Token | added |
Token ID | 2087 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
penn | +0.011 |
imet | +0.011 |
pen | +0.011 |
ian | +0.011 |
Ow | +0.010 |
Neg logit contributions | |
late | -0.010 |
quiet | -0.009 |
seek | -0.009 |
Lisa | -0.009 |
Lisa | -0.008 |
Token | shells |
Token ID | 19679 |
Feature activation | +7.26e-01 |
Pos logit contributions | |
penn | +0.004 |
imet | +0.004 |
Ow | +0.004 |
pen | +0.003 |
Pe | +0.003 |
Neg logit contributions | |
late | -0.005 |
seek | -0.005 |
Lisa | -0.004 |
quiet | -0.004 |
Lisa | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | +3.33e-01 |
Pos logit contributions | |
penn | +0.022 |
imet | +0.021 |
pen | +0.020 |
ian | +0.019 |
ome | +0.019 |
Neg logit contributions | |
late | -0.024 |
seek | -0.022 |
quiet | -0.021 |
Lisa | -0.021 |
Lisa | -0.019 |
Token | stones |
Token ID | 14966 |
Feature activation | +8.58e-01 |
Pos logit contributions | |
penn | +0.008 |
imet | +0.007 |
pen | +0.007 |
ian | +0.006 |
trek | +0.006 |
Neg logit contributions | |
late | -0.013 |
quiet | -0.012 |
seek | -0.012 |
Lisa | -0.011 |
Lisa | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | +5.40e-01 |
Pos logit contributions | |
penn | +0.025 |
imet | +0.025 |
pen | +0.024 |
ome | +0.022 |
ian | +0.022 |
Neg logit contributions | |
late | -0.028 |
seek | -0.026 |
Lisa | -0.025 |
quiet | -0.024 |
Lisa | -0.023 |
Token | seaw |
Token ID | 32085 |
Feature activation | +3.13e+00 |
Pos logit contributions | |
penn | +0.013 |
imet | +0.013 |
pen | +0.012 |
ian | +0.012 |
trek | +0.011 |
Neg logit contributions | |
late | -0.020 |
seek | -0.018 |
quiet | -0.018 |
Lisa | -0.017 |
Lisa | -0.017 |
Token | eed |
Token ID | 2308 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
pen | +0.069 |
penn | +0.069 |
imet | +0.069 |
ian | +0.068 |
anybody | +0.067 |
Neg logit contributions | |
late | -0.110 |
quiet | -0.105 |
seek | -0.103 |
too | -0.099 |
Lisa | -0.098 |
Token | to |
Token ID | 284 |
Feature activation | -5.33e-01 |
Pos logit contributions | |
penn | +0.041 |
imet | +0.039 |
pen | +0.038 |
ian | +0.036 |
ome | +0.036 |
Neg logit contributions | |
late | -0.045 |
seek | -0.041 |
Lisa | -0.039 |
quiet | -0.039 |
Lisa | -0.037 |
Token | make |
Token ID | 787 |
Feature activation | -8.64e-01 |
Pos logit contributions | |
late | +0.020 |
seek | +0.017 |
Happy | +0.016 |
quiet | +0.016 |
lic | +0.016 |
Neg logit contributions | |
penn | -0.013 |
pen | -0.012 |
imet | -0.012 |
bank | -0.012 |
trek | -0.011 |
Token | it |
Token ID | 340 |
Feature activation | +3.43e-02 |
Pos logit contributions | |
late | +0.026 |
seek | +0.024 |
quiet | +0.022 |
Lisa | +0.021 |
Lisa | +0.020 |
Neg logit contributions | |
penn | -0.027 |
pen | -0.026 |
imet | -0.026 |
Ow | -0.025 |
Pe | -0.024 |
Token | look |
Token ID | 804 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
penn | +0.001 |
imet | +0.001 |
bank | +0.001 |
rest | +0.001 |
pen | +0.001 |
Neg logit contributions | |
late | -0.001 |
Lisa | -0.001 |
seek | -0.001 |
Sarah | -0.001 |
Lisa | -0.001 |
Token | ugg |
Token ID | 6837 |
Feature activation | -4.68e-01 |
Pos logit contributions | |
penn | +0.001 |
imet | +0.001 |
ian | +0.001 |
pen | +0.001 |
trigger | +0.000 |
Neg logit contributions | |
late | -0.002 |
seek | -0.002 |
quiet | -0.002 |
too | -0.002 |
together | -0.002 |
Token | led |
Token ID | 992 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
late | +0.020 |
seek | +0.019 |
quiet | +0.019 |
Lisa | +0.018 |
Lisa | +0.018 |
Neg logit contributions | |
penn | -0.009 |
ian | -0.009 |
pen | -0.008 |
imet | -0.008 |
ome | -0.008 |
Token | up |
Token ID | 510 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
penn | +0.044 |
imet | +0.043 |
pen | +0.043 |
ian | +0.042 |
Ow | +0.041 |
Neg logit contributions | |
late | -0.046 |
seek | -0.041 |
quiet | -0.040 |
Lisa | -0.039 |
Lisa | -0.038 |
Token | in |
Token ID | 287 |
Feature activation | +3.15e-01 |
Pos logit contributions | |
penn | +0.047 |
imet | +0.046 |
pen | +0.046 |
brow | +0.044 |
ian | +0.043 |
Neg logit contributions | |
late | -0.042 |
Lisa | -0.038 |
Lisa | -0.038 |
seek | -0.038 |
quiet | -0.037 |
Token | some |
Token ID | 617 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
penn | +0.009 |
pen | +0.008 |
brow | +0.008 |
Pe | +0.008 |
trek | +0.008 |
Neg logit contributions | |
lic | -0.010 |
late | -0.010 |
Lisa | -0.010 |
aky | -0.009 |
seek | -0.009 |
Token | seaw |
Token ID | 32085 |
Feature activation | +3.07e+00 |
Pos logit contributions | |
penn | +0.000 |
pen | +0.000 |
imet | +0.000 |
ian | +0.000 |
brow | +0.000 |
Neg logit contributions | |
late | -0.000 |
seek | -0.000 |
Lisa | -0.000 |
quiet | -0.000 |
Lisa | -0.000 |
Token | eed |
Token ID | 2308 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
penn | +0.072 |
anybody | +0.071 |
imet | +0.070 |
Sy | +0.070 |
Ow | +0.068 |
Neg logit contributions | |
late | -0.107 |
seek | -0.100 |
quiet | -0.100 |
aky | -0.094 |
too | -0.093 |
Token | and |
Token ID | 290 |
Feature activation | +5.58e-01 |
Pos logit contributions | |
penn | +0.038 |
pen | +0.037 |
imet | +0.035 |
ian | +0.035 |
ome | +0.035 |
Neg logit contributions | |
late | -0.034 |
Lisa | -0.031 |
seek | -0.030 |
quiet | -0.029 |
Lisa | -0.029 |
Token | felt |
Token ID | 2936 |
Feature activation | +9.17e-01 |
Pos logit contributions | |
penn | +0.014 |
imet | +0.013 |
pen | +0.013 |
ian | +0.013 |
brow | +0.013 |
Neg logit contributions | |
late | -0.020 |
quiet | -0.017 |
seek | -0.017 |
Lisa | -0.017 |
Lisa | -0.017 |
Token | very |
Token ID | 845 |
Feature activation | -7.45e-01 |
Pos logit contributions | |
penn | +0.036 |
pen | +0.035 |
imet | +0.035 |
Pe | +0.034 |
ian | +0.034 |
Neg logit contributions | |
late | -0.019 |
seek | -0.017 |
Lisa | -0.016 |
Lisa | -0.016 |
quiet | -0.015 |
Token | cozy |
Token ID | 37438 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
late | +0.013 |
Lisa | +0.011 |
seek | +0.011 |
Lisa | +0.011 |
quiet | +0.010 |
Neg logit contributions | |
penn | -0.031 |
pen | -0.030 |
imet | -0.030 |
brow | -0.029 |
Mont | -0.029 |
Token | like |
Token ID | 588 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
late | +0.002 |
Lisa | +0.002 |
Happy | +0.002 |
seek | +0.002 |
lic | +0.002 |
Neg logit contributions | |
penn | -0.003 |
Pe | -0.002 |
pen | -0.002 |
knees | -0.002 |
bank | -0.002 |
Token | fish |
Token ID | 5916 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
late | +0.006 |
seek | +0.005 |
quiet | +0.005 |
Lisa | +0.005 |
Lisa | +0.005 |
Neg logit contributions | |
penn | -0.004 |
imet | -0.004 |
pen | -0.004 |
ian | -0.004 |
Ow | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
penn | +0.000 |
pen | +0.000 |
Pe | +0.000 |
bank | +0.000 |
ome | +0.000 |
Neg logit contributions | |
late | -0.001 |
Lisa | -0.001 |
Sarah | -0.001 |
Lisa | -0.000 |
Happy | -0.000 |
Token | shells |
Token ID | 19679 |
Feature activation | +4.61e-01 |
Pos logit contributions | |
penn | +0.003 |
imet | +0.002 |
pen | +0.002 |
Ow | +0.002 |
Pe | +0.002 |
Neg logit contributions | |
late | -0.004 |
quiet | -0.004 |
seek | -0.004 |
Lisa | -0.004 |
Lisa | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
penn | +0.013 |
pen | +0.011 |
ome | +0.010 |
imet | +0.010 |
bank | +0.010 |
Neg logit contributions | |
late | -0.017 |
Lisa | -0.016 |
seek | -0.016 |
Lisa | -0.015 |
Sarah | -0.015 |
Token | seaw |
Token ID | 32085 |
Feature activation | +3.03e+00 |
Pos logit contributions | |
penn | +0.004 |
imet | +0.004 |
pen | +0.004 |
Ow | +0.004 |
Pe | +0.004 |
Neg logit contributions | |
late | -0.007 |
seek | -0.006 |
quiet | -0.006 |
Lisa | -0.006 |
Lisa | -0.006 |
Token | eed |
Token ID | 2308 |
Feature activation | +9.86e-01 |
Pos logit contributions | |
imet | +0.069 |
ian | +0.068 |
penn | +0.067 |
pen | +0.067 |
Higher | +0.067 |
Neg logit contributions | |
late | -0.105 |
quiet | -0.101 |
seek | -0.100 |
too | -0.095 |
lic | -0.092 |
Token | . |
Token ID | 13 |
Feature activation | -4.84e-01 |
Pos logit contributions | |
penn | +0.027 |
pen | +0.024 |
imet | +0.023 |
ome | +0.023 |
bank | +0.023 |
Neg logit contributions | |
late | -0.036 |
Lisa | -0.033 |
seek | -0.033 |
Sarah | -0.032 |
Lisa | -0.031 |
Token | But |
Token ID | 887 |
Feature activation | +3.69e-01 |
Pos logit contributions | |
late | +0.013 |
quiet | +0.011 |
seek | +0.011 |
Lisa | +0.010 |
Lisa | +0.009 |
Neg logit contributions | |
penn | -0.017 |
imet | -0.017 |
pen | -0.017 |
ian | -0.017 |
Ow | -0.016 |
Token | they |
Token ID | 484 |
Feature activation | -9.20e-01 |
Pos logit contributions | |
penn | +0.012 |
Pe | +0.011 |
imet | +0.011 |
Ow | +0.010 |
brow | +0.010 |
Neg logit contributions | |
late | -0.011 |
lic | -0.010 |
seek | -0.010 |
quiet | -0.010 |
Lisa | -0.009 |
Token | did |
Token ID | 750 |
Feature activation | -6.34e-01 |
Pos logit contributions | |
late | +0.029 |
seek | +0.025 |
quiet | +0.024 |
Lisa | +0.024 |
Lisa | +0.023 |
Neg logit contributions | |
penn | -0.028 |
pen | -0.027 |
imet | -0.026 |
ian | -0.026 |
Ow | -0.026 |
Token | like |
Token ID | 588 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
late | +0.004 |
Lisa | +0.003 |
Happy | +0.003 |
seek | +0.003 |
lic | +0.003 |
Neg logit contributions | |
penn | -0.004 |
Pe | -0.003 |
bank | -0.003 |
pen | -0.003 |
knees | -0.003 |
Token | fish |
Token ID | 5916 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
late | +0.007 |
seek | +0.007 |
quiet | +0.006 |
Lisa | +0.006 |
Lisa | +0.006 |
Neg logit contributions | |
penn | -0.005 |
imet | -0.005 |
pen | -0.005 |
Ow | -0.005 |
Pe | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
late | +0.005 |
Lisa | +0.004 |
Sarah | +0.004 |
Lisa | +0.004 |
Happy | +0.004 |
Neg logit contributions | |
penn | -0.003 |
pen | -0.002 |
Pe | -0.002 |
bank | -0.002 |
ome | -0.002 |
Token | shells |
Token ID | 19679 |
Feature activation | +3.68e-01 |
Pos logit contributions | |
penn | +0.000 |
imet | +0.000 |
Ow | +0.000 |
Pe | +0.000 |
pen | +0.000 |
Neg logit contributions | |
late | -0.000 |
quiet | -0.000 |
seek | -0.000 |
Lisa | -0.000 |
Lisa | -0.000 |
Token | and |
Token ID | 290 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
penn | +0.010 |
pen | +0.009 |
ome | +0.008 |
bank | +0.008 |
imet | +0.008 |
Neg logit contributions | |
late | -0.014 |
Lisa | -0.013 |
seek | -0.012 |
Lisa | -0.012 |
Sarah | -0.012 |
Token | seaw |
Token ID | 32085 |
Feature activation | +2.90e+00 |
Pos logit contributions | |
penn | +0.003 |
imet | +0.002 |
pen | +0.002 |
Ow | +0.002 |
Pe | +0.002 |
Neg logit contributions | |
late | -0.004 |
seek | -0.004 |
quiet | -0.004 |
Lisa | -0.004 |
Lisa | -0.004 |
Token | eed |
Token ID | 2308 |
Feature activation | +8.52e-01 |
Pos logit contributions | |
imet | +0.067 |
ian | +0.066 |
Higher | +0.066 |
pen | +0.066 |
penn | +0.065 |
Neg logit contributions | |
late | -0.099 |
quiet | -0.095 |
seek | -0.094 |
too | -0.089 |
lic | -0.087 |
Token | . |
Token ID | 13 |
Feature activation | -7.48e-01 |
Pos logit contributions | |
penn | +0.024 |
pen | +0.020 |
imet | +0.020 |
ome | +0.020 |
bank | +0.020 |
Neg logit contributions | |
late | -0.032 |
Lisa | -0.030 |
seek | -0.028 |
Sarah | -0.028 |
Lisa | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
late | +0.023 |
seek | +0.022 |
quiet | +0.021 |
Happy | +0.019 |
aky | +0.019 |
Neg logit contributions | |
imet | -0.022 |
Ow | -0.022 |
penn | -0.021 |
Pe | -0.021 |
pen | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
late | +0.007 |
seek | +0.007 |
quiet | +0.007 |
pans | +0.006 |
Lisa | +0.006 |
Neg logit contributions | |
Ow | -0.006 |
pen | -0.006 |
Pe | -0.006 |
imet | -0.006 |
penn | -0.006 |
Token | One |
Token ID | 3198 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
late | +0.008 |
Lisa | +0.007 |
pleased | +0.007 |
too | +0.007 |
quiet | +0.007 |
Neg logit contributions | |
Twe | -0.005 |
Whoever | -0.005 |
pen | -0.005 |
penn | -0.005 |
trek | -0.005 |
Token | of |
Token ID | 286 |
Feature activation | -6.96e-01 |
Pos logit contributions | |
late | +0.011 |
seek | +0.010 |
quiet | +0.010 |
Lisa | +0.009 |
Lisa | +0.009 |
Neg logit contributions | |
penn | -0.016 |
pen | -0.015 |
imet | -0.015 |
ome | -0.014 |
ian | -0.014 |
Token | shells |
Token ID | 19679 |
Feature activation | +1.98e-02 |
Pos logit contributions | |
late | +0.018 |
seek | +0.016 |
quiet | +0.015 |
Lisa | +0.015 |
Lisa | +0.014 |
Neg logit contributions | |
penn | -0.025 |
pen | -0.024 |
imet | -0.024 |
ian | -0.023 |
Ow | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
penn | +0.001 |
pen | +0.001 |
imet | +0.001 |
ome | +0.001 |
ian | +0.001 |
Neg logit contributions | |
late | -0.001 |
Lisa | -0.001 |
quiet | -0.001 |
seek | -0.001 |
Lisa | -0.001 |
Token | rocks |
Token ID | 12586 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
late | +0.007 |
Lisa | +0.006 |
seek | +0.006 |
quiet | +0.006 |
Lisa | +0.006 |
Neg logit contributions | |
penn | -0.005 |
pen | -0.004 |
imet | -0.004 |
Pe | -0.004 |
ian | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | -3.17e-02 |
Pos logit contributions | |
penn | +0.005 |
pen | +0.005 |
imet | +0.004 |
ian | +0.004 |
ome | +0.004 |
Neg logit contributions | |
late | -0.005 |
seek | -0.004 |
quiet | -0.004 |
Lisa | -0.004 |
Lisa | -0.004 |
Token | seaw |
Token ID | 32085 |
Feature activation | +2.84e+00 |
Pos logit contributions | |
late | +0.001 |
quiet | +0.001 |
seek | +0.001 |
Lisa | +0.001 |
Lisa | +0.001 |
Neg logit contributions | |
penn | -0.001 |
imet | -0.001 |
pen | -0.001 |
Pe | -0.001 |
ian | -0.001 |
Token | eed |
Token ID | 2308 |
Feature activation | +5.98e-01 |
Pos logit contributions | |
penn | +0.063 |
Ow | +0.063 |
imet | +0.063 |
anybody | +0.062 |
pen | +0.062 |
Neg logit contributions | |
late | -0.096 |
seek | -0.093 |
quiet | -0.092 |
lic | -0.087 |
too | -0.086 |
Token | , |
Token ID | 11 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
penn | +0.018 |
pen | +0.017 |
imet | +0.016 |
ian | +0.016 |
ome | +0.016 |
Neg logit contributions | |
late | -0.021 |
Lisa | -0.018 |
seek | -0.018 |
quiet | -0.018 |
Lisa | -0.017 |
Token | but |
Token ID | 475 |
Feature activation | +5.48e-01 |
Pos logit contributions | |
late | +0.004 |
seek | +0.003 |
Lisa | +0.003 |
Happy | +0.003 |
quiet | +0.003 |
Neg logit contributions | |
penn | -0.004 |
imet | -0.004 |
pen | -0.004 |
Pe | -0.004 |
trek | -0.004 |
Token | Jane |
Token ID | 12091 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
penn | +0.014 |
anybody | +0.013 |
imet | +0.013 |
Pe | +0.013 |
brow | +0.012 |
Neg logit contributions | |
late | -0.020 |
lic | -0.017 |
seek | -0.017 |
aky | -0.017 |
quiet | -0.017 |
Token | was |
Token ID | 373 |
Feature activation | -4.11e-01 |
Pos logit contributions | |
late | +0.005 |
seek | +0.004 |
quiet | +0.004 |
Lisa | +0.004 |
Lisa | +0.004 |
Neg logit contributions | |
penn | -0.003 |
pen | -0.003 |
imet | -0.003 |
ian | -0.003 |
brow | -0.003 |
Token | shore |
Token ID | 15191 |
Feature activation | +5.20e-01 |
Pos logit contributions | |
late | +0.006 |
Lisa | +0.006 |
seek | +0.006 |
Lisa | +0.006 |
quiet | +0.006 |
Neg logit contributions | |
penn | -0.003 |
imet | -0.003 |
pen | -0.003 |
ian | -0.003 |
bank | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
pen | +0.014 |
penn | +0.014 |
bank | +0.013 |
imet | +0.013 |
ome | +0.013 |
Neg logit contributions | |
late | -0.018 |
Lisa | -0.016 |
seek | -0.016 |
Happy | -0.015 |
quiet | -0.015 |
Token | collecting |
Token ID | 13157 |
Feature activation | -4.67e-01 |
Pos logit contributions | |
late | +0.005 |
seek | +0.004 |
Lisa | +0.004 |
Happy | +0.004 |
lic | +0.004 |
Neg logit contributions | |
penn | -0.005 |
imet | -0.005 |
pen | -0.005 |
brow | -0.004 |
Pe | -0.004 |
Token | shells |
Token ID | 19679 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
late | +0.017 |
seek | +0.015 |
quiet | +0.015 |
Lisa | +0.014 |
Lisa | +0.014 |
Neg logit contributions | |
penn | -0.012 |
imet | -0.012 |
pen | -0.012 |
ian | -0.011 |
Ow | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
penn | +0.004 |
pen | +0.004 |
imet | +0.003 |
ome | +0.003 |
ian | +0.003 |
Neg logit contributions | |
late | -0.004 |
Lisa | -0.003 |
seek | -0.003 |
quiet | -0.003 |
Lisa | -0.003 |
Token | seaw |
Token ID | 32085 |
Feature activation | +2.83e+00 |
Pos logit contributions | |
penn | +0.001 |
imet | +0.001 |
pen | +0.001 |
trek | +0.001 |
Pe | +0.001 |
Neg logit contributions | |
late | -0.001 |
quiet | -0.001 |
seek | -0.001 |
Lisa | -0.001 |
Lisa | -0.001 |
Token | eed |
Token ID | 2308 |
Feature activation | +9.00e-01 |
Pos logit contributions | |
penn | +0.062 |
imet | +0.061 |
pen | +0.060 |
ian | +0.060 |
Ow | +0.060 |
Neg logit contributions | |
late | -0.102 |
quiet | -0.096 |
seek | -0.096 |
too | -0.090 |
lic | -0.089 |
Token | . |
Token ID | 13 |
Feature activation | -7.93e-03 |
Pos logit contributions | |
penn | +0.027 |
pen | +0.026 |
imet | +0.026 |
ome | +0.025 |
ian | +0.025 |
Neg logit contributions | |
late | -0.030 |
Lisa | -0.027 |
seek | -0.027 |
Lisa | -0.026 |
quiet | -0.026 |
Token | Jimmy |
Token ID | 12963 |
Feature activation | -4.70e-03 |
Pos logit contributions | |
late | +0.000 |
quiet | +0.000 |
Lisa | +0.000 |
seek | +0.000 |
too | +0.000 |
Neg logit contributions | |
penn | -0.000 |
pen | -0.000 |
ian | -0.000 |
imet | -0.000 |
trigger | -0.000 |
Token | took |
Token ID | 1718 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
late | +0.000 |
seek | +0.000 |
quiet | +0.000 |
Lisa | +0.000 |
Lisa | +0.000 |
Neg logit contributions | |
pen | -0.000 |
penn | -0.000 |
ian | -0.000 |
imet | -0.000 |
brow | -0.000 |
Token | time |
Token ID | 640 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
penn | +0.005 |
imet | +0.005 |
pen | +0.005 |
ian | +0.005 |
Ow | +0.005 |
Neg logit contributions | |
late | -0.004 |
seek | -0.004 |
quiet | -0.004 |
Lisa | -0.003 |
Lisa | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | +3.20e-01 |
Pos logit contributions | |
penn | +0.022 |
imet | +0.021 |
pen | +0.021 |
ian | +0.021 |
brow | +0.020 |
Neg logit contributions | |
late | -0.019 |
seek | -0.016 |
quiet | -0.016 |
Lisa | -0.015 |
Lisa | -0.015 |
Token | different |
Token ID | 1180 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
penn | +0.008 |
pen | +0.008 |
ian | +0.007 |
imet | +0.007 |
trigger | +0.007 |
Neg logit contributions | |
late | -0.011 |
Lisa | -0.011 |
Lisa | -0.010 |
seek | -0.010 |
quiet | -0.010 |
Token | fish |
Token ID | 5916 |
Feature activation | +3.87e-01 |
Pos logit contributions | |
penn | +0.009 |
pen | +0.008 |
imet | +0.008 |
ian | +0.008 |
vered | +0.008 |
Neg logit contributions | |
late | -0.009 |
seek | -0.008 |
Lisa | -0.008 |
Lisa | -0.008 |
quiet | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
pen | +0.011 |
penn | +0.011 |
imet | +0.010 |
ome | +0.010 |
vered | +0.010 |
Neg logit contributions | |
late | -0.013 |
Lisa | -0.011 |
seek | -0.011 |
quiet | -0.011 |
Lisa | -0.011 |
Token | colorful |
Token ID | 20239 |
Feature activation | +4.01e-01 |
Pos logit contributions | |
late | +0.006 |
aky | +0.005 |
Lisa | +0.005 |
quiet | +0.004 |
lic | +0.004 |
Neg logit contributions | |
penn | -0.005 |
imet | -0.004 |
knees | -0.004 |
trek | -0.004 |
pen | -0.004 |
Token | seaw |
Token ID | 32085 |
Feature activation | +2.83e+00 |
Pos logit contributions | |
penn | +0.011 |
pen | +0.011 |
imet | +0.011 |
ian | +0.011 |
Ow | +0.010 |
Neg logit contributions | |
late | -0.014 |
seek | -0.012 |
quiet | -0.012 |
Lisa | -0.012 |
Lisa | -0.011 |
Token | eed |
Token ID | 2308 |
Feature activation | +9.43e-01 |
Pos logit contributions | |
penn | +0.067 |
imet | +0.064 |
Ow | +0.064 |
Sy | +0.064 |
Pe | +0.063 |
Neg logit contributions | |
late | -0.096 |
seek | -0.092 |
quiet | -0.090 |
lic | -0.085 |
too | -0.084 |
Token | below |
Token ID | 2174 |
Feature activation | +4.30e-01 |
Pos logit contributions | |
penn | +0.029 |
pen | +0.027 |
imet | +0.027 |
ian | +0.026 |
ome | +0.026 |
Neg logit contributions | |
late | -0.031 |
Lisa | -0.027 |
seek | -0.027 |
quiet | -0.026 |
Lisa | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -4.60e-01 |
Pos logit contributions | |
pen | +0.013 |
penn | +0.012 |
imet | +0.012 |
ian | +0.012 |
brow | +0.012 |
Neg logit contributions | |
late | -0.014 |
seek | -0.012 |
quiet | -0.012 |
Lisa | -0.011 |
lic | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | -3.76e-01 |
Pos logit contributions | |
late | +0.012 |
quiet | +0.010 |
seek | +0.010 |
Lisa | +0.010 |
Lisa | +0.009 |
Neg logit contributions | |
penn | -0.016 |
pen | -0.016 |
imet | -0.016 |
ian | -0.016 |
Ow | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | -3.63e-01 |
Pos logit contributions | |
late | +0.011 |
seek | +0.010 |
quiet | +0.010 |
Lisa | +0.009 |
too | +0.009 |
Neg logit contributions | |
pen | -0.011 |
imet | -0.011 |
Ow | -0.011 |
penn | -0.011 |
Pe | -0.010 |
Token | decorated |
Token ID | 24789 |
Feature activation | -7.42e-01 |
Pos logit contributions | |
penn | +0.005 |
imet | +0.005 |
knees | +0.005 |
bank | +0.005 |
Pe | +0.005 |
Neg logit contributions | |
late | -0.008 |
lic | -0.006 |
quiet | -0.006 |
Lisa | -0.006 |
aky | -0.006 |
Token | it |
Token ID | 340 |
Feature activation | +5.76e-01 |
Pos logit contributions | |
late | +0.031 |
seek | +0.029 |
aky | +0.028 |
lic | +0.027 |
quiet | +0.027 |
Neg logit contributions | |
penn | -0.016 |
pen | -0.014 |
Ow | -0.014 |
Pe | -0.013 |
ware | -0.013 |
Token | with |
Token ID | 351 |
Feature activation | +9.52e-01 |
Pos logit contributions | |
penn | +0.018 |
pen | +0.017 |
imet | +0.016 |
bank | +0.016 |
knees | +0.016 |
Neg logit contributions | |
late | -0.017 |
Lisa | -0.015 |
seek | -0.015 |
Lisa | -0.015 |
Sarah | -0.014 |
Token | shells |
Token ID | 19679 |
Feature activation | +6.23e-01 |
Pos logit contributions | |
penn | +0.029 |
pen | +0.025 |
Ow | +0.025 |
imet | +0.025 |
brow | +0.025 |
Neg logit contributions | |
late | -0.031 |
seek | -0.028 |
Lisa | -0.027 |
lic | -0.027 |
Lisa | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | +2.58e-01 |
Pos logit contributions | |
penn | +0.019 |
pen | +0.017 |
imet | +0.017 |
ome | +0.016 |
ian | +0.016 |
Neg logit contributions | |
late | -0.021 |
seek | -0.020 |
Lisa | -0.020 |
quiet | -0.019 |
Lisa | -0.018 |
Token | seaw |
Token ID | 32085 |
Feature activation | +2.82e+00 |
Pos logit contributions | |
penn | +0.006 |
imet | +0.005 |
pen | +0.005 |
brow | +0.005 |
trek | +0.005 |
Neg logit contributions | |
late | -0.010 |
seek | -0.009 |
quiet | -0.009 |
Lisa | -0.009 |
Lisa | -0.009 |
Token | eed |
Token ID | 2308 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
penn | +0.059 |
imet | +0.059 |
pen | +0.058 |
ian | +0.057 |
Ow | +0.055 |
Neg logit contributions | |
late | -0.106 |
seek | -0.099 |
quiet | -0.099 |
too | -0.093 |
Lisa | -0.093 |
Token | . |
Token ID | 13 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
penn | +0.033 |
imet | +0.030 |
pen | +0.029 |
ian | +0.028 |
ome | +0.028 |
Neg logit contributions | |
late | -0.040 |
seek | -0.038 |
Lisa | -0.037 |
quiet | -0.035 |
Sarah | -0.035 |
Token | They |
Token ID | 1119 |
Feature activation | -7.54e-02 |
Pos logit contributions | |
late | +0.006 |
seek | +0.006 |
quiet | +0.006 |
aky | +0.005 |
Lisa | +0.005 |
Neg logit contributions | |
imet | -0.007 |
penn | -0.007 |
Ow | -0.006 |
pen | -0.006 |
Pe | -0.006 |
Token | had |
Token ID | 550 |
Feature activation | -4.93e-01 |
Pos logit contributions | |
late | +0.003 |
quiet | +0.002 |
seek | +0.002 |
Lisa | +0.002 |
Lisa | +0.002 |
Neg logit contributions | |
penn | -0.002 |
pen | -0.002 |
ian | -0.002 |
imet | -0.002 |
trek | -0.002 |
Token | a |
Token ID | 257 |
Feature activation | -2.26e-01 |
Pos logit contributions | |
late | +0.011 |
seek | +0.009 |
Lisa | +0.008 |
Lisa | +0.008 |
quiet | +0.008 |
Neg logit contributions | |
penn | -0.020 |
imet | -0.019 |
Pe | -0.018 |
pen | -0.018 |
brow | -0.018 |
Token | a |
Token ID | 257 |
Feature activation | -5.94e-02 |
Pos logit contributions | |
late | +0.002 |
lic | +0.002 |
seek | +0.002 |
aking | +0.002 |
quiet | +0.002 |
Neg logit contributions | |
penn | -0.002 |
pen | -0.002 |
Pe | -0.002 |
growth | -0.002 |
Ow | -0.002 |
Token | place |
Token ID | 1295 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
late | +0.002 |
Lisa | +0.001 |
seek | +0.001 |
Lisa | +0.001 |
quiet | +0.001 |
Neg logit contributions | |
penn | -0.002 |
pen | -0.002 |
imet | -0.002 |
ian | -0.002 |
bank | -0.002 |
Token | she |
Token ID | 673 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
pen | +0.025 |
penn | +0.025 |
ome | +0.024 |
brow | +0.023 |
bank | +0.023 |
Neg logit contributions | |
late | -0.033 |
Lisa | -0.030 |
Happy | -0.030 |
lic | -0.030 |
pans | -0.030 |
Token | had |
Token ID | 550 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
penn | +0.004 |
imet | +0.004 |
pen | +0.004 |
ian | +0.004 |
Ow | +0.004 |
Neg logit contributions | |
late | -0.004 |
quiet | -0.004 |
seek | -0.004 |
Lisa | -0.004 |
Lisa | -0.003 |
Token | never |
Token ID | 1239 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
penn | +0.007 |
pen | +0.007 |
imet | +0.007 |
ian | +0.007 |
Ow | +0.006 |
Neg logit contributions | |
late | -0.008 |
seek | -0.007 |
quiet | -0.007 |
Lisa | -0.007 |
Lisa | -0.007 |
Token | looked |
Token ID | 3114 |
Feature activation | +2.82e+00 |
Pos logit contributions | |
penn | +0.051 |
imet | +0.050 |
pen | +0.049 |
ian | +0.049 |
Ow | +0.048 |
Neg logit contributions | |
late | -0.043 |
seek | -0.037 |
quiet | -0.037 |
Lisa | -0.034 |
too | -0.034 |
Token | before |
Token ID | 878 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
penn | +0.091 |
pen | +0.085 |
imet | +0.084 |
ome | +0.081 |
Pe | +0.080 |
Neg logit contributions | |
late | -0.080 |
Lisa | -0.078 |
seek | -0.077 |
Lisa | -0.073 |
quiet | -0.071 |
Token | . |
Token ID | 13 |
Feature activation | +7.66e-01 |
Pos logit contributions | |
penn | +0.055 |
pen | +0.054 |
imet | +0.053 |
Pe | +0.050 |
ian | +0.050 |
Neg logit contributions | |
late | -0.051 |
seek | -0.049 |
Lisa | -0.049 |
quiet | -0.048 |
Lisa | -0.047 |
Token | She |
Token ID | 1375 |
Feature activation | +3.90e-01 |
Pos logit contributions | |
imet | +0.019 |
Pe | +0.018 |
penn | +0.018 |
pen | +0.018 |
Ow | +0.017 |
Neg logit contributions | |
seek | -0.026 |
late | -0.025 |
quiet | -0.025 |
aky | -0.023 |
pans | -0.023 |
Token | was |
Token ID | 373 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
penn | +0.012 |
imet | +0.012 |
pen | +0.012 |
ian | +0.011 |
Ow | +0.011 |
Neg logit contributions | |
late | -0.012 |
quiet | -0.011 |
seek | -0.011 |
Lisa | -0.010 |
too | -0.010 |
Token | very |
Token ID | 845 |
Feature activation | -7.86e-01 |
Pos logit contributions | |
penn | +0.004 |
Pe | +0.004 |
Mont | +0.004 |
Twe | +0.003 |
brow | +0.003 |
Neg logit contributions | |
Lisa | -0.003 |
seek | -0.003 |
pans | -0.002 |
late | -0.002 |
apes | -0.002 |
Token | Joe |
Token ID | 5689 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
late | +0.008 |
seek | +0.008 |
quiet | +0.007 |
Lisa | +0.007 |
Lisa | +0.007 |
Neg logit contributions | |
imet | -0.009 |
penn | -0.009 |
pen | -0.009 |
Ow | -0.009 |
ian | -0.009 |
Token | a |
Token ID | 257 |
Feature activation | +9.30e-01 |
Pos logit contributions | |
penn | +0.038 |
imet | +0.037 |
pen | +0.037 |
ian | +0.036 |
Ow | +0.036 |
Neg logit contributions | |
late | -0.027 |
quiet | -0.023 |
seek | -0.023 |
Lisa | -0.021 |
too | -0.020 |
Token | tasty |
Token ID | 25103 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
penn | +0.030 |
imet | +0.030 |
pen | +0.030 |
Ow | +0.029 |
ian | +0.029 |
Neg logit contributions | |
late | -0.025 |
seek | -0.022 |
Lisa | -0.021 |
Lisa | -0.021 |
quiet | -0.020 |
Token | piece |
Token ID | 3704 |
Feature activation | +5.38e-01 |
Pos logit contributions | |
penn | +0.007 |
imet | +0.007 |
pen | +0.007 |
ian | +0.007 |
yle | +0.007 |
Neg logit contributions | |
late | -0.007 |
quiet | -0.007 |
seek | -0.007 |
too | -0.006 |
Lisa | -0.006 |
Token | of |
Token ID | 286 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
penn | +0.018 |
pen | +0.018 |
imet | +0.017 |
ome | +0.017 |
ian | +0.017 |
Neg logit contributions | |
late | -0.014 |
seek | -0.013 |
quiet | -0.013 |
Lisa | -0.012 |
Lisa | -0.012 |
Token | seaw |
Token ID | 32085 |
Feature activation | +2.81e+00 |
Pos logit contributions | |
late | +0.007 |
seek | +0.006 |
Lisa | +0.006 |
quiet | +0.006 |
Lisa | +0.006 |
Neg logit contributions | |
penn | -0.005 |
pen | -0.005 |
imet | -0.005 |
Ow | -0.004 |
ian | -0.004 |
Token | eed |
Token ID | 2308 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
penn | +0.065 |
imet | +0.063 |
Pe | +0.062 |
ian | +0.061 |
anybody | +0.061 |
Neg logit contributions | |
late | -0.099 |
seek | -0.095 |
quiet | -0.093 |
too | -0.088 |
aky | -0.086 |
Token | , |
Token ID | 11 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
penn | +0.045 |
imet | +0.043 |
pen | +0.043 |
ian | +0.041 |
ome | +0.041 |
Neg logit contributions | |
late | -0.038 |
Lisa | -0.034 |
seek | -0.034 |
quiet | -0.033 |
Lisa | -0.033 |
Token | and |
Token ID | 290 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
penn | +0.005 |
imet | +0.005 |
brow | +0.004 |
pen | +0.004 |
mun | +0.004 |
Neg logit contributions | |
late | -0.004 |
seek | -0.004 |
Lisa | -0.004 |
Happy | -0.004 |
quiet | -0.004 |
Token | Joe |
Token ID | 5689 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
Pe | +0.005 |
trek | +0.004 |
penn | +0.004 |
imet | +0.004 |
knees | +0.004 |
Neg logit contributions | |
late | -0.007 |
aky | -0.006 |
Lisa | -0.006 |
seek | -0.006 |
Happy | -0.006 |
Token | ate |
Token ID | 15063 |
Feature activation | -9.85e-02 |
Pos logit contributions | |
penn | +0.033 |
imet | +0.033 |
pen | +0.033 |
ian | +0.032 |
Ow | +0.031 |
Neg logit contributions | |
late | -0.043 |
quiet | -0.038 |
seek | -0.038 |
Lisa | -0.036 |
Lisa | -0.035 |
Token | sea |
Token ID | 5417 |
Feature activation | +2.73e-01 |
Pos logit contributions | |
penn | +0.003 |
bank | +0.003 |
pen | +0.003 |
imet | +0.003 |
trek | +0.003 |
Neg logit contributions | |
late | -0.006 |
Lisa | -0.006 |
seek | -0.005 |
Lisa | -0.005 |
lic | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
ome | +0.006 |
penn | +0.006 |
pen | +0.006 |
bank | +0.006 |
imet | +0.006 |
Neg logit contributions | |
late | -0.010 |
Lisa | -0.010 |
Happy | -0.009 |
Sarah | -0.009 |
seek | -0.009 |
Token | She |
Token ID | 1375 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
late | +0.006 |
Lisa | +0.005 |
quiet | +0.005 |
too | +0.005 |
seek | +0.005 |
Neg logit contributions | |
penn | -0.009 |
pen | -0.009 |
trigger | -0.009 |
ian | -0.009 |
yle | -0.008 |
Token | took |
Token ID | 1718 |
Feature activation | +3.14e-01 |
Pos logit contributions | |
penn | +0.003 |
pen | +0.003 |
imet | +0.003 |
ian | +0.003 |
trek | +0.003 |
Neg logit contributions | |
late | -0.005 |
seek | -0.004 |
quiet | -0.004 |
Lisa | -0.004 |
Lisa | -0.004 |
Token | some |
Token ID | 617 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
penn | +0.010 |
imet | +0.010 |
pen | +0.010 |
ian | +0.009 |
Ow | +0.009 |
Neg logit contributions | |
late | -0.009 |
seek | -0.008 |
quiet | -0.008 |
Lisa | -0.008 |
Lisa | -0.008 |
Token | seaw |
Token ID | 32085 |
Feature activation | +2.79e+00 |
Pos logit contributions | |
penn | +0.004 |
imet | +0.004 |
pen | +0.004 |
ian | +0.004 |
trek | +0.004 |
Neg logit contributions | |
late | -0.005 |
seek | -0.004 |
quiet | -0.004 |
Lisa | -0.004 |
Lisa | -0.004 |
Token | eed |
Token ID | 2308 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
penn | +0.057 |
imet | +0.056 |
pen | +0.055 |
ian | +0.054 |
Ow | +0.053 |
Neg logit contributions | |
late | -0.110 |
seek | -0.102 |
quiet | -0.102 |
Lisa | -0.096 |
too | -0.095 |
Token | , |
Token ID | 11 |
Feature activation | +3.10e-01 |
Pos logit contributions | |
penn | +0.050 |
imet | +0.049 |
pen | +0.048 |
ian | +0.047 |
ome | +0.046 |
Neg logit contributions | |
late | -0.045 |
seek | -0.039 |
quiet | -0.038 |
Lisa | -0.038 |
Lisa | -0.036 |
Token | made |
Token ID | 925 |
Feature activation | -5.14e-01 |
Pos logit contributions | |
penn | +0.008 |
imet | +0.008 |
pen | +0.007 |
trek | +0.007 |
brow | +0.007 |
Neg logit contributions | |
late | -0.011 |
seek | -0.010 |
quiet | -0.010 |
Lisa | -0.009 |
Lisa | -0.009 |
Token | a |
Token ID | 257 |
Feature activation | -3.37e-02 |
Pos logit contributions | |
late | +0.016 |
seek | +0.015 |
quiet | +0.013 |
Lisa | +0.013 |
Lisa | +0.013 |
Neg logit contributions | |
penn | -0.016 |
pen | -0.015 |
imet | -0.015 |
Ow | -0.014 |
ian | -0.014 |
Token | big |
Token ID | 1263 |
Feature activation | -3.60e-01 |
Pos logit contributions | |
late | +0.001 |
Lisa | +0.001 |
Lisa | +0.001 |
seek | +0.001 |
Happy | +0.001 |
Neg logit contributions | |
penn | -0.001 |
imet | -0.001 |
bank | -0.001 |
trek | -0.001 |
pen | -0.001 |
Token | ate |
Token ID | 378 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
late | +0.043 |
seek | +0.040 |
Lisa | +0.037 |
quiet | +0.036 |
Lisa | +0.035 |
Neg logit contributions | |
penn | -0.028 |
pen | -0.027 |
ian | -0.025 |
yle | -0.025 |
Ow | -0.024 |
Token | it |
Token ID | 340 |
Feature activation | +3.96e-01 |
Pos logit contributions | |
late | +0.010 |
seek | +0.010 |
aky | +0.009 |
lic | +0.009 |
Lisa | +0.008 |
Neg logit contributions | |
penn | -0.008 |
Pe | -0.007 |
yle | -0.007 |
pen | -0.007 |
knees | -0.006 |
Token | with |
Token ID | 351 |
Feature activation | +5.99e-01 |
Pos logit contributions | |
penn | +0.012 |
imet | +0.011 |
bank | +0.011 |
pen | +0.011 |
knees | +0.011 |
Neg logit contributions | |
late | -0.012 |
Lisa | -0.011 |
Lisa | -0.010 |
Sarah | -0.010 |
seek | -0.010 |
Token | shells |
Token ID | 19679 |
Feature activation | +5.77e-01 |
Pos logit contributions | |
penn | +0.019 |
pen | +0.017 |
imet | +0.017 |
Ow | +0.017 |
brow | +0.017 |
Neg logit contributions | |
late | -0.019 |
seek | -0.017 |
Lisa | -0.016 |
quiet | -0.015 |
Lisa | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
penn | +0.017 |
imet | +0.015 |
pen | +0.015 |
ian | +0.015 |
ome | +0.014 |
Neg logit contributions | |
late | -0.020 |
seek | -0.019 |
Lisa | -0.019 |
quiet | -0.018 |
Lisa | -0.017 |
Token | seaw |
Token ID | 32085 |
Feature activation | +2.79e+00 |
Pos logit contributions | |
penn | +0.004 |
imet | +0.003 |
pen | +0.003 |
ian | +0.003 |
Ow | +0.003 |
Neg logit contributions | |
late | -0.006 |
seek | -0.005 |
quiet | -0.005 |
Lisa | -0.005 |
Lisa | -0.005 |
Token | eed |
Token ID | 2308 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
penn | +0.058 |
pen | +0.058 |
imet | +0.058 |
ian | +0.057 |
Ow | +0.055 |
Neg logit contributions | |
late | -0.105 |
quiet | -0.098 |
seek | -0.097 |
too | -0.092 |
Lisa | -0.092 |
Token | . |
Token ID | 13 |
Feature activation | -3.45e-01 |
Pos logit contributions | |
penn | +0.031 |
imet | +0.028 |
pen | +0.027 |
ian | +0.027 |
ome | +0.025 |
Neg logit contributions | |
late | -0.039 |
seek | -0.037 |
Lisa | -0.036 |
quiet | -0.034 |
Sarah | -0.034 |
Token | They |
Token ID | 1119 |
Feature activation | -6.98e-01 |
Pos logit contributions | |
late | +0.010 |
seek | +0.009 |
quiet | +0.009 |
Lisa | +0.008 |
Lisa | +0.008 |
Neg logit contributions | |
penn | -0.011 |
imet | -0.011 |
pen | -0.011 |
Ow | -0.011 |
ian | -0.011 |
Token | find |
Token ID | 1064 |
Feature activation | -8.65e-02 |
Pos logit contributions | |
late | +0.026 |
seek | +0.022 |
quiet | +0.022 |
Lisa | +0.021 |
Lisa | +0.021 |
Neg logit contributions | |
penn | -0.018 |
pen | -0.018 |
imet | -0.017 |
ian | -0.017 |
trek | -0.016 |
Token | a |
Token ID | 257 |
Feature activation | -5.43e-01 |
Pos logit contributions | |
late | +0.002 |
quiet | +0.002 |
seek | +0.002 |
Lisa | +0.002 |
too | +0.002 |
Neg logit contributions | |
imet | -0.003 |
pen | -0.003 |
penn | -0.003 |
ian | -0.003 |
Ow | -0.003 |
Token | They |
Token ID | 1119 |
Feature activation | -3.78e-01 |
Pos logit contributions | |
late | +0.006 |
Lisa | +0.005 |
quiet | +0.005 |
too | +0.005 |
seek | +0.004 |
Neg logit contributions | |
penn | -0.010 |
pen | -0.010 |
ian | -0.010 |
yle | -0.010 |
imet | -0.010 |
Token | went |
Token ID | 1816 |
Feature activation | -3.50e-01 |
Pos logit contributions | |
late | +0.011 |
quiet | +0.010 |
seek | +0.010 |
too | +0.009 |
Lisa | +0.009 |
Neg logit contributions | |
imet | -0.012 |
penn | -0.012 |
pen | -0.012 |
Ow | -0.011 |
ian | -0.011 |
Token | past |
Token ID | 1613 |
Feature activation | +6.76e-01 |
Pos logit contributions | |
Lisa | +0.012 |
late | +0.011 |
Lisa | +0.011 |
Happy | +0.010 |
aky | +0.010 |
Neg logit contributions | |
penn | -0.009 |
bank | -0.009 |
trek | -0.009 |
imet | -0.009 |
kay | -0.008 |
Token | rocks |
Token ID | 12586 |
Feature activation | +6.08e-01 |
Pos logit contributions | |
imet | +0.026 |
penn | +0.026 |
pen | +0.026 |
ian | +0.026 |
ome | +0.025 |
Neg logit contributions | |
late | -0.015 |
quiet | -0.013 |
seek | -0.012 |
Lisa | -0.012 |
too | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
penn | +0.021 |
pen | +0.020 |
imet | +0.020 |
ian | +0.019 |
ome | +0.019 |
Neg logit contributions | |
late | -0.017 |
seek | -0.015 |
quiet | -0.014 |
Lisa | -0.014 |
Lisa | -0.014 |
Token | seaw |
Token ID | 32085 |
Feature activation | +2.79e+00 |
Pos logit contributions | |
penn | +0.004 |
imet | +0.003 |
pen | +0.003 |
ian | +0.003 |
Ow | +0.003 |
Neg logit contributions | |
late | -0.004 |
quiet | -0.004 |
seek | -0.004 |
Lisa | -0.004 |
Lisa | -0.004 |
Token | eed |
Token ID | 2308 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
imet | +0.074 |
Higher | +0.073 |
anybody | +0.071 |
ian | +0.071 |
penn | +0.070 |
Neg logit contributions | |
late | -0.084 |
quiet | -0.084 |
seek | -0.083 |
lic | -0.079 |
aky | -0.079 |
Token | . |
Token ID | 13 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
penn | +0.036 |
pen | +0.035 |
imet | +0.035 |
ian | +0.034 |
ome | +0.033 |
Neg logit contributions | |
late | -0.033 |
seek | -0.029 |
quiet | -0.028 |
Lisa | -0.028 |
Lisa | -0.027 |
Token | They |
Token ID | 1119 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
late | +0.003 |
Lisa | +0.002 |
quiet | +0.002 |
too | +0.002 |
seek | +0.002 |
Neg logit contributions | |
penn | -0.004 |
pen | -0.003 |
ian | -0.003 |
yle | -0.003 |
imet | -0.003 |
Token | saw |
Token ID | 2497 |
Feature activation | -4.94e-01 |
Pos logit contributions | |
late | +0.003 |
quiet | +0.003 |
seek | +0.003 |
Lisa | +0.003 |
too | +0.003 |
Neg logit contributions | |
imet | -0.003 |
penn | -0.003 |
pen | -0.003 |
ian | -0.003 |
Ow | -0.003 |
Token | other |
Token ID | 584 |
Feature activation | -2.97e-01 |
Pos logit contributions | |
late | +0.012 |
quiet | +0.011 |
seek | +0.010 |
Lisa | +0.010 |
too | +0.010 |
Neg logit contributions | |
imet | -0.018 |
pen | -0.018 |
ian | -0.018 |
penn | -0.018 |
ome | -0.017 |
Token | friends |
Token ID | 2460 |
Feature activation | +3.28e-01 |
Pos logit contributions | |
late | +0.001 |
seek | +0.001 |
Lisa | +0.001 |
lic | +0.001 |
Lisa | +0.001 |
Neg logit contributions | |
penn | -0.001 |
pen | -0.001 |
ian | -0.001 |
Ow | -0.001 |
brow | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | +3.88e-01 |
Pos logit contributions | |
penn | +0.011 |
pen | +0.011 |
imet | +0.011 |
ian | +0.011 |
ome | +0.011 |
Neg logit contributions | |
late | -0.009 |
seek | -0.008 |
quiet | -0.007 |
Lisa | -0.007 |
Lisa | -0.007 |
Token | he |
Token ID | 339 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
penn | +0.012 |
imet | +0.012 |
pen | +0.011 |
ian | +0.011 |
Ow | +0.011 |
Neg logit contributions | |
late | -0.012 |
seek | -0.010 |
quiet | -0.010 |
Lisa | -0.010 |
Lisa | -0.009 |
Token | never |
Token ID | 1239 |
Feature activation | +8.60e-01 |
Pos logit contributions | |
imet | +0.007 |
trigger | +0.007 |
penn | +0.007 |
brow | +0.007 |
Ow | +0.007 |
Neg logit contributions | |
late | -0.005 |
quiet | -0.005 |
too | -0.005 |
soon | -0.004 |
seek | -0.004 |
Token | hid |
Token ID | 24519 |
Feature activation | +9.41e-01 |
Pos logit contributions | |
penn | +0.027 |
imet | +0.027 |
ian | +0.026 |
pen | +0.026 |
Ow | +0.025 |
Neg logit contributions | |
late | -0.027 |
quiet | -0.024 |
seek | -0.024 |
too | -0.022 |
Lisa | -0.022 |
Token | anything |
Token ID | 1997 |
Feature activation | +2.78e+00 |
Pos logit contributions | |
penn | +0.027 |
pen | +0.025 |
imet | +0.024 |
anybody | +0.024 |
Ow | +0.024 |
Neg logit contributions | |
late | -0.030 |
seek | -0.029 |
Lisa | -0.028 |
quiet | -0.027 |
lic | -0.027 |
Token | under |
Token ID | 739 |
Feature activation | +5.36e-01 |
Pos logit contributions | |
penn | +0.081 |
pen | +0.079 |
imet | +0.079 |
ome | +0.074 |
anybody | +0.072 |
Neg logit contributions | |
late | -0.084 |
seek | -0.078 |
Lisa | -0.077 |
Lisa | -0.076 |
quiet | -0.075 |
Token | his |
Token ID | 465 |
Feature activation | -7.15e-02 |
Pos logit contributions | |
penn | +0.016 |
pen | +0.015 |
imet | +0.015 |
Ow | +0.015 |
ian | +0.015 |
Neg logit contributions | |
late | -0.017 |
seek | -0.016 |
quiet | -0.015 |
Lisa | -0.014 |
Lisa | -0.014 |
Token | bed |
Token ID | 3996 |
Feature activation | +6.11e-01 |
Pos logit contributions | |
late | +0.002 |
seek | +0.002 |
quiet | +0.002 |
Lisa | +0.002 |
Lisa | +0.002 |
Neg logit contributions | |
penn | -0.002 |
pen | -0.002 |
imet | -0.002 |
ian | -0.002 |
Ow | -0.002 |
Token | again |
Token ID | 757 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
pen | +0.016 |
ome | +0.015 |
penn | +0.014 |
ian | +0.013 |
bank | +0.013 |
Neg logit contributions | |
late | -0.020 |
pans | -0.020 |
seek | -0.019 |
Lisa | -0.019 |
quiet | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | +9.13e-01 |
Pos logit contributions | |
pen | +0.023 |
penn | +0.020 |
anyone | +0.019 |
bank | +0.018 |
anybody | +0.018 |
Neg logit contributions | |
lic | -0.042 |
Lisa | -0.039 |
Kate | -0.037 |
Lisa | -0.036 |
Happy | -0.036 |
Token | her |
Token ID | 607 |
Feature activation | +2.81e-01 |
Pos logit contributions | |
penn | +0.018 |
pen | +0.018 |
imet | +0.018 |
ian | +0.017 |
Ow | +0.017 |
Neg logit contributions | |
late | -0.015 |
seek | -0.014 |
quiet | -0.013 |
Lisa | -0.012 |
Lisa | -0.012 |
Token | grandma |
Token ID | 49890 |
Feature activation | +6.01e-01 |
Pos logit contributions | |
imet | +0.009 |
penn | +0.009 |
pen | +0.009 |
ian | +0.009 |
Ow | +0.008 |
Neg logit contributions | |
late | -0.008 |
quiet | -0.007 |
seek | -0.007 |
Lisa | -0.007 |
too | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +4.84e-01 |
Pos logit contributions | |
penn | +0.019 |
pen | +0.019 |
imet | +0.019 |
ian | +0.018 |
Ow | +0.018 |
Neg logit contributions | |
late | -0.017 |
seek | -0.015 |
quiet | -0.015 |
Lisa | -0.014 |
Lisa | -0.014 |
Token | On |
Token ID | 1550 |
Feature activation | +6.53e-01 |
Pos logit contributions | |
imet | +0.013 |
penn | +0.012 |
pen | +0.012 |
ian | +0.012 |
Ow | +0.012 |
Neg logit contributions | |
late | -0.016 |
seek | -0.015 |
quiet | -0.015 |
aky | -0.013 |
Lisa | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | +8.37e-01 |
Pos logit contributions | |
penn | +0.021 |
ian | +0.020 |
pen | +0.019 |
imet | +0.019 |
Pe | +0.019 |
Neg logit contributions | |
late | -0.019 |
seek | -0.018 |
quiet | -0.017 |
lic | -0.015 |
aky | -0.015 |
Token | way |
Token ID | 835 |
Feature activation | +2.78e+00 |
Pos logit contributions | |
penn | +0.024 |
imet | +0.024 |
pen | +0.024 |
ian | +0.023 |
Ow | +0.023 |
Neg logit contributions | |
late | -0.027 |
seek | -0.024 |
quiet | -0.024 |
Lisa | -0.023 |
Lisa | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | -3.61e-01 |
Pos logit contributions | |
penn | +0.091 |
pen | +0.091 |
imet | +0.090 |
ian | +0.087 |
Ow | +0.086 |
Neg logit contributions | |
late | -0.079 |
seek | -0.069 |
quiet | -0.069 |
Lisa | -0.067 |
too | -0.064 |
Token | she |
Token ID | 673 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
late | +0.010 |
seek | +0.009 |
quiet | +0.009 |
Lisa | +0.009 |
Lisa | +0.008 |
Neg logit contributions | |
penn | -0.012 |
imet | -0.011 |
pen | -0.011 |
ian | -0.011 |
Ow | -0.011 |
Token | found |
Token ID | 1043 |
Feature activation | +5.70e-02 |
Pos logit contributions | |
penn | +0.006 |
imet | +0.006 |
trigger | +0.006 |
Ow | +0.006 |
pen | +0.006 |
Neg logit contributions | |
late | -0.004 |
quiet | -0.003 |
seek | -0.003 |
Lisa | -0.003 |
too | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | +4.10e-01 |
Pos logit contributions | |
imet | +0.002 |
pen | +0.002 |
ian | +0.002 |
penn | +0.002 |
unn | +0.002 |
Neg logit contributions | |
late | -0.001 |
quiet | -0.001 |
Lisa | -0.001 |
too | -0.001 |
seek | -0.001 |
Token | shiny |
Token ID | 22441 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
penn | +0.016 |
pen | +0.015 |
imet | +0.015 |
ian | +0.015 |
Ow | +0.015 |
Neg logit contributions | |
late | -0.010 |
seek | -0.008 |
quiet | -0.008 |
Lisa | -0.008 |
Lisa | -0.007 |
Token | things |
Token ID | 1243 |
Feature activation | +9.73e-01 |
Pos logit contributions | |
late | +0.004 |
quiet | +0.004 |
seek | +0.003 |
Lisa | +0.003 |
too | +0.003 |
Neg logit contributions | |
imet | -0.004 |
pen | -0.004 |
penn | -0.004 |
ian | -0.004 |
Ow | -0.004 |
Token | â |
Token ID | 22940 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
penn | +0.026 |
pen | +0.025 |
imet | +0.025 |
ome | +0.025 |
Mont | +0.023 |
Neg logit contributions | |
late | -0.032 |
Lisa | -0.031 |
seek | -0.030 |
quiet | -0.030 |
Lisa | -0.029 |
Token | € |
Token ID | 26391 |
Feature activation | +6.73e-02 |
Pos logit contributions | |
penn | +0.004 |
imet | +0.003 |
pen | +0.003 |
ian | +0.003 |
Ow | +0.003 |
Neg logit contributions | |
late | -0.004 |
seek | -0.004 |
quiet | -0.004 |
Lisa | -0.004 |
Lisa | -0.004 |
Token | � |
Token ID | 447 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
pen | +0.002 |
imet | +0.002 |
penn | +0.002 |
ian | +0.002 |
Ow | +0.001 |
Neg logit contributions | |
late | -0.002 |
quiet | -0.002 |
seek | -0.002 |
too | -0.002 |
Lisa | -0.002 |
Token | � |
Token ID | 251 |
Feature activation | +5.10e-01 |
Pos logit contributions | |
penn | +0.002 |
imet | +0.002 |
pen | +0.002 |
trigger | +0.002 |
ian | +0.002 |
Neg logit contributions | |
late | -0.007 |
quiet | -0.006 |
seek | -0.006 |
Lisa | -0.006 |
too | -0.006 |
Token | col |
Token ID | 4033 |
Feature activation | +2.77e+00 |
Pos logit contributions | |
penn | +0.015 |
imet | +0.015 |
Pe | +0.015 |
ian | +0.015 |
pen | +0.014 |
Neg logit contributions | |
late | -0.016 |
seek | -0.015 |
quiet | -0.014 |
Lisa | -0.014 |
Lisa | -0.013 |
Token | ouring |
Token ID | 21823 |
Feature activation | +7.48e-01 |
Pos logit contributions | |
penn | +0.074 |
bank | +0.071 |
trigger | +0.070 |
endings | +0.070 |
lur | +0.069 |
Neg logit contributions | |
lic | -0.079 |
Lisa | -0.075 |
Kate | -0.074 |
too | -0.071 |
Happy | -0.070 |
Token | books |
Token ID | 3835 |
Feature activation | +6.52e-01 |
Pos logit contributions | |
penn | +0.024 |
ian | +0.022 |
pen | +0.022 |
Pe | +0.021 |
yle | +0.021 |
Neg logit contributions | |
late | -0.025 |
seek | -0.022 |
Lisa | -0.020 |
quiet | -0.020 |
Happy | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
pen | +0.019 |
penn | +0.018 |
ome | +0.018 |
ian | +0.017 |
imet | +0.017 |
Neg logit contributions | |
late | -0.022 |
seek | -0.019 |
Lisa | -0.019 |
quiet | -0.019 |
Lisa | -0.018 |
Token | stickers |
Token ID | 28568 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
penn | +0.007 |
Pe | +0.007 |
pen | +0.007 |
imet | +0.007 |
brow | +0.007 |
Neg logit contributions | |
late | -0.008 |
seek | -0.006 |
quiet | -0.006 |
aky | -0.006 |
Lisa | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | +7.22e-01 |
Pos logit contributions | |
penn | +0.006 |
pen | +0.006 |
ian | +0.006 |
imet | +0.006 |
ome | +0.006 |
Neg logit contributions | |
late | -0.006 |
seek | -0.005 |
quiet | -0.005 |
Lisa | -0.005 |
Lisa | -0.005 |
Token | Like |
Token ID | 4525 |
Feature activation | +9.30e-02 |
Pos logit contributions | |
late | +0.004 |
seek | +0.004 |
quiet | +0.004 |
impressed | +0.003 |
pleased | +0.003 |
Neg logit contributions | |
Pe | -0.003 |
imet | -0.003 |
Ow | -0.003 |
pen | -0.003 |
penn | -0.003 |
Token | shells |
Token ID | 19679 |
Feature activation | +6.03e-01 |
Pos logit contributions | |
penn | +0.003 |
imet | +0.003 |
pen | +0.003 |
Pe | +0.003 |
ian | +0.003 |
Neg logit contributions | |
late | -0.003 |
seek | -0.003 |
Lisa | -0.003 |
quiet | -0.002 |
Lisa | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
penn | +0.015 |
pen | +0.013 |
imet | +0.013 |
ome | +0.013 |
ian | +0.012 |
Neg logit contributions | |
late | -0.023 |
Lisa | -0.023 |
seek | -0.021 |
Kate | -0.021 |
Maria | -0.021 |
Token | rocks |
Token ID | 12586 |
Feature activation | +5.72e-01 |
Pos logit contributions | |
penn | +0.006 |
imet | +0.005 |
pen | +0.005 |
Pe | +0.005 |
Ow | +0.005 |
Neg logit contributions | |
late | -0.011 |
Lisa | -0.009 |
seek | -0.009 |
Lisa | -0.009 |
quiet | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
penn | +0.017 |
imet | +0.014 |
pen | +0.014 |
ome | +0.014 |
bank | +0.013 |
Neg logit contributions | |
Lisa | -0.019 |
late | -0.019 |
seek | -0.018 |
Maria | -0.018 |
Lisa | -0.018 |
Token | seaw |
Token ID | 32085 |
Feature activation | +2.74e+00 |
Pos logit contributions | |
penn | +0.004 |
imet | +0.004 |
pen | +0.003 |
Pe | +0.003 |
Ow | +0.003 |
Neg logit contributions | |
late | -0.007 |
Lisa | -0.006 |
seek | -0.006 |
Lisa | -0.006 |
quiet | -0.006 |
Token | eed |
Token ID | 2308 |
Feature activation | +8.79e-01 |
Pos logit contributions | |
penn | +0.058 |
imet | +0.058 |
pen | +0.057 |
ian | +0.056 |
Ow | +0.055 |
Neg logit contributions | |
late | -0.102 |
quiet | -0.095 |
seek | -0.094 |
too | -0.089 |
Lisa | -0.087 |
Token | ." |
Token ID | 526 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
penn | +0.025 |
imet | +0.021 |
ome | +0.021 |
pen | +0.020 |
bank | +0.020 |
Neg logit contributions | |
Lisa | -0.032 |
late | -0.031 |
Sarah | -0.031 |
Maria | -0.030 |
seek | -0.030 |
Token |
|
Token ID | 198 |
Feature activation | +2.97e-01 |
Pos logit contributions | |
late | +0.004 |
seek | +0.004 |
quiet | +0.004 |
lic | +0.004 |
aky | +0.003 |
Neg logit contributions | |
penn | -0.005 |
Pe | -0.005 |
imet | -0.005 |
pen | -0.005 |
Ow | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
pen | +0.009 |
Ow | +0.008 |
penn | +0.008 |
Pe | +0.008 |
imet | +0.008 |
Neg logit contributions | |
late | -0.009 |
seek | -0.008 |
quiet | -0.008 |
too | -0.007 |
lic | -0.007 |
Token | M |
Token ID | 44 |
Feature activation | +6.05e-02 |
Pos logit contributions | |
Twe | +0.005 |
pen | +0.005 |
penn | +0.005 |
Pe | +0.005 |
Mont | +0.005 |
Neg logit contributions | |
late | -0.005 |
too | -0.005 |
lic | -0.005 |
awesome | -0.004 |
quiet | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | +6.46e-01 |
Pos logit contributions | |
imet | +0.062 |
pen | +0.060 |
penn | +0.059 |
ian | +0.058 |
trigger | +0.058 |
Neg logit contributions | |
late | -0.025 |
quiet | -0.022 |
too | -0.021 |
seek | -0.019 |
Lisa | -0.019 |
Token | and |
Token ID | 290 |
Feature activation | +6.94e-01 |
Pos logit contributions | |
penn | +0.024 |
imet | +0.024 |
pen | +0.023 |
ian | +0.023 |
Ow | +0.023 |
Neg logit contributions | |
late | -0.016 |
quiet | -0.014 |
seek | -0.013 |
Lisa | -0.012 |
too | -0.012 |
Token | she |
Token ID | 673 |
Feature activation | +4.91e-01 |
Pos logit contributions | |
penn | +0.018 |
ian | +0.016 |
pen | +0.016 |
Pe | +0.016 |
imet | +0.016 |
Neg logit contributions | |
late | -0.025 |
seek | -0.022 |
quiet | -0.021 |
Lisa | -0.020 |
Lisa | -0.020 |
Token | had |
Token ID | 550 |
Feature activation | +7.56e-02 |
Pos logit contributions | |
penn | +0.012 |
imet | +0.012 |
pen | +0.012 |
ian | +0.012 |
Ow | +0.011 |
Neg logit contributions | |
late | -0.018 |
quiet | -0.016 |
seek | -0.016 |
Lisa | -0.015 |
Lisa | -0.015 |
Token | long |
Token ID | 890 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
penn | +0.003 |
imet | +0.003 |
pen | +0.003 |
ian | +0.003 |
Ow | +0.003 |
Neg logit contributions | |
late | -0.002 |
seek | -0.002 |
quiet | -0.002 |
Lisa | -0.002 |
Lisa | -0.002 |
Token | blonde |
Token ID | 21541 |
Feature activation | +2.71e+00 |
Pos logit contributions | |
imet | +0.030 |
pen | +0.030 |
penn | +0.030 |
ian | +0.029 |
trigger | +0.028 |
Neg logit contributions | |
late | -0.045 |
quiet | -0.042 |
seek | -0.041 |
Lisa | -0.039 |
too | -0.038 |
Token | hair |
Token ID | 4190 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
imet | +0.093 |
uable | +0.093 |
trigger | +0.092 |
pen | +0.090 |
ian | +0.089 |
Neg logit contributions | |
quiet | -0.074 |
too | -0.066 |
late | -0.065 |
friendly | -0.062 |
happy | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | -8.66e-02 |
Pos logit contributions | |
imet | +0.042 |
penn | +0.042 |
pen | +0.041 |
Ow | +0.041 |
ian | +0.040 |
Neg logit contributions | |
late | -0.025 |
quiet | -0.022 |
seek | -0.021 |
too | -0.020 |
Lisa | -0.019 |
Token | One |
Token ID | 1881 |
Feature activation | -4.26e-01 |
Pos logit contributions | |
late | +0.003 |
seek | +0.003 |
quiet | +0.003 |
aky | +0.003 |
pleased | +0.002 |
Neg logit contributions | |
imet | -0.002 |
Ow | -0.002 |
pen | -0.002 |
Higher | -0.002 |
ian | -0.002 |
Token | day |
Token ID | 1110 |
Feature activation | +5.58e-01 |
Pos logit contributions | |
late | +0.012 |
seek | +0.010 |
quiet | +0.010 |
Lisa | +0.010 |
Lisa | +0.009 |
Neg logit contributions | |
penn | -0.014 |
imet | -0.013 |
pen | -0.013 |
ian | -0.013 |
Ow | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | -6.53e-01 |
Pos logit contributions | |
penn | +0.019 |
pen | +0.018 |
imet | +0.018 |
ian | +0.018 |
Ow | +0.017 |
Neg logit contributions | |
late | -0.016 |
quiet | -0.014 |
seek | -0.014 |
Lisa | -0.014 |
too | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
pen | +0.021 |
penn | +0.020 |
imet | +0.020 |
ian | +0.020 |
ome | +0.020 |
Neg logit contributions | |
late | -0.021 |
quiet | -0.018 |
Lisa | -0.018 |
seek | -0.018 |
Lisa | -0.017 |
Token | he |
Token ID | 339 |
Feature activation | +3.94e-01 |
Pos logit contributions | |
penn | +0.005 |
imet | +0.005 |
pen | +0.005 |
Pe | +0.005 |
trek | +0.005 |
Neg logit contributions | |
late | -0.005 |
Lisa | -0.004 |
seek | -0.004 |
lic | -0.004 |
Happy | -0.004 |
Token | just |
Token ID | 655 |
Feature activation | +8.63e-01 |
Pos logit contributions | |
pen | +0.008 |
ian | +0.008 |
penn | +0.008 |
vered | +0.008 |
trek | +0.008 |
Neg logit contributions | |
late | -0.016 |
Lisa | -0.014 |
Lisa | -0.014 |
quiet | -0.013 |
seek | -0.013 |
Token | enjoyed |
Token ID | 8359 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
penn | +0.020 |
pen | +0.020 |
imet | +0.020 |
ian | +0.019 |
brow | +0.019 |
Neg logit contributions | |
late | -0.032 |
seek | -0.028 |
Lisa | -0.028 |
quiet | -0.028 |
Lisa | -0.027 |
Token | having |
Token ID | 1719 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
late | +0.005 |
lic | +0.004 |
Lisa | +0.004 |
aky | +0.004 |
seek | +0.004 |
Neg logit contributions | |
penn | -0.004 |
pen | -0.004 |
Pe | -0.003 |
imet | -0.003 |
yle | -0.003 |
Token | something |
Token ID | 1223 |
Feature activation | +2.70e+00 |
Pos logit contributions | |
late | +0.006 |
Lisa | +0.005 |
seek | +0.005 |
aky | +0.005 |
quiet | +0.005 |
Neg logit contributions | |
penn | -0.007 |
pen | -0.007 |
trek | -0.006 |
Pe | -0.006 |
imet | -0.006 |
Token | special |
Token ID | 2041 |
Feature activation | +6.36e-01 |
Pos logit contributions | |
penn | +0.096 |
imet | +0.092 |
pen | +0.089 |
ome | +0.087 |
Pe | +0.086 |
Neg logit contributions | |
late | -0.069 |
Lisa | -0.061 |
seek | -0.060 |
Lisa | -0.059 |
quiet | -0.057 |
Token | to |
Token ID | 284 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
penn | +0.019 |
imet | +0.018 |
pen | +0.018 |
ome | +0.018 |
ian | +0.017 |
Neg logit contributions | |
late | -0.020 |
Lisa | -0.018 |
quiet | -0.018 |
Lisa | -0.017 |
seek | -0.017 |
Token | jog |
Token ID | 48342 |
Feature activation | +7.97e-02 |
Pos logit contributions | |
pen | +0.005 |
penn | +0.005 |
rest | +0.005 |
anybody | +0.005 |
Mont | +0.004 |
Neg logit contributions | |
late | -0.011 |
lic | -0.009 |
Lisa | -0.009 |
Lisa | -0.009 |
aky | -0.009 |
Token | with |
Token ID | 351 |
Feature activation | +8.36e-01 |
Pos logit contributions | |
pen | +0.002 |
ome | +0.002 |
bank | +0.002 |
penn | +0.002 |
ian | +0.002 |
Neg logit contributions | |
Lisa | -0.003 |
Lisa | -0.002 |
late | -0.002 |
Maria | -0.002 |
pans | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
pen | +0.024 |
penn | +0.024 |
anybody | +0.022 |
imet | +0.022 |
ian | +0.022 |
Neg logit contributions | |
late | -0.027 |
Lisa | -0.023 |
seek | -0.023 |
lic | -0.023 |
Lisa | -0.022 |
Token | necklace |
Token ID | 38987 |
Feature activation | -7.88e-01 |
Pos logit contributions | |
penn | +0.014 |
imet | +0.014 |
pen | +0.013 |
ian | +0.013 |
Ow | +0.013 |
Neg logit contributions | |
late | -0.016 |
seek | -0.014 |
Lisa | -0.014 |
quiet | -0.014 |
Lisa | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
late | +0.027 |
Lisa | +0.024 |
quiet | +0.023 |
Happy | +0.023 |
seek | +0.022 |
Neg logit contributions | |
pen | -0.020 |
penn | -0.020 |
ome | -0.020 |
imet | -0.020 |
bank | -0.020 |
Token | They |
Token ID | 1119 |
Feature activation | -6.23e-03 |
Pos logit contributions | |
late | +0.004 |
quiet | +0.003 |
seek | +0.003 |
Lisa | +0.003 |
too | +0.003 |
Neg logit contributions | |
penn | -0.004 |
pen | -0.004 |
imet | -0.004 |
ian | -0.004 |
ome | -0.004 |
Token | found |
Token ID | 1043 |
Feature activation | +2.80e-02 |
Pos logit contributions | |
late | +0.000 |
seek | +0.000 |
quiet | +0.000 |
Lisa | +0.000 |
Lisa | +0.000 |
Neg logit contributions | |
penn | -0.000 |
pen | -0.000 |
ian | -0.000 |
imet | -0.000 |
trek | -0.000 |
Token | some |
Token ID | 617 |
Feature activation | +3.64e-01 |
Pos logit contributions | |
imet | +0.001 |
pen | +0.001 |
ian | +0.001 |
penn | +0.001 |
uf | +0.001 |
Neg logit contributions | |
late | -0.001 |
quiet | -0.001 |
Lisa | -0.001 |
seek | -0.001 |
too | -0.001 |
Token | seaw |
Token ID | 32085 |
Feature activation | +2.68e+00 |
Pos logit contributions | |
penn | +0.011 |
imet | +0.011 |
pen | +0.011 |
ian | +0.010 |
Ow | +0.010 |
Neg logit contributions | |
late | -0.011 |
seek | -0.010 |
quiet | -0.010 |
Lisa | -0.009 |
Lisa | -0.009 |
Token | eed |
Token ID | 2308 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
penn | +0.061 |
anybody | +0.061 |
Ow | +0.060 |
Sy | +0.060 |
imet | +0.060 |
Neg logit contributions | |
late | -0.091 |
seek | -0.089 |
quiet | -0.085 |
too | -0.080 |
lic | -0.080 |
Token | to |
Token ID | 284 |
Feature activation | -2.84e-01 |
Pos logit contributions | |
penn | +0.038 |
imet | +0.036 |
pen | +0.035 |
ome | +0.034 |
ian | +0.034 |
Neg logit contributions | |
late | -0.038 |
Lisa | -0.034 |
quiet | -0.033 |
seek | -0.033 |
Lisa | -0.032 |
Token | use |
Token ID | 779 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
late | +0.011 |
quiet | +0.009 |
Lisa | +0.009 |
seek | +0.009 |
Lisa | +0.009 |
Neg logit contributions | |
pen | -0.007 |
penn | -0.006 |
imet | -0.006 |
ian | -0.006 |
Pe | -0.006 |
Token | as |
Token ID | 355 |
Feature activation | +6.53e-01 |
Pos logit contributions | |
late | +0.007 |
seek | +0.006 |
Lisa | +0.006 |
quiet | +0.006 |
Lisa | +0.006 |
Neg logit contributions | |
penn | -0.006 |
imet | -0.006 |
pen | -0.006 |
ian | -0.006 |
Pe | -0.005 |
Token | string |
Token ID | 4731 |
Feature activation | +7.97e-01 |
Pos logit contributions | |
penn | +0.022 |
imet | +0.022 |
pen | +0.021 |
ian | +0.021 |
Ow | +0.020 |
Neg logit contributions | |
late | -0.017 |
seek | -0.016 |
quiet | -0.015 |
Lisa | -0.015 |
Lisa | -0.014 |
Token | something |
Token ID | 1223 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
late | +0.035 |
aky | +0.030 |
lic | +0.030 |
quiet | +0.030 |
aking | +0.029 |
Neg logit contributions | |
penn | -0.026 |
Pe | -0.025 |
imet | -0.025 |
ian | -0.024 |
pen | -0.023 |
Token | amazing |
Token ID | 4998 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
late | +0.006 |
seek | +0.005 |
Lisa | +0.005 |
quiet | +0.005 |
Lisa | +0.005 |
Neg logit contributions | |
penn | -0.006 |
imet | -0.006 |
pen | -0.005 |
ian | -0.005 |
ome | -0.005 |
Token | happened |
Token ID | 3022 |
Feature activation | -8.17e-01 |
Pos logit contributions | |
late | +0.058 |
seek | +0.054 |
quiet | +0.054 |
Lisa | +0.052 |
Lisa | +0.051 |
Neg logit contributions | |
imet | -0.009 |
ian | -0.008 |
pen | -0.007 |
ome | -0.007 |
penn | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -8.20e-01 |
Pos logit contributions | |
late | +0.028 |
Lisa | +0.027 |
seek | +0.026 |
quiet | +0.026 |
pans | +0.026 |
Neg logit contributions | |
pen | -0.018 |
imet | -0.018 |
vered | -0.016 |
brow | -0.016 |
ance | -0.016 |
Token | A |
Token ID | 317 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
seek | +0.028 |
quiet | +0.028 |
late | +0.028 |
aky | +0.026 |
quieter | +0.025 |
Neg logit contributions | |
Pe | -0.019 |
imet | -0.019 |
pen | -0.019 |
Ow | -0.018 |
ian | -0.018 |
Token | very |
Token ID | 845 |
Feature activation | -3.17e+00 |
Pos logit contributions | |
late | +0.036 |
aking | +0.030 |
They | +0.030 |
seek | +0.030 |
Lisa | +0.030 |
Neg logit contributions | |
penn | -0.037 |
ian | -0.035 |
pen | -0.035 |
imet | -0.034 |
stomach | -0.034 |
Token | small |
Token ID | 1402 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
They | +0.083 |
late | +0.080 |
Lisa | +0.073 |
apes | +0.073 |
Happy | +0.070 |
Neg logit contributions | |
penn | -0.097 |
stomach | -0.090 |
tum | -0.089 |
ome | -0.087 |
trek | -0.086 |
Token | bird |
Token ID | 6512 |
Feature activation | -5.73e-01 |
Pos logit contributions | |
late | +0.036 |
seek | +0.031 |
quiet | +0.031 |
Lisa | +0.030 |
Happy | +0.030 |
Neg logit contributions | |
penn | -0.027 |
pen | -0.026 |
ian | -0.026 |
imet | -0.025 |
ome | -0.024 |
Token | saw |
Token ID | 2497 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
late | +0.024 |
seek | +0.021 |
quiet | +0.021 |
Lisa | +0.021 |
Lisa | +0.021 |
Neg logit contributions | |
pen | -0.011 |
imet | -0.010 |
ian | -0.010 |
penn | -0.010 |
vered | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
late | +0.033 |
seek | +0.031 |
quiet | +0.029 |
Lisa | +0.026 |
aky | +0.026 |
Neg logit contributions | |
pen | -0.031 |
penn | -0.031 |
imet | -0.031 |
Ow | -0.030 |
ian | -0.030 |
Token | start |
Token ID | 923 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
penn | +0.003 |
imet | +0.003 |
pen | +0.003 |
ian | +0.003 |
Ow | +0.002 |
Neg logit contributions | |
late | -0.003 |
seek | -0.003 |
quiet | -0.003 |
Lisa | -0.003 |
Lisa | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -4.46e-02 |
Pos logit contributions | |
late | +0.003 |
quiet | +0.002 |
seek | +0.002 |
Lisa | +0.002 |
Lisa | +0.002 |
Neg logit contributions | |
penn | -0.004 |
imet | -0.004 |
pen | -0.004 |
ian | -0.004 |
Ow | -0.004 |
Token | But |
Token ID | 1537 |
Feature activation | +4.36e-01 |
Pos logit contributions | |
late | +0.001 |
seek | +0.001 |
quiet | +0.001 |
Lisa | +0.001 |
too | +0.001 |
Neg logit contributions | |
pen | -0.002 |
penn | -0.002 |
imet | -0.002 |
Ow | -0.002 |
ian | -0.002 |
Token | then |
Token ID | 788 |
Feature activation | -6.95e-01 |
Pos logit contributions | |
penn | +0.014 |
imet | +0.013 |
pen | +0.013 |
Pe | +0.013 |
Ow | +0.013 |
Neg logit contributions | |
late | -0.013 |
seek | -0.011 |
quiet | -0.011 |
lic | -0.010 |
aky | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
late | +0.022 |
seek | +0.019 |
quiet | +0.019 |
Lisa | +0.018 |
Lisa | +0.017 |
Neg logit contributions | |
penn | -0.021 |
imet | -0.020 |
pen | -0.020 |
ian | -0.020 |
Ow | -0.019 |
Token | they |
Token ID | 484 |
Feature activation | -2.36e+00 |
Pos logit contributions | |
late | +0.038 |
seek | +0.033 |
quiet | +0.033 |
aky | +0.031 |
impressed | +0.031 |
Neg logit contributions | |
penn | -0.037 |
imet | -0.036 |
Ow | -0.035 |
Pe | -0.035 |
pen | -0.034 |
Token | both |
Token ID | 1111 |
Feature activation | -3.15e+00 |
Pos logit contributions | |
late | +0.077 |
seek | +0.067 |
quiet | +0.064 |
Lisa | +0.061 |
Lisa | +0.060 |
Neg logit contributions | |
pen | -0.068 |
penn | -0.067 |
imet | -0.067 |
ian | -0.065 |
Ow | -0.064 |
Token | want |
Token ID | 765 |
Feature activation | -4.48e-01 |
Pos logit contributions | |
late | +0.105 |
seek | +0.095 |
quiet | +0.088 |
Lisa | +0.084 |
Lisa | +0.084 |
Neg logit contributions | |
penn | -0.087 |
pen | -0.086 |
imet | -0.086 |
ian | -0.083 |
Ow | -0.081 |
Token | to |
Token ID | 284 |
Feature activation | -7.29e-01 |
Pos logit contributions | |
late | +0.014 |
seek | +0.013 |
quiet | +0.012 |
aky | +0.011 |
Lisa | +0.011 |
Neg logit contributions | |
penn | -0.014 |
imet | -0.013 |
pen | -0.013 |
Ow | -0.013 |
Pe | -0.012 |
Token | color |
Token ID | 3124 |
Feature activation | -3.72e-02 |
Pos logit contributions | |
late | +0.029 |
seek | +0.024 |
quiet | +0.024 |
Lisa | +0.023 |
Lisa | +0.023 |
Neg logit contributions | |
pen | -0.017 |
penn | -0.016 |
imet | -0.016 |
ian | -0.016 |
Ow | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
seek | +0.001 |
late | +0.001 |
persuasive | +0.001 |
pleased | +0.001 |
aky | +0.001 |
Neg logit contributions | |
penn | -0.001 |
Pe | -0.001 |
ware | -0.001 |
anyone | -0.001 |
Ow | -0.001 |
Token | same |
Token ID | 976 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
late | +0.001 |
seek | +0.001 |
Lisa | +0.001 |
lic | +0.000 |
ores | +0.000 |
Neg logit contributions | |
penn | -0.001 |
imet | -0.001 |
Ow | -0.001 |
ian | -0.001 |
pen | -0.001 |
Token | What |
Token ID | 2061 |
Feature activation | +4.91e-01 |
Pos logit contributions | |
late | +0.035 |
seek | +0.031 |
quiet | +0.030 |
pleased | +0.030 |
lic | +0.030 |
Neg logit contributions | |
Ow | -0.036 |
penn | -0.035 |
yle | -0.034 |
trigger | -0.033 |
Pe | -0.033 |
Token | did |
Token ID | 750 |
Feature activation | -8.69e-01 |
Pos logit contributions | |
penn | +0.012 |
imet | +0.012 |
pen | +0.012 |
ian | +0.011 |
Ow | +0.011 |
Neg logit contributions | |
late | -0.018 |
quiet | -0.016 |
seek | -0.016 |
Lisa | -0.016 |
Lisa | -0.015 |
Token | you |
Token ID | 345 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
late | +0.025 |
quiet | +0.022 |
seek | +0.022 |
Lisa | +0.021 |
too | +0.019 |
Neg logit contributions | |
penn | -0.029 |
pen | -0.029 |
imet | -0.028 |
ian | -0.028 |
Ow | -0.027 |
Token | do |
Token ID | 466 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
late | +0.036 |
seek | +0.031 |
Lisa | +0.030 |
Lisa | +0.030 |
Happy | +0.029 |
Neg logit contributions | |
Ow | -0.034 |
penn | -0.034 |
pen | -0.034 |
ian | -0.034 |
Pe | -0.032 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
late | +0.051 |
aky | +0.045 |
quiet | +0.045 |
Lisa | +0.044 |
pleased | +0.043 |
Neg logit contributions | |
pen | -0.066 |
penn | -0.065 |
ian | -0.064 |
imet | -0.063 |
ome | -0.060 |
Token | she |
Token ID | 673 |
Feature activation | -3.12e+00 |
Pos logit contributions | |
seek | +0.005 |
aking | +0.005 |
lic | +0.005 |
aky | +0.005 |
late | +0.005 |
Neg logit contributions | |
Pe | -0.003 |
penn | -0.003 |
Twe | -0.003 |
anyone | -0.003 |
tum | -0.002 |
Token | asks |
Token ID | 7893 |
Feature activation | -3.95e-01 |
Pos logit contributions | |
late | +0.101 |
seek | +0.092 |
quiet | +0.088 |
Lisa | +0.083 |
Lisa | +0.081 |
Neg logit contributions | |
penn | -0.089 |
imet | -0.087 |
pen | -0.087 |
ian | -0.087 |
Ow | -0.084 |
Token | . |
Token ID | 13 |
Feature activation | -4.52e-01 |
Pos logit contributions | |
late | +0.008 |
Lisa | +0.007 |
quiet | +0.006 |
too | +0.006 |
seek | +0.005 |
Neg logit contributions | |
imet | -0.017 |
penn | -0.017 |
pen | -0.016 |
ian | -0.016 |
Ow | -0.016 |
Token | " |
Token ID | 366 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
late | +0.013 |
seek | +0.012 |
quiet | +0.012 |
Lisa | +0.011 |
aky | +0.010 |
Neg logit contributions | |
imet | -0.014 |
penn | -0.014 |
pen | -0.013 |
Ow | -0.013 |
ian | -0.013 |
Token | That |
Token ID | 2504 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
late | +0.038 |
seek | +0.037 |
quiet | +0.036 |
Lisa | +0.034 |
too | +0.033 |
Neg logit contributions | |
pen | -0.047 |
Ow | -0.046 |
penn | -0.046 |
yle | -0.045 |
imet | -0.044 |
Token | was |
Token ID | 373 |
Feature activation | -3.21e-01 |
Pos logit contributions | |
late | +0.054 |
seek | +0.050 |
quiet | +0.049 |
Lisa | +0.047 |
pleased | +0.046 |
Neg logit contributions | |
penn | -0.031 |
imet | -0.030 |
ian | -0.030 |
Ow | -0.030 |
pen | -0.029 |
Token | said |
Token ID | 531 |
Feature activation | +3.44e-01 |
Pos logit contributions | |
late | +0.018 |
seek | +0.015 |
quiet | +0.015 |
Lisa | +0.014 |
Happy | +0.014 |
Neg logit contributions | |
pen | -0.022 |
ian | -0.022 |
Pe | -0.021 |
imet | -0.021 |
penn | -0.021 |
Token | , |
Token ID | 11 |
Feature activation | +2.85e-02 |
Pos logit contributions | |
penn | +0.009 |
imet | +0.009 |
pen | +0.009 |
ian | +0.009 |
brow | +0.009 |
Neg logit contributions | |
late | -0.011 |
seek | -0.010 |
Lisa | -0.010 |
quiet | -0.010 |
Happy | -0.009 |
Token | pointing |
Token ID | 10609 |
Feature activation | +6.00e-01 |
Pos logit contributions | |
penn | +0.001 |
imet | +0.001 |
pen | +0.001 |
ian | +0.001 |
Ow | +0.001 |
Neg logit contributions | |
late | -0.001 |
seek | -0.001 |
quiet | -0.001 |
Lisa | -0.001 |
Lisa | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | -8.10e-01 |
Pos logit contributions | |
Ow | +0.012 |
Pe | +0.012 |
anybody | +0.011 |
imet | +0.011 |
anyone | +0.011 |
Neg logit contributions | |
seek | -0.023 |
late | -0.022 |
Lisa | -0.021 |
Happy | -0.020 |
lic | -0.020 |
Token | a |
Token ID | 257 |
Feature activation | -6.85e-01 |
Pos logit contributions | |
late | +0.027 |
aky | +0.025 |
aking | +0.025 |
lic | +0.024 |
seek | +0.024 |
Neg logit contributions | |
anyone | -0.021 |
anybody | -0.021 |
Pe | -0.020 |
growth | -0.020 |
Ow | -0.020 |
Token | very |
Token ID | 845 |
Feature activation | -3.06e+00 |
Pos logit contributions | |
late | +0.022 |
aking | +0.019 |
Lisa | +0.019 |
ores | +0.019 |
They | +0.019 |
Neg logit contributions | |
penn | -0.018 |
stomach | -0.018 |
je | -0.018 |
pen | -0.017 |
trigger | -0.017 |
Token | big |
Token ID | 1263 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
They | +0.078 |
late | +0.077 |
Lisa | +0.076 |
Lisa | +0.074 |
circles | +0.073 |
Neg logit contributions | |
penn | -0.091 |
stomach | -0.085 |
tum | -0.084 |
penny | -0.082 |
Twe | -0.082 |
Token | tree |
Token ID | 5509 |
Feature activation | -4.26e-01 |
Pos logit contributions | |
late | +0.040 |
Lisa | +0.035 |
They | +0.035 |
Happy | +0.035 |
lic | +0.034 |
Neg logit contributions | |
penn | -0.025 |
pen | -0.024 |
ian | -0.024 |
stomach | -0.023 |
trigger | -0.023 |
Token | with |
Token ID | 351 |
Feature activation | +2.59e-01 |
Pos logit contributions | |
late | +0.013 |
Lisa | +0.012 |
seek | +0.012 |
quiet | +0.011 |
Lisa | +0.011 |
Neg logit contributions | |
pen | -0.012 |
penn | -0.012 |
imet | -0.012 |
bank | -0.012 |
ome | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | -8.71e-01 |
Pos logit contributions | |
penn | +0.008 |
Ow | +0.008 |
pen | +0.008 |
Pe | +0.008 |
imet | +0.008 |
Neg logit contributions | |
late | -0.008 |
seek | -0.007 |
Lisa | -0.006 |
quiet | -0.006 |
Lisa | -0.006 |
Token | feather |
Token ID | 17972 |
Feature activation | -4.60e-01 |
Pos logit contributions | |
late | +0.028 |
ores | +0.024 |
Lisa | +0.024 |
Lisa | +0.024 |
They | +0.023 |
Neg logit contributions | |
stomach | -0.024 |
penn | -0.024 |
tum | -0.023 |
trigger | -0.023 |
Ow | -0.023 |
Token | " |
Token ID | 1 |
Feature activation | -8.82e-01 |
Pos logit contributions | |
late | +0.001 |
seek | +0.001 |
quiet | +0.001 |
Lisa | +0.001 |
too | +0.001 |
Neg logit contributions | |
pen | -0.002 |
penn | -0.002 |
imet | -0.002 |
ian | -0.002 |
ome | -0.001 |
Token | What |
Token ID | 2061 |
Feature activation | +9.44e-01 |
Pos logit contributions | |
late | +0.029 |
seek | +0.027 |
Lisa | +0.027 |
pleased | +0.026 |
quiet | +0.026 |
Neg logit contributions | |
penn | -0.021 |
Ow | -0.021 |
pen | -0.020 |
imet | -0.020 |
Whoever | -0.020 |
Token | is |
Token ID | 318 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
penn | +0.027 |
pen | +0.026 |
imet | +0.026 |
ian | +0.025 |
Ow | +0.025 |
Neg logit contributions | |
late | -0.031 |
seek | -0.028 |
quiet | -0.028 |
Lisa | -0.026 |
Lisa | -0.025 |
Token | famous |
Token ID | 5863 |
Feature activation | -9.71e-01 |
Pos logit contributions | |
late | +0.039 |
quiet | +0.035 |
seek | +0.034 |
Lisa | +0.033 |
too | +0.031 |
Neg logit contributions | |
pen | -0.039 |
penn | -0.039 |
imet | -0.039 |
ian | -0.038 |
Ow | -0.037 |
Token | ?" |
Token ID | 1701 |
Feature activation | +4.99e-02 |
Pos logit contributions | |
late | +0.027 |
Lisa | +0.026 |
seek | +0.024 |
quiet | +0.024 |
Maria | +0.024 |
Neg logit contributions | |
ian | -0.031 |
penn | -0.031 |
imet | -0.030 |
pen | -0.030 |
ome | -0.028 |
Token | she |
Token ID | 673 |
Feature activation | -3.04e+00 |
Pos logit contributions | |
imet | +0.002 |
pen | +0.002 |
penn | +0.002 |
ian | +0.002 |
trigger | +0.002 |
Neg logit contributions | |
late | -0.001 |
quiet | -0.001 |
Lisa | -0.001 |
seek | -0.001 |
too | -0.001 |
Token | asks |
Token ID | 7893 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
late | +0.111 |
seek | +0.104 |
quiet | +0.094 |
lic | +0.093 |
Happy | +0.091 |
Neg logit contributions | |
ian | -0.073 |
penn | -0.071 |
pen | -0.071 |
imet | -0.070 |
Ow | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
late | +0.002 |
Lisa | +0.002 |
too | +0.002 |
quiet | +0.002 |
Jill | +0.001 |
Neg logit contributions | |
imet | -0.006 |
penn | -0.006 |
ian | -0.006 |
trek | -0.006 |
uable | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +4.37e-02 |
Pos logit contributions | |
late | +0.004 |
quiet | +0.003 |
seek | +0.003 |
Lisa | +0.003 |
too | +0.003 |
Neg logit contributions | |
penn | -0.006 |
pen | -0.006 |
imet | -0.006 |
ian | -0.006 |
Ow | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +3.81e-02 |
Pos logit contributions | |
penn | +0.002 |
imet | +0.002 |
pen | +0.002 |
ian | +0.002 |
ome | +0.002 |
Neg logit contributions | |
late | -0.001 |
quiet | -0.001 |
Lisa | -0.001 |
seek | -0.001 |
Lisa | -0.001 |
Token | Ben |
Token ID | 11696 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
penn | +0.001 |
pen | +0.001 |
imet | +0.001 |
ian | +0.001 |
Ow | +0.001 |
Neg logit contributions | |
late | -0.001 |
seek | -0.001 |
quiet | -0.001 |
Lisa | -0.001 |
too | -0.001 |
Token | something |
Token ID | 1223 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
late | +0.043 |
quiet | +0.036 |
seek | +0.036 |
Lisa | +0.034 |
aky | +0.034 |
Neg logit contributions | |
penn | -0.036 |
imet | -0.035 |
Pe | -0.035 |
pen | -0.034 |
ian | -0.034 |
Token | scary |
Token ID | 14343 |
Feature activation | -2.89e-01 |
Pos logit contributions | |
penn | +0.007 |
imet | +0.007 |
pen | +0.007 |
ian | +0.007 |
Ow | +0.007 |
Neg logit contributions | |
late | -0.005 |
seek | -0.005 |
quiet | -0.005 |
Lisa | -0.004 |
Lisa | -0.004 |
Token | happened |
Token ID | 3022 |
Feature activation | -5.34e-01 |
Pos logit contributions | |
late | +0.011 |
quiet | +0.010 |
seek | +0.010 |
Lisa | +0.010 |
too | +0.010 |
Neg logit contributions | |
penn | -0.006 |
imet | -0.006 |
pen | -0.006 |
ian | -0.006 |
Ow | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
late | +0.017 |
seek | +0.016 |
Lisa | +0.016 |
quiet | +0.016 |
Happy | +0.015 |
Neg logit contributions | |
imet | -0.014 |
pen | -0.014 |
brow | -0.013 |
vered | -0.013 |
ian | -0.013 |
Token | A |
Token ID | 317 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
late | +0.044 |
seek | +0.041 |
quiet | +0.040 |
Lisa | +0.037 |
Lisa | +0.036 |
Neg logit contributions | |
imet | -0.032 |
Ow | -0.032 |
Pe | -0.032 |
penn | -0.031 |
pen | -0.030 |
Token | really |
Token ID | 1107 |
Feature activation | -3.03e+00 |
Pos logit contributions | |
late | +0.043 |
Lisa | +0.036 |
Lisa | +0.036 |
seek | +0.036 |
Happy | +0.034 |
Neg logit contributions | |
penn | -0.045 |
ian | -0.042 |
pen | -0.041 |
trigger | -0.041 |
imet | -0.041 |
Token | big |
Token ID | 1263 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
late | +0.076 |
They | +0.074 |
amer | +0.074 |
apes | +0.073 |
Lisa | +0.072 |
Neg logit contributions | |
tum | -0.083 |
eyel | -0.083 |
penn | -0.081 |
rest | -0.081 |
stomach | -0.081 |
Token | spider |
Token ID | 19230 |
Feature activation | -9.83e-01 |
Pos logit contributions | |
late | +0.045 |
seek | +0.039 |
quiet | +0.039 |
Lisa | +0.038 |
Happy | +0.038 |
Neg logit contributions | |
penn | -0.030 |
pen | -0.029 |
imet | -0.029 |
ian | -0.028 |
trigger | -0.027 |
Token | appeared |
Token ID | 4120 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
late | +0.038 |
quiet | +0.033 |
seek | +0.032 |
Lisa | +0.032 |
Lisa | +0.032 |
Neg logit contributions | |
pen | -0.022 |
imet | -0.021 |
ian | -0.021 |
penn | -0.020 |
ome | -0.020 |
Token | ! |
Token ID | 0 |
Feature activation | -6.41e-01 |
Pos logit contributions | |
late | +0.034 |
seek | +0.031 |
quiet | +0.030 |
Lisa | +0.029 |
Lisa | +0.028 |
Neg logit contributions | |
pen | -0.038 |
imet | -0.038 |
penn | -0.037 |
brow | -0.037 |
ian | -0.036 |
Token | It |
Token ID | 632 |
Feature activation | -8.96e-02 |
Pos logit contributions | |
late | +0.021 |
seek | +0.019 |
quiet | +0.019 |
aky | +0.017 |
Lisa | +0.017 |
Neg logit contributions | |
imet | -0.018 |
Pe | -0.018 |
Ow | -0.018 |
penn | -0.017 |
pen | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
late | +0.008 |
quiet | +0.007 |
seek | +0.007 |
Lisa | +0.007 |
too | +0.007 |
Neg logit contributions | |
penn | -0.008 |
pen | -0.008 |
ian | -0.008 |
imet | -0.008 |
yle | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
late | +0.003 |
seek | +0.003 |
quiet | +0.003 |
Lisa | +0.003 |
too | +0.003 |
Neg logit contributions | |
pen | -0.003 |
imet | -0.003 |
Ow | -0.003 |
penn | -0.003 |
Pe | -0.003 |
Token | Suddenly |
Token ID | 38582 |
Feature activation | -9.85e-01 |
Pos logit contributions | |
late | +0.007 |
Lisa | +0.007 |
quiet | +0.007 |
seek | +0.006 |
too | +0.006 |
Neg logit contributions | |
Mont | -0.006 |
pen | -0.006 |
Pe | -0.006 |
trek | -0.006 |
Ow | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
late | +0.027 |
quiet | +0.023 |
seek | +0.023 |
too | +0.022 |
Lisa | +0.022 |
Neg logit contributions | |
pen | -0.035 |
penn | -0.034 |
imet | -0.034 |
ian | -0.034 |
yle | -0.034 |
Token | a |
Token ID | 257 |
Feature activation | -6.93e-01 |
Pos logit contributions | |
late | +0.071 |
seek | +0.064 |
quiet | +0.064 |
Lisa | +0.061 |
Lisa | +0.059 |
Neg logit contributions | |
penn | -0.056 |
imet | -0.054 |
pen | -0.053 |
ian | -0.052 |
Ow | -0.052 |
Token | very |
Token ID | 845 |
Feature activation | -2.99e+00 |
Pos logit contributions | |
late | +0.019 |
seek | +0.016 |
Lisa | +0.015 |
Lisa | +0.015 |
aking | +0.014 |
Neg logit contributions | |
penn | -0.022 |
ian | -0.022 |
pen | -0.022 |
imet | -0.021 |
trigger | -0.021 |
Token | angry |
Token ID | 7954 |
Feature activation | -8.53e-01 |
Pos logit contributions | |
late | +0.070 |
They | +0.069 |
Lisa | +0.067 |
circles | +0.066 |
Lisa | +0.065 |
Neg logit contributions | |
penn | -0.097 |
bank | -0.088 |
trigger | -0.088 |
ome | -0.086 |
penny | -0.086 |
Token | voice |
Token ID | 3809 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
late | +0.025 |
quiet | +0.022 |
seek | +0.022 |
Lisa | +0.021 |
too | +0.020 |
Neg logit contributions | |
penn | -0.027 |
imet | -0.027 |
pen | -0.027 |
ian | -0.026 |
vered | -0.025 |
Token | stopped |
Token ID | 5025 |
Feature activation | -8.63e-01 |
Pos logit contributions | |
late | +0.004 |
seek | +0.003 |
quiet | +0.003 |
Lisa | +0.003 |
Lisa | +0.003 |
Neg logit contributions | |
penn | -0.004 |
pen | -0.004 |
imet | -0.004 |
ian | -0.004 |
Pe | -0.004 |
Token | him |
Token ID | 683 |
Feature activation | -2.84e-01 |
Pos logit contributions | |
late | +0.023 |
seek | +0.023 |
Lisa | +0.022 |
quiet | +0.021 |
Lisa | +0.021 |
Neg logit contributions | |
pen | -0.027 |
Pe | -0.027 |
penn | -0.027 |
imet | -0.027 |
ian | -0.026 |
Token | in |
Token ID | 287 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
late | +0.007 |
seek | +0.006 |
quiet | +0.006 |
Lisa | +0.006 |
Lisa | +0.006 |
Neg logit contributions | |
pen | -0.010 |
penn | -0.010 |
imet | -0.010 |
ian | -0.010 |
brow | -0.009 |
Token | big |
Token ID | 1263 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
late | +0.039 |
seek | +0.035 |
quiet | +0.034 |
Lisa | +0.033 |
Lisa | +0.033 |
Neg logit contributions | |
penn | -0.026 |
imet | -0.025 |
pen | -0.024 |
trigger | -0.024 |
ian | -0.023 |
Token | red |
Token ID | 2266 |
Feature activation | +5.05e-02 |
Pos logit contributions | |
late | +0.047 |
seek | +0.041 |
quiet | +0.041 |
Lisa | +0.039 |
Lisa | +0.039 |
Neg logit contributions | |
penn | -0.033 |
imet | -0.032 |
pen | -0.032 |
ian | -0.031 |
trigger | -0.030 |
Token | car |
Token ID | 1097 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
imet | +0.001 |
penn | +0.001 |
unn | +0.001 |
pen | +0.001 |
ian | +0.001 |
Neg logit contributions | |
late | -0.002 |
quiet | -0.001 |
seek | -0.001 |
too | -0.001 |
Lisa | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -9.99e-01 |
Pos logit contributions | |
late | +0.049 |
seek | +0.047 |
pleased | +0.045 |
Lisa | +0.044 |
pans | +0.043 |
Neg logit contributions | |
pen | -0.039 |
imet | -0.038 |
bank | -0.037 |
penn | -0.036 |
trigger | -0.036 |
Token | They |
Token ID | 1119 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
late | +0.033 |
seek | +0.032 |
aky | +0.029 |
quiet | +0.029 |
impressed | +0.027 |
Neg logit contributions | |
Ow | -0.028 |
imet | -0.026 |
Pe | -0.026 |
penn | -0.026 |
brow | -0.026 |
Token | both |
Token ID | 1111 |
Feature activation | -2.99e+00 |
Pos logit contributions | |
late | +0.062 |
seek | +0.055 |
quiet | +0.055 |
Lisa | +0.052 |
Lisa | +0.051 |
Neg logit contributions | |
penn | -0.049 |
pen | -0.048 |
imet | -0.048 |
ian | -0.046 |
Ow | -0.045 |
Token | want |
Token ID | 765 |
Feature activation | -4.85e-01 |
Pos logit contributions | |
late | +0.096 |
seek | +0.085 |
quiet | +0.083 |
Lisa | +0.080 |
Lisa | +0.077 |
Neg logit contributions | |
penn | -0.086 |
pen | -0.086 |
imet | -0.085 |
ian | -0.082 |
Ow | -0.081 |
Token | to |
Token ID | 284 |
Feature activation | -4.37e-01 |
Pos logit contributions | |
late | +0.014 |
seek | +0.013 |
quiet | +0.012 |
aky | +0.011 |
Lisa | +0.011 |
Neg logit contributions | |
penn | -0.016 |
imet | -0.015 |
pen | -0.015 |
Ow | -0.015 |
ian | -0.014 |
Token | play |
Token ID | 711 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
late | +0.014 |
quiet | +0.013 |
seek | +0.013 |
Lisa | +0.012 |
Lisa | +0.012 |
Neg logit contributions | |
penn | -0.012 |
imet | -0.012 |
pen | -0.012 |
ian | -0.012 |
Ow | -0.011 |
Token | with |
Token ID | 351 |
Feature activation | -3.33e-02 |
Pos logit contributions | |
late | +0.042 |
seek | +0.039 |
lic | +0.037 |
Lisa | +0.037 |
aky | +0.037 |
Neg logit contributions | |
pen | -0.047 |
penn | -0.046 |
bank | -0.042 |
imet | -0.042 |
trigger | -0.041 |
Token | it |
Token ID | 340 |
Feature activation | -8.24e-01 |
Pos logit contributions | |
late | +0.001 |
quiet | +0.001 |
seek | +0.001 |
Lisa | +0.001 |
too | +0.001 |
Neg logit contributions | |
imet | -0.001 |
penn | -0.001 |
pen | -0.001 |
ian | -0.001 |
Ow | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
late | +0.025 |
quiet | +0.022 |
seek | +0.021 |
Lisa | +0.020 |
too | +0.019 |
Neg logit contributions | |
penn | -0.037 |
imet | -0.037 |
pen | -0.036 |
ian | -0.036 |
Ow | -0.035 |
Token | slide |
Token ID | 10649 |
Feature activation | +8.49e-02 |
Pos logit contributions | |
penn | +0.005 |
imet | +0.005 |
ome | +0.005 |
unn | +0.005 |
pen | +0.004 |
Neg logit contributions | |
quiet | -0.005 |
late | -0.005 |
seek | -0.005 |
Lisa | -0.004 |
too | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
penn | +0.002 |
imet | +0.002 |
Pe | +0.002 |
Twe | +0.002 |
pen | +0.002 |
Neg logit contributions | |
late | -0.003 |
seek | -0.003 |
Happy | -0.003 |
aky | -0.003 |
lic | -0.003 |
Token | In |
Token ID | 554 |
Feature activation | -9.74e-01 |
Pos logit contributions | |
late | +0.006 |
quiet | +0.005 |
Lisa | +0.005 |
seek | +0.005 |
too | +0.005 |
Neg logit contributions | |
penn | -0.007 |
pen | -0.007 |
unn | -0.007 |
ian | -0.007 |
yle | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
late | +0.025 |
seek | +0.022 |
quiet | +0.022 |
Lisa | +0.020 |
Lisa | +0.019 |
Neg logit contributions | |
penn | -0.035 |
imet | -0.034 |
pen | -0.034 |
ian | -0.033 |
Ow | -0.032 |
Token | few |
Token ID | 1178 |
Feature activation | -2.99e+00 |
Pos logit contributions | |
late | +0.055 |
seek | +0.045 |
Lisa | +0.043 |
ores | +0.041 |
akers | +0.041 |
Neg logit contributions | |
pen | -0.056 |
trek | -0.053 |
ian | -0.052 |
brow | -0.052 |
penn | -0.051 |
Token | moments |
Token ID | 7188 |
Feature activation | +8.68e-03 |
Pos logit contributions | |
late | +0.100 |
seek | +0.082 |
quiet | +0.078 |
Lisa | +0.075 |
lic | +0.075 |
Neg logit contributions | |
penn | -0.091 |
pen | -0.085 |
vered | -0.085 |
imet | -0.082 |
ian | -0.081 |
Token | , |
Token ID | 11 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
penn | +0.000 |
imet | +0.000 |
pen | +0.000 |
ian | +0.000 |
Ow | +0.000 |
Neg logit contributions | |
late | -0.000 |
quiet | -0.000 |
seek | -0.000 |
Lisa | -0.000 |
too | -0.000 |
Token | he |
Token ID | 339 |
Feature activation | -4.13e-01 |
Pos logit contributions | |
late | +0.050 |
quiet | +0.044 |
seek | +0.044 |
Lisa | +0.042 |
Lisa | +0.040 |
Neg logit contributions | |
penn | -0.051 |
imet | -0.051 |
pen | -0.050 |
ian | -0.049 |
Ow | -0.048 |
Token | had |
Token ID | 550 |
Feature activation | -7.56e-01 |
Pos logit contributions | |
late | +0.014 |
seek | +0.013 |
quiet | +0.012 |
Lisa | +0.012 |
Lisa | +0.012 |
Neg logit contributions | |
pen | -0.011 |
imet | -0.011 |
penn | -0.011 |
ian | -0.011 |
brow | -0.010 |
Token | slid |
Token ID | 27803 |
Feature activation | -3.91e-01 |
Pos logit contributions | |
late | +0.024 |
seek | +0.022 |
Lisa | +0.021 |
Lisa | +0.021 |
aky | +0.021 |
Neg logit contributions | |
penn | -0.020 |
imet | -0.020 |
Pe | -0.020 |
pen | -0.020 |
Ow | -0.019 |
Token | it |
Token ID | 340 |
Feature activation | -6.13e-02 |
Pos logit contributions | |
late | +0.007 |
quiet | +0.006 |
seek | +0.006 |
Lisa | +0.006 |
too | +0.006 |
Neg logit contributions | |
penn | -0.008 |
imet | -0.008 |
pen | -0.008 |
ian | -0.007 |
Ow | -0.007 |
Token | was |
Token ID | 373 |
Feature activation | +9.25e-01 |
Pos logit contributions | |
late | +0.002 |
seek | +0.002 |
quiet | +0.002 |
Lisa | +0.002 |
Lisa | +0.002 |
Neg logit contributions | |
pen | -0.001 |
penn | -0.001 |
imet | -0.001 |
ian | -0.001 |
brow | -0.001 |
Token | written |
Token ID | 3194 |
Feature activation | +9.85e-02 |
Pos logit contributions | |
penn | +0.026 |
Pe | +0.026 |
brow | +0.025 |
pen | +0.024 |
eyel | +0.024 |
Neg logit contributions | |
late | -0.028 |
seek | -0.026 |
Lisa | -0.026 |
Happy | -0.026 |
aking | -0.024 |
Token | in |
Token ID | 287 |
Feature activation | -5.84e-01 |
Pos logit contributions | |
pen | +0.003 |
penn | +0.003 |
imet | +0.003 |
ian | +0.003 |
brow | +0.003 |
Neg logit contributions | |
late | -0.003 |
seek | -0.003 |
quiet | -0.002 |
Lisa | -0.002 |
Lisa | -0.002 |
Token | a |
Token ID | 257 |
Feature activation | -7.44e-01 |
Pos logit contributions | |
late | +0.018 |
seek | +0.016 |
lic | +0.015 |
quiet | +0.015 |
aking | +0.014 |
Neg logit contributions | |
penn | -0.018 |
pen | -0.017 |
growth | -0.017 |
Ow | -0.017 |
Pe | -0.017 |
Token | very |
Token ID | 845 |
Feature activation | -2.98e+00 |
Pos logit contributions | |
late | +0.023 |
ores | +0.019 |
seek | +0.019 |
Lisa | +0.019 |
They | +0.018 |
Neg logit contributions | |
penn | -0.022 |
pen | -0.022 |
ian | -0.021 |
trigger | -0.020 |
brow | -0.020 |
Token | fancy |
Token ID | 14996 |
Feature activation | -7.57e-01 |
Pos logit contributions | |
late | +0.084 |
They | +0.072 |
seek | +0.070 |
Lisa | +0.070 |
Lisa | +0.069 |
Neg logit contributions | |
penn | -0.097 |
stomach | -0.086 |
tum | -0.085 |
brow | -0.084 |
imet | -0.084 |
Token | handwriting |
Token ID | 44396 |
Feature activation | -3.79e-01 |
Pos logit contributions | |
late | +0.028 |
seek | +0.023 |
They | +0.023 |
quiet | +0.023 |
lic | +0.023 |
Neg logit contributions | |
pen | -0.019 |
penn | -0.019 |
ian | -0.019 |
imet | -0.018 |
brow | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
late | +0.012 |
quiet | +0.010 |
seek | +0.010 |
Lisa | +0.010 |
Lisa | +0.010 |
Neg logit contributions | |
pen | -0.012 |
ian | -0.011 |
penn | -0.011 |
imet | -0.011 |
ome | -0.011 |
Token | The |
Token ID | 383 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
late | +0.010 |
Lisa | +0.009 |
quiet | +0.008 |
seek | +0.008 |
Lisa | +0.008 |
Neg logit contributions | |
penn | -0.010 |
pen | -0.010 |
imet | -0.009 |
trigger | -0.009 |
ian | -0.009 |
Token | poem |
Token ID | 21247 |
Feature activation | -7.67e-01 |
Pos logit contributions | |
penn | +0.006 |
ian | +0.006 |
pen | +0.006 |
imet | +0.005 |
brow | +0.005 |
Neg logit contributions | |
late | -0.008 |
seek | -0.007 |
quiet | -0.007 |
Lisa | -0.007 |
Lisa | -0.006 |
Token | What |
Token ID | 2061 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
late | +0.015 |
quiet | +0.014 |
too | +0.014 |
seek | +0.013 |
awesome | +0.013 |
Neg logit contributions | |
penn | -0.014 |
Ow | -0.013 |
yle | -0.013 |
Pe | -0.012 |
cash | -0.012 |
Token | did |
Token ID | 750 |
Feature activation | -4.90e-01 |
Pos logit contributions | |
penn | +0.026 |
imet | +0.026 |
pen | +0.026 |
ian | +0.025 |
Ow | +0.024 |
Neg logit contributions | |
late | -0.040 |
seek | -0.036 |
quiet | -0.036 |
Lisa | -0.034 |
Lisa | -0.033 |
Token | you |
Token ID | 345 |
Feature activation | -8.08e-01 |
Pos logit contributions | |
late | +0.013 |
quiet | +0.012 |
seek | +0.012 |
Lisa | +0.011 |
too | +0.010 |
Neg logit contributions | |
pen | -0.017 |
imet | -0.017 |
penn | -0.017 |
ian | -0.016 |
brow | -0.016 |
Token | do |
Token ID | 466 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
late | +0.031 |
seek | +0.026 |
Lisa | +0.026 |
Lisa | +0.026 |
Happy | +0.026 |
Neg logit contributions | |
Ow | -0.018 |
pen | -0.018 |
ian | -0.017 |
penn | -0.017 |
uf | -0.017 |
Token | ?" |
Token ID | 1701 |
Feature activation | -3.36e-01 |
Pos logit contributions | |
late | +0.046 |
quiet | +0.041 |
aky | +0.040 |
pleased | +0.040 |
Lisa | +0.040 |
Neg logit contributions | |
pen | -0.058 |
penn | -0.057 |
ian | -0.056 |
imet | -0.055 |
ome | -0.053 |
Token | she |
Token ID | 673 |
Feature activation | -2.98e+00 |
Pos logit contributions | |
late | +0.011 |
seek | +0.011 |
quiet | +0.010 |
lic | +0.010 |
aky | +0.009 |
Neg logit contributions | |
penn | -0.009 |
Pe | -0.008 |
Ow | -0.008 |
imet | -0.008 |
ian | -0.008 |
Token | asks |
Token ID | 7893 |
Feature activation | -2.60e-01 |
Pos logit contributions | |
late | +0.098 |
seek | +0.091 |
quiet | +0.085 |
Lisa | +0.079 |
Happy | +0.079 |
Neg logit contributions | |
penn | -0.082 |
ian | -0.080 |
pen | -0.079 |
imet | -0.078 |
Ow | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | -3.65e-01 |
Pos logit contributions | |
late | +0.005 |
quiet | +0.005 |
Lisa | +0.004 |
too | +0.004 |
seek | +0.004 |
Neg logit contributions | |
imet | -0.011 |
penn | -0.011 |
pen | -0.011 |
ian | -0.011 |
Ow | -0.010 |
Token | " |
Token ID | 366 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
late | +0.010 |
seek | +0.010 |
quiet | +0.009 |
Lisa | +0.008 |
Lisa | +0.008 |
Neg logit contributions | |
penn | -0.011 |
imet | -0.011 |
pen | -0.011 |
ian | -0.011 |
Ow | -0.011 |
Token | You |
Token ID | 1639 |
Feature activation | -6.75e-01 |
Pos logit contributions | |
late | +0.031 |
quiet | +0.030 |
seek | +0.030 |
Lisa | +0.028 |
too | +0.027 |
Neg logit contributions | |
pen | -0.039 |
Ow | -0.039 |
penn | -0.038 |
imet | -0.037 |
yle | -0.037 |
Token | made |
Token ID | 925 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
late | +0.025 |
seek | +0.022 |
quiet | +0.022 |
Lisa | +0.021 |
Lisa | +0.020 |
Neg logit contributions | |
penn | -0.017 |
pen | -0.016 |
imet | -0.016 |
ian | -0.016 |
Ow | -0.016 |
Token | What |
Token ID | 2061 |
Feature activation | +8.87e-01 |
Pos logit contributions | |
late | +0.025 |
Lisa | +0.024 |
seek | +0.023 |
quiet | +0.022 |
too | +0.022 |
Neg logit contributions | |
penn | -0.028 |
Whoever | -0.026 |
yle | -0.026 |
pen | -0.026 |
imet | -0.025 |
Token | does |
Token ID | 857 |
Feature activation | -8.48e-01 |
Pos logit contributions | |
penn | +0.026 |
imet | +0.026 |
pen | +0.026 |
ian | +0.025 |
Ow | +0.024 |
Neg logit contributions | |
late | -0.028 |
quiet | -0.025 |
seek | -0.025 |
Lisa | -0.024 |
Lisa | -0.023 |
Token | he |
Token ID | 339 |
Feature activation | -3.72e-01 |
Pos logit contributions | |
late | +0.028 |
aky | +0.027 |
seek | +0.027 |
aking | +0.027 |
Lisa | +0.026 |
Neg logit contributions | |
penn | -0.020 |
ian | -0.020 |
Pe | -0.019 |
anybody | -0.019 |
imet | -0.019 |
Token | know |
Token ID | 760 |
Feature activation | -7.40e-01 |
Pos logit contributions | |
late | +0.015 |
Lisa | +0.013 |
Kate | +0.013 |
Maria | +0.013 |
pans | +0.013 |
Neg logit contributions | |
uable | -0.007 |
pen | -0.007 |
infection | -0.007 |
ian | -0.007 |
our | -0.007 |
Token | ?" |
Token ID | 1701 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
late | +0.017 |
seek | +0.014 |
quiet | +0.014 |
Lisa | +0.013 |
Lisa | +0.012 |
Neg logit contributions | |
penn | -0.029 |
imet | -0.028 |
pen | -0.028 |
ian | -0.028 |
Ow | -0.027 |
Token | Lily |
Token ID | 20037 |
Feature activation | -2.96e+00 |
Pos logit contributions | |
penn | +0.005 |
Pe | +0.005 |
imet | +0.004 |
Ow | +0.004 |
Peck | +0.004 |
Neg logit contributions | |
seek | -0.006 |
late | -0.006 |
aking | -0.005 |
lic | -0.005 |
akers | -0.005 |
Token | asks |
Token ID | 7893 |
Feature activation | -3.80e-01 |
Pos logit contributions | |
late | +0.114 |
seek | +0.111 |
lic | +0.101 |
Happy | +0.100 |
quiet | +0.099 |
Neg logit contributions | |
ian | -0.063 |
pen | -0.062 |
imet | -0.058 |
penn | -0.058 |
Ow | -0.057 |
Token | . |
Token ID | 13 |
Feature activation | -5.00e-01 |
Pos logit contributions | |
late | +0.008 |
Lisa | +0.007 |
quiet | +0.007 |
too | +0.006 |
seek | +0.006 |
Neg logit contributions | |
penn | -0.016 |
imet | -0.015 |
pen | -0.015 |
ian | -0.015 |
trek | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
late | +0.013 |
seek | +0.012 |
quiet | +0.012 |
Lisa | +0.011 |
Lisa | +0.010 |
Neg logit contributions | |
penn | -0.017 |
imet | -0.017 |
pen | -0.017 |
ian | -0.016 |
Ow | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | -4.78e-02 |
Pos logit contributions | |
late | +0.003 |
quiet | +0.002 |
seek | +0.002 |
Lisa | +0.002 |
Lisa | +0.002 |
Neg logit contributions | |
penn | -0.004 |
imet | -0.004 |
pen | -0.004 |
ian | -0.004 |
ome | -0.004 |
Token | " |
Token ID | 1 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
late | +0.001 |
seek | +0.001 |
quiet | +0.001 |
Lisa | +0.001 |
too | +0.001 |
Neg logit contributions | |
pen | -0.002 |
penn | -0.002 |
imet | -0.002 |
ian | -0.002 |
Ow | -0.002 |
Token | ". |
Token ID | 1911 |
Feature activation | +6.87e-01 |
Pos logit contributions | |
late | +0.008 |
seek | +0.007 |
quiet | +0.007 |
Lisa | +0.007 |
Lisa | +0.007 |
Neg logit contributions | |
ian | -0.009 |
pen | -0.009 |
imet | -0.009 |
penn | -0.008 |
ance | -0.008 |
Token | L |
Token ID | 406 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
penn | +0.024 |
imet | +0.024 |
pen | +0.024 |
ian | +0.023 |
ome | +0.022 |
Neg logit contributions | |
late | -0.018 |
quiet | -0.016 |
seek | -0.015 |
Lisa | -0.015 |
too | -0.014 |
Token | ila |
Token ID | 10102 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
late | +0.004 |
quiet | +0.003 |
seek | +0.003 |
Lisa | +0.003 |
Lisa | +0.003 |
Neg logit contributions | |
penn | -0.007 |
imet | -0.007 |
pen | -0.007 |
ian | -0.007 |
Ow | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | -3.61e-01 |
Pos logit contributions | |
late | +0.038 |
seek | +0.033 |
quiet | +0.033 |
Lisa | +0.031 |
Lisa | +0.030 |
Neg logit contributions | |
penn | -0.040 |
imet | -0.040 |
pen | -0.039 |
ian | -0.038 |
Ow | -0.037 |
Token | Ben |
Token ID | 3932 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
late | +0.010 |
quiet | +0.009 |
Lisa | +0.009 |
seek | +0.009 |
too | +0.008 |
Neg logit contributions | |
penn | -0.012 |
pen | -0.011 |
imet | -0.011 |
ian | -0.011 |
vered | -0.011 |
Token | both |
Token ID | 1111 |
Feature activation | -2.94e+00 |
Pos logit contributions | |
late | +0.030 |
quiet | +0.028 |
seek | +0.027 |
Lisa | +0.026 |
too | +0.025 |
Neg logit contributions | |
penn | -0.032 |
imet | -0.032 |
pen | -0.031 |
ian | -0.030 |
Ow | -0.030 |
Token | want |
Token ID | 765 |
Feature activation | -2.58e-01 |
Pos logit contributions | |
late | +0.074 |
quiet | +0.069 |
seek | +0.066 |
Lisa | +0.061 |
too | +0.060 |
Neg logit contributions | |
penn | -0.105 |
imet | -0.104 |
pen | -0.101 |
ian | -0.100 |
Ow | -0.098 |
Token | to |
Token ID | 284 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
late | +0.008 |
seek | +0.008 |
aky | +0.007 |
quiet | +0.007 |
Lisa | +0.007 |
Neg logit contributions | |
penn | -0.008 |
pen | -0.007 |
imet | -0.007 |
ian | -0.007 |
Ow | -0.007 |
Token | play |
Token ID | 711 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
late | +0.007 |
seek | +0.006 |
quiet | +0.006 |
Lisa | +0.006 |
Lisa | +0.006 |
Neg logit contributions | |
pen | -0.004 |
penn | -0.004 |
imet | -0.004 |
ian | -0.004 |
Ow | -0.004 |
Token | with |
Token ID | 351 |
Feature activation | +2.01e-01 |
Pos logit contributions | |
late | +0.046 |
Lisa | +0.041 |
aky | +0.041 |
Maria | +0.041 |
arge | +0.041 |
Neg logit contributions | |
pen | -0.034 |
penn | -0.031 |
ian | -0.027 |
anyone | -0.027 |
Peck | -0.027 |
Token | it |
Token ID | 340 |
Feature activation | -6.95e-01 |
Pos logit contributions | |
penn | +0.006 |
pen | +0.006 |
imet | +0.006 |
ian | +0.006 |
Ow | +0.005 |
Neg logit contributions | |
late | -0.006 |
seek | -0.006 |
quiet | -0.006 |
Lisa | -0.005 |
Lisa | -0.005 |
Token | ! |
Token ID | 0 |
Feature activation | -3.61e-01 |
Pos logit contributions | |
pen | +0.020 |
penn | +0.019 |
imet | +0.019 |
ome | +0.018 |
ian | +0.018 |
Neg logit contributions | |
late | -0.018 |
seek | -0.016 |
Lisa | -0.016 |
Lisa | -0.016 |
quiet | -0.016 |
Token | It |
Token ID | 632 |
Feature activation | -6.72e-03 |
Pos logit contributions | |
late | +0.012 |
seek | +0.011 |
quiet | +0.011 |
Lisa | +0.010 |
Lisa | +0.010 |
Neg logit contributions | |
penn | -0.010 |
imet | -0.010 |
pen | -0.010 |
ian | -0.009 |
Ow | -0.009 |
Token | was |
Token ID | 373 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
late | +0.000 |
Lisa | +0.000 |
Lisa | +0.000 |
seek | +0.000 |
Happy | +0.000 |
Neg logit contributions | |
ian | -0.000 |
trek | -0.000 |
Pe | -0.000 |
lur | -0.000 |
pen | -0.000 |
Token | making |
Token ID | 1642 |
Feature activation | -9.56e-01 |
Pos logit contributions | |
Pe | +0.005 |
Peck | +0.004 |
Sy | +0.004 |
Twe | +0.004 |
Mont | +0.004 |
Neg logit contributions | |
Happy | -0.005 |
Lisa | -0.004 |
seek | -0.004 |
late | -0.004 |
aking | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | -9.15e-01 |
Pos logit contributions | |
late | +0.027 |
seek | +0.027 |
Lisa | +0.024 |
quiet | +0.024 |
Lisa | +0.023 |
Neg logit contributions | |
penn | -0.030 |
Ow | -0.029 |
Pe | -0.028 |
pen | -0.028 |
imet | -0.028 |
Token | very |
Token ID | 845 |
Feature activation | -2.93e+00 |
Pos logit contributions | |
late | +0.021 |
seek | +0.018 |
Lisa | +0.017 |
Lisa | +0.016 |
Happy | +0.015 |
Neg logit contributions | |
penn | -0.033 |
Ow | -0.033 |
imet | -0.033 |
trigger | -0.033 |
pen | -0.032 |
Token | loud |
Token ID | 7812 |
Feature activation | -6.87e-01 |
Pos logit contributions | |
late | +0.066 |
Lisa | +0.062 |
Lisa | +0.061 |
seek | +0.061 |
Sarah | +0.060 |
Neg logit contributions | |
penn | -0.100 |
tum | -0.097 |
bank | -0.094 |
trigger | -0.094 |
knees | -0.093 |
Token | noise |
Token ID | 7838 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
late | +0.020 |
seek | +0.018 |
Lisa | +0.018 |
Lisa | +0.017 |
quiet | +0.016 |
Neg logit contributions | |
imet | -0.020 |
pen | -0.020 |
ian | -0.020 |
penn | -0.020 |
trigger | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -7.55e-01 |
Pos logit contributions | |
pen | +0.005 |
imet | +0.004 |
ome | +0.004 |
ian | +0.004 |
penn | +0.004 |
Neg logit contributions | |
late | -0.006 |
seek | -0.005 |
Lisa | -0.005 |
quiet | -0.005 |
Happy | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | -3.29e-01 |
Pos logit contributions | |
late | +0.025 |
seek | +0.023 |
quiet | +0.022 |
Lisa | +0.021 |
Lisa | +0.021 |
Neg logit contributions | |
penn | -0.021 |
imet | -0.021 |
pen | -0.020 |
ian | -0.020 |
Ow | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -3.75e-01 |
Pos logit contributions | |
late | +0.009 |
seek | +0.009 |
quiet | +0.008 |
Lisa | +0.008 |
too | +0.008 |
Neg logit contributions | |
imet | -0.010 |
pen | -0.010 |
Ow | -0.010 |
penn | -0.010 |
Pe | -0.010 |
Token | her |
Token ID | 607 |
Feature activation | -9.72e-01 |
Pos logit contributions | |
late | +0.027 |
aky | +0.025 |
aking | +0.024 |
seek | +0.024 |
lic | +0.024 |
Neg logit contributions | |
penn | -0.021 |
anybody | -0.020 |
anyone | -0.020 |
ian | -0.020 |
Pe | -0.019 |
Token | mom |
Token ID | 1995 |
Feature activation | -6.17e-01 |
Pos logit contributions | |
late | +0.036 |
seek | +0.032 |
Happy | +0.031 |
They | +0.030 |
lic | +0.030 |
Neg logit contributions | |
penn | -0.023 |
tum | -0.022 |
stomach | -0.021 |
imet | -0.021 |
ian | -0.021 |
Token | said |
Token ID | 531 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
late | +0.021 |
seek | +0.019 |
quiet | +0.018 |
awesome | +0.017 |
Happy | +0.017 |
Neg logit contributions | |
ian | -0.017 |
pen | -0.017 |
imet | -0.016 |
penn | -0.015 |
brow | -0.015 |
Token | in |
Token ID | 287 |
Feature activation | -9.56e-01 |
Pos logit contributions | |
penn | +0.005 |
imet | +0.005 |
pen | +0.005 |
ian | +0.004 |
Ow | +0.004 |
Neg logit contributions | |
late | -0.005 |
seek | -0.005 |
quiet | -0.005 |
Lisa | -0.005 |
Lisa | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
late | +0.028 |
seek | +0.027 |
quiet | +0.024 |
lic | +0.023 |
aking | +0.023 |
Neg logit contributions | |
penn | -0.029 |
pen | -0.029 |
ian | -0.028 |
growth | -0.027 |
Ow | -0.027 |
Token | much |
Token ID | 881 |
Feature activation | -2.92e+00 |
Pos logit contributions | |
late | +0.037 |
seek | +0.035 |
amer | +0.033 |
aking | +0.032 |
ores | +0.032 |
Neg logit contributions | |
penn | -0.040 |
tum | -0.039 |
brow | -0.039 |
stomach | -0.038 |
bank | -0.038 |
Token | gent |
Token ID | 25049 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
seek | +0.088 |
Lisa | +0.084 |
akers | +0.083 |
amer | +0.080 |
o | +0.079 |
Neg logit contributions | |
risk | -0.078 |
penn | -0.078 |
tum | -0.070 |
anybody | -0.070 |
stomach | -0.069 |
Token | ler |
Token ID | 1754 |
Feature activation | -3.43e-01 |
Pos logit contributions | |
late | +0.047 |
seek | +0.044 |
quiet | +0.043 |
Lisa | +0.042 |
Lisa | +0.040 |
Neg logit contributions | |
ian | -0.025 |
penn | -0.024 |
pen | -0.024 |
imet | -0.022 |
ome | -0.021 |
Token | voice |
Token ID | 3809 |
Feature activation | +3.89e-01 |
Pos logit contributions | |
late | +0.012 |
seek | +0.011 |
Lisa | +0.010 |
Happy | +0.010 |
quiet | +0.010 |
Neg logit contributions | |
ian | -0.009 |
imet | -0.009 |
pen | -0.008 |
penn | -0.008 |
ome | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
ian | +0.008 |
ome | +0.008 |
imet | +0.008 |
penn | +0.008 |
uable | +0.007 |
Neg logit contributions | |
late | -0.015 |
seek | -0.015 |
quiet | -0.013 |
Lisa | -0.013 |
pans | -0.013 |
Token | " |
Token ID | 366 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
Pe | +0.003 |
imet | +0.002 |
chief | +0.002 |
Twe | +0.002 |
brow | +0.002 |
Neg logit contributions | |
seek | -0.005 |
late | -0.005 |
aky | -0.005 |
aking | -0.004 |
Lisa | -0.004 |
Token | L |
Token ID | 406 |
Feature activation | -7.50e-01 |
Pos logit contributions | |
late | +0.010 |
seek | +0.009 |
aky | +0.009 |
aking | +0.008 |
aming | +0.008 |
Neg logit contributions | |
Pe | -0.007 |
penn | -0.007 |
Twe | -0.007 |
Ow | -0.006 |
cash | -0.006 |
Token | ila |
Token ID | 10102 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
late | +0.020 |
quiet | +0.018 |
seek | +0.018 |
Lisa | +0.015 |
quieter | +0.015 |
Neg logit contributions | |
imet | -0.025 |
penn | -0.025 |
ian | -0.025 |
ome | -0.025 |
pen | -0.024 |
Token | and |
Token ID | 290 |
Feature activation | -4.56e-01 |
Pos logit contributions | |
late | +0.053 |
seek | +0.048 |
quiet | +0.048 |
Happy | +0.046 |
ushes | +0.045 |
Neg logit contributions | |
pen | -0.063 |
ian | -0.063 |
ome | -0.059 |
imet | -0.058 |
penn | -0.058 |
Token | Tom |
Token ID | 4186 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
late | +0.015 |
seek | +0.013 |
quiet | +0.013 |
aky | +0.012 |
Happy | +0.012 |
Neg logit contributions | |
penn | -0.013 |
imet | -0.013 |
Pe | -0.013 |
Ow | -0.013 |
pen | -0.012 |
Token | ?" |
Token ID | 1701 |
Feature activation | -3.38e-01 |
Pos logit contributions | |
late | +0.031 |
seek | +0.029 |
Lisa | +0.028 |
quiet | +0.027 |
Happy | +0.026 |
Neg logit contributions | |
pen | -0.043 |
ian | -0.041 |
penn | -0.039 |
imet | -0.039 |
Pe | -0.038 |
Token | she |
Token ID | 673 |
Feature activation | -2.92e+00 |
Pos logit contributions | |
seek | +0.012 |
late | +0.012 |
aky | +0.011 |
lic | +0.011 |
aking | +0.011 |
Neg logit contributions | |
Pe | -0.009 |
penn | -0.008 |
Twe | -0.008 |
yle | -0.007 |
Ow | -0.007 |
Token | asks |
Token ID | 7893 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
late | +0.109 |
seek | +0.105 |
lic | +0.096 |
Happy | +0.094 |
ores | +0.093 |
Neg logit contributions | |
ian | -0.065 |
penn | -0.062 |
Pe | -0.061 |
pen | -0.060 |
Ow | -0.059 |
Token | . |
Token ID | 13 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
late | +0.006 |
quiet | +0.005 |
seek | +0.005 |
Lisa | +0.005 |
too | +0.005 |
Neg logit contributions | |
penn | -0.010 |
imet | -0.010 |
pen | -0.010 |
ian | -0.009 |
Ow | -0.009 |
Token | " |
Token ID | 366 |
Feature activation | -7.50e-01 |
Pos logit contributions | |
seek | +0.011 |
late | +0.011 |
quiet | +0.010 |
aky | +0.010 |
awesome | +0.010 |
Neg logit contributions | |
Pe | -0.009 |
imet | -0.009 |
Ow | -0.009 |
ian | -0.009 |
penn | -0.009 |
Token | Did |
Token ID | 11633 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
late | +0.019 |
seek | +0.019 |
quiet | +0.019 |
Lisa | +0.019 |
awesome | +0.018 |
Neg logit contributions | |
pen | -0.023 |
penn | -0.023 |
Ow | -0.023 |
yle | -0.022 |
imet | -0.022 |
Token | you |
Token ID | 345 |
Feature activation | -6.18e-01 |
Pos logit contributions | |
penn | +0.004 |
imet | +0.004 |
pen | +0.004 |
ian | +0.004 |
Ow | +0.004 |
Neg logit contributions | |
late | -0.004 |
seek | -0.003 |
quiet | -0.003 |
Lisa | -0.003 |
Lisa | -0.003 |
Token | What |
Token ID | 2061 |
Feature activation | +7.16e-01 |
Pos logit contributions | |
late | +0.024 |
seek | +0.021 |
quiet | +0.020 |
pleased | +0.020 |
lic | +0.020 |
Neg logit contributions | |
penn | -0.024 |
yle | -0.023 |
Ow | -0.023 |
pen | -0.023 |
ian | -0.022 |
Token | are |
Token ID | 389 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
penn | +0.020 |
imet | +0.020 |
pen | +0.020 |
ian | +0.019 |
brow | +0.019 |
Neg logit contributions | |
late | -0.024 |
quiet | -0.022 |
seek | -0.021 |
Lisa | -0.021 |
Lisa | -0.020 |
Token | you |
Token ID | 345 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
late | +0.027 |
quiet | +0.025 |
Lisa | +0.024 |
seek | +0.022 |
too | +0.022 |
Neg logit contributions | |
pen | -0.071 |
growth | -0.069 |
imet | -0.068 |
vered | -0.068 |
trigger | -0.067 |
Token | doing |
Token ID | 1804 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
late | +0.029 |
seek | +0.026 |
Lisa | +0.026 |
Happy | +0.026 |
lic | +0.025 |
Neg logit contributions | |
penn | -0.032 |
pen | -0.031 |
Ow | -0.031 |
ian | -0.030 |
uf | -0.029 |
Token | ?" |
Token ID | 1701 |
Feature activation | -3.92e-01 |
Pos logit contributions | |
late | +0.051 |
lic | +0.045 |
aky | +0.045 |
quiet | +0.044 |
seek | +0.043 |
Neg logit contributions | |
pen | -0.063 |
penn | -0.060 |
ian | -0.060 |
yle | -0.057 |
imet | -0.057 |
Token | she |
Token ID | 673 |
Feature activation | -2.91e+00 |
Pos logit contributions | |
late | +0.012 |
seek | +0.012 |
lic | +0.010 |
aky | +0.010 |
aking | +0.010 |
Neg logit contributions | |
penn | -0.011 |
Pe | -0.011 |
ian | -0.011 |
imet | -0.010 |
pen | -0.010 |
Token | asks |
Token ID | 7893 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
late | +0.094 |
seek | +0.087 |
quiet | +0.082 |
Lisa | +0.077 |
Lisa | +0.076 |
Neg logit contributions | |
penn | -0.082 |
ian | -0.081 |
pen | -0.080 |
imet | -0.080 |
Ow | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | -4.37e-01 |
Pos logit contributions | |
late | +0.006 |
Lisa | +0.006 |
quiet | +0.005 |
too | +0.005 |
seek | +0.005 |
Neg logit contributions | |
imet | -0.014 |
penn | -0.014 |
pen | -0.013 |
Ow | -0.013 |
ian | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
late | +0.012 |
seek | +0.011 |
quiet | +0.010 |
Lisa | +0.009 |
Lisa | +0.009 |
Neg logit contributions | |
penn | -0.014 |
imet | -0.014 |
pen | -0.014 |
ian | -0.014 |
Ow | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | +2.62e-02 |
Pos logit contributions | |
penn | +0.001 |
imet | +0.001 |
pen | +0.001 |
ian | +0.001 |
Ow | +0.001 |
Neg logit contributions | |
late | -0.001 |
quiet | -0.000 |
seek | -0.000 |
Lisa | -0.000 |
Lisa | -0.000 |
Token | " |
Token ID | 1 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
penn | +0.001 |
pen | +0.001 |
ian | +0.001 |
Pe | +0.001 |
imet | +0.001 |
Neg logit contributions | |
late | -0.001 |
seek | -0.000 |
lic | -0.000 |
quiet | -0.000 |
too | -0.000 |
Token | making |
Token ID | 1642 |
Feature activation | -9.54e-01 |
Pos logit contributions | |
late | +0.030 |
seek | +0.029 |
Happy | +0.028 |
Lisa | +0.027 |
Lisa | +0.025 |
Neg logit contributions | |
penn | -0.025 |
pen | -0.023 |
yle | -0.023 |
Ow | -0.023 |
Pe | -0.023 |
Token | such |
Token ID | 884 |
Feature activation | -6.99e-02 |
Pos logit contributions | |
late | +0.029 |
seek | +0.028 |
Lisa | +0.025 |
quiet | +0.025 |
aky | +0.024 |
Neg logit contributions | |
penn | -0.030 |
Ow | -0.028 |
pen | -0.028 |
imet | -0.027 |
Pe | -0.027 |
Token | a |
Token ID | 257 |
Feature activation | -2.88e-01 |
Pos logit contributions | |
late | +0.002 |
seek | +0.002 |
Lisa | +0.002 |
Lisa | +0.002 |
quiet | +0.002 |
Neg logit contributions | |
penn | -0.002 |
Ow | -0.002 |
pen | -0.002 |
imet | -0.002 |
ian | -0.002 |
Token | mess |
Token ID | 2085 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
late | +0.008 |
seek | +0.007 |
Lisa | +0.007 |
Lisa | +0.006 |
quiet | +0.006 |
Neg logit contributions | |
penn | -0.010 |
pen | -0.009 |
Ow | -0.009 |
imet | -0.009 |
ian | -0.009 |
Token | ?" |
Token ID | 1701 |
Feature activation | -3.29e-01 |
Pos logit contributions | |
catchy | +0.065 |
Maria | +0.065 |
Lisa | +0.064 |
pleased | +0.063 |
seek | +0.063 |
Neg logit contributions | |
pen | -0.047 |
ian | -0.045 |
penn | -0.044 |
our | -0.044 |
ors | -0.042 |
Token | she |
Token ID | 673 |
Feature activation | -2.90e+00 |
Pos logit contributions | |
late | +0.010 |
seek | +0.009 |
quiet | +0.009 |
lic | +0.008 |
Lisa | +0.008 |
Neg logit contributions | |
penn | -0.010 |
Ow | -0.009 |
pen | -0.009 |
imet | -0.009 |
Pe | -0.009 |
Token | asks |
Token ID | 7893 |
Feature activation | -8.24e-02 |
Pos logit contributions | |
late | +0.102 |
seek | +0.098 |
quiet | +0.090 |
lic | +0.087 |
pleased | +0.085 |
Neg logit contributions | |
penn | -0.071 |
ian | -0.071 |
pen | -0.068 |
imet | -0.067 |
Ow | -0.067 |
Token | and |
Token ID | 290 |
Feature activation | -9.25e-02 |
Pos logit contributions | |
late | +0.002 |
Lisa | +0.001 |
quiet | +0.001 |
too | +0.001 |
seek | +0.001 |
Neg logit contributions | |
imet | -0.003 |
pen | -0.003 |
penn | -0.003 |
ian | -0.003 |
trek | -0.003 |
Token | looks |
Token ID | 3073 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
late | +0.002 |
quiet | +0.002 |
Lisa | +0.002 |
seek | +0.002 |
too | +0.002 |
Neg logit contributions | |
pen | -0.004 |
penn | -0.004 |
imet | -0.004 |
ian | -0.003 |
Ow | -0.003 |
Token | cross |
Token ID | 3272 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
penn | +0.005 |
imet | +0.005 |
pen | +0.005 |
ian | +0.005 |
Ow | +0.005 |
Neg logit contributions | |
late | -0.005 |
seek | -0.005 |
Lisa | -0.005 |
quiet | -0.004 |
Lisa | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -3.79e-01 |
Pos logit contributions | |
seek | +0.044 |
lic | +0.039 |
late | +0.038 |
pleased | +0.038 |
impressed | +0.037 |
Neg logit contributions | |
penn | -0.028 |
pen | -0.027 |
ian | -0.025 |
brow | -0.025 |
vered | -0.025 |
Token | apart |
Token ID | 5475 |
Feature activation | -5.65e-01 |
Pos logit contributions | |
late | +0.024 |
Lisa | +0.023 |
seek | +0.023 |
Lisa | +0.022 |
quiet | +0.022 |
Neg logit contributions | |
imet | -0.023 |
pen | -0.022 |
penn | -0.022 |
ian | -0.022 |
trigger | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | -8.52e-01 |
Pos logit contributions | |
Lisa | +0.019 |
late | +0.019 |
lic | +0.018 |
seek | +0.018 |
Happy | +0.018 |
Neg logit contributions | |
penn | -0.014 |
pen | -0.014 |
imet | -0.013 |
Pe | -0.012 |
ian | -0.012 |
Token | But |
Token ID | 887 |
Feature activation | +9.51e-02 |
Pos logit contributions | |
late | +0.026 |
seek | +0.024 |
quiet | +0.023 |
Lisa | +0.022 |
Lisa | +0.021 |
Neg logit contributions | |
penn | -0.025 |
imet | -0.025 |
pen | -0.025 |
ian | -0.024 |
Ow | -0.024 |
Token | then |
Token ID | 788 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
penn | +0.003 |
imet | +0.003 |
Pe | +0.003 |
pen | +0.003 |
brow | +0.003 |
Neg logit contributions | |
late | -0.003 |
seek | -0.003 |
Lisa | -0.002 |
lic | -0.002 |
quiet | -0.002 |
Token | a |
Token ID | 257 |
Feature activation | -8.33e-01 |
Pos logit contributions | |
late | +0.035 |
seek | +0.031 |
quiet | +0.030 |
lic | +0.029 |
Lisa | +0.029 |
Neg logit contributions | |
penn | -0.030 |
imet | -0.028 |
pen | -0.028 |
ian | -0.027 |
Pe | -0.027 |
Token | very |
Token ID | 845 |
Feature activation | -2.90e+00 |
Pos logit contributions | |
late | +0.022 |
Lisa | +0.019 |
seek | +0.018 |
aking | +0.018 |
Lisa | +0.018 |
Neg logit contributions | |
penn | -0.026 |
trigger | -0.025 |
ian | -0.025 |
imet | -0.025 |
pen | -0.024 |
Token | special |
Token ID | 2041 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
late | +0.069 |
Lisa | +0.066 |
Lisa | +0.065 |
They | +0.065 |
apes | +0.062 |
Neg logit contributions | |
penn | -0.092 |
trigger | -0.084 |
imet | -0.082 |
ome | -0.082 |
bank | -0.081 |
Token | thing |
Token ID | 1517 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
late | +0.034 |
seek | +0.030 |
quiet | +0.029 |
Lisa | +0.029 |
Lisa | +0.028 |
Neg logit contributions | |
penn | -0.032 |
ian | -0.031 |
imet | -0.030 |
pen | -0.030 |
ome | -0.029 |
Token | happened |
Token ID | 3022 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
imet | +0.003 |
penn | +0.003 |
pen | +0.003 |
ian | +0.003 |
ome | +0.003 |
Neg logit contributions | |
late | -0.007 |
seek | -0.006 |
quiet | -0.006 |
Lisa | -0.006 |
Lisa | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -7.22e-01 |
Pos logit contributions | |
late | +0.006 |
seek | +0.006 |
Lisa | +0.006 |
aky | +0.005 |
Happy | +0.005 |
Neg logit contributions | |
pen | -0.004 |
imet | -0.004 |
chief | -0.004 |
penn | -0.004 |
brow | -0.004 |
Token | A |
Token ID | 317 |
Feature activation | -9.16e-01 |
Pos logit contributions | |
late | +0.028 |
seek | +0.026 |
quiet | +0.026 |
Lisa | +0.024 |
Lisa | +0.024 |
Neg logit contributions | |
imet | -0.015 |
penn | -0.015 |
pen | -0.015 |
ian | -0.014 |
Ow | -0.014 |
Token | What |
Token ID | 1867 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
imet | +0.022 |
penn | +0.022 |
pen | +0.022 |
Ow | +0.022 |
ian | +0.022 |
Neg logit contributions | |
late | -0.028 |
quiet | -0.027 |
seek | -0.026 |
Lisa | -0.024 |
Lisa | -0.024 |
Token | did |
Token ID | 750 |
Feature activation | -6.08e-01 |
Pos logit contributions | |
penn | +0.035 |
imet | +0.034 |
pen | +0.034 |
ian | +0.033 |
Ow | +0.032 |
Neg logit contributions | |
late | -0.043 |
quiet | -0.039 |
seek | -0.039 |
Lisa | -0.037 |
Lisa | -0.035 |
Token | you |
Token ID | 345 |
Feature activation | -8.91e-01 |
Pos logit contributions | |
late | +0.018 |
seek | +0.016 |
quiet | +0.016 |
Lisa | +0.015 |
Lisa | +0.014 |
Neg logit contributions | |
penn | -0.019 |
imet | -0.019 |
pen | -0.018 |
ian | -0.018 |
Ow | -0.018 |
Token | create |
Token ID | 2251 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
late | +0.038 |
Lisa | +0.034 |
ushes | +0.033 |
Lisa | +0.033 |
lic | +0.033 |
Neg logit contributions | |
pen | -0.016 |
rest | -0.015 |
our | -0.014 |
ian | -0.014 |
Ow | -0.013 |
Token | ?" |
Token ID | 1701 |
Feature activation | -7.27e-01 |
Pos logit contributions | |
late | +0.042 |
Lisa | +0.039 |
seek | +0.039 |
lic | +0.038 |
pleased | +0.038 |
Neg logit contributions | |
pen | -0.052 |
penn | -0.050 |
anybody | -0.047 |
ian | -0.046 |
Ow | -0.046 |
Token | she |
Token ID | 673 |
Feature activation | -2.90e+00 |
Pos logit contributions | |
late | +0.022 |
seek | +0.021 |
quiet | +0.020 |
Lisa | +0.018 |
Lisa | +0.018 |
Neg logit contributions | |
penn | -0.022 |
pen | -0.021 |
imet | -0.021 |
ian | -0.020 |
Pe | -0.020 |
Token | asks |
Token ID | 7893 |
Feature activation | -6.30e-01 |
Pos logit contributions | |
late | +0.100 |
seek | +0.093 |
quiet | +0.087 |
Lisa | +0.081 |
Lisa | +0.081 |
Neg logit contributions | |
penn | -0.075 |
ian | -0.074 |
pen | -0.073 |
imet | -0.072 |
Ow | -0.070 |
Token | , |
Token ID | 11 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
late | +0.012 |
Lisa | +0.011 |
too | +0.010 |
quiet | +0.010 |
seek | +0.008 |
Neg logit contributions | |
imet | -0.027 |
penn | -0.026 |
pen | -0.026 |
Ow | -0.025 |
ian | -0.025 |
Token | looking |
Token ID | 2045 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
late | +0.001 |
seek | +0.001 |
Happy | +0.001 |
Lisa | +0.001 |
lic | +0.001 |
Neg logit contributions | |
brow | -0.001 |
penn | -0.001 |
pen | -0.001 |
Pe | -0.001 |
knees | -0.001 |
Token | at |
Token ID | 379 |
Feature activation | -4.33e-01 |
Pos logit contributions | |
late | +0.006 |
seek | +0.006 |
Lisa | +0.005 |
quiet | +0.005 |
Lisa | +0.005 |
Neg logit contributions | |
penn | -0.006 |
pen | -0.006 |
ian | -0.006 |
imet | -0.006 |
ome | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | +5.10e-02 |
Pos logit contributions | |
late | +0.011 |
seek | +0.010 |
quiet | +0.010 |
Lisa | +0.009 |
Lisa | +0.009 |
Neg logit contributions | |
penn | -0.015 |
imet | -0.015 |
pen | -0.015 |
ian | -0.015 |
Ow | -0.014 |
Token | for |
Token ID | 329 |
Feature activation | -6.08e-01 |
Pos logit contributions | |
late | +0.010 |
seek | +0.009 |
quiet | +0.009 |
Lisa | +0.009 |
aky | +0.008 |
Neg logit contributions | |
pen | -0.009 |
penn | -0.009 |
ome | -0.009 |
imet | -0.009 |
ian | -0.008 |
Token | him |
Token ID | 683 |
Feature activation | +4.29e-01 |
Pos logit contributions | |
late | +0.021 |
seek | +0.019 |
quiet | +0.018 |
lic | +0.018 |
Lisa | +0.018 |
Neg logit contributions | |
penn | -0.016 |
pen | -0.015 |
anybody | -0.015 |
Pe | -0.015 |
imet | -0.014 |
Token | ! |
Token ID | 0 |
Feature activation | -8.43e-02 |
Pos logit contributions | |
pen | +0.011 |
penn | +0.011 |
imet | +0.011 |
ome | +0.011 |
ian | +0.010 |
Neg logit contributions | |
late | -0.015 |
Lisa | -0.013 |
seek | -0.013 |
pans | -0.013 |
Happy | -0.012 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +4.61e-03 |
Pos logit contributions | |
late | +0.002 |
seek | +0.002 |
quiet | +0.002 |
Lisa | +0.002 |
Lisa | +0.002 |
Neg logit contributions | |
imet | -0.003 |
penn | -0.003 |
pen | -0.003 |
Ow | -0.003 |
ian | -0.003 |
Token | Little |
Token ID | 22253 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
pen | +0.000 |
imet | +0.000 |
penn | +0.000 |
ian | +0.000 |
Ow | +0.000 |
Neg logit contributions | |
late | -0.000 |
quiet | -0.000 |
seek | -0.000 |
Lisa | -0.000 |
Lisa | -0.000 |
Token | Billy |
Token ID | 15890 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
penn | +0.036 |
imet | +0.036 |
pen | +0.035 |
ian | +0.035 |
Ow | +0.033 |
Neg logit contributions | |
late | -0.042 |
seek | -0.037 |
quiet | -0.037 |
Lisa | -0.035 |
Lisa | -0.034 |
Token | was |
Token ID | 373 |
Feature activation | +5.77e-01 |
Pos logit contributions | |
penn | +0.050 |
imet | +0.049 |
pen | +0.047 |
trigger | +0.046 |
Ow | +0.046 |
Neg logit contributions | |
late | -0.048 |
quiet | -0.043 |
seek | -0.042 |
Lisa | -0.040 |
too | -0.038 |
Token | walking |
Token ID | 6155 |
Feature activation | -3.47e-02 |
Pos logit contributions | |
penn | +0.016 |
pen | +0.016 |
brow | +0.016 |
Pe | +0.016 |
imet | +0.015 |
Neg logit contributions | |
seek | -0.017 |
late | -0.016 |
Lisa | -0.016 |
Lisa | -0.015 |
Happy | -0.015 |
Token | in |
Token ID | 287 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
late | +0.001 |
Lisa | +0.001 |
seek | +0.001 |
Lisa | +0.001 |
lic | +0.001 |
Neg logit contributions | |
penn | -0.001 |
pen | -0.001 |
trek | -0.001 |
imet | -0.001 |
ian | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | -2.39e-01 |
Pos logit contributions | |
late | +0.035 |
lic | +0.033 |
seek | +0.032 |
aky | +0.031 |
Happy | +0.031 |
Neg logit contributions | |
trek | -0.032 |
pen | -0.032 |
penn | -0.031 |
Pe | -0.031 |
Higher | -0.031 |
Token | park |
Token ID | 3952 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
late | +0.008 |
seek | +0.008 |
lic | +0.008 |
Lisa | +0.007 |
Lisa | +0.007 |
Neg logit contributions | |
penn | -0.006 |
pen | -0.005 |
Pe | -0.005 |
Ow | -0.005 |
trek | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | +6.31e-01 |
Pos logit contributions | |
penn | +0.028 |
imet | +0.028 |
pen | +0.028 |
ian | +0.027 |
Ow | +0.027 |
Neg logit contributions | |
late | -0.024 |
quiet | -0.021 |
seek | -0.021 |
Lisa | -0.020 |
too | -0.019 |
Token | plane |
Token ID | 6614 |
Feature activation | -2.66e-01 |
Pos logit contributions | |
penn | +0.012 |
ian | +0.010 |
imet | +0.010 |
pen | +0.010 |
trigger | +0.009 |
Neg logit contributions | |
late | -0.027 |
seek | -0.026 |
Lisa | -0.024 |
quiet | -0.024 |
Lisa | -0.024 |
Token | couldn |
Token ID | 3521 |
Feature activation | +7.19e-01 |
Pos logit contributions | |
late | +0.010 |
seek | +0.009 |
Lisa | +0.009 |
quiet | +0.009 |
Lisa | +0.008 |
Neg logit contributions | |
penn | -0.006 |
pen | -0.006 |
ian | -0.006 |
imet | -0.006 |
ome | -0.006 |
Token | 't |
Token ID | 470 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
penn | +0.021 |
imet | +0.020 |
pen | +0.020 |
ian | +0.019 |
Ow | +0.019 |
Neg logit contributions | |
late | -0.024 |
quiet | -0.021 |
seek | -0.021 |
Lisa | -0.020 |
too | -0.020 |
Token | fly |
Token ID | 6129 |
Feature activation | -3.25e-02 |
Pos logit contributions | |
penn | +0.026 |
imet | +0.025 |
pen | +0.025 |
ian | +0.024 |
Ow | +0.024 |
Neg logit contributions | |
late | -0.039 |
quiet | -0.036 |
seek | -0.036 |
Lisa | -0.034 |
too | -0.033 |
Token | without |
Token ID | 1231 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
late | +0.001 |
Lisa | +0.001 |
seek | +0.001 |
pans | +0.001 |
Lisa | +0.001 |
Neg logit contributions | |
pen | -0.001 |
anybody | -0.001 |
penn | -0.001 |
anyone | -0.001 |
imet | -0.001 |
Token | fuel |
Token ID | 5252 |
Feature activation | -3.54e-02 |
Pos logit contributions | |
pen | +0.051 |
imet | +0.051 |
ian | +0.050 |
ome | +0.050 |
penn | +0.050 |
Neg logit contributions | |
late | -0.047 |
quiet | -0.043 |
seek | -0.040 |
too | -0.039 |
Lisa | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | +3.35e-02 |
Pos logit contributions | |
late | +0.001 |
quiet | +0.001 |
seek | +0.001 |
Lisa | +0.001 |
too | +0.001 |
Neg logit contributions | |
penn | -0.001 |
imet | -0.001 |
pen | -0.001 |
ian | -0.001 |
Ow | -0.001 |
Token | So |
Token ID | 1406 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
penn | +0.001 |
imet | +0.001 |
ian | +0.001 |
pen | +0.001 |
trigger | +0.001 |
Neg logit contributions | |
late | -0.001 |
Lisa | -0.001 |
quiet | -0.001 |
too | -0.001 |
seek | -0.001 |
Token | every |
Token ID | 790 |
Feature activation | -3.53e-01 |
Pos logit contributions | |
late | +0.007 |
quiet | +0.006 |
seek | +0.006 |
Lisa | +0.006 |
too | +0.006 |
Neg logit contributions | |
pen | -0.008 |
imet | -0.008 |
penn | -0.008 |
ian | -0.008 |
Ow | -0.008 |
Token | day |
Token ID | 1110 |
Feature activation | -6.09e-04 |
Pos logit contributions | |
late | +0.011 |
seek | +0.010 |
quiet | +0.009 |
Lisa | +0.009 |
Lisa | +0.009 |
Neg logit contributions | |
penn | -0.010 |
imet | -0.010 |
ian | -0.010 |
trek | -0.010 |
pen | -0.010 |
Token | to |
Token ID | 284 |
Feature activation | +2.62e-01 |
Pos logit contributions | |
bank | +0.001 |
trek | +0.001 |
penn | +0.001 |
kay | +0.001 |
imet | +0.001 |
Neg logit contributions | |
Lisa | -0.002 |
late | -0.002 |
Happy | -0.002 |
aky | -0.002 |
Maria | -0.002 |
Token | her |
Token ID | 607 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
penn | +0.007 |
pen | +0.007 |
imet | +0.007 |
Ow | +0.007 |
ian | +0.007 |
Neg logit contributions | |
late | -0.009 |
seek | -0.008 |
quiet | -0.008 |
Lisa | -0.007 |
Lisa | -0.007 |
Token | mom |
Token ID | 1995 |
Feature activation | +4.34e-01 |
Pos logit contributions | |
late | +0.008 |
Happy | +0.007 |
lic | +0.007 |
Lisa | +0.007 |
seek | +0.006 |
Neg logit contributions | |
knees | -0.006 |
penn | -0.005 |
brow | -0.005 |
bank | -0.005 |
pen | -0.005 |
Token | to |
Token ID | 284 |
Feature activation | +4.07e-01 |
Pos logit contributions | |
pen | +0.013 |
ian | +0.012 |
imet | +0.012 |
Pe | +0.012 |
penn | +0.011 |
Neg logit contributions | |
late | -0.014 |
seek | -0.013 |
Lisa | -0.012 |
quiet | -0.012 |
Happy | -0.012 |
Token | ask |
Token ID | 1265 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
penn | +0.011 |
imet | +0.010 |
pen | +0.010 |
ian | +0.010 |
Ow | +0.010 |
Neg logit contributions | |
late | -0.014 |
quiet | -0.013 |
seek | -0.013 |
Lisa | -0.012 |
Lisa | -0.012 |
Token | what |
Token ID | 644 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
penn | +0.037 |
pen | +0.036 |
imet | +0.036 |
ian | +0.036 |
brow | +0.034 |
Neg logit contributions | |
late | -0.033 |
seek | -0.029 |
quiet | -0.028 |
Lisa | -0.026 |
Lisa | -0.026 |
Token | to |
Token ID | 284 |
Feature activation | -5.01e-01 |
Pos logit contributions | |
penn | +0.043 |
imet | +0.041 |
pen | +0.040 |
ian | +0.039 |
Pe | +0.038 |
Neg logit contributions | |
late | -0.052 |
seek | -0.047 |
quiet | -0.046 |
Lisa | -0.043 |
Lisa | -0.042 |
Token | do |
Token ID | 466 |
Feature activation | -7.91e-01 |
Pos logit contributions | |
late | +0.019 |
seek | +0.016 |
quiet | +0.016 |
Lisa | +0.016 |
Lisa | +0.015 |
Neg logit contributions | |
penn | -0.012 |
pen | -0.012 |
imet | -0.011 |
Ow | -0.011 |
ian | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
late | +0.028 |
lic | +0.026 |
Lisa | +0.026 |
quiet | +0.025 |
seek | +0.025 |
Neg logit contributions | |
penn | -0.018 |
pen | -0.017 |
imet | -0.016 |
ome | -0.016 |
anybody | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
penn | +0.008 |
imet | +0.008 |
pen | +0.008 |
ian | +0.008 |
Ow | +0.008 |
Neg logit contributions | |
late | -0.007 |
quiet | -0.006 |
seek | -0.006 |
Lisa | -0.006 |
Lisa | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
pen | +0.005 |
imet | +0.005 |
penn | +0.005 |
Ow | +0.005 |
ian | +0.005 |
Neg logit contributions | |
late | -0.005 |
seek | -0.005 |
quiet | -0.005 |
Lisa | -0.004 |
too | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | +4.73e-01 |
Pos logit contributions | |
penn | +0.012 |
pen | +0.012 |
ian | +0.011 |
imet | +0.011 |
ome | +0.011 |
Neg logit contributions | |
late | -0.010 |
seek | -0.009 |
quiet | -0.009 |
Lisa | -0.008 |
Lisa | -0.008 |
Token | quickly |
Token ID | 2952 |
Feature activation | +5.69e-01 |
Pos logit contributions | |
penn | +0.012 |
imet | +0.011 |
pen | +0.011 |
ian | +0.011 |
Ow | +0.010 |
Neg logit contributions | |
late | -0.017 |
quiet | -0.016 |
seek | -0.016 |
Lisa | -0.015 |
Lisa | -0.014 |
Token | hopped |
Token ID | 45230 |
Feature activation | +5.58e-01 |
Pos logit contributions | |
penn | +0.015 |
Ow | +0.015 |
imet | +0.015 |
pen | +0.015 |
ian | +0.014 |
Neg logit contributions | |
late | -0.019 |
quiet | -0.018 |
seek | -0.018 |
Lisa | -0.017 |
too | -0.016 |
Token | on |
Token ID | 319 |
Feature activation | +7.10e-01 |
Pos logit contributions | |
imet | +0.018 |
penn | +0.018 |
pen | +0.017 |
ian | +0.017 |
brow | +0.017 |
Neg logit contributions | |
late | -0.015 |
seek | -0.014 |
quiet | -0.014 |
Lisa | -0.013 |
Lisa | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
penn | +0.023 |
imet | +0.023 |
pen | +0.023 |
ian | +0.022 |
Ow | +0.022 |
Neg logit contributions | |
late | -0.020 |
seek | -0.018 |
quiet | -0.017 |
Lisa | -0.016 |
Lisa | -0.016 |
Token | Mark |
Token ID | 2940 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
late | +0.011 |
seek | +0.010 |
quiet | +0.010 |
aky | +0.009 |
too | +0.009 |
Neg logit contributions | |
imet | -0.008 |
Ow | -0.007 |
pen | -0.007 |
penn | -0.007 |
ian | -0.007 |
Token | truly |
Token ID | 4988 |
Feature activation | -1.99e-01 |
Pos logit contributions | |
Ow | +0.051 |
trigger | +0.050 |
penn | +0.049 |
imet | +0.049 |
uf | +0.048 |
Neg logit contributions | |
quiet | -0.044 |
late | -0.044 |
seek | -0.041 |
too | -0.040 |
soon | -0.040 |
Token | enjoyed |
Token ID | 8359 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
late | +0.007 |
seek | +0.006 |
Lisa | +0.006 |
quiet | +0.006 |
Lisa | +0.006 |
Neg logit contributions | |
penn | -0.005 |
pen | -0.005 |
ian | -0.005 |
imet | -0.005 |
vered | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
late | +0.005 |
seek | +0.004 |
quiet | +0.004 |
Lisa | +0.004 |
Lisa | +0.004 |
Neg logit contributions | |
penn | -0.004 |
pen | -0.003 |
imet | -0.003 |
Pe | -0.003 |
ian | -0.003 |
Token | ride |
Token ID | 6594 |
Feature activation | +3.23e-01 |
Pos logit contributions | |
penn | +0.023 |
imet | +0.023 |
pen | +0.023 |
ian | +0.022 |
Ow | +0.021 |
Neg logit contributions | |
late | -0.051 |
quiet | -0.047 |
seek | -0.047 |
Lisa | -0.044 |
too | -0.043 |
Token | , |
Token ID | 11 |
Feature activation | -4.89e-02 |
Pos logit contributions | |
pen | +0.010 |
penn | +0.010 |
ian | +0.010 |
imet | +0.010 |
ome | +0.010 |
Neg logit contributions | |
late | -0.009 |
seek | -0.008 |
quiet | -0.008 |
Lisa | -0.008 |
Happy | -0.007 |
Token | that |
Token ID | 326 |
Feature activation | -5.23e-01 |
Pos logit contributions | |
penn | +0.006 |
imet | +0.006 |
Pe | +0.005 |
pen | +0.005 |
ome | +0.005 |
Neg logit contributions | |
late | -0.008 |
aky | -0.008 |
seek | -0.008 |
quiet | -0.008 |
Lisa | -0.007 |
Token | Mom |
Token ID | 11254 |
Feature activation | -7.96e-02 |
Pos logit contributions | |
late | +0.019 |
lic | +0.019 |
aky | +0.018 |
seek | +0.017 |
aking | +0.017 |
Neg logit contributions | |
penn | -0.011 |
anybody | -0.011 |
anyone | -0.011 |
Pe | -0.010 |
rest | -0.010 |
Token | my |
Token ID | 1820 |
Feature activation | +6.02e-01 |
Pos logit contributions | |
late | +0.003 |
seek | +0.003 |
educational | +0.003 |
quiet | +0.003 |
Happy | +0.002 |
Neg logit contributions | |
ian | -0.002 |
wouldn | -0.002 |
pen | -0.002 |
Pe | -0.002 |
imet | -0.002 |
Token | had |
Token ID | 550 |
Feature activation | +1.45e-02 |
Pos logit contributions | |
ian | +0.015 |
pen | +0.014 |
penn | +0.014 |
imet | +0.014 |
ome | +0.013 |
Neg logit contributions | |
late | -0.022 |
seek | -0.020 |
quiet | -0.019 |
Lisa | -0.019 |
Lisa | -0.019 |
Token | made |
Token ID | 925 |
Feature activation | -3.27e-01 |
Pos logit contributions | |
penn | +0.000 |
pen | +0.000 |
Pe | +0.000 |
imet | +0.000 |
ian | +0.000 |
Neg logit contributions | |
late | -0.000 |
seek | -0.000 |
Lisa | -0.000 |
quiet | -0.000 |
Lisa | -0.000 |
Token | something |
Token ID | 1223 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
seek | +0.011 |
late | +0.010 |
aky | +0.009 |
lic | +0.009 |
quiet | +0.009 |
Neg logit contributions | |
pen | -0.008 |
penn | -0.008 |
anybody | -0.008 |
Ow | -0.008 |
anyone | -0.008 |
Token | special |
Token ID | 2041 |
Feature activation | +3.44e-01 |
Pos logit contributions | |
penn | +0.057 |
imet | +0.052 |
pen | +0.051 |
ome | +0.050 |
Pe | +0.050 |
Neg logit contributions | |
late | -0.043 |
seek | -0.039 |
Lisa | -0.037 |
Lisa | -0.036 |
quiet | -0.035 |
Token | just |
Token ID | 655 |
Feature activation | +8.04e-01 |
Pos logit contributions | |
penn | +0.009 |
ome | +0.008 |
imet | +0.008 |
pen | +0.008 |
ian | +0.007 |
Neg logit contributions | |
late | -0.012 |
Lisa | -0.011 |
lic | -0.011 |
quiet | -0.011 |
aky | -0.011 |
Token | for |
Token ID | 329 |
Feature activation | +1.88e-02 |
Pos logit contributions | |
penn | +0.025 |
imet | +0.024 |
Pe | +0.024 |
brow | +0.023 |
pen | +0.023 |
Neg logit contributions | |
late | -0.022 |
seek | -0.022 |
Lisa | -0.021 |
quiet | -0.019 |
lic | -0.019 |
Token | her |
Token ID | 607 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
penn | +0.000 |
anybody | +0.000 |
anyone | +0.000 |
pen | +0.000 |
brow | +0.000 |
Neg logit contributions | |
late | -0.001 |
lic | -0.001 |
aky | -0.001 |
aking | -0.001 |
seek | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -7.56e-02 |
Pos logit contributions | |
penn | +0.006 |
imet | +0.006 |
ian | +0.005 |
pen | +0.005 |
brow | +0.005 |
Neg logit contributions | |
late | -0.006 |
seek | -0.006 |
Lisa | -0.005 |
quiet | -0.005 |
Lisa | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
late | +0.024 |
quiet | +0.020 |
seek | +0.020 |
Lisa | +0.020 |
too | +0.018 |
Neg logit contributions | |
imet | -0.038 |
penn | -0.038 |
pen | -0.038 |
ian | -0.037 |
Ow | -0.036 |
Token | friendly |
Token ID | 8030 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
penn | +0.004 |
ian | +0.004 |
brow | +0.004 |
imet | +0.004 |
pen | +0.004 |
Neg logit contributions | |
late | -0.006 |
lic | -0.005 |
They | -0.005 |
Lisa | -0.005 |
Lisa | -0.005 |
Token | bee |
Token ID | 20697 |
Feature activation | +4.87e-01 |
Pos logit contributions | |
penn | +0.004 |
pen | +0.004 |
imet | +0.004 |
ian | +0.004 |
Ow | +0.003 |
Neg logit contributions | |
late | -0.004 |
seek | -0.004 |
quiet | -0.004 |
Lisa | -0.004 |
Lisa | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | +3.96e-01 |
Pos logit contributions | |
pen | +0.014 |
ian | +0.013 |
penn | +0.013 |
imet | +0.013 |
trek | +0.013 |
Neg logit contributions | |
late | -0.016 |
Lisa | -0.014 |
seek | -0.013 |
quiet | -0.013 |
lic | -0.013 |
Token | went |
Token ID | 1816 |
Feature activation | +3.51e-01 |
Pos logit contributions | |
penn | +0.009 |
imet | +0.009 |
pen | +0.008 |
Pe | +0.008 |
ian | +0.008 |
Neg logit contributions | |
late | -0.015 |
quiet | -0.013 |
seek | -0.013 |
Lisa | -0.013 |
Lisa | -0.013 |
Token | inside |
Token ID | 2641 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
penn | +0.010 |
imet | +0.009 |
pen | +0.009 |
ance | +0.009 |
trek | +0.009 |
Neg logit contributions | |
late | -0.011 |
Lisa | -0.010 |
seek | -0.010 |
quiet | -0.010 |
impressed | -0.010 |
Token | with |
Token ID | 351 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
pen | +0.039 |
penn | +0.039 |
Pe | +0.038 |
brow | +0.036 |
imet | +0.036 |
Neg logit contributions | |
late | -0.047 |
seek | -0.046 |
Lisa | -0.046 |
aky | -0.045 |
lic | -0.045 |
Token | the |
Token ID | 262 |
Feature activation | -6.67e-02 |
Pos logit contributions | |
penn | +0.031 |
pen | +0.029 |
Pe | +0.029 |
ian | +0.028 |
brow | +0.028 |
Neg logit contributions | |
late | -0.031 |
seek | -0.028 |
Lisa | -0.027 |
quiet | -0.026 |
lic | -0.026 |
Token | fork |
Token ID | 15563 |
Feature activation | +6.62e-01 |
Pos logit contributions | |
late | +0.002 |
seek | +0.002 |
lic | +0.002 |
Lisa | +0.002 |
Lisa | +0.002 |
Neg logit contributions | |
penn | -0.002 |
ian | -0.002 |
imet | -0.002 |
pen | -0.002 |
brow | -0.002 |
Token | to |
Token ID | 284 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
penn | +0.019 |
pen | +0.018 |
brow | +0.018 |
ian | +0.018 |
imet | +0.018 |
Neg logit contributions | |
late | -0.022 |
Lisa | -0.019 |
seek | -0.019 |
quiet | -0.018 |
Happy | -0.018 |
Token | show |
Token ID | 905 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
pen | +0.004 |
penn | +0.004 |
imet | +0.004 |
ian | +0.004 |
Pe | +0.003 |
Neg logit contributions | |
late | -0.007 |
quiet | -0.006 |
seek | -0.006 |
Lisa | -0.006 |
Lisa | -0.006 |
Token | ies |
Token ID | 444 |
Feature activation | +5.68e-01 |
Pos logit contributions | |
late | +0.016 |
seek | +0.014 |
quiet | +0.014 |
Lisa | +0.014 |
Lisa | +0.014 |
Neg logit contributions | |
penn | -0.018 |
ian | -0.016 |
pen | -0.015 |
imet | -0.015 |
ome | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +8.90e-01 |
Pos logit contributions | |
penn | +0.016 |
ome | +0.014 |
pen | +0.014 |
cand | +0.013 |
rest | +0.013 |
Neg logit contributions | |
Lisa | -0.020 |
lic | -0.019 |
late | -0.018 |
Maria | -0.018 |
Lisa | -0.018 |
Token | I |
Token ID | 314 |
Feature activation | +2.99e-01 |
Pos logit contributions | |
penn | +0.024 |
imet | +0.023 |
pen | +0.023 |
ian | +0.022 |
Ow | +0.022 |
Neg logit contributions | |
late | -0.030 |
quiet | -0.028 |
seek | -0.027 |
Lisa | -0.026 |
Lisa | -0.025 |
Token | don |
Token ID | 836 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
pen | +0.008 |
penn | +0.008 |
bank | +0.007 |
imet | +0.007 |
ian | +0.007 |
Neg logit contributions | |
late | -0.010 |
Lisa | -0.010 |
Lisa | -0.009 |
seek | -0.009 |
quiet | -0.009 |
Token | 't |
Token ID | 470 |
Feature activation | +9.64e-01 |
Pos logit contributions | |
penn | +0.004 |
imet | +0.004 |
pen | +0.004 |
ian | +0.004 |
Ow | +0.004 |
Neg logit contributions | |
late | -0.006 |
quiet | -0.005 |
seek | -0.005 |
Lisa | -0.005 |
Lisa | -0.005 |
Token | need |
Token ID | 761 |
Feature activation | +7.21e-01 |
Pos logit contributions | |
penn | +0.026 |
pen | +0.026 |
Pe | +0.024 |
rest | +0.024 |
risk | +0.023 |
Neg logit contributions | |
late | -0.031 |
Lisa | -0.028 |
Lisa | -0.028 |
lic | -0.027 |
quiet | -0.027 |
Token | them |
Token ID | 606 |
Feature activation | +4.32e-01 |
Pos logit contributions | |
pen | +0.026 |
ian | +0.026 |
imet | +0.025 |
trek | +0.025 |
vered | +0.025 |
Neg logit contributions | |
late | -0.018 |
too | -0.016 |
quiet | -0.015 |
seek | -0.014 |
Lisa | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | +9.47e-01 |
Pos logit contributions | |
penn | +0.013 |
allowance | +0.011 |
ome | +0.011 |
imet | +0.010 |
pen | +0.010 |
Neg logit contributions | |
Lisa | -0.015 |
late | -0.014 |
lic | -0.014 |
Maria | -0.014 |
seek | -0.013 |
Token | I |
Token ID | 314 |
Feature activation | +3.83e-01 |
Pos logit contributions | |
penn | +0.025 |
imet | +0.025 |
pen | +0.024 |
Ow | +0.024 |
Pe | +0.024 |
Neg logit contributions | |
late | -0.032 |
quiet | -0.029 |
seek | -0.029 |
Lisa | -0.026 |
impressed | -0.026 |
Token | have |
Token ID | 423 |
Feature activation | +7.96e-02 |
Pos logit contributions | |
pen | +0.011 |
penn | +0.011 |
imet | +0.010 |
ian | +0.010 |
Ow | +0.010 |
Neg logit contributions | |
late | -0.012 |
Lisa | -0.011 |
seek | -0.011 |
quiet | -0.011 |
Lisa | -0.011 |
Token | enough |
Token ID | 1576 |
Feature activation | -2.54e-01 |
Pos logit contributions | |
penn | +0.003 |
imet | +0.002 |
pen | +0.002 |
Pe | +0.002 |
Ow | +0.002 |
Neg logit contributions | |
late | -0.002 |
seek | -0.002 |
quiet | -0.002 |
Lisa | -0.002 |
Lisa | -0.002 |
Token | her |
Token ID | 607 |
Feature activation | -6.41e-01 |
Pos logit contributions | |
penn | +0.024 |
imet | +0.024 |
pen | +0.024 |
ian | +0.023 |
Ow | +0.023 |
Neg logit contributions | |
late | -0.021 |
quiet | -0.019 |
seek | -0.018 |
Lisa | -0.018 |
too | -0.017 |
Token | close |
Token ID | 1969 |
Feature activation | +3.13e-01 |
Pos logit contributions | |
late | +0.021 |
quiet | +0.019 |
seek | +0.018 |
Lisa | +0.017 |
too | +0.017 |
Neg logit contributions | |
penn | -0.019 |
imet | -0.018 |
pen | -0.018 |
ian | -0.018 |
Ow | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +7.28e-02 |
Pos logit contributions | |
penn | +0.010 |
imet | +0.010 |
pen | +0.010 |
ian | +0.010 |
brow | +0.010 |
Neg logit contributions | |
late | -0.009 |
seek | -0.008 |
quiet | -0.008 |
Lisa | -0.007 |
Lisa | -0.007 |
Token | Grand |
Token ID | 5675 |
Feature activation | +3.19e-01 |
Pos logit contributions | |
imet | +0.002 |
penn | +0.002 |
pen | +0.002 |
Ow | +0.002 |
Pe | +0.002 |
Neg logit contributions | |
late | -0.002 |
seek | -0.002 |
quiet | -0.002 |
aky | -0.002 |
Lisa | -0.002 |
Token | ma |
Token ID | 2611 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
penn | +0.012 |
ian | +0.012 |
pen | +0.012 |
imet | +0.012 |
ome | +0.012 |
Neg logit contributions | |
late | -0.007 |
seek | -0.006 |
quiet | -0.006 |
Lisa | -0.005 |
too | -0.005 |
Token | whispered |
Token ID | 25029 |
Feature activation | +2.87e-01 |
Pos logit contributions | |
pen | +0.004 |
ian | +0.004 |
penn | +0.004 |
imet | +0.004 |
brow | +0.003 |
Neg logit contributions | |
late | -0.005 |
seek | -0.004 |
quiet | -0.004 |
Lisa | -0.004 |
Lisa | -0.004 |
Token | in |
Token ID | 287 |
Feature activation | -1.93e-01 |
Pos logit contributions | |
penn | +0.009 |
imet | +0.008 |
pen | +0.008 |
ian | +0.008 |
brow | +0.008 |
Neg logit contributions | |
late | -0.009 |
seek | -0.008 |
quiet | -0.008 |
Lisa | -0.007 |
Lisa | -0.007 |
Token | Mar |
Token ID | 1526 |
Feature activation | +2.21e-02 |
Pos logit contributions | |
late | +0.006 |
aking | +0.006 |
seek | +0.006 |
lic | +0.006 |
aky | +0.005 |
Neg logit contributions | |
penn | -0.005 |
Pe | -0.005 |
growth | -0.005 |
pen | -0.005 |
brow | -0.005 |
Token | cy |
Token ID | 948 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
ian | +0.001 |
imet | +0.001 |
pen | +0.001 |
penn | +0.001 |
ome | +0.001 |
Neg logit contributions | |
late | -0.001 |
seek | -0.001 |
quiet | -0.001 |
Lisa | -0.001 |
too | -0.001 |
Token | 's |
Token ID | 338 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
late | +0.004 |
seek | +0.004 |
quiet | +0.004 |
Lisa | +0.003 |
Lisa | +0.003 |
Neg logit contributions | |
penn | -0.004 |
pen | -0.004 |
imet | -0.004 |
ian | -0.004 |
brow | -0.003 |
Token | ear |
Token ID | 1027 |
Feature activation | +7.71e-01 |
Pos logit contributions | |
late | +0.004 |
seek | +0.004 |
quiet | +0.003 |
Lisa | +0.003 |
Lisa | +0.003 |
Neg logit contributions | |
penn | -0.003 |
imet | -0.003 |
pen | -0.003 |
ian | -0.003 |
brow | -0.003 |
Token | them |
Token ID | 606 |
Feature activation | -4.99e-02 |
Pos logit contributions | |
late | +0.028 |
seek | +0.028 |
lic | +0.028 |
Lisa | +0.027 |
impressed | +0.025 |
Neg logit contributions | |
penn | -0.029 |
Ow | -0.025 |
Pe | -0.024 |
imet | -0.024 |
pen | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
Lisa | +0.001 |
late | +0.001 |
seek | +0.001 |
lic | +0.001 |
Lisa | +0.001 |
Neg logit contributions | |
penn | -0.002 |
pen | -0.001 |
imet | -0.001 |
ome | -0.001 |
ian | -0.001 |
Token | made |
Token ID | 925 |
Feature activation | -3.50e-01 |
Pos logit contributions | |
penn | +0.005 |
imet | +0.005 |
pen | +0.005 |
Pe | +0.005 |
Ow | +0.005 |
Neg logit contributions | |
late | -0.006 |
seek | -0.006 |
Lisa | -0.005 |
quiet | -0.005 |
lic | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | -9.30e-02 |
Pos logit contributions | |
seek | +0.011 |
late | +0.011 |
Lisa | +0.009 |
quiet | +0.009 |
aky | +0.009 |
Neg logit contributions | |
penn | -0.011 |
Ow | -0.010 |
imet | -0.010 |
Pe | -0.010 |
pen | -0.009 |
Token | mess |
Token ID | 2085 |
Feature activation | -9.14e-01 |
Pos logit contributions | |
late | +0.003 |
seek | +0.002 |
Lisa | +0.002 |
quiet | +0.002 |
Lisa | +0.002 |
Neg logit contributions | |
penn | -0.003 |
imet | -0.003 |
pen | -0.003 |
Ow | -0.003 |
trigger | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
catchy | +0.031 |
seek | +0.031 |
Lisa | +0.031 |
Rachel | +0.030 |
Maria | +0.030 |
Neg logit contributions | |
penn | -0.020 |
pen | -0.019 |
ian | -0.019 |
ome | -0.018 |
ors | -0.018 |
Token | Go |
Token ID | 1514 |
Feature activation | +7.11e-01 |
Pos logit contributions | |
penn | +0.005 |
imet | +0.005 |
pen | +0.005 |
Ow | +0.005 |
Pe | +0.005 |
Neg logit contributions | |
late | -0.006 |
quiet | -0.006 |
seek | -0.006 |
Lisa | -0.005 |
too | -0.005 |
Token | away |
Token ID | 1497 |
Feature activation | +3.78e-01 |
Pos logit contributions | |
pen | +0.022 |
imet | +0.021 |
penn | +0.021 |
ian | +0.021 |
Ow | +0.021 |
Neg logit contributions | |
late | -0.020 |
seek | -0.019 |
Lisa | -0.018 |
quiet | -0.018 |
impressed | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | +6.19e-01 |
Pos logit contributions | |
penn | +0.010 |
Pe | +0.009 |
pen | +0.009 |
Ow | +0.009 |
uf | +0.008 |
Neg logit contributions | |
lic | -0.012 |
catchy | -0.011 |
late | -0.011 |
Lisa | -0.011 |
Happy | -0.011 |
Token | don |
Token ID | 836 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
penn | +0.016 |
imet | +0.015 |
pen | +0.014 |
Pe | +0.014 |
bank | +0.014 |
Neg logit contributions | |
late | -0.022 |
seek | -0.019 |
quiet | -0.019 |
lic | -0.019 |
Lisa | -0.019 |
Token | 't |
Token ID | 470 |
Feature activation | +6.15e-01 |
Pos logit contributions | |
penn | +0.000 |
imet | +0.000 |
pen | +0.000 |
ian | +0.000 |
Ow | +0.000 |
Neg logit contributions | |
late | -0.000 |
quiet | -0.000 |
seek | -0.000 |
Lisa | -0.000 |
too | -0.000 |
Token | beside |
Token ID | 13970 |
Feature activation | +5.54e-01 |
Pos logit contributions | |
penn | +0.029 |
imet | +0.029 |
ian | +0.028 |
brow | +0.028 |
pen | +0.028 |
Neg logit contributions | |
late | -0.027 |
seek | -0.024 |
quiet | -0.024 |
Lisa | -0.023 |
Lisa | -0.022 |
Token | it |
Token ID | 340 |
Feature activation | +9.42e-03 |
Pos logit contributions | |
penn | +0.017 |
imet | +0.017 |
pen | +0.017 |
ian | +0.016 |
Ow | +0.016 |
Neg logit contributions | |
late | -0.017 |
quiet | -0.015 |
seek | -0.015 |
Lisa | -0.014 |
too | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | +3.77e-02 |
Pos logit contributions | |
pen | +0.000 |
penn | +0.000 |
brow | +0.000 |
imet | +0.000 |
ian | +0.000 |
Neg logit contributions | |
late | -0.000 |
seek | -0.000 |
Lisa | -0.000 |
quiet | -0.000 |
Happy | -0.000 |
Token | She |
Token ID | 1375 |
Feature activation | +3.62e-01 |
Pos logit contributions | |
penn | +0.001 |
imet | +0.001 |
pen | +0.001 |
ian | +0.001 |
Ow | +0.001 |
Neg logit contributions | |
late | -0.001 |
quiet | -0.001 |
seek | -0.001 |
Lisa | -0.001 |
Lisa | -0.001 |
Token | started |
Token ID | 2067 |
Feature activation | +4.46e-02 |
Pos logit contributions | |
penn | +0.010 |
imet | +0.010 |
pen | +0.010 |
ian | +0.009 |
brow | +0.009 |
Neg logit contributions | |
late | -0.012 |
seek | -0.011 |
quiet | -0.011 |
Lisa | -0.010 |
Lisa | -0.010 |
Token | to |
Token ID | 284 |
Feature activation | +9.77e-01 |
Pos logit contributions | |
penn | +0.001 |
imet | +0.001 |
pen | +0.001 |
ian | +0.001 |
Ow | +0.001 |
Neg logit contributions | |
late | -0.001 |
seek | -0.001 |
quiet | -0.001 |
Lisa | -0.001 |
Lisa | -0.001 |
Token | pray |
Token ID | 12472 |
Feature activation | +2.93e-01 |
Pos logit contributions | |
penn | +0.023 |
imet | +0.023 |
pen | +0.022 |
ian | +0.022 |
Ow | +0.021 |
Neg logit contributions | |
late | -0.037 |
seek | -0.033 |
quiet | -0.033 |
Lisa | -0.032 |
Lisa | -0.031 |
Token | , |
Token ID | 11 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
penn | +0.007 |
ian | +0.007 |
brow | +0.007 |
pen | +0.007 |
Pe | +0.007 |
Neg logit contributions | |
late | -0.010 |
seek | -0.009 |
Lisa | -0.009 |
Happy | -0.009 |
lic | -0.009 |
Token | wondering |
Token ID | 11263 |
Feature activation | +9.39e-01 |
Pos logit contributions | |
Pe | +0.004 |
imet | +0.004 |
brow | +0.004 |
penn | +0.003 |
pen | +0.003 |
Neg logit contributions | |
late | -0.004 |
seek | -0.004 |
Lisa | -0.003 |
Happy | -0.003 |
aky | -0.003 |
Token | if |
Token ID | 611 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
penn | +0.028 |
imet | +0.026 |
pen | +0.026 |
ian | +0.026 |
Pe | +0.025 |
Neg logit contributions | |
late | -0.030 |
seek | -0.028 |
quiet | -0.026 |
Lisa | -0.025 |
Happy | -0.024 |
Token | the |
Token ID | 262 |
Feature activation | +5.99e-02 |
Pos logit contributions | |
penn | +0.004 |
imet | +0.004 |
pen | +0.004 |
ian | +0.004 |
Ow | +0.004 |
Neg logit contributions | |
late | -0.005 |
seek | -0.004 |
quiet | -0.004 |
Lisa | -0.004 |
Lisa | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
penn | +0.026 |
imet | +0.025 |
pen | +0.025 |
Ow | +0.024 |
ian | +0.024 |
Neg logit contributions | |
late | -0.022 |
seek | -0.021 |
quiet | -0.019 |
Lisa | -0.019 |
Lisa | -0.019 |
Token | wood |
Token ID | 4898 |
Feature activation | +6.91e-01 |
Pos logit contributions | |
penn | +0.007 |
pen | +0.006 |
imet | +0.006 |
ian | +0.006 |
Ow | +0.006 |
Neg logit contributions | |
late | -0.009 |
quiet | -0.008 |
seek | -0.008 |
Lisa | -0.007 |
too | -0.007 |
Token | from |
Token ID | 422 |
Feature activation | -2.53e-02 |
Pos logit contributions | |
penn | +0.021 |
imet | +0.020 |
bank | +0.019 |
pen | +0.019 |
ian | +0.019 |
Neg logit contributions | |
late | -0.022 |
Lisa | -0.020 |
seek | -0.020 |
Happy | -0.019 |
Lisa | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | +3.74e-01 |
Pos logit contributions | |
late | +0.001 |
seek | +0.001 |
quiet | +0.001 |
Lisa | +0.001 |
Lisa | +0.001 |
Neg logit contributions | |
penn | -0.001 |
imet | -0.001 |
pen | -0.001 |
ian | -0.001 |
Ow | -0.001 |
Token | trees |
Token ID | 7150 |
Feature activation | +5.79e-01 |
Pos logit contributions | |
penn | +0.010 |
imet | +0.010 |
pen | +0.010 |
ian | +0.009 |
Ow | +0.009 |
Neg logit contributions | |
late | -0.013 |
quiet | -0.012 |
seek | -0.012 |
Lisa | -0.011 |
too | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -3.62e-01 |
Pos logit contributions | |
pen | +0.017 |
penn | +0.017 |
imet | +0.017 |
bank | +0.016 |
ian | +0.016 |
Neg logit contributions | |
late | -0.018 |
seek | -0.016 |
Lisa | -0.016 |
quiet | -0.016 |
Lisa | -0.016 |
Token | After |
Token ID | 2293 |
Feature activation | +8.93e-01 |
Pos logit contributions | |
late | +0.010 |
quiet | +0.008 |
seek | +0.008 |
Lisa | +0.008 |
too | +0.008 |
Neg logit contributions | |
penn | -0.012 |
pen | -0.012 |
imet | -0.012 |
ian | -0.012 |
Ow | -0.012 |
Token | he |
Token ID | 339 |
Feature activation | +4.57e-01 |
Pos logit contributions | |
penn | +0.029 |
imet | +0.028 |
pen | +0.027 |
ian | +0.026 |
Pe | +0.026 |
Neg logit contributions | |
late | -0.026 |
seek | -0.023 |
quiet | -0.023 |
Lisa | -0.022 |
Lisa | -0.022 |
Token | was |
Token ID | 373 |
Feature activation | +2.91e-01 |
Pos logit contributions | |
trigger | +0.016 |
penn | +0.015 |
brow | +0.015 |
Ow | +0.015 |
imet | +0.015 |
Neg logit contributions | |
late | -0.012 |
quiet | -0.011 |
soon | -0.011 |
seek | -0.011 |
too | -0.011 |
Token | done |
Token ID | 1760 |
Feature activation | +2.72e-01 |
Pos logit contributions | |
pen | +0.009 |
imet | +0.009 |
penn | +0.009 |
trek | +0.009 |
Pe | +0.009 |
Neg logit contributions | |
late | -0.007 |
Lisa | -0.007 |
seek | -0.006 |
Lisa | -0.006 |
Happy | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
penn | +0.008 |
imet | +0.008 |
pen | +0.008 |
ian | +0.008 |
Ow | +0.008 |
Neg logit contributions | |
late | -0.008 |
quiet | -0.007 |
seek | -0.007 |
Lisa | -0.007 |
too | -0.007 |
Token | want |
Token ID | 765 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
late | +0.014 |
seek | +0.013 |
Lisa | +0.012 |
quiet | +0.012 |
Lisa | +0.012 |
Neg logit contributions | |
pen | -0.011 |
penn | -0.010 |
imet | -0.010 |
ian | -0.010 |
Ow | -0.010 |
Token | it |
Token ID | 340 |
Feature activation | +8.15e-02 |
Pos logit contributions | |
penn | +0.004 |
imet | +0.004 |
Ow | +0.004 |
pen | +0.004 |
Pe | +0.004 |
Neg logit contributions | |
late | -0.004 |
seek | -0.004 |
quiet | -0.004 |
lic | -0.004 |
Lisa | -0.004 |
Token | too |
Token ID | 1165 |
Feature activation | -2.77e-01 |
Pos logit contributions | |
anyone | +0.002 |
anybody | +0.002 |
Twe | +0.002 |
penn | +0.002 |
Pe | +0.002 |
Neg logit contributions | |
late | -0.003 |
Lisa | -0.003 |
Maria | -0.002 |
pans | -0.002 |
lic | -0.002 |
Token | !" |
Token ID | 2474 |
Feature activation | +3.18e-01 |
Pos logit contributions | |
late | +0.007 |
seek | +0.007 |
Lisa | +0.006 |
Lisa | +0.006 |
lic | +0.006 |
Neg logit contributions | |
penn | -0.009 |
Ow | -0.009 |
pen | -0.009 |
imet | -0.008 |
brow | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -8.61e-03 |
Pos logit contributions | |
imet | +0.009 |
Ow | +0.009 |
Pe | +0.009 |
penn | +0.009 |
pen | +0.009 |
Neg logit contributions | |
late | -0.010 |
seek | -0.010 |
quiet | -0.009 |
aky | -0.008 |
awesome | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
late | +0.000 |
seek | +0.000 |
quiet | +0.000 |
Lisa | +0.000 |
too | +0.000 |
Neg logit contributions | |
imet | -0.000 |
pen | -0.000 |
penn | -0.000 |
Ow | -0.000 |
ian | -0.000 |
Token | Mom |
Token ID | 29252 |
Feature activation | -5.47e-01 |
Pos logit contributions | |
late | +0.004 |
quiet | +0.003 |
seek | +0.003 |
Lisa | +0.003 |
too | +0.003 |
Neg logit contributions | |
pen | -0.004 |
Twe | -0.004 |
Ow | -0.004 |
imet | -0.004 |
penn | -0.004 |
Token | came |
Token ID | 1625 |
Feature activation | -3.74e-01 |
Pos logit contributions | |
late | +0.017 |
seek | +0.015 |
quiet | +0.014 |
Lisa | +0.013 |
Happy | +0.013 |
Neg logit contributions | |
ian | -0.017 |
pen | -0.017 |
imet | -0.017 |
penn | -0.016 |
brow | -0.016 |
Token | in |
Token ID | 287 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
educational | +0.011 |
late | +0.011 |
seek | +0.011 |
Happy | +0.011 |
catchy | +0.011 |
Neg logit contributions | |
brow | -0.010 |
knees | -0.010 |
pen | -0.010 |
Pe | -0.009 |
yle | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | +7.56e-01 |
Pos logit contributions | |
pen | +0.005 |
penn | +0.005 |
Pe | +0.005 |
uf | +0.005 |
cash | +0.005 |
Neg logit contributions | |
seek | -0.006 |
lic | -0.006 |
aky | -0.006 |
catchy | -0.006 |
awesome | -0.006 |
Token | said |
Token ID | 531 |
Feature activation | +5.53e-01 |
Pos logit contributions | |
ian | +0.022 |
pen | +0.022 |
imet | +0.022 |
penn | +0.021 |
Pe | +0.021 |
Neg logit contributions | |
late | -0.023 |
seek | -0.021 |
quiet | -0.020 |
awesome | -0.019 |
aky | -0.018 |
Token | new |
Token ID | 649 |
Feature activation | +3.87e-02 |
Pos logit contributions | |
late | +0.001 |
quiet | +0.001 |
seek | +0.001 |
Lisa | +0.001 |
too | +0.001 |
Neg logit contributions | |
pen | -0.001 |
penn | -0.001 |
imet | -0.001 |
ian | -0.001 |
Ow | -0.001 |
Token | doll |
Token ID | 3654 |
Feature activation | -4.32e-01 |
Pos logit contributions | |
penn | +0.001 |
imet | +0.001 |
pen | +0.001 |
ian | +0.001 |
Ow | +0.001 |
Neg logit contributions | |
late | -0.001 |
seek | -0.001 |
quiet | -0.001 |
Lisa | -0.001 |
Lisa | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
late | +0.013 |
quiet | +0.012 |
seek | +0.012 |
Lisa | +0.011 |
Lisa | +0.011 |
Neg logit contributions | |
penn | -0.013 |
pen | -0.013 |
imet | -0.013 |
ian | -0.013 |
ome | -0.012 |
Token | She |
Token ID | 1375 |
Feature activation | -3.16e-01 |
Pos logit contributions | |
late | +0.007 |
Lisa | +0.007 |
quiet | +0.006 |
seek | +0.006 |
Lisa | +0.006 |
Neg logit contributions | |
penn | -0.009 |
imet | -0.009 |
pen | -0.009 |
ian | -0.009 |
ome | -0.009 |
Token | is |
Token ID | 318 |
Feature activation | -4.46e-01 |
Pos logit contributions | |
late | +0.010 |
quiet | +0.009 |
seek | +0.008 |
Lisa | +0.008 |
too | +0.008 |
Neg logit contributions | |
penn | -0.010 |
imet | -0.010 |
pen | -0.009 |
ian | -0.009 |
Ow | -0.009 |
Token | very |
Token ID | 845 |
Feature activation | -7.96e-01 |
Pos logit contributions | |
late | +0.011 |
seek | +0.010 |
Lisa | +0.009 |
Happy | +0.009 |
quiet | +0.009 |
Neg logit contributions | |
penn | -0.016 |
Pe | -0.015 |
Peck | -0.015 |
pen | -0.015 |
yle | -0.015 |
Token | happy |
Token ID | 3772 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
late | +0.018 |
seek | +0.016 |
Lisa | +0.014 |
Lisa | +0.014 |
Maria | +0.013 |
Neg logit contributions | |
penn | -0.031 |
yle | -0.028 |
Pe | -0.028 |
ian | -0.027 |
pen | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | -4.41e-01 |
Pos logit contributions | |
late | +0.004 |
lic | +0.004 |
seek | +0.004 |
Lisa | +0.004 |
aky | +0.004 |
Neg logit contributions | |
penn | -0.004 |
pen | -0.003 |
imet | -0.003 |
knees | -0.003 |
ome | -0.003 |
Token | She |
Token ID | 1375 |
Feature activation | -4.76e-01 |
Pos logit contributions | |
late | +0.011 |
Lisa | +0.010 |
quiet | +0.010 |
seek | +0.009 |
Lisa | +0.009 |
Neg logit contributions | |
penn | -0.016 |
imet | -0.016 |
pen | -0.015 |
ian | -0.015 |
ome | -0.015 |
Token | shows |
Token ID | 2523 |
Feature activation | +5.68e-01 |
Pos logit contributions | |
late | +0.015 |
quiet | +0.013 |
seek | +0.013 |
Lisa | +0.012 |
too | +0.012 |
Neg logit contributions | |
penn | -0.014 |
imet | -0.014 |
pen | -0.014 |
ian | -0.014 |
Ow | -0.013 |
Token | her |
Token ID | 607 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
imet | +0.019 |
penn | +0.019 |
pen | +0.019 |
ian | +0.018 |
ome | +0.018 |
Neg logit contributions | |
late | -0.015 |
quiet | -0.014 |
Lisa | -0.013 |
seek | -0.013 |
too | -0.013 |
Token | dark |
Token ID | 3223 |
Feature activation | +5.14e-01 |
Pos logit contributions | |
penn | +0.000 |
imet | +0.000 |
pen | +0.000 |
Ow | +0.000 |
ian | +0.000 |
Neg logit contributions | |
late | -0.000 |
seek | -0.000 |
quiet | -0.000 |
Lisa | -0.000 |
Lisa | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
pen | +0.014 |
imet | +0.014 |
penn | +0.014 |
ian | +0.014 |
Pe | +0.013 |
Neg logit contributions | |
late | -0.015 |
seek | -0.014 |
quiet | -0.014 |
Lisa | -0.014 |
pleased | -0.013 |
Token | Then |
Token ID | 3244 |
Feature activation | -7.01e-01 |
Pos logit contributions | |
late | +0.004 |
quiet | +0.004 |
seek | +0.004 |
Lisa | +0.004 |
too | +0.004 |
Neg logit contributions | |
penn | -0.004 |
pen | -0.004 |
imet | -0.004 |
ian | -0.004 |
yle | -0.004 |
Token | Lucy |
Token ID | 22162 |
Feature activation | -3.45e-01 |
Pos logit contributions | |
late | +0.022 |
seek | +0.019 |
quiet | +0.019 |
Lisa | +0.018 |
Lisa | +0.017 |
Neg logit contributions | |
penn | -0.021 |
imet | -0.021 |
pen | -0.021 |
ian | -0.020 |
Ow | -0.020 |
Token | felt |
Token ID | 2936 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
late | +0.012 |
seek | +0.011 |
quiet | +0.011 |
Lisa | +0.010 |
Lisa | +0.010 |
Neg logit contributions | |
pen | -0.009 |
ian | -0.009 |
imet | -0.009 |
penn | -0.009 |
brow | -0.008 |
Token | very |
Token ID | 845 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
penn | +0.007 |
imet | +0.007 |
pen | +0.007 |
ian | +0.007 |
Ow | +0.007 |
Neg logit contributions | |
late | -0.005 |
seek | -0.004 |
quiet | -0.004 |
Lisa | -0.004 |
Lisa | -0.004 |
Token | sleepy |
Token ID | 42023 |
Feature activation | -2.64e-01 |
Pos logit contributions | |
late | +0.027 |
quiet | +0.023 |
seek | +0.021 |
too | +0.019 |
Lisa | +0.019 |
Neg logit contributions | |
imet | -0.049 |
pen | -0.049 |
penn | -0.049 |
ian | -0.048 |
Ow | -0.047 |
Token | . |
Token ID | 13 |
Feature activation | -5.36e-01 |
Pos logit contributions | |
late | +0.008 |
seek | +0.008 |
Lisa | +0.008 |
lic | +0.008 |
Happy | +0.007 |
Neg logit contributions | |
pen | -0.007 |
penn | -0.007 |
imet | -0.007 |
ian | -0.007 |
ome | -0.007 |
Token | They |
Token ID | 1119 |
Feature activation | -3.83e-01 |
Pos logit contributions | |
late | +0.014 |
Lisa | +0.012 |
quiet | +0.012 |
seek | +0.011 |
too | +0.011 |
Neg logit contributions | |
penn | -0.020 |
pen | -0.019 |
imet | -0.019 |
ian | -0.018 |
trigger | -0.018 |
Token | all |
Token ID | 477 |
Feature activation | -7.94e-01 |
Pos logit contributions | |
late | +0.014 |
quiet | +0.013 |
seek | +0.013 |
Lisa | +0.012 |
too | +0.012 |
Neg logit contributions | |
penn | -0.010 |
imet | -0.010 |
pen | -0.009 |
ian | -0.009 |
Ow | -0.009 |
Token | agreed |
Token ID | 4987 |
Feature activation | -5.69e-01 |
Pos logit contributions | |
late | +0.031 |
seek | +0.028 |
quiet | +0.028 |
Lisa | +0.027 |
Lisa | +0.026 |
Neg logit contributions | |
penn | -0.017 |
imet | -0.017 |
pen | -0.017 |
ian | -0.016 |
Ow | -0.016 |
Token | , |
Token ID | 11 |
Feature activation | -3.58e-01 |
Pos logit contributions | |
yle | +0.039 |
bank | +0.038 |
bars | +0.038 |
iped | +0.038 |
ian | +0.038 |
Neg logit contributions | |
late | -0.013 |
too | -0.011 |
together | -0.010 |
quiet | -0.010 |
seek | -0.008 |
Token | Benny |
Token ID | 44275 |
Feature activation | +2.97e-01 |
Pos logit contributions | |
late | +0.012 |
quiet | +0.010 |
seek | +0.010 |
too | +0.010 |
together | +0.009 |
Neg logit contributions | |
penn | -0.011 |
imet | -0.011 |
pen | -0.011 |
ome | -0.011 |
ian | -0.011 |
Token | saw |
Token ID | 2497 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
penn | +0.010 |
imet | +0.009 |
pen | +0.009 |
ian | +0.009 |
Ow | +0.009 |
Neg logit contributions | |
late | -0.009 |
quiet | -0.008 |
seek | -0.008 |
Lisa | -0.007 |
too | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | -3.59e-02 |
Pos logit contributions | |
late | +0.004 |
quiet | +0.004 |
too | +0.003 |
Lisa | +0.003 |
seek | +0.003 |
Neg logit contributions | |
imet | -0.010 |
ian | -0.010 |
penn | -0.010 |
pen | -0.010 |
ome | -0.010 |
Token | big |
Token ID | 1263 |
Feature activation | -7.37e-01 |
Pos logit contributions | |
late | +0.001 |
quiet | +0.001 |
seek | +0.001 |
Lisa | +0.001 |
too | +0.001 |
Neg logit contributions | |
imet | -0.001 |
penn | -0.001 |
pen | -0.001 |
ian | -0.001 |
Ow | -0.001 |
Token | parade |
Token ID | 16134 |
Feature activation | -7.80e-01 |
Pos logit contributions | |
late | +0.024 |
quiet | +0.021 |
seek | +0.021 |
Lisa | +0.020 |
too | +0.019 |
Neg logit contributions | |
penn | -0.021 |
imet | -0.021 |
pen | -0.021 |
ian | -0.020 |
Ow | -0.020 |
Token | in |
Token ID | 287 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
late | +0.022 |
seek | +0.019 |
quiet | +0.019 |
Lisa | +0.018 |
Lisa | +0.018 |
Neg logit contributions | |
pen | -0.025 |
penn | -0.024 |
imet | -0.024 |
ian | -0.024 |
ome | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | +3.80e-01 |
Pos logit contributions | |
pleased | +0.034 |
seek | +0.034 |
lic | +0.034 |
late | +0.034 |
aky | +0.033 |
Neg logit contributions | |
growth | -0.030 |
Pe | -0.029 |
pen | -0.029 |
Ow | -0.029 |
penn | -0.028 |
Token | town |
Token ID | 3240 |
Feature activation | -7.15e-01 |
Pos logit contributions | |
penn | +0.010 |
pen | +0.010 |
ian | +0.009 |
Ow | +0.009 |
imet | +0.009 |
Neg logit contributions | |
late | -0.013 |
seek | -0.012 |
Lisa | -0.011 |
quiet | -0.011 |
Lisa | -0.011 |
Token | square |
Token ID | 6616 |
Feature activation | -7.93e-01 |
Pos logit contributions | |
late | +0.024 |
seek | +0.023 |
Lisa | +0.022 |
Lisa | +0.021 |
Happy | +0.021 |
Neg logit contributions | |
pen | -0.017 |
bank | -0.017 |
penn | -0.016 |
yle | -0.015 |
je | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
late | +0.025 |
Lisa | +0.023 |
Lisa | +0.023 |
seek | +0.023 |
pleased | +0.022 |
Neg logit contributions | |
pen | -0.022 |
penn | -0.022 |
bank | -0.021 |
ome | -0.020 |
trek | -0.020 |
Token | fight |
Token ID | 1907 |
Feature activation | -7.16e-01 |
Pos logit contributions | |
late | +0.030 |
seek | +0.027 |
quiet | +0.026 |
Lisa | +0.025 |
Lisa | +0.025 |
Neg logit contributions | |
penn | -0.025 |
pen | -0.024 |
imet | -0.024 |
ian | -0.023 |
Ow | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -6.58e-01 |
Pos logit contributions | |
seek | +0.017 |
late | +0.017 |
pleased | +0.014 |
pans | +0.014 |
Happy | +0.014 |
Neg logit contributions | |
pen | -0.024 |
Ow | -0.024 |
penn | -0.023 |
imet | -0.023 |
anyone | -0.023 |
Token | They |
Token ID | 1119 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
late | +0.022 |
seek | +0.020 |
quiet | +0.020 |
Lisa | +0.018 |
Lisa | +0.018 |
Neg logit contributions | |
penn | -0.018 |
imet | -0.018 |
pen | -0.018 |
Ow | -0.017 |
ian | -0.017 |
Token | fight |
Token ID | 1907 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
late | +0.049 |
seek | +0.043 |
quiet | +0.043 |
Lisa | +0.041 |
Lisa | +0.039 |
Neg logit contributions | |
penn | -0.043 |
pen | -0.042 |
imet | -0.042 |
ian | -0.040 |
Ow | -0.039 |
Token | over |
Token ID | 625 |
Feature activation | -7.84e-01 |
Pos logit contributions | |
seek | +0.027 |
late | +0.025 |
pleased | +0.022 |
aky | +0.021 |
pans | +0.021 |
Neg logit contributions | |
pen | -0.035 |
Ow | -0.034 |
imet | -0.033 |
penn | -0.033 |
trigger | -0.033 |
Token | toys |
Token ID | 14958 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
late | +0.027 |
seek | +0.025 |
pleased | +0.023 |
lic | +0.022 |
Happy | +0.022 |
Neg logit contributions | |
penn | -0.022 |
pen | -0.020 |
vered | -0.019 |
Ow | -0.019 |
anyone | -0.019 |
Token | and |
Token ID | 290 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
late | +0.056 |
seek | +0.053 |
lic | +0.051 |
Happy | +0.050 |
pleased | +0.050 |
Neg logit contributions | |
pen | -0.043 |
penn | -0.043 |
imet | -0.040 |
ome | -0.040 |
bank | -0.039 |
Token | snacks |
Token ID | 28487 |
Feature activation | -9.56e-01 |
Pos logit contributions | |
late | +0.050 |
lic | +0.044 |
Happy | +0.042 |
aky | +0.042 |
seek | +0.041 |
Neg logit contributions | |
penn | -0.026 |
bank | -0.025 |
knees | -0.025 |
pen | -0.024 |
imet | -0.024 |
Token | and |
Token ID | 290 |
Feature activation | -6.34e-01 |
Pos logit contributions | |
late | +0.033 |
seek | +0.032 |
Maria | +0.030 |
Lisa | +0.030 |
lic | +0.030 |
Neg logit contributions | |
penn | -0.023 |
pen | -0.023 |
imet | -0.022 |
ome | -0.022 |
bank | -0.021 |
Token | TV |
Token ID | 3195 |
Feature activation | -5.49e-01 |
Pos logit contributions | |
late | +0.024 |
lic | +0.021 |
seek | +0.020 |
Happy | +0.020 |
Lisa | +0.020 |
Neg logit contributions | |
penn | -0.015 |
bank | -0.014 |
imet | -0.014 |
knees | -0.014 |
allowance | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -7.46e-01 |
Pos logit contributions | |
late | +0.018 |
seek | +0.018 |
Maria | +0.017 |
pleased | +0.016 |
Lisa | +0.016 |
Neg logit contributions | |
penn | -0.014 |
pen | -0.014 |
imet | -0.014 |
bank | -0.013 |
ian | -0.013 |
Token | felt |
Token ID | 2936 |
Feature activation | +5.49e-01 |
Pos logit contributions | |
late | +0.005 |
seek | +0.004 |
quiet | +0.004 |
Lisa | +0.004 |
Lisa | +0.004 |
Neg logit contributions | |
pen | -0.004 |
penn | -0.004 |
ian | -0.004 |
imet | -0.004 |
brow | -0.004 |
Token | really |
Token ID | 1107 |
Feature activation | -7.26e-01 |
Pos logit contributions | |
imet | +0.021 |
penn | +0.021 |
pen | +0.021 |
ian | +0.021 |
Ow | +0.020 |
Neg logit contributions | |
late | -0.013 |
quiet | -0.011 |
seek | -0.010 |
too | -0.010 |
Lisa | -0.009 |
Token | scared |
Token ID | 12008 |
Feature activation | -4.41e-01 |
Pos logit contributions | |
late | +0.018 |
seek | +0.016 |
quiet | +0.015 |
Lisa | +0.015 |
Lisa | +0.014 |
Neg logit contributions | |
penn | -0.026 |
imet | -0.025 |
pen | -0.025 |
ian | -0.025 |
Ow | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
seek | +0.013 |
late | +0.013 |
Lisa | +0.012 |
quiet | +0.012 |
Happy | +0.012 |
Neg logit contributions | |
penn | -0.012 |
pen | -0.012 |
imet | -0.012 |
Pe | -0.012 |
anybody | -0.012 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | -4.67e-01 |
Pos logit contributions | |
late | +0.009 |
quiet | +0.008 |
seek | +0.008 |
Lisa | +0.008 |
too | +0.007 |
Neg logit contributions | |
penn | -0.010 |
pen | -0.010 |
imet | -0.010 |
ian | -0.010 |
trigger | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
late | +0.012 |
quiet | +0.011 |
seek | +0.011 |
Lisa | +0.010 |
too | +0.010 |
Neg logit contributions | |
pen | -0.016 |
penn | -0.016 |
imet | -0.016 |
yle | -0.016 |
unn | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | +5.45e-01 |
Pos logit contributions | |
late | +0.049 |
quiet | +0.043 |
seek | +0.043 |
Lisa | +0.041 |
Lisa | +0.039 |
Neg logit contributions | |
penn | -0.049 |
imet | -0.049 |
pen | -0.048 |
ian | -0.047 |
Ow | -0.046 |
Token | door |
Token ID | 3420 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
penn | +0.014 |
imet | +0.014 |
ian | +0.013 |
pen | +0.013 |
Ow | +0.013 |
Neg logit contributions | |
late | -0.019 |
seek | -0.017 |
quiet | -0.017 |
Lisa | -0.016 |
Lisa | -0.016 |
Token | slammed |
Token ID | 21376 |
Feature activation | -7.64e-01 |
Pos logit contributions | |
late | +0.011 |
quiet | +0.010 |
seek | +0.010 |
Lisa | +0.009 |
Lisa | +0.009 |
Neg logit contributions | |
penn | -0.009 |
imet | -0.009 |
pen | -0.009 |
ian | -0.009 |
Ow | -0.009 |
Token | shut |
Token ID | 4423 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
late | +0.022 |
quiet | +0.019 |
seek | +0.018 |
Lisa | +0.017 |
too | +0.017 |
Neg logit contributions | |
penn | -0.026 |
imet | -0.026 |
pen | -0.025 |
Ow | -0.024 |
unn | -0.024 |
Token | and |
Token ID | 290 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
late | +0.028 |
seek | +0.026 |
quiet | +0.024 |
Lisa | +0.024 |
Happy | +0.023 |
Neg logit contributions | |
penn | -0.033 |
ian | -0.032 |
pen | -0.032 |
brow | -0.032 |
imet | -0.031 |
Token | saw |
Token ID | 2497 |
Feature activation | -2.75e-01 |
Pos logit contributions | |
late | +0.010 |
quiet | +0.009 |
seek | +0.009 |
Lisa | +0.008 |
Lisa | +0.008 |
Neg logit contributions | |
penn | -0.012 |
imet | -0.011 |
pen | -0.011 |
ian | -0.011 |
Ow | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | -4.61e-01 |
Pos logit contributions | |
late | +0.006 |
quiet | +0.005 |
seek | +0.005 |
Lisa | +0.005 |
too | +0.005 |
Neg logit contributions | |
imet | -0.011 |
pen | -0.011 |
ian | -0.011 |
penn | -0.011 |
ome | -0.010 |
Token | big |
Token ID | 1263 |
Feature activation | -9.37e-01 |
Pos logit contributions | |
late | +0.011 |
seek | +0.009 |
quiet | +0.009 |
Lisa | +0.009 |
Lisa | +0.008 |
Neg logit contributions | |
penn | -0.017 |
pen | -0.016 |
imet | -0.016 |
ian | -0.016 |
Ow | -0.016 |
Token | tree |
Token ID | 5509 |
Feature activation | -6.19e-01 |
Pos logit contributions | |
late | +0.033 |
seek | +0.029 |
quiet | +0.029 |
Lisa | +0.027 |
Lisa | +0.027 |
Neg logit contributions | |
penn | -0.025 |
imet | -0.024 |
pen | -0.024 |
ian | -0.023 |
Ow | -0.022 |
Token | with |
Token ID | 351 |
Feature activation | +9.88e-02 |
Pos logit contributions | |
late | +0.016 |
seek | +0.014 |
quiet | +0.014 |
Lisa | +0.013 |
Lisa | +0.013 |
Neg logit contributions | |
penn | -0.021 |
pen | -0.021 |
imet | -0.020 |
ian | -0.020 |
ome | -0.019 |
Token | many |
Token ID | 867 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
imet | +0.004 |
pen | +0.004 |
ian | +0.004 |
ome | +0.003 |
penn | +0.003 |
Neg logit contributions | |
late | -0.002 |
quiet | -0.002 |
seek | -0.002 |
Lisa | -0.002 |
too | -0.002 |
Token | leaves |
Token ID | 5667 |
Feature activation | -5.13e-01 |
Pos logit contributions | |
late | +0.058 |
quiet | +0.055 |
seek | +0.053 |
Lisa | +0.050 |
too | +0.050 |
Neg logit contributions | |
imet | -0.048 |
pen | -0.047 |
ian | -0.046 |
penn | -0.045 |
Ow | -0.044 |
Token | . |
Token ID | 13 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
late | +0.014 |
seek | +0.013 |
quiet | +0.012 |
Lisa | +0.012 |
Lisa | +0.012 |
Neg logit contributions | |
penn | -0.017 |
imet | -0.017 |
pen | -0.017 |
ian | -0.016 |
Ow | -0.016 |
Token | Lily |
Token ID | 20037 |
Feature activation | -6.37e-01 |
Pos logit contributions | |
penn | +0.004 |
imet | +0.004 |
pen | +0.004 |
ian | +0.004 |
ome | +0.004 |
Neg logit contributions | |
late | -0.004 |
quiet | -0.003 |
seek | -0.003 |
Lisa | -0.003 |
too | -0.003 |
Token | wanted |
Token ID | 2227 |
Feature activation | -4.62e-01 |
Pos logit contributions | |
late | +0.020 |
quiet | +0.018 |
seek | +0.018 |
Lisa | +0.017 |
too | +0.016 |
Neg logit contributions | |
penn | -0.019 |
imet | -0.019 |
pen | -0.019 |
ian | -0.018 |
Ow | -0.018 |
Token | to |
Token ID | 284 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
late | +0.012 |
quiet | +0.010 |
seek | +0.010 |
Lisa | +0.009 |
too | +0.009 |
Neg logit contributions | |
penn | -0.017 |
imet | -0.016 |
pen | -0.016 |
ian | -0.016 |
Ow | -0.016 |
Token | with |
Token ID | 351 |
Feature activation | +3.18e-01 |
Pos logit contributions | |
late | +0.013 |
seek | +0.013 |
Lisa | +0.011 |
lic | +0.011 |
Happy | +0.011 |
Neg logit contributions | |
penn | -0.013 |
imet | -0.012 |
pen | -0.012 |
ian | -0.012 |
vered | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | +6.45e-02 |
Pos logit contributions | |
penn | +0.011 |
imet | +0.011 |
pen | +0.010 |
ian | +0.010 |
Ow | +0.010 |
Neg logit contributions | |
late | -0.009 |
seek | -0.008 |
quiet | -0.007 |
Lisa | -0.007 |
Lisa | -0.007 |
Token | screws |
Token ID | 23742 |
Feature activation | -4.59e-01 |
Pos logit contributions | |
penn | +0.001 |
imet | +0.001 |
pen | +0.001 |
ian | +0.001 |
Ow | +0.001 |
Neg logit contributions | |
late | -0.003 |
seek | -0.002 |
quiet | -0.002 |
Lisa | -0.002 |
Lisa | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -6.04e-01 |
Pos logit contributions | |
late | +0.014 |
seek | +0.013 |
quiet | +0.012 |
Lisa | +0.012 |
Lisa | +0.011 |
Neg logit contributions | |
penn | -0.015 |
imet | -0.014 |
pen | -0.014 |
ian | -0.014 |
ome | -0.013 |
Token | They |
Token ID | 1119 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
late | +0.018 |
seek | +0.018 |
quiet | +0.017 |
aky | +0.016 |
pleased | +0.015 |
Neg logit contributions | |
imet | -0.018 |
Ow | -0.018 |
Pe | -0.017 |
penn | -0.017 |
Higher | -0.017 |
Token | work |
Token ID | 670 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
late | +0.038 |
seek | +0.033 |
quiet | +0.032 |
Lisa | +0.031 |
Lisa | +0.031 |
Neg logit contributions | |
penn | -0.026 |
pen | -0.026 |
imet | -0.025 |
ian | -0.025 |
trigger | -0.023 |
Token | hard |
Token ID | 1327 |
Feature activation | +4.94e-01 |
Pos logit contributions | |
seek | +0.039 |
late | +0.038 |
quiet | +0.033 |
Lisa | +0.033 |
Lisa | +0.033 |
Neg logit contributions | |
penn | -0.058 |
imet | -0.057 |
pen | -0.056 |
ian | -0.054 |
brow | -0.052 |
Token | and |
Token ID | 290 |
Feature activation | -8.38e-01 |
Pos logit contributions | |
penn | +0.016 |
imet | +0.016 |
pen | +0.016 |
ian | +0.015 |
ome | +0.014 |
Neg logit contributions | |
seek | -0.014 |
late | -0.013 |
quiet | -0.013 |
Lisa | -0.012 |
Lisa | -0.012 |
Token | have |
Token ID | 423 |
Feature activation | -6.61e-01 |
Pos logit contributions | |
late | +0.031 |
seek | +0.028 |
lic | +0.027 |
aky | +0.027 |
Maria | +0.026 |
Neg logit contributions | |
penn | -0.020 |
imet | -0.020 |
pen | -0.018 |
allowance | -0.018 |
ian | -0.017 |
Token | fun |
Token ID | 1257 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
late | +0.014 |
seek | +0.013 |
aky | +0.011 |
amer | +0.011 |
Lisa | +0.010 |
Neg logit contributions | |
penn | -0.026 |
imet | -0.025 |
pen | -0.024 |
trek | -0.024 |
brow | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | -5.89e-01 |
Pos logit contributions | |
late | +0.003 |
seek | +0.003 |
quiet | +0.003 |
Lisa | +0.003 |
Lisa | +0.003 |
Neg logit contributions | |
penn | -0.003 |
pen | -0.003 |
imet | -0.003 |
ian | -0.003 |
yle | -0.003 |
Token | being |
Token ID | 852 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
penn | +0.003 |
imet | +0.003 |
pen | +0.003 |
ian | +0.003 |
Ow | +0.003 |
Neg logit contributions | |
late | -0.003 |
quiet | -0.003 |
seek | -0.003 |
Lisa | -0.002 |
too | -0.002 |
Token | disturbed |
Token ID | 24069 |
Feature activation | -6.05e-01 |
Pos logit contributions | |
late | +0.038 |
seek | +0.037 |
Lisa | +0.035 |
Happy | +0.033 |
THE | +0.031 |
Neg logit contributions | |
penn | -0.045 |
pen | -0.042 |
Ow | -0.040 |
Pe | -0.040 |
rab | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | +3.77e-01 |
Pos logit contributions | |
late | +0.018 |
seek | +0.017 |
quiet | +0.016 |
Lisa | +0.015 |
pleased | +0.015 |
Neg logit contributions | |
penn | -0.019 |
pen | -0.018 |
Ow | -0.017 |
imet | -0.017 |
Pe | -0.017 |
Token | He |
Token ID | 679 |
Feature activation | +9.31e-02 |
Pos logit contributions | |
Ow | +0.010 |
Pe | +0.010 |
imet | +0.009 |
pen | +0.009 |
penn | +0.009 |
Neg logit contributions | |
seek | -0.012 |
quiet | -0.012 |
late | -0.012 |
aky | -0.011 |
aking | -0.010 |
Token | scratched |
Token ID | 37468 |
Feature activation | -3.03e-01 |
Pos logit contributions | |
imet | +0.003 |
penn | +0.003 |
pen | +0.003 |
ian | +0.003 |
Ow | +0.003 |
Neg logit contributions | |
late | -0.003 |
quiet | -0.003 |
seek | -0.003 |
too | -0.003 |
Lisa | -0.003 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
late | +0.009 |
seek | +0.008 |
quiet | +0.007 |
impressed | +0.007 |
pleased | +0.007 |
Neg logit contributions | |
penn | -0.010 |
Ow | -0.010 |
pen | -0.010 |
ian | -0.010 |
imet | -0.009 |
Token | 's |
Token ID | 338 |
Feature activation | -6.16e-01 |
Pos logit contributions | |
late | +0.042 |
quiet | +0.038 |
too | +0.036 |
seek | +0.036 |
Lisa | +0.034 |
Neg logit contributions | |
imet | -0.057 |
penn | -0.057 |
ian | -0.055 |
pen | -0.054 |
Ow | -0.054 |
Token | hand |
Token ID | 1021 |
Feature activation | +8.58e-02 |
Pos logit contributions | |
late | +0.022 |
quiet | +0.020 |
seek | +0.020 |
Lisa | +0.019 |
too | +0.018 |
Neg logit contributions | |
penn | -0.016 |
imet | -0.016 |
pen | -0.016 |
ian | -0.015 |
Ow | -0.015 |
Token | with |
Token ID | 351 |
Feature activation | +8.83e-01 |
Pos logit contributions | |
penn | +0.003 |
pen | +0.003 |
imet | +0.003 |
ian | +0.003 |
brow | +0.003 |
Neg logit contributions | |
late | -0.002 |
seek | -0.002 |
quiet | -0.002 |
Lisa | -0.002 |
Lisa | -0.002 |
Token | his |
Token ID | 465 |
Feature activation | -8.25e-01 |
Pos logit contributions | |
penn | +0.029 |
imet | +0.029 |
pen | +0.028 |
ian | +0.028 |
Ow | +0.027 |
Neg logit contributions | |
late | -0.025 |
quiet | -0.022 |
seek | -0.022 |
Lisa | -0.021 |
too | -0.020 |
Token | claws |
Token ID | 28421 |
Feature activation | +9.39e-01 |
Pos logit contributions | |
late | +0.031 |
seek | +0.028 |
quiet | +0.028 |
Lisa | +0.027 |
Lisa | +0.026 |
Neg logit contributions | |
penn | -0.020 |
imet | -0.019 |
pen | -0.019 |
ian | -0.018 |
Ow | -0.018 |