Token | . |
Token ID | 13 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
living | +0.049 |
mun | +0.038 |
have | +0.038 |
w | +0.038 |
d | +0.037 |
Neg logit contributions | |
horizon | -0.050 |
heights | -0.042 |
peak | -0.042 |
edly | -0.042 |
itself | -0.041 |
Token |
|
Token ID | 198 |
Feature activation | -7.20e-03 |
Pos logit contributions | |
living | +0.030 |
mun | +0.024 |
have | +0.024 |
Good | +0.023 |
d | +0.022 |
Neg logit contributions | |
horizon | -0.025 |
heights | -0.021 |
peak | -0.021 |
comb | -0.021 |
edly | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | +8.34e-04 |
Pos logit contributions | |
living | +0.013 |
mun | +0.010 |
have | +0.010 |
w | +0.010 |
d | +0.009 |
Neg logit contributions | |
horizon | -0.015 |
heights | -0.012 |
peak | -0.012 |
edly | -0.012 |
itself | -0.012 |
Token | To |
Token ID | 2514 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
horizon | +0.002 |
itself | +0.001 |
peak | +0.001 |
edly | +0.001 |
heights | +0.001 |
Neg logit contributions | |
living | -0.001 |
mun | -0.001 |
w | -0.001 |
d | -0.001 |
have | -0.001 |
Token | her |
Token ID | 607 |
Feature activation | +8.65e-03 |
Pos logit contributions | |
horizon | +0.030 |
heights | +0.026 |
peak | +0.025 |
edly | +0.025 |
ement | +0.025 |
Neg logit contributions | |
living | -0.024 |
w | -0.018 |
have | -0.018 |
mun | -0.018 |
d | -0.018 |
Token | amaz |
Token ID | 40642 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
horizon | +0.015 |
edly | +0.013 |
itself | +0.013 |
comb | +0.013 |
heights | +0.013 |
Neg logit contributions | |
living | -0.018 |
d | -0.014 |
mun | -0.013 |
have | -0.013 |
loves | -0.013 |
Token | ement |
Token ID | 972 |
Feature activation | +4.95e-02 |
Pos logit contributions | |
horizon | +0.095 |
comb | +0.079 |
alks | +0.066 |
peak | +0.065 |
edly | +0.063 |
Neg logit contributions | |
living | -0.298 |
w | -0.255 |
mun | -0.246 |
loves | -0.245 |
have | -0.244 |
Token | , |
Token ID | 11 |
Feature activation | +1.30e-03 |
Pos logit contributions | |
horizon | +0.093 |
heights | +0.080 |
itself | +0.078 |
peak | +0.078 |
edly | +0.077 |
Neg logit contributions | |
living | -0.093 |
d | -0.072 |
mun | -0.072 |
w | -0.071 |
have | -0.070 |
Token | the |
Token ID | 262 |
Feature activation | +7.84e-03 |
Pos logit contributions | |
horizon | +0.002 |
itself | +0.002 |
peak | +0.002 |
heights | +0.002 |
edly | +0.002 |
Neg logit contributions | |
living | -0.002 |
mun | -0.002 |
w | -0.002 |
have | -0.002 |
d | -0.002 |
Token | clothes |
Token ID | 8242 |
Feature activation | +1.25e-04 |
Pos logit contributions | |
horizon | +0.015 |
peak | +0.013 |
heights | +0.012 |
itself | +0.012 |
edly | +0.012 |
Neg logit contributions | |
living | -0.013 |
w | -0.012 |
mun | -0.012 |
have | -0.012 |
haw | -0.011 |
Token | came |
Token ID | 1625 |
Feature activation | -4.45e-03 |
Pos logit contributions | |
horizon | +0.000 |
heights | +0.000 |
itself | +0.000 |
edly | +0.000 |
peak | +0.000 |
Neg logit contributions | |
living | -0.000 |
mun | -0.000 |
w | -0.000 |
d | -0.000 |
have | -0.000 |
Token | animals |
Token ID | 4695 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
living | +0.045 |
mun | +0.037 |
w | +0.036 |
have | +0.036 |
d | +0.033 |
Neg logit contributions | |
horizon | -0.054 |
peak | -0.046 |
heights | -0.046 |
itself | -0.046 |
edly | -0.044 |
Token | were |
Token ID | 547 |
Feature activation | +8.87e-03 |
Pos logit contributions | |
living | +0.016 |
mun | +0.013 |
w | +0.013 |
have | +0.013 |
d | +0.012 |
Neg logit contributions | |
horizon | -0.024 |
itself | -0.021 |
peak | -0.021 |
heights | -0.021 |
edly | -0.020 |
Token | watching |
Token ID | 4964 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
horizon | +0.020 |
edly | +0.017 |
peak | +0.017 |
itself | +0.017 |
heights | +0.017 |
Neg logit contributions | |
living | -0.014 |
mun | -0.011 |
have | -0.011 |
w | -0.010 |
d | -0.010 |
Token | him |
Token ID | 683 |
Feature activation | +1.43e-02 |
Pos logit contributions | |
living | +0.024 |
mun | +0.019 |
d | +0.019 |
have | +0.018 |
w | +0.018 |
Neg logit contributions | |
horizon | -0.027 |
edly | -0.023 |
peak | -0.023 |
comb | -0.022 |
heights | -0.022 |
Token | with |
Token ID | 351 |
Feature activation | +7.98e-03 |
Pos logit contributions | |
horizon | +0.029 |
edly | +0.025 |
ement | +0.024 |
heights | +0.024 |
peak | +0.024 |
Neg logit contributions | |
living | -0.025 |
have | -0.020 |
mun | -0.020 |
d | -0.019 |
w | -0.019 |
Token | amaz |
Token ID | 40642 |
Feature activation | +9.42e-02 |
Pos logit contributions | |
horizon | +0.018 |
edly | +0.016 |
comb | +0.015 |
peak | +0.015 |
ement | +0.015 |
Neg logit contributions | |
living | -0.013 |
d | -0.010 |
mun | -0.009 |
have | -0.009 |
w | -0.009 |
Token | ement |
Token ID | 972 |
Feature activation | +3.53e-02 |
Pos logit contributions | |
horizon | +0.081 |
comb | +0.060 |
peak | +0.052 |
edly | +0.046 |
heights | +0.046 |
Neg logit contributions | |
living | -0.266 |
w | -0.236 |
mun | -0.231 |
have | -0.230 |
sm | -0.228 |
Token | . |
Token ID | 13 |
Feature activation | -4.37e-03 |
Pos logit contributions | |
horizon | +0.067 |
peak | +0.056 |
heights | +0.056 |
itself | +0.055 |
edly | +0.055 |
Neg logit contributions | |
living | -0.066 |
mun | -0.051 |
d | -0.051 |
w | -0.051 |
have | -0.051 |
Token | |
Token ID | 220 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
living | +0.008 |
mun | +0.006 |
Good | +0.006 |
have | +0.006 |
Did | +0.006 |
Neg logit contributions | |
horizon | -0.009 |
comb | -0.008 |
edly | -0.007 |
heights | -0.007 |
peak | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | -3.66e-03 |
Pos logit contributions | |
living | +0.036 |
mun | +0.028 |
have | +0.027 |
w | +0.027 |
d | +0.026 |
Neg logit contributions | |
horizon | -0.046 |
heights | -0.039 |
edly | -0.039 |
peak | -0.039 |
comb | -0.038 |
Token |
|
Token ID | 198 |
Feature activation | +3.08e-03 |
Pos logit contributions | |
living | +0.006 |
mun | +0.005 |
have | +0.005 |
w | +0.005 |
d | +0.005 |
Neg logit contributions | |
horizon | -0.007 |
heights | -0.006 |
peak | -0.006 |
edly | -0.006 |
itself | -0.006 |
Token | at |
Token ID | 379 |
Feature activation | +9.82e-03 |
Pos logit contributions | |
living | +0.018 |
mun | +0.015 |
w | +0.015 |
have | +0.015 |
d | +0.014 |
Neg logit contributions | |
horizon | -0.017 |
itself | -0.014 |
peak | -0.014 |
heights | -0.014 |
edly | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | +4.03e-03 |
Pos logit contributions | |
horizon | +0.024 |
comb | +0.023 |
edly | +0.022 |
peak | +0.022 |
� | +0.021 |
Neg logit contributions | |
living | -0.013 |
d | -0.008 |
mun | -0.008 |
sm | -0.008 |
w | -0.008 |
Token | bright |
Token ID | 6016 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
horizon | +0.008 |
itself | +0.007 |
peak | +0.007 |
heights | +0.007 |
edly | +0.007 |
Neg logit contributions | |
living | -0.007 |
mun | -0.006 |
w | -0.006 |
have | -0.006 |
haw | -0.005 |
Token | colours |
Token ID | 18915 |
Feature activation | -5.37e-03 |
Pos logit contributions | |
horizon | +0.029 |
edly | +0.027 |
comb | +0.027 |
peak | +0.025 |
heights | +0.025 |
Neg logit contributions | |
living | -0.029 |
d | -0.023 |
mun | -0.022 |
w | -0.021 |
have | -0.020 |
Token | with |
Token ID | 351 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
living | +0.010 |
d | +0.008 |
watching | +0.007 |
mun | +0.007 |
sm | +0.007 |
Neg logit contributions | |
horizon | -0.010 |
heights | -0.009 |
edly | -0.009 |
alks | -0.009 |
peak | -0.009 |
Token | amaz |
Token ID | 40642 |
Feature activation | +9.22e-02 |
Pos logit contributions | |
horizon | +0.025 |
comb | +0.023 |
edly | +0.022 |
peak | +0.021 |
heights | +0.021 |
Neg logit contributions | |
living | -0.024 |
d | -0.019 |
mun | -0.018 |
sm | -0.017 |
have | -0.017 |
Token | ement |
Token ID | 972 |
Feature activation | +4.53e-02 |
Pos logit contributions | |
horizon | +0.095 |
comb | +0.089 |
peak | +0.072 |
edly | +0.069 |
current | +0.068 |
Neg logit contributions | |
living | -0.234 |
sm | -0.202 |
d | -0.197 |
w | -0.196 |
mun | -0.191 |
Token | and |
Token ID | 290 |
Feature activation | +1.72e-02 |
Pos logit contributions | |
horizon | +0.085 |
peak | +0.072 |
heights | +0.072 |
� | +0.069 |
itself | +0.069 |
Neg logit contributions | |
living | -0.085 |
d | -0.068 |
w | -0.062 |
mun | -0.062 |
watching | -0.062 |
Token | awe |
Token ID | 25030 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
horizon | +0.037 |
peak | +0.032 |
itself | +0.032 |
heights | +0.032 |
edly | +0.032 |
Neg logit contributions | |
living | -0.027 |
mun | -0.022 |
have | -0.021 |
w | -0.021 |
d | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | +2.37e-03 |
Pos logit contributions | |
horizon | +0.067 |
comb | +0.058 |
peak | +0.057 |
heights | +0.055 |
� | +0.055 |
Neg logit contributions | |
living | -0.061 |
d | -0.047 |
sm | -0.046 |
mun | -0.044 |
have | -0.044 |
Token | He |
Token ID | 679 |
Feature activation | -8.78e-04 |
Pos logit contributions | |
horizon | +0.004 |
peak | +0.004 |
heights | +0.004 |
comb | +0.004 |
edly | +0.004 |
Neg logit contributions | |
living | -0.005 |
mun | -0.004 |
Good | -0.004 |
have | -0.003 |
d | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | +1.88e-03 |
Pos logit contributions | |
horizon | +0.015 |
heights | +0.013 |
peak | +0.012 |
ement | +0.012 |
� | +0.012 |
Neg logit contributions | |
living | -0.012 |
d | -0.009 |
mun | -0.009 |
have | -0.009 |
w | -0.008 |
Token | her |
Token ID | 607 |
Feature activation | -9.26e-04 |
Pos logit contributions | |
horizon | +0.004 |
peak | +0.003 |
itself | +0.003 |
heights | +0.003 |
edly | +0.003 |
Neg logit contributions | |
living | -0.003 |
mun | -0.003 |
have | -0.003 |
w | -0.003 |
d | -0.002 |
Token | eyes |
Token ID | 2951 |
Feature activation | +3.21e-02 |
Pos logit contributions | |
living | +0.002 |
d | +0.002 |
w | +0.002 |
mun | +0.002 |
loves | +0.001 |
Neg logit contributions | |
horizon | -0.002 |
itself | -0.001 |
heights | -0.001 |
peak | -0.001 |
edly | -0.001 |
Token | wide |
Token ID | 3094 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
horizon | +0.062 |
heights | +0.052 |
itself | +0.052 |
peak | +0.052 |
edly | +0.051 |
Neg logit contributions | |
living | -0.058 |
mun | -0.047 |
w | -0.046 |
have | -0.046 |
d | -0.045 |
Token | with |
Token ID | 351 |
Feature activation | +3.74e-03 |
Pos logit contributions | |
horizon | +0.041 |
heights | +0.035 |
peak | +0.034 |
alks | +0.034 |
� | +0.034 |
Neg logit contributions | |
living | -0.036 |
d | -0.028 |
have | -0.026 |
mun | -0.026 |
sm | -0.025 |
Token | amaz |
Token ID | 40642 |
Feature activation | +8.93e-02 |
Pos logit contributions | |
horizon | +0.007 |
edly | +0.006 |
comb | +0.006 |
peak | +0.006 |
heights | +0.006 |
Neg logit contributions | |
living | -0.007 |
d | -0.006 |
w | -0.005 |
mun | -0.005 |
have | -0.005 |
Token | ement |
Token ID | 972 |
Feature activation | +3.92e-02 |
Pos logit contributions | |
horizon | +0.092 |
comb | +0.071 |
peak | +0.067 |
heights | +0.063 |
alks | +0.061 |
Neg logit contributions | |
living | -0.232 |
w | -0.204 |
sm | -0.202 |
d | -0.200 |
have | -0.196 |
Token | . |
Token ID | 13 |
Feature activation | +2.78e-03 |
Pos logit contributions | |
horizon | +0.078 |
heights | +0.066 |
peak | +0.066 |
itself | +0.064 |
comb | +0.062 |
Neg logit contributions | |
living | -0.070 |
d | -0.055 |
w | -0.054 |
mun | -0.053 |
have | -0.053 |
Token |
|
Token ID | 198 |
Feature activation | +1.96e-03 |
Pos logit contributions | |
horizon | +0.006 |
heights | +0.005 |
peak | +0.005 |
comb | +0.005 |
edly | +0.005 |
Neg logit contributions | |
living | -0.005 |
Good | -0.003 |
mun | -0.003 |
d | -0.003 |
Did | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +6.09e-03 |
Pos logit contributions | |
horizon | +0.004 |
heights | +0.003 |
peak | +0.003 |
edly | +0.003 |
comb | +0.003 |
Neg logit contributions | |
living | -0.003 |
mun | -0.003 |
have | -0.003 |
w | -0.003 |
d | -0.003 |
Token | Her |
Token ID | 9360 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
horizon | +0.012 |
itself | +0.011 |
edly | +0.010 |
peak | +0.010 |
heights | +0.010 |
Neg logit contributions | |
living | -0.011 |
mun | -0.009 |
w | -0.008 |
have | -0.008 |
d | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | +7.82e-03 |
Pos logit contributions | |
horizon | +0.030 |
heights | +0.025 |
peak | +0.025 |
edly | +0.023 |
itself | +0.023 |
Neg logit contributions | |
living | -0.023 |
d | -0.018 |
mun | -0.018 |
have | -0.017 |
sm | -0.017 |
Token | her |
Token ID | 607 |
Feature activation | +1.54e-03 |
Pos logit contributions | |
horizon | +0.017 |
peak | +0.014 |
itself | +0.014 |
heights | +0.014 |
edly | +0.014 |
Neg logit contributions | |
living | -0.013 |
mun | -0.010 |
have | -0.010 |
w | -0.010 |
d | -0.009 |
Token | eyes |
Token ID | 2951 |
Feature activation | +3.45e-02 |
Pos logit contributions | |
horizon | +0.003 |
itself | +0.002 |
edly | +0.002 |
heights | +0.002 |
peak | +0.002 |
Neg logit contributions | |
living | -0.003 |
d | -0.002 |
loves | -0.002 |
mun | -0.002 |
sm | -0.002 |
Token | wide |
Token ID | 3094 |
Feature activation | +2.57e-02 |
Pos logit contributions | |
horizon | +0.064 |
heights | +0.054 |
itself | +0.054 |
peak | +0.053 |
edly | +0.052 |
Neg logit contributions | |
living | -0.065 |
mun | -0.053 |
have | -0.052 |
w | -0.052 |
d | -0.051 |
Token | with |
Token ID | 351 |
Feature activation | +4.19e-03 |
Pos logit contributions | |
horizon | +0.050 |
heights | +0.043 |
peak | +0.042 |
alks | +0.042 |
� | +0.041 |
Neg logit contributions | |
living | -0.045 |
d | -0.036 |
mun | -0.033 |
have | -0.033 |
sm | -0.032 |
Token | amaz |
Token ID | 40642 |
Feature activation | +8.92e-02 |
Pos logit contributions | |
horizon | +0.008 |
edly | +0.007 |
comb | +0.006 |
heights | +0.006 |
peak | +0.006 |
Neg logit contributions | |
living | -0.008 |
d | -0.007 |
sm | -0.006 |
mun | -0.006 |
stri | -0.006 |
Token | ement |
Token ID | 972 |
Feature activation | +3.93e-02 |
Pos logit contributions | |
horizon | +0.096 |
comb | +0.076 |
peak | +0.069 |
heights | +0.065 |
Everywhere | +0.065 |
Neg logit contributions | |
living | -0.225 |
sm | -0.198 |
w | -0.197 |
d | -0.196 |
have | -0.189 |
Token | . |
Token ID | 13 |
Feature activation | +2.97e-03 |
Pos logit contributions | |
horizon | +0.078 |
heights | +0.067 |
peak | +0.066 |
itself | +0.065 |
comb | +0.063 |
Neg logit contributions | |
living | -0.069 |
d | -0.055 |
w | -0.052 |
mun | -0.052 |
have | -0.051 |
Token |
|
Token ID | 198 |
Feature activation | +5.98e-04 |
Pos logit contributions | |
horizon | +0.007 |
peak | +0.006 |
heights | +0.006 |
comb | +0.006 |
edly | +0.006 |
Neg logit contributions | |
living | -0.005 |
mun | -0.003 |
Good | -0.003 |
d | -0.003 |
L | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +5.72e-03 |
Pos logit contributions | |
horizon | +0.001 |
heights | +0.001 |
peak | +0.001 |
comb | +0.001 |
edly | +0.001 |
Neg logit contributions | |
living | -0.001 |
mun | -0.001 |
have | -0.001 |
w | -0.001 |
sm | -0.001 |
Token | " |
Token ID | 1 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
horizon | +0.012 |
itself | +0.011 |
peak | +0.010 |
heights | +0.010 |
edly | +0.010 |
Neg logit contributions | |
living | -0.009 |
mun | -0.007 |
w | -0.007 |
have | -0.007 |
d | -0.007 |
Token | magical |
Token ID | 10883 |
Feature activation | +6.97e-03 |
Pos logit contributions | |
horizon | +0.006 |
itself | +0.006 |
heights | +0.005 |
peak | +0.005 |
edly | +0.005 |
Neg logit contributions | |
living | -0.005 |
mun | -0.004 |
w | -0.004 |
have | -0.004 |
sm | -0.004 |
Token | noise |
Token ID | 7838 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
horizon | +0.012 |
peak | +0.010 |
itself | +0.010 |
heights | +0.010 |
edly | +0.009 |
Neg logit contributions | |
living | -0.014 |
mun | -0.012 |
have | -0.012 |
w | -0.012 |
d | -0.011 |
Token | ! |
Token ID | 0 |
Feature activation | +7.43e-03 |
Pos logit contributions | |
horizon | +0.020 |
itself | +0.017 |
peak | +0.017 |
heights | +0.016 |
edly | +0.016 |
Neg logit contributions | |
living | -0.021 |
mun | -0.018 |
have | -0.018 |
w | -0.018 |
haw | -0.016 |
Token | To |
Token ID | 1675 |
Feature activation | +1.72e-03 |
Pos logit contributions | |
horizon | +0.013 |
heights | +0.011 |
comb | +0.011 |
peak | +0.011 |
edly | +0.011 |
Neg logit contributions | |
living | -0.015 |
mun | -0.012 |
w | -0.012 |
have | -0.012 |
d | -0.012 |
Token | his |
Token ID | 465 |
Feature activation | -9.39e-03 |
Pos logit contributions | |
horizon | +0.004 |
heights | +0.004 |
edly | +0.004 |
comb | +0.004 |
ement | +0.004 |
Neg logit contributions | |
living | -0.002 |
d | -0.001 |
have | -0.001 |
w | -0.001 |
mun | -0.001 |
Token | amaz |
Token ID | 40642 |
Feature activation | +8.65e-02 |
Pos logit contributions | |
living | +0.018 |
d | +0.014 |
loves | +0.013 |
stri | +0.013 |
w | +0.013 |
Neg logit contributions | |
horizon | -0.018 |
edly | -0.016 |
itself | -0.015 |
comb | -0.015 |
rily | -0.015 |
Token | ement |
Token ID | 972 |
Feature activation | +4.60e-02 |
Pos logit contributions | |
horizon | +0.086 |
comb | +0.077 |
umbs | +0.064 |
edly | +0.061 |
Everywhere | +0.060 |
Neg logit contributions | |
living | -0.215 |
w | -0.193 |
d | -0.189 |
happened | -0.182 |
stri | -0.181 |
Token | he |
Token ID | 339 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
horizon | +0.085 |
heights | +0.072 |
peak | +0.070 |
itself | +0.070 |
edly | +0.069 |
Neg logit contributions | |
living | -0.086 |
d | -0.070 |
w | -0.070 |
mun | -0.069 |
have | -0.067 |
Token | had |
Token ID | 550 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
horizon | +0.030 |
peak | +0.027 |
itself | +0.026 |
heights | +0.026 |
ement | +0.026 |
Neg logit contributions | |
living | -0.014 |
mun | -0.011 |
have | -0.010 |
w | -0.010 |
haw | -0.009 |
Token | revealed |
Token ID | 4602 |
Feature activation | +5.56e-03 |
Pos logit contributions | |
living | +0.019 |
mun | +0.017 |
w | +0.015 |
have | +0.015 |
haw | +0.014 |
Neg logit contributions | |
horizon | -0.025 |
itself | -0.022 |
peak | -0.022 |
heights | -0.022 |
edly | -0.021 |
Token | a |
Token ID | 257 |
Feature activation | +3.45e-03 |
Pos logit contributions | |
horizon | +0.012 |
edly | +0.010 |
comb | +0.010 |
� | +0.010 |
peak | +0.010 |
Neg logit contributions | |
living | -0.009 |
d | -0.007 |
w | -0.007 |
have | -0.006 |
mun | -0.006 |
Token | He |
Token ID | 679 |
Feature activation | +5.83e-03 |
Pos logit contributions | |
horizon | +0.008 |
comb | +0.008 |
edly | +0.007 |
heights | +0.007 |
iously | +0.007 |
Neg logit contributions | |
living | -0.007 |
Good | -0.005 |
mun | -0.005 |
Enjoy | -0.005 |
Are | -0.004 |
Token | tried |
Token ID | 3088 |
Feature activation | -9.30e-03 |
Pos logit contributions | |
horizon | +0.013 |
peak | +0.012 |
itself | +0.012 |
edly | +0.012 |
heights | +0.011 |
Neg logit contributions | |
living | -0.008 |
w | -0.006 |
mun | -0.006 |
have | -0.006 |
sm | -0.006 |
Token | to |
Token ID | 284 |
Feature activation | +9.03e-03 |
Pos logit contributions | |
living | +0.016 |
mun | +0.013 |
w | +0.013 |
have | +0.013 |
sm | +0.012 |
Neg logit contributions | |
horizon | -0.018 |
itself | -0.016 |
heights | -0.016 |
peak | -0.016 |
edly | -0.015 |
Token | find |
Token ID | 1064 |
Feature activation | -4.92e-03 |
Pos logit contributions | |
horizon | +0.021 |
ement | +0.019 |
edly | +0.018 |
heights | +0.018 |
peak | +0.018 |
Neg logit contributions | |
living | -0.013 |
have | -0.010 |
d | -0.010 |
mun | -0.009 |
w | -0.009 |
Token | a |
Token ID | 257 |
Feature activation | +3.52e-03 |
Pos logit contributions | |
living | +0.008 |
mun | +0.007 |
w | +0.007 |
have | +0.007 |
d | +0.006 |
Neg logit contributions | |
horizon | -0.010 |
itself | -0.009 |
heights | -0.008 |
peak | -0.008 |
edly | -0.008 |
Token | low |
Token ID | 1877 |
Feature activation | +8.62e-02 |
Pos logit contributions | |
horizon | +0.007 |
peak | +0.006 |
heights | +0.006 |
itself | +0.006 |
comb | +0.005 |
Neg logit contributions | |
living | -0.006 |
w | -0.005 |
mun | -0.005 |
have | -0.005 |
sm | -0.005 |
Token | stone |
Token ID | 7815 |
Feature activation | -6.46e-03 |
Pos logit contributions | |
horizon | +0.168 |
edly | +0.149 |
ement | +0.146 |
comb | +0.142 |
heights | +0.140 |
Neg logit contributions | |
living | -0.147 |
d | -0.124 |
w | -0.117 |
have | -0.116 |
mun | -0.116 |
Token | to |
Token ID | 284 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
living | +0.011 |
d | +0.008 |
stole | +0.007 |
w | +0.007 |
watching | +0.007 |
Neg logit contributions | |
horizon | -0.015 |
ement | -0.013 |
edly | -0.013 |
heights | -0.012 |
� | -0.012 |
Token | sit |
Token ID | 1650 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
horizon | +0.026 |
edly | +0.023 |
ement | +0.022 |
heights | +0.022 |
comb | +0.022 |
Neg logit contributions | |
living | -0.018 |
have | -0.014 |
d | -0.014 |
mun | -0.014 |
w | -0.014 |
Token | on |
Token ID | 319 |
Feature activation | +8.46e-03 |
Pos logit contributions | |
horizon | +0.028 |
heights | +0.025 |
fingertips | +0.025 |
peak | +0.025 |
delight | +0.024 |
Neg logit contributions | |
living | -0.012 |
w | -0.009 |
d | -0.008 |
P | -0.008 |
have | -0.008 |
Token | but |
Token ID | 475 |
Feature activation | +2.47e-02 |
Pos logit contributions | |
horizon | +0.021 |
comb | +0.020 |
edly | +0.019 |
fingertips | +0.018 |
umbs | +0.018 |
Neg logit contributions | |
living | -0.011 |
d | -0.007 |
watching | -0.006 |
your | -0.006 |
L | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +3.18e-03 |
Pos logit contributions | |
living | +0.016 |
have | +0.011 |
d | +0.011 |
loves | +0.011 |
loved | +0.011 |
Neg logit contributions | |
horizon | -0.022 |
peak | -0.018 |
heights | -0.018 |
ement | -0.018 |
edly | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | -7.64e-03 |
Pos logit contributions | |
horizon | +0.007 |
comb | +0.007 |
edly | +0.006 |
rily | +0.006 |
heights | +0.006 |
Neg logit contributions | |
living | -0.005 |
mun | -0.004 |
loves | -0.004 |
have | -0.003 |
Good | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +1.46e-03 |
Pos logit contributions | |
living | +0.013 |
mun | +0.010 |
w | +0.010 |
have | +0.010 |
d | +0.010 |
Neg logit contributions | |
horizon | -0.016 |
edly | -0.014 |
heights | -0.014 |
peak | -0.013 |
comb | -0.013 |
Token | On |
Token ID | 2202 |
Feature activation | +4.04e-02 |
Pos logit contributions | |
horizon | +0.003 |
peak | +0.003 |
itself | +0.003 |
heights | +0.003 |
edly | +0.003 |
Neg logit contributions | |
living | -0.002 |
mun | -0.002 |
w | -0.002 |
have | -0.002 |
d | -0.002 |
Token | one |
Token ID | 530 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
horizon | +0.089 |
edly | +0.077 |
heights | +0.077 |
itself | +0.076 |
peak | +0.076 |
Neg logit contributions | |
living | -0.064 |
mun | -0.048 |
w | -0.048 |
have | -0.047 |
d | -0.045 |
Token | sunny |
Token ID | 27737 |
Feature activation | +8.57e-02 |
Pos logit contributions | |
horizon | +0.034 |
itself | +0.030 |
edly | +0.030 |
heights | +0.030 |
ement | +0.029 |
Neg logit contributions | |
living | -0.024 |
w | -0.017 |
d | -0.016 |
mun | -0.016 |
have | -0.016 |
Token | day |
Token ID | 1110 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
horizon | +0.201 |
peak | +0.174 |
itself | +0.173 |
edly | +0.172 |
ement | +0.172 |
Neg logit contributions | |
living | -0.129 |
w | -0.094 |
mun | -0.093 |
have | -0.093 |
d | -0.086 |
Token | , |
Token ID | 11 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
horizon | +0.025 |
peak | +0.021 |
itself | +0.020 |
heights | +0.020 |
edly | +0.020 |
Neg logit contributions | |
living | -0.034 |
mun | -0.030 |
have | -0.030 |
w | -0.029 |
d | -0.027 |
Token | Mark |
Token ID | 2940 |
Feature activation | +2.76e-02 |
Pos logit contributions | |
horizon | +0.019 |
itself | +0.015 |
peak | +0.015 |
heights | +0.015 |
edly | +0.015 |
Neg logit contributions | |
living | -0.027 |
mun | -0.023 |
w | -0.023 |
have | -0.023 |
d | -0.022 |
Token | was |
Token ID | 373 |
Feature activation | +8.91e-03 |
Pos logit contributions | |
horizon | +0.066 |
peak | +0.058 |
itself | +0.057 |
edly | +0.057 |
heights | +0.056 |
Neg logit contributions | |
living | -0.036 |
mun | -0.027 |
w | -0.026 |
have | -0.026 |
haw | -0.024 |
Token | walking |
Token ID | 6155 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
horizon | +0.018 |
peak | +0.015 |
itself | +0.015 |
heights | +0.015 |
edly | +0.015 |
Neg logit contributions | |
living | -0.014 |
w | -0.013 |
mun | -0.012 |
haw | -0.012 |
have | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | +7.32e-04 |
Pos logit contributions | |
horizon | +0.047 |
peak | +0.040 |
heights | +0.040 |
itself | +0.040 |
comb | +0.039 |
Neg logit contributions | |
living | -0.034 |
mun | -0.027 |
have | -0.026 |
w | -0.026 |
d | -0.025 |
Token | thought |
Token ID | 1807 |
Feature activation | -7.56e-03 |
Pos logit contributions | |
horizon | +0.002 |
edly | +0.001 |
peak | +0.001 |
itself | +0.001 |
comb | +0.001 |
Neg logit contributions | |
living | -0.001 |
have | -0.001 |
w | -0.001 |
mun | -0.001 |
sm | -0.001 |
Token | for |
Token ID | 329 |
Feature activation | +2.96e-02 |
Pos logit contributions | |
living | +0.013 |
d | +0.010 |
mun | +0.010 |
have | +0.010 |
w | +0.009 |
Neg logit contributions | |
horizon | -0.016 |
heights | -0.013 |
peak | -0.013 |
edly | -0.013 |
comb | -0.013 |
Token | a |
Token ID | 257 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
horizon | +0.066 |
itself | +0.057 |
peak | +0.057 |
heights | +0.057 |
edly | +0.057 |
Neg logit contributions | |
living | -0.045 |
mun | -0.035 |
w | -0.034 |
have | -0.034 |
d | -0.032 |
Token | few |
Token ID | 1178 |
Feature activation | +5.73e-02 |
Pos logit contributions | |
horizon | +0.019 |
peak | +0.017 |
heights | +0.016 |
itself | +0.016 |
comb | +0.016 |
Neg logit contributions | |
living | -0.019 |
mun | -0.017 |
w | -0.017 |
have | -0.017 |
asc | -0.015 |
Token | minutes |
Token ID | 2431 |
Feature activation | +8.52e-02 |
Pos logit contributions | |
horizon | +0.106 |
itself | +0.087 |
peak | +0.086 |
comb | +0.085 |
edly | +0.084 |
Neg logit contributions | |
living | -0.112 |
mun | -0.091 |
have | -0.090 |
d | -0.088 |
w | -0.088 |
Token | and |
Token ID | 290 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
horizon | +0.164 |
peak | +0.139 |
heights | +0.138 |
itself | +0.134 |
edly | +0.133 |
Neg logit contributions | |
living | -0.161 |
mun | -0.127 |
d | -0.126 |
have | -0.123 |
w | -0.120 |
Token | then |
Token ID | 788 |
Feature activation | +3.15e-02 |
Pos logit contributions | |
horizon | +0.032 |
itself | +0.025 |
peak | +0.025 |
heights | +0.025 |
edly | +0.024 |
Neg logit contributions | |
living | -0.053 |
mun | -0.046 |
w | -0.046 |
have | -0.045 |
d | -0.043 |
Token | he |
Token ID | 339 |
Feature activation | +1.02e-03 |
Pos logit contributions | |
horizon | +0.064 |
itself | +0.056 |
peak | +0.056 |
edly | +0.056 |
comb | +0.055 |
Neg logit contributions | |
living | -0.051 |
mun | -0.042 |
w | -0.041 |
have | -0.040 |
sm | -0.038 |
Token | had |
Token ID | 550 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
horizon | +0.002 |
peak | +0.002 |
itself | +0.002 |
edly | +0.002 |
comb | +0.002 |
Neg logit contributions | |
living | -0.001 |
mun | -0.001 |
w | -0.001 |
have | -0.001 |
sm | -0.001 |
Token | an |
Token ID | 281 |
Feature activation | +5.45e-03 |
Pos logit contributions | |
living | +0.021 |
mun | +0.018 |
w | +0.017 |
have | +0.017 |
haw | +0.015 |
Neg logit contributions | |
horizon | -0.026 |
itself | -0.023 |
heights | -0.022 |
peak | -0.022 |
edly | -0.021 |
Token | day |
Token ID | 1110 |
Feature activation | -1.25e-02 |
Pos logit contributions | |
horizon | +0.014 |
peak | +0.012 |
itself | +0.011 |
edly | +0.011 |
comb | +0.011 |
Neg logit contributions | |
living | -0.012 |
mun | -0.011 |
have | -0.010 |
w | -0.010 |
Did | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | +8.78e-03 |
Pos logit contributions | |
living | +0.028 |
have | +0.024 |
mun | +0.024 |
w | +0.023 |
haw | +0.022 |
Neg logit contributions | |
horizon | -0.017 |
comb | -0.015 |
itself | -0.014 |
peak | -0.014 |
edly | -0.014 |
Token | her |
Token ID | 607 |
Feature activation | -1.24e-03 |
Pos logit contributions | |
horizon | +0.018 |
peak | +0.015 |
ement | +0.014 |
itself | +0.014 |
heights | +0.014 |
Neg logit contributions | |
living | -0.016 |
mun | -0.013 |
w | -0.013 |
d | -0.013 |
have | -0.012 |
Token | mom |
Token ID | 1995 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
living | +0.002 |
d | +0.002 |
mun | +0.002 |
w | +0.002 |
have | +0.002 |
Neg logit contributions | |
horizon | -0.002 |
itself | -0.002 |
heights | -0.002 |
edly | -0.002 |
peak | -0.002 |
Token | gave |
Token ID | 2921 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
horizon | +0.023 |
comb | +0.020 |
itself | +0.020 |
peak | +0.020 |
ement | +0.019 |
Neg logit contributions | |
living | -0.013 |
w | -0.011 |
mun | -0.011 |
haw | -0.010 |
have | -0.010 |
Token | Luna |
Token ID | 23694 |
Feature activation | +8.52e-02 |
Pos logit contributions | |
horizon | +0.023 |
peak | +0.020 |
itself | +0.020 |
heights | +0.020 |
edly | +0.019 |
Neg logit contributions | |
living | -0.018 |
mun | -0.015 |
w | -0.014 |
have | -0.014 |
d | -0.013 |
Token | a |
Token ID | 257 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
horizon | +0.183 |
itself | +0.159 |
peak | +0.158 |
heights | +0.157 |
edly | +0.156 |
Neg logit contributions | |
living | -0.133 |
mun | -0.107 |
w | -0.104 |
have | -0.103 |
d | -0.097 |
Token | mask |
Token ID | 9335 |
Feature activation | +2.87e-02 |
Pos logit contributions | |
horizon | +0.027 |
peak | +0.024 |
comb | +0.022 |
� | +0.022 |
fingertips | +0.021 |
Neg logit contributions | |
living | -0.028 |
mun | -0.026 |
have | -0.026 |
w | -0.026 |
Did | -0.024 |
Token | to |
Token ID | 284 |
Feature activation | +2.21e-02 |
Pos logit contributions | |
horizon | +0.057 |
heights | +0.050 |
peak | +0.048 |
itself | +0.047 |
fingertips | +0.046 |
Neg logit contributions | |
living | -0.054 |
d | -0.040 |
w | -0.040 |
sm | -0.039 |
mun | -0.039 |
Token | wear |
Token ID | 5806 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
horizon | +0.050 |
peak | +0.044 |
itself | +0.043 |
heights | +0.043 |
edly | +0.043 |
Neg logit contributions | |
living | -0.032 |
mun | -0.025 |
have | -0.024 |
w | -0.024 |
d | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | +3.13e-03 |
Pos logit contributions | |
horizon | +0.032 |
itself | +0.027 |
peak | +0.026 |
heights | +0.026 |
edly | +0.026 |
Neg logit contributions | |
living | -0.033 |
mun | -0.028 |
w | -0.027 |
have | -0.027 |
d | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | +7.07e-03 |
Pos logit contributions | |
horizon | +0.067 |
peak | +0.056 |
comb | +0.055 |
heights | +0.055 |
itself | +0.055 |
Neg logit contributions | |
living | -0.062 |
mun | -0.051 |
d | -0.050 |
w | -0.050 |
have | -0.048 |
Token | Amy |
Token ID | 14235 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
horizon | +0.014 |
heights | +0.013 |
peak | +0.012 |
comb | +0.012 |
iously | +0.011 |
Neg logit contributions | |
living | -0.013 |
Good | -0.010 |
mun | -0.010 |
L | -0.009 |
Enjoy | -0.009 |
Token | watched |
Token ID | 7342 |
Feature activation | -5.12e-04 |
Pos logit contributions | |
horizon | +0.036 |
heights | +0.031 |
itself | +0.031 |
peak | +0.031 |
edly | +0.031 |
Neg logit contributions | |
living | -0.023 |
mun | -0.017 |
w | -0.017 |
have | -0.017 |
d | -0.016 |
Token | them |
Token ID | 606 |
Feature activation | +9.52e-03 |
Pos logit contributions | |
living | +0.001 |
mun | +0.001 |
d | +0.001 |
have | +0.001 |
w | +0.001 |
Neg logit contributions | |
horizon | -0.001 |
peak | -0.001 |
heights | -0.001 |
edly | -0.001 |
comb | -0.001 |
Token | with |
Token ID | 351 |
Feature activation | +6.78e-03 |
Pos logit contributions | |
horizon | +0.019 |
edly | +0.016 |
ement | +0.016 |
heights | +0.016 |
itself | +0.016 |
Neg logit contributions | |
living | -0.017 |
mun | -0.013 |
d | -0.013 |
have | -0.013 |
w | -0.012 |
Token | amaz |
Token ID | 40642 |
Feature activation | +8.52e-02 |
Pos logit contributions | |
horizon | +0.015 |
edly | +0.014 |
comb | +0.013 |
iously | +0.013 |
ement | +0.013 |
Neg logit contributions | |
living | -0.011 |
d | -0.008 |
mun | -0.007 |
sm | -0.007 |
watching | -0.007 |
Token | ement |
Token ID | 972 |
Feature activation | +3.53e-02 |
Pos logit contributions | |
horizon | +0.079 |
comb | +0.059 |
peak | +0.054 |
alks | +0.046 |
heights | +0.046 |
Neg logit contributions | |
living | -0.234 |
d | -0.206 |
mun | -0.205 |
w | -0.205 |
have | -0.203 |
Token | . |
Token ID | 13 |
Feature activation | +1.56e-03 |
Pos logit contributions | |
horizon | +0.061 |
peak | +0.051 |
heights | +0.051 |
itself | +0.049 |
comb | +0.048 |
Neg logit contributions | |
living | -0.072 |
d | -0.058 |
mun | -0.057 |
w | -0.057 |
have | -0.056 |
Token | She |
Token ID | 1375 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
horizon | +0.003 |
heights | +0.003 |
comb | +0.003 |
peak | +0.003 |
ement | +0.002 |
Neg logit contributions | |
living | -0.003 |
Good | -0.002 |
mun | -0.002 |
Enjoy | -0.002 |
loves | -0.002 |
Token | smiled |
Token ID | 13541 |
Feature activation | -6.60e-03 |
Pos logit contributions | |
horizon | +0.026 |
itself | +0.023 |
heights | +0.023 |
peak | +0.023 |
edly | +0.023 |
Neg logit contributions | |
living | -0.013 |
mun | -0.010 |
have | -0.010 |
w | -0.009 |
d | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
living | +0.013 |
d | +0.010 |
mer | +0.009 |
loves | +0.009 |
loved | +0.009 |
Neg logit contributions | |
horizon | -0.012 |
peak | -0.010 |
heights | -0.010 |
� | -0.010 |
comb | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
living | +0.077 |
mun | +0.059 |
d | +0.059 |
w | +0.059 |
have | +0.059 |
Neg logit contributions | |
horizon | -0.078 |
heights | -0.068 |
peak | -0.067 |
edly | -0.067 |
itself | -0.065 |
Token | He |
Token ID | 679 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
living | +0.029 |
mun | +0.024 |
have | +0.024 |
Good | +0.023 |
d | +0.023 |
Neg logit contributions | |
horizon | -0.023 |
heights | -0.019 |
peak | -0.019 |
comb | -0.019 |
edly | -0.019 |
Token | pulled |
Token ID | 5954 |
Feature activation | -9.04e-04 |
Pos logit contributions | |
living | +0.015 |
mun | +0.012 |
w | +0.011 |
have | +0.011 |
sm | +0.010 |
Neg logit contributions | |
horizon | -0.028 |
peak | -0.025 |
itself | -0.024 |
heights | -0.024 |
edly | -0.024 |
Token | his |
Token ID | 465 |
Feature activation | +6.08e-03 |
Pos logit contributions | |
living | +0.001 |
d | +0.001 |
w | +0.001 |
have | +0.001 |
sm | +0.001 |
Neg logit contributions | |
horizon | -0.002 |
edly | -0.002 |
comb | -0.002 |
� | -0.002 |
ement | -0.002 |
Token | hat |
Token ID | 6877 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
horizon | +0.011 |
edly | +0.010 |
� | +0.010 |
ement | +0.010 |
heights | +0.010 |
Neg logit contributions | |
living | -0.012 |
d | -0.009 |
w | -0.009 |
have | -0.008 |
loves | -0.008 |
Token | low |
Token ID | 1877 |
Feature activation | +8.47e-02 |
Pos logit contributions | |
horizon | +0.037 |
heights | +0.034 |
edly | +0.033 |
ement | +0.032 |
� | +0.032 |
Neg logit contributions | |
living | -0.021 |
d | -0.018 |
w | -0.016 |
took | -0.015 |
would | -0.014 |
Token | and |
Token ID | 290 |
Feature activation | +5.46e-03 |
Pos logit contributions | |
horizon | +0.168 |
heights | +0.143 |
� | +0.142 |
fingertips | +0.142 |
ement | +0.141 |
Neg logit contributions | |
living | -0.146 |
d | -0.115 |
w | -0.104 |
mun | -0.099 |
have | -0.098 |
Token | wished |
Token ID | 16555 |
Feature activation | +2.02e-03 |
Pos logit contributions | |
horizon | +0.011 |
itself | +0.010 |
peak | +0.010 |
edly | +0.010 |
heights | +0.010 |
Neg logit contributions | |
living | -0.009 |
mun | -0.007 |
have | -0.007 |
w | -0.007 |
sm | -0.007 |
Token | it |
Token ID | 340 |
Feature activation | +2.33e-02 |
Pos logit contributions | |
horizon | +0.004 |
peak | +0.004 |
heights | +0.004 |
edly | +0.004 |
itself | +0.004 |
Neg logit contributions | |
living | -0.003 |
mun | -0.002 |
w | -0.002 |
have | -0.002 |
d | -0.002 |
Token | was |
Token ID | 373 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
horizon | +0.058 |
itself | +0.053 |
peak | +0.052 |
edly | +0.051 |
heights | +0.051 |
Neg logit contributions | |
living | -0.028 |
mun | -0.022 |
have | -0.020 |
w | -0.019 |
P | -0.017 |
Token | his |
Token ID | 465 |
Feature activation | -3.36e-03 |
Pos logit contributions | |
horizon | +0.045 |
itself | +0.039 |
peak | +0.039 |
heights | +0.039 |
edly | +0.038 |
Neg logit contributions | |
living | -0.034 |
mun | -0.028 |
w | -0.027 |
have | -0.027 |
d | -0.025 |
Token | She |
Token ID | 1375 |
Feature activation | +1.52e-02 |
Pos logit contributions | |
horizon | +0.008 |
comb | +0.007 |
peak | +0.007 |
rily | +0.007 |
edly | +0.007 |
Neg logit contributions | |
living | -0.008 |
Good | -0.006 |
mun | -0.006 |
L | -0.006 |
d | -0.006 |
Token | said |
Token ID | 531 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
horizon | +0.034 |
itself | +0.030 |
peak | +0.030 |
edly | +0.030 |
comb | +0.029 |
Neg logit contributions | |
living | -0.021 |
w | -0.016 |
mun | -0.016 |
have | -0.016 |
haw | -0.014 |
Token | she |
Token ID | 673 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
horizon | +0.026 |
itself | +0.023 |
edly | +0.021 |
heights | +0.021 |
peak | +0.021 |
Neg logit contributions | |
living | -0.032 |
mun | -0.028 |
w | -0.028 |
have | -0.027 |
haw | -0.025 |
Token | would |
Token ID | 561 |
Feature activation | +8.50e-03 |
Pos logit contributions | |
horizon | +0.047 |
itself | +0.042 |
edly | +0.042 |
peak | +0.042 |
ement | +0.041 |
Neg logit contributions | |
living | -0.027 |
mun | -0.021 |
w | -0.020 |
have | -0.019 |
P | -0.019 |
Token | lend |
Token ID | 22096 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
horizon | +0.020 |
edly | +0.017 |
heights | +0.017 |
ement | +0.017 |
peak | +0.017 |
Neg logit contributions | |
living | -0.013 |
have | -0.009 |
d | -0.009 |
mun | -0.009 |
w | -0.009 |
Token | Ann |
Token ID | 5506 |
Feature activation | +8.44e-02 |
Pos logit contributions | |
living | +0.023 |
mun | +0.018 |
w | +0.018 |
have | +0.018 |
d | +0.017 |
Neg logit contributions | |
horizon | -0.028 |
peak | -0.023 |
heights | -0.023 |
itself | -0.023 |
edly | -0.023 |
Token | a |
Token ID | 257 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
horizon | +0.167 |
itself | +0.144 |
heights | +0.142 |
peak | +0.142 |
edly | +0.140 |
Neg logit contributions | |
living | -0.145 |
mun | -0.119 |
w | -0.116 |
have | -0.115 |
d | -0.110 |
Token | pair |
Token ID | 5166 |
Feature activation | +7.09e-02 |
Pos logit contributions | |
horizon | +0.040 |
peak | +0.035 |
heights | +0.033 |
itself | +0.032 |
fingertips | +0.032 |
Neg logit contributions | |
living | -0.037 |
mun | -0.033 |
w | -0.032 |
have | -0.031 |
Did | -0.030 |
Token | of |
Token ID | 286 |
Feature activation | +4.56e-02 |
Pos logit contributions | |
horizon | +0.172 |
heights | +0.155 |
peak | +0.146 |
edly | +0.141 |
fingertips | +0.141 |
Neg logit contributions | |
living | -0.104 |
w | -0.080 |
sm | -0.078 |
d | -0.074 |
have | -0.071 |
Token | shoes |
Token ID | 10012 |
Feature activation | +3.24e-02 |
Pos logit contributions | |
horizon | +0.078 |
heights | +0.068 |
itself | +0.067 |
peak | +0.065 |
ement | +0.064 |
Neg logit contributions | |
living | -0.086 |
w | -0.072 |
mun | -0.072 |
have | -0.068 |
Good | -0.067 |
Token | so |
Token ID | 523 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
horizon | +0.073 |
heights | +0.065 |
peak | +0.063 |
fingertips | +0.062 |
edly | +0.061 |
Neg logit contributions | |
living | -0.051 |
w | -0.038 |
d | -0.038 |
have | -0.036 |
mun | -0.034 |
Token | , |
Token ID | 11 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
horizon | +0.022 |
itself | +0.022 |
heights | +0.020 |
ement | +0.019 |
peak | +0.019 |
Neg logit contributions | |
living | -0.043 |
have | -0.039 |
mun | -0.039 |
w | -0.039 |
P | -0.036 |
Token | " |
Token ID | 366 |
Feature activation | +1.52e-02 |
Pos logit contributions | |
horizon | +0.028 |
itself | +0.024 |
edly | +0.023 |
peak | +0.023 |
heights | +0.023 |
Neg logit contributions | |
living | -0.027 |
w | -0.023 |
mun | -0.023 |
have | -0.023 |
sm | -0.021 |
Token | Is |
Token ID | 3792 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
horizon | +0.030 |
itself | +0.025 |
edly | +0.025 |
peak | +0.025 |
heights | +0.025 |
Neg logit contributions | |
living | -0.028 |
mun | -0.022 |
have | -0.022 |
w | -0.022 |
d | -0.021 |
Token | your |
Token ID | 534 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
horizon | +0.045 |
heights | +0.040 |
itself | +0.039 |
peak | +0.038 |
ement | +0.036 |
Neg logit contributions | |
living | -0.054 |
mun | -0.046 |
have | -0.044 |
w | -0.043 |
sm | -0.042 |
Token | name |
Token ID | 1438 |
Feature activation | -1.60e-04 |
Pos logit contributions | |
living | +0.017 |
d | +0.013 |
w | +0.013 |
have | +0.013 |
loves | +0.012 |
Neg logit contributions | |
horizon | -0.021 |
edly | -0.019 |
itself | -0.019 |
peak | -0.018 |
rily | -0.018 |
Token | Kitty |
Token ID | 21168 |
Feature activation | +8.42e-02 |
Pos logit contributions | |
living | +0.000 |
mun | +0.000 |
have | +0.000 |
w | +0.000 |
d | +0.000 |
Neg logit contributions | |
horizon | -0.000 |
itself | -0.000 |
peak | -0.000 |
heights | -0.000 |
edly | -0.000 |
Token | ?" |
Token ID | 1701 |
Feature activation | +3.41e-02 |
Pos logit contributions | |
horizon | +0.170 |
itself | +0.146 |
peak | +0.144 |
heights | +0.144 |
edly | +0.142 |
Neg logit contributions | |
living | -0.142 |
mun | -0.116 |
w | -0.114 |
have | -0.113 |
d | -0.107 |
Token | The |
Token ID | 383 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
horizon | +0.067 |
rily | +0.057 |
peak | +0.057 |
ement | +0.055 |
comb | +0.055 |
Neg logit contributions | |
living | -0.066 |
mun | -0.052 |
d | -0.048 |
Good | -0.048 |
loves | -0.047 |
Token | black |
Token ID | 2042 |
Feature activation | +2.39e-02 |
Pos logit contributions | |
horizon | +0.029 |
itself | +0.025 |
heights | +0.025 |
peak | +0.024 |
edly | +0.024 |
Neg logit contributions | |
living | -0.026 |
mun | -0.022 |
w | -0.021 |
have | -0.021 |
d | -0.020 |
Token | cat |
Token ID | 3797 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
horizon | +0.050 |
peak | +0.043 |
heights | +0.042 |
itself | +0.042 |
edly | +0.042 |
Neg logit contributions | |
living | -0.041 |
w | -0.033 |
mun | -0.032 |
d | -0.031 |
have | -0.031 |
Token | laughed |
Token ID | 13818 |
Feature activation | +5.55e-03 |
Pos logit contributions | |
horizon | +0.040 |
itself | +0.035 |
peak | +0.035 |
heights | +0.035 |
edly | +0.034 |
Neg logit contributions | |
living | -0.026 |
mun | -0.020 |
w | -0.020 |
have | -0.020 |
d | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | +2.90e-03 |
Pos logit contributions | |
horizon | +0.010 |
itself | +0.009 |
heights | +0.009 |
peak | +0.009 |
ement | +0.009 |
Neg logit contributions | |
living | -0.010 |
w | -0.009 |
have | -0.008 |
mun | -0.008 |
haw | -0.008 |
Token | world |
Token ID | 995 |
Feature activation | +8.98e-03 |
Pos logit contributions | |
horizon | +0.004 |
heights | +0.003 |
peak | +0.003 |
itself | +0.003 |
destination | +0.003 |
Neg logit contributions | |
living | -0.006 |
w | -0.006 |
mun | -0.006 |
have | -0.006 |
haw | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +7.85e-04 |
Pos logit contributions | |
horizon | +0.017 |
edly | +0.014 |
comb | +0.014 |
heights | +0.014 |
fingertips | +0.014 |
Neg logit contributions | |
living | -0.018 |
mun | -0.014 |
sm | -0.013 |
d | -0.013 |
have | -0.013 |
Token | On |
Token ID | 1550 |
Feature activation | +2.54e-02 |
Pos logit contributions | |
horizon | +0.001 |
edly | +0.001 |
comb | +0.001 |
peak | +0.001 |
itself | +0.001 |
Neg logit contributions | |
living | -0.002 |
mun | -0.001 |
have | -0.001 |
d | -0.001 |
Good | -0.001 |
Token | one |
Token ID | 530 |
Feature activation | +9.99e-03 |
Pos logit contributions | |
horizon | +0.054 |
edly | +0.048 |
comb | +0.047 |
ement | +0.046 |
peak | +0.046 |
Neg logit contributions | |
living | -0.042 |
mun | -0.031 |
w | -0.030 |
have | -0.030 |
d | -0.030 |
Token | sunny |
Token ID | 27737 |
Feature activation | +8.35e-02 |
Pos logit contributions | |
horizon | +0.022 |
edly | +0.020 |
itself | +0.019 |
rily | +0.019 |
ement | +0.019 |
Neg logit contributions | |
living | -0.017 |
d | -0.012 |
mun | -0.011 |
w | -0.011 |
L | -0.011 |
Token | day |
Token ID | 1110 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
horizon | +0.185 |
edly | +0.160 |
peak | +0.159 |
itself | +0.159 |
comb | +0.158 |
Neg logit contributions | |
living | -0.133 |
mun | -0.100 |
w | -0.099 |
have | -0.098 |
d | -0.094 |
Token | , |
Token ID | 11 |
Feature activation | +9.85e-03 |
Pos logit contributions | |
horizon | +0.022 |
itself | +0.019 |
peak | +0.018 |
edly | +0.018 |
heights | +0.018 |
Neg logit contributions | |
living | -0.038 |
have | -0.033 |
mun | -0.032 |
w | -0.032 |
haw | -0.031 |
Token | they |
Token ID | 484 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
horizon | +0.015 |
itself | +0.013 |
heights | +0.013 |
� | +0.013 |
peak | +0.013 |
Neg logit contributions | |
living | -0.020 |
mun | -0.017 |
w | -0.017 |
have | -0.017 |
haw | -0.016 |
Token | decided |
Token ID | 3066 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
horizon | +0.040 |
peak | +0.035 |
itself | +0.034 |
comb | +0.034 |
edly | +0.033 |
Neg logit contributions | |
living | -0.027 |
mun | -0.022 |
w | -0.022 |
have | -0.021 |
sm | -0.020 |
Token | to |
Token ID | 284 |
Feature activation | +7.04e-03 |
Pos logit contributions | |
horizon | +0.022 |
itself | +0.019 |
heights | +0.018 |
peak | +0.018 |
edly | +0.018 |
Neg logit contributions | |
living | -0.021 |
mun | -0.018 |
w | -0.017 |
have | -0.017 |
sm | -0.016 |
Token | saw |
Token ID | 2497 |
Feature activation | +3.77e-03 |
Pos logit contributions | |
living | +0.025 |
mun | +0.019 |
have | +0.018 |
w | +0.017 |
sm | +0.016 |
Neg logit contributions | |
horizon | -0.053 |
peak | -0.047 |
itself | -0.046 |
heights | -0.046 |
comb | -0.045 |
Token | a |
Token ID | 257 |
Feature activation | -1.59e-02 |
Pos logit contributions | |
horizon | +0.008 |
itself | +0.007 |
peak | +0.007 |
heights | +0.007 |
edly | +0.007 |
Neg logit contributions | |
living | -0.006 |
mun | -0.005 |
have | -0.005 |
w | -0.005 |
d | -0.004 |
Token | tree |
Token ID | 5509 |
Feature activation | +5.21e-03 |
Pos logit contributions | |
living | +0.028 |
have | +0.026 |
w | +0.025 |
mun | +0.025 |
haw | +0.023 |
Neg logit contributions | |
horizon | -0.030 |
peak | -0.026 |
heights | -0.024 |
comb | -0.024 |
itself | -0.023 |
Token | with |
Token ID | 351 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
horizon | +0.010 |
edly | +0.009 |
heights | +0.009 |
itself | +0.009 |
peak | +0.009 |
Neg logit contributions | |
living | -0.009 |
w | -0.007 |
mun | -0.007 |
d | -0.007 |
have | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | -3.88e-03 |
Pos logit contributions | |
horizon | +0.045 |
edly | +0.039 |
peak | +0.039 |
heights | +0.038 |
itself | +0.038 |
Neg logit contributions | |
living | -0.030 |
mun | -0.022 |
w | -0.022 |
have | -0.022 |
d | -0.021 |
Token | low |
Token ID | 1877 |
Feature activation | +8.34e-02 |
Pos logit contributions | |
living | +0.008 |
mun | +0.006 |
have | +0.006 |
w | +0.006 |
d | +0.006 |
Neg logit contributions | |
horizon | -0.007 |
peak | -0.006 |
itself | -0.006 |
heights | -0.006 |
edly | -0.005 |
Token | branch |
Token ID | 8478 |
Feature activation | +3.06e-02 |
Pos logit contributions | |
horizon | +0.107 |
ement | +0.085 |
edly | +0.083 |
heights | +0.077 |
rily | +0.077 |
Neg logit contributions | |
living | -0.201 |
w | -0.174 |
d | -0.174 |
have | -0.166 |
mun | -0.163 |
Token | . |
Token ID | 13 |
Feature activation | +2.40e-03 |
Pos logit contributions | |
horizon | +0.063 |
heights | +0.055 |
edly | +0.053 |
peak | +0.053 |
fingertips | +0.053 |
Neg logit contributions | |
living | -0.054 |
w | -0.040 |
d | -0.039 |
sm | -0.038 |
mun | -0.037 |
Token | He |
Token ID | 679 |
Feature activation | +2.57e-03 |
Pos logit contributions | |
horizon | +0.004 |
heights | +0.004 |
comb | +0.004 |
peak | +0.004 |
edly | +0.004 |
Neg logit contributions | |
living | -0.005 |
Good | -0.004 |
mun | -0.004 |
stole | -0.003 |
d | -0.003 |
Token | hung |
Token ID | 9174 |
Feature activation | +7.34e-03 |
Pos logit contributions | |
horizon | +0.006 |
itself | +0.005 |
peak | +0.005 |
heights | +0.005 |
edly | +0.005 |
Neg logit contributions | |
living | -0.003 |
mun | -0.003 |
have | -0.003 |
w | -0.003 |
d | -0.002 |
Token | his |
Token ID | 465 |
Feature activation | +6.37e-04 |
Pos logit contributions | |
horizon | +0.016 |
heights | +0.013 |
edly | +0.013 |
� | +0.013 |
peak | +0.013 |
Neg logit contributions | |
living | -0.012 |
d | -0.010 |
w | -0.009 |
have | -0.009 |
sm | -0.008 |
Token | seek |
Token ID | 5380 |
Feature activation | +3.24e-02 |
Pos logit contributions | |
horizon | +0.007 |
itself | +0.007 |
peak | +0.006 |
edly | +0.006 |
comb | +0.006 |
Neg logit contributions | |
living | -0.010 |
sm | -0.009 |
have | -0.009 |
w | -0.009 |
mun | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +4.95e-03 |
Pos logit contributions | |
horizon | +0.073 |
peak | +0.060 |
heights | +0.059 |
alks | +0.058 |
itself | +0.058 |
Neg logit contributions | |
living | -0.055 |
mun | -0.041 |
d | -0.041 |
have | -0.040 |
w | -0.038 |
Token | After |
Token ID | 2293 |
Feature activation | +5.00e-02 |
Pos logit contributions | |
horizon | +0.009 |
comb | +0.008 |
heights | +0.008 |
edly | +0.008 |
ement | +0.008 |
Neg logit contributions | |
living | -0.010 |
mun | -0.007 |
d | -0.007 |
Good | -0.007 |
have | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | +3.76e-03 |
Pos logit contributions | |
horizon | +0.090 |
itself | +0.088 |
heights | +0.085 |
� | +0.083 |
peak | +0.081 |
Neg logit contributions | |
living | -0.080 |
w | -0.074 |
mun | -0.073 |
have | -0.071 |
asc | -0.068 |
Token | few |
Token ID | 1178 |
Feature activation | +4.39e-02 |
Pos logit contributions | |
horizon | +0.006 |
peak | +0.006 |
� | +0.005 |
comb | +0.005 |
edly | +0.005 |
Neg logit contributions | |
living | -0.007 |
have | -0.007 |
mun | -0.006 |
w | -0.006 |
asc | -0.006 |
Token | minutes |
Token ID | 2431 |
Feature activation | +8.32e-02 |
Pos logit contributions | |
horizon | +0.076 |
itself | +0.062 |
peak | +0.062 |
edly | +0.061 |
heights | +0.061 |
Neg logit contributions | |
living | -0.090 |
mun | -0.076 |
have | -0.074 |
w | -0.074 |
d | -0.072 |
Token | , |
Token ID | 11 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
horizon | +0.165 |
heights | +0.137 |
edly | +0.137 |
peak | +0.135 |
itself | +0.134 |
Neg logit contributions | |
living | -0.154 |
mun | -0.124 |
d | -0.121 |
w | -0.120 |
have | -0.118 |
Token | one |
Token ID | 530 |
Feature activation | +5.41e-03 |
Pos logit contributions | |
horizon | +0.034 |
itself | +0.029 |
peak | +0.029 |
heights | +0.029 |
edly | +0.029 |
Neg logit contributions | |
living | -0.027 |
mun | -0.021 |
w | -0.021 |
have | -0.021 |
d | -0.020 |
Token | of |
Token ID | 286 |
Feature activation | +4.85e-02 |
Pos logit contributions | |
horizon | +0.009 |
peak | +0.008 |
itself | +0.008 |
comb | +0.008 |
heights | +0.008 |
Neg logit contributions | |
living | -0.010 |
mun | -0.009 |
have | -0.009 |
w | -0.009 |
haw | -0.008 |
Token | them |
Token ID | 606 |
Feature activation | +3.48e-02 |
Pos logit contributions | |
horizon | +0.101 |
edly | +0.086 |
comb | +0.085 |
peak | +0.085 |
heights | +0.083 |
Neg logit contributions | |
living | -0.084 |
mun | -0.067 |
w | -0.065 |
have | -0.065 |
d | -0.064 |
Token | found |
Token ID | 1043 |
Feature activation | +8.96e-03 |
Pos logit contributions | |
horizon | +0.079 |
peak | +0.071 |
itself | +0.069 |
comb | +0.068 |
heights | +0.068 |
Neg logit contributions | |
living | -0.048 |
mun | -0.038 |
w | -0.037 |
have | -0.036 |
haw | -0.035 |
Token | borrow |
Token ID | 8804 |
Feature activation | +9.06e-04 |
Pos logit contributions | |
horizon | +0.026 |
edly | +0.023 |
heights | +0.023 |
ement | +0.022 |
itself | +0.022 |
Neg logit contributions | |
living | -0.015 |
have | -0.011 |
d | -0.011 |
mun | -0.010 |
w | -0.010 |
Token | it |
Token ID | 340 |
Feature activation | +3.51e-02 |
Pos logit contributions | |
horizon | +0.002 |
edly | +0.002 |
comb | +0.002 |
peak | +0.002 |
heights | +0.002 |
Neg logit contributions | |
living | -0.001 |
have | -0.001 |
d | -0.001 |
w | -0.001 |
mun | -0.001 |
Token | for |
Token ID | 329 |
Feature activation | +3.37e-02 |
Pos logit contributions | |
horizon | +0.085 |
heights | +0.073 |
fingertips | +0.072 |
peak | +0.072 |
edly | +0.070 |
Neg logit contributions | |
living | -0.052 |
d | -0.036 |
have | -0.035 |
w | -0.034 |
sm | -0.032 |
Token | a |
Token ID | 257 |
Feature activation | +4.44e-03 |
Pos logit contributions | |
horizon | +0.084 |
peak | +0.071 |
ement | +0.069 |
comb | +0.069 |
edly | +0.068 |
Neg logit contributions | |
living | -0.046 |
have | -0.033 |
d | -0.032 |
w | -0.031 |
watching | -0.030 |
Token | few |
Token ID | 1178 |
Feature activation | +4.47e-02 |
Pos logit contributions | |
horizon | +0.009 |
peak | +0.008 |
comb | +0.007 |
heights | +0.007 |
itself | +0.007 |
Neg logit contributions | |
living | -0.008 |
w | -0.007 |
mun | -0.007 |
have | -0.006 |
asc | -0.006 |
Token | minutes |
Token ID | 2431 |
Feature activation | +8.29e-02 |
Pos logit contributions | |
horizon | +0.087 |
comb | +0.071 |
itself | +0.070 |
peak | +0.070 |
edly | +0.068 |
Neg logit contributions | |
living | -0.086 |
have | -0.070 |
mun | -0.068 |
d | -0.066 |
w | -0.066 |
Token | . |
Token ID | 13 |
Feature activation | +3.09e-03 |
Pos logit contributions | |
horizon | +0.178 |
heights | +0.155 |
peak | +0.154 |
edly | +0.147 |
comb | +0.147 |
Neg logit contributions | |
living | -0.139 |
d | -0.104 |
mun | -0.101 |
have | -0.100 |
watching | -0.097 |
Token | The |
Token ID | 383 |
Feature activation | +8.85e-03 |
Pos logit contributions | |
horizon | +0.007 |
peak | +0.006 |
heights | +0.006 |
comb | +0.006 |
edly | +0.006 |
Neg logit contributions | |
living | -0.005 |
mun | -0.004 |
Good | -0.004 |
L | -0.004 |
Enjoy | -0.004 |
Token | lady |
Token ID | 10846 |
Feature activation | +2.08e-03 |
Pos logit contributions | |
horizon | +0.015 |
peak | +0.013 |
itself | +0.013 |
heights | +0.013 |
edly | +0.013 |
Neg logit contributions | |
living | -0.018 |
mun | -0.015 |
w | -0.015 |
have | -0.015 |
d | -0.014 |
Token | was |
Token ID | 373 |
Feature activation | +9.81e-03 |
Pos logit contributions | |
horizon | +0.005 |
heights | +0.004 |
itself | +0.004 |
peak | +0.004 |
edly | +0.004 |
Neg logit contributions | |
living | -0.003 |
mun | -0.002 |
have | -0.002 |
w | -0.002 |
d | -0.002 |
Token | kind |
Token ID | 1611 |
Feature activation | -2.09e-03 |
Pos logit contributions | |
horizon | +0.021 |
heights | +0.017 |
peak | +0.017 |
edly | +0.017 |
itself | +0.017 |
Neg logit contributions | |
living | -0.018 |
have | -0.014 |
mun | -0.013 |
d | -0.013 |
watching | -0.012 |
Token | treats |
Token ID | 18432 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
horizon | +0.069 |
edly | +0.061 |
itself | +0.060 |
peak | +0.059 |
heights | +0.057 |
Neg logit contributions | |
living | -0.047 |
mun | -0.035 |
d | -0.034 |
have | -0.032 |
sm | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | +8.46e-03 |
Pos logit contributions | |
horizon | +0.026 |
heights | +0.023 |
edly | +0.022 |
peak | +0.022 |
ement | +0.021 |
Neg logit contributions | |
living | -0.018 |
d | -0.013 |
sm | -0.012 |
have | -0.012 |
mun | -0.012 |
Token | After |
Token ID | 2293 |
Feature activation | +4.83e-02 |
Pos logit contributions | |
horizon | +0.015 |
comb | +0.014 |
peak | +0.013 |
heights | +0.013 |
fingertips | +0.013 |
Neg logit contributions | |
living | -0.016 |
Good | -0.013 |
Enjoy | -0.012 |
loves | -0.012 |
mun | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | +1.65e-03 |
Pos logit contributions | |
horizon | +0.102 |
itself | +0.088 |
peak | +0.088 |
heights | +0.088 |
edly | +0.087 |
Neg logit contributions | |
living | -0.078 |
mun | -0.063 |
w | -0.061 |
have | -0.061 |
d | -0.058 |
Token | few |
Token ID | 1178 |
Feature activation | +3.98e-02 |
Pos logit contributions | |
horizon | +0.003 |
peak | +0.003 |
comb | +0.003 |
heights | +0.003 |
itself | +0.003 |
Neg logit contributions | |
living | -0.003 |
w | -0.003 |
have | -0.003 |
mun | -0.003 |
asc | -0.002 |
Token | minutes |
Token ID | 2431 |
Feature activation | +8.29e-02 |
Pos logit contributions | |
horizon | +0.068 |
itself | +0.056 |
peak | +0.055 |
edly | +0.055 |
heights | +0.054 |
Neg logit contributions | |
living | -0.083 |
mun | -0.070 |
have | -0.068 |
w | -0.067 |
d | -0.066 |
Token | in |
Token ID | 287 |
Feature activation | -7.82e-04 |
Pos logit contributions | |
horizon | +0.183 |
edly | +0.160 |
heights | +0.157 |
itself | +0.151 |
peak | +0.150 |
Neg logit contributions | |
living | -0.136 |
d | -0.105 |
mun | -0.102 |
w | -0.101 |
watching | -0.096 |
Token | the |
Token ID | 262 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
living | +0.001 |
watching | +0.001 |
d | +0.001 |
taking | +0.001 |
talking | +0.001 |
Neg logit contributions | |
edly | -0.002 |
horizon | -0.002 |
iously | -0.002 |
comb | -0.002 |
rily | -0.002 |
Token | oven |
Token ID | 14361 |
Feature activation | -2.91e-02 |
Pos logit contributions | |
living | +0.030 |
mun | +0.024 |
have | +0.024 |
w | +0.024 |
d | +0.023 |
Neg logit contributions | |
horizon | -0.027 |
itself | -0.024 |
edly | -0.023 |
heights | -0.023 |
peak | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | +6.25e-03 |
Pos logit contributions | |
living | +0.047 |
d | +0.034 |
mun | +0.032 |
w | +0.031 |
came | +0.031 |
Neg logit contributions | |
horizon | -0.062 |
edly | -0.059 |
heights | -0.058 |
itself | -0.058 |
rily | -0.057 |
Token | the |
Token ID | 262 |
Feature activation | +2.11e-03 |
Pos logit contributions | |
horizon | +0.012 |
edly | +0.010 |
peak | +0.010 |
heights | +0.010 |
itself | +0.010 |
Neg logit contributions | |
living | -0.011 |
mun | -0.009 |
w | -0.009 |
have | -0.009 |
d | -0.009 |
Token | create |
Token ID | 2251 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
living | +0.026 |
mun | +0.022 |
w | +0.022 |
have | +0.021 |
sm | +0.021 |
Neg logit contributions | |
horizon | -0.029 |
itself | -0.026 |
peak | -0.026 |
heights | -0.025 |
comb | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +3.87e-03 |
Pos logit contributions | |
living | +0.017 |
mun | +0.012 |
have | +0.012 |
d | +0.012 |
w | +0.012 |
Neg logit contributions | |
horizon | -0.025 |
edly | -0.021 |
peak | -0.021 |
comb | -0.021 |
heights | -0.020 |
Token | After |
Token ID | 2293 |
Feature activation | +4.87e-02 |
Pos logit contributions | |
horizon | +0.009 |
lights | +0.008 |
comb | +0.008 |
edly | +0.008 |
ement | +0.008 |
Neg logit contributions | |
living | -0.006 |
Enjoy | -0.004 |
mun | -0.004 |
d | -0.004 |
Good | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | +8.60e-04 |
Pos logit contributions | |
horizon | +0.100 |
itself | +0.090 |
heights | +0.088 |
peak | +0.087 |
edly | +0.085 |
Neg logit contributions | |
living | -0.075 |
mun | -0.065 |
w | -0.065 |
have | -0.063 |
haw | -0.060 |
Token | few |
Token ID | 1178 |
Feature activation | +4.07e-02 |
Pos logit contributions | |
horizon | +0.002 |
peak | +0.002 |
comb | +0.001 |
� | +0.001 |
itself | +0.001 |
Neg logit contributions | |
living | -0.001 |
have | -0.001 |
w | -0.001 |
asc | -0.001 |
mun | -0.001 |
Token | minutes |
Token ID | 2431 |
Feature activation | +8.27e-02 |
Pos logit contributions | |
horizon | +0.068 |
itself | +0.055 |
peak | +0.055 |
edly | +0.054 |
heights | +0.054 |
Neg logit contributions | |
living | -0.086 |
mun | -0.072 |
have | -0.071 |
w | -0.071 |
d | -0.069 |
Token | of |
Token ID | 286 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
horizon | +0.191 |
edly | +0.168 |
heights | +0.162 |
peak | +0.159 |
rily | +0.158 |
Neg logit contributions | |
living | -0.131 |
d | -0.101 |
mun | -0.093 |
watching | -0.091 |
sm | -0.089 |
Token | working |
Token ID | 1762 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
horizon | +0.051 |
itself | +0.045 |
heights | +0.045 |
peak | +0.044 |
edly | +0.043 |
Neg logit contributions | |
living | -0.037 |
mun | -0.030 |
w | -0.030 |
have | -0.029 |
d | -0.027 |
Token | she |
Token ID | 673 |
Feature activation | +1.35e-02 |
Pos logit contributions | |
living | +0.029 |
d | +0.022 |
mun | +0.022 |
w | +0.020 |
have | +0.020 |
Neg logit contributions | |
horizon | -0.042 |
edly | -0.037 |
comb | -0.035 |
peak | -0.035 |
fingertips | -0.034 |
Token | had |
Token ID | 550 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
horizon | +0.034 |
itself | +0.030 |
peak | +0.030 |
heights | +0.030 |
edly | +0.029 |
Neg logit contributions | |
living | -0.017 |
mun | -0.012 |
w | -0.012 |
have | -0.012 |
d | -0.011 |
Token | created |
Token ID | 2727 |
Feature activation | -2.54e-03 |
Pos logit contributions | |
living | +0.025 |
mun | +0.020 |
w | +0.020 |
have | +0.019 |
haw | +0.018 |
Neg logit contributions | |
horizon | -0.035 |
itself | -0.031 |
peak | -0.030 |
heights | -0.030 |
edly | -0.030 |
Token | keep |
Token ID | 1394 |
Feature activation | -8.48e-03 |
Pos logit contributions | |
living | +0.047 |
mun | +0.037 |
have | +0.036 |
w | +0.036 |
d | +0.034 |
Neg logit contributions | |
horizon | -0.066 |
edly | -0.057 |
peak | -0.056 |
itself | -0.056 |
heights | -0.056 |
Token | people |
Token ID | 661 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
living | +0.012 |
mun | +0.010 |
w | +0.010 |
have | +0.009 |
d | +0.009 |
Neg logit contributions | |
horizon | -0.019 |
itself | -0.017 |
heights | -0.017 |
peak | -0.016 |
edly | -0.016 |
Token | safe |
Token ID | 3338 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
living | +0.039 |
mun | +0.032 |
w | +0.031 |
have | +0.031 |
d | +0.030 |
Neg logit contributions | |
horizon | -0.042 |
itself | -0.035 |
peak | -0.035 |
heights | -0.035 |
edly | -0.035 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.47e-02 |
Pos logit contributions | |
horizon | +0.023 |
peak | +0.020 |
ement | +0.020 |
heights | +0.020 |
� | +0.020 |
Neg logit contributions | |
living | -0.020 |
have | -0.015 |
mun | -0.015 |
d | -0.015 |
w | -0.015 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
horizon | +0.028 |
heights | +0.024 |
peak | +0.024 |
itself | +0.023 |
comb | +0.023 |
Neg logit contributions | |
living | -0.028 |
mun | -0.023 |
have | -0.022 |
w | -0.021 |
d | -0.021 |
Token | P |
Token ID | 47 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
living | +0.039 |
mun | +0.031 |
have | +0.030 |
w | +0.030 |
d | +0.028 |
Neg logit contributions | |
horizon | -0.057 |
itself | -0.050 |
peak | -0.049 |
heights | -0.049 |
edly | -0.048 |
Token | ete |
Token ID | 14471 |
Feature activation | -5.24e-02 |
Pos logit contributions | |
living | +0.159 |
w | +0.114 |
have | +0.112 |
mun | +0.112 |
P | +0.108 |
Neg logit contributions | |
horizon | -0.290 |
edly | -0.263 |
heights | -0.259 |
itself | -0.256 |
comb | -0.248 |
Token | was |
Token ID | 373 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
living | +0.110 |
mun | +0.091 |
w | +0.090 |
have | +0.090 |
d | +0.089 |
Neg logit contributions | |
horizon | -0.087 |
heights | -0.071 |
peak | -0.071 |
itself | -0.070 |
edly | -0.069 |
Token | a |
Token ID | 257 |
Feature activation | -3.21e-02 |
Pos logit contributions | |
living | +0.020 |
mun | +0.017 |
w | +0.017 |
have | +0.016 |
haw | +0.015 |
Neg logit contributions | |
horizon | -0.031 |
peak | -0.027 |
itself | -0.027 |
heights | -0.026 |
edly | -0.026 |
Token | little |
Token ID | 1310 |
Feature activation | +4.12e-03 |
Pos logit contributions | |
living | +0.052 |
w | +0.044 |
mun | +0.043 |
have | +0.043 |
haw | +0.039 |
Neg logit contributions | |
horizon | -0.066 |
peak | -0.057 |
heights | -0.056 |
itself | -0.055 |
edly | -0.055 |
Token | boy |
Token ID | 2933 |
Feature activation | +3.63e-04 |
Pos logit contributions | |
horizon | +0.009 |
peak | +0.008 |
heights | +0.008 |
itself | +0.008 |
edly | +0.008 |
Neg logit contributions | |
living | -0.006 |
mun | -0.005 |
w | -0.005 |
have | -0.005 |
d | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | -8.58e-03 |
Pos logit contributions | |
horizon | +0.014 |
ement | +0.012 |
edly | +0.012 |
heights | +0.012 |
peak | +0.011 |
Neg logit contributions | |
living | -0.011 |
d | -0.008 |
have | -0.008 |
loves | -0.007 |
sm | -0.007 |
Token | went |
Token ID | 1816 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
living | +0.014 |
mun | +0.011 |
w | +0.011 |
have | +0.011 |
sm | +0.010 |
Neg logit contributions | |
horizon | -0.018 |
itself | -0.016 |
edly | -0.015 |
peak | -0.015 |
heights | -0.015 |
Token | home |
Token ID | 1363 |
Feature activation | -4.75e-02 |
Pos logit contributions | |
living | +0.036 |
d | +0.028 |
have | +0.028 |
mun | +0.027 |
w | +0.027 |
Neg logit contributions | |
horizon | -0.032 |
heights | -0.027 |
comb | -0.026 |
peak | -0.026 |
ement | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -3.33e-03 |
Pos logit contributions | |
living | +0.087 |
mun | +0.069 |
have | +0.067 |
d | +0.066 |
w | +0.065 |
Neg logit contributions | |
horizon | -0.094 |
heights | -0.079 |
peak | -0.078 |
itself | -0.077 |
edly | -0.076 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.87e-02 |
Pos logit contributions | |
living | +0.006 |
mun | +0.005 |
have | +0.005 |
w | +0.005 |
d | +0.004 |
Neg logit contributions | |
horizon | -0.007 |
heights | -0.006 |
peak | -0.006 |
edly | -0.006 |
itself | -0.005 |
Token | P |
Token ID | 47 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
living | +0.040 |
mun | +0.032 |
have | +0.031 |
w | +0.031 |
d | +0.029 |
Neg logit contributions | |
horizon | -0.066 |
itself | -0.058 |
peak | -0.057 |
heights | -0.056 |
edly | -0.056 |
Token | ete |
Token ID | 14471 |
Feature activation | -4.74e-02 |
Pos logit contributions | |
living | +0.136 |
w | +0.095 |
have | +0.092 |
mun | +0.091 |
d | +0.086 |
Neg logit contributions | |
horizon | -0.289 |
edly | -0.265 |
heights | -0.259 |
itself | -0.259 |
comb | -0.252 |
Token | was |
Token ID | 373 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
living | +0.083 |
mun | +0.070 |
have | +0.068 |
w | +0.068 |
sm | +0.063 |
Neg logit contributions | |
horizon | -0.091 |
itself | -0.078 |
peak | -0.077 |
heights | -0.076 |
edly | -0.075 |
Token | a |
Token ID | 257 |
Feature activation | -3.22e-02 |
Pos logit contributions | |
living | +0.020 |
mun | +0.018 |
w | +0.018 |
haw | +0.017 |
have | +0.017 |
Neg logit contributions | |
horizon | -0.027 |
peak | -0.023 |
itself | -0.023 |
heights | -0.022 |
edly | -0.022 |
Token | gentle |
Token ID | 10296 |
Feature activation | -1.63e-03 |
Pos logit contributions | |
living | +0.053 |
w | +0.047 |
mun | +0.047 |
have | +0.046 |
haw | +0.042 |
Neg logit contributions | |
horizon | -0.065 |
peak | -0.056 |
heights | -0.053 |
edly | -0.052 |
comb | -0.051 |
Token | boy |
Token ID | 2933 |
Feature activation | -1.39e-03 |
Pos logit contributions | |
living | +0.003 |
mun | +0.002 |
d | +0.002 |
have | +0.002 |
w | +0.002 |
Neg logit contributions | |
horizon | -0.003 |
itself | -0.003 |
heights | -0.003 |
peak | -0.003 |
edly | -0.003 |
Token | discover |
Token ID | 7073 |
Feature activation | -4.09e-04 |
Pos logit contributions | |
living | +0.030 |
w | +0.025 |
mun | +0.024 |
have | +0.023 |
sm | +0.023 |
Neg logit contributions | |
horizon | -0.035 |
peak | -0.031 |
itself | -0.030 |
heights | -0.030 |
edly | -0.029 |
Token | more |
Token ID | 517 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
living | +0.001 |
mun | +0.001 |
w | +0.001 |
have | +0.001 |
d | +0.001 |
Neg logit contributions | |
horizon | -0.001 |
itself | -0.001 |
peak | -0.001 |
heights | -0.001 |
edly | -0.001 |
Token | surprises |
Token ID | 24072 |
Feature activation | +9.87e-03 |
Pos logit contributions | |
horizon | +0.037 |
peak | +0.032 |
itself | +0.032 |
edly | +0.032 |
comb | +0.031 |
Neg logit contributions | |
living | -0.030 |
mun | -0.023 |
have | -0.023 |
w | -0.022 |
d | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -6.05e-03 |
Pos logit contributions | |
horizon | +0.021 |
peak | +0.018 |
heights | +0.018 |
edly | +0.017 |
itself | +0.017 |
Neg logit contributions | |
living | -0.017 |
mun | -0.013 |
have | -0.013 |
w | -0.012 |
d | -0.012 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.63e-02 |
Pos logit contributions | |
living | +0.010 |
Good | +0.008 |
mun | +0.008 |
have | +0.008 |
Enjoy | +0.007 |
Neg logit contributions | |
horizon | -0.013 |
heights | -0.011 |
comb | -0.011 |
peak | -0.011 |
edly | -0.011 |
Token | P |
Token ID | 47 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
living | +0.044 |
mun | +0.035 |
w | +0.034 |
have | +0.034 |
d | +0.032 |
Neg logit contributions | |
horizon | -0.055 |
heights | -0.047 |
peak | -0.047 |
itself | -0.047 |
edly | -0.047 |
Token | ete |
Token ID | 14471 |
Feature activation | -4.55e-02 |
Pos logit contributions | |
living | +0.141 |
w | +0.101 |
mun | +0.101 |
have | +0.100 |
d | +0.093 |
Neg logit contributions | |
horizon | -0.285 |
edly | -0.256 |
itself | -0.253 |
heights | -0.252 |
peak | -0.246 |
Token | was |
Token ID | 373 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
living | +0.092 |
mun | +0.078 |
w | +0.076 |
have | +0.076 |
d | +0.073 |
Neg logit contributions | |
horizon | -0.077 |
itself | -0.063 |
peak | -0.063 |
heights | -0.063 |
edly | -0.062 |
Token | walking |
Token ID | 6155 |
Feature activation | -3.23e-02 |
Pos logit contributions | |
living | +0.020 |
mun | +0.016 |
w | +0.016 |
have | +0.015 |
d | +0.014 |
Neg logit contributions | |
horizon | -0.030 |
itself | -0.026 |
peak | -0.026 |
heights | -0.025 |
edly | -0.025 |
Token | in |
Token ID | 287 |
Feature activation | -2.46e-02 |
Pos logit contributions | |
living | +0.058 |
mun | +0.049 |
w | +0.048 |
have | +0.048 |
d | +0.045 |
Neg logit contributions | |
horizon | -0.061 |
itself | -0.052 |
peak | -0.052 |
edly | -0.051 |
heights | -0.051 |
Token | the |
Token ID | 262 |
Feature activation | -1.95e-02 |
Pos logit contributions | |
living | +0.039 |
d | +0.024 |
watching | +0.024 |
mun | +0.023 |
have | +0.022 |
Neg logit contributions | |
horizon | -0.057 |
edly | -0.054 |
comb | -0.053 |
ement | -0.052 |
rily | -0.049 |
Token | and |
Token ID | 290 |
Feature activation | -5.37e-03 |
Pos logit contributions | |
living | +0.009 |
have | +0.007 |
mun | +0.007 |
d | +0.007 |
w | +0.007 |
Neg logit contributions | |
horizon | -0.011 |
itself | -0.009 |
peak | -0.009 |
edly | -0.009 |
ement | -0.009 |
Token | visit |
Token ID | 3187 |
Feature activation | +8.69e-04 |
Pos logit contributions | |
living | +0.009 |
w | +0.007 |
mun | +0.007 |
have | +0.007 |
sm | +0.007 |
Neg logit contributions | |
horizon | -0.011 |
peak | -0.009 |
itself | -0.009 |
heights | -0.009 |
comb | -0.009 |
Token | again |
Token ID | 757 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
horizon | +0.002 |
peak | +0.001 |
heights | +0.001 |
edly | +0.001 |
comb | +0.001 |
Neg logit contributions | |
living | -0.002 |
mun | -0.001 |
d | -0.001 |
have | -0.001 |
w | -0.001 |
Token | ! |
Token ID | 0 |
Feature activation | -1.53e-03 |
Pos logit contributions | |
horizon | +0.020 |
peak | +0.017 |
itself | +0.017 |
heights | +0.017 |
edly | +0.017 |
Neg logit contributions | |
living | -0.020 |
mun | -0.016 |
have | -0.016 |
w | -0.015 |
d | -0.015 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
living | +0.002 |
Good | +0.002 |
mun | +0.002 |
have | +0.002 |
L | +0.002 |
Neg logit contributions | |
horizon | -0.003 |
heights | -0.003 |
peak | -0.003 |
ement | -0.003 |
edly | -0.003 |
Token | P |
Token ID | 47 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
living | +0.039 |
mun | +0.031 |
w | +0.030 |
have | +0.030 |
d | +0.028 |
Neg logit contributions | |
horizon | -0.052 |
peak | -0.045 |
itself | -0.045 |
heights | -0.045 |
edly | -0.044 |
Token | olly |
Token ID | 5098 |
Feature activation | -3.94e-02 |
Pos logit contributions | |
living | +0.147 |
w | +0.107 |
have | +0.106 |
mun | +0.106 |
d | +0.098 |
Neg logit contributions | |
horizon | -0.279 |
edly | -0.251 |
itself | -0.247 |
heights | -0.247 |
comb | -0.239 |
Token | was |
Token ID | 373 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
living | +0.080 |
mun | +0.068 |
w | +0.067 |
have | +0.067 |
d | +0.064 |
Neg logit contributions | |
horizon | -0.066 |
itself | -0.054 |
peak | -0.054 |
heights | -0.054 |
edly | -0.053 |
Token | walking |
Token ID | 6155 |
Feature activation | -2.81e-02 |
Pos logit contributions | |
living | +0.019 |
mun | +0.015 |
w | +0.015 |
have | +0.015 |
haw | +0.014 |
Neg logit contributions | |
horizon | -0.026 |
itself | -0.023 |
peak | -0.022 |
edly | -0.022 |
heights | -0.022 |
Token | in |
Token ID | 287 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
living | +0.050 |
mun | +0.042 |
w | +0.041 |
have | +0.041 |
d | +0.038 |
Neg logit contributions | |
horizon | -0.054 |
itself | -0.046 |
peak | -0.045 |
edly | -0.045 |
heights | -0.045 |
Token | the |
Token ID | 262 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
living | +0.034 |
d | +0.021 |
watching | +0.021 |
mun | +0.020 |
sm | +0.019 |
Neg logit contributions | |
horizon | -0.051 |
edly | -0.048 |
comb | -0.046 |
ement | -0.046 |
rily | -0.044 |
Token | party |
Token ID | 2151 |
Feature activation | +3.75e-04 |
Pos logit contributions | |
living | +0.020 |
d | +0.015 |
have | +0.014 |
mun | +0.014 |
w | +0.014 |
Neg logit contributions | |
horizon | -0.028 |
heights | -0.024 |
peak | -0.024 |
fingertips | -0.023 |
edly | -0.023 |
Token | perfect |
Token ID | 2818 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
horizon | +0.001 |
peak | +0.001 |
heights | +0.001 |
fingertips | +0.001 |
edly | +0.001 |
Neg logit contributions | |
living | -0.001 |
d | -0.000 |
have | -0.000 |
loves | -0.000 |
w | -0.000 |
Token | ." |
Token ID | 526 |
Feature activation | +3.90e-03 |
Pos logit contributions | |
living | +0.033 |
w | +0.025 |
d | +0.025 |
have | +0.025 |
mun | +0.024 |
Neg logit contributions | |
horizon | -0.040 |
peak | -0.034 |
heights | -0.034 |
itself | -0.032 |
ement | -0.032 |
Token |
|
Token ID | 198 |
Feature activation | -1.14e-02 |
Pos logit contributions | |
horizon | +0.009 |
peak | +0.009 |
heights | +0.008 |
comb | +0.008 |
ement | +0.008 |
Neg logit contributions | |
living | -0.006 |
mun | -0.004 |
L | -0.004 |
Good | -0.004 |
Are | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -2.78e-03 |
Pos logit contributions | |
living | +0.020 |
mun | +0.016 |
have | +0.016 |
w | +0.016 |
d | +0.015 |
Neg logit contributions | |
horizon | -0.023 |
peak | -0.020 |
heights | -0.020 |
edly | -0.019 |
comb | -0.019 |
Token | P |
Token ID | 47 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
living | +0.003 |
mun | +0.003 |
haw | +0.003 |
d | +0.003 |
w | +0.003 |
Neg logit contributions | |
horizon | -0.006 |
itself | -0.006 |
edly | -0.006 |
heights | -0.005 |
� | -0.005 |
Token | inky |
Token ID | 29246 |
Feature activation | -4.58e-02 |
Pos logit contributions | |
living | +0.196 |
mun | +0.160 |
w | +0.157 |
have | +0.156 |
d | +0.149 |
Neg logit contributions | |
horizon | -0.228 |
peak | -0.194 |
itself | -0.193 |
heights | -0.193 |
edly | -0.190 |
Token | smiled |
Token ID | 13541 |
Feature activation | -2.26e-02 |
Pos logit contributions | |
living | +0.070 |
w | +0.059 |
mun | +0.058 |
have | +0.057 |
haw | +0.053 |
Neg logit contributions | |
horizon | -0.094 |
itself | -0.084 |
edly | -0.083 |
peak | -0.080 |
comb | -0.080 |
Token | and |
Token ID | 290 |
Feature activation | -2.30e-02 |
Pos logit contributions | |
living | +0.043 |
d | +0.034 |
loves | +0.031 |
have | +0.031 |
sm | +0.031 |
Neg logit contributions | |
horizon | -0.045 |
peak | -0.039 |
heights | -0.039 |
� | -0.039 |
comb | -0.038 |
Token | said |
Token ID | 531 |
Feature activation | +1.10e-03 |
Pos logit contributions | |
living | +0.033 |
mun | +0.027 |
have | +0.026 |
w | +0.026 |
sm | +0.023 |
Neg logit contributions | |
horizon | -0.051 |
edly | -0.046 |
itself | -0.046 |
peak | -0.044 |
comb | -0.044 |
Token | , |
Token ID | 11 |
Feature activation | -1.75e-03 |
Pos logit contributions | |
horizon | +0.001 |
itself | +0.001 |
edly | +0.001 |
ement | +0.001 |
heights | +0.001 |
Neg logit contributions | |
living | -0.003 |
mun | -0.002 |
have | -0.002 |
w | -0.002 |
Did | -0.002 |
Token | with |
Token ID | 351 |
Feature activation | +4.03e-03 |
Pos logit contributions | |
horizon | +0.014 |
heights | +0.012 |
peak | +0.011 |
comb | +0.011 |
fingertips | +0.011 |
Neg logit contributions | |
living | -0.008 |
sm | -0.005 |
d | -0.005 |
mun | -0.005 |
loves | -0.005 |
Token | his |
Token ID | 465 |
Feature activation | +1.19e-04 |
Pos logit contributions | |
horizon | +0.009 |
comb | +0.008 |
edly | +0.008 |
peak | +0.008 |
heights | +0.008 |
Neg logit contributions | |
living | -0.006 |
mun | -0.005 |
d | -0.004 |
have | -0.004 |
w | -0.004 |
Token | friends |
Token ID | 2460 |
Feature activation | +8.91e-03 |
Pos logit contributions | |
horizon | +0.000 |
edly | +0.000 |
comb | +0.000 |
itself | +0.000 |
peak | +0.000 |
Neg logit contributions | |
living | -0.000 |
mun | -0.000 |
d | -0.000 |
have | -0.000 |
w | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | +2.65e-03 |
Pos logit contributions | |
horizon | +0.018 |
heights | +0.016 |
peak | +0.015 |
itself | +0.015 |
edly | +0.015 |
Neg logit contributions | |
living | -0.016 |
mun | -0.012 |
d | -0.012 |
have | -0.011 |
sm | -0.011 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.47e-02 |
Pos logit contributions | |
horizon | +0.006 |
heights | +0.005 |
peak | +0.005 |
comb | +0.005 |
edly | +0.005 |
Neg logit contributions | |
living | -0.004 |
mun | -0.003 |
Good | -0.003 |
sm | -0.003 |
have | -0.003 |
Token | P |
Token ID | 47 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
living | +0.036 |
mun | +0.028 |
have | +0.028 |
w | +0.028 |
d | +0.026 |
Neg logit contributions | |
horizon | -0.055 |
itself | -0.048 |
peak | -0.048 |
heights | -0.048 |
edly | -0.047 |
Token | ete |
Token ID | 14471 |
Feature activation | -4.48e-02 |
Pos logit contributions | |
living | +0.130 |
w | +0.094 |
have | +0.089 |
mun | +0.088 |
P | +0.083 |
Neg logit contributions | |
horizon | -0.293 |
edly | -0.268 |
heights | -0.262 |
itself | -0.258 |
comb | -0.254 |
Token | was |
Token ID | 373 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
living | +0.081 |
mun | +0.066 |
w | +0.065 |
have | +0.064 |
d | +0.062 |
Neg logit contributions | |
horizon | -0.086 |
itself | -0.073 |
peak | -0.073 |
heights | -0.073 |
edly | -0.072 |
Token | walking |
Token ID | 6155 |
Feature activation | -2.97e-02 |
Pos logit contributions | |
living | +0.018 |
mun | +0.015 |
w | +0.015 |
have | +0.014 |
haw | +0.014 |
Neg logit contributions | |
horizon | -0.025 |
peak | -0.022 |
itself | -0.022 |
heights | -0.022 |
edly | -0.021 |
Token | in |
Token ID | 287 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
living | +0.053 |
mun | +0.044 |
w | +0.043 |
have | +0.043 |
d | +0.041 |
Neg logit contributions | |
horizon | -0.058 |
itself | -0.049 |
peak | -0.049 |
heights | -0.049 |
edly | -0.048 |
Token | the |
Token ID | 262 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
living | +0.036 |
d | +0.024 |
mun | +0.024 |
watching | +0.023 |
have | +0.023 |
Neg logit contributions | |
horizon | -0.049 |
edly | -0.046 |
comb | -0.045 |
ement | -0.043 |
rily | -0.042 |
Token | just |
Token ID | 655 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
horizon | +0.000 |
itself | +0.000 |
peak | +0.000 |
heights | +0.000 |
edly | +0.000 |
Neg logit contributions | |
living | -0.000 |
mun | -0.000 |
w | -0.000 |
have | -0.000 |
d | -0.000 |
Token | in |
Token ID | 287 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
living | +0.052 |
mun | +0.045 |
w | +0.044 |
have | +0.043 |
P | +0.041 |
Neg logit contributions | |
horizon | -0.047 |
itself | -0.040 |
heights | -0.040 |
peak | -0.039 |
edly | -0.039 |
Token | case |
Token ID | 1339 |
Feature activation | -2.39e-02 |
Pos logit contributions | |
living | +0.028 |
d | +0.018 |
sm | +0.017 |
mun | +0.016 |
L | +0.016 |
Neg logit contributions | |
horizon | -0.043 |
edly | -0.040 |
overed | -0.037 |
iously | -0.037 |
rily | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | -7.72e-04 |
Pos logit contributions | |
living | +0.043 |
mun | +0.033 |
w | +0.032 |
d | +0.031 |
sm | +0.030 |
Neg logit contributions | |
horizon | -0.050 |
peak | -0.042 |
edly | -0.041 |
heights | -0.041 |
comb | -0.040 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
living | +0.001 |
mun | +0.001 |
Good | +0.001 |
have | +0.001 |
w | +0.001 |
Neg logit contributions | |
horizon | -0.001 |
heights | -0.001 |
peak | -0.001 |
itself | -0.001 |
edly | -0.001 |
Token | P |
Token ID | 47 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
living | +0.042 |
mun | +0.033 |
w | +0.032 |
have | +0.032 |
d | +0.031 |
Neg logit contributions | |
horizon | -0.050 |
heights | -0.043 |
peak | -0.043 |
itself | -0.042 |
edly | -0.042 |
Token | ete |
Token ID | 14471 |
Feature activation | -4.18e-02 |
Pos logit contributions | |
living | +0.128 |
w | +0.089 |
mun | +0.084 |
have | +0.083 |
P | +0.079 |
Neg logit contributions | |
horizon | -0.288 |
edly | -0.264 |
heights | -0.258 |
itself | -0.256 |
peak | -0.249 |
Token | was |
Token ID | 373 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
living | +0.077 |
mun | +0.066 |
have | +0.065 |
w | +0.064 |
haw | +0.059 |
Neg logit contributions | |
horizon | -0.076 |
itself | -0.064 |
peak | -0.063 |
heights | -0.062 |
edly | -0.062 |
Token | worried |
Token ID | 7960 |
Feature activation | -1.49e-02 |
Pos logit contributions | |
living | +0.015 |
mun | +0.012 |
w | +0.012 |
have | +0.011 |
haw | +0.011 |
Neg logit contributions | |
horizon | -0.023 |
itself | -0.020 |
peak | -0.020 |
heights | -0.020 |
edly | -0.020 |
Token | about |
Token ID | 546 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
living | +0.029 |
mun | +0.023 |
have | +0.023 |
w | +0.022 |
d | +0.022 |
Neg logit contributions | |
horizon | -0.027 |
heights | -0.023 |
peak | -0.023 |
itself | -0.022 |
edly | -0.022 |
Token | what |
Token ID | 644 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
horizon | +0.039 |
itself | +0.034 |
heights | +0.034 |
peak | +0.033 |
edly | +0.032 |
Neg logit contributions | |
living | -0.037 |
mun | -0.031 |
w | -0.031 |
have | -0.030 |
d | -0.029 |
Token | our |
Token ID | 674 |
Feature activation | -1.66e-02 |
Pos logit contributions | |
living | +0.018 |
mun | +0.013 |
w | +0.013 |
have | +0.013 |
d | +0.012 |
Neg logit contributions | |
horizon | -0.030 |
peak | -0.026 |
comb | -0.026 |
edly | -0.026 |
heights | -0.026 |
Token | toys |
Token ID | 14958 |
Feature activation | -7.18e-03 |
Pos logit contributions | |
living | +0.026 |
mun | +0.023 |
w | +0.022 |
have | +0.022 |
sm | +0.020 |
Neg logit contributions | |
horizon | -0.034 |
heights | -0.029 |
peak | -0.029 |
itself | -0.029 |
edly | -0.028 |
Token | ." |
Token ID | 526 |
Feature activation | -5.29e-04 |
Pos logit contributions | |
living | +0.012 |
d | +0.009 |
w | +0.008 |
have | +0.008 |
loves | +0.008 |
Neg logit contributions | |
horizon | -0.016 |
peak | -0.014 |
fingertips | -0.014 |
heights | -0.014 |
ement | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | -5.78e-03 |
Pos logit contributions | |
living | +0.001 |
mun | +0.001 |
have | +0.001 |
d | +0.001 |
w | +0.001 |
Neg logit contributions | |
horizon | -0.001 |
peak | -0.001 |
heights | -0.001 |
comb | -0.001 |
itself | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +1.49e-03 |
Pos logit contributions | |
living | +0.010 |
mun | +0.008 |
w | +0.008 |
have | +0.008 |
d | +0.008 |
Neg logit contributions | |
horizon | -0.012 |
peak | -0.010 |
heights | -0.010 |
edly | -0.010 |
itself | -0.010 |
Token | P |
Token ID | 47 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
horizon | +0.003 |
itself | +0.003 |
edly | +0.003 |
peak | +0.002 |
heights | +0.002 |
Neg logit contributions | |
living | -0.002 |
mun | -0.002 |
w | -0.002 |
have | -0.002 |
d | -0.002 |
Token | apa |
Token ID | 32678 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
living | +0.244 |
mun | +0.211 |
w | +0.209 |
have | +0.207 |
d | +0.200 |
Neg logit contributions | |
horizon | -0.169 |
itself | -0.134 |
peak | -0.133 |
heights | -0.132 |
edly | -0.131 |
Token | says |
Token ID | 1139 |
Feature activation | +7.63e-03 |
Pos logit contributions | |
living | +0.039 |
mun | +0.032 |
w | +0.032 |
have | +0.032 |
sm | +0.029 |
Neg logit contributions | |
horizon | -0.039 |
itself | -0.034 |
edly | -0.033 |
peak | -0.033 |
heights | -0.032 |
Token | : |
Token ID | 25 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
horizon | +0.012 |
itself | +0.011 |
edly | +0.010 |
peak | +0.010 |
ement | +0.010 |
Neg logit contributions | |
living | -0.016 |
mun | -0.014 |
have | -0.013 |
w | -0.013 |
d | -0.012 |
Token | " |
Token ID | 366 |
Feature activation | +4.83e-03 |
Pos logit contributions | |
living | +0.035 |
w | +0.031 |
mun | +0.031 |
have | +0.031 |
haw | +0.029 |
Neg logit contributions | |
horizon | -0.031 |
itself | -0.026 |
� | -0.025 |
edly | -0.025 |
peak | -0.025 |
Token | You |
Token ID | 1639 |
Feature activation | -2.58e-02 |
Pos logit contributions | |
horizon | +0.011 |
itself | +0.009 |
edly | +0.009 |
heights | +0.009 |
peak | +0.009 |
Neg logit contributions | |
living | -0.007 |
mun | -0.006 |
have | -0.005 |
d | -0.005 |
w | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | -7.59e-04 |
Pos logit contributions | |
living | +0.008 |
mun | +0.007 |
w | +0.007 |
have | +0.007 |
d | +0.006 |
Neg logit contributions | |
horizon | -0.012 |
itself | -0.010 |
peak | -0.010 |
edly | -0.010 |
heights | -0.010 |
Token | perfect |
Token ID | 2818 |
Feature activation | +6.46e-04 |
Pos logit contributions | |
living | +0.001 |
mun | +0.001 |
w | +0.001 |
have | +0.001 |
sm | +0.001 |
Neg logit contributions | |
horizon | -0.001 |
heights | -0.001 |
peak | -0.001 |
itself | -0.001 |
edly | -0.001 |
Token | haircut |
Token ID | 45548 |
Feature activation | -6.04e-03 |
Pos logit contributions | |
horizon | +0.001 |
itself | +0.001 |
peak | +0.001 |
heights | +0.001 |
edly | +0.001 |
Neg logit contributions | |
living | -0.001 |
mun | -0.001 |
w | -0.001 |
have | -0.001 |
d | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -6.97e-04 |
Pos logit contributions | |
living | +0.011 |
mun | +0.008 |
d | +0.008 |
w | +0.008 |
have | +0.008 |
Neg logit contributions | |
horizon | -0.013 |
heights | -0.011 |
peak | -0.011 |
itself | -0.010 |
edly | -0.010 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.40e-02 |
Pos logit contributions | |
living | +0.001 |
mun | +0.001 |
w | +0.001 |
d | +0.001 |
have | +0.001 |
Neg logit contributions | |
horizon | -0.001 |
heights | -0.001 |
peak | -0.001 |
edly | -0.001 |
itself | -0.001 |
Token | P |
Token ID | 47 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
living | +0.037 |
mun | +0.030 |
have | +0.029 |
w | +0.029 |
d | +0.028 |
Neg logit contributions | |
horizon | -0.052 |
itself | -0.045 |
peak | -0.045 |
heights | -0.044 |
edly | -0.044 |
Token | ete |
Token ID | 14471 |
Feature activation | -4.24e-02 |
Pos logit contributions | |
living | +0.129 |
w | +0.092 |
have | +0.086 |
mun | +0.084 |
d | +0.083 |
Neg logit contributions | |
horizon | -0.286 |
edly | -0.261 |
heights | -0.256 |
itself | -0.254 |
peak | -0.248 |
Token | went |
Token ID | 1816 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
living | +0.075 |
mun | +0.064 |
have | +0.062 |
w | +0.062 |
haw | +0.057 |
Neg logit contributions | |
horizon | -0.080 |
itself | -0.068 |
peak | -0.067 |
edly | -0.066 |
heights | -0.066 |
Token | to |
Token ID | 284 |
Feature activation | -3.18e-02 |
Pos logit contributions | |
living | +0.040 |
mun | +0.037 |
w | +0.037 |
have | +0.036 |
Did | +0.034 |
Neg logit contributions | |
horizon | -0.040 |
peak | -0.033 |
itself | -0.032 |
heights | -0.032 |
edly | -0.030 |
Token | the |
Token ID | 262 |
Feature activation | -8.09e-03 |
Pos logit contributions | |
living | +0.056 |
mun | +0.047 |
w | +0.047 |
have | +0.045 |
d | +0.043 |
Neg logit contributions | |
horizon | -0.062 |
itself | -0.053 |
peak | -0.053 |
heights | -0.052 |
edly | -0.050 |
Token | store |
Token ID | 3650 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
living | +0.015 |
w | +0.014 |
mun | +0.014 |
have | +0.013 |
haw | +0.013 |
Neg logit contributions | |
horizon | -0.014 |
peak | -0.012 |
heights | -0.012 |
itself | -0.011 |
edly | -0.011 |
Token | never |
Token ID | 1239 |
Feature activation | -2.81e-02 |
Pos logit contributions | |
living | +0.019 |
mun | +0.015 |
have | +0.015 |
w | +0.014 |
d | +0.014 |
Neg logit contributions | |
horizon | -0.023 |
edly | -0.020 |
peak | -0.019 |
heights | -0.019 |
itself | -0.019 |
Token | be |
Token ID | 307 |
Feature activation | -9.79e-03 |
Pos logit contributions | |
living | +0.053 |
w | +0.051 |
mun | +0.048 |
haw | +0.047 |
P | +0.046 |
Neg logit contributions | |
horizon | -0.048 |
heights | -0.041 |
itself | -0.041 |
peak | -0.041 |
side | -0.037 |
Token | peace |
Token ID | 4167 |
Feature activation | -1.29e-02 |
Pos logit contributions | |
living | +0.017 |
mun | +0.014 |
w | +0.014 |
have | +0.013 |
d | +0.013 |
Neg logit contributions | |
horizon | -0.019 |
itself | -0.017 |
heights | -0.017 |
peak | -0.016 |
edly | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +3.28e-05 |
Pos logit contributions | |
living | +0.025 |
mun | +0.020 |
have | +0.020 |
d | +0.019 |
w | +0.019 |
Neg logit contributions | |
horizon | -0.023 |
peak | -0.020 |
heights | -0.020 |
itself | -0.020 |
edly | -0.019 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.38e-02 |
Pos logit contributions | |
horizon | +0.000 |
comb | +0.000 |
heights | +0.000 |
peak | +0.000 |
rily | +0.000 |
Neg logit contributions | |
living | -0.000 |
Good | -0.000 |
sm | -0.000 |
loves | -0.000 |
mun | -0.000 |
Token | P |
Token ID | 47 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
living | +0.042 |
mun | +0.033 |
w | +0.032 |
have | +0.032 |
d | +0.031 |
Neg logit contributions | |
horizon | -0.048 |
heights | -0.041 |
peak | -0.041 |
itself | -0.040 |
edly | -0.040 |
Token | olly |
Token ID | 5098 |
Feature activation | -3.85e-02 |
Pos logit contributions | |
living | +0.114 |
w | +0.076 |
d | +0.068 |
have | +0.065 |
loves | +0.064 |
Neg logit contributions | |
horizon | -0.296 |
edly | -0.282 |
heights | -0.269 |
itself | -0.265 |
comb | -0.262 |
Token | was |
Token ID | 373 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
living | +0.082 |
mun | +0.070 |
w | +0.069 |
have | +0.069 |
d | +0.066 |
Neg logit contributions | |
horizon | -0.061 |
itself | -0.049 |
peak | -0.049 |
heights | -0.049 |
edly | -0.048 |
Token | ready |
Token ID | 3492 |
Feature activation | -2.07e-02 |
Pos logit contributions | |
living | +0.022 |
mun | +0.019 |
w | +0.018 |
have | +0.018 |
d | +0.017 |
Neg logit contributions | |
horizon | -0.030 |
itself | -0.026 |
peak | -0.026 |
edly | -0.025 |
heights | -0.025 |
Token | to |
Token ID | 284 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
living | +0.036 |
mun | +0.031 |
have | +0.029 |
w | +0.029 |
d | +0.027 |
Neg logit contributions | |
horizon | -0.041 |
itself | -0.035 |
peak | -0.034 |
heights | -0.034 |
edly | -0.034 |
Token | plan |
Token ID | 1410 |
Feature activation | +8.59e-04 |
Pos logit contributions | |
living | +0.017 |
have | +0.012 |
mun | +0.012 |
w | +0.012 |
d | +0.012 |
Neg logit contributions | |
horizon | -0.025 |
edly | -0.023 |
heights | -0.022 |
ement | -0.022 |
itself | -0.022 |
Token | fun |
Token ID | 1257 |
Feature activation | +3.72e-02 |
Pos logit contributions | |
horizon | +0.024 |
peak | +0.020 |
heights | +0.020 |
itself | +0.020 |
edly | +0.020 |
Neg logit contributions | |
living | -0.018 |
mun | -0.014 |
have | -0.014 |
w | -0.014 |
d | -0.014 |
Token | playing |
Token ID | 2712 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
horizon | +0.075 |
itself | +0.063 |
peak | +0.063 |
heights | +0.063 |
edly | +0.062 |
Neg logit contributions | |
living | -0.065 |
mun | -0.053 |
w | -0.051 |
have | -0.051 |
d | -0.049 |
Token | together |
Token ID | 1978 |
Feature activation | +2.26e-04 |
Pos logit contributions | |
living | +0.037 |
mun | +0.027 |
have | +0.027 |
w | +0.026 |
d | +0.026 |
Neg logit contributions | |
horizon | -0.059 |
peak | -0.051 |
heights | -0.049 |
comb | -0.049 |
edly | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | +3.64e-03 |
Pos logit contributions | |
horizon | +0.000 |
peak | +0.000 |
heights | +0.000 |
edly | +0.000 |
alks | +0.000 |
Neg logit contributions | |
living | -0.000 |
mun | -0.000 |
d | -0.000 |
have | -0.000 |
w | -0.000 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
horizon | +0.009 |
heights | +0.008 |
edly | +0.008 |
peak | +0.008 |
comb | +0.008 |
Neg logit contributions | |
living | -0.005 |
Good | -0.004 |
mun | -0.004 |
Enjoy | -0.004 |
loves | -0.003 |
Token | P |
Token ID | 47 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
living | +0.033 |
mun | +0.026 |
w | +0.025 |
have | +0.025 |
d | +0.023 |
Neg logit contributions | |
horizon | -0.049 |
itself | -0.042 |
peak | -0.042 |
heights | -0.042 |
edly | -0.042 |
Token | ete |
Token ID | 14471 |
Feature activation | -4.22e-02 |
Pos logit contributions | |
living | +0.100 |
w | +0.061 |
P | +0.059 |
have | +0.054 |
mun | +0.052 |
Neg logit contributions | |
horizon | -0.308 |
edly | -0.295 |
heights | -0.282 |
itself | -0.277 |
comb | -0.271 |
Token | was |
Token ID | 373 |
Feature activation | -1.14e-02 |
Pos logit contributions | |
living | +0.071 |
mun | +0.058 |
have | +0.057 |
w | +0.057 |
d | +0.053 |
Neg logit contributions | |
horizon | -0.085 |
itself | -0.073 |
peak | -0.073 |
heights | -0.072 |
edly | -0.072 |
Token | feeling |
Token ID | 4203 |
Feature activation | -6.65e-04 |
Pos logit contributions | |
living | +0.017 |
mun | +0.015 |
w | +0.015 |
have | +0.014 |
haw | +0.014 |
Neg logit contributions | |
horizon | -0.024 |
peak | -0.021 |
itself | -0.021 |
heights | -0.020 |
edly | -0.020 |
Token | nervous |
Token ID | 10927 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
living | +0.001 |
w | +0.001 |
mun | +0.001 |
have | +0.001 |
haw | +0.001 |
Neg logit contributions | |
horizon | -0.001 |
itself | -0.001 |
peak | -0.001 |
edly | -0.001 |
heights | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -9.58e-03 |
Pos logit contributions | |
living | +0.029 |
mun | +0.024 |
w | +0.024 |
have | +0.024 |
d | +0.023 |
Neg logit contributions | |
horizon | -0.025 |
itself | -0.021 |
peak | -0.021 |
heights | -0.021 |
edly | -0.020 |
Token | a |
Token ID | 257 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
living | +0.041 |
mun | +0.033 |
w | +0.032 |
have | +0.032 |
d | +0.030 |
Neg logit contributions | |
horizon | -0.055 |
itself | -0.047 |
peak | -0.047 |
heights | -0.047 |
edly | -0.046 |
Token | bad |
Token ID | 2089 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
living | +0.017 |
w | +0.017 |
have | +0.016 |
mun | +0.016 |
asc | +0.015 |
Neg logit contributions | |
horizon | -0.023 |
peak | -0.020 |
� | -0.020 |
destination | -0.020 |
heights | -0.019 |
Token | ending |
Token ID | 7464 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
horizon | +0.037 |
peak | +0.031 |
itself | +0.031 |
heights | +0.031 |
edly | +0.031 |
Neg logit contributions | |
living | -0.037 |
mun | -0.031 |
have | -0.030 |
w | -0.030 |
d | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -4.73e-03 |
Pos logit contributions | |
horizon | +0.022 |
peak | +0.019 |
heights | +0.018 |
itself | +0.018 |
edly | +0.018 |
Neg logit contributions | |
living | -0.023 |
mun | -0.019 |
w | -0.018 |
have | -0.018 |
d | -0.018 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.08e-02 |
Pos logit contributions | |
living | +0.008 |
mun | +0.006 |
have | +0.006 |
w | +0.006 |
Good | +0.006 |
Neg logit contributions | |
horizon | -0.010 |
heights | -0.009 |
peak | -0.009 |
comb | -0.008 |
itself | -0.008 |
Token | P |
Token ID | 47 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
living | +0.026 |
have | +0.022 |
mun | +0.021 |
w | +0.021 |
haw | +0.021 |
Neg logit contributions | |
horizon | -0.047 |
itself | -0.042 |
� | -0.041 |
peak | -0.041 |
heights | -0.041 |
Token | oppy |
Token ID | 26696 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
living | +0.129 |
w | +0.094 |
mun | +0.091 |
have | +0.089 |
d | +0.085 |
Neg logit contributions | |
horizon | -0.277 |
edly | -0.251 |
itself | -0.246 |
heights | -0.244 |
peak | -0.241 |
Token | was |
Token ID | 373 |
Feature activation | -9.45e-03 |
Pos logit contributions | |
living | +0.050 |
mun | +0.044 |
have | +0.043 |
w | +0.043 |
haw | +0.040 |
Neg logit contributions | |
horizon | -0.040 |
itself | -0.033 |
edly | -0.033 |
peak | -0.032 |
heights | -0.032 |
Token | a |
Token ID | 257 |
Feature activation | -3.02e-02 |
Pos logit contributions | |
living | +0.014 |
mun | +0.013 |
w | +0.013 |
haw | +0.013 |
have | +0.012 |
Neg logit contributions | |
horizon | -0.019 |
itself | -0.016 |
peak | -0.016 |
edly | -0.016 |
heights | -0.015 |
Token | tall |
Token ID | 7331 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
living | +0.050 |
w | +0.045 |
mun | +0.045 |
have | +0.045 |
haw | +0.042 |
Neg logit contributions | |
horizon | -0.060 |
peak | -0.051 |
heights | -0.048 |
comb | -0.047 |
edly | -0.047 |
Token | girl |
Token ID | 2576 |
Feature activation | +7.70e-03 |
Pos logit contributions | |
horizon | +0.033 |
itself | +0.029 |
edly | +0.028 |
heights | +0.028 |
peak | +0.028 |
Neg logit contributions | |
living | -0.033 |
mun | -0.026 |
d | -0.025 |
w | -0.024 |
have | -0.024 |
Token | a |
Token ID | 257 |
Feature activation | -3.64e-02 |
Pos logit contributions | |
living | +0.006 |
d | +0.004 |
watching | +0.003 |
have | +0.003 |
mun | +0.003 |
Neg logit contributions | |
horizon | -0.010 |
comb | -0.010 |
edly | -0.010 |
ement | -0.009 |
rily | -0.009 |
Token | new |
Token ID | 649 |
Feature activation | +2.26e-03 |
Pos logit contributions | |
living | +0.062 |
mun | +0.047 |
d | +0.047 |
have | +0.046 |
w | +0.045 |
Neg logit contributions | |
horizon | -0.073 |
edly | -0.065 |
itself | -0.064 |
heights | -0.063 |
ement | -0.063 |
Token | adventure |
Token ID | 8855 |
Feature activation | +8.00e-03 |
Pos logit contributions | |
horizon | +0.004 |
edly | +0.003 |
ement | +0.003 |
� | +0.003 |
comb | +0.003 |
Neg logit contributions | |
living | -0.005 |
d | -0.004 |
have | -0.004 |
mun | -0.004 |
w | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +2.82e-03 |
Pos logit contributions | |
horizon | +0.017 |
� | +0.015 |
edly | +0.015 |
heights | +0.015 |
comb | +0.015 |
Neg logit contributions | |
living | -0.013 |
d | -0.009 |
mun | -0.008 |
watching | -0.008 |
sm | -0.008 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
horizon | +0.006 |
heights | +0.005 |
edly | +0.005 |
comb | +0.005 |
ement | +0.005 |
Neg logit contributions | |
living | -0.005 |
mun | -0.004 |
Good | -0.004 |
have | -0.004 |
d | -0.004 |
Token | P |
Token ID | 47 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
living | +0.035 |
mun | +0.029 |
have | +0.028 |
w | +0.028 |
d | +0.027 |
Neg logit contributions | |
horizon | -0.044 |
itself | -0.038 |
peak | -0.038 |
heights | -0.037 |
edly | -0.037 |
Token | olly |
Token ID | 5098 |
Feature activation | -3.51e-02 |
Pos logit contributions | |
living | +0.118 |
w | +0.083 |
have | +0.076 |
P | +0.073 |
mun | +0.073 |
Neg logit contributions | |
horizon | -0.284 |
edly | -0.265 |
heights | -0.257 |
itself | -0.253 |
comb | -0.248 |
Token | had |
Token ID | 550 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
living | +0.070 |
mun | +0.060 |
have | +0.059 |
w | +0.059 |
d | +0.055 |
Neg logit contributions | |
horizon | -0.060 |
itself | -0.050 |
peak | -0.049 |
edly | -0.049 |
heights | -0.049 |
Token | been |
Token ID | 587 |
Feature activation | -2.52e-02 |
Pos logit contributions | |
living | +0.048 |
mun | +0.040 |
w | +0.038 |
have | +0.038 |
d | +0.036 |
Neg logit contributions | |
horizon | -0.064 |
itself | -0.056 |
peak | -0.055 |
heights | -0.055 |
edly | -0.054 |
Token | playing |
Token ID | 2712 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
living | +0.039 |
w | +0.038 |
mun | +0.038 |
haw | +0.037 |
have | +0.036 |
Neg logit contributions | |
horizon | -0.047 |
itself | -0.040 |
peak | -0.040 |
heights | -0.039 |
edly | -0.038 |
Token | in |
Token ID | 287 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
living | +0.033 |
mun | +0.027 |
w | +0.027 |
have | +0.027 |
haw | +0.025 |
Neg logit contributions | |
horizon | -0.038 |
itself | -0.033 |
peak | -0.033 |
heights | -0.033 |
edly | -0.032 |
Token | else |
Token ID | 2073 |
Feature activation | -7.24e-03 |
Pos logit contributions | |
horizon | +0.018 |
itself | +0.016 |
peak | +0.016 |
heights | +0.016 |
edly | +0.016 |
Neg logit contributions | |
living | -0.013 |
mun | -0.010 |
w | -0.010 |
have | -0.010 |
haw | -0.009 |
Token | to |
Token ID | 284 |
Feature activation | +3.88e-03 |
Pos logit contributions | |
living | +0.013 |
mun | +0.011 |
w | +0.011 |
have | +0.011 |
d | +0.010 |
Neg logit contributions | |
horizon | -0.014 |
itself | -0.012 |
peak | -0.012 |
heights | -0.012 |
edly | -0.012 |
Token | do |
Token ID | 466 |
Feature activation | +3.66e-02 |
Pos logit contributions | |
horizon | +0.009 |
edly | +0.008 |
heights | +0.008 |
peak | +0.008 |
ement | +0.008 |
Neg logit contributions | |
living | -0.005 |
have | -0.004 |
d | -0.004 |
mun | -0.004 |
w | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +1.29e-03 |
Pos logit contributions | |
horizon | +0.071 |
peak | +0.059 |
heights | +0.059 |
edly | +0.058 |
comb | +0.057 |
Neg logit contributions | |
living | -0.068 |
have | -0.055 |
mun | -0.054 |
w | -0.054 |
d | -0.052 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
horizon | +0.003 |
heights | +0.002 |
peak | +0.002 |
comb | +0.002 |
edly | +0.002 |
Neg logit contributions | |
living | -0.002 |
mun | -0.002 |
Good | -0.002 |
sm | -0.002 |
d | -0.002 |
Token | P |
Token ID | 47 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
living | +0.033 |
mun | +0.026 |
w | +0.026 |
have | +0.026 |
d | +0.024 |
Neg logit contributions | |
horizon | -0.043 |
itself | -0.037 |
peak | -0.037 |
heights | -0.036 |
edly | -0.036 |
Token | olly |
Token ID | 5098 |
Feature activation | -3.47e-02 |
Pos logit contributions | |
living | +0.106 |
w | +0.067 |
P | +0.063 |
have | +0.062 |
prise | +0.062 |
Neg logit contributions | |
horizon | -0.296 |
edly | -0.285 |
heights | -0.271 |
itself | -0.268 |
comb | -0.263 |
Token | was |
Token ID | 373 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
living | +0.072 |
mun | +0.060 |
w | +0.059 |
have | +0.059 |
d | +0.057 |
Neg logit contributions | |
horizon | -0.058 |
itself | -0.048 |
peak | -0.047 |
heights | -0.047 |
edly | -0.047 |
Token | playing |
Token ID | 2712 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
living | +0.015 |
mun | +0.013 |
w | +0.013 |
have | +0.012 |
haw | +0.012 |
Neg logit contributions | |
horizon | -0.022 |
itself | -0.019 |
peak | -0.019 |
edly | -0.018 |
heights | -0.018 |
Token | at |
Token ID | 379 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
living | +0.044 |
w | +0.038 |
mun | +0.038 |
haw | +0.037 |
have | +0.037 |
Neg logit contributions | |
horizon | -0.051 |
itself | -0.047 |
heights | -0.045 |
peak | -0.045 |
edly | -0.045 |
Token | the |
Token ID | 262 |
Feature activation | -1.74e-04 |
Pos logit contributions | |
horizon | +0.040 |
edly | +0.036 |
ement | +0.036 |
comb | +0.035 |
peak | +0.035 |
Neg logit contributions | |
living | -0.031 |
mun | -0.022 |
have | -0.021 |
w | -0.021 |
d | -0.021 |
Token | best |
Token ID | 1266 |
Feature activation | -3.37e-03 |
Pos logit contributions | |
living | +0.014 |
mun | +0.011 |
have | +0.010 |
w | +0.010 |
d | +0.010 |
Neg logit contributions | |
horizon | -0.022 |
itself | -0.020 |
edly | -0.019 |
peak | -0.019 |
heights | -0.019 |
Token | of |
Token ID | 286 |
Feature activation | +2.68e-02 |
Pos logit contributions | |
living | +0.006 |
mun | +0.005 |
d | +0.005 |
sm | +0.004 |
have | +0.004 |
Neg logit contributions | |
horizon | -0.007 |
peak | -0.006 |
edly | -0.006 |
heights | -0.006 |
itself | -0.006 |
Token | friends |
Token ID | 2460 |
Feature activation | -5.07e-03 |
Pos logit contributions | |
horizon | +0.058 |
itself | +0.051 |
heights | +0.050 |
peak | +0.050 |
edly | +0.050 |
Neg logit contributions | |
living | -0.041 |
mun | -0.033 |
w | -0.032 |
have | -0.032 |
d | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | -4.41e-03 |
Pos logit contributions | |
living | +0.009 |
mun | +0.007 |
d | +0.007 |
w | +0.007 |
sm | +0.007 |
Neg logit contributions | |
horizon | -0.010 |
peak | -0.009 |
heights | -0.009 |
itself | -0.008 |
edly | -0.008 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
living | +0.007 |
mun | +0.005 |
Good | +0.005 |
sm | +0.005 |
d | +0.005 |
Neg logit contributions | |
horizon | -0.010 |
peak | -0.009 |
heights | -0.009 |
comb | -0.008 |
edly | -0.008 |
Token | P |
Token ID | 47 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
living | +0.035 |
mun | +0.027 |
w | +0.026 |
have | +0.026 |
d | +0.025 |
Neg logit contributions | |
horizon | -0.053 |
peak | -0.046 |
itself | -0.046 |
heights | -0.046 |
edly | -0.046 |
Token | ete |
Token ID | 14471 |
Feature activation | -3.92e-02 |
Pos logit contributions | |
living | +0.124 |
w | +0.088 |
mun | +0.083 |
have | +0.082 |
d | +0.078 |
Neg logit contributions | |
horizon | -0.278 |
edly | -0.255 |
heights | -0.248 |
itself | -0.247 |
peak | -0.241 |
Token | was |
Token ID | 373 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
living | +0.077 |
mun | +0.067 |
have | +0.066 |
w | +0.065 |
sm | +0.061 |
Neg logit contributions | |
horizon | -0.066 |
itself | -0.055 |
peak | -0.054 |
edly | -0.053 |
heights | -0.053 |
Token | walking |
Token ID | 6155 |
Feature activation | -3.30e-02 |
Pos logit contributions | |
living | +0.015 |
mun | +0.013 |
w | +0.013 |
have | +0.012 |
haw | +0.012 |
Neg logit contributions | |
horizon | -0.022 |
itself | -0.019 |
peak | -0.019 |
heights | -0.019 |
edly | -0.019 |
Token | in |
Token ID | 287 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
living | +0.060 |
mun | +0.049 |
have | +0.048 |
w | +0.048 |
d | +0.046 |
Neg logit contributions | |
horizon | -0.064 |
heights | -0.054 |
itself | -0.054 |
peak | -0.054 |
edly | -0.053 |
Token | the |
Token ID | 262 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
living | +0.039 |
d | +0.025 |
mun | +0.024 |
watching | +0.024 |
have | +0.023 |
Neg logit contributions | |
horizon | -0.057 |
edly | -0.054 |
comb | -0.053 |
ement | -0.051 |
iously | -0.050 |
Token | is |
Token ID | 318 |
Feature activation | +5.98e-03 |
Pos logit contributions | |
living | +0.064 |
mun | +0.051 |
have | +0.050 |
w | +0.049 |
sm | +0.046 |
Neg logit contributions | |
horizon | -0.118 |
itself | -0.108 |
heights | -0.104 |
peak | -0.101 |
edly | -0.101 |
Token | good |
Token ID | 922 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
horizon | +0.012 |
itself | +0.011 |
heights | +0.011 |
peak | +0.011 |
edly | +0.010 |
Neg logit contributions | |
living | -0.009 |
mun | -0.008 |
w | -0.008 |
have | -0.007 |
haw | -0.007 |
Token | ." |
Token ID | 526 |
Feature activation | +6.00e-03 |
Pos logit contributions | |
living | +0.022 |
w | +0.017 |
have | +0.017 |
mun | +0.017 |
d | +0.016 |
Neg logit contributions | |
horizon | -0.026 |
peak | -0.022 |
heights | -0.021 |
itself | -0.021 |
ement | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
horizon | +0.012 |
peak | +0.011 |
comb | +0.010 |
itself | +0.010 |
heights | +0.010 |
Neg logit contributions | |
living | -0.011 |
mun | -0.009 |
w | -0.008 |
d | -0.008 |
have | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -2.06e-03 |
Pos logit contributions | |
living | +0.019 |
mun | +0.016 |
w | +0.015 |
have | +0.015 |
d | +0.015 |
Neg logit contributions | |
horizon | -0.021 |
peak | -0.018 |
heights | -0.018 |
itself | -0.018 |
edly | -0.017 |
Token | P |
Token ID | 47 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
living | +0.003 |
mun | +0.002 |
have | +0.002 |
d | +0.002 |
w | +0.002 |
Neg logit contributions | |
horizon | -0.005 |
itself | -0.004 |
heights | -0.004 |
edly | -0.004 |
peak | -0.004 |
Token | ao |
Token ID | 5488 |
Feature activation | -8.39e-03 |
Pos logit contributions | |
living | +0.199 |
mun | +0.164 |
have | +0.159 |
w | +0.159 |
d | +0.152 |
Neg logit contributions | |
horizon | -0.198 |
itself | -0.168 |
peak | -0.168 |
heights | -0.167 |
edly | -0.165 |
Token | was |
Token ID | 373 |
Feature activation | -5.76e-03 |
Pos logit contributions | |
living | +0.012 |
mun | +0.009 |
have | +0.009 |
w | +0.009 |
Good | +0.008 |
Neg logit contributions | |
horizon | -0.019 |
edly | -0.017 |
itself | -0.017 |
peak | -0.016 |
heights | -0.016 |
Token | surprised |
Token ID | 6655 |
Feature activation | -7.51e-03 |
Pos logit contributions | |
living | +0.009 |
have | +0.007 |
mun | +0.007 |
d | +0.007 |
w | +0.006 |
Neg logit contributions | |
horizon | -0.013 |
peak | -0.011 |
heights | -0.011 |
comb | -0.011 |
edly | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -1.72e-03 |
Pos logit contributions | |
living | +0.014 |
mun | +0.010 |
d | +0.010 |
have | +0.010 |
w | +0.010 |
Neg logit contributions | |
horizon | -0.015 |
peak | -0.013 |
comb | -0.013 |
edly | -0.012 |
heights | -0.012 |
Token | He |
Token ID | 679 |
Feature activation | -6.15e-03 |
Pos logit contributions | |
living | +0.003 |
Good | +0.002 |
d | +0.002 |
L | +0.002 |
Enjoy | +0.002 |
Neg logit contributions | |
horizon | -0.003 |
peak | -0.003 |
comb | -0.003 |
destination | -0.003 |
ement | -0.003 |
Token | animals |
Token ID | 4695 |
Feature activation | +6.26e-03 |
Pos logit contributions | |
horizon | +0.019 |
itself | +0.016 |
peak | +0.016 |
heights | +0.016 |
edly | +0.015 |
Neg logit contributions | |
living | -0.019 |
mun | -0.016 |
w | -0.015 |
have | -0.015 |
d | -0.014 |
Token | brought |
Token ID | 3181 |
Feature activation | -2.90e-03 |
Pos logit contributions | |
horizon | +0.013 |
itself | +0.011 |
peak | +0.011 |
heights | +0.011 |
edly | +0.011 |
Neg logit contributions | |
living | -0.011 |
mun | -0.009 |
w | -0.008 |
have | -0.008 |
d | -0.008 |
Token | him |
Token ID | 683 |
Feature activation | +3.35e-02 |
Pos logit contributions | |
living | +0.005 |
mun | +0.004 |
have | +0.004 |
w | +0.004 |
d | +0.004 |
Neg logit contributions | |
horizon | -0.006 |
peak | -0.005 |
edly | -0.005 |
comb | -0.005 |
heights | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +3.47e-03 |
Pos logit contributions | |
horizon | +0.066 |
peak | +0.056 |
itself | +0.055 |
heights | +0.055 |
edly | +0.055 |
Neg logit contributions | |
living | -0.060 |
mun | -0.048 |
have | -0.047 |
w | -0.047 |
d | -0.045 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.37e-02 |
Pos logit contributions | |
horizon | +0.007 |
heights | +0.006 |
edly | +0.006 |
peak | +0.006 |
comb | +0.006 |
Neg logit contributions | |
living | -0.006 |
mun | -0.005 |
Good | -0.004 |
have | -0.004 |
w | -0.004 |
Token | P |
Token ID | 47 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
living | +0.035 |
mun | +0.028 |
have | +0.027 |
w | +0.027 |
d | +0.026 |
Neg logit contributions | |
horizon | -0.053 |
itself | -0.046 |
peak | -0.046 |
heights | -0.045 |
edly | -0.045 |
Token | olly |
Token ID | 5098 |
Feature activation | -3.59e-02 |
Pos logit contributions | |
living | +0.113 |
w | +0.078 |
have | +0.071 |
mun | +0.070 |
d | +0.067 |
Neg logit contributions | |
horizon | -0.285 |
edly | -0.267 |
heights | -0.259 |
itself | -0.257 |
comb | -0.249 |
Token | was |
Token ID | 373 |
Feature activation | -8.39e-03 |
Pos logit contributions | |
living | +0.073 |
mun | +0.062 |
have | +0.061 |
w | +0.061 |
d | +0.058 |
Neg logit contributions | |
horizon | -0.060 |
itself | -0.049 |
peak | -0.049 |
heights | -0.049 |
edly | -0.048 |
Token | a |
Token ID | 257 |
Feature activation | -2.85e-02 |
Pos logit contributions | |
living | +0.013 |
mun | +0.011 |
w | +0.011 |
have | +0.011 |
haw | +0.010 |
Neg logit contributions | |
horizon | -0.017 |
itself | -0.015 |
peak | -0.015 |
edly | -0.015 |
heights | -0.015 |
Token | little |
Token ID | 1310 |
Feature activation | +4.42e-03 |
Pos logit contributions | |
living | +0.046 |
w | +0.039 |
mun | +0.038 |
have | +0.038 |
haw | +0.035 |
Neg logit contributions | |
horizon | -0.060 |
peak | -0.051 |
heights | -0.050 |
itself | -0.049 |
edly | -0.049 |
Token | cat |
Token ID | 3797 |
Feature activation | +6.25e-03 |
Pos logit contributions | |
horizon | +0.010 |
peak | +0.009 |
itself | +0.009 |
heights | +0.009 |
edly | +0.009 |
Neg logit contributions | |
living | -0.006 |
mun | -0.005 |
w | -0.005 |
have | -0.005 |
d | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | -9.34e-03 |
Pos logit contributions | |
living | +0.008 |
d | +0.005 |
sm | +0.005 |
have | +0.005 |
watching | +0.005 |
Neg logit contributions | |
horizon | -0.011 |
edly | -0.010 |
comb | -0.010 |
peak | -0.009 |
heights | -0.009 |
Token | deep |
Token ID | 2769 |
Feature activation | +6.78e-03 |
Pos logit contributions | |
w | +0.017 |
mun | +0.017 |
living | +0.016 |
have | +0.016 |
haw | +0.015 |
Neg logit contributions | |
horizon | -0.017 |
peak | -0.014 |
ement | -0.013 |
side | -0.013 |
destination | -0.013 |
Token | garage |
Token ID | 15591 |
Feature activation | -2.07e-02 |
Pos logit contributions | |
horizon | +0.012 |
comb | +0.011 |
edly | +0.011 |
itself | +0.011 |
� | +0.010 |
Neg logit contributions | |
living | -0.013 |
d | -0.010 |
have | -0.009 |
mun | -0.009 |
stri | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +5.72e-03 |
Pos logit contributions | |
living | +0.037 |
d | +0.027 |
mun | +0.027 |
have | +0.026 |
stole | +0.025 |
Neg logit contributions | |
horizon | -0.042 |
heights | -0.036 |
itself | -0.036 |
peak | -0.035 |
edly | -0.035 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.24e-02 |
Pos logit contributions | |
horizon | +0.011 |
heights | +0.010 |
peak | +0.010 |
comb | +0.010 |
itself | +0.010 |
Neg logit contributions | |
living | -0.010 |
mun | -0.008 |
have | -0.008 |
Good | -0.008 |
d | -0.008 |
Token | P |
Token ID | 47 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
living | +0.038 |
mun | +0.031 |
w | +0.031 |
have | +0.031 |
d | +0.029 |
Neg logit contributions | |
horizon | -0.045 |
itself | -0.038 |
peak | -0.038 |
heights | -0.038 |
edly | -0.038 |
Token | olly |
Token ID | 5098 |
Feature activation | -3.46e-02 |
Pos logit contributions | |
living | +0.125 |
w | +0.088 |
have | +0.085 |
mun | +0.083 |
d | +0.080 |
Neg logit contributions | |
horizon | -0.274 |
edly | -0.250 |
heights | -0.246 |
itself | -0.245 |
comb | -0.240 |
Token | and |
Token ID | 290 |
Feature activation | +5.18e-04 |
Pos logit contributions | |
living | +0.058 |
mun | +0.048 |
have | +0.048 |
w | +0.047 |
d | +0.043 |
Neg logit contributions | |
horizon | -0.069 |
itself | -0.059 |
peak | -0.059 |
edly | -0.058 |
heights | -0.058 |
Token | Jason |
Token ID | 8982 |
Feature activation | -2.65e-02 |
Pos logit contributions | |
horizon | +0.001 |
itself | +0.001 |
peak | +0.001 |
heights | +0.001 |
edly | +0.001 |
Neg logit contributions | |
living | -0.001 |
mun | -0.001 |
w | -0.001 |
have | -0.001 |
d | -0.001 |
Token | were |
Token ID | 547 |
Feature activation | +2.45e-03 |
Pos logit contributions | |
living | +0.038 |
w | +0.031 |
mun | +0.031 |
have | +0.031 |
haw | +0.027 |
Neg logit contributions | |
horizon | -0.058 |
peak | -0.052 |
itself | -0.051 |
comb | -0.050 |
heights | -0.049 |
Token | best |
Token ID | 1266 |
Feature activation | -3.24e-02 |
Pos logit contributions | |
horizon | +0.006 |
ement | +0.005 |
peak | +0.005 |
itself | +0.005 |
heights | +0.005 |
Neg logit contributions | |
living | -0.003 |
w | -0.003 |
mun | -0.003 |
haw | -0.002 |
have | -0.002 |
Token | good |
Token ID | 922 |
Feature activation | -1.38e-03 |
Pos logit contributions | |
living | +0.054 |
mun | +0.047 |
w | +0.046 |
have | +0.044 |
haw | +0.038 |
Neg logit contributions | |
horizon | -0.084 |
peak | -0.074 |
itself | -0.073 |
heights | -0.072 |
comb | -0.071 |
Token | friend |
Token ID | 1545 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
living | +0.002 |
mun | +0.002 |
have | +0.002 |
w | +0.002 |
Did | +0.002 |
Neg logit contributions | |
horizon | -0.003 |
peak | -0.002 |
comb | -0.002 |
heights | -0.002 |
edly | -0.002 |
Token | ." |
Token ID | 526 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
living | +0.020 |
w | +0.015 |
mun | +0.015 |
d | +0.014 |
have | +0.014 |
Neg logit contributions | |
horizon | -0.029 |
peak | -0.025 |
heights | -0.024 |
fingertips | -0.024 |
alks | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | -6.48e-03 |
Pos logit contributions | |
horizon | +0.021 |
itself | +0.018 |
peak | +0.018 |
heights | +0.018 |
edly | +0.018 |
Neg logit contributions | |
living | -0.022 |
mun | -0.018 |
w | -0.017 |
have | -0.017 |
d | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | +3.09e-04 |
Pos logit contributions | |
living | +0.012 |
mun | +0.010 |
w | +0.009 |
have | +0.009 |
d | +0.009 |
Neg logit contributions | |
horizon | -0.012 |
itself | -0.011 |
peak | -0.010 |
heights | -0.010 |
edly | -0.010 |
Token | P |
Token ID | 47 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
horizon | +0.001 |
itself | +0.001 |
edly | +0.001 |
rily | +0.001 |
� | +0.001 |
Neg logit contributions | |
living | -0.000 |
w | -0.000 |
mun | -0.000 |
d | -0.000 |
haw | -0.000 |
Token | inky |
Token ID | 29246 |
Feature activation | -3.44e-02 |
Pos logit contributions | |
living | +0.226 |
mun | +0.192 |
w | +0.189 |
have | +0.188 |
d | +0.181 |
Neg logit contributions | |
horizon | -0.169 |
itself | -0.137 |
peak | -0.137 |
heights | -0.136 |
edly | -0.134 |
Token | smiled |
Token ID | 13541 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
living | +0.063 |
mun | +0.052 |
w | +0.051 |
have | +0.051 |
d | +0.049 |
Neg logit contributions | |
horizon | -0.065 |
itself | -0.055 |
peak | -0.054 |
heights | -0.054 |
edly | -0.054 |
Token | and |
Token ID | 290 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
living | +0.030 |
d | +0.025 |
loved | +0.022 |
sm | +0.022 |
mer | +0.022 |
Neg logit contributions | |
horizon | -0.035 |
heights | -0.031 |
peak | -0.030 |
� | -0.029 |
fingertips | -0.029 |
Token | said |
Token ID | 531 |
Feature activation | +4.18e-03 |
Pos logit contributions | |
living | +0.021 |
mun | +0.016 |
w | +0.015 |
have | +0.015 |
d | +0.015 |
Neg logit contributions | |
horizon | -0.033 |
peak | -0.029 |
heights | -0.029 |
itself | -0.029 |
edly | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | -1.79e-04 |
Pos logit contributions | |
horizon | +0.005 |
itself | +0.005 |
edly | +0.005 |
peak | +0.004 |
ement | +0.004 |
Neg logit contributions | |
living | -0.009 |
w | -0.008 |
mun | -0.008 |
have | -0.008 |
Did | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | -3.24e-02 |
Pos logit contributions | |
horizon | +0.024 |
ement | +0.021 |
comb | +0.021 |
edly | +0.021 |
peak | +0.020 |
Neg logit contributions | |
living | -0.014 |
d | -0.009 |
sm | -0.008 |
watching | -0.008 |
w | -0.008 |
Token | bush |
Token ID | 24484 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
living | +0.070 |
d | +0.052 |
sm | +0.049 |
mun | +0.048 |
ye | +0.048 |
Neg logit contributions | |
horizon | -0.055 |
edly | -0.052 |
ement | -0.050 |
heights | -0.050 |
iously | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | -3.37e-03 |
Pos logit contributions | |
horizon | +0.044 |
heights | +0.040 |
ement | +0.039 |
fingertips | +0.039 |
edly | +0.039 |
Neg logit contributions | |
living | -0.028 |
d | -0.017 |
sm | -0.016 |
stole | -0.015 |
mun | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | -4.51e-03 |
Pos logit contributions | |
living | +0.007 |
mun | +0.006 |
w | +0.005 |
have | +0.005 |
d | +0.005 |
Neg logit contributions | |
horizon | -0.006 |
peak | -0.005 |
heights | -0.005 |
itself | -0.005 |
ement | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +2.89e-03 |
Pos logit contributions | |
living | +0.008 |
mun | +0.006 |
w | +0.006 |
have | +0.006 |
d | +0.006 |
Neg logit contributions | |
horizon | -0.009 |
heights | -0.008 |
peak | -0.008 |
edly | -0.008 |
itself | -0.008 |
Token | P |
Token ID | 47 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
horizon | +0.005 |
itself | +0.005 |
edly | +0.005 |
comb | +0.005 |
peak | +0.005 |
Neg logit contributions | |
living | -0.005 |
mun | -0.004 |
actors | -0.004 |
have | -0.004 |
icated | -0.004 |
Token | oppy |
Token ID | 26696 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
living | +0.194 |
mun | +0.158 |
have | +0.153 |
w | +0.151 |
d | +0.146 |
Neg logit contributions | |
horizon | -0.194 |
peak | -0.169 |
itself | -0.168 |
heights | -0.168 |
comb | -0.164 |
Token | cries |
Token ID | 24691 |
Feature activation | -8.81e-03 |
Pos logit contributions | |
living | +0.029 |
w | +0.023 |
mun | +0.023 |
have | +0.023 |
sm | +0.020 |
Neg logit contributions | |
horizon | -0.051 |
heights | -0.046 |
itself | -0.045 |
peak | -0.045 |
edly | -0.044 |
Token | . |
Token ID | 13 |
Feature activation | -5.24e-03 |
Pos logit contributions | |
living | +0.019 |
mun | +0.016 |
w | +0.015 |
have | +0.015 |
d | +0.015 |
Neg logit contributions | |
horizon | -0.015 |
peak | -0.012 |
heights | -0.012 |
itself | -0.012 |
edly | -0.012 |
Token | She |
Token ID | 1375 |
Feature activation | +3.37e-03 |
Pos logit contributions | |
living | +0.010 |
mun | +0.008 |
d | +0.008 |
w | +0.008 |
have | +0.008 |
Neg logit contributions | |
horizon | -0.010 |
peak | -0.008 |
ement | -0.008 |
heights | -0.008 |
itself | -0.008 |
Token | has |
Token ID | 468 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
horizon | +0.006 |
heights | +0.006 |
peak | +0.006 |
itself | +0.005 |
branches | +0.005 |
Neg logit contributions | |
living | -0.005 |
mun | -0.005 |
have | -0.004 |
w | -0.004 |
sm | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | -8.11e-03 |
Pos logit contributions | |
living | +0.016 |
mun | +0.013 |
w | +0.013 |
have | +0.013 |
d | +0.013 |
Neg logit contributions | |
horizon | -0.015 |
itself | -0.012 |
peak | -0.012 |
heights | -0.012 |
edly | -0.012 |
Token | she |
Token ID | 673 |
Feature activation | -8.64e-03 |
Pos logit contributions | |
living | +0.015 |
mun | +0.012 |
w | +0.012 |
have | +0.012 |
d | +0.011 |
Neg logit contributions | |
horizon | -0.015 |
itself | -0.013 |
peak | -0.013 |
heights | -0.013 |
edly | -0.013 |
Token | gave |
Token ID | 2921 |
Feature activation | -4.47e-04 |
Pos logit contributions | |
living | +0.013 |
w | +0.011 |
mun | +0.011 |
have | +0.011 |
haw | +0.010 |
Neg logit contributions | |
horizon | -0.018 |
peak | -0.016 |
comb | -0.015 |
itself | -0.015 |
edly | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | +3.29e-02 |
Pos logit contributions | |
living | +0.001 |
w | +0.001 |
mun | +0.001 |
have | +0.001 |
d | +0.001 |
Neg logit contributions | |
horizon | -0.001 |
edly | -0.001 |
peak | -0.001 |
comb | -0.001 |
heights | -0.001 |
Token | arm |
Token ID | 3211 |
Feature activation | +5.04e-02 |
Pos logit contributions | |
horizon | +0.059 |
edly | +0.051 |
itself | +0.051 |
heights | +0.049 |
rily | +0.049 |
Neg logit contributions | |
living | -0.067 |
w | -0.052 |
d | -0.052 |
have | -0.051 |
mun | -0.050 |
Token | chair |
Token ID | 16337 |
Feature activation | +6.87e-02 |
Pos logit contributions | |
horizon | +0.078 |
heights | +0.064 |
itself | +0.061 |
peak | +0.061 |
ement | +0.061 |
Neg logit contributions | |
living | -0.114 |
w | -0.093 |
d | -0.092 |
mun | -0.092 |
sm | -0.091 |
Token | a |
Token ID | 257 |
Feature activation | +8.41e-03 |
Pos logit contributions | |
horizon | +0.163 |
heights | +0.142 |
peak | +0.140 |
itself | +0.139 |
edly | +0.137 |
Neg logit contributions | |
living | -0.099 |
d | -0.072 |
have | -0.070 |
mun | -0.070 |
sm | -0.069 |
Token | final |
Token ID | 2457 |
Feature activation | +3.45e-02 |
Pos logit contributions | |
horizon | +0.014 |
peak | +0.013 |
comb | +0.012 |
delight | +0.012 |
� | +0.012 |
Neg logit contributions | |
living | -0.015 |
mun | -0.015 |
have | -0.014 |
w | -0.014 |
Did | -0.013 |
Token | squeeze |
Token ID | 21229 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
horizon | +0.060 |
peak | +0.054 |
itself | +0.052 |
comb | +0.052 |
heights | +0.050 |
Neg logit contributions | |
living | -0.064 |
mun | -0.059 |
have | -0.053 |
w | -0.053 |
Did | -0.052 |
Token | and |
Token ID | 290 |
Feature activation | +6.19e-03 |
Pos logit contributions | |
horizon | +0.028 |
heights | +0.024 |
peak | +0.024 |
alks | +0.023 |
ement | +0.023 |
Neg logit contributions | |
living | -0.024 |
w | -0.018 |
d | -0.018 |
mun | -0.017 |
sm | -0.017 |
Token | smiled |
Token ID | 13541 |
Feature activation | -6.07e-03 |
Pos logit contributions | |
horizon | +0.013 |
itself | +0.011 |
peak | +0.011 |
edly | +0.011 |
heights | +0.011 |
Neg logit contributions | |
living | -0.010 |
mun | -0.008 |
have | -0.008 |
w | -0.008 |
sm | -0.007 |
Token | it |
Token ID | 340 |
Feature activation | +2.27e-02 |
Pos logit contributions | |
living | +0.014 |
mun | +0.011 |
have | +0.011 |
w | +0.011 |
d | +0.010 |
Neg logit contributions | |
horizon | -0.018 |
peak | -0.015 |
edly | -0.015 |
heights | -0.015 |
itself | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
horizon | +0.047 |
peak | +0.039 |
heights | +0.038 |
edly | +0.038 |
ement | +0.038 |
Neg logit contributions | |
living | -0.041 |
have | -0.031 |
d | -0.031 |
mun | -0.031 |
w | -0.030 |
Token | |
Token ID | 220 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
horizon | +0.022 |
heights | +0.019 |
peak | +0.019 |
comb | +0.019 |
edly | +0.018 |
Neg logit contributions | |
living | -0.018 |
mun | -0.014 |
Good | -0.013 |
L | -0.013 |
have | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | -2.14e-03 |
Pos logit contributions | |
living | +0.032 |
mun | +0.022 |
have | +0.022 |
w | +0.021 |
L | +0.020 |
Neg logit contributions | |
horizon | -0.053 |
heights | -0.046 |
peak | -0.046 |
edly | -0.045 |
comb | -0.044 |
Token |
|
Token ID | 198 |
Feature activation | +5.03e-03 |
Pos logit contributions | |
living | +0.004 |
mun | +0.003 |
have | +0.003 |
w | +0.003 |
sm | +0.003 |
Neg logit contributions | |
horizon | -0.005 |
heights | -0.004 |
peak | -0.004 |
edly | -0.004 |
comb | -0.004 |
Token | When |
Token ID | 2215 |
Feature activation | +5.57e-02 |
Pos logit contributions | |
horizon | +0.008 |
itself | +0.007 |
edly | +0.007 |
peak | +0.007 |
heights | +0.007 |
Neg logit contributions | |
living | -0.010 |
mun | -0.008 |
w | -0.008 |
d | -0.008 |
haw | -0.008 |
Token | she |
Token ID | 673 |
Feature activation | +8.29e-03 |
Pos logit contributions | |
horizon | +0.109 |
itself | +0.092 |
peak | +0.092 |
heights | +0.091 |
edly | +0.090 |
Neg logit contributions | |
living | -0.101 |
mun | -0.083 |
w | -0.081 |
have | -0.081 |
d | -0.077 |
Token | started |
Token ID | 2067 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
horizon | +0.019 |
peak | +0.017 |
itself | +0.017 |
heights | +0.017 |
edly | +0.017 |
Neg logit contributions | |
living | -0.012 |
mun | -0.009 |
w | -0.009 |
have | -0.009 |
d | -0.008 |
Token | her |
Token ID | 607 |
Feature activation | +3.02e-03 |
Pos logit contributions | |
horizon | +0.030 |
edly | +0.026 |
peak | +0.025 |
itself | +0.025 |
ement | +0.025 |
Neg logit contributions | |
living | -0.020 |
mun | -0.015 |
d | -0.014 |
have | -0.014 |
w | -0.013 |
Token | search |
Token ID | 2989 |
Feature activation | +2.57e-02 |
Pos logit contributions | |
horizon | +0.007 |
itself | +0.006 |
edly | +0.006 |
heights | +0.006 |
comb | +0.006 |
Neg logit contributions | |
living | -0.005 |
d | -0.003 |
mun | -0.003 |
have | -0.003 |
loves | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | +4.79e-03 |
Pos logit contributions | |
horizon | +0.053 |
edly | +0.046 |
heights | +0.046 |
peak | +0.045 |
comb | +0.045 |
Neg logit contributions | |
living | -0.045 |
mun | -0.034 |
d | -0.034 |
w | -0.034 |
have | -0.032 |
Token | He |
Token ID | 679 |
Feature activation | -6.55e-04 |
Pos logit contributions | |
horizon | +0.015 |
comb | +0.013 |
edly | +0.013 |
heights | +0.013 |
peak | +0.013 |
Neg logit contributions | |
living | -0.018 |
mun | -0.015 |
have | -0.014 |
d | -0.014 |
Good | -0.014 |
Token | watched |
Token ID | 7342 |
Feature activation | -2.39e-03 |
Pos logit contributions | |
living | +0.001 |
w | +0.001 |
mun | +0.001 |
have | +0.001 |
sm | +0.001 |
Neg logit contributions | |
horizon | -0.001 |
peak | -0.001 |
itself | -0.001 |
edly | -0.001 |
heights | -0.001 |
Token | them |
Token ID | 606 |
Feature activation | +2.88e-03 |
Pos logit contributions | |
living | +0.005 |
mun | +0.004 |
have | +0.004 |
w | +0.004 |
d | +0.004 |
Neg logit contributions | |
horizon | -0.004 |
peak | -0.004 |
itself | -0.004 |
heights | -0.004 |
edly | -0.004 |
Token | for |
Token ID | 329 |
Feature activation | +1.37e-02 |
Pos logit contributions | |
horizon | +0.006 |
edly | +0.005 |
itself | +0.005 |
heights | +0.005 |
peak | +0.005 |
Neg logit contributions | |
living | -0.005 |
mun | -0.004 |
have | -0.004 |
d | -0.004 |
w | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | -7.59e-03 |
Pos logit contributions | |
horizon | +0.032 |
peak | +0.028 |
comb | +0.028 |
edly | +0.027 |
ement | +0.027 |
Neg logit contributions | |
living | -0.020 |
have | -0.015 |
mun | -0.014 |
d | -0.014 |
w | -0.014 |
Token | while |
Token ID | 981 |
Feature activation | +5.19e-02 |
Pos logit contributions | |
living | +0.014 |
w | +0.013 |
have | +0.013 |
mun | +0.012 |
asc | +0.012 |
Neg logit contributions | |
horizon | -0.014 |
peak | -0.012 |
� | -0.011 |
comb | -0.011 |
heights | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | +4.02e-03 |
Pos logit contributions | |
horizon | +0.095 |
comb | +0.082 |
peak | +0.080 |
edly | +0.080 |
heights | +0.079 |
Neg logit contributions | |
living | -0.100 |
d | -0.078 |
watching | -0.075 |
mun | -0.074 |
Did | -0.071 |
Token | saw |
Token ID | 2497 |
Feature activation | +2.06e-03 |
Pos logit contributions | |
horizon | +0.007 |
itself | +0.006 |
peak | +0.006 |
heights | +0.006 |
edly | +0.006 |
Neg logit contributions | |
living | -0.008 |
mun | -0.007 |
w | -0.007 |
have | -0.007 |
sm | -0.006 |
Token | that |
Token ID | 326 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
horizon | +0.004 |
itself | +0.003 |
heights | +0.003 |
peak | +0.003 |
edly | +0.003 |
Neg logit contributions | |
living | -0.004 |
w | -0.003 |
mun | -0.003 |
have | -0.003 |
d | -0.003 |
Token | their |
Token ID | 511 |
Feature activation | +6.35e-03 |
Pos logit contributions | |
living | +0.027 |
mun | +0.022 |
w | +0.022 |
have | +0.022 |
d | +0.021 |
Neg logit contributions | |
horizon | -0.030 |
itself | -0.026 |
peak | -0.026 |
heights | -0.026 |
edly | -0.025 |
Token | kicks |
Token ID | 17364 |
Feature activation | +8.47e-03 |
Pos logit contributions | |
horizon | +0.012 |
itself | +0.011 |
comb | +0.010 |
edly | +0.010 |
peak | +0.010 |
Neg logit contributions | |
living | -0.012 |
mun | -0.009 |
d | -0.009 |
have | -0.009 |
w | -0.009 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
horizon | +0.011 |
comb | +0.010 |
heights | +0.010 |
ement | +0.010 |
peak | +0.010 |
Neg logit contributions | |
living | -0.010 |
mun | -0.008 |
Good | -0.008 |
Enjoy | -0.007 |
d | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | +1.21e-03 |
Pos logit contributions | |
horizon | +0.040 |
itself | +0.033 |
peak | +0.033 |
heights | +0.033 |
edly | +0.032 |
Neg logit contributions | |
living | -0.052 |
mun | -0.044 |
have | -0.044 |
w | -0.043 |
sm | -0.040 |
Token | he |
Token ID | 339 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
horizon | +0.002 |
itself | +0.002 |
peak | +0.002 |
heights | +0.002 |
edly | +0.002 |
Neg logit contributions | |
living | -0.002 |
mun | -0.002 |
have | -0.002 |
w | -0.002 |
d | -0.002 |
Token | heard |
Token ID | 2982 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
living | +0.015 |
mun | +0.012 |
have | +0.011 |
w | +0.011 |
Good | +0.009 |
Neg logit contributions | |
horizon | -0.035 |
peak | -0.032 |
itself | -0.031 |
edly | -0.031 |
heights | -0.030 |
Token | one |
Token ID | 530 |
Feature activation | +2.56e-03 |
Pos logit contributions | |
horizon | +0.026 |
itself | +0.023 |
peak | +0.022 |
heights | +0.022 |
edly | +0.022 |
Neg logit contributions | |
living | -0.020 |
mun | -0.016 |
w | -0.016 |
have | -0.016 |
d | -0.015 |
Token | of |
Token ID | 286 |
Feature activation | +5.23e-02 |
Pos logit contributions | |
horizon | +0.005 |
itself | +0.004 |
heights | +0.004 |
edly | +0.004 |
� | +0.004 |
Neg logit contributions | |
living | -0.005 |
mun | -0.004 |
have | -0.004 |
d | -0.004 |
sm | -0.003 |
Token | his |
Token ID | 465 |
Feature activation | +3.29e-03 |
Pos logit contributions | |
horizon | +0.118 |
edly | +0.103 |
comb | +0.103 |
peak | +0.099 |
ement | +0.098 |
Neg logit contributions | |
living | -0.081 |
mun | -0.060 |
d | -0.059 |
have | -0.057 |
sm | -0.057 |
Token | friends |
Token ID | 2460 |
Feature activation | +9.24e-03 |
Pos logit contributions | |
horizon | +0.006 |
comb | +0.005 |
itself | +0.005 |
rily | +0.005 |
edly | +0.005 |
Neg logit contributions | |
living | -0.006 |
d | -0.005 |
stri | -0.004 |
loves | -0.004 |
mun | -0.004 |
Token | calling |
Token ID | 4585 |
Feature activation | +1.26e-03 |
Pos logit contributions | |
horizon | +0.020 |
peak | +0.017 |
itself | +0.017 |
heights | +0.017 |
edly | +0.016 |
Neg logit contributions | |
living | -0.015 |
mun | -0.012 |
w | -0.012 |
have | -0.012 |
d | -0.011 |
Token | for |
Token ID | 329 |
Feature activation | +2.20e-02 |
Pos logit contributions | |
horizon | +0.003 |
itself | +0.002 |
peak | +0.002 |
heights | +0.002 |
edly | +0.002 |
Neg logit contributions | |
living | -0.002 |
mun | -0.002 |
w | -0.002 |
have | -0.002 |
d | -0.002 |
Token | him |
Token ID | 683 |
Feature activation | +3.02e-02 |
Pos logit contributions | |
horizon | +0.047 |
peak | +0.040 |
edly | +0.040 |
itself | +0.040 |
heights | +0.040 |
Neg logit contributions | |
living | -0.035 |
mun | -0.028 |
have | -0.027 |
w | -0.027 |
d | -0.026 |
Token | pick |
Token ID | 2298 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
horizon | +0.030 |
ement | +0.027 |
edly | +0.026 |
heights | +0.024 |
rily | +0.024 |
Neg logit contributions | |
living | -0.019 |
d | -0.015 |
have | -0.013 |
mun | -0.013 |
w | -0.013 |
Token | up |
Token ID | 510 |
Feature activation | +3.47e-02 |
Pos logit contributions | |
living | +0.018 |
mun | +0.014 |
d | +0.014 |
w | +0.013 |
have | +0.013 |
Neg logit contributions | |
horizon | -0.025 |
edly | -0.022 |
peak | -0.021 |
comb | -0.021 |
fingertips | -0.021 |
Token | sticks |
Token ID | 16461 |
Feature activation | +6.86e-03 |
Pos logit contributions | |
horizon | +0.078 |
peak | +0.065 |
edly | +0.064 |
comb | +0.064 |
ement | +0.063 |
Neg logit contributions | |
living | -0.056 |
mun | -0.044 |
d | -0.043 |
have | -0.041 |
sm | -0.040 |
Token | and |
Token ID | 290 |
Feature activation | +1.87e-02 |
Pos logit contributions | |
horizon | +0.016 |
fingertips | +0.014 |
� | +0.014 |
peak | +0.014 |
heights | +0.014 |
Neg logit contributions | |
living | -0.010 |
d | -0.007 |
sm | -0.007 |
mun | -0.006 |
w | -0.006 |
Token | give |
Token ID | 1577 |
Feature activation | +7.11e-04 |
Pos logit contributions | |
horizon | +0.035 |
peak | +0.029 |
ement | +0.029 |
heights | +0.028 |
itself | +0.028 |
Neg logit contributions | |
living | -0.036 |
mun | -0.030 |
w | -0.029 |
have | -0.029 |
d | -0.028 |
Token | them |
Token ID | 606 |
Feature activation | +5.16e-02 |
Pos logit contributions | |
horizon | +0.001 |
peak | +0.001 |
edly | +0.001 |
heights | +0.001 |
comb | +0.001 |
Neg logit contributions | |
living | -0.001 |
mun | -0.001 |
w | -0.001 |
d | -0.001 |
have | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
horizon | +0.103 |
peak | +0.087 |
itself | +0.087 |
heights | +0.087 |
edly | +0.086 |
Neg logit contributions | |
living | -0.090 |
mun | -0.073 |
w | -0.071 |
have | -0.071 |
d | -0.068 |
Token | Bird |
Token ID | 14506 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
horizon | +0.044 |
edly | +0.037 |
ement | +0.037 |
peak | +0.036 |
heights | +0.036 |
Neg logit contributions | |
living | -0.033 |
mun | -0.026 |
d | -0.025 |
have | -0.024 |
w | -0.024 |
Token | ie |
Token ID | 494 |
Feature activation | +3.33e-02 |
Pos logit contributions | |
horizon | +0.040 |
itself | +0.035 |
peak | +0.035 |
heights | +0.035 |
edly | +0.035 |
Neg logit contributions | |
living | -0.027 |
mun | -0.021 |
w | -0.021 |
have | -0.021 |
d | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | +4.08e-04 |
Pos logit contributions | |
horizon | +0.067 |
peak | +0.055 |
edly | +0.055 |
heights | +0.055 |
itself | +0.054 |
Neg logit contributions | |
living | -0.061 |
mun | -0.048 |
have | -0.047 |
d | -0.046 |
w | -0.046 |
Token | Together |
Token ID | 17083 |
Feature activation | +2.49e-02 |
Pos logit contributions | |
horizon | +0.001 |
heights | +0.001 |
peak | +0.001 |
ement | +0.001 |
fingertips | +0.001 |
Neg logit contributions | |
living | -0.001 |
mun | -0.001 |
Good | -0.001 |
d | -0.001 |
have | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -2.46e-04 |
Pos logit contributions | |
horizon | +0.017 |
edly | +0.015 |
ement | +0.014 |
heights | +0.014 |
peak | +0.014 |
Neg logit contributions | |
living | -0.014 |
mun | -0.010 |
d | -0.010 |
have | -0.010 |
stole | -0.010 |
Token | He |
Token ID | 679 |
Feature activation | +4.52e-03 |
Pos logit contributions | |
living | +0.000 |
L | +0.000 |
loves | +0.000 |
took | +0.000 |
d | +0.000 |
Neg logit contributions | |
horizon | -0.001 |
comb | -0.001 |
peak | -0.000 |
ement | -0.000 |
fingertips | -0.000 |
Token | tried |
Token ID | 3088 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
horizon | +0.010 |
itself | +0.009 |
edly | +0.009 |
peak | +0.009 |
comb | +0.009 |
Neg logit contributions | |
living | -0.006 |
mun | -0.005 |
w | -0.005 |
have | -0.005 |
Good | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | +6.48e-03 |
Pos logit contributions | |
living | +0.020 |
mun | +0.015 |
have | +0.015 |
w | +0.015 |
d | +0.015 |
Neg logit contributions | |
horizon | -0.027 |
peak | -0.023 |
edly | -0.023 |
heights | -0.023 |
ement | -0.023 |
Token | restore |
Token ID | 11169 |
Feature activation | -5.28e-03 |
Pos logit contributions | |
horizon | +0.015 |
ement | +0.013 |
edly | +0.013 |
heights | +0.013 |
peak | +0.013 |
Neg logit contributions | |
living | -0.010 |
have | -0.008 |
mun | -0.007 |
d | -0.007 |
w | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
living | +0.008 |
d | +0.006 |
w | +0.005 |
mun | +0.005 |
sm | +0.005 |
Neg logit contributions | |
horizon | -0.013 |
edly | -0.012 |
comb | -0.011 |
� | -0.011 |
peak | -0.011 |
Token | snake |
Token ID | 17522 |
Feature activation | +1.54e-03 |
Pos logit contributions | |
horizon | +0.019 |
itself | +0.016 |
peak | +0.016 |
heights | +0.016 |
edly | +0.015 |
Neg logit contributions | |
living | -0.029 |
mun | -0.025 |
have | -0.025 |
w | -0.025 |
sm | -0.023 |
Token | by |
Token ID | 416 |
Feature activation | +6.59e-03 |
Pos logit contributions | |
horizon | +0.004 |
ement | +0.003 |
edly | +0.003 |
heights | +0.003 |
peak | +0.003 |
Neg logit contributions | |
living | -0.003 |
d | -0.002 |
stole | -0.002 |
sm | -0.002 |
mun | -0.001 |
Token | giving |
Token ID | 3501 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
horizon | +0.014 |
peak | +0.011 |
edly | +0.011 |
ement | +0.011 |
comb | +0.011 |
Neg logit contributions | |
living | -0.012 |
d | -0.009 |
w | -0.009 |
have | -0.009 |
mun | -0.009 |
Token | it |
Token ID | 340 |
Feature activation | +2.53e-02 |
Pos logit contributions | |
living | +0.020 |
w | +0.017 |
d | +0.016 |
mun | +0.016 |
have | +0.016 |
Neg logit contributions | |
horizon | -0.020 |
edly | -0.017 |
comb | -0.017 |
peak | -0.017 |
ement | -0.017 |
Token | food |
Token ID | 2057 |
Feature activation | +2.33e-02 |
Pos logit contributions | |
horizon | +0.050 |
peak | +0.042 |
heights | +0.042 |
edly | +0.042 |
ement | +0.041 |
Neg logit contributions | |
living | -0.046 |
w | -0.036 |
have | -0.036 |
mun | -0.035 |
d | -0.035 |
Token | was |
Token ID | 373 |
Feature activation | +2.43e-03 |
Pos logit contributions | |
horizon | +0.020 |
heights | +0.018 |
itself | +0.018 |
peak | +0.018 |
comb | +0.017 |
Neg logit contributions | |
living | -0.010 |
mun | -0.008 |
d | -0.007 |
have | -0.007 |
w | -0.007 |
Token | sad |
Token ID | 6507 |
Feature activation | +7.34e-03 |
Pos logit contributions | |
horizon | +0.006 |
comb | +0.005 |
peak | +0.005 |
heights | +0.005 |
alks | +0.005 |
Neg logit contributions | |
living | -0.004 |
have | -0.003 |
watching | -0.003 |
mun | -0.003 |
d | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
horizon | +0.014 |
peak | +0.012 |
itself | +0.012 |
comb | +0.012 |
heights | +0.012 |
Neg logit contributions | |
living | -0.014 |
d | -0.011 |
mun | -0.011 |
have | -0.010 |
w | -0.010 |
Token | but |
Token ID | 475 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
living | +0.030 |
d | +0.024 |
mun | +0.024 |
watching | +0.024 |
Did | +0.024 |
Neg logit contributions | |
horizon | -0.030 |
peak | -0.027 |
alks | -0.027 |
comb | -0.026 |
alk | -0.025 |
Token | she |
Token ID | 673 |
Feature activation | +9.35e-03 |
Pos logit contributions | |
horizon | +0.049 |
peak | +0.043 |
heights | +0.042 |
alks | +0.042 |
edly | +0.041 |
Neg logit contributions | |
living | -0.042 |
mun | -0.033 |
d | -0.032 |
have | -0.032 |
w | -0.031 |
Token | ordered |
Token ID | 6149 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
horizon | +0.024 |
heights | +0.021 |
peak | +0.021 |
itself | +0.021 |
alks | +0.020 |
Neg logit contributions | |
living | -0.011 |
mun | -0.008 |
have | -0.008 |
d | -0.008 |
w | -0.008 |
Token | chocolate |
Token ID | 11311 |
Feature activation | +8.67e-03 |
Pos logit contributions | |
horizon | +0.041 |
peak | +0.035 |
heights | +0.035 |
comb | +0.034 |
edly | +0.034 |
Neg logit contributions | |
living | -0.027 |
mun | -0.020 |
d | -0.019 |
have | -0.019 |
w | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +6.58e-03 |
Pos logit contributions | |
horizon | +0.019 |
heights | +0.016 |
itself | +0.016 |
peak | +0.016 |
fingertips | +0.016 |
Neg logit contributions | |
living | -0.014 |
mun | -0.011 |
d | -0.011 |
w | -0.011 |
have | -0.010 |
Token | She |
Token ID | 1375 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
horizon | +0.012 |
peak | +0.010 |
heights | +0.010 |
comb | +0.010 |
itself | +0.010 |
Neg logit contributions | |
living | -0.013 |
Good | -0.011 |
mun | -0.010 |
L | -0.010 |
Enjoy | -0.010 |
Token | ate |
Token ID | 15063 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
horizon | +0.043 |
heights | +0.038 |
itself | +0.038 |
peak | +0.037 |
ement | +0.037 |
Neg logit contributions | |
living | -0.021 |
mun | -0.015 |
d | -0.015 |
have | -0.015 |
w | -0.014 |
Token | it |
Token ID | 340 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
horizon | +0.025 |
peak | +0.022 |
heights | +0.022 |
edly | +0.021 |
comb | +0.021 |
Neg logit contributions | |
living | -0.018 |
d | -0.014 |
mun | -0.014 |
w | -0.014 |
have | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | +7.03e-03 |
Pos logit contributions | |
horizon | +0.048 |
peak | +0.041 |
heights | +0.040 |
comb | +0.040 |
itself | +0.040 |
Neg logit contributions | |
living | -0.044 |
d | -0.033 |
mun | -0.033 |
sm | -0.033 |
have | -0.032 |
Token | colors |
Token ID | 7577 |
Feature activation | -6.63e-03 |
Pos logit contributions | |
horizon | +0.016 |
itself | +0.014 |
heights | +0.014 |
peak | +0.014 |
edly | +0.014 |
Neg logit contributions | |
living | -0.010 |
w | -0.008 |
mun | -0.008 |
have | -0.007 |
sm | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -1.62e-03 |
Pos logit contributions | |
living | +0.012 |
mun | +0.009 |
d | +0.009 |
have | +0.009 |
w | +0.009 |
Neg logit contributions | |
horizon | -0.013 |
heights | -0.011 |
peak | -0.011 |
edly | -0.011 |
itself | -0.011 |
Token | He |
Token ID | 679 |
Feature activation | -1.65e-03 |
Pos logit contributions | |
living | +0.002 |
Good | +0.002 |
loves | +0.002 |
Enjoy | +0.002 |
L | +0.002 |
Neg logit contributions | |
horizon | -0.004 |
comb | -0.004 |
rily | -0.003 |
peak | -0.003 |
edly | -0.003 |
Token | laughed |
Token ID | 13818 |
Feature activation | +1.54e-03 |
Pos logit contributions | |
living | +0.002 |
mun | +0.002 |
w | +0.002 |
have | +0.002 |
sm | +0.001 |
Neg logit contributions | |
horizon | -0.004 |
peak | -0.003 |
itself | -0.003 |
edly | -0.003 |
heights | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | -4.23e-03 |
Pos logit contributions | |
horizon | +0.003 |
peak | +0.002 |
heights | +0.002 |
comb | +0.002 |
� | +0.002 |
Neg logit contributions | |
living | -0.003 |
d | -0.003 |
have | -0.002 |
mun | -0.002 |
watching | -0.002 |
Token | cheered |
Token ID | 34767 |
Feature activation | +2.48e-03 |
Pos logit contributions | |
living | +0.006 |
mun | +0.004 |
have | +0.004 |
w | +0.004 |
sm | +0.004 |
Neg logit contributions | |
horizon | -0.010 |
itself | -0.009 |
edly | -0.009 |
peak | -0.009 |
heights | -0.008 |
Token | during |
Token ID | 1141 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
horizon | +0.005 |
peak | +0.004 |
heights | +0.004 |
comb | +0.004 |
edly | +0.004 |
Neg logit contributions | |
living | -0.004 |
have | -0.004 |
d | -0.004 |
mun | -0.003 |
w | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
horizon | +0.024 |
itself | +0.020 |
peak | +0.020 |
edly | +0.020 |
heights | +0.020 |
Neg logit contributions | |
living | -0.019 |
mun | -0.015 |
have | -0.015 |
w | -0.014 |
d | -0.014 |
Token | movie |
Token ID | 3807 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
living | +0.025 |
mun | +0.022 |
w | +0.022 |
have | +0.021 |
stole | +0.021 |
Neg logit contributions | |
horizon | -0.033 |
peak | -0.027 |
ement | -0.026 |
heights | -0.025 |
side | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -1.53e-03 |
Pos logit contributions | |
horizon | +0.027 |
heights | +0.024 |
itself | +0.024 |
edly | +0.023 |
peak | +0.023 |
Neg logit contributions | |
living | -0.023 |
d | -0.017 |
mun | -0.016 |
watching | -0.015 |
have | -0.015 |
Token | Tom |
Token ID | 4186 |
Feature activation | -1.35e-03 |
Pos logit contributions | |
living | +0.037 |
d | +0.030 |
have | +0.028 |
w | +0.028 |
mun | +0.027 |
Neg logit contributions | |
horizon | -0.050 |
edly | -0.046 |
peak | -0.044 |
comb | -0.042 |
itself | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
living | +0.003 |
mun | +0.002 |
have | +0.002 |
w | +0.002 |
d | +0.002 |
Neg logit contributions | |
horizon | -0.002 |
itself | -0.002 |
heights | -0.002 |
peak | -0.002 |
edly | -0.002 |
Token | We |
Token ID | 775 |
Feature activation | -1.50e-02 |
Pos logit contributions | |
living | +0.028 |
mun | +0.023 |
have | +0.022 |
w | +0.022 |
d | +0.021 |
Neg logit contributions | |
horizon | -0.039 |
itself | -0.034 |
peak | -0.033 |
heights | -0.033 |
edly | -0.033 |
Token | can |
Token ID | 460 |
Feature activation | -1.63e-03 |
Pos logit contributions | |
living | +0.022 |
mun | +0.017 |
have | +0.016 |
sm | +0.015 |
w | +0.015 |
Neg logit contributions | |
horizon | -0.032 |
itself | -0.029 |
peak | -0.029 |
comb | -0.029 |
� | -0.028 |
Token | still |
Token ID | 991 |
Feature activation | +2.94e-03 |
Pos logit contributions | |
living | +0.002 |
w | +0.002 |
have | +0.002 |
d | +0.002 |
mun | +0.002 |
Neg logit contributions | |
horizon | -0.004 |
edly | -0.003 |
heights | -0.003 |
ement | -0.003 |
itself | -0.003 |
Token | have |
Token ID | 423 |
Feature activation | +3.07e-03 |
Pos logit contributions | |
horizon | +0.007 |
ement | +0.006 |
edly | +0.006 |
heights | +0.006 |
rily | +0.006 |
Neg logit contributions | |
living | -0.004 |
have | -0.003 |
d | -0.003 |
w | -0.003 |
mun | -0.003 |
Token | fun |
Token ID | 1257 |
Feature activation | +2.67e-02 |
Pos logit contributions | |
horizon | +0.007 |
itself | +0.006 |
heights | +0.006 |
peak | +0.006 |
edly | +0.006 |
Neg logit contributions | |
living | -0.005 |
mun | -0.004 |
w | -0.004 |
have | -0.004 |
d | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
horizon | +0.049 |
itself | +0.042 |
heights | +0.042 |
peak | +0.041 |
edly | +0.041 |
Neg logit contributions | |
living | -0.049 |
mun | -0.042 |
w | -0.041 |
have | -0.041 |
haw | -0.038 |
Token | Let |
Token ID | 3914 |
Feature activation | -1.78e-02 |
Pos logit contributions | |
living | +0.021 |
mun | +0.017 |
w | +0.017 |
have | +0.017 |
Good | +0.016 |
Neg logit contributions | |
horizon | -0.021 |
peak | -0.018 |
heights | -0.017 |
edly | -0.017 |
itself | -0.017 |
Token | 's |
Token ID | 338 |
Feature activation | -2.82e-02 |
Pos logit contributions | |
living | +0.021 |
mun | +0.015 |
w | +0.015 |
have | +0.015 |
d | +0.014 |
Neg logit contributions | |
horizon | -0.046 |
peak | -0.041 |
itself | -0.041 |
heights | -0.040 |
edly | -0.040 |
Token | see |
Token ID | 766 |
Feature activation | -3.74e-03 |
Pos logit contributions | |
living | +0.041 |
have | +0.030 |
d | +0.030 |
w | +0.029 |
mun | +0.029 |
Neg logit contributions | |
horizon | -0.066 |
edly | -0.059 |
ement | -0.058 |
heights | -0.057 |
itself | -0.056 |
Token | road |
Token ID | 2975 |
Feature activation | +4.35e-03 |
Pos logit contributions | |
horizon | +0.003 |
peak | +0.003 |
ement | +0.002 |
heights | +0.002 |
destination | +0.002 |
Neg logit contributions | |
living | -0.003 |
w | -0.003 |
mun | -0.003 |
have | -0.003 |
sm | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | -3.95e-03 |
Pos logit contributions | |
horizon | +0.009 |
edly | +0.008 |
heights | +0.008 |
itself | +0.007 |
comb | +0.007 |
Neg logit contributions | |
living | -0.008 |
mun | -0.006 |
d | -0.006 |
have | -0.006 |
w | -0.006 |
Token | there |
Token ID | 612 |
Feature activation | -1.93e-03 |
Pos logit contributions | |
living | +0.007 |
mun | +0.006 |
w | +0.006 |
have | +0.006 |
d | +0.005 |
Neg logit contributions | |
horizon | -0.008 |
itself | -0.007 |
peak | -0.007 |
heights | -0.006 |
edly | -0.006 |
Token | was |
Token ID | 373 |
Feature activation | +4.91e-03 |
Pos logit contributions | |
living | +0.002 |
mun | +0.001 |
d | +0.001 |
w | +0.001 |
have | +0.001 |
Neg logit contributions | |
horizon | -0.005 |
edly | -0.005 |
heights | -0.005 |
peak | -0.005 |
itself | -0.005 |
Token | so |
Token ID | 523 |
Feature activation | +1.60e-03 |
Pos logit contributions | |
horizon | +0.010 |
itself | +0.009 |
peak | +0.009 |
heights | +0.009 |
edly | +0.009 |
Neg logit contributions | |
living | -0.008 |
mun | -0.006 |
w | -0.006 |
have | -0.006 |
d | -0.006 |
Token | much |
Token ID | 881 |
Feature activation | +3.58e-03 |
Pos logit contributions | |
horizon | +0.003 |
itself | +0.003 |
edly | +0.003 |
peak | +0.003 |
heights | +0.003 |
Neg logit contributions | |
living | -0.003 |
mun | -0.002 |
have | -0.002 |
w | -0.002 |
d | -0.002 |
Token | traffic |
Token ID | 4979 |
Feature activation | -3.16e-03 |
Pos logit contributions | |
horizon | +0.008 |
itself | +0.007 |
peak | +0.007 |
edly | +0.007 |
heights | +0.007 |
Neg logit contributions | |
living | -0.006 |
mun | -0.004 |
have | -0.004 |
w | -0.004 |
d | -0.004 |
Token | ! |
Token ID | 0 |
Feature activation | +5.06e-03 |
Pos logit contributions | |
living | +0.005 |
mun | +0.004 |
w | +0.004 |
have | +0.004 |
d | +0.004 |
Neg logit contributions | |
horizon | -0.006 |
itself | -0.006 |
peak | -0.006 |
heights | -0.005 |
edly | -0.005 |
Token | It |
Token ID | 632 |
Feature activation | -1.97e-03 |
Pos logit contributions | |
horizon | +0.012 |
comb | +0.011 |
heights | +0.010 |
peak | +0.010 |
edly | +0.010 |
Neg logit contributions | |
living | -0.008 |
mun | -0.006 |
Good | -0.006 |
d | -0.005 |
have | -0.005 |
Token | made |
Token ID | 925 |
Feature activation | +6.03e-03 |
Pos logit contributions | |
living | +0.002 |
mun | +0.002 |
w | +0.002 |
have | +0.002 |
d | +0.002 |
Neg logit contributions | |
horizon | -0.005 |
peak | -0.004 |
itself | -0.004 |
heights | -0.004 |
edly | -0.004 |
Token | Tim |
Token ID | 5045 |
Feature activation | +2.17e-04 |
Pos logit contributions | |
horizon | +0.013 |
comb | +0.012 |
edly | +0.012 |
peak | +0.011 |
heights | +0.011 |
Neg logit contributions | |
living | -0.010 |
have | -0.007 |
mun | -0.007 |
d | -0.007 |
w | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -2.23e-03 |
Pos logit contributions | |
horizon | +0.011 |
itself | +0.011 |
side | +0.009 |
edly | +0.009 |
peak | +0.009 |
Neg logit contributions | |
mun | -0.023 |
living | -0.023 |
have | -0.023 |
w | -0.022 |
haw | -0.022 |
Token | Tim |
Token ID | 5045 |
Feature activation | +9.19e-03 |
Pos logit contributions | |
living | +0.004 |
mun | +0.003 |
w | +0.003 |
have | +0.003 |
d | +0.003 |
Neg logit contributions | |
horizon | -0.005 |
edly | -0.004 |
heights | -0.004 |
peak | -0.004 |
itself | -0.004 |
Token | my |
Token ID | 1820 |
Feature activation | -5.55e-03 |
Pos logit contributions | |
horizon | +0.023 |
itself | +0.021 |
peak | +0.020 |
heights | +0.020 |
comb | +0.019 |
Neg logit contributions | |
living | -0.010 |
mun | -0.008 |
have | -0.007 |
w | -0.007 |
haw | -0.006 |
Token | loved |
Token ID | 6151 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
mun | +0.006 |
living | +0.006 |
have | +0.005 |
prise | +0.005 |
haw | +0.005 |
Neg logit contributions | |
horizon | -0.012 |
itself | -0.011 |
edly | -0.011 |
case | -0.011 |
comb | -0.011 |
Token | to |
Token ID | 284 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
w | +0.020 |
mun | +0.020 |
living | +0.020 |
have | +0.019 |
a | +0.018 |
Neg logit contributions | |
horizon | -0.016 |
heights | -0.015 |
itself | -0.014 |
peak | -0.013 |
branches | -0.013 |
Token | play |
Token ID | 711 |
Feature activation | -1.65e-02 |
Pos logit contributions | |
living | +0.016 |
mun | +0.013 |
have | +0.012 |
w | +0.012 |
d | +0.011 |
Neg logit contributions | |
horizon | -0.027 |
itself | -0.023 |
peak | -0.023 |
heights | -0.023 |
edly | -0.023 |
Token | with |
Token ID | 351 |
Feature activation | +3.55e-03 |
Pos logit contributions | |
living | +0.029 |
w | +0.026 |
mun | +0.026 |
haw | +0.025 |
have | +0.025 |
Neg logit contributions | |
horizon | -0.027 |
itself | -0.025 |
heights | -0.025 |
edly | -0.024 |
peak | -0.024 |
Token | his |
Token ID | 465 |
Feature activation | +5.59e-04 |
Pos logit contributions | |
horizon | +0.006 |
heights | +0.006 |
itself | +0.006 |
peak | +0.005 |
branches | +0.005 |
Neg logit contributions | |
living | -0.006 |
mun | -0.006 |
w | -0.005 |
have | -0.005 |
Good | -0.005 |
Token | sword |
Token ID | 8429 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
horizon | +0.001 |
heights | +0.001 |
peak | +0.001 |
itself | +0.001 |
ement | +0.001 |
Neg logit contributions | |
living | -0.001 |
mun | -0.001 |
have | -0.001 |
w | -0.001 |
sm | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
living | +0.018 |
d | +0.013 |
w | +0.013 |
have | +0.012 |
mun | +0.012 |
Neg logit contributions | |
horizon | -0.022 |
itself | -0.018 |
heights | -0.018 |
peak | -0.018 |
ement | -0.018 |
Token | pretend |
Token ID | 16614 |
Feature activation | -4.42e-04 |
Pos logit contributions | |
horizon | +0.030 |
peak | +0.026 |
heights | +0.026 |
itself | +0.026 |
comb | +0.024 |
Neg logit contributions | |
living | -0.028 |
w | -0.024 |
mun | -0.024 |
have | -0.024 |
sm | -0.024 |
Token | ily |
Token ID | 813 |
Feature activation | +7.78e-03 |
Pos logit contributions | |
horizon | +0.013 |
peak | +0.011 |
itself | +0.011 |
heights | +0.011 |
edly | +0.011 |
Neg logit contributions | |
living | -0.012 |
mun | -0.009 |
w | -0.009 |
have | -0.009 |
d | -0.009 |
Token | saw |
Token ID | 2497 |
Feature activation | +3.06e-03 |
Pos logit contributions | |
horizon | +0.017 |
itself | +0.015 |
peak | +0.015 |
heights | +0.015 |
edly | +0.014 |
Neg logit contributions | |
living | -0.012 |
mun | -0.010 |
have | -0.009 |
w | -0.009 |
d | -0.009 |
Token | a |
Token ID | 257 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
horizon | +0.006 |
peak | +0.005 |
itself | +0.005 |
edly | +0.005 |
heights | +0.005 |
Neg logit contributions | |
living | -0.005 |
mun | -0.004 |
w | -0.004 |
have | -0.004 |
d | -0.004 |
Token | painting |
Token ID | 12036 |
Feature activation | -1.00e-02 |
Pos logit contributions | |
living | +0.029 |
have | +0.028 |
w | +0.027 |
mun | +0.027 |
sm | +0.025 |
Neg logit contributions | |
horizon | -0.029 |
peak | -0.025 |
fingertips | -0.023 |
delight | -0.022 |
comb | -0.022 |
Token | of |
Token ID | 286 |
Feature activation | +3.66e-02 |
Pos logit contributions | |
living | +0.014 |
w | +0.010 |
mun | +0.009 |
have | +0.009 |
d | +0.009 |
Neg logit contributions | |
horizon | -0.024 |
itself | -0.021 |
heights | -0.021 |
peak | -0.021 |
edly | -0.021 |
Token | a |
Token ID | 257 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
horizon | +0.075 |
itself | +0.065 |
heights | +0.064 |
peak | +0.064 |
edly | +0.062 |
Neg logit contributions | |
living | -0.060 |
mun | -0.050 |
w | -0.049 |
have | -0.048 |
d | -0.045 |
Token | girl |
Token ID | 2576 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
living | +0.026 |
have | +0.025 |
mun | +0.024 |
w | +0.023 |
sm | +0.021 |
Neg logit contributions | |
horizon | -0.027 |
peak | -0.023 |
heights | -0.021 |
fingertips | -0.021 |
itself | -0.021 |
Token | with |
Token ID | 351 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
horizon | +0.042 |
itself | +0.037 |
heights | +0.036 |
peak | +0.036 |
edly | +0.036 |
Neg logit contributions | |
living | -0.025 |
mun | -0.020 |
w | -0.020 |
have | -0.020 |
haw | -0.019 |
Token | a |
Token ID | 257 |
Feature activation | -5.56e-03 |
Pos logit contributions | |
horizon | +0.035 |
itself | +0.031 |
peak | +0.030 |
heights | +0.030 |
edly | +0.030 |
Neg logit contributions | |
living | -0.026 |
mun | -0.021 |
have | -0.021 |
w | -0.021 |
d | -0.019 |
Token | smile |
Token ID | 8212 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
living | +0.010 |
mun | +0.009 |
have | +0.009 |
w | +0.008 |
sm | +0.008 |
Neg logit contributions | |
horizon | -0.010 |
peak | -0.009 |
heights | -0.008 |
itself | -0.008 |
fingertips | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -2.06e-03 |
Pos logit contributions | |
horizon | +0.021 |
itself | +0.018 |
edly | +0.017 |
peak | +0.017 |
heights | +0.017 |
Neg logit contributions | |
living | -0.020 |
mun | -0.017 |
have | -0.016 |
w | -0.016 |
d | -0.015 |
Token | See |
Token ID | 6214 |
Feature activation | -3.54e-02 |
Pos logit contributions | |
horizon | +0.012 |
edly | +0.010 |
itself | +0.010 |
peak | +0.010 |
heights | +0.010 |
Neg logit contributions | |
living | -0.007 |
mun | -0.005 |
w | -0.005 |
have | -0.005 |
d | -0.005 |
Token | you |
Token ID | 345 |
Feature activation | -2.97e-02 |
Pos logit contributions | |
living | +0.074 |
mun | +0.065 |
have | +0.062 |
w | +0.060 |
Did | +0.059 |
Neg logit contributions | |
horizon | -0.056 |
itself | -0.050 |
heights | -0.049 |
peak | -0.048 |
ement | -0.047 |
Token | soon |
Token ID | 2582 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
living | +0.068 |
mun | +0.060 |
w | +0.055 |
have | +0.055 |
Mechan | +0.055 |
Neg logit contributions | |
horizon | -0.043 |
itself | -0.035 |
peak | -0.034 |
heights | -0.034 |
edly | -0.033 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
horizon | +0.039 |
itself | +0.033 |
peak | +0.033 |
heights | +0.033 |
edly | +0.032 |
Neg logit contributions | |
living | -0.036 |
mun | -0.030 |
w | -0.029 |
have | -0.029 |
d | -0.028 |
Token | Tom |
Token ID | 4186 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
horizon | +0.039 |
comb | +0.035 |
edly | +0.034 |
itself | +0.034 |
peak | +0.034 |
Neg logit contributions | |
living | -0.031 |
mun | -0.024 |
d | -0.023 |
w | -0.022 |
loves | -0.022 |
Token | says |
Token ID | 1139 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
living | +0.021 |
mun | +0.016 |
have | +0.015 |
d | +0.014 |
loves | +0.013 |
Neg logit contributions | |
horizon | -0.039 |
comb | -0.035 |
edly | -0.035 |
peak | -0.035 |
itself | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -3.34e-03 |
Pos logit contributions | |
living | +0.020 |
d | +0.015 |
mun | +0.015 |
have | +0.015 |
sm | +0.014 |
Neg logit contributions | |
horizon | -0.027 |
heights | -0.024 |
peak | -0.023 |
comb | -0.022 |
� | -0.022 |
Token |
|
Token ID | 198 |
Feature activation | -2.98e-04 |
Pos logit contributions | |
living | +0.005 |
Are | +0.003 |
Good | +0.003 |
L | +0.003 |
Did | +0.003 |
Neg logit contributions | |
horizon | -0.008 |
peak | -0.008 |
heights | -0.007 |
iously | -0.007 |
comb | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | +2.94e-03 |
Pos logit contributions | |
living | +0.001 |
mun | +0.000 |
have | +0.000 |
w | +0.000 |
d | +0.000 |
Neg logit contributions | |
horizon | -0.001 |
peak | -0.000 |
heights | -0.000 |
itself | -0.000 |
edly | -0.000 |
Token | They |
Token ID | 2990 |
Feature activation | +8.76e-03 |
Pos logit contributions | |
horizon | +0.006 |
itself | +0.005 |
edly | +0.005 |
peak | +0.005 |
heights | +0.005 |
Neg logit contributions | |
living | -0.005 |
mun | -0.004 |
w | -0.004 |
have | -0.004 |
stole | -0.004 |
Token | get |
Token ID | 651 |
Feature activation | +1.21e-03 |
Pos logit contributions | |
horizon | +0.018 |
peak | +0.016 |
itself | +0.016 |
heights | +0.016 |
edly | +0.015 |
Neg logit contributions | |
living | -0.014 |
mun | -0.012 |
w | -0.011 |
have | -0.011 |
d | -0.011 |
Token | " |
Token ID | 366 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
horizon | +0.021 |
itself | +0.018 |
edly | +0.017 |
heights | +0.017 |
peak | +0.017 |
Neg logit contributions | |
living | -0.022 |
w | -0.019 |
mun | -0.019 |
have | -0.018 |
haw | -0.017 |
Token | Wow |
Token ID | 22017 |
Feature activation | -8.58e-03 |
Pos logit contributions | |
horizon | +0.029 |
itself | +0.026 |
heights | +0.025 |
edly | +0.025 |
peak | +0.024 |
Neg logit contributions | |
living | -0.019 |
mun | -0.015 |
have | -0.014 |
w | -0.014 |
d | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | -7.74e-03 |
Pos logit contributions | |
living | +0.020 |
mun | +0.019 |
have | +0.018 |
Say | +0.018 |
Did | +0.017 |
Neg logit contributions | |
horizon | -0.010 |
itself | -0.009 |
peak | -0.008 |
heights | -0.008 |
ement | -0.008 |
Token | Tim |
Token ID | 5045 |
Feature activation | +7.08e-03 |
Pos logit contributions | |
living | +0.017 |
mun | +0.015 |
have | +0.014 |
w | +0.014 |
P | +0.014 |
Neg logit contributions | |
horizon | -0.011 |
itself | -0.011 |
heights | -0.010 |
peak | -0.009 |
ement | -0.009 |
Token | ! |
Token ID | 0 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
horizon | +0.011 |
itself | +0.009 |
heights | +0.009 |
peak | +0.009 |
ement | +0.009 |
Neg logit contributions | |
living | -0.015 |
mun | -0.013 |
have | -0.013 |
w | -0.012 |
haw | -0.012 |
Token | You |
Token ID | 921 |
Feature activation | -2.37e-02 |
Pos logit contributions | |
living | +0.024 |
mun | +0.022 |
have | +0.021 |
w | +0.020 |
sm | +0.019 |
Neg logit contributions | |
horizon | -0.024 |
itself | -0.021 |
� | -0.020 |
peak | -0.020 |
heights | -0.019 |
Token | found |
Token ID | 1043 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
living | +0.037 |
mun | +0.032 |
have | +0.029 |
w | +0.028 |
sm | +0.028 |
Neg logit contributions | |
horizon | -0.049 |
itself | -0.044 |
peak | -0.044 |
heights | -0.042 |
comb | -0.042 |
Token | a |
Token ID | 257 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
living | +0.018 |
mun | +0.016 |
have | +0.015 |
w | +0.015 |
d | +0.013 |
Neg logit contributions | |
horizon | -0.022 |
itself | -0.020 |
heights | -0.019 |
peak | -0.019 |
edly | -0.018 |
Token | beautiful |
Token ID | 4950 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
living | +0.020 |
mun | +0.019 |
have | +0.018 |
w | +0.017 |
sm | +0.016 |
Neg logit contributions | |
horizon | -0.023 |
peak | -0.020 |
heights | -0.019 |
itself | -0.019 |
destination | -0.018 |
Token | treasure |
Token ID | 14068 |
Feature activation | -6.67e-03 |
Pos logit contributions | |
horizon | +0.033 |
peak | +0.027 |
itself | +0.027 |
heights | +0.026 |
comb | +0.025 |
Neg logit contributions | |
living | -0.042 |
mun | -0.038 |
have | -0.037 |
w | -0.036 |
sm | -0.034 |
Token | !" |
Token ID | 2474 |
Feature activation | +2.08e-02 |
Pos logit contributions | |
living | +0.011 |
w | +0.009 |
d | +0.008 |
mun | +0.008 |
have | +0.008 |
Neg logit contributions | |
horizon | -0.014 |
itself | -0.012 |
edly | -0.012 |
heights | -0.012 |
peak | -0.012 |
Token | It |
Token ID | 1026 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
horizon | +0.012 |
peak | +0.011 |
heights | +0.011 |
ement | +0.010 |
comb | +0.010 |
Neg logit contributions | |
living | -0.007 |
have | -0.005 |
w | -0.005 |
mun | -0.005 |
Did | -0.005 |
Token | is |
Token ID | 318 |
Feature activation | +2.22e-03 |
Pos logit contributions | |
living | +0.027 |
mun | +0.024 |
have | +0.023 |
prise | +0.023 |
w | +0.023 |
Neg logit contributions | |
horizon | -0.036 |
heights | -0.033 |
peak | -0.033 |
itself | -0.032 |
branches | -0.030 |
Token | so |
Token ID | 523 |
Feature activation | -4.47e-03 |
Pos logit contributions | |
horizon | +0.004 |
edly | +0.004 |
ement | +0.003 |
peak | +0.003 |
comb | +0.003 |
Neg logit contributions | |
living | -0.004 |
d | -0.004 |
have | -0.003 |
w | -0.003 |
watching | -0.003 |
Token | big |
Token ID | 1263 |
Feature activation | +7.21e-03 |
Pos logit contributions | |
living | +0.008 |
mun | +0.007 |
have | +0.006 |
w | +0.006 |
Did | +0.006 |
Neg logit contributions | |
horizon | -0.009 |
itself | -0.008 |
peak | -0.007 |
heights | -0.007 |
ement | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | -7.56e-03 |
Pos logit contributions | |
horizon | +0.014 |
peak | +0.012 |
fingertips | +0.012 |
edly | +0.012 |
comb | +0.012 |
Neg logit contributions | |
living | -0.014 |
d | -0.011 |
w | -0.010 |
have | -0.010 |
mun | -0.010 |
Token | funny |
Token ID | 8258 |
Feature activation | -1.78e-02 |
Pos logit contributions | |
living | +0.013 |
mun | +0.011 |
have | +0.011 |
w | +0.010 |
sm | +0.010 |
Neg logit contributions | |
horizon | -0.015 |
itself | -0.013 |
peak | -0.013 |
heights | -0.013 |
edly | -0.012 |
Token | !" |
Token ID | 2474 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
living | +0.030 |
mun | +0.023 |
w | +0.023 |
have | +0.023 |
d | +0.022 |
Neg logit contributions | |
horizon | -0.038 |
peak | -0.032 |
edly | -0.031 |
itself | -0.031 |
heights | -0.031 |
Token |
|
Token ID | 198 |
Feature activation | -4.46e-03 |
Pos logit contributions | |
comb | +0.061 |
horizon | +0.060 |
rily | +0.056 |
edly | +0.054 |
ement | +0.053 |
Neg logit contributions | |
living | -0.024 |
L | -0.022 |
d | -0.020 |
took | -0.019 |
saw | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | +3.09e-03 |
Pos logit contributions | |
living | +0.007 |
w | +0.006 |
mun | +0.006 |
have | +0.006 |
d | +0.005 |
Neg logit contributions | |
horizon | -0.010 |
edly | -0.008 |
comb | -0.008 |
heights | -0.008 |
peak | -0.008 |
Token | " |
Token ID | 1 |
Feature activation | +5.97e-03 |
Pos logit contributions | |
horizon | +0.007 |
comb | +0.006 |
peak | +0.006 |
edly | +0.006 |
itself | +0.006 |
Neg logit contributions | |
living | -0.004 |
have | -0.003 |
w | -0.003 |
mun | -0.003 |
d | -0.003 |
Token | Let |
Token ID | 5756 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
horizon | +0.014 |
peak | +0.012 |
itself | +0.012 |
heights | +0.012 |
comb | +0.012 |
Neg logit contributions | |
living | -0.009 |
mun | -0.007 |
have | -0.007 |
w | -0.007 |
d | -0.006 |
Token | with |
Token ID | 351 |
Feature activation | +5.71e-03 |
Pos logit contributions | |
living | +0.043 |
mun | +0.036 |
w | +0.036 |
have | +0.035 |
haw | +0.034 |
Neg logit contributions | |
horizon | -0.057 |
itself | -0.052 |
heights | -0.051 |
peak | -0.051 |
edly | -0.050 |
Token | their |
Token ID | 511 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
horizon | +0.013 |
peak | +0.011 |
edly | +0.011 |
itself | +0.011 |
heights | +0.011 |
Neg logit contributions | |
living | -0.009 |
mun | -0.007 |
w | -0.007 |
have | -0.006 |
d | -0.006 |
Token | toys |
Token ID | 14958 |
Feature activation | +6.50e-03 |
Pos logit contributions | |
horizon | +0.051 |
itself | +0.045 |
peak | +0.044 |
heights | +0.044 |
edly | +0.044 |
Neg logit contributions | |
living | -0.033 |
mun | -0.025 |
w | -0.025 |
have | -0.025 |
d | -0.023 |
Token | in |
Token ID | 287 |
Feature activation | -1.61e-02 |
Pos logit contributions | |
horizon | +0.015 |
itself | +0.013 |
peak | +0.013 |
heights | +0.012 |
edly | +0.012 |
Neg logit contributions | |
living | -0.011 |
mun | -0.008 |
d | -0.007 |
w | -0.007 |
have | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | -9.94e-03 |
Pos logit contributions | |
living | +0.024 |
d | +0.014 |
mun | +0.014 |
L | +0.014 |
have | +0.013 |
Neg logit contributions | |
horizon | -0.041 |
edly | -0.038 |
comb | -0.037 |
iously | -0.036 |
ement | -0.035 |
Token | living |
Token ID | 2877 |
Feature activation | -7.20e-02 |
Pos logit contributions | |
living | +0.016 |
mun | +0.015 |
w | +0.015 |
have | +0.015 |
haw | +0.015 |
Neg logit contributions | |
horizon | -0.019 |
peak | -0.016 |
heights | -0.016 |
itself | -0.016 |
side | -0.015 |
Token | room |
Token ID | 2119 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
living | +0.137 |
mun | +0.111 |
have | +0.109 |
w | +0.109 |
d | +0.105 |
Neg logit contributions | |
horizon | -0.135 |
peak | -0.113 |
itself | -0.113 |
heights | -0.112 |
edly | -0.111 |
Token | . |
Token ID | 13 |
Feature activation | -1.14e-02 |
Pos logit contributions | |
horizon | +0.020 |
itself | +0.017 |
peak | +0.016 |
heights | +0.016 |
edly | +0.016 |
Neg logit contributions | |
living | -0.019 |
mun | -0.015 |
have | -0.015 |
d | -0.014 |
w | -0.014 |
Token | Tom |
Token ID | 4186 |
Feature activation | +7.71e-03 |
Pos logit contributions | |
living | +0.018 |
loves | +0.014 |
d | +0.013 |
L | +0.012 |
Enjoy | +0.012 |
Neg logit contributions | |
horizon | -0.025 |
comb | -0.022 |
edly | -0.022 |
tal | -0.022 |
itself | -0.022 |
Token | had |
Token ID | 550 |
Feature activation | -2.35e-02 |
Pos logit contributions | |
horizon | +0.018 |
itself | +0.015 |
peak | +0.015 |
edly | +0.015 |
comb | +0.015 |
Neg logit contributions | |
living | -0.009 |
w | -0.008 |
have | -0.008 |
mun | -0.008 |
haw | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | -4.10e-03 |
Pos logit contributions | |
living | +0.040 |
mun | +0.038 |
w | +0.036 |
have | +0.034 |
haw | +0.033 |
Neg logit contributions | |
horizon | -0.041 |
itself | -0.039 |
heights | -0.037 |
peak | -0.036 |
edly | -0.036 |
Token | was |
Token ID | 373 |
Feature activation | +7.67e-03 |
Pos logit contributions | |
living | +0.002 |
mun | +0.002 |
have | +0.002 |
w | +0.002 |
d | +0.002 |
Neg logit contributions | |
horizon | -0.005 |
peak | -0.004 |
itself | -0.004 |
heights | -0.004 |
edly | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | -4.48e-03 |
Pos logit contributions | |
horizon | +0.018 |
edly | +0.016 |
ement | +0.016 |
peak | +0.016 |
comb | +0.016 |
Neg logit contributions | |
living | -0.011 |
d | -0.008 |
watching | -0.007 |
mun | -0.007 |
w | -0.007 |
Token | television |
Token ID | 5581 |
Feature activation | -2.81e-02 |
Pos logit contributions | |
living | +0.009 |
mun | +0.007 |
w | +0.007 |
have | +0.007 |
d | +0.007 |
Neg logit contributions | |
horizon | -0.008 |
itself | -0.007 |
edly | -0.007 |
peak | -0.007 |
heights | -0.007 |
Token | in |
Token ID | 287 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
living | +0.046 |
mun | +0.035 |
d | +0.034 |
w | +0.034 |
have | +0.033 |
Neg logit contributions | |
horizon | -0.060 |
itself | -0.053 |
edly | -0.052 |
heights | -0.052 |
peak | -0.051 |
Token | the |
Token ID | 262 |
Feature activation | -2.05e-02 |
Pos logit contributions | |
living | +0.031 |
d | +0.023 |
w | +0.022 |
sm | +0.021 |
have | +0.020 |
Neg logit contributions | |
horizon | -0.051 |
edly | -0.045 |
ement | -0.045 |
comb | -0.044 |
iously | -0.044 |
Token | living |
Token ID | 2877 |
Feature activation | -7.69e-02 |
Pos logit contributions | |
living | +0.043 |
mun | +0.039 |
have | +0.039 |
w | +0.038 |
haw | +0.036 |
Neg logit contributions | |
horizon | -0.033 |
peak | -0.026 |
itself | -0.025 |
heights | -0.025 |
edly | -0.024 |
Token | room |
Token ID | 2119 |
Feature activation | +2.17e-03 |
Pos logit contributions | |
living | +0.145 |
mun | +0.117 |
w | +0.116 |
have | +0.115 |
d | +0.112 |
Neg logit contributions | |
horizon | -0.146 |
peak | -0.123 |
heights | -0.122 |
itself | -0.122 |
comb | -0.121 |
Token | . |
Token ID | 13 |
Feature activation | -1.81e-04 |
Pos logit contributions | |
horizon | +0.004 |
heights | +0.004 |
peak | +0.004 |
itself | +0.004 |
edly | +0.004 |
Neg logit contributions | |
living | -0.004 |
d | -0.003 |
mun | -0.003 |
w | -0.003 |
have | -0.003 |
Token | It |
Token ID | 632 |
Feature activation | -3.72e-03 |
Pos logit contributions | |
living | +0.000 |
Good | +0.000 |
Did | +0.000 |
d | +0.000 |
L | +0.000 |
Neg logit contributions | |
horizon | -0.000 |
peak | -0.000 |
heights | -0.000 |
comb | -0.000 |
iously | -0.000 |
Token | was |
Token ID | 373 |
Feature activation | +6.04e-03 |
Pos logit contributions | |
living | +0.005 |
mun | +0.004 |
have | +0.004 |
w | +0.004 |
Enjoy | +0.003 |
Neg logit contributions | |
horizon | -0.009 |
peak | -0.008 |
itself | -0.008 |
edly | -0.007 |
comb | -0.007 |
Token | on |
Token ID | 319 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
horizon | +0.013 |
edly | +0.012 |
heights | +0.012 |
peak | +0.012 |
itself | +0.012 |
Neg logit contributions | |
living | -0.010 |
d | -0.007 |
w | -0.007 |
mun | -0.007 |
have | -0.007 |
Token | more |
Token ID | 517 |
Feature activation | +4.76e-03 |
Pos logit contributions | |
horizon | +0.013 |
itself | +0.011 |
peak | +0.011 |
heights | +0.011 |
edly | +0.011 |
Neg logit contributions | |
living | -0.011 |
mun | -0.009 |
w | -0.009 |
have | -0.009 |
d | -0.008 |
Token | fun |
Token ID | 1257 |
Feature activation | +2.74e-02 |
Pos logit contributions | |
horizon | +0.010 |
itself | +0.009 |
peak | +0.009 |
heights | +0.009 |
edly | +0.009 |
Neg logit contributions | |
living | -0.007 |
mun | -0.006 |
have | -0.006 |
w | -0.006 |
d | -0.005 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
horizon | +0.055 |
peak | +0.047 |
edly | +0.047 |
heights | +0.046 |
itself | +0.046 |
Neg logit contributions | |
living | -0.048 |
have | -0.038 |
mun | -0.038 |
w | -0.037 |
d | -0.036 |
Token |
|
Token ID | 198 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
horizon | +0.043 |
comb | +0.040 |
peak | +0.040 |
heights | +0.038 |
edly | +0.037 |
Neg logit contributions | |
living | -0.027 |
mun | -0.021 |
L | -0.019 |
have | -0.019 |
P | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | -2.38e-03 |
Pos logit contributions | |
living | +0.021 |
mun | +0.018 |
w | +0.017 |
have | +0.017 |
d | +0.016 |
Neg logit contributions | |
horizon | -0.022 |
peak | -0.019 |
itself | -0.019 |
heights | -0.019 |
edly | -0.019 |
Token | P |
Token ID | 47 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
living | +0.003 |
mun | +0.002 |
w | +0.002 |
d | +0.002 |
sm | +0.002 |
Neg logit contributions | |
horizon | -0.006 |
itself | -0.005 |
edly | -0.005 |
heights | -0.005 |
peak | -0.005 |
Token | enny |
Token ID | 11870 |
Feature activation | -1.95e-02 |
Pos logit contributions | |
living | +0.129 |
w | +0.105 |
have | +0.104 |
mun | +0.099 |
P | +0.096 |
Neg logit contributions | |
horizon | -0.250 |
peak | -0.217 |
heights | -0.214 |
comb | -0.213 |
edly | -0.211 |
Token | said |
Token ID | 531 |
Feature activation | +8.45e-03 |
Pos logit contributions | |
living | +0.028 |
mun | +0.022 |
have | +0.021 |
w | +0.021 |
d | +0.021 |
Neg logit contributions | |
horizon | -0.045 |
heights | -0.039 |
peak | -0.039 |
itself | -0.039 |
edly | -0.038 |
Token | , |
Token ID | 11 |
Feature activation | +6.02e-03 |
Pos logit contributions | |
horizon | +0.013 |
itself | +0.011 |
edly | +0.011 |
peak | +0.011 |
heights | +0.011 |
Neg logit contributions | |
living | -0.018 |
mun | -0.015 |
w | -0.015 |
have | -0.015 |
d | -0.014 |
Token | " |
Token ID | 366 |
Feature activation | +8.52e-03 |
Pos logit contributions | |
horizon | +0.012 |
itself | +0.010 |
peak | +0.010 |
heights | +0.010 |
edly | +0.010 |
Neg logit contributions | |
living | -0.011 |
mun | -0.009 |
w | -0.009 |
have | -0.008 |
d | -0.008 |
Token | Okay |
Token ID | 16454 |
Feature activation | -8.69e-03 |
Pos logit contributions | |
horizon | +0.019 |
peak | +0.017 |
itself | +0.017 |
heights | +0.017 |
comb | +0.016 |
Neg logit contributions | |
living | -0.012 |
mun | -0.010 |
have | -0.010 |
w | -0.010 |
d | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -1.87e-03 |
Pos logit contributions | |
horizon | +0.013 |
heights | +0.011 |
itself | +0.011 |
peak | +0.011 |
fingertips | +0.011 |
Neg logit contributions | |
living | -0.009 |
d | -0.006 |
sm | -0.006 |
w | -0.006 |
mun | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | -9.20e-03 |
Pos logit contributions | |
living | +0.003 |
mun | +0.002 |
Good | +0.002 |
Did | +0.002 |
d | +0.002 |
Neg logit contributions | |
horizon | -0.004 |
peak | -0.004 |
comb | -0.004 |
heights | -0.004 |
itself | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -1.94e-03 |
Pos logit contributions | |
living | +0.015 |
mun | +0.010 |
w | +0.010 |
have | +0.010 |
L | +0.010 |
Neg logit contributions | |
horizon | -0.021 |
heights | -0.019 |
peak | -0.019 |
comb | -0.018 |
edly | -0.018 |
Token | " |
Token ID | 1 |
Feature activation | +5.29e-03 |
Pos logit contributions | |
living | +0.003 |
mun | +0.003 |
haw | +0.003 |
have | +0.003 |
stri | +0.003 |
Neg logit contributions | |
horizon | -0.003 |
edly | -0.003 |
itself | -0.003 |
peak | -0.003 |
comb | -0.003 |
Token | Stop |
Token ID | 19485 |
Feature activation | -3.84e-02 |
Pos logit contributions | |
horizon | +0.011 |
edly | +0.009 |
itself | +0.009 |
heights | +0.009 |
� | +0.008 |
Neg logit contributions | |
living | -0.009 |
mun | -0.007 |
loves | -0.007 |
d | -0.007 |
haw | -0.006 |
Token | fighting |
Token ID | 4330 |
Feature activation | -6.33e-02 |
Pos logit contributions | |
living | +0.082 |
mun | +0.073 |
P | +0.071 |
have | +0.071 |
w | +0.068 |
Neg logit contributions | |
horizon | -0.056 |
heights | -0.051 |
itself | -0.049 |
peak | -0.045 |
ement | -0.044 |
Token | , |
Token ID | 11 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
living | +0.107 |
w | +0.086 |
have | +0.083 |
mun | +0.082 |
d | +0.081 |
Neg logit contributions | |
horizon | -0.133 |
edly | -0.117 |
comb | -0.113 |
peak | -0.112 |
heights | -0.110 |
Token | Anna |
Token ID | 11735 |
Feature activation | -1.80e-03 |
Pos logit contributions | |
living | +0.035 |
mun | +0.027 |
w | +0.027 |
have | +0.027 |
d | +0.027 |
Neg logit contributions | |
horizon | -0.042 |
edly | -0.035 |
peak | -0.035 |
itself | -0.035 |
heights | -0.035 |
Token | and |
Token ID | 290 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
living | +0.003 |
w | +0.002 |
d | +0.002 |
mun | +0.002 |
have | +0.002 |
Neg logit contributions | |
horizon | -0.004 |
peak | -0.003 |
edly | -0.003 |
heights | -0.003 |
ement | -0.003 |
Token | Ben |
Token ID | 3932 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
living | +0.023 |
mun | +0.019 |
w | +0.018 |
have | +0.018 |
d | +0.017 |
Neg logit contributions | |
horizon | -0.025 |
itself | -0.021 |
peak | -0.021 |
heights | -0.021 |
edly | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
living | +0.031 |
mun | +0.026 |
have | +0.025 |
sm | +0.024 |
Did | +0.024 |
Neg logit contributions | |
horizon | -0.024 |
itself | -0.021 |
heights | -0.021 |
peak | -0.020 |
comb | -0.019 |
Token | " |
Token ID | 366 |
Feature activation | +6.61e-04 |
Pos logit contributions | |
living | +0.046 |
w | +0.042 |
mun | +0.042 |
have | +0.041 |
haw | +0.038 |
Neg logit contributions | |
horizon | -0.040 |
itself | -0.034 |
� | -0.032 |
edly | -0.032 |
peak | -0.032 |
Token | Sorry |
Token ID | 14385 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
horizon | +0.002 |
edly | +0.001 |
itself | +0.001 |
peak | +0.001 |
heights | +0.001 |
Neg logit contributions | |
living | -0.001 |
mun | -0.001 |
have | -0.001 |
w | -0.001 |
d | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
living | +0.042 |
mun | +0.037 |
w | +0.037 |
have | +0.036 |
sm | +0.035 |
Neg logit contributions | |
horizon | -0.030 |
itself | -0.025 |
heights | -0.025 |
peak | -0.024 |
ement | -0.023 |
Token | Mom |
Token ID | 11254 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
living | +0.027 |
mun | +0.024 |
w | +0.023 |
have | +0.023 |
sm | +0.022 |
Neg logit contributions | |
horizon | -0.019 |
itself | -0.015 |
peak | -0.015 |
heights | -0.015 |
comb | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
living | +0.038 |
mun | +0.033 |
have | +0.032 |
w | +0.032 |
haw | +0.031 |
Neg logit contributions | |
horizon | -0.029 |
itself | -0.024 |
heights | -0.024 |
peak | -0.023 |
comb | -0.023 |
Token | Sorry |
Token ID | 19061 |
Feature activation | -6.30e-02 |
Pos logit contributions | |
living | +0.016 |
mun | +0.013 |
w | +0.013 |
have | +0.012 |
d | +0.012 |
Neg logit contributions | |
horizon | -0.024 |
itself | -0.021 |
peak | -0.021 |
heights | -0.021 |
edly | -0.020 |
Token | , |
Token ID | 11 |
Feature activation | -3.53e-02 |
Pos logit contributions | |
living | +0.146 |
mun | +0.140 |
have | +0.134 |
w | +0.133 |
P | +0.131 |
Neg logit contributions | |
horizon | -0.070 |
itself | -0.062 |
peak | -0.057 |
destination | -0.056 |
comb | -0.056 |
Token | Tim |
Token ID | 5045 |
Feature activation | -2.91e-02 |
Pos logit contributions | |
living | +0.070 |
mun | +0.057 |
have | +0.056 |
w | +0.056 |
d | +0.055 |
Neg logit contributions | |
horizon | -0.064 |
itself | -0.053 |
peak | -0.053 |
edly | -0.053 |
heights | -0.053 |
Token | . |
Token ID | 13 |
Feature activation | -2.60e-02 |
Pos logit contributions | |
living | +0.054 |
w | +0.043 |
mun | +0.043 |
have | +0.043 |
d | +0.042 |
Neg logit contributions | |
horizon | -0.056 |
peak | -0.048 |
heights | -0.048 |
itself | -0.047 |
edly | -0.047 |
Token | Sorry |
Token ID | 19061 |
Feature activation | -6.68e-02 |
Pos logit contributions | |
living | +0.037 |
mun | +0.029 |
have | +0.028 |
w | +0.028 |
d | +0.026 |
Neg logit contributions | |
horizon | -0.060 |
itself | -0.052 |
peak | -0.052 |
heights | -0.052 |
edly | -0.051 |
Token | , |
Token ID | 11 |
Feature activation | -4.03e-02 |
Pos logit contributions | |
living | +0.148 |
mun | +0.142 |
have | +0.135 |
w | +0.133 |
P | +0.131 |
Neg logit contributions | |
horizon | -0.083 |
itself | -0.073 |
peak | -0.068 |
comb | -0.066 |
destination | -0.065 |