Token | " |
Token ID | 366 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
day | +0.014 |
good | +0.012 |
time | +0.012 |
end | +0.012 |
team | +0.012 |
Neg logit contributions | |
iqu | -0.049 |
ridge | -0.049 |
umes | -0.047 |
yles | -0.046 |
rol | -0.045 |
Token | You |
Token ID | 1639 |
Feature activation | +3.66e-02 |
Pos logit contributions | |
, | +0.037 |
day | +0.036 |
time | +0.035 |
end | +0.035 |
team | +0.034 |
Neg logit contributions | |
ridge | -0.042 |
iqu | -0.041 |
yles | -0.040 |
umes | -0.038 |
cher | -0.038 |
Token | are |
Token ID | 389 |
Feature activation | +3.72e-02 |
Pos logit contributions | |
day | +0.064 |
afternoon | +0.059 |
girl | +0.057 |
time | +0.056 |
their | +0.055 |
Neg logit contributions | |
iqu | -0.081 |
ridge | -0.080 |
vell | -0.077 |
umes | -0.077 |
yles | -0.074 |
Token | welcome |
Token ID | 7062 |
Feature activation | +6.20e-02 |
Pos logit contributions | |
day | +0.043 |
team | +0.040 |
girl | +0.039 |
moment | +0.038 |
afternoon | +0.038 |
Neg logit contributions | |
vell | -0.096 |
iqu | -0.094 |
ridge | -0.092 |
umes | -0.092 |
cher | -0.090 |
Token | , |
Token ID | 11 |
Feature activation | +2.09e-02 |
Pos logit contributions | |
day | +0.043 |
afternoon | +0.037 |
end | +0.030 |
team | +0.030 |
boy | +0.025 |
Neg logit contributions | |
ridge | -0.198 |
iqu | -0.197 |
umes | -0.188 |
angs | -0.186 |
paste | -0.183 |
Token | Anna |
Token ID | 11735 |
Feature activation | +9.88e-02 |
Pos logit contributions | |
day | +0.024 |
afternoon | +0.020 |
team | +0.019 |
town | +0.019 |
, | +0.018 |
Neg logit contributions | |
ridge | -0.056 |
iqu | -0.056 |
yles | -0.055 |
umes | -0.054 |
paste | -0.053 |
Token | . |
Token ID | 13 |
Feature activation | +5.53e-02 |
Pos logit contributions | |
day | +0.121 |
team | +0.120 |
afternoon | +0.118 |
time | +0.104 |
town | +0.103 |
Neg logit contributions | |
ridge | -0.231 |
ache | -0.217 |
paste | -0.217 |
ker | -0.215 |
pled | -0.215 |
Token | You |
Token ID | 921 |
Feature activation | +3.36e-02 |
Pos logit contributions | |
day | +0.109 |
end | +0.107 |
team | +0.105 |
good | +0.105 |
time | +0.104 |
Neg logit contributions | |
ridge | -0.106 |
iqu | -0.106 |
umes | -0.098 |
yles | -0.097 |
rol | -0.096 |
Token | are |
Token ID | 389 |
Feature activation | +3.09e-02 |
Pos logit contributions | |
day | +0.054 |
afternoon | +0.049 |
girl | +0.049 |
boy | +0.047 |
time | +0.046 |
Neg logit contributions | |
ridge | -0.077 |
iqu | -0.076 |
vell | -0.074 |
umes | -0.073 |
cher | -0.072 |
Token | a |
Token ID | 257 |
Feature activation | +5.43e-02 |
Pos logit contributions | |
day | +0.033 |
team | +0.030 |
moment | +0.029 |
girl | +0.029 |
afternoon | +0.029 |
Neg logit contributions | |
iqu | -0.084 |
vell | -0.084 |
ridge | -0.083 |
umes | -0.082 |
cher | -0.080 |
Token | good |
Token ID | 922 |
Feature activation | +5.15e-02 |
Pos logit contributions | |
day | +0.043 |
They | +0.041 |
afternoon | +0.040 |
time | +0.035 |
, | +0.032 |
Neg logit contributions | |
vell | -0.146 |
iqu | -0.142 |
ache | -0.139 |
ridge | -0.139 |
rou | -0.136 |
Token | " |
Token ID | 366 |
Feature activation | +6.15e-03 |
Pos logit contributions | |
day | +0.007 |
good | +0.006 |
time | +0.005 |
end | +0.005 |
team | +0.005 |
Neg logit contributions | |
iqu | -0.021 |
ridge | -0.021 |
umes | -0.020 |
yles | -0.020 |
angs | -0.019 |
Token | You |
Token ID | 1639 |
Feature activation | +2.55e-02 |
Pos logit contributions | |
, | +0.011 |
day | +0.011 |
time | +0.010 |
end | +0.010 |
good | +0.010 |
Neg logit contributions | |
ridge | -0.013 |
iqu | -0.013 |
yles | -0.013 |
umes | -0.013 |
paste | -0.012 |
Token | are |
Token ID | 389 |
Feature activation | +3.01e-02 |
Pos logit contributions | |
day | +0.033 |
good | +0.031 |
, | +0.031 |
time | +0.029 |
end | +0.029 |
Neg logit contributions | |
iqu | -0.073 |
ridge | -0.070 |
umes | -0.068 |
yles | -0.067 |
rol | -0.067 |
Token | right |
Token ID | 826 |
Feature activation | +5.94e-02 |
Pos logit contributions | |
day | +0.030 |
team | +0.027 |
girl | +0.026 |
boy | +0.025 |
afternoon | +0.024 |
Neg logit contributions | |
iqu | -0.089 |
ridge | -0.088 |
vell | -0.086 |
umes | -0.085 |
angs | -0.082 |
Token | , |
Token ID | 11 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
day | +0.029 |
team | +0.020 |
afternoon | +0.017 |
good | +0.017 |
girl | +0.017 |
Neg logit contributions | |
iqu | -0.212 |
ridge | -0.205 |
umes | -0.200 |
yles | -0.197 |
angs | -0.197 |
Token | Anna |
Token ID | 11735 |
Feature activation | +9.33e-02 |
Pos logit contributions | |
day | +0.021 |
afternoon | +0.017 |
team | +0.017 |
town | +0.015 |
end | +0.015 |
Neg logit contributions | |
ridge | -0.048 |
iqu | -0.048 |
yles | -0.047 |
umes | -0.047 |
paste | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | +4.55e-02 |
Pos logit contributions | |
team | +0.104 |
day | +0.103 |
afternoon | +0.099 |
moment | +0.089 |
town | +0.088 |
Neg logit contributions | |
ridge | -0.233 |
iqu | -0.223 |
ker | -0.218 |
paste | -0.217 |
ache | -0.216 |
Token | The |
Token ID | 383 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
day | +0.078 |
good | +0.074 |
, | +0.072 |
end | +0.072 |
time | +0.071 |
Neg logit contributions | |
iqu | -0.107 |
ridge | -0.105 |
umes | -0.099 |
yles | -0.098 |
rol | -0.097 |
Token | book |
Token ID | 1492 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
They | +0.031 |
their | +0.031 |
, | +0.029 |
day | +0.029 |
good | +0.028 |
Neg logit contributions | |
ridge | -0.037 |
iqu | -0.036 |
vell | -0.036 |
cher | -0.034 |
yles | -0.033 |
Token | is |
Token ID | 318 |
Feature activation | -1.68e-03 |
Pos logit contributions | |
iqu | +0.039 |
ridge | +0.039 |
angs | +0.037 |
umes | +0.037 |
paste | +0.036 |
Neg logit contributions | |
day | -0.013 |
girl | -0.011 |
good | -0.011 |
team | -0.011 |
afternoon | -0.011 |
Token | not |
Token ID | 407 |
Feature activation | -1.31e-03 |
Pos logit contributions | |
iqu | +0.005 |
ridge | +0.005 |
yles | +0.005 |
umes | +0.005 |
rol | +0.005 |
Neg logit contributions | |
day | -0.002 |
good | -0.002 |
, | -0.002 |
time | -0.002 |
and | -0.002 |
Token | " |
Token ID | 366 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
day | +0.009 |
good | +0.008 |
, | +0.008 |
time | +0.007 |
end | +0.007 |
Neg logit contributions | |
iqu | -0.037 |
ridge | -0.036 |
umes | -0.034 |
yles | -0.034 |
rol | -0.034 |
Token | You |
Token ID | 1639 |
Feature activation | +3.41e-02 |
Pos logit contributions | |
day | +0.025 |
, | +0.025 |
good | +0.024 |
end | +0.023 |
team | +0.023 |
Neg logit contributions | |
iqu | -0.037 |
ridge | -0.036 |
yles | -0.034 |
umes | -0.034 |
rol | -0.033 |
Token | are |
Token ID | 389 |
Feature activation | +3.49e-02 |
Pos logit contributions | |
day | +0.064 |
good | +0.060 |
, | +0.059 |
girl | +0.058 |
time | +0.058 |
Neg logit contributions | |
iqu | -0.077 |
ridge | -0.074 |
umes | -0.070 |
yles | -0.069 |
angs | -0.068 |
Token | welcome |
Token ID | 7062 |
Feature activation | +5.89e-02 |
Pos logit contributions | |
day | +0.037 |
team | +0.035 |
girl | +0.033 |
boy | +0.032 |
moment | +0.031 |
Neg logit contributions | |
iqu | -0.098 |
ridge | -0.097 |
vell | -0.094 |
umes | -0.093 |
yles | -0.089 |
Token | , |
Token ID | 11 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
day | +0.025 |
team | +0.015 |
afternoon | +0.015 |
end | +0.013 |
good | +0.012 |
Neg logit contributions | |
iqu | -0.215 |
ridge | -0.208 |
umes | -0.204 |
angs | -0.200 |
yles | -0.198 |
Token | Jill |
Token ID | 22772 |
Feature activation | +9.17e-02 |
Pos logit contributions | |
day | +0.029 |
good | +0.026 |
team | +0.026 |
, | +0.026 |
time | +0.026 |
Neg logit contributions | |
iqu | -0.042 |
ridge | -0.039 |
umes | -0.038 |
yles | -0.038 |
angs | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | +5.27e-02 |
Pos logit contributions | |
team | +0.098 |
day | +0.092 |
town | +0.088 |
afternoon | +0.087 |
end | +0.083 |
Neg logit contributions | |
ridge | -0.237 |
iqu | -0.230 |
ker | -0.227 |
ache | -0.225 |
rou | -0.222 |
Token | You |
Token ID | 921 |
Feature activation | +2.91e-02 |
Pos logit contributions | |
day | +0.093 |
good | +0.089 |
team | +0.088 |
end | +0.087 |
, | +0.087 |
Neg logit contributions | |
iqu | -0.118 |
ridge | -0.116 |
umes | -0.108 |
yles | -0.107 |
rol | -0.106 |
Token | are |
Token ID | 389 |
Feature activation | +2.85e-02 |
Pos logit contributions | |
day | +0.042 |
good | +0.038 |
, | +0.037 |
girl | +0.037 |
time | +0.037 |
Neg logit contributions | |
iqu | -0.076 |
ridge | -0.075 |
umes | -0.071 |
yles | -0.070 |
vell | -0.070 |
Token | the |
Token ID | 262 |
Feature activation | +5.47e-02 |
Pos logit contributions | |
day | +0.030 |
team | +0.026 |
girl | +0.026 |
, | +0.025 |
time | +0.025 |
Neg logit contributions | |
iqu | -0.084 |
ridge | -0.083 |
umes | -0.080 |
vell | -0.079 |
yles | -0.078 |
Token | best |
Token ID | 1266 |
Feature activation | +5.63e-02 |
Pos logit contributions | |
day | +0.043 |
, | +0.039 |
good | +0.035 |
their | +0.034 |
and | +0.034 |
Neg logit contributions | |
iqu | -0.179 |
ridge | -0.173 |
umes | -0.168 |
yles | -0.165 |
vell | -0.164 |
Token | " |
Token ID | 366 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
day | +0.006 |
time | +0.005 |
good | +0.005 |
end | +0.005 |
team | +0.005 |
Neg logit contributions | |
ridge | -0.018 |
iqu | -0.018 |
umes | -0.018 |
yles | -0.017 |
rings | -0.017 |
Token | You |
Token ID | 1639 |
Feature activation | +2.86e-02 |
Pos logit contributions | |
, | +0.020 |
day | +0.019 |
good | +0.018 |
team | +0.018 |
end | +0.018 |
Neg logit contributions | |
ridge | -0.025 |
iqu | -0.024 |
yles | -0.024 |
umes | -0.023 |
paste | -0.023 |
Token | 're |
Token ID | 821 |
Feature activation | +2.91e-02 |
Pos logit contributions | |
day | +0.047 |
good | +0.043 |
, | +0.043 |
girl | +0.042 |
time | +0.042 |
Neg logit contributions | |
iqu | -0.069 |
ridge | -0.068 |
umes | -0.065 |
yles | -0.064 |
angs | -0.063 |
Token | welcome |
Token ID | 7062 |
Feature activation | +5.88e-02 |
Pos logit contributions | |
day | +0.028 |
team | +0.027 |
girl | +0.026 |
boy | +0.025 |
moment | +0.023 |
Neg logit contributions | |
iqu | -0.083 |
ridge | -0.083 |
umes | -0.080 |
vell | -0.080 |
yles | -0.077 |
Token | , |
Token ID | 11 |
Feature activation | +1.98e-02 |
Pos logit contributions | |
day | +0.040 |
afternoon | +0.033 |
team | +0.029 |
end | +0.025 |
boy | +0.024 |
Neg logit contributions | |
iqu | -0.191 |
ridge | -0.187 |
umes | -0.183 |
angs | -0.178 |
rings | -0.175 |
Token | Anna |
Token ID | 11735 |
Feature activation | +9.07e-02 |
Pos logit contributions | |
day | +0.023 |
afternoon | +0.020 |
team | +0.019 |
town | +0.019 |
They | +0.018 |
Neg logit contributions | |
ridge | -0.051 |
iqu | -0.051 |
yles | -0.050 |
umes | -0.050 |
paste | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | +5.33e-02 |
Pos logit contributions | |
day | +0.087 |
team | +0.084 |
afternoon | +0.082 |
end | +0.072 |
time | +0.071 |
Neg logit contributions | |
ridge | -0.246 |
ker | -0.229 |
ache | -0.228 |
umes | -0.228 |
iqu | -0.227 |
Token | I |
Token ID | 314 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
day | +0.098 |
good | +0.094 |
team | +0.093 |
end | +0.093 |
, | +0.092 |
Neg logit contributions | |
iqu | -0.113 |
ridge | -0.113 |
umes | -0.105 |
yles | -0.104 |
rol | -0.103 |
Token | love |
Token ID | 1842 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
day | +0.030 |
good | +0.027 |
girl | +0.026 |
, | +0.026 |
team | +0.026 |
Neg logit contributions | |
iqu | -0.063 |
ridge | -0.062 |
umes | -0.059 |
yles | -0.058 |
rol | -0.058 |
Token | you |
Token ID | 345 |
Feature activation | +4.51e-02 |
Pos logit contributions | |
day | +0.030 |
good | +0.028 |
, | +0.028 |
end | +0.027 |
girl | +0.027 |
Neg logit contributions | |
iqu | -0.044 |
ridge | -0.043 |
yles | -0.041 |
umes | -0.041 |
angs | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | +4.21e-02 |
Pos logit contributions | |
day | +0.055 |
girl | +0.052 |
boy | +0.051 |
afternoon | +0.050 |
moment | +0.048 |
Neg logit contributions | |
ridge | -0.114 |
umes | -0.109 |
rings | -0.103 |
cher | -0.103 |
iqu | -0.102 |
Token | love |
Token ID | 1842 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
day | +0.052 |
team | +0.048 |
afternoon | +0.047 |
girl | +0.047 |
time | +0.045 |
Neg logit contributions | |
iqu | -0.100 |
ridge | -0.099 |
umes | -0.094 |
cher | -0.092 |
yles | -0.091 |
Token | you |
Token ID | 345 |
Feature activation | +4.97e-02 |
Pos logit contributions | |
day | +0.033 |
good | +0.032 |
girl | +0.031 |
end | +0.031 |
, | +0.031 |
Neg logit contributions | |
iqu | -0.045 |
ridge | -0.044 |
yles | -0.042 |
umes | -0.041 |
rol | -0.040 |
Token | , |
Token ID | 11 |
Feature activation | +7.17e-03 |
Pos logit contributions | |
day | +0.075 |
afternoon | +0.069 |
moment | +0.067 |
girl | +0.066 |
boy | +0.065 |
Neg logit contributions | |
ridge | -0.113 |
umes | -0.106 |
bar | -0.102 |
iqu | -0.102 |
ker | -0.101 |
Token | too |
Token ID | 1165 |
Feature activation | +5.93e-02 |
Pos logit contributions | |
day | +0.009 |
afternoon | +0.008 |
good | +0.007 |
time | +0.007 |
, | +0.007 |
Neg logit contributions | |
iqu | -0.019 |
ridge | -0.019 |
yles | -0.018 |
umes | -0.018 |
ocol | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
day | +0.040 |
afternoon | +0.032 |
girl | +0.031 |
end | +0.028 |
team | +0.028 |
Neg logit contributions | |
iqu | -0.196 |
ridge | -0.184 |
umes | -0.183 |
angs | -0.181 |
yles | -0.180 |
Token | Anna |
Token ID | 11735 |
Feature activation | +8.99e-02 |
Pos logit contributions | |
day | +0.022 |
afternoon | +0.019 |
town | +0.018 |
good | +0.018 |
team | +0.017 |
Neg logit contributions | |
ridge | -0.041 |
iqu | -0.040 |
yles | -0.040 |
umes | -0.039 |
rings | -0.038 |
Token | and |
Token ID | 290 |
Feature activation | +3.41e-02 |
Pos logit contributions | |
day | +0.085 |
team | +0.083 |
afternoon | +0.081 |
end | +0.074 |
good | +0.072 |
Neg logit contributions | |
ridge | -0.239 |
ache | -0.227 |
paste | -0.227 |
ker | -0.225 |
cher | -0.223 |
Token | Ben |
Token ID | 3932 |
Feature activation | +6.33e-02 |
Pos logit contributions | |
day | +0.042 |
time | +0.037 |
good | +0.037 |
end | +0.036 |
, | +0.035 |
Neg logit contributions | |
ridge | -0.095 |
iqu | -0.095 |
yles | -0.091 |
umes | -0.090 |
rings | -0.087 |
Token | . |
Token ID | 13 |
Feature activation | +4.87e-02 |
Pos logit contributions | |
team | +0.082 |
day | +0.080 |
afternoon | +0.079 |
moment | +0.073 |
end | +0.072 |
Neg logit contributions | |
ridge | -0.137 |
ache | -0.128 |
cher | -0.127 |
angs | -0.124 |
paste | -0.123 |
Token | You |
Token ID | 921 |
Feature activation | +2.66e-02 |
Pos logit contributions | |
day | +0.081 |
good | +0.077 |
, | +0.075 |
end | +0.075 |
team | +0.075 |
Neg logit contributions | |
iqu | -0.112 |
ridge | -0.111 |
umes | -0.104 |
yles | -0.104 |
rol | -0.103 |
Token | are |
Token ID | 389 |
Feature activation | +2.96e-02 |
Pos logit contributions | |
day | +0.043 |
afternoon | +0.040 |
girl | +0.040 |
their | +0.038 |
boy | +0.038 |
Neg logit contributions | |
ridge | -0.058 |
vell | -0.057 |
iqu | -0.056 |
umes | -0.055 |
cher | -0.055 |
Token | says |
Token ID | 1139 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
day | +0.014 |
good | +0.013 |
, | +0.013 |
time | +0.012 |
end | +0.012 |
Neg logit contributions | |
iqu | -0.029 |
ridge | -0.028 |
umes | -0.027 |
yles | -0.027 |
rol | -0.026 |
Token | , |
Token ID | 11 |
Feature activation | -4.06e-04 |
Pos logit contributions | |
iqu | +0.067 |
ridge | +0.066 |
umes | +0.064 |
yles | +0.063 |
angs | +0.063 |
Neg logit contributions | |
day | -0.006 |
end | -0.003 |
afternoon | -0.003 |
time | -0.003 |
girl | -0.003 |
Token | " |
Token ID | 366 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
iqu | +0.001 |
ridge | +0.001 |
umes | +0.001 |
yles | +0.001 |
rol | +0.001 |
Neg logit contributions | |
day | -0.000 |
good | -0.000 |
, | -0.000 |
time | -0.000 |
end | -0.000 |
Token | Sure |
Token ID | 19457 |
Feature activation | +2.89e-02 |
Pos logit contributions | |
iqu | +0.028 |
ridge | +0.027 |
umes | +0.026 |
yles | +0.026 |
rol | +0.026 |
Neg logit contributions | |
day | -0.021 |
, | -0.021 |
good | -0.020 |
time | -0.020 |
end | -0.020 |
Token | , |
Token ID | 11 |
Feature activation | +5.85e-03 |
Pos logit contributions | |
day | +0.015 |
team | +0.013 |
afternoon | +0.012 |
They | +0.011 |
end | +0.010 |
Neg logit contributions | |
iqu | -0.094 |
ridge | -0.094 |
umes | -0.089 |
angs | -0.088 |
rings | -0.088 |
Token | Anna |
Token ID | 11735 |
Feature activation | +8.98e-02 |
Pos logit contributions | |
day | +0.007 |
good | +0.006 |
end | +0.006 |
team | +0.006 |
, | +0.006 |
Neg logit contributions | |
iqu | -0.016 |
ridge | -0.016 |
umes | -0.016 |
yles | -0.015 |
rol | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +3.08e-02 |
Pos logit contributions | |
team | +0.091 |
day | +0.091 |
end | +0.087 |
afternoon | +0.085 |
girl | +0.083 |
Neg logit contributions | |
ridge | -0.232 |
iqu | -0.230 |
ker | -0.219 |
angs | -0.217 |
cher | -0.217 |
Token | You |
Token ID | 921 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
day | +0.045 |
good | +0.043 |
, | +0.041 |
time | +0.041 |
end | +0.041 |
Neg logit contributions | |
iqu | -0.080 |
ridge | -0.078 |
umes | -0.075 |
yles | -0.073 |
rol | -0.073 |
Token | can |
Token ID | 460 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
day | +0.023 |
good | +0.021 |
, | +0.021 |
time | +0.020 |
girl | +0.020 |
Neg logit contributions | |
iqu | -0.049 |
ridge | -0.048 |
umes | -0.046 |
yles | -0.045 |
rol | -0.045 |
Token | call |
Token ID | 869 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
iqu | +0.051 |
ridge | +0.049 |
umes | +0.048 |
yles | +0.048 |
rol | +0.047 |
Neg logit contributions | |
day | -0.022 |
, | -0.021 |
good | -0.021 |
and | -0.020 |
end | -0.020 |
Token | grandma |
Token ID | 49890 |
Feature activation | +1.53e-02 |
Pos logit contributions | |
iqu | +0.042 |
ridge | +0.040 |
umes | +0.039 |
yles | +0.038 |
rol | +0.038 |
Neg logit contributions | |
day | -0.016 |
good | -0.015 |
, | -0.015 |
time | -0.014 |
end | -0.014 |
Token | " |
Token ID | 366 |
Feature activation | +1.37e-02 |
Pos logit contributions | |
day | +0.008 |
good | +0.007 |
time | +0.006 |
, | +0.006 |
end | +0.006 |
Neg logit contributions | |
iqu | -0.029 |
ridge | -0.028 |
umes | -0.027 |
yles | -0.027 |
rol | -0.026 |
Token | You |
Token ID | 1639 |
Feature activation | +3.27e-02 |
Pos logit contributions | |
, | +0.024 |
day | +0.024 |
good | +0.023 |
time | +0.023 |
team | +0.023 |
Neg logit contributions | |
iqu | -0.029 |
ridge | -0.029 |
yles | -0.028 |
umes | -0.028 |
rol | -0.027 |
Token | 're |
Token ID | 821 |
Feature activation | +3.21e-02 |
Pos logit contributions | |
day | +0.067 |
, | +0.062 |
good | +0.062 |
girl | +0.061 |
end | +0.061 |
Neg logit contributions | |
iqu | -0.067 |
ridge | -0.064 |
umes | -0.061 |
yles | -0.060 |
angs | -0.059 |
Token | welcome |
Token ID | 7062 |
Feature activation | +5.84e-02 |
Pos logit contributions | |
day | +0.032 |
team | +0.031 |
girl | +0.030 |
boy | +0.029 |
afternoon | +0.026 |
Neg logit contributions | |
iqu | -0.091 |
ridge | -0.089 |
vell | -0.088 |
umes | -0.088 |
rol | -0.085 |
Token | , |
Token ID | 11 |
Feature activation | +2.44e-02 |
Pos logit contributions | |
day | +0.032 |
afternoon | +0.024 |
team | +0.021 |
end | +0.018 |
boy | +0.016 |
Neg logit contributions | |
iqu | -0.204 |
ridge | -0.195 |
umes | -0.193 |
angs | -0.190 |
rings | -0.186 |
Token | Lily |
Token ID | 20037 |
Feature activation | +8.96e-02 |
Pos logit contributions | |
day | +0.025 |
afternoon | +0.020 |
good | +0.020 |
, | +0.020 |
team | +0.020 |
Neg logit contributions | |
iqu | -0.072 |
ridge | -0.070 |
umes | -0.069 |
yles | -0.068 |
bar | -0.066 |
Token | . |
Token ID | 13 |
Feature activation | +5.23e-02 |
Pos logit contributions | |
day | +0.096 |
team | +0.093 |
afternoon | +0.089 |
town | +0.086 |
time | +0.085 |
Neg logit contributions | |
ridge | -0.231 |
iqu | -0.221 |
ker | -0.219 |
ache | -0.214 |
ous | -0.213 |
Token | I |
Token ID | 314 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
day | +0.098 |
team | +0.096 |
end | +0.095 |
good | +0.095 |
time | +0.094 |
Neg logit contributions | |
iqu | -0.106 |
ridge | -0.104 |
umes | -0.100 |
yles | -0.097 |
rol | -0.096 |
Token | 'm |
Token ID | 1101 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
day | +0.036 |
good | +0.033 |
girl | +0.032 |
team | +0.032 |
, | +0.032 |
Neg logit contributions | |
iqu | -0.062 |
ridge | -0.061 |
umes | -0.058 |
yles | -0.057 |
rol | -0.057 |
Token | proud |
Token ID | 6613 |
Feature activation | +5.74e-02 |
Pos logit contributions | |
day | +0.022 |
good | +0.021 |
, | +0.020 |
time | +0.019 |
and | +0.019 |
Neg logit contributions | |
iqu | -0.055 |
ridge | -0.053 |
umes | -0.051 |
yles | -0.051 |
angs | -0.050 |
Token | of |
Token ID | 286 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
day | +0.082 |
team | +0.077 |
good | +0.077 |
end | +0.076 |
girl | +0.075 |
Neg logit contributions | |
iqu | -0.141 |
ridge | -0.134 |
umes | -0.130 |
yles | -0.129 |
angs | -0.128 |
Token | " |
Token ID | 366 |
Feature activation | +1.79e-02 |
Pos logit contributions | |
day | +0.007 |
good | +0.006 |
, | +0.006 |
time | +0.006 |
end | +0.006 |
Neg logit contributions | |
iqu | -0.028 |
ridge | -0.027 |
umes | -0.026 |
yles | -0.026 |
rol | -0.026 |
Token | You |
Token ID | 1639 |
Feature activation | +3.44e-02 |
Pos logit contributions | |
, | +0.033 |
day | +0.032 |
time | +0.031 |
end | +0.031 |
good | +0.031 |
Neg logit contributions | |
ridge | -0.038 |
iqu | -0.037 |
yles | -0.036 |
umes | -0.035 |
paste | -0.034 |
Token | 're |
Token ID | 821 |
Feature activation | +3.80e-02 |
Pos logit contributions | |
day | +0.069 |
afternoon | +0.063 |
girl | +0.062 |
, | +0.062 |
team | +0.062 |
Neg logit contributions | |
iqu | -0.069 |
ridge | -0.067 |
umes | -0.065 |
vell | -0.064 |
angs | -0.062 |
Token | welcome |
Token ID | 7062 |
Feature activation | +6.12e-02 |
Pos logit contributions | |
day | +0.041 |
team | +0.040 |
girl | +0.039 |
boy | +0.038 |
moment | +0.036 |
Neg logit contributions | |
iqu | -0.101 |
vell | -0.100 |
ridge | -0.099 |
umes | -0.098 |
cher | -0.094 |
Token | , |
Token ID | 11 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
day | +0.040 |
afternoon | +0.033 |
team | +0.028 |
end | +0.025 |
boy | +0.022 |
Neg logit contributions | |
iqu | -0.203 |
ridge | -0.196 |
umes | -0.192 |
angs | -0.189 |
ocol | -0.184 |
Token | Lily |
Token ID | 20037 |
Feature activation | +8.93e-02 |
Pos logit contributions | |
day | +0.030 |
afternoon | +0.025 |
, | +0.025 |
time | +0.024 |
good | +0.024 |
Neg logit contributions | |
iqu | -0.067 |
ridge | -0.064 |
umes | -0.063 |
yles | -0.063 |
bar | -0.061 |
Token | and |
Token ID | 290 |
Feature activation | +4.24e-02 |
Pos logit contributions | |
team | +0.086 |
day | +0.082 |
afternoon | +0.078 |
town | +0.074 |
their | +0.073 |
Neg logit contributions | |
ker | -0.228 |
ridge | -0.227 |
pled | -0.221 |
ache | -0.221 |
paste | -0.218 |
Token | Ben |
Token ID | 3932 |
Feature activation | +7.15e-02 |
Pos logit contributions | |
day | +0.052 |
time | +0.047 |
good | +0.046 |
end | +0.045 |
, | +0.044 |
Neg logit contributions | |
iqu | -0.119 |
ridge | -0.117 |
yles | -0.112 |
umes | -0.112 |
rings | -0.107 |
Token | . |
Token ID | 13 |
Feature activation | +5.25e-02 |
Pos logit contributions | |
team | +0.099 |
day | +0.087 |
afternoon | +0.085 |
town | +0.082 |
end | +0.081 |
Neg logit contributions | |
ridge | -0.153 |
iqu | -0.150 |
pled | -0.149 |
angs | -0.143 |
ache | -0.142 |
Token | I |
Token ID | 314 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
team | +0.102 |
day | +0.102 |
end | +0.101 |
time | +0.100 |
good | +0.097 |
Neg logit contributions | |
ridge | -0.099 |
iqu | -0.098 |
umes | -0.092 |
rings | -0.090 |
rol | -0.089 |
Token | 'm |
Token ID | 1101 |
Feature activation | +2.28e-02 |
Pos logit contributions | |
day | +0.036 |
team | +0.034 |
girl | +0.032 |
afternoon | +0.032 |
time | +0.032 |
Neg logit contributions | |
iqu | -0.064 |
ridge | -0.063 |
umes | -0.059 |
cher | -0.058 |
rol | -0.058 |
Token | " |
Token ID | 366 |
Feature activation | +1.67e-02 |
Pos logit contributions | |
day | +0.006 |
good | +0.005 |
, | +0.005 |
time | +0.005 |
end | +0.005 |
Neg logit contributions | |
iqu | -0.025 |
ridge | -0.025 |
umes | -0.024 |
yles | -0.024 |
rol | -0.023 |
Token | You |
Token ID | 1639 |
Feature activation | +3.54e-02 |
Pos logit contributions | |
, | +0.031 |
day | +0.030 |
time | +0.029 |
team | +0.029 |
end | +0.029 |
Neg logit contributions | |
iqu | -0.034 |
ridge | -0.034 |
yles | -0.033 |
umes | -0.032 |
rol | -0.031 |
Token | 're |
Token ID | 821 |
Feature activation | +3.50e-02 |
Pos logit contributions | |
day | +0.076 |
afternoon | +0.070 |
girl | +0.069 |
, | +0.069 |
time | +0.068 |
Neg logit contributions | |
iqu | -0.063 |
ridge | -0.063 |
umes | -0.060 |
vell | -0.059 |
yles | -0.058 |
Token | welcome |
Token ID | 7062 |
Feature activation | +5.72e-02 |
Pos logit contributions | |
day | +0.037 |
team | +0.036 |
girl | +0.035 |
boy | +0.034 |
moment | +0.032 |
Neg logit contributions | |
iqu | -0.093 |
ridge | -0.092 |
vell | -0.092 |
umes | -0.091 |
rol | -0.087 |
Token | , |
Token ID | 11 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
day | +0.033 |
afternoon | +0.025 |
team | +0.023 |
end | +0.019 |
boy | +0.016 |
Neg logit contributions | |
iqu | -0.198 |
ridge | -0.190 |
umes | -0.187 |
angs | -0.183 |
rings | -0.180 |
Token | Lily |
Token ID | 20037 |
Feature activation | +8.92e-02 |
Pos logit contributions | |
day | +0.019 |
, | +0.016 |
good | +0.016 |
time | +0.015 |
team | +0.015 |
Neg logit contributions | |
iqu | -0.064 |
ridge | -0.062 |
umes | -0.061 |
yles | -0.060 |
rol | -0.059 |
Token | . |
Token ID | 13 |
Feature activation | +5.15e-02 |
Pos logit contributions | |
team | +0.095 |
day | +0.091 |
afternoon | +0.087 |
town | +0.084 |
time | +0.080 |
Neg logit contributions | |
ridge | -0.219 |
ker | -0.219 |
iqu | -0.214 |
ache | -0.213 |
pled | -0.209 |
Token | I |
Token ID | 314 |
Feature activation | +2.63e-02 |
Pos logit contributions | |
day | +0.096 |
end | +0.095 |
team | +0.095 |
time | +0.094 |
good | +0.093 |
Neg logit contributions | |
iqu | -0.102 |
ridge | -0.101 |
umes | -0.095 |
yles | -0.094 |
rings | -0.093 |
Token | love |
Token ID | 1842 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
day | +0.037 |
team | +0.034 |
girl | +0.033 |
time | +0.033 |
afternoon | +0.033 |
Neg logit contributions | |
iqu | -0.063 |
ridge | -0.063 |
umes | -0.060 |
yles | -0.058 |
rol | -0.058 |
Token | you |
Token ID | 345 |
Feature activation | +4.37e-02 |
Pos logit contributions | |
day | +0.025 |
good | +0.023 |
, | +0.023 |
time | +0.022 |
girl | +0.022 |
Neg logit contributions | |
iqu | -0.041 |
ridge | -0.039 |
umes | -0.038 |
yles | -0.037 |
rol | -0.037 |
Token | too |
Token ID | 1165 |
Feature activation | +5.36e-02 |
Pos logit contributions | |
day | +0.063 |
afternoon | +0.061 |
girl | +0.060 |
time | +0.059 |
boy | +0.059 |
Neg logit contributions | |
ridge | -0.096 |
umes | -0.091 |
rings | -0.088 |
cher | -0.086 |
ents | -0.084 |
Token | I |
Token ID | 40 |
Feature activation | +2.80e-02 |
Pos logit contributions | |
, | +0.010 |
day | +0.010 |
good | +0.009 |
time | +0.009 |
end | +0.009 |
Neg logit contributions | |
iqu | -0.013 |
ridge | -0.013 |
yles | -0.012 |
umes | -0.012 |
paste | -0.011 |
Token | love |
Token ID | 1842 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
day | +0.037 |
good | +0.034 |
, | +0.033 |
girl | +0.033 |
time | +0.032 |
Neg logit contributions | |
iqu | -0.077 |
ridge | -0.074 |
umes | -0.071 |
yles | -0.071 |
angs | -0.069 |
Token | you |
Token ID | 345 |
Feature activation | +4.39e-02 |
Pos logit contributions | |
day | +0.023 |
good | +0.022 |
, | +0.022 |
time | +0.021 |
girl | +0.021 |
Neg logit contributions | |
iqu | -0.039 |
ridge | -0.038 |
umes | -0.036 |
yles | -0.036 |
rol | -0.035 |
Token | too |
Token ID | 1165 |
Feature activation | +5.42e-02 |
Pos logit contributions | |
day | +0.068 |
girl | +0.062 |
afternoon | +0.061 |
boy | +0.060 |
moment | +0.059 |
Neg logit contributions | |
ridge | -0.100 |
iqu | -0.096 |
umes | -0.095 |
bar | -0.092 |
angs | -0.090 |
Token | , |
Token ID | 11 |
Feature activation | +8.80e-03 |
Pos logit contributions | |
day | +0.031 |
girl | +0.027 |
afternoon | +0.024 |
boy | +0.023 |
team | +0.021 |
Neg logit contributions | |
iqu | -0.182 |
angs | -0.171 |
umes | -0.170 |
ridge | -0.169 |
ocol | -0.166 |
Token | Anna |
Token ID | 11735 |
Feature activation | +8.90e-02 |
Pos logit contributions | |
day | +0.011 |
afternoon | +0.009 |
town | +0.008 |
good | +0.008 |
, | +0.008 |
Neg logit contributions | |
ridge | -0.023 |
iqu | -0.023 |
yles | -0.023 |
umes | -0.022 |
paste | -0.021 |
Token | and |
Token ID | 290 |
Feature activation | +3.22e-02 |
Pos logit contributions | |
day | +0.081 |
team | +0.076 |
afternoon | +0.076 |
good | +0.068 |
their | +0.066 |
Neg logit contributions | |
ridge | -0.243 |
paste | -0.231 |
umes | -0.231 |
ache | -0.230 |
ker | -0.229 |
Token | Ben |
Token ID | 3932 |
Feature activation | +6.22e-02 |
Pos logit contributions | |
day | +0.036 |
good | +0.032 |
, | +0.031 |
time | +0.030 |
end | +0.030 |
Neg logit contributions | |
iqu | -0.096 |
ridge | -0.094 |
umes | -0.091 |
yles | -0.091 |
angs | -0.088 |
Token | . |
Token ID | 13 |
Feature activation | +4.28e-02 |
Pos logit contributions | |
day | +0.081 |
team | +0.080 |
afternoon | +0.080 |
town | +0.072 |
moment | +0.072 |
Neg logit contributions | |
ridge | -0.135 |
ache | -0.127 |
angs | -0.125 |
paste | -0.125 |
iqu | -0.124 |
Token | You |
Token ID | 921 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
day | +0.068 |
good | +0.064 |
, | +0.062 |
end | +0.062 |
team | +0.062 |
Neg logit contributions | |
iqu | -0.103 |
ridge | -0.101 |
umes | -0.096 |
yles | -0.095 |
ocol | -0.093 |
Token | are |
Token ID | 389 |
Feature activation | +2.23e-02 |
Pos logit contributions | |
day | +0.026 |
good | +0.023 |
, | +0.023 |
girl | +0.023 |
time | +0.022 |
Neg logit contributions | |
iqu | -0.048 |
ridge | -0.047 |
umes | -0.045 |
yles | -0.044 |
angs | -0.044 |
Token | " |
Token ID | 366 |
Feature activation | +1.27e-02 |
Pos logit contributions | |
day | +0.002 |
good | +0.001 |
time | +0.001 |
end | +0.001 |
team | +0.001 |
Neg logit contributions | |
iqu | -0.005 |
ridge | -0.005 |
umes | -0.005 |
yles | -0.005 |
rol | -0.005 |
Token | That |
Token ID | 2504 |
Feature activation | +4.08e-02 |
Pos logit contributions | |
, | +0.024 |
day | +0.023 |
time | +0.023 |
end | +0.022 |
good | +0.022 |
Neg logit contributions | |
iqu | -0.026 |
ridge | -0.026 |
yles | -0.025 |
umes | -0.024 |
paste | -0.024 |
Token | 's |
Token ID | 338 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
day | +0.043 |
good | +0.042 |
team | +0.042 |
their | +0.041 |
time | +0.040 |
Neg logit contributions | |
iqu | -0.109 |
ridge | -0.107 |
bar | -0.099 |
ous | -0.098 |
umes | -0.098 |
Token | good |
Token ID | 922 |
Feature activation | +4.81e-02 |
Pos logit contributions | |
team | +0.033 |
girl | +0.029 |
day | +0.028 |
boy | +0.027 |
town | +0.026 |
Neg logit contributions | |
iqu | -0.079 |
vell | -0.074 |
umes | -0.072 |
ridge | -0.071 |
yles | -0.068 |
Token | , |
Token ID | 11 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
day | +0.035 |
afternoon | +0.033 |
end | +0.031 |
time | +0.030 |
town | +0.029 |
Neg logit contributions | |
iqu | -0.139 |
angs | -0.135 |
ous | -0.133 |
ridge | -0.132 |
paste | -0.131 |
Token | Anna |
Token ID | 11735 |
Feature activation | +8.90e-02 |
Pos logit contributions | |
day | +0.014 |
afternoon | +0.011 |
town | +0.010 |
team | +0.010 |
end | +0.010 |
Neg logit contributions | |
iqu | -0.028 |
ridge | -0.028 |
yles | -0.028 |
umes | -0.027 |
paste | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | +3.94e-02 |
Pos logit contributions | |
team | +0.091 |
day | +0.085 |
afternoon | +0.082 |
town | +0.076 |
end | +0.072 |
Neg logit contributions | |
ridge | -0.228 |
ker | -0.223 |
paste | -0.223 |
pled | -0.221 |
ache | -0.220 |
Token | Ben |
Token ID | 3932 |
Feature activation | +6.77e-02 |
Pos logit contributions | |
day | +0.051 |
time | +0.045 |
end | +0.044 |
good | +0.043 |
, | +0.043 |
Neg logit contributions | |
iqu | -0.107 |
ridge | -0.105 |
yles | -0.102 |
umes | -0.101 |
rings | -0.097 |
Token | . |
Token ID | 13 |
Feature activation | +4.39e-02 |
Pos logit contributions | |
team | +0.097 |
afternoon | +0.089 |
day | +0.089 |
town | +0.082 |
sunny | +0.082 |
Neg logit contributions | |
pled | -0.136 |
ridge | -0.135 |
iqu | -0.135 |
ache | -0.134 |
angs | -0.127 |
Token | I |
Token ID | 314 |
Feature activation | +2.77e-02 |
Pos logit contributions | |
day | +0.080 |
good | +0.076 |
time | +0.076 |
end | +0.076 |
team | +0.074 |
Neg logit contributions | |
iqu | -0.097 |
ridge | -0.094 |
umes | -0.089 |
yles | -0.088 |
angs | -0.086 |
Token | am |
Token ID | 716 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
day | +0.039 |
afternoon | +0.035 |
time | +0.034 |
team | +0.034 |
girl | +0.033 |
Neg logit contributions | |
iqu | -0.070 |
ridge | -0.068 |
umes | -0.065 |
yles | -0.063 |
cher | -0.063 |
Token | " |
Token ID | 1 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
day | +0.015 |
time | +0.014 |
good | +0.014 |
team | +0.014 |
, | +0.013 |
Neg logit contributions | |
iqu | -0.035 |
ridge | -0.035 |
umes | -0.035 |
yles | -0.034 |
cher | -0.034 |
Token | You |
Token ID | 1639 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
iqu | +0.028 |
ridge | +0.028 |
yles | +0.027 |
umes | +0.027 |
rol | +0.026 |
Neg logit contributions | |
day | -0.019 |
, | -0.019 |
good | -0.018 |
time | -0.017 |
team | -0.017 |
Token | are |
Token ID | 389 |
Feature activation | +2.85e-02 |
Pos logit contributions | |
day | +0.037 |
good | +0.035 |
, | +0.034 |
girl | +0.033 |
time | +0.033 |
Neg logit contributions | |
iqu | -0.057 |
ridge | -0.055 |
umes | -0.053 |
yles | -0.052 |
rol | -0.051 |
Token | welcome |
Token ID | 7062 |
Feature activation | +5.63e-02 |
Pos logit contributions | |
day | +0.027 |
girl | +0.024 |
team | +0.024 |
moment | +0.023 |
boy | +0.023 |
Neg logit contributions | |
iqu | -0.083 |
ridge | -0.083 |
umes | -0.082 |
vell | -0.081 |
yles | -0.078 |
Token | , |
Token ID | 11 |
Feature activation | +6.16e-03 |
Pos logit contributions | |
day | +0.036 |
afternoon | +0.028 |
end | +0.027 |
team | +0.023 |
good | +0.022 |
Neg logit contributions | |
iqu | -0.188 |
ridge | -0.187 |
umes | -0.181 |
angs | -0.176 |
yles | -0.175 |
Token | Anna |
Token ID | 11735 |
Feature activation | +8.90e-02 |
Pos logit contributions | |
day | +0.007 |
afternoon | +0.006 |
, | +0.006 |
town | +0.006 |
moment | +0.006 |
Neg logit contributions | |
iqu | -0.016 |
ridge | -0.016 |
yles | -0.016 |
umes | -0.016 |
ocol | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | +3.58e-02 |
Pos logit contributions | |
day | +0.066 |
team | +0.065 |
afternoon | +0.057 |
good | +0.057 |
end | +0.055 |
Neg logit contributions | |
ridge | -0.263 |
iqu | -0.254 |
ker | -0.251 |
ache | -0.250 |
umes | -0.249 |
Token | Ben |
Token ID | 3932 |
Feature activation | +6.38e-02 |
Pos logit contributions | |
day | +0.044 |
good | +0.039 |
time | +0.039 |
, | +0.038 |
end | +0.037 |
Neg logit contributions | |
iqu | -0.101 |
ridge | -0.099 |
yles | -0.096 |
umes | -0.094 |
angs | -0.092 |
Token | !" |
Token ID | 2474 |
Feature activation | +2.81e-02 |
Pos logit contributions | |
team | +0.083 |
day | +0.074 |
afternoon | +0.071 |
their | +0.070 |
end | +0.069 |
Neg logit contributions | |
ridge | -0.147 |
ache | -0.144 |
iqu | -0.142 |
ker | -0.140 |
pled | -0.136 |
Token | Mom |
Token ID | 11254 |
Feature activation | -1.17e-03 |
Pos logit contributions | |
day | +0.041 |
afternoon | +0.040 |
team | +0.039 |
time | +0.037 |
whole | +0.037 |
Neg logit contributions | |
iqu | -0.058 |
yles | -0.056 |
rings | -0.056 |
ents | -0.054 |
ridge | -0.052 |
Token | says |
Token ID | 1139 |
Feature activation | +5.77e-03 |
Pos logit contributions | |
iqu | +0.003 |
ridge | +0.003 |
umes | +0.003 |
yles | +0.003 |
rol | +0.003 |
Neg logit contributions | |
day | -0.002 |
good | -0.002 |
, | -0.001 |
time | -0.001 |
team | -0.001 |
Token | " |
Token ID | 366 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
day | +0.009 |
time | +0.008 |
good | +0.008 |
end | +0.007 |
girl | +0.007 |
Neg logit contributions | |
iqu | -0.027 |
ridge | -0.027 |
umes | -0.026 |
yles | -0.026 |
rol | -0.025 |
Token | You |
Token ID | 1639 |
Feature activation | +3.78e-02 |
Pos logit contributions | |
day | +0.029 |
, | +0.029 |
good | +0.027 |
time | +0.027 |
end | +0.027 |
Neg logit contributions | |
ridge | -0.035 |
iqu | -0.035 |
yles | -0.034 |
umes | -0.033 |
rol | -0.033 |
Token | 're |
Token ID | 821 |
Feature activation | +3.76e-02 |
Pos logit contributions | |
day | +0.076 |
girl | +0.070 |
afternoon | +0.069 |
, | +0.069 |
team | +0.068 |
Neg logit contributions | |
iqu | -0.073 |
ridge | -0.073 |
umes | -0.070 |
vell | -0.069 |
yles | -0.068 |
Token | welcome |
Token ID | 7062 |
Feature activation | +6.28e-02 |
Pos logit contributions | |
day | +0.040 |
girl | +0.039 |
team | +0.039 |
boy | +0.038 |
moment | +0.035 |
Neg logit contributions | |
ridge | -0.100 |
iqu | -0.099 |
vell | -0.097 |
umes | -0.097 |
rol | -0.093 |
Token | , |
Token ID | 11 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
day | +0.042 |
afternoon | +0.034 |
team | +0.030 |
end | +0.026 |
boy | +0.024 |
Neg logit contributions | |
iqu | -0.205 |
ridge | -0.199 |
umes | -0.197 |
angs | -0.191 |
ocol | -0.188 |
Token | Sara |
Token ID | 24799 |
Feature activation | +8.88e-02 |
Pos logit contributions | |
day | +0.033 |
afternoon | +0.028 |
team | +0.028 |
, | +0.027 |
good | +0.027 |
Neg logit contributions | |
iqu | -0.058 |
ridge | -0.056 |
yles | -0.055 |
umes | -0.055 |
ocol | -0.053 |
Token | and |
Token ID | 290 |
Feature activation | +4.58e-02 |
Pos logit contributions | |
day | +0.075 |
team | +0.073 |
afternoon | +0.069 |
end | +0.067 |
time | +0.065 |
Neg logit contributions | |
ridge | -0.246 |
ker | -0.234 |
ache | -0.234 |
paste | -0.227 |
umes | -0.227 |
Token | Ben |
Token ID | 3932 |
Feature activation | +7.36e-02 |
Pos logit contributions | |
day | +0.052 |
good | +0.046 |
time | +0.045 |
, | +0.045 |
end | +0.043 |
Neg logit contributions | |
iqu | -0.135 |
ridge | -0.133 |
umes | -0.127 |
yles | -0.127 |
angs | -0.123 |
Token | . |
Token ID | 13 |
Feature activation | +5.25e-02 |
Pos logit contributions | |
team | +0.095 |
day | +0.089 |
afternoon | +0.087 |
town | +0.084 |
end | +0.083 |
Neg logit contributions | |
ridge | -0.164 |
iqu | -0.149 |
pled | -0.148 |
ache | -0.146 |
rou | -0.146 |
Token | You |
Token ID | 921 |
Feature activation | +2.97e-02 |
Pos logit contributions | |
day | +0.100 |
end | +0.097 |
team | +0.097 |
good | +0.096 |
time | +0.096 |
Neg logit contributions | |
ridge | -0.105 |
iqu | -0.104 |
umes | -0.097 |
rol | -0.096 |
yles | -0.095 |
Token | are |
Token ID | 389 |
Feature activation | +3.06e-02 |
Pos logit contributions | |
day | +0.046 |
girl | +0.041 |
afternoon | +0.041 |
, | +0.040 |
good | +0.040 |
Neg logit contributions | |
ridge | -0.072 |
iqu | -0.071 |
umes | -0.068 |
vell | -0.067 |
yles | -0.066 |
Token | see |
Token ID | 766 |
Feature activation | +4.18e-03 |
Pos logit contributions | |
day | +0.006 |
good | +0.005 |
, | +0.005 |
time | +0.005 |
end | +0.005 |
Neg logit contributions | |
iqu | -0.017 |
ridge | -0.016 |
umes | -0.016 |
yles | -0.016 |
rol | -0.016 |
Token | what |
Token ID | 644 |
Feature activation | -3.08e-03 |
Pos logit contributions | |
day | +0.006 |
, | +0.005 |
good | +0.005 |
and | +0.005 |
time | +0.005 |
Neg logit contributions | |
iqu | -0.012 |
ridge | -0.011 |
umes | -0.011 |
yles | -0.011 |
rol | -0.011 |
Token | we |
Token ID | 356 |
Feature activation | -4.15e-03 |
Pos logit contributions | |
yles | +0.007 |
iqu | +0.007 |
ridge | +0.007 |
umes | +0.007 |
rings | +0.006 |
Neg logit contributions | |
end | -0.004 |
day | -0.004 |
girl | -0.004 |
afternoon | -0.004 |
, | -0.004 |
Token | can |
Token ID | 460 |
Feature activation | -2.05e-02 |
Pos logit contributions | |
iqu | +0.012 |
ridge | +0.011 |
umes | +0.011 |
yles | +0.011 |
rol | +0.011 |
Neg logit contributions | |
day | -0.005 |
good | -0.005 |
, | -0.005 |
girl | -0.005 |
time | -0.005 |
Token | come |
Token ID | 1282 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
iqu | +0.061 |
ridge | +0.059 |
umes | +0.057 |
yles | +0.057 |
rol | +0.056 |
Neg logit contributions | |
day | -0.023 |
good | -0.021 |
, | -0.021 |
end | -0.020 |
time | -0.020 |
Token | up |
Token ID | 510 |
Feature activation | +8.87e-02 |
Pos logit contributions | |
day | +0.009 |
good | +0.008 |
girl | +0.008 |
end | +0.008 |
time | +0.008 |
Neg logit contributions | |
iqu | -0.036 |
ridge | -0.035 |
umes | -0.035 |
yles | -0.034 |
angs | -0.034 |
Token | with |
Token ID | 351 |
Feature activation | +3.15e-02 |
Pos logit contributions | |
end | +0.082 |
day | +0.081 |
girl | +0.079 |
good | +0.072 |
boy | +0.070 |
Neg logit contributions | |
iqu | -0.242 |
umes | -0.242 |
yles | -0.241 |
ridge | -0.239 |
vell | -0.228 |
Token | ." |
Token ID | 526 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
day | +0.039 |
girl | +0.036 |
good | +0.035 |
boy | +0.034 |
end | +0.034 |
Neg logit contributions | |
iqu | -0.083 |
ridge | -0.083 |
yles | -0.082 |
umes | -0.081 |
ocol | -0.076 |
Token |
|
Token ID | 198 |
Feature activation | +1.38e-03 |
Pos logit contributions | |
end | +0.012 |
team | +0.011 |
day | +0.011 |
good | +0.010 |
afternoon | +0.009 |
Neg logit contributions | |
ridge | -0.030 |
rol | -0.029 |
iqu | -0.029 |
yles | -0.028 |
umes | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | -7.90e-03 |
Pos logit contributions | |
team | +0.001 |
end | +0.001 |
day | +0.001 |
girl | +0.001 |
good | +0.001 |
Neg logit contributions | |
iqu | -0.004 |
ridge | -0.004 |
rol | -0.004 |
yles | -0.004 |
umes | -0.003 |
Token | They |
Token ID | 2990 |
Feature activation | +3.91e-02 |
Pos logit contributions | |
iqu | +0.014 |
umes | +0.014 |
rol | +0.014 |
ridge | +0.014 |
yles | +0.013 |
Neg logit contributions | |
day | -0.015 |
team | -0.014 |
end | -0.013 |
good | -0.013 |
, | -0.013 |
Token | " |
Token ID | 366 |
Feature activation | +8.77e-03 |
Pos logit contributions | |
day | +0.006 |
, | +0.005 |
good | +0.005 |
and | +0.004 |
girl | +0.004 |
Neg logit contributions | |
iqu | -0.028 |
ridge | -0.027 |
umes | -0.026 |
yles | -0.026 |
angs | -0.026 |
Token | You |
Token ID | 1639 |
Feature activation | +3.09e-02 |
Pos logit contributions | |
, | +0.016 |
day | +0.016 |
time | +0.015 |
end | +0.015 |
good | +0.015 |
Neg logit contributions | |
iqu | -0.019 |
ridge | -0.019 |
yles | -0.018 |
umes | -0.018 |
rol | -0.018 |
Token | 're |
Token ID | 821 |
Feature activation | +2.99e-02 |
Pos logit contributions | |
day | +0.061 |
afternoon | +0.056 |
time | +0.055 |
girl | +0.054 |
end | +0.054 |
Neg logit contributions | |
iqu | -0.063 |
ridge | -0.062 |
umes | -0.059 |
vell | -0.059 |
yles | -0.057 |
Token | welcome |
Token ID | 7062 |
Feature activation | +6.03e-02 |
Pos logit contributions | |
day | +0.032 |
team | +0.029 |
girl | +0.029 |
boy | +0.027 |
afternoon | +0.027 |
Neg logit contributions | |
iqu | -0.083 |
ridge | -0.082 |
vell | -0.080 |
umes | -0.080 |
rol | -0.078 |
Token | , |
Token ID | 11 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
day | +0.036 |
afternoon | +0.028 |
end | +0.025 |
team | +0.024 |
time | +0.019 |
Neg logit contributions | |
iqu | -0.206 |
ridge | -0.200 |
umes | -0.196 |
angs | -0.190 |
rol | -0.189 |
Token | Lily |
Token ID | 20037 |
Feature activation | +8.86e-02 |
Pos logit contributions | |
day | +0.021 |
, | +0.017 |
end | +0.017 |
time | +0.017 |
good | +0.017 |
Neg logit contributions | |
iqu | -0.062 |
ridge | -0.060 |
umes | -0.059 |
yles | -0.059 |
rol | -0.057 |
Token | . |
Token ID | 13 |
Feature activation | +4.97e-02 |
Pos logit contributions | |
team | +0.083 |
day | +0.081 |
afternoon | +0.073 |
town | +0.072 |
time | +0.072 |
Neg logit contributions | |
ridge | -0.238 |
ker | -0.231 |
iqu | -0.229 |
ache | -0.225 |
ous | -0.224 |
Token | We |
Token ID | 775 |
Feature activation | +2.88e-02 |
Pos logit contributions | |
day | +0.091 |
end | +0.089 |
time | +0.089 |
good | +0.087 |
team | +0.086 |
Neg logit contributions | |
iqu | -0.105 |
ridge | -0.104 |
rol | -0.097 |
umes | -0.096 |
yles | -0.096 |
Token | 're |
Token ID | 821 |
Feature activation | +2.74e-02 |
Pos logit contributions | |
day | +0.039 |
girl | +0.036 |
team | +0.036 |
new | +0.036 |
boy | +0.036 |
Neg logit contributions | |
iqu | -0.070 |
ridge | -0.069 |
umes | -0.067 |
rol | -0.066 |
yles | -0.064 |
Token | glad |
Token ID | 9675 |
Feature activation | +4.24e-02 |
Pos logit contributions | |
day | +0.037 |
end | +0.033 |
team | +0.033 |
time | +0.033 |
girl | +0.033 |
Neg logit contributions | |
iqu | -0.073 |
ridge | -0.071 |
umes | -0.069 |
yles | -0.069 |
rol | -0.068 |
Token | you |
Token ID | 345 |
Feature activation | +2.23e-02 |
Pos logit contributions | |
day | +0.077 |
end | +0.073 |
good | +0.071 |
afternoon | +0.071 |
time | +0.071 |
Neg logit contributions | |
iqu | -0.091 |
ridge | -0.088 |
yles | -0.084 |
umes | -0.082 |
rings | -0.080 |
Token | . |
Token ID | 13 |
Feature activation | +4.56e-02 |
Pos logit contributions | |
day | +0.007 |
time | +0.006 |
afternoon | +0.006 |
end | +0.006 |
team | +0.006 |
Neg logit contributions | |
iqu | -0.029 |
ridge | -0.028 |
umes | -0.027 |
angs | -0.027 |
rings | -0.027 |
Token | " |
Token ID | 366 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
day | +0.064 |
afternoon | +0.060 |
end | +0.058 |
time | +0.056 |
team | +0.054 |
Neg logit contributions | |
iqu | -0.110 |
ridge | -0.109 |
yles | -0.104 |
paste | -0.104 |
umes | -0.104 |
Token | Thank |
Token ID | 10449 |
Feature activation | +1.87e-02 |
Pos logit contributions | |
day | +0.028 |
, | +0.027 |
good | +0.027 |
time | +0.026 |
end | +0.026 |
Neg logit contributions | |
iqu | -0.040 |
ridge | -0.039 |
yles | -0.037 |
umes | -0.037 |
rol | -0.036 |
Token | you |
Token ID | 345 |
Feature activation | +6.89e-02 |
Pos logit contributions | |
day | +0.036 |
time | +0.034 |
good | +0.034 |
team | +0.033 |
, | +0.033 |
Neg logit contributions | |
ridge | -0.037 |
iqu | -0.037 |
yles | -0.034 |
ocol | -0.034 |
umes | -0.034 |
Token | , |
Token ID | 11 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
day | +0.078 |
afternoon | +0.074 |
time | +0.073 |
moment | +0.068 |
team | +0.066 |
Neg logit contributions | |
ridge | -0.180 |
ous | -0.168 |
angs | -0.165 |
rol | -0.165 |
iqu | -0.165 |
Token | Anna |
Token ID | 11735 |
Feature activation | +8.85e-02 |
Pos logit contributions | |
day | +0.028 |
good | +0.023 |
, | +0.023 |
team | +0.023 |
afternoon | +0.023 |
Neg logit contributions | |
iqu | -0.063 |
ridge | -0.063 |
yles | -0.059 |
umes | -0.059 |
ocol | -0.058 |
Token | . |
Token ID | 13 |
Feature activation | +4.94e-02 |
Pos logit contributions | |
day | +0.116 |
team | +0.116 |
afternoon | +0.113 |
time | +0.111 |
moment | +0.104 |
Neg logit contributions | |
ridge | -0.214 |
cher | -0.195 |
ache | -0.190 |
paste | -0.190 |
ents | -0.184 |
Token | You |
Token ID | 921 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
day | +0.090 |
good | +0.085 |
, | +0.084 |
time | +0.082 |
end | +0.082 |
Neg logit contributions | |
iqu | -0.111 |
ridge | -0.108 |
umes | -0.102 |
yles | -0.101 |
rol | -0.101 |
Token | are |
Token ID | 389 |
Feature activation | +2.70e-02 |
Pos logit contributions | |
day | +0.035 |
, | +0.031 |
good | +0.031 |
girl | +0.031 |
time | +0.030 |
Neg logit contributions | |
iqu | -0.068 |
ridge | -0.066 |
umes | -0.063 |
yles | -0.062 |
angs | -0.062 |
Token | a |
Token ID | 257 |
Feature activation | +5.53e-02 |
Pos logit contributions | |
day | +0.028 |
team | +0.024 |
girl | +0.024 |
time | +0.023 |
, | +0.023 |
Neg logit contributions | |
iqu | -0.079 |
ridge | -0.079 |
umes | -0.075 |
vell | -0.074 |
yles | -0.074 |
Token | good |
Token ID | 922 |
Feature activation | +4.86e-02 |
Pos logit contributions | |
day | +0.037 |
They | +0.035 |
afternoon | +0.034 |
time | +0.030 |
Wednesday | +0.027 |
Neg logit contributions | |
ridge | -0.161 |
vell | -0.161 |
iqu | -0.159 |
ache | -0.156 |
cher | -0.152 |
Token | , |
Token ID | 11 |
Feature activation | +7.92e-03 |
Pos logit contributions | |
iqu | +0.016 |
ridge | +0.016 |
umes | +0.016 |
paste | +0.015 |
ous | +0.015 |
Neg logit contributions | |
day | -0.003 |
afternoon | -0.002 |
end | -0.002 |
team | -0.002 |
town | -0.002 |
Token | " |
Token ID | 366 |
Feature activation | +5.44e-03 |
Pos logit contributions | |
day | +0.007 |
good | +0.006 |
, | +0.006 |
time | +0.005 |
end | +0.005 |
Neg logit contributions | |
iqu | -0.026 |
ridge | -0.025 |
umes | -0.024 |
yles | -0.024 |
rol | -0.024 |
Token | Thank |
Token ID | 10449 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
, | +0.008 |
day | +0.008 |
good | +0.008 |
end | +0.008 |
team | +0.008 |
Neg logit contributions | |
iqu | -0.013 |
ridge | -0.013 |
yles | -0.013 |
umes | -0.012 |
rol | -0.012 |
Token | you |
Token ID | 345 |
Feature activation | +6.67e-02 |
Pos logit contributions | |
day | +0.033 |
boy | +0.031 |
team | +0.031 |
good | +0.031 |
girl | +0.030 |
Neg logit contributions | |
iqu | -0.033 |
ridge | -0.032 |
umes | -0.031 |
yles | -0.030 |
ocol | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
day | +0.053 |
afternoon | +0.046 |
time | +0.044 |
town | +0.043 |
boy | +0.041 |
Neg logit contributions | |
ridge | -0.198 |
iqu | -0.197 |
umes | -0.196 |
ous | -0.193 |
ker | -0.187 |
Token | Lily |
Token ID | 20037 |
Feature activation | +8.84e-02 |
Pos logit contributions | |
day | +0.018 |
good | +0.016 |
, | +0.016 |
time | +0.015 |
end | +0.015 |
Neg logit contributions | |
iqu | -0.063 |
ridge | -0.061 |
umes | -0.059 |
yles | -0.058 |
rol | -0.058 |
Token | . |
Token ID | 13 |
Feature activation | +3.88e-02 |
Pos logit contributions | |
team | +0.093 |
day | +0.089 |
town | +0.085 |
time | +0.085 |
afternoon | +0.082 |
Neg logit contributions | |
ridge | -0.232 |
iqu | -0.224 |
rou | -0.217 |
ous | -0.215 |
ker | -0.213 |
Token | You |
Token ID | 921 |
Feature activation | +2.64e-02 |
Pos logit contributions | |
day | +0.062 |
good | +0.059 |
, | +0.058 |
time | +0.057 |
end | +0.057 |
Neg logit contributions | |
iqu | -0.096 |
ridge | -0.093 |
umes | -0.089 |
yles | -0.088 |
rol | -0.087 |
Token | are |
Token ID | 389 |
Feature activation | +2.88e-02 |
Pos logit contributions | |
day | +0.043 |
girl | +0.039 |
, | +0.039 |
good | +0.039 |
team | +0.038 |
Neg logit contributions | |
iqu | -0.063 |
ridge | -0.062 |
umes | -0.059 |
vell | -0.058 |
yles | -0.058 |
Token | such |
Token ID | 884 |
Feature activation | +4.39e-02 |
Pos logit contributions | |
day | +0.035 |
team | +0.032 |
girl | +0.031 |
boy | +0.030 |
time | +0.030 |
Neg logit contributions | |
iqu | -0.077 |
ridge | -0.077 |
umes | -0.075 |
vell | -0.073 |
yles | -0.073 |
Token | a |
Token ID | 257 |
Feature activation | +3.96e-02 |
Pos logit contributions | |
day | +0.034 |
, | +0.030 |
end | +0.028 |
girl | +0.028 |
team | +0.027 |
Neg logit contributions | |
iqu | -0.143 |
ridge | -0.137 |
umes | -0.135 |
yles | -0.132 |
vell | -0.132 |
Token | , |
Token ID | 11 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
day | +0.004 |
afternoon | +0.003 |
place | +0.003 |
end | +0.003 |
girl | +0.003 |
Neg logit contributions | |
ridge | -0.015 |
iqu | -0.015 |
paste | -0.015 |
umes | -0.015 |
rings | -0.014 |
Token | " |
Token ID | 366 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
day | +0.011 |
good | +0.009 |
time | +0.009 |
end | +0.009 |
girl | +0.009 |
Neg logit contributions | |
iqu | -0.033 |
ridge | -0.033 |
umes | -0.032 |
yles | -0.031 |
rol | -0.030 |
Token | Thank |
Token ID | 10449 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
, | +0.031 |
day | +0.029 |
time | +0.029 |
end | +0.028 |
good | +0.028 |
Neg logit contributions | |
ridge | -0.034 |
yles | -0.034 |
iqu | -0.034 |
umes | -0.032 |
cher | -0.031 |
Token | you |
Token ID | 345 |
Feature activation | +7.09e-02 |
Pos logit contributions | |
day | +0.050 |
place | +0.047 |
afternoon | +0.046 |
boy | +0.046 |
team | +0.046 |
Neg logit contributions | |
iqu | -0.039 |
ridge | -0.039 |
ocol | -0.038 |
umes | -0.037 |
yles | -0.036 |
Token | , |
Token ID | 11 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
day | +0.060 |
afternoon | +0.059 |
moment | +0.046 |
town | +0.046 |
time | +0.045 |
Neg logit contributions | |
umes | -0.195 |
ridge | -0.194 |
iqu | -0.192 |
ker | -0.191 |
ous | -0.190 |
Token | Lily |
Token ID | 20037 |
Feature activation | +8.80e-02 |
Pos logit contributions | |
day | +0.021 |
good | +0.018 |
time | +0.017 |
, | +0.017 |
afternoon | +0.017 |
Neg logit contributions | |
iqu | -0.057 |
ridge | -0.055 |
umes | -0.054 |
yles | -0.054 |
rol | -0.052 |
Token | , |
Token ID | 11 |
Feature activation | +2.15e-02 |
Pos logit contributions | |
day | +0.098 |
afternoon | +0.098 |
team | +0.096 |
town | +0.090 |
sunny | +0.085 |
Neg logit contributions | |
ker | -0.207 |
ridge | -0.202 |
pled | -0.198 |
ous | -0.195 |
ache | -0.195 |
Token | thank |
Token ID | 5875 |
Feature activation | -2.73e-04 |
Pos logit contributions | |
day | +0.028 |
girl | +0.025 |
time | +0.025 |
, | +0.025 |
good | +0.025 |
Neg logit contributions | |
iqu | -0.056 |
ridge | -0.055 |
umes | -0.054 |
yles | -0.053 |
ocol | -0.051 |
Token | you |
Token ID | 345 |
Feature activation | +5.71e-02 |
Pos logit contributions | |
ridge | +0.000 |
iqu | +0.000 |
umes | +0.000 |
yles | +0.000 |
ocol | +0.000 |
Neg logit contributions | |
day | -0.001 |
good | -0.001 |
boy | -0.001 |
girl | -0.001 |
afternoon | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
day | +0.054 |
afternoon | +0.053 |
time | +0.046 |
town | +0.045 |
girl | +0.045 |
Neg logit contributions | |
umes | -0.149 |
iqu | -0.148 |
ridge | -0.147 |
ous | -0.146 |
cher | -0.145 |
Token | Ben |
Token ID | 3932 |
Feature activation | +6.99e-02 |
Pos logit contributions | |
day | +0.025 |
afternoon | +0.021 |
time | +0.021 |
, | +0.021 |
good | +0.021 |
Neg logit contributions | |
iqu | -0.059 |
ridge | -0.057 |
umes | -0.057 |
yles | -0.056 |
rol | -0.054 |
Token | smiled |
Token ID | 13541 |
Feature activation | +5.05e-02 |
Pos logit contributions | |
day | +0.037 |
time | +0.036 |
end | +0.035 |
good | +0.034 |
team | +0.032 |
Neg logit contributions | |
iqu | -0.072 |
ridge | -0.070 |
yles | -0.067 |
umes | -0.067 |
rol | -0.066 |
Token | . |
Token ID | 13 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
day | +0.050 |
time | +0.045 |
afternoon | +0.042 |
end | +0.041 |
team | +0.037 |
Neg logit contributions | |
iqu | -0.148 |
ous | -0.139 |
vell | -0.137 |
umes | -0.137 |
ridge | -0.136 |
Token | They |
Token ID | 1119 |
Feature activation | +4.51e-02 |
Pos logit contributions | |
day | +0.020 |
end | +0.019 |
girl | +0.016 |
team | +0.016 |
time | +0.016 |
Neg logit contributions | |
umes | -0.061 |
yles | -0.061 |
iqu | -0.061 |
rol | -0.060 |
ridge | -0.060 |
Token | had |
Token ID | 550 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
day | +0.062 |
good | +0.057 |
time | +0.056 |
girl | +0.055 |
, | +0.054 |
Neg logit contributions | |
iqu | -0.120 |
ridge | -0.116 |
umes | -0.111 |
rol | -0.110 |
yles | -0.109 |
Token | come |
Token ID | 1282 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
day | +0.015 |
girl | +0.014 |
end | +0.014 |
team | +0.013 |
, | +0.013 |
Neg logit contributions | |
ridge | -0.030 |
iqu | -0.030 |
umes | -0.028 |
rol | -0.028 |
yles | -0.028 |
Token | up |
Token ID | 510 |
Feature activation | +8.77e-02 |
Pos logit contributions | |
day | +0.011 |
good | +0.009 |
time | +0.009 |
girl | +0.009 |
end | +0.009 |
Neg logit contributions | |
iqu | -0.035 |
ridge | -0.034 |
umes | -0.034 |
yles | -0.033 |
rol | -0.033 |
Token | with |
Token ID | 351 |
Feature activation | +2.73e-02 |
Pos logit contributions | |
day | +0.089 |
girl | +0.088 |
end | +0.085 |
boy | +0.082 |
good | +0.080 |
Neg logit contributions | |
iqu | -0.244 |
ridge | -0.233 |
yles | -0.229 |
umes | -0.227 |
rol | -0.226 |
Token | a |
Token ID | 257 |
Feature activation | +2.67e-02 |
Pos logit contributions | |
day | +0.023 |
girl | +0.020 |
team | +0.019 |
boy | +0.018 |
end | +0.018 |
Neg logit contributions | |
iqu | -0.086 |
ridge | -0.083 |
yles | -0.081 |
umes | -0.081 |
rol | -0.078 |
Token | creative |
Token ID | 7325 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
day | +0.026 |
They | +0.024 |
end | +0.023 |
, | +0.023 |
their | +0.022 |
Neg logit contributions | |
iqu | -0.077 |
ridge | -0.073 |
umes | -0.073 |
vell | -0.070 |
bar | -0.069 |
Token | way |
Token ID | 835 |
Feature activation | +3.71e-02 |
Pos logit contributions | |
They | +0.028 |
| +0.028 |
their | +0.027 |
day | +0.025 |
many | +0.025 |
Neg logit contributions | |
iqu | -0.072 |
bar | -0.069 |
ker | -0.068 |
ridge | -0.067 |
ocol | -0.066 |
Token | to |
Token ID | 284 |
Feature activation | +5.60e-04 |
Pos logit contributions | |
day | +0.040 |
good | +0.038 |
girl | +0.036 |
end | +0.035 |
team | +0.035 |
Neg logit contributions | |
iqu | -0.110 |
ridge | -0.105 |
umes | -0.103 |
rol | -0.102 |
yles | -0.101 |
Token | said |
Token ID | 531 |
Feature activation | -9.56e-04 |
Pos logit contributions | |
day | +0.057 |
end | +0.052 |
time | +0.051 |
afternoon | +0.050 |
good | +0.050 |
Neg logit contributions | |
iqu | -0.075 |
ridge | -0.072 |
umes | -0.069 |
paste | -0.067 |
cher | -0.066 |
Token | , |
Token ID | 11 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
ridge | +0.003 |
iqu | +0.003 |
paste | +0.003 |
bar | +0.003 |
cher | +0.003 |
Neg logit contributions | |
day | -0.001 |
afternoon | -0.001 |
time | -0.000 |
girl | -0.000 |
place | -0.000 |
Token | " |
Token ID | 366 |
Feature activation | +1.43e-02 |
Pos logit contributions | |
day | +0.016 |
time | +0.014 |
good | +0.014 |
end | +0.014 |
girl | +0.013 |
Neg logit contributions | |
ridge | -0.047 |
iqu | -0.046 |
umes | -0.045 |
yles | -0.044 |
bar | -0.043 |
Token | OK |
Token ID | 11380 |
Feature activation | +6.08e-02 |
Pos logit contributions | |
, | +0.024 |
day | +0.023 |
time | +0.023 |
end | +0.022 |
good | +0.021 |
Neg logit contributions | |
ridge | -0.032 |
iqu | -0.031 |
yles | -0.030 |
cher | -0.030 |
bar | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
afternoon | +0.059 |
day | +0.057 |
town | +0.048 |
country | +0.045 |
their | +0.043 |
Neg logit contributions | |
iqu | -0.149 |
ridge | -0.148 |
bar | -0.144 |
ker | -0.143 |
umes | -0.142 |
Token | Lily |
Token ID | 20037 |
Feature activation | +8.76e-02 |
Pos logit contributions | |
day | +0.016 |
good | +0.014 |
, | +0.014 |
time | +0.014 |
end | +0.013 |
Neg logit contributions | |
iqu | -0.045 |
ridge | -0.044 |
umes | -0.042 |
yles | -0.042 |
rol | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | +5.39e-02 |
Pos logit contributions | |
afternoon | +0.093 |
day | +0.088 |
town | +0.083 |
their | +0.080 |
team | +0.077 |
Neg logit contributions | |
iqu | -0.207 |
ker | -0.207 |
ridge | -0.206 |
ache | -0.202 |
ous | -0.202 |
Token | I |
Token ID | 314 |
Feature activation | +3.20e-02 |
Pos logit contributions | |
day | +0.107 |
time | +0.105 |
and | +0.100 |
good | +0.100 |
end | +0.100 |
Neg logit contributions | |
ridge | -0.103 |
iqu | -0.100 |
bar | -0.094 |
cher | -0.093 |
rings | -0.091 |
Token | will |
Token ID | 481 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
day | +0.048 |
girl | +0.044 |
afternoon | +0.044 |
their | +0.043 |
boy | +0.041 |
Neg logit contributions | |
iqu | -0.076 |
ridge | -0.073 |
umes | -0.071 |
angs | -0.068 |
cher | -0.068 |
Token | do |
Token ID | 466 |
Feature activation | +4.78e-02 |
Pos logit contributions | |
day | +0.026 |
good | +0.022 |
girl | +0.022 |
, | +0.021 |
boy | +0.021 |
Neg logit contributions | |
iqu | -0.070 |
ridge | -0.067 |
umes | -0.065 |
yles | -0.064 |
angs | -0.063 |
Token | that |
Token ID | 326 |
Feature activation | +2.50e-02 |
Pos logit contributions | |
day | +0.067 |
girl | +0.061 |
team | +0.059 |
boy | +0.058 |
time | +0.058 |
Neg logit contributions | |
iqu | -0.129 |
ridge | -0.125 |
yles | -0.120 |
umes | -0.119 |
angs | -0.115 |
Token | delicious |
Token ID | 12625 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
iqu | +0.047 |
ridge | +0.045 |
umes | +0.043 |
yles | +0.043 |
rol | +0.043 |
Neg logit contributions | |
day | -0.019 |
, | -0.017 |
good | -0.017 |
end | -0.016 |
team | -0.016 |
Token | vegetables |
Token ID | 13701 |
Feature activation | -4.24e-03 |
Pos logit contributions | |
iqu | +0.032 |
ridge | +0.031 |
umes | +0.029 |
yles | +0.029 |
ocol | +0.029 |
Neg logit contributions | |
day | -0.016 |
end | -0.015 |
good | -0.015 |
team | -0.015 |
girl | -0.014 |
Token | and |
Token ID | 290 |
Feature activation | -3.23e-02 |
Pos logit contributions | |
iqu | +0.016 |
ridge | +0.016 |
yles | +0.015 |
umes | +0.015 |
rol | +0.015 |
Neg logit contributions | |
day | -0.002 |
, | -0.001 |
good | -0.001 |
and | -0.001 |
time | -0.001 |
Token | bright |
Token ID | 6016 |
Feature activation | +2.54e-03 |
Pos logit contributions | |
iqu | +0.087 |
ridge | +0.084 |
umes | +0.080 |
yles | +0.080 |
rol | +0.079 |
Neg logit contributions | |
day | -0.046 |
good | -0.043 |
, | -0.043 |
time | -0.041 |
end | -0.041 |
Token | - |
Token ID | 12 |
Feature activation | -6.67e-02 |
Pos logit contributions | |
day | +0.003 |
good | +0.003 |
end | +0.003 |
team | +0.003 |
girl | +0.003 |
Neg logit contributions | |
iqu | -0.007 |
ridge | -0.007 |
yles | -0.006 |
umes | -0.006 |
ocol | -0.006 |
Token | col |
Token ID | 4033 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
iqu | +0.131 |
ridge | +0.119 |
yles | +0.113 |
umes | +0.113 |
angs | +0.108 |
Neg logit contributions | |
day | -0.153 |
good | -0.143 |
, | -0.141 |
and | -0.139 |
girl | -0.137 |
Token | oured |
Token ID | 8167 |
Feature activation | -7.85e-03 |
Pos logit contributions | |
cher | +0.182 |
iqu | +0.155 |
outage | +0.155 |
angs | +0.154 |
paste | +0.152 |
Neg logit contributions | |
, | -0.297 |
and | -0.285 |
happily | -0.269 |
loves | -0.263 |
They | -0.261 |
Token | flowers |
Token ID | 12734 |
Feature activation | -6.26e-03 |
Pos logit contributions | |
iqu | +0.019 |
ridge | +0.019 |
umes | +0.018 |
yles | +0.018 |
ocol | +0.018 |
Neg logit contributions | |
day | -0.012 |
end | -0.011 |
good | -0.011 |
time | -0.011 |
, | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
iqu | +0.024 |
ridge | +0.023 |
yles | +0.023 |
umes | +0.023 |
rol | +0.023 |
Neg logit contributions | |
day | -0.002 |
, | -0.002 |
and | -0.001 |
good | -0.001 |
time | -0.001 |
Token | |
Token ID | 220 |
Feature activation | -1.59e-02 |
Pos logit contributions | |
ridge | +0.035 |
iqu | +0.035 |
bar | +0.034 |
yles | +0.034 |
rol | +0.033 |
Neg logit contributions | |
good | -0.014 |
day | -0.013 |
end | -0.012 |
girl | -0.012 |
, | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
iqu | +0.057 |
ridge | +0.055 |
yles | +0.054 |
umes | +0.054 |
rol | +0.054 |
Neg logit contributions | |
day | -0.007 |
good | -0.006 |
, | -0.005 |
girl | -0.005 |
end | -0.005 |
Token | tried |
Token ID | 3088 |
Feature activation | -8.46e-03 |
Pos logit contributions | |
day | +0.012 |
good | +0.011 |
, | +0.010 |
end | +0.010 |
time | +0.010 |
Neg logit contributions | |
iqu | -0.023 |
ridge | -0.022 |
umes | -0.021 |
yles | -0.021 |
rol | -0.021 |
Token | to |
Token ID | 284 |
Feature activation | -3.62e-02 |
Pos logit contributions | |
iqu | +0.026 |
ridge | +0.025 |
umes | +0.024 |
yles | +0.024 |
rol | +0.024 |
Neg logit contributions | |
day | -0.009 |
good | -0.008 |
, | -0.007 |
girl | -0.007 |
end | -0.007 |
Token | jump |
Token ID | 4391 |
Feature activation | -1.85e-03 |
Pos logit contributions | |
iqu | +0.105 |
ridge | +0.101 |
umes | +0.097 |
yles | +0.097 |
rol | +0.095 |
Neg logit contributions | |
day | -0.049 |
, | -0.047 |
good | -0.046 |
and | -0.045 |
end | -0.045 |
Token | on |
Token ID | 319 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
iqu | +0.006 |
ridge | +0.005 |
rol | +0.005 |
umes | +0.005 |
bar | +0.005 |
Neg logit contributions | |
day | -0.002 |
girl | -0.002 |
team | -0.002 |
good | -0.002 |
time | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | -2.80e-02 |
Pos logit contributions | |
iqu | +0.036 |
ridge | +0.035 |
umes | +0.033 |
yles | +0.033 |
rol | +0.033 |
Neg logit contributions | |
day | -0.008 |
good | -0.007 |
girl | -0.007 |
, | -0.007 |
end | -0.006 |
Token | ostr |
Token ID | 42387 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
iqu | +0.075 |
ridge | +0.072 |
umes | +0.070 |
angs | +0.068 |
bar | +0.068 |
Neg logit contributions | |
day | -0.039 |
, | -0.036 |
good | -0.036 |
time | -0.035 |
girl | -0.034 |
Token | ich |
Token ID | 488 |
Feature activation | -4.20e-02 |
Pos logit contributions | |
iqu | +0.200 |
ridge | +0.189 |
yles | +0.185 |
umes | +0.182 |
rol | +0.181 |
Neg logit contributions | |
, | -0.300 |
team | -0.295 |
and | -0.292 |
day | -0.288 |
end | -0.282 |
Token | 's |
Token ID | 338 |
Feature activation | -3.24e-02 |
Pos logit contributions | |
iqu | +0.158 |
ridge | +0.154 |
umes | +0.149 |
yles | +0.149 |
rol | +0.147 |
Neg logit contributions | |
day | -0.016 |
, | -0.012 |
good | -0.011 |
and | -0.009 |
end | -0.009 |
Token | back |
Token ID | 736 |
Feature activation | -4.29e-03 |
Pos logit contributions | |
iqu | +0.085 |
ridge | +0.084 |
umes | +0.079 |
bar | +0.079 |
angs | +0.078 |
Neg logit contributions | |
day | -0.046 |
good | -0.042 |
time | -0.041 |
girl | -0.041 |
, | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | -3.88e-03 |
Pos logit contributions | |
iqu | +0.016 |
ridge | +0.015 |
umes | +0.015 |
yles | +0.015 |
rol | +0.015 |
Neg logit contributions | |
day | -0.002 |
good | -0.001 |
, | -0.001 |
time | -0.001 |
girl | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -1.14e-02 |
Pos logit contributions | |
iqu | +0.013 |
ridge | +0.013 |
umes | +0.012 |
yles | +0.012 |
rol | +0.012 |
Neg logit contributions | |
day | -0.003 |
good | -0.002 |
end | -0.002 |
, | -0.002 |
girl | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -2.28e-02 |
Pos logit contributions | |
day | +0.005 |
team | +0.004 |
end | +0.004 |
good | +0.004 |
girl | +0.004 |
Neg logit contributions | |
iqu | -0.024 |
ridge | -0.024 |
yles | -0.023 |
umes | -0.023 |
rings | -0.022 |
Token | they |
Token ID | 484 |
Feature activation | +1.73e-03 |
Pos logit contributions | |
iqu | +0.077 |
ridge | +0.074 |
umes | +0.072 |
yles | +0.071 |
angs | +0.071 |
Neg logit contributions | |
day | -0.019 |
, | -0.018 |
good | -0.017 |
girl | -0.016 |
and | -0.015 |
Token | realized |
Token ID | 6939 |
Feature activation | -3.50e-02 |
Pos logit contributions | |
day | +0.002 |
good | +0.002 |
, | +0.002 |
girl | +0.002 |
time | +0.002 |
Neg logit contributions | |
iqu | -0.005 |
ridge | -0.005 |
umes | -0.004 |
yles | -0.004 |
rol | -0.004 |
Token | that |
Token ID | 326 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
iqu | +0.109 |
ridge | +0.107 |
umes | +0.103 |
rol | +0.102 |
yles | +0.102 |
Neg logit contributions | |
, | -0.037 |
day | -0.036 |
and | -0.034 |
good | -0.034 |
their | -0.034 |
Token | the |
Token ID | 262 |
Feature activation | -2.35e-02 |
Pos logit contributions | |
iqu | +0.089 |
ridge | +0.086 |
umes | +0.083 |
yles | +0.082 |
angs | +0.082 |
Neg logit contributions | |
day | -0.023 |
good | -0.020 |
, | -0.020 |
time | -0.018 |
and | -0.018 |
Token | ostr |
Token ID | 42387 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
iqu | +0.061 |
ridge | +0.059 |
umes | +0.056 |
yles | +0.056 |
rol | +0.055 |
Neg logit contributions | |
day | -0.036 |
good | -0.034 |
, | -0.034 |
time | -0.032 |
girl | -0.032 |
Token | ich |
Token ID | 488 |
Feature activation | -5.34e-02 |
Pos logit contributions | |
iqu | +0.208 |
rol | +0.200 |
paste | +0.200 |
yles | +0.198 |
ridge | +0.196 |
Neg logit contributions | |
, | -0.265 |
and | -0.257 |
friends | -0.255 |
day | -0.254 |
team | -0.245 |
Token | was |
Token ID | 373 |
Feature activation | -2.89e-02 |
Pos logit contributions | |
iqu | +0.169 |
ridge | +0.164 |
umes | +0.158 |
yles | +0.157 |
angs | +0.155 |
Neg logit contributions | |
day | -0.049 |
good | -0.045 |
, | -0.042 |
end | -0.041 |
girl | -0.041 |
Token | very |
Token ID | 845 |
Feature activation | -8.71e-04 |
Pos logit contributions | |
iqu | +0.077 |
ridge | +0.074 |
umes | +0.071 |
yles | +0.071 |
rol | +0.070 |
Neg logit contributions | |
day | -0.044 |
good | -0.042 |
, | -0.041 |
time | -0.040 |
end | -0.039 |
Token | messy |
Token ID | 24097 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
iqu | +0.002 |
ridge | +0.002 |
umes | +0.002 |
yles | +0.002 |
rol | +0.002 |
Neg logit contributions | |
day | -0.001 |
good | -0.001 |
, | -0.001 |
time | -0.001 |
girl | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -6.96e-03 |
Pos logit contributions | |
day | +0.006 |
good | +0.005 |
, | +0.004 |
end | +0.004 |
girl | +0.004 |
Neg logit contributions | |
iqu | -0.039 |
ridge | -0.038 |
umes | -0.037 |
yles | -0.036 |
rol | -0.036 |
Token | saved |
Token ID | 7448 |
Feature activation | -6.82e-03 |
Pos logit contributions | |
day | +0.041 |
girl | +0.037 |
afternoon | +0.037 |
their | +0.037 |
boy | +0.035 |
Neg logit contributions | |
iqu | -0.059 |
ridge | -0.058 |
vell | -0.057 |
cher | -0.055 |
umes | -0.055 |
Token | my |
Token ID | 616 |
Feature activation | -6.00e-04 |
Pos logit contributions | |
iqu | +0.020 |
ridge | +0.020 |
umes | +0.019 |
yles | +0.019 |
rol | +0.018 |
Neg logit contributions | |
day | -0.008 |
good | -0.007 |
, | -0.007 |
time | -0.007 |
end | -0.007 |
Token | life |
Token ID | 1204 |
Feature activation | +4.89e-03 |
Pos logit contributions | |
iqu | +0.001 |
ridge | +0.001 |
umes | +0.001 |
yles | +0.001 |
angs | +0.001 |
Neg logit contributions | |
day | -0.001 |
, | -0.001 |
good | -0.001 |
team | -0.001 |
They | -0.001 |
Token | ," |
Token ID | 553 |
Feature activation | +5.74e-03 |
Pos logit contributions | |
day | +0.006 |
good | +0.006 |
end | +0.005 |
time | +0.005 |
girl | +0.005 |
Neg logit contributions | |
iqu | -0.013 |
ridge | -0.013 |
umes | -0.013 |
angs | -0.012 |
rol | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | -2.80e-02 |
Pos logit contributions | |
day | +0.007 |
time | +0.006 |
afternoon | +0.006 |
good | +0.006 |
moment | +0.005 |
Neg logit contributions | |
iqu | -0.015 |
ridge | -0.015 |
yles | -0.015 |
rings | -0.015 |
ocol | -0.014 |
Token | ostr |
Token ID | 42387 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
iqu | +0.059 |
ridge | +0.056 |
umes | +0.053 |
yles | +0.053 |
rol | +0.052 |
Neg logit contributions | |
day | -0.058 |
good | -0.055 |
, | -0.055 |
girl | -0.053 |
end | -0.053 |
Token | ich |
Token ID | 488 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
iqu | +0.210 |
ridge | +0.204 |
paste | +0.198 |
yles | +0.198 |
umes | +0.197 |
Neg logit contributions | |
, | -0.270 |
and | -0.261 |
friends | -0.258 |
team | -0.249 |
friend | -0.248 |
Token | says |
Token ID | 1139 |
Feature activation | -4.95e-03 |
Pos logit contributions | |
iqu | +0.051 |
ridge | +0.050 |
angs | +0.046 |
yles | +0.046 |
umes | +0.045 |
Neg logit contributions | |
day | -0.036 |
good | -0.033 |
time | -0.032 |
afternoon | -0.032 |
team | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
iqu | +0.016 |
ridge | +0.016 |
umes | +0.016 |
angs | +0.015 |
yles | +0.015 |
Neg logit contributions | |
day | -0.004 |
afternoon | -0.003 |
time | -0.003 |
end | -0.003 |
good | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
day | +0.025 |
end | +0.021 |
afternoon | +0.021 |
team | +0.019 |
time | +0.019 |
Neg logit contributions | |
ridge | -0.070 |
iqu | -0.070 |
yles | -0.068 |
umes | -0.067 |
angs | -0.067 |
Token |
|
Token ID | 198 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
day | +0.017 |
team | +0.014 |
afternoon | +0.014 |
good | +0.014 |
girl | +0.014 |
Neg logit contributions | |
iqu | -0.064 |
yles | -0.061 |
ridge | -0.061 |
angs | -0.060 |
ents | -0.059 |
Token | a |
Token ID | 257 |
Feature activation | -2.37e-02 |
Pos logit contributions | |
iqu | +0.102 |
ridge | +0.102 |
rol | +0.098 |
umes | +0.098 |
vell | +0.097 |
Neg logit contributions | |
, | -0.019 |
day | -0.018 |
and | -0.017 |
the | -0.016 |
their | -0.015 |
Token | strange |
Token ID | 6283 |
Feature activation | -1.95e-02 |
Pos logit contributions | |
iqu | +0.076 |
ridge | +0.075 |
yles | +0.072 |
umes | +0.072 |
rol | +0.071 |
Neg logit contributions | |
day | -0.022 |
good | -0.020 |
, | -0.020 |
girl | -0.019 |
end | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | -2.82e-02 |
Pos logit contributions | |
iqu | +0.053 |
ridge | +0.051 |
umes | +0.049 |
yles | +0.049 |
rol | +0.048 |
Neg logit contributions | |
day | -0.028 |
good | -0.026 |
, | -0.026 |
time | -0.024 |
girl | -0.024 |
Token | raspberry |
Token ID | 38973 |
Feature activation | -4.88e-02 |
Pos logit contributions | |
iqu | +0.084 |
ridge | +0.084 |
yles | +0.080 |
umes | +0.080 |
rol | +0.080 |
Neg logit contributions | |
day | -0.035 |
good | -0.033 |
, | -0.032 |
and | -0.031 |
time | -0.031 |
Token | - |
Token ID | 12 |
Feature activation | -7.83e-02 |
Pos logit contributions | |
iqu | +0.149 |
ridge | +0.140 |
yles | +0.140 |
umes | +0.138 |
rol | +0.138 |
Neg logit contributions | |
day | -0.058 |
, | -0.057 |
and | -0.054 |
good | -0.052 |
time | -0.051 |
Token | col |
Token ID | 4033 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
iqu | +0.153 |
yles | +0.138 |
cher | +0.136 |
umes | +0.133 |
ridge | +0.133 |
Neg logit contributions | |
day | -0.177 |
and | -0.171 |
good | -0.166 |
, | -0.166 |
| -0.163 |
Token | oured |
Token ID | 8167 |
Feature activation | -2.54e-02 |
Pos logit contributions | |
cher | +0.185 |
vell | +0.168 |
outage | +0.163 |
odic | +0.154 |
rol | +0.152 |
Neg logit contributions | |
, | -0.263 |
and | -0.254 |
end | -0.232 |
time | -0.232 |
team | -0.232 |
Token | object |
Token ID | 2134 |
Feature activation | -5.65e-02 |
Pos logit contributions | |
iqu | +0.061 |
ridge | +0.058 |
umes | +0.056 |
yles | +0.055 |
rol | +0.054 |
Neg logit contributions | |
day | -0.044 |
good | -0.041 |
, | -0.041 |
girl | -0.040 |
time | -0.040 |
Token | hiding |
Token ID | 11816 |
Feature activation | -4.07e-03 |
Pos logit contributions | |
iqu | +0.199 |
ridge | +0.197 |
yles | +0.194 |
rol | +0.192 |
umes | +0.187 |
Neg logit contributions | |
, | -0.044 |
and | -0.040 |
day | -0.037 |
. | -0.032 |
good | -0.025 |
Token | under |
Token ID | 739 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
iqu | +0.011 |
ridge | +0.010 |
umes | +0.010 |
yles | +0.010 |
rol | +0.010 |
Neg logit contributions | |
day | -0.006 |
good | -0.005 |
, | -0.005 |
girl | -0.005 |
time | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | -4.22e-02 |
Pos logit contributions | |
iqu | +0.047 |
ridge | +0.046 |
umes | +0.045 |
rol | +0.044 |
yles | +0.044 |
Neg logit contributions | |
day | -0.007 |
, | -0.006 |
good | -0.006 |
time | -0.005 |
and | -0.005 |
Token | its |
Token ID | 663 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
iqu | +0.148 |
ridge | +0.141 |
umes | +0.135 |
yles | +0.133 |
angs | +0.133 |
Neg logit contributions | |
day | -0.060 |
, | -0.059 |
good | -0.054 |
and | -0.054 |
their | -0.054 |
Token | clean |
Token ID | 3424 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
iqu | +0.068 |
ridge | +0.065 |
umes | +0.063 |
yles | +0.062 |
rol | +0.062 |
Neg logit contributions | |
day | -0.034 |
good | -0.031 |
, | -0.031 |
time | -0.030 |
girl | -0.030 |
Token | feathers |
Token ID | 27400 |
Feature activation | -3.37e-02 |
Pos logit contributions | |
iqu | +0.046 |
ridge | +0.045 |
umes | +0.043 |
rol | +0.043 |
yles | +0.043 |
Neg logit contributions | |
day | -0.022 |
good | -0.021 |
team | -0.019 |
, | -0.019 |
end | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -9.30e-03 |
Pos logit contributions | |
iqu | +0.126 |
ridge | +0.122 |
yles | +0.119 |
umes | +0.119 |
rol | +0.117 |
Neg logit contributions | |
day | -0.014 |
, | -0.012 |
good | -0.011 |
and | -0.010 |
time | -0.008 |
Token | The |
Token ID | 383 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
iqu | +0.030 |
ridge | +0.029 |
umes | +0.028 |
yles | +0.028 |
rol | +0.028 |
Neg logit contributions | |
day | -0.008 |
good | -0.007 |
, | -0.007 |
time | -0.007 |
girl | -0.007 |
Token | ostr |
Token ID | 42387 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
iqu | +0.033 |
ridge | +0.032 |
umes | +0.031 |
yles | +0.031 |
rol | +0.030 |
Neg logit contributions | |
day | -0.016 |
good | -0.015 |
, | -0.015 |
time | -0.014 |
girl | -0.014 |
Token | ich |
Token ID | 488 |
Feature activation | -5.01e-02 |
Pos logit contributions | |
iqu | +0.205 |
ridge | +0.195 |
yles | +0.191 |
umes | +0.189 |
paste | +0.188 |
Neg logit contributions | |
, | -0.275 |
and | -0.264 |
friends | -0.263 |
day | -0.258 |
girl | -0.253 |
Token | was |
Token ID | 373 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
iqu | +0.143 |
ridge | +0.139 |
umes | +0.132 |
yles | +0.131 |
rol | +0.130 |
Neg logit contributions | |
day | -0.062 |
good | -0.057 |
end | -0.054 |
time | -0.053 |
, | -0.053 |
Token | friendly |
Token ID | 8030 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
iqu | +0.075 |
ridge | +0.074 |
yles | +0.071 |
umes | +0.070 |
vell | +0.069 |
Neg logit contributions | |
good | -0.027 |
day | -0.026 |
, | -0.026 |
and | -0.025 |
time | -0.024 |
Token | and |
Token ID | 290 |
Feature activation | -8.60e-03 |
Pos logit contributions | |
day | +0.010 |
good | +0.008 |
end | +0.008 |
afternoon | +0.007 |
team | +0.007 |
Neg logit contributions | |
ridge | -0.069 |
iqu | -0.069 |
angs | -0.066 |
umes | -0.066 |
yles | -0.065 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
iqu | +0.025 |
ridge | +0.024 |
umes | +0.023 |
yles | +0.023 |
rol | +0.023 |
Neg logit contributions | |
day | -0.011 |
good | -0.010 |
, | -0.010 |
girl | -0.009 |
time | -0.009 |
Token | as |
Token ID | 355 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
day | +0.005 |
good | +0.005 |
girl | +0.005 |
end | +0.004 |
boy | +0.004 |
Neg logit contributions | |
iqu | -0.028 |
ridge | -0.028 |
umes | -0.026 |
bar | -0.026 |
rol | -0.026 |
Token | he |
Token ID | 339 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
day | +0.012 |
good | +0.011 |
, | +0.011 |
time | +0.010 |
girl | +0.010 |
Neg logit contributions | |
iqu | -0.032 |
ridge | -0.031 |
umes | -0.030 |
yles | -0.030 |
rol | -0.029 |
Token | opened |
Token ID | 4721 |
Feature activation | -4.03e-03 |
Pos logit contributions | |
iqu | +0.037 |
ridge | +0.036 |
umes | +0.035 |
yles | +0.034 |
rol | +0.034 |
Neg logit contributions | |
day | -0.019 |
good | -0.018 |
, | -0.018 |
time | -0.017 |
end | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | -4.67e-03 |
Pos logit contributions | |
iqu | +0.013 |
ridge | +0.013 |
umes | +0.012 |
yles | +0.012 |
rol | +0.012 |
Neg logit contributions | |
day | -0.003 |
good | -0.003 |
, | -0.003 |
girl | -0.003 |
end | -0.003 |
Token | ch |
Token ID | 442 |
Feature activation | -4.94e-02 |
Pos logit contributions | |
iqu | +0.012 |
ridge | +0.012 |
yles | +0.011 |
umes | +0.011 |
rol | +0.011 |
Neg logit contributions | |
day | -0.007 |
good | -0.007 |
, | -0.007 |
end | -0.007 |
girl | -0.006 |
Token | ocol |
Token ID | 4668 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
iqu | +0.059 |
ridge | +0.054 |
umes | +0.049 |
yles | +0.048 |
rol | +0.046 |
Neg logit contributions | |
day | -0.146 |
good | -0.141 |
, | -0.140 |
time | -0.137 |
girl | -0.137 |
Token | ates |
Token ID | 689 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
iqu | +0.277 |
umes | +0.242 |
yles | +0.233 |
ridge | +0.229 |
wheel | +0.227 |
Neg logit contributions | |
day | -0.224 |
, | -0.220 |
time | -0.214 |
and | -0.213 |
end | -0.203 |
Token | and |
Token ID | 290 |
Feature activation | -2.47e-02 |
Pos logit contributions | |
iqu | +0.077 |
ridge | +0.074 |
yles | +0.072 |
umes | +0.071 |
angs | +0.071 |
Neg logit contributions | |
day | -0.014 |
, | -0.014 |
and | -0.012 |
good | -0.011 |
end | -0.010 |
Token | started |
Token ID | 2067 |
Feature activation | -2.76e-03 |
Pos logit contributions | |
iqu | +0.068 |
ridge | +0.064 |
umes | +0.063 |
rol | +0.062 |
angs | +0.062 |
Neg logit contributions | |
day | -0.035 |
, | -0.034 |
good | -0.033 |
and | -0.032 |
their | -0.032 |
Token | eating |
Token ID | 6600 |
Feature activation | -6.41e-03 |
Pos logit contributions | |
iqu | +0.008 |
ridge | +0.007 |
umes | +0.007 |
yles | +0.007 |
angs | +0.007 |
Neg logit contributions | |
day | -0.004 |
, | -0.004 |
good | -0.003 |
and | -0.003 |
their | -0.003 |
Token | them |
Token ID | 606 |
Feature activation | +9.87e-03 |
Pos logit contributions | |
iqu | +0.020 |
ridge | +0.019 |
umes | +0.018 |
yles | +0.018 |
rol | +0.018 |
Neg logit contributions | |
day | -0.007 |
good | -0.006 |
, | -0.006 |
girl | -0.006 |
end | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | +4.71e-03 |
Pos logit contributions | |
iqu | +0.132 |
ridge | +0.127 |
umes | +0.123 |
angs | +0.122 |
yles | +0.122 |
Neg logit contributions | |
day | -0.037 |
good | -0.032 |
girl | -0.032 |
end | -0.031 |
, | -0.031 |
Token | ostr |
Token ID | 42387 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
day | +0.007 |
good | +0.006 |
, | +0.006 |
girl | +0.006 |
time | +0.006 |
Neg logit contributions | |
iqu | -0.013 |
ridge | -0.013 |
yles | -0.012 |
umes | -0.012 |
rol | -0.012 |
Token | ich |
Token ID | 488 |
Feature activation | -3.30e-02 |
Pos logit contributions | |
iqu | +0.192 |
ridge | +0.183 |
yles | +0.180 |
umes | +0.177 |
paste | +0.177 |
Neg logit contributions | |
, | -0.252 |
team | -0.241 |
and | -0.238 |
day | -0.237 |
friends | -0.234 |
Token | . |
Token ID | 13 |
Feature activation | +1.90e-04 |
Pos logit contributions | |
iqu | +0.114 |
ridge | +0.111 |
umes | +0.107 |
yles | +0.106 |
angs | +0.106 |
Neg logit contributions | |
day | -0.021 |
good | -0.018 |
, | -0.016 |
girl | -0.016 |
end | -0.016 |
Token | The |
Token ID | 383 |
Feature activation | -9.92e-03 |
Pos logit contributions | |
day | +0.000 |
good | +0.000 |
girl | +0.000 |
end | +0.000 |
afternoon | +0.000 |
Neg logit contributions | |
iqu | -0.001 |
ridge | -0.001 |
yles | -0.001 |
angs | -0.001 |
umes | -0.001 |
Token | ostr |
Token ID | 42387 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
iqu | +0.018 |
ridge | +0.017 |
umes | +0.016 |
yles | +0.016 |
rol | +0.016 |
Neg logit contributions | |
day | -0.022 |
, | -0.021 |
good | -0.021 |
time | -0.020 |
team | -0.020 |
Token | ich |
Token ID | 488 |
Feature activation | -5.35e-02 |
Pos logit contributions | |
iqu | +0.210 |
ridge | +0.204 |
paste | +0.196 |
umes | +0.193 |
rol | +0.193 |
Neg logit contributions | |
, | -0.271 |
and | -0.263 |
day | -0.256 |
friends | -0.253 |
boy | -0.252 |
Token | had |
Token ID | 550 |
Feature activation | -4.87e-02 |
Pos logit contributions | |
iqu | +0.154 |
ridge | +0.152 |
yles | +0.143 |
umes | +0.143 |
angs | +0.142 |
Neg logit contributions | |
day | -0.055 |
good | -0.053 |
end | -0.051 |
team | -0.049 |
time | -0.048 |
Token | a |
Token ID | 257 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
iqu | +0.170 |
ridge | +0.165 |
umes | +0.160 |
yles | +0.160 |
rol | +0.157 |
Neg logit contributions | |
day | -0.035 |
good | -0.033 |
, | -0.031 |
time | -0.028 |
and | -0.028 |
Token | long |
Token ID | 890 |
Feature activation | -6.62e-04 |
Pos logit contributions | |
iqu | +0.042 |
ridge | +0.041 |
yles | +0.039 |
umes | +0.039 |
rol | +0.039 |
Neg logit contributions | |
day | -0.010 |
good | -0.009 |
, | -0.008 |
time | -0.008 |
girl | -0.007 |
Token | neck |
Token ID | 7393 |
Feature activation | -4.08e-02 |
Pos logit contributions | |
iqu | +0.002 |
ridge | +0.002 |
bar | +0.002 |
angs | +0.002 |
umes | +0.002 |
Neg logit contributions | |
day | -0.001 |
good | -0.001 |
girl | -0.001 |
end | -0.001 |
team | -0.001 |
Token | big |
Token ID | 1263 |
Feature activation | -7.03e-03 |
Pos logit contributions | |
iqu | +0.029 |
ridge | +0.028 |
umes | +0.027 |
yles | +0.026 |
rol | +0.026 |
Neg logit contributions | |
day | -0.012 |
good | -0.011 |
, | -0.011 |
time | -0.011 |
end | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
iqu | +0.018 |
ridge | +0.017 |
umes | +0.017 |
ocol | +0.017 |
bar | +0.016 |
Neg logit contributions | |
day | -0.009 |
good | -0.009 |
time | -0.008 |
team | -0.008 |
their | -0.008 |
Token | beautiful |
Token ID | 4950 |
Feature activation | -3.37e-03 |
Pos logit contributions | |
iqu | +0.049 |
ridge | +0.047 |
umes | +0.046 |
yles | +0.045 |
rol | +0.045 |
Neg logit contributions | |
day | -0.020 |
good | -0.018 |
, | -0.018 |
end | -0.017 |
time | -0.017 |
Token | rainbow |
Token ID | 27223 |
Feature activation | -2.77e-02 |
Pos logit contributions | |
iqu | +0.008 |
ridge | +0.008 |
umes | +0.008 |
yles | +0.008 |
rol | +0.007 |
Neg logit contributions | |
day | -0.005 |
good | -0.005 |
, | -0.005 |
end | -0.005 |
team | -0.005 |
Token | - |
Token ID | 12 |
Feature activation | -6.22e-02 |
Pos logit contributions | |
iqu | +0.086 |
ridge | +0.083 |
umes | +0.080 |
yles | +0.079 |
rol | +0.078 |
Neg logit contributions | |
day | -0.030 |
, | -0.027 |
good | -0.026 |
and | -0.025 |
girl | -0.025 |
Token | col |
Token ID | 4033 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
iqu | +0.103 |
ridge | +0.096 |
yles | +0.087 |
ocol | +0.084 |
angs | +0.083 |
Neg logit contributions | |
day | -0.159 |
and | -0.153 |
, | -0.151 |
good | -0.149 |
girl | -0.148 |
Token | oured |
Token ID | 8167 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
cher | +0.174 |
odic | +0.165 |
iqu | +0.163 |
angs | +0.162 |
paste | +0.161 |
Neg logit contributions | |
, | -0.266 |
and | -0.255 |
girl | -0.240 |
. | -0.230 |
back | -0.229 |
Token | butterfly |
Token ID | 35113 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
iqu | +0.033 |
ridge | +0.032 |
umes | +0.031 |
rol | +0.031 |
yles | +0.031 |
Neg logit contributions | |
day | -0.016 |
good | -0.015 |
, | -0.015 |
time | -0.014 |
end | -0.014 |
Token | ! |
Token ID | 0 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
iqu | +0.079 |
ridge | +0.077 |
umes | +0.075 |
yles | +0.074 |
rol | +0.073 |
Neg logit contributions | |
day | -0.026 |
good | -0.023 |
, | -0.022 |
end | -0.022 |
time | -0.022 |
Token | The |
Token ID | 383 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
iqu | +0.045 |
ridge | +0.044 |
umes | +0.042 |
yles | +0.042 |
angs | +0.041 |
Neg logit contributions | |
day | -0.016 |
, | -0.014 |
good | -0.014 |
girl | -0.013 |
and | -0.013 |
Token | little |
Token ID | 1310 |
Feature activation | -1.50e-02 |
Pos logit contributions | |
iqu | +0.030 |
ridge | +0.028 |
umes | +0.028 |
yles | +0.027 |
vell | +0.027 |
Neg logit contributions | |
day | -0.013 |
, | -0.012 |
good | -0.012 |
their | -0.012 |
time | -0.011 |
Token | feathers |
Token ID | 27400 |
Feature activation | -3.81e-02 |
Pos logit contributions | |
iqu | +0.019 |
ridge | +0.018 |
umes | +0.018 |
yles | +0.017 |
rol | +0.017 |
Neg logit contributions | |
day | -0.012 |
, | -0.011 |
good | -0.011 |
time | -0.011 |
girl | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
iqu | +0.155 |
ridge | +0.152 |
yles | +0.149 |
umes | +0.148 |
ocol | +0.145 |
Neg logit contributions | |
, | -0.009 |
and | -0.006 |
day | -0.005 |
good | -0.002 |
. | -0.000 |
Token | Ben |
Token ID | 3932 |
Feature activation | +2.16e-03 |
Pos logit contributions | |
day | +0.010 |
good | +0.008 |
girl | +0.008 |
end | +0.008 |
afternoon | +0.008 |
Neg logit contributions | |
iqu | -0.030 |
ridge | -0.030 |
yles | -0.029 |
angs | -0.029 |
umes | -0.028 |
Token | thought |
Token ID | 1807 |
Feature activation | -2.47e-02 |
Pos logit contributions | |
day | +0.003 |
end | +0.003 |
good | +0.003 |
afternoon | +0.003 |
time | +0.003 |
Neg logit contributions | |
ridge | -0.005 |
iqu | -0.005 |
yles | -0.005 |
umes | -0.005 |
ents | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | -1.25e-02 |
Pos logit contributions | |
iqu | +0.073 |
ridge | +0.072 |
yles | +0.070 |
angs | +0.069 |
umes | +0.069 |
Neg logit contributions | |
day | -0.025 |
girl | -0.023 |
good | -0.022 |
end | -0.022 |
team | -0.021 |
Token | ostr |
Token ID | 42387 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
iqu | +0.027 |
ridge | +0.026 |
yles | +0.025 |
umes | +0.025 |
rol | +0.024 |
Neg logit contributions | |
day | -0.026 |
good | -0.025 |
, | -0.024 |
girl | -0.024 |
end | -0.024 |
Token | ich |
Token ID | 488 |
Feature activation | -4.21e-02 |
Pos logit contributions | |
iqu | +0.211 |
ridge | +0.205 |
rol | +0.202 |
paste | +0.198 |
ocol | +0.197 |
Neg logit contributions | |
, | -0.258 |
and | -0.252 |
friends | -0.247 |
day | -0.247 |
friend | -0.243 |
Token | was |
Token ID | 373 |
Feature activation | -2.77e-02 |
Pos logit contributions | |
iqu | +0.124 |
ridge | +0.122 |
angs | +0.115 |
umes | +0.115 |
yles | +0.114 |
Neg logit contributions | |
day | -0.043 |
end | -0.039 |
good | -0.039 |
team | -0.038 |
girl | -0.037 |
Token | very |
Token ID | 845 |
Feature activation | +1.43e-02 |
Pos logit contributions | |
iqu | +0.087 |
ridge | +0.084 |
umes | +0.081 |
yles | +0.081 |
angs | +0.080 |
Neg logit contributions | |
day | -0.027 |
, | -0.024 |
good | -0.024 |
girl | -0.023 |
end | -0.023 |
Token | original |
Token ID | 2656 |
Feature activation | +3.03e-02 |
Pos logit contributions | |
day | +0.016 |
end | +0.015 |
, | +0.014 |
team | +0.014 |
girl | +0.014 |
Neg logit contributions | |
iqu | -0.042 |
angs | -0.038 |
ridge | -0.038 |
umes | -0.038 |
vell | -0.038 |
Token | . |
Token ID | 13 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
day | +0.015 |
good | +0.012 |
end | +0.011 |
time | +0.011 |
girl | +0.010 |
Neg logit contributions | |
iqu | -0.108 |
ridge | -0.105 |
umes | -0.102 |
yles | -0.101 |
angs | -0.101 |
Token | his |
Token ID | 465 |
Feature activation | -1.92e-02 |
Pos logit contributions | |
iqu | +0.034 |
ridge | +0.033 |
yles | +0.032 |
umes | +0.032 |
rol | +0.031 |
Neg logit contributions | |
day | -0.010 |
good | -0.009 |
, | -0.008 |
girl | -0.008 |
team | -0.008 |
Token | friends |
Token ID | 2460 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
iqu | +0.068 |
ridge | +0.065 |
umes | +0.063 |
angs | +0.063 |
yles | +0.062 |
Neg logit contributions | |
day | -0.010 |
good | -0.008 |
end | -0.008 |
, | -0.008 |
girl | -0.007 |
Token | dropped |
Token ID | 5710 |
Feature activation | -2.83e-02 |
Pos logit contributions | |
iqu | +0.070 |
ridge | +0.068 |
umes | +0.065 |
yles | +0.065 |
rol | +0.064 |
Neg logit contributions | |
day | -0.031 |
, | -0.029 |
good | -0.028 |
and | -0.027 |
time | -0.027 |
Token | her |
Token ID | 607 |
Feature activation | -1.65e-02 |
Pos logit contributions | |
iqu | +0.101 |
ridge | +0.098 |
umes | +0.095 |
yles | +0.095 |
rol | +0.093 |
Neg logit contributions | |
day | -0.019 |
, | -0.017 |
good | -0.016 |
and | -0.016 |
their | -0.015 |
Token | lip |
Token ID | 10645 |
Feature activation | -6.42e-02 |
Pos logit contributions | |
iqu | +0.047 |
ridge | +0.045 |
umes | +0.044 |
yles | +0.043 |
rol | +0.043 |
Neg logit contributions | |
day | -0.021 |
good | -0.019 |
, | -0.019 |
time | -0.018 |
end | -0.018 |
Token | bal |
Token ID | 3652 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
iqu | +0.258 |
yles | +0.248 |
ridge | +0.246 |
ocol | +0.246 |
rol | +0.242 |
Neg logit contributions | |
, | -0.022 |
and | -0.021 |
day | -0.016 |
good | -0.008 |
time | -0.004 |
Token | m |
Token ID | 76 |
Feature activation | -2.93e-02 |
Pos logit contributions | |
ridge | +0.297 |
iqu | +0.295 |
ocol | +0.278 |
angs | +0.263 |
paste | +0.262 |
Neg logit contributions | |
day | -0.165 |
, | -0.164 |
and | -0.150 |
girl | -0.146 |
| -0.144 |
Token | on |
Token ID | 319 |
Feature activation | +1.51e-03 |
Pos logit contributions | |
iqu | +0.072 |
ridge | +0.069 |
umes | +0.066 |
yles | +0.066 |
rol | +0.065 |
Neg logit contributions | |
day | -0.049 |
good | -0.046 |
, | -0.046 |
time | -0.044 |
girl | -0.044 |
Token | the |
Token ID | 262 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
day | +0.001 |
, | +0.001 |
good | +0.001 |
time | +0.000 |
and | +0.000 |
Neg logit contributions | |
iqu | -0.006 |
ridge | -0.005 |
umes | -0.005 |
yles | -0.005 |
rol | -0.005 |
Token | ground |
Token ID | 2323 |
Feature activation | -3.17e-02 |
Pos logit contributions | |
iqu | +0.068 |
ridge | +0.065 |
umes | +0.063 |
yles | +0.063 |
rol | +0.062 |
Neg logit contributions | |
day | -0.039 |
good | -0.036 |
, | -0.036 |
end | -0.034 |
time | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
iqu | +0.115 |
ridge | +0.112 |
umes | +0.109 |
yles | +0.108 |
rol | +0.107 |
Neg logit contributions | |
day | -0.015 |
good | -0.012 |
, | -0.011 |
girl | -0.010 |
time | -0.010 |
Token | couldn |
Token ID | 3521 |
Feature activation | -4.09e-02 |
Pos logit contributions | |
day | +0.006 |
good | +0.006 |
team | +0.005 |
end | +0.005 |
girl | +0.005 |
Neg logit contributions | |
iqu | -0.013 |
ridge | -0.012 |
rol | -0.012 |
bar | -0.012 |
angs | -0.011 |
Token | 't |
Token ID | 470 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
iqu | +0.068 |
ridge | +0.063 |
umes | +0.060 |
yles | +0.059 |
angs | +0.057 |
Neg logit contributions | |
day | -0.103 |
, | -0.099 |
good | -0.098 |
time | -0.096 |
and | -0.096 |
Token | see |
Token ID | 766 |
Feature activation | -2.39e-02 |
Pos logit contributions | |
iqu | +0.045 |
ridge | +0.044 |
umes | +0.041 |
yles | +0.041 |
rol | +0.041 |
Neg logit contributions | |
day | -0.028 |
good | -0.026 |
, | -0.026 |
girl | -0.025 |
time | -0.025 |
Token | what |
Token ID | 644 |
Feature activation | -3.56e-02 |
Pos logit contributions | |
iqu | +0.076 |
ridge | +0.074 |
umes | +0.071 |
rol | +0.070 |
yles | +0.069 |
Neg logit contributions | |
day | -0.025 |
, | -0.023 |
good | -0.022 |
time | -0.021 |
and | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
iqu | +0.095 |
ridge | +0.091 |
yles | +0.088 |
umes | +0.087 |
angs | +0.086 |
Neg logit contributions | |
day | -0.046 |
end | -0.043 |
good | -0.043 |
girl | -0.043 |
team | -0.042 |
Token | ostr |
Token ID | 42387 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
iqu | +0.049 |
ridge | +0.048 |
umes | +0.046 |
yles | +0.045 |
rol | +0.045 |
Neg logit contributions | |
day | -0.027 |
good | -0.025 |
, | -0.025 |
girl | -0.024 |
time | -0.024 |
Token | ich |
Token ID | 488 |
Feature activation | -6.48e-02 |
Pos logit contributions | |
iqu | +0.200 |
rol | +0.182 |
umes | +0.182 |
yles | +0.182 |
angs | +0.182 |
Neg logit contributions | |
, | -0.269 |
and | -0.258 |
day | -0.257 |
friends | -0.252 |
girl | -0.247 |
Token | was |
Token ID | 373 |
Feature activation | -2.68e-02 |
Pos logit contributions | |
iqu | +0.213 |
ridge | +0.205 |
umes | +0.200 |
yles | +0.198 |
rol | +0.198 |
Neg logit contributions | |
day | -0.060 |
, | -0.056 |
and | -0.051 |
good | -0.049 |
time | -0.048 |
Token | hiding |
Token ID | 11816 |
Feature activation | +8.78e-03 |
Pos logit contributions | |
iqu | +0.074 |
ridge | +0.071 |
umes | +0.068 |
yles | +0.068 |
rol | +0.067 |
Neg logit contributions | |
day | -0.039 |
, | -0.037 |
good | -0.036 |
and | -0.035 |
time | -0.035 |
Token | , |
Token ID | 11 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
day | +0.008 |
good | +0.007 |
girl | +0.007 |
end | +0.007 |
time | +0.007 |
Neg logit contributions | |
iqu | -0.027 |
ridge | -0.026 |
umes | -0.026 |
yles | -0.025 |
rol | -0.025 |
Token | but |
Token ID | 475 |
Feature activation | -2.00e-02 |
Pos logit contributions | |
iqu | +0.069 |
ridge | +0.066 |
umes | +0.064 |
yles | +0.064 |
rol | +0.063 |
Neg logit contributions | |
day | -0.031 |
good | -0.029 |
, | -0.028 |
time | -0.027 |
end | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | -2.79e-02 |
Pos logit contributions | |
day | +0.006 |
end | +0.005 |
good | +0.005 |
team | +0.005 |
time | +0.005 |
Neg logit contributions | |
iqu | -0.020 |
ridge | -0.019 |
umes | -0.019 |
yles | -0.018 |
angs | -0.018 |
Token | she |
Token ID | 673 |
Feature activation | +3.51e-03 |
Pos logit contributions | |
iqu | +0.070 |
ridge | +0.067 |
yles | +0.065 |
umes | +0.065 |
rol | +0.064 |
Neg logit contributions | |
day | -0.042 |
good | -0.039 |
team | -0.038 |
end | -0.038 |
time | -0.037 |
Token | saw |
Token ID | 2497 |
Feature activation | -2.65e-02 |
Pos logit contributions | |
day | +0.004 |
good | +0.004 |
, | +0.004 |
time | +0.004 |
end | +0.004 |
Neg logit contributions | |
iqu | -0.010 |
ridge | -0.010 |
umes | -0.010 |
yles | -0.009 |
rol | -0.009 |
Token | that |
Token ID | 326 |
Feature activation | -2.61e-02 |
Pos logit contributions | |
iqu | +0.090 |
ridge | +0.090 |
umes | +0.086 |
rol | +0.085 |
vell | +0.085 |
Neg logit contributions | |
, | -0.023 |
day | -0.022 |
their | -0.022 |
and | -0.022 |
the | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
iqu | +0.084 |
ridge | +0.082 |
umes | +0.079 |
rol | +0.078 |
yles | +0.078 |
Neg logit contributions | |
day | -0.026 |
, | -0.023 |
good | -0.023 |
girl | -0.021 |
and | -0.021 |
Token | pr |
Token ID | 778 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
iqu | +0.039 |
ridge | +0.039 |
yles | +0.037 |
umes | +0.037 |
rol | +0.037 |
Neg logit contributions | |
day | -0.021 |
good | -0.019 |
girl | -0.018 |
time | -0.018 |
, | -0.018 |
Token | une |
Token ID | 1726 |
Feature activation | -6.09e-02 |
Pos logit contributions | |
yles | +0.194 |
iqu | +0.179 |
cher | +0.163 |
rol | +0.153 |
vell | +0.153 |
Neg logit contributions | |
, | -0.289 |
and | -0.276 |
day | -0.269 |
end | -0.267 |
girl | -0.259 |
Token | had |
Token ID | 550 |
Feature activation | -5.14e-02 |
Pos logit contributions | |
iqu | +0.206 |
ridge | +0.199 |
umes | +0.194 |
yles | +0.193 |
rol | +0.191 |
Neg logit contributions | |
day | -0.048 |
, | -0.042 |
good | -0.041 |
and | -0.038 |
time | -0.037 |
Token | disappeared |
Token ID | 12120 |
Feature activation | -1.50e-02 |
Pos logit contributions | |
iqu | +0.126 |
ridge | +0.120 |
yles | +0.119 |
umes | +0.115 |
rol | +0.113 |
Neg logit contributions | |
good | -0.089 |
day | -0.089 |
time | -0.086 |
, | -0.086 |
and | -0.084 |
Token | ! |
Token ID | 0 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
iqu | +0.048 |
ridge | +0.047 |
yles | +0.046 |
umes | +0.045 |
rol | +0.045 |
Neg logit contributions | |
day | -0.014 |
, | -0.013 |
good | -0.012 |
and | -0.012 |
time | -0.011 |
Token | She |
Token ID | 1375 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
iqu | +0.045 |
ridge | +0.044 |
umes | +0.043 |
yles | +0.042 |
rol | +0.042 |
Neg logit contributions | |
day | -0.019 |
good | -0.017 |
end | -0.017 |
, | -0.017 |
time | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | -3.44e-03 |
Pos logit contributions | |
iqu | +0.078 |
ridge | +0.075 |
umes | +0.074 |
yles | +0.074 |
angs | +0.072 |
Neg logit contributions | |
day | -0.012 |
good | -0.010 |
, | -0.009 |
girl | -0.008 |
end | -0.008 |
Token | ostr |
Token ID | 42387 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
iqu | +0.010 |
ridge | +0.009 |
umes | +0.009 |
yles | +0.009 |
rol | +0.009 |
Neg logit contributions | |
day | -0.004 |
good | -0.004 |
, | -0.004 |
time | -0.004 |
and | -0.004 |
Token | ich |
Token ID | 488 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
iqu | +0.174 |
paste | +0.165 |
umes | +0.165 |
ridge | +0.165 |
rol | +0.163 |
Neg logit contributions | |
, | -0.256 |
and | -0.244 |
team | -0.240 |
boy | -0.237 |
day | -0.236 |
Token | . |
Token ID | 13 |
Feature activation | +3.92e-03 |
Pos logit contributions | |
iqu | +0.063 |
ridge | +0.062 |
umes | +0.060 |
yles | +0.059 |
bar | +0.059 |
Neg logit contributions | |
day | -0.018 |
good | -0.018 |
time | -0.016 |
girl | -0.015 |
end | -0.015 |
Token | The |
Token ID | 383 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
day | +0.003 |
good | +0.003 |
, | +0.003 |
time | +0.002 |
end | +0.002 |
Neg logit contributions | |
iqu | -0.013 |
ridge | -0.013 |
umes | -0.012 |
yles | -0.012 |
rol | -0.012 |
Token | ostr |
Token ID | 42387 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
iqu | +0.030 |
ridge | +0.028 |
umes | +0.027 |
yles | +0.027 |
rol | +0.026 |
Neg logit contributions | |
day | -0.032 |
good | -0.030 |
, | -0.030 |
girl | -0.030 |
end | -0.030 |
Token | ich |
Token ID | 488 |
Feature activation | -4.40e-02 |
Pos logit contributions | |
iqu | +0.199 |
paste | +0.194 |
ridge | +0.193 |
rol | +0.193 |
umes | +0.192 |
Neg logit contributions | |
, | -0.267 |
and | -0.255 |
friends | -0.248 |
day | -0.244 |
boy | -0.244 |
Token | did |
Token ID | 750 |
Feature activation | -7.58e-03 |
Pos logit contributions | |
iqu | +0.126 |
ridge | +0.124 |
umes | +0.118 |
yles | +0.117 |
rol | +0.115 |
Neg logit contributions | |
day | -0.053 |
good | -0.049 |
end | -0.046 |
time | -0.045 |
team | -0.045 |
Token | not |
Token ID | 407 |
Feature activation | -2.47e-02 |
Pos logit contributions | |
iqu | +0.018 |
ridge | +0.017 |
yles | +0.017 |
umes | +0.017 |
angs | +0.016 |
Neg logit contributions | |
day | -0.013 |
girl | -0.012 |
team | -0.011 |
boy | -0.011 |
afternoon | -0.011 |
Token | like |
Token ID | 588 |
Feature activation | -5.30e-03 |
Pos logit contributions | |
iqu | +0.059 |
ridge | +0.058 |
angs | +0.056 |
vell | +0.055 |
umes | +0.055 |
Neg logit contributions | |
day | -0.038 |
good | -0.034 |
afternoon | -0.034 |
girl | -0.034 |
, | -0.034 |
Token | them |
Token ID | 606 |
Feature activation | -1.37e-03 |
Pos logit contributions | |
iqu | +0.018 |
ridge | +0.017 |
umes | +0.017 |
rol | +0.017 |
yles | +0.016 |
Neg logit contributions | |
day | -0.005 |
, | -0.004 |
good | -0.004 |
time | -0.004 |
and | -0.004 |
Token | out |
Token ID | 503 |
Feature activation | +2.71e-02 |
Pos logit contributions | |
iqu | +0.020 |
ridge | +0.020 |
umes | +0.019 |
yles | +0.019 |
angs | +0.019 |
Neg logit contributions | |
day | -0.007 |
end | -0.007 |
team | -0.006 |
good | -0.006 |
time | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | -3.72e-03 |
Pos logit contributions | |
day | +0.023 |
afternoon | +0.019 |
end | +0.018 |
time | +0.018 |
girl | +0.017 |
Neg logit contributions | |
iqu | -0.087 |
ridge | -0.083 |
umes | -0.081 |
angs | -0.081 |
rol | -0.080 |
Token | bread |
Token ID | 8509 |
Feature activation | -2.38e-02 |
Pos logit contributions | |
iqu | +0.009 |
ridge | +0.009 |
umes | +0.009 |
yles | +0.009 |
rol | +0.008 |
Neg logit contributions | |
day | -0.006 |
good | -0.006 |
, | -0.006 |
girl | -0.005 |
time | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +5.22e-03 |
Pos logit contributions | |
iqu | +0.087 |
ridge | +0.085 |
umes | +0.082 |
yles | +0.082 |
rol | +0.081 |
Neg logit contributions | |
day | -0.012 |
, | -0.009 |
good | -0.009 |
and | -0.008 |
end | -0.007 |
Token | The |
Token ID | 383 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
day | +0.005 |
good | +0.005 |
, | +0.004 |
end | +0.004 |
girl | +0.004 |
Neg logit contributions | |
iqu | -0.016 |
ridge | -0.016 |
umes | -0.015 |
yles | -0.015 |
angs | -0.015 |
Token | ostr |
Token ID | 42387 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
iqu | +0.023 |
ridge | +0.022 |
umes | +0.020 |
yles | +0.020 |
rol | +0.020 |
Neg logit contributions | |
day | -0.025 |
, | -0.024 |
good | -0.024 |
time | -0.023 |
team | -0.023 |
Token | ich |
Token ID | 488 |
Feature activation | -4.83e-02 |
Pos logit contributions | |
iqu | +0.206 |
ridge | +0.197 |
rol | +0.191 |
paste | +0.191 |
yles | +0.188 |
Neg logit contributions | |
, | -0.256 |
and | -0.251 |
friends | -0.244 |
day | -0.240 |
girl | -0.237 |
Token | came |
Token ID | 1625 |
Feature activation | -4.01e-02 |
Pos logit contributions | |
iqu | +0.132 |
ridge | +0.131 |
umes | +0.124 |
yles | +0.122 |
rol | +0.121 |
Neg logit contributions | |
day | -0.064 |
good | -0.059 |
end | -0.057 |
time | -0.056 |
, | -0.055 |
Token | closer |
Token ID | 5699 |
Feature activation | +3.14e-02 |
Pos logit contributions | |
iqu | +0.111 |
ridge | +0.108 |
yles | +0.104 |
umes | +0.102 |
ocol | +0.100 |
Neg logit contributions | |
day | -0.056 |
, | -0.055 |
good | -0.053 |
and | -0.053 |
end | -0.049 |
Token | and |
Token ID | 290 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
day | +0.018 |
good | +0.014 |
girl | +0.014 |
team | +0.014 |
end | +0.014 |
Neg logit contributions | |
iqu | -0.105 |
ridge | -0.104 |
umes | -0.100 |
yles | -0.100 |
rings | -0.099 |
Token | pe |
Token ID | 613 |
Feature activation | -6.34e-02 |
Pos logit contributions | |
iqu | +0.052 |
ridge | +0.049 |
rol | +0.048 |
umes | +0.048 |
yles | +0.047 |
Neg logit contributions | |
day | -0.028 |
, | -0.027 |
good | -0.027 |
and | -0.026 |
girl | -0.026 |
Token | fence |
Token ID | 13990 |
Feature activation | -3.51e-02 |
Pos logit contributions | |
iqu | +0.022 |
ridge | +0.021 |
umes | +0.020 |
bar | +0.020 |
angs | +0.019 |
Neg logit contributions | |
day | -0.012 |
, | -0.012 |
good | -0.012 |
team | -0.012 |
They | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -3.85e-04 |
Pos logit contributions | |
iqu | +0.129 |
ridge | +0.127 |
umes | +0.123 |
rol | +0.122 |
yles | +0.121 |
Neg logit contributions | |
day | -0.013 |
good | -0.011 |
girl | -0.008 |
, | -0.008 |
time | -0.008 |
Token | He |
Token ID | 679 |
Feature activation | +2.66e-03 |
Pos logit contributions | |
iqu | +0.001 |
ridge | +0.001 |
umes | +0.001 |
yles | +0.001 |
rol | +0.001 |
Neg logit contributions | |
day | -0.000 |
, | -0.000 |
good | -0.000 |
time | -0.000 |
girl | -0.000 |
Token | puts |
Token ID | 7584 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
day | +0.004 |
good | +0.003 |
, | +0.003 |
end | +0.003 |
time | +0.003 |
Neg logit contributions | |
iqu | -0.007 |
ridge | -0.007 |
umes | -0.007 |
yles | -0.007 |
angs | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
iqu | +0.049 |
ridge | +0.047 |
umes | +0.046 |
yles | +0.046 |
angs | +0.045 |
Neg logit contributions | |
day | -0.011 |
good | -0.009 |
, | -0.009 |
time | -0.008 |
and | -0.008 |
Token | ostr |
Token ID | 42387 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
iqu | +0.056 |
ridge | +0.054 |
yles | +0.052 |
umes | +0.052 |
rol | +0.051 |
Neg logit contributions | |
day | -0.036 |
good | -0.033 |
, | -0.033 |
girl | -0.032 |
end | -0.032 |
Token | ich |
Token ID | 488 |
Feature activation | -3.38e-02 |
Pos logit contributions | |
iqu | +0.195 |
ridge | +0.186 |
yles | +0.186 |
rol | +0.185 |
umes | +0.180 |
Neg logit contributions | |
, | -0.264 |
and | -0.257 |
friends | -0.251 |
team | -0.250 |
day | -0.248 |
Token | in |
Token ID | 287 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
iqu | +0.099 |
ridge | +0.098 |
umes | +0.094 |
rol | +0.092 |
yles | +0.092 |
Neg logit contributions | |
day | -0.037 |
good | -0.034 |
time | -0.032 |
, | -0.032 |
girl | -0.032 |
Token | the |
Token ID | 262 |
Feature activation | +4.19e-05 |
Pos logit contributions | |
day | +0.012 |
good | +0.010 |
, | +0.010 |
girl | +0.010 |
end | +0.009 |
Neg logit contributions | |
iqu | -0.058 |
ridge | -0.057 |
umes | -0.055 |
yles | -0.055 |
rol | -0.054 |
Token | box |
Token ID | 3091 |
Feature activation | -2.26e-02 |
Pos logit contributions | |
day | +0.000 |
, | +0.000 |
team | +0.000 |
good | +0.000 |
and | +0.000 |
Neg logit contributions | |
iqu | -0.000 |
ridge | -0.000 |
bar | -0.000 |
umes | -0.000 |
angs | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | +3.96e-03 |
Pos logit contributions | |
iqu | +0.080 |
ridge | +0.078 |
umes | +0.076 |
bar | +0.075 |
rol | +0.075 |
Neg logit contributions | |
day | -0.011 |
good | -0.009 |
girl | -0.008 |
end | -0.008 |
time | -0.007 |
Token | She |
Token ID | 1375 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
iqu | +0.025 |
ridge | +0.024 |
umes | +0.024 |
yles | +0.023 |
rol | +0.023 |
Neg logit contributions | |
day | -0.008 |
good | -0.007 |
end | -0.007 |
, | -0.007 |
time | -0.006 |
Token | made |
Token ID | 925 |
Feature activation | -1.29e-02 |
Pos logit contributions | |
day | +0.014 |
good | +0.013 |
, | +0.012 |
end | +0.012 |
girl | +0.012 |
Neg logit contributions | |
iqu | -0.029 |
ridge | -0.028 |
umes | -0.026 |
yles | -0.026 |
rol | -0.026 |
Token | sure |
Token ID | 1654 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
iqu | +0.043 |
ridge | +0.042 |
umes | +0.041 |
yles | +0.040 |
vell | +0.040 |
Neg logit contributions | |
day | -0.011 |
good | -0.011 |
and | -0.010 |
, | -0.010 |
their | -0.009 |
Token | that |
Token ID | 326 |
Feature activation | -1.95e-02 |
Pos logit contributions | |
iqu | +0.044 |
ridge | +0.043 |
umes | +0.041 |
rol | +0.041 |
yles | +0.041 |
Neg logit contributions | |
day | -0.012 |
good | -0.011 |
, | -0.011 |
their | -0.010 |
and | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | -2.27e-02 |
Pos logit contributions | |
iqu | +0.061 |
ridge | +0.058 |
umes | +0.056 |
rol | +0.056 |
angs | +0.056 |
Neg logit contributions | |
day | -0.022 |
, | -0.021 |
good | -0.020 |
and | -0.019 |
time | -0.019 |
Token | m |
Token ID | 285 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
iqu | +0.062 |
ridge | +0.060 |
umes | +0.057 |
yles | +0.057 |
rol | +0.056 |
Neg logit contributions | |
day | -0.032 |
good | -0.029 |
, | -0.029 |
time | -0.028 |
girl | -0.028 |
Token | ule |
Token ID | 2261 |
Feature activation | -5.89e-02 |
Pos logit contributions | |
iqu | +0.205 |
yles | +0.203 |
vell | +0.194 |
umes | +0.189 |
cher | +0.181 |
Neg logit contributions | |
day | -0.278 |
, | -0.263 |
and | -0.260 |
girl | -0.248 |
end | -0.245 |
Token | was |
Token ID | 373 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
iqu | +0.181 |
ridge | +0.175 |
umes | +0.169 |
yles | +0.168 |
rol | +0.166 |
Neg logit contributions | |
day | -0.061 |
good | -0.055 |
, | -0.055 |
time | -0.052 |
girl | -0.051 |
Token | always |
Token ID | 1464 |
Feature activation | -3.53e-03 |
Pos logit contributions | |
iqu | +0.055 |
ridge | +0.053 |
yles | +0.052 |
umes | +0.051 |
angs | +0.050 |
Neg logit contributions | |
day | -0.023 |
good | -0.022 |
, | -0.021 |
time | -0.020 |
and | -0.020 |
Token | nice |
Token ID | 3621 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
iqu | +0.010 |
ridge | +0.010 |
umes | +0.009 |
yles | +0.009 |
rol | +0.009 |
Neg logit contributions | |
day | -0.005 |
good | -0.004 |
, | -0.004 |
end | -0.004 |
girl | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
day | +0.013 |
good | +0.011 |
end | +0.010 |
, | +0.010 |
girl | +0.010 |
Neg logit contributions | |
iqu | -0.069 |
ridge | -0.067 |
umes | -0.065 |
yles | -0.065 |
angs | -0.064 |
Token | , |
Token ID | 11 |
Feature activation | -3.21e-02 |
Pos logit contributions | |
iqu | +0.148 |
ridge | +0.145 |
umes | +0.140 |
yles | +0.140 |
rol | +0.139 |
Neg logit contributions | |
day | -0.011 |
, | -0.010 |
good | -0.008 |
and | -0.006 |
time | -0.005 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.75e-02 |
Pos logit contributions | |
iqu | +0.117 |
ridge | +0.111 |
rol | +0.108 |
yles | +0.108 |
angs | +0.106 |
Neg logit contributions | |
, | -0.023 |
day | -0.022 |
good | -0.021 |
girl | -0.019 |
boy | -0.018 |
Token | my |
Token ID | 1820 |
Feature activation | +1.27e-02 |
Pos logit contributions | |
iqu | +0.048 |
ridge | +0.046 |
yles | +0.044 |
umes | +0.044 |
angs | +0.043 |
Neg logit contributions | |
, | -0.026 |
day | -0.026 |
and | -0.024 |
girl | -0.023 |
good | -0.023 |
Token | ! |
Token ID | 0 |
Feature activation | -6.48e-03 |
Pos logit contributions | |
day | +0.011 |
good | +0.010 |
end | +0.009 |
time | +0.009 |
team | +0.009 |
Neg logit contributions | |
iqu | -0.041 |
ridge | -0.040 |
umes | -0.039 |
yles | -0.038 |
angs | -0.038 |
Token | The |
Token ID | 383 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
iqu | +0.019 |
ridge | +0.018 |
yles | +0.018 |
rol | +0.017 |
umes | +0.017 |
Neg logit contributions | |
day | -0.008 |
, | -0.008 |
They | -0.007 |
girl | -0.007 |
good | -0.007 |
Token | ostr |
Token ID | 42387 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
iqu | +0.036 |
ridge | +0.034 |
yles | +0.032 |
umes | +0.032 |
rol | +0.031 |
Neg logit contributions | |
day | -0.040 |
good | -0.038 |
, | -0.038 |
end | -0.037 |
girl | -0.037 |
Token | ich |
Token ID | 488 |
Feature activation | -5.69e-02 |
Pos logit contributions | |
iqu | +0.204 |
ridge | +0.194 |
paste | +0.186 |
rol | +0.185 |
umes | +0.182 |
Neg logit contributions | |
, | -0.265 |
and | -0.258 |
day | -0.253 |
friends | -0.245 |
boy | -0.244 |
Token | is |
Token ID | 318 |
Feature activation | -2.25e-02 |
Pos logit contributions | |
iqu | +0.180 |
ridge | +0.175 |
umes | +0.169 |
yles | +0.168 |
rol | +0.166 |
Neg logit contributions | |
day | -0.053 |
good | -0.047 |
, | -0.045 |
end | -0.044 |
time | -0.043 |
Token | so |
Token ID | 523 |
Feature activation | +2.43e-03 |
Pos logit contributions | |
iqu | +0.067 |
ridge | +0.065 |
yles | +0.063 |
umes | +0.062 |
rol | +0.061 |
Neg logit contributions | |
day | -0.028 |
good | -0.027 |
, | -0.027 |
and | -0.025 |
time | -0.025 |
Token | fat |
Token ID | 3735 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
day | +0.002 |
good | +0.002 |
, | +0.002 |
time | +0.002 |
girl | +0.001 |
Neg logit contributions | |
iqu | -0.008 |
ridge | -0.008 |
umes | -0.008 |
yles | -0.008 |
rol | -0.008 |
Token | !" |
Token ID | 2474 |
Feature activation | +5.94e-03 |
Pos logit contributions | |
day | +0.025 |
good | +0.024 |
, | +0.023 |
end | +0.023 |
time | +0.022 |
Neg logit contributions | |
iqu | -0.049 |
ridge | -0.047 |
umes | -0.045 |
yles | -0.045 |
rol | -0.044 |
Token | por |
Token ID | 16964 |
Feature activation | -9.77e-02 |
Pos logit contributions | |
iqu | +0.043 |
ridge | +0.042 |
yles | +0.041 |
umes | +0.040 |
rol | +0.040 |
Neg logit contributions | |
day | -0.021 |
good | -0.020 |
, | -0.020 |
girl | -0.019 |
time | -0.019 |
Token | cup |
Token ID | 25244 |
Feature activation | -3.07e-02 |
Pos logit contributions | |
yles | +0.171 |
iqu | +0.167 |
rol | +0.148 |
umes | +0.140 |
rings | +0.138 |
Neg logit contributions | |
, | -0.238 |
and | -0.224 |
good | -0.217 |
day | -0.217 |
end | -0.212 |
Token | ine |
Token ID | 500 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
iqu | +0.070 |
yles | +0.066 |
ridge | +0.064 |
umes | +0.063 |
rol | +0.063 |
Neg logit contributions | |
, | -0.060 |
day | -0.058 |
and | -0.057 |
time | -0.055 |
good | -0.054 |
Token | ! |
Token ID | 0 |
Feature activation | -7.00e-04 |
Pos logit contributions | |
iqu | +0.080 |
ridge | +0.078 |
umes | +0.075 |
yles | +0.075 |
rol | +0.074 |
Neg logit contributions | |
day | -0.012 |
good | -0.009 |
, | -0.009 |
time | -0.008 |
girl | -0.008 |
Token | The |
Token ID | 383 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
iqu | +0.002 |
ridge | +0.002 |
umes | +0.002 |
yles | +0.002 |
rol | +0.002 |
Neg logit contributions | |
day | -0.000 |
, | -0.000 |
good | -0.000 |
and | -0.000 |
girl | -0.000 |
Token | por |
Token ID | 16964 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
iqu | +0.037 |
ridge | +0.035 |
yles | +0.035 |
umes | +0.034 |
rol | +0.034 |
Neg logit contributions | |
day | -0.028 |
good | -0.026 |
, | -0.026 |
girl | -0.026 |
end | -0.026 |
Token | cup |
Token ID | 25244 |
Feature activation | -4.79e-02 |
Pos logit contributions | |
iqu | +0.198 |
yles | +0.188 |
rol | +0.169 |
umes | +0.163 |
bar | +0.162 |
Neg logit contributions | |
, | -0.279 |
and | -0.274 |
day | -0.262 |
" | -0.254 |
good | -0.253 |
Token | ine |
Token ID | 500 |
Feature activation | -3.70e-02 |
Pos logit contributions | |
iqu | +0.098 |
yles | +0.093 |
ridge | +0.091 |
umes | +0.090 |
vell | +0.089 |
Neg logit contributions | |
, | -0.099 |
day | -0.097 |
and | -0.096 |
time | -0.092 |
good | -0.090 |
Token | qu |
Token ID | 627 |
Feature activation | -5.17e-02 |
Pos logit contributions | |
iqu | +0.104 |
ridge | +0.102 |
umes | +0.097 |
yles | +0.095 |
angs | +0.094 |
Neg logit contributions | |
day | -0.048 |
good | -0.043 |
end | -0.042 |
girl | -0.041 |
time | -0.040 |
Token | illed |
Token ID | 2967 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
iqu | +0.119 |
ridge | +0.113 |
yles | +0.110 |
umes | +0.109 |
vell | +0.105 |
Neg logit contributions | |
day | -0.091 |
, | -0.087 |
good | -0.087 |
time | -0.085 |
and | -0.085 |
Token | and |
Token ID | 290 |
Feature activation | -4.57e-03 |
Pos logit contributions | |
day | +0.007 |
good | +0.006 |
girl | +0.005 |
end | +0.005 |
team | +0.005 |
Neg logit contributions | |
iqu | -0.038 |
ridge | -0.037 |
umes | -0.036 |
angs | -0.036 |
yles | -0.035 |
Token | the |
Token ID | 262 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
iqu | +0.022 |
ridge | +0.021 |
yles | +0.021 |
umes | +0.021 |
angs | +0.020 |
Neg logit contributions | |
day | -0.004 |
good | -0.003 |
, | -0.003 |
girl | -0.003 |
end | -0.003 |
Token | ostr |
Token ID | 42387 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
iqu | +0.031 |
ridge | +0.030 |
umes | +0.029 |
yles | +0.029 |
angs | +0.028 |
Neg logit contributions | |
day | -0.015 |
good | -0.014 |
, | -0.013 |
time | -0.013 |
end | -0.013 |
Token | ich |
Token ID | 488 |
Feature activation | -2.25e-02 |
Pos logit contributions | |
iqu | +0.186 |
ridge | +0.177 |
paste | +0.176 |
umes | +0.175 |
rol | +0.171 |
Neg logit contributions | |
, | -0.253 |
and | -0.242 |
team | -0.238 |
friends | -0.237 |
boy | -0.234 |
Token | . |
Token ID | 13 |
Feature activation | +6.09e-03 |
Pos logit contributions | |
iqu | +0.080 |
ridge | +0.079 |
umes | +0.076 |
yles | +0.075 |
rol | +0.075 |
Neg logit contributions | |
day | -0.011 |
good | -0.010 |
time | -0.008 |
girl | -0.008 |
end | -0.007 |
Token | The |
Token ID | 383 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
day | +0.005 |
good | +0.004 |
end | +0.004 |
time | +0.004 |
girl | +0.004 |
Neg logit contributions | |
iqu | -0.020 |
ridge | -0.019 |
umes | -0.019 |
yles | -0.019 |
rol | -0.019 |
Token | ostr |
Token ID | 42387 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
iqu | +0.038 |
ridge | +0.036 |
yles | +0.034 |
umes | +0.034 |
rol | +0.033 |
Neg logit contributions | |
day | -0.041 |
good | -0.039 |
, | -0.038 |
girl | -0.038 |
end | -0.038 |
Token | ich |
Token ID | 488 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
iqu | +0.204 |
paste | +0.201 |
ridge | +0.199 |
rol | +0.196 |
umes | +0.194 |
Neg logit contributions | |
, | -0.255 |
and | -0.245 |
friends | -0.239 |
boy | -0.234 |
friend | -0.233 |
Token | sees |
Token ID | 7224 |
Feature activation | -2.38e-02 |
Pos logit contributions | |
iqu | +0.107 |
ridge | +0.104 |
angs | +0.097 |
yles | +0.097 |
umes | +0.096 |
Neg logit contributions | |
day | -0.054 |
good | -0.048 |
afternoon | -0.046 |
time | -0.045 |
girl | -0.045 |
Token | them |
Token ID | 606 |
Feature activation | +5.52e-04 |
Pos logit contributions | |
iqu | +0.080 |
ridge | +0.078 |
umes | +0.076 |
yles | +0.074 |
rol | +0.074 |
Neg logit contributions | |
day | -0.019 |
, | -0.018 |
good | -0.017 |
and | -0.016 |
their | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
day | +0.000 |
, | +0.000 |
good | +0.000 |
and | +0.000 |
end | +0.000 |
Neg logit contributions | |
iqu | -0.002 |
ridge | -0.002 |
umes | -0.002 |
yles | -0.002 |
rol | -0.002 |
Token | gets |
Token ID | 3011 |
Feature activation | -2.86e-02 |
Pos logit contributions | |
iqu | +0.049 |
ridge | +0.048 |
umes | +0.046 |
yles | +0.046 |
angs | +0.045 |
Neg logit contributions | |
day | -0.024 |
good | -0.022 |
, | -0.022 |
time | -0.021 |
end | -0.021 |
Token | He |
Token ID | 679 |
Feature activation | -7.57e-03 |
Pos logit contributions | |
day | +0.017 |
end | +0.014 |
good | +0.014 |
team | +0.014 |
time | +0.014 |
Neg logit contributions | |
iqu | -0.053 |
ridge | -0.051 |
yles | -0.049 |
umes | -0.049 |
angs | -0.049 |
Token | looked |
Token ID | 3114 |
Feature activation | +1.64e-03 |
Pos logit contributions | |
iqu | +0.023 |
ridge | +0.023 |
umes | +0.022 |
yles | +0.022 |
rol | +0.022 |
Neg logit contributions | |
and | -0.007 |
, | -0.007 |
day | -0.007 |
good | -0.007 |
girl | -0.007 |
Token | wealthy |
Token ID | 11574 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
day | +0.002 |
team | +0.002 |
end | +0.002 |
good | +0.002 |
, | +0.002 |
Neg logit contributions | |
iqu | -0.004 |
ridge | -0.004 |
umes | -0.004 |
yles | -0.004 |
angs | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
day | +0.008 |
good | +0.006 |
time | +0.005 |
end | +0.005 |
afternoon | +0.005 |
Neg logit contributions | |
iqu | -0.049 |
ridge | -0.048 |
umes | -0.046 |
yles | -0.046 |
angs | -0.046 |
Token | He |
Token ID | 679 |
Feature activation | +1.30e-03 |
Pos logit contributions | |
day | +0.019 |
end | +0.016 |
good | +0.016 |
team | +0.015 |
, | +0.014 |
Neg logit contributions | |
iqu | -0.073 |
ridge | -0.071 |
umes | -0.069 |
yles | -0.069 |
angs | -0.068 |
Token | smiled |
Token ID | 13541 |
Feature activation | +4.33e-02 |
Pos logit contributions | |
day | +0.001 |
, | +0.001 |
good | +0.001 |
and | +0.001 |
girl | +0.001 |
Neg logit contributions | |
iqu | -0.004 |
ridge | -0.004 |
umes | -0.004 |
yles | -0.004 |
rol | -0.004 |
Token | when |
Token ID | 618 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
day | +0.044 |
afternoon | +0.037 |
team | +0.036 |
time | +0.034 |
good | +0.032 |
Neg logit contributions | |
iqu | -0.130 |
angs | -0.120 |
ridge | -0.117 |
yles | -0.117 |
vell | -0.116 |
Token | he |
Token ID | 339 |
Feature activation | +7.38e-03 |
Pos logit contributions | |
day | +0.016 |
team | +0.014 |
good | +0.014 |
afternoon | +0.014 |
time | +0.014 |
Neg logit contributions | |
iqu | -0.025 |
ridge | -0.024 |
yles | -0.024 |
umes | -0.023 |
rings | -0.023 |
Token | saw |
Token ID | 2497 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
, | +0.007 |
and | +0.007 |
day | +0.007 |
good | +0.006 |
end | +0.006 |
Neg logit contributions | |
iqu | -0.024 |
ridge | -0.023 |
umes | -0.022 |
rol | -0.022 |
angs | -0.022 |
Token | Tom |
Token ID | 4186 |
Feature activation | -5.01e-04 |
Pos logit contributions | |
iqu | +0.050 |
ridge | +0.049 |
umes | +0.047 |
yles | +0.047 |
rol | +0.047 |
Neg logit contributions | |
day | -0.010 |
, | -0.009 |
good | -0.009 |
their | -0.008 |
and | -0.008 |
Token | in |
Token ID | 287 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
iqu | +0.002 |
ridge | +0.002 |
umes | +0.002 |
yles | +0.002 |
rol | +0.002 |
Neg logit contributions | |
day | -0.000 |
, | -0.000 |
good | -0.000 |
and | -0.000 |
time | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | +3.58e-02 |
Pos logit contributions | |
day | +0.013 |
good | +0.011 |
afternoon | +0.011 |
time | +0.010 |
end | +0.010 |
Neg logit contributions | |
iqu | -0.055 |
ridge | -0.053 |
umes | -0.052 |
yles | -0.052 |
angs | -0.052 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
day | +0.024 |
end | +0.022 |
good | +0.020 |
team | +0.019 |
, | +0.019 |
Neg logit contributions | |
iqu | -0.117 |
yles | -0.113 |
ridge | -0.112 |
umes | -0.111 |
angs | -0.110 |
Token | Tom |
Token ID | 13787 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
good | +0.027 |
day | +0.026 |
girl | +0.026 |
end | +0.026 |
team | +0.026 |
Neg logit contributions | |
ridge | -0.047 |
yles | -0.045 |
rol | -0.045 |
angs | -0.044 |
iqu | -0.044 |
Token | loves |
Token ID | 10408 |
Feature activation | +5.53e-03 |
Pos logit contributions | |
team | +0.052 |
good | +0.050 |
day | +0.048 |
new | +0.047 |
girl | +0.047 |
Neg logit contributions | |
ridge | -0.052 |
pped | -0.047 |
angs | -0.047 |
iqu | -0.047 |
pled | -0.046 |
Token | to |
Token ID | 284 |
Feature activation | -1.48e-03 |
Pos logit contributions | |
day | +0.006 |
good | +0.006 |
, | +0.006 |
end | +0.006 |
time | +0.006 |
Neg logit contributions | |
iqu | -0.016 |
ridge | -0.015 |
yles | -0.015 |
umes | -0.015 |
rol | -0.014 |
Token | play |
Token ID | 711 |
Feature activation | +4.57e-02 |
Pos logit contributions | |
iqu | +0.004 |
ridge | +0.004 |
yles | +0.004 |
umes | +0.004 |
angs | +0.004 |
Neg logit contributions | |
day | -0.002 |
good | -0.002 |
girl | -0.002 |
, | -0.002 |
end | -0.002 |
Token | ball |
Token ID | 2613 |
Feature activation | +2.77e-03 |
Pos logit contributions | |
day | +0.055 |
girl | +0.050 |
moment | +0.050 |
good | +0.048 |
afternoon | +0.046 |
Neg logit contributions | |
ridge | -0.118 |
iqu | -0.111 |
ocol | -0.111 |
vell | -0.110 |
bar | -0.110 |
Token | with |
Token ID | 351 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
good | +0.002 |
day | +0.002 |
girl | +0.002 |
moment | +0.002 |
time | +0.002 |
Neg logit contributions | |
iqu | -0.009 |
ridge | -0.009 |
bar | -0.008 |
rol | -0.008 |
angs | -0.008 |
Token | his |
Token ID | 465 |
Feature activation | +6.76e-03 |
Pos logit contributions | |
day | +0.013 |
girl | +0.012 |
good | +0.011 |
end | +0.011 |
moment | +0.011 |
Neg logit contributions | |
iqu | -0.036 |
ridge | -0.034 |
yles | -0.034 |
ocol | -0.034 |
angs | -0.033 |
Token | dad |
Token ID | 9955 |
Feature activation | +1.45e-02 |
Pos logit contributions | |
day | +0.007 |
good | +0.007 |
time | +0.007 |
end | +0.007 |
, | +0.006 |
Neg logit contributions | |
iqu | -0.018 |
yles | -0.017 |
umes | -0.017 |
bar | -0.017 |
rings | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | -4.56e-03 |
Pos logit contributions | |
girl | +0.011 |
good | +0.011 |
day | +0.010 |
boy | +0.010 |
moment | +0.009 |
Neg logit contributions | |
ridge | -0.042 |
bar | -0.040 |
umes | -0.039 |
rings | -0.039 |
iqu | -0.038 |
Token | felt |
Token ID | 2936 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
day | +0.040 |
good | +0.036 |
time | +0.035 |
end | +0.035 |
team | +0.034 |
Neg logit contributions | |
iqu | -0.090 |
ridge | -0.089 |
umes | -0.083 |
yles | -0.083 |
bar | -0.082 |
Token | warm |
Token ID | 5814 |
Feature activation | +3.93e-02 |
Pos logit contributions | |
day | +0.017 |
team | +0.016 |
girl | +0.016 |
boy | +0.015 |
, | +0.014 |
Neg logit contributions | |
iqu | -0.042 |
ridge | -0.041 |
rol | -0.040 |
umes | -0.039 |
yles | -0.039 |
Token | and |
Token ID | 290 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
day | +0.016 |
good | +0.012 |
team | +0.010 |
end | +0.010 |
time | +0.010 |
Neg logit contributions | |
iqu | -0.144 |
ridge | -0.142 |
umes | -0.136 |
yles | -0.136 |
rol | -0.135 |
Token | safe |
Token ID | 3338 |
Feature activation | +4.26e-02 |
Pos logit contributions | |
day | +0.009 |
good | +0.008 |
, | +0.008 |
time | +0.007 |
girl | +0.007 |
Neg logit contributions | |
iqu | -0.033 |
ridge | -0.032 |
umes | -0.031 |
yles | -0.031 |
rol | -0.031 |
Token | once |
Token ID | 1752 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
day | +0.032 |
good | +0.028 |
girl | +0.026 |
time | +0.026 |
end | +0.026 |
Neg logit contributions | |
iqu | -0.140 |
ridge | -0.137 |
umes | -0.132 |
yles | -0.131 |
rol | -0.131 |
Token | again |
Token ID | 757 |
Feature activation | +5.16e-02 |
Pos logit contributions | |
day | +0.032 |
team | +0.032 |
They | +0.031 |
girl | +0.030 |
good | +0.029 |
Neg logit contributions | |
iqu | -0.066 |
ridge | -0.064 |
rol | -0.061 |
bar | -0.061 |
yles | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
day | +0.034 |
good | +0.029 |
girl | +0.028 |
end | +0.026 |
team | +0.025 |
Neg logit contributions | |
iqu | -0.173 |
ridge | -0.170 |
umes | -0.162 |
yles | -0.162 |
bar | -0.159 |
Token | |
Token ID | 220 |
Feature activation | +8.28e-04 |
Pos logit contributions | |
girl | +0.021 |
day | +0.020 |
good | +0.020 |
end | +0.019 |
team | +0.018 |
Neg logit contributions | |
iqu | -0.038 |
ridge | -0.037 |
yles | -0.036 |
angs | -0.034 |
rol | -0.034 |
Token |
|
Token ID | 198 |
Feature activation | +3.73e-03 |
Pos logit contributions | |
day | +0.000 |
good | +0.000 |
, | +0.000 |
end | +0.000 |
girl | +0.000 |
Neg logit contributions | |
iqu | -0.003 |
ridge | -0.003 |
umes | -0.003 |
yles | -0.003 |
rol | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -7.99e-03 |
Pos logit contributions | |
girl | +0.003 |
day | +0.003 |
good | +0.003 |
boy | +0.003 |
team | +0.002 |
Neg logit contributions | |
iqu | -0.011 |
ridge | -0.011 |
yles | -0.011 |
rol | -0.011 |
umes | -0.011 |
Token | Jack |
Token ID | 14295 |
Feature activation | +4.92e-02 |
Pos logit contributions | |
iqu | +0.017 |
ridge | +0.017 |
yles | +0.016 |
pped | +0.016 |
rol | +0.016 |
Neg logit contributions | |
girl | -0.013 |
day | -0.013 |
, | -0.013 |
good | -0.013 |
boy | -0.013 |
Token | And |
Token ID | 843 |
Feature activation | +1.89e-03 |
Pos logit contributions | |
day | +0.025 |
good | +0.023 |
, | +0.023 |
girl | +0.022 |
time | +0.022 |
Neg logit contributions | |
iqu | -0.057 |
ridge | -0.055 |
umes | -0.052 |
yles | -0.052 |
rol | -0.052 |
Token | we |
Token ID | 356 |
Feature activation | +8.04e-03 |
Pos logit contributions | |
day | +0.003 |
good | +0.003 |
, | +0.003 |
time | +0.003 |
end | +0.003 |
Neg logit contributions | |
iqu | -0.005 |
ridge | -0.005 |
umes | -0.005 |
yles | -0.005 |
rol | -0.005 |
Token | have |
Token ID | 423 |
Feature activation | +5.72e-03 |
Pos logit contributions | |
day | +0.009 |
, | +0.008 |
good | +0.008 |
end | +0.008 |
time | +0.008 |
Neg logit contributions | |
iqu | -0.025 |
ridge | -0.024 |
yles | -0.023 |
umes | -0.023 |
rol | -0.023 |
Token | to |
Token ID | 284 |
Feature activation | -9.40e-03 |
Pos logit contributions | |
day | +0.007 |
girl | +0.007 |
end | +0.006 |
, | +0.006 |
afternoon | +0.006 |
Neg logit contributions | |
iqu | -0.016 |
ridge | -0.015 |
umes | -0.015 |
angs | -0.015 |
yles | -0.015 |
Token | be |
Token ID | 307 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
iqu | +0.026 |
ridge | +0.026 |
umes | +0.024 |
bar | +0.024 |
yles | +0.024 |
Neg logit contributions | |
day | -0.011 |
girl | -0.010 |
boy | -0.009 |
good | -0.009 |
, | -0.009 |
Token | quiet |
Token ID | 5897 |
Feature activation | +5.40e-02 |
Pos logit contributions | |
day | +0.025 |
moment | +0.022 |
end | +0.022 |
girl | +0.022 |
team | +0.022 |
Neg logit contributions | |
cher | -0.037 |
iqu | -0.036 |
eking | -0.035 |
umes | -0.034 |
ridge | -0.034 |
Token | ." |
Token ID | 526 |
Feature activation | +3.94e-02 |
Pos logit contributions | |
day | +0.042 |
girl | +0.036 |
good | +0.034 |
team | +0.034 |
end | +0.034 |
Neg logit contributions | |
iqu | -0.167 |
ridge | -0.165 |
umes | -0.162 |
angs | -0.160 |
yles | -0.160 |
Token |
|
Token ID | 198 |
Feature activation | +2.04e-02 |
Pos logit contributions | |
day | +0.041 |
end | +0.037 |
afternoon | +0.037 |
team | +0.036 |
good | +0.033 |
Neg logit contributions | |
iqu | -0.108 |
yles | -0.105 |
ridge | -0.104 |
rings | -0.103 |
paste | -0.102 |
Token |
|
Token ID | 198 |
Feature activation | +5.88e-03 |
Pos logit contributions | |
day | +0.016 |
team | +0.014 |
afternoon | +0.013 |
good | +0.013 |
girl | +0.013 |
Neg logit contributions | |
iqu | -0.063 |
ridge | -0.059 |
yles | -0.059 |
angs | -0.059 |
bar | -0.058 |
Token | They |
Token ID | 2990 |
Feature activation | +3.92e-02 |
Pos logit contributions | |
day | +0.010 |
time | +0.009 |
team | +0.009 |
good | +0.008 |
, | +0.008 |
Neg logit contributions | |
bar | -0.013 |
iqu | -0.012 |
cher | -0.012 |
ents | -0.012 |
pped | -0.012 |
Token | ran |
Token ID | 4966 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
day | +0.058 |
good | +0.053 |
, | +0.051 |
time | +0.050 |
girl | +0.050 |
Neg logit contributions | |
iqu | -0.103 |
ridge | -0.099 |
umes | -0.095 |
yles | -0.094 |
rol | -0.094 |
Token | day |
Token ID | 1110 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
They | +0.014 |
town | +0.012 |
team | +0.011 |
girl | +0.011 |
good | +0.011 |
Neg logit contributions | |
ous | -0.064 |
angs | -0.064 |
iqu | -0.064 |
itten | -0.061 |
ker | -0.061 |
Token | , |
Token ID | 11 |
Feature activation | -3.85e-02 |
Pos logit contributions | |
iqu | +0.051 |
ridge | +0.049 |
angs | +0.048 |
yles | +0.048 |
rol | +0.048 |
Neg logit contributions | |
girl | -0.008 |
day | -0.007 |
end | -0.007 |
team | -0.006 |
afternoon | -0.006 |
Token | Bob |
Token ID | 5811 |
Feature activation | -5.87e-03 |
Pos logit contributions | |
iqu | +0.070 |
ridge | +0.067 |
yles | +0.066 |
rol | +0.064 |
umes | +0.062 |
Neg logit contributions | |
end | -0.072 |
day | -0.072 |
girl | -0.071 |
team | -0.069 |
afternoon | -0.069 |
Token | o |
Token ID | 78 |
Feature activation | -3.85e-03 |
Pos logit contributions | |
iqu | +0.009 |
ridge | +0.009 |
angs | +0.008 |
umes | +0.008 |
rol | +0.008 |
Neg logit contributions | |
day | -0.014 |
good | -0.014 |
end | -0.014 |
girl | -0.013 |
team | -0.013 |
Token | woke |
Token ID | 19092 |
Feature activation | -4.45e-02 |
Pos logit contributions | |
ridge | +0.009 |
iqu | +0.008 |
yles | +0.008 |
umes | +0.008 |
pped | +0.008 |
Neg logit contributions | |
good | -0.006 |
day | -0.005 |
end | -0.005 |
afternoon | -0.005 |
their | -0.005 |
Token | up |
Token ID | 510 |
Feature activation | +4.42e-02 |
Pos logit contributions | |
iqu | +0.130 |
ridge | +0.127 |
umes | +0.123 |
yles | +0.121 |
rol | +0.121 |
Neg logit contributions | |
day | -0.054 |
, | -0.050 |
good | -0.050 |
and | -0.047 |
time | -0.047 |
Token | early |
Token ID | 1903 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
day | +0.036 |
girl | +0.034 |
team | +0.033 |
good | +0.033 |
end | +0.032 |
Neg logit contributions | |
iqu | -0.138 |
ridge | -0.131 |
umes | -0.127 |
yles | -0.127 |
angs | -0.126 |
Token | . |
Token ID | 13 |
Feature activation | -2.68e-02 |
Pos logit contributions | |
good | +0.018 |
girl | +0.017 |
end | +0.017 |
day | +0.016 |
team | +0.016 |
Neg logit contributions | |
ridge | -0.073 |
iqu | -0.073 |
umes | -0.070 |
yles | -0.069 |
bar | -0.069 |
Token | He |
Token ID | 679 |
Feature activation | +4.95e-03 |
Pos logit contributions | |
iqu | +0.078 |
ridge | +0.076 |
angs | +0.073 |
rol | +0.072 |
umes | +0.072 |
Neg logit contributions | |
day | -0.030 |
end | -0.028 |
good | -0.028 |
girl | -0.026 |
, | -0.026 |
Token | saw |
Token ID | 2497 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
day | +0.006 |
good | +0.006 |
end | +0.006 |
team | +0.006 |
, | +0.005 |
Neg logit contributions | |
iqu | -0.013 |
ridge | -0.013 |
yles | -0.013 |
umes | -0.012 |
rol | -0.012 |
Token | a |
Token ID | 257 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
iqu | +0.112 |
ridge | +0.109 |
umes | +0.107 |
yles | +0.105 |
rol | +0.104 |
Neg logit contributions | |
day | -0.016 |
good | -0.014 |
, | -0.013 |
their | -0.011 |
and | -0.011 |
Token | 's |
Token ID | 338 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
iqu | +0.030 |
ridge | +0.029 |
umes | +0.028 |
yles | +0.028 |
rol | +0.028 |
Neg logit contributions | |
day | -0.009 |
, | -0.008 |
good | -0.008 |
time | -0.008 |
and | -0.008 |
Token | coming |
Token ID | 2406 |
Feature activation | -2.39e-02 |
Pos logit contributions | |
iqu | +0.053 |
ridge | +0.052 |
yles | +0.050 |
umes | +0.049 |
rol | +0.049 |
Neg logit contributions | |
day | -0.028 |
good | -0.026 |
, | -0.026 |
time | -0.025 |
and | -0.025 |
Token | to |
Token ID | 284 |
Feature activation | -2.51e-02 |
Pos logit contributions | |
iqu | +0.075 |
ridge | +0.073 |
yles | +0.070 |
umes | +0.069 |
rol | +0.068 |
Neg logit contributions | |
day | -0.026 |
, | -0.025 |
and | -0.025 |
good | -0.023 |
girl | -0.022 |
Token | visit |
Token ID | 3187 |
Feature activation | +1.06e-03 |
Pos logit contributions | |
iqu | +0.083 |
ridge | +0.079 |
umes | +0.078 |
rol | +0.077 |
yles | +0.076 |
Neg logit contributions | |
day | -0.024 |
good | -0.022 |
, | -0.021 |
their | -0.021 |
end | -0.020 |
Token | us |
Token ID | 514 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
day | +0.001 |
good | +0.001 |
end | +0.001 |
team | +0.001 |
, | +0.001 |
Neg logit contributions | |
iqu | -0.003 |
ridge | -0.003 |
umes | -0.003 |
yles | -0.003 |
rol | -0.003 |
Token | tomorrow |
Token ID | 9439 |
Feature activation | +3.05e-02 |
Pos logit contributions | |
day | +0.020 |
good | +0.019 |
, | +0.018 |
time | +0.018 |
end | +0.018 |
Neg logit contributions | |
iqu | -0.050 |
ridge | -0.048 |
umes | -0.047 |
yles | -0.046 |
rol | -0.046 |
Token | and |
Token ID | 290 |
Feature activation | -1.26e-03 |
Pos logit contributions | |
day | +0.037 |
good | +0.034 |
end | +0.033 |
, | +0.033 |
girl | +0.032 |
Neg logit contributions | |
iqu | -0.087 |
ridge | -0.084 |
umes | -0.081 |
yles | -0.080 |
rol | -0.079 |
Token | we |
Token ID | 356 |
Feature activation | -9.14e-03 |
Pos logit contributions | |
iqu | +0.004 |
ridge | +0.003 |
umes | +0.003 |
rol | +0.003 |
yles | +0.003 |
Neg logit contributions | |
day | -0.002 |
, | -0.002 |
good | -0.002 |
girl | -0.002 |
and | -0.002 |
Token | need |
Token ID | 761 |
Feature activation | -1.97e-02 |
Pos logit contributions | |
iqu | +0.025 |
ridge | +0.024 |
umes | +0.023 |
yles | +0.023 |
angs | +0.022 |
Neg logit contributions | |
day | -0.013 |
, | -0.013 |
good | -0.012 |
end | -0.012 |
time | -0.012 |
Token | to |
Token ID | 284 |
Feature activation | -2.00e-02 |
Pos logit contributions | |
iqu | +0.063 |
ridge | +0.061 |
umes | +0.059 |
yles | +0.059 |
rol | +0.058 |
Neg logit contributions | |
day | -0.019 |
good | -0.017 |
, | -0.017 |
time | -0.016 |
and | -0.016 |
Token | pick |
Token ID | 2298 |
Feature activation | -7.27e-03 |
Pos logit contributions | |
iqu | +0.061 |
ridge | +0.058 |
yles | +0.057 |
umes | +0.057 |
rol | +0.056 |
Neg logit contributions | |
day | -0.024 |
, | -0.023 |
good | -0.023 |
time | -0.022 |
end | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | -2.40e-02 |
Pos logit contributions | |
iqu | +0.001 |
ridge | +0.001 |
umes | +0.001 |
yles | +0.001 |
rol | +0.001 |
Neg logit contributions | |
day | -0.000 |
good | -0.000 |
, | -0.000 |
time | -0.000 |
and | -0.000 |
Token | give |
Token ID | 1577 |
Feature activation | -2.04e-02 |
Pos logit contributions | |
iqu | +0.068 |
ridge | +0.066 |
umes | +0.063 |
yles | +0.063 |
rol | +0.062 |
Neg logit contributions | |
day | -0.032 |
good | -0.029 |
, | -0.029 |
end | -0.028 |
time | -0.028 |
Token | Sam |
Token ID | 3409 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
iqu | +0.059 |
ridge | +0.057 |
umes | +0.055 |
yles | +0.054 |
rol | +0.054 |
Neg logit contributions | |
day | -0.026 |
good | -0.024 |
, | -0.024 |
time | -0.022 |
girl | -0.022 |
Token | any |
Token ID | 597 |
Feature activation | +6.67e-03 |
Pos logit contributions | |
day | +0.012 |
girl | +0.010 |
afternoon | +0.010 |
town | +0.010 |
team | +0.010 |
Neg logit contributions | |
iqu | -0.028 |
angs | -0.028 |
bar | -0.026 |
ridge | -0.026 |
rings | -0.026 |
Token | of |
Token ID | 286 |
Feature activation | +1.92e-03 |
Pos logit contributions | |
day | +0.011 |
good | +0.010 |
, | +0.010 |
girl | +0.010 |
end | +0.010 |
Neg logit contributions | |
iqu | -0.016 |
ridge | -0.016 |
umes | -0.015 |
angs | -0.015 |
yles | -0.015 |
Token | his |
Token ID | 465 |
Feature activation | -5.73e-03 |
Pos logit contributions | |
day | +0.002 |
, | +0.002 |
good | +0.002 |
time | +0.001 |
end | +0.001 |
Neg logit contributions | |
iqu | -0.006 |
ridge | -0.006 |
umes | -0.006 |
yles | -0.006 |
rol | -0.006 |
Token | meat |
Token ID | 6174 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
iqu | +0.011 |
ridge | +0.011 |
angs | +0.010 |
ocol | +0.010 |
umes | +0.010 |
Neg logit contributions | |
day | -0.011 |
girl | -0.011 |
, | -0.011 |
good | -0.011 |
their | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +9.00e-03 |
Pos logit contributions | |
iqu | +0.064 |
ridge | +0.063 |
umes | +0.061 |
angs | +0.061 |
yles | +0.060 |
Neg logit contributions | |
day | -0.012 |
good | -0.009 |
girl | -0.009 |
time | -0.009 |
end | -0.008 |
Token | He |
Token ID | 679 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
day | +0.009 |
good | +0.008 |
end | +0.008 |
girl | +0.008 |
, | +0.008 |
Neg logit contributions | |
iqu | -0.028 |
ridge | -0.026 |
umes | -0.025 |
yles | -0.025 |
angs | -0.025 |
Token | thinks |
Token ID | 6834 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
day | +0.017 |
good | +0.015 |
, | +0.015 |
girl | +0.015 |
team | +0.015 |
Neg logit contributions | |
iqu | -0.034 |
ridge | -0.032 |
yles | -0.030 |
umes | -0.030 |
angs | -0.030 |
Token | Sam |
Token ID | 3409 |
Feature activation | -2.62e-03 |
Pos logit contributions | |
iqu | +0.070 |
ridge | +0.067 |
umes | +0.065 |
yles | +0.064 |
rol | +0.064 |
Neg logit contributions | |
day | -0.030 |
good | -0.028 |
, | -0.027 |
time | -0.026 |
end | -0.026 |
Token | put |
Token ID | 1234 |
Feature activation | +7.10e-03 |
Pos logit contributions | |
day | +0.037 |
good | +0.035 |
time | +0.032 |
girl | +0.032 |
end | +0.031 |
Neg logit contributions | |
ridge | -0.098 |
iqu | -0.096 |
bar | -0.091 |
rol | -0.090 |
umes | -0.090 |
Token | the |
Token ID | 262 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
day | +0.006 |
team | +0.005 |
girl | +0.005 |
good | +0.005 |
end | +0.004 |
Neg logit contributions | |
iqu | -0.022 |
ridge | -0.022 |
yles | -0.021 |
rol | -0.021 |
umes | -0.021 |
Token | stones |
Token ID | 14966 |
Feature activation | -5.69e-04 |
Pos logit contributions | |
iqu | +0.021 |
vell | +0.021 |
umes | +0.020 |
ridge | +0.020 |
bar | +0.019 |
Neg logit contributions | |
day | -0.020 |
They | -0.019 |
their | -0.019 |
Wednesday | -0.019 |
Sunday | -0.018 |
Token | together |
Token ID | 1978 |
Feature activation | +4.16e-02 |
Pos logit contributions | |
iqu | +0.002 |
ridge | +0.002 |
bar | +0.002 |
umes | +0.002 |
rol | +0.002 |
Neg logit contributions | |
day | -0.001 |
good | -0.001 |
girl | -0.001 |
time | -0.001 |
team | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | +6.07e-03 |
Pos logit contributions | |
day | +0.026 |
girl | +0.025 |
good | +0.023 |
boy | +0.022 |
team | +0.021 |
Neg logit contributions | |
iqu | -0.136 |
ridge | -0.135 |
umes | -0.130 |
yles | -0.129 |
rol | -0.129 |
Token | suddenly |
Token ID | 6451 |
Feature activation | -6.89e-03 |
Pos logit contributions | |
day | +0.008 |
good | +0.007 |
, | +0.007 |
time | +0.007 |
end | +0.007 |
Neg logit contributions | |
iqu | -0.017 |
ridge | -0.017 |
umes | -0.016 |
yles | -0.016 |
rol | -0.016 |
Token | , |
Token ID | 11 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
iqu | +0.019 |
ridge | +0.019 |
yles | +0.018 |
bar | +0.018 |
umes | +0.018 |
Neg logit contributions | |
day | -0.008 |
girl | -0.007 |
good | -0.007 |
afternoon | -0.007 |
end | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | -9.66e-03 |
Pos logit contributions | |
iqu | +0.033 |
ridge | +0.032 |
umes | +0.031 |
yles | +0.031 |
rol | +0.030 |
Neg logit contributions | |
day | -0.011 |
good | -0.010 |
, | -0.010 |
end | -0.009 |
time | -0.009 |
Token | forest |
Token ID | 8222 |
Feature activation | -2.60e-02 |
Pos logit contributions | |
iqu | +0.022 |
ridge | +0.021 |
bar | +0.021 |
vell | +0.021 |
umes | +0.020 |
Neg logit contributions | |
day | -0.015 |
their | -0.015 |
, | -0.014 |
good | -0.014 |
They | -0.014 |
Token | lit |
Token ID | 6578 |
Feature activation | -9.86e-03 |
Pos logit contributions | |
iqu | +0.072 |
ridge | +0.069 |
umes | +0.067 |
rol | +0.066 |
yles | +0.066 |
Neg logit contributions | |
day | -0.034 |
good | -0.031 |
girl | -0.030 |
, | -0.029 |
time | -0.029 |
Token | up |
Token ID | 510 |
Feature activation | +5.16e-02 |
Pos logit contributions | |
iqu | +0.027 |
ridge | +0.026 |
umes | +0.025 |
yles | +0.025 |
rol | +0.025 |
Neg logit contributions | |
day | -0.013 |
good | -0.012 |
girl | -0.012 |
, | -0.012 |
end | -0.011 |
Token | went |
Token ID | 1816 |
Feature activation | +5.18e-03 |
Pos logit contributions | |
iqu | +0.076 |
ridge | +0.073 |
umes | +0.070 |
yles | +0.070 |
rol | +0.069 |
Neg logit contributions | |
day | -0.031 |
good | -0.028 |
, | -0.027 |
girl | -0.026 |
end | -0.026 |
Token | away |
Token ID | 1497 |
Feature activation | +2.07e-02 |
Pos logit contributions | |
day | +0.006 |
, | +0.005 |
good | +0.005 |
time | +0.005 |
and | +0.005 |
Neg logit contributions | |
iqu | -0.016 |
ridge | -0.016 |
yles | -0.015 |
umes | -0.015 |
rol | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | -5.56e-04 |
Pos logit contributions | |
day | +0.016 |
team | +0.014 |
girl | +0.013 |
afternoon | +0.013 |
end | +0.013 |
Neg logit contributions | |
iqu | -0.066 |
ridge | -0.063 |
umes | -0.062 |
ous | -0.060 |
vell | -0.060 |
Token | and |
Token ID | 290 |
Feature activation | -1.92e-03 |
Pos logit contributions | |
iqu | +0.002 |
ridge | +0.002 |
umes | +0.002 |
yles | +0.002 |
rol | +0.002 |
Neg logit contributions | |
day | -0.001 |
good | -0.000 |
, | -0.000 |
end | -0.000 |
time | -0.000 |
Token | she |
Token ID | 673 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
iqu | +0.005 |
ridge | +0.005 |
umes | +0.005 |
yles | +0.005 |
angs | +0.005 |
Neg logit contributions | |
day | -0.003 |
, | -0.003 |
good | -0.003 |
and | -0.003 |
girl | -0.002 |
Token | had |
Token ID | 550 |
Feature activation | -1.84e-03 |
Pos logit contributions | |
day | +0.033 |
good | +0.030 |
time | +0.029 |
girl | +0.029 |
end | +0.028 |
Neg logit contributions | |
ridge | -0.072 |
iqu | -0.071 |
yles | -0.066 |
umes | -0.066 |
rol | -0.065 |
Token | more |
Token ID | 517 |
Feature activation | +3.48e-02 |
Pos logit contributions | |
iqu | +0.005 |
ridge | +0.005 |
umes | +0.005 |
yles | +0.005 |
rol | +0.005 |
Neg logit contributions | |
day | -0.002 |
girl | -0.002 |
team | -0.002 |
end | -0.002 |
boy | -0.002 |
Token | energy |
Token ID | 2568 |
Feature activation | +2.78e-02 |
Pos logit contributions | |
day | +0.048 |
team | +0.045 |
end | +0.044 |
girl | +0.044 |
, | +0.044 |
Neg logit contributions | |
iqu | -0.093 |
ridge | -0.089 |
yles | -0.083 |
umes | -0.083 |
angs | -0.083 |
Token | to |
Token ID | 284 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
day | +0.027 |
good | +0.024 |
, | +0.024 |
time | +0.023 |
girl | +0.023 |
Neg logit contributions | |
iqu | -0.087 |
ridge | -0.084 |
umes | -0.082 |
yles | -0.081 |
rol | -0.080 |
Token | play |
Token ID | 711 |
Feature activation | +4.77e-02 |
Pos logit contributions | |
iqu | +0.038 |
ridge | +0.037 |
umes | +0.036 |
yles | +0.035 |
rol | +0.035 |
Neg logit contributions | |
day | -0.017 |
good | -0.015 |
, | -0.015 |
girl | -0.015 |
time | -0.015 |
Token | with |
Token ID | 351 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
day | +0.052 |
girl | +0.046 |
end | +0.044 |
boy | +0.043 |
good | +0.042 |
Neg logit contributions | |
ridge | -0.132 |
iqu | -0.126 |
yles | -0.126 |
umes | -0.123 |
paste | -0.123 |
Token | and |
Token ID | 290 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
iqu | +0.027 |
ridge | +0.026 |
umes | +0.026 |
yles | +0.026 |
angs | +0.025 |
Neg logit contributions | |
day | -0.002 |
, | -0.001 |
and | -0.001 |
good | -0.001 |
time | -0.000 |
Token | th |
Token ID | 294 |
Feature activation | -6.16e-02 |
Pos logit contributions | |
iqu | +0.098 |
ridge | +0.094 |
umes | +0.091 |
yles | +0.090 |
angs | +0.089 |
Neg logit contributions | |
day | -0.041 |
, | -0.038 |
good | -0.037 |
girl | -0.036 |
and | -0.035 |
Token | orns |
Token ID | 19942 |
Feature activation | -1.76e-02 |
Pos logit contributions | |
iqu | +0.113 |
ridge | +0.102 |
umes | +0.102 |
vell | +0.098 |
cher | +0.096 |
Neg logit contributions | |
day | -0.140 |
, | -0.140 |
and | -0.138 |
time | -0.134 |
good | -0.133 |
Token | , |
Token ID | 11 |
Feature activation | -3.85e-02 |
Pos logit contributions | |
iqu | +0.063 |
ridge | +0.061 |
yles | +0.060 |
umes | +0.059 |
ocol | +0.058 |
Neg logit contributions | |
day | -0.011 |
, | -0.011 |
and | -0.009 |
good | -0.008 |
time | -0.008 |
Token | but |
Token ID | 475 |
Feature activation | -1.99e-02 |
Pos logit contributions | |
iqu | +0.123 |
ridge | +0.118 |
umes | +0.114 |
yles | +0.114 |
angs | +0.113 |
Neg logit contributions | |
day | -0.038 |
, | -0.036 |
and | -0.035 |
good | -0.035 |
time | -0.032 |
Token | she |
Token ID | 673 |
Feature activation | -5.27e-03 |
Pos logit contributions | |
iqu | +0.060 |
ridge | +0.058 |
angs | +0.055 |
umes | +0.055 |
rol | +0.054 |
Neg logit contributions | |
day | -0.025 |
, | -0.023 |
good | -0.023 |
and | -0.021 |
their | -0.021 |
Token | never |
Token ID | 1239 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
iqu | +0.016 |
ridge | +0.015 |
umes | +0.015 |
yles | +0.015 |
rol | +0.014 |
Neg logit contributions | |
day | -0.006 |
, | -0.006 |
good | -0.006 |
and | -0.006 |
time | -0.005 |
Token | gave |
Token ID | 2921 |
Feature activation | -3.50e-02 |
Pos logit contributions | |
iqu | +0.083 |
ridge | +0.080 |
umes | +0.077 |
yles | +0.077 |
rol | +0.076 |
Neg logit contributions | |
day | -0.041 |
, | -0.039 |
good | -0.038 |
time | -0.036 |
girl | -0.036 |
Token | up |
Token ID | 510 |
Feature activation | +3.87e-02 |
Pos logit contributions | |
iqu | +0.107 |
ridge | +0.105 |
umes | +0.100 |
yles | +0.099 |
rol | +0.099 |
Neg logit contributions | |
day | -0.037 |
good | -0.036 |
, | -0.035 |
and | -0.033 |
time | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -3.90e-03 |
Pos logit contributions | |
day | +0.024 |
good | +0.020 |
, | +0.019 |
girl | +0.018 |
end | +0.018 |
Neg logit contributions | |
iqu | -0.134 |
ridge | -0.130 |
umes | -0.127 |
yles | -0.126 |
rol | -0.124 |
Token | After |
Token ID | 2293 |
Feature activation | +2.20e-02 |
Pos logit contributions | |
iqu | +0.013 |
ridge | +0.013 |
umes | +0.012 |
yles | +0.012 |
rol | +0.012 |
Neg logit contributions | |
day | -0.003 |
good | -0.003 |
, | -0.002 |
end | -0.002 |
team | -0.002 |
Token | spark |
Token ID | 9009 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
iqu | +0.057 |
ridge | +0.055 |
yles | +0.053 |
rol | +0.052 |
umes | +0.052 |
Neg logit contributions | |
day | -0.023 |
good | -0.022 |
, | -0.022 |
girl | -0.021 |
and | -0.020 |
Token | ly |
Token ID | 306 |
Feature activation | -2.83e-02 |
Pos logit contributions | |
iqu | +0.070 |
ridge | +0.068 |
yles | +0.065 |
umes | +0.065 |
rol | +0.064 |
Neg logit contributions | |
day | -0.034 |
, | -0.032 |
good | -0.031 |
and | -0.030 |
girl | -0.030 |
Token | cape |
Token ID | 36023 |
Feature activation | -2.47e-02 |
Pos logit contributions | |
iqu | +0.069 |
ridge | +0.066 |
umes | +0.063 |
yles | +0.062 |
rol | +0.061 |
Neg logit contributions | |
day | -0.048 |
good | -0.045 |
, | -0.045 |
time | -0.043 |
girl | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | -9.26e-03 |
Pos logit contributions | |
iqu | +0.096 |
ridge | +0.092 |
yles | +0.090 |
ocol | +0.089 |
umes | +0.088 |
Neg logit contributions | |
, | -0.013 |
and | -0.013 |
day | -0.011 |
. | -0.007 |
good | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
iqu | +0.026 |
ridge | +0.026 |
yles | +0.025 |
umes | +0.025 |
rol | +0.025 |
Neg logit contributions | |
day | -0.010 |
end | -0.010 |
good | -0.010 |
team | -0.010 |
girl | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
iqu | +0.032 |
yles | +0.032 |
rol | +0.031 |
ridge | +0.031 |
umes | +0.031 |
Neg logit contributions | |
good | -0.007 |
team | -0.006 |
day | -0.006 |
end | -0.006 |
girl | -0.006 |
Token | One |
Token ID | 3198 |
Feature activation | +2.86e-02 |
Pos logit contributions | |
iqu | +0.038 |
rol | +0.037 |
umes | +0.037 |
ridge | +0.036 |
yles | +0.036 |
Neg logit contributions | |
team | -0.034 |
good | -0.032 |
day | -0.032 |
boy | -0.031 |
end | -0.031 |
Token | day |
Token ID | 1110 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
They | +0.016 |
team | +0.012 |
good | +0.011 |
town | +0.011 |
, | +0.011 |
Neg logit contributions | |
iqu | -0.079 |
angs | -0.076 |
ous | -0.075 |
ridge | -0.074 |
bar | -0.074 |
Token | Lucy |
Token ID | 22162 |
Feature activation | +2.14e-02 |
Pos logit contributions | |
iqu | +0.040 |
ridge | +0.040 |
yles | +0.039 |
umes | +0.039 |
rings | +0.038 |
Neg logit contributions | |
team | -0.007 |
day | -0.007 |
end | -0.007 |
girl | -0.006 |
boy | -0.006 |
Token | wanted |
Token ID | 2227 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
day | +0.022 |
good | +0.022 |
end | +0.020 |
team | +0.019 |
time | +0.018 |
Neg logit contributions | |
iqu | -0.063 |
ridge | -0.062 |
yles | -0.060 |
umes | -0.060 |
angs | -0.058 |
Token | to |
Token ID | 284 |
Feature activation | -2.45e-02 |
Pos logit contributions | |
iqu | +0.056 |
ridge | +0.054 |
umes | +0.052 |
rol | +0.052 |
yles | +0.051 |
Neg logit contributions | |
day | -0.016 |
, | -0.014 |
good | -0.014 |
and | -0.013 |
time | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | -1.92e-02 |
Pos logit contributions | |
iqu | +0.038 |
ridge | +0.037 |
yles | +0.037 |
umes | +0.037 |
rol | +0.037 |
Neg logit contributions | |
day | -0.008 |
good | -0.008 |
team | -0.007 |
end | -0.007 |
girl | -0.007 |
Token | The |
Token ID | 464 |
Feature activation | -7.89e-05 |
Pos logit contributions | |
umes | +0.046 |
iqu | +0.046 |
ridge | +0.045 |
bar | +0.043 |
pped | +0.043 |
Neg logit contributions | |
day | -0.029 |
good | -0.027 |
time | -0.025 |
, | -0.025 |
their | -0.024 |
Token | pig |
Token ID | 12967 |
Feature activation | -3.80e-02 |
Pos logit contributions | |
iqu | +0.000 |
ridge | +0.000 |
umes | +0.000 |
bar | +0.000 |
yles | +0.000 |
Neg logit contributions | |
day | -0.000 |
, | -0.000 |
team | -0.000 |
good | -0.000 |
afternoon | -0.000 |
Token | replied |
Token ID | 8712 |
Feature activation | -6.81e-03 |
Pos logit contributions | |
iqu | +0.070 |
ridge | +0.067 |
umes | +0.061 |
bar | +0.060 |
angs | +0.060 |
Neg logit contributions | |
day | -0.085 |
good | -0.083 |
time | -0.081 |
end | -0.080 |
afternoon | -0.078 |
Token | , |
Token ID | 11 |
Feature activation | -1.43e-03 |
Pos logit contributions | |
iqu | +0.023 |
ridge | +0.022 |
umes | +0.021 |
bar | +0.021 |
ous | +0.021 |
Neg logit contributions | |
day | -0.004 |
afternoon | -0.003 |
time | -0.003 |
girl | -0.003 |
end | -0.003 |
Token | " |
Token ID | 366 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
ridge | +0.004 |
iqu | +0.004 |
umes | +0.004 |
yles | +0.004 |
bar | +0.004 |
Neg logit contributions | |
day | -0.002 |
end | -0.001 |
time | -0.001 |
good | -0.001 |
afternoon | -0.001 |
Token | Thank |
Token ID | 10449 |
Feature activation | -1.23e-03 |
Pos logit contributions | |
iqu | +0.040 |
ridge | +0.038 |
umes | +0.037 |
yles | +0.036 |
angs | +0.036 |
Neg logit contributions | |
day | -0.022 |
good | -0.020 |
, | -0.019 |
girl | -0.019 |
end | -0.019 |
Token | you |
Token ID | 345 |
Feature activation | +5.65e-02 |
Pos logit contributions | |
iqu | +0.003 |
ridge | +0.003 |
umes | +0.003 |
yles | +0.003 |
rol | +0.002 |
Neg logit contributions | |
day | -0.002 |
good | -0.002 |
time | -0.002 |
, | -0.002 |
end | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -4.91e-03 |
Pos logit contributions | |
day | +0.029 |
afternoon | +0.021 |
time | +0.021 |
girl | +0.020 |
good | +0.018 |
Neg logit contributions | |
iqu | -0.193 |
ridge | -0.192 |
umes | -0.187 |
angs | -0.184 |
rol | -0.183 |
Token | Lily |
Token ID | 20037 |
Feature activation | +7.45e-02 |
Pos logit contributions | |
iqu | +0.017 |
ridge | +0.016 |
yles | +0.016 |
umes | +0.016 |
rol | +0.016 |
Neg logit contributions | |
day | -0.004 |
good | -0.003 |
, | -0.003 |
girl | -0.003 |
and | -0.003 |
Token | ! |
Token ID | 0 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
day | +0.101 |
team | +0.093 |
time | +0.092 |
end | +0.092 |
good | +0.092 |
Neg logit contributions | |
ridge | -0.182 |
iqu | -0.182 |
angs | -0.169 |
ker | -0.165 |
rou | -0.165 |
Token | She |
Token ID | 1375 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
iqu | +0.018 |
ridge | +0.018 |
umes | +0.017 |
yles | +0.017 |
rol | +0.016 |
Neg logit contributions | |
day | -0.007 |
end | -0.006 |
good | -0.006 |
time | -0.006 |
team | -0.006 |
Token | picked |
Token ID | 6497 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
day | +0.025 |
good | +0.022 |
time | +0.022 |
, | +0.022 |
team | +0.022 |
Neg logit contributions | |
iqu | -0.040 |
ridge | -0.038 |
umes | -0.036 |
yles | -0.036 |
angs | -0.035 |
Token | up |
Token ID | 510 |
Feature activation | +3.09e-02 |
Pos logit contributions | |
iqu | +0.080 |
ridge | +0.076 |
umes | +0.074 |
yles | +0.073 |
vell | +0.072 |
Neg logit contributions | |
day | -0.034 |
good | -0.032 |
, | -0.032 |
and | -0.030 |
their | -0.029 |
Token | Fl |
Token ID | 1610 |
Feature activation | -4.59e-02 |
Pos logit contributions | |
day | +0.036 |
end | +0.031 |
team | +0.031 |
good | +0.031 |
time | +0.030 |
Neg logit contributions | |
iqu | -0.090 |
ridge | -0.088 |
umes | -0.083 |
yles | -0.083 |
rol | -0.082 |
Token | uffy |
Token ID | 15352 |
Feature activation | +4.39e-03 |
Pos logit contributions | |
iqu | +0.091 |
ridge | +0.086 |
umes | +0.081 |
yles | +0.080 |
angs | +0.079 |
Neg logit contributions | |
day | -0.100 |
good | -0.095 |
, | -0.093 |
end | -0.092 |
time | -0.092 |
Token | and |
Token ID | 290 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
day | +0.003 |
time | +0.003 |
good | +0.003 |
girl | +0.003 |
team | +0.003 |
Neg logit contributions | |
ridge | -0.014 |
iqu | -0.014 |
umes | -0.013 |
ker | -0.013 |
bar | -0.013 |
Token | gave |
Token ID | 2921 |
Feature activation | -4.03e-02 |
Pos logit contributions | |
iqu | +0.049 |
ridge | +0.047 |
umes | +0.045 |
yles | +0.045 |
rol | +0.045 |
Neg logit contributions | |
day | -0.026 |
, | -0.025 |
good | -0.024 |
and | -0.023 |
girl | -0.023 |
Token | her |
Token ID | 607 |
Feature activation | +4.05e-03 |
Pos logit contributions | |
iqu | +0.107 |
ridge | +0.101 |
umes | +0.099 |
yles | +0.096 |
vell | +0.096 |
Neg logit contributions | |
day | -0.064 |
good | -0.061 |
, | -0.061 |
their | -0.058 |
and | -0.057 |
Token | a |
Token ID | 257 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
day | +0.004 |
, | +0.003 |
good | +0.003 |
girl | +0.003 |
end | +0.003 |
Neg logit contributions | |
iqu | -0.013 |
ridge | -0.012 |
umes | -0.012 |
angs | -0.012 |
yles | -0.012 |
Token | big |
Token ID | 1263 |
Feature activation | +8.20e-03 |
Pos logit contributions | |
day | +0.010 |
, | +0.009 |
good | +0.009 |
end | +0.009 |
girl | +0.008 |
Neg logit contributions | |
iqu | -0.035 |
ridge | -0.034 |
umes | -0.033 |
angs | -0.032 |
yles | -0.032 |
Token | hug |
Token ID | 16225 |
Feature activation | -1.14e-03 |
Pos logit contributions | |
day | +0.010 |
afternoon | +0.009 |
good | +0.009 |
time | +0.009 |
team | +0.009 |
Neg logit contributions | |
iqu | -0.022 |
ridge | -0.022 |
angs | -0.021 |
bar | -0.020 |
umes | -0.020 |
Token | , |
Token ID | 11 |
Feature activation | -1.92e-02 |
Pos logit contributions | |
day | +0.001 |
good | +0.001 |
, | +0.001 |
girl | +0.001 |
end | +0.001 |
Neg logit contributions | |
iqu | -0.007 |
ridge | -0.007 |
umes | -0.006 |
yles | -0.006 |
angs | -0.006 |
Token | because |
Token ID | 780 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
iqu | +0.057 |
ridge | +0.056 |
umes | +0.053 |
yles | +0.053 |
rol | +0.052 |
Neg logit contributions | |
day | -0.023 |
good | -0.021 |
, | -0.021 |
and | -0.020 |
time | -0.020 |
Token | if |
Token ID | 611 |
Feature activation | -9.11e-03 |
Pos logit contributions | |
iqu | +0.034 |
ridge | +0.033 |
angs | +0.031 |
rol | +0.031 |
umes | +0.031 |
Neg logit contributions | |
day | -0.012 |
, | -0.011 |
good | -0.011 |
and | -0.010 |
their | -0.010 |
Token | they |
Token ID | 484 |
Feature activation | -5.27e-04 |
Pos logit contributions | |
iqu | +0.028 |
ridge | +0.027 |
umes | +0.026 |
rol | +0.026 |
angs | +0.026 |
Neg logit contributions | |
day | -0.010 |
good | -0.009 |
, | -0.009 |
time | -0.009 |
girl | -0.009 |
Token | do |
Token ID | 466 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
iqu | +0.001 |
ridge | +0.001 |
umes | +0.001 |
yles | +0.001 |
rol | +0.001 |
Neg logit contributions | |
day | -0.001 |
good | -0.001 |
, | -0.001 |
time | -0.001 |
end | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | -2.58e-02 |
Pos logit contributions | |
day | +0.024 |
girl | +0.023 |
boy | +0.022 |
team | +0.022 |
afternoon | +0.021 |
Neg logit contributions | |
iqu | -0.045 |
ridge | -0.043 |
umes | -0.042 |
yles | -0.041 |
angs | -0.040 |
Token | they |
Token ID | 484 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
iqu | +0.085 |
ridge | +0.083 |
umes | +0.080 |
yles | +0.080 |
rol | +0.079 |
Neg logit contributions | |
day | -0.020 |
good | -0.017 |
, | -0.017 |
end | -0.016 |
time | -0.016 |
Token | 're |
Token ID | 821 |
Feature activation | -3.13e-03 |
Pos logit contributions | |
day | +0.017 |
good | +0.016 |
, | +0.016 |
end | +0.016 |
time | +0.015 |
Neg logit contributions | |
iqu | -0.031 |
ridge | -0.030 |
umes | -0.029 |
yles | -0.029 |
rol | -0.028 |
Token | more |
Token ID | 517 |
Feature activation | +3.72e-02 |
Pos logit contributions | |
iqu | +0.008 |
ridge | +0.008 |
umes | +0.008 |
yles | +0.008 |
rol | +0.008 |
Neg logit contributions | |
day | -0.004 |
girl | -0.004 |
team | -0.004 |
, | -0.004 |
good | -0.004 |
Token | likely |
Token ID | 1884 |
Feature activation | -1.73e-03 |
Pos logit contributions | |
day | +0.048 |
good | +0.043 |
, | +0.043 |
end | +0.042 |
girl | +0.041 |
Neg logit contributions | |
iqu | -0.107 |
ridge | -0.102 |
umes | -0.098 |
yles | -0.098 |
angs | -0.096 |
Token | to |
Token ID | 284 |
Feature activation | -2.19e-02 |
Pos logit contributions | |
iqu | +0.005 |
ridge | +0.005 |
umes | +0.005 |
yles | +0.005 |
angs | +0.004 |
Neg logit contributions | |
day | -0.002 |
girl | -0.002 |
end | -0.002 |
team | -0.002 |
good | -0.002 |
Token | to |
Token ID | 284 |
Feature activation | -3.37e-02 |
Pos logit contributions | |
good | +0.019 |
day | +0.019 |
girl | +0.018 |
end | +0.018 |
time | +0.017 |
Neg logit contributions | |
iqu | -0.062 |
ridge | -0.061 |
rol | -0.059 |
yles | -0.058 |
bar | -0.058 |
Token | play |
Token ID | 711 |
Feature activation | +2.04e-02 |
Pos logit contributions | |
iqu | +0.108 |
ridge | +0.105 |
umes | +0.102 |
yles | +0.101 |
rol | +0.100 |
Neg logit contributions | |
day | -0.031 |
good | -0.028 |
, | -0.028 |
time | -0.026 |
end | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
day | +0.018 |
girl | +0.017 |
team | +0.015 |
afternoon | +0.015 |
boy | +0.015 |
Neg logit contributions | |
ridge | -0.057 |
iqu | -0.056 |
pped | -0.055 |
rol | -0.054 |
bar | -0.054 |
Token | The |
Token ID | 383 |
Feature activation | -5.63e-03 |
Pos logit contributions | |
iqu | +0.060 |
ridge | +0.058 |
yles | +0.056 |
angs | +0.055 |
bar | +0.054 |
Neg logit contributions | |
day | -0.021 |
end | -0.021 |
good | -0.021 |
girl | -0.021 |
afternoon | -0.019 |
Token | rain |
Token ID | 6290 |
Feature activation | -5.81e-02 |
Pos logit contributions | |
iqu | +0.014 |
ridge | +0.013 |
rol | +0.013 |
bar | +0.013 |
umes | +0.013 |
Neg logit contributions | |
day | -0.007 |
, | -0.007 |
team | -0.007 |
good | -0.007 |
their | -0.007 |
Token | made |
Token ID | 925 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
iqu | +0.149 |
rol | +0.144 |
ridge | +0.143 |
angs | +0.140 |
umes | +0.139 |
Neg logit contributions | |
team | -0.068 |
good | -0.066 |
day | -0.064 |
their | -0.063 |
girl | -0.062 |
Token | him |
Token ID | 683 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
iqu | +0.034 |
ridge | +0.033 |
angs | +0.032 |
rol | +0.031 |
yles | +0.031 |
Neg logit contributions | |
day | -0.023 |
girl | -0.022 |
, | -0.021 |
team | -0.021 |
boy | -0.021 |
Token | wet |
Token ID | 9583 |
Feature activation | +1.22e-03 |
Pos logit contributions | |
iqu | +0.033 |
ridge | +0.031 |
umes | +0.030 |
angs | +0.030 |
rol | +0.030 |
Neg logit contributions | |
day | -0.013 |
good | -0.011 |
girl | -0.011 |
, | -0.011 |
time | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
day | +0.001 |
good | +0.000 |
team | +0.000 |
time | +0.000 |
girl | +0.000 |
Neg logit contributions | |
iqu | -0.004 |
ridge | -0.004 |
rol | -0.004 |
umes | -0.004 |
yles | -0.004 |
Token | but |
Token ID | 475 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
iqu | +0.052 |
ridge | +0.050 |
umes | +0.048 |
yles | +0.048 |
rol | +0.047 |
Neg logit contributions | |
day | -0.021 |
good | -0.019 |
team | -0.018 |
end | -0.018 |
, | -0.018 |
Token | he |
Token ID | 339 |
Feature activation | -1.42e-03 |
Pos logit contributions | |
iqu | +0.027 |
ridge | +0.025 |
yles | +0.025 |
bar | +0.024 |
angs | +0.023 |
Neg logit contributions | |
girl | -0.022 |
team | -0.022 |
day | -0.022 |
good | -0.021 |
town | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | -2.04e-02 |
Pos logit contributions | |
iqu | +0.038 |
ridge | +0.037 |
umes | +0.036 |
yles | +0.036 |
rol | +0.035 |
Neg logit contributions | |
day | -0.004 |
, | -0.004 |
good | -0.003 |
time | -0.003 |
and | -0.003 |
Token | It |
Token ID | 632 |
Feature activation | -3.94e-02 |
Pos logit contributions | |
iqu | +0.058 |
ridge | +0.056 |
umes | +0.054 |
yles | +0.054 |
angs | +0.052 |
Neg logit contributions | |
day | -0.025 |
good | -0.022 |
end | -0.022 |
girl | -0.021 |
, | -0.021 |
Token | was |
Token ID | 373 |
Feature activation | -4.08e-02 |
Pos logit contributions | |
iqu | +0.150 |
ridge | +0.148 |
yles | +0.145 |
umes | +0.145 |
ocol | +0.144 |
Neg logit contributions | |
day | -0.017 |
, | -0.016 |
and | -0.012 |
good | -0.012 |
time | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | -2.02e-02 |
Pos logit contributions | |
iqu | +0.151 |
ridge | +0.150 |
yles | +0.148 |
umes | +0.146 |
rol | +0.143 |
Neg logit contributions | |
day | -0.022 |
good | -0.020 |
, | -0.019 |
and | -0.018 |
time | -0.017 |
Token | big |
Token ID | 1263 |
Feature activation | -5.25e-03 |
Pos logit contributions | |
iqu | +0.060 |
ridge | +0.060 |
yles | +0.058 |
rol | +0.058 |
umes | +0.057 |
Neg logit contributions | |
day | -0.025 |
good | -0.023 |
, | -0.023 |
girl | -0.022 |
boy | -0.021 |
Token | re |
Token ID | 302 |
Feature activation | -7.33e-02 |
Pos logit contributions | |
iqu | +0.015 |
ridge | +0.015 |
umes | +0.014 |
yles | +0.014 |
angs | +0.014 |
Neg logit contributions | |
day | -0.006 |
good | -0.006 |
, | -0.005 |
time | -0.005 |
end | -0.005 |
Token | ind |
Token ID | 521 |
Feature activation | -2.98e-02 |
Pos logit contributions | |
iqu | +0.158 |
ridge | +0.154 |
umes | +0.149 |
yles | +0.148 |
rol | +0.145 |
Neg logit contributions | |
day | -0.141 |
, | -0.136 |
and | -0.132 |
team | -0.130 |
boy | -0.128 |
Token | eer |
Token ID | 28153 |
Feature activation | -2.71e-02 |
Pos logit contributions | |
iqu | +0.060 |
ridge | +0.059 |
angs | +0.055 |
umes | +0.054 |
yles | +0.054 |
Neg logit contributions | |
day | -0.059 |
good | -0.057 |
, | -0.056 |
time | -0.055 |
team | -0.055 |
Token | ! |
Token ID | 0 |
Feature activation | -4.44e-03 |
Pos logit contributions | |
iqu | +0.083 |
yles | +0.079 |
ridge | +0.078 |
umes | +0.077 |
rol | +0.076 |
Neg logit contributions | |
, | -0.034 |
and | -0.032 |
day | -0.032 |
good | -0.026 |
. | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
iqu | +0.012 |
ridge | +0.012 |
yles | +0.011 |
umes | +0.011 |
bar | +0.011 |
Neg logit contributions | |
day | -0.005 |
end | -0.005 |
good | -0.005 |
girl | -0.005 |
afternoon | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | -2.58e-02 |
Pos logit contributions | |
iqu | +0.052 |
yles | +0.049 |
ridge | +0.049 |
rol | +0.048 |
umes | +0.048 |
Neg logit contributions | |
day | -0.011 |
girl | -0.011 |
good | -0.010 |
end | -0.010 |
team | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -6.39e-03 |
Pos logit contributions | |
day | +0.005 |
good | +0.005 |
end | +0.005 |
girl | +0.004 |
team | +0.004 |
Neg logit contributions | |
iqu | -0.020 |
ridge | -0.020 |
yles | -0.020 |
umes | -0.020 |
bar | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | -1.66e-02 |
Pos logit contributions | |
iqu | +0.021 |
ridge | +0.020 |
yles | +0.020 |
umes | +0.020 |
rol | +0.020 |
Neg logit contributions | |
day | -0.004 |
good | -0.003 |
girl | -0.003 |
end | -0.003 |
team | -0.003 |
Token | But |
Token ID | 1537 |
Feature activation | -5.69e-03 |
Pos logit contributions | |
iqu | +0.043 |
ridge | +0.041 |
umes | +0.041 |
bar | +0.040 |
yles | +0.040 |
Neg logit contributions | |
day | -0.022 |
good | -0.021 |
, | -0.021 |
girl | -0.020 |
end | -0.020 |
Token | , |
Token ID | 11 |
Feature activation | -2.52e-02 |
Pos logit contributions | |
iqu | +0.016 |
ridge | +0.016 |
umes | +0.015 |
yles | +0.015 |
bar | +0.015 |
Neg logit contributions | |
day | -0.006 |
good | -0.006 |
girl | -0.006 |
team | -0.006 |
time | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | -4.92e-03 |
Pos logit contributions | |
iqu | +0.083 |
ridge | +0.082 |
umes | +0.079 |
yles | +0.079 |
rol | +0.077 |
Neg logit contributions | |
day | -0.018 |
good | -0.016 |
, | -0.015 |
end | -0.015 |
time | -0.014 |
Token | c |
Token ID | 269 |
Feature activation | -8.45e-02 |
Pos logit contributions | |
iqu | +0.013 |
ridge | +0.012 |
umes | +0.012 |
yles | +0.012 |
rol | +0.012 |
Neg logit contributions | |
day | -0.008 |
good | -0.007 |
, | -0.007 |
girl | -0.007 |
end | -0.007 |
Token | actus |
Token ID | 34144 |
Feature activation | -4.49e-02 |
Pos logit contributions | |
iqu | +0.176 |
ridge | +0.172 |
yles | +0.164 |
umes | +0.162 |
rol | +0.160 |
Neg logit contributions | |
day | -0.170 |
, | -0.161 |
and | -0.160 |
good | -0.159 |
girl | -0.156 |
Token | had |
Token ID | 550 |
Feature activation | -4.42e-02 |
Pos logit contributions | |
iqu | +0.121 |
ridge | +0.120 |
umes | +0.112 |
angs | +0.111 |
bar | +0.110 |
Neg logit contributions | |
day | -0.055 |
good | -0.055 |
their | -0.051 |
end | -0.051 |
girl | -0.050 |
Token | a |
Token ID | 257 |
Feature activation | -6.86e-03 |
Pos logit contributions | |
iqu | +0.137 |
ridge | +0.132 |
umes | +0.128 |
yles | +0.127 |
rol | +0.125 |
Neg logit contributions | |
day | -0.046 |
good | -0.042 |
, | -0.041 |
time | -0.039 |
end | -0.038 |
Token | friend |
Token ID | 1545 |
Feature activation | -2.26e-02 |
Pos logit contributions | |
iqu | +0.020 |
ridge | +0.020 |
umes | +0.019 |
angs | +0.018 |
yles | +0.018 |
Neg logit contributions | |
day | -0.007 |
, | -0.007 |
good | -0.007 |
girl | -0.006 |
end | -0.006 |
Token | - |
Token ID | 532 |
Feature activation | -3.40e-02 |
Pos logit contributions | |
iqu | +0.073 |
ridge | +0.071 |
umes | +0.068 |
angs | +0.068 |
rol | +0.067 |
Neg logit contributions | |
day | -0.019 |
good | -0.017 |
end | -0.015 |
girl | -0.015 |
team | -0.015 |
Token | ate |
Token ID | 15063 |
Feature activation | -3.94e-03 |
Pos logit contributions | |
day | +0.006 |
good | +0.005 |
, | +0.005 |
time | +0.005 |
end | +0.005 |
Neg logit contributions | |
iqu | -0.014 |
ridge | -0.014 |
umes | -0.013 |
yles | -0.013 |
rol | -0.013 |
Token | them |
Token ID | 606 |
Feature activation | +2.96e-02 |
Pos logit contributions | |
iqu | +0.011 |
ridge | +0.010 |
yles | +0.010 |
umes | +0.010 |
rol | +0.010 |
Neg logit contributions | |
day | -0.005 |
team | -0.004 |
girl | -0.004 |
end | -0.004 |
boy | -0.004 |
Token | together |
Token ID | 1978 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
girl | +0.038 |
boy | +0.036 |
day | +0.035 |
end | +0.033 |
team | +0.032 |
Neg logit contributions | |
ridge | -0.074 |
rol | -0.072 |
iqu | -0.072 |
umes | -0.071 |
vell | -0.070 |
Token | . |
Token ID | 13 |
Feature activation | +3.39e-03 |
Pos logit contributions | |
girl | +0.032 |
day | +0.030 |
boy | +0.029 |
end | +0.026 |
team | +0.025 |
Neg logit contributions | |
ridge | -0.123 |
iqu | -0.120 |
umes | -0.119 |
yles | -0.116 |
rol | -0.115 |
Token | The |
Token ID | 383 |
Feature activation | +7.61e-03 |
Pos logit contributions | |
day | +0.003 |
good | +0.003 |
end | +0.003 |
girl | +0.002 |
team | +0.002 |
Neg logit contributions | |
iqu | -0.011 |
ridge | -0.010 |
rol | -0.010 |
umes | -0.010 |
yles | -0.010 |
Token | p |
Token ID | 279 |
Feature activation | -7.36e-02 |
Pos logit contributions | |
their | +0.005 |
They | +0.004 |
day | +0.004 |
, | +0.004 |
and | +0.004 |
Neg logit contributions | |
iqu | -0.025 |
ridge | -0.023 |
umes | -0.023 |
rol | -0.023 |
ocol | -0.022 |
Token | ears |
Token ID | 4127 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
iqu | +0.116 |
ridge | +0.109 |
umes | +0.101 |
yles | +0.100 |
rol | +0.097 |
Neg logit contributions | |
day | -0.188 |
good | -0.181 |
, | -0.180 |
girl | -0.176 |
time | -0.176 |
Token | were |
Token ID | 547 |
Feature activation | -7.22e-03 |
Pos logit contributions | |
iqu | +0.058 |
ridge | +0.058 |
rol | +0.055 |
umes | +0.054 |
yles | +0.053 |
Neg logit contributions | |
day | -0.035 |
good | -0.033 |
time | -0.032 |
girl | -0.032 |
team | -0.032 |
Token | sweet |
Token ID | 6029 |
Feature activation | +4.09e-02 |
Pos logit contributions | |
iqu | +0.022 |
ridge | +0.021 |
umes | +0.020 |
rol | +0.020 |
yles | +0.020 |
Neg logit contributions | |
day | -0.007 |
girl | -0.007 |
team | -0.006 |
, | -0.006 |
good | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | -4.32e-03 |
Pos logit contributions | |
day | +0.022 |
end | +0.017 |
good | +0.017 |
team | +0.015 |
their | +0.015 |
Neg logit contributions | |
ridge | -0.133 |
iqu | -0.132 |
umes | -0.125 |
rol | -0.124 |
ous | -0.124 |
Token | juicy |
Token ID | 35075 |
Feature activation | +8.23e-03 |
Pos logit contributions | |
iqu | +0.012 |
ridge | +0.011 |
umes | +0.011 |
yles | +0.011 |
rol | +0.011 |
Neg logit contributions | |
day | -0.006 |
good | -0.006 |
, | -0.005 |
time | -0.005 |
end | -0.005 |
Token | day |
Token ID | 1110 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
good | +0.009 |
They | +0.009 |
team | +0.007 |
girl | +0.007 |
their | +0.006 |
Neg logit contributions | |
bar | -0.056 |
iqu | -0.054 |
ker | -0.053 |
ous | -0.053 |
cream | -0.052 |
Token | , |
Token ID | 11 |
Feature activation | -3.77e-02 |
Pos logit contributions | |
iqu | +0.061 |
ridge | +0.059 |
bar | +0.058 |
yles | +0.057 |
angs | +0.057 |
Neg logit contributions | |
day | -0.006 |
girl | -0.005 |
team | -0.004 |
end | -0.004 |
good | -0.004 |
Token | he |
Token ID | 339 |
Feature activation | +7.81e-04 |
Pos logit contributions | |
iqu | +0.100 |
ridge | +0.096 |
umes | +0.092 |
yles | +0.091 |
rol | +0.090 |
Neg logit contributions | |
day | -0.058 |
, | -0.054 |
good | -0.054 |
time | -0.052 |
girl | -0.051 |
Token | lost |
Token ID | 2626 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
day | +0.001 |
good | +0.001 |
end | +0.001 |
, | +0.001 |
girl | +0.001 |
Neg logit contributions | |
iqu | -0.002 |
ridge | -0.002 |
yles | -0.002 |
umes | -0.002 |
bar | -0.002 |
Token | his |
Token ID | 465 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
iqu | +0.079 |
ridge | +0.077 |
umes | +0.074 |
yles | +0.074 |
angs | +0.073 |
Neg logit contributions | |
day | -0.020 |
good | -0.017 |
, | -0.017 |
team | -0.016 |
end | -0.016 |
Token | er |
Token ID | 1931 |
Feature activation | -8.49e-02 |
Pos logit contributions | |
iqu | +0.050 |
ridge | +0.048 |
umes | +0.047 |
yles | +0.046 |
angs | +0.046 |
Neg logit contributions | |
day | -0.016 |
good | -0.015 |
, | -0.015 |
girl | -0.014 |
end | -0.014 |
Token | aser |
Token ID | 6005 |
Feature activation | -3.51e-02 |
Pos logit contributions | |
iqu | +0.122 |
ridge | +0.119 |
yles | +0.104 |
umes | +0.098 |
rol | +0.096 |
Neg logit contributions | |
, | -0.226 |
and | -0.223 |
day | -0.223 |
girl | -0.218 |
team | -0.217 |
Token | while |
Token ID | 981 |
Feature activation | +3.44e-03 |
Pos logit contributions | |
iqu | +0.119 |
ridge | +0.115 |
umes | +0.112 |
yles | +0.111 |
rol | +0.110 |
Neg logit contributions | |
day | -0.026 |
good | -0.022 |
, | -0.021 |
girl | -0.020 |
time | -0.020 |
Token | he |
Token ID | 339 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
day | +0.005 |
, | +0.005 |
good | +0.005 |
and | +0.005 |
time | +0.005 |
Neg logit contributions | |
iqu | -0.009 |
ridge | -0.009 |
umes | -0.009 |
rol | -0.008 |
angs | -0.008 |
Token | was |
Token ID | 373 |
Feature activation | -2.38e-02 |
Pos logit contributions | |
iqu | +0.044 |
ridge | +0.042 |
umes | +0.041 |
yles | +0.041 |
rol | +0.040 |
Neg logit contributions | |
day | -0.019 |
, | -0.018 |
good | -0.017 |
and | -0.017 |
time | -0.017 |
Token | drawing |
Token ID | 8263 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
iqu | +0.066 |
ridge | +0.064 |
umes | +0.061 |
yles | +0.061 |
rol | +0.060 |
Neg logit contributions | |
day | -0.034 |
, | -0.033 |
good | -0.033 |
time | -0.031 |
and | -0.031 |
Token | But |
Token ID | 887 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
iqu | +0.043 |
ridge | +0.042 |
umes | +0.041 |
yles | +0.041 |
rings | +0.041 |
Neg logit contributions | |
day | -0.016 |
good | -0.015 |
team | -0.014 |
end | -0.013 |
girl | -0.013 |
Token | one |
Token ID | 530 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
iqu | +0.062 |
ridge | +0.060 |
umes | +0.058 |
angs | +0.058 |
rol | +0.057 |
Neg logit contributions | |
day | -0.016 |
, | -0.014 |
good | -0.014 |
end | -0.013 |
time | -0.013 |
Token | day |
Token ID | 1110 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
They | +0.004 |
good | +0.003 |
their | +0.002 |
team | +0.002 |
town | +0.002 |
Neg logit contributions | |
iqu | -0.032 |
ridge | -0.032 |
yles | -0.031 |
angs | -0.031 |
paste | -0.031 |
Token | , |
Token ID | 11 |
Feature activation | -4.38e-02 |
Pos logit contributions | |
iqu | +0.092 |
ridge | +0.090 |
umes | +0.088 |
yles | +0.088 |
rings | +0.087 |
Neg logit contributions | |
day | -0.008 |
team | -0.006 |
girl | -0.006 |
end | -0.005 |
good | -0.005 |
Token | she |
Token ID | 673 |
Feature activation | -9.60e-03 |
Pos logit contributions | |
iqu | +0.130 |
ridge | +0.125 |
umes | +0.121 |
yles | +0.120 |
rol | +0.119 |
Neg logit contributions | |
day | -0.053 |
good | -0.049 |
, | -0.049 |
girl | -0.046 |
time | -0.046 |
Token | accidentally |
Token ID | 14716 |
Feature activation | -7.13e-02 |
Pos logit contributions | |
iqu | +0.024 |
ridge | +0.024 |
yles | +0.022 |
umes | +0.022 |
angs | +0.022 |
Neg logit contributions | |
day | -0.015 |
good | -0.014 |
, | -0.014 |
end | -0.013 |
girl | -0.013 |
Token | knocked |
Token ID | 13642 |
Feature activation | -5.44e-02 |
Pos logit contributions | |
iqu | +0.174 |
rol | +0.158 |
ocol | +0.156 |
yles | +0.156 |
umes | +0.156 |
Neg logit contributions | |
, | -0.130 |
day | -0.126 |
and | -0.125 |
good | -0.123 |
their | -0.120 |
Token | over |
Token ID | 625 |
Feature activation | +3.64e-03 |
Pos logit contributions | |
iqu | +0.177 |
ridge | +0.168 |
umes | +0.161 |
vell | +0.160 |
yles | +0.159 |
Neg logit contributions | |
day | -0.062 |
, | -0.055 |
and | -0.054 |
good | -0.054 |
time | -0.052 |
Token | some |
Token ID | 617 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
day | +0.003 |
good | +0.002 |
, | +0.002 |
end | +0.002 |
team | +0.002 |
Neg logit contributions | |
iqu | -0.012 |
ridge | -0.012 |
umes | -0.011 |
yles | -0.011 |
angs | -0.011 |
Token | furniture |
Token ID | 13091 |
Feature activation | -2.60e-02 |
Pos logit contributions | |
iqu | +0.036 |
ridge | +0.034 |
umes | +0.033 |
yles | +0.033 |
rol | +0.032 |
Neg logit contributions | |
day | -0.018 |
good | -0.017 |
, | -0.017 |
time | -0.016 |
end | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -3.34e-02 |
Pos logit contributions | |
iqu | +0.098 |
ridge | +0.094 |
yles | +0.092 |
umes | +0.092 |
rol | +0.091 |
Neg logit contributions | |
day | -0.011 |
, | -0.010 |
and | -0.009 |
good | -0.008 |
end | -0.007 |