Token | jumped |
Token ID | 11687 |
Feature activation | -5.34e-01 |
Pos logit contributions | |
block | +0.021 |
most | +0.020 |
bucket | +0.019 |
palm | +0.019 |
foot | +0.018 |
Neg logit contributions | |
ting | -0.024 |
ils | -0.023 |
Midnight | -0.023 |
obe | -0.022 |
rewarded | -0.021 |
Token | onto |
Token ID | 4291 |
Feature activation | +5.76e-01 |
Pos logit contributions | |
ils | +0.012 |
ting | +0.011 |
rewarded | +0.011 |
Midnight | +0.010 |
obe | +0.010 |
Neg logit contributions | |
block | -0.019 |
bucket | -0.018 |
most | -0.018 |
ladder | -0.018 |
railing | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
most | +0.025 |
adder | +0.025 |
path | +0.024 |
block | +0.024 |
bucket | +0.024 |
Neg logit contributions | |
ting | -0.007 |
Midnight | -0.006 |
from | -0.005 |
ils | -0.005 |
rewarded | -0.004 |
Token | zip |
Token ID | 19974 |
Feature activation | +3.20e+00 |
Pos logit contributions | |
adder | +0.004 |
most | +0.004 |
block | +0.004 |
brick | +0.004 |
bucket | +0.004 |
Neg logit contributions | |
ting | -0.007 |
ils | -0.007 |
rewarded | -0.007 |
Midnight | -0.007 |
punished | -0.006 |
Token | - |
Token ID | 12 |
Feature activation | +2.40e+00 |
Pos logit contributions | |
block | +0.109 |
most | +0.108 |
bucket | +0.106 |
brick | +0.104 |
ladder | +0.104 |
Neg logit contributions | |
ting | -0.069 |
ils | -0.067 |
rewarded | -0.065 |
Midnight | -0.061 |
boys | -0.058 |
Token | line |
Token ID | 1370 |
Feature activation | +6.02e+00 |
Pos logit contributions | |
bucket | +0.071 |
most | +0.069 |
ladder | +0.069 |
key | +0.068 |
brick | +0.068 |
Neg logit contributions | |
ting | -0.063 |
ils | -0.059 |
rewarded | -0.054 |
minded | -0.053 |
ups | -0.052 |
Token | and |
Token ID | 290 |
Feature activation | +9.00e-02 |
Pos logit contributions | |
block | +0.210 |
most | +0.209 |
bucket | +0.205 |
brick | +0.202 |
ladder | +0.202 |
Neg logit contributions | |
ting | -0.122 |
ils | -0.118 |
rewarded | -0.114 |
Midnight | -0.106 |
boys | -0.101 |
Token | z |
Token ID | 1976 |
Feature activation | +5.58e-01 |
Pos logit contributions | |
block | +0.002 |
bucket | +0.002 |
shovel | +0.002 |
ladder | +0.002 |
palm | +0.002 |
Neg logit contributions | |
obe | -0.003 |
ils | -0.003 |
ior | -0.003 |
respected | -0.003 |
Mr | -0.003 |
Token | ipped |
Token ID | 3949 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
brick | +0.019 |
ladder | +0.018 |
palm | +0.018 |
tactic | +0.017 |
most | +0.017 |
Neg logit contributions | |
ils | -0.013 |
ting | -0.013 |
rewarded | -0.012 |
ating | -0.011 |
punished | -0.010 |
Token | around |
Token ID | 1088 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
block | +0.007 |
most | +0.007 |
bucket | +0.007 |
ladder | +0.007 |
brick | +0.007 |
Neg logit contributions | |
ils | -0.005 |
ting | -0.005 |
rewarded | -0.005 |
Midnight | -0.005 |
Claus | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | +5.62e-01 |
Pos logit contributions | |
block | +0.044 |
most | +0.044 |
bucket | +0.043 |
ladder | +0.042 |
brick | +0.042 |
Neg logit contributions | |
ils | -0.027 |
rewarded | -0.027 |
ting | -0.027 |
Midnight | -0.025 |
punished | -0.024 |
Token | back |
Token ID | 736 |
Feature activation | +5.24e-01 |
Pos logit contributions | |
ils | +0.023 |
ior | +0.023 |
obe | +0.023 |
rewarded | +0.023 |
respected | +0.023 |
Neg logit contributions | |
block | -0.027 |
bucket | -0.027 |
ladder | -0.026 |
second | -0.025 |
most | -0.025 |
Token | on |
Token ID | 319 |
Feature activation | -8.87e-02 |
Pos logit contributions | |
block | +0.016 |
most | +0.016 |
ladder | +0.016 |
bucket | +0.016 |
brick | +0.015 |
Neg logit contributions | |
ils | -0.014 |
rewarded | -0.013 |
ting | -0.013 |
Midnight | -0.013 |
punished | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
ils | +0.003 |
rewarded | +0.003 |
ting | +0.003 |
Claus | +0.002 |
punished | +0.002 |
Neg logit contributions | |
block | -0.002 |
most | -0.002 |
brick | -0.002 |
ladder | -0.002 |
bucket | -0.002 |
Token | zip |
Token ID | 19974 |
Feature activation | +3.09e+00 |
Pos logit contributions | |
ladder | +0.003 |
palm | +0.003 |
block | +0.003 |
bucket | +0.003 |
brick | +0.003 |
Neg logit contributions | |
rewarded | -0.005 |
ting | -0.005 |
ils | -0.005 |
respected | -0.005 |
punished | -0.005 |
Token | - |
Token ID | 12 |
Feature activation | +2.46e+00 |
Pos logit contributions | |
block | +0.101 |
most | +0.098 |
ladder | +0.097 |
bucket | +0.096 |
brick | +0.094 |
Neg logit contributions | |
ils | -0.076 |
rewarded | -0.074 |
ting | -0.073 |
Midnight | -0.072 |
obe | -0.069 |
Token | line |
Token ID | 1370 |
Feature activation | +5.91e+00 |
Pos logit contributions | |
block | +0.065 |
most | +0.065 |
brick | +0.061 |
adder | +0.061 |
bucket | +0.061 |
Neg logit contributions | |
ting | -0.071 |
ils | -0.070 |
rewarded | -0.069 |
Midnight | -0.066 |
boys | -0.065 |
Token | , |
Token ID | 11 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
block | +0.192 |
most | +0.187 |
ladder | +0.184 |
brick | +0.181 |
bucket | +0.181 |
Neg logit contributions | |
ils | -0.153 |
ting | -0.144 |
Midnight | -0.141 |
rewarded | -0.135 |
ior | -0.131 |
Token | z |
Token ID | 1976 |
Feature activation | +8.68e-01 |
Pos logit contributions | |
ils | +0.003 |
ting | +0.003 |
obe | +0.003 |
boys | +0.003 |
rewarded | +0.003 |
Neg logit contributions | |
block | -0.003 |
bucket | -0.003 |
ladder | -0.003 |
palm | -0.003 |
most | -0.003 |
Token | ipping |
Token ID | 4501 |
Feature activation | -9.34e-02 |
Pos logit contributions | |
block | +0.020 |
most | +0.020 |
bucket | +0.019 |
adder | +0.019 |
brick | +0.019 |
Neg logit contributions | |
ting | -0.028 |
ils | -0.027 |
rewarded | -0.027 |
Midnight | -0.026 |
boys | -0.025 |
Token | all |
Token ID | 477 |
Feature activation | +2.98e+00 |
Pos logit contributions | |
Claus | +0.003 |
obe | +0.003 |
boys | +0.003 |
rewarded | +0.003 |
punished | +0.003 |
Neg logit contributions | |
block | -0.003 |
bucket | -0.002 |
most | -0.002 |
palm | -0.002 |
rope | -0.002 |
Token | around |
Token ID | 1088 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
block | +0.099 |
most | +0.096 |
brick | +0.094 |
bucket | +0.094 |
ladder | +0.092 |
Neg logit contributions | |
ils | -0.068 |
ting | -0.067 |
Midnight | -0.064 |
rewarded | -0.064 |
boys | -0.063 |
Token | stick |
Token ID | 4859 |
Feature activation | -5.56e-01 |
Pos logit contributions | |
block | +0.017 |
bucket | +0.016 |
foot | +0.015 |
nail | +0.014 |
brick | +0.014 |
Neg logit contributions | |
ting | -0.019 |
ils | -0.018 |
yer | -0.017 |
lish | -0.017 |
rewarded | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
ting | +0.014 |
ils | +0.014 |
rewarded | +0.013 |
Claus | +0.012 |
obe | +0.012 |
Neg logit contributions | |
block | -0.018 |
bucket | -0.017 |
most | -0.017 |
ladder | -0.017 |
brick | -0.017 |
Token | mustache |
Token ID | 49303 |
Feature activation | -6.64e-01 |
Pos logit contributions | |
block | +0.004 |
bucket | +0.003 |
ladder | +0.003 |
brick | +0.003 |
most | +0.003 |
Neg logit contributions | |
ting | -0.009 |
ils | -0.009 |
rewarded | -0.009 |
obe | -0.008 |
Midnight | -0.008 |
Token | on |
Token ID | 319 |
Feature activation | +1.98e+00 |
Pos logit contributions | |
ting | +0.017 |
ils | +0.017 |
rewarded | +0.017 |
Midnight | +0.016 |
obe | +0.015 |
Neg logit contributions | |
block | -0.022 |
most | -0.020 |
ladder | -0.020 |
bucket | -0.020 |
foot | -0.019 |
Token | her |
Token ID | 607 |
Feature activation | +7.12e-01 |
Pos logit contributions | |
block | +0.058 |
brick | +0.054 |
nail | +0.053 |
most | +0.053 |
bucket | +0.052 |
Neg logit contributions | |
rewarded | -0.057 |
ils | -0.057 |
ting | -0.057 |
Claus | -0.053 |
punished | -0.052 |
Token | upper |
Token ID | 6727 |
Feature activation | +5.89e+00 |
Pos logit contributions | |
block | +0.015 |
most | +0.015 |
bucket | +0.014 |
brick | +0.014 |
adder | +0.014 |
Neg logit contributions | |
ting | -0.024 |
ils | -0.024 |
rewarded | -0.024 |
Midnight | -0.023 |
boys | -0.022 |
Token | lip |
Token ID | 10645 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
ladder | +0.130 |
bucket | +0.130 |
railing | +0.128 |
most | +0.126 |
brick | +0.125 |
Neg logit contributions | |
ting | -0.171 |
obe | -0.169 |
ior | -0.161 |
rewarded | -0.161 |
respected | -0.160 |
Token | with |
Token ID | 351 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
block | +0.043 |
foot | +0.041 |
most | +0.041 |
ladder | +0.040 |
bucket | +0.039 |
Neg logit contributions | |
ting | -0.033 |
ils | -0.033 |
rewarded | -0.032 |
minded | -0.029 |
Midnight | -0.028 |
Token | some |
Token ID | 617 |
Feature activation | +4.26e-01 |
Pos logit contributions | |
block | +0.007 |
bucket | +0.007 |
most | +0.007 |
brick | +0.007 |
palm | +0.007 |
Neg logit contributions | |
ting | -0.009 |
ils | -0.009 |
rewarded | -0.009 |
Claus | -0.008 |
Midnight | -0.008 |
Token | tape |
Token ID | 9154 |
Feature activation | -7.85e-01 |
Pos logit contributions | |
block | +0.010 |
bucket | +0.010 |
ladder | +0.009 |
brick | +0.009 |
palm | +0.009 |
Neg logit contributions | |
ting | -0.015 |
ils | -0.014 |
Claus | -0.014 |
rewarded | -0.014 |
Midnight | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
ting | +0.022 |
ils | +0.021 |
rewarded | +0.020 |
obe | +0.020 |
Mr | +0.020 |
Neg logit contributions | |
block | -0.025 |
bucket | -0.024 |
ladder | -0.023 |
foot | -0.022 |
most | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
ting | +0.024 |
ils | +0.023 |
rewarded | +0.023 |
Midnight | +0.021 |
obe | +0.021 |
Neg logit contributions | |
block | -0.033 |
bucket | -0.032 |
most | -0.031 |
brick | -0.031 |
ladder | -0.031 |
Token | camera |
Token ID | 4676 |
Feature activation | -8.77e-02 |
Pos logit contributions | |
ting | +0.006 |
ils | +0.006 |
rewarded | +0.006 |
Midnight | +0.006 |
boys | +0.005 |
Neg logit contributions | |
most | -0.004 |
block | -0.004 |
bucket | -0.004 |
adder | -0.004 |
brick | -0.004 |
Token | back |
Token ID | 736 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
ting | +0.002 |
ils | +0.002 |
rewarded | +0.002 |
Midnight | +0.002 |
obe | +0.002 |
Neg logit contributions | |
block | -0.003 |
bucket | -0.003 |
key | -0.003 |
most | -0.003 |
ladder | -0.003 |
Token | on |
Token ID | 319 |
Feature activation | +8.07e-01 |
Pos logit contributions | |
block | +0.057 |
most | +0.052 |
bucket | +0.051 |
ladder | +0.051 |
brick | +0.050 |
Neg logit contributions | |
rewarded | -0.041 |
ting | -0.040 |
Midnight | -0.039 |
ils | -0.039 |
obe | -0.038 |
Token | the |
Token ID | 262 |
Feature activation | +3.46e-01 |
Pos logit contributions | |
block | +0.027 |
most | +0.025 |
bucket | +0.025 |
brick | +0.025 |
palm | +0.024 |
Neg logit contributions | |
ils | -0.020 |
ting | -0.020 |
rewarded | -0.020 |
obe | -0.018 |
punished | -0.017 |
Token | tripod |
Token ID | 47439 |
Feature activation | +5.81e+00 |
Pos logit contributions | |
block | +0.007 |
bucket | +0.007 |
ladder | +0.007 |
railing | +0.007 |
palm | +0.007 |
Neg logit contributions | |
rewarded | -0.013 |
ting | -0.013 |
ils | -0.012 |
treated | -0.012 |
Midnight | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -5.06e-01 |
Pos logit contributions | |
block | +0.212 |
most | +0.210 |
bucket | +0.205 |
brick | +0.202 |
ladder | +0.201 |
Neg logit contributions | |
ting | -0.113 |
ils | -0.109 |
rewarded | -0.106 |
Midnight | -0.099 |
punished | -0.093 |
Token |
|
Token ID | 198 |
Feature activation | -5.26e-01 |
Pos logit contributions | |
ting | +0.014 |
ils | +0.013 |
rewarded | +0.013 |
Midnight | +0.013 |
boys | +0.012 |
Neg logit contributions | |
most | -0.014 |
block | -0.014 |
bucket | -0.014 |
brick | -0.013 |
adder | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | -6.62e-01 |
Pos logit contributions | |
ting | +0.013 |
rewarded | +0.013 |
ils | +0.013 |
Midnight | +0.012 |
Mr | +0.012 |
Neg logit contributions | |
most | -0.015 |
bucket | -0.015 |
brick | -0.015 |
block | -0.015 |
adder | -0.015 |
Token | " |
Token ID | 1 |
Feature activation | -3.70e-01 |
Pos logit contributions | |
ting | +0.017 |
ils | +0.017 |
rewarded | +0.016 |
Midnight | +0.015 |
boys | +0.015 |
Neg logit contributions | |
block | -0.020 |
most | -0.020 |
bucket | -0.019 |
brick | -0.019 |
ladder | -0.019 |
Token | Go |
Token ID | 5247 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
ting | +0.009 |
ils | +0.009 |
rewarded | +0.009 |
Midnight | +0.008 |
punished | +0.008 |
Neg logit contributions | |
most | -0.011 |
block | -0.011 |
bucket | -0.011 |
ladder | -0.011 |
brick | -0.011 |
Token | robe |
Token ID | 25481 |
Feature activation | -5.70e-01 |
Pos logit contributions | |
bucket | +0.001 |
block | +0.001 |
foot | +0.001 |
most | +0.001 |
railing | +0.001 |
Neg logit contributions | |
rewarded | -0.002 |
ils | -0.002 |
ting | -0.002 |
ior | -0.002 |
Mr | -0.002 |
Token | back |
Token ID | 736 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
ils | +0.013 |
ting | +0.013 |
rewarded | +0.013 |
Midnight | +0.012 |
Claus | +0.012 |
Neg logit contributions | |
block | -0.019 |
most | -0.018 |
bucket | -0.018 |
ladder | -0.018 |
foot | -0.018 |
Token | onto |
Token ID | 4291 |
Feature activation | +8.28e-01 |
Pos logit contributions | |
block | +0.050 |
most | +0.049 |
bucket | +0.048 |
ladder | +0.047 |
brick | +0.047 |
Neg logit contributions | |
ils | -0.031 |
ting | -0.030 |
rewarded | -0.030 |
Midnight | -0.028 |
punished | -0.026 |
Token | the |
Token ID | 262 |
Feature activation | +6.32e-01 |
Pos logit contributions | |
block | +0.028 |
most | +0.028 |
bucket | +0.028 |
brick | +0.027 |
palm | +0.027 |
Neg logit contributions | |
ting | -0.018 |
ils | -0.018 |
rewarded | -0.017 |
Midnight | -0.015 |
boys | -0.015 |
Token | clothes |
Token ID | 8242 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
bucket | +0.013 |
foot | +0.012 |
block | +0.012 |
palm | +0.012 |
railing | +0.012 |
Neg logit contributions | |
ils | -0.025 |
rewarded | -0.024 |
ting | -0.023 |
respected | -0.023 |
obe | -0.022 |
Token | line |
Token ID | 1370 |
Feature activation | +5.74e+00 |
Pos logit contributions | |
block | +0.041 |
bucket | +0.040 |
most | +0.039 |
ladder | +0.039 |
foot | +0.039 |
Neg logit contributions | |
ils | -0.033 |
ting | -0.030 |
rewarded | -0.029 |
Midnight | -0.029 |
ior | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
block | +0.193 |
most | +0.185 |
bucket | +0.182 |
foot | +0.181 |
ladder | +0.180 |
Neg logit contributions | |
ils | -0.143 |
ting | -0.134 |
Midnight | -0.130 |
rewarded | -0.129 |
ior | -0.119 |
Token | He |
Token ID | 679 |
Feature activation | -4.17e-01 |
Pos logit contributions | |
block | +0.003 |
most | +0.003 |
bucket | +0.003 |
ladder | +0.003 |
brick | +0.003 |
Neg logit contributions | |
ting | -0.003 |
ils | -0.003 |
rewarded | -0.003 |
Midnight | -0.003 |
boys | -0.003 |
Token | used |
Token ID | 973 |
Feature activation | -7.05e-01 |
Pos logit contributions | |
ting | +0.013 |
ils | +0.012 |
rewarded | +0.012 |
obe | +0.011 |
Midnight | +0.011 |
Neg logit contributions | |
block | -0.011 |
bucket | -0.011 |
most | -0.011 |
ladder | -0.010 |
palm | -0.010 |
Token | all |
Token ID | 477 |
Feature activation | +2.65e+00 |
Pos logit contributions | |
ting | +0.013 |
rewarded | +0.012 |
ils | +0.012 |
Midnight | +0.012 |
from | +0.011 |
Neg logit contributions | |
most | -0.026 |
adder | -0.025 |
railing | -0.024 |
ladder | -0.024 |
block | -0.024 |
Token | his |
Token ID | 465 |
Feature activation | +2.56e-01 |
Pos logit contributions | |
most | +0.100 |
adder | +0.099 |
key | +0.099 |
railing | +0.099 |
ladder | +0.099 |
Neg logit contributions | |
ting | -0.042 |
ils | -0.037 |
from | -0.037 |
rewarded | -0.035 |
thanks | -0.033 |
Token | I |
Token ID | 40 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
most | +0.008 |
block | +0.008 |
bucket | +0.008 |
ladder | +0.008 |
brick | +0.007 |
Neg logit contributions | |
ting | -0.005 |
ils | -0.004 |
rewarded | -0.004 |
Midnight | -0.004 |
Mr | -0.004 |
Token | fell |
Token ID | 3214 |
Feature activation | -8.88e-01 |
Pos logit contributions | |
block | +0.027 |
most | +0.026 |
bucket | +0.026 |
ladder | +0.025 |
brick | +0.025 |
Neg logit contributions | |
ting | -0.044 |
ils | -0.043 |
rewarded | -0.042 |
Midnight | -0.040 |
obe | -0.039 |
Token | off |
Token ID | 572 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
ils | +0.027 |
rewarded | +0.024 |
yer | +0.022 |
thel | +0.022 |
boys | +0.022 |
Neg logit contributions | |
block | -0.027 |
foot | -0.027 |
bucket | -0.026 |
railing | -0.025 |
most | -0.025 |
Token | the |
Token ID | 262 |
Feature activation | +3.15e-02 |
Pos logit contributions | |
ils | +0.034 |
obe | +0.033 |
boys | +0.032 |
Claus | +0.031 |
respected | +0.031 |
Neg logit contributions | |
block | -0.032 |
bucket | -0.030 |
foot | -0.030 |
ladder | -0.029 |
most | -0.029 |
Token | clothes |
Token ID | 8242 |
Feature activation | +8.88e-01 |
Pos logit contributions | |
bucket | +0.001 |
block | +0.001 |
ladder | +0.001 |
foot | +0.001 |
railing | +0.001 |
Neg logit contributions | |
ils | -0.001 |
rewarded | -0.001 |
ting | -0.001 |
respected | -0.001 |
obe | -0.001 |
Token | line |
Token ID | 1370 |
Feature activation | +5.45e+00 |
Pos logit contributions | |
block | +0.027 |
bucket | +0.026 |
ladder | +0.025 |
foot | +0.025 |
railing | +0.025 |
Neg logit contributions | |
ils | -0.028 |
ior | -0.024 |
minded | -0.023 |
obe | -0.023 |
Midnight | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | -8.81e-02 |
Pos logit contributions | |
block | +0.168 |
ladder | +0.154 |
bucket | +0.152 |
foot | +0.152 |
most | +0.151 |
Neg logit contributions | |
ils | -0.168 |
ting | -0.152 |
Midnight | -0.151 |
rewarded | -0.144 |
obe | -0.144 |
Token | I |
Token ID | 314 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
ils | +0.003 |
obe | +0.003 |
ting | +0.003 |
ior | +0.003 |
yer | +0.003 |
Neg logit contributions | |
bucket | -0.002 |
block | -0.002 |
shovel | -0.002 |
foot | -0.002 |
ladder | -0.002 |
Token | can |
Token ID | 460 |
Feature activation | -9.66e-01 |
Pos logit contributions | |
block | +0.030 |
bucket | +0.029 |
most | +0.029 |
ladder | +0.027 |
brick | +0.027 |
Neg logit contributions | |
ting | -0.042 |
ils | -0.040 |
rewarded | -0.039 |
Midnight | -0.038 |
boys | -0.037 |
Token | 't |
Token ID | 470 |
Feature activation | -4.68e-02 |
Pos logit contributions | |
ting | +0.030 |
ils | +0.030 |
rewarded | +0.029 |
obe | +0.027 |
Midnight | +0.027 |
Neg logit contributions | |
block | -0.024 |
bucket | -0.024 |
most | -0.023 |
ladder | -0.023 |
foot | -0.023 |
Token | get |
Token ID | 651 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
ting | +0.001 |
ils | +0.001 |
rewarded | +0.001 |
Midnight | +0.001 |
boys | +0.001 |
Neg logit contributions | |
block | -0.001 |
most | -0.001 |
bucket | -0.001 |
brick | -0.001 |
ladder | -0.001 |
Token | was |
Token ID | 373 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
block | +0.018 |
bucket | +0.017 |
most | +0.017 |
ladder | +0.017 |
brick | +0.017 |
Neg logit contributions | |
ting | -0.020 |
Midnight | -0.018 |
obe | -0.018 |
ils | -0.018 |
Mr | -0.018 |
Token | always |
Token ID | 1464 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
block | +0.040 |
bucket | +0.037 |
palm | +0.037 |
second | +0.037 |
most | +0.037 |
Neg logit contributions | |
ting | -0.048 |
ils | -0.045 |
obe | -0.044 |
ior | -0.043 |
Midnight | -0.042 |
Token | there |
Token ID | 612 |
Feature activation | +1.98e+00 |
Pos logit contributions | |
ting | +0.043 |
ils | +0.042 |
obe | +0.041 |
Midnight | +0.041 |
ior | +0.040 |
Neg logit contributions | |
bucket | -0.029 |
block | -0.028 |
key | -0.027 |
second | -0.026 |
brick | -0.026 |
Token | teaching |
Token ID | 7743 |
Feature activation | +4.90e-01 |
Pos logit contributions | |
key | +0.053 |
most | +0.052 |
block | +0.052 |
bucket | +0.052 |
palm | +0.051 |
Neg logit contributions | |
ting | -0.057 |
ils | -0.056 |
ior | -0.055 |
Midnight | -0.054 |
Mr | -0.054 |
Token | the |
Token ID | 262 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
second | +0.011 |
block | +0.011 |
most | +0.011 |
palm | +0.011 |
key | +0.011 |
Neg logit contributions | |
Claus | -0.015 |
ting | -0.014 |
rewarded | -0.014 |
ior | -0.014 |
ils | -0.014 |
Token | 3 |
Token ID | 513 |
Feature activation | +5.44e+00 |
Pos logit contributions | |
most | +0.029 |
block | +0.028 |
bucket | +0.027 |
adder | +0.027 |
brick | +0.027 |
Neg logit contributions | |
ting | -0.044 |
ils | -0.043 |
rewarded | -0.042 |
Midnight | -0.041 |
boys | -0.039 |
Token | - |
Token ID | 12 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
most | +0.138 |
bucket | +0.134 |
block | +0.131 |
adder | +0.131 |
railing | +0.129 |
Neg logit contributions | |
ils | -0.166 |
rewarded | -0.156 |
ting | -0.155 |
treated | -0.143 |
Midnight | -0.140 |
Token | year |
Token ID | 1941 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
block | +0.072 |
most | +0.071 |
adder | +0.069 |
brick | +0.068 |
palm | +0.067 |
Neg logit contributions | |
ting | -0.074 |
ils | -0.070 |
rewarded | -0.070 |
Midnight | -0.069 |
boys | -0.066 |
Token | - |
Token ID | 12 |
Feature activation | +2.98e+00 |
Pos logit contributions | |
block | +0.056 |
most | +0.056 |
bucket | +0.055 |
ladder | +0.054 |
brick | +0.054 |
Neg logit contributions | |
ting | -0.065 |
ils | -0.063 |
rewarded | -0.062 |
Midnight | -0.060 |
boys | -0.058 |
Token | old |
Token ID | 727 |
Feature activation | +7.92e-01 |
Pos logit contributions | |
block | +0.088 |
most | +0.088 |
adder | +0.085 |
brick | +0.084 |
palm | +0.083 |
Neg logit contributions | |
ting | -0.077 |
ils | -0.073 |
rewarded | -0.072 |
Midnight | -0.071 |
boys | -0.067 |
Token | girl |
Token ID | 2576 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
block | +0.020 |
second | +0.020 |
palm | +0.020 |
foot | +0.020 |
hand | +0.019 |
Neg logit contributions | |
Mr | -0.022 |
ting | -0.022 |
ils | -0.022 |
Midnight | -0.021 |
Claus | -0.021 |
Token | he |
Token ID | 339 |
Feature activation | -7.62e-01 |
Pos logit contributions | |
ting | +0.003 |
ils | +0.003 |
Claus | +0.003 |
ior | +0.003 |
obe | +0.003 |
Neg logit contributions | |
block | -0.003 |
key | -0.003 |
brick | -0.003 |
ladder | -0.003 |
bucket | -0.003 |
Token | had |
Token ID | 550 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
obe | +0.028 |
ting | +0.028 |
ior | +0.026 |
respected | +0.026 |
boys | +0.026 |
Neg logit contributions | |
key | -0.015 |
block | -0.014 |
foot | -0.013 |
palm | -0.013 |
second | -0.013 |
Token | used |
Token ID | 973 |
Feature activation | -6.42e-01 |
Pos logit contributions | |
block | +0.030 |
bucket | +0.029 |
key | +0.029 |
foot | +0.028 |
hand | +0.027 |
Neg logit contributions | |
ting | -0.041 |
ior | -0.038 |
Claus | -0.037 |
Midnight | -0.037 |
boys | -0.037 |
Token | the |
Token ID | 262 |
Feature activation | +2.67e-01 |
Pos logit contributions | |
ting | +0.019 |
ils | +0.018 |
Claus | +0.017 |
rewarded | +0.017 |
boys | +0.017 |
Neg logit contributions | |
block | -0.018 |
bucket | -0.017 |
brick | -0.017 |
most | -0.017 |
key | -0.017 |
Token | squee |
Token ID | 13170 |
Feature activation | -9.67e-01 |
Pos logit contributions | |
block | +0.006 |
bucket | +0.006 |
ladder | +0.006 |
brick | +0.006 |
most | +0.005 |
Neg logit contributions | |
ting | -0.009 |
ils | -0.009 |
rewarded | -0.009 |
Midnight | -0.008 |
boys | -0.008 |
Token | gee |
Token ID | 29622 |
Feature activation | +5.38e+00 |
Pos logit contributions | |
ting | +0.029 |
rewarded | +0.028 |
ils | +0.027 |
Midnight | +0.026 |
Claus | +0.026 |
Neg logit contributions | |
block | -0.026 |
bucket | -0.024 |
most | -0.024 |
foot | -0.024 |
ladder | -0.023 |
Token | to |
Token ID | 284 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
block | +0.157 |
most | +0.145 |
foot | +0.144 |
ladder | +0.144 |
brick | +0.144 |
Neg logit contributions | |
ils | -0.159 |
rewarded | -0.156 |
ting | -0.152 |
Midnight | -0.142 |
respected | -0.140 |
Token | good |
Token ID | 922 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
block | +0.031 |
nail | +0.029 |
foot | +0.029 |
bucket | +0.029 |
key | +0.028 |
Neg logit contributions | |
ils | -0.041 |
ting | -0.041 |
minded | -0.038 |
Claus | -0.038 |
rewarded | -0.037 |
Token | use |
Token ID | 779 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
block | +0.005 |
foot | +0.004 |
bucket | +0.004 |
ladder | +0.004 |
key | +0.004 |
Neg logit contributions | |
Claus | -0.006 |
ils | -0.006 |
ting | -0.006 |
Midnight | -0.006 |
Mr | -0.006 |
Token | ! |
Token ID | 0 |
Feature activation | +8.73e-01 |
Pos logit contributions | |
Claus | +0.028 |
ils | +0.028 |
ting | +0.028 |
boys | +0.027 |
rewarded | +0.027 |
Neg logit contributions | |
block | -0.031 |
bucket | -0.029 |
foot | -0.029 |
ladder | -0.029 |
nail | -0.028 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -7.09e-01 |
Pos logit contributions | |
most | +0.020 |
block | +0.020 |
brick | +0.018 |
ladder | +0.018 |
bucket | +0.018 |
Neg logit contributions | |
ils | -0.030 |
ting | -0.029 |
rewarded | -0.028 |
boys | -0.027 |
punished | -0.026 |
Token | as |
Token ID | 355 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
ils | +0.013 |
ting | +0.013 |
Midnight | +0.013 |
obe | +0.013 |
Mr | +0.013 |
Neg logit contributions | |
block | -0.015 |
bucket | -0.015 |
foot | -0.015 |
ladder | -0.014 |
palm | -0.014 |
Token | he |
Token ID | 339 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
block | +0.007 |
most | +0.007 |
brick | +0.007 |
ladder | +0.007 |
bucket | +0.007 |
Neg logit contributions | |
ils | -0.008 |
ting | -0.008 |
rewarded | -0.008 |
obe | -0.007 |
Midnight | -0.007 |
Token | explored |
Token ID | 18782 |
Feature activation | +3.76e-02 |
Pos logit contributions | |
ting | +0.037 |
ils | +0.035 |
rewarded | +0.035 |
Midnight | +0.033 |
boys | +0.031 |
Neg logit contributions | |
block | -0.046 |
most | -0.045 |
bucket | -0.045 |
brick | -0.044 |
ladder | -0.044 |
Token | the |
Token ID | 262 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
block | +0.001 |
bucket | +0.001 |
palm | +0.001 |
brick | +0.001 |
ladder | +0.001 |
Neg logit contributions | |
Midnight | -0.001 |
Claus | -0.001 |
ils | -0.001 |
rewarded | -0.001 |
punished | -0.001 |
Token | shore |
Token ID | 15191 |
Feature activation | +2.30e+00 |
Pos logit contributions | |
bucket | +0.003 |
block | +0.003 |
palm | +0.003 |
ladder | +0.002 |
railing | +0.002 |
Neg logit contributions | |
Midnight | -0.006 |
ting | -0.006 |
ils | -0.006 |
rewarded | -0.006 |
obe | -0.006 |
Token | line |
Token ID | 1370 |
Feature activation | +5.35e+00 |
Pos logit contributions | |
block | +0.072 |
bucket | +0.069 |
most | +0.068 |
palm | +0.066 |
ladder | +0.065 |
Neg logit contributions | |
Midnight | -0.069 |
punished | -0.062 |
ils | -0.062 |
obe | -0.059 |
rewarded | -0.059 |
Token | . |
Token ID | 13 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
block | +0.179 |
most | +0.174 |
bucket | +0.170 |
ladder | +0.168 |
brick | +0.168 |
Neg logit contributions | |
ils | -0.131 |
Midnight | -0.128 |
ting | -0.122 |
rewarded | -0.119 |
punished | -0.116 |
Token |
|
Token ID | 198 |
Feature activation | -5.89e-02 |
Pos logit contributions | |
chie | +0.001 |
key | +0.001 |
most | +0.001 |
railing | +0.001 |
adder | +0.001 |
Neg logit contributions | |
Midnight | -0.001 |
ting | -0.001 |
rewarded | -0.001 |
Rudolph | -0.001 |
ils | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -3.77e-01 |
Pos logit contributions | |
Mr | +0.001 |
Midnight | +0.001 |
Rudolph | +0.001 |
ting | +0.001 |
rewarded | +0.001 |
Neg logit contributions | |
chie | -0.002 |
railing | -0.002 |
bucket | -0.002 |
key | -0.002 |
shovel | -0.002 |
Token | One |
Token ID | 3198 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
ils | +0.011 |
ting | +0.011 |
rewarded | +0.011 |
Mr | +0.011 |
Midnight | +0.011 |
Neg logit contributions | |
ladder | -0.009 |
most | -0.009 |
bucket | -0.009 |
adder | -0.009 |
palm | -0.009 |
Token | day |
Token ID | 1110 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
most | +0.029 |
block | +0.029 |
adder | +0.028 |
ladder | +0.028 |
bucket | +0.028 |
Neg logit contributions | |
ting | -0.030 |
ils | -0.029 |
rewarded | -0.028 |
Midnight | -0.026 |
treated | -0.026 |
Token | was |
Token ID | 373 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
block | +0.044 |
most | +0.042 |
palm | +0.042 |
bucket | +0.041 |
ladder | +0.041 |
Neg logit contributions | |
ting | -0.061 |
ils | -0.061 |
rewarded | -0.059 |
minded | -0.058 |
ior | -0.057 |
Token | something |
Token ID | 1223 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
block | +0.039 |
most | +0.038 |
bucket | +0.037 |
brick | +0.037 |
palm | +0.037 |
Neg logit contributions | |
ting | -0.060 |
ils | -0.060 |
rewarded | -0.057 |
obe | -0.054 |
Midnight | -0.054 |
Token | written |
Token ID | 3194 |
Feature activation | -5.62e-01 |
Pos logit contributions | |
ting | +0.076 |
ils | +0.075 |
ior | +0.073 |
rewarded | +0.072 |
obe | +0.072 |
Neg logit contributions | |
bucket | -0.046 |
block | -0.045 |
brick | -0.044 |
palm | -0.044 |
nail | -0.044 |
Token | on |
Token ID | 319 |
Feature activation | +2.42e+00 |
Pos logit contributions | |
ils | +0.013 |
obe | +0.012 |
rewarded | +0.012 |
ior | +0.012 |
ting | +0.012 |
Neg logit contributions | |
block | -0.019 |
most | -0.018 |
key | -0.018 |
foot | -0.017 |
bucket | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
block | +0.065 |
palm | +0.063 |
brick | +0.063 |
most | +0.062 |
nail | +0.062 |
Neg logit contributions | |
rewarded | -0.072 |
ils | -0.070 |
ting | -0.070 |
punished | -0.066 |
obe | -0.065 |
Token | pole |
Token ID | 16825 |
Feature activation | +5.31e+00 |
Pos logit contributions | |
most | +0.021 |
adder | +0.020 |
block | +0.019 |
path | +0.018 |
run | +0.018 |
Neg logit contributions | |
ting | -0.078 |
ils | -0.073 |
rewarded | -0.068 |
Claus | -0.068 |
boys | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | -2.70e-01 |
Pos logit contributions | |
block | +0.172 |
most | +0.164 |
brick | +0.159 |
ladder | +0.158 |
bucket | +0.156 |
Neg logit contributions | |
ils | -0.138 |
rewarded | -0.129 |
ior | -0.129 |
Midnight | -0.129 |
minded | -0.128 |
Token | It |
Token ID | 632 |
Feature activation | +6.85e-01 |
Pos logit contributions | |
ting | +0.007 |
ils | +0.007 |
rewarded | +0.007 |
Midnight | +0.007 |
boys | +0.006 |
Neg logit contributions | |
block | -0.008 |
most | -0.007 |
bucket | -0.007 |
adder | -0.007 |
brick | -0.007 |
Token | said |
Token ID | 531 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
block | +0.018 |
most | +0.018 |
bucket | +0.018 |
brick | +0.017 |
ladder | +0.017 |
Neg logit contributions | |
ting | -0.020 |
ils | -0.020 |
rewarded | -0.018 |
Midnight | -0.018 |
obe | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
block | +0.034 |
ladder | +0.034 |
most | +0.034 |
brick | +0.034 |
palm | +0.033 |
Neg logit contributions | |
ils | -0.033 |
ting | -0.032 |
obe | -0.031 |
ior | -0.031 |
rewarded | -0.030 |
Token | " |
Token ID | 366 |
Feature activation | +2.57e-01 |
Pos logit contributions | |
most | +0.003 |
block | +0.003 |
ladder | +0.003 |
brick | +0.003 |
bucket | +0.003 |
Neg logit contributions | |
ils | -0.006 |
rewarded | -0.005 |
ting | -0.005 |
punished | -0.005 |
obe | -0.005 |
Token | when |
Token ID | 618 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
ils | +0.010 |
rewarded | +0.009 |
ting | +0.009 |
obe | +0.009 |
ups | +0.008 |
Neg logit contributions | |
block | -0.008 |
bucket | -0.008 |
brick | -0.008 |
ladder | -0.008 |
most | -0.008 |
Token | it |
Token ID | 340 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
ils | +0.053 |
ting | +0.049 |
rewarded | +0.048 |
obe | +0.048 |
ior | +0.046 |
Neg logit contributions | |
bucket | -0.044 |
ladder | -0.043 |
block | -0.042 |
brick | -0.042 |
palm | -0.041 |
Token | hugged |
Token ID | 41130 |
Feature activation | -5.24e-01 |
Pos logit contributions | |
ting | +0.022 |
rewarded | +0.022 |
ils | +0.020 |
punished | +0.019 |
Claus | +0.018 |
Neg logit contributions | |
adder | -0.034 |
most | -0.034 |
brick | -0.033 |
palm | -0.033 |
block | -0.033 |
Token | the |
Token ID | 262 |
Feature activation | +6.11e-01 |
Pos logit contributions | |
rewarded | +0.018 |
ils | +0.017 |
obe | +0.017 |
punished | +0.017 |
Claus | +0.017 |
Neg logit contributions | |
palm | -0.012 |
block | -0.012 |
brick | -0.012 |
foot | -0.012 |
bucket | -0.012 |
Token | shore |
Token ID | 15191 |
Feature activation | +2.52e+00 |
Pos logit contributions | |
most | +0.012 |
block | +0.012 |
bucket | +0.012 |
brick | +0.012 |
adder | +0.012 |
Neg logit contributions | |
ting | -0.021 |
ils | -0.021 |
rewarded | -0.021 |
Midnight | -0.020 |
boys | -0.019 |
Token | line |
Token ID | 1370 |
Feature activation | +5.31e+00 |
Pos logit contributions | |
ladder | +0.066 |
bucket | +0.066 |
block | +0.064 |
foot | +0.063 |
palm | +0.062 |
Neg logit contributions | |
ils | -0.087 |
obe | -0.085 |
Midnight | -0.080 |
respected | -0.079 |
punished | -0.078 |
Token | again |
Token ID | 757 |
Feature activation | -6.39e-01 |
Pos logit contributions | |
block | +0.159 |
bucket | +0.156 |
brick | +0.155 |
palm | +0.155 |
ladder | +0.153 |
Neg logit contributions | |
ils | -0.154 |
ting | -0.145 |
Midnight | -0.140 |
obe | -0.139 |
rewarded | -0.131 |
Token | , |
Token ID | 11 |
Feature activation | +5.67e-01 |
Pos logit contributions | |
ils | +0.025 |
punished | +0.022 |
ups | +0.022 |
boys | +0.022 |
Midnight | +0.021 |
Neg logit contributions | |
most | -0.013 |
bucket | -0.013 |
foot | -0.013 |
block | -0.013 |
palm | -0.013 |
Token | it |
Token ID | 340 |
Feature activation | -2.26e-01 |
Pos logit contributions | |
ladder | +0.013 |
bucket | +0.013 |
block | +0.013 |
palm | +0.012 |
brick | +0.012 |
Neg logit contributions | |
ils | -0.020 |
rewarded | -0.019 |
obe | -0.019 |
ting | -0.018 |
ior | -0.018 |
Token | noticed |
Token ID | 6810 |
Feature activation | +7.51e-01 |
Pos logit contributions | |
ting | +0.007 |
ils | +0.007 |
obe | +0.006 |
ior | +0.006 |
Midnight | +0.006 |
Neg logit contributions | |
bucket | -0.006 |
block | -0.006 |
most | -0.006 |
foot | -0.006 |
palm | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | +3.77e-02 |
Pos logit contributions | |
block | +0.018 |
brick | +0.018 |
bucket | +0.018 |
palm | +0.017 |
most | +0.017 |
Neg logit contributions | |
ils | -0.024 |
rewarded | -0.023 |
Claus | -0.022 |
ting | -0.022 |
obe | -0.022 |
Token | dust |
Token ID | 8977 |
Feature activation | -2.97e+00 |
Pos logit contributions | |
ting | +0.062 |
obe | +0.061 |
rewarded | +0.060 |
respected | +0.058 |
punished | +0.058 |
Neg logit contributions | |
block | -0.057 |
most | -0.057 |
bucket | -0.056 |
brick | -0.053 |
palm | -0.053 |
Token | onto |
Token ID | 4291 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
ior | +0.083 |
obe | +0.082 |
Mr | +0.080 |
rewarded | +0.076 |
punished | +0.075 |
Neg logit contributions | |
bucket | -0.085 |
block | -0.084 |
rope | -0.083 |
most | -0.082 |
ladder | -0.081 |
Token | the |
Token ID | 262 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
block | +0.049 |
most | +0.048 |
brick | +0.047 |
bucket | +0.047 |
palm | +0.047 |
Neg logit contributions | |
ils | -0.038 |
ting | -0.038 |
rewarded | -0.038 |
obe | -0.035 |
respected | -0.035 |
Token | l |
Token ID | 300 |
Feature activation | +5.03e-01 |
Pos logit contributions | |
block | +0.025 |
bucket | +0.024 |
most | +0.024 |
ladder | +0.023 |
palm | +0.023 |
Neg logit contributions | |
ting | -0.046 |
ils | -0.045 |
rewarded | -0.045 |
Midnight | -0.042 |
obe | -0.042 |
Token | ily |
Token ID | 813 |
Feature activation | +3.61e+00 |
Pos logit contributions | |
brick | +0.014 |
tactic | +0.013 |
ladder | +0.013 |
bucket | +0.013 |
block | +0.013 |
Neg logit contributions | |
ils | -0.014 |
ting | -0.014 |
rewarded | -0.013 |
Midnight | -0.012 |
from | -0.011 |
Token | pad |
Token ID | 14841 |
Feature activation | +5.29e+00 |
Pos logit contributions | |
most | +0.083 |
block | +0.081 |
bucket | +0.077 |
palm | +0.076 |
foot | +0.074 |
Neg logit contributions | |
rewarded | -0.120 |
ting | -0.120 |
punished | -0.118 |
ils | -0.118 |
obe | -0.116 |
Token | and |
Token ID | 290 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
block | +0.194 |
brick | +0.193 |
most | +0.193 |
bucket | +0.193 |
adder | +0.191 |
Neg logit contributions | |
ting | -0.087 |
ils | -0.082 |
rewarded | -0.080 |
boys | -0.070 |
Midnight | -0.070 |
Token | the |
Token ID | 262 |
Feature activation | +3.86e-01 |
Pos logit contributions | |
bucket | +0.031 |
shovel | +0.029 |
palm | +0.029 |
block | +0.029 |
key | +0.028 |
Neg logit contributions | |
obe | -0.030 |
respected | -0.028 |
ils | -0.028 |
yer | -0.027 |
ior | -0.027 |
Token | pond |
Token ID | 16723 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
block | +0.008 |
bucket | +0.008 |
most | +0.007 |
ladder | +0.007 |
palm | +0.007 |
Neg logit contributions | |
ting | -0.014 |
ils | -0.014 |
rewarded | -0.014 |
Midnight | -0.013 |
obe | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -7.03e-03 |
Pos logit contributions | |
bucket | +0.041 |
block | +0.038 |
ladder | +0.037 |
palm | +0.036 |
most | +0.036 |
Neg logit contributions | |
ior | -0.040 |
obe | -0.038 |
Mr | -0.038 |
ting | -0.038 |
Midnight | -0.037 |
Token |
|
Token ID | 198 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
ils | +0.000 |
punished | +0.000 |
ting | +0.000 |
rewarded | +0.000 |
boys | +0.000 |
Neg logit contributions | |
most | -0.000 |
block | -0.000 |
ladder | -0.000 |
brick | -0.000 |
bucket | -0.000 |
Token | Tom |
Token ID | 4186 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
block | +0.025 |
most | +0.025 |
bucket | +0.024 |
brick | +0.024 |
ladder | +0.024 |
Neg logit contributions | |
ils | -0.028 |
ting | -0.028 |
rewarded | -0.027 |
Midnight | -0.025 |
Claus | -0.025 |
Token | said |
Token ID | 531 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
block | +0.041 |
most | +0.040 |
bucket | +0.039 |
brick | +0.038 |
palm | +0.038 |
Neg logit contributions | |
ting | -0.024 |
ils | -0.024 |
rewarded | -0.022 |
Midnight | -0.021 |
obe | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | +5.62e-01 |
Pos logit contributions | |
block | +0.042 |
most | +0.041 |
bucket | +0.041 |
foot | +0.041 |
ladder | +0.041 |
Neg logit contributions | |
ils | -0.038 |
ting | -0.036 |
rewarded | -0.034 |
Midnight | -0.032 |
obe | -0.031 |
Token | showing |
Token ID | 4478 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
block | +0.016 |
most | +0.016 |
bucket | +0.015 |
ladder | +0.015 |
brick | +0.015 |
Neg logit contributions | |
ils | -0.016 |
ting | -0.015 |
rewarded | -0.015 |
Midnight | -0.014 |
obe | -0.013 |
Token | a |
Token ID | 257 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
most | +0.053 |
block | +0.052 |
bucket | +0.051 |
ladder | +0.051 |
brick | +0.050 |
Neg logit contributions | |
ting | -0.036 |
ils | -0.035 |
rewarded | -0.034 |
Midnight | -0.032 |
boys | -0.030 |
Token | red |
Token ID | 2266 |
Feature activation | +5.27e+00 |
Pos logit contributions | |
most | +0.047 |
adder | +0.047 |
path | +0.041 |
chie | +0.040 |
railing | +0.040 |
Neg logit contributions | |
rewarded | -0.054 |
Claus | -0.051 |
ting | -0.049 |
from | -0.049 |
Midnight | -0.049 |
Token | car |
Token ID | 1097 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
adder | +0.111 |
most | +0.110 |
chie | +0.100 |
Int | +0.097 |
bucket | +0.096 |
Neg logit contributions | |
ting | -0.172 |
from | -0.160 |
Midnight | -0.159 |
thank | -0.145 |
ils | -0.145 |
Token | that |
Token ID | 326 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
block | +0.087 |
most | +0.082 |
ladder | +0.081 |
bucket | +0.081 |
brick | +0.081 |
Neg logit contributions | |
ils | -0.051 |
ting | -0.049 |
rewarded | -0.048 |
Midnight | -0.048 |
punished | -0.043 |
Token | made |
Token ID | 925 |
Feature activation | +5.85e-01 |
Pos logit contributions | |
block | +0.048 |
bucket | +0.046 |
most | +0.045 |
brick | +0.045 |
ladder | +0.044 |
Neg logit contributions | |
ting | -0.046 |
ils | -0.044 |
obe | -0.041 |
rewarded | -0.041 |
respected | -0.040 |
Token | noises |
Token ID | 26782 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
block | +0.016 |
bucket | +0.015 |
most | +0.015 |
brick | +0.015 |
foot | +0.015 |
Neg logit contributions | |
ils | -0.017 |
ting | -0.017 |
rewarded | -0.017 |
Midnight | -0.016 |
obe | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
ils | +0.007 |
Midnight | +0.006 |
ior | +0.006 |
rewarded | +0.006 |
obe | +0.006 |
Neg logit contributions | |
block | -0.007 |
foot | -0.007 |
bucket | -0.007 |
most | -0.007 |
second | -0.007 |
Token | touched |
Token ID | 12615 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
block | +0.040 |
most | +0.040 |
bucket | +0.039 |
brick | +0.038 |
ladder | +0.037 |
Neg logit contributions | |
ting | -0.048 |
ils | -0.047 |
rewarded | -0.045 |
Midnight | -0.044 |
boys | -0.042 |
Token | the |
Token ID | 262 |
Feature activation | +8.23e-01 |
Pos logit contributions | |
rewarded | +0.004 |
ils | +0.004 |
Claus | +0.004 |
respected | +0.003 |
obe | +0.003 |
Neg logit contributions | |
block | -0.005 |
palm | -0.004 |
brick | -0.004 |
foot | -0.004 |
nail | -0.004 |
Token | fire |
Token ID | 2046 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
most | +0.013 |
block | +0.013 |
adder | +0.012 |
brick | +0.011 |
bucket | +0.011 |
Neg logit contributions | |
ting | -0.032 |
ils | -0.031 |
rewarded | -0.030 |
Midnight | -0.030 |
boys | -0.029 |
Token | behind |
Token ID | 2157 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
ils | +0.042 |
rewarded | +0.039 |
Midnight | +0.039 |
treated | +0.036 |
ting | +0.036 |
Neg logit contributions | |
block | -0.044 |
bucket | -0.040 |
ladder | -0.039 |
brick | -0.038 |
most | -0.036 |
Token | the |
Token ID | 262 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
block | +0.085 |
bucket | +0.078 |
brick | +0.077 |
palm | +0.074 |
most | +0.074 |
Neg logit contributions | |
rewarded | -0.082 |
ils | -0.080 |
minded | -0.079 |
obe | -0.078 |
Claus | -0.076 |
Token | gate |
Token ID | 8946 |
Feature activation | +5.25e+00 |
Pos logit contributions | |
block | +0.024 |
bucket | +0.023 |
ladder | +0.022 |
most | +0.022 |
brick | +0.022 |
Neg logit contributions | |
ils | -0.065 |
ting | -0.064 |
rewarded | -0.064 |
Midnight | -0.061 |
obe | -0.060 |
Token | . |
Token ID | 13 |
Feature activation | -5.25e-01 |
Pos logit contributions | |
block | +0.171 |
ladder | +0.162 |
most | +0.157 |
bucket | +0.156 |
key | +0.148 |
Neg logit contributions | |
ils | -0.147 |
Midnight | -0.140 |
ting | -0.139 |
rewarded | -0.128 |
Rudolph | -0.128 |
Token | It |
Token ID | 632 |
Feature activation | +4.92e-01 |
Pos logit contributions | |
ils | +0.019 |
ting | +0.019 |
rewarded | +0.018 |
obe | +0.017 |
punished | +0.017 |
Neg logit contributions | |
block | -0.011 |
most | -0.011 |
bucket | -0.010 |
brick | -0.010 |
ladder | -0.010 |
Token | was |
Token ID | 373 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
block | +0.014 |
most | +0.013 |
bucket | +0.013 |
ladder | +0.013 |
brick | +0.013 |
Neg logit contributions | |
ting | -0.014 |
ils | -0.014 |
rewarded | -0.013 |
obe | -0.013 |
Midnight | -0.013 |
Token | terrible |
Token ID | 7818 |
Feature activation | -7.76e-01 |
Pos logit contributions | |
block | +0.025 |
bucket | +0.024 |
key | +0.023 |
brick | +0.023 |
most | +0.023 |
Neg logit contributions | |
ting | -0.039 |
ils | -0.039 |
minded | -0.036 |
respected | -0.036 |
obe | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -5.67e-01 |
Pos logit contributions | |
Mr | +0.027 |
rewarded | +0.026 |
otton | +0.025 |
ppa | +0.025 |
ils | +0.025 |
Neg logit contributions | |
block | -0.017 |
ladder | -0.017 |
shovel | -0.015 |
brick | -0.015 |
bucket | -0.015 |
Token | it |
Token ID | 340 |
Feature activation | -2.62e-01 |
Pos logit contributions | |
ils | +0.006 |
rewarded | +0.005 |
obe | +0.005 |
punished | +0.005 |
ting | +0.005 |
Neg logit contributions | |
block | -0.004 |
bucket | -0.004 |
ladder | -0.004 |
palm | -0.004 |
brick | -0.004 |
Token | would |
Token ID | 561 |
Feature activation | -2.60e-01 |
Pos logit contributions | |
ting | +0.007 |
ils | +0.006 |
Midnight | +0.006 |
obe | +0.006 |
rewarded | +0.006 |
Neg logit contributions | |
block | -0.008 |
bucket | -0.008 |
most | -0.008 |
ladder | -0.008 |
foot | -0.007 |
Token | hug |
Token ID | 16225 |
Feature activation | -6.17e-01 |
Pos logit contributions | |
ils | +0.010 |
ting | +0.010 |
rewarded | +0.009 |
Midnight | +0.009 |
obe | +0.009 |
Neg logit contributions | |
block | -0.005 |
bucket | -0.005 |
ladder | -0.005 |
most | -0.005 |
key | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | +7.84e-01 |
Pos logit contributions | |
punished | +0.021 |
rewarded | +0.021 |
ils | +0.020 |
glers | +0.019 |
obe | +0.019 |
Neg logit contributions | |
block | -0.014 |
most | -0.014 |
palm | -0.014 |
foot | -0.014 |
hand | -0.013 |
Token | shore |
Token ID | 15191 |
Feature activation | +2.32e+00 |
Pos logit contributions | |
bucket | +0.017 |
block | +0.017 |
brick | +0.016 |
palm | +0.016 |
ladder | +0.016 |
Neg logit contributions | |
ting | -0.027 |
ils | -0.027 |
rewarded | -0.026 |
Midnight | -0.025 |
obe | -0.025 |
Token | line |
Token ID | 1370 |
Feature activation | +5.25e+00 |
Pos logit contributions | |
block | +0.068 |
bucket | +0.064 |
most | +0.063 |
palm | +0.063 |
foot | +0.062 |
Neg logit contributions | |
Midnight | -0.070 |
ils | -0.067 |
punished | -0.065 |
obe | -0.065 |
Mr | -0.064 |
Token | and |
Token ID | 290 |
Feature activation | +9.74e-01 |
Pos logit contributions | |
block | +0.159 |
brick | +0.148 |
palm | +0.148 |
most | +0.148 |
ladder | +0.147 |
Neg logit contributions | |
ils | -0.149 |
Midnight | -0.145 |
ting | -0.137 |
punished | -0.135 |
obe | -0.135 |
Token | admire |
Token ID | 21099 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
bucket | +0.023 |
palm | +0.022 |
block | +0.022 |
shovel | +0.022 |
foot | +0.021 |
Neg logit contributions | |
ils | -0.035 |
ting | -0.034 |
obe | -0.032 |
lish | -0.032 |
Mr | -0.031 |
Token | the |
Token ID | 262 |
Feature activation | +4.54e-02 |
Pos logit contributions | |
ils | +0.011 |
Claus | +0.011 |
obe | +0.010 |
rewarded | +0.010 |
ting | +0.010 |
Neg logit contributions | |
palm | -0.008 |
block | -0.008 |
brick | -0.008 |
bucket | -0.008 |
foot | -0.008 |
Token | wrecked |
Token ID | 44770 |
Feature activation | +9.95e-01 |
Pos logit contributions | |
bucket | +0.001 |
block | +0.001 |
palm | +0.001 |
brick | +0.001 |
ladder | +0.001 |
Neg logit contributions | |
ils | -0.002 |
Midnight | -0.002 |
ting | -0.002 |
rewarded | -0.001 |
obe | -0.001 |
Token | boats |
Token ID | 16659 |
Feature activation | +4.54e-01 |
Pos logit contributions | |
bucket | +0.015 |
block | +0.015 |
palm | +0.014 |
ladder | +0.014 |
brick | +0.013 |
Neg logit contributions | |
ils | -0.041 |
Midnight | -0.041 |
ting | -0.041 |
rewarded | -0.041 |
obe | -0.039 |
Token | tie |
Token ID | 9839 |
Feature activation | +7.47e-02 |
Pos logit contributions | |
block | +0.003 |
bucket | +0.002 |
ladder | +0.002 |
nail | +0.002 |
most | +0.002 |
Neg logit contributions | |
ting | -0.003 |
ils | -0.003 |
rewarded | -0.003 |
Midnight | -0.003 |
obe | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | +6.20e-01 |
Pos logit contributions | |
block | +0.002 |
bucket | +0.002 |
most | +0.002 |
brick | +0.002 |
ladder | +0.002 |
Neg logit contributions | |
ting | -0.002 |
ils | -0.002 |
rewarded | -0.002 |
obe | -0.002 |
Midnight | -0.002 |
Token | yarn |
Token ID | 21181 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
most | +0.011 |
adder | +0.011 |
block | +0.010 |
brick | +0.010 |
bucket | +0.010 |
Neg logit contributions | |
ting | -0.022 |
ils | -0.022 |
rewarded | -0.021 |
Midnight | -0.021 |
boys | -0.020 |
Token | on |
Token ID | 319 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
ting | +0.025 |
ils | +0.024 |
rewarded | +0.024 |
Midnight | +0.022 |
punished | +0.021 |
Neg logit contributions | |
block | -0.045 |
most | -0.044 |
bucket | -0.044 |
ladder | -0.043 |
brick | -0.043 |
Token | the |
Token ID | 262 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
most | +0.071 |
block | +0.069 |
adder | +0.069 |
bucket | +0.069 |
ladder | +0.068 |
Neg logit contributions | |
ting | -0.044 |
ils | -0.042 |
rewarded | -0.040 |
Midnight | -0.039 |
boys | -0.037 |
Token | round |
Token ID | 2835 |
Feature activation | +5.24e+00 |
Pos logit contributions | |
block | +0.026 |
bucket | +0.025 |
most | +0.025 |
ladder | +0.024 |
palm | +0.024 |
Neg logit contributions | |
ting | -0.044 |
ils | -0.044 |
rewarded | -0.044 |
Midnight | -0.041 |
obe | -0.041 |
Token | block |
Token ID | 2512 |
Feature activation | +4.00e+00 |
Pos logit contributions | |
adder | +0.111 |
most | +0.110 |
bucket | +0.102 |
brick | +0.101 |
ladder | +0.101 |
Neg logit contributions | |
ting | -0.172 |
ils | -0.167 |
rewarded | -0.154 |
Midnight | -0.153 |
from | -0.148 |
Token | . |
Token ID | 13 |
Feature activation | -5.56e-01 |
Pos logit contributions | |
most | +0.153 |
adder | +0.153 |
bucket | +0.151 |
palm | +0.150 |
brick | +0.149 |
Neg logit contributions | |
ting | -0.056 |
ils | -0.056 |
rewarded | -0.052 |
from | -0.047 |
boys | -0.046 |
Token | They |
Token ID | 1119 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
ils | +0.015 |
ting | +0.015 |
rewarded | +0.014 |
punished | +0.014 |
obe | +0.013 |
Neg logit contributions | |
block | -0.017 |
most | -0.017 |
bucket | -0.016 |
ladder | -0.016 |
brick | -0.016 |
Token | wrap |
Token ID | 14441 |
Feature activation | -5.81e-01 |
Pos logit contributions | |
block | +0.004 |
bucket | +0.004 |
ladder | +0.004 |
nail | +0.004 |
key | +0.004 |
Neg logit contributions | |
ting | -0.005 |
ils | -0.005 |
rewarded | -0.004 |
Midnight | -0.004 |
obe | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | +8.68e-01 |
Pos logit contributions | |
ting | +0.015 |
rewarded | +0.015 |
ils | +0.015 |
obe | +0.015 |
punished | +0.014 |
Neg logit contributions | |
block | -0.018 |
bucket | -0.017 |
brick | -0.017 |
most | -0.017 |
ladder | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
block | +0.019 |
bucket | +0.018 |
brick | +0.018 |
palm | +0.018 |
most | +0.018 |
Neg logit contributions | |
ting | -0.023 |
rewarded | -0.022 |
ils | -0.022 |
obe | -0.022 |
Midnight | -0.022 |
Token | funny |
Token ID | 8258 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
most | +0.022 |
block | +0.021 |
adder | +0.020 |
bucket | +0.020 |
brick | +0.020 |
Neg logit contributions | |
ting | -0.038 |
ils | -0.038 |
rewarded | -0.036 |
Midnight | -0.035 |
boys | -0.035 |
Token | lines |
Token ID | 3951 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
block | +0.055 |
bucket | +0.054 |
most | +0.054 |
brick | +0.052 |
ladder | +0.051 |
Neg logit contributions | |
ting | -0.071 |
rewarded | -0.068 |
ils | -0.068 |
Midnight | -0.066 |
punished | -0.063 |
Token | on |
Token ID | 319 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
block | +0.055 |
bucket | +0.052 |
most | +0.051 |
ladder | +0.051 |
brick | +0.050 |
Neg logit contributions | |
ior | -0.044 |
obe | -0.043 |
ting | -0.043 |
Mr | -0.042 |
Midnight | -0.042 |
Token | the |
Token ID | 262 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
block | +0.063 |
most | +0.061 |
brick | +0.060 |
bucket | +0.060 |
nail | +0.059 |
Neg logit contributions | |
ting | -0.081 |
rewarded | -0.080 |
ils | -0.079 |
punished | -0.075 |
obe | -0.075 |
Token | pole |
Token ID | 16825 |
Feature activation | +5.23e+00 |
Pos logit contributions | |
most | +0.029 |
block | +0.029 |
bucket | +0.028 |
brick | +0.027 |
adder | +0.027 |
Neg logit contributions | |
ting | -0.064 |
ils | -0.062 |
rewarded | -0.061 |
Midnight | -0.059 |
boys | -0.058 |
Token | . |
Token ID | 13 |
Feature activation | +8.37e-02 |
Pos logit contributions | |
block | +0.169 |
most | +0.159 |
brick | +0.156 |
bucket | +0.155 |
ladder | +0.152 |
Neg logit contributions | |
ting | -0.135 |
ils | -0.135 |
Midnight | -0.132 |
rewarded | -0.129 |
Mr | -0.128 |
Token |
|
Token ID | 198 |
Feature activation | +3.81e-02 |
Pos logit contributions | |
block | +0.002 |
most | +0.002 |
bucket | +0.002 |
brick | +0.002 |
ladder | +0.002 |
Neg logit contributions | |
ting | -0.002 |
ils | -0.002 |
rewarded | -0.002 |
Midnight | -0.002 |
boys | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | -2.57e-01 |
Pos logit contributions | |
most | +0.001 |
block | +0.001 |
bucket | +0.001 |
brick | +0.001 |
adder | +0.001 |
Neg logit contributions | |
ting | -0.001 |
ils | -0.001 |
rewarded | -0.001 |
Midnight | -0.001 |
boys | -0.001 |
Token | The |
Token ID | 464 |
Feature activation | -5.42e-01 |
Pos logit contributions | |
ting | +0.008 |
ils | +0.008 |
rewarded | +0.008 |
Claus | +0.008 |
boys | +0.008 |
Neg logit contributions | |
most | -0.006 |
block | -0.006 |
palm | -0.006 |
nail | -0.006 |
second | -0.006 |
Token | girl |
Token ID | 2576 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
ting | +0.013 |
ils | +0.013 |
Midnight | +0.012 |
rewarded | +0.012 |
boys | +0.012 |
Neg logit contributions | |
most | -0.018 |
adder | -0.016 |
run | -0.016 |
chie | -0.016 |
path | -0.015 |
Token | quickly |
Token ID | 2952 |
Feature activation | +3.11e-01 |
Pos logit contributions | |
ting | +0.001 |
ils | +0.001 |
Midnight | +0.001 |
obe | +0.001 |
Rudolph | +0.001 |
Neg logit contributions | |
block | -0.001 |
bucket | -0.001 |
most | -0.001 |
brick | -0.001 |
ladder | -0.001 |
Token | unt |
Token ID | 1418 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
bucket | +0.008 |
block | +0.008 |
most | +0.008 |
shovel | +0.008 |
ladder | +0.008 |
Neg logit contributions | |
ting | -0.009 |
ils | -0.008 |
Midnight | -0.008 |
minded | -0.008 |
ior | -0.008 |
Token | ied |
Token ID | 798 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
ting | +0.027 |
ils | +0.026 |
rewarded | +0.025 |
Midnight | +0.024 |
boys | +0.022 |
Neg logit contributions | |
block | -0.041 |
most | -0.041 |
bucket | -0.040 |
brick | -0.039 |
ladder | -0.039 |
Token | the |
Token ID | 262 |
Feature activation | +5.97e-01 |
Pos logit contributions | |
ils | +0.004 |
rewarded | +0.004 |
obe | +0.004 |
Claus | +0.004 |
ting | +0.004 |
Neg logit contributions | |
block | -0.004 |
brick | -0.004 |
bucket | -0.004 |
ladder | -0.004 |
palm | -0.004 |
Token | sho |
Token ID | 7354 |
Feature activation | +2.20e+00 |
Pos logit contributions | |
adder | +0.012 |
most | +0.012 |
block | +0.010 |
bucket | +0.010 |
palm | +0.010 |
Neg logit contributions | |
ting | -0.021 |
ils | -0.020 |
rewarded | -0.020 |
Midnight | -0.020 |
boys | -0.019 |
Token | el |
Token ID | 417 |
Feature activation | +5.19e+00 |
Pos logit contributions | |
bucket | +0.057 |
block | +0.057 |
most | +0.056 |
ladder | +0.055 |
adder | +0.054 |
Neg logit contributions | |
ting | -0.068 |
ils | -0.062 |
rewarded | -0.060 |
Midnight | -0.059 |
boys | -0.056 |
Token | ace |
Token ID | 558 |
Feature activation | +3.14e+00 |
Pos logit contributions | |
most | +0.130 |
block | +0.130 |
bucket | +0.127 |
adder | +0.124 |
brick | +0.124 |
Neg logit contributions | |
ting | -0.155 |
ils | -0.152 |
rewarded | -0.148 |
Midnight | -0.140 |
boys | -0.136 |
Token | and |
Token ID | 290 |
Feature activation | +8.22e-01 |
Pos logit contributions | |
block | +0.100 |
ladder | +0.098 |
most | +0.097 |
bucket | +0.096 |
foot | +0.094 |
Neg logit contributions | |
ting | -0.078 |
ils | -0.078 |
rewarded | -0.074 |
Midnight | -0.074 |
Mr | -0.070 |
Token | the |
Token ID | 262 |
Feature activation | -9.92e-02 |
Pos logit contributions | |
bucket | +0.026 |
block | +0.025 |
shovel | +0.025 |
ladder | +0.024 |
most | +0.024 |
Neg logit contributions | |
ils | -0.020 |
ting | -0.020 |
obe | -0.019 |
respected | -0.019 |
rewarded | -0.018 |
Token | depend |
Token ID | 4745 |
Feature activation | -7.91e-02 |
Pos logit contributions | |
ting | +0.003 |
ils | +0.003 |
rewarded | +0.003 |
Midnight | +0.003 |
obe | +0.003 |
Neg logit contributions | |
block | -0.002 |
most | -0.002 |
bucket | -0.002 |
brick | -0.002 |
ladder | -0.002 |
Token | able |
Token ID | 540 |
Feature activation | -4.93e-02 |
Pos logit contributions | |
ting | +0.002 |
ils | +0.002 |
rewarded | +0.002 |
Midnight | +0.001 |
punished | +0.001 |
Neg logit contributions | |
most | -0.003 |
block | -0.003 |
bucket | -0.003 |
ladder | -0.003 |
brick | -0.003 |
Token | made |
Token ID | 925 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
ting | +0.018 |
ils | +0.017 |
rewarded | +0.016 |
Midnight | +0.016 |
obe | +0.015 |
Neg logit contributions | |
block | -0.020 |
bucket | -0.020 |
most | -0.019 |
palm | -0.018 |
ladder | -0.018 |
Token | a |
Token ID | 257 |
Feature activation | -5.43e-01 |
Pos logit contributions | |
ting | +0.006 |
ils | +0.006 |
rewarded | +0.005 |
Midnight | +0.005 |
boys | +0.005 |
Neg logit contributions | |
most | -0.006 |
block | -0.006 |
bucket | -0.006 |
brick | -0.006 |
adder | -0.006 |
Token | v |
Token ID | 410 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
rewarded | +0.016 |
ting | +0.016 |
boys | +0.015 |
Midnight | +0.015 |
ils | +0.015 |
Neg logit contributions | |
adder | -0.013 |
most | -0.013 |
path | -0.012 |
block | -0.011 |
brick | -0.011 |
Token | room |
Token ID | 3823 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
most | +0.028 |
bucket | +0.027 |
ladder | +0.027 |
palm | +0.027 |
brick | +0.027 |
Neg logit contributions | |
ting | -0.028 |
ils | -0.027 |
rewarded | -0.026 |
obe | -0.025 |
boys | -0.025 |
Token | - |
Token ID | 12 |
Feature activation | +2.48e+00 |
Pos logit contributions | |
block | +0.065 |
ladder | +0.059 |
most | +0.059 |
bucket | +0.059 |
brick | +0.057 |
Neg logit contributions | |
ils | -0.070 |
ting | -0.068 |
rewarded | -0.067 |
Midnight | -0.066 |
respected | -0.061 |
Token | v |
Token ID | 85 |
Feature activation | +5.17e+00 |
Pos logit contributions | |
bucket | +0.078 |
most | +0.076 |
key | +0.076 |
ladder | +0.076 |
palm | +0.075 |
Neg logit contributions | |
ting | -0.058 |
ils | -0.056 |
ups | -0.055 |
rewarded | -0.051 |
punished | -0.050 |
Token | room |
Token ID | 3823 |
Feature activation | +2.28e+00 |
Pos logit contributions | |
most | +0.160 |
ladder | +0.158 |
palm | +0.152 |
coconut | +0.151 |
adder | +0.151 |
Neg logit contributions | |
ting | -0.122 |
ils | -0.117 |
ating | -0.113 |
rewarded | -0.110 |
obe | -0.106 |
Token | sound |
Token ID | 2128 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
block | +0.065 |
most | +0.060 |
ladder | +0.060 |
bucket | +0.059 |
foot | +0.057 |
Neg logit contributions | |
ils | -0.066 |
ting | -0.065 |
rewarded | -0.064 |
Midnight | -0.062 |
respected | -0.057 |
Token | and |
Token ID | 290 |
Feature activation | +8.57e-01 |
Pos logit contributions | |
ils | +0.003 |
Midnight | +0.003 |
rewarded | +0.003 |
ior | +0.003 |
ting | +0.003 |
Neg logit contributions | |
block | -0.004 |
most | -0.003 |
foot | -0.003 |
adder | -0.003 |
second | -0.003 |
Token | laughed |
Token ID | 13818 |
Feature activation | +6.45e-01 |
Pos logit contributions | |
block | +0.022 |
bucket | +0.021 |
most | +0.021 |
ladder | +0.020 |
palm | +0.020 |
Neg logit contributions | |
ting | -0.027 |
ils | -0.026 |
rewarded | -0.025 |
Midnight | -0.024 |
obe | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | -6.10e-01 |
Pos logit contributions | |
block | +0.019 |
second | +0.017 |
foot | +0.017 |
most | +0.017 |
bucket | +0.017 |
Neg logit contributions | |
ils | -0.020 |
rewarded | -0.018 |
punished | -0.017 |
ting | -0.017 |
Midnight | -0.017 |
Token | keep |
Token ID | 1394 |
Feature activation | -2.24e+00 |
Pos logit contributions | |
ting | +0.005 |
ils | +0.005 |
rewarded | +0.004 |
Midnight | +0.004 |
obe | +0.004 |
Neg logit contributions | |
block | -0.004 |
bucket | -0.004 |
brick | -0.004 |
ladder | -0.004 |
most | -0.004 |
Token | going |
Token ID | 1016 |
Feature activation | -3.14e-01 |
Pos logit contributions | |
ting | +0.049 |
ils | +0.046 |
rewarded | +0.045 |
Midnight | +0.042 |
boys | +0.040 |
Neg logit contributions | |
most | -0.077 |
block | -0.076 |
adder | -0.074 |
bucket | -0.073 |
brick | -0.073 |
Token | beyond |
Token ID | 3675 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
ils | +0.009 |
Midnight | +0.008 |
ting | +0.008 |
rewarded | +0.008 |
Mr | +0.008 |
Neg logit contributions | |
block | -0.009 |
bucket | -0.009 |
ladder | -0.009 |
most | -0.009 |
key | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | -3.39e-01 |
Pos logit contributions | |
ting | +0.005 |
ils | +0.005 |
Midnight | +0.005 |
rewarded | +0.005 |
Claus | +0.004 |
Neg logit contributions | |
block | -0.006 |
bucket | -0.006 |
palm | -0.006 |
most | -0.006 |
brick | -0.005 |
Token | shore |
Token ID | 15191 |
Feature activation | +2.21e+00 |
Pos logit contributions | |
ils | +0.012 |
ting | +0.012 |
Midnight | +0.012 |
rewarded | +0.012 |
boys | +0.011 |
Neg logit contributions | |
bucket | -0.007 |
block | -0.007 |
palm | -0.007 |
ladder | -0.007 |
brick | -0.007 |
Token | line |
Token ID | 1370 |
Feature activation | +5.15e+00 |
Pos logit contributions | |
block | +0.069 |
most | +0.067 |
bucket | +0.067 |
palm | +0.065 |
foot | +0.065 |
Neg logit contributions | |
Midnight | -0.064 |
ils | -0.060 |
punished | -0.054 |
Mr | -0.053 |
rewarded | -0.053 |
Token | and |
Token ID | 290 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
block | +0.172 |
most | +0.168 |
bucket | +0.164 |
ladder | +0.163 |
brick | +0.162 |
Neg logit contributions | |
ils | -0.127 |
Midnight | -0.121 |
ting | -0.118 |
rewarded | -0.113 |
punished | -0.106 |
Token | swim |
Token ID | 9422 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
ils | +0.007 |
ting | +0.007 |
Midnight | +0.006 |
rewarded | +0.006 |
obe | +0.006 |
Neg logit contributions | |
bucket | -0.005 |
block | -0.005 |
ladder | -0.005 |
palm | -0.005 |
foot | -0.005 |
Token | out |
Token ID | 503 |
Feature activation | -8.65e-01 |
Pos logit contributions | |
Midnight | +0.047 |
ils | +0.043 |
obe | +0.043 |
ica | +0.042 |
ting | +0.042 |
Neg logit contributions | |
bucket | -0.042 |
block | -0.040 |
second | -0.039 |
foot | -0.038 |
most | -0.037 |
Token | further |
Token ID | 2252 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
ils | +0.027 |
Midnight | +0.027 |
Claus | +0.026 |
obe | +0.026 |
ior | +0.025 |
Neg logit contributions | |
block | -0.023 |
most | -0.023 |
bucket | -0.022 |
ladder | -0.022 |
foot | -0.022 |
Token | in |
Token ID | 287 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
Midnight | +0.004 |
ils | +0.004 |
ting | +0.004 |
rewarded | +0.004 |
Mr | +0.004 |
Neg logit contributions | |
bucket | -0.004 |
block | -0.004 |
palm | -0.004 |
ladder | -0.004 |
foot | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | -3.19e-01 |
Pos logit contributions | |
ils | +0.009 |
rewarded | +0.009 |
obe | +0.008 |
ting | +0.008 |
respected | +0.008 |
Neg logit contributions | |
bucket | -0.006 |
block | -0.006 |
second | -0.006 |
most | -0.006 |
brick | -0.006 |
Token | famous |
Token ID | 5863 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
ils | +0.011 |
ting | +0.011 |
rewarded | +0.011 |
obe | +0.011 |
respected | +0.011 |
Neg logit contributions | |
block | -0.007 |
bucket | -0.007 |
brick | -0.006 |
ladder | -0.006 |
palm | -0.006 |
Token | chef |
Token ID | 21221 |
Feature activation | -3.02e+00 |
Pos logit contributions | |
block | +0.025 |
bucket | +0.025 |
brick | +0.024 |
ladder | +0.024 |
key | +0.023 |
Neg logit contributions | |
ting | -0.042 |
rewarded | -0.041 |
ils | -0.041 |
obe | -0.040 |
respected | -0.040 |
Token | was |
Token ID | 373 |
Feature activation | +4.21e-01 |
Pos logit contributions | |
ting | +0.108 |
obe | +0.104 |
Midnight | +0.101 |
minded | +0.099 |
lish | +0.098 |
Neg logit contributions | |
block | -0.072 |
bucket | -0.070 |
key | -0.062 |
brick | -0.061 |
shovel | -0.061 |
Token | making |
Token ID | 1642 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
block | +0.011 |
bucket | +0.011 |
ladder | +0.010 |
second | +0.010 |
foot | +0.010 |
Neg logit contributions | |
ils | -0.013 |
ting | -0.013 |
ior | -0.013 |
obe | -0.012 |
minded | -0.012 |
Token | las |
Token ID | 39990 |
Feature activation | -6.16e+00 |
Pos logit contributions | |
ting | +0.067 |
ils | +0.065 |
Claus | +0.064 |
Midnight | +0.062 |
obe | +0.061 |
Neg logit contributions | |
block | -0.039 |
bucket | -0.038 |
brick | -0.037 |
foot | -0.035 |
second | -0.035 |
Token | agna |
Token ID | 48669 |
Feature activation | -4.07e+00 |
Pos logit contributions | |
ting | +0.162 |
Midnight | +0.154 |
rewarded | +0.142 |
treated | +0.140 |
from | +0.138 |
Neg logit contributions | |
block | -0.181 |
most | -0.178 |
bucket | -0.174 |
adder | -0.169 |
ruler | -0.165 |
Token | . |
Token ID | 13 |
Feature activation | -2.75e-01 |
Pos logit contributions | |
Midnight | +0.123 |
ting | +0.119 |
ior | +0.111 |
ils | +0.110 |
Mr | +0.110 |
Neg logit contributions | |
block | -0.119 |
bucket | -0.109 |
ladder | -0.106 |
ruler | -0.103 |
foot | -0.102 |
Token | That |
Token ID | 1320 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
ils | +0.010 |
ting | +0.009 |
boys | +0.009 |
rewarded | +0.009 |
obe | +0.009 |
Neg logit contributions | |
block | -0.006 |
most | -0.006 |
brick | -0.006 |
bucket | -0.006 |
ladder | -0.006 |
Token | is |
Token ID | 318 |
Feature activation | +6.04e-01 |
Pos logit contributions | |
block | +0.040 |
bucket | +0.039 |
brick | +0.037 |
key | +0.037 |
pole | +0.036 |
Neg logit contributions | |
ting | -0.042 |
obe | -0.039 |
Midnight | -0.037 |
boys | -0.036 |
respected | -0.036 |
Token | a |
Token ID | 257 |
Feature activation | -7.06e-01 |
Pos logit contributions | |
block | +0.013 |
second | +0.013 |
brick | +0.013 |
bucket | +0.012 |
key | +0.012 |
Neg logit contributions | |
ting | -0.023 |
ils | -0.021 |
obe | -0.021 |
respected | -0.021 |
boys | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
ting | +0.119 |
Midnight | +0.101 |
boys | +0.099 |
minded | +0.097 |
Mr | +0.097 |
Neg logit contributions | |
bucket | -0.114 |
block | -0.110 |
shovel | -0.101 |
ladder | -0.100 |
hand | -0.094 |
Token |
|
Token ID | 198 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
ils | +0.008 |
ting | +0.007 |
rewarded | +0.007 |
obe | +0.007 |
punished | +0.007 |
Neg logit contributions | |
block | -0.005 |
most | -0.005 |
brick | -0.005 |
bucket | -0.005 |
ladder | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | -3.32e-01 |
Pos logit contributions | |
rewarded | +0.000 |
ting | +0.000 |
ils | +0.000 |
Midnight | +0.000 |
Mr | +0.000 |
Neg logit contributions | |
bucket | -0.001 |
most | -0.001 |
block | -0.001 |
ladder | -0.001 |
brick | -0.001 |
Token | " |
Token ID | 1 |
Feature activation | -4.70e-01 |
Pos logit contributions | |
ting | +0.009 |
ils | +0.008 |
rewarded | +0.008 |
Midnight | +0.008 |
boys | +0.007 |
Neg logit contributions | |
most | -0.010 |
block | -0.010 |
bucket | -0.009 |
brick | -0.009 |
ladder | -0.009 |
Token | Bro |
Token ID | 15783 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
ting | +0.012 |
ils | +0.012 |
rewarded | +0.012 |
Midnight | +0.011 |
boys | +0.011 |
Neg logit contributions | |
block | -0.014 |
most | -0.014 |
bucket | -0.013 |
brick | -0.013 |
ladder | -0.013 |
Token | ccoli |
Token ID | 34544 |
Feature activation | -5.97e+00 |
Pos logit contributions | |
ting | +0.059 |
ils | +0.055 |
Midnight | +0.053 |
rewarded | +0.053 |
obe | +0.051 |
Neg logit contributions | |
block | -0.053 |
bucket | -0.050 |
ladder | -0.050 |
most | -0.050 |
adder | -0.049 |
Token | is |
Token ID | 318 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
ting | +0.238 |
ils | +0.214 |
obe | +0.213 |
minded | +0.211 |
eline | +0.208 |
Neg logit contributions | |
block | -0.142 |
bucket | -0.122 |
ladder | -0.116 |
pole | -0.112 |
foot | -0.106 |
Token | y |
Token ID | 331 |
Feature activation | -5.59e-01 |
Pos logit contributions | |
block | +0.005 |
bucket | +0.005 |
brick | +0.005 |
second | +0.005 |
foot | +0.004 |
Neg logit contributions | |
ting | -0.010 |
ils | -0.009 |
minded | -0.009 |
respected | -0.009 |
obe | -0.008 |
Token | ucky |
Token ID | 5309 |
Feature activation | -7.21e-01 |
Pos logit contributions | |
ting | +0.015 |
rewarded | +0.015 |
ils | +0.015 |
Midnight | +0.014 |
punished | +0.014 |
Neg logit contributions | |
block | -0.016 |
most | -0.016 |
bucket | -0.016 |
brick | -0.016 |
adder | -0.015 |
Token | !" |
Token ID | 2474 |
Feature activation | +7.47e-01 |
Pos logit contributions | |
ting | +0.023 |
minded | +0.021 |
rewarded | +0.021 |
respected | +0.020 |
ils | +0.020 |
Neg logit contributions | |
block | -0.021 |
ladder | -0.019 |
bucket | -0.019 |
foot | -0.018 |
brick | -0.018 |
Token | Ben |
Token ID | 3932 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
block | +0.018 |
bucket | +0.018 |
ladder | +0.018 |
second | +0.018 |
most | +0.017 |
Neg logit contributions | |
ils | -0.025 |
ting | -0.023 |
rewarded | -0.023 |
ating | -0.023 |
obe | -0.022 |
Token | that |
Token ID | 326 |
Feature activation | -9.92e-01 |
Pos logit contributions | |
bucket | +0.007 |
ladder | +0.007 |
block | +0.007 |
foot | +0.007 |
brick | +0.007 |
Neg logit contributions | |
ils | -0.006 |
rewarded | -0.006 |
treated | -0.005 |
boys | -0.005 |
obe | -0.005 |
Token | day |
Token ID | 1110 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
ting | +0.031 |
ils | +0.031 |
rewarded | +0.030 |
Midnight | +0.028 |
boys | +0.028 |
Neg logit contributions | |
block | -0.024 |
most | -0.024 |
bucket | -0.023 |
brick | -0.023 |
ladder | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
minded | +0.053 |
rewarded | +0.051 |
ils | +0.051 |
obe | +0.050 |
boys | +0.050 |
Neg logit contributions | |
bucket | -0.033 |
ladder | -0.033 |
block | -0.031 |
foot | -0.030 |
second | -0.030 |
Token | Lily |
Token ID | 20037 |
Feature activation | -5.04e-01 |
Pos logit contributions | |
ils | +0.036 |
rewarded | +0.036 |
obe | +0.036 |
minded | +0.035 |
ting | +0.034 |
Neg logit contributions | |
second | -0.022 |
bucket | -0.021 |
block | -0.020 |
foot | -0.020 |
ladder | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | -5.71e-01 |
Pos logit contributions | |
ting | +0.018 |
ils | +0.018 |
ating | +0.017 |
ises | +0.017 |
ual | +0.017 |
Neg logit contributions | |
block | -0.010 |
bucket | -0.010 |
palm | -0.009 |
foot | -0.009 |
most | -0.009 |
Token | Midnight |
Token ID | 27960 |
Feature activation | -5.70e+00 |
Pos logit contributions | |
rewarded | +0.016 |
ups | +0.015 |
ils | +0.014 |
rim | +0.014 |
ating | +0.013 |
Neg logit contributions | |
second | -0.016 |
foot | -0.015 |
hand | -0.015 |
key | -0.015 |
rope | -0.014 |
Token | went |
Token ID | 1816 |
Feature activation | -9.76e-01 |
Pos logit contributions | |
ating | +0.204 |
ils | +0.195 |
minded | +0.194 |
ior | +0.194 |
ups | +0.192 |
Neg logit contributions | |
bucket | -0.124 |
palm | -0.109 |
shovel | -0.108 |
second | -0.102 |
most | -0.099 |
Token | back |
Token ID | 736 |
Feature activation | -5.42e-01 |
Pos logit contributions | |
ils | +0.026 |
ting | +0.025 |
rewarded | +0.025 |
Midnight | +0.023 |
punished | +0.023 |
Neg logit contributions | |
bucket | -0.029 |
block | -0.029 |
most | -0.028 |
ladder | -0.028 |
brick | -0.027 |
Token | home |
Token ID | 1363 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
ting | +0.013 |
ils | +0.013 |
rewarded | +0.012 |
Midnight | +0.012 |
boys | +0.011 |
Neg logit contributions | |
block | -0.017 |
most | -0.017 |
bucket | -0.016 |
brick | -0.016 |
adder | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -5.86e-01 |
Pos logit contributions | |
ils | +0.026 |
ting | +0.025 |
Midnight | +0.023 |
rewarded | +0.022 |
boys | +0.021 |
Neg logit contributions | |
block | -0.036 |
bucket | -0.036 |
most | -0.036 |
ladder | -0.036 |
foot | -0.035 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
ils | +0.019 |
rewarded | +0.019 |
ating | +0.018 |
ting | +0.018 |
obe | +0.017 |
Neg logit contributions | |
bucket | -0.014 |
second | -0.013 |
Int | -0.013 |
ladder | -0.013 |
hand | -0.013 |
Token | to |
Token ID | 284 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
ting | +0.003 |
rewarded | +0.003 |
ils | +0.003 |
Claus | +0.003 |
obe | +0.003 |
Neg logit contributions | |
block | -0.003 |
brick | -0.003 |
ladder | -0.003 |
bucket | -0.003 |
palm | -0.003 |
Token | eat |
Token ID | 4483 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
block | +0.004 |
bucket | +0.003 |
palm | +0.003 |
foot | +0.003 |
nail | +0.003 |
Neg logit contributions | |
ting | -0.005 |
ils | -0.004 |
minded | -0.004 |
rewarded | -0.004 |
ups | -0.004 |
Token | broccoli |
Token ID | 44653 |
Feature activation | -4.02e+00 |
Pos logit contributions | |
rewarded | +0.075 |
Claus | +0.072 |
ting | +0.072 |
minded | +0.068 |
ils | +0.068 |
Neg logit contributions | |
block | -0.061 |
bucket | -0.060 |
ladder | -0.057 |
palm | -0.057 |
brick | -0.055 |
Token | . |
Token ID | 13 |
Feature activation | -7.03e-01 |
Pos logit contributions | |
ior | +0.132 |
ting | +0.126 |
rewarded | +0.125 |
respected | +0.123 |
pty | +0.122 |
Neg logit contributions | |
block | -0.102 |
bucket | -0.099 |
key | -0.091 |
ladder | -0.091 |
foot | -0.090 |
Token | Bro |
Token ID | 2806 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
ting | +0.017 |
ils | +0.017 |
rewarded | +0.017 |
Midnight | +0.016 |
boys | +0.015 |
Neg logit contributions | |
block | -0.021 |
most | -0.021 |
bucket | -0.021 |
brick | -0.021 |
ladder | -0.020 |
Token | ccoli |
Token ID | 34544 |
Feature activation | -5.65e+00 |
Pos logit contributions | |
ting | +0.092 |
rewarded | +0.075 |
including | +0.074 |
bothering | +0.071 |
from | +0.071 |
Neg logit contributions | |
block | -0.066 |
bucket | -0.061 |
shovel | -0.056 |
finger | -0.056 |
ig | -0.056 |
Token | was |
Token ID | 373 |
Feature activation | +6.16e-01 |
Pos logit contributions | |
ting | +0.217 |
rewarded | +0.212 |
minded | +0.210 |
obe | +0.208 |
eline | +0.205 |
Neg logit contributions | |
block | -0.114 |
bucket | -0.102 |
foot | -0.093 |
pole | -0.090 |
ladder | -0.087 |
Token | green |
Token ID | 4077 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
block | +0.014 |
bucket | +0.013 |
brick | +0.013 |
second | +0.013 |
most | +0.013 |
Neg logit contributions | |
ting | -0.022 |
ils | -0.020 |
boys | -0.020 |
minded | -0.020 |
rewarded | -0.019 |
Token | and |
Token ID | 290 |
Feature activation | +7.43e-01 |
Pos logit contributions | |
block | +0.038 |
bucket | +0.035 |
foot | +0.033 |
palm | +0.033 |
ladder | +0.032 |
Neg logit contributions | |
rewarded | -0.046 |
ting | -0.044 |
ils | -0.043 |
punished | -0.043 |
minded | -0.043 |
Token | bitter |
Token ID | 12922 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
block | +0.017 |
bucket | +0.016 |
brick | +0.016 |
ladder | +0.015 |
palm | +0.015 |
Neg logit contributions | |
ting | -0.027 |
ils | -0.025 |
rewarded | -0.024 |
obe | -0.024 |
respected | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | +8.69e-01 |
Pos logit contributions | |
rewarded | +0.058 |
Claus | +0.056 |
ica | +0.055 |
eline | +0.054 |
Mr | +0.054 |
Neg logit contributions | |
block | -0.035 |
bucket | -0.033 |
foot | -0.032 |
brick | -0.031 |
palm | -0.031 |
Token | said |
Token ID | 531 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
obe | +0.016 |
ting | +0.016 |
ils | +0.015 |
minded | +0.015 |
respected | +0.015 |
Neg logit contributions | |
block | -0.009 |
palm | -0.008 |
bucket | -0.008 |
ladder | -0.008 |
foot | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
ladder | +0.010 |
bucket | +0.010 |
foot | +0.010 |
block | +0.010 |
second | +0.010 |
Neg logit contributions | |
obe | -0.014 |
ting | -0.013 |
ils | -0.013 |
ior | -0.013 |
Claus | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
ils | +0.005 |
ting | +0.005 |
rewarded | +0.004 |
boys | +0.004 |
punished | +0.004 |
Neg logit contributions | |
most | -0.004 |
block | -0.004 |
bucket | -0.004 |
brick | -0.004 |
ladder | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
block | +0.003 |
most | +0.003 |
bucket | +0.003 |
brick | +0.003 |
ladder | +0.003 |
Neg logit contributions | |
ting | -0.003 |
ils | -0.003 |
rewarded | -0.003 |
Midnight | -0.003 |
boys | -0.003 |
Token | So |
Token ID | 2396 |
Feature activation | +4.44e-01 |
Pos logit contributions | |
ting | +0.005 |
ils | +0.005 |
rewarded | +0.005 |
Claus | +0.004 |
Midnight | +0.004 |
Neg logit contributions | |
most | -0.004 |
block | -0.004 |
bucket | -0.004 |
brick | -0.004 |
adder | -0.004 |
Token | Midnight |
Token ID | 27960 |
Feature activation | -5.65e+00 |
Pos logit contributions | |
bucket | +0.010 |
second | +0.009 |
foot | +0.009 |
hand | +0.009 |
key | +0.009 |
Neg logit contributions | |
ils | -0.015 |
Claus | -0.015 |
boys | -0.015 |
girls | -0.015 |
yer | -0.014 |
Token | held |
Token ID | 2714 |
Feature activation | -8.72e-01 |
Pos logit contributions | |
respected | +0.183 |
obe | +0.183 |
yourselves | +0.182 |
ting | +0.181 |
ating | +0.181 |
Neg logit contributions | |
bucket | -0.119 |
block | -0.114 |
rope | -0.109 |
circle | -0.106 |
peek | -0.106 |
Token | the |
Token ID | 262 |
Feature activation | -3.17e-02 |
Pos logit contributions | |
rewarded | +0.024 |
ils | +0.024 |
obe | +0.023 |
Claus | +0.022 |
boys | +0.022 |
Neg logit contributions | |
block | -0.024 |
bucket | -0.024 |
foot | -0.024 |
palm | -0.024 |
most | -0.023 |
Token | furry |
Token ID | 40857 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
ting | +0.001 |
ils | +0.001 |
rewarded | +0.001 |
Midnight | +0.001 |
obe | +0.001 |
Neg logit contributions | |
block | -0.001 |
bucket | -0.001 |
most | -0.001 |
brick | -0.001 |
ladder | -0.001 |
Token | toy |
Token ID | 13373 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
block | +0.046 |
most | +0.045 |
bucket | +0.044 |
brick | +0.043 |
ladder | +0.043 |
Neg logit contributions | |
ting | -0.077 |
ils | -0.076 |
rewarded | -0.075 |
Midnight | -0.071 |
punished | -0.070 |
Token | and |
Token ID | 290 |
Feature activation | +5.98e-01 |
Pos logit contributions | |
block | +0.033 |
bucket | +0.032 |
brick | +0.032 |
most | +0.032 |
ladder | +0.031 |
Neg logit contributions | |
ils | -0.027 |
rewarded | -0.026 |
ting | -0.026 |
obe | -0.024 |
punished | -0.024 |
Token | to |
Token ID | 284 |
Feature activation | +2.41e-01 |
Pos logit contributions | |
block | +0.011 |
bucket | +0.011 |
foot | +0.010 |
palm | +0.010 |
ladder | +0.010 |
Neg logit contributions | |
ting | -0.009 |
ils | -0.009 |
Midnight | -0.009 |
Mr | -0.009 |
ior | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
block | +0.006 |
bucket | +0.006 |
ladder | +0.006 |
foot | +0.006 |
nail | +0.006 |
Neg logit contributions | |
ting | -0.008 |
ils | -0.008 |
rewarded | -0.007 |
minded | -0.007 |
obe | -0.007 |
Token | man |
Token ID | 582 |
Feature activation | -3.07e-01 |
Pos logit contributions | |
ils | +0.010 |
ting | +0.010 |
rewarded | +0.009 |
respected | +0.009 |
obe | +0.009 |
Neg logit contributions | |
block | -0.006 |
bucket | -0.005 |
ladder | -0.005 |
brick | -0.005 |
palm | -0.005 |
Token | " |
Token ID | 366 |
Feature activation | +8.40e-01 |
Pos logit contributions | |
ils | +0.009 |
ting | +0.009 |
Midnight | +0.008 |
Mr | +0.008 |
lish | +0.008 |
Neg logit contributions | |
block | -0.009 |
hand | -0.008 |
bucket | -0.008 |
most | -0.008 |
ladder | -0.008 |
Token | Bro |
Token ID | 15783 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
block | +0.026 |
most | +0.026 |
brick | +0.024 |
adder | +0.024 |
bucket | +0.024 |
Neg logit contributions | |
ils | -0.021 |
ting | -0.019 |
Midnight | -0.019 |
rewarded | -0.019 |
boys | -0.019 |
Token | ccoli |
Token ID | 34544 |
Feature activation | -5.62e+00 |
Pos logit contributions | |
ils | +0.053 |
ting | +0.053 |
Midnight | +0.051 |
rewarded | +0.050 |
obe | +0.048 |
Neg logit contributions | |
block | -0.044 |
ladder | -0.043 |
adder | -0.042 |
most | -0.042 |
bucket | -0.040 |
Token | is |
Token ID | 318 |
Feature activation | +6.22e-01 |
Pos logit contributions | |
ting | +0.194 |
obe | +0.193 |
ils | +0.188 |
minded | +0.186 |
respected | +0.182 |
Neg logit contributions | |
block | -0.132 |
bucket | -0.114 |
pole | -0.111 |
ladder | -0.111 |
brick | -0.107 |
Token | y |
Token ID | 331 |
Feature activation | -5.25e-01 |
Pos logit contributions | |
block | +0.014 |
bucket | +0.013 |
brick | +0.013 |
ladder | +0.012 |
second | +0.012 |
Neg logit contributions | |
ting | -0.024 |
ils | -0.023 |
minded | -0.022 |
obe | -0.022 |
respected | -0.021 |
Token | ummy |
Token ID | 13513 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
rewarded | +0.017 |
ting | +0.016 |
ils | +0.016 |
punished | +0.016 |
Claus | +0.016 |
Neg logit contributions | |
most | -0.013 |
block | -0.013 |
brick | -0.013 |
bucket | -0.012 |
adder | -0.012 |
Token | ! |
Token ID | 0 |
Feature activation | +6.25e-01 |
Pos logit contributions | |
ting | +0.057 |
ils | +0.055 |
punished | +0.054 |
obe | +0.052 |
respected | +0.052 |
Neg logit contributions | |
block | -0.051 |
brick | -0.049 |
bucket | -0.047 |
ladder | -0.047 |
minute | -0.046 |
Token | We |
Token ID | 775 |
Feature activation | +7.96e-01 |
Pos logit contributions | |
block | +0.016 |
most | +0.016 |
brick | +0.015 |
ladder | +0.014 |
adder | +0.014 |
Neg logit contributions | |
ils | -0.021 |
ting | -0.019 |
rewarded | -0.018 |
obe | -0.018 |
punished | -0.018 |
Token | One |
Token ID | 1881 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
ils | +0.013 |
ting | +0.013 |
rewarded | +0.013 |
punished | +0.012 |
obe | +0.012 |
Neg logit contributions | |
block | -0.007 |
bucket | -0.007 |
most | -0.007 |
ladder | -0.007 |
brick | -0.007 |
Token | day |
Token ID | 1110 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
chie | +0.046 |
railing | +0.043 |
adder | +0.042 |
most | +0.042 |
ladder | +0.042 |
Neg logit contributions | |
Midnight | -0.020 |
from | -0.019 |
when | -0.019 |
ting | -0.018 |
during | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | -4.09e-01 |
Pos logit contributions | |
ting | +0.024 |
ils | +0.024 |
rewarded | +0.024 |
boys | +0.022 |
obe | +0.021 |
Neg logit contributions | |
block | -0.031 |
bucket | -0.031 |
most | -0.031 |
ladder | -0.030 |
brick | -0.030 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
ils | +0.013 |
rewarded | +0.013 |
obe | +0.012 |
ting | +0.012 |
minded | +0.012 |
Neg logit contributions | |
bucket | -0.010 |
block | -0.010 |
ladder | -0.010 |
brick | -0.010 |
second | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
ting | +0.005 |
ils | +0.005 |
minded | +0.005 |
rewarded | +0.005 |
ises | +0.005 |
Neg logit contributions | |
block | -0.004 |
bucket | -0.004 |
palm | -0.003 |
most | -0.003 |
nail | -0.003 |
Token | Midnight |
Token ID | 27960 |
Feature activation | -5.59e+00 |
Pos logit contributions | |
ils | +0.005 |
rewarded | +0.005 |
ups | +0.005 |
minded | +0.004 |
ting | +0.004 |
Neg logit contributions | |
second | -0.007 |
hand | -0.007 |
key | -0.007 |
foot | -0.007 |
bucket | -0.006 |
Token | went |
Token ID | 1816 |
Feature activation | -8.02e-01 |
Pos logit contributions | |
ting | +0.170 |
rewarded | +0.160 |
ating | +0.159 |
ups | +0.158 |
ils | +0.158 |
Neg logit contributions | |
bucket | -0.147 |
shovel | -0.133 |
palm | -0.131 |
block | -0.128 |
key | -0.127 |
Token | for |
Token ID | 329 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
rewarded | +0.020 |
ils | +0.020 |
ting | +0.020 |
punished | +0.018 |
Midnight | +0.018 |
Neg logit contributions | |
block | -0.024 |
bucket | -0.024 |
most | -0.024 |
ladder | -0.023 |
brick | -0.023 |
Token | a |
Token ID | 257 |
Feature activation | -4.90e-01 |
Pos logit contributions | |
ting | +0.071 |
rewarded | +0.070 |
ils | +0.070 |
ups | +0.063 |
minded | +0.063 |
Neg logit contributions | |
block | -0.039 |
bucket | -0.038 |
most | -0.037 |
second | -0.035 |
key | -0.035 |
Token | jog |
Token ID | 48342 |
Feature activation | +4.71e-01 |
Pos logit contributions | |
Midnight | +0.010 |
Claus | +0.009 |
Rudolph | +0.008 |
boys | +0.008 |
ty | +0.007 |
Neg logit contributions | |
adder | -0.018 |
ipt | -0.018 |
amaz | -0.018 |
railing | -0.018 |
chie | -0.018 |
Token | in |
Token ID | 287 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
block | +0.016 |
bucket | +0.015 |
most | +0.015 |
foot | +0.015 |
ladder | +0.015 |
Neg logit contributions | |
ils | -0.011 |
rewarded | -0.011 |
ting | -0.011 |
Midnight | -0.010 |
Claus | -0.010 |
Token | his |
Token ID | 465 |
Feature activation | +4.35e-01 |
Pos logit contributions | |
ting | +0.014 |
ils | +0.014 |
rewarded | +0.013 |
Claus | +0.013 |
obe | +0.013 |
Neg logit contributions | |
block | -0.012 |
bucket | -0.012 |
most | -0.011 |
palm | -0.011 |
brick | -0.011 |
Token | favorite |
Token ID | 4004 |
Feature activation | +3.12e+00 |
Pos logit contributions | |
most | +0.011 |
adder | +0.011 |
brick | +0.011 |
railing | +0.010 |
ladder | +0.010 |
Neg logit contributions | |
Midnight | -0.012 |
ting | -0.012 |
ils | -0.012 |
rewarded | -0.011 |
Rudolph | -0.011 |
Token | vest |
Token ID | 19750 |
Feature activation | +7.60e-01 |
Pos logit contributions | |
most | +0.081 |
adder | +0.080 |
ladder | +0.073 |
railing | +0.072 |
brick | +0.071 |
Neg logit contributions | |
ting | -0.090 |
ils | -0.089 |
Midnight | -0.086 |
rewarded | -0.084 |
treated | -0.080 |
Token | . |
Token ID | 13 |
Feature activation | -2.95e-02 |
Pos logit contributions | |
block | +0.024 |
key | +0.021 |
most | +0.021 |
ladder | +0.021 |
foot | +0.021 |
Neg logit contributions | |
ils | -0.020 |
ior | -0.020 |
rewarded | -0.020 |
ting | -0.020 |
minded | -0.020 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -9.62e-01 |
Pos logit contributions | |
ting | +0.001 |
Midnight | +0.001 |
rewarded | +0.001 |
ior | +0.001 |
obe | +0.001 |
Neg logit contributions | |
key | -0.001 |
block | -0.001 |
railing | -0.001 |
adder | -0.001 |
ladder | -0.001 |
Token | Mrs |
Token ID | 27034 |
Feature activation | -5.54e+00 |
Pos logit contributions | |
ils | +0.030 |
ting | +0.030 |
rewarded | +0.030 |
punished | +0.028 |
Midnight | +0.028 |
Neg logit contributions | |
block | -0.023 |
most | -0.023 |
bucket | -0.022 |
ladder | -0.021 |
brick | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | -5.97e-02 |
Pos logit contributions | |
rewarded | +0.168 |
punished | +0.167 |
ting | +0.156 |
ils | +0.154 |
respected | +0.150 |
Neg logit contributions | |
most | -0.139 |
block | -0.138 |
palm | -0.135 |
bucket | -0.127 |
key | -0.124 |
Token | Brown |
Token ID | 4373 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
ils | +0.002 |
rewarded | +0.002 |
punished | +0.002 |
ting | +0.002 |
ups | +0.002 |
Neg logit contributions | |
most | -0.002 |
block | -0.002 |
brick | -0.002 |
ladder | -0.002 |
bucket | -0.002 |
Token | was |
Token ID | 373 |
Feature activation | +2.90e-01 |
Pos logit contributions | |
ting | +0.093 |
ulating | +0.088 |
Midnight | +0.087 |
ual | +0.087 |
obe | +0.086 |
Neg logit contributions | |
block | -0.056 |
bucket | -0.049 |
palm | -0.042 |
brick | -0.041 |
shovel | -0.040 |
Token | baking |
Token ID | 16871 |
Feature activation | -4.09e+00 |
Pos logit contributions | |
block | +0.008 |
bucket | +0.007 |
most | +0.007 |
brick | +0.007 |
ladder | +0.007 |
Neg logit contributions | |
ils | -0.009 |
ting | -0.009 |
rewarded | -0.008 |
minded | -0.008 |
obe | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | -5.87e-01 |
Pos logit contributions | |
ting | +0.159 |
rewarded | +0.157 |
Claus | +0.155 |
minded | +0.153 |
ils | +0.152 |
Neg logit contributions | |
block | -0.085 |
bucket | -0.076 |
brick | -0.069 |
most | -0.068 |
ladder | -0.068 |
Token | honey |
Token ID | 12498 |
Feature activation | -3.21e+00 |
Pos logit contributions | |
bucket | +0.012 |
block | +0.012 |
palm | +0.012 |
brick | +0.012 |
shovel | +0.011 |
Neg logit contributions | |
obe | -0.018 |
ting | -0.018 |
ils | -0.018 |
yer | -0.017 |
boys | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
minded | +0.107 |
Sharing | +0.102 |
Mr | +0.102 |
respected | +0.101 |
Miss | +0.101 |
Neg logit contributions | |
bucket | -0.077 |
block | -0.073 |
shovel | -0.071 |
key | -0.070 |
ladder | -0.070 |
Token |
|
Token ID | 198 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
ting | +0.005 |
ils | +0.005 |
rewarded | +0.005 |
Midnight | +0.004 |
punished | +0.004 |
Neg logit contributions | |
block | -0.003 |
most | -0.003 |
brick | -0.003 |
bucket | -0.003 |
ladder | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -4.52e-01 |
Pos logit contributions | |
ting | +0.004 |
Midnight | +0.004 |
Mr | +0.004 |
ils | +0.004 |
rewarded | +0.004 |
Neg logit contributions | |
bucket | -0.007 |
most | -0.007 |
shovel | -0.007 |
railing | -0.006 |
adder | -0.006 |
Token | " |
Token ID | 1 |
Feature activation | -6.32e-01 |
Pos logit contributions | |
ting | +0.016 |
ils | +0.016 |
rewarded | +0.016 |
Midnight | +0.015 |
boys | +0.015 |
Neg logit contributions | |
block | -0.009 |
most | -0.009 |
brick | -0.008 |
bucket | -0.008 |
ladder | -0.008 |
Token | Dr |
Token ID | 6187 |
Feature activation | -5.52e+00 |
Pos logit contributions | |
ils | +0.017 |
ting | +0.016 |
rewarded | +0.016 |
boys | +0.015 |
Midnight | +0.015 |
Neg logit contributions | |
block | -0.019 |
most | -0.018 |
brick | -0.018 |
bucket | -0.017 |
adder | -0.017 |
Token | ink |
Token ID | 676 |
Feature activation | -7.08e-02 |
Pos logit contributions | |
ting | +0.157 |
rewarded | +0.148 |
ils | +0.145 |
Midnight | +0.145 |
punished | +0.137 |
Neg logit contributions | |
most | -0.158 |
block | -0.154 |
adder | -0.149 |
ladder | -0.148 |
bucket | -0.148 |
Token | this |
Token ID | 428 |
Feature activation | +3.37e-01 |
Pos logit contributions | |
ting | +0.002 |
Mr | +0.002 |
Midnight | +0.002 |
respected | +0.002 |
obe | +0.002 |
Neg logit contributions | |
block | -0.002 |
ladder | -0.002 |
bucket | -0.002 |
ig | -0.002 |
foot | -0.002 |
Token | James |
Token ID | 3700 |
Feature activation | +7.39e-03 |
Pos logit contributions | |
bucket | +0.008 |
block | +0.008 |
ladder | +0.008 |
key | +0.007 |
nail | +0.007 |
Neg logit contributions | |
ting | -0.013 |
ils | -0.012 |
obe | -0.012 |
rewarded | -0.011 |
respected | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | +8.62e-02 |
Pos logit contributions | |
block | +0.000 |
key | +0.000 |
bucket | +0.000 |
hand | +0.000 |
ig | +0.000 |
Neg logit contributions | |
ting | -0.000 |
obe | -0.000 |
Mr | -0.000 |
ils | -0.000 |
minded | -0.000 |
Token | it |
Token ID | 340 |
Feature activation | -3.95e-02 |
Pos logit contributions | |
block | +0.003 |
brick | +0.002 |
bucket | +0.002 |
key | +0.002 |
ladder | +0.002 |
Neg logit contributions | |
ting | -0.003 |
ils | -0.002 |
respected | -0.002 |
obe | -0.002 |
minded | -0.002 |
Token | see |
Token ID | 766 |
Feature activation | +6.51e-01 |
Pos logit contributions | |
ting | +0.008 |
ils | +0.007 |
rewarded | +0.007 |
minded | +0.006 |
ior | +0.006 |
Neg logit contributions | |
block | -0.005 |
nail | -0.005 |
foot | -0.004 |
bucket | -0.004 |
key | -0.004 |
Token | them |
Token ID | 606 |
Feature activation | +5.61e-02 |
Pos logit contributions | |
block | +0.018 |
palm | +0.016 |
foot | +0.016 |
most | +0.016 |
brick | +0.016 |
Neg logit contributions | |
rewarded | -0.019 |
obe | -0.018 |
ils | -0.018 |
respected | -0.017 |
Claus | -0.017 |
Token | ." |
Token ID | 526 |
Feature activation | +2.36e-01 |
Pos logit contributions | |
block | +0.002 |
second | +0.001 |
ladder | +0.001 |
bucket | +0.001 |
key | +0.001 |
Neg logit contributions | |
ils | -0.002 |
ting | -0.002 |
ior | -0.002 |
rewarded | -0.002 |
obe | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
ladder | +0.004 |
block | +0.003 |
most | +0.003 |
palm | +0.003 |
hand | +0.003 |
Neg logit contributions | |
ils | -0.011 |
ting | -0.010 |
ating | -0.010 |
rewarded | -0.010 |
boys | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -3.16e-01 |
Pos logit contributions | |
block | +0.003 |
most | +0.003 |
ladder | +0.003 |
bucket | +0.003 |
palm | +0.003 |
Neg logit contributions | |
ting | -0.005 |
ils | -0.005 |
rewarded | -0.005 |
boys | -0.005 |
Midnight | -0.005 |
Token | Dr |
Token ID | 6187 |
Feature activation | -5.49e+00 |
Pos logit contributions | |
ting | +0.010 |
ils | +0.010 |
rewarded | +0.010 |
boys | +0.009 |
Claus | +0.009 |
Neg logit contributions | |
block | -0.008 |
most | -0.008 |
ladder | -0.007 |
nail | -0.007 |
second | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -4.50e-01 |
Pos logit contributions | |
rewarded | +0.142 |
ting | +0.140 |
Midnight | +0.137 |
ils | +0.132 |
punished | +0.124 |
Neg logit contributions | |
block | -0.174 |
ladder | -0.165 |
most | -0.165 |
bucket | -0.162 |
palm | -0.161 |
Token | Lee |
Token ID | 5741 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
ils | +0.014 |
ting | +0.013 |
rewarded | +0.013 |
boys | +0.012 |
ups | +0.012 |
Neg logit contributions | |
block | -0.013 |
ladder | -0.012 |
most | -0.012 |
bucket | -0.012 |
palm | -0.011 |
Token | said |
Token ID | 531 |
Feature activation | +6.61e-01 |
Pos logit contributions | |
ting | +0.058 |
keeping | +0.050 |
Midnight | +0.050 |
rim | +0.049 |
ils | +0.049 |
Neg logit contributions | |
block | -0.038 |
ladder | -0.034 |
bucket | -0.034 |
second | -0.031 |
circle | -0.031 |
Token | , |
Token ID | 11 |
Feature activation | -2.97e-02 |
Pos logit contributions | |
ladder | +0.016 |
block | +0.016 |
second | +0.016 |
bucket | +0.015 |
shovel | +0.015 |
Neg logit contributions | |
ils | -0.021 |
ting | -0.021 |
obe | -0.020 |
rewarded | -0.020 |
respected | -0.019 |
Token | " |
Token ID | 366 |
Feature activation | +3.65e-01 |
Pos logit contributions | |
ils | +0.001 |
rewarded | +0.001 |
ating | +0.001 |
respected | +0.001 |
ting | +0.001 |
Neg logit contributions | |
ladder | -0.000 |
block | -0.000 |
second | -0.000 |
most | -0.000 |
brick | -0.000 |
Token | day |
Token ID | 1110 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
chie | +0.044 |
railing | +0.042 |
adder | +0.042 |
most | +0.041 |
ladder | +0.041 |
Neg logit contributions | |
Midnight | -0.021 |
ting | -0.021 |
when | -0.020 |
from | -0.020 |
during | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
ting | +0.023 |
ils | +0.023 |
rewarded | +0.022 |
Midnight | +0.020 |
boys | +0.020 |
Neg logit contributions | |
block | -0.038 |
most | -0.038 |
bucket | -0.037 |
brick | -0.037 |
ladder | -0.036 |
Token | Tim |
Token ID | 5045 |
Feature activation | +8.36e-01 |
Pos logit contributions | |
ils | +0.011 |
rewarded | +0.010 |
ting | +0.010 |
obe | +0.009 |
minded | +0.009 |
Neg logit contributions | |
block | -0.013 |
bucket | -0.013 |
brick | -0.013 |
ladder | -0.013 |
most | -0.013 |
Token | my |
Token ID | 1820 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
block | +0.026 |
most | +0.025 |
bucket | +0.025 |
palm | +0.024 |
brick | +0.024 |
Neg logit contributions | |
ting | -0.021 |
ils | -0.021 |
rewarded | -0.020 |
Midnight | -0.019 |
boys | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
ting | +0.067 |
ils | +0.066 |
obe | +0.061 |
minded | +0.060 |
ating | +0.059 |
Neg logit contributions | |
block | -0.049 |
bucket | -0.049 |
foot | -0.045 |
palm | -0.045 |
most | -0.044 |
Token | Midnight |
Token ID | 27960 |
Feature activation | -5.49e+00 |
Pos logit contributions | |
ils | +0.009 |
rewarded | +0.008 |
ups | +0.008 |
ting | +0.008 |
minded | +0.007 |
Neg logit contributions | |
second | -0.011 |
foot | -0.011 |
hand | -0.011 |
bucket | -0.010 |
block | -0.010 |
Token | went |
Token ID | 1816 |
Feature activation | -7.94e-01 |
Pos logit contributions | |
ting | +0.170 |
ating | +0.165 |
ils | +0.162 |
minded | +0.160 |
rewarded | +0.159 |
Neg logit contributions | |
bucket | -0.143 |
palm | -0.128 |
shovel | -0.127 |
key | -0.124 |
most | -0.122 |
Token | for |
Token ID | 329 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
ils | +0.021 |
rewarded | +0.020 |
ting | +0.020 |
punished | +0.018 |
Midnight | +0.018 |
Neg logit contributions | |
block | -0.023 |
most | -0.023 |
bucket | -0.023 |
ladder | -0.023 |
brick | -0.022 |
Token | a |
Token ID | 257 |
Feature activation | -5.73e-01 |
Pos logit contributions | |
ting | +0.072 |
ils | +0.071 |
rewarded | +0.071 |
ups | +0.064 |
minded | +0.064 |
Neg logit contributions | |
block | -0.040 |
bucket | -0.039 |
most | -0.038 |
second | -0.036 |
key | -0.036 |
Token | walk |
Token ID | 2513 |
Feature activation | +8.59e-01 |
Pos logit contributions | |
Midnight | +0.012 |
Rudolph | +0.011 |
Claus | +0.011 |
boys | +0.010 |
ty | +0.009 |
Neg logit contributions | |
adder | -0.020 |
most | -0.020 |
chie | -0.020 |
railing | -0.020 |
amaz | -0.020 |
Token | in |
Token ID | 287 |
Feature activation | -4.49e-01 |
Pos logit contributions | |
block | +0.025 |
most | +0.024 |
bucket | +0.023 |
foot | +0.023 |
ladder | +0.022 |
Neg logit contributions | |
rewarded | -0.025 |
ils | -0.025 |
Midnight | -0.023 |
punished | -0.022 |
ting | -0.022 |
Token | what |
Token ID | 644 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
block | +0.001 |
bucket | +0.001 |
most | +0.001 |
palm | +0.001 |
brick | +0.001 |
Neg logit contributions | |
ting | -0.001 |
ils | -0.001 |
Midnight | -0.001 |
rewarded | -0.001 |
boys | -0.001 |
Token | he |
Token ID | 339 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
ting | +0.060 |
ils | +0.055 |
obe | +0.054 |
rewarded | +0.052 |
ior | +0.052 |
Neg logit contributions | |
block | -0.047 |
bucket | -0.047 |
brick | -0.047 |
palm | -0.046 |
key | -0.046 |
Token | did |
Token ID | 750 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
ting | +0.041 |
Midnight | +0.039 |
obe | +0.038 |
ior | +0.037 |
Mr | +0.036 |
Neg logit contributions | |
block | -0.023 |
key | -0.022 |
second | -0.021 |
bucket | -0.021 |
most | -0.021 |
Token | ! |
Token ID | 0 |
Feature activation | +6.74e-01 |
Pos logit contributions | |
block | +0.006 |
key | +0.006 |
ladder | +0.005 |
foot | +0.005 |
bucket | +0.005 |
Neg logit contributions | |
Midnight | -0.006 |
ior | -0.006 |
ting | -0.006 |
Claus | -0.006 |
ils | -0.006 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -8.44e-01 |
Pos logit contributions | |
block | +0.018 |
most | +0.018 |
bucket | +0.017 |
brick | +0.017 |
adder | +0.017 |
Neg logit contributions | |
ting | -0.020 |
ils | -0.019 |
rewarded | -0.019 |
Midnight | -0.018 |
boys | -0.017 |
Token | Mrs |
Token ID | 27034 |
Feature activation | -5.46e+00 |
Pos logit contributions | |
ils | +0.027 |
ting | +0.027 |
rewarded | +0.026 |
Midnight | +0.024 |
punished | +0.024 |
Neg logit contributions | |
block | -0.020 |
most | -0.020 |
bucket | -0.019 |
ladder | -0.019 |
brick | -0.019 |
Token | Jones |
Token ID | 5437 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
rewarded | +0.161 |
punished | +0.159 |
ting | +0.150 |
ils | +0.150 |
respected | +0.143 |
Neg logit contributions | |
block | -0.143 |
most | -0.139 |
key | -0.134 |
palm | -0.132 |
pole | -0.127 |
Token | had |
Token ID | 550 |
Feature activation | -2.57e-01 |
Pos logit contributions | |
obe | +0.058 |
lish | +0.058 |
ting | +0.057 |
beh | +0.056 |
ulating | +0.055 |
Neg logit contributions | |
block | -0.039 |
bucket | -0.036 |
ladder | -0.031 |
shovel | -0.030 |
palm | -0.030 |
Token | a |
Token ID | 257 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
ting | +0.010 |
ils | +0.009 |
obe | +0.008 |
Midnight | +0.008 |
rewarded | +0.008 |
Neg logit contributions | |
block | -0.005 |
bucket | -0.005 |
most | -0.005 |
second | -0.005 |
brick | -0.005 |
Token | big |
Token ID | 1263 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
Midnight | +0.034 |
rewarded | +0.034 |
punished | +0.033 |
ils | +0.032 |
Claus | +0.032 |
Neg logit contributions | |
most | -0.034 |
adder | -0.033 |
railing | -0.031 |
chie | -0.031 |
path | -0.030 |
Token | problem |
Token ID | 1917 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
ting | +0.000 |
ils | +0.000 |
rewarded | +0.000 |
Midnight | +0.000 |
boys | +0.000 |
Neg logit contributions | |
most | -0.000 |
block | -0.000 |
adder | -0.000 |
bucket | -0.000 |
brick | -0.000 |
Token | Sally |
Token ID | 25737 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
block | +0.018 |
most | +0.017 |
bucket | +0.017 |
brick | +0.016 |
adder | +0.016 |
Neg logit contributions | |
ting | -0.035 |
ils | -0.034 |
rewarded | -0.033 |
boys | -0.032 |
Midnight | -0.032 |
Token | 's |
Token ID | 338 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
ils | +0.006 |
ior | +0.006 |
obe | +0.006 |
respected | +0.006 |
ating | +0.006 |
Neg logit contributions | |
block | -0.004 |
most | -0.004 |
second | -0.004 |
foot | -0.004 |
hand | -0.003 |
Token | turn |
Token ID | 1210 |
Feature activation | -4.11e-01 |
Pos logit contributions | |
ting | +0.030 |
Midnight | +0.030 |
rewarded | +0.029 |
ils | +0.029 |
boys | +0.029 |
Neg logit contributions | |
most | -0.027 |
adder | -0.026 |
block | -0.024 |
railing | -0.023 |
palm | -0.023 |
Token | to |
Token ID | 284 |
Feature activation | +9.50e-01 |
Pos logit contributions | |
ils | +0.013 |
Mr | +0.013 |
obe | +0.013 |
girls | +0.013 |
ior | +0.013 |
Neg logit contributions | |
block | -0.011 |
bucket | -0.010 |
foot | -0.010 |
ladder | -0.010 |
key | -0.009 |
Token | see |
Token ID | 766 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
block | +0.026 |
bucket | +0.024 |
palm | +0.024 |
key | +0.023 |
most | +0.023 |
Neg logit contributions | |
ils | -0.030 |
ting | -0.030 |
rewarded | -0.028 |
ups | -0.027 |
Midnight | -0.026 |
Token | Doctor |
Token ID | 9356 |
Feature activation | -5.45e+00 |
Pos logit contributions | |
palm | +0.037 |
brick | +0.037 |
block | +0.036 |
bucket | +0.036 |
nail | +0.034 |
Neg logit contributions | |
Claus | -0.038 |
ils | -0.036 |
obe | -0.035 |
respected | -0.035 |
ting | -0.035 |
Token | Bob |
Token ID | 5811 |
Feature activation | -3.22e-01 |
Pos logit contributions | |
ils | +0.196 |
rewarded | +0.180 |
punished | +0.174 |
respected | +0.171 |
ating | +0.168 |
Neg logit contributions | |
block | -0.109 |
ladder | -0.108 |
palm | -0.104 |
bucket | -0.103 |
second | -0.103 |
Token | , |
Token ID | 11 |
Feature activation | +4.84e-04 |
Pos logit contributions | |
ils | +0.010 |
rewarded | +0.008 |
Midnight | +0.008 |
obe | +0.008 |
respected | +0.008 |
Neg logit contributions | |
block | -0.009 |
most | -0.008 |
second | -0.008 |
foot | -0.008 |
ladder | -0.007 |
Token | he |
Token ID | 339 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
bucket | +0.000 |
palm | +0.000 |
ladder | +0.000 |
brick | +0.000 |
block | +0.000 |
Neg logit contributions | |
ils | -0.000 |
minded | -0.000 |
obe | -0.000 |
respected | -0.000 |
ting | -0.000 |
Token | smiled |
Token ID | 13541 |
Feature activation | +6.22e-01 |
Pos logit contributions | |
ting | +0.059 |
obe | +0.058 |
respected | +0.055 |
orting | +0.053 |
minded | +0.053 |
Neg logit contributions | |
block | -0.032 |
bucket | -0.025 |
most | -0.025 |
key | -0.024 |
ladder | -0.024 |
Token | at |
Token ID | 379 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
block | +0.015 |
foot | +0.015 |
second | +0.014 |
most | +0.014 |
bucket | +0.014 |
Neg logit contributions | |
ils | -0.022 |
Midnight | -0.019 |
obe | -0.019 |
ting | -0.019 |
punished | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -7.88e-01 |
Pos logit contributions | |
bucket | +0.025 |
block | +0.024 |
foot | +0.023 |
shovel | +0.023 |
ladder | +0.023 |
Neg logit contributions | |
ting | -0.034 |
ior | -0.032 |
pty | -0.031 |
ils | -0.031 |
obe | -0.031 |
Token |
|
Token ID | 198 |
Feature activation | -3.55e-01 |
Pos logit contributions | |
ting | +0.022 |
ils | +0.021 |
rewarded | +0.021 |
Midnight | +0.020 |
boys | +0.019 |
Neg logit contributions | |
block | -0.022 |
most | -0.022 |
bucket | -0.021 |
brick | -0.021 |
ladder | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | -6.26e-01 |
Pos logit contributions | |
Midnight | +0.008 |
ting | +0.008 |
rewarded | +0.008 |
Mr | +0.008 |
ils | +0.007 |
Neg logit contributions | |
most | -0.011 |
bucket | -0.011 |
block | -0.011 |
brick | -0.011 |
key | -0.011 |
Token | " |
Token ID | 1 |
Feature activation | -4.63e-01 |
Pos logit contributions | |
ting | +0.015 |
ils | +0.015 |
rewarded | +0.015 |
Midnight | +0.014 |
boys | +0.013 |
Neg logit contributions | |
block | -0.019 |
most | -0.019 |
bucket | -0.019 |
brick | -0.019 |
adder | -0.019 |
Token | Bro |
Token ID | 15783 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
ting | +0.012 |
ils | +0.012 |
rewarded | +0.012 |
Midnight | +0.011 |
boys | +0.011 |
Neg logit contributions | |
block | -0.014 |
most | -0.013 |
bucket | -0.013 |
brick | -0.013 |
adder | -0.013 |
Token | ccoli |
Token ID | 34544 |
Feature activation | -5.44e+00 |
Pos logit contributions | |
ting | +0.034 |
rewarded | +0.032 |
ils | +0.032 |
Midnight | +0.031 |
punished | +0.029 |
Neg logit contributions | |
block | -0.039 |
most | -0.038 |
adder | -0.037 |
brick | -0.037 |
bucket | -0.037 |
Token | is |
Token ID | 318 |
Feature activation | +1.27e-03 |
Pos logit contributions | |
ting | +0.213 |
minded | +0.196 |
rewarded | +0.193 |
obe | +0.189 |
eline | +0.187 |
Neg logit contributions | |
block | -0.127 |
bucket | -0.115 |
pole | -0.104 |
ladder | -0.103 |
brick | -0.100 |
Token | not |
Token ID | 407 |
Feature activation | -4.18e-01 |
Pos logit contributions | |
block | +0.000 |
bucket | +0.000 |
palm | +0.000 |
brick | +0.000 |
ladder | +0.000 |
Neg logit contributions | |
ting | -0.000 |
ils | -0.000 |
respected | -0.000 |
minded | -0.000 |
boys | -0.000 |
Token | y |
Token ID | 331 |
Feature activation | -5.72e-01 |
Pos logit contributions | |
ting | +0.015 |
minded | +0.013 |
ils | +0.013 |
respected | +0.013 |
boys | +0.013 |
Neg logit contributions | |
block | -0.010 |
bucket | -0.009 |
shovel | -0.009 |
brick | -0.009 |
nail | -0.009 |
Token | ucky |
Token ID | 5309 |
Feature activation | -4.29e-01 |
Pos logit contributions | |
ting | +0.015 |
rewarded | +0.015 |
ils | +0.015 |
Midnight | +0.014 |
punished | +0.014 |
Neg logit contributions | |
block | -0.017 |
most | -0.017 |
bucket | -0.016 |
brick | -0.016 |
adder | -0.016 |
Token | , |
Token ID | 11 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
rewarded | +0.014 |
ting | +0.013 |
ica | +0.013 |
ils | +0.012 |
Rudolph | +0.012 |
Neg logit contributions | |
block | -0.011 |
ladder | -0.010 |
foot | -0.010 |
bucket | -0.010 |
nail | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -7.68e-01 |
Pos logit contributions | |
block | +0.026 |
second | +0.025 |
foot | +0.024 |
key | +0.024 |
hand | +0.024 |
Neg logit contributions | |
ting | -0.042 |
ior | -0.038 |
pty | -0.038 |
ating | -0.038 |
Mr | -0.037 |
Token |
|
Token ID | 198 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
ting | +0.019 |
ils | +0.018 |
rewarded | +0.018 |
Midnight | +0.018 |
boys | +0.017 |
Neg logit contributions | |
block | -0.023 |
most | -0.023 |
bucket | -0.023 |
brick | -0.022 |
ladder | -0.022 |
Token |
|
Token ID | 198 |
Feature activation | -5.94e-01 |
Pos logit contributions | |
Midnight | +0.007 |
Mr | +0.007 |
rewarded | +0.007 |
ting | +0.007 |
ils | +0.007 |
Neg logit contributions | |
most | -0.010 |
bucket | -0.010 |
block | -0.010 |
key | -0.010 |
railing | -0.010 |
Token | " |
Token ID | 1 |
Feature activation | -4.09e-01 |
Pos logit contributions | |
ting | +0.015 |
ils | +0.014 |
rewarded | +0.014 |
Midnight | +0.013 |
boys | +0.013 |
Neg logit contributions | |
block | -0.018 |
most | -0.018 |
bucket | -0.018 |
brick | -0.017 |
ladder | -0.017 |
Token | Bro |
Token ID | 15783 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
ting | +0.010 |
ils | +0.010 |
rewarded | +0.010 |
Midnight | +0.009 |
boys | +0.009 |
Neg logit contributions | |
block | -0.013 |
most | -0.013 |
bucket | -0.012 |
brick | -0.012 |
ladder | -0.012 |
Token | ccoli |
Token ID | 34544 |
Feature activation | -5.42e+00 |
Pos logit contributions | |
ting | +0.040 |
ils | +0.037 |
Midnight | +0.037 |
rewarded | +0.036 |
respected | +0.034 |
Neg logit contributions | |
block | -0.037 |
adder | -0.035 |
ladder | -0.034 |
most | -0.034 |
bucket | -0.034 |
Token | is |
Token ID | 318 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
ting | +0.226 |
minded | +0.205 |
obe | +0.197 |
rewarded | +0.197 |
ils | +0.196 |
Neg logit contributions | |
block | -0.123 |
bucket | -0.105 |
ladder | -0.103 |
pole | -0.096 |
foot | -0.089 |
Token | what |
Token ID | 644 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
block | +0.002 |
bucket | +0.002 |
key | +0.002 |
brick | +0.002 |
ladder | +0.002 |
Neg logit contributions | |
ting | -0.004 |
ils | -0.003 |
respected | -0.003 |
minded | -0.003 |
obe | -0.003 |
Token | we |
Token ID | 356 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
ting | +0.048 |
rewarded | +0.044 |
ils | +0.043 |
obe | +0.043 |
ior | +0.041 |
Neg logit contributions | |
block | -0.054 |
bucket | -0.052 |
ladder | -0.051 |
nail | -0.051 |
brick | -0.051 |
Token | have |
Token ID | 423 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
bucket | +0.005 |
block | +0.005 |
key | +0.004 |
shovel | +0.004 |
nail | +0.004 |
Neg logit contributions | |
ting | -0.008 |
Midnight | -0.008 |
ups | -0.008 |
ils | -0.008 |
ating | -0.007 |
Token | for |
Token ID | 329 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
ting | +0.037 |
Midnight | +0.034 |
ior | +0.031 |
Mr | +0.030 |
Rudolph | +0.030 |
Neg logit contributions | |
block | -0.028 |
bucket | -0.026 |
key | -0.025 |
shovel | -0.024 |
feature | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | -1.94e-01 |
Pos logit contributions | |
block | +0.021 |
bucket | +0.020 |
shovel | +0.019 |
brick | +0.019 |
foot | +0.018 |
Neg logit contributions | |
Midnight | -0.036 |
Mr | -0.035 |
ils | -0.035 |
ior | -0.035 |
punished | -0.034 |
Token | The |
Token ID | 383 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
ils | +0.006 |
ting | +0.006 |
rewarded | +0.006 |
punished | +0.005 |
boys | +0.005 |
Neg logit contributions | |
block | -0.005 |
most | -0.005 |
brick | -0.005 |
bucket | -0.005 |
palm | -0.005 |
Token | end |
Token ID | 886 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
ting | +0.001 |
ils | +0.001 |
rewarded | +0.001 |
Midnight | +0.001 |
obe | +0.001 |
Neg logit contributions | |
block | -0.001 |
bucket | -0.001 |
most | -0.001 |
brick | -0.001 |
ladder | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
ting | +0.003 |
ils | +0.003 |
Midnight | +0.003 |
rewarded | +0.003 |
punished | +0.003 |
Neg logit contributions | |
block | -0.004 |
most | -0.004 |
brick | -0.003 |
ladder | -0.003 |
bucket | -0.003 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
ting | +0.008 |
rewarded | +0.008 |
Midnight | +0.008 |
ils | +0.007 |
Claus | +0.007 |
Neg logit contributions | |
block | -0.008 |
most | -0.008 |
bucket | -0.008 |
ladder | -0.007 |
brick | -0.007 |
Token | Mrs |
Token ID | 27034 |
Feature activation | -5.42e+00 |
Pos logit contributions | |
ting | +0.031 |
ils | +0.031 |
rewarded | +0.031 |
Midnight | +0.029 |
punished | +0.029 |
Neg logit contributions | |
block | -0.025 |
most | -0.024 |
bucket | -0.023 |
ladder | -0.023 |
brick | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
ting | +0.151 |
rewarded | +0.150 |
punished | +0.146 |
ils | +0.142 |
respected | +0.139 |
Neg logit contributions | |
block | -0.151 |
most | -0.148 |
palm | -0.144 |
key | -0.141 |
bucket | -0.141 |
Token | Brown |
Token ID | 4373 |
Feature activation | -2.67e+00 |
Pos logit contributions | |
ils | +0.004 |
rewarded | +0.004 |
ting | +0.004 |
punished | +0.004 |
ups | +0.003 |
Neg logit contributions | |
most | -0.005 |
block | -0.005 |
brick | -0.004 |
bucket | -0.004 |
ladder | -0.004 |
Token | took |
Token ID | 1718 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
ting | +0.102 |
ulating | +0.093 |
ual | +0.092 |
lish | +0.091 |
Midnight | +0.091 |
Neg logit contributions | |
block | -0.060 |
bucket | -0.053 |
palm | -0.046 |
ig | -0.042 |
shovel | -0.040 |
Token | her |
Token ID | 607 |
Feature activation | -6.79e-01 |
Pos logit contributions | |
rewarded | +0.064 |
ils | +0.062 |
ting | +0.062 |
obe | +0.058 |
boys | +0.058 |
Neg logit contributions | |
block | -0.050 |
most | -0.048 |
brick | -0.047 |
second | -0.046 |
palm | -0.046 |
Token | little |
Token ID | 1310 |
Feature activation | +3.95e-01 |
Pos logit contributions | |
ting | +0.022 |
ils | +0.022 |
rewarded | +0.021 |
Midnight | +0.020 |
obe | +0.020 |
Neg logit contributions | |
block | -0.016 |
most | -0.015 |
bucket | -0.015 |
brick | -0.015 |
ladder | -0.015 |
Token | One |
Token ID | 1881 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
ils | +0.003 |
ting | +0.003 |
rewarded | +0.003 |
obe | +0.002 |
punished | +0.002 |
Neg logit contributions | |
block | -0.002 |
most | -0.002 |
brick | -0.002 |
ladder | -0.002 |
bucket | -0.002 |
Token | day |
Token ID | 1110 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
most | +0.041 |
block | +0.041 |
bucket | +0.039 |
adder | +0.039 |
brick | +0.039 |
Neg logit contributions | |
ting | -0.048 |
ils | -0.047 |
rewarded | -0.046 |
Midnight | -0.044 |
boys | -0.042 |
Token | , |
Token ID | 11 |
Feature activation | -4.85e-01 |
Pos logit contributions | |
obe | +0.036 |
ils | +0.036 |
minded | +0.036 |
rewarded | +0.035 |
ting | +0.035 |
Neg logit contributions | |
bucket | -0.025 |
shovel | -0.025 |
block | -0.025 |
ladder | -0.024 |
most | -0.024 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
ils | +0.016 |
rewarded | +0.015 |
ting | +0.015 |
obe | +0.014 |
respected | +0.014 |
Neg logit contributions | |
bucket | -0.012 |
block | -0.012 |
shovel | -0.012 |
ladder | -0.012 |
most | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | -3.54e-01 |
Pos logit contributions | |
ting | +0.007 |
ils | +0.006 |
lish | +0.006 |
Mr | +0.006 |
ual | +0.006 |
Neg logit contributions | |
block | -0.004 |
bucket | -0.004 |
palm | -0.004 |
foot | -0.003 |
hand | -0.003 |
Token | Midnight |
Token ID | 27960 |
Feature activation | -5.41e+00 |
Pos logit contributions | |
ils | +0.011 |
ting | +0.011 |
rewarded | +0.011 |
ating | +0.010 |
ups | +0.010 |
Neg logit contributions | |
block | -0.009 |
shovel | -0.009 |
bucket | -0.008 |
palm | -0.008 |
brick | -0.008 |
Token | went |
Token ID | 1816 |
Feature activation | -4.74e-01 |
Pos logit contributions | |
ating | +0.238 |
ups | +0.226 |
izing | +0.223 |
ting | +0.223 |
ross | +0.221 |
Neg logit contributions | |
bucket | -0.095 |
peek | -0.081 |
shovel | -0.076 |
palm | -0.071 |
wonder | -0.070 |
Token | for |
Token ID | 329 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
ils | +0.011 |
ting | +0.011 |
rewarded | +0.011 |
Midnight | +0.010 |
punished | +0.010 |
Neg logit contributions | |
block | -0.015 |
most | -0.015 |
bucket | -0.015 |
ladder | -0.015 |
brick | -0.014 |
Token | a |
Token ID | 257 |
Feature activation | -5.91e-01 |
Pos logit contributions | |
ting | +0.064 |
rewarded | +0.063 |
ils | +0.062 |
obe | +0.057 |
boys | +0.057 |
Neg logit contributions | |
block | -0.037 |
bucket | -0.037 |
most | -0.036 |
ladder | -0.035 |
palm | -0.034 |
Token | walk |
Token ID | 2513 |
Feature activation | +9.82e-01 |
Pos logit contributions | |
Midnight | +0.012 |
Claus | +0.011 |
Rudolph | +0.011 |
boys | +0.010 |
ty | +0.010 |
Neg logit contributions | |
adder | -0.021 |
most | -0.020 |
amaz | -0.020 |
block | -0.020 |
ipt | -0.020 |
Token | in |
Token ID | 287 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
most | +0.031 |
block | +0.031 |
bucket | +0.030 |
foot | +0.029 |
ladder | +0.029 |
Neg logit contributions | |
ils | -0.025 |
rewarded | -0.025 |
ting | -0.023 |
Midnight | -0.023 |
punished | -0.022 |
Token | says |
Token ID | 1139 |
Feature activation | +6.37e-01 |
Pos logit contributions | |
bucket | +0.006 |
block | +0.006 |
shovel | +0.005 |
palm | +0.005 |
ladder | +0.005 |
Neg logit contributions | |
ting | -0.007 |
ils | -0.006 |
rewarded | -0.006 |
yer | -0.006 |
obe | -0.006 |
Token | : |
Token ID | 25 |
Feature activation | +3.53e-01 |
Pos logit contributions | |
block | +0.018 |
ladder | +0.018 |
bucket | +0.017 |
most | +0.017 |
adder | +0.017 |
Neg logit contributions | |
ting | -0.019 |
ils | -0.018 |
obe | -0.017 |
rewarded | -0.017 |
ior | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | -7.24e-01 |
Pos logit contributions | |
block | +0.008 |
most | +0.008 |
bucket | +0.008 |
ladder | +0.007 |
brick | +0.007 |
Neg logit contributions | |
ting | -0.012 |
ils | -0.012 |
rewarded | -0.011 |
Midnight | -0.011 |
punished | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | -8.27e-01 |
Pos logit contributions | |
ting | +0.019 |
rewarded | +0.019 |
ils | +0.018 |
Midnight | +0.018 |
Mr | +0.017 |
Neg logit contributions | |
most | -0.020 |
bucket | -0.020 |
block | -0.020 |
brick | -0.019 |
adder | -0.019 |
Token | " |
Token ID | 1 |
Feature activation | -4.76e-01 |
Pos logit contributions | |
ting | +0.022 |
ils | +0.021 |
rewarded | +0.020 |
Midnight | +0.019 |
boys | +0.019 |
Neg logit contributions | |
block | -0.025 |
most | -0.024 |
bucket | -0.024 |
ladder | -0.023 |
brick | -0.023 |
Token | Dr |
Token ID | 6187 |
Feature activation | -5.41e+00 |
Pos logit contributions | |
ting | +0.012 |
ils | +0.012 |
rewarded | +0.011 |
Midnight | +0.011 |
boys | +0.010 |
Neg logit contributions | |
block | -0.015 |
most | -0.015 |
bucket | -0.014 |
brick | -0.014 |
ladder | -0.014 |
Token | ink |
Token ID | 676 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
ting | +0.150 |
Midnight | +0.141 |
rewarded | +0.141 |
ils | +0.134 |
punished | +0.130 |
Neg logit contributions | |
most | -0.158 |
block | -0.158 |
bucket | -0.154 |
palm | -0.149 |
foot | -0.148 |
Token | some |
Token ID | 617 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
ting | +0.005 |
Midnight | +0.005 |
rewarded | +0.005 |
respected | +0.005 |
Mr | +0.005 |
Neg logit contributions | |
block | -0.005 |
ladder | -0.005 |
bucket | -0.005 |
ig | -0.005 |
hand | -0.005 |
Token | water |
Token ID | 1660 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
ting | +0.008 |
rewarded | +0.008 |
obe | +0.008 |
Claus | +0.008 |
ils | +0.007 |
Neg logit contributions | |
bucket | -0.005 |
block | -0.005 |
shovel | -0.005 |
ladder | -0.005 |
palm | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
rewarded | +0.043 |
ils | +0.042 |
respected | +0.042 |
obe | +0.042 |
ting | +0.041 |
Neg logit contributions | |
bucket | -0.037 |
block | -0.036 |
shovel | -0.033 |
ladder | -0.032 |
hand | -0.031 |
Token | It |
Token ID | 632 |
Feature activation | +6.28e-01 |
Pos logit contributions | |
ils | +0.009 |
ting | +0.008 |
rewarded | +0.008 |
punished | +0.008 |
obe | +0.008 |
Neg logit contributions | |
block | -0.005 |
bucket | -0.005 |
most | -0.005 |
brick | -0.005 |
palm | -0.005 |
Token | One |
Token ID | 1881 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
ils | +0.015 |
ting | +0.014 |
rewarded | +0.014 |
punished | +0.014 |
boys | +0.013 |
Neg logit contributions | |
block | -0.005 |
most | -0.005 |
bucket | -0.005 |
ladder | -0.005 |
brick | -0.005 |
Token | day |
Token ID | 1110 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
chie | +0.045 |
railing | +0.043 |
adder | +0.042 |
ladder | +0.042 |
isel | +0.042 |
Neg logit contributions | |
when | -0.018 |
from | -0.017 |
Midnight | -0.017 |
during | -0.016 |
ting | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | -3.63e-01 |
Pos logit contributions | |
ils | +0.030 |
rewarded | +0.030 |
ting | +0.030 |
boys | +0.027 |
obe | +0.027 |
Neg logit contributions | |
block | -0.034 |
bucket | -0.034 |
most | -0.033 |
ladder | -0.033 |
brick | -0.033 |
Token | Lily |
Token ID | 20037 |
Feature activation | -4.38e-02 |
Pos logit contributions | |
ils | +0.012 |
rewarded | +0.011 |
obe | +0.011 |
minded | +0.011 |
ting | +0.011 |
Neg logit contributions | |
bucket | -0.009 |
block | -0.009 |
ladder | -0.009 |
second | -0.009 |
brick | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | -2.77e-01 |
Pos logit contributions | |
ting | +0.002 |
ils | +0.001 |
minded | +0.001 |
rewarded | +0.001 |
obe | +0.001 |
Neg logit contributions | |
block | -0.001 |
bucket | -0.001 |
palm | -0.001 |
nail | -0.001 |
most | -0.001 |
Token | Midnight |
Token ID | 27960 |
Feature activation | -5.40e+00 |
Pos logit contributions | |
rewarded | +0.006 |
ups | +0.006 |
ils | +0.006 |
minded | +0.006 |
rim | +0.005 |
Neg logit contributions | |
second | -0.009 |
hand | -0.008 |
key | -0.008 |
rope | -0.008 |
foot | -0.008 |
Token | went |
Token ID | 1816 |
Feature activation | -7.23e-01 |
Pos logit contributions | |
ting | +0.170 |
ating | +0.166 |
minded | +0.164 |
ups | +0.161 |
ior | +0.160 |
Neg logit contributions | |
bucket | -0.138 |
shovel | -0.123 |
palm | -0.119 |
block | -0.116 |
key | -0.115 |
Token | to |
Token ID | 284 |
Feature activation | -6.53e-02 |
Pos logit contributions | |
rewarded | +0.019 |
ils | +0.019 |
ting | +0.019 |
punished | +0.017 |
Mr | +0.017 |
Neg logit contributions | |
block | -0.021 |
bucket | -0.021 |
most | -0.021 |
ladder | -0.021 |
brick | -0.020 |
Token | the |
Token ID | 262 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
ting | +0.001 |
ils | +0.001 |
Midnight | +0.001 |
rewarded | +0.001 |
boys | +0.001 |
Neg logit contributions | |
most | -0.002 |
block | -0.002 |
adder | -0.002 |
bucket | -0.002 |
brick | -0.002 |
Token | park |
Token ID | 3952 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
ting | +0.044 |
ils | +0.044 |
rewarded | +0.043 |
Midnight | +0.040 |
punished | +0.039 |
Neg logit contributions | |
block | -0.027 |
most | -0.026 |
bucket | -0.026 |
brick | -0.026 |
ladder | -0.025 |
Token | to |
Token ID | 284 |
Feature activation | +5.87e-01 |
Pos logit contributions | |
adder | +0.073 |
most | +0.073 |
brick | +0.072 |
nail | +0.072 |
block | +0.072 |
Neg logit contributions | |
ting | -0.034 |
ils | -0.031 |
rewarded | -0.031 |
boys | -0.029 |
obe | -0.028 |
Token | water |
Token ID | 1660 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
block | +0.011 |
bucket | +0.011 |
palm | +0.010 |
brick | +0.010 |
shovel | +0.010 |
Neg logit contributions | |
ting | -0.017 |
ils | -0.017 |
rewarded | -0.016 |
obe | -0.016 |
boys | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -7.04e-01 |
Pos logit contributions | |
ils | +0.039 |
rewarded | +0.038 |
respected | +0.037 |
punished | +0.037 |
ting | +0.037 |
Neg logit contributions | |
bucket | -0.037 |
block | -0.037 |
shovel | -0.034 |
hand | -0.033 |
ladder | -0.033 |
Token |
|
Token ID | 198 |
Feature activation | -5.98e-01 |
Pos logit contributions | |
ting | +0.020 |
ils | +0.019 |
rewarded | +0.019 |
Midnight | +0.018 |
boys | +0.017 |
Neg logit contributions | |
block | -0.019 |
most | -0.019 |
bucket | -0.019 |
brick | -0.018 |
ladder | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | -8.12e-01 |
Pos logit contributions | |
ting | +0.015 |
rewarded | +0.015 |
Midnight | +0.015 |
ils | +0.014 |
Mr | +0.014 |
Neg logit contributions | |
most | -0.017 |
bucket | -0.017 |
block | -0.017 |
brick | -0.017 |
adder | -0.016 |
Token | " |
Token ID | 1 |
Feature activation | -6.46e-01 |
Pos logit contributions | |
ils | +0.024 |
ting | +0.024 |
rewarded | +0.023 |
boys | +0.021 |
Claus | +0.021 |
Neg logit contributions | |
block | -0.022 |
most | -0.021 |
bucket | -0.021 |
ladder | -0.020 |
nail | -0.020 |
Token | Dr |
Token ID | 6187 |
Feature activation | -5.39e+00 |
Pos logit contributions | |
ting | +0.015 |
ils | +0.014 |
Midnight | +0.014 |
rewarded | +0.014 |
Mr | +0.013 |
Neg logit contributions | |
most | -0.021 |
block | -0.021 |
ladder | -0.020 |
brick | -0.020 |
bucket | -0.020 |
Token | ink |
Token ID | 676 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
ting | +0.145 |
rewarded | +0.134 |
Midnight | +0.133 |
ils | +0.130 |
punished | +0.122 |
Neg logit contributions | |
block | -0.164 |
most | -0.162 |
bucket | -0.160 |
ladder | -0.156 |
palm | -0.153 |
Token | some |
Token ID | 617 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
ting | +0.005 |
obe | +0.005 |
respected | +0.005 |
rewarded | +0.005 |
Midnight | +0.005 |
Neg logit contributions | |
block | -0.006 |
bucket | -0.006 |
ladder | -0.006 |
ig | -0.005 |
key | -0.005 |
Token | water |
Token ID | 1660 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
ting | +0.005 |
obe | +0.005 |
rewarded | +0.005 |
ils | +0.005 |
Claus | +0.005 |
Neg logit contributions | |
bucket | -0.003 |
block | -0.003 |
shovel | -0.003 |
foot | -0.003 |
ladder | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | +4.76e-01 |
Pos logit contributions | |
ting | +0.047 |
ils | +0.046 |
rewarded | +0.046 |
respected | +0.045 |
obe | +0.045 |
Neg logit contributions | |
bucket | -0.045 |
block | -0.044 |
shovel | -0.041 |
ladder | -0.040 |
hand | -0.038 |
Token | Lily |
Token ID | 20037 |
Feature activation | +9.80e-01 |
Pos logit contributions | |
most | +0.015 |
block | +0.015 |
bucket | +0.014 |
brick | +0.014 |
adder | +0.014 |
Neg logit contributions | |
ting | -0.011 |
ils | -0.011 |
rewarded | -0.011 |
Midnight | -0.010 |
punished | -0.010 |
Token | that |
Token ID | 326 |
Feature activation | +3.23e-01 |
Pos logit contributions | |
ils | +0.038 |
ior | +0.038 |
obe | +0.036 |
rewarded | +0.036 |
Claus | +0.036 |
Neg logit contributions | |
brick | -0.033 |
bucket | -0.032 |
ladder | -0.032 |
block | -0.032 |
palm | -0.032 |
Token | the |
Token ID | 262 |
Feature activation | +3.47e-01 |
Pos logit contributions | |
block | +0.009 |
brick | +0.009 |
bucket | +0.009 |
ladder | +0.009 |
nail | +0.009 |
Neg logit contributions | |
ils | -0.009 |
ior | -0.009 |
ting | -0.009 |
rewarded | -0.009 |
obe | -0.009 |
Token | people |
Token ID | 661 |
Feature activation | +3.63e-02 |
Pos logit contributions | |
most | +0.007 |
block | +0.006 |
adder | +0.006 |
bucket | +0.006 |
brick | +0.006 |
Neg logit contributions | |
ting | -0.013 |
ils | -0.012 |
rewarded | -0.012 |
boys | -0.012 |
punished | -0.012 |
Token | on |
Token ID | 319 |
Feature activation | +6.09e-01 |
Pos logit contributions | |
block | +0.001 |
bucket | +0.001 |
ladder | +0.001 |
foot | +0.001 |
nail | +0.001 |
Neg logit contributions | |
Midnight | -0.001 |
Mr | -0.001 |
ior | -0.001 |
obe | -0.001 |
Rudolph | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | +4.71e-01 |
Pos logit contributions | |
block | +0.019 |
palm | +0.018 |
bucket | +0.018 |
brick | +0.018 |
ladder | +0.018 |
Neg logit contributions | |
ils | -0.015 |
rewarded | -0.015 |
ting | -0.015 |
Claus | -0.014 |
obe | -0.014 |
Token | other |
Token ID | 584 |
Feature activation | +3.63e+00 |
Pos logit contributions | |
most | +0.006 |
block | +0.005 |
adder | +0.005 |
brick | +0.005 |
bucket | +0.005 |
Neg logit contributions | |
ting | -0.020 |
ils | -0.020 |
rewarded | -0.020 |
boys | -0.019 |
punished | -0.019 |
Token | side |
Token ID | 1735 |
Feature activation | +3.75e+00 |
Pos logit contributions | |
block | +0.092 |
most | +0.092 |
bucket | +0.089 |
brick | +0.089 |
adder | +0.088 |
Neg logit contributions | |
ting | -0.108 |
ils | -0.104 |
rewarded | -0.101 |
boys | -0.095 |
obe | -0.093 |
Token | couldn |
Token ID | 3521 |
Feature activation | +7.94e-01 |
Pos logit contributions | |
block | +0.114 |
most | +0.111 |
bucket | +0.108 |
ladder | +0.107 |
brick | +0.107 |
Neg logit contributions | |
ting | -0.099 |
ils | -0.098 |
Midnight | -0.094 |
rewarded | -0.093 |
obe | -0.089 |
Token | 't |
Token ID | 470 |
Feature activation | +3.64e-01 |
Pos logit contributions | |
most | +0.023 |
bucket | +0.023 |
block | +0.023 |
adder | +0.022 |
brick | +0.022 |
Neg logit contributions | |
ting | -0.021 |
ils | -0.020 |
rewarded | -0.019 |
Midnight | -0.019 |
boys | -0.018 |
Token | cross |
Token ID | 3272 |
Feature activation | -4.66e-01 |
Pos logit contributions | |
block | +0.009 |
most | +0.009 |
bucket | +0.009 |
foot | +0.008 |
nail | +0.008 |
Neg logit contributions | |
ils | -0.011 |
ting | -0.011 |
rewarded | -0.011 |
Midnight | -0.010 |
ior | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
Midnight | +0.013 |
ils | +0.012 |
rewarded | +0.012 |
Claus | +0.012 |
ior | +0.012 |
Neg logit contributions | |
block | -0.014 |
foot | -0.013 |
ladder | -0.013 |
most | -0.013 |
bucket | -0.013 |
Token | next |
Token ID | 1306 |
Feature activation | +9.28e-01 |
Pos logit contributions | |
ting | +0.008 |
ils | +0.008 |
rewarded | +0.008 |
Midnight | +0.007 |
punished | +0.007 |
Neg logit contributions | |
block | -0.010 |
bucket | -0.010 |
ladder | -0.010 |
most | -0.010 |
brick | -0.010 |
Token | day |
Token ID | 1110 |
Feature activation | +9.97e-01 |
Pos logit contributions | |
most | +0.025 |
adder | +0.025 |
block | +0.024 |
bucket | +0.024 |
railing | +0.024 |
Neg logit contributions | |
ils | -0.024 |
ting | -0.024 |
rewarded | -0.023 |
Midnight | -0.022 |
obe | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | +6.29e-02 |
Pos logit contributions | |
most | +0.030 |
block | +0.029 |
foot | +0.027 |
bucket | +0.027 |
second | +0.027 |
Neg logit contributions | |
obe | -0.027 |
Midnight | -0.027 |
punished | -0.027 |
ils | -0.027 |
ior | -0.026 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -9.29e-01 |
Pos logit contributions | |
block | +0.002 |
most | +0.002 |
bucket | +0.001 |
brick | +0.001 |
ladder | +0.001 |
Neg logit contributions | |
ting | -0.002 |
ils | -0.002 |
rewarded | -0.002 |
Midnight | -0.002 |
boys | -0.002 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.03e-01 |
Pos logit contributions | |
ting | +0.027 |
ils | +0.026 |
rewarded | +0.026 |
Midnight | +0.025 |
boys | +0.024 |
Neg logit contributions | |
most | -0.024 |
block | -0.024 |
bucket | -0.023 |
brick | -0.023 |
adder | -0.023 |
Token | there |
Token ID | 612 |
Feature activation | +3.08e+00 |
Pos logit contributions | |
most | +0.010 |
feature | +0.010 |
railing | +0.010 |
chie | +0.010 |
spiral | +0.010 |
Neg logit contributions | |
rewarded | -0.011 |
from | -0.011 |
Mr | -0.011 |
Midnight | -0.010 |
Mrs | -0.010 |
Token | was |
Token ID | 373 |
Feature activation | +8.63e-01 |
Pos logit contributions | |
most | +0.110 |
adder | +0.107 |
railing | +0.107 |
block | +0.104 |
bucket | +0.103 |
Neg logit contributions | |
ting | -0.058 |
rewarded | -0.056 |
ils | -0.054 |
Midnight | -0.053 |
treated | -0.049 |
Token | a |
Token ID | 257 |
Feature activation | -5.40e-01 |
Pos logit contributions | |
chie | +0.031 |
most | +0.031 |
ladder | +0.031 |
railing | +0.031 |
adder | +0.029 |
Neg logit contributions | |
Midnight | -0.017 |
ting | -0.015 |
Rudolph | -0.015 |
when | -0.015 |
ils | -0.015 |
Token | girl |
Token ID | 2576 |
Feature activation | +7.97e-02 |
Pos logit contributions | |
ting | +0.017 |
Midnight | +0.017 |
rewarded | +0.016 |
Claus | +0.016 |
ils | +0.016 |
Neg logit contributions | |
most | -0.012 |
adder | -0.012 |
railing | -0.011 |
block | -0.011 |
bucket | -0.011 |
Token | who |
Token ID | 508 |
Feature activation | -3.57e-01 |
Pos logit contributions | |
block | +0.003 |
most | +0.003 |
bucket | +0.003 |
brick | +0.003 |
palm | +0.003 |
Neg logit contributions | |
ting | -0.002 |
ils | -0.002 |
rewarded | -0.002 |
Midnight | -0.002 |
punished | -0.001 |
Token | hated |
Token ID | 16563 |
Feature activation | -3.23e-01 |
Pos logit contributions | |
ting | +0.009 |
ils | +0.008 |
rewarded | +0.008 |
boys | +0.008 |
minded | +0.008 |
Neg logit contributions | |
block | -0.011 |
palm | -0.011 |
bucket | -0.011 |
most | -0.011 |
ladder | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | -4.52e-01 |
Pos logit contributions | |
block | +0.040 |
bucket | +0.037 |
most | +0.036 |
ladder | +0.036 |
palm | +0.035 |
Neg logit contributions | |
ting | -0.037 |
ils | -0.037 |
rewarded | -0.035 |
Midnight | -0.033 |
punished | -0.032 |
Token | Ben |
Token ID | 3932 |
Feature activation | -8.95e-01 |
Pos logit contributions | |
ils | +0.011 |
rewarded | +0.010 |
ting | +0.010 |
punished | +0.009 |
treated | +0.009 |
Neg logit contributions | |
block | -0.015 |
most | -0.015 |
bucket | -0.015 |
palm | -0.014 |
ladder | -0.014 |
Token | say |
Token ID | 910 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
ting | +0.035 |
ils | +0.034 |
ups | +0.033 |
rewarded | +0.033 |
Midnight | +0.032 |
Neg logit contributions | |
bucket | -0.017 |
block | -0.016 |
palm | -0.014 |
second | -0.014 |
shovel | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
block | +0.038 |
bucket | +0.037 |
most | +0.036 |
ladder | +0.036 |
brick | +0.036 |
Neg logit contributions | |
ting | -0.031 |
ils | -0.030 |
rewarded | -0.029 |
obe | -0.028 |
Midnight | -0.027 |
Token | " |
Token ID | 366 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
block | +0.001 |
most | +0.001 |
ladder | +0.001 |
bucket | +0.001 |
brick | +0.001 |
Neg logit contributions | |
ils | -0.001 |
ting | -0.001 |
rewarded | -0.001 |
obe | -0.001 |
punished | -0.001 |
Token | Yes |
Token ID | 5297 |
Feature activation | +3.06e+00 |
Pos logit contributions | |
ting | +0.003 |
ils | +0.003 |
rewarded | +0.003 |
Midnight | +0.003 |
boys | +0.003 |
Neg logit contributions | |
most | -0.003 |
block | -0.003 |
bucket | -0.003 |
brick | -0.003 |
adder | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
bucket | +0.092 |
block | +0.091 |
most | +0.090 |
ladder | +0.089 |
brick | +0.088 |
Neg logit contributions | |
ils | -0.082 |
ting | -0.080 |
rewarded | -0.076 |
obe | -0.072 |
treated | -0.072 |
Token | Mom |
Token ID | 11254 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
block | +0.029 |
most | +0.028 |
bucket | +0.028 |
brick | +0.027 |
ladder | +0.027 |
Neg logit contributions | |
ting | -0.031 |
ils | -0.030 |
rewarded | -0.029 |
Midnight | -0.028 |
boys | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
block | +0.027 |
bucket | +0.022 |
most | +0.022 |
adder | +0.022 |
brick | +0.022 |
Neg logit contributions | |
ils | -0.038 |
ting | -0.036 |
Midnight | -0.036 |
punished | -0.035 |
rewarded | -0.033 |
Token | Dad |
Token ID | 17415 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
bucket | +0.036 |
block | +0.034 |
ladder | +0.033 |
palm | +0.033 |
shovel | +0.032 |
Neg logit contributions | |
ting | -0.056 |
ils | -0.056 |
rewarded | -0.052 |
obe | -0.050 |
boys | -0.050 |
Token | . |
Token ID | 13 |
Feature activation | -5.26e-01 |
Pos logit contributions | |
block | +0.012 |
key | +0.010 |
bucket | +0.010 |
hand | +0.009 |
chance | +0.009 |
Neg logit contributions | |
ting | -0.022 |
Midnight | -0.021 |
Mr | -0.020 |
bing | -0.019 |
lish | -0.019 |
Token | on |
Token ID | 319 |
Feature activation | -4.09e-01 |
Pos logit contributions | |
ils | +0.027 |
Midnight | +0.026 |
ting | +0.026 |
Mr | +0.026 |
boys | +0.025 |
Neg logit contributions | |
block | -0.023 |
bucket | -0.023 |
ladder | -0.023 |
most | -0.023 |
key | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | -4.09e-01 |
Pos logit contributions | |
ting | +0.012 |
ils | +0.011 |
rewarded | +0.011 |
Claus | +0.010 |
Midnight | +0.010 |
Neg logit contributions | |
most | -0.012 |
block | -0.011 |
bucket | -0.011 |
ladder | -0.011 |
key | -0.011 |
Token | journey |
Token ID | 7002 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
ting | +0.014 |
ils | +0.014 |
rewarded | +0.014 |
Midnight | +0.013 |
boys | +0.013 |
Neg logit contributions | |
block | -0.009 |
most | -0.009 |
bucket | -0.008 |
ladder | -0.008 |
brick | -0.008 |
Token | she |
Token ID | 673 |
Feature activation | -2.47e-01 |
Pos logit contributions | |
most | +0.044 |
ladder | +0.044 |
block | +0.043 |
key | +0.043 |
second | +0.042 |
Neg logit contributions | |
ils | -0.041 |
Midnight | -0.040 |
ior | -0.039 |
ting | -0.039 |
Mr | -0.037 |
Token | always |
Token ID | 1464 |
Feature activation | -5.99e-01 |
Pos logit contributions | |
Midnight | +0.007 |
ting | +0.007 |
ils | +0.007 |
obe | +0.007 |
Mr | +0.007 |
Neg logit contributions | |
most | -0.006 |
block | -0.006 |
bucket | -0.006 |
key | -0.006 |
foot | -0.006 |
Token | wanted |
Token ID | 2227 |
Feature activation | +3.26e+00 |
Pos logit contributions | |
ils | +0.017 |
Midnight | +0.017 |
ting | +0.017 |
Mr | +0.016 |
boys | +0.015 |
Neg logit contributions | |
block | -0.016 |
bucket | -0.016 |
key | -0.016 |
most | -0.016 |
foot | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
most | +0.073 |
key | +0.072 |
second | +0.071 |
block | +0.070 |
foot | +0.069 |
Neg logit contributions | |
Claus | -0.108 |
ior | -0.103 |
ils | -0.102 |
ting | -0.102 |
Midnight | -0.101 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -8.07e-01 |
Pos logit contributions | |
ting | +0.003 |
rewarded | +0.003 |
Midnight | +0.003 |
ils | +0.003 |
obe | +0.003 |
Neg logit contributions | |
bucket | -0.004 |
block | -0.004 |
most | -0.004 |
nail | -0.004 |
ladder | -0.004 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.19e-01 |
Pos logit contributions | |
ils | +0.025 |
ting | +0.025 |
rewarded | +0.025 |
punished | +0.024 |
boys | +0.023 |
Neg logit contributions | |
block | -0.020 |
most | -0.019 |
bucket | -0.019 |
ladder | -0.018 |
brick | -0.018 |
Token | there |
Token ID | 612 |
Feature activation | +2.91e+00 |
Pos logit contributions | |
most | +0.011 |
feature | +0.010 |
railing | +0.010 |
chie | +0.010 |
block | +0.010 |
Neg logit contributions | |
from | -0.011 |
Mr | -0.011 |
rewarded | -0.011 |
when | -0.011 |
Midnight | -0.011 |
Token | was |
Token ID | 373 |
Feature activation | +9.07e-01 |
Pos logit contributions | |
most | +0.101 |
adder | +0.098 |
railing | +0.097 |
block | +0.097 |
bucket | +0.096 |
Neg logit contributions | |
ting | -0.059 |
rewarded | -0.056 |
ils | -0.055 |
Midnight | -0.053 |
treated | -0.049 |
Token | zip |
Token ID | 19974 |
Feature activation | -6.07e-01 |
Pos logit contributions | |
block | +0.005 |
bucket | +0.005 |
most | +0.005 |
palm | +0.005 |
brick | +0.005 |
Neg logit contributions | |
ils | -0.006 |
ting | -0.006 |
rewarded | -0.005 |
Midnight | -0.005 |
Claus | -0.005 |
Token | around |
Token ID | 1088 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
ils | +0.015 |
rewarded | +0.014 |
Claus | +0.014 |
ting | +0.014 |
obe | +0.013 |
Neg logit contributions | |
block | -0.019 |
bucket | -0.018 |
ladder | -0.018 |
key | -0.018 |
most | -0.018 |
Token | on |
Token ID | 319 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
block | +0.032 |
nail | +0.032 |
foot | +0.031 |
palm | +0.031 |
most | +0.031 |
Neg logit contributions | |
Claus | -0.037 |
ils | -0.037 |
rewarded | -0.036 |
ior | -0.035 |
respected | -0.034 |
Token | the |
Token ID | 262 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
block | +0.031 |
most | +0.030 |
palm | +0.030 |
bucket | +0.030 |
ladder | +0.030 |
Neg logit contributions | |
ils | -0.026 |
rewarded | -0.026 |
Claus | -0.025 |
ting | -0.025 |
obe | -0.023 |
Token | yellow |
Token ID | 7872 |
Feature activation | +4.90e+00 |
Pos logit contributions | |
most | +0.025 |
block | +0.025 |
adder | +0.025 |
bucket | +0.024 |
brick | +0.024 |
Neg logit contributions | |
ting | -0.032 |
ils | -0.031 |
rewarded | -0.031 |
Midnight | -0.029 |
boys | -0.029 |
Token | loop |
Token ID | 9052 |
Feature activation | +4.37e+00 |
Pos logit contributions | |
adder | +0.104 |
coconut | +0.098 |
most | +0.093 |
shovel | +0.090 |
brick | +0.089 |
Neg logit contributions | |
ting | -0.168 |
from | -0.147 |
ils | -0.146 |
rewarded | -0.144 |
boys | -0.143 |
Token | again |
Token ID | 757 |
Feature activation | +2.26e-01 |
Pos logit contributions | |
block | +0.141 |
most | +0.138 |
ladder | +0.134 |
bucket | +0.132 |
brick | +0.131 |
Neg logit contributions | |
ils | -0.106 |
ting | -0.102 |
rewarded | -0.101 |
Midnight | -0.095 |
punished | -0.093 |
Token | . |
Token ID | 13 |
Feature activation | -3.77e-02 |
Pos logit contributions | |
most | +0.007 |
block | +0.006 |
foot | +0.006 |
palm | +0.006 |
bucket | +0.006 |
Neg logit contributions | |
ils | -0.006 |
ting | -0.006 |
punished | -0.006 |
Midnight | -0.006 |
rewarded | -0.006 |
Token | And |
Token ID | 843 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
ting | +0.001 |
ils | +0.001 |
rewarded | +0.001 |
Midnight | +0.001 |
boys | +0.001 |
Neg logit contributions | |
block | -0.001 |
most | -0.001 |
bucket | -0.001 |
brick | -0.001 |
ladder | -0.001 |
Token | that |
Token ID | 326 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
ils | +0.033 |
ting | +0.032 |
obe | +0.031 |
ior | +0.030 |
Claus | +0.030 |
Neg logit contributions | |
palm | -0.036 |
foot | -0.036 |
most | -0.035 |
ladder | -0.034 |
key | -0.034 |
Token | 's |
Token ID | 338 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
ting | +0.006 |
girls | +0.005 |
ils | +0.005 |
boys | +0.005 |
Mr | +0.005 |
Neg logit contributions | |
block | -0.004 |
palm | -0.004 |
foot | -0.004 |
key | -0.004 |
bucket | -0.004 |
Token | some |
Token ID | 617 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
block | +0.003 |
most | +0.003 |
bucket | +0.003 |
brick | +0.003 |
ladder | +0.003 |
Neg logit contributions | |
ting | -0.001 |
ils | -0.001 |
rewarded | -0.001 |
Midnight | -0.001 |
boys | -0.001 |
Token | more |
Token ID | 517 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
ting | +0.004 |
ils | +0.004 |
rewarded | +0.004 |
Midnight | +0.004 |
Claus | +0.003 |
Neg logit contributions | |
block | -0.004 |
bucket | -0.004 |
most | -0.004 |
ladder | -0.004 |
brick | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -6.19e-01 |
Pos logit contributions | |
block | +0.008 |
bucket | +0.007 |
most | +0.007 |
ladder | +0.007 |
brick | +0.007 |
Neg logit contributions | |
ting | -0.004 |
rewarded | -0.004 |
ils | -0.004 |
Midnight | -0.004 |
Claus | -0.004 |
Token | They |
Token ID | 1119 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
ting | +0.013 |
ils | +0.013 |
rewarded | +0.012 |
Midnight | +0.011 |
boys | +0.011 |
Neg logit contributions | |
block | -0.021 |
most | -0.021 |
bucket | -0.021 |
brick | -0.021 |
ladder | -0.021 |
Token | are |
Token ID | 389 |
Feature activation | +6.80e-02 |
Pos logit contributions | |
block | +0.003 |
bucket | +0.003 |
palm | +0.003 |
ladder | +0.003 |
shovel | +0.003 |
Neg logit contributions | |
ting | -0.005 |
ils | -0.004 |
ating | -0.004 |
obe | -0.004 |
rewarded | -0.004 |
Token | happy |
Token ID | 3772 |
Feature activation | +9.80e-01 |
Pos logit contributions | |
block | +0.001 |
bucket | +0.001 |
foot | +0.001 |
palm | +0.001 |
second | +0.001 |
Neg logit contributions | |
ting | -0.003 |
ils | -0.003 |
obe | -0.003 |
minded | -0.003 |
ior | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -5.35e-01 |
Pos logit contributions | |
block | +0.028 |
bucket | +0.027 |
shovel | +0.026 |
palm | +0.025 |
most | +0.025 |
Neg logit contributions | |
ils | -0.029 |
rewarded | -0.028 |
ting | -0.028 |
Midnight | -0.027 |
punished | -0.027 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
ting | +0.014 |
ils | +0.013 |
rewarded | +0.013 |
Midnight | +0.012 |
punished | +0.012 |
Neg logit contributions | |
block | -0.016 |
most | -0.016 |
bucket | -0.016 |
brick | -0.016 |
ladder | -0.016 |
Token | Tom |
Token ID | 13787 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
ting | +0.032 |
ils | +0.031 |
rewarded | +0.030 |
Midnight | +0.029 |
boys | +0.028 |
Neg logit contributions | |
block | -0.036 |
most | -0.036 |
bucket | -0.035 |
brick | -0.034 |
ladder | -0.034 |
Token | and |
Token ID | 290 |
Feature activation | -4.97e-01 |
Pos logit contributions | |
block | +0.004 |
bucket | +0.004 |
palm | +0.003 |
shovel | +0.003 |
hand | +0.003 |
Neg logit contributions | |
ting | -0.006 |
ils | -0.005 |
Midnight | -0.005 |
ups | -0.005 |
lish | -0.005 |
Token | Mia |
Token ID | 32189 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
ting | +0.013 |
ils | +0.013 |
rewarded | +0.012 |
Midnight | +0.012 |
boys | +0.011 |
Neg logit contributions | |
block | -0.015 |
most | -0.014 |
bucket | -0.014 |
brick | -0.014 |
ladder | -0.014 |
Token | could |
Token ID | 714 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
ting | +0.019 |
ils | +0.018 |
Midnight | +0.017 |
rewarded | +0.016 |
obe | +0.016 |
Neg logit contributions | |
block | -0.024 |
most | -0.023 |
bucket | -0.023 |
brick | -0.022 |
ladder | -0.022 |
Token | see |
Token ID | 766 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
ils | +0.008 |
ting | +0.007 |
ior | +0.007 |
minded | +0.007 |
lish | +0.007 |
Neg logit contributions | |
block | -0.004 |
key | -0.004 |
ladder | -0.004 |
bucket | -0.004 |
nail | -0.004 |
Token | how |
Token ID | 703 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
block | +0.042 |
brick | +0.041 |
ladder | +0.041 |
palm | +0.040 |
nail | +0.040 |
Neg logit contributions | |
obe | -0.044 |
rewarded | -0.042 |
Claus | -0.042 |
ils | -0.041 |
respected | -0.041 |
Token | determined |
Token ID | 5295 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
ting | +0.038 |
ils | +0.037 |
rewarded | +0.035 |
Midnight | +0.034 |
boys | +0.034 |
Neg logit contributions | |
block | -0.030 |
most | -0.030 |
bucket | -0.030 |
ladder | -0.030 |
brick | -0.030 |
Token | Lucy |
Token ID | 22162 |
Feature activation | +2.95e-01 |
Pos logit contributions | |
ladder | +0.035 |
key | +0.031 |
inspection | +0.031 |
rope | +0.030 |
most | +0.029 |
Neg logit contributions | |
ting | -0.061 |
minded | -0.060 |
Claus | -0.060 |
obe | -0.060 |
rewarded | -0.060 |
Token | was |
Token ID | 373 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
block | +0.006 |
most | +0.006 |
ladder | +0.005 |
nail | +0.005 |
bucket | +0.005 |
Neg logit contributions | |
ting | -0.010 |
obe | -0.010 |
Mr | -0.010 |
respected | -0.009 |
ior | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | +4.16e-01 |
Pos logit contributions | |
block | +0.033 |
key | +0.032 |
bucket | +0.032 |
ladder | +0.032 |
nail | +0.031 |
Neg logit contributions | |
ils | -0.038 |
ting | -0.036 |
ior | -0.036 |
Mr | -0.036 |
Midnight | -0.035 |
Token | so |
Token ID | 523 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
block | +0.010 |
most | +0.010 |
ladder | +0.010 |
bucket | +0.010 |
palm | +0.010 |
Neg logit contributions | |
ils | -0.014 |
ting | -0.012 |
Claus | -0.012 |
rewarded | -0.012 |
yer | -0.012 |
Token | she |
Token ID | 673 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
ladder | +0.029 |
second | +0.026 |
brick | +0.026 |
palm | +0.025 |
block | +0.025 |
Neg logit contributions | |
ting | -0.042 |
ils | -0.041 |
obe | -0.040 |
ior | -0.038 |
Claus | -0.038 |
Token | agreed |
Token ID | 4987 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
ting | +0.040 |
obe | +0.037 |
respected | +0.036 |
ior | +0.033 |
ils | +0.033 |
Neg logit contributions | |
block | -0.040 |
bucket | -0.039 |
key | -0.039 |
most | -0.037 |
ladder | -0.037 |
Token | to |
Token ID | 284 |
Feature activation | +9.17e-01 |
Pos logit contributions | |
second | +0.051 |
key | +0.048 |
ladder | +0.047 |
block | +0.047 |
bucket | +0.047 |
Neg logit contributions | |
ting | -0.065 |
ils | -0.064 |
ior | -0.063 |
boys | -0.061 |
Claus | -0.060 |
Token | They |
Token ID | 1119 |
Feature activation | -3.33e-01 |
Pos logit contributions | |
ting | +0.006 |
ils | +0.006 |
rewarded | +0.006 |
Midnight | +0.006 |
punished | +0.006 |
Neg logit contributions | |
block | -0.005 |
most | -0.005 |
ladder | -0.005 |
brick | -0.005 |
bucket | -0.005 |
Token | sat |
Token ID | 3332 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
ting | +0.008 |
ils | +0.007 |
rewarded | +0.007 |
Midnight | +0.007 |
obe | +0.006 |
Neg logit contributions | |
block | -0.011 |
most | -0.011 |
bucket | -0.011 |
brick | -0.011 |
ladder | -0.011 |
Token | down |
Token ID | 866 |
Feature activation | +7.17e-01 |
Pos logit contributions | |
ils | +0.029 |
rewarded | +0.028 |
ting | +0.028 |
Midnight | +0.026 |
boys | +0.024 |
Neg logit contributions | |
block | -0.045 |
bucket | -0.044 |
most | -0.043 |
ladder | -0.042 |
brick | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | -5.56e-01 |
Pos logit contributions | |
block | +0.025 |
most | +0.024 |
bucket | +0.024 |
ladder | +0.024 |
brick | +0.024 |
Neg logit contributions | |
ting | -0.015 |
ils | -0.015 |
rewarded | -0.015 |
Midnight | -0.014 |
boys | -0.013 |
Token | Sue |
Token ID | 26089 |
Feature activation | +8.36e-01 |
Pos logit contributions | |
ting | +0.015 |
ils | +0.015 |
rewarded | +0.015 |
ups | +0.014 |
obe | +0.014 |
Neg logit contributions | |
block | -0.016 |
bucket | -0.016 |
shovel | -0.015 |
most | -0.014 |
brick | -0.014 |
Token | showed |
Token ID | 3751 |
Feature activation | +2.78e-01 |
Pos logit contributions | |
block | +0.023 |
bucket | +0.020 |
circle | +0.018 |
shovel | +0.018 |
most | +0.018 |
Neg logit contributions | |
ting | -0.027 |
rim | -0.027 |
yourselves | -0.026 |
lish | -0.026 |
Mr | -0.025 |
Token | Tim |
Token ID | 5045 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
block | +0.010 |
brick | +0.009 |
most | +0.009 |
bucket | +0.009 |
ladder | +0.009 |
Neg logit contributions | |
ting | -0.006 |
rewarded | -0.006 |
ils | -0.006 |
Claus | -0.005 |
Midnight | -0.005 |
Token | how |
Token ID | 703 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
block | +0.070 |
bucket | +0.066 |
brick | +0.065 |
most | +0.064 |
nail | +0.063 |
Neg logit contributions | |
ting | -0.056 |
rewarded | -0.052 |
ils | -0.052 |
Midnight | -0.048 |
Claus | -0.048 |
Token | to |
Token ID | 284 |
Feature activation | +6.79e-01 |
Pos logit contributions | |
ting | +0.029 |
ils | +0.028 |
rewarded | +0.027 |
Midnight | +0.026 |
boys | +0.025 |
Neg logit contributions | |
block | -0.034 |
most | -0.033 |
bucket | -0.033 |
brick | -0.032 |
ladder | -0.032 |
Token | move |
Token ID | 1445 |
Feature activation | -8.83e-01 |
Pos logit contributions | |
block | +0.019 |
bucket | +0.018 |
most | +0.018 |
nail | +0.018 |
ladder | +0.017 |
Neg logit contributions | |
ting | -0.020 |
ils | -0.019 |
rewarded | -0.019 |
Midnight | -0.018 |
Claus | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | -4.79e-01 |
Pos logit contributions | |
ils | +0.024 |
rewarded | +0.023 |
ting | +0.023 |
Claus | +0.023 |
obe | +0.022 |
Neg logit contributions | |
block | -0.027 |
bucket | -0.025 |
foot | -0.025 |
nail | -0.025 |
second | -0.024 |
Token | rain |
Token ID | 6290 |
Feature activation | -2.58e-01 |
Pos logit contributions | |
ils | +0.003 |
ting | +0.003 |
rewarded | +0.003 |
Claus | +0.003 |
Midnight | +0.003 |
Neg logit contributions | |
bucket | -0.002 |
ladder | -0.001 |
block | -0.001 |
brick | -0.001 |
palm | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
rewarded | +0.007 |
ils | +0.007 |
ting | +0.007 |
Mr | +0.006 |
Rudolph | +0.006 |
Neg logit contributions | |
block | -0.008 |
ladder | -0.007 |
bucket | -0.007 |
most | -0.007 |
brick | -0.007 |
Token | it |
Token ID | 340 |
Feature activation | -3.07e-01 |
Pos logit contributions | |
ils | +0.004 |
rewarded | +0.004 |
ting | +0.004 |
Claus | +0.004 |
yer | +0.003 |
Neg logit contributions | |
palm | -0.003 |
bucket | -0.003 |
ladder | -0.003 |
block | -0.003 |
brick | -0.003 |
Token | made |
Token ID | 925 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
ting | +0.008 |
ils | +0.008 |
rewarded | +0.008 |
Midnight | +0.008 |
Claus | +0.007 |
Neg logit contributions | |
block | -0.009 |
most | -0.009 |
bucket | -0.008 |
ladder | -0.008 |
brick | -0.008 |
Token | everything |
Token ID | 2279 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
key | +0.004 |
brick | +0.004 |
block | +0.004 |
second | +0.004 |
nail | +0.004 |
Neg logit contributions | |
ils | -0.005 |
ting | -0.005 |
Claus | -0.005 |
rewarded | -0.005 |
boys | -0.005 |
Token | difficult |
Token ID | 2408 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
ting | +0.006 |
ils | +0.006 |
rewarded | +0.005 |
Midnight | +0.005 |
obe | +0.005 |
Neg logit contributions | |
block | -0.004 |
bucket | -0.004 |
ladder | -0.004 |
brick | -0.004 |
most | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
foot | +0.004 |
most | +0.004 |
block | +0.004 |
nail | +0.004 |
minute | +0.004 |
Neg logit contributions | |
Mr | -0.006 |
Claus | -0.006 |
rewarded | -0.006 |
yer | -0.006 |
ior | -0.005 |
Token | |
Token ID | 220 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
ting | +0.003 |
ils | +0.003 |
rewarded | +0.003 |
Midnight | +0.003 |
boys | +0.003 |
Neg logit contributions | |
block | -0.003 |
most | -0.003 |
bucket | -0.003 |
brick | -0.003 |
adder | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +6.25e-02 |
Pos logit contributions | |
most | +0.031 |
block | +0.031 |
bucket | +0.030 |
palm | +0.030 |
brick | +0.030 |
Neg logit contributions | |
ting | -0.028 |
rewarded | -0.027 |
Midnight | -0.026 |
ils | -0.026 |
obe | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | -2.23e-01 |
Pos logit contributions | |
block | +0.002 |
most | +0.002 |
bucket | +0.002 |
adder | +0.002 |
brick | +0.002 |
Neg logit contributions | |
ting | -0.002 |
rewarded | -0.002 |
Midnight | -0.002 |
ils | -0.002 |
Mr | -0.001 |
Token | Despite |
Token ID | 8332 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
ting | +0.006 |
ils | +0.006 |
rewarded | +0.006 |
Midnight | +0.006 |
boys | +0.006 |
Neg logit contributions | |
block | -0.006 |
most | -0.006 |
bucket | -0.006 |
brick | -0.006 |
ladder | -0.006 |
Token | if |
Token ID | 611 |
Feature activation | -3.46e-01 |
Pos logit contributions | |
block | +0.007 |
nail | +0.007 |
key | +0.007 |
foot | +0.007 |
shovel | +0.007 |
Neg logit contributions | |
ils | -0.008 |
yer | -0.008 |
ting | -0.008 |
obe | -0.007 |
Claus | -0.007 |
Token | you |
Token ID | 345 |
Feature activation | +3.69e-01 |
Pos logit contributions | |
ting | +0.009 |
obe | +0.008 |
ils | +0.008 |
ior | +0.008 |
rewarded | +0.008 |
Neg logit contributions | |
bucket | -0.011 |
ladder | -0.011 |
brick | -0.010 |
key | -0.010 |
block | -0.010 |
Token | are |
Token ID | 389 |
Feature activation | -8.09e-01 |
Pos logit contributions | |
block | +0.008 |
bucket | +0.008 |
key | +0.007 |
most | +0.007 |
nail | +0.007 |
Neg logit contributions | |
ting | -0.013 |
Midnight | -0.012 |
Mr | -0.012 |
ils | -0.012 |
ior | -0.012 |
Token | small |
Token ID | 1402 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
ting | +0.031 |
ils | +0.028 |
obe | +0.028 |
ior | +0.028 |
respected | +0.028 |
Neg logit contributions | |
block | -0.017 |
second | -0.017 |
key | -0.016 |
foot | -0.015 |
ruler | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +5.85e-02 |
Pos logit contributions | |
ils | +0.005 |
ting | +0.005 |
Midnight | +0.005 |
Mr | +0.005 |
rewarded | +0.004 |
Neg logit contributions | |
block | -0.005 |
foot | -0.005 |
key | -0.005 |
bucket | -0.005 |
ladder | -0.005 |
Token | You |
Token ID | 921 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
block | +0.002 |
most | +0.002 |
brick | +0.002 |
adder | +0.002 |
bucket | +0.002 |
Neg logit contributions | |
ils | -0.002 |
ting | -0.002 |
rewarded | -0.001 |
punished | -0.001 |
Midnight | -0.001 |
Token | just |
Token ID | 655 |
Feature activation | +8.18e-01 |
Pos logit contributions | |
block | +0.043 |
key | +0.040 |
bucket | +0.039 |
nail | +0.038 |
most | +0.038 |
Neg logit contributions | |
ting | -0.069 |
ils | -0.064 |
obe | -0.063 |
Midnight | -0.062 |
ior | -0.060 |
Token | have |
Token ID | 423 |
Feature activation | -4.03e-01 |
Pos logit contributions | |
block | +0.017 |
bucket | +0.015 |
nail | +0.015 |
key | +0.015 |
brick | +0.015 |
Neg logit contributions | |
ting | -0.030 |
ils | -0.028 |
Mr | -0.028 |
Midnight | -0.028 |
obe | -0.028 |
Token | to |
Token ID | 284 |
Feature activation | +4.85e-01 |
Pos logit contributions | |
ting | +0.013 |
ils | +0.012 |
obe | +0.011 |
Midnight | +0.011 |
Claus | +0.011 |
Neg logit contributions | |
block | -0.010 |
bucket | -0.010 |
most | -0.010 |
second | -0.010 |
brick | -0.010 |
Token | look |
Token ID | 804 |
Feature activation | +7.50e-02 |
Pos logit contributions | |
block | +0.009 |
key | +0.009 |
palm | +0.009 |
bucket | +0.009 |
foot | +0.009 |
Neg logit contributions | |
ting | -0.020 |
ils | -0.018 |
led | -0.017 |
Midnight | -0.017 |
Claus | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | +7.64e-02 |
Pos logit contributions | |
second | +0.002 |
block | +0.002 |
bucket | +0.002 |
key | +0.002 |
brick | +0.002 |
Neg logit contributions | |
obe | -0.003 |
ting | -0.002 |
ils | -0.002 |
boys | -0.002 |
Midnight | -0.002 |
Token | Sam |
Token ID | 3409 |
Feature activation | +4.30e-01 |
Pos logit contributions | |
ting | +0.014 |
ils | +0.013 |
rewarded | +0.013 |
Midnight | +0.012 |
punished | +0.011 |
Neg logit contributions | |
most | -0.022 |
block | -0.022 |
adder | -0.021 |
bucket | -0.021 |
ladder | -0.021 |
Token | , |
Token ID | 11 |
Feature activation | -3.89e-01 |
Pos logit contributions | |
block | +0.013 |
foot | +0.011 |
second | +0.011 |
bucket | +0.011 |
nail | +0.011 |
Neg logit contributions | |
ting | -0.012 |
ils | -0.012 |
rewarded | -0.012 |
thank | -0.011 |
Merry | -0.011 |
Token | but |
Token ID | 475 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
ting | +0.011 |
rewarded | +0.011 |
ils | +0.011 |
Midnight | +0.011 |
boys | +0.010 |
Neg logit contributions | |
most | -0.010 |
block | -0.010 |
bucket | -0.010 |
adder | -0.010 |
brick | -0.010 |
Token | Sam |
Token ID | 3409 |
Feature activation | -3.33e-01 |
Pos logit contributions | |
ting | +0.003 |
rewarded | +0.003 |
ils | +0.003 |
Midnight | +0.003 |
boys | +0.003 |
Neg logit contributions | |
most | -0.005 |
adder | -0.005 |
block | -0.005 |
railing | -0.005 |
ladder | -0.005 |
Token | was |
Token ID | 373 |
Feature activation | +9.72e-01 |
Pos logit contributions | |
ting | +0.013 |
minded | +0.011 |
Merry | +0.011 |
ils | +0.011 |
Rudolph | +0.011 |
Neg logit contributions | |
block | -0.006 |
foot | -0.006 |
nail | -0.006 |
bucket | -0.006 |
key | -0.005 |
Token | not |
Token ID | 407 |
Feature activation | -3.31e-01 |
Pos logit contributions | |
most | +0.029 |
bucket | +0.028 |
block | +0.028 |
adder | +0.027 |
brick | +0.027 |
Neg logit contributions | |
ting | -0.025 |
rewarded | -0.025 |
ils | -0.024 |
Midnight | -0.023 |
punished | -0.023 |
Token | there |
Token ID | 612 |
Feature activation | +2.32e+00 |
Pos logit contributions | |
ting | +0.010 |
ils | +0.009 |
rewarded | +0.009 |
Midnight | +0.009 |
punished | +0.009 |
Neg logit contributions | |
most | -0.009 |
block | -0.009 |
bucket | -0.008 |
brick | -0.008 |
ladder | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -6.16e-01 |
Pos logit contributions | |
block | +0.074 |
most | +0.068 |
brick | +0.068 |
foot | +0.067 |
key | +0.067 |
Neg logit contributions | |
ils | -0.060 |
ting | -0.059 |
rewarded | -0.058 |
Midnight | -0.054 |
ior | -0.054 |
Token | He |
Token ID | 679 |
Feature activation | -8.44e-01 |
Pos logit contributions | |
ting | +0.012 |
Midnight | +0.012 |
rewarded | +0.012 |
Mr | +0.011 |
obe | +0.011 |
Neg logit contributions | |
adder | -0.020 |
railing | -0.020 |
key | -0.020 |
bucket | -0.020 |
most | -0.020 |
Token | looked |
Token ID | 3114 |
Feature activation | +2.91e-01 |
Pos logit contributions | |
ting | +0.024 |
ils | +0.023 |
rewarded | +0.022 |
Midnight | +0.021 |
obe | +0.021 |
Neg logit contributions | |
block | -0.024 |
most | -0.023 |
bucket | -0.023 |
palm | -0.022 |
nail | -0.022 |
Token | for |
Token ID | 329 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
block | +0.008 |
bucket | +0.008 |
brick | +0.008 |
foot | +0.008 |
shovel | +0.008 |
Neg logit contributions | |
ting | -0.008 |
ils | -0.008 |
Claus | -0.008 |
boys | -0.007 |
rewarded | -0.007 |
Token | day |
Token ID | 1110 |
Feature activation | -8.92e-01 |
Pos logit contributions | |
ting | +0.084 |
rewarded | +0.080 |
ils | +0.078 |
Midnight | +0.078 |
Mr | +0.073 |
Neg logit contributions | |
bucket | -0.069 |
block | -0.067 |
ladder | -0.067 |
most | -0.066 |
palm | -0.065 |
Token | ." |
Token ID | 526 |
Feature activation | +4.04e-01 |
Pos logit contributions | |
ils | +0.023 |
Midnight | +0.023 |
obe | +0.023 |
ting | +0.023 |
punished | +0.020 |
Neg logit contributions | |
ladder | -0.028 |
block | -0.028 |
most | -0.027 |
bucket | -0.027 |
palm | -0.026 |
Token | |
Token ID | 220 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
block | +0.010 |
most | +0.010 |
bucket | +0.009 |
ladder | +0.009 |
brick | +0.009 |
Neg logit contributions | |
ils | -0.013 |
ting | -0.013 |
rewarded | -0.013 |
Midnight | -0.012 |
boys | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | +4.63e-01 |
Pos logit contributions | |
most | +0.038 |
block | +0.037 |
bucket | +0.037 |
brick | +0.037 |
palm | +0.036 |
Neg logit contributions | |
ting | -0.038 |
rewarded | -0.037 |
ils | -0.036 |
Midnight | -0.035 |
boys | -0.033 |
Token |
|
Token ID | 198 |
Feature activation | +1.74e-02 |
Pos logit contributions | |
most | +0.014 |
bucket | +0.014 |
brick | +0.013 |
adder | +0.013 |
block | +0.013 |
Neg logit contributions | |
ting | -0.011 |
rewarded | -0.011 |
ils | -0.011 |
Midnight | -0.011 |
Mr | -0.011 |
Token | The |
Token ID | 464 |
Feature activation | -6.81e-01 |
Pos logit contributions | |
block | +0.000 |
most | +0.000 |
bucket | +0.000 |
ladder | +0.000 |
brick | +0.000 |
Neg logit contributions | |
ting | -0.001 |
ils | -0.001 |
rewarded | -0.000 |
Midnight | -0.000 |
boys | -0.000 |
Token | farmer |
Token ID | 18739 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
ils | +0.019 |
ting | +0.018 |
rewarded | +0.018 |
Midnight | +0.017 |
respected | +0.017 |
Neg logit contributions | |
block | -0.020 |
ladder | -0.019 |
bucket | -0.019 |
most | -0.019 |
brick | -0.018 |
Token | came |
Token ID | 1625 |
Feature activation | +3.15e-01 |
Pos logit contributions | |
ting | +0.044 |
ils | +0.041 |
Midnight | +0.040 |
obe | +0.040 |
respected | +0.038 |
Neg logit contributions | |
block | -0.039 |
bucket | -0.036 |
most | -0.035 |
palm | -0.034 |
ladder | -0.034 |
Token | out |
Token ID | 503 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
block | +0.010 |
most | +0.010 |
bucket | +0.010 |
brick | +0.009 |
ladder | +0.009 |
Neg logit contributions | |
ils | -0.008 |
ting | -0.008 |
rewarded | -0.007 |
Midnight | -0.007 |
boys | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
ils | +0.037 |
Midnight | +0.034 |
rewarded | +0.032 |
ting | +0.032 |
boys | +0.032 |
Neg logit contributions | |
most | -0.038 |
block | -0.038 |
brick | -0.035 |
foot | -0.035 |
bucket | -0.035 |
Token | feed |
Token ID | 3745 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
ils | +0.006 |
ting | +0.006 |
rewarded | +0.006 |
Midnight | +0.005 |
boys | +0.005 |
Neg logit contributions | |
block | -0.006 |
bucket | -0.006 |
most | -0.006 |
brick | -0.006 |
palm | -0.006 |
Token | he |
Token ID | 339 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
adder | +0.010 |
most | +0.010 |
brick | +0.010 |
chie | +0.009 |
block | +0.009 |
Neg logit contributions | |
ting | -0.004 |
rewarded | -0.004 |
ils | -0.004 |
from | -0.004 |
Midnight | -0.004 |
Token | cannot |
Token ID | 2314 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
ting | +0.045 |
ils | +0.044 |
rewarded | +0.043 |
Midnight | +0.041 |
boys | +0.040 |
Neg logit contributions | |
block | -0.041 |
most | -0.041 |
bucket | -0.040 |
ladder | -0.039 |
brick | -0.039 |
Token | find |
Token ID | 1064 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
rewarded | +0.002 |
ils | +0.002 |
treated | +0.002 |
from | +0.002 |
share | +0.002 |
Neg logit contributions | |
most | -0.004 |
ladder | -0.004 |
adder | -0.004 |
brick | -0.004 |
railing | -0.004 |
Token | any |
Token ID | 597 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
ting | +0.002 |
ils | +0.002 |
rewarded | +0.002 |
him | +0.002 |
that | +0.001 |
Neg logit contributions | |
most | -0.004 |
ladder | -0.004 |
railing | -0.004 |
path | -0.004 |
stair | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -4.48e-01 |
Pos logit contributions | |
ting | +0.006 |
ils | +0.006 |
rewarded | +0.006 |
Midnight | +0.005 |
boys | +0.005 |
Neg logit contributions | |
most | -0.007 |
adder | -0.007 |
ladder | -0.007 |
railing | -0.007 |
bucket | -0.007 |
Token | He |
Token ID | 679 |
Feature activation | -6.95e-01 |
Pos logit contributions | |
ting | +0.009 |
rewarded | +0.008 |
ils | +0.008 |
Midnight | +0.008 |
minded | +0.007 |
Neg logit contributions | |
brick | -0.015 |
adder | -0.015 |
most | -0.015 |
block | -0.015 |
nail | -0.015 |
Token | sees |
Token ID | 7224 |
Feature activation | +7.72e-01 |
Pos logit contributions | |
ting | +0.020 |
ils | +0.019 |
rewarded | +0.019 |
Midnight | +0.018 |
boys | +0.017 |
Neg logit contributions | |
block | -0.019 |
most | -0.019 |
bucket | -0.018 |
ladder | -0.018 |
brick | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | -2.17e-01 |
Pos logit contributions | |
most | +0.022 |
block | +0.021 |
ladder | +0.021 |
adder | +0.021 |
bucket | +0.021 |
Neg logit contributions | |
ting | -0.021 |
ils | -0.020 |
rewarded | -0.020 |
Midnight | -0.019 |
boys | -0.018 |
Token | wall |
Token ID | 3355 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
ting | +0.008 |
ils | +0.007 |
rewarded | +0.007 |
Midnight | +0.007 |
boys | +0.007 |
Neg logit contributions | |
most | -0.004 |
adder | -0.004 |
block | -0.004 |
ladder | -0.004 |
railing | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -7.58e-01 |
Pos logit contributions | |
block | +0.049 |
most | +0.046 |
bucket | +0.046 |
ladder | +0.046 |
key | +0.045 |
Neg logit contributions | |
ting | -0.033 |
ils | -0.031 |
rewarded | -0.030 |
Midnight | -0.029 |
obe | -0.028 |
Token | It |
Token ID | 632 |
Feature activation | +5.12e-01 |
Pos logit contributions | |
ting | +0.013 |
Midnight | +0.013 |
rewarded | +0.012 |
Rudolph | +0.012 |
ils | +0.011 |
Neg logit contributions | |
adder | -0.027 |
brick | -0.027 |
chie | -0.027 |
most | -0.027 |
railing | -0.026 |
Token | tell |
Token ID | 1560 |
Feature activation | +3.22e-01 |
Pos logit contributions | |
ting | +0.020 |
obe | +0.019 |
ating | +0.018 |
ils | +0.018 |
minded | +0.017 |
Neg logit contributions | |
bucket | -0.016 |
block | -0.015 |
shovel | -0.015 |
hand | -0.015 |
palm | -0.015 |
Token | her |
Token ID | 607 |
Feature activation | -3.22e-01 |
Pos logit contributions | |
palm | +0.007 |
foot | +0.006 |
second | +0.006 |
block | +0.006 |
brick | +0.006 |
Neg logit contributions | |
Claus | -0.011 |
ting | -0.011 |
rewarded | -0.010 |
respected | -0.010 |
boys | -0.010 |
Token | he |
Token ID | 339 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
ting | +0.010 |
ils | +0.009 |
ior | +0.009 |
obe | +0.009 |
respected | +0.009 |
Neg logit contributions | |
block | -0.009 |
bucket | -0.009 |
ladder | -0.009 |
palm | -0.009 |
foot | -0.008 |
Token | loves |
Token ID | 10408 |
Feature activation | -7.93e-01 |
Pos logit contributions | |
ting | +0.063 |
obe | +0.057 |
ior | +0.053 |
Mr | +0.053 |
Midnight | +0.053 |
Neg logit contributions | |
block | -0.028 |
key | -0.026 |
bucket | -0.025 |
palm | -0.024 |
foot | -0.023 |
Token | her |
Token ID | 607 |
Feature activation | -7.52e-01 |
Pos logit contributions | |
rewarded | +0.023 |
Claus | +0.023 |
ting | +0.023 |
yer | +0.022 |
ils | +0.022 |
Neg logit contributions | |
block | -0.020 |
palm | -0.019 |
brick | -0.019 |
foot | -0.018 |
bucket | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -5.04e-01 |
Pos logit contributions | |
ting | +0.022 |
ils | +0.021 |
rewarded | +0.019 |
respected | +0.018 |
obe | +0.018 |
Neg logit contributions | |
block | -0.022 |
bucket | -0.022 |
palm | -0.021 |
key | -0.021 |
ladder | -0.021 |
Token | She |
Token ID | 1375 |
Feature activation | -4.74e-01 |
Pos logit contributions | |
ting | +0.013 |
ils | +0.013 |
rewarded | +0.012 |
Midnight | +0.012 |
boys | +0.011 |
Neg logit contributions | |
block | -0.015 |
most | -0.015 |
bucket | -0.014 |
brick | -0.014 |
ladder | -0.014 |
Token | wishes |
Token ID | 12802 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
ting | +0.016 |
ating | +0.013 |
ils | +0.013 |
obe | +0.013 |
respected | +0.013 |
Neg logit contributions | |
block | -0.013 |
bucket | -0.013 |
palm | -0.012 |
ladder | -0.012 |
key | -0.012 |
Token | Daddy |
Token ID | 26926 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
ting | +0.010 |
ior | +0.010 |
obe | +0.010 |
respected | +0.010 |
ils | +0.010 |
Neg logit contributions | |
key | -0.007 |
ladder | -0.006 |
block | -0.006 |
bucket | -0.006 |
shovel | -0.006 |
Token | would |
Token ID | 561 |
Feature activation | -2.72e-01 |
Pos logit contributions | |
ting | +0.044 |
obe | +0.042 |
ulating | +0.040 |
ils | +0.039 |
ual | +0.039 |
Neg logit contributions | |
block | -0.032 |
key | -0.029 |
most | -0.028 |
second | -0.026 |
ladder | -0.026 |
Token | see |
Token ID | 766 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
ting | +0.011 |
ils | +0.010 |
led | +0.009 |
ior | +0.009 |
lish | +0.009 |
Neg logit contributions | |
key | -0.006 |
block | -0.006 |
second | -0.005 |
bucket | -0.005 |
foot | -0.005 |
Token | time |
Token ID | 640 |
Feature activation | +7.28e-01 |
Pos logit contributions | |
Midnight | +0.019 |
ils | +0.019 |
obe | +0.018 |
rewarded | +0.018 |
Mr | +0.018 |
Neg logit contributions | |
railing | -0.027 |
chie | -0.027 |
most | -0.026 |
ladder | -0.026 |
brick | -0.026 |
Token | , |
Token ID | 11 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
railing | +0.036 |
most | +0.035 |
adder | +0.035 |
palm | +0.035 |
feature | +0.034 |
Neg logit contributions | |
from | -0.003 |
when | -0.002 |
Midnight | -0.002 |
Mr | -0.002 |
ting | -0.001 |
Token | there |
Token ID | 612 |
Feature activation | +2.72e+00 |
Pos logit contributions | |
ting | +0.003 |
when | +0.003 |
Midnight | +0.003 |
from | +0.003 |
during | +0.002 |
Neg logit contributions | |
most | -0.005 |
adder | -0.005 |
railing | -0.005 |
feature | -0.005 |
ladder | -0.005 |
Token | was |
Token ID | 373 |
Feature activation | +8.63e-01 |
Pos logit contributions | |
most | +0.100 |
adder | +0.100 |
railing | +0.099 |
ladder | +0.097 |
chie | +0.097 |
Neg logit contributions | |
ting | -0.046 |
ils | -0.044 |
rewarded | -0.044 |
Midnight | -0.042 |
treated | -0.040 |
Token | a |
Token ID | 257 |
Feature activation | -7.90e-01 |
Pos logit contributions | |
ladder | +0.031 |
chie | +0.031 |
railing | +0.031 |
most | +0.031 |
adder | +0.029 |
Neg logit contributions | |
Midnight | -0.018 |
ting | -0.016 |
Rudolph | -0.015 |
Mr | -0.015 |
ils | -0.015 |
Token | lazy |
Token ID | 16931 |
Feature activation | -9.77e-01 |
Pos logit contributions | |
ting | +0.025 |
Midnight | +0.025 |
rewarded | +0.024 |
ils | +0.024 |
Claus | +0.023 |
Neg logit contributions | |
most | -0.018 |
adder | -0.017 |
block | -0.016 |
railing | -0.016 |
bucket | -0.016 |
Token | fox |
Token ID | 21831 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
ils | +0.037 |
rewarded | +0.037 |
ting | +0.036 |
respected | +0.035 |
punished | +0.034 |
Neg logit contributions | |
block | -0.020 |
bucket | -0.019 |
ladder | -0.019 |
brick | -0.019 |
palm | -0.018 |
Token | living |
Token ID | 2877 |
Feature activation | -2.43e+00 |
Pos logit contributions | |
rewarded | +0.058 |
respected | +0.056 |
ting | +0.055 |
ils | +0.055 |
ior | +0.054 |
Neg logit contributions | |
block | -0.049 |
palm | -0.045 |
foot | -0.044 |
most | -0.043 |
brick | -0.043 |
Token | in |
Token ID | 287 |
Feature activation | -6.26e-02 |
Pos logit contributions | |
rewarded | +0.055 |
ting | +0.054 |
ils | +0.052 |
Midnight | +0.052 |
obe | +0.049 |
Neg logit contributions | |
block | -0.083 |
bucket | -0.080 |
foot | -0.079 |
palm | -0.076 |
nail | -0.076 |
Token | a |
Token ID | 257 |
Feature activation | +3.79e-01 |
Pos logit contributions | |
ting | +0.001 |
Midnight | +0.001 |
ils | +0.001 |
rewarded | +0.001 |
Rudolph | +0.001 |
Neg logit contributions | |
most | -0.002 |
adder | -0.002 |
bucket | -0.002 |
block | -0.002 |
ladder | -0.002 |
Token | big |
Token ID | 1263 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
adder | +0.008 |
most | +0.008 |
block | +0.008 |
brick | +0.008 |
railing | +0.008 |
Neg logit contributions | |
ting | -0.012 |
rewarded | -0.012 |
Midnight | -0.012 |
ils | -0.012 |
boys | -0.011 |
Token | mouth |
Token ID | 5422 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
Midnight | +0.004 |
rewarded | +0.004 |
ils | +0.004 |
when | +0.004 |
treated | +0.004 |
Neg logit contributions | |
adder | -0.004 |
chie | -0.004 |
inspection | -0.003 |
most | -0.003 |
path | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -2.89e-01 |
Pos logit contributions | |
ils | +0.034 |
ting | +0.034 |
rewarded | +0.032 |
Midnight | +0.031 |
boys | +0.030 |
Neg logit contributions | |
block | -0.056 |
most | -0.055 |
bucket | -0.054 |
palm | -0.053 |
brick | -0.053 |
Token | She |
Token ID | 1375 |
Feature activation | -5.22e-01 |
Pos logit contributions | |
ting | +0.008 |
Midnight | +0.008 |
rewarded | +0.007 |
ils | +0.007 |
Mr | +0.007 |
Neg logit contributions | |
most | -0.008 |
adder | -0.008 |
block | -0.008 |
chie | -0.008 |
bucket | -0.008 |
Token | tried |
Token ID | 3088 |
Feature activation | +9.50e-01 |
Pos logit contributions | |
ting | +0.013 |
ils | +0.012 |
rewarded | +0.012 |
Midnight | +0.011 |
boys | +0.011 |
Neg logit contributions | |
most | -0.016 |
block | -0.016 |
bucket | -0.016 |
brick | -0.015 |
ladder | -0.015 |
Token | to |
Token ID | 284 |
Feature activation | +8.44e-01 |
Pos logit contributions | |
adder | +0.036 |
most | +0.036 |
chie | +0.036 |
bucket | +0.035 |
railing | +0.035 |
Neg logit contributions | |
ting | -0.017 |
Midnight | -0.014 |
ils | -0.014 |
rewarded | -0.014 |
from | -0.013 |
Token | chew |
Token ID | 34722 |
Feature activation | -3.16e+00 |
Pos logit contributions | |
most | +0.022 |
block | +0.021 |
adder | +0.021 |
brick | +0.020 |
bucket | +0.020 |
Neg logit contributions | |
ting | -0.025 |
ils | -0.024 |
rewarded | -0.024 |
Midnight | -0.023 |
boys | -0.022 |
Token | it |
Token ID | 340 |
Feature activation | +5.41e-01 |
Pos logit contributions | |
ting | +0.082 |
ils | +0.080 |
rewarded | +0.079 |
Midnight | +0.074 |
boys | +0.072 |
Neg logit contributions | |
block | -0.093 |
most | -0.092 |
bucket | -0.090 |
brick | -0.088 |
ladder | -0.088 |
Token | , |
Token ID | 11 |
Feature activation | -4.68e-01 |
Pos logit contributions | |
block | +0.018 |
most | +0.018 |
bucket | +0.017 |
ladder | +0.017 |
brick | +0.017 |
Neg logit contributions | |
ting | -0.012 |
ils | -0.012 |
rewarded | -0.012 |
Midnight | -0.011 |
Claus | -0.011 |
Token | but |
Token ID | 475 |
Feature activation | +3.12e-01 |
Pos logit contributions | |
ting | +0.013 |
rewarded | +0.012 |
ils | +0.012 |
Midnight | +0.012 |
punished | +0.011 |
Neg logit contributions | |
most | -0.013 |
block | -0.013 |
adder | -0.012 |
bucket | -0.012 |
brick | -0.012 |
Token | it |
Token ID | 340 |
Feature activation | -4.45e-01 |
Pos logit contributions | |
most | +0.010 |
adder | +0.010 |
block | +0.010 |
bucket | +0.010 |
brick | +0.010 |
Neg logit contributions | |
ting | -0.007 |
Midnight | -0.007 |
ils | -0.007 |
rewarded | -0.007 |
punished | -0.006 |
Token | was |
Token ID | 373 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
ils | +0.008 |
ting | +0.008 |
rewarded | +0.008 |
Midnight | +0.008 |
punished | +0.007 |
Neg logit contributions | |
adder | -0.016 |
most | -0.016 |
railing | -0.016 |
block | -0.015 |
brick | -0.015 |
Token | Smith |
Token ID | 4176 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
ils | +0.002 |
ting | +0.002 |
rewarded | +0.002 |
punished | +0.002 |
Midnight | +0.002 |
Neg logit contributions | |
most | -0.003 |
block | -0.003 |
bucket | -0.003 |
brick | -0.003 |
ladder | -0.003 |
Token | because |
Token ID | 780 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
ting | +0.035 |
ils | +0.031 |
rewarded | +0.031 |
obe | +0.030 |
punished | +0.030 |
Neg logit contributions | |
block | -0.038 |
bucket | -0.034 |
ladder | -0.034 |
most | -0.034 |
palm | -0.034 |
Token | she |
Token ID | 673 |
Feature activation | -9.77e-01 |
Pos logit contributions | |
brick | +0.003 |
block | +0.003 |
ladder | +0.003 |
key | +0.003 |
most | +0.003 |
Neg logit contributions | |
ils | -0.004 |
obe | -0.004 |
ting | -0.004 |
rewarded | -0.004 |
punished | -0.004 |
Token | always |
Token ID | 1464 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
ting | +0.033 |
ils | +0.030 |
respected | +0.030 |
obe | +0.030 |
orting | +0.029 |
Neg logit contributions | |
block | -0.022 |
most | -0.019 |
palm | -0.019 |
key | -0.018 |
bucket | -0.018 |
Token | had |
Token ID | 550 |
Feature activation | +7.33e-01 |
Pos logit contributions | |
ting | +0.049 |
ils | +0.047 |
Mr | +0.044 |
ups | +0.044 |
Midnight | +0.043 |
Neg logit contributions | |
block | -0.034 |
bucket | -0.033 |
key | -0.033 |
shovel | -0.032 |
palm | -0.031 |
Token | cookies |
Token ID | 14746 |
Feature activation | -3.29e+00 |
Pos logit contributions | |
palm | +0.013 |
bucket | +0.012 |
block | +0.012 |
second | +0.012 |
key | +0.012 |
Neg logit contributions | |
ting | -0.029 |
obe | -0.027 |
respected | -0.026 |
Claus | -0.026 |
ior | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
ior | +0.105 |
ting | +0.104 |
pty | +0.099 |
minded | +0.097 |
ica | +0.096 |
Neg logit contributions | |
block | -0.073 |
key | -0.070 |
bucket | -0.063 |
most | -0.060 |
second | -0.059 |
Token | milk |
Token ID | 7545 |
Feature activation | -2.92e+00 |
Pos logit contributions | |
shovel | +0.026 |
bucket | +0.026 |
block | +0.025 |
palm | +0.025 |
brick | +0.024 |
Neg logit contributions | |
ting | -0.038 |
rewarded | -0.035 |
yer | -0.035 |
minded | -0.034 |
obe | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | +2.23e-02 |
Pos logit contributions | |
rewarded | +0.099 |
respected | +0.097 |
ica | +0.097 |
ting | +0.096 |
pty | +0.096 |
Neg logit contributions | |
block | -0.061 |
bucket | -0.058 |
ladder | -0.056 |
key | -0.053 |
shovel | -0.051 |
Token |
|
Token ID | 198 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
block | +0.000 |
most | +0.000 |
bucket | +0.000 |
ladder | +0.000 |
brick | +0.000 |
Neg logit contributions | |
ting | -0.001 |
ils | -0.001 |
rewarded | -0.001 |
punished | -0.001 |
boys | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -4.47e-01 |
Pos logit contributions | |
Midnight | +0.004 |
ting | +0.004 |
ils | +0.004 |
rewarded | +0.004 |
Mr | +0.004 |
Neg logit contributions | |
most | -0.006 |
bucket | -0.006 |
railing | -0.006 |
chie | -0.006 |
shovel | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
rewarded | +0.037 |
ils | +0.037 |
ting | +0.037 |
respected | +0.035 |
obe | +0.035 |
Neg logit contributions | |
palm | -0.037 |
block | -0.037 |
brick | -0.037 |
bucket | -0.036 |
nail | -0.036 |
Token | frog |
Token ID | 21264 |
Feature activation | +8.59e-01 |
Pos logit contributions | |
block | +0.002 |
bucket | +0.002 |
ladder | +0.002 |
brick | +0.002 |
palm | +0.002 |
Neg logit contributions | |
ting | -0.004 |
ils | -0.004 |
rewarded | -0.004 |
Midnight | -0.004 |
respected | -0.004 |
Token | 's |
Token ID | 338 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
block | +0.026 |
palm | +0.025 |
brick | +0.024 |
bucket | +0.024 |
most | +0.024 |
Neg logit contributions | |
respected | -0.022 |
ting | -0.022 |
obe | -0.022 |
Mr | -0.022 |
ior | -0.021 |
Token | habitat |
Token ID | 20018 |
Feature activation | +6.79e-02 |
Pos logit contributions | |
ting | +0.009 |
ils | +0.009 |
rewarded | +0.008 |
Midnight | +0.008 |
Mr | +0.008 |
Neg logit contributions | |
bucket | -0.006 |
block | -0.006 |
palm | -0.005 |
brick | -0.005 |
ladder | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | -9.06e-02 |
Pos logit contributions | |
block | +0.002 |
most | +0.002 |
bucket | +0.002 |
palm | +0.002 |
foot | +0.002 |
Neg logit contributions | |
Midnight | -0.002 |
ting | -0.002 |
ils | -0.002 |
boys | -0.002 |
Claus | -0.002 |
Token | diet |
Token ID | 5496 |
Feature activation | -3.43e+00 |
Pos logit contributions | |
ils | +0.003 |
ting | +0.002 |
Claus | +0.002 |
Midnight | +0.002 |
obe | +0.002 |
Neg logit contributions | |
block | -0.003 |
palm | -0.003 |
bucket | -0.003 |
brick | -0.003 |
most | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
Midnight | +0.097 |
ting | +0.094 |
Claus | +0.093 |
punished | +0.093 |
rewarded | +0.093 |
Neg logit contributions | |
block | -0.089 |
most | -0.087 |
palm | -0.086 |
foot | -0.083 |
bucket | -0.083 |
Token | how |
Token ID | 703 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
block | +0.033 |
palm | +0.033 |
bucket | +0.032 |
brick | +0.032 |
shovel | +0.031 |
Neg logit contributions | |
ting | -0.032 |
ils | -0.032 |
Midnight | -0.032 |
obe | -0.031 |
yer | -0.031 |
Token | it |
Token ID | 340 |
Feature activation | -3.93e-01 |
Pos logit contributions | |
ting | +0.041 |
obe | +0.038 |
ils | +0.037 |
Midnight | +0.037 |
boys | +0.036 |
Neg logit contributions | |
ladder | -0.026 |
brick | -0.025 |
bucket | -0.025 |
block | -0.025 |
palm | -0.025 |
Token | looked |
Token ID | 3114 |
Feature activation | +9.19e-01 |
Pos logit contributions | |
ting | +0.012 |
ils | +0.011 |
obe | +0.011 |
respected | +0.011 |
Midnight | +0.011 |
Neg logit contributions | |
block | -0.010 |
bucket | -0.010 |
most | -0.010 |
ladder | -0.009 |
nail | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +6.07e-02 |
Pos logit contributions | |
bucket | +0.021 |
second | +0.021 |
block | +0.020 |
palm | +0.020 |
key | +0.019 |
Neg logit contributions | |
obe | -0.030 |
ils | -0.030 |
Midnight | -0.029 |
ior | -0.029 |
boys | -0.028 |
Token | some |
Token ID | 617 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
ting | +0.029 |
ils | +0.029 |
rewarded | +0.028 |
Midnight | +0.026 |
boys | +0.026 |
Neg logit contributions | |
block | -0.024 |
most | -0.023 |
bucket | -0.023 |
brick | -0.023 |
ladder | -0.022 |
Token | cookies |
Token ID | 14746 |
Feature activation | -2.88e+00 |
Pos logit contributions | |
ting | +0.006 |
ils | +0.006 |
rewarded | +0.006 |
Claus | +0.006 |
obe | +0.006 |
Neg logit contributions | |
block | -0.006 |
bucket | -0.006 |
ladder | -0.005 |
brick | -0.005 |
nail | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
ils | +0.082 |
ting | +0.079 |
rewarded | +0.077 |
ior | +0.077 |
minded | +0.076 |
Neg logit contributions | |
bucket | -0.081 |
block | -0.081 |
shovel | -0.076 |
key | -0.073 |
brick | -0.073 |
Token | asked |
Token ID | 1965 |
Feature activation | +3.07e-01 |
Pos logit contributions | |
bucket | +0.041 |
block | +0.040 |
shovel | +0.039 |
brick | +0.039 |
ladder | +0.038 |
Neg logit contributions | |
ting | -0.038 |
ils | -0.035 |
obe | -0.033 |
rewarded | -0.033 |
minded | -0.032 |
Token | for |
Token ID | 329 |
Feature activation | -2.55e+00 |
Pos logit contributions | |
brick | +0.008 |
block | +0.008 |
shovel | +0.008 |
key | +0.008 |
bucket | +0.008 |
Neg logit contributions | |
ting | -0.008 |
ils | -0.008 |
Claus | -0.008 |
rewarded | -0.008 |
respected | -0.007 |
Token | forgiveness |
Token ID | 26027 |
Feature activation | -3.11e+00 |
Pos logit contributions | |
ting | +0.082 |
ils | +0.077 |
rewarded | +0.076 |
obe | +0.074 |
Claus | +0.074 |
Neg logit contributions | |
block | -0.061 |
bucket | -0.060 |
second | -0.058 |
brick | -0.057 |
shovel | -0.056 |
Token | . |
Token ID | 13 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
rewarded | +0.091 |
ting | +0.091 |
Claus | +0.087 |
Midnight | +0.085 |
obe | +0.085 |
Neg logit contributions | |
block | -0.075 |
foot | -0.071 |
second | -0.066 |
bucket | -0.066 |
most | -0.065 |
Token | His |
Token ID | 2399 |
Feature activation | +2.56e-01 |
Pos logit contributions | |
most | +0.004 |
block | +0.004 |
bucket | +0.004 |
adder | +0.004 |
railing | +0.004 |
Neg logit contributions | |
ting | -0.003 |
rewarded | -0.003 |
Midnight | -0.003 |
ils | -0.003 |
obe | -0.003 |
Token | friend |
Token ID | 1545 |
Feature activation | +6.94e-01 |
Pos logit contributions | |
most | +0.005 |
railing | +0.005 |
block | +0.005 |
adder | +0.005 |
bucket | +0.005 |
Neg logit contributions | |
rewarded | -0.009 |
ting | -0.009 |
ils | -0.009 |
boys | -0.008 |
Midnight | -0.008 |
Token | forg |
Token ID | 5784 |
Feature activation | -7.98e-01 |
Pos logit contributions | |
block | +0.017 |
second | +0.016 |
bucket | +0.015 |
brick | +0.015 |
nail | +0.015 |
Neg logit contributions | |
ting | -0.024 |
ils | -0.021 |
obe | -0.020 |
falls | -0.019 |
iling | -0.019 |
Token | ave |
Token ID | 1015 |
Feature activation | +4.82e-01 |
Pos logit contributions | |
ils | +0.010 |
ting | +0.009 |
rewarded | +0.009 |
Midnight | +0.007 |
punished | +0.007 |
Neg logit contributions | |
block | -0.035 |
most | -0.035 |
bucket | -0.035 |
ladder | -0.034 |
brick | -0.034 |
Token | back |
Token ID | 736 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
ils | +0.015 |
ting | +0.014 |
rewarded | +0.014 |
Midnight | +0.013 |
boys | +0.013 |
Neg logit contributions | |
block | -0.017 |
most | -0.017 |
ladder | -0.016 |
bucket | -0.016 |
brick | -0.016 |
Token | inside |
Token ID | 2641 |
Feature activation | -7.73e-01 |
Pos logit contributions | |
ting | +0.005 |
ils | +0.005 |
rewarded | +0.005 |
Midnight | +0.005 |
boys | +0.005 |
Neg logit contributions | |
block | -0.007 |
most | -0.007 |
bucket | -0.007 |
ladder | -0.007 |
brick | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | -2.97e-01 |
Pos logit contributions | |
ils | +0.018 |
ting | +0.018 |
rewarded | +0.017 |
boys | +0.016 |
Midnight | +0.016 |
Neg logit contributions | |
block | -0.025 |
most | -0.025 |
bucket | -0.025 |
ladder | -0.024 |
brick | -0.024 |
Token | went |
Token ID | 1816 |
Feature activation | -9.11e-01 |
Pos logit contributions | |
ting | +0.008 |
ils | +0.008 |
obe | +0.008 |
respected | +0.008 |
ior | +0.007 |
Neg logit contributions | |
bucket | -0.008 |
block | -0.008 |
most | -0.008 |
palm | -0.008 |
ladder | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | -4.57e-02 |
Pos logit contributions | |
ils | +0.027 |
ting | +0.026 |
rewarded | +0.025 |
boys | +0.024 |
Midnight | +0.024 |
Neg logit contributions | |
block | -0.025 |
bucket | -0.024 |
ladder | -0.024 |
most | -0.024 |
foot | -0.022 |
Token | bed |
Token ID | 3996 |
Feature activation | -4.01e+00 |
Pos logit contributions | |
ting | +0.001 |
ils | +0.001 |
rewarded | +0.001 |
Midnight | +0.001 |
boys | +0.001 |
Neg logit contributions | |
block | -0.001 |
most | -0.001 |
bucket | -0.001 |
brick | -0.001 |
ladder | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | -3.28e-01 |
Pos logit contributions | |
ils | +0.107 |
ting | +0.106 |
ior | +0.099 |
rewarded | +0.097 |
boys | +0.096 |
Neg logit contributions | |
ladder | -0.114 |
block | -0.112 |
most | -0.110 |
foot | -0.110 |
bucket | -0.109 |
Token | dreaming |
Token ID | 28368 |
Feature activation | +8.40e-01 |
Pos logit contributions | |
ils | +0.010 |
ting | +0.009 |
rewarded | +0.009 |
Midnight | +0.008 |
boys | +0.008 |
Neg logit contributions | |
block | -0.009 |
most | -0.009 |
bucket | -0.009 |
ladder | -0.008 |
brick | -0.008 |
Token | of |
Token ID | 286 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
block | +0.020 |
most | +0.020 |
ladder | +0.019 |
bucket | +0.019 |
foot | +0.019 |
Neg logit contributions | |
ting | -0.027 |
ils | -0.027 |
boys | -0.026 |
Claus | -0.025 |
rewarded | -0.025 |
Token | all |
Token ID | 477 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
ting | +0.025 |
ils | +0.025 |
rewarded | +0.024 |
Midnight | +0.023 |
boys | +0.022 |
Neg logit contributions | |
block | -0.032 |
most | -0.032 |
bucket | -0.031 |
brick | -0.031 |
ladder | -0.031 |
Token | the |
Token ID | 262 |
Feature activation | -2.70e-01 |
Pos logit contributions | |
block | +0.073 |
brick | +0.072 |
most | +0.071 |
ladder | +0.070 |
palm | +0.070 |
Neg logit contributions | |
ils | -0.055 |
rewarded | -0.054 |
ting | -0.053 |
Claus | -0.053 |
boys | -0.050 |