster | -0.028 |
fin | -0.027 |
Token | noise |
Token ID | 7838 |
Feature activation | -2.71e-03 |
Pos logit contributions | |
ird | +0.015 |
rell | +0.014 |
Karen | +0.014 |
aren | +0.014 |
Ellen | +0.013 |
Neg logit contributions | |
ble | -0.022 |
bled | -0.020 |
balls | -0.019 |
ster | -0.018 |
fin | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -8.38e-02 |
Pos logit contributions | |
ird | +0.004 |
aren | +0.004 |
rell | +0.004 |
Karen | +0.004 |
Ellen | +0.004 |
Neg logit contributions | |
ble | -0.006 |
bled | -0.006 |
balls | -0.006 |
ster | -0.005 |
fin | -0.005 |
Token | Mad |
Token ID | 4627 |
Feature activation | -5.60e-02 |
Pos logit contributions | |
ird | +0.166 |
hell | +0.160 |
old | +0.153 |
ited | +0.152 |
ock | +0.150 |
Neg logit contributions | |
ble | -0.143 |
bled | -0.123 |
balls | -0.102 |
next | -0.100 |
ster | -0.100 |
Token | eline |
Token ID | 4470 |
Feature activation | +8.45e-03 |
Pos logit contributions | |
ird | +0.064 |
rell | +0.062 |
hell | +0.058 |
Aud | +0.055 |
ives | +0.053 |
Neg logit contributions | |
ble | -0.145 |
bled | -0.137 |
ster | -0.133 |
bing | -0.119 |
ich | -0.113 |
Token | panicked |
Token ID | 38225 |
Feature activation | -8.58e-03 |
Pos logit contributions | |
ble | +0.019 |
bled | +0.018 |
balls | +0.017 |
ster | +0.016 |
fin | +0.016 |
Neg logit contributions | |
ird | -0.014 |
Karen | -0.013 |
aren | -0.013 |
rell | -0.013 |
Ellen | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
ird | +0.016 |
aren | +0.015 |
rell | +0.015 |
Karen | +0.014 |
old | +0.013 |
Neg logit contributions | |
ble | -0.019 |
bled | -0.016 |
balls | -0.015 |
ster | -0.015 |
fin | -0.015 |
Token | looked |
Token ID | 3114 |
Feature activation | -2.40e-02 |
Pos logit contributions | |
ird | +0.020 |
Karen | +0.019 |
rell | +0.019 |
aren | +0.019 |
Ellen | +0.018 |
Neg logit contributions | |
ble | -0.026 |
bled | -0.024 |
balls | -0.022 |
ster | -0.022 |
fin | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | +5.75e-03 |
Pos logit contributions | |
Karen | +0.002 |
ird | +0.002 |
rell | +0.002 |
aren | +0.002 |
Ellen | +0.002 |
Neg logit contributions | |
ble | -0.003 |
bled | -0.003 |
ster | -0.003 |
fin | -0.002 |
balls | -0.002 |
Token | bake |
Token ID | 28450 |
Feature activation | -2.06e-03 |
Pos logit contributions | |
ble | +0.013 |
bled | +0.012 |
balls | +0.011 |
ster | +0.011 |
fin | +0.011 |
Neg logit contributions | |
ird | -0.010 |
rell | -0.009 |
Karen | -0.009 |
aren | -0.009 |
Ellen | -0.008 |
Token | cookies |
Token ID | 14746 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
ird | +0.004 |
aren | +0.004 |
rell | +0.004 |
Karen | +0.003 |
Ellen | +0.003 |
Neg logit contributions | |
ble | -0.004 |
bled | -0.004 |
balls | -0.004 |
ster | -0.004 |
diving | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | -1.50e-02 |
Pos logit contributions | |
ird | +0.021 |
aren | +0.020 |
rell | +0.020 |
Karen | +0.020 |
Ellen | +0.019 |
Neg logit contributions | |
ble | -0.021 |
bled | -0.019 |
balls | -0.018 |
ster | -0.017 |
fin | -0.016 |
Token | cakes |
Token ID | 30849 |
Feature activation | -9.29e-03 |
Pos logit contributions | |
ird | +0.029 |
aren | +0.028 |
rell | +0.028 |
Karen | +0.027 |
Ellen | +0.026 |
Neg logit contributions | |
ble | -0.030 |
bled | -0.027 |
balls | -0.026 |
ster | -0.024 |
fin | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | -8.33e-02 |
Pos logit contributions | |
ird | +0.017 |
aren | +0.016 |
rell | +0.016 |
Karen | +0.016 |
Ellen | +0.015 |
Neg logit contributions | |
ble | -0.020 |
bled | -0.018 |
balls | -0.017 |
ster | -0.016 |
fin | -0.015 |
Token | One |
Token ID | 1881 |
Feature activation | -8.63e-04 |
Pos logit contributions | |
ird | +0.159 |
ives | +0.155 |
hell | +0.152 |
ited | +0.151 |
aback | +0.145 |
Neg logit contributions | |
ble | -0.122 |
next | -0.109 |
balls | -0.097 |
bled | -0.088 |
diving | -0.088 |
Token | day |
Token ID | 1110 |
Feature activation | -4.22e-02 |
Pos logit contributions | |
ird | +0.002 |
Karen | +0.002 |
aren | +0.002 |
rell | +0.002 |
Ellen | +0.001 |
Neg logit contributions | |
ble | -0.002 |
bled | -0.002 |
balls | -0.002 |
ster | -0.001 |
fin | -0.001 |
Token | while |
Token ID | 981 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
ird | +0.091 |
ives | +0.086 |
hell | +0.085 |
rell | +0.085 |
old | +0.084 |
Neg logit contributions | |
ble | -0.069 |
bled | -0.061 |
balls | -0.058 |
diving | -0.052 |
ster | -0.049 |
Token | baking |
Token ID | 16871 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
ird | +0.041 |
rell | +0.039 |
hell | +0.038 |
aren | +0.038 |
aback | +0.037 |
Neg logit contributions | |
ble | -0.035 |
bled | -0.031 |
balls | -0.029 |
diving | -0.027 |
ster | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | -5.38e-02 |
Pos logit contributions | |
ble | +0.023 |
bled | +0.020 |
balls | +0.019 |
ster | +0.019 |
fin | +0.018 |
Neg logit contributions | |
ird | -0.019 |
aren | -0.018 |
rell | -0.017 |
Karen | -0.017 |
Ellen | -0.016 |
Token | she |
Token ID | 673 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
ird | +0.020 |
rell | +0.019 |
aren | +0.018 |
Karen | +0.017 |
ives | +0.017 |
Neg logit contributions | |
ble | -0.026 |
bled | -0.023 |
balls | -0.022 |
ster | -0.021 |
fin | -0.021 |
Token | went |
Token ID | 1816 |
Feature activation | -2.36e-03 |
Pos logit contributions | |
ble | +0.036 |
bled | +0.033 |
balls | +0.031 |
ster | +0.031 |
fin | +0.030 |
Neg logit contributions | |
ird | -0.028 |
Karen | -0.025 |
rell | -0.025 |
aren | -0.025 |
Ellen | -0.023 |
Token | on |
Token ID | 319 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
ird | +0.004 |
rell | +0.004 |
Karen | +0.004 |
aren | +0.004 |
Ellen | +0.004 |
Neg logit contributions | |
ble | -0.005 |
bled | -0.005 |
ster | -0.004 |
balls | -0.004 |
fin | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
ird | +0.034 |
rell | +0.031 |
aren | +0.030 |
ives | +0.030 |
hell | +0.029 |
Neg logit contributions | |
ble | -0.023 |
diving | -0.020 |
bled | -0.019 |
fin | -0.018 |
digging | -0.018 |
Token | search |
Token ID | 2989 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
ird | +0.051 |
rell | +0.047 |
aren | +0.046 |
Karen | +0.046 |
Ellen | +0.042 |
Neg logit contributions | |
ble | -0.068 |
bled | -0.060 |
balls | -0.059 |
ster | -0.058 |
fin | -0.057 |
Token | . |
Token ID | 13 |
Feature activation | -8.27e-02 |
Pos logit contributions | |
ird | +0.037 |
aren | +0.035 |
rell | +0.035 |
Karen | +0.033 |
Aud | +0.033 |
Neg logit contributions | |
ble | -0.031 |
bled | -0.028 |
balls | -0.026 |
ster | -0.024 |
diving | -0.024 |
Token | She |
Token ID | 1375 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
hell | +0.167 |
ird | +0.165 |
ives | +0.158 |
vious | +0.144 |
rell | +0.141 |
Neg logit contributions | |
ble | -0.136 |
next | -0.104 |
bled | -0.103 |
balls | -0.099 |
ster | -0.097 |
Token | walked |
Token ID | 6807 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
ble | +0.038 |
bled | +0.034 |
balls | +0.032 |
ster | +0.032 |
fin | +0.031 |
Neg logit contributions | |
ird | -0.026 |
rell | -0.024 |
Karen | -0.024 |
aren | -0.024 |
Ellen | -0.022 |
Token | around |
Token ID | 1088 |
Feature activation | -8.38e-03 |
Pos logit contributions | |
ird | +0.030 |
Karen | +0.028 |
rell | +0.028 |
aren | +0.028 |
Ellen | +0.026 |
Neg logit contributions | |
ble | -0.040 |
bled | -0.036 |
balls | -0.034 |
ster | -0.034 |
fin | -0.033 |
Token | the |
Token ID | 262 |
Feature activation | -6.98e-03 |
Pos logit contributions | |
ird | +0.014 |
Karen | +0.013 |
aren | +0.013 |
rell | +0.013 |
Ellen | +0.012 |
Neg logit contributions | |
ble | -0.019 |
bled | -0.017 |
balls | -0.016 |
ster | -0.016 |
fin | -0.016 |
Token | garden |
Token ID | 11376 |
Feature activation | -2.66e-03 |
Pos logit contributions | |
ird | +0.010 |
aren | +0.010 |
Karen | +0.010 |
rell | +0.010 |
old | +0.009 |
Neg logit contributions | |
ble | -0.017 |
bled | -0.016 |
balls | -0.015 |
fin | -0.014 |
ster | -0.014 |
Token | mum |
Token ID | 25682 |
Feature activation | -1.68e-02 |
Pos logit contributions | |
ble | +0.018 |
bled | +0.015 |
balls | +0.014 |
ster | +0.014 |
fin | +0.014 |
Neg logit contributions | |
ird | -0.026 |
rell | -0.025 |
aren | -0.024 |
Karen | -0.024 |
hell | -0.023 |
Token | wouldn |
Token ID | 3636 |
Feature activation | -2.06e-02 |
Pos logit contributions | |
rell | +0.030 |
ird | +0.029 |
hell | +0.028 |
aren | +0.027 |
Karen | +0.027 |
Neg logit contributions | |
ble | -0.034 |
bled | -0.031 |
ster | -0.030 |
balls | -0.029 |
fin | -0.027 |
Token | 't |
Token ID | 470 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
ird | +0.038 |
rell | +0.035 |
Karen | +0.035 |
aren | +0.035 |
Ellen | +0.032 |
Neg logit contributions | |
ble | -0.044 |
bled | -0.040 |
balls | -0.037 |
ster | -0.037 |
fin | -0.036 |
Token | permit |
Token ID | 8749 |
Feature activation | -5.05e-02 |
Pos logit contributions | |
ird | +0.026 |
aren | +0.025 |
rell | +0.024 |
Karen | +0.024 |
chairs | +0.023 |
Neg logit contributions | |
ble | -0.029 |
bled | -0.024 |
balls | -0.023 |
ster | -0.023 |
fin | -0.023 |
Token | it |
Token ID | 340 |
Feature activation | -2.50e-02 |
Pos logit contributions | |
ird | +0.103 |
ives | +0.093 |
hell | +0.090 |
old | +0.087 |
rell | +0.087 |
Neg logit contributions | |
ble | -0.092 |
bled | -0.082 |
balls | -0.081 |
diving | -0.073 |
next | -0.072 |
Token | . |
Token ID | 13 |
Feature activation | -8.21e-02 |
Pos logit contributions | |
ird | +0.050 |
aren | +0.048 |
rell | +0.045 |
Karen | +0.044 |
old | +0.044 |
Neg logit contributions | |
ble | -0.048 |
bled | -0.042 |
balls | -0.041 |
ster | -0.037 |
diving | -0.036 |
Token | She |
Token ID | 1375 |
Feature activation | -4.45e-03 |
Pos logit contributions | |
ird | +0.157 |
hell | +0.154 |
ock | +0.148 |
ives | +0.144 |
isi | +0.137 |
Neg logit contributions | |
ble | -0.134 |
balls | -0.129 |
bled | -0.118 |
next | -0.109 |
digging | -0.102 |
Token | said |
Token ID | 531 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
ird | +0.007 |
rell | +0.007 |
aren | +0.007 |
Karen | +0.007 |
hell | +0.006 |
Neg logit contributions | |
ble | -0.010 |
bled | -0.009 |
balls | -0.009 |
ster | -0.009 |
fin | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | +9.34e-03 |
Pos logit contributions | |
ird | +0.030 |
Karen | +0.028 |
rell | +0.027 |
aren | +0.027 |
Ellen | +0.025 |
Neg logit contributions | |
ble | -0.045 |
bled | -0.041 |
balls | -0.039 |
ster | -0.038 |
fin | -0.038 |
Token | garden |
Token ID | 11376 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
ble | +0.018 |
bled | +0.016 |
balls | +0.015 |
ster | +0.015 |
fin | +0.015 |
Neg logit contributions | |
ird | -0.019 |
rell | -0.018 |
aren | -0.018 |
Karen | -0.018 |
Ellen | -0.017 |
Token | was |
Token ID | 373 |
Feature activation | +8.94e-03 |
Pos logit contributions | |
ble | +0.024 |
bled | +0.021 |
balls | +0.020 |
ster | +0.020 |
fin | +0.019 |
Neg logit contributions | |
ird | -0.024 |
rell | -0.022 |
aren | -0.022 |
Karen | -0.022 |
Ellen | -0.021 |
Token | eline |
Token ID | 4470 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
rell | +0.079 |
ird | +0.078 |
ives | +0.075 |
Aud | +0.074 |
hell | +0.070 |
Neg logit contributions | |
ble | -0.157 |
ster | -0.145 |
bled | -0.142 |
bing | -0.129 |
next | -0.125 |
Token | who |
Token ID | 508 |
Feature activation | -3.32e-03 |
Pos logit contributions | |
rell | +0.029 |
ird | +0.029 |
aren | +0.027 |
Aud | +0.027 |
Karen | +0.026 |
Neg logit contributions | |
ble | -0.032 |
ster | -0.028 |
bled | -0.027 |
balls | -0.026 |
fin | -0.025 |
Token | was |
Token ID | 373 |
Feature activation | -9.21e-03 |
Pos logit contributions | |
ird | +0.006 |
rell | +0.006 |
aren | +0.006 |
Karen | +0.006 |
Ellen | +0.005 |
Neg logit contributions | |
ble | -0.007 |
bled | -0.006 |
ster | -0.005 |
balls | -0.005 |
fin | -0.005 |
Token | very |
Token ID | 845 |
Feature activation | +9.33e-03 |
Pos logit contributions | |
ird | +0.016 |
Karen | +0.015 |
aren | +0.014 |
rell | +0.014 |
Ellen | +0.014 |
Neg logit contributions | |
ble | -0.021 |
bled | -0.019 |
balls | -0.018 |
ster | -0.018 |
fin | -0.018 |
Token | thoughtful |
Token ID | 22677 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
ble | +0.021 |
bled | +0.020 |
balls | +0.019 |
ster | +0.018 |
fin | +0.018 |
Neg logit contributions | |
ird | -0.016 |
Karen | -0.014 |
aren | -0.014 |
rell | -0.014 |
Stella | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -8.18e-02 |
Pos logit contributions | |
ird | +0.058 |
rell | +0.057 |
aren | +0.055 |
Aud | +0.054 |
Ell | +0.053 |
Neg logit contributions | |
ble | -0.057 |
ster | -0.050 |
balls | -0.049 |
bled | -0.047 |
fin | -0.045 |
Token | She |
Token ID | 1375 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
ird | +0.153 |
old | +0.149 |
ives | +0.147 |
hell | +0.146 |
Aud | +0.137 |
Neg logit contributions | |
ble | -0.131 |
next | -0.116 |
bled | -0.116 |
bing | -0.092 |
balls | -0.091 |
Token | loved |
Token ID | 6151 |
Feature activation | -6.56e-03 |
Pos logit contributions | |
ble | +0.037 |
bled | +0.033 |
balls | +0.032 |
ster | +0.031 |
fin | +0.031 |
Neg logit contributions | |
ird | -0.023 |
Karen | -0.021 |
aren | -0.021 |
rell | -0.021 |
Ellen | -0.019 |
Token | to |
Token ID | 284 |
Feature activation | -7.06e-03 |
Pos logit contributions | |
Karen | +0.011 |
ird | +0.011 |
rell | +0.010 |
aren | +0.010 |
Ellen | +0.010 |
Neg logit contributions | |
ble | -0.015 |
bled | -0.014 |
ster | -0.013 |
fin | -0.012 |
balls | -0.012 |
Token | do |
Token ID | 466 |
Feature activation | -9.40e-03 |
Pos logit contributions | |
ird | +0.013 |
rell | +0.012 |
Karen | +0.012 |
aren | +0.012 |
Ellen | +0.011 |
Neg logit contributions | |
ble | -0.016 |
bled | -0.014 |
balls | -0.013 |
ster | -0.013 |
fin | -0.013 |
Token | nice |
Token ID | 3621 |
Feature activation | +4.09e-03 |
Pos logit contributions | |
ird | +0.019 |
aren | +0.018 |
rell | +0.018 |
Karen | +0.018 |
Stella | +0.016 |
Neg logit contributions | |
ble | -0.018 |
bled | -0.016 |
balls | -0.015 |
ster | -0.015 |
fin | -0.014 |
Token | was |
Token ID | 373 |
Feature activation | -1.00e-02 |
Pos logit contributions | |
ird | +0.048 |
rell | +0.045 |
aren | +0.044 |
Aud | +0.043 |
ives | +0.043 |
Neg logit contributions | |
ble | -0.032 |
bled | -0.027 |
balls | -0.024 |
ster | -0.023 |
fin | -0.023 |
Token | a |
Token ID | 257 |
Feature activation | +3.45e-03 |
Pos logit contributions | |
ird | +0.022 |
rell | +0.021 |
aren | +0.021 |
Karen | +0.020 |
chairs | +0.019 |
Neg logit contributions | |
ble | -0.018 |
bled | -0.016 |
fin | -0.014 |
balls | -0.014 |
diving | -0.013 |
Token | girl |
Token ID | 2576 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
ble | +0.009 |
bled | +0.009 |
balls | +0.008 |
ster | +0.007 |
fin | +0.007 |
Neg logit contributions | |
ird | -0.005 |
Karen | -0.004 |
aren | -0.004 |
Stella | -0.004 |
Ellen | -0.004 |
Token | named |
Token ID | 3706 |
Feature activation | -2.26e-02 |
Pos logit contributions | |
ird | +0.021 |
rell | +0.020 |
aren | +0.019 |
Karen | +0.019 |
Ellen | +0.018 |
Neg logit contributions | |
ble | -0.030 |
bled | -0.027 |
ster | -0.026 |
balls | -0.026 |
fin | -0.026 |
Token | Karen |
Token ID | 18678 |
Feature activation | -3.41e-02 |
Pos logit contributions | |
rell | +0.050 |
ird | +0.049 |
hell | +0.048 |
chairs | +0.046 |
aback | +0.046 |
Neg logit contributions | |
ble | -0.033 |
bled | -0.029 |
balls | -0.028 |
next | -0.026 |
digging | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -8.18e-02 |
Pos logit contributions | |
rell | +0.080 |
ird | +0.076 |
Aud | +0.074 |
ives | +0.073 |
aback | +0.072 |
Neg logit contributions | |
ble | -0.051 |
ster | -0.043 |
next | -0.038 |
bled | -0.037 |
fin | -0.034 |
Token | She |
Token ID | 1375 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
ird | +0.183 |
old | +0.156 |
ives | +0.155 |
hell | +0.151 |
lier | +0.143 |
Neg logit contributions | |
ble | -0.114 |
next | -0.100 |
bled | -0.092 |
balls | -0.087 |
diving | -0.078 |
Token | wore |
Token ID | 12408 |
Feature activation | -2.11e-02 |
Pos logit contributions | |
ble | +0.078 |
bled | +0.073 |
balls | +0.069 |
ster | +0.066 |
fin | +0.066 |
Neg logit contributions | |
ird | -0.039 |
Karen | -0.035 |
aren | -0.034 |
rell | -0.034 |
Ell | -0.032 |
Token | a |
Token ID | 257 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
ird | +0.049 |
aren | +0.047 |
old | +0.043 |
rell | +0.043 |
Karen | +0.043 |
Neg logit contributions | |
ble | -0.036 |
bled | -0.033 |
balls | -0.029 |
fin | -0.027 |
ster | -0.027 |
Token | cozy |
Token ID | 37438 |
Feature activation | +3.84e-03 |
Pos logit contributions | |
ird | +0.045 |
rell | +0.042 |
aren | +0.042 |
Karen | +0.042 |
Ellen | +0.038 |
Neg logit contributions | |
ble | -0.058 |
bled | -0.052 |
balls | -0.049 |
ster | -0.049 |
fin | -0.048 |
Token | scarf |
Token ID | 36953 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
ble | +0.009 |
bled | +0.008 |
balls | +0.008 |
ster | +0.007 |
fin | +0.007 |
Neg logit contributions | |
ird | -0.006 |
rell | -0.006 |
aren | -0.006 |
Karen | -0.006 |
Ellen | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | -9.95e-03 |
Pos logit contributions | |
ird | +0.038 |
rell | +0.035 |
aren | +0.033 |
hell | +0.032 |
Karen | +0.032 |
Neg logit contributions | |
ble | -0.048 |
bled | -0.043 |
balls | -0.041 |
ster | -0.039 |
fin | -0.039 |
Token | voice |
Token ID | 3809 |
Feature activation | -4.61e-02 |
Pos logit contributions | |
ird | +0.018 |
rell | +0.017 |
aren | +0.016 |
Karen | +0.016 |
Ellen | +0.015 |
Neg logit contributions | |
ble | -0.022 |
bled | -0.019 |
balls | -0.019 |
ster | -0.018 |
fin | -0.018 |
Token | said |
Token ID | 531 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
ird | +0.081 |
rell | +0.080 |
aren | +0.074 |
hell | +0.074 |
Aud | +0.071 |
Neg logit contributions | |
ble | -0.098 |
bled | -0.086 |
ster | -0.081 |
fin | -0.078 |
shark | -0.076 |
Token | " |
Token ID | 366 |
Feature activation | -2.79e-02 |
Pos logit contributions | |
ird | +0.023 |
rell | +0.021 |
aren | +0.021 |
Karen | +0.020 |
Ellen | +0.019 |
Neg logit contributions | |
ble | -0.029 |
bled | -0.026 |
balls | -0.024 |
ster | -0.024 |
fin | -0.024 |
Token | Mine |
Token ID | 24461 |
Feature activation | +3.80e-03 |
Pos logit contributions | |
ird | +0.050 |
rell | +0.045 |
aren | +0.043 |
Karen | +0.043 |
hell | +0.042 |
Neg logit contributions | |
ble | -0.060 |
bled | -0.052 |
ster | -0.049 |
balls | -0.049 |
fin | -0.046 |
Token | !". |
Token ID | 48220 |
Feature activation | -8.13e-02 |
Pos logit contributions | |
ble | +0.010 |
bled | +0.010 |
balls | +0.009 |
fin | +0.009 |
ster | +0.009 |
Neg logit contributions | |
ird | -0.005 |
Karen | -0.004 |
rell | -0.004 |
aren | -0.004 |
Ellen | -0.004 |
Token | A |
Token ID | 317 |
Feature activation | -1.97e-02 |
Pos logit contributions | |
ird | +0.150 |
hell | +0.135 |
rell | +0.134 |
irst | +0.128 |
ock | +0.126 |
Neg logit contributions | |
ble | -0.158 |
bled | -0.132 |
balls | -0.130 |
fin | -0.121 |
next | -0.121 |
Token | small |
Token ID | 1402 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
ird | +0.040 |
rell | +0.039 |
aren | +0.037 |
Karen | +0.036 |
hell | +0.035 |
Neg logit contributions | |
ble | -0.038 |
fin | -0.032 |
ster | -0.032 |
bled | -0.032 |
balls | -0.031 |
Token | girl |
Token ID | 2576 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
ird | +0.045 |
rell | +0.044 |
aren | +0.041 |
Karen | +0.041 |
hell | +0.039 |
Neg logit contributions | |
ble | -0.042 |
fin | -0.036 |
ster | -0.036 |
bled | -0.035 |
shark | -0.035 |
Token | with |
Token ID | 351 |
Feature activation | -3.22e-02 |
Pos logit contributions | |
rell | +0.071 |
ird | +0.067 |
aren | +0.064 |
hell | +0.063 |
ives | +0.062 |
Neg logit contributions | |
ble | -0.065 |
bled | -0.056 |
ster | -0.055 |
fin | -0.055 |
balls | -0.052 |
Token | curly |
Token ID | 45731 |
Feature activation | -3.97e-02 |
Pos logit contributions | |
ird | +0.073 |
rell | +0.070 |
aren | +0.070 |
chairs | +0.065 |
Karen | +0.065 |
Neg logit contributions | |
ble | -0.052 |
fin | -0.047 |
bled | -0.047 |
shark | -0.040 |
ster | -0.039 |
Token | was |
Token ID | 373 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
ird | +0.038 |
aren | +0.036 |
rell | +0.036 |
Karen | +0.035 |
old | +0.033 |
Neg logit contributions | |
ble | -0.033 |
bled | -0.029 |
balls | -0.028 |
ster | -0.027 |
fin | -0.026 |
Token | time |
Token ID | 640 |
Feature activation | -8.04e-03 |
Pos logit contributions | |
ble | +0.023 |
bled | +0.020 |
balls | +0.019 |
ster | +0.019 |
fin | +0.019 |
Neg logit contributions | |
ird | -0.018 |
rell | -0.017 |
aren | -0.017 |
Karen | -0.017 |
Ellen | -0.015 |
Token | to |
Token ID | 284 |
Feature activation | -9.10e-03 |
Pos logit contributions | |
ird | +0.016 |
Karen | +0.015 |
aren | +0.015 |
rell | +0.015 |
Ellen | +0.014 |
Neg logit contributions | |
ble | -0.016 |
bled | -0.014 |
balls | -0.013 |
ster | -0.013 |
fin | -0.013 |
Token | go |
Token ID | 467 |
Feature activation | -6.82e-03 |
Pos logit contributions | |
ird | +0.018 |
rell | +0.017 |
aren | +0.017 |
Karen | +0.016 |
chairs | +0.016 |
Neg logit contributions | |
ble | -0.019 |
bled | -0.016 |
fin | -0.015 |
ster | -0.015 |
balls | -0.015 |
Token | home |
Token ID | 1363 |
Feature activation | -2.10e-04 |
Pos logit contributions | |
ird | +0.013 |
rell | +0.012 |
Karen | +0.012 |
aren | +0.012 |
Ellen | +0.011 |
Neg logit contributions | |
ble | -0.015 |
bled | -0.013 |
balls | -0.012 |
ster | -0.012 |
fin | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -8.12e-02 |
Pos logit contributions | |
ird | +0.000 |
rell | +0.000 |
aren | +0.000 |
Karen | +0.000 |
Ellen | +0.000 |
Neg logit contributions | |
ble | -0.000 |
bled | -0.000 |
balls | -0.000 |
ster | -0.000 |
fin | -0.000 |
Token | So |
Token ID | 1406 |
Feature activation | -5.68e-02 |
Pos logit contributions | |
hell | +0.141 |
ird | +0.126 |
ock | +0.124 |
ives | +0.119 |
ha | +0.109 |
Neg logit contributions | |
ble | -0.145 |
balls | -0.142 |
bled | -0.135 |
next | -0.131 |
bing | -0.123 |
Token | Karen |
Token ID | 18678 |
Feature activation | -5.15e-03 |
Pos logit contributions | |
ives | +0.106 |
hell | +0.105 |
ird | +0.103 |
rell | +0.101 |
old | +0.097 |
Neg logit contributions | |
ble | -0.108 |
bled | -0.098 |
balls | -0.090 |
next | -0.087 |
diving | -0.082 |
Token | and |
Token ID | 290 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
ird | +0.009 |
rell | +0.009 |
Karen | +0.008 |
aren | +0.008 |
Ellen | +0.008 |
Neg logit contributions | |
ble | -0.011 |
bled | -0.010 |
balls | -0.010 |
ster | -0.010 |
fin | -0.009 |
Token | her |
Token ID | 607 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
ird | +0.020 |
rell | +0.019 |
aren | +0.018 |
hell | +0.018 |
ives | +0.017 |
Neg logit contributions | |
ble | -0.027 |
bled | -0.024 |
balls | -0.023 |
ster | -0.022 |
fin | -0.022 |
Token | mom |
Token ID | 1995 |
Feature activation | -4.29e-02 |
Pos logit contributions | |
ble | +0.019 |
bled | +0.017 |
balls | +0.016 |
ster | +0.015 |
fin | +0.014 |
Neg logit contributions | |
ird | -0.032 |
rell | -0.031 |
aren | -0.031 |
Karen | -0.031 |
Ellen | -0.029 |
Token | big |
Token ID | 1263 |
Feature activation | -2.93e-02 |
Pos logit contributions | |
ird | +0.029 |
Karen | +0.026 |
aren | +0.026 |
rell | +0.025 |
Ellen | +0.024 |
Neg logit contributions | |
ble | -0.038 |
bled | -0.035 |
balls | -0.033 |
ster | -0.032 |
fin | -0.031 |
Token | fan |
Token ID | 4336 |
Feature activation | -9.45e-03 |
Pos logit contributions | |
ird | +0.058 |
rell | +0.055 |
aren | +0.053 |
Karen | +0.053 |
hell | +0.051 |
Neg logit contributions | |
ble | -0.056 |
fin | -0.048 |
ster | -0.048 |
bled | -0.048 |
shark | -0.047 |
Token | of |
Token ID | 286 |
Feature activation | -2.47e-02 |
Pos logit contributions | |
ird | +0.014 |
Karen | +0.013 |
rell | +0.013 |
aren | +0.012 |
Ellen | +0.011 |
Neg logit contributions | |
ble | -0.024 |
bled | -0.022 |
balls | -0.020 |
ster | -0.020 |
fin | -0.020 |
Token | fire |
Token ID | 2046 |
Feature activation | +1.22e-02 |
Pos logit contributions | |
ird | +0.044 |
rell | +0.041 |
aren | +0.041 |
Karen | +0.040 |
hell | +0.037 |
Neg logit contributions | |
ble | -0.055 |
bled | -0.050 |
balls | -0.047 |
ster | -0.045 |
fin | -0.045 |
Token | men |
Token ID | 3653 |
Feature activation | -1.76e-02 |
Pos logit contributions | |
ble | +0.033 |
bled | +0.031 |
balls | +0.028 |
ster | +0.028 |
fin | +0.028 |
Neg logit contributions | |
ird | -0.015 |
rell | -0.014 |
Karen | -0.014 |
aren | -0.013 |
chairs | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -8.11e-02 |
Pos logit contributions | |
ird | +0.028 |
rell | +0.026 |
aren | +0.026 |
Karen | +0.025 |
Ellen | +0.023 |
Neg logit contributions | |
ble | -0.041 |
bled | -0.037 |
balls | -0.036 |
ster | -0.035 |
fin | -0.035 |
Token | She |
Token ID | 1375 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
ird | +0.185 |
hell | +0.171 |
ives | +0.166 |
ited | +0.156 |
ha | +0.153 |
Neg logit contributions | |
next | -0.091 |
ble | -0.085 |
balls | -0.078 |
Then | -0.072 |
He | -0.064 |
Token | often |
Token ID | 1690 |
Feature activation | -2.73e-03 |
Pos logit contributions | |
ble | +0.038 |
bled | +0.034 |
balls | +0.032 |
ster | +0.032 |
fin | +0.031 |
Neg logit contributions | |
ird | -0.029 |
rell | -0.026 |
Karen | -0.026 |
aren | -0.026 |
Ellen | -0.024 |
Token | asked |
Token ID | 1965 |
Feature activation | -3.27e-02 |
Pos logit contributions | |
ird | +0.005 |
rell | +0.005 |
aren | +0.005 |
Karen | +0.005 |
Ellen | +0.004 |
Neg logit contributions | |
ble | -0.006 |
bled | -0.005 |
balls | -0.005 |
ster | -0.005 |
fin | -0.005 |
Token | her |
Token ID | 607 |
Feature activation | -1.37e-04 |
Pos logit contributions | |
ird | +0.067 |
rell | +0.065 |
aren | +0.064 |
hell | +0.063 |
Aud | +0.059 |
Neg logit contributions | |
ble | -0.054 |
bled | -0.051 |
balls | -0.047 |
ster | -0.042 |
next | -0.042 |
Token | daddy |
Token ID | 35822 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
ird | +0.000 |
rell | +0.000 |
aren | +0.000 |
Karen | +0.000 |
hell | +0.000 |
Neg logit contributions | |
ble | -0.000 |
bled | -0.000 |
balls | -0.000 |
ster | -0.000 |
fin | -0.000 |
Token | veil |
Token ID | 30615 |
Feature activation | +6.21e-03 |
Pos logit contributions | |
ble | +0.007 |
bled | +0.006 |
balls | +0.006 |
ster | +0.006 |
fin | +0.006 |
Neg logit contributions | |
ird | -0.005 |
rell | -0.004 |
Karen | -0.004 |
aren | -0.004 |
Ellen | -0.004 |
Token | on |
Token ID | 319 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
ble | +0.012 |
bled | +0.011 |
balls | +0.010 |
fin | +0.010 |
ster | +0.010 |
Neg logit contributions | |
ird | -0.013 |
rell | -0.012 |
Karen | -0.012 |
aren | -0.011 |
Ellen | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
ird | +0.038 |
aren | +0.035 |
rell | +0.034 |
old | +0.033 |
Karen | +0.033 |
Neg logit contributions | |
ble | -0.032 |
bled | -0.027 |
fin | -0.025 |
balls | -0.024 |
ster | -0.024 |
Token | kitchen |
Token ID | 9592 |
Feature activation | +8.92e-03 |
Pos logit contributions | |
ird | +0.041 |
aren | +0.038 |
Karen | +0.038 |
rell | +0.038 |
Ellen | +0.036 |
Neg logit contributions | |
ble | -0.037 |
bled | -0.032 |
balls | -0.030 |
ster | -0.030 |
fin | -0.029 |
Token | table |
Token ID | 3084 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
ble | +0.020 |
bled | +0.019 |
balls | +0.017 |
ster | +0.017 |
fin | +0.017 |
Neg logit contributions | |
ird | -0.015 |
rell | -0.014 |
Karen | -0.014 |
aren | -0.014 |
Ellen | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -8.00e-02 |
Pos logit contributions | |
ble | +0.051 |
bled | +0.047 |
balls | +0.045 |
ster | +0.044 |
fin | +0.044 |
Neg logit contributions | |
ird | -0.026 |
rell | -0.023 |
aren | -0.022 |
Karen | -0.022 |
hell | -0.020 |
Token | She |
Token ID | 1375 |
Feature activation | +5.56e-03 |
Pos logit contributions | |
ird | +0.168 |
ives | +0.149 |
hell | +0.148 |
old | +0.143 |
ited | +0.142 |
Neg logit contributions | |
ble | -0.133 |
bled | -0.106 |
balls | -0.102 |
next | -0.093 |
diving | -0.086 |
Token | asked |
Token ID | 1965 |
Feature activation | -4.07e-02 |
Pos logit contributions | |
ble | +0.013 |
bled | +0.012 |
balls | +0.012 |
ster | +0.011 |
fin | +0.011 |
Neg logit contributions | |
ird | -0.009 |
Karen | -0.008 |
aren | -0.008 |
rell | -0.008 |
Ellen | -0.007 |
Token | her |
Token ID | 607 |
Feature activation | +9.28e-03 |
Pos logit contributions | |
ives | +0.096 |
ird | +0.095 |
rell | +0.092 |
hell | +0.090 |
ited | +0.088 |
Neg logit contributions | |
ble | -0.059 |
bled | -0.054 |
next | -0.042 |
his | -0.041 |
balls | -0.038 |
Token | mom |
Token ID | 1995 |
Feature activation | -3.23e-02 |
Pos logit contributions | |
ble | +0.012 |
ster | +0.010 |
bled | +0.010 |
balls | +0.009 |
fin | +0.008 |
Neg logit contributions | |
ird | -0.025 |
rell | -0.025 |
aren | -0.024 |
Karen | -0.023 |
chairs | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | +1.20e-03 |
Pos logit contributions | |
ird | +0.050 |
rell | +0.045 |
aren | +0.045 |
Karen | +0.044 |
Ellen | +0.041 |
Neg logit contributions | |
ble | -0.078 |
bled | -0.071 |
balls | -0.068 |
ster | -0.068 |
fin | -0.065 |
Token | very |
Token ID | 845 |
Feature activation | +2.68e-02 |
Pos logit contributions | |
ird | +0.014 |
Karen | +0.012 |
aren | +0.012 |
rell | +0.012 |
Ellen | +0.011 |
Neg logit contributions | |
ble | -0.020 |
bled | -0.019 |
balls | -0.018 |
ster | -0.017 |
fin | -0.016 |
Token | happy |
Token ID | 3772 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
ble | +0.066 |
bled | +0.061 |
balls | +0.057 |
ster | +0.055 |
fin | +0.055 |
Neg logit contributions | |
ird | -0.041 |
aren | -0.037 |
Karen | -0.037 |
Stella | -0.035 |
old | -0.034 |
Token | and |
Token ID | 290 |
Feature activation | +5.83e-03 |
Pos logit contributions | |
ble | +0.039 |
bled | +0.036 |
balls | +0.034 |
ster | +0.032 |
fin | +0.032 |
Neg logit contributions | |
ird | -0.028 |
Karen | -0.026 |
aren | -0.026 |
rell | -0.026 |
Ellen | -0.024 |
Token | playful |
Token ID | 34264 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
ble | +0.014 |
bled | +0.013 |
balls | +0.012 |
ster | +0.012 |
fin | +0.012 |
Neg logit contributions | |
ird | -0.009 |
Karen | -0.008 |
rell | -0.008 |
aren | -0.008 |
Ellen | -0.008 |
Token | girl |
Token ID | 2576 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
ird | +0.026 |
rell | +0.024 |
aren | +0.024 |
Karen | +0.023 |
hell | +0.022 |
Neg logit contributions | |
ble | -0.029 |
bled | -0.025 |
balls | -0.025 |
ster | -0.025 |
fin | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -7.96e-02 |
Pos logit contributions | |
ird | +0.029 |
rell | +0.027 |
aren | +0.027 |
Karen | +0.026 |
hell | +0.025 |
Neg logit contributions | |
ble | -0.041 |
bled | -0.037 |
balls | -0.035 |
ster | -0.035 |
fin | -0.035 |
Token | She |
Token ID | 1375 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
ird | +0.161 |
hell | +0.144 |
ives | +0.139 |
ha | +0.139 |
rell | +0.134 |
Neg logit contributions | |
next | -0.115 |
ble | -0.104 |
balls | -0.097 |
bled | -0.095 |
diving | -0.090 |
Token | loved |
Token ID | 6151 |
Feature activation | +6.46e-03 |
Pos logit contributions | |
ble | +0.037 |
bled | +0.033 |
balls | +0.031 |
ster | +0.031 |
fin | +0.030 |
Neg logit contributions | |
ird | -0.028 |
rell | -0.026 |
Karen | -0.026 |
aren | -0.026 |
Ellen | -0.024 |
Token | to |
Token ID | 284 |
Feature activation | -7.91e-03 |
Pos logit contributions | |
ble | +0.016 |
bled | +0.014 |
ster | +0.013 |
fin | +0.012 |
balls | +0.012 |
Neg logit contributions | |
Karen | -0.010 |
rell | -0.010 |
ird | -0.010 |
Ellen | -0.009 |
Stella | -0.009 |
Token | play |
Token ID | 711 |
Feature activation | +2.83e-03 |
Pos logit contributions | |
ird | +0.014 |
rell | +0.013 |
aren | +0.013 |
Karen | +0.013 |
Ellen | +0.012 |
Neg logit contributions | |
ble | -0.017 |
bled | -0.015 |
balls | -0.015 |
ster | -0.014 |
fin | -0.014 |
Token | with |
Token ID | 351 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
ble | +0.006 |
bled | +0.005 |
ster | +0.005 |
balls | +0.005 |
fin | +0.005 |
Neg logit contributions | |
ird | -0.005 |
rell | -0.005 |
Karen | -0.005 |
aren | -0.005 |
Ellen | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | -1.25e-03 |
Pos logit contributions | |
ird | +0.110 |
rell | +0.103 |
hell | +0.099 |
ives | +0.099 |
aren | +0.095 |
Neg logit contributions | |
ble | -0.099 |
bled | -0.091 |
balls | -0.084 |
diving | -0.078 |
ster | -0.077 |
Token | hallway |
Token ID | 23959 |
Feature activation | -3.32e-03 |
Pos logit contributions | |
ird | +0.002 |
aren | +0.002 |
rell | +0.002 |
Karen | +0.002 |
Stella | +0.002 |
Neg logit contributions | |
ble | -0.003 |
bled | -0.002 |
balls | -0.002 |
fin | -0.002 |
ster | -0.002 |
Token | seemed |
Token ID | 3947 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
ird | +0.007 |
aren | +0.006 |
rell | +0.006 |
Karen | +0.006 |
Ellen | +0.006 |
Neg logit contributions | |
ble | -0.006 |
bled | -0.006 |
balls | -0.005 |
ster | -0.005 |
fin | -0.005 |
Token | less |
Token ID | 1342 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
ird | +0.026 |
rell | +0.024 |
aren | +0.024 |
Karen | +0.024 |
Ellen | +0.022 |
Neg logit contributions | |
ble | -0.030 |
bled | -0.027 |
balls | -0.025 |
ster | -0.025 |
fin | -0.025 |
Token | scary |
Token ID | 14343 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
ird | +0.020 |
aren | +0.019 |
Karen | +0.019 |
rell | +0.019 |
Ellen | +0.017 |
Neg logit contributions | |
ble | -0.025 |
bled | -0.022 |
fin | -0.021 |
balls | -0.021 |
ster | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -7.89e-02 |
Pos logit contributions | |
ird | +0.019 |
aren | +0.017 |
rell | +0.017 |
Karen | +0.017 |
Ellen | +0.016 |
Neg logit contributions | |
ble | -0.028 |
bled | -0.026 |
balls | -0.025 |
ster | -0.024 |
fin | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | -6.08e-02 |
Pos logit contributions | |
ird | +0.159 |
hell | +0.156 |
ives | +0.153 |
ock | +0.147 |
ited | +0.145 |
Neg logit contributions | |
ble | -0.113 |
balls | -0.098 |
bled | -0.096 |
Then | -0.095 |
next | -0.095 |
Token |
|
Token ID | 198 |
Feature activation | -5.56e-02 |
Pos logit contributions | |
rell | +0.147 |
ird | +0.144 |
ives | +0.142 |
vious | +0.138 |
hell | +0.138 |
Neg logit contributions | |
ble | -0.081 |
bled | -0.059 |
next | -0.056 |
balls | -0.052 |
diving | -0.044 |
Token | She |
Token ID | 3347 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
ives | +0.095 |
ird | +0.094 |
hell | +0.092 |
rell | +0.092 |
forts | +0.081 |
Neg logit contributions | |
ble | -0.111 |
bled | -0.098 |
next | -0.091 |
balls | -0.088 |
bing | -0.088 |
Token | t |
Token ID | 256 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
ble | +0.026 |
bled | +0.024 |
balls | +0.023 |
ster | +0.022 |
fin | +0.022 |
Neg logit contributions | |
ird | -0.020 |
rell | -0.019 |
aren | -0.019 |
Karen | -0.019 |
Ellen | -0.017 |
Token | ipt |
Token ID | 10257 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
ird | +0.034 |
rell | +0.032 |
Karen | +0.032 |
aren | +0.032 |
Ellen | +0.031 |
Neg logit contributions | |
ble | -0.016 |
bled | -0.013 |
balls | -0.011 |
ster | -0.011 |
fin | -0.010 |
Token | Lucy |
Token ID | 22162 |
Feature activation | -3.69e-03 |
Pos logit contributions | |
rell | +0.027 |
hell | +0.024 |
ird | +0.024 |
chairs | +0.023 |
ives | +0.023 |
Neg logit contributions | |
ble | -0.031 |
bled | -0.029 |
Mitch | -0.027 |
balls | -0.027 |
Shark | -0.026 |
Token | went |
Token ID | 1816 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
ird | +0.008 |
rell | +0.008 |
aren | +0.007 |
Karen | +0.007 |
hell | +0.007 |
Neg logit contributions | |
ble | -0.006 |
bled | -0.005 |
ster | -0.005 |
fin | -0.005 |
balls | -0.005 |
Token | for |
Token ID | 329 |
Feature activation | -3.79e-02 |
Pos logit contributions | |
ird | +0.017 |
rell | +0.016 |
Karen | +0.016 |
Ellen | +0.015 |
aren | +0.015 |
Neg logit contributions | |
ble | -0.023 |
bled | -0.022 |
ster | -0.020 |
balls | -0.020 |
fin | -0.019 |
Token | a |
Token ID | 257 |
Feature activation | -4.43e-02 |
Pos logit contributions | |
ird | +0.096 |
rell | +0.094 |
hell | +0.093 |
ives | +0.090 |
aren | +0.090 |
Neg logit contributions | |
ble | -0.044 |
diving | -0.038 |
bled | -0.036 |
digging | -0.035 |
balls | -0.034 |
Token | walk |
Token ID | 2513 |
Feature activation | -2.25e-02 |
Pos logit contributions | |
ird | +0.065 |
rell | +0.061 |
aren | +0.059 |
Karen | +0.058 |
Ellen | +0.053 |
Neg logit contributions | |
ble | -0.108 |
bled | -0.098 |
balls | -0.096 |
ster | -0.094 |
fin | -0.093 |
Token | . |
Token ID | 13 |
Feature activation | -7.89e-02 |
Pos logit contributions | |
ird | +0.046 |
rell | +0.043 |
aren | +0.043 |
Karen | +0.041 |
hell | +0.040 |
Neg logit contributions | |
ble | -0.043 |
bled | -0.037 |
balls | -0.035 |
ster | -0.034 |
fin | -0.034 |
Token | She |
Token ID | 1375 |
Feature activation | +5.66e-03 |
Pos logit contributions | |
ird | +0.166 |
hell | +0.150 |
ha | +0.147 |
Aud | +0.138 |
ock | +0.138 |
Neg logit contributions | |
ble | -0.113 |
balls | -0.109 |
bled | -0.103 |
next | -0.095 |
digging | -0.094 |
Token | saw |
Token ID | 2497 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
ble | +0.012 |
bled | +0.011 |
balls | +0.010 |
ster | +0.010 |
fin | +0.010 |
Neg logit contributions | |
ird | -0.010 |
rell | -0.009 |
aren | -0.009 |
Karen | -0.009 |
hell | -0.009 |
Token | lots |
Token ID | 6041 |
Feature activation | -3.48e-02 |
Pos logit contributions | |
ird | +0.047 |
rell | +0.045 |
aren | +0.043 |
hell | +0.040 |
ives | +0.040 |
Neg logit contributions | |
ble | -0.032 |
bled | -0.028 |
balls | -0.024 |
ster | -0.024 |
fin | -0.023 |
Token | of |
Token ID | 286 |
Feature activation | -4.66e-02 |
Pos logit contributions | |
ird | +0.068 |
aren | +0.059 |
Karen | +0.057 |
rell | +0.057 |
Ell | +0.056 |
Neg logit contributions | |
ble | -0.069 |
diving | -0.058 |
bled | -0.057 |
shark | -0.056 |
balls | -0.056 |
Token | interesting |
Token ID | 3499 |
Feature activation | -2.39e-02 |
Pos logit contributions | |
ird | +0.107 |
rell | +0.097 |
aren | +0.095 |
Aud | +0.092 |
ives | +0.090 |
Neg logit contributions | |
ble | -0.077 |
shark | -0.064 |
diving | -0.062 |
fin | -0.060 |
digging | -0.058 |
Token | back |
Token ID | 736 |
Feature activation | +5.07e-03 |
Pos logit contributions | |
ble | +0.007 |
bled | +0.006 |
balls | +0.006 |
ster | +0.006 |
fin | +0.006 |
Neg logit contributions | |
ird | -0.006 |
rell | -0.006 |
Karen | -0.006 |
aren | -0.006 |
Ellen | -0.005 |
Token | to |
Token ID | 284 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
ble | +0.010 |
bled | +0.008 |
balls | +0.008 |
ster | +0.008 |
fin | +0.008 |
Neg logit contributions | |
ird | -0.011 |
rell | -0.010 |
aren | -0.010 |
Karen | -0.010 |
hell | -0.009 |
Token | his |
Token ID | 465 |
Feature activation | +3.66e-02 |
Pos logit contributions | |
ble | +0.033 |
bled | +0.032 |
balls | +0.029 |
ster | +0.028 |
fin | +0.025 |
Neg logit contributions | |
Karen | -0.028 |
ird | -0.027 |
rell | -0.026 |
Stella | -0.026 |
Ellen | -0.026 |
Token | family |
Token ID | 1641 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
bled | +0.084 |
ble | +0.083 |
balls | +0.072 |
ster | +0.067 |
bling | +0.065 |
Neg logit contributions | |
ird | -0.049 |
Karen | -0.048 |
aren | -0.046 |
Eve | -0.046 |
rell | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
ird | +0.025 |
rell | +0.024 |
aren | +0.024 |
Karen | +0.023 |
Ellen | +0.021 |
Neg logit contributions | |
ble | -0.031 |
bled | -0.027 |
balls | -0.027 |
ster | -0.026 |
fin | -0.026 |
Token | While |
Token ID | 2893 |
Feature activation | +6.32e-02 |
Pos logit contributions | |
ble | +0.050 |
bled | +0.046 |
iflower | +0.041 |
fish | +0.041 |
ster | +0.041 |
Neg logit contributions | |
rell | -0.015 |
ird | -0.014 |
Karen | -0.014 |
aback | -0.014 |
Ellen | -0.014 |
Token | he |
Token ID | 339 |
Feature activation | +3.27e-02 |
Pos logit contributions | |
ble | +0.156 |
bled | +0.134 |
ster | +0.132 |
castle | +0.132 |
urg | +0.127 |
Neg logit contributions | |
Ellie | -0.079 |
ird | -0.074 |
Ellen | -0.073 |
Stella | -0.072 |
Karen | -0.070 |
Token | was |
Token ID | 373 |
Feature activation | -6.37e-03 |
Pos logit contributions | |
ble | +0.077 |
bled | +0.069 |
balls | +0.068 |
ster | +0.064 |
fin | +0.061 |
Neg logit contributions | |
ird | -0.051 |
Karen | -0.045 |
aren | -0.044 |
rell | -0.044 |
ha | -0.040 |
Token | swimming |
Token ID | 14899 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
ird | +0.013 |
aren | +0.012 |
rell | +0.012 |
Karen | +0.012 |
Ellen | +0.012 |
Neg logit contributions | |
ble | -0.012 |
bled | -0.011 |
balls | -0.010 |
ster | -0.010 |
fin | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | +5.97e-02 |
Pos logit contributions | |
ble | +0.031 |
bled | +0.029 |
balls | +0.027 |
ster | +0.027 |
fin | +0.026 |
Neg logit contributions | |
ird | -0.021 |
rell | -0.019 |
Karen | -0.019 |
aren | -0.019 |
Ellen | -0.017 |
Token | he |
Token ID | 339 |
Feature activation | +4.57e-02 |
Pos logit contributions | |
ble | +0.136 |
ster | +0.125 |
castle | +0.125 |
bled | +0.121 |
iflower | +0.118 |
Neg logit contributions | |
Ellie | -0.074 |
Ellen | -0.068 |
ird | -0.067 |
Beat | -0.067 |
Stella | -0.067 |
Token | ocean |
Token ID | 9151 |
Feature activation | +5.32e-02 |
Pos logit contributions | |
ble | +0.092 |
bled | +0.091 |
balls | +0.081 |
ster | +0.071 |
fin | +0.065 |
Neg logit contributions | |
ird | -0.093 |
rell | -0.093 |
aren | -0.082 |
Eve | -0.080 |
hell | -0.080 |
Token | . |
Token ID | 13 |
Feature activation | +2.71e-02 |
Pos logit contributions | |
ble | +0.137 |
bled | +0.135 |
ster | +0.117 |
balls | +0.116 |
fin | +0.112 |
Neg logit contributions | |
ird | -0.060 |
rell | -0.056 |
hell | -0.054 |
Eve | -0.050 |
he | -0.048 |
Token | One |
Token ID | 1881 |
Feature activation | +4.28e-02 |
Pos logit contributions | |
ble | +0.075 |
bled | +0.071 |
ster | +0.066 |
balls | +0.066 |
fin | +0.063 |
Neg logit contributions | |
rell | -0.032 |
ird | -0.031 |
Karen | -0.029 |
aren | -0.027 |
Stella | -0.026 |
Token | day |
Token ID | 1110 |
Feature activation | +3.46e-02 |
Pos logit contributions | |
bled | +0.113 |
ble | +0.112 |
balls | +0.107 |
ster | +0.100 |
bing | +0.091 |
Neg logit contributions | |
rell | -0.044 |
ird | -0.043 |
Karen | -0.042 |
aren | -0.042 |
old | -0.041 |
Token | , |
Token ID | 11 |
Feature activation | +3.35e-02 |
Pos logit contributions | |
ble | +0.095 |
bled | +0.087 |
ster | +0.083 |
balls | +0.082 |
fin | +0.081 |
Neg logit contributions | |
ird | -0.041 |
rell | -0.038 |
Karen | -0.036 |
aren | -0.034 |
Stella | -0.034 |
Token | Nem |
Token ID | 22547 |
Feature activation | +5.93e-02 |
Pos logit contributions | |
ble | +0.072 |
bled | +0.065 |
ster | +0.064 |
balls | +0.063 |
fin | +0.061 |
Neg logit contributions | |
Stella | -0.053 |
Karen | -0.051 |
Mad | -0.048 |
rell | -0.047 |
Ellie | -0.047 |
Token | o |
Token ID | 78 |
Feature activation | +5.80e-02 |
Pos logit contributions | |
ble | +0.132 |
bled | +0.131 |
balls | +0.115 |
fin | +0.105 |
diving | +0.103 |
Neg logit contributions | |
ird | -0.091 |
aren | -0.086 |
old | -0.084 |
rell | -0.083 |
ives | -0.081 |
Token | saw |
Token ID | 2497 |
Feature activation | +3.93e-02 |
Pos logit contributions | |
ble | +0.156 |
bled | +0.146 |
balls | +0.137 |
ster | +0.129 |
fin | +0.125 |
Neg logit contributions | |
ird | -0.074 |
rell | -0.065 |
aren | -0.062 |
Karen | -0.061 |
old | -0.059 |
Token | a |
Token ID | 257 |
Feature activation | +3.16e-02 |
Pos logit contributions | |
ble | +0.092 |
bled | +0.086 |
balls | +0.085 |
ster | +0.082 |
fin | +0.076 |
Neg logit contributions | |
Karen | -0.051 |
rell | -0.050 |
ird | -0.049 |
old | -0.047 |
Stella | -0.045 |
Token | plastic |
Token ID | 7309 |
Feature activation | +6.34e-02 |
Pos logit contributions | |
bled | +0.085 |
ble | +0.078 |
balls | +0.075 |
bing | +0.069 |
ster | +0.066 |
Neg logit contributions | |
ird | -0.030 |
hell | -0.030 |
ha | -0.030 |
Karen | -0.029 |
aren | -0.027 |
Token | bag |
Token ID | 6131 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
bled | +0.122 |
ble | +0.103 |
bing | +0.095 |
Miles | +0.093 |
balls | +0.093 |
Neg logit contributions | |
ird | -0.101 |
keys | -0.095 |
hell | -0.094 |
bars | -0.092 |
umbrella | -0.092 |
Token | ery |
Token ID | 1924 |
Feature activation | +5.48e-02 |
Pos logit contributions | |
ble | +0.063 |
bled | +0.056 |
fin | +0.055 |
balls | +0.054 |
ster | +0.053 |
Neg logit contributions | |
ird | -0.070 |
rell | -0.070 |
hell | -0.068 |
aren | -0.064 |
ives | -0.064 |
Token | became |
Token ID | 2627 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
ble | +0.127 |
bled | +0.118 |
balls | +0.109 |
fin | +0.106 |
ster | +0.105 |
Neg logit contributions | |
rell | -0.077 |
ird | -0.076 |
old | -0.067 |
hell | -0.067 |
ives | -0.067 |
Token | one |
Token ID | 530 |
Feature activation | -8.63e-03 |
Pos logit contributions | |
ble | +0.024 |
bled | +0.022 |
balls | +0.021 |
ster | +0.020 |
fin | +0.020 |
Neg logit contributions | |
ird | -0.019 |
Karen | -0.017 |
rell | -0.017 |
aren | -0.017 |
Ellen | -0.016 |
Token | of |
Token ID | 286 |
Feature activation | +2.34e-03 |
Pos logit contributions | |
ird | +0.009 |
aren | +0.008 |
Karen | +0.008 |
rell | +0.008 |
old | +0.007 |
Neg logit contributions | |
ble | -0.025 |
bled | -0.024 |
balls | -0.023 |
ster | -0.022 |
fin | -0.022 |
Token | her |
Token ID | 607 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
ble | +0.005 |
bled | +0.005 |
balls | +0.005 |
ster | +0.005 |
fin | +0.005 |
Neg logit contributions | |
ird | -0.004 |
Karen | -0.004 |
rell | -0.004 |
aren | -0.004 |
Ellen | -0.003 |
Token | favourite |
Token ID | 12507 |
Feature activation | +5.00e-02 |
Pos logit contributions | |
ble | +0.069 |
bled | +0.064 |
balls | +0.057 |
fin | +0.054 |
ster | +0.054 |
Neg logit contributions | |
ird | -0.056 |
aren | -0.051 |
Karen | -0.051 |
rell | -0.050 |
old | -0.048 |
Token | snacks |
Token ID | 28487 |
Feature activation | +5.52e-03 |
Pos logit contributions | |
ble | +0.104 |
bled | +0.100 |
balls | +0.087 |
ster | +0.081 |
fin | +0.077 |
Neg logit contributions | |
ird | -0.090 |
rell | -0.085 |
hell | -0.085 |
Karen | -0.080 |
aren | -0.079 |
Token | . |
Token ID | 13 |
Feature activation | -2.65e-02 |
Pos logit contributions | |
ble | +0.013 |
bled | +0.011 |
balls | +0.011 |
ster | +0.011 |
fin | +0.010 |
Neg logit contributions | |
ird | -0.009 |
rell | -0.009 |
Karen | -0.009 |
aren | -0.009 |
Ellen | -0.008 |
Token | |
Token ID | 220 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
ird | +0.053 |
keys | +0.048 |
rell | +0.048 |
Aud | +0.046 |
hell | +0.046 |
Neg logit contributions | |
ble | -0.041 |
balls | -0.037 |
Miles | -0.036 |
emo | -0.036 |
Mitch | -0.034 |
Token |
|
Token ID | 198 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
ird | +0.022 |
aren | +0.021 |
rell | +0.021 |
Karen | +0.021 |
Ellen | +0.020 |
Neg logit contributions | |
ble | -0.026 |
bled | -0.023 |
balls | -0.022 |
ster | -0.020 |
emo | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
ird | +0.032 |
rell | +0.030 |
aren | +0.028 |
Karen | +0.028 |
Ellen | +0.027 |
Neg logit contributions | |
ble | -0.031 |
balls | -0.027 |
bled | -0.027 |
emo | -0.025 |
ster | -0.024 |
Token | careful |
Token ID | 8161 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
ble | +0.064 |
bled | +0.060 |
ster | +0.058 |
fin | +0.054 |
shark | +0.052 |
Neg logit contributions | |
ird | -0.032 |
Karen | -0.031 |
Alice | -0.030 |
rell | -0.028 |
Stella | -0.028 |
Token | not |
Token ID | 407 |
Feature activation | +2.67e-02 |
Pos logit contributions | |
ble | +0.031 |
bled | +0.029 |
ster | +0.027 |
fin | +0.026 |
shark | +0.025 |
Neg logit contributions | |
ird | -0.025 |
rell | -0.025 |
Karen | -0.024 |
hell | -0.023 |
aren | -0.023 |
Token | to |
Token ID | 284 |
Feature activation | +3.07e-02 |
Pos logit contributions | |
ble | +0.066 |
bled | +0.065 |
ster | +0.057 |
fin | +0.054 |
balls | +0.054 |
Neg logit contributions | |
ird | -0.034 |
rell | -0.032 |
Karen | -0.031 |
hell | -0.030 |
old | -0.030 |
Token | touch |
Token ID | 3638 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
ble | +0.065 |
bled | +0.065 |
ster | +0.058 |
balls | +0.058 |
fin | +0.054 |
Neg logit contributions | |
ird | -0.054 |
rell | -0.049 |
Karen | -0.049 |
aren | -0.047 |
Stella | -0.045 |
Token | the |
Token ID | 262 |
Feature activation | +3.63e-02 |
Pos logit contributions | |
ble | +0.047 |
bled | +0.043 |
balls | +0.040 |
ster | +0.040 |
fin | +0.039 |
Neg logit contributions | |
ird | -0.037 |
rell | -0.034 |
Karen | -0.034 |
aren | -0.034 |
Ellen | -0.031 |
Token | hot |
Token ID | 3024 |
Feature activation | +4.92e-02 |
Pos logit contributions | |
ble | +0.069 |
bled | +0.064 |
balls | +0.056 |
fin | +0.055 |
ster | +0.055 |
Neg logit contributions | |
ird | -0.072 |
rell | -0.069 |
Karen | -0.066 |
aren | -0.066 |
hell | -0.064 |
Token | stove |
Token ID | 27635 |
Feature activation | +2.75e-02 |
Pos logit contributions | |
ble | +0.101 |
bled | +0.095 |
balls | +0.085 |
ster | +0.082 |
fin | +0.082 |
Neg logit contributions | |
ird | -0.092 |
rell | -0.089 |
Karen | -0.084 |
aren | -0.084 |
hell | -0.079 |
Token | . |
Token ID | 13 |
Feature activation | +7.75e-03 |
Pos logit contributions | |
ble | +0.075 |
bled | +0.072 |
balls | +0.065 |
ster | +0.065 |
fin | +0.064 |
Neg logit contributions | |
ird | -0.032 |
rell | -0.028 |
Karen | -0.027 |
hell | -0.026 |
aren | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
ble | +0.018 |
bled | +0.017 |
balls | +0.016 |
ster | +0.015 |
fin | +0.015 |
Neg logit contributions | |
ird | -0.013 |
Karen | -0.012 |
rell | -0.012 |
aren | -0.012 |
Ellen | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | +2.97e-02 |
Pos logit contributions | |
ble | +0.026 |
bled | +0.023 |
balls | +0.022 |
fin | +0.021 |
ster | +0.021 |
Neg logit contributions | |
ird | -0.020 |
rell | -0.019 |
aren | -0.018 |
Karen | -0.018 |
Ellen | -0.017 |
Token | They |
Token ID | 2990 |
Feature activation | +1.35e-02 |
Pos logit contributions | |
ble | +0.066 |
bled | +0.062 |
ster | +0.057 |
fin | +0.056 |
balls | +0.056 |
Neg logit contributions | |
ird | -0.051 |
aren | -0.047 |
Karen | -0.047 |
Alice | -0.046 |
Ell | -0.045 |
Token | big |
Token ID | 1263 |
Feature activation | +2.68e-03 |
Pos logit contributions | |
ble | +0.029 |
bled | +0.027 |
balls | +0.025 |
ster | +0.024 |
fin | +0.024 |
Neg logit contributions | |
ird | -0.024 |
Karen | -0.022 |
aren | -0.022 |
rell | -0.021 |
Ellen | -0.020 |
Token | sea |
Token ID | 5417 |
Feature activation | +4.70e-02 |
Pos logit contributions | |
ble | +0.006 |
bled | +0.006 |
balls | +0.005 |
ster | +0.005 |
fin | +0.005 |
Neg logit contributions | |
ird | -0.004 |
Karen | -0.004 |
rell | -0.004 |
aren | -0.004 |
Ellen | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -1.47e-03 |
Pos logit contributions | |
bled | +0.127 |
ble | +0.127 |
balls | +0.109 |
fin | +0.104 |
ster | +0.103 |
Neg logit contributions | |
ird | -0.051 |
hell | -0.044 |
chair | -0.044 |
aren | -0.043 |
rell | -0.042 |
Token | In |
Token ID | 554 |
Feature activation | +4.30e-02 |
Pos logit contributions | |
ird | +0.003 |
aren | +0.002 |
rell | +0.002 |
Karen | +0.002 |
old | +0.002 |
Neg logit contributions | |
ble | -0.003 |
bled | -0.003 |
balls | -0.003 |
fin | -0.003 |
ster | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | +3.77e-02 |
Pos logit contributions | |
balls | +0.118 |
bled | +0.113 |
ble | +0.111 |
ster | +0.105 |
ider | +0.101 |
Neg logit contributions | |
rell | -0.037 |
Karen | -0.034 |
Stella | -0.030 |
Alice | -0.030 |
course | -0.030 |
Token | sea |
Token ID | 5417 |
Feature activation | +5.24e-02 |
Pos logit contributions | |
bled | +0.077 |
balls | +0.077 |
ble | +0.076 |
ster | +0.066 |
bing | +0.064 |
Neg logit contributions | |
rell | -0.055 |
he | -0.053 |
ird | -0.050 |
hell | -0.050 |
aren | -0.047 |
Token | , |
Token ID | 11 |
Feature activation | +3.41e-02 |
Pos logit contributions | |
bled | +0.157 |
ble | +0.154 |
balls | +0.147 |
fin | +0.136 |
ster | +0.135 |
Neg logit contributions | |
ird | -0.037 |
aren | -0.029 |
old | -0.029 |
rell | -0.028 |
hell | -0.026 |
Token | there |
Token ID | 612 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
ble | +0.084 |
balls | +0.083 |
ster | +0.080 |
bled | +0.080 |
fin | +0.079 |
Neg logit contributions | |
Karen | -0.038 |
ird | -0.037 |
Stella | -0.033 |
Ellie | -0.032 |
Ellen | -0.032 |
Token | was |
Token ID | 373 |
Feature activation | +3.11e-02 |
Pos logit contributions | |
ble | +0.024 |
bled | +0.023 |
balls | +0.022 |
ster | +0.021 |
fin | +0.021 |
Neg logit contributions | |
ird | -0.017 |
Karen | -0.015 |
rell | -0.015 |
aren | -0.015 |
hell | -0.014 |
Token | a |
Token ID | 257 |
Feature activation | +4.14e-02 |
Pos logit contributions | |
ble | +0.085 |
bled | +0.080 |
ster | +0.079 |
balls | +0.077 |
bing | +0.074 |
Neg logit contributions | |
Karen | -0.030 |
rell | -0.029 |
ird | -0.028 |
Stella | -0.026 |
aren | -0.026 |
Token | small |
Token ID | 1402 |
Feature activation | +3.83e-02 |
Pos logit contributions | |
bled | +0.108 |
ble | +0.101 |
balls | +0.096 |
bing | +0.092 |
ster | +0.085 |
Neg logit contributions | |
ird | -0.043 |
ha | -0.038 |
Karen | -0.038 |
hell | -0.037 |
aren | -0.036 |
Token | and |
Token ID | 290 |
Feature activation | +6.04e-03 |
Pos logit contributions | |
ble | +0.078 |
bled | +0.075 |
balls | +0.069 |
ster | +0.066 |
fin | +0.065 |
Neg logit contributions | |
ird | -0.038 |
Karen | -0.034 |
aren | -0.033 |
rell | -0.033 |
Ellen | -0.030 |
Token | smiles |
Token ID | 21845 |
Feature activation | +1.29e-02 |
Pos logit contributions | |
ble | +0.016 |
bled | +0.015 |
balls | +0.013 |
fin | +0.013 |
emo | +0.013 |
Neg logit contributions | |
ird | -0.007 |
rell | -0.007 |
old | -0.007 |
Karen | -0.006 |
Alice | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +3.70e-02 |
Pos logit contributions | |
ble | +0.035 |
bled | +0.033 |
balls | +0.031 |
fin | +0.030 |
ster | +0.030 |
Neg logit contributions | |
ird | -0.014 |
rell | -0.013 |
Karen | -0.013 |
Ellen | -0.012 |
aren | -0.012 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +3.49e-02 |
Pos logit contributions | |
ble | +0.093 |
bled | +0.087 |
balls | +0.080 |
ster | +0.079 |
fin | +0.076 |
Neg logit contributions | |
ird | -0.053 |
Karen | -0.050 |
aren | -0.048 |
rell | -0.048 |
Ellen | -0.045 |
Token | S |
Token ID | 50 |
Feature activation | -2.59e-03 |
Pos logit contributions | |
ble | +0.077 |
bled | +0.071 |
balls | +0.066 |
ster | +0.065 |
fin | +0.064 |
Neg logit contributions | |
ird | -0.061 |
rell | -0.057 |
aren | -0.056 |
Karen | -0.056 |
Ellen | -0.052 |
Token | ara |
Token ID | 3301 |
Feature activation | +3.26e-02 |
Pos logit contributions | |
ird | +0.005 |
rell | +0.004 |
Karen | +0.004 |
aren | +0.004 |
Ellen | +0.004 |
Neg logit contributions | |
ble | -0.005 |
bled | -0.005 |
ster | -0.005 |
balls | -0.005 |
fin | -0.004 |
Token | liked |
Token ID | 8288 |
Feature activation | +2.19e-02 |
Pos logit contributions | |
ble | +0.084 |
bled | +0.078 |
balls | +0.074 |
fin | +0.070 |
ster | +0.070 |
Neg logit contributions | |
ird | -0.045 |
aren | -0.041 |
rell | -0.041 |
Karen | -0.040 |
old | -0.039 |
Token | to |
Token ID | 284 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
ble | +0.057 |
bled | +0.054 |
ster | +0.050 |
fin | +0.047 |
balls | +0.046 |
Neg logit contributions | |
Karen | -0.029 |
rell | -0.026 |
Stella | -0.026 |
Mad | -0.026 |
Ellie | -0.025 |
Token | play |
Token ID | 711 |
Feature activation | +2.35e-02 |
Pos logit contributions | |
ble | +0.044 |
bled | +0.042 |
balls | +0.038 |
ster | +0.038 |
fin | +0.036 |
Neg logit contributions | |
ird | -0.032 |
rell | -0.029 |
Karen | -0.029 |
aren | -0.027 |
Ellen | -0.027 |
Token | with |
Token ID | 351 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
ble | +0.049 |
bled | +0.046 |
ster | +0.042 |
balls | +0.041 |
fin | +0.041 |
Neg logit contributions | |
ird | -0.040 |
rell | -0.035 |
he | -0.035 |
Ellie | -0.034 |
hell | -0.034 |
Token | her |
Token ID | 607 |
Feature activation | +2.54e-02 |
Pos logit contributions | |
ble | +0.036 |
bled | +0.033 |
ster | +0.032 |
balls | +0.031 |
fin | +0.030 |
Neg logit contributions | |
ird | -0.016 |
Karen | -0.015 |
rell | -0.015 |
Ellen | -0.014 |
aren | -0.014 |
Token | - |
Token ID | 12 |
Feature activation | -2.45e-02 |
Pos logit contributions | |
ble | +0.001 |
bled | +0.001 |
balls | +0.001 |
ster | +0.001 |
fin | +0.001 |
Neg logit contributions | |
ird | -0.001 |
rell | -0.001 |
aren | -0.001 |
Karen | -0.001 |
Ellen | -0.001 |
Token | ray |
Token ID | 2433 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
ird | +0.047 |
rell | +0.043 |
aren | +0.043 |
Karen | +0.043 |
Stella | +0.040 |
Neg logit contributions | |
ble | -0.050 |
bled | -0.045 |
ster | -0.043 |
balls | -0.042 |
fin | -0.041 |
Token | that |
Token ID | 326 |
Feature activation | +7.52e-03 |
Pos logit contributions | |
ble | +0.041 |
bled | +0.038 |
balls | +0.035 |
ster | +0.033 |
fin | +0.033 |
Neg logit contributions | |
ird | -0.032 |
rell | -0.028 |
Karen | -0.027 |
aren | -0.027 |
old | -0.026 |
Token | he |
Token ID | 339 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
ble | +0.016 |
bled | +0.015 |
balls | +0.014 |
ster | +0.014 |
fin | +0.013 |
Neg logit contributions | |
ird | -0.013 |
rell | -0.012 |
Karen | -0.012 |
aren | -0.012 |
Ellen | -0.011 |
Token | shone |
Token ID | 44193 |
Feature activation | +9.02e-04 |
Pos logit contributions | |
ble | +0.036 |
bled | +0.033 |
balls | +0.031 |
ster | +0.030 |
fin | +0.030 |
Neg logit contributions | |
ird | -0.031 |
rell | -0.029 |
Karen | -0.029 |
aren | -0.029 |
Ellen | -0.027 |
Token | on |
Token ID | 319 |
Feature activation | +9.26e-03 |
Pos logit contributions | |
ble | +0.002 |
bled | +0.002 |
balls | +0.001 |
ster | +0.001 |
fin | +0.001 |
Neg logit contributions | |
ird | -0.002 |
Karen | -0.002 |
rell | -0.002 |
aren | -0.002 |
Ellen | -0.002 |
Token | the |
Token ID | 262 |
Feature activation | -8.08e-03 |
Pos logit contributions | |
ble | +0.019 |
bled | +0.017 |
balls | +0.016 |
ster | +0.016 |
fin | +0.015 |
Neg logit contributions | |
ird | -0.018 |
Karen | -0.017 |
rell | -0.017 |
aren | -0.016 |
Ellen | -0.016 |
Token | tree |
Token ID | 5509 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
ird | +0.016 |
rell | +0.015 |
Karen | +0.015 |
aren | +0.015 |
Ellen | +0.014 |
Neg logit contributions | |
ble | -0.017 |
bled | -0.015 |
balls | -0.014 |
ster | -0.014 |
fin | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -7.80e-03 |
Pos logit contributions | |
ble | +0.026 |
bled | +0.024 |
balls | +0.023 |
ster | +0.022 |
fin | +0.022 |
Neg logit contributions | |
ird | -0.017 |
rell | -0.016 |
Karen | -0.016 |
aren | -0.016 |
Ellen | -0.014 |
Token | With |
Token ID | 2080 |
Feature activation | -4.12e-03 |
Pos logit contributions | |
ird | +0.016 |
aren | +0.015 |
rell | +0.015 |
Karen | +0.014 |
hell | +0.014 |
Neg logit contributions | |
ble | -0.014 |
bled | -0.013 |
balls | -0.013 |
fin | -0.012 |
ster | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
ird | +0.008 |
aren | +0.008 |
rell | +0.008 |
Karen | +0.008 |
Ellen | +0.007 |
Neg logit contributions | |
ble | -0.008 |
bled | -0.007 |
balls | -0.007 |
ster | -0.006 |
fin | -0.006 |
Token | Sam |
Token ID | 16305 |
Feature activation | -5.87e-03 |
Pos logit contributions | |
ble | +0.063 |
bled | +0.056 |
balls | +0.053 |
ster | +0.052 |
fin | +0.051 |
Neg logit contributions | |
ird | -0.056 |
Karen | -0.052 |
rell | -0.052 |
aren | -0.052 |
Ellen | -0.049 |
Token | and |
Token ID | 290 |
Feature activation | +6.44e-03 |
Pos logit contributions | |
ird | +0.011 |
Karen | +0.010 |
rell | +0.010 |
aren | +0.010 |
Ellen | +0.010 |
Neg logit contributions | |
ble | -0.012 |
bled | -0.011 |
ster | -0.010 |
fin | -0.010 |
balls | -0.010 |
Token | Mia |
Token ID | 32189 |
Feature activation | -2.35e-02 |
Pos logit contributions | |
ble | +0.014 |
bled | +0.013 |
balls | +0.012 |
ster | +0.012 |
fin | +0.011 |
Neg logit contributions | |
ird | -0.012 |
Karen | -0.011 |
aren | -0.011 |
rell | -0.011 |
Ellen | -0.010 |
Token | are |
Token ID | 389 |
Feature activation | -8.02e-03 |
Pos logit contributions | |
ird | +0.043 |
Karen | +0.041 |
rell | +0.041 |
Ellen | +0.040 |
aren | +0.040 |
Neg logit contributions | |
ble | -0.046 |
bled | -0.040 |
ster | -0.039 |
fin | -0.037 |
balls | -0.036 |
Token | friends |
Token ID | 2460 |
Feature activation | -3.24e-02 |
Pos logit contributions | |
ird | +0.016 |
Karen | +0.015 |
aren | +0.015 |
rell | +0.015 |
Ellen | +0.014 |
Neg logit contributions | |
ble | -0.016 |
bled | -0.014 |
balls | -0.013 |
ster | -0.013 |
fin | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | +4.39e-03 |
Pos logit contributions | |
ird | +0.060 |
rell | +0.058 |
aren | +0.055 |
Karen | +0.055 |
Ellen | +0.052 |
Neg logit contributions | |
ble | -0.069 |
bled | -0.060 |
ster | -0.057 |
balls | -0.057 |
fin | -0.057 |
Token | They |
Token ID | 1119 |
Feature activation | +1.77e-03 |
Pos logit contributions | |
ble | +0.007 |
balls | +0.006 |
bled | +0.006 |
fin | +0.006 |
ster | +0.006 |
Neg logit contributions | |
ird | -0.010 |
rell | -0.009 |
aren | -0.008 |
hell | -0.008 |
Karen | -0.008 |
Token | like |
Token ID | 588 |
Feature activation | +8.32e-03 |
Pos logit contributions | |
ble | +0.004 |
bled | +0.004 |
balls | +0.004 |
ster | +0.003 |
fin | +0.003 |
Neg logit contributions | |
ird | -0.003 |
aren | -0.003 |
Karen | -0.003 |
rell | -0.003 |
chairs | -0.002 |
Token | to |
Token ID | 284 |
Feature activation | +8.65e-03 |
Pos logit contributions | |
ble | +0.020 |
bled | +0.018 |
ster | +0.016 |
balls | +0.016 |
fin | +0.016 |
Neg logit contributions | |
ird | -0.013 |
Karen | -0.013 |
rell | -0.012 |
aren | -0.012 |
Stella | -0.012 |
Token | play |
Token ID | 711 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
ble | +0.018 |
bled | +0.017 |
balls | +0.016 |
ster | +0.015 |
fin | +0.015 |
Neg logit contributions | |
ird | -0.016 |
rell | -0.015 |
aren | -0.015 |
Karen | -0.015 |
Ellen | -0.014 |
Token | outside |
Token ID | 2354 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
ble | +0.031 |
bled | +0.028 |
ster | +0.026 |
fin | +0.026 |
balls | +0.025 |
Neg logit contributions | |
ird | -0.027 |
Karen | -0.023 |
rell | -0.023 |
Ellen | -0.023 |
Ellie | -0.022 |
Token | oven |
Token ID | 14361 |
Feature activation | +6.29e-03 |
Pos logit contributions | |
ble | +0.016 |
bled | +0.015 |
balls | +0.014 |
ster | +0.014 |
fin | +0.013 |
Neg logit contributions | |
ird | -0.014 |
Karen | -0.013 |
rell | -0.013 |
aren | -0.013 |
Ellen | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
ble | +0.014 |
bled | +0.013 |
balls | +0.012 |
ster | +0.012 |
fin | +0.012 |
Neg logit contributions | |
ird | -0.011 |
aren | -0.010 |
rell | -0.010 |
Karen | -0.010 |
Ellen | -0.009 |
Token | took |
Token ID | 1718 |
Feature activation | -2.33e-02 |
Pos logit contributions | |
ird | +0.019 |
rell | +0.018 |
aren | +0.018 |
Karen | +0.018 |
hell | +0.017 |
Neg logit contributions | |
ble | -0.025 |
bled | -0.023 |
balls | -0.022 |
ster | -0.022 |
fin | -0.021 |
Token | out |
Token ID | 503 |
Feature activation | +3.93e-03 |
Pos logit contributions | |
ird | +0.048 |
aren | +0.048 |
rell | +0.047 |
hell | +0.044 |
Karen | +0.044 |
Neg logit contributions | |
ble | -0.040 |
bled | -0.038 |
balls | -0.036 |
ster | -0.031 |
fin | -0.031 |
Token | the |
Token ID | 262 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
ble | +0.006 |
bled | +0.005 |
balls | +0.005 |
fin | +0.005 |
diving | +0.004 |
Neg logit contributions | |
ird | -0.009 |
rell | -0.009 |
aren | -0.009 |
hell | -0.008 |
Karen | -0.008 |
Token | cake |
Token ID | 12187 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
ird | +0.024 |
rell | +0.023 |
aren | +0.023 |
Karen | +0.023 |
Ellen | +0.021 |
Neg logit contributions | |
ble | -0.025 |
bled | -0.022 |
balls | -0.022 |
ster | -0.021 |
fin | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -2.98e-03 |
Pos logit contributions | |
ble | +0.044 |
bled | +0.040 |
balls | +0.039 |
ster | +0.038 |
fin | +0.037 |
Neg logit contributions | |
ird | -0.031 |
aren | -0.029 |
Karen | -0.029 |
rell | -0.029 |
Ellen | -0.027 |
Token | She |
Token ID | 1375 |
Feature activation | -2.43e-03 |
Pos logit contributions | |
ird | +0.005 |
rell | +0.004 |
aren | +0.004 |
Karen | +0.004 |
hell | +0.004 |
Neg logit contributions | |
ble | -0.007 |
bled | -0.006 |
balls | -0.006 |
fin | -0.006 |
ster | -0.006 |
Token | checked |
Token ID | 10667 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
ird | +0.004 |
rell | +0.004 |
aren | +0.004 |
Karen | +0.004 |
hell | +0.003 |
Neg logit contributions | |
ble | -0.005 |
bled | -0.005 |
balls | -0.005 |
ster | -0.005 |
fin | -0.005 |
Token | if |
Token ID | 611 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
ird | +0.023 |
rell | +0.022 |
aren | +0.021 |
Karen | +0.020 |
Ell | +0.020 |
Neg logit contributions | |
ble | -0.022 |
bled | -0.020 |
balls | -0.018 |
ster | -0.018 |
fin | -0.017 |
Token | Sara |
Token ID | 24799 |
Feature activation | +3.81e-03 |
Pos logit contributions | |
ble | +0.031 |
bled | +0.028 |
balls | +0.026 |
ster | +0.026 |
fin | +0.026 |
Neg logit contributions | |
ird | -0.022 |
Karen | -0.021 |
aren | -0.021 |
rell | -0.020 |
Ellen | -0.019 |
Token | told |
Token ID | 1297 |
Feature activation | +2.57e-02 |
Pos logit contributions | |
ble | +0.069 |
bled | +0.063 |
balls | +0.060 |
ster | +0.057 |
fin | +0.056 |
Neg logit contributions | |
ird | -0.041 |
Karen | -0.037 |
aren | -0.037 |
rell | -0.036 |
Ellen | -0.033 |
Token | his |
Token ID | 465 |
Feature activation | +3.73e-02 |
Pos logit contributions | |
ble | +0.036 |
bled | +0.030 |
balls | +0.027 |
ster | +0.026 |
fin | +0.025 |
Neg logit contributions | |
ird | -0.067 |
rell | -0.064 |
Karen | -0.064 |
aren | -0.064 |
Ellen | -0.061 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.92e-02 |
Pos logit contributions | |
ble | +0.072 |
bled | +0.068 |
balls | +0.059 |
ster | +0.056 |
bing | +0.055 |
Neg logit contributions | |
ird | -0.071 |
Karen | -0.070 |
aren | -0.067 |
old | -0.066 |
rell | -0.065 |
Token | about |
Token ID | 546 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
ird | +0.030 |
rell | +0.027 |
aren | +0.027 |
Karen | +0.026 |
Ellen | +0.024 |
Neg logit contributions | |
ble | -0.047 |
bled | -0.043 |
balls | -0.041 |
ster | -0.040 |
fin | -0.039 |
Token | the |
Token ID | 262 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
ble | +0.041 |
bled | +0.038 |
balls | +0.035 |
ster | +0.035 |
fin | +0.033 |
Neg logit contributions | |
Karen | -0.037 |
ird | -0.035 |
aren | -0.034 |
rell | -0.034 |
Ellen | -0.033 |
Token | monster |
Token ID | 9234 |
Feature activation | +3.38e-03 |
Pos logit contributions | |
ble | +0.035 |
bled | +0.033 |
balls | +0.030 |
ster | +0.029 |
fin | +0.028 |
Neg logit contributions | |
ird | -0.029 |
rell | -0.028 |
Karen | -0.027 |
aren | -0.027 |
hell | -0.026 |
Token | he |
Token ID | 339 |
Feature activation | +3.02e-03 |
Pos logit contributions | |
ble | +0.007 |
bled | +0.007 |
balls | +0.006 |
ster | +0.006 |
fin | +0.006 |
Neg logit contributions | |
ird | -0.006 |
rell | -0.006 |
aren | -0.005 |
Karen | -0.005 |
Ellen | -0.005 |
Token | found |
Token ID | 1043 |
Feature activation | +1.23e-03 |
Pos logit contributions | |
ble | +0.007 |
bled | +0.006 |
ster | +0.006 |
balls | +0.006 |
fin | +0.006 |
Neg logit contributions | |
ird | -0.005 |
rell | -0.005 |
Karen | -0.005 |
aren | -0.005 |
Ellen | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +6.25e-03 |
Pos logit contributions | |
ble | +0.003 |
bled | +0.002 |
balls | +0.002 |
ster | +0.002 |
fin | +0.002 |
Neg logit contributions | |
ird | -0.002 |
rell | -0.002 |
aren | -0.002 |
Karen | -0.002 |
Ellen | -0.002 |
Token | His |
Token ID | 2399 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
ble | +0.015 |
bled | +0.013 |
ster | +0.012 |
balls | +0.012 |
fin | +0.012 |
Neg logit contributions | |
ird | -0.011 |
Karen | -0.010 |
rell | -0.010 |
aren | -0.010 |
Stella | -0.009 |
Token | mom |
Token ID | 1995 |
Feature activation | -2.85e-02 |
Pos logit contributions | |
ble | +0.042 |
bled | +0.038 |
balls | +0.034 |
ster | +0.033 |
fin | +0.032 |
Neg logit contributions | |
ird | -0.047 |
Karen | -0.045 |
aren | -0.044 |
rell | -0.043 |
Ellen | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
ble | +0.071 |
bled | +0.068 |
balls | +0.063 |
ster | +0.063 |
fin | +0.061 |
Neg logit contributions | |
ird | -0.029 |
rell | -0.025 |
Karen | -0.025 |
hell | -0.023 |
Ellen | -0.022 |
Token |
|
Token ID | 198 |
Feature activation | +3.03e-02 |
Pos logit contributions | |
ble | +0.037 |
bled | +0.034 |
balls | +0.031 |
ster | +0.031 |
fin | +0.029 |
Neg logit contributions | |
ird | -0.039 |
Karen | -0.036 |
rell | -0.036 |
aren | -0.036 |
Ellen | -0.034 |
Token |
|
Token ID | 198 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
ble | +0.079 |
bled | +0.073 |
balls | +0.069 |
ster | +0.069 |
fin | +0.067 |
Neg logit contributions | |
ird | -0.042 |
Karen | -0.037 |
aren | -0.037 |
rell | -0.036 |
old | -0.033 |
Token | Anna |
Token ID | 31160 |
Feature activation | -1.76e-03 |
Pos logit contributions | |
ble | +0.087 |
bled | +0.082 |
balls | +0.076 |
ster | +0.074 |
fin | +0.072 |
Neg logit contributions | |
ird | -0.063 |
aren | -0.057 |
Karen | -0.056 |
old | -0.056 |
Alice | -0.055 |
Token | heard |
Token ID | 2982 |
Feature activation | +1.00e-02 |
Pos logit contributions | |
ird | +0.003 |
rell | +0.003 |
Karen | +0.003 |
aren | +0.003 |
Ellen | +0.003 |
Neg logit contributions | |
ble | -0.004 |
bled | -0.003 |
ster | -0.003 |
balls | -0.003 |
fin | -0.003 |
Token | him |
Token ID | 683 |
Feature activation | -7.43e-03 |
Pos logit contributions | |
ble | +0.019 |
bled | +0.017 |
balls | +0.016 |
fin | +0.015 |
ster | +0.015 |
Neg logit contributions | |
ird | -0.021 |
aren | -0.020 |
rell | -0.020 |
Karen | -0.019 |
Ell | -0.018 |
Token | cry |
Token ID | 3960 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
ird | +0.011 |
rell | +0.010 |
Karen | +0.010 |
aren | +0.009 |
Ellen | +0.009 |
Neg logit contributions | |
ble | -0.019 |
bled | -0.017 |
balls | -0.017 |
ster | -0.016 |
fin | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | +5.74e-03 |
Pos logit contributions | |
ble | +0.031 |
bled | +0.029 |
balls | +0.027 |
ster | +0.027 |
fin | +0.026 |
Neg logit contributions | |
ird | -0.015 |
rell | -0.014 |
Karen | -0.013 |
aren | -0.012 |
hell | -0.012 |
Token | came |
Token ID | 1625 |
Feature activation | +8.33e-03 |
Pos logit contributions | |
ble | +0.014 |
bled | +0.012 |
balls | +0.012 |
ster | +0.011 |
fin | +0.011 |
Neg logit contributions | |
ird | -0.009 |
Karen | -0.009 |
rell | -0.008 |
aren | -0.008 |
Ellen | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | -3.80e-03 |
Pos logit contributions | |
ble | +0.018 |
bled | +0.016 |
ster | +0.015 |
balls | +0.015 |
fin | +0.015 |
Neg logit contributions | |
ird | -0.014 |
Karen | -0.014 |
rell | -0.014 |
aren | -0.013 |
Ellen | -0.013 |
Token | see |
Token ID | 766 |
Feature activation | +7.70e-04 |
Pos logit contributions | |
ird | +0.007 |
rell | +0.007 |
aren | +0.006 |
Karen | +0.006 |
hell | +0.006 |
Neg logit contributions | |
ble | -0.008 |
bled | -0.007 |
balls | -0.007 |
ster | -0.007 |
fin | -0.007 |
Token | green |
Token ID | 4077 |
Feature activation | -1.82e-03 |
Pos logit contributions | |
ird | +0.006 |
rell | +0.006 |
Karen | +0.006 |
aren | +0.006 |
Stella | +0.005 |
Neg logit contributions | |
ble | -0.008 |
bled | -0.007 |
balls | -0.007 |
ster | -0.007 |
fin | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | +4.70e-03 |
Pos logit contributions | |
ird | +0.003 |
rell | +0.002 |
Karen | +0.002 |
aren | +0.002 |
Ellen | +0.002 |
Neg logit contributions | |
ble | -0.005 |
bled | -0.004 |
balls | -0.004 |
ster | -0.004 |
fin | -0.004 |
Token | brown |
Token ID | 7586 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
ble | +0.011 |
bled | +0.010 |
ster | +0.009 |
balls | +0.009 |
fin | +0.009 |
Neg logit contributions | |
ird | -0.007 |
rell | -0.007 |
Karen | -0.007 |
aren | -0.007 |
Stella | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | -6.60e-03 |
Pos logit contributions | |
ird | +0.018 |
aren | +0.016 |
Karen | +0.016 |
rell | +0.016 |
Ellen | +0.015 |
Neg logit contributions | |
ble | -0.035 |
bled | -0.032 |
balls | -0.030 |
ster | -0.030 |
fin | -0.029 |
Token | like |
Token ID | 588 |
Feature activation | -8.97e-03 |
Pos logit contributions | |
ird | +0.010 |
Karen | +0.010 |
rell | +0.010 |
aren | +0.009 |
Stella | +0.009 |
Neg logit contributions | |
ble | -0.015 |
bled | -0.014 |
balls | -0.014 |
ster | -0.013 |
fin | -0.013 |
Token | a |
Token ID | 257 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
ird | +0.018 |
aren | +0.017 |
rell | +0.016 |
Karen | +0.016 |
old | +0.015 |
Neg logit contributions | |
ble | -0.018 |
bled | -0.016 |
balls | -0.015 |
fin | -0.015 |
ster | -0.015 |
Token | leaf |
Token ID | 12835 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
ird | +0.022 |
rell | +0.020 |
Karen | +0.020 |
aren | +0.020 |
Ellen | +0.019 |
Neg logit contributions | |
ble | -0.025 |
bled | -0.023 |
balls | -0.022 |
ster | -0.021 |
fin | -0.021 |
Token | ," |
Token ID | 553 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
ird | +0.016 |
aren | +0.014 |
Karen | +0.014 |
rell | +0.014 |
Ellen | +0.013 |
Neg logit contributions | |
ble | -0.038 |
bled | -0.034 |
balls | -0.034 |
ster | -0.033 |
fin | -0.033 |
Token | Lily |
Token ID | 20037 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
ble | +0.032 |
balls | +0.029 |
bled | +0.028 |
fin | +0.027 |
ster | +0.027 |
Neg logit contributions | |
ird | -0.031 |
rell | -0.029 |
aren | -0.029 |
Karen | -0.027 |
hell | -0.026 |
Token | said |
Token ID | 531 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
aren | +0.041 |
rell | +0.040 |
ird | +0.040 |
Karen | +0.040 |
irst | +0.038 |
Neg logit contributions | |
ble | -0.046 |
ster | -0.040 |
fin | -0.040 |
balls | -0.039 |
bled | -0.038 |
Token | , |
Token ID | 11 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
ird | +0.028 |
aren | +0.026 |
rell | +0.026 |
Karen | +0.026 |
Ellen | +0.024 |
Neg logit contributions | |
ble | -0.043 |
bled | -0.040 |
balls | -0.038 |
ster | -0.037 |
fin | -0.036 |
Token | cat |
Token ID | 3797 |
Feature activation | -4.28e-03 |
Pos logit contributions | |
rell | +0.013 |
ird | +0.013 |
aren | +0.012 |
hell | +0.012 |
Karen | +0.012 |
Neg logit contributions | |
ble | -0.007 |
ster | -0.007 |
balls | -0.006 |
next | -0.006 |
rod | -0.006 |
Token | a |
Token ID | 257 |
Feature activation | -1.92e-02 |
Pos logit contributions | |
ird | +0.009 |
aren | +0.009 |
rell | +0.008 |
Karen | +0.008 |
Ellen | +0.008 |
Neg logit contributions | |
ble | -0.008 |
bled | -0.007 |
balls | -0.007 |
ster | -0.006 |
fin | -0.006 |
Token | toy |
Token ID | 13373 |
Feature activation | +4.62e-03 |
Pos logit contributions | |
ird | +0.034 |
rell | +0.032 |
aren | +0.032 |
Karen | +0.032 |
Ellen | +0.029 |
Neg logit contributions | |
ble | -0.042 |
bled | -0.036 |
balls | -0.036 |
fin | -0.036 |
ster | -0.036 |
Token | mouse |
Token ID | 10211 |
Feature activation | -8.98e-03 |
Pos logit contributions | |
ble | +0.011 |
bled | +0.010 |
balls | +0.010 |
ster | +0.009 |
fin | +0.009 |
Neg logit contributions | |
ird | -0.007 |
rell | -0.007 |
Karen | -0.007 |
aren | -0.007 |
Ellen | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -2.27e-02 |
Pos logit contributions | |
ird | +0.016 |
aren | +0.015 |
rell | +0.015 |
Karen | +0.015 |
Ellen | +0.014 |
Neg logit contributions | |
ble | -0.019 |
bled | -0.017 |
balls | -0.017 |
ster | -0.017 |
fin | -0.016 |
Token | The |
Token ID | 383 |
Feature activation | -2.68e-03 |
Pos logit contributions | |
ird | +0.042 |
aren | +0.039 |
rell | +0.039 |
Karen | +0.038 |
hell | +0.037 |
Neg logit contributions | |
ble | -0.046 |
bled | -0.041 |
balls | -0.040 |
ster | -0.038 |
fin | -0.038 |
Token | cat |
Token ID | 3797 |
Feature activation | -9.60e-03 |
Pos logit contributions | |
ird | +0.005 |
rell | +0.005 |
aren | +0.005 |
Karen | +0.005 |
hell | +0.004 |
Neg logit contributions | |
ble | -0.006 |
ster | -0.005 |
bled | -0.005 |
balls | -0.005 |
fin | -0.005 |
Token | seemed |
Token ID | 3947 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
ird | +0.019 |
rell | +0.018 |
aren | +0.018 |
Karen | +0.018 |
Ellen | +0.016 |
Neg logit contributions | |
ble | -0.019 |
bled | -0.017 |
balls | -0.016 |
ster | -0.016 |
fin | -0.016 |
Token | to |
Token ID | 284 |
Feature activation | -8.36e-03 |
Pos logit contributions | |
ird | +0.032 |
rell | +0.030 |
aren | +0.030 |
Karen | +0.030 |
Ellen | +0.028 |
Neg logit contributions | |
ble | -0.035 |
bled | -0.031 |
balls | -0.029 |
ster | -0.029 |
fin | -0.028 |
Token | like |
Token ID | 588 |
Feature activation | -3.22e-03 |
Pos logit contributions | |
ird | +0.016 |
aren | +0.015 |
chairs | +0.015 |
chair | +0.014 |
rell | +0.014 |
Neg logit contributions | |
ble | -0.018 |
fin | -0.015 |
bled | -0.014 |
ster | -0.013 |
diving | -0.013 |
Token | it |
Token ID | 340 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
ird | +0.006 |
rell | +0.006 |
aren | +0.006 |
Karen | +0.006 |
Ellen | +0.005 |
Neg logit contributions | |
ble | -0.007 |
bled | -0.006 |
balls | -0.006 |
ster | -0.005 |
fin | -0.005 |
Token | again |
Token ID | 757 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
ble | +0.049 |
bled | +0.045 |
balls | +0.043 |
ster | +0.042 |
fin | +0.041 |
Neg logit contributions | |
ird | -0.030 |
rell | -0.027 |
Karen | -0.027 |
aren | -0.027 |
Ellen | -0.025 |
Token | tomorrow |
Token ID | 9439 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
ble | +0.038 |
bled | +0.035 |
balls | +0.034 |
ster | +0.033 |
fin | +0.032 |
Neg logit contributions | |
ird | -0.023 |
Karen | -0.021 |
rell | -0.021 |
aren | -0.021 |
Ellen | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -5.48e-04 |
Pos logit contributions | |
ble | +0.034 |
bled | +0.031 |
balls | +0.029 |
ster | +0.029 |
fin | +0.028 |
Neg logit contributions | |
ird | -0.021 |
rell | -0.020 |
aren | -0.019 |
Karen | -0.019 |
Ellen | -0.018 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +6.71e-03 |
Pos logit contributions | |
ird | +0.001 |
rell | +0.001 |
aren | +0.001 |
Karen | +0.001 |
hell | +0.001 |
Neg logit contributions | |
ble | -0.001 |
bled | -0.001 |
balls | -0.001 |
ster | -0.001 |
fin | -0.001 |
Token | Once |
Token ID | 7454 |
Feature activation | +3.35e-03 |
Pos logit contributions | |
ble | +0.013 |
bled | +0.011 |
balls | +0.011 |
ster | +0.011 |
fin | +0.010 |
Neg logit contributions | |
ird | -0.013 |
rell | -0.012 |
Karen | -0.012 |
aren | -0.012 |
Ellen | -0.011 |
Token | upon |
Token ID | 2402 |
Feature activation | -8.63e-03 |
Pos logit contributions | |
ble | +0.006 |
bled | +0.005 |
balls | +0.004 |
ster | +0.004 |
fin | +0.004 |
Neg logit contributions | |
ird | -0.008 |
rell | -0.008 |
Karen | -0.007 |
aren | -0.007 |
Ellen | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | -1.95e-02 |
Pos logit contributions | |
ird | +0.018 |
aren | +0.018 |
rell | +0.017 |
Karen | +0.017 |
Ellen | +0.016 |
Neg logit contributions | |
ble | -0.016 |
bled | -0.014 |
balls | -0.013 |
fin | -0.013 |
ster | -0.013 |
Token | time |
Token ID | 640 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
ird | +0.030 |
Karen | +0.029 |
rell | +0.029 |
aren | +0.028 |
Ellen | +0.027 |
Neg logit contributions | |
ble | -0.044 |
balls | -0.039 |
bled | -0.039 |
fin | -0.039 |
ster | -0.038 |
Token | , |
Token ID | 11 |
Feature activation | +4.99e-03 |
Pos logit contributions | |
ird | +0.026 |
rell | +0.024 |
aren | +0.023 |
Karen | +0.023 |
chairs | +0.022 |
Neg logit contributions | |
ble | -0.031 |
bled | -0.027 |
balls | -0.026 |
ster | -0.025 |
fin | -0.025 |
Token | there |
Token ID | 612 |
Feature activation | +6.30e-04 |
Pos logit contributions | |
ble | +0.010 |
bled | +0.009 |
balls | +0.009 |
ster | +0.009 |
fin | +0.009 |
Neg logit contributions | |
ird | -0.009 |
Karen | -0.008 |
aren | -0.008 |
rell | -0.008 |
Ellen | -0.008 |
Token | was |
Token ID | 373 |
Feature activation | +5.99e-03 |
Pos logit contributions | |
ble | +0.001 |
bled | +0.001 |
balls | +0.001 |
ster | +0.001 |
fin | +0.001 |
Neg logit contributions | |
ird | -0.001 |
rell | -0.001 |
aren | -0.001 |
Karen | -0.001 |
Ellen | -0.001 |
Token | very |
Token ID | 845 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
ble | +0.013 |
bled | +0.012 |
balls | +0.011 |
ster | +0.011 |
fin | +0.011 |
Neg logit contributions | |
ird | -0.010 |
Karen | -0.009 |
aren | -0.009 |
rell | -0.009 |
Ellen | -0.008 |
Token | bright |
Token ID | 6016 |
Feature activation | -9.84e-04 |
Pos logit contributions | |
ble | +0.070 |
bled | +0.066 |
balls | +0.061 |
ster | +0.060 |
fin | +0.058 |
Neg logit contributions | |
ird | -0.044 |
Karen | -0.042 |
aren | -0.042 |
rell | -0.040 |
hell | -0.038 |
Token | star |
Token ID | 3491 |
Feature activation | +2.58e-02 |
Pos logit contributions | |
ird | +0.002 |
rell | +0.002 |
aren | +0.002 |
Karen | +0.002 |
Ellen | +0.002 |
Neg logit contributions | |
ble | -0.002 |
bled | -0.002 |
balls | -0.002 |
ster | -0.002 |
fin | -0.002 |
Token | that |
Token ID | 326 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
bled | +0.052 |
ble | +0.052 |
balls | +0.048 |
emo | +0.047 |
fin | +0.044 |
Neg logit contributions | |
ird | -0.044 |
chair | -0.039 |
rell | -0.038 |
hell | -0.037 |
Karen | -0.037 |
Token | was |
Token ID | 373 |
Feature activation | +3.69e-03 |
Pos logit contributions | |
ble | +0.036 |
bled | +0.032 |
balls | +0.031 |
ster | +0.030 |
fin | +0.029 |
Neg logit contributions | |
ird | -0.028 |
Karen | -0.026 |
aren | -0.026 |
rell | -0.026 |
Ellen | -0.024 |
Token | cheerful |
Token ID | 37999 |
Feature activation | +2.16e-03 |
Pos logit contributions | |
ble | +0.007 |
bled | +0.007 |
balls | +0.007 |
ster | +0.007 |
fin | +0.006 |
Neg logit contributions | |
ird | -0.006 |
Karen | -0.006 |
rell | -0.006 |
aren | -0.006 |
Ellen | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | +4.27e-03 |
Pos logit contributions | |
ble | +0.005 |
bled | +0.005 |
balls | +0.005 |
ster | +0.005 |
fin | +0.005 |
Neg logit contributions | |
ird | -0.003 |
rell | -0.003 |
Karen | -0.003 |
aren | -0.003 |
Ellen | -0.003 |
Token | shiny |
Token ID | 22441 |
Feature activation | -3.70e-03 |
Pos logit contributions | |
ble | +0.010 |
bled | +0.010 |
ster | +0.009 |
balls | +0.009 |
fin | +0.009 |
Neg logit contributions | |
ird | -0.006 |
Karen | -0.006 |
Ellen | -0.005 |
rell | -0.005 |
Stella | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +2.19e-03 |
Pos logit contributions | |
ird | +0.005 |
rell | +0.005 |
Karen | +0.005 |
aren | +0.005 |
Ellen | +0.004 |
Neg logit contributions | |
ble | -0.009 |
bled | -0.009 |
balls | -0.008 |
ster | -0.008 |
fin | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | +6.92e-03 |
Pos logit contributions | |
ble | +0.004 |
bled | +0.004 |
balls | +0.004 |
fin | +0.004 |
ster | +0.004 |
Neg logit contributions | |
ird | -0.004 |
rell | -0.004 |
aren | -0.004 |
Karen | -0.004 |
hell | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
ble | +0.013 |
bled | +0.011 |
balls | +0.011 |
emo | +0.011 |
fin | +0.011 |
Neg logit contributions | |
rell | -0.014 |
ird | -0.014 |
aren | -0.012 |
Karen | -0.012 |
chairs | -0.012 |
Token | ." |
Token ID | 526 |
Feature activation | +3.77e-02 |
Pos logit contributions | |
ird | +0.005 |
aren | +0.005 |
Karen | +0.005 |
rell | +0.005 |
Ellen | +0.005 |
Neg logit contributions | |
ble | -0.007 |
bled | -0.007 |
balls | -0.006 |
ster | -0.006 |
fin | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +3.40e-02 |
Pos logit contributions | |
ble | +0.101 |
bled | +0.094 |
balls | +0.088 |
ster | +0.087 |
fin | +0.085 |
Neg logit contributions | |
ird | -0.050 |
Karen | -0.045 |
aren | -0.045 |
rell | -0.044 |
Ellen | -0.040 |
Token |
|
Token ID | 198 |
Feature activation | +4.87e-02 |
Pos logit contributions | |
ble | +0.086 |
bled | +0.079 |
balls | +0.076 |
ster | +0.074 |
fin | +0.073 |
Neg logit contributions | |
ird | -0.049 |
rell | -0.045 |
Karen | -0.044 |
aren | -0.044 |
Ellen | -0.040 |
Token | Ben |
Token ID | 11696 |
Feature activation | -6.88e-03 |
Pos logit contributions | |
ble | +0.106 |
bled | +0.098 |
balls | +0.091 |
ster | +0.086 |
fin | +0.085 |
Neg logit contributions | |
ird | -0.086 |
aren | -0.081 |
old | -0.077 |
Karen | -0.077 |
Aud | -0.075 |
Token | and |
Token ID | 290 |
Feature activation | +3.44e-02 |
Pos logit contributions | |
ird | +0.011 |
rell | +0.010 |
Karen | +0.010 |
aren | +0.010 |
Ellen | +0.009 |
Neg logit contributions | |
ble | -0.016 |
bled | -0.015 |
balls | -0.014 |
ster | -0.014 |
fin | -0.014 |
Token | Sam |
Token ID | 3409 |
Feature activation | -4.79e-02 |
Pos logit contributions | |
ble | +0.066 |
bled | +0.059 |
balls | +0.054 |
ster | +0.054 |
fin | +0.052 |
Neg logit contributions | |
ird | -0.071 |
Karen | -0.067 |
rell | -0.067 |
aren | -0.067 |
Ellen | -0.063 |
Token | looked |
Token ID | 3114 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
rell | +0.088 |
Karen | +0.087 |
ird | +0.087 |
aren | +0.086 |
chairs | +0.085 |
Neg logit contributions | |
ble | -0.095 |
bled | -0.088 |
ster | -0.082 |
fin | -0.079 |
balls | -0.078 |
Token | at |
Token ID | 379 |
Feature activation | +8.01e-03 |
Pos logit contributions | |
ird | +0.042 |
Karen | +0.040 |
aren | +0.040 |
rell | +0.039 |
Ell | +0.037 |
Neg logit contributions | |
ble | -0.026 |
bled | -0.024 |
balls | -0.022 |
ster | -0.022 |
fin | -0.021 |
Token | Bob |
Token ID | 5811 |
Feature activation | +1.00e-02 |
Pos logit contributions | |
ble | +0.015 |
bled | +0.014 |
balls | +0.013 |
fin | +0.012 |
ster | +0.012 |
Neg logit contributions | |
ird | -0.017 |
rell | -0.016 |
aren | -0.016 |
Karen | -0.015 |
Ellen | -0.014 |
Token | o |
Token ID | 78 |
Feature activation | +8.19e-03 |
Pos logit contributions | |
ble | +0.027 |
bled | +0.025 |
balls | +0.023 |
ster | +0.023 |
fin | +0.023 |
Neg logit contributions | |
ird | -0.013 |
rell | -0.012 |
Karen | -0.011 |
aren | -0.011 |
hell | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | +1.28e-02 |
Pos logit contributions | |
ble | +0.021 |
bled | +0.020 |
balls | +0.019 |
ster | +0.018 |
fin | +0.018 |
Neg logit contributions | |
ird | -0.011 |
rell | -0.010 |
Karen | -0.009 |
aren | -0.009 |
hell | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | -8.44e-03 |
Pos logit contributions | |
ird | +0.012 |
Karen | +0.011 |
rell | +0.011 |
aren | +0.011 |
Ellen | +0.010 |
Neg logit contributions | |
ble | -0.016 |
bled | -0.015 |
balls | -0.014 |
ster | -0.013 |
fin | -0.013 |
Token | picked |
Token ID | 6497 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
ird | +0.014 |
Karen | +0.013 |
rell | +0.013 |
aren | +0.013 |
Ellen | +0.012 |
Neg logit contributions | |
ble | -0.019 |
bled | -0.018 |
balls | -0.017 |
ster | -0.016 |
fin | -0.016 |
Token | up |
Token ID | 510 |
Feature activation | -8.55e-03 |
Pos logit contributions | |
ird | +0.030 |
rell | +0.027 |
aren | +0.027 |
Karen | +0.027 |
Ellen | +0.025 |
Neg logit contributions | |
ble | -0.030 |
bled | -0.027 |
balls | -0.025 |
ster | -0.024 |
fin | -0.024 |
Token | her |
Token ID | 607 |
Feature activation | +3.46e-02 |
Pos logit contributions | |
ird | +0.018 |
rell | +0.016 |
aren | +0.016 |
Ell | +0.015 |
Karen | +0.015 |
Neg logit contributions | |
ble | -0.016 |
bled | -0.014 |
balls | -0.014 |
ster | -0.013 |
fin | -0.013 |
Token | steak |
Token ID | 26320 |
Feature activation | +3.04e-02 |
Pos logit contributions | |
ble | +0.070 |
bled | +0.065 |
balls | +0.056 |
fin | +0.055 |
ster | +0.055 |
Neg logit contributions | |
ird | -0.061 |
rell | -0.055 |
hell | -0.055 |
Alice | -0.054 |
Stella | -0.054 |
Token | . |
Token ID | 13 |
Feature activation | -4.49e-02 |
Pos logit contributions | |
ble | +0.077 |
bled | +0.072 |
fin | +0.068 |
balls | +0.067 |
ster | +0.067 |
Neg logit contributions | |
ird | -0.039 |
Karen | -0.035 |
rell | -0.034 |
aren | -0.034 |
hell | -0.033 |
Token | She |
Token ID | 1375 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
ird | +0.091 |
hell | +0.086 |
aren | +0.080 |
rell | +0.079 |
irst | +0.079 |
Neg logit contributions | |
ble | -0.074 |
bled | -0.066 |
balls | -0.066 |
next | -0.055 |
diving | -0.052 |
Token | took |
Token ID | 1718 |
Feature activation | -4.04e-02 |
Pos logit contributions | |
ble | +0.039 |
bled | +0.035 |
balls | +0.033 |
ster | +0.033 |
fin | +0.032 |
Neg logit contributions | |
ird | -0.029 |
Karen | -0.026 |
aren | -0.026 |
rell | -0.025 |
Stella | -0.024 |
Token | it |
Token ID | 340 |
Feature activation | -2.17e-03 |
Pos logit contributions | |
ird | +0.098 |
aren | +0.096 |
rell | +0.093 |
Karen | +0.090 |
Ell | +0.087 |
Neg logit contributions | |
ble | -0.062 |
bled | -0.057 |
balls | -0.050 |
ster | -0.045 |
fin | -0.045 |
Token | to |
Token ID | 284 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
ird | +0.004 |
aren | +0.004 |
rell | +0.004 |
Karen | +0.004 |
Ellen | +0.004 |
Neg logit contributions | |
ble | -0.004 |
bled | -0.004 |
balls | -0.004 |
ster | -0.004 |
fin | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | +3.32e-03 |
Pos logit contributions | |
ird | +0.029 |
rell | +0.028 |
aren | +0.027 |
Karen | +0.026 |
hell | +0.025 |
Neg logit contributions | |
ble | -0.036 |
bled | -0.032 |
balls | -0.030 |
fin | -0.030 |
ster | -0.030 |
Token | the |
Token ID | 262 |
Feature activation | -3.11e-02 |
Pos logit contributions | |
ird | +0.050 |
rell | +0.048 |
aren | +0.045 |
hell | +0.045 |
chairs | +0.045 |
Neg logit contributions | |
ble | -0.035 |
bled | -0.030 |
balls | -0.028 |
fin | -0.027 |
diving | -0.026 |
Token | vendor |
Token ID | 18371 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
ird | +0.053 |
rell | +0.049 |
Karen | +0.048 |
aren | +0.048 |
Ellen | +0.044 |
Neg logit contributions | |
ble | -0.071 |
bled | -0.065 |
balls | -0.061 |
ster | -0.060 |
fin | -0.059 |
Token | and |
Token ID | 290 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
ird | +0.027 |
rell | +0.025 |
Karen | +0.025 |
aren | +0.025 |
Ellen | +0.023 |
Neg logit contributions | |
ble | -0.036 |
bled | -0.032 |
balls | -0.031 |
ster | -0.031 |
fin | -0.030 |
Token | asked |
Token ID | 1965 |
Feature activation | -2.90e-03 |
Pos logit contributions | |
ird | +0.031 |
rell | +0.029 |
aren | +0.028 |
Karen | +0.028 |
hell | +0.027 |
Neg logit contributions | |
ble | -0.039 |
bled | -0.036 |
balls | -0.034 |
ster | -0.033 |
fin | -0.033 |
Token | for |
Token ID | 329 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
ird | +0.006 |
rell | +0.005 |
aren | +0.005 |
Karen | +0.005 |
Ellen | +0.005 |
Neg logit contributions | |
ble | -0.006 |
bled | -0.005 |
balls | -0.005 |
ster | -0.005 |
fin | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | -4.37e-02 |
Pos logit contributions | |
ird | +0.021 |
rell | +0.020 |
aren | +0.020 |
Karen | +0.019 |
Ellen | +0.018 |
Neg logit contributions | |
ble | -0.020 |
bled | -0.019 |
balls | -0.018 |
fin | -0.017 |
ster | -0.016 |
Token | red |
Token ID | 2266 |
Feature activation | -2.79e-02 |
Pos logit contributions | |
ird | +0.081 |
rell | +0.080 |
Karen | +0.075 |
aren | +0.075 |
chairs | +0.074 |
Neg logit contributions | |
ble | -0.087 |
balls | -0.077 |
bled | -0.075 |
ster | -0.075 |
fin | -0.074 |
Token | balloon |
Token ID | 21190 |
Feature activation | +3.83e-03 |
Pos logit contributions | |
ird | +0.053 |
rell | +0.049 |
Karen | +0.048 |
aren | +0.048 |
Ellen | +0.046 |
Neg logit contributions | |
ble | -0.058 |
bled | -0.051 |
balls | -0.049 |
ster | -0.048 |
fin | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
ble | +0.009 |
bled | +0.008 |
balls | +0.008 |
ster | +0.008 |
fin | +0.008 |
Neg logit contributions | |
ird | -0.006 |
Karen | -0.005 |
rell | -0.005 |
aren | -0.005 |
Ellen | -0.005 |
Token | The |
Token ID | 383 |
Feature activation | -2.76e-02 |
Pos logit contributions | |
ble | +0.031 |
bled | +0.028 |
ster | +0.026 |
balls | +0.026 |
fin | +0.025 |
Neg logit contributions | |
ird | -0.016 |
Karen | -0.015 |
aren | -0.015 |
rell | -0.015 |
Stella | -0.014 |
Token | vendor |
Token ID | 18371 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
ird | +0.054 |
rell | +0.050 |
Karen | +0.046 |
aren | +0.046 |
Ell | +0.045 |
Neg logit contributions | |
ble | -0.057 |
bled | -0.051 |
balls | -0.049 |
ster | -0.048 |
shark | -0.046 |
Token | that |
Token ID | 326 |
Feature activation | +2.46e-03 |
Pos logit contributions | |
ird | +0.003 |
rell | +0.003 |
aren | +0.003 |
Karen | +0.003 |
Ellen | +0.003 |
Neg logit contributions | |
ble | -0.003 |
bled | -0.003 |
balls | -0.003 |
ster | -0.003 |
fin | -0.003 |
Token | was |
Token ID | 373 |
Feature activation | -6.54e-03 |
Pos logit contributions | |
ble | +0.006 |
bled | +0.005 |
balls | +0.005 |
ster | +0.005 |
fin | +0.005 |
Neg logit contributions | |
ird | -0.004 |
rell | -0.004 |
aren | -0.004 |
Karen | -0.004 |
Ellen | -0.003 |
Token | filled |
Token ID | 5901 |
Feature activation | +9.34e-03 |
Pos logit contributions | |
ird | +0.011 |
Karen | +0.010 |
rell | +0.010 |
aren | +0.010 |
Ellen | +0.010 |
Neg logit contributions | |
ble | -0.015 |
bled | -0.013 |
balls | -0.013 |
ster | -0.012 |
fin | -0.012 |
Token | with |
Token ID | 351 |
Feature activation | -2.64e-02 |
Pos logit contributions | |
ble | +0.018 |
bled | +0.017 |
balls | +0.016 |
ster | +0.016 |
fin | +0.016 |
Neg logit contributions | |
ird | -0.017 |
Karen | -0.016 |
Stella | -0.016 |
aren | -0.016 |
hell | -0.015 |
Token | toys |
Token ID | 14958 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
ird | +0.055 |
aren | +0.053 |
rell | +0.050 |
Karen | +0.049 |
Alice | +0.047 |
Neg logit contributions | |
ble | -0.051 |
bled | -0.044 |
fin | -0.043 |
balls | -0.042 |
ster | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | -6.74e-02 |
Pos logit contributions | |
ird | +0.018 |
aren | +0.017 |
Karen | +0.017 |
rell | +0.016 |
Ellen | +0.015 |
Neg logit contributions | |
ble | -0.023 |
bled | -0.020 |
balls | -0.020 |
ster | -0.019 |
fin | -0.018 |
Token | As |
Token ID | 1081 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
ird | +0.141 |
ives | +0.130 |
hell | +0.130 |
ited | +0.126 |
ha | +0.124 |
Neg logit contributions | |
ble | -0.084 |
balls | -0.077 |
next | -0.073 |
diving | -0.067 |
Finn | -0.067 |
Token | Lucy |
Token ID | 22162 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
ird | +0.030 |
rell | +0.029 |
ives | +0.028 |
hell | +0.027 |
aren | +0.027 |
Neg logit contributions | |
ble | -0.029 |
bled | -0.026 |
balls | -0.025 |
diving | -0.024 |
shark | -0.023 |
Token | walked |
Token ID | 6807 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
ble | +0.025 |
bled | +0.022 |
balls | +0.021 |
ster | +0.020 |
fin | +0.020 |
Neg logit contributions | |
ird | -0.022 |
rell | -0.021 |
Karen | -0.021 |
aren | -0.021 |
Ellen | -0.019 |
Token | through |
Token ID | 832 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
ird | +0.023 |
rell | +0.021 |
Karen | +0.021 |
aren | +0.021 |
Ellen | +0.019 |
Neg logit contributions | |
ble | -0.030 |
bled | -0.027 |
balls | -0.025 |
ster | -0.025 |
fin | -0.025 |
Token | the |
Token ID | 262 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
ird | +0.034 |
rell | +0.032 |
aren | +0.031 |
ives | +0.030 |
hell | +0.030 |
Neg logit contributions | |
ble | -0.022 |
bled | -0.019 |
diving | -0.018 |
balls | -0.017 |
ster | -0.017 |
Token | wanted |
Token ID | 2227 |
Feature activation | -3.38e-02 |
Pos logit contributions | |
ble | +0.014 |
bled | +0.013 |
balls | +0.012 |
ster | +0.012 |
fin | +0.012 |
Neg logit contributions | |
ird | -0.010 |
Karen | -0.009 |
aren | -0.009 |
rell | -0.009 |
Ellen | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | -3.12e-02 |
Pos logit contributions | |
ird | +0.086 |
rell | +0.079 |
aren | +0.077 |
Karen | +0.075 |
Ellen | +0.074 |
Neg logit contributions | |
ble | -0.050 |
bled | -0.042 |
balls | -0.041 |
ster | -0.038 |
diving | -0.037 |
Token | get |
Token ID | 651 |
Feature activation | -2.17e-02 |
Pos logit contributions | |
ird | +0.068 |
aren | +0.065 |
rell | +0.062 |
chairs | +0.062 |
Karen | +0.061 |
Neg logit contributions | |
ble | -0.056 |
bled | -0.043 |
balls | -0.041 |
fin | -0.041 |
ster | -0.040 |
Token | some |
Token ID | 617 |
Feature activation | -6.83e-02 |
Pos logit contributions | |
ird | +0.044 |
rell | +0.040 |
aren | +0.040 |
Karen | +0.039 |
Ellen | +0.037 |
Neg logit contributions | |
ble | -0.043 |
bled | -0.039 |
balls | -0.036 |
ster | -0.035 |
fin | -0.035 |
Token | of |
Token ID | 286 |
Feature activation | -2.95e-02 |
Pos logit contributions | |
Ellen | +0.129 |
ird | +0.125 |
Ell | +0.124 |
Aud | +0.122 |
Kathy | +0.120 |
Neg logit contributions | |
ble | -0.135 |
bled | -0.106 |
diving | -0.103 |
balls | -0.103 |
next | -0.103 |
Token | the |
Token ID | 262 |
Feature activation | -5.96e-02 |
Pos logit contributions | |
ird | +0.072 |
aren | +0.063 |
rell | +0.063 |
chair | +0.061 |
chairs | +0.061 |
Neg logit contributions | |
ble | -0.049 |
bled | -0.041 |
balls | -0.037 |
diving | -0.036 |
fin | -0.036 |
Token | cher |
Token ID | 21382 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
ird | +0.117 |
Aud | +0.109 |
Ellen | +0.107 |
old | +0.105 |
jo | +0.101 |
Neg logit contributions | |
ble | -0.118 |
next | -0.096 |
bled | -0.093 |
balls | -0.091 |
ster | -0.084 |
Token | ries |
Token ID | 1678 |
Feature activation | -2.83e-02 |
Pos logit contributions | |
ird | +0.037 |
Karen | +0.035 |
Ellen | +0.034 |
rell | +0.034 |
aren | +0.034 |
Neg logit contributions | |
ble | -0.036 |
bled | -0.031 |
ster | -0.028 |
balls | -0.028 |
diving | -0.025 |
Token | but |
Token ID | 475 |
Feature activation | -2.24e-02 |
Pos logit contributions | |
ird | +0.054 |
aren | +0.051 |
rell | +0.050 |
Karen | +0.050 |
Ellen | +0.048 |
Neg logit contributions | |
ble | -0.060 |
bled | -0.052 |
balls | -0.049 |
ster | -0.048 |
fin | -0.046 |
Token | she |
Token ID | 673 |
Feature activation | +9.09e-03 |
Pos logit contributions | |
ird | +0.040 |
rell | +0.037 |
aren | +0.036 |
Karen | +0.036 |
Ellen | +0.033 |
Neg logit contributions | |
ble | -0.050 |
bled | -0.045 |
balls | -0.043 |
ster | -0.042 |
fin | -0.041 |
Token | couldn |
Token ID | 3521 |
Feature activation | -3.68e-02 |
Pos logit contributions | |
ble | +0.019 |
bled | +0.017 |
balls | +0.016 |
fin | +0.016 |
ster | +0.016 |
Neg logit contributions | |
ird | -0.017 |
Karen | -0.016 |
aren | -0.016 |
rell | -0.015 |
Stella | -0.014 |