Token | make |
Token ID | 787 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
ide | +0.015 |
winning | +0.012 |
icated | +0.012 |
seem | +0.012 |
hip | +0.011 |
Neg logit contributions | |
rolley | -0.020 |
butcher | -0.019 |
frying | -0.019 |
microwave | -0.019 |
piping | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
ide | +0.052 |
winning | +0.048 |
icated | +0.046 |
seem | +0.045 |
hip | +0.043 |
Neg logit contributions | |
rolley | -0.070 |
microwave | -0.068 |
butcher | -0.066 |
pan | -0.065 |
piping | -0.065 |
Token | most |
Token ID | 749 |
Feature activation | +4.81e+00 |
Pos logit contributions | |
ide | +0.055 |
winning | +0.048 |
icated | +0.047 |
seem | +0.046 |
hip | +0.045 |
Neg logit contributions | |
basket | -0.037 |
bag | -0.037 |
pan | -0.036 |
pouch | -0.035 |
rolley | -0.035 |
Token | beautiful |
Token ID | 4950 |
Feature activation | +5.50e+00 |
Pos logit contributions | |
ide | +0.145 |
hip | +0.137 |
seem | +0.137 |
icated | +0.130 |
wered | +0.128 |
Neg logit contributions | |
rolley | -0.126 |
pan | -0.118 |
etchup | -0.117 |
piping | -0.113 |
frying | -0.112 |
Token | rain |
Token ID | 6290 |
Feature activation | +6.32e+00 |
Pos logit contributions | |
ide | +0.144 |
winning | +0.133 |
icated | +0.131 |
seem | +0.126 |
hip | +0.121 |
Neg logit contributions | |
rolley | -0.168 |
pan | -0.167 |
basket | -0.163 |
pouch | -0.163 |
bag | -0.161 |
Token | bows |
Token ID | 25435 |
Feature activation | +9.37e+00 |
Pos logit contributions | |
winning | +0.143 |
ide | +0.142 |
icated | +0.140 |
aching | +0.131 |
wered | +0.127 |
Neg logit contributions | |
rolley | -0.196 |
pan | -0.190 |
bag | -0.185 |
frying | -0.185 |
pouch | -0.181 |
Token | . |
Token ID | 13 |
Feature activation | +7.09e-01 |
Pos logit contributions | |
winning | +0.247 |
icated | +0.245 |
ide | +0.237 |
aching | +0.232 |
hip | +0.226 |
Neg logit contributions | |
pan | -0.257 |
rolley | -0.242 |
piping | -0.239 |
waitress | -0.237 |
frying | -0.235 |
Token | The |
Token ID | 383 |
Feature activation | +7.82e-01 |
Pos logit contributions | |
ide | +0.021 |
winning | +0.019 |
seem | +0.019 |
icated | +0.018 |
shooting | +0.018 |
Neg logit contributions | |
rolley | -0.021 |
butcher | -0.020 |
microwave | -0.020 |
piping | -0.019 |
frying | -0.019 |
Token | sun |
Token ID | 4252 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
ide | +0.028 |
winning | +0.026 |
icated | +0.025 |
seem | +0.024 |
hip | +0.024 |
Neg logit contributions | |
bag | -0.018 |
pan | -0.018 |
microwave | -0.018 |
basket | -0.018 |
rolley | -0.018 |
Token | be |
Token ID | 1350 |
Feature activation | -2.74e-01 |
Pos logit contributions | |
icated | +0.037 |
ide | +0.037 |
winning | +0.036 |
seem | +0.034 |
hip | +0.033 |
Neg logit contributions | |
pan | -0.031 |
rolley | -0.030 |
microwave | -0.029 |
frying | -0.029 |
pouch | -0.029 |
Token | ams |
Token ID | 4105 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
rolley | +0.007 |
piping | +0.006 |
aked | +0.006 |
etchup | +0.006 |
pan | +0.006 |
Neg logit contributions | |
ide | -0.009 |
winning | -0.009 |
seem | -0.008 |
shooting | -0.008 |
icated | -0.008 |
Token | learn |
Token ID | 2193 |
Feature activation | +3.30e+00 |
Pos logit contributions | |
ide | +0.031 |
planets | +0.029 |
icated | +0.027 |
winning | +0.027 |
shooting | +0.026 |
Neg logit contributions | |
use | -0.036 |
butcher | -0.036 |
frying | -0.036 |
discount | -0.036 |
pouch | -0.035 |
Token | new |
Token ID | 649 |
Feature activation | +3.12e+00 |
Pos logit contributions | |
ide | +0.071 |
winning | +0.063 |
seem | +0.062 |
icated | +0.058 |
shooting | +0.056 |
Neg logit contributions | |
rolley | -0.119 |
butcher | -0.115 |
microwave | -0.114 |
frying | -0.109 |
piping | -0.109 |
Token | things |
Token ID | 1243 |
Feature activation | +4.93e+00 |
Pos logit contributions | |
ide | +0.073 |
winning | +0.063 |
seem | +0.061 |
icated | +0.059 |
shooting | +0.057 |
Neg logit contributions | |
rolley | -0.107 |
butcher | -0.103 |
microwave | -0.102 |
piping | -0.099 |
waitress | -0.099 |
Token | too |
Token ID | 1165 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
ide | +0.154 |
winning | +0.142 |
icated | +0.138 |
planets | +0.135 |
shooting | +0.134 |
Neg logit contributions | |
rolley | -0.118 |
butcher | -0.113 |
microwave | -0.112 |
use | -0.110 |
discount | -0.107 |
Token | . |
Token ID | 13 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
ide | +0.004 |
winning | +0.004 |
seem | +0.004 |
icated | +0.004 |
shooting | +0.004 |
Neg logit contributions | |
rolley | -0.004 |
butcher | -0.003 |
microwave | -0.003 |
piping | -0.003 |
frying | -0.003 |
Token | Playing |
Token ID | 23911 |
Feature activation | +8.97e+00 |
Pos logit contributions | |
ide | +0.039 |
winning | +0.036 |
seem | +0.035 |
icated | +0.034 |
shooting | +0.034 |
Neg logit contributions | |
rolley | -0.038 |
butcher | -0.036 |
microwave | -0.036 |
piping | -0.034 |
frying | -0.034 |
Token | is |
Token ID | 318 |
Feature activation | +3.14e+00 |
Pos logit contributions | |
ide | +0.215 |
seem | +0.194 |
winning | +0.187 |
icated | +0.182 |
textures | +0.177 |
Neg logit contributions | |
microwave | -0.281 |
butcher | -0.272 |
P | -0.270 |
rolley | -0.269 |
� | -0.264 |
Token | fun |
Token ID | 1257 |
Feature activation | +5.54e+00 |
Pos logit contributions | |
ide | +0.078 |
seem | +0.061 |
icated | +0.060 |
hip | +0.060 |
planets | +0.058 |
Neg logit contributions | |
butcher | -0.102 |
microwave | -0.100 |
rolley | -0.096 |
frying | -0.096 |
paramed | -0.094 |
Token | , |
Token ID | 11 |
Feature activation | +3.01e+00 |
Pos logit contributions | |
ide | +0.140 |
winning | +0.125 |
icated | +0.121 |
seem | +0.121 |
wered | +0.116 |
Neg logit contributions | |
rolley | -0.165 |
microwave | -0.161 |
butcher | -0.157 |
pan | -0.155 |
bag | -0.153 |
Token | but |
Token ID | 475 |
Feature activation | +3.18e+00 |
Pos logit contributions | |
ide | +0.078 |
seem | +0.061 |
hip | +0.061 |
planets | +0.060 |
communicate | +0.059 |
Neg logit contributions | |
microwave | -0.092 |
bag | -0.085 |
pan | -0.084 |
paramed | -0.084 |
cart | -0.083 |
Token | learning |
Token ID | 4673 |
Feature activation | +4.73e+00 |
Pos logit contributions | |
ide | +0.083 |
icated | +0.063 |
hip | +0.062 |
seem | +0.061 |
tunes | +0.061 |
Neg logit contributions | |
pouch | -0.094 |
etchup | -0.093 |
iw | -0.093 |
butcher | -0.092 |
pan | -0.091 |
Token | Winning |
Token ID | 40983 |
Feature activation | +8.56e+00 |
Pos logit contributions | |
ide | +0.003 |
winning | +0.003 |
seem | +0.002 |
icated | +0.002 |
sharing | +0.002 |
Neg logit contributions | |
rolley | -0.004 |
butcher | -0.004 |
microwave | -0.004 |
frying | -0.004 |
piping | -0.004 |
Token | is |
Token ID | 318 |
Feature activation | +3.14e+00 |
Pos logit contributions | |
ide | +0.319 |
wered | +0.293 |
hip | +0.279 |
seem | +0.271 |
oted | +0.271 |
Neg logit contributions | |
� | -0.153 |
microwave | -0.143 |
bag | -0.138 |
pan | -0.132 |
rolley | -0.128 |
Token | not |
Token ID | 407 |
Feature activation | +2.52e+00 |
Pos logit contributions | |
ide | +0.076 |
seem | +0.058 |
icated | +0.057 |
planets | +0.057 |
hip | +0.056 |
Neg logit contributions | |
butcher | -0.101 |
discount | -0.099 |
P | -0.099 |
Found | -0.097 |
use | -0.096 |
Token | important |
Token ID | 1593 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
ide | +0.068 |
seem | +0.058 |
bee | +0.056 |
planets | +0.055 |
hip | +0.052 |
Neg logit contributions | |
discount | -0.068 |
butcher | -0.066 |
Found | -0.064 |
rolley | -0.064 |
microwave | -0.062 |
Token | . |
Token ID | 13 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
rolley | +0.005 |
butcher | +0.005 |
microwave | +0.004 |
waitress | +0.004 |
frying | +0.004 |
Neg logit contributions | |
ide | -0.004 |
seem | -0.003 |
winning | -0.003 |
icated | -0.003 |
planets | -0.003 |
Token | Winning |
Token ID | 40983 |
Feature activation | +8.86e+00 |
Pos logit contributions | |
ide | +0.036 |
winning | +0.032 |
seem | +0.031 |
happening | +0.031 |
shooting | +0.030 |
Neg logit contributions | |
rolley | -0.041 |
butcher | -0.039 |
microwave | -0.039 |
piping | -0.038 |
etchup | -0.037 |
Token | is |
Token ID | 318 |
Feature activation | +3.30e+00 |
Pos logit contributions | |
ide | +0.335 |
wered | +0.305 |
hip | +0.283 |
oted | +0.281 |
winning | +0.280 |
Neg logit contributions | |
� | -0.151 |
microwave | -0.136 |
bag | -0.135 |
pan | -0.132 |
rolley | -0.127 |
Token | not |
Token ID | 407 |
Feature activation | +2.64e+00 |
Pos logit contributions | |
ide | +0.075 |
seem | +0.055 |
planets | +0.055 |
icated | +0.055 |
tunes | +0.053 |
Neg logit contributions | |
butcher | -0.110 |
discount | -0.109 |
P | -0.107 |
Found | -0.106 |
use | -0.105 |
Token | important |
Token ID | 1593 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
ide | +0.069 |
seem | +0.057 |
planets | +0.053 |
bee | +0.052 |
winning | +0.051 |
Neg logit contributions | |
discount | -0.075 |
butcher | -0.073 |
rolley | -0.072 |
Found | -0.071 |
microwave | -0.069 |
Token | . |
Token ID | 13 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
rolley | +0.001 |
butcher | +0.001 |
microwave | +0.001 |
waitress | +0.001 |
frying | +0.001 |
Neg logit contributions | |
ide | -0.001 |
winning | -0.001 |
seem | -0.001 |
icated | -0.001 |
planets | -0.001 |
Token | Playing |
Token ID | 23911 |
Feature activation | +8.61e+00 |
Pos logit contributions | |
ide | +0.041 |
winning | +0.036 |
seem | +0.035 |
happening | +0.034 |
icated | +0.034 |
Neg logit contributions | |
rolley | -0.046 |
butcher | -0.044 |
microwave | -0.044 |
piping | -0.043 |
etchup | -0.043 |
Token | more |
Token ID | 517 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
ide | +0.061 |
icated | +0.052 |
seem | +0.051 |
winning | +0.049 |
planets | +0.047 |
Neg logit contributions | |
rolley | -0.080 |
butcher | -0.080 |
microwave | -0.077 |
frying | -0.076 |
piping | -0.075 |
Token | important |
Token ID | 1593 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
ide | +0.026 |
icated | +0.022 |
seem | +0.022 |
winning | +0.021 |
hip | +0.020 |
Neg logit contributions | |
rolley | -0.041 |
butcher | -0.040 |
frying | -0.039 |
microwave | -0.039 |
piping | -0.038 |
Token | than |
Token ID | 621 |
Feature activation | +2.86e+00 |
Pos logit contributions | |
rolley | +0.005 |
butcher | +0.005 |
microwave | +0.005 |
frying | +0.005 |
waitress | +0.005 |
Neg logit contributions | |
ide | -0.003 |
winning | -0.002 |
seem | -0.002 |
icated | -0.002 |
shooting | -0.002 |
Token | winning |
Token ID | 5442 |
Feature activation | +5.21e+00 |
Pos logit contributions | |
ide | +0.061 |
seem | +0.046 |
hip | +0.046 |
icated | +0.045 |
rhythms | +0.045 |
Neg logit contributions | |
butcher | -0.100 |
etchup | -0.098 |
microwave | -0.097 |
frying | -0.096 |
rolley | -0.096 |
Token | . |
Token ID | 13 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
ide | +0.127 |
icated | +0.112 |
seem | +0.110 |
winning | +0.107 |
textures | +0.103 |
Neg logit contributions | |
rolley | -0.162 |
butcher | -0.159 |
microwave | -0.153 |
pouch | -0.151 |
frying | -0.151 |
Token | Winning |
Token ID | 40983 |
Feature activation | +8.84e+00 |
Pos logit contributions | |
ide | +0.040 |
winning | +0.037 |
seem | +0.036 |
icated | +0.035 |
shooting | +0.035 |
Neg logit contributions | |
rolley | -0.041 |
butcher | -0.039 |
microwave | -0.039 |
piping | -0.037 |
frying | -0.037 |
Token | is |
Token ID | 318 |
Feature activation | +3.56e+00 |
Pos logit contributions | |
ide | +0.244 |
icated | +0.208 |
wered | +0.204 |
winning | +0.203 |
seem | +0.202 |
Neg logit contributions | |
rolley | -0.238 |
microwave | -0.229 |
butcher | -0.227 |
pan | -0.226 |
bag | -0.225 |
Token | nice |
Token ID | 3621 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
ide | +0.076 |
icated | +0.059 |
planets | +0.059 |
seem | +0.058 |
tunes | +0.056 |
Neg logit contributions | |
butcher | -0.121 |
discount | -0.117 |
rolley | -0.117 |
P | -0.115 |
etchup | -0.115 |
Token | , |
Token ID | 11 |
Feature activation | +2.91e+00 |
Pos logit contributions | |
ide | +0.031 |
icated | +0.027 |
seem | +0.027 |
winning | +0.027 |
wered | +0.025 |
Neg logit contributions | |
rolley | -0.049 |
butcher | -0.047 |
microwave | -0.045 |
bag | -0.045 |
pan | -0.044 |
Token | but |
Token ID | 475 |
Feature activation | +3.28e+00 |
Pos logit contributions | |
ide | +0.064 |
seem | +0.055 |
planets | +0.053 |
hip | +0.051 |
icated | +0.050 |
Neg logit contributions | |
etchup | -0.092 |
rolley | -0.090 |
butcher | -0.089 |
iw | -0.088 |
pan | -0.087 |
Token | it |
Token ID | 340 |
Feature activation | +9.98e-01 |
Pos logit contributions | |
ide | +0.076 |
icated | +0.057 |
planets | +0.056 |
bee | +0.055 |
hip | +0.053 |
Neg logit contributions | |
rolley | -0.108 |
etchup | -0.107 |
iw | -0.106 |
butcher | -0.106 |
pouch | -0.106 |
Token | , |
Token ID | 11 |
Feature activation | +5.42e-01 |
Pos logit contributions | |
butcher | +0.050 |
microwave | +0.049 |
pan | +0.049 |
frying | +0.048 |
rolley | +0.048 |
Neg logit contributions | |
ide | -0.047 |
icated | -0.045 |
planets | -0.044 |
winning | -0.043 |
textures | -0.041 |
Token | " |
Token ID | 366 |
Feature activation | -6.70e-01 |
Pos logit contributions | |
ide | +0.018 |
icated | +0.017 |
planets | +0.017 |
textures | +0.016 |
sizes | +0.015 |
Neg logit contributions | |
butcher | -0.014 |
microwave | -0.013 |
paramed | -0.013 |
pan | -0.013 |
frying | -0.013 |
Token | But |
Token ID | 1537 |
Feature activation | +2.32e+00 |
Pos logit contributions | |
rolley | +0.015 |
P | +0.015 |
butcher | +0.015 |
microwave | +0.014 |
etchup | +0.013 |
Neg logit contributions | |
ide | -0.024 |
winning | -0.022 |
planets | -0.021 |
icated | -0.021 |
shooting | -0.021 |
Token | why |
Token ID | 1521 |
Feature activation | -5.40e-01 |
Pos logit contributions | |
ide | +0.080 |
hip | +0.066 |
tunes | +0.065 |
icated | +0.065 |
seem | +0.063 |
Neg logit contributions | |
pan | -0.053 |
microwave | -0.053 |
butcher | -0.050 |
frying | -0.050 |
piping | -0.049 |
Token | ? |
Token ID | 30 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
� | +0.016 |
microwave | +0.016 |
butcher | +0.016 |
bag | +0.016 |
frying | +0.015 |
Neg logit contributions | |
ide | -0.014 |
tunes | -0.012 |
planets | -0.011 |
hip | -0.011 |
textures | -0.011 |
Token | Winning |
Token ID | 40983 |
Feature activation | +8.75e+00 |
Pos logit contributions | |
ide | +0.006 |
winning | +0.006 |
seem | +0.005 |
icated | +0.005 |
shooting | +0.005 |
Neg logit contributions | |
rolley | -0.006 |
butcher | -0.006 |
microwave | -0.006 |
piping | -0.006 |
frying | -0.006 |
Token | doesn |
Token ID | 1595 |
Feature activation | +2.50e+00 |
Pos logit contributions | |
ide | +0.251 |
winning | +0.209 |
seem | +0.206 |
hip | +0.206 |
wered | +0.205 |
Neg logit contributions | |
microwave | -0.225 |
� | -0.221 |
rolley | -0.218 |
pan | -0.215 |
frying | -0.212 |
Token | 't |
Token ID | 470 |
Feature activation | +3.10e+00 |
Pos logit contributions | |
ide | +0.055 |
textures | +0.049 |
planets | +0.048 |
winning | +0.047 |
shapes | +0.046 |
Neg logit contributions | |
rolley | -0.083 |
microwave | -0.082 |
butcher | -0.082 |
� | -0.078 |
piping | -0.077 |
Token | mean |
Token ID | 1612 |
Feature activation | +9.36e-01 |
Pos logit contributions | |
ide | +0.071 |
planets | +0.062 |
winning | +0.058 |
pretending | +0.057 |
aching | +0.056 |
Neg logit contributions | |
discount | -0.091 |
rolley | -0.087 |
Found | -0.084 |
microwave | -0.084 |
butcher | -0.084 |
Token | you |
Token ID | 345 |
Feature activation | +2.82e+00 |
Pos logit contributions | |
ide | +0.024 |
winning | +0.022 |
seem | +0.021 |
icated | +0.021 |
shooting | +0.021 |
Neg logit contributions | |
rolley | -0.029 |
butcher | -0.028 |
microwave | -0.028 |
piping | -0.027 |
frying | -0.027 |
Token | have |
Token ID | 423 |
Feature activation | +9.26e-01 |
Pos logit contributions | |
ide | +0.070 |
planets | +0.063 |
winning | +0.060 |
patterns | +0.060 |
textures | +0.059 |
Neg logit contributions | |
use | -0.082 |
� | -0.082 |
rolley | -0.079 |
butcher | -0.079 |
discount | -0.078 |
Token | . |
Token ID | 13 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
rolley | +0.001 |
butcher | +0.001 |
microwave | +0.001 |
waitress | +0.001 |
frying | +0.001 |
Neg logit contributions | |
ide | -0.001 |
winning | -0.001 |
seem | -0.001 |
icated | -0.001 |
planets | -0.001 |
Token | Playing |
Token ID | 23911 |
Feature activation | +8.61e+00 |
Pos logit contributions | |
ide | +0.041 |
winning | +0.036 |
seem | +0.035 |
happening | +0.034 |
icated | +0.034 |
Neg logit contributions | |
rolley | -0.046 |
butcher | -0.044 |
microwave | -0.044 |
piping | -0.043 |
etchup | -0.043 |
Token | is |
Token ID | 318 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
ide | +0.248 |
seem | +0.215 |
icated | +0.211 |
textures | +0.204 |
wered | +0.199 |
Neg logit contributions | |
P | -0.230 |
� | -0.226 |
'," | -0.211 |
microwave | -0.208 |
etchup | -0.208 |
Token | important |
Token ID | 1593 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
ide | +0.067 |
seem | +0.049 |
icated | +0.047 |
planets | +0.047 |
hip | +0.046 |
Neg logit contributions | |
butcher | -0.086 |
microwave | -0.083 |
discount | -0.083 |
Found | -0.082 |
P | -0.082 |
Token | . |
Token ID | 13 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
ide | +0.005 |
winning | +0.004 |
seem | +0.004 |
icated | +0.004 |
planets | +0.004 |
Neg logit contributions | |
rolley | -0.007 |
butcher | -0.007 |
microwave | -0.007 |
frying | -0.006 |
pouch | -0.006 |
Token | Playing |
Token ID | 23911 |
Feature activation | +8.64e+00 |
Pos logit contributions | |
ide | +0.042 |
winning | +0.038 |
seem | +0.036 |
happening | +0.035 |
icated | +0.035 |
Neg logit contributions | |
rolley | -0.046 |
butcher | -0.044 |
microwave | -0.043 |
piping | -0.042 |
etchup | -0.042 |
Token | is |
Token ID | 318 |
Feature activation | +2.72e+00 |
Pos logit contributions | |
ide | +0.247 |
seem | +0.216 |
icated | +0.210 |
textures | +0.202 |
matter | +0.195 |
Neg logit contributions | |
� | -0.233 |
P | -0.233 |
microwave | -0.218 |
'," | -0.217 |
etchup | -0.212 |
Token | important |
Token ID | 1593 |
Feature activation | +2.81e-01 |
Pos logit contributions | |
ide | +0.073 |
seem | +0.056 |
icated | +0.053 |
planets | +0.051 |
hip | +0.051 |
Neg logit contributions | |
butcher | -0.086 |
microwave | -0.083 |
discount | -0.082 |
P | -0.081 |
paramed | -0.081 |
Token | ." |
Token ID | 526 |
Feature activation | -5.10e-01 |
Pos logit contributions | |
ide | +0.007 |
winning | +0.006 |
seem | +0.006 |
icated | +0.006 |
shooting | +0.005 |
Neg logit contributions | |
rolley | -0.010 |
butcher | -0.009 |
microwave | -0.009 |
frying | -0.009 |
pouch | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
rolley | +0.018 |
butcher | +0.017 |
microwave | +0.017 |
frying | +0.017 |
piping | +0.016 |
Neg logit contributions | |
ide | -0.012 |
winning | -0.011 |
seem | -0.011 |
icated | -0.010 |
sharing | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | +9.06e-01 |
Pos logit contributions | |
ide | +0.035 |
winning | +0.033 |
seem | +0.032 |
icated | +0.031 |
textures | +0.030 |
Neg logit contributions | |
rolley | -0.036 |
butcher | -0.035 |
microwave | -0.034 |
piping | -0.033 |
frying | -0.033 |
Token | and |
Token ID | 290 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
ide | +0.074 |
winning | +0.070 |
icated | +0.067 |
seem | +0.063 |
textures | +0.062 |
Neg logit contributions | |
rolley | -0.106 |
butcher | -0.104 |
microwave | -0.103 |
frying | -0.101 |
piping | -0.101 |
Token | even |
Token ID | 772 |
Feature activation | -3.30e+00 |
Pos logit contributions | |
ide | +0.031 |
winning | +0.029 |
seem | +0.028 |
shapes | +0.027 |
icated | +0.027 |
Neg logit contributions | |
rolley | -0.052 |
butcher | -0.049 |
microwave | -0.049 |
piping | -0.047 |
frying | -0.047 |
Token | friendly |
Token ID | 8030 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
rolley | +0.111 |
butcher | +0.105 |
microwave | +0.104 |
piping | +0.100 |
frying | +0.099 |
Neg logit contributions | |
ide | -0.083 |
winning | -0.078 |
seem | -0.076 |
icated | -0.073 |
shooting | -0.072 |
Token | animals |
Token ID | 4695 |
Feature activation | +3.96e+00 |
Pos logit contributions | |
ide | +0.045 |
winning | +0.041 |
seem | +0.039 |
icated | +0.039 |
shooting | +0.037 |
Neg logit contributions | |
rolley | -0.064 |
butcher | -0.062 |
microwave | -0.061 |
waitress | -0.059 |
piping | -0.059 |
Token | . |
Token ID | 13 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
ide | +0.092 |
winning | +0.086 |
icated | +0.084 |
seem | +0.080 |
textures | +0.079 |
Neg logit contributions | |
rolley | -0.130 |
butcher | -0.126 |
microwave | -0.122 |
frying | -0.122 |
pouch | -0.121 |
Token | Playing |
Token ID | 23911 |
Feature activation | +8.63e+00 |
Pos logit contributions | |
ide | +0.030 |
winning | +0.028 |
seem | +0.027 |
icated | +0.026 |
sharing | +0.025 |
Neg logit contributions | |
rolley | -0.038 |
butcher | -0.036 |
microwave | -0.036 |
frying | -0.035 |
piping | -0.035 |
Token | in |
Token ID | 287 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
ide | +0.200 |
winning | +0.188 |
seem | +0.183 |
icated | +0.178 |
hip | +0.176 |
Neg logit contributions | |
P | -0.274 |
butcher | -0.263 |
microwave | -0.260 |
pan | -0.259 |
rolley | -0.257 |
Token | the |
Token ID | 262 |
Feature activation | +4.70e-01 |
Pos logit contributions | |
ide | +0.046 |
winning | +0.041 |
seem | +0.041 |
icated | +0.040 |
hip | +0.037 |
Neg logit contributions | |
rolley | -0.068 |
butcher | -0.067 |
microwave | -0.066 |
piping | -0.064 |
frying | -0.063 |
Token | garden |
Token ID | 11376 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
ide | +0.013 |
winning | +0.012 |
seem | +0.011 |
icated | +0.011 |
shooting | +0.011 |
Neg logit contributions | |
rolley | -0.015 |
butcher | -0.014 |
microwave | -0.014 |
piping | -0.013 |
frying | -0.013 |
Token | made |
Token ID | 925 |
Feature activation | +1.85e+00 |
Pos logit contributions | |
ide | +0.050 |
winning | +0.048 |
icated | +0.047 |
seem | +0.044 |
shooting | +0.043 |
Neg logit contributions | |
rolley | -0.065 |
microwave | -0.063 |
butcher | -0.062 |
frying | -0.061 |
pan | -0.061 |
Token | the |
Token ID | 262 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
ide | +0.044 |
winning | +0.042 |
icated | +0.039 |
seem | +0.038 |
hip | +0.038 |
Neg logit contributions | |
rolley | -0.057 |
butcher | -0.056 |
waitress | -0.056 |
microwave | -0.055 |
pan | -0.054 |
Token | Winning |
Token ID | 40983 |
Feature activation | +8.86e+00 |
Pos logit contributions | |
ide | +0.036 |
winning | +0.032 |
seem | +0.031 |
happening | +0.031 |
shooting | +0.030 |
Neg logit contributions | |
rolley | -0.041 |
butcher | -0.039 |
microwave | -0.039 |
piping | -0.038 |
etchup | -0.037 |
Token | is |
Token ID | 318 |
Feature activation | +3.30e+00 |
Pos logit contributions | |
ide | +0.335 |
wered | +0.305 |
hip | +0.283 |
oted | +0.281 |
winning | +0.280 |
Neg logit contributions | |
� | -0.151 |
microwave | -0.136 |
bag | -0.135 |
pan | -0.132 |
rolley | -0.127 |
Token | not |
Token ID | 407 |
Feature activation | +2.64e+00 |
Pos logit contributions | |
ide | +0.075 |
seem | +0.055 |
planets | +0.055 |
icated | +0.055 |
tunes | +0.053 |
Neg logit contributions | |
butcher | -0.110 |
discount | -0.109 |
P | -0.107 |
Found | -0.106 |
use | -0.105 |
Token | important |
Token ID | 1593 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
ide | +0.069 |
seem | +0.057 |
planets | +0.053 |
bee | +0.052 |
winning | +0.051 |
Neg logit contributions | |
discount | -0.075 |
butcher | -0.073 |
rolley | -0.072 |
Found | -0.071 |
microwave | -0.069 |
Token | . |
Token ID | 13 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
rolley | +0.001 |
butcher | +0.001 |
microwave | +0.001 |
waitress | +0.001 |
frying | +0.001 |
Neg logit contributions | |
ide | -0.001 |
winning | -0.001 |
seem | -0.001 |
icated | -0.001 |
planets | -0.001 |
Token | Playing |
Token ID | 23911 |
Feature activation | +8.61e+00 |
Pos logit contributions | |
ide | +0.041 |
winning | +0.036 |
seem | +0.035 |
happening | +0.034 |
icated | +0.034 |
Neg logit contributions | |
rolley | -0.046 |
butcher | -0.044 |
microwave | -0.044 |
piping | -0.043 |
etchup | -0.043 |
Token | is |
Token ID | 318 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
ide | +0.248 |
seem | +0.215 |
icated | +0.211 |
textures | +0.204 |
wered | +0.199 |
Neg logit contributions | |
P | -0.230 |
� | -0.226 |
'," | -0.211 |
microwave | -0.208 |
etchup | -0.208 |
Token | important |
Token ID | 1593 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
ide | +0.067 |
seem | +0.049 |
icated | +0.047 |
planets | +0.047 |
hip | +0.046 |
Neg logit contributions | |
butcher | -0.086 |
microwave | -0.083 |
discount | -0.083 |
Found | -0.082 |
P | -0.082 |
Token | . |
Token ID | 13 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
ide | +0.005 |
winning | +0.004 |
seem | +0.004 |
icated | +0.004 |
planets | +0.004 |
Neg logit contributions | |
rolley | -0.007 |
butcher | -0.007 |
microwave | -0.007 |
frying | -0.006 |
pouch | -0.006 |
Token | Playing |
Token ID | 23911 |
Feature activation | +8.64e+00 |
Pos logit contributions | |
ide | +0.042 |
winning | +0.038 |
seem | +0.036 |
happening | +0.035 |
icated | +0.035 |
Neg logit contributions | |
rolley | -0.046 |
butcher | -0.044 |
microwave | -0.043 |
piping | -0.042 |
etchup | -0.042 |
Token | is |
Token ID | 318 |
Feature activation | +2.72e+00 |
Pos logit contributions | |
ide | +0.247 |
seem | +0.216 |
icated | +0.210 |
textures | +0.202 |
matter | +0.195 |
Neg logit contributions | |
� | -0.233 |
P | -0.233 |
microwave | -0.218 |
'," | -0.217 |
etchup | -0.212 |
Token | . |
Token ID | 13 |
Feature activation | +5.68e-01 |
Pos logit contributions | |
rolley | +0.108 |
butcher | +0.102 |
microwave | +0.102 |
piping | +0.101 |
frying | +0.099 |
Neg logit contributions | |
winning | -0.033 |
ide | -0.029 |
sharing | -0.029 |
being | -0.029 |
shooting | -0.027 |
Token | Playing |
Token ID | 23911 |
Feature activation | +8.21e+00 |
Pos logit contributions | |
ide | +0.014 |
winning | +0.013 |
icated | +0.012 |
seem | +0.012 |
sharing | +0.012 |
Neg logit contributions | |
rolley | -0.019 |
butcher | -0.018 |
microwave | -0.018 |
frying | -0.018 |
piping | -0.018 |
Token | is |
Token ID | 318 |
Feature activation | +2.45e+00 |
Pos logit contributions | |
ide | +0.238 |
seem | +0.215 |
textures | +0.196 |
matter | +0.195 |
icated | +0.192 |
Neg logit contributions | |
� | -0.225 |
P | -0.223 |
microwave | -0.218 |
etchup | -0.216 |
pan | -0.206 |
Token | important |
Token ID | 1593 |
Feature activation | +8.32e-02 |
Pos logit contributions | |
ide | +0.067 |
seem | +0.055 |
hip | +0.050 |
icated | +0.050 |
textures | +0.049 |
Neg logit contributions | |
butcher | -0.077 |
microwave | -0.075 |
waitress | -0.071 |
frying | -0.071 |
discount | -0.071 |
Token | . |
Token ID | 13 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
ide | +0.002 |
winning | +0.002 |
seem | +0.002 |
icated | +0.002 |
shooting | +0.002 |
Neg logit contributions | |
rolley | -0.003 |
butcher | -0.003 |
microwave | -0.003 |
waitress | -0.003 |
frying | -0.003 |
Token | Playing |
Token ID | 23911 |
Feature activation | +8.60e+00 |
Pos logit contributions | |
ide | +0.040 |
winning | +0.037 |
seem | +0.035 |
icated | +0.034 |
shooting | +0.034 |
Neg logit contributions | |
rolley | -0.047 |
butcher | -0.045 |
microwave | -0.045 |
piping | -0.043 |
frying | -0.043 |
Token | is |
Token ID | 318 |
Feature activation | +2.73e+00 |
Pos logit contributions | |
ide | +0.256 |
seem | +0.228 |
icated | +0.213 |
textures | +0.210 |
matter | +0.208 |
Neg logit contributions | |
� | -0.229 |
P | -0.218 |
microwave | -0.216 |
etchup | -0.210 |
'," | -0.203 |
Token | important |
Token ID | 1593 |
Feature activation | +2.87e-01 |
Pos logit contributions | |
ide | +0.075 |
seem | +0.062 |
hip | +0.056 |
icated | +0.056 |
textures | +0.054 |
Neg logit contributions | |
butcher | -0.084 |
microwave | -0.082 |
frying | -0.078 |
discount | -0.078 |
etchup | -0.078 |
Token | ." |
Token ID | 526 |
Feature activation | -4.74e-01 |
Pos logit contributions | |
ide | +0.007 |
winning | +0.006 |
seem | +0.006 |
icated | +0.006 |
shooting | +0.006 |
Neg logit contributions | |
rolley | -0.009 |
butcher | -0.009 |
microwave | -0.009 |
waitress | -0.009 |
frying | -0.009 |
Token | They |
Token ID | 1119 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
rolley | +0.019 |
frying | +0.018 |
butcher | +0.017 |
piping | +0.017 |
microwave | +0.017 |
Neg logit contributions | |
ide | -0.009 |
winning | -0.008 |
seem | -0.008 |
sharing | -0.008 |
icated | -0.008 |
Token | smile |
Token ID | 8212 |
Feature activation | +7.69e-01 |
Pos logit contributions | |
ide | +0.057 |
winning | +0.049 |
aching | +0.047 |
izing | +0.047 |
happening | +0.047 |
Neg logit contributions | |
bag | -0.044 |
microwave | -0.043 |
pouch | -0.042 |
basket | -0.041 |
waitress | -0.041 |
Token | when |
Token ID | 618 |
Feature activation | -1.93e-01 |
Pos logit contributions | |
rolley | +0.021 |
butcher | +0.019 |
microwave | +0.019 |
frying | +0.019 |
etchup | +0.019 |
Neg logit contributions | |
ide | -0.011 |
winning | -0.010 |
seem | -0.010 |
icated | -0.010 |
shooting | -0.009 |
Token | you |
Token ID | 345 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
rolley | +0.006 |
butcher | +0.006 |
microwave | +0.006 |
frying | +0.005 |
pan | +0.005 |
Neg logit contributions | |
ide | -0.005 |
icated | -0.004 |
winning | -0.004 |
hip | -0.004 |
seem | -0.004 |
Token | 're |
Token ID | 821 |
Feature activation | -6.96e-01 |
Pos logit contributions | |
use | +0.009 |
� | +0.008 |
rolley | +0.008 |
bag | +0.008 |
butcher | +0.008 |
Neg logit contributions | |
ide | -0.008 |
winning | -0.007 |
planets | -0.007 |
textures | -0.007 |
happening | -0.006 |
Token | tired |
Token ID | 10032 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
butcher | +0.018 |
frying | +0.018 |
microwave | +0.018 |
rolley | +0.017 |
pan | +0.017 |
Neg logit contributions | |
ide | -0.021 |
textures | -0.018 |
icated | -0.018 |
planets | -0.018 |
tunes | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
rolley | +0.053 |
butcher | +0.051 |
microwave | +0.050 |
frying | +0.049 |
piping | +0.049 |
Neg logit contributions | |
ide | -0.031 |
winning | -0.028 |
icated | -0.026 |
seem | -0.026 |
shooting | -0.025 |
Token | Winning |
Token ID | 40983 |
Feature activation | +8.59e+00 |
Pos logit contributions | |
ide | +0.033 |
happening | +0.030 |
winning | +0.029 |
seem | +0.029 |
planets | +0.029 |
Neg logit contributions | |
rolley | -0.030 |
butcher | -0.029 |
microwave | -0.028 |
etchup | -0.027 |
piping | -0.026 |
Token | isn |
Token ID | 2125 |
Feature activation | +5.48e+00 |
Pos logit contributions | |
ide | +0.274 |
wered | +0.245 |
hip | +0.226 |
icated | +0.220 |
oted | +0.220 |
Neg logit contributions | |
� | -0.186 |
bag | -0.182 |
microwave | -0.178 |
rolley | -0.177 |
pan | -0.177 |
Token | 't |
Token ID | 470 |
Feature activation | +3.06e+00 |
Pos logit contributions | |
textures | +0.161 |
planets | +0.152 |
patterns | +0.151 |
shapes | +0.149 |
tunes | +0.147 |
Neg logit contributions | |
rolley | -0.141 |
P | -0.139 |
� | -0.137 |
izz | -0.135 |
paramed | -0.134 |
Token | everything |
Token ID | 2279 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
ide | +0.089 |
bee | +0.080 |
igo | +0.076 |
iding | +0.072 |
bees | +0.071 |
Neg logit contributions | |
discount | -0.071 |
butcher | -0.070 |
rolley | -0.069 |
Found | -0.066 |
trouble | -0.066 |
Token | , |
Token ID | 11 |
Feature activation | +2.46e+00 |
Pos logit contributions | |
ide | +0.078 |
wered | +0.070 |
oted | +0.066 |
icated | +0.066 |
hip | +0.066 |
Neg logit contributions | |
pan | -0.063 |
frying | -0.063 |
etchup | -0.062 |
rolley | -0.059 |
� | -0.059 |
Token | and |
Token ID | 290 |
Feature activation | +2.20e+00 |
Pos logit contributions | |
ide | +0.063 |
planets | +0.050 |
hip | +0.048 |
seem | +0.046 |
ility | +0.046 |
Neg logit contributions | |
pan | -0.065 |
paramed | -0.064 |
cart | -0.064 |
etchup | -0.064 |
microwave | -0.063 |
Token | been |
Token ID | 587 |
Feature activation | -5.93e-01 |
Pos logit contributions | |
ide | +0.016 |
winning | +0.015 |
seem | +0.014 |
icated | +0.014 |
shooting | +0.013 |
Neg logit contributions | |
rolley | -0.033 |
butcher | -0.032 |
microwave | -0.031 |
piping | -0.030 |
frying | -0.030 |
Token | raining |
Token ID | 43079 |
Feature activation | +3.20e+00 |
Pos logit contributions | |
rolley | +0.023 |
butcher | +0.021 |
pouch | +0.020 |
piping | +0.020 |
discount | +0.020 |
Neg logit contributions | |
winning | -0.013 |
ide | -0.013 |
being | -0.012 |
not | -0.012 |
sharing | -0.012 |
Token | all |
Token ID | 477 |
Feature activation | -9.47e-01 |
Pos logit contributions | |
ide | +0.090 |
winning | +0.082 |
seem | +0.079 |
icated | +0.078 |
shooting | +0.076 |
Neg logit contributions | |
rolley | -0.098 |
butcher | -0.094 |
microwave | -0.092 |
piping | -0.090 |
frying | -0.089 |
Token | week |
Token ID | 1285 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
rolley | +0.033 |
butcher | +0.032 |
microwave | +0.032 |
piping | +0.031 |
frying | +0.031 |
Neg logit contributions | |
ide | -0.022 |
winning | -0.020 |
seem | -0.019 |
icated | -0.019 |
shooting | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +8.22e-01 |
Pos logit contributions | |
rolley | +0.053 |
butcher | +0.050 |
microwave | +0.049 |
piping | +0.049 |
P | +0.047 |
Neg logit contributions | |
ide | -0.049 |
winning | -0.047 |
seem | -0.044 |
sharing | -0.043 |
being | -0.043 |
Token | Playing |
Token ID | 23911 |
Feature activation | +8.59e+00 |
Pos logit contributions | |
winning | +0.021 |
ide | +0.021 |
seem | +0.020 |
icated | +0.020 |
sharing | +0.019 |
Neg logit contributions | |
rolley | -0.025 |
butcher | -0.024 |
microwave | -0.024 |
piping | -0.024 |
P | -0.023 |
Token | in |
Token ID | 287 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
ide | +0.209 |
winning | +0.197 |
seem | +0.192 |
icated | +0.187 |
shooting | +0.177 |
Neg logit contributions | |
rolley | -0.280 |
microwave | -0.273 |
butcher | -0.267 |
frying | -0.264 |
pan | -0.260 |
Token | the |
Token ID | 262 |
Feature activation | +4.66e-02 |
Pos logit contributions | |
ide | +0.042 |
winning | +0.041 |
being | +0.037 |
sharing | +0.036 |
seem | +0.036 |
Neg logit contributions | |
rolley | -0.082 |
butcher | -0.079 |
microwave | -0.077 |
piping | -0.076 |
waitress | -0.075 |
Token | rain |
Token ID | 6290 |
Feature activation | +4.25e+00 |
Pos logit contributions | |
winning | +0.001 |
ide | +0.001 |
shapes | +0.001 |
textures | +0.001 |
being | +0.001 |
Neg logit contributions | |
rolley | -0.002 |
butcher | -0.002 |
piping | -0.002 |
etchup | -0.002 |
microwave | -0.002 |
Token | is |
Token ID | 318 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
ide | +0.116 |
winning | +0.109 |
seem | +0.104 |
icated | +0.103 |
shooting | +0.100 |
Neg logit contributions | |
rolley | -0.130 |
microwave | -0.124 |
butcher | -0.123 |
frying | -0.119 |
piping | -0.118 |
Token | fun |
Token ID | 1257 |
Feature activation | +5.08e+00 |
Pos logit contributions | |
ide | +0.055 |
winning | +0.049 |
seem | +0.048 |
icated | +0.047 |
planets | +0.045 |
Neg logit contributions | |
rolley | -0.080 |
microwave | -0.078 |
butcher | -0.077 |
frying | -0.075 |
piping | -0.074 |
Token | You |
Token ID | 921 |
Feature activation | +1.91e-02 |
Pos logit contributions | |
rolley | +0.001 |
butcher | +0.001 |
microwave | +0.001 |
frying | +0.001 |
piping | +0.001 |
Neg logit contributions | |
ide | -0.000 |
winning | -0.000 |
seem | -0.000 |
icated | -0.000 |
sharing | -0.000 |
Token | are |
Token ID | 389 |
Feature activation | +6.64e-01 |
Pos logit contributions | |
ide | +0.001 |
winning | +0.000 |
tunes | +0.000 |
textures | +0.000 |
happening | +0.000 |
Neg logit contributions | |
use | -0.001 |
P | -0.001 |
rolley | -0.001 |
� | -0.001 |
bag | -0.001 |
Token | my |
Token ID | 616 |
Feature activation | +2.31e-01 |
Pos logit contributions | |
ide | +0.024 |
textures | +0.020 |
tunes | +0.020 |
hip | +0.020 |
seem | +0.019 |
Neg logit contributions | |
P | -0.016 |
butcher | -0.015 |
Found | -0.015 |
microwave | -0.014 |
waitress | -0.014 |
Token | friend |
Token ID | 1545 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
ide | +0.008 |
seem | +0.007 |
icated | +0.007 |
hip | +0.006 |
winning | +0.006 |
Neg logit contributions | |
butcher | -0.006 |
rolley | -0.006 |
microwave | -0.006 |
bag | -0.006 |
waitress | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
rolley | +0.054 |
butcher | +0.053 |
microwave | +0.052 |
piping | +0.050 |
frying | +0.050 |
Neg logit contributions | |
ide | -0.034 |
winning | -0.031 |
seem | -0.030 |
icated | -0.029 |
shooting | -0.028 |
Token | Winning |
Token ID | 40983 |
Feature activation | +8.56e+00 |
Pos logit contributions | |
ide | +0.003 |
winning | +0.003 |
seem | +0.002 |
icated | +0.002 |
sharing | +0.002 |
Neg logit contributions | |
rolley | -0.004 |
butcher | -0.004 |
microwave | -0.004 |
frying | -0.004 |
piping | -0.004 |
Token | is |
Token ID | 318 |
Feature activation | +3.14e+00 |
Pos logit contributions | |
ide | +0.319 |
wered | +0.293 |
hip | +0.279 |
seem | +0.271 |
oted | +0.271 |
Neg logit contributions | |
� | -0.153 |
microwave | -0.143 |
bag | -0.138 |
pan | -0.132 |
rolley | -0.128 |
Token | not |
Token ID | 407 |
Feature activation | +2.52e+00 |
Pos logit contributions | |
ide | +0.076 |
seem | +0.058 |
icated | +0.057 |
planets | +0.057 |
hip | +0.056 |
Neg logit contributions | |
butcher | -0.101 |
discount | -0.099 |
P | -0.099 |
Found | -0.097 |
use | -0.096 |
Token | important |
Token ID | 1593 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
ide | +0.068 |
seem | +0.058 |
bee | +0.056 |
planets | +0.055 |
hip | +0.052 |
Neg logit contributions | |
discount | -0.068 |
butcher | -0.066 |
Found | -0.064 |
rolley | -0.064 |
microwave | -0.062 |
Token | . |
Token ID | 13 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
rolley | +0.005 |
butcher | +0.005 |
microwave | +0.004 |
waitress | +0.004 |
frying | +0.004 |
Neg logit contributions | |
ide | -0.004 |
seem | -0.003 |
winning | -0.003 |
icated | -0.003 |
planets | -0.003 |
Token | Winning |
Token ID | 40983 |
Feature activation | +8.86e+00 |
Pos logit contributions | |
ide | +0.036 |
winning | +0.032 |
seem | +0.031 |
happening | +0.031 |
shooting | +0.030 |
Neg logit contributions | |
rolley | -0.041 |
butcher | -0.039 |
microwave | -0.039 |
piping | -0.038 |
etchup | -0.037 |
Token | I |
Token ID | 314 |
Feature activation | -2.79e+00 |
Pos logit contributions | |
ide | +0.101 |
tunes | +0.082 |
hip | +0.082 |
icated | +0.080 |
melodies | +0.076 |
Neg logit contributions | |
pan | -0.069 |
microwave | -0.065 |
etchup | -0.063 |
bag | -0.062 |
ha | -0.062 |
Token | am |
Token ID | 716 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
rolley | +0.113 |
butcher | +0.108 |
microwave | +0.107 |
piping | +0.105 |
frying | +0.104 |
Neg logit contributions | |
ide | -0.051 |
winning | -0.047 |
seem | -0.046 |
icated | -0.042 |
shooting | -0.041 |
Token | sleepy |
Token ID | 42023 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
ide | +0.029 |
winning | +0.028 |
seem | +0.026 |
icated | +0.025 |
shooting | +0.025 |
Neg logit contributions | |
rolley | -0.055 |
butcher | -0.052 |
microwave | -0.052 |
piping | -0.051 |
frying | -0.050 |
Token | now |
Token ID | 783 |
Feature activation | -2.38e+00 |
Pos logit contributions | |
ide | +0.026 |
seem | +0.022 |
winning | +0.021 |
sometimes | +0.021 |
being | +0.021 |
Neg logit contributions | |
rolley | -0.058 |
butcher | -0.055 |
piping | -0.055 |
microwave | -0.054 |
pouch | -0.053 |
Token | . |
Token ID | 13 |
Feature activation | +5.32e-01 |
Pos logit contributions | |
rolley | +0.088 |
butcher | +0.083 |
microwave | +0.082 |
piping | +0.081 |
frying | +0.080 |
Neg logit contributions | |
ide | -0.052 |
winning | -0.049 |
seem | -0.046 |
icated | -0.045 |
being | -0.045 |
Token | Playing |
Token ID | 23911 |
Feature activation | +8.54e+00 |
Pos logit contributions | |
winning | +0.013 |
ide | +0.013 |
icated | +0.013 |
seem | +0.012 |
sharing | +0.012 |
Neg logit contributions | |
rolley | -0.017 |
butcher | -0.016 |
frying | -0.016 |
microwave | -0.015 |
piping | -0.015 |
Token | with |
Token ID | 351 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
ide | +0.211 |
seem | +0.197 |
winning | +0.193 |
hip | +0.184 |
icated | +0.182 |
Neg logit contributions | |
microwave | -0.280 |
pan | -0.258 |
rolley | -0.253 |
P | -0.249 |
� | -0.248 |
Token | stones |
Token ID | 14966 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
ide | +0.046 |
winning | +0.043 |
seem | +0.041 |
icated | +0.041 |
shooting | +0.040 |
Neg logit contributions | |
rolley | -0.060 |
butcher | -0.057 |
microwave | -0.057 |
piping | -0.055 |
frying | -0.055 |
Token | is |
Token ID | 318 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
ide | +0.058 |
winning | +0.055 |
icated | +0.050 |
seem | +0.049 |
shooting | +0.049 |
Neg logit contributions | |
rolley | -0.075 |
microwave | -0.073 |
butcher | -0.072 |
piping | -0.071 |
pouch | -0.070 |
Token | hard |
Token ID | 1327 |
Feature activation | -4.78e-01 |
Pos logit contributions | |
ide | +0.052 |
winning | +0.045 |
seem | +0.045 |
icated | +0.044 |
hip | +0.042 |
Neg logit contributions | |
rolley | -0.068 |
microwave | -0.068 |
butcher | -0.067 |
frying | -0.065 |
piping | -0.065 |
Token | work |
Token ID | 670 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
rolley | +0.018 |
butcher | +0.018 |
microwave | +0.018 |
piping | +0.017 |
frying | +0.017 |
Neg logit contributions | |
ide | -0.009 |
winning | -0.009 |
seem | -0.008 |
icated | -0.008 |
shooting | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
icated | +0.014 |
winning | +0.014 |
ide | +0.014 |
matter | +0.014 |
seem | +0.013 |
Neg logit contributions | |
rolley | -0.019 |
pan | -0.019 |
microwave | -0.019 |
bag | -0.018 |
pouch | -0.018 |
Token | It |
Token ID | 632 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
ide | +0.004 |
winning | +0.004 |
seem | +0.004 |
shapes | +0.004 |
icated | +0.004 |
Neg logit contributions | |
rolley | -0.006 |
butcher | -0.006 |
microwave | -0.006 |
piping | -0.005 |
etchup | -0.005 |
Token | said |
Token ID | 531 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
ide | +0.055 |
winning | +0.051 |
icated | +0.048 |
seem | +0.048 |
sharing | +0.046 |
Neg logit contributions | |
rolley | -0.082 |
butcher | -0.079 |
microwave | -0.079 |
frying | -0.076 |
piping | -0.076 |
Token | ' |
Token ID | 705 |
Feature activation | -2.53e+00 |
Pos logit contributions | |
butcher | +0.062 |
microwave | +0.061 |
frying | +0.060 |
paramed | +0.060 |
piping | +0.060 |
Neg logit contributions | |
ide | -0.057 |
icated | -0.051 |
winning | -0.050 |
textures | -0.050 |
matter | -0.048 |
Token | No |
Token ID | 2949 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
rolley | +0.073 |
microwave | +0.073 |
butcher | +0.070 |
frying | +0.068 |
waitress | +0.068 |
Neg logit contributions | |
ide | -0.072 |
seem | -0.068 |
winning | -0.067 |
icated | -0.065 |
shooting | -0.062 |
Token | Playing |
Token ID | 23911 |
Feature activation | +8.54e+00 |
Pos logit contributions | |
rolley | +0.035 |
butcher | +0.034 |
microwave | +0.034 |
frying | +0.032 |
piping | +0.032 |
Neg logit contributions | |
ide | -0.029 |
winning | -0.027 |
planets | -0.025 |
icated | -0.025 |
seem | -0.025 |
Token | All |
Token ID | 1439 |
Feature activation | +4.04e+00 |
Pos logit contributions | |
ide | +0.241 |
winning | +0.225 |
seem | +0.218 |
icated | +0.214 |
shooting | +0.205 |
Neg logit contributions | |
rolley | -0.253 |
butcher | -0.243 |
microwave | -0.239 |
frying | -0.231 |
piping | -0.226 |
Token | owed |
Token ID | 6972 |
Feature activation | +3.03e+00 |
Pos logit contributions | |
ide | +0.098 |
winning | +0.092 |
seem | +0.085 |
icated | +0.083 |
shooting | +0.083 |
Neg logit contributions | |
rolley | -0.134 |
butcher | -0.126 |
microwave | -0.124 |
frying | -0.121 |
pan | -0.120 |
Token | '. |
Token ID | 4458 |
Feature activation | -8.54e-01 |
Pos logit contributions | |
ide | +0.079 |
winning | +0.074 |
seem | +0.071 |
icated | +0.069 |
shooting | +0.068 |
Neg logit contributions | |
rolley | -0.096 |
butcher | -0.092 |
microwave | -0.091 |
frying | -0.088 |
piping | -0.087 |
Token | Henry |
Token ID | 8616 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
rolley | +0.031 |
butcher | +0.029 |
microwave | +0.028 |
paramed | +0.028 |
frying | +0.028 |
Neg logit contributions | |
ide | -0.019 |
seem | -0.018 |
winning | -0.017 |
being | -0.016 |
icated | -0.016 |
Token | frowned |
Token ID | 32198 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
ide | +0.025 |
winning | +0.025 |
icated | +0.024 |
seem | +0.022 |
sharing | +0.022 |
Neg logit contributions | |
rolley | -0.033 |
butcher | -0.031 |
microwave | -0.031 |
frying | -0.030 |
pouch | -0.030 |
Token | I |
Token ID | 314 |
Feature activation | -2.80e+00 |
Pos logit contributions | |
ide | +0.046 |
winning | +0.043 |
seem | +0.041 |
icated | +0.040 |
shooting | +0.039 |
Neg logit contributions | |
rolley | -0.066 |
butcher | -0.063 |
microwave | -0.063 |
piping | -0.060 |
frying | -0.060 |
Token | love |
Token ID | 1842 |
Feature activation | +3.31e+00 |
Pos logit contributions | |
rolley | +0.086 |
butcher | +0.082 |
microwave | +0.082 |
piping | +0.079 |
frying | +0.079 |
Neg logit contributions | |
ide | -0.077 |
winning | -0.074 |
seem | -0.072 |
sharing | -0.068 |
icated | -0.068 |
Token | you |
Token ID | 345 |
Feature activation | +2.20e+00 |
Pos logit contributions | |
ide | +0.085 |
seem | +0.073 |
winning | +0.071 |
icated | +0.068 |
hip | +0.068 |
Neg logit contributions | |
rolley | -0.104 |
butcher | -0.104 |
microwave | -0.101 |
etchup | -0.101 |
piping | -0.099 |
Token | both |
Token ID | 1111 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
ide | +0.059 |
winning | +0.054 |
seem | +0.053 |
icated | +0.051 |
shooting | +0.051 |
Neg logit contributions | |
rolley | -0.067 |
butcher | -0.065 |
microwave | -0.064 |
waitress | -0.062 |
piping | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | +9.22e-01 |
Pos logit contributions | |
rolley | +0.008 |
butcher | +0.007 |
microwave | +0.007 |
waitress | +0.007 |
piping | +0.007 |
Neg logit contributions | |
ide | -0.009 |
winning | -0.008 |
seem | -0.008 |
icated | -0.008 |
shooting | -0.008 |
Token | Playing |
Token ID | 23911 |
Feature activation | +8.52e+00 |
Pos logit contributions | |
ide | +0.026 |
winning | +0.025 |
seem | +0.024 |
icated | +0.024 |
sharing | +0.023 |
Neg logit contributions | |
rolley | -0.027 |
butcher | -0.026 |
microwave | -0.026 |
frying | -0.025 |
piping | -0.025 |
Token | together |
Token ID | 1978 |
Feature activation | +2.47e+00 |
Pos logit contributions | |
ide | +0.214 |
seem | +0.199 |
icated | +0.184 |
winning | +0.179 |
textures | +0.178 |
Neg logit contributions | |
P | -0.269 |
microwave | -0.258 |
etchup | -0.252 |
� | -0.252 |
butcher | -0.251 |
Token | is |
Token ID | 318 |
Feature activation | +2.65e+00 |
Pos logit contributions | |
winning | +0.066 |
ide | +0.066 |
icated | +0.061 |
aching | +0.058 |
happening | +0.058 |
Neg logit contributions | |
� | -0.064 |
rolley | -0.063 |
use | -0.063 |
pouch | -0.062 |
bag | -0.062 |
Token | more |
Token ID | 517 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
ide | +0.067 |
seem | +0.059 |
winning | +0.056 |
icated | +0.056 |
hip | +0.053 |
Neg logit contributions | |
butcher | -0.085 |
rolley | -0.084 |
microwave | -0.083 |
piping | -0.081 |
waitress | -0.079 |
Token | fun |
Token ID | 1257 |
Feature activation | +5.03e+00 |
Pos logit contributions | |
ide | +0.036 |
seem | +0.032 |
winning | +0.030 |
icated | +0.030 |
hip | +0.028 |
Neg logit contributions | |
rolley | -0.040 |
butcher | -0.039 |
microwave | -0.038 |
P | -0.038 |
piping | -0.037 |
Token | than |
Token ID | 621 |
Feature activation | +3.48e+00 |
Pos logit contributions | |
ide | +0.147 |
winning | +0.133 |
seem | +0.132 |
icated | +0.129 |
shooting | +0.123 |
Neg logit contributions | |
rolley | -0.139 |
microwave | -0.131 |
butcher | -0.131 |
pan | -0.128 |
etchup | -0.127 |
Token | you |
Token ID | 345 |
Feature activation | +6.23e-01 |
Pos logit contributions | |
rolley | +0.021 |
butcher | +0.021 |
microwave | +0.020 |
frying | +0.020 |
piping | +0.020 |
Neg logit contributions | |
ide | -0.017 |
winning | -0.015 |
seem | -0.014 |
icated | -0.014 |
shooting | -0.013 |
Token | to |
Token ID | 284 |
Feature activation | +2.41e-01 |
Pos logit contributions | |
ide | +0.018 |
winning | +0.015 |
icated | +0.014 |
shooting | +0.014 |
hip | +0.014 |
Neg logit contributions | |
rolley | -0.017 |
butcher | -0.016 |
bag | -0.016 |
frying | -0.016 |
microwave | -0.016 |
Token | come |
Token ID | 1282 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
ide | +0.006 |
planets | +0.005 |
icated | +0.005 |
sizes | +0.005 |
winning | +0.005 |
Neg logit contributions | |
pouch | -0.007 |
butcher | -0.007 |
use | -0.007 |
frying | -0.007 |
rolley | -0.007 |
Token | inside |
Token ID | 2641 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
ide | +0.001 |
winning | +0.001 |
seem | +0.001 |
shooting | +0.001 |
icated | +0.001 |
Neg logit contributions | |
rolley | -0.001 |
butcher | -0.001 |
microwave | -0.001 |
piping | -0.001 |
frying | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +8.53e-01 |
Pos logit contributions | |
rolley | +0.008 |
butcher | +0.008 |
microwave | +0.008 |
piping | +0.008 |
frying | +0.008 |
Neg logit contributions | |
ide | -0.004 |
winning | -0.004 |
seem | -0.004 |
icated | -0.004 |
shooting | -0.003 |
Token | Playing |
Token ID | 23911 |
Feature activation | +8.51e+00 |
Pos logit contributions | |
ide | +0.024 |
winning | +0.022 |
seem | +0.022 |
icated | +0.021 |
shooting | +0.021 |
Neg logit contributions | |
rolley | -0.026 |
butcher | -0.024 |
microwave | -0.024 |
piping | -0.023 |
frying | -0.023 |
Token | outside |
Token ID | 2354 |
Feature activation | +3.65e+00 |
Pos logit contributions | |
ide | +0.230 |
icated | +0.207 |
winning | +0.205 |
seem | +0.205 |
involve | +0.191 |
Neg logit contributions | |
microwave | -0.236 |
rolley | -0.232 |
P | -0.229 |
butcher | -0.225 |
pan | -0.223 |
Token | is |
Token ID | 318 |
Feature activation | +2.46e+00 |
Pos logit contributions | |
ide | +0.097 |
winning | +0.090 |
icated | +0.085 |
seem | +0.083 |
shooting | +0.083 |
Neg logit contributions | |
rolley | -0.113 |
microwave | -0.106 |
butcher | -0.105 |
frying | -0.102 |
pan | -0.101 |
Token | fun |
Token ID | 1257 |
Feature activation | +5.12e+00 |
Pos logit contributions | |
ide | +0.063 |
icated | +0.051 |
seem | +0.051 |
winning | +0.050 |
planets | +0.049 |
Neg logit contributions | |
butcher | -0.079 |
rolley | -0.078 |
microwave | -0.077 |
frying | -0.075 |
paramed | -0.074 |
Token | , |
Token ID | 11 |
Feature activation | +2.73e+00 |
Pos logit contributions | |
ide | +0.131 |
icated | +0.119 |
winning | +0.113 |
wered | +0.110 |
seem | +0.110 |
Neg logit contributions | |
rolley | -0.153 |
microwave | -0.146 |
pan | -0.146 |
butcher | -0.144 |
bag | -0.144 |
Token | but |
Token ID | 475 |
Feature activation | +2.86e+00 |
Pos logit contributions | |
ide | +0.067 |
icated | +0.052 |
hip | +0.049 |
planets | +0.049 |
bee | +0.047 |
Neg logit contributions | |
microwave | -0.084 |
pan | -0.082 |
paramed | -0.081 |
rolley | -0.081 |
bag | -0.080 |
Token | Sing |
Token ID | 5573 |
Feature activation | +5.31e+00 |
Pos logit contributions | |
ide | +0.007 |
winning | +0.007 |
seem | +0.007 |
icated | +0.007 |
shooting | +0.006 |
Neg logit contributions | |
rolley | -0.008 |
butcher | -0.008 |
microwave | -0.008 |
piping | -0.007 |
frying | -0.007 |
Token | ing |
Token ID | 278 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
ide | +0.153 |
winning | +0.152 |
textures | +0.147 |
matter | +0.146 |
seem | +0.146 |
Neg logit contributions | |
microwave | -0.128 |
rolley | -0.126 |
etchup | -0.122 |
butcher | -0.121 |
� | -0.120 |
Token | is |
Token ID | 318 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
ide | +0.047 |
winning | +0.045 |
seem | +0.042 |
icated | +0.041 |
shooting | +0.040 |
Neg logit contributions | |
rolley | -0.044 |
microwave | -0.044 |
butcher | -0.042 |
pan | -0.041 |
piping | -0.041 |
Token | boring |
Token ID | 14262 |
Feature activation | +2.05e+00 |
Pos logit contributions | |
ide | +0.037 |
seem | +0.031 |
hip | +0.029 |
icated | +0.029 |
winning | +0.029 |
Neg logit contributions | |
microwave | -0.036 |
butcher | -0.036 |
piping | -0.035 |
waitress | -0.034 |
frying | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | +9.53e-01 |
Pos logit contributions | |
ide | +0.054 |
winning | +0.050 |
seem | +0.049 |
icated | +0.047 |
shooting | +0.047 |
Neg logit contributions | |
rolley | -0.062 |
microwave | -0.060 |
butcher | -0.060 |
waitress | -0.057 |
frying | -0.056 |
Token | Playing |
Token ID | 23911 |
Feature activation | +8.49e+00 |
Pos logit contributions | |
ide | +0.028 |
winning | +0.026 |
seem | +0.025 |
icated | +0.024 |
shooting | +0.024 |
Neg logit contributions | |
rolley | -0.027 |
microwave | -0.026 |
butcher | -0.026 |
piping | -0.025 |
frying | -0.025 |
Token | with |
Token ID | 351 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
ide | +0.221 |
seem | +0.204 |
textures | +0.188 |
winning | +0.185 |
icated | +0.184 |
Neg logit contributions | |
microwave | -0.245 |
P | -0.243 |
� | -0.243 |
pan | -0.237 |
etchup | -0.232 |
Token | the |
Token ID | 262 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
ide | +0.050 |
seem | +0.044 |
winning | +0.043 |
icated | +0.040 |
hip | +0.039 |
Neg logit contributions | |
rolley | -0.053 |
microwave | -0.052 |
butcher | -0.052 |
pouch | -0.050 |
piping | -0.050 |
Token | ball |
Token ID | 2613 |
Feature activation | -4.54e-02 |
Pos logit contributions | |
ide | +0.003 |
seem | +0.003 |
winning | +0.003 |
icated | +0.003 |
hip | +0.003 |
Neg logit contributions | |
microwave | -0.003 |
rolley | -0.003 |
butcher | -0.003 |
bag | -0.003 |
pouch | -0.003 |
Token | is |
Token ID | 318 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
rolley | +0.001 |
microwave | +0.001 |
butcher | +0.001 |
pouch | +0.001 |
pan | +0.001 |
Neg logit contributions | |
ide | -0.001 |
winning | -0.001 |
icated | -0.001 |
seem | -0.001 |
shooting | -0.001 |
Token | fun |
Token ID | 1257 |
Feature activation | +4.82e+00 |
Pos logit contributions | |
ide | +0.057 |
seem | +0.047 |
icated | +0.044 |
hip | +0.044 |
winning | +0.043 |
Neg logit contributions | |
microwave | -0.064 |
butcher | -0.063 |
frying | -0.062 |
piping | -0.062 |
pouch | -0.061 |
Token | of |
Token ID | 286 |
Feature activation | -2.17e-01 |
Pos logit contributions | |
piping | +0.152 |
frying | +0.147 |
rolley | +0.145 |
microwave | +0.144 |
butcher | +0.141 |
Neg logit contributions | |
sometimes | -0.056 |
winning | -0.055 |
being | -0.054 |
seem | -0.054 |
not | -0.053 |
Token | the |
Token ID | 262 |
Feature activation | -6.61e-01 |
Pos logit contributions | |
rolley | +0.008 |
butcher | +0.008 |
microwave | +0.008 |
piping | +0.007 |
frying | +0.007 |
Neg logit contributions | |
ide | -0.005 |
winning | -0.004 |
seem | -0.004 |
icated | -0.004 |
textures | -0.004 |
Token | story |
Token ID | 1621 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
butcher | +0.019 |
microwave | +0.019 |
waitress | +0.018 |
paramed | +0.018 |
pan | +0.017 |
Neg logit contributions | |
ide | -0.020 |
winning | -0.018 |
icated | -0.018 |
hip | -0.018 |
textures | -0.017 |
Token | is |
Token ID | 318 |
Feature activation | +9.62e-01 |
Pos logit contributions | |
ide | +0.036 |
winning | +0.033 |
seem | +0.032 |
icated | +0.031 |
shooting | +0.031 |
Neg logit contributions | |
rolley | -0.046 |
butcher | -0.044 |
microwave | -0.044 |
piping | -0.042 |
frying | -0.042 |
Token | : |
Token ID | 25 |
Feature activation | -5.02e-01 |
Pos logit contributions | |
ide | +0.022 |
winning | +0.021 |
seem | +0.020 |
icated | +0.019 |
shooting | +0.019 |
Neg logit contributions | |
rolley | -0.034 |
butcher | -0.032 |
microwave | -0.032 |
piping | -0.031 |
frying | -0.031 |
Token | Playing |
Token ID | 23911 |
Feature activation | +8.45e+00 |
Pos logit contributions | |
butcher | +0.012 |
etchup | +0.012 |
use | +0.012 |
microwave | +0.011 |
paramed | +0.011 |
Neg logit contributions | |
ide | -0.018 |
textures | -0.014 |
planets | -0.014 |
happening | -0.013 |
patterns | -0.013 |
Token | with |
Token ID | 351 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
ide | +0.218 |
seem | +0.189 |
winning | +0.183 |
icated | +0.182 |
textures | +0.169 |
Neg logit contributions | |
microwave | -0.259 |
butcher | -0.259 |
rolley | -0.258 |
P | -0.256 |
etchup | -0.245 |
Token | fire |
Token ID | 2046 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
ide | +0.042 |
winning | +0.038 |
seem | +0.037 |
icated | +0.036 |
shooting | +0.035 |
Neg logit contributions | |
rolley | -0.052 |
butcher | -0.050 |
microwave | -0.050 |
frying | -0.048 |
piping | -0.048 |
Token | is |
Token ID | 318 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
ide | +0.060 |
winning | +0.056 |
icated | +0.054 |
seem | +0.052 |
shooting | +0.050 |
Neg logit contributions | |
rolley | -0.063 |
microwave | -0.060 |
butcher | -0.059 |
etchup | -0.058 |
pan | -0.057 |
Token | stupid |
Token ID | 8531 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
ide | +0.049 |
icated | +0.036 |
seem | +0.036 |
textures | +0.034 |
hip | +0.034 |
Neg logit contributions | |
butcher | -0.072 |
microwave | -0.069 |
rolley | -0.068 |
frying | -0.068 |
paramed | -0.066 |
Token | and |
Token ID | 290 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
ide | +0.027 |
winning | +0.022 |
seem | +0.022 |
icated | +0.022 |
wered | +0.021 |
Neg logit contributions | |
rolley | -0.037 |
butcher | -0.036 |
microwave | -0.036 |
bag | -0.035 |
etchup | -0.035 |
Token | to |
Token ID | 284 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
rolley | +0.023 |
microwave | +0.022 |
butcher | +0.022 |
piping | +0.021 |
frying | +0.021 |
Neg logit contributions | |
winning | -0.016 |
being | -0.015 |
sharing | -0.014 |
not | -0.014 |
shapes | -0.014 |
Token | be |
Token ID | 307 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
ide | +0.005 |
winning | +0.004 |
icated | +0.004 |
seem | +0.004 |
planets | +0.004 |
Neg logit contributions | |
rolley | -0.008 |
butcher | -0.008 |
microwave | -0.008 |
frying | -0.007 |
pouch | -0.007 |
Token | careful |
Token ID | 8161 |
Feature activation | +2.95e+00 |
Pos logit contributions | |
winning | +0.025 |
being | +0.024 |
not | +0.023 |
sharing | +0.023 |
shooting | +0.022 |
Neg logit contributions | |
rolley | -0.049 |
microwave | -0.046 |
butcher | -0.045 |
waitress | -0.044 |
piping | -0.044 |
Token | too |
Token ID | 1165 |
Feature activation | -7.78e-01 |
Pos logit contributions | |
ide | +0.068 |
icated | +0.067 |
seem | +0.064 |
planets | +0.064 |
winning | +0.063 |
Neg logit contributions | |
rolley | -0.092 |
frying | -0.088 |
pan | -0.088 |
microwave | -0.085 |
butcher | -0.085 |
Token | . |
Token ID | 13 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
rolley | +0.027 |
butcher | +0.026 |
microwave | +0.026 |
piping | +0.025 |
frying | +0.025 |
Neg logit contributions | |
ide | -0.019 |
winning | -0.017 |
seem | -0.016 |
icated | -0.016 |
shooting | -0.016 |
Token | Playing |
Token ID | 23911 |
Feature activation | +8.44e+00 |
Pos logit contributions | |
ide | +0.032 |
winning | +0.031 |
seem | +0.030 |
icated | +0.029 |
sharing | +0.028 |
Neg logit contributions | |
rolley | -0.036 |
butcher | -0.034 |
microwave | -0.033 |
frying | -0.032 |
piping | -0.032 |
Token | with |
Token ID | 351 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
ide | +0.214 |
seem | +0.193 |
winning | +0.189 |
icated | +0.186 |
textures | +0.175 |
Neg logit contributions | |
rolley | -0.268 |
microwave | -0.267 |
butcher | -0.261 |
pan | -0.255 |
P | -0.254 |
Token | the |
Token ID | 262 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
ide | +0.041 |
winning | +0.039 |
seem | +0.037 |
icated | +0.037 |
shooting | +0.036 |
Neg logit contributions | |
rolley | -0.059 |
butcher | -0.056 |
microwave | -0.055 |
piping | -0.054 |
frying | -0.053 |
Token | hose |
Token ID | 31489 |
Feature activation | +7.03e-01 |
Pos logit contributions | |
ide | +0.004 |
winning | +0.004 |
seem | +0.004 |
icated | +0.003 |
sharing | +0.003 |
Neg logit contributions | |
rolley | -0.003 |
butcher | -0.003 |
microwave | -0.003 |
frying | -0.003 |
piping | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
ide | +0.016 |
winning | +0.015 |
icated | +0.014 |
seem | +0.014 |
shooting | +0.013 |
Neg logit contributions | |
rolley | -0.024 |
microwave | -0.024 |
butcher | -0.023 |
frying | -0.023 |
piping | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | -7.36e-01 |
Pos logit contributions | |
ide | +0.034 |
winning | +0.032 |
seem | +0.030 |
icated | +0.029 |
shooting | +0.028 |
Neg logit contributions | |
rolley | -0.063 |
butcher | -0.061 |
microwave | -0.060 |
piping | -0.059 |
frying | -0.059 |
Token | moral |
Token ID | 6573 |
Feature activation | -3.75e+00 |
Pos logit contributions | |
ide | +0.007 |
seem | +0.006 |
icated | +0.006 |
winning | +0.006 |
hip | +0.006 |
Neg logit contributions | |
microwave | -0.007 |
butcher | -0.007 |
rolley | -0.007 |
piping | -0.007 |
waitress | -0.007 |
Token | of |
Token ID | 286 |
Feature activation | -3.44e-02 |
Pos logit contributions | |
rolley | +0.133 |
butcher | +0.130 |
microwave | +0.129 |
piping | +0.128 |
frying | +0.128 |
Neg logit contributions | |
ide | -0.083 |
winning | -0.076 |
seem | -0.073 |
icated | -0.072 |
shapes | -0.070 |
Token | the |
Token ID | 262 |
Feature activation | -3.82e-01 |
Pos logit contributions | |
rolley | +0.001 |
butcher | +0.001 |
microwave | +0.001 |
piping | +0.001 |
frying | +0.001 |
Neg logit contributions | |
ide | -0.001 |
winning | -0.001 |
seem | -0.001 |
icated | -0.001 |
textures | -0.001 |
Token | story |
Token ID | 1621 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
microwave | +0.011 |
butcher | +0.011 |
waitress | +0.011 |
paramed | +0.011 |
piping | +0.011 |
Neg logit contributions | |
ide | -0.011 |
hip | -0.010 |
textures | -0.009 |
winning | -0.009 |
icated | -0.009 |
Token | : |
Token ID | 25 |
Feature activation | -6.31e-01 |
Pos logit contributions | |
ide | +0.037 |
winning | +0.035 |
seem | +0.033 |
icated | +0.033 |
shooting | +0.032 |
Neg logit contributions | |
rolley | -0.053 |
butcher | -0.051 |
microwave | -0.050 |
piping | -0.048 |
waitress | -0.048 |
Token | Playing |
Token ID | 23911 |
Feature activation | +8.43e+00 |
Pos logit contributions | |
rolley | +0.016 |
butcher | +0.016 |
microwave | +0.016 |
discount | +0.015 |
etchup | +0.015 |
Neg logit contributions | |
ide | -0.019 |
happening | -0.016 |
seem | -0.016 |
textures | -0.016 |
planets | -0.016 |
Token | games |
Token ID | 1830 |
Feature activation | +4.69e+00 |
Pos logit contributions | |
ide | +0.189 |
seem | +0.171 |
winning | +0.169 |
icated | +0.164 |
textures | +0.155 |
Neg logit contributions | |
rolley | -0.284 |
butcher | -0.283 |
microwave | -0.281 |
P | -0.274 |
frying | -0.267 |
Token | and |
Token ID | 290 |
Feature activation | +2.37e+00 |
Pos logit contributions | |
ide | +0.130 |
icated | +0.118 |
winning | +0.116 |
aching | +0.106 |
tunes | +0.105 |
Neg logit contributions | |
rolley | -0.127 |
bag | -0.126 |
microwave | -0.121 |
pouch | -0.121 |
pan | -0.120 |
Token | having |
Token ID | 1719 |
Feature activation | +3.43e+00 |
Pos logit contributions | |
ide | +0.056 |
hip | +0.043 |
icated | +0.042 |
seem | +0.041 |
winning | +0.040 |
Neg logit contributions | |
butcher | -0.078 |
frying | -0.077 |
rolley | -0.075 |
microwave | -0.075 |
piping | -0.075 |
Token | fun |
Token ID | 1257 |
Feature activation | +5.75e+00 |
Pos logit contributions | |
ide | +0.090 |
winning | +0.074 |
seem | +0.074 |
icated | +0.074 |
happening | +0.073 |
Neg logit contributions | |
rolley | -0.104 |
butcher | -0.103 |
piping | -0.102 |
microwave | -0.101 |
frying | -0.098 |
Token | are |
Token ID | 389 |
Feature activation | +3.03e+00 |
Pos logit contributions | |
ide | +0.158 |
icated | +0.148 |
seem | +0.145 |
wered | +0.144 |
oted | +0.132 |
Neg logit contributions | |
rolley | -0.149 |
pan | -0.145 |
frying | -0.141 |
butcher | -0.141 |
etchup | -0.140 |
Token |
|
Token ID | 198 |
Feature activation | +5.01e-01 |
Pos logit contributions | |
ide | +0.017 |
winning | +0.015 |
seem | +0.014 |
shooting | +0.014 |
icated | +0.014 |
Neg logit contributions | |
rolley | -0.026 |
butcher | -0.024 |
microwave | -0.024 |
frying | -0.024 |
waitress | -0.024 |
Token | He |
Token ID | 1544 |
Feature activation | -6.78e-01 |
Pos logit contributions | |
ide | +0.011 |
winning | +0.010 |
seem | +0.009 |
icated | +0.009 |
shooting | +0.009 |
Neg logit contributions | |
rolley | -0.019 |
microwave | -0.018 |
butcher | -0.018 |
frying | -0.017 |
piping | -0.017 |
Token | pointed |
Token ID | 6235 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
rolley | +0.022 |
butcher | +0.021 |
microwave | +0.020 |
piping | +0.020 |
frying | +0.020 |
Neg logit contributions | |
ide | -0.018 |
winning | -0.016 |
seem | -0.016 |
shooting | -0.015 |
icated | -0.015 |
Token | his |
Token ID | 465 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
rolley | +0.044 |
butcher | +0.042 |
microwave | +0.041 |
piping | +0.040 |
frying | +0.040 |
Neg logit contributions | |
ide | -0.034 |
winning | -0.031 |
seem | -0.030 |
icated | -0.029 |
shooting | -0.029 |
Token | magn |
Token ID | 7842 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
rolley | +0.050 |
microwave | +0.050 |
butcher | +0.049 |
piping | +0.047 |
frying | +0.047 |
Neg logit contributions | |
ide | -0.068 |
winning | -0.062 |
icated | -0.060 |
seem | -0.060 |
sharing | -0.058 |
Token | ifying |
Token ID | 4035 |
Feature activation | -8.93e+00 |
Pos logit contributions | |
rolley | +0.026 |
butcher | +0.022 |
microwave | +0.022 |
piping | +0.020 |
frying | +0.020 |
Neg logit contributions | |
ide | -0.093 |
winning | -0.090 |
seem | -0.088 |
icated | -0.087 |
sharing | -0.085 |
Token | glass |
Token ID | 5405 |
Feature activation | -2.98e+00 |
Pos logit contributions | |
rolley | +0.302 |
butcher | +0.295 |
stret | +0.283 |
use | +0.283 |
piping | +0.282 |
Neg logit contributions | |
shapes | -0.196 |
color | -0.195 |
planet | -0.182 |
glow | -0.179 |
not | -0.177 |
Token | at |
Token ID | 379 |
Feature activation | -5.49e-01 |
Pos logit contributions | |
rolley | +0.105 |
butcher | +0.101 |
microwave | +0.099 |
piping | +0.097 |
frying | +0.096 |
Neg logit contributions | |
ide | -0.069 |
winning | -0.063 |
seem | -0.060 |
icated | -0.058 |
shooting | -0.057 |
Token | different |
Token ID | 1180 |
Feature activation | +2.70e+00 |
Pos logit contributions | |
rolley | +0.018 |
microwave | +0.018 |
butcher | +0.018 |
pan | +0.017 |
bag | +0.017 |
Neg logit contributions | |
ide | -0.013 |
winning | -0.012 |
icated | -0.012 |
seem | -0.012 |
sharing | -0.011 |
Token | classmates |
Token ID | 28999 |
Feature activation | +9.79e-01 |
Pos logit contributions | |
ide | +0.054 |
winning | +0.049 |
seem | +0.047 |
icated | +0.047 |
hip | +0.045 |
Neg logit contributions | |
rolley | -0.098 |
butcher | -0.093 |
frying | -0.092 |
microwave | -0.092 |
pan | -0.091 |
Token | and |
Token ID | 290 |
Feature activation | +6.24e-01 |
Pos logit contributions | |
ide | +0.025 |
icated | +0.025 |
textures | +0.025 |
wered | +0.024 |
matter | +0.024 |
Neg logit contributions | |
pan | -0.027 |
bag | -0.026 |
butcher | -0.026 |
microwave | -0.026 |
frying | -0.026 |
Token | hat |
Token ID | 6877 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
rolley | +0.028 |
butcher | +0.025 |
microwave | +0.024 |
paramed | +0.024 |
enn | +0.024 |
Neg logit contributions | |
winning | -0.018 |
ide | -0.018 |
shapes | -0.017 |
being | -0.017 |
shooting | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | +5.23e-02 |
Pos logit contributions | |
rolley | +0.065 |
microwave | +0.064 |
butcher | +0.063 |
pouch | +0.061 |
frying | +0.061 |
Neg logit contributions | |
ide | -0.028 |
winning | -0.027 |
seem | -0.025 |
icated | -0.025 |
textures | -0.023 |
Token | grabbed |
Token ID | 13176 |
Feature activation | -3.08e+00 |
Pos logit contributions | |
shapes | +0.001 |
planets | +0.001 |
songs | +0.001 |
not | +0.001 |
patterns | +0.001 |
Neg logit contributions | |
rolley | -0.002 |
butcher | -0.002 |
enn | -0.002 |
paramed | -0.002 |
piping | -0.002 |
Token | her |
Token ID | 607 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
rolley | +0.132 |
butcher | +0.124 |
enn | +0.118 |
frying | +0.118 |
paramed | +0.118 |
Neg logit contributions | |
ide | -0.048 |
shapes | -0.047 |
seem | -0.044 |
winning | -0.044 |
shooting | -0.043 |
Token | magn |
Token ID | 7842 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
rolley | +0.054 |
butcher | +0.050 |
microwave | +0.048 |
piping | +0.047 |
frying | +0.047 |
Neg logit contributions | |
ide | -0.047 |
winning | -0.045 |
shapes | -0.043 |
seem | -0.043 |
shooting | -0.042 |
Token | ifying |
Token ID | 4035 |
Feature activation | -8.48e+00 |
Pos logit contributions | |
rolley | +0.023 |
butcher | +0.022 |
microwave | +0.021 |
piping | +0.019 |
frying | +0.019 |
Neg logit contributions | |
ide | -0.068 |
winning | -0.063 |
seem | -0.062 |
icated | -0.061 |
shooting | -0.061 |
Token | glass |
Token ID | 5405 |
Feature activation | -2.62e+00 |
Pos logit contributions | |
rolley | +0.316 |
butcher | +0.310 |
ilia | +0.294 |
use | +0.285 |
aked | +0.283 |
Neg logit contributions | |
color | -0.157 |
not | -0.150 |
glow | -0.148 |
shapes | -0.147 |
planet | -0.146 |
Token | . |
Token ID | 13 |
Feature activation | -3.51e-01 |
Pos logit contributions | |
rolley | +0.103 |
butcher | +0.102 |
piping | +0.100 |
microwave | +0.097 |
frying | +0.097 |
Neg logit contributions | |
ide | -0.052 |
shapes | -0.042 |
textures | -0.038 |
winning | -0.037 |
patterns | -0.037 |
Token | She |
Token ID | 1375 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
rolley | +0.013 |
butcher | +0.012 |
microwave | +0.011 |
frying | +0.011 |
piping | +0.011 |
Neg logit contributions | |
ide | -0.008 |
winning | -0.008 |
seem | -0.008 |
shapes | -0.007 |
sharing | -0.007 |
Token | was |
Token ID | 373 |
Feature activation | +5.17e-02 |
Pos logit contributions | |
rolley | +0.047 |
butcher | +0.045 |
microwave | +0.043 |
piping | +0.043 |
frying | +0.042 |
Neg logit contributions | |
ide | -0.034 |
winning | -0.030 |
shapes | -0.030 |
seem | -0.030 |
planets | -0.029 |
Token | ready |
Token ID | 3492 |
Feature activation | -3.18e+00 |
Pos logit contributions | |
winning | +0.001 |
ide | +0.001 |
being | +0.001 |
sharing | +0.001 |
shooting | +0.001 |
Neg logit contributions | |
rolley | -0.002 |
butcher | -0.002 |
microwave | -0.002 |
use | -0.002 |
piping | -0.002 |
Token | He |
Token ID | 679 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
rolley | +0.024 |
butcher | +0.023 |
microwave | +0.023 |
piping | +0.022 |
waitress | +0.021 |
Neg logit contributions | |
ide | -0.022 |
winning | -0.020 |
icated | -0.020 |
seem | -0.020 |
happening | -0.019 |
Token | took |
Token ID | 1718 |
Feature activation | -3.89e+00 |
Pos logit contributions | |
ide | +0.005 |
winning | +0.005 |
seem | +0.004 |
icated | +0.004 |
shooting | +0.004 |
Neg logit contributions | |
rolley | -0.007 |
butcher | -0.007 |
microwave | -0.007 |
piping | -0.007 |
frying | -0.007 |
Token | out |
Token ID | 503 |
Feature activation | -3.40e+00 |
Pos logit contributions | |
rolley | +0.145 |
butcher | +0.140 |
microwave | +0.136 |
frying | +0.136 |
piping | +0.132 |
Neg logit contributions | |
ide | -0.084 |
winning | -0.077 |
seem | -0.075 |
shapes | -0.075 |
shooting | -0.072 |
Token | a |
Token ID | 257 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
rolley | +0.106 |
microwave | +0.105 |
butcher | +0.103 |
bag | +0.100 |
waitress | +0.099 |
Neg logit contributions | |
ide | -0.089 |
winning | -0.082 |
seem | -0.080 |
icated | -0.079 |
hip | -0.077 |
Token | magn |
Token ID | 7842 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
rolley | +0.079 |
butcher | +0.070 |
waitress | +0.067 |
etchup | +0.067 |
microwave | +0.067 |
Neg logit contributions | |
shooting | -0.060 |
winning | -0.057 |
ide | -0.056 |
seem | -0.054 |
shapes | -0.053 |
Token | ifying |
Token ID | 4035 |
Feature activation | -8.36e+00 |
Pos logit contributions | |
rolley | +0.016 |
microwave | +0.013 |
butcher | +0.013 |
piping | +0.012 |
frying | +0.012 |
Neg logit contributions | |
ide | -0.057 |
winning | -0.056 |
seem | -0.054 |
icated | -0.054 |
sharing | -0.053 |
Token | glass |
Token ID | 5405 |
Feature activation | -2.40e+00 |
Pos logit contributions | |
rolley | +0.280 |
butcher | +0.271 |
use | +0.256 |
piping | +0.256 |
aked | +0.255 |
Neg logit contributions | |
planet | -0.197 |
shooting | -0.186 |
appear | -0.182 |
ide | -0.179 |
color | -0.178 |
Token | and |
Token ID | 290 |
Feature activation | -4.33e-01 |
Pos logit contributions | |
rolley | +0.090 |
butcher | +0.087 |
microwave | +0.085 |
piping | +0.085 |
frying | +0.083 |
Neg logit contributions | |
ide | -0.050 |
winning | -0.044 |
seem | -0.042 |
shooting | -0.040 |
icated | -0.040 |
Token | a |
Token ID | 257 |
Feature activation | -2.10e+00 |
Pos logit contributions | |
rolley | +0.014 |
butcher | +0.013 |
microwave | +0.013 |
piping | +0.013 |
frying | +0.013 |
Neg logit contributions | |
ide | -0.011 |
winning | -0.010 |
seem | -0.010 |
shapes | -0.010 |
shooting | -0.010 |
Token | notebook |
Token ID | 20922 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
rolley | +0.071 |
butcher | +0.063 |
aked | +0.061 |
microwave | +0.061 |
etchup | +0.061 |
Neg logit contributions | |
shooting | -0.052 |
winning | -0.052 |
shapes | -0.049 |
ide | -0.048 |
appear | -0.047 |
Token | and |
Token ID | 290 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
rolley | +0.055 |
microwave | +0.053 |
butcher | +0.052 |
frying | +0.051 |
pouch | +0.050 |
Neg logit contributions | |
ide | -0.040 |
winning | -0.037 |
icated | -0.036 |
seem | -0.035 |
textures | -0.033 |
Token | help |
Token ID | 1037 |
Feature activation | +4.45e-01 |
Pos logit contributions | |
ide | +0.007 |
winning | +0.007 |
seem | +0.007 |
being | +0.006 |
sharing | +0.006 |
Neg logit contributions | |
rolley | -0.011 |
butcher | -0.010 |
microwave | -0.010 |
piping | -0.010 |
frying | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
ide | +0.009 |
seem | +0.007 |
being | +0.007 |
winning | +0.007 |
sharing | +0.007 |
Neg logit contributions | |
rolley | -0.018 |
pouch | -0.017 |
butcher | -0.017 |
microwave | -0.017 |
piping | -0.017 |
Token | fixed |
Token ID | 5969 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
not | +0.031 |
always | +0.029 |
when | +0.028 |
being | +0.028 |
seem | +0.028 |
Neg logit contributions | |
rolley | -0.070 |
piping | -0.065 |
microwave | -0.065 |
butcher | -0.064 |
enn | -0.063 |
Token | the |
Token ID | 262 |
Feature activation | -7.96e-03 |
Pos logit contributions | |
rolley | +0.008 |
butcher | +0.008 |
microwave | +0.008 |
piping | +0.008 |
frying | +0.008 |
Neg logit contributions | |
ide | -0.004 |
winning | -0.004 |
seem | -0.003 |
shapes | -0.003 |
icated | -0.003 |
Token | cash |
Token ID | 5003 |
Feature activation | -5.37e+00 |
Pos logit contributions | |
rolley | +0.000 |
butcher | +0.000 |
microwave | +0.000 |
piping | +0.000 |
P | +0.000 |
Neg logit contributions | |
ide | -0.000 |
winning | -0.000 |
shooting | -0.000 |
shapes | -0.000 |
icated | -0.000 |
Token | register |
Token ID | 7881 |
Feature activation | -8.28e+00 |
Pos logit contributions | |
rolley | +0.158 |
butcher | +0.152 |
microwave | +0.150 |
piping | +0.149 |
frying | +0.147 |
Neg logit contributions | |
ide | -0.155 |
winning | -0.141 |
seem | -0.138 |
being | -0.134 |
icated | -0.134 |
Token | . |
Token ID | 13 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
rolley | +0.291 |
butcher | +0.281 |
microwave | +0.278 |
frying | +0.270 |
piping | +0.268 |
Neg logit contributions | |
ide | -0.190 |
winning | -0.173 |
icated | -0.167 |
seem | -0.165 |
shooting | -0.156 |
Token | The |
Token ID | 383 |
Feature activation | +2.77e-01 |
Pos logit contributions | |
rolley | +0.006 |
butcher | +0.006 |
frying | +0.006 |
piping | +0.006 |
microwave | +0.006 |
Neg logit contributions | |
ide | -0.004 |
winning | -0.004 |
seem | -0.004 |
sharing | -0.003 |
being | -0.003 |
Token | shop |
Token ID | 6128 |
Feature activation | -3.57e+00 |
Pos logit contributions | |
winning | +0.009 |
being | +0.009 |
shapes | +0.009 |
shooting | +0.009 |
sharing | +0.009 |
Neg logit contributions | |
rolley | -0.007 |
enn | -0.006 |
piping | -0.006 |
etchup | -0.006 |
butcher | -0.005 |
Token | keeper |
Token ID | 13884 |
Feature activation | -2.59e+00 |
Pos logit contributions | |
rolley | +0.078 |
butcher | +0.074 |
microwave | +0.073 |
piping | +0.072 |
frying | +0.069 |
Neg logit contributions | |
ide | -0.127 |
winning | -0.120 |
seem | -0.117 |
sharing | -0.115 |
shapes | -0.115 |
Token | was |
Token ID | 373 |
Feature activation | +3.69e-01 |
Pos logit contributions | |
rolley | +0.080 |
butcher | +0.076 |
microwave | +0.075 |
piping | +0.073 |
frying | +0.073 |
Neg logit contributions | |
ide | -0.071 |
winning | -0.066 |
seem | -0.063 |
icated | -0.062 |
shooting | -0.061 |
Token | again |
Token ID | 757 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
ide | +0.047 |
icated | +0.042 |
winning | +0.041 |
seem | +0.040 |
shapes | +0.038 |
Neg logit contributions | |
rolley | -0.071 |
butcher | -0.069 |
microwave | -0.069 |
frying | -0.067 |
waitress | -0.066 |
Token | . |
Token ID | 13 |
Feature activation | +4.97e-01 |
Pos logit contributions | |
rolley | +0.061 |
butcher | +0.059 |
microwave | +0.059 |
piping | +0.057 |
frying | +0.057 |
Neg logit contributions | |
ide | -0.034 |
winning | -0.031 |
seem | -0.029 |
icated | -0.028 |
shooting | -0.027 |
Token | But |
Token ID | 887 |
Feature activation | +2.59e+00 |
Pos logit contributions | |
ide | +0.013 |
winning | +0.012 |
seem | +0.012 |
icated | +0.012 |
shooting | +0.011 |
Neg logit contributions | |
rolley | -0.016 |
butcher | -0.015 |
microwave | -0.015 |
frying | -0.015 |
piping | -0.015 |
Token | unb |
Token ID | 22619 |
Feature activation | +2.89e-01 |
Pos logit contributions | |
ide | +0.069 |
icated | +0.061 |
shapes | +0.057 |
textures | +0.057 |
patterns | +0.056 |
Neg logit contributions | |
butcher | -0.079 |
pouch | -0.079 |
rolley | -0.078 |
microwave | -0.076 |
pan | -0.076 |
Token | ek |
Token ID | 988 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
ide | +0.010 |
winning | +0.010 |
seem | +0.010 |
icated | +0.009 |
shooting | +0.009 |
Neg logit contributions | |
rolley | -0.007 |
butcher | -0.006 |
microwave | -0.006 |
frying | -0.006 |
piping | -0.006 |
Token | nown |
Token ID | 3408 |
Feature activation | -8.25e+00 |
Pos logit contributions | |
rolley | +0.036 |
butcher | +0.034 |
microwave | +0.033 |
frying | +0.032 |
piping | +0.032 |
Neg logit contributions | |
ide | -0.038 |
winning | -0.037 |
seem | -0.035 |
icated | -0.034 |
sharing | -0.034 |
Token | to |
Token ID | 284 |
Feature activation | -4.10e+00 |
Pos logit contributions | |
paramed | +0.270 |
butcher | +0.269 |
piping | +0.266 |
rolley | +0.266 |
waitress | +0.265 |
Neg logit contributions | |
ide | -0.195 |
winning | -0.186 |
icated | -0.179 |
shooting | -0.176 |
being | -0.173 |
Token | him |
Token ID | 683 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
butcher | +0.145 |
rolley | +0.142 |
microwave | +0.141 |
waitress | +0.135 |
frying | +0.133 |
Neg logit contributions | |
ide | -0.089 |
winning | -0.078 |
icated | -0.075 |
seem | -0.074 |
shapes | -0.073 |
Token | , |
Token ID | 11 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
butcher | +0.051 |
rolley | +0.050 |
microwave | +0.049 |
waitress | +0.047 |
frying | +0.047 |
Neg logit contributions | |
ide | -0.036 |
winning | -0.033 |
icated | -0.032 |
seem | -0.031 |
textures | -0.031 |
Token | the |
Token ID | 262 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
rolley | +0.062 |
butcher | +0.060 |
microwave | +0.059 |
piping | +0.058 |
frying | +0.058 |
Neg logit contributions | |
ide | -0.031 |
winning | -0.028 |
seem | -0.026 |
icated | -0.026 |
shooting | -0.025 |
Token | king |
Token ID | 5822 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
microwave | +0.064 |
butcher | +0.062 |
rolley | +0.062 |
pan | +0.059 |
bag | +0.059 |
Neg logit contributions | |
ide | -0.064 |
seem | -0.054 |
winning | -0.054 |
icated | -0.054 |
hip | -0.051 |
Token | She |
Token ID | 1375 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
rolley | +0.006 |
butcher | +0.006 |
microwave | +0.006 |
frying | +0.006 |
piping | +0.006 |
Neg logit contributions | |
ide | -0.004 |
winning | -0.003 |
seem | -0.003 |
shooting | -0.003 |
shapes | -0.003 |
Token | takes |
Token ID | 2753 |
Feature activation | -3.44e+00 |
Pos logit contributions | |
rolley | +0.026 |
butcher | +0.025 |
microwave | +0.025 |
piping | +0.024 |
frying | +0.023 |
Neg logit contributions | |
ide | -0.033 |
winning | -0.031 |
seem | -0.030 |
icated | -0.029 |
shooting | -0.029 |
Token | out |
Token ID | 503 |
Feature activation | -3.65e+00 |
Pos logit contributions | |
rolley | +0.128 |
butcher | +0.124 |
frying | +0.117 |
microwave | +0.117 |
piping | +0.116 |
Neg logit contributions | |
ide | -0.071 |
shapes | -0.068 |
winning | -0.067 |
seem | -0.067 |
shooting | -0.065 |
Token | a |
Token ID | 257 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
rolley | +0.123 |
microwave | +0.119 |
butcher | +0.118 |
piping | +0.114 |
frying | +0.113 |
Neg logit contributions | |
ide | -0.089 |
winning | -0.083 |
seem | -0.078 |
icated | -0.078 |
sharing | -0.074 |
Token | magn |
Token ID | 7842 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
rolley | +0.088 |
butcher | +0.079 |
waitress | +0.077 |
etchup | +0.075 |
P | +0.075 |
Neg logit contributions | |
shooting | -0.061 |
ide | -0.058 |
winning | -0.058 |
seem | -0.057 |
shapes | -0.057 |
Token | ifying |
Token ID | 4035 |
Feature activation | -8.23e+00 |
Pos logit contributions | |
rolley | +0.018 |
microwave | +0.015 |
butcher | +0.014 |
piping | +0.014 |
frying | +0.013 |
Neg logit contributions | |
ide | -0.066 |
winning | -0.065 |
seem | -0.063 |
sharing | -0.062 |
icated | -0.062 |
Token | glass |
Token ID | 5405 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
rolley | +0.273 |
butcher | +0.268 |
use | +0.255 |
aked | +0.246 |
ip | +0.246 |
Neg logit contributions | |
shapes | -0.185 |
planet | -0.185 |
ide | -0.183 |
shooting | -0.180 |
color | -0.180 |
Token | . |
Token ID | 13 |
Feature activation | -7.84e-01 |
Pos logit contributions | |
rolley | +0.084 |
butcher | +0.083 |
piping | +0.079 |
frying | +0.077 |
waitress | +0.077 |
Neg logit contributions | |
ide | -0.050 |
shapes | -0.043 |
seem | -0.040 |
winning | -0.040 |
shooting | -0.039 |
Token | She |
Token ID | 1375 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
rolley | +0.026 |
butcher | +0.025 |
microwave | +0.024 |
frying | +0.024 |
piping | +0.024 |
Neg logit contributions | |
ide | -0.020 |
winning | -0.019 |
seem | -0.018 |
icated | -0.017 |
shooting | -0.017 |
Token | looks |
Token ID | 3073 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
rolley | +0.043 |
butcher | +0.041 |
microwave | +0.040 |
piping | +0.039 |
frying | +0.039 |
Neg logit contributions | |
ide | -0.048 |
winning | -0.045 |
seem | -0.044 |
icated | -0.043 |
shooting | -0.042 |
Token | at |
Token ID | 379 |
Feature activation | +3.43e-01 |
Pos logit contributions | |
rolley | +0.092 |
butcher | +0.087 |
microwave | +0.083 |
pouch | +0.083 |
enn | +0.082 |
Neg logit contributions | |
ide | -0.038 |
seem | -0.035 |
winning | -0.035 |
shooting | -0.035 |
shapes | -0.034 |
Token | on |
Token ID | 319 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
ide | +0.017 |
winning | +0.015 |
seem | +0.014 |
icated | +0.014 |
shooting | +0.014 |
Neg logit contributions | |
rolley | -0.032 |
butcher | -0.031 |
microwave | -0.031 |
frying | -0.030 |
piping | -0.030 |
Token | display |
Token ID | 3359 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
rolley | +0.050 |
butcher | +0.048 |
microwave | +0.048 |
piping | +0.047 |
frying | +0.046 |
Neg logit contributions | |
ide | -0.026 |
winning | -0.024 |
seem | -0.023 |
icated | -0.022 |
shooting | -0.022 |
Token | near |
Token ID | 1474 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
ide | +0.028 |
winning | +0.025 |
seem | +0.024 |
icated | +0.023 |
shooting | +0.023 |
Neg logit contributions | |
rolley | -0.050 |
butcher | -0.048 |
microwave | -0.048 |
piping | -0.046 |
frying | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | -6.69e-02 |
Pos logit contributions | |
ide | +0.035 |
winning | +0.032 |
seem | +0.030 |
icated | +0.029 |
shooting | +0.028 |
Neg logit contributions | |
rolley | -0.063 |
butcher | -0.061 |
microwave | -0.060 |
piping | -0.059 |
frying | -0.059 |
Token | cash |
Token ID | 5003 |
Feature activation | -5.28e+00 |
Pos logit contributions | |
rolley | +0.002 |
aked | +0.002 |
enn | +0.002 |
piping | +0.002 |
etchup | +0.002 |
Neg logit contributions | |
ide | -0.002 |
winning | -0.002 |
shapes | -0.002 |
shooting | -0.002 |
planet | -0.002 |
Token | register |
Token ID | 7881 |
Feature activation | -8.00e+00 |
Pos logit contributions | |
rolley | +0.132 |
butcher | +0.124 |
microwave | +0.123 |
piping | +0.117 |
frying | +0.117 |
Neg logit contributions | |
ide | -0.176 |
winning | -0.167 |
seem | -0.161 |
icated | -0.159 |
shooting | -0.156 |
Token | . |
Token ID | 13 |
Feature activation | +2.78e-01 |
Pos logit contributions | |
rolley | +0.288 |
butcher | +0.278 |
microwave | +0.275 |
frying | +0.267 |
piping | +0.267 |
Neg logit contributions | |
ide | -0.176 |
winning | -0.160 |
icated | -0.152 |
seem | -0.152 |
shooting | -0.144 |
Token | They |
Token ID | 1119 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
ide | +0.006 |
winning | +0.005 |
seem | +0.005 |
sharing | +0.005 |
being | +0.005 |
Neg logit contributions | |
rolley | -0.010 |
butcher | -0.010 |
piping | -0.010 |
use | -0.010 |
etchup | -0.010 |
Token | looked |
Token ID | 3114 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
ide | +0.044 |
winning | +0.041 |
seem | +0.039 |
icated | +0.038 |
shooting | +0.037 |
Neg logit contributions | |
rolley | -0.063 |
butcher | -0.060 |
microwave | -0.060 |
piping | -0.058 |
frying | -0.058 |
Token | so |
Token ID | 523 |
Feature activation | -3.87e+00 |
Pos logit contributions | |
rolley | +0.062 |
butcher | +0.059 |
microwave | +0.058 |
piping | +0.056 |
frying | +0.056 |
Neg logit contributions | |
ide | -0.033 |
winning | -0.031 |
seem | -0.029 |
shooting | -0.029 |
sharing | -0.028 |
Token | y |
Token ID | 331 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
butcher | +0.106 |
frying | +0.105 |
rolley | +0.105 |
piping | +0.104 |
microwave | +0.102 |
Neg logit contributions | |
ide | -0.107 |
icated | -0.102 |
winning | -0.102 |
hip | -0.099 |
planets | -0.095 |
Token | She |
Token ID | 1375 |
Feature activation | -3.61e-01 |
Pos logit contributions | |
ide | +0.006 |
winning | +0.006 |
seem | +0.006 |
shapes | +0.005 |
being | +0.005 |
Neg logit contributions | |
rolley | -0.008 |
microwave | -0.008 |
piping | -0.008 |
frying | -0.008 |
butcher | -0.008 |
Token | took |
Token ID | 1718 |
Feature activation | -3.77e+00 |
Pos logit contributions | |
rolley | +0.011 |
butcher | +0.011 |
microwave | +0.011 |
piping | +0.011 |
frying | +0.011 |
Neg logit contributions | |
ide | -0.009 |
winning | -0.009 |
seem | -0.008 |
shooting | -0.008 |
icated | -0.008 |
Token | out |
Token ID | 503 |
Feature activation | -3.08e+00 |
Pos logit contributions | |
rolley | +0.120 |
butcher | +0.115 |
microwave | +0.114 |
frying | +0.110 |
piping | +0.109 |
Neg logit contributions | |
ide | -0.100 |
winning | -0.093 |
seem | -0.090 |
icated | -0.088 |
shooting | -0.087 |
Token | a |
Token ID | 257 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
rolley | +0.096 |
butcher | +0.093 |
microwave | +0.092 |
piping | +0.089 |
waitress | +0.089 |
Neg logit contributions | |
ide | -0.081 |
winning | -0.075 |
seem | -0.072 |
icated | -0.072 |
hip | -0.069 |
Token | magn |
Token ID | 7842 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
rolley | +0.069 |
butcher | +0.064 |
microwave | +0.062 |
waitress | +0.061 |
etchup | +0.060 |
Neg logit contributions | |
ide | -0.049 |
shooting | -0.049 |
winning | -0.048 |
seem | -0.047 |
shapes | -0.044 |
Token | ifying |
Token ID | 4035 |
Feature activation | -7.99e+00 |
Pos logit contributions | |
rolley | +0.011 |
microwave | +0.008 |
pouch | +0.008 |
frying | +0.007 |
enn | +0.007 |
Neg logit contributions | |
winning | -0.056 |
ide | -0.056 |
seem | -0.055 |
sharing | -0.054 |
being | -0.054 |
Token | glass |
Token ID | 5405 |
Feature activation | -2.23e+00 |
Pos logit contributions | |
rolley | +0.274 |
butcher | +0.263 |
piping | +0.252 |
use | +0.247 |
frying | +0.244 |
Neg logit contributions | |
planet | -0.186 |
shooting | -0.181 |
appear | -0.173 |
ide | -0.171 |
color | -0.168 |
Token | and |
Token ID | 290 |
Feature activation | -4.26e-01 |
Pos logit contributions | |
rolley | +0.085 |
butcher | +0.082 |
microwave | +0.081 |
piping | +0.080 |
frying | +0.079 |
Neg logit contributions | |
ide | -0.045 |
winning | -0.040 |
seem | -0.038 |
icated | -0.037 |
shooting | -0.037 |
Token | looked |
Token ID | 3114 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
rolley | +0.014 |
butcher | +0.014 |
microwave | +0.013 |
piping | +0.013 |
frying | +0.013 |
Neg logit contributions | |
ide | -0.010 |
winning | -0.010 |
seem | -0.010 |
shapes | -0.010 |
shooting | -0.010 |
Token | very |
Token ID | 845 |
Feature activation | -4.45e+00 |
Pos logit contributions | |
rolley | +0.071 |
butcher | +0.068 |
microwave | +0.068 |
piping | +0.065 |
frying | +0.065 |
Neg logit contributions | |
ide | -0.047 |
winning | -0.043 |
seem | -0.041 |
icated | -0.040 |
shooting | -0.039 |
Token | carefully |
Token ID | 7773 |
Feature activation | -2.31e+00 |
Pos logit contributions | |
rolley | +0.140 |
frying | +0.135 |
butcher | +0.135 |
microwave | +0.135 |
piping | +0.132 |
Neg logit contributions | |
ide | -0.113 |
icated | -0.103 |
winning | -0.102 |
hip | -0.095 |
seem | -0.094 |
Token | borrowed |
Token ID | 22546 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
rolley | +0.000 |
butcher | +0.000 |
microwave | +0.000 |
frying | +0.000 |
pouch | +0.000 |
Neg logit contributions | |
ide | -0.000 |
winning | -0.000 |
seem | -0.000 |
icated | -0.000 |
hip | -0.000 |
Token | her |
Token ID | 607 |
Feature activation | -4.28e-01 |
Pos logit contributions | |
rolley | +0.047 |
butcher | +0.046 |
microwave | +0.045 |
piping | +0.044 |
pouch | +0.044 |
Neg logit contributions | |
ide | -0.033 |
winning | -0.029 |
seem | -0.027 |
icated | -0.027 |
sharing | -0.026 |
Token | mom |
Token ID | 1995 |
Feature activation | -3.98e+00 |
Pos logit contributions | |
rolley | +0.013 |
butcher | +0.012 |
microwave | +0.012 |
frying | +0.012 |
piping | +0.012 |
Neg logit contributions | |
ide | -0.012 |
winning | -0.011 |
seem | -0.011 |
icated | -0.011 |
sharing | -0.010 |
Token | 's |
Token ID | 338 |
Feature activation | -3.72e-01 |
Pos logit contributions | |
rolley | +0.122 |
butcher | +0.120 |
frying | +0.115 |
pan | +0.115 |
bag | +0.114 |
Neg logit contributions | |
winning | -0.106 |
seem | -0.097 |
ide | -0.097 |
rhythms | -0.095 |
tunes | -0.092 |
Token | magn |
Token ID | 7842 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
rolley | +0.010 |
butcher | +0.010 |
microwave | +0.010 |
frying | +0.009 |
piping | +0.009 |
Neg logit contributions | |
ide | -0.011 |
winning | -0.011 |
seem | -0.010 |
icated | -0.010 |
shooting | -0.010 |
Token | ifying |
Token ID | 4035 |
Feature activation | -7.90e+00 |
Pos logit contributions | |
rolley | +0.013 |
pouch | +0.010 |
microwave | +0.010 |
butcher | +0.009 |
frying | +0.009 |
Neg logit contributions | |
ide | -0.059 |
winning | -0.058 |
seem | -0.057 |
icated | -0.056 |
textures | -0.056 |
Token | glass |
Token ID | 5405 |
Feature activation | -2.39e+00 |
Pos logit contributions | |
rolley | +0.284 |
butcher | +0.261 |
stret | +0.257 |
piping | +0.256 |
use | +0.250 |
Neg logit contributions | |
color | -0.168 |
shapes | -0.165 |
winning | -0.164 |
not | -0.162 |
shooting | -0.162 |
Token | to |
Token ID | 284 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
rolley | +0.086 |
butcher | +0.083 |
microwave | +0.082 |
piping | +0.080 |
frying | +0.079 |
Neg logit contributions | |
ide | -0.053 |
winning | -0.049 |
seem | -0.047 |
icated | -0.045 |
shooting | -0.044 |
Token | take |
Token ID | 1011 |
Feature activation | -3.41e+00 |
Pos logit contributions | |
rolley | +0.042 |
butcher | +0.041 |
microwave | +0.040 |
frying | +0.039 |
piping | +0.039 |
Neg logit contributions | |
ide | -0.031 |
winning | -0.029 |
icated | -0.027 |
seem | -0.027 |
shooting | -0.026 |
Token | a |
Token ID | 257 |
Feature activation | -9.74e-01 |
Pos logit contributions | |
rolley | +0.113 |
butcher | +0.108 |
microwave | +0.107 |
piping | +0.104 |
frying | +0.103 |
Neg logit contributions | |
ide | -0.086 |
winning | -0.080 |
seem | -0.077 |
icated | -0.075 |
shooting | -0.073 |
Token | closer |
Token ID | 5699 |
Feature activation | -2.55e+00 |
Pos logit contributions | |
rolley | +0.029 |
butcher | +0.028 |
microwave | +0.028 |
piping | +0.027 |
frying | +0.027 |
Neg logit contributions | |
ide | -0.027 |
winning | -0.026 |
seem | -0.024 |
icated | -0.024 |
shooting | -0.023 |
Token | Finally |
Token ID | 9461 |
Feature activation | -3.26e+00 |
Pos logit contributions | |
ide | +0.001 |
winning | +0.001 |
seem | +0.001 |
icated | +0.001 |
shooting | +0.001 |
Neg logit contributions | |
rolley | -0.001 |
butcher | -0.001 |
microwave | -0.001 |
frying | -0.001 |
piping | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | -7.72e-01 |
Pos logit contributions | |
rolley | +0.133 |
butcher | +0.127 |
microwave | +0.125 |
frying | +0.122 |
piping | +0.121 |
Neg logit contributions | |
ide | -0.059 |
winning | -0.054 |
seem | -0.052 |
shooting | -0.048 |
being | -0.048 |
Token | it |
Token ID | 340 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
rolley | +0.030 |
butcher | +0.029 |
microwave | +0.029 |
piping | +0.028 |
frying | +0.028 |
Neg logit contributions | |
ide | -0.015 |
winning | -0.014 |
seem | -0.013 |
icated | -0.012 |
shooting | -0.012 |
Token | swoop |
Token ID | 38527 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
rolley | +0.050 |
microwave | +0.048 |
butcher | +0.047 |
frying | +0.046 |
pouch | +0.046 |
Neg logit contributions | |
ide | -0.039 |
winning | -0.038 |
icated | -0.036 |
seem | -0.035 |
sharing | -0.034 |
Token | ed |
Token ID | 276 |
Feature activation | -5.21e+00 |
Pos logit contributions | |
rolley | +0.016 |
butcher | +0.014 |
microwave | +0.014 |
frying | +0.012 |
pouch | +0.012 |
Neg logit contributions | |
ide | -0.066 |
winning | -0.064 |
icated | -0.063 |
seem | -0.063 |
sharing | -0.061 |
Token | back |
Token ID | 736 |
Feature activation | -7.86e+00 |
Pos logit contributions | |
rolley | +0.126 |
microwave | +0.120 |
butcher | +0.119 |
piping | +0.114 |
frying | +0.113 |
Neg logit contributions | |
ide | -0.173 |
winning | -0.165 |
seem | -0.161 |
icated | -0.160 |
sharing | -0.153 |
Token | down |
Token ID | 866 |
Feature activation | -5.76e+00 |
Pos logit contributions | |
rolley | +0.257 |
butcher | +0.249 |
microwave | +0.244 |
piping | +0.235 |
waitress | +0.234 |
Neg logit contributions | |
ide | -0.200 |
winning | -0.189 |
shooting | -0.176 |
seem | -0.176 |
icated | -0.171 |
Token | to |
Token ID | 284 |
Feature activation | -8.02e-01 |
Pos logit contributions | |
rolley | +0.247 |
butcher | +0.241 |
microwave | +0.237 |
piping | +0.233 |
frying | +0.232 |
Neg logit contributions | |
ide | -0.089 |
winning | -0.080 |
seem | -0.072 |
shooting | -0.070 |
being | -0.067 |
Token | the |
Token ID | 262 |
Feature activation | -2.39e-01 |
Pos logit contributions | |
rolley | +0.030 |
butcher | +0.029 |
microwave | +0.028 |
frying | +0.027 |
piping | +0.027 |
Neg logit contributions | |
ide | -0.017 |
winning | -0.016 |
seem | -0.015 |
icated | -0.015 |
shooting | -0.014 |
Token | lake |
Token ID | 13546 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
rolley | +0.008 |
microwave | +0.008 |
butcher | +0.008 |
frying | +0.008 |
piping | +0.008 |
Neg logit contributions | |
ide | -0.006 |
winning | -0.005 |
seem | -0.005 |
icated | -0.005 |
sharing | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | +9.35e-01 |
Pos logit contributions | |
icated | +0.035 |
winning | +0.035 |
ide | +0.035 |
seem | +0.033 |
textures | +0.031 |
Neg logit contributions | |
rolley | -0.060 |
microwave | -0.057 |
pouch | -0.057 |
pan | -0.057 |
butcher | -0.056 |
Token |
|
Token ID | 198 |
Feature activation | +6.60e-01 |
Pos logit contributions | |
ide | +0.020 |
winning | +0.018 |
seem | +0.018 |
icated | +0.017 |
shooting | +0.017 |
Neg logit contributions | |
rolley | -0.026 |
butcher | -0.025 |
microwave | -0.025 |
frying | -0.024 |
piping | -0.024 |
Token | He |
Token ID | 1544 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
ide | +0.018 |
winning | +0.018 |
seem | +0.018 |
planets | +0.017 |
icated | +0.017 |
Neg logit contributions | |
rolley | -0.020 |
butcher | -0.019 |
P | -0.019 |
microwave | -0.018 |
frying | -0.017 |
Token | brought |
Token ID | 3181 |
Feature activation | -2.42e+00 |
Pos logit contributions | |
ide | +0.004 |
winning | +0.004 |
seem | +0.004 |
shooting | +0.004 |
shapes | +0.004 |
Neg logit contributions | |
rolley | -0.005 |
butcher | -0.004 |
microwave | -0.004 |
piping | -0.004 |
frying | -0.004 |
Token | his |
Token ID | 465 |
Feature activation | -8.38e-02 |
Pos logit contributions | |
rolley | +0.093 |
butcher | +0.089 |
microwave | +0.088 |
piping | +0.086 |
frying | +0.086 |
Neg logit contributions | |
ide | -0.049 |
winning | -0.044 |
seem | -0.042 |
icated | -0.041 |
shapes | -0.041 |
Token | magn |
Token ID | 7842 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
rolley | +0.003 |
butcher | +0.002 |
microwave | +0.002 |
frying | +0.002 |
piping | +0.002 |
Neg logit contributions | |
ide | -0.002 |
winning | -0.002 |
seem | -0.002 |
shapes | -0.002 |
shooting | -0.002 |
Token | ifying |
Token ID | 4035 |
Feature activation | -7.84e+00 |
Pos logit contributions | |
rolley | +0.012 |
pouch | +0.010 |
microwave | +0.009 |
piping | +0.008 |
frying | +0.008 |
Neg logit contributions | |
ide | -0.050 |
winning | -0.049 |
seem | -0.049 |
icated | -0.048 |
being | -0.048 |
Token | glass |
Token ID | 5405 |
Feature activation | -2.10e+00 |
Pos logit contributions | |
rolley | +0.270 |
butcher | +0.257 |
piping | +0.250 |
use | +0.248 |
frying | +0.242 |
Neg logit contributions | |
color | -0.169 |
shapes | -0.168 |
not | -0.165 |
ide | -0.165 |
winning | -0.162 |
Token | and |
Token ID | 290 |
Feature activation | +3.43e-02 |
Pos logit contributions | |
rolley | +0.070 |
butcher | +0.068 |
microwave | +0.067 |
piping | +0.066 |
frying | +0.065 |
Neg logit contributions | |
ide | -0.052 |
winning | -0.047 |
seem | -0.046 |
icated | -0.044 |
shapes | -0.044 |
Token | looked |
Token ID | 3114 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
ide | +0.001 |
winning | +0.001 |
seem | +0.001 |
shapes | +0.001 |
shooting | +0.001 |
Neg logit contributions | |
rolley | -0.001 |
butcher | -0.001 |
microwave | -0.001 |
piping | -0.001 |
frying | -0.001 |
Token | closely |
Token ID | 7173 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
rolley | +0.059 |
microwave | +0.057 |
butcher | +0.056 |
frying | +0.055 |
pan | +0.055 |
Neg logit contributions | |
ide | -0.048 |
winning | -0.045 |
icated | -0.043 |
seem | -0.042 |
hip | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | -1.87e-01 |
Pos logit contributions | |
rolley | +0.004 |
microwave | +0.004 |
pan | +0.004 |
butcher | +0.004 |
bag | +0.004 |
Neg logit contributions | |
ide | -0.004 |
winning | -0.004 |
icated | -0.004 |
seem | -0.004 |
hip | -0.004 |
Token | on |
Token ID | 261 |
Feature activation | -2.81e+00 |
Pos logit contributions | |
rolley | +0.026 |
butcher | +0.025 |
microwave | +0.024 |
piping | +0.023 |
frying | +0.023 |
Neg logit contributions | |
ide | -0.030 |
winning | -0.029 |
seem | -0.028 |
icated | -0.027 |
sharing | -0.027 |
Token | zo |
Token ID | 10872 |
Feature activation | +3.65e+00 |
Pos logit contributions | |
rolley | +0.078 |
microwave | +0.075 |
butcher | +0.074 |
pouch | +0.072 |
pan | +0.071 |
Neg logit contributions | |
ide | -0.082 |
winning | -0.081 |
icated | -0.077 |
seem | -0.077 |
sharing | -0.072 |
Token | and |
Token ID | 290 |
Feature activation | +8.27e-01 |
Pos logit contributions | |
ide | +0.104 |
winning | +0.102 |
icated | +0.096 |
sharing | +0.092 |
seem | +0.091 |
Neg logit contributions | |
rolley | -0.097 |
butcher | -0.093 |
pouch | -0.092 |
microwave | -0.091 |
use | -0.091 |
Token | Gret |
Token ID | 35877 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
ide | +0.022 |
winning | +0.020 |
seem | +0.019 |
icated | +0.019 |
shooting | +0.018 |
Neg logit contributions | |
rolley | -0.026 |
butcher | -0.025 |
microwave | -0.025 |
piping | -0.024 |
frying | -0.024 |
Token | chen |
Token ID | 6607 |
Feature activation | -7.07e-02 |
Pos logit contributions | |
rolley | +0.038 |
butcher | +0.034 |
microwave | +0.034 |
piping | +0.034 |
P | +0.033 |
Neg logit contributions | |
ide | -0.036 |
winning | -0.036 |
seem | -0.035 |
sharing | -0.033 |
shooting | -0.033 |
Token | quickly |
Token ID | 2952 |
Feature activation | -7.83e+00 |
Pos logit contributions | |
rolley | +0.002 |
butcher | +0.002 |
microwave | +0.002 |
pouch | +0.002 |
piping | +0.002 |
Neg logit contributions | |
ide | -0.002 |
winning | -0.002 |
seem | -0.002 |
icated | -0.002 |
sharing | -0.002 |
Token | became |
Token ID | 2627 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
rolley | +0.248 |
butcher | +0.238 |
microwave | +0.235 |
piping | +0.225 |
waitress | +0.222 |
Neg logit contributions | |
ide | -0.211 |
seem | -0.195 |
winning | -0.194 |
shooting | -0.186 |
shapes | -0.186 |
Token | the |
Token ID | 262 |
Feature activation | -4.57e-01 |
Pos logit contributions | |
ide | +0.024 |
winning | +0.023 |
seem | +0.021 |
shooting | +0.021 |
sharing | +0.021 |
Neg logit contributions | |
rolley | -0.038 |
microwave | -0.036 |
butcher | -0.036 |
frying | -0.035 |
piping | -0.035 |
Token | best |
Token ID | 1266 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
butcher | +0.013 |
rolley | +0.013 |
microwave | +0.013 |
waitress | +0.013 |
P | +0.012 |
Neg logit contributions | |
ide | -0.013 |
seem | -0.012 |
winning | -0.011 |
icated | -0.011 |
hip | -0.011 |
Token | of |
Token ID | 286 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
rolley | +0.028 |
butcher | +0.026 |
microwave | +0.026 |
waitress | +0.025 |
pan | +0.025 |
Neg logit contributions | |
ide | -0.031 |
winning | -0.030 |
seem | -0.030 |
icated | -0.029 |
happening | -0.028 |
Token | friends |
Token ID | 2460 |
Feature activation | +2.42e+00 |
Pos logit contributions | |
ide | +0.042 |
seem | +0.038 |
winning | +0.037 |
icated | +0.037 |
hip | +0.035 |
Neg logit contributions | |
rolley | -0.050 |
butcher | -0.048 |
microwave | -0.047 |
piping | -0.047 |
waitress | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
rolley | +0.112 |
microwave | +0.108 |
butcher | +0.107 |
piping | +0.104 |
waitress | +0.103 |
Neg logit contributions | |
ide | -0.090 |
winning | -0.086 |
seem | -0.081 |
icated | -0.080 |
hip | -0.077 |
Token | necklace |
Token ID | 38987 |
Feature activation | -6.52e-01 |
Pos logit contributions | |
rolley | +0.061 |
butcher | +0.058 |
microwave | +0.058 |
piping | +0.056 |
frying | +0.055 |
Neg logit contributions | |
ide | -0.057 |
winning | -0.053 |
seem | -0.051 |
icated | -0.050 |
shooting | -0.049 |
Token | and |
Token ID | 290 |
Feature activation | +5.85e-02 |
Pos logit contributions | |
rolley | +0.024 |
pouch | +0.023 |
microwave | +0.023 |
bag | +0.023 |
butcher | +0.023 |
Neg logit contributions | |
winning | -0.013 |
icated | -0.012 |
ide | -0.012 |
seem | -0.012 |
rhythms | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
shapes | +0.001 |
planets | +0.001 |
ide | +0.001 |
patterns | +0.001 |
not | +0.001 |
Neg logit contributions | |
rolley | -0.002 |
butcher | -0.002 |
piping | -0.002 |
paramed | -0.002 |
microwave | -0.002 |
Token | magn |
Token ID | 7842 |
Feature activation | -9.10e-01 |
Pos logit contributions | |
rolley | +0.055 |
butcher | +0.052 |
microwave | +0.051 |
frying | +0.049 |
piping | +0.049 |
Neg logit contributions | |
ide | -0.048 |
winning | -0.045 |
seem | -0.043 |
icated | -0.043 |
shooting | -0.042 |
Token | ifying |
Token ID | 4035 |
Feature activation | -7.63e+00 |
Pos logit contributions | |
rolley | +0.013 |
microwave | +0.011 |
butcher | +0.011 |
piping | +0.011 |
pouch | +0.011 |
Neg logit contributions | |
ide | -0.039 |
winning | -0.039 |
seem | -0.037 |
icated | -0.037 |
sharing | -0.037 |
Token | glass |
Token ID | 5405 |
Feature activation | -2.23e+00 |
Pos logit contributions | |
butcher | +0.269 |
rolley | +0.265 |
use | +0.255 |
aked | +0.247 |
dollars | +0.245 |
Neg logit contributions | |
planet | -0.163 |
shapes | -0.152 |
color | -0.151 |
shooting | -0.147 |
appear | -0.147 |
Token | and |
Token ID | 290 |
Feature activation | -1.61e-01 |
Pos logit contributions | |
rolley | +0.085 |
butcher | +0.082 |
microwave | +0.080 |
piping | +0.079 |
frying | +0.078 |
Neg logit contributions | |
ide | -0.046 |
winning | -0.040 |
seem | -0.039 |
shapes | -0.037 |
icated | -0.037 |
Token | started |
Token ID | 2067 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
rolley | +0.006 |
butcher | +0.006 |
microwave | +0.005 |
piping | +0.005 |
paramed | +0.005 |
Neg logit contributions | |
shapes | -0.004 |
ide | -0.003 |
planets | -0.003 |
winning | -0.003 |
being | -0.003 |
Token | to |
Token ID | 284 |
Feature activation | +8.94e-02 |
Pos logit contributions | |
rolley | +0.035 |
microwave | +0.034 |
butcher | +0.034 |
frying | +0.034 |
pan | +0.032 |
Neg logit contributions | |
ide | -0.047 |
icated | -0.043 |
winning | -0.042 |
seem | -0.042 |
hip | -0.040 |
Token | examine |
Token ID | 10716 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
ide | +0.002 |
winning | +0.002 |
icated | +0.002 |
happening | +0.002 |
hip | +0.002 |
Neg logit contributions | |
frying | -0.003 |
microwave | -0.003 |
pouch | -0.003 |
pan | -0.003 |
piping | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | +6.57e-01 |
Pos logit contributions | |
pan | +0.057 |
microwave | +0.056 |
frying | +0.055 |
butcher | +0.055 |
piping | +0.055 |
Neg logit contributions | |
ide | -0.061 |
planets | -0.059 |
icated | -0.059 |
winning | -0.055 |
textures | -0.054 |
Token | handing |
Token ID | 22786 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
ide | +0.021 |
icated | +0.020 |
planets | +0.019 |
textures | +0.019 |
matter | +0.018 |
Neg logit contributions | |
butcher | -0.017 |
microwave | -0.017 |
pan | -0.017 |
frying | -0.017 |
paramed | -0.016 |
Token | him |
Token ID | 683 |
Feature activation | +6.12e-01 |
Pos logit contributions | |
rolley | +0.083 |
frying | +0.079 |
butcher | +0.078 |
piping | +0.078 |
microwave | +0.078 |
Neg logit contributions | |
ide | -0.025 |
winning | -0.022 |
icated | -0.020 |
seem | -0.020 |
being | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | -9.71e-01 |
Pos logit contributions | |
ide | +0.014 |
winning | +0.013 |
seem | +0.012 |
icated | +0.012 |
shooting | +0.012 |
Neg logit contributions | |
rolley | -0.022 |
butcher | -0.021 |
microwave | -0.021 |
piping | -0.020 |
frying | -0.020 |
Token | magn |
Token ID | 7842 |
Feature activation | -8.84e-01 |
Pos logit contributions | |
rolley | +0.028 |
butcher | +0.027 |
microwave | +0.026 |
piping | +0.025 |
frying | +0.025 |
Neg logit contributions | |
ide | -0.028 |
winning | -0.027 |
seem | -0.026 |
icated | -0.025 |
shooting | -0.025 |
Token | ifying |
Token ID | 4035 |
Feature activation | -7.62e+00 |
Pos logit contributions | |
rolley | +0.011 |
microwave | +0.009 |
butcher | +0.009 |
piping | +0.008 |
frying | +0.008 |
Neg logit contributions | |
ide | -0.041 |
winning | -0.039 |
seem | -0.039 |
icated | -0.038 |
sharing | -0.038 |
Token | glass |
Token ID | 5405 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
rolley | +0.266 |
butcher | +0.257 |
use | +0.244 |
aked | +0.243 |
ilia | +0.237 |
Neg logit contributions | |
planet | -0.162 |
shapes | -0.159 |
color | -0.157 |
ide | -0.153 |
shooting | -0.149 |
Token | . |
Token ID | 13 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
rolley | +0.084 |
butcher | +0.082 |
microwave | +0.080 |
piping | +0.079 |
frying | +0.078 |
Neg logit contributions | |
ide | -0.046 |
winning | -0.040 |
seem | -0.039 |
shapes | -0.038 |
icated | -0.037 |
Token | She |
Token ID | 1375 |
Feature activation | -8.94e-01 |
Pos logit contributions | |
rolley | +0.008 |
butcher | +0.008 |
microwave | +0.008 |
frying | +0.008 |
piping | +0.008 |
Neg logit contributions | |
ide | -0.007 |
winning | -0.006 |
seem | -0.006 |
icated | -0.006 |
shooting | -0.006 |
Token | looked |
Token ID | 3114 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
rolley | +0.028 |
butcher | +0.027 |
microwave | +0.026 |
piping | +0.026 |
frying | +0.025 |
Neg logit contributions | |
ide | -0.024 |
winning | -0.022 |
seem | -0.021 |
shooting | -0.021 |
shapes | -0.021 |
Token | around |
Token ID | 1088 |
Feature activation | +2.93e+00 |
Pos logit contributions | |
rolley | +0.068 |
butcher | +0.065 |
microwave | +0.063 |
piping | +0.062 |
frying | +0.061 |
Neg logit contributions | |
ide | -0.043 |
winning | -0.040 |
seem | -0.038 |
shooting | -0.037 |
sharing | -0.036 |
Token | When |
Token ID | 1649 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
ide | +0.018 |
winning | +0.018 |
seem | +0.017 |
icated | +0.016 |
sharing | +0.016 |
Neg logit contributions | |
rolley | -0.026 |
butcher | -0.025 |
microwave | -0.024 |
frying | -0.024 |
piping | -0.024 |
Token | Daisy |
Token ID | 40355 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
ide | +0.018 |
winning | +0.018 |
seem | +0.017 |
being | +0.016 |
shooting | +0.016 |
Neg logit contributions | |
rolley | -0.042 |
butcher | -0.040 |
microwave | -0.039 |
piping | -0.039 |
frying | -0.038 |
Token | put |
Token ID | 1234 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
ide | +0.026 |
seem | +0.025 |
winning | +0.023 |
shooting | +0.022 |
being | +0.022 |
Neg logit contributions | |
rolley | -0.035 |
butcher | -0.032 |
microwave | -0.032 |
piping | -0.031 |
waitress | -0.030 |
Token | the |
Token ID | 262 |
Feature activation | -6.93e-01 |
Pos logit contributions | |
rolley | +0.083 |
butcher | +0.077 |
paramed | +0.076 |
waitress | +0.073 |
frying | +0.073 |
Neg logit contributions | |
ide | -0.033 |
winning | -0.033 |
being | -0.032 |
shapes | -0.032 |
shooting | -0.032 |
Token | magn |
Token ID | 7842 |
Feature activation | -9.19e-01 |
Pos logit contributions | |
rolley | +0.018 |
butcher | +0.017 |
microwave | +0.016 |
piping | +0.016 |
frying | +0.016 |
Neg logit contributions | |
ide | -0.021 |
winning | -0.021 |
shapes | -0.020 |
shooting | -0.020 |
seem | -0.020 |
Token | ifying |
Token ID | 4035 |
Feature activation | -7.57e+00 |
Pos logit contributions | |
rolley | +0.014 |
butcher | +0.013 |
microwave | +0.013 |
frying | +0.012 |
piping | +0.012 |
Neg logit contributions | |
ide | -0.039 |
winning | -0.037 |
seem | -0.037 |
icated | -0.036 |
shooting | -0.036 |
Token | glass |
Token ID | 5405 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
butcher | +0.239 |
rolley | +0.229 |
use | +0.221 |
paramed | +0.219 |
waitress | +0.212 |
Neg logit contributions | |
ide | -0.191 |
shapes | -0.182 |
planet | -0.180 |
color | -0.178 |
appear | -0.170 |
Token | down |
Token ID | 866 |
Feature activation | -4.37e+00 |
Pos logit contributions | |
rolley | +0.064 |
butcher | +0.064 |
paramed | +0.061 |
waitress | +0.059 |
microwave | +0.059 |
Neg logit contributions | |
ide | -0.030 |
shapes | -0.025 |
seem | -0.023 |
sometimes | -0.022 |
patterns | -0.021 |
Token | , |
Token ID | 11 |
Feature activation | +8.75e-03 |
Pos logit contributions | |
rolley | +0.176 |
butcher | +0.169 |
microwave | +0.166 |
frying | +0.162 |
piping | +0.161 |
Neg logit contributions | |
ide | -0.083 |
winning | -0.073 |
seem | -0.071 |
shapes | -0.066 |
being | -0.065 |
Token | the |
Token ID | 262 |
Feature activation | -6.51e-01 |
Pos logit contributions | |
ide | +0.000 |
winning | +0.000 |
icated | +0.000 |
seem | +0.000 |
shooting | +0.000 |
Neg logit contributions | |
rolley | -0.000 |
butcher | -0.000 |
microwave | -0.000 |
piping | -0.000 |
frying | -0.000 |
Token | cater |
Token ID | 20825 |
Feature activation | -2.39e-01 |
Pos logit contributions | |
rolley | +0.019 |
butcher | +0.018 |
microwave | +0.018 |
piping | +0.017 |
frying | +0.017 |
Neg logit contributions | |
ide | -0.019 |
winning | -0.018 |
seem | -0.017 |
icated | -0.017 |
shooting | -0.017 |
Token | cried |
Token ID | 16896 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
ide | +0.032 |
winning | +0.031 |
seem | +0.030 |
icated | +0.029 |
shooting | +0.029 |
Neg logit contributions | |
rolley | -0.022 |
butcher | -0.020 |
microwave | -0.020 |
piping | -0.019 |
frying | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | +6.44e-01 |
Pos logit contributions | |
rolley | +0.062 |
butcher | +0.059 |
microwave | +0.059 |
piping | +0.057 |
frying | +0.057 |
Neg logit contributions | |
ide | -0.032 |
winning | -0.029 |
seem | -0.028 |
icated | -0.026 |
shooting | -0.026 |
Token | dropping |
Token ID | 12047 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
ide | +0.018 |
icated | +0.017 |
planets | +0.015 |
seem | +0.015 |
hip | +0.015 |
Neg logit contributions | |
pan | -0.020 |
frying | -0.020 |
microwave | -0.020 |
butcher | -0.019 |
pouch | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
rolley | +0.055 |
butcher | +0.053 |
microwave | +0.052 |
piping | +0.052 |
frying | +0.051 |
Neg logit contributions | |
ide | -0.027 |
winning | -0.025 |
seem | -0.024 |
shapes | -0.023 |
shooting | -0.023 |
Token | magn |
Token ID | 7842 |
Feature activation | -9.29e-01 |
Pos logit contributions | |
rolley | +0.053 |
butcher | +0.051 |
microwave | +0.051 |
frying | +0.048 |
pan | +0.048 |
Neg logit contributions | |
ide | -0.064 |
winning | -0.059 |
seem | -0.057 |
icated | -0.056 |
sharing | -0.056 |
Token | ifying |
Token ID | 4035 |
Feature activation | -7.54e+00 |
Pos logit contributions | |
rolley | +0.012 |
butcher | +0.011 |
microwave | +0.011 |
piping | +0.010 |
frying | +0.010 |
Neg logit contributions | |
ide | -0.042 |
winning | -0.040 |
seem | -0.039 |
icated | -0.039 |
shooting | -0.038 |
Token | glass |
Token ID | 5405 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
use | +0.279 |
butcher | +0.272 |
rolley | +0.270 |
ip | +0.268 |
dollars | +0.267 |
Neg logit contributions | |
planet | -0.147 |
color | -0.138 |
shapes | -0.134 |
glow | -0.133 |
ide | -0.123 |
Token | . |
Token ID | 13 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
rolley | +0.083 |
butcher | +0.080 |
piping | +0.078 |
microwave | +0.077 |
frying | +0.076 |
Neg logit contributions | |
ide | -0.047 |
winning | -0.040 |
seem | -0.039 |
shapes | -0.038 |
icated | -0.037 |
Token |
|
Token ID | 198 |
Feature activation | +3.73e-01 |
Pos logit contributions | |
rolley | +0.009 |
butcher | +0.008 |
microwave | +0.008 |
piping | +0.008 |
frying | +0.008 |
Neg logit contributions | |
ide | -0.005 |
winning | -0.004 |
seem | -0.004 |
shooting | -0.004 |
being | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
ide | +0.008 |
winning | +0.007 |
seem | +0.007 |
shooting | +0.006 |
not | +0.006 |
Neg logit contributions | |
rolley | -0.014 |
microwave | -0.013 |
butcher | -0.013 |
frying | -0.013 |
piping | -0.013 |
Token | " |
Token ID | 1 |
Feature activation | -2.75e+00 |
Pos logit contributions | |
ide | +0.007 |
winning | +0.006 |
seem | +0.006 |
icated | +0.006 |
shooting | +0.006 |
Neg logit contributions | |
rolley | -0.008 |
microwave | -0.008 |
butcher | -0.007 |
piping | -0.007 |
frying | -0.007 |
Token | truck |
Token ID | 7779 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
butcher | +0.004 |
microwave | +0.004 |
rolley | +0.004 |
waitress | +0.004 |
bag | +0.004 |
Neg logit contributions | |
ide | -0.005 |
seem | -0.004 |
icated | -0.004 |
winning | -0.004 |
aching | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +7.26e-02 |
Pos logit contributions | |
rolley | +0.055 |
butcher | +0.055 |
microwave | +0.054 |
carrier | +0.053 |
bag | +0.053 |
Neg logit contributions | |
ide | -0.047 |
icated | -0.045 |
winning | -0.044 |
rhythms | -0.042 |
textures | -0.041 |
Token | He |
Token ID | 679 |
Feature activation | -2.50e-01 |
Pos logit contributions | |
ide | +0.002 |
winning | +0.001 |
seem | +0.001 |
icated | +0.001 |
shapes | +0.001 |
Neg logit contributions | |
rolley | -0.003 |
butcher | -0.003 |
frying | -0.003 |
piping | -0.003 |
microwave | -0.002 |
Token | opened |
Token ID | 4721 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
rolley | +0.008 |
butcher | +0.008 |
microwave | +0.008 |
piping | +0.008 |
frying | +0.008 |
Neg logit contributions | |
ide | -0.006 |
winning | -0.006 |
seem | -0.006 |
shooting | -0.005 |
shapes | -0.005 |
Token | it |
Token ID | 340 |
Feature activation | -3.03e+00 |
Pos logit contributions | |
rolley | +0.040 |
microwave | +0.038 |
butcher | +0.038 |
piping | +0.037 |
pan | +0.037 |
Neg logit contributions | |
ide | -0.025 |
winning | -0.023 |
seem | -0.022 |
icated | -0.022 |
sharing | -0.021 |
Token | quickly |
Token ID | 2952 |
Feature activation | -7.45e+00 |
Pos logit contributions | |
rolley | +0.113 |
microwave | +0.110 |
butcher | +0.106 |
bag | +0.104 |
pan | +0.104 |
Neg logit contributions | |
ide | -0.060 |
winning | -0.060 |
icated | -0.058 |
seem | -0.054 |
sharing | -0.052 |
Token | and |
Token ID | 290 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
rolley | +0.325 |
butcher | +0.316 |
microwave | +0.311 |
piping | +0.305 |
frying | +0.303 |
Neg logit contributions | |
ide | -0.109 |
winning | -0.096 |
seem | -0.090 |
icated | -0.083 |
shooting | -0.083 |
Token | looked |
Token ID | 3114 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
rolley | +0.005 |
butcher | +0.005 |
piping | +0.005 |
microwave | +0.004 |
paramed | +0.004 |
Neg logit contributions | |
ide | -0.003 |
shooting | -0.003 |
when | -0.003 |
seem | -0.003 |
being | -0.003 |
Token | inside |
Token ID | 2641 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
rolley | +0.086 |
butcher | +0.083 |
microwave | +0.080 |
piping | +0.079 |
paramed | +0.078 |
Neg logit contributions | |
ide | -0.042 |
winning | -0.040 |
shooting | -0.039 |
shapes | -0.038 |
seem | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | -7.51e-02 |
Pos logit contributions | |
rolley | +0.042 |
microwave | +0.041 |
pan | +0.040 |
frying | +0.039 |
butcher | +0.039 |
Neg logit contributions | |
ide | -0.038 |
icated | -0.036 |
seem | -0.034 |
winning | -0.034 |
hip | -0.033 |
Token | But |
Token ID | 887 |
Feature activation | +2.27e+00 |
Pos logit contributions | |
rolley | +0.003 |
butcher | +0.003 |
frying | +0.003 |
piping | +0.003 |
microwave | +0.003 |
Neg logit contributions | |
ide | -0.002 |
winning | -0.001 |
seem | -0.001 |
icated | -0.001 |
shapes | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
butcher | +0.049 |
rolley | +0.049 |
bag | +0.048 |
microwave | +0.048 |
use | +0.046 |
Neg logit contributions | |
ide | -0.038 |
winning | -0.037 |
planets | -0.037 |
icated | -0.036 |
textures | -0.034 |
Token | handed |
Token ID | 10158 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
being | +0.022 |
sharing | +0.022 |
winning | +0.022 |
ide | +0.021 |
not | +0.021 |
Neg logit contributions | |
rolley | -0.037 |
butcher | -0.034 |
microwave | -0.034 |
piping | -0.033 |
frying | -0.033 |
Token | him |
Token ID | 683 |
Feature activation | +6.13e-01 |
Pos logit contributions | |
rolley | +0.084 |
butcher | +0.081 |
microwave | +0.080 |
piping | +0.078 |
frying | +0.078 |
Neg logit contributions | |
ide | -0.043 |
winning | -0.039 |
seem | -0.037 |
icated | -0.036 |
shooting | -0.034 |
Token | the |
Token ID | 262 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
ide | +0.015 |
winning | +0.014 |
icated | +0.013 |
seem | +0.013 |
shooting | +0.013 |
Neg logit contributions | |
rolley | -0.020 |
butcher | -0.020 |
microwave | -0.019 |
frying | -0.019 |
pan | -0.019 |
Token | magn |
Token ID | 7842 |
Feature activation | -6.71e-01 |
Pos logit contributions | |
rolley | +0.034 |
pouch | +0.033 |
microwave | +0.033 |
piping | +0.032 |
butcher | +0.032 |
Neg logit contributions | |
ide | -0.035 |
winning | -0.031 |
icated | -0.030 |
seem | -0.029 |
sharing | -0.028 |
Token | ifying |
Token ID | 4035 |
Feature activation | -7.42e+00 |
Pos logit contributions | |
rolley | +0.010 |
microwave | +0.008 |
butcher | +0.008 |
piping | +0.008 |
pouch | +0.008 |
Neg logit contributions | |
ide | -0.029 |
winning | -0.029 |
seem | -0.028 |
icated | -0.027 |
sharing | -0.027 |
Token | glass |
Token ID | 5405 |
Feature activation | -2.10e+00 |
Pos logit contributions | |
butcher | +0.274 |
rolley | +0.271 |
stret | +0.258 |
aked | +0.253 |
frying | +0.252 |
Neg logit contributions | |
planet | -0.154 |
color | -0.146 |
shapes | -0.144 |
shooting | -0.135 |
glow | -0.134 |
Token | . |
Token ID | 13 |
Feature activation | -8.14e-02 |
Pos logit contributions | |
rolley | +0.078 |
butcher | +0.075 |
microwave | +0.074 |
piping | +0.073 |
frying | +0.072 |
Neg logit contributions | |
ide | -0.044 |
winning | -0.040 |
seem | -0.038 |
icated | -0.037 |
shooting | -0.036 |
Token |
|
Token ID | 198 |
Feature activation | +4.37e-01 |
Pos logit contributions | |
rolley | +0.003 |
frying | +0.003 |
butcher | +0.003 |
microwave | +0.003 |
paramed | +0.003 |
Neg logit contributions | |
ide | -0.002 |
winning | -0.002 |
seem | -0.001 |
sharing | -0.001 |
being | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +3.27e-01 |
Pos logit contributions | |
ide | +0.011 |
winning | +0.009 |
seem | +0.009 |
icated | +0.009 |
shooting | +0.009 |
Neg logit contributions | |
rolley | -0.015 |
butcher | -0.014 |
microwave | -0.014 |
frying | -0.014 |
piping | -0.014 |
Token | He |
Token ID | 1544 |
Feature activation | -8.83e-01 |
Pos logit contributions | |
ide | +0.009 |
winning | +0.008 |
seem | +0.008 |
icated | +0.008 |
shooting | +0.007 |
Neg logit contributions | |
rolley | -0.010 |
butcher | -0.010 |
microwave | -0.010 |
piping | -0.010 |
frying | -0.010 |
Token | said |
Token ID | 531 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
ide | +0.033 |
winning | +0.031 |
seem | +0.031 |
icated | +0.030 |
shooting | +0.029 |
Neg logit contributions | |
rolley | -0.025 |
butcher | -0.024 |
microwave | -0.024 |
piping | -0.023 |
frying | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | +9.77e-01 |
Pos logit contributions | |
rolley | +0.057 |
microwave | +0.055 |
butcher | +0.055 |
frying | +0.054 |
piping | +0.054 |
Neg logit contributions | |
ide | -0.041 |
icated | -0.039 |
winning | -0.038 |
planets | -0.037 |
seem | -0.035 |
Token | holding |
Token ID | 4769 |
Feature activation | -8.78e-01 |
Pos logit contributions | |
icated | +0.029 |
ide | +0.028 |
planets | +0.026 |
hips | +0.025 |
sizes | +0.025 |
Neg logit contributions | |
pan | -0.029 |
frying | -0.029 |
microwave | -0.026 |
paramed | -0.026 |
butcher | -0.026 |
Token | the |
Token ID | 262 |
Feature activation | -9.23e-01 |
Pos logit contributions | |
rolley | +0.032 |
butcher | +0.031 |
microwave | +0.031 |
piping | +0.030 |
frying | +0.030 |
Neg logit contributions | |
ide | -0.019 |
winning | -0.018 |
seem | -0.017 |
icated | -0.016 |
shooting | -0.016 |
Token | magn |
Token ID | 7842 |
Feature activation | -5.72e-01 |
Pos logit contributions | |
rolley | +0.025 |
butcher | +0.024 |
microwave | +0.023 |
frying | +0.022 |
piping | +0.022 |
Neg logit contributions | |
ide | -0.029 |
winning | -0.027 |
seem | -0.027 |
icated | -0.026 |
sharing | -0.025 |
Token | ifying |
Token ID | 4035 |
Feature activation | -7.39e+00 |
Pos logit contributions | |
rolley | +0.007 |
microwave | +0.006 |
butcher | +0.005 |
piping | +0.005 |
frying | +0.005 |
Neg logit contributions | |
ide | -0.026 |
winning | -0.026 |
seem | -0.025 |
icated | -0.025 |
sharing | -0.025 |
Token | glass |
Token ID | 5405 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
rolley | +0.274 |
use | +0.261 |
butcher | +0.258 |
stret | +0.249 |
aked | +0.249 |
Neg logit contributions | |
planet | -0.147 |
color | -0.142 |
shapes | -0.142 |
glow | -0.135 |
appear | -0.132 |
Token | over |
Token ID | 625 |
Feature activation | -3.27e+00 |
Pos logit contributions | |
rolley | +0.059 |
microwave | +0.057 |
butcher | +0.056 |
frying | +0.055 |
pouch | +0.054 |
Neg logit contributions | |
ide | -0.039 |
winning | -0.037 |
seem | -0.035 |
icated | -0.035 |
shooting | -0.033 |
Token | a |
Token ID | 257 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
rolley | +0.106 |
butcher | +0.104 |
microwave | +0.102 |
pan | +0.100 |
etchup | +0.099 |
Neg logit contributions | |
ide | -0.080 |
icated | -0.075 |
seem | -0.075 |
winning | -0.073 |
hip | -0.068 |
Token | black |
Token ID | 2042 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
rolley | +0.046 |
butcher | +0.043 |
microwave | +0.042 |
piping | +0.041 |
frying | +0.041 |
Neg logit contributions | |
ide | -0.041 |
winning | -0.038 |
seem | -0.037 |
shooting | -0.036 |
icated | -0.036 |
Token | ant |
Token ID | 1885 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
rolley | +0.031 |
butcher | +0.029 |
microwave | +0.029 |
frying | +0.028 |
piping | +0.028 |
Neg logit contributions | |
ide | -0.030 |
winning | -0.029 |
seem | -0.027 |
icated | -0.027 |
shooting | -0.026 |
Token | do |
Token ID | 466 |
Feature activation | +3.34e+00 |
Pos logit contributions | |
ide | +0.067 |
icated | +0.066 |
textures | +0.064 |
planets | +0.063 |
rhythms | +0.060 |
Neg logit contributions | |
frying | -0.088 |
pan | -0.087 |
butcher | -0.086 |
rolley | -0.085 |
bag | -0.085 |
Token | . |
Token ID | 13 |
Feature activation | +6.29e-01 |
Pos logit contributions | |
ide | +0.084 |
icated | +0.081 |
winning | +0.073 |
textures | +0.073 |
planets | +0.073 |
Neg logit contributions | |
rolley | -0.100 |
butcher | -0.094 |
pan | -0.094 |
microwave | -0.094 |
frying | -0.093 |
Token | She |
Token ID | 1375 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
ide | +0.015 |
winning | +0.014 |
seem | +0.013 |
being | +0.012 |
icated | +0.012 |
Neg logit contributions | |
rolley | -0.022 |
frying | -0.021 |
microwave | -0.021 |
butcher | -0.021 |
piping | -0.020 |
Token | opened |
Token ID | 4721 |
Feature activation | -8.89e-01 |
Pos logit contributions | |
rolley | +0.004 |
butcher | +0.004 |
microwave | +0.004 |
piping | +0.004 |
frying | +0.004 |
Neg logit contributions | |
ide | -0.003 |
winning | -0.003 |
seem | -0.003 |
shooting | -0.003 |
icated | -0.003 |
Token | her |
Token ID | 607 |
Feature activation | -2.46e+00 |
Pos logit contributions | |
rolley | +0.034 |
butcher | +0.033 |
microwave | +0.033 |
frying | +0.032 |
piping | +0.032 |
Neg logit contributions | |
ide | -0.018 |
winning | -0.016 |
seem | -0.015 |
shooting | -0.015 |
icated | -0.014 |
Token | basket |
Token ID | 7988 |
Feature activation | -7.38e+00 |
Pos logit contributions | |
rolley | +0.070 |
butcher | +0.065 |
waitress | +0.062 |
paramed | +0.061 |
izz | +0.061 |
Neg logit contributions | |
winning | -0.072 |
shooting | -0.071 |
shapes | -0.069 |
planets | -0.067 |
patterns | -0.067 |
Token | and |
Token ID | 290 |
Feature activation | -9.70e-02 |
Pos logit contributions | |
rolley | +0.302 |
butcher | +0.289 |
microwave | +0.289 |
pouch | +0.279 |
frying | +0.279 |
Neg logit contributions | |
ide | -0.126 |
winning | -0.116 |
icated | -0.111 |
seem | -0.111 |
shooting | -0.100 |
Token | put |
Token ID | 1234 |
Feature activation | -2.30e+00 |
Pos logit contributions | |
rolley | +0.003 |
butcher | +0.003 |
microwave | +0.003 |
piping | +0.003 |
frying | +0.003 |
Neg logit contributions | |
ide | -0.002 |
winning | -0.002 |
shooting | -0.002 |
seem | -0.002 |
being | -0.002 |
Token | her |
Token ID | 607 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
rolley | +0.098 |
butcher | +0.094 |
paramed | +0.090 |
frying | +0.089 |
enn | +0.089 |
Neg logit contributions | |
ide | -0.034 |
shapes | -0.033 |
being | -0.031 |
seem | -0.030 |
winning | -0.030 |
Token | blue |
Token ID | 4171 |
Feature activation | -6.86e-01 |
Pos logit contributions | |
rolley | +0.030 |
butcher | +0.029 |
microwave | +0.028 |
frying | +0.027 |
piping | +0.027 |
Neg logit contributions | |
ide | -0.034 |
winning | -0.033 |
seem | -0.031 |
icated | -0.031 |
shooting | -0.030 |
Token | berries |
Token ID | 20853 |
Feature activation | -7.32e-01 |
Pos logit contributions | |
rolley | +0.019 |
butcher | +0.018 |
microwave | +0.017 |
frying | +0.017 |
piping | +0.017 |
Neg logit contributions | |
ide | -0.021 |
winning | -0.020 |
seem | -0.019 |
shapes | -0.018 |
shooting | -0.018 |
Token | forest |
Token ID | 8222 |
Feature activation | +3.20e+00 |
Pos logit contributions | |
rolley | +0.054 |
butcher | +0.052 |
microwave | +0.051 |
piping | +0.050 |
waitress | +0.050 |
Neg logit contributions | |
ide | -0.037 |
winning | -0.036 |
shapes | -0.034 |
shooting | -0.034 |
planets | -0.034 |
Token | , |
Token ID | 11 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
icated | +0.093 |
matter | +0.084 |
ide | +0.079 |
winning | +0.077 |
seem | +0.074 |
Neg logit contributions | |
bag | -0.089 |
rolley | -0.088 |
pan | -0.087 |
basket | -0.086 |
pouch | -0.086 |
Token | having |
Token ID | 1719 |
Feature activation | +3.22e+00 |
Pos logit contributions | |
ide | +0.043 |
icated | +0.043 |
matter | +0.041 |
hip | +0.039 |
seem | +0.038 |
Neg logit contributions | |
pan | -0.039 |
bag | -0.039 |
basket | -0.039 |
frying | -0.037 |
pouch | -0.036 |
Token | lots |
Token ID | 6041 |
Feature activation | -9.32e-01 |
Pos logit contributions | |
ide | +0.095 |
hip | +0.083 |
icated | +0.082 |
winning | +0.082 |
happening | +0.081 |
Neg logit contributions | |
rolley | -0.087 |
P | -0.085 |
butcher | -0.085 |
microwave | -0.085 |
pouch | -0.084 |
Token | of |
Token ID | 286 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
rolley | +0.018 |
etchup | +0.018 |
butcher | +0.016 |
P | +0.016 |
piping | +0.016 |
Neg logit contributions | |
ide | -0.033 |
winning | -0.032 |
hip | -0.032 |
icated | -0.030 |
happening | -0.030 |
Token | fun |
Token ID | 1257 |
Feature activation | +5.21e+00 |
Pos logit contributions | |
ide | +0.036 |
hip | +0.031 |
winning | +0.030 |
icated | +0.030 |
seem | +0.029 |
Neg logit contributions | |
P | -0.040 |
microwave | -0.040 |
rolley | -0.040 |
butcher | -0.040 |
pouch | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | +7.51e-01 |
Pos logit contributions | |
hip | +0.151 |
icated | +0.148 |
ide | +0.147 |
hips | +0.139 |
winning | +0.136 |
Neg logit contributions | |
P | -0.137 |
etchup | -0.122 |
pan | -0.121 |
rolley | -0.118 |
aked | -0.116 |
Token |
|
Token ID | 198 |
Feature activation | +7.18e-01 |
Pos logit contributions | |
ide | +0.021 |
winning | +0.020 |
seem | +0.019 |
icated | +0.019 |
shooting | +0.018 |
Neg logit contributions | |
rolley | -0.023 |
butcher | -0.022 |
microwave | -0.021 |
frying | -0.020 |
piping | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | +5.32e-01 |
Pos logit contributions | |
ide | +0.020 |
winning | +0.019 |
seem | +0.018 |
icated | +0.018 |
shooting | +0.017 |
Neg logit contributions | |
rolley | -0.022 |
butcher | -0.021 |
microwave | -0.020 |
piping | -0.020 |
frying | -0.019 |
Token | As |
Token ID | 1722 |
Feature activation | +7.07e-01 |
Pos logit contributions | |
ide | +0.014 |
winning | +0.014 |
seem | +0.013 |
textures | +0.013 |
icated | +0.013 |
Neg logit contributions | |
P | -0.016 |
rolley | -0.016 |
butcher | -0.016 |
microwave | -0.015 |
waitress | -0.015 |
Token | they |
Token ID | 484 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
ide | +0.017 |
winning | +0.015 |
seem | +0.015 |
textures | +0.015 |
icated | +0.015 |
Neg logit contributions | |
butcher | -0.024 |
rolley | -0.023 |
microwave | -0.023 |
etchup | -0.023 |
waitress | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | +4.22e-01 |
Pos logit contributions | |
seem | +0.018 |
winning | +0.018 |
ide | +0.018 |
icated | +0.016 |
textures | +0.015 |
Neg logit contributions | |
rolley | -0.036 |
butcher | -0.035 |
microwave | -0.034 |
pan | -0.034 |
pouch | -0.033 |
Token | Charlie |
Token ID | 11526 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
ide | +0.009 |
winning | +0.009 |
seem | +0.008 |
icated | +0.008 |
shooting | +0.008 |
Neg logit contributions | |
rolley | -0.015 |
butcher | -0.015 |
microwave | -0.014 |
piping | -0.014 |
frying | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | +4.28e-01 |
Pos logit contributions | |
ide | +0.048 |
winning | +0.046 |
icated | +0.045 |
seem | +0.043 |
planets | +0.043 |
Neg logit contributions | |
rolley | -0.050 |
butcher | -0.050 |
etchup | -0.048 |
pouch | -0.048 |
frying | -0.048 |
Token | who |
Token ID | 508 |
Feature activation | -6.35e-01 |
Pos logit contributions | |
ide | +0.010 |
winning | +0.009 |
seem | +0.009 |
icated | +0.009 |
patterns | +0.008 |
Neg logit contributions | |
microwave | -0.014 |
rolley | -0.014 |
butcher | -0.014 |
piping | -0.013 |
pan | -0.013 |
Token | were |
Token ID | 547 |
Feature activation | +5.89e-01 |
Pos logit contributions | |
butcher | +0.020 |
rolley | +0.019 |
P | +0.018 |
waitress | +0.018 |
frying | +0.018 |
Neg logit contributions | |
winning | -0.015 |
ide | -0.015 |
icated | -0.015 |
izing | -0.014 |
planets | -0.014 |
Token | playing |
Token ID | 2712 |
Feature activation | +5.76e+00 |
Pos logit contributions | |
ide | +0.013 |
icated | +0.011 |
seem | +0.011 |
winning | +0.011 |
textures | +0.011 |
Neg logit contributions | |
rolley | -0.021 |
butcher | -0.021 |
microwave | -0.020 |
frying | -0.020 |
piping | -0.020 |
Token | together |
Token ID | 1978 |
Feature activation | +2.54e+00 |
Pos logit contributions | |
ide | +0.144 |
seem | +0.139 |
icated | +0.135 |
winning | +0.131 |
textures | +0.123 |
Neg logit contributions | |
P | -0.184 |
rolley | -0.181 |
microwave | -0.179 |
butcher | -0.176 |
pan | -0.172 |
Token | in |
Token ID | 287 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
icated | +0.072 |
seem | +0.071 |
ide | +0.066 |
hip | +0.064 |
rhythms | +0.062 |
Neg logit contributions | |
pan | -0.069 |
P | -0.063 |
microwave | -0.063 |
rolley | -0.062 |
basket | -0.061 |
Token | the |
Token ID | 262 |
Feature activation | +2.78e-01 |
Pos logit contributions | |
ide | +0.039 |
seem | +0.035 |
winning | +0.034 |
icated | +0.034 |
hip | +0.033 |
Neg logit contributions | |
rolley | -0.049 |
piping | -0.048 |
microwave | -0.048 |
butcher | -0.048 |
etchup | -0.046 |
Token | garden |
Token ID | 11376 |
Feature activation | +2.28e+00 |
Pos logit contributions | |
ide | +0.007 |
winning | +0.007 |
seem | +0.007 |
icated | +0.007 |
shooting | +0.006 |
Neg logit contributions | |
rolley | -0.009 |
butcher | -0.008 |
microwave | -0.008 |
piping | -0.008 |
frying | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +3.46e-01 |
Pos logit contributions | |
icated | +0.062 |
seem | +0.061 |
ide | +0.059 |
winning | +0.058 |
matter | +0.057 |
Neg logit contributions | |
rolley | -0.066 |
pan | -0.063 |
butcher | -0.062 |
microwave | -0.062 |
etchup | -0.060 |
Token | avi |
Token ID | 15820 |
Feature activation | +4.43e-01 |
Pos logit contributions | |
winning | +0.005 |
seem | +0.005 |
ide | +0.005 |
shooting | +0.005 |
sharing | +0.005 |
Neg logit contributions | |
rolley | -0.004 |
butcher | -0.004 |
microwave | -0.004 |
piping | -0.004 |
frying | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
ide | +0.009 |
winning | +0.009 |
seem | +0.008 |
icated | +0.008 |
shooting | +0.008 |
Neg logit contributions | |
rolley | -0.016 |
butcher | -0.016 |
microwave | -0.016 |
frying | -0.015 |
piping | -0.015 |
Token | Jay |
Token ID | 9180 |
Feature activation | +2.96e+00 |
Pos logit contributions | |
rolley | +0.005 |
microwave | +0.005 |
butcher | +0.005 |
frying | +0.005 |
etchup | +0.005 |
Neg logit contributions | |
ide | -0.003 |
seem | -0.002 |
winning | -0.002 |
icated | -0.002 |
being | -0.002 |
Token | a |
Token ID | 64 |
Feature activation | -2.86e+00 |
Pos logit contributions | |
ide | +0.065 |
winning | +0.059 |
seem | +0.057 |
icated | +0.055 |
shooting | +0.054 |
Neg logit contributions | |
rolley | -0.107 |
butcher | -0.104 |
microwave | -0.103 |
frying | -0.099 |
piping | -0.099 |
Token | were |
Token ID | 547 |
Feature activation | -5.54e-01 |
Pos logit contributions | |
rolley | +0.105 |
butcher | +0.102 |
microwave | +0.101 |
piping | +0.098 |
frying | +0.098 |
Neg logit contributions | |
ide | -0.059 |
winning | -0.055 |
seem | -0.051 |
icated | -0.051 |
shooting | -0.049 |
Token | playing |
Token ID | 2712 |
Feature activation | +5.39e+00 |
Pos logit contributions | |
rolley | +0.023 |
butcher | +0.023 |
microwave | +0.022 |
frying | +0.022 |
piping | +0.022 |
Neg logit contributions | |
ide | -0.009 |
winning | -0.009 |
shooting | -0.007 |
sharing | -0.007 |
being | -0.007 |
Token | in |
Token ID | 287 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
seem | +0.147 |
icated | +0.145 |
ide | +0.141 |
sizes | +0.131 |
winning | +0.129 |
Neg logit contributions | |
P | -0.166 |
pan | -0.157 |
microwave | -0.153 |
rolley | -0.153 |
bag | -0.147 |
Token | the |
Token ID | 262 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
ide | +0.061 |
hip | +0.060 |
ust | +0.059 |
seem | +0.059 |
icated | +0.058 |
Neg logit contributions | |
pan | -0.051 |
rolley | -0.050 |
piping | -0.049 |
microwave | -0.048 |
etchup | -0.047 |
Token | garden |
Token ID | 11376 |
Feature activation | +2.71e+00 |
Pos logit contributions | |
ide | +0.033 |
winning | +0.031 |
seem | +0.030 |
icated | +0.030 |
shooting | +0.029 |
Neg logit contributions | |
rolley | -0.034 |
butcher | -0.032 |
microwave | -0.032 |
piping | -0.031 |
frying | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
icated | +0.081 |
seem | +0.077 |
matter | +0.073 |
ide | +0.072 |
winning | +0.072 |
Neg logit contributions | |
rolley | -0.075 |
pan | -0.074 |
microwave | -0.068 |
basket | -0.068 |
bag | -0.068 |
Token | Everything |
Token ID | 11391 |
Feature activation | +4.60e+00 |
Pos logit contributions | |
ide | +0.006 |
winning | +0.006 |
icated | +0.006 |
seem | +0.005 |
shooting | +0.005 |
Neg logit contributions | |
rolley | -0.007 |
butcher | -0.006 |
microwave | -0.006 |
frying | -0.006 |
piping | -0.006 |
Token | that |
Token ID | 326 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
ide | +0.037 |
winning | +0.035 |
seem | +0.033 |
icated | +0.032 |
shooting | +0.032 |
Neg logit contributions | |
rolley | -0.056 |
butcher | -0.053 |
microwave | -0.052 |
piping | -0.050 |
frying | -0.050 |
Token | making |
Token ID | 1642 |
Feature activation | +2.82e+00 |
Pos logit contributions | |
ide | +0.005 |
winning | +0.004 |
seem | +0.004 |
icated | +0.004 |
shooting | +0.004 |
Neg logit contributions | |
rolley | -0.008 |
butcher | -0.007 |
microwave | -0.007 |
frying | -0.007 |
piping | -0.007 |
Token | new |
Token ID | 649 |
Feature activation | +4.00e+00 |
Pos logit contributions | |
ide | +0.066 |
winning | +0.059 |
seem | +0.058 |
icated | +0.056 |
hip | +0.055 |
Neg logit contributions | |
rolley | -0.092 |
microwave | -0.091 |
butcher | -0.090 |
frying | -0.088 |
pan | -0.087 |
Token | friends |
Token ID | 2460 |
Feature activation | +3.64e+00 |
Pos logit contributions | |
ide | +0.089 |
winning | +0.079 |
seem | +0.079 |
icated | +0.075 |
hip | +0.072 |
Neg logit contributions | |
rolley | -0.142 |
microwave | -0.135 |
butcher | -0.135 |
piping | -0.132 |
frying | -0.131 |
Token | is |
Token ID | 318 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
ide | +0.091 |
icated | +0.084 |
winning | +0.082 |
wered | +0.079 |
textures | +0.077 |
Neg logit contributions | |
rolley | -0.105 |
pan | -0.102 |
piping | -0.101 |
frying | -0.098 |
� | -0.098 |
Token | fun |
Token ID | 1257 |
Feature activation | +5.65e+00 |
Pos logit contributions | |
ide | +0.059 |
icated | +0.051 |
seem | +0.051 |
winning | +0.049 |
textures | +0.047 |
Neg logit contributions | |
rolley | -0.084 |
butcher | -0.083 |
microwave | -0.082 |
piping | -0.081 |
frying | -0.080 |
Token | . |
Token ID | 13 |
Feature activation | +5.77e-01 |
Pos logit contributions | |
ide | +0.144 |
seem | +0.134 |
icated | +0.132 |
wered | +0.127 |
winning | +0.126 |
Neg logit contributions | |
rolley | -0.167 |
pan | -0.160 |
microwave | -0.155 |
frying | -0.153 |
butcher | -0.153 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.90e-01 |
Pos logit contributions | |
ide | +0.015 |
winning | +0.014 |
seem | +0.014 |
icated | +0.013 |
shooting | +0.013 |
Neg logit contributions | |
rolley | -0.018 |
butcher | -0.018 |
microwave | -0.017 |
frying | -0.017 |
piping | -0.017 |
Token | L |
Token ID | 43 |
Feature activation | -4.26e-01 |
Pos logit contributions | |
rolley | +0.015 |
butcher | +0.015 |
microwave | +0.014 |
frying | +0.014 |
piping | +0.014 |
Neg logit contributions | |
ide | -0.013 |
winning | -0.012 |
seem | -0.012 |
icated | -0.012 |
shooting | -0.011 |
Token | ily |
Token ID | 813 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
rolley | +0.013 |
butcher | +0.012 |
microwave | +0.012 |
piping | +0.011 |
frying | +0.011 |
Neg logit contributions | |
ide | -0.012 |
winning | -0.012 |
seem | -0.011 |
icated | -0.011 |
shooting | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | -6.35e-01 |
Pos logit contributions | |
rolley | +0.045 |
butcher | +0.044 |
microwave | +0.043 |
piping | +0.042 |
frying | +0.042 |
Neg logit contributions | |
ide | -0.018 |
winning | -0.016 |
seem | -0.015 |
icated | -0.014 |
shooting | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -4.43e-01 |
Pos logit contributions | |
rolley | +0.014 |
butcher | +0.013 |
microwave | +0.013 |
P | +0.012 |
frying | +0.012 |
Neg logit contributions | |
ide | -0.011 |
winning | -0.011 |
seem | -0.011 |
icated | -0.010 |
patterns | -0.010 |
Token | Tim |
Token ID | 5045 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
microwave | +0.013 |
butcher | +0.013 |
rolley | +0.012 |
etchup | +0.012 |
piping | +0.012 |
Neg logit contributions | |
tunes | -0.011 |
ide | -0.011 |
happening | -0.011 |
planets | -0.011 |
seem | -0.011 |
Token | my |
Token ID | 1820 |
Feature activation | +6.17e-01 |
Pos logit contributions | |
winning | +0.069 |
ide | +0.067 |
seem | +0.066 |
shooting | +0.062 |
rhythms | +0.060 |
Neg logit contributions | |
rolley | -0.079 |
microwave | -0.074 |
butcher | -0.074 |
P | -0.072 |
piping | -0.071 |
Token | liked |
Token ID | 8288 |
Feature activation | +3.74e+00 |
Pos logit contributions | |
ide | +0.015 |
winning | +0.014 |
seem | +0.014 |
icated | +0.013 |
textures | +0.012 |
Neg logit contributions | |
rolley | -0.020 |
butcher | -0.020 |
P | -0.019 |
microwave | -0.019 |
frying | -0.019 |
Token | to |
Token ID | 284 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
ide | +0.132 |
seem | +0.120 |
winning | +0.117 |
icated | +0.115 |
hip | +0.114 |
Neg logit contributions | |
butcher | -0.087 |
rolley | -0.086 |
microwave | -0.085 |
frying | -0.081 |
waitress | -0.079 |
Token | play |
Token ID | 711 |
Feature activation | +6.10e+00 |
Pos logit contributions | |
ide | +0.051 |
winning | +0.047 |
icated | +0.043 |
seem | +0.042 |
shooting | +0.041 |
Neg logit contributions | |
rolley | -0.088 |
microwave | -0.086 |
butcher | -0.086 |
frying | -0.084 |
piping | -0.084 |
Token | with |
Token ID | 351 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
ide | +0.154 |
seem | +0.150 |
icated | +0.147 |
winning | +0.146 |
sizes | +0.139 |
Neg logit contributions | |
microwave | -0.190 |
P | -0.189 |
rolley | -0.188 |
butcher | -0.186 |
pan | -0.180 |
Token | his |
Token ID | 465 |
Feature activation | +8.38e-01 |
Pos logit contributions | |
ide | +0.052 |
seem | +0.050 |
winning | +0.049 |
hip | +0.046 |
icated | +0.044 |
Neg logit contributions | |
microwave | -0.068 |
rolley | -0.068 |
butcher | -0.067 |
P | -0.064 |
pan | -0.064 |
Token | toys |
Token ID | 14958 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
ide | +0.023 |
seem | +0.021 |
winning | +0.021 |
icated | +0.020 |
hip | +0.019 |
Neg logit contributions | |
rolley | -0.026 |
butcher | -0.026 |
microwave | -0.025 |
frying | -0.024 |
pouch | -0.024 |
Token | all |
Token ID | 477 |
Feature activation | -5.39e-01 |
Pos logit contributions | |
icated | +0.049 |
ide | +0.048 |
seem | +0.046 |
winning | +0.046 |
rhythms | +0.044 |
Neg logit contributions | |
microwave | -0.059 |
bag | -0.059 |
rolley | -0.058 |
pan | -0.058 |
butcher | -0.057 |
Token | day |
Token ID | 1110 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
rolley | +0.013 |
P | +0.013 |
pan | +0.012 |
etchup | +0.012 |
enn | +0.012 |
Neg logit contributions | |
seem | -0.015 |
winning | -0.015 |
jokes | -0.015 |
communicate | -0.015 |
ide | -0.015 |
Token | a |
Token ID | 257 |
Feature activation | -3.94e-01 |
Pos logit contributions | |
rolley | +0.008 |
microwave | +0.008 |
butcher | +0.008 |
piping | +0.008 |
frying | +0.008 |
Neg logit contributions | |
ide | -0.003 |
winning | -0.003 |
seem | -0.003 |
shapes | -0.002 |
being | -0.002 |
Token | way |
Token ID | 835 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
rolley | +0.014 |
piping | +0.013 |
P | +0.013 |
microwave | +0.012 |
aked | +0.012 |
Neg logit contributions | |
winning | -0.009 |
shooting | -0.009 |
ide | -0.009 |
seem | -0.008 |
shapes | -0.008 |
Token | back |
Token ID | 736 |
Feature activation | -4.84e+00 |
Pos logit contributions | |
rolley | +0.025 |
butcher | +0.022 |
pouch | +0.022 |
microwave | +0.022 |
enn | +0.021 |
Neg logit contributions | |
ide | -0.039 |
icated | -0.038 |
winning | -0.037 |
tunes | -0.037 |
planets | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | -4.46e-02 |
Pos logit contributions | |
rolley | +0.153 |
microwave | +0.151 |
butcher | +0.150 |
waitress | +0.146 |
piping | +0.146 |
Neg logit contributions | |
ide | -0.133 |
winning | -0.123 |
seem | -0.118 |
shooting | -0.113 |
being | -0.113 |
Token | He |
Token ID | 679 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
rolley | +0.001 |
butcher | +0.001 |
microwave | +0.001 |
piping | +0.001 |
frying | +0.001 |
Neg logit contributions | |
ide | -0.001 |
winning | -0.001 |
seem | -0.001 |
icated | -0.001 |
shooting | -0.001 |
Token | sw |
Token ID | 1509 |
Feature activation | +7.59e-01 |
Pos logit contributions | |
ide | +0.007 |
winning | +0.007 |
icated | +0.007 |
seem | +0.007 |
sharing | +0.006 |
Neg logit contributions | |
rolley | -0.007 |
butcher | -0.007 |
microwave | -0.007 |
frying | -0.007 |
pouch | -0.007 |
Token | am |
Token ID | 321 |
Feature activation | -2.83e-01 |
Pos logit contributions | |
ide | +0.024 |
winning | +0.023 |
seem | +0.022 |
icated | +0.022 |
shooting | +0.022 |
Neg logit contributions | |
rolley | -0.020 |
butcher | -0.019 |
microwave | -0.019 |
frying | -0.018 |
pouch | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | +7.75e-01 |
Pos logit contributions | |
rolley | +0.007 |
bag | +0.007 |
pouch | +0.007 |
butcher | +0.007 |
microwave | +0.007 |
Neg logit contributions | |
icated | -0.009 |
ide | -0.008 |
winning | -0.008 |
seem | -0.007 |
sharing | -0.007 |
Token | sw |
Token ID | 1509 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
ide | +0.026 |
icated | +0.024 |
winning | +0.024 |
hip | +0.023 |
seem | +0.023 |
Neg logit contributions | |
pouch | -0.018 |
rolley | -0.018 |
microwave | -0.017 |
bag | -0.017 |
pan | -0.017 |
Token | am |
Token ID | 321 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
ide | +0.036 |
winning | +0.034 |
seem | +0.033 |
icated | +0.033 |
shooting | +0.032 |
Neg logit contributions | |
rolley | -0.028 |
butcher | -0.026 |
microwave | -0.026 |
pouch | -0.025 |
frying | -0.025 |
Token | but |
Token ID | 475 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
pouch | +0.007 |
rolley | +0.007 |
bag | +0.006 |
pan | +0.006 |
butcher | +0.006 |
Neg logit contributions | |
icated | -0.007 |
ide | -0.007 |
winning | -0.006 |
sharing | -0.006 |
planets | -0.006 |
Token | make |
Token ID | 787 |
Feature activation | +3.08e+00 |
Pos logit contributions | |
ide | +0.024 |
icated | +0.023 |
planets | +0.022 |
winning | +0.021 |
sizes | +0.021 |
Neg logit contributions | |
rolley | -0.027 |
butcher | -0.027 |
piping | -0.026 |
waitress | -0.026 |
microwave | -0.026 |
Token | her |
Token ID | 607 |
Feature activation | +1.75e+00 |
Pos logit contributions | |
ide | +0.078 |
winning | +0.069 |
icated | +0.067 |
seem | +0.067 |
textures | +0.063 |
Neg logit contributions | |
rolley | -0.100 |
butcher | -0.097 |
microwave | -0.097 |
piping | -0.093 |
frying | -0.092 |
Token | friend |
Token ID | 1545 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
ide | +0.049 |
winning | +0.045 |
icated | +0.043 |
seem | +0.043 |
shooting | +0.041 |
Neg logit contributions | |
rolley | -0.053 |
butcher | -0.052 |
microwave | -0.051 |
piping | -0.050 |
pouch | -0.049 |
Token | feel |
Token ID | 1254 |
Feature activation | +7.82e-01 |
Pos logit contributions | |
ide | +0.005 |
icated | +0.004 |
winning | +0.004 |
happening | +0.004 |
planets | +0.004 |
Neg logit contributions | |
rolley | -0.006 |
butcher | -0.006 |
piping | -0.006 |
bag | -0.006 |
microwave | -0.006 |
Token | better |
Token ID | 1365 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
ide | +0.026 |
winning | +0.023 |
icated | +0.023 |
planets | +0.022 |
seem | +0.022 |
Neg logit contributions | |
rolley | -0.020 |
butcher | -0.019 |
microwave | -0.019 |
piping | -0.018 |
frying | -0.018 |
Token | by |
Token ID | 416 |
Feature activation | +4.36e+00 |
Pos logit contributions | |
rolley | +0.076 |
butcher | +0.073 |
microwave | +0.072 |
frying | +0.070 |
piping | +0.069 |
Neg logit contributions | |
ide | -0.043 |
winning | -0.039 |
seem | -0.037 |
icated | -0.037 |
shooting | -0.035 |
Token | sharing |
Token ID | 7373 |
Feature activation | +2.71e+00 |
Pos logit contributions | |
ide | +0.106 |
seem | +0.081 |
planets | +0.080 |
hip | +0.078 |
icated | +0.077 |
Neg logit contributions | |
piping | -0.139 |
pan | -0.137 |
paramed | -0.133 |
frying | -0.131 |
etchup | -0.130 |
Token | her |
Token ID | 607 |
Feature activation | +3.55e-01 |
Pos logit contributions | |
ide | +0.072 |
icated | +0.065 |
seem | +0.064 |
winning | +0.063 |
planets | +0.061 |
Neg logit contributions | |
butcher | -0.077 |
rolley | -0.074 |
microwave | -0.074 |
bag | -0.073 |
pan | -0.073 |
Token | toy |
Token ID | 13373 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
ide | +0.011 |
seem | +0.010 |
icated | +0.010 |
winning | +0.010 |
hip | +0.010 |
Neg logit contributions | |
rolley | -0.009 |
bag | -0.009 |
pouch | -0.009 |
butcher | -0.008 |
discount | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +4.26e-01 |
Pos logit contributions | |
rolley | +0.062 |
butcher | +0.060 |
microwave | +0.058 |
pouch | +0.058 |
waitress | +0.057 |
Neg logit contributions | |
ide | -0.045 |
seem | -0.042 |
winning | -0.042 |
icated | -0.041 |
shooting | -0.038 |
Token | From |
Token ID | 3574 |
Feature activation | +9.16e-01 |
Pos logit contributions | |
ide | +0.015 |
winning | +0.014 |
shooting | +0.014 |
icated | +0.014 |
seem | +0.014 |
Neg logit contributions | |
rolley | -0.010 |
butcher | -0.009 |
microwave | -0.009 |
waitress | -0.008 |
piping | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | +4.16e-01 |
Pos logit contributions | |
ide | +0.013 |
winning | +0.012 |
seem | +0.012 |
icated | +0.011 |
shooting | +0.011 |
Neg logit contributions | |
rolley | -0.017 |
butcher | -0.016 |
microwave | -0.016 |
frying | -0.016 |
piping | -0.016 |
Token | The |
Token ID | 464 |
Feature activation | +3.27e-01 |
Pos logit contributions | |
ide | +0.010 |
winning | +0.010 |
seem | +0.010 |
icated | +0.009 |
planets | +0.009 |
Neg logit contributions | |
rolley | -0.014 |
butcher | -0.013 |
P | -0.013 |
microwave | -0.013 |
piping | -0.012 |
Token | bird |
Token ID | 6512 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
ide | +0.010 |
seem | +0.008 |
winning | +0.008 |
icated | +0.008 |
hip | +0.008 |
Neg logit contributions | |
butcher | -0.010 |
rolley | -0.009 |
waitress | -0.009 |
microwave | -0.009 |
pan | -0.009 |
Token | 's |
Token ID | 338 |
Feature activation | +5.12e-01 |
Pos logit contributions | |
ide | +0.039 |
winning | +0.038 |
seem | +0.036 |
icated | +0.036 |
textures | +0.034 |
Neg logit contributions | |
rolley | -0.037 |
butcher | -0.035 |
microwave | -0.035 |
pouch | -0.034 |
P | -0.033 |
Token | friends |
Token ID | 2460 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
ide | +0.017 |
icated | +0.015 |
seem | +0.015 |
winning | +0.014 |
hip | +0.013 |
Neg logit contributions | |
pouch | -0.013 |
butcher | -0.013 |
rolley | -0.013 |
bag | -0.013 |
microwave | -0.013 |
Token | were |
Token ID | 547 |
Feature activation | +9.39e-01 |
Pos logit contributions | |
ide | +0.033 |
winning | +0.033 |
icated | +0.031 |
textures | +0.030 |
patterns | +0.030 |
Neg logit contributions | |
rolley | -0.034 |
butcher | -0.032 |
pan | -0.032 |
pouch | -0.032 |
microwave | -0.031 |
Token | kind |
Token ID | 1611 |
Feature activation | -1.57e-03 |
Pos logit contributions | |
ide | +0.022 |
winning | +0.021 |
seem | +0.020 |
icated | +0.019 |
shooting | +0.019 |
Neg logit contributions | |
rolley | -0.032 |
butcher | -0.031 |
microwave | -0.031 |
piping | -0.030 |
frying | -0.030 |
Token | and |
Token ID | 290 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
rolley | +0.000 |
bag | +0.000 |
butcher | +0.000 |
pouch | +0.000 |
waitress | +0.000 |
Neg logit contributions | |
ide | -0.000 |
icated | -0.000 |
winning | -0.000 |
seem | -0.000 |
wered | -0.000 |
Token | wanted |
Token ID | 2227 |
Feature activation | -9.87e-01 |
Pos logit contributions | |
ide | +0.053 |
winning | +0.049 |
seem | +0.047 |
icated | +0.047 |
shooting | +0.045 |
Neg logit contributions | |
rolley | -0.063 |
butcher | -0.060 |
microwave | -0.059 |
piping | -0.057 |
frying | -0.057 |
Token | to |
Token ID | 284 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
butcher | +0.021 |
rolley | +0.020 |
microwave | +0.020 |
waitress | +0.019 |
frying | +0.019 |
Neg logit contributions | |
ide | -0.036 |
icated | -0.032 |
winning | -0.031 |
seem | -0.031 |
hip | -0.030 |
Token | help |
Token ID | 1037 |
Feature activation | +8.40e-01 |
Pos logit contributions | |
ide | +0.052 |
winning | +0.048 |
icated | +0.047 |
planets | +0.045 |
seem | +0.044 |
Neg logit contributions | |
rolley | -0.048 |
butcher | -0.046 |
microwave | -0.045 |
piping | -0.045 |
frying | -0.044 |
Token | stupid |
Token ID | 8531 |
Feature activation | +5.31e-01 |
Pos logit contributions | |
ide | +0.037 |
planets | +0.035 |
winning | +0.035 |
icated | +0.034 |
aching | +0.033 |
Neg logit contributions | |
rolley | -0.036 |
microwave | -0.035 |
piping | -0.034 |
bag | -0.034 |
pan | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -6.95e-01 |
Pos logit contributions | |
ide | +0.014 |
winning | +0.012 |
icated | +0.012 |
planets | +0.012 |
shooting | +0.012 |
Neg logit contributions | |
rolley | -0.016 |
butcher | -0.016 |
microwave | -0.016 |
waitress | -0.015 |
piping | -0.015 |
Token | He |
Token ID | 679 |
Feature activation | -7.73e-01 |
Pos logit contributions | |
rolley | +0.028 |
frying | +0.026 |
piping | +0.026 |
butcher | +0.026 |
microwave | +0.026 |
Neg logit contributions | |
winning | -0.013 |
ide | -0.013 |
seem | -0.012 |
shapes | -0.011 |
being | -0.011 |
Token | does |
Token ID | 857 |
Feature activation | +2.49e+00 |
Pos logit contributions | |
rolley | +0.021 |
butcher | +0.020 |
microwave | +0.020 |
piping | +0.019 |
frying | +0.019 |
Neg logit contributions | |
ide | -0.024 |
winning | -0.022 |
seem | -0.021 |
icated | -0.021 |
shooting | -0.020 |
Token | not |
Token ID | 407 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
ide | +0.056 |
textures | +0.042 |
planets | +0.042 |
icated | +0.040 |
sizes | +0.040 |
Neg logit contributions | |
microwave | -0.089 |
discount | -0.088 |
butcher | -0.088 |
piping | -0.087 |
� | -0.087 |
Token | see |
Token ID | 766 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
ide | +0.032 |
winning | +0.030 |
icated | +0.027 |
seem | +0.026 |
shooting | +0.026 |
Neg logit contributions | |
rolley | -0.049 |
butcher | -0.048 |
microwave | -0.047 |
piping | -0.046 |
frying | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
ide | +0.025 |
winning | +0.023 |
seem | +0.022 |
icated | +0.021 |
shooting | +0.021 |
Neg logit contributions | |
rolley | -0.047 |
butcher | -0.045 |
microwave | -0.044 |
frying | -0.043 |
piping | -0.043 |
Token | teacher |
Token ID | 4701 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
microwave | +0.004 |
butcher | +0.004 |
rolley | +0.004 |
bag | +0.004 |
pan | +0.004 |
Neg logit contributions | |
ide | -0.006 |
seem | -0.005 |
winning | -0.005 |
icated | -0.005 |
aching | -0.005 |
Token | coming |
Token ID | 2406 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
microwave | +0.033 |
butcher | +0.033 |
rolley | +0.032 |
frying | +0.032 |
pan | +0.032 |
Neg logit contributions | |
icated | -0.027 |
ide | -0.027 |
textures | -0.027 |
planets | -0.025 |
winning | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -3.87e-01 |
Pos logit contributions | |
rolley | +0.052 |
butcher | +0.051 |
microwave | +0.050 |
piping | +0.049 |
frying | +0.049 |
Neg logit contributions | |
ide | -0.037 |
winning | -0.033 |
seem | -0.031 |
shooting | -0.030 |
being | -0.030 |
Token | He |
Token ID | 679 |
Feature activation | -6.65e-01 |
Pos logit contributions | |
rolley | +0.016 |
frying | +0.015 |
butcher | +0.015 |
piping | +0.014 |
pouch | +0.014 |
Neg logit contributions | |
winning | -0.007 |
ide | -0.007 |
seem | -0.006 |
shapes | -0.006 |
being | -0.006 |
Token | " |
Token ID | 366 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
ide | +0.002 |
icated | +0.002 |
textures | +0.002 |
planets | +0.002 |
matter | +0.002 |
Neg logit contributions | |
microwave | -0.002 |
butcher | -0.002 |
paramed | -0.002 |
frying | -0.002 |
pan | -0.002 |
Token | We |
Token ID | 1135 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
rolley | +0.024 |
P | +0.022 |
butcher | +0.021 |
microwave | +0.021 |
etchup | +0.020 |
Neg logit contributions | |
ide | -0.045 |
winning | -0.043 |
seem | -0.042 |
icated | -0.041 |
planets | -0.041 |
Token | told |
Token ID | 1297 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
rolley | +0.048 |
butcher | +0.046 |
microwave | +0.046 |
piping | +0.045 |
frying | +0.045 |
Neg logit contributions | |
ide | -0.028 |
winning | -0.024 |
icated | -0.023 |
seem | -0.022 |
shooting | -0.022 |
Token | you |
Token ID | 345 |
Feature activation | +5.86e-01 |
Pos logit contributions | |
butcher | +0.033 |
microwave | +0.033 |
rolley | +0.032 |
waitress | +0.032 |
piping | +0.032 |
Neg logit contributions | |
ide | -0.037 |
winning | -0.033 |
textures | -0.032 |
sizes | -0.031 |
seem | -0.031 |
Token | to |
Token ID | 284 |
Feature activation | +7.66e-01 |
Pos logit contributions | |
ide | +0.016 |
hip | +0.014 |
tunes | +0.013 |
planets | +0.013 |
aching | +0.013 |
Neg logit contributions | |
butcher | -0.016 |
microwave | -0.016 |
bag | -0.016 |
rolley | -0.015 |
waitress | -0.015 |
Token | be |
Token ID | 307 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
ide | +0.017 |
winning | +0.016 |
icated | +0.016 |
planets | +0.016 |
seem | +0.015 |
Neg logit contributions | |
butcher | -0.025 |
rolley | -0.024 |
use | -0.024 |
microwave | -0.024 |
frying | -0.024 |
Token | careful |
Token ID | 8161 |
Feature activation | +2.96e+00 |
Pos logit contributions | |
winning | +0.021 |
not | +0.020 |
being | +0.020 |
sharing | +0.020 |
songs | +0.018 |
Neg logit contributions | |
rolley | -0.047 |
microwave | -0.042 |
butcher | -0.042 |
piping | -0.042 |
pouch | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | +7.72e-01 |
Pos logit contributions | |
icated | +0.083 |
planets | +0.083 |
ide | +0.082 |
matter | +0.081 |
seem | +0.080 |
Neg logit contributions | |
pan | -0.074 |
rolley | -0.074 |
frying | -0.070 |
'," | -0.070 |
microwave | -0.070 |
Token | Why |
Token ID | 4162 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
ide | +0.023 |
winning | +0.021 |
seem | +0.021 |
icated | +0.020 |
shooting | +0.020 |
Neg logit contributions | |
rolley | -0.022 |
butcher | -0.021 |
microwave | -0.021 |
piping | -0.020 |
frying | -0.020 |
Token | didn |
Token ID | 1422 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
ide | +0.045 |
tunes | +0.041 |
hip | +0.039 |
galaxies | +0.038 |
winning | +0.038 |
Neg logit contributions | |
� | -0.048 |
bag | -0.044 |
microwave | -0.044 |
pan | -0.044 |
use | -0.042 |
Token | 't |
Token ID | 470 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
ide | +0.055 |
textures | +0.050 |
winning | +0.049 |
sizes | +0.047 |
planets | +0.047 |
Neg logit contributions | |
butcher | -0.081 |
microwave | -0.080 |
rolley | -0.080 |
piping | -0.078 |
frying | -0.076 |
Token | ed |
Token ID | 276 |
Feature activation | -2.94e+00 |
Pos logit contributions | |
ide | +0.029 |
winning | +0.028 |
seem | +0.027 |
icated | +0.027 |
shooting | +0.026 |
Neg logit contributions | |
rolley | -0.008 |
butcher | -0.007 |
microwave | -0.007 |
piping | -0.006 |
frying | -0.006 |
Token | loudly |
Token ID | 23112 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
rolley | +0.107 |
butcher | +0.104 |
microwave | +0.104 |
frying | +0.099 |
pouch | +0.099 |
Neg logit contributions | |
ide | -0.063 |
winning | -0.057 |
seem | -0.055 |
icated | -0.055 |
textures | -0.050 |
Token | and |
Token ID | 290 |
Feature activation | +2.63e-01 |
Pos logit contributions | |
rolley | +0.068 |
microwave | +0.067 |
butcher | +0.067 |
frying | +0.064 |
waitress | +0.064 |
Neg logit contributions | |
ide | -0.033 |
winning | -0.030 |
icated | -0.028 |
seem | -0.028 |
textures | -0.026 |
Token | jumped |
Token ID | 11687 |
Feature activation | -9.43e-01 |
Pos logit contributions | |
winning | +0.006 |
ide | +0.006 |
seem | +0.006 |
shooting | +0.006 |
being | +0.006 |
Neg logit contributions | |
rolley | -0.009 |
butcher | -0.008 |
microwave | -0.008 |
piping | -0.008 |
waitress | -0.008 |
Token | on |
Token ID | 319 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
rolley | +0.033 |
piping | +0.031 |
butcher | +0.030 |
microwave | +0.029 |
frying | +0.029 |
Neg logit contributions | |
ide | -0.024 |
winning | -0.022 |
shooting | -0.021 |
happening | -0.021 |
seem | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | -3.65e-01 |
Pos logit contributions | |
rolley | +0.098 |
butcher | +0.095 |
microwave | +0.095 |
piping | +0.092 |
frying | +0.091 |
Neg logit contributions | |
ide | -0.055 |
winning | -0.050 |
seem | -0.048 |
icated | -0.047 |
shooting | -0.045 |
Token | basket |
Token ID | 7988 |
Feature activation | -5.05e+00 |
Pos logit contributions | |
rolley | +0.012 |
butcher | +0.011 |
microwave | +0.011 |
piping | +0.011 |
frying | +0.011 |
Neg logit contributions | |
ide | -0.009 |
winning | -0.009 |
seem | -0.008 |
icated | -0.008 |
shapes | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -3.25e-01 |
Pos logit contributions | |
rolley | +0.158 |
microwave | +0.154 |
butcher | +0.150 |
pan | +0.150 |
pouch | +0.150 |
Neg logit contributions | |
icated | -0.121 |
ide | -0.119 |
winning | -0.117 |
planets | -0.116 |
seem | -0.115 |
Token | It |
Token ID | 632 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
rolley | +0.014 |
frying | +0.013 |
use | +0.013 |
butcher | +0.013 |
P | +0.012 |
Neg logit contributions | |
winning | -0.005 |
seem | -0.005 |
not | -0.005 |
ide | -0.005 |
shooting | -0.005 |
Token | knocked |
Token ID | 13642 |
Feature activation | -4.76e+00 |
Pos logit contributions | |
ide | +0.037 |
winning | +0.035 |
seem | +0.034 |
icated | +0.032 |
shooting | +0.032 |
Neg logit contributions | |
rolley | -0.046 |
butcher | -0.043 |
microwave | -0.043 |
piping | -0.042 |
frying | -0.041 |
Token | over |
Token ID | 625 |
Feature activation | -3.95e+00 |
Pos logit contributions | |
rolley | +0.210 |
butcher | +0.204 |
frying | +0.202 |
piping | +0.196 |
waitress | +0.195 |
Neg logit contributions | |
ide | -0.072 |
sometimes | -0.067 |
winning | -0.066 |
being | -0.066 |
icated | -0.065 |
Token | goodbye |
Token ID | 24829 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
rolley | +0.012 |
butcher | +0.012 |
microwave | +0.012 |
piping | +0.011 |
frying | +0.011 |
Neg logit contributions | |
ide | -0.012 |
winning | -0.011 |
seem | -0.011 |
icated | -0.010 |
shooting | -0.010 |
Token | to |
Token ID | 284 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
rolley | +0.039 |
butcher | +0.037 |
microwave | +0.036 |
frying | +0.035 |
pouch | +0.035 |
Neg logit contributions | |
ide | -0.041 |
winning | -0.039 |
icated | -0.038 |
seem | -0.038 |
planets | -0.035 |
Token | their |
Token ID | 511 |
Feature activation | +7.57e-01 |
Pos logit contributions | |
ide | +0.003 |
winning | +0.002 |
seem | +0.002 |
icated | +0.002 |
shooting | +0.002 |
Neg logit contributions | |
rolley | -0.006 |
butcher | -0.005 |
microwave | -0.005 |
piping | -0.005 |
frying | -0.005 |
Token | pictures |
Token ID | 5986 |
Feature activation | +3.02e+00 |
Pos logit contributions | |
ide | +0.022 |
winning | +0.021 |
seem | +0.020 |
icated | +0.020 |
shooting | +0.020 |
Neg logit contributions | |
rolley | -0.022 |
butcher | -0.021 |
microwave | -0.020 |
piping | -0.020 |
frying | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
ide | +0.065 |
icated | +0.065 |
winning | +0.064 |
seem | +0.062 |
textures | +0.061 |
Neg logit contributions | |
rolley | -0.101 |
microwave | -0.100 |
pan | -0.096 |
carrier | -0.096 |
bag | -0.095 |
Token | hope |
Token ID | 2911 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
ide | +0.034 |
winning | +0.032 |
seem | +0.031 |
icated | +0.029 |
shooting | +0.029 |
Neg logit contributions | |
rolley | -0.051 |
butcher | -0.048 |
microwave | -0.048 |
piping | -0.047 |
frying | -0.046 |
Token | grandma |
Token ID | 49890 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
rolley | +0.003 |
butcher | +0.003 |
microwave | +0.003 |
frying | +0.003 |
waitress | +0.003 |
Neg logit contributions | |
ide | -0.003 |
icated | -0.003 |
winning | -0.002 |
seem | -0.002 |
textures | -0.002 |
Token | will |
Token ID | 481 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
rolley | +0.006 |
butcher | +0.005 |
microwave | +0.005 |
waitress | +0.005 |
piping | +0.005 |
Neg logit contributions | |
ide | -0.004 |
winning | -0.004 |
icated | -0.004 |
shooting | -0.003 |
seem | -0.003 |
Token | get |
Token ID | 651 |
Feature activation | -3.43e+00 |
Pos logit contributions | |
rolley | +0.009 |
butcher | +0.009 |
microwave | +0.009 |
piping | +0.009 |
pouch | +0.009 |
Neg logit contributions | |
ide | -0.006 |
icated | -0.005 |
winning | -0.005 |
textures | -0.005 |
shapes | -0.005 |
Token | them |
Token ID | 606 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
rolley | +0.130 |
butcher | +0.121 |
microwave | +0.119 |
P | +0.117 |
piping | +0.116 |
Neg logit contributions | |
being | -0.069 |
not | -0.067 |
winning | -0.066 |
when | -0.064 |
shooting | -0.063 |
Token | soon |
Token ID | 2582 |
Feature activation | -3.38e+00 |
Pos logit contributions | |
ide | +0.036 |
winning | +0.034 |
seem | +0.033 |
icated | +0.033 |
shooting | +0.032 |
Neg logit contributions | |
rolley | -0.025 |
butcher | -0.024 |
microwave | -0.024 |
piping | -0.022 |
frying | -0.022 |
Token | ? |
Token ID | 30 |
Feature activation | -2.23e-01 |
Pos logit contributions | |
ide | +0.103 |
winning | +0.094 |
icated | +0.091 |
textures | +0.091 |
seem | +0.091 |
Neg logit contributions | |
rolley | -0.071 |
microwave | -0.067 |
butcher | -0.067 |
etchup | -0.066 |
P | -0.065 |
Token |
|
Token ID | 198 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
rolley | +0.008 |
frying | +0.008 |
butcher | +0.007 |
microwave | +0.007 |
waitress | +0.007 |
Neg logit contributions | |
winning | -0.005 |
ide | -0.005 |
seem | -0.004 |
icated | -0.004 |
sharing | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | +6.84e-02 |
Pos logit contributions | |
ide | +0.005 |
winning | +0.005 |
seem | +0.005 |
icated | +0.004 |
shooting | +0.004 |
Neg logit contributions | |
rolley | -0.007 |
butcher | -0.006 |
microwave | -0.006 |
frying | -0.006 |
piping | -0.006 |
Token | S |
Token ID | 50 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
ide | +0.002 |
winning | +0.002 |
seem | +0.002 |
icated | +0.002 |
shooting | +0.002 |
Neg logit contributions | |
rolley | -0.002 |
microwave | -0.002 |
butcher | -0.002 |
piping | -0.002 |
frying | -0.002 |
Token | ara |
Token ID | 3301 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
rolley | +0.035 |
butcher | +0.032 |
piping | +0.030 |
microwave | +0.030 |
frying | +0.030 |
Neg logit contributions | |
ide | -0.067 |
winning | -0.062 |
seem | -0.061 |
icated | -0.060 |
sharing | -0.059 |
Token | and |
Token ID | 290 |
Feature activation | -7.66e-01 |
Pos logit contributions | |
rolley | +0.038 |
microwave | +0.037 |
butcher | +0.036 |
frying | +0.035 |
piping | +0.035 |
Neg logit contributions | |
ide | -0.033 |
winning | -0.031 |
seem | -0.030 |
icated | -0.030 |
textures | -0.028 |
Token | Ben |
Token ID | 3932 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
butcher | +0.024 |
rolley | +0.023 |
microwave | +0.023 |
frying | +0.022 |
waitress | +0.022 |
Neg logit contributions | |
ide | -0.021 |
winning | -0.019 |
icated | -0.018 |
seem | -0.018 |
shooting | -0.017 |
Token | see |
Token ID | 766 |
Feature activation | +4.56e-01 |
Pos logit contributions | |
ide | +0.006 |
winning | +0.006 |
icated | +0.006 |
textures | +0.005 |
sharing | +0.005 |
Neg logit contributions | |
use | -0.006 |
microwave | -0.006 |
butcher | -0.006 |
bag | -0.006 |
pouch | -0.006 |
Token | a |
Token ID | 257 |
Feature activation | -6.81e-01 |
Pos logit contributions | |
ide | +0.011 |
winning | +0.010 |
seem | +0.009 |
icated | +0.009 |
shooting | +0.009 |
Neg logit contributions | |
rolley | -0.016 |
butcher | -0.015 |
microwave | -0.015 |
piping | -0.015 |
frying | -0.015 |
Token | friendly |
Token ID | 8030 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
rolley | +0.022 |
butcher | +0.020 |
microwave | +0.020 |
piping | +0.019 |
frying | +0.019 |
Neg logit contributions | |
ide | -0.018 |
winning | -0.017 |
shooting | -0.016 |
seem | -0.016 |
shapes | -0.016 |
Token | dog |
Token ID | 3290 |
Feature activation | +9.26e-01 |
Pos logit contributions | |
ide | +0.036 |
winning | +0.033 |
icated | +0.033 |
seem | +0.032 |
planets | +0.031 |
Neg logit contributions | |
butcher | -0.028 |
rolley | -0.028 |
microwave | -0.027 |
waitress | -0.027 |
pan | -0.026 |
Token | moral |
Token ID | 6573 |
Feature activation | -3.34e+00 |
Pos logit contributions | |
ide | +0.023 |
winning | +0.021 |
seem | +0.021 |
icated | +0.020 |
shooting | +0.020 |
Neg logit contributions | |
rolley | -0.026 |
butcher | -0.025 |
microwave | -0.024 |
piping | -0.023 |
frying | -0.023 |
Token | of |
Token ID | 286 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
rolley | +0.124 |
piping | +0.122 |
microwave | +0.121 |
butcher | +0.120 |
frying | +0.120 |
Neg logit contributions | |
ide | -0.065 |
winning | -0.063 |
seem | -0.062 |
icated | -0.060 |
being | -0.059 |
Token | the |
Token ID | 262 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
ide | +0.003 |
seem | +0.003 |
winning | +0.003 |
textures | +0.003 |
sizes | +0.003 |
Neg logit contributions | |
butcher | -0.003 |
pan | -0.003 |
rolley | -0.003 |
microwave | -0.003 |
piping | -0.003 |
Token | story |
Token ID | 1621 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
butcher | +0.005 |
microwave | +0.005 |
waitress | +0.005 |
pan | +0.004 |
paramed | +0.004 |
Neg logit contributions | |
ide | -0.005 |
hip | -0.004 |
winning | -0.004 |
textures | -0.004 |
icated | -0.004 |
Token | is |
Token ID | 318 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
ide | +0.047 |
winning | +0.044 |
icated | +0.043 |
seem | +0.042 |
shooting | +0.042 |
Neg logit contributions | |
rolley | -0.045 |
butcher | -0.042 |
microwave | -0.042 |
pouch | -0.040 |
waitress | -0.039 |
Token | that |
Token ID | 326 |
Feature activation | -2.92e-01 |
Pos logit contributions | |
ide | +0.034 |
winning | +0.030 |
seem | +0.029 |
icated | +0.029 |
textures | +0.028 |
Neg logit contributions | |
rolley | -0.048 |
butcher | -0.046 |
microwave | -0.045 |
frying | -0.044 |
piping | -0.044 |
Token | even |
Token ID | 772 |
Feature activation | -2.76e+00 |
Pos logit contributions | |
butcher | +0.007 |
pan | +0.007 |
trouble | +0.007 |
bag | +0.007 |
pouch | +0.007 |
Neg logit contributions | |
ide | -0.009 |
hip | -0.007 |
bee | -0.007 |
igo | -0.007 |
wered | -0.007 |
Token | when |
Token ID | 618 |
Feature activation | +6.28e-01 |
Pos logit contributions | |
rolley | +0.091 |
butcher | +0.090 |
microwave | +0.088 |
frying | +0.086 |
etchup | +0.086 |
Neg logit contributions | |
ide | -0.067 |
winning | -0.058 |
seem | -0.054 |
hip | -0.054 |
icated | -0.053 |
Token | we |
Token ID | 356 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
ide | +0.017 |
icated | +0.014 |
hip | +0.014 |
winning | +0.013 |
seem | +0.013 |
Neg logit contributions | |
butcher | -0.019 |
rolley | -0.019 |
frying | -0.018 |
microwave | -0.018 |
etchup | -0.018 |
Token | are |
Token ID | 389 |
Feature activation | +5.33e-01 |
Pos logit contributions | |
use | +0.051 |
butcher | +0.050 |
rolley | +0.050 |
pouch | +0.049 |
microwave | +0.048 |
Neg logit contributions | |
ide | -0.047 |
winning | -0.043 |
icated | -0.041 |
patterns | -0.038 |
shooting | -0.038 |
Token | tired |
Token ID | 10032 |
Feature activation | -8.11e-01 |
Pos logit contributions | |
ide | +0.014 |
winning | +0.012 |
icated | +0.012 |
seem | +0.012 |
textures | +0.011 |
Neg logit contributions | |
rolley | -0.017 |
butcher | -0.017 |
microwave | -0.016 |
frying | -0.016 |
piping | -0.016 |
Token | again |
Token ID | 757 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
rolley | +0.009 |
butcher | +0.009 |
microwave | +0.008 |
pan | +0.008 |
frying | +0.008 |
Neg logit contributions | |
ide | -0.006 |
icated | -0.006 |
winning | -0.006 |
seem | -0.005 |
textures | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
rolley | +0.079 |
butcher | +0.076 |
microwave | +0.076 |
piping | +0.074 |
frying | +0.073 |
Neg logit contributions | |
ide | -0.043 |
winning | -0.039 |
seem | -0.037 |
icated | -0.036 |
shooting | -0.035 |
Token | the |
Token ID | 262 |
Feature activation | -6.21e-01 |
Pos logit contributions | |
ide | +0.036 |
icated | +0.029 |
seem | +0.026 |
textures | +0.026 |
hip | +0.025 |
Neg logit contributions | |
microwave | -0.051 |
pan | -0.051 |
butcher | -0.051 |
pouch | -0.051 |
rolley | -0.051 |
Token | little |
Token ID | 1310 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
pan | +0.017 |
rolley | +0.016 |
microwave | +0.016 |
pouch | +0.016 |
bag | +0.016 |
Neg logit contributions | |
ide | -0.020 |
seem | -0.017 |
icated | -0.017 |
winning | -0.017 |
hip | -0.017 |
Token | boat |
Token ID | 8848 |
Feature activation | -3.60e-01 |
Pos logit contributions | |
rolley | +0.028 |
pan | +0.027 |
pouch | +0.026 |
microwave | +0.026 |
bag | +0.026 |
Neg logit contributions | |
ide | -0.042 |
winning | -0.039 |
icated | -0.037 |
seem | -0.037 |
hip | -0.036 |
Token | went |
Token ID | 1816 |
Feature activation | -4.18e+00 |
Pos logit contributions | |
rolley | +0.011 |
microwave | +0.011 |
butcher | +0.011 |
pan | +0.010 |
pouch | +0.010 |
Neg logit contributions | |
ide | -0.009 |
icated | -0.009 |
winning | -0.009 |
seem | -0.008 |
textures | -0.008 |
Token | back |
Token ID | 736 |
Feature activation | -4.63e+00 |
Pos logit contributions | |
rolley | +0.111 |
butcher | +0.106 |
microwave | +0.104 |
piping | +0.100 |
frying | +0.100 |
Neg logit contributions | |
ide | -0.132 |
winning | -0.126 |
seem | -0.121 |
icated | -0.118 |
shooting | -0.118 |
Token | out |
Token ID | 503 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
rolley | +0.143 |
butcher | +0.138 |
microwave | +0.138 |
piping | +0.135 |
waitress | +0.133 |
Neg logit contributions | |
ide | -0.125 |
winning | -0.122 |
shooting | -0.114 |
seem | -0.113 |
being | -0.111 |
Token | onto |
Token ID | 4291 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
rolley | +0.094 |
butcher | +0.091 |
microwave | +0.090 |
piping | +0.087 |
frying | +0.086 |
Neg logit contributions | |
ide | -0.060 |
winning | -0.056 |
seem | -0.053 |
icated | -0.051 |
shooting | -0.051 |
Token | the |
Token ID | 262 |
Feature activation | +6.43e-02 |
Pos logit contributions | |
rolley | +0.068 |
butcher | +0.065 |
microwave | +0.064 |
piping | +0.063 |
frying | +0.063 |
Neg logit contributions | |
ide | -0.028 |
winning | -0.026 |
shooting | -0.024 |
seem | -0.024 |
icated | -0.023 |
Token | sea |
Token ID | 5417 |
Feature activation | +2.80e+00 |
Pos logit contributions | |
ide | +0.002 |
winning | +0.002 |
seem | +0.002 |
icated | +0.002 |
sharing | +0.002 |
Neg logit contributions | |
rolley | -0.002 |
microwave | -0.002 |
butcher | -0.002 |
pan | -0.002 |
pouch | -0.002 |
Token | are |
Token ID | 389 |
Feature activation | +7.22e-01 |
Pos logit contributions | |
rolley | +0.070 |
butcher | +0.068 |
microwave | +0.068 |
piping | +0.066 |
frying | +0.065 |
Neg logit contributions | |
ide | -0.039 |
winning | -0.035 |
icated | -0.033 |
seem | -0.032 |
shooting | -0.031 |
Token | sorry |
Token ID | 7926 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
ide | +0.019 |
seem | +0.016 |
icated | +0.016 |
winning | +0.016 |
textures | +0.015 |
Neg logit contributions | |
butcher | -0.023 |
microwave | -0.023 |
rolley | -0.023 |
waitress | -0.022 |
frying | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -3.69e-01 |
Pos logit contributions | |
microwave | +0.041 |
rolley | +0.040 |
butcher | +0.040 |
waitress | +0.039 |
frying | +0.038 |
Neg logit contributions | |
ide | -0.040 |
winning | -0.038 |
planets | -0.036 |
shooting | -0.035 |
wered | -0.035 |
Token |
|
Token ID | 198 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
rolley | +0.014 |
butcher | +0.013 |
microwave | +0.013 |
frying | +0.013 |
waitress | +0.013 |
Neg logit contributions | |
ide | -0.007 |
winning | -0.007 |
seem | -0.007 |
sharing | -0.007 |
icated | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
rolley | +0.007 |
butcher | +0.007 |
microwave | +0.007 |
frying | +0.006 |
piping | +0.006 |
Neg logit contributions | |
ide | -0.005 |
winning | -0.005 |
seem | -0.004 |
icated | -0.004 |
shooting | -0.004 |
Token | Mom |
Token ID | 29252 |
Feature activation | -4.88e+00 |
Pos logit contributions | |
rolley | +0.010 |
microwave | +0.009 |
butcher | +0.009 |
frying | +0.009 |
piping | +0.009 |
Neg logit contributions | |
ide | -0.007 |
winning | -0.007 |
icated | -0.006 |
seem | -0.006 |
shooting | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
rolley | +0.141 |
butcher | +0.134 |
microwave | +0.128 |
piping | +0.126 |
frying | +0.126 |
Neg logit contributions | |
ide | -0.143 |
winning | -0.134 |
seem | -0.130 |
icated | -0.127 |
shooting | -0.126 |
Token | Dad |
Token ID | 17415 |
Feature activation | -3.63e+00 |
Pos logit contributions | |
rolley | +0.047 |
butcher | +0.042 |
microwave | +0.042 |
piping | +0.041 |
pouch | +0.041 |
Neg logit contributions | |
ide | -0.041 |
winning | -0.040 |
seem | -0.039 |
being | -0.039 |
sharing | -0.038 |
Token | hug |
Token ID | 16225 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
rolley | +0.101 |
microwave | +0.100 |
butcher | +0.098 |
frying | +0.096 |
piping | +0.095 |
Neg logit contributions | |
ide | -0.107 |
winning | -0.100 |
icated | -0.097 |
planets | -0.094 |
textures | -0.094 |
Token | them |
Token ID | 606 |
Feature activation | +4.99e-01 |
Pos logit contributions | |
rolley | +0.042 |
butcher | +0.041 |
microwave | +0.041 |
waitress | +0.040 |
piping | +0.040 |
Neg logit contributions | |
ide | -0.032 |
winning | -0.031 |
seem | -0.029 |
icated | -0.029 |
shooting | -0.028 |
Token | and |
Token ID | 290 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
ide | +0.012 |
seem | +0.012 |
winning | +0.012 |
icated | +0.011 |
planets | +0.011 |
Neg logit contributions | |
rolley | -0.016 |
microwave | -0.016 |
butcher | -0.015 |
pan | -0.015 |
waitress | -0.015 |
Token | breakfast |
Token ID | 12607 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
rolley | +0.056 |
butcher | +0.052 |
microwave | +0.052 |
pouch | +0.050 |
use | +0.049 |
Neg logit contributions | |
ide | -0.042 |
winning | -0.040 |
being | -0.040 |
seem | -0.039 |
sharing | -0.038 |
Token | . |
Token ID | 13 |
Feature activation | +4.56e-01 |
Pos logit contributions | |
rolley | +0.039 |
butcher | +0.038 |
microwave | +0.038 |
piping | +0.037 |
frying | +0.036 |
Neg logit contributions | |
ide | -0.020 |
winning | -0.018 |
seem | -0.017 |
icated | -0.016 |
sharing | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | +5.89e-01 |
Pos logit contributions | |
ide | +0.013 |
winning | +0.013 |
seem | +0.012 |
icated | +0.012 |
shooting | +0.012 |
Neg logit contributions | |
rolley | -0.013 |
butcher | -0.013 |
microwave | -0.013 |
frying | -0.012 |
piping | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | +4.92e-01 |
Pos logit contributions | |
ide | +0.014 |
winning | +0.012 |
seem | +0.012 |
shooting | +0.011 |
icated | +0.011 |
Neg logit contributions | |
rolley | -0.021 |
microwave | -0.021 |
butcher | -0.021 |
frying | -0.020 |
piping | -0.020 |
Token | " |
Token ID | 1 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
ide | +0.014 |
winning | +0.014 |
icated | +0.013 |
seem | +0.013 |
shooting | +0.013 |
Neg logit contributions | |
rolley | -0.014 |
butcher | -0.013 |
P | -0.013 |
microwave | -0.013 |
piping | -0.012 |
Token | Mom |
Token ID | 29252 |
Feature activation | -5.07e+00 |
Pos logit contributions | |
rolley | +0.057 |
microwave | +0.053 |
butcher | +0.052 |
piping | +0.051 |
pouch | +0.050 |
Neg logit contributions | |
ide | -0.067 |
winning | -0.062 |
seem | -0.060 |
icated | -0.059 |
textures | -0.058 |
Token | , |
Token ID | 11 |
Feature activation | -8.19e-01 |
Pos logit contributions | |
rolley | +0.202 |
butcher | +0.194 |
microwave | +0.191 |
piping | +0.186 |
frying | +0.186 |
Neg logit contributions | |
ide | -0.095 |
winning | -0.086 |
seem | -0.080 |
icated | -0.079 |
sharing | -0.078 |
Token | I |
Token ID | 314 |
Feature activation | -3.08e+00 |
Pos logit contributions | |
rolley | +0.031 |
butcher | +0.029 |
microwave | +0.029 |
P | +0.028 |
pouch | +0.028 |
Neg logit contributions | |
ide | -0.017 |
seem | -0.015 |
winning | -0.015 |
icated | -0.015 |
sharing | -0.015 |
Token | want |
Token ID | 765 |
Feature activation | +2.83e-01 |
Pos logit contributions | |
rolley | +0.133 |
butcher | +0.122 |
microwave | +0.121 |
piping | +0.121 |
waitress | +0.119 |
Neg logit contributions | |
seem | -0.052 |
don | -0.044 |
ide | -0.043 |
not | -0.043 |
always | -0.043 |
Token | to |
Token ID | 284 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
being | +0.006 |
seem | +0.006 |
icated | +0.006 |
not | +0.006 |
ide | +0.006 |
Neg logit contributions | |
rolley | -0.010 |
pouch | -0.010 |
aked | -0.009 |
carrier | -0.009 |
enn | -0.009 |
Token | eat |
Token ID | 4483 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
ide | +0.028 |
seem | +0.027 |
winning | +0.024 |
not | +0.024 |
sharing | +0.023 |
Neg logit contributions | |
rolley | -0.059 |
butcher | -0.054 |
microwave | -0.054 |
enn | -0.054 |
etchup | -0.053 |
Token | sleep |
Token ID | 3993 |
Feature activation | +4.18e+00 |
Pos logit contributions | |
ide | +0.009 |
winning | +0.009 |
seem | +0.008 |
icated | +0.008 |
shooting | +0.008 |
Neg logit contributions | |
rolley | -0.015 |
butcher | -0.015 |
microwave | -0.015 |
piping | -0.014 |
frying | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
ide | +0.076 |
winning | +0.069 |
seem | +0.062 |
sharing | +0.062 |
being | +0.062 |
Neg logit contributions | |
rolley | -0.170 |
butcher | -0.162 |
piping | -0.160 |
microwave | -0.160 |
frying | -0.156 |
Token | He |
Token ID | 679 |
Feature activation | -5.35e-01 |
Pos logit contributions | |
rolley | +0.005 |
butcher | +0.005 |
frying | +0.005 |
microwave | +0.005 |
piping | +0.005 |
Neg logit contributions | |
ide | -0.003 |
winning | -0.003 |
seem | -0.003 |
icated | -0.003 |
shooting | -0.003 |
Token | hopes |
Token ID | 7176 |
Feature activation | -7.47e-01 |
Pos logit contributions | |
rolley | +0.014 |
microwave | +0.014 |
butcher | +0.014 |
frying | +0.013 |
piping | +0.013 |
Neg logit contributions | |
ide | -0.017 |
winning | -0.015 |
icated | -0.015 |
seem | -0.015 |
shooting | -0.014 |
Token | his |
Token ID | 465 |
Feature activation | -1.68e-02 |
Pos logit contributions | |
rolley | +0.026 |
butcher | +0.024 |
microwave | +0.024 |
piping | +0.024 |
frying | +0.023 |
Neg logit contributions | |
ide | -0.018 |
winning | -0.017 |
seem | -0.016 |
icated | -0.016 |
shooting | -0.015 |
Token | tum |
Token ID | 11814 |
Feature activation | -4.68e+00 |
Pos logit contributions | |
rolley | +0.001 |
butcher | +0.000 |
microwave | +0.000 |
piping | +0.000 |
frying | +0.000 |
Neg logit contributions | |
ide | -0.000 |
winning | -0.000 |
seem | -0.000 |
icated | -0.000 |
shooting | -0.000 |
Token | my |
Token ID | 1820 |
Feature activation | -5.71e-02 |
Pos logit contributions | |
rolley | +0.133 |
microwave | +0.131 |
etchup | +0.129 |
piping | +0.120 |
butcher | +0.119 |
Neg logit contributions | |
winning | -0.132 |
ide | -0.126 |
shooting | -0.123 |
seem | -0.122 |
tunes | -0.120 |
Token | will |
Token ID | 481 |
Feature activation | -3.32e-01 |
Pos logit contributions | |
rolley | +0.002 |
butcher | +0.002 |
microwave | +0.002 |
piping | +0.002 |
frying | +0.002 |
Neg logit contributions | |
ide | -0.001 |
winning | -0.001 |
icated | -0.001 |
seem | -0.001 |
shooting | -0.001 |
Token | feel |
Token ID | 1254 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
rolley | +0.013 |
butcher | +0.013 |
microwave | +0.013 |
waitress | +0.012 |
piping | +0.012 |
Neg logit contributions | |
ide | -0.006 |
winning | -0.005 |
icated | -0.005 |
planets | -0.005 |
shapes | -0.005 |
Token | better |
Token ID | 1365 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
ide | +0.035 |
winning | +0.033 |
seem | +0.032 |
icated | +0.031 |
sharing | +0.031 |
Neg logit contributions | |
rolley | -0.036 |
butcher | -0.034 |
microwave | -0.033 |
piping | -0.032 |
frying | -0.032 |
Token | soon |
Token ID | 2582 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
rolley | +0.066 |
butcher | +0.064 |
microwave | +0.063 |
frying | +0.061 |
piping | +0.060 |
Neg logit contributions | |
ide | -0.042 |
winning | -0.038 |
icated | -0.036 |
seem | -0.036 |
shooting | -0.035 |
Token | play |
Token ID | 711 |
Feature activation | +5.57e+00 |
Pos logit contributions | |
ide | +0.020 |
winning | +0.019 |
seem | +0.018 |
icated | +0.017 |
shooting | +0.017 |
Neg logit contributions | |
rolley | -0.033 |
butcher | -0.032 |
microwave | -0.031 |
piping | -0.030 |
frying | -0.030 |
Token | peek |
Token ID | 27185 |
Feature activation | +4.64e-01 |
Pos logit contributions | |
ide | +0.107 |
winning | +0.097 |
seem | +0.093 |
icated | +0.091 |
shooting | +0.085 |
Neg logit contributions | |
rolley | -0.214 |
microwave | -0.207 |
butcher | -0.207 |
frying | -0.200 |
piping | -0.200 |
Token | - |
Token ID | 12 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
ide | +0.011 |
winning | +0.010 |
shapes | +0.010 |
seem | +0.010 |
songs | +0.009 |
Neg logit contributions | |
butcher | -0.016 |
rolley | -0.016 |
microwave | -0.016 |
piping | -0.015 |
etchup | -0.015 |
Token | a |
Token ID | 64 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
rolley | +0.039 |
butcher | +0.039 |
microwave | +0.037 |
frying | +0.035 |
piping | +0.035 |
Neg logit contributions | |
ide | -0.069 |
winning | -0.063 |
seem | -0.063 |
icated | -0.063 |
shooting | -0.060 |
Token | - |
Token ID | 12 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
rolley | +0.060 |
microwave | +0.058 |
butcher | +0.057 |
frying | +0.056 |
piping | +0.056 |
Neg logit contributions | |
ide | -0.052 |
winning | -0.050 |
seem | -0.048 |
icated | -0.046 |
shooting | -0.044 |
Token | bo |
Token ID | 2127 |
Feature activation | -4.76e+00 |
Pos logit contributions | |
rolley | +0.048 |
butcher | +0.048 |
microwave | +0.046 |
frying | +0.044 |
pouch | +0.043 |
Neg logit contributions | |
ide | -0.073 |
winning | -0.067 |
icated | -0.067 |
seem | -0.066 |
shooting | -0.064 |
Token | o |
Token ID | 78 |
Feature activation | +2.20e-02 |
Pos logit contributions | |
rolley | +0.122 |
butcher | +0.116 |
microwave | +0.114 |
piping | +0.110 |
frying | +0.109 |
Neg logit contributions | |
ide | -0.155 |
winning | -0.146 |
seem | -0.141 |
icated | -0.139 |
shooting | -0.136 |
Token | . |
Token ID | 13 |
Feature activation | +4.87e-01 |
Pos logit contributions | |
ide | +0.001 |
winning | +0.001 |
seem | +0.001 |
icated | +0.001 |
textures | +0.001 |
Neg logit contributions | |
rolley | -0.001 |
microwave | -0.001 |
butcher | -0.001 |
pan | -0.001 |
frying | -0.001 |
Token | Every |
Token ID | 3887 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
ide | +0.013 |
winning | +0.012 |
seem | +0.012 |
icated | +0.011 |
shooting | +0.011 |
Neg logit contributions | |
rolley | -0.015 |
butcher | -0.015 |
microwave | -0.014 |
piping | -0.014 |
frying | -0.014 |
Token | time |
Token ID | 640 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
ide | +0.004 |
winning | +0.003 |
seem | +0.003 |
icated | +0.003 |
shooting | +0.003 |
Neg logit contributions | |
rolley | -0.006 |
butcher | -0.006 |
microwave | -0.006 |
piping | -0.006 |
frying | -0.006 |
Token | she |
Token ID | 673 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
ide | +0.029 |
winning | +0.026 |
icated | +0.025 |
hip | +0.025 |
seem | +0.024 |
Neg logit contributions | |
rolley | -0.040 |
butcher | -0.039 |
microwave | -0.039 |
piping | -0.038 |
frying | -0.038 |
Token | Sam |
Token ID | 3409 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
ide | +0.005 |
winning | +0.004 |
seem | +0.004 |
icated | +0.004 |
textures | +0.004 |
Neg logit contributions | |
rolley | -0.007 |
microwave | -0.007 |
butcher | -0.007 |
pan | -0.007 |
piping | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
planets | +0.049 |
ide | +0.049 |
textures | +0.049 |
icated | +0.048 |
winning | +0.048 |
Neg logit contributions | |
bag | -0.065 |
pouch | -0.064 |
pan | -0.064 |
microwave | -0.063 |
butcher | -0.063 |
Token | wanted |
Token ID | 2227 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
rolley | +0.007 |
butcher | +0.007 |
piping | +0.007 |
microwave | +0.006 |
frying | +0.006 |
Neg logit contributions | |
being | -0.005 |
shooting | -0.005 |
not | -0.005 |
winning | -0.004 |
seem | -0.004 |
Token | his |
Token ID | 465 |
Feature activation | +5.39e-01 |
Pos logit contributions | |
rolley | +0.047 |
butcher | +0.047 |
microwave | +0.047 |
frying | +0.044 |
piping | +0.044 |
Neg logit contributions | |
ide | -0.050 |
winning | -0.045 |
seem | -0.044 |
icated | -0.043 |
planets | -0.042 |
Token | money |
Token ID | 1637 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
ide | +0.017 |
winning | +0.015 |
icated | +0.015 |
seem | +0.015 |
sharing | +0.014 |
Neg logit contributions | |
rolley | -0.014 |
butcher | -0.014 |
microwave | -0.014 |
pouch | -0.014 |
bag | -0.014 |
Token | back |
Token ID | 736 |
Feature activation | -4.84e+00 |
Pos logit contributions | |
rolley | +0.069 |
butcher | +0.067 |
microwave | +0.066 |
pouch | +0.063 |
frying | +0.063 |
Neg logit contributions | |
ide | -0.055 |
winning | -0.050 |
icated | -0.049 |
seem | -0.048 |
planets | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
rolley | +0.165 |
butcher | +0.157 |
microwave | +0.155 |
pouch | +0.152 |
piping | +0.150 |
Neg logit contributions | |
ide | -0.110 |
winning | -0.102 |
seem | -0.101 |
icated | -0.100 |
planets | -0.094 |
Token |
|
Token ID | 198 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
rolley | +0.001 |
butcher | +0.001 |
frying | +0.001 |
piping | +0.001 |
microwave | +0.001 |
Neg logit contributions | |
ide | -0.000 |
winning | -0.000 |
seem | -0.000 |
shooting | -0.000 |
icated | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | +7.15e-02 |
Pos logit contributions | |
ide | +0.005 |
winning | +0.005 |
seem | +0.005 |
icated | +0.004 |
shooting | +0.004 |
Neg logit contributions | |
rolley | -0.007 |
butcher | -0.007 |
microwave | -0.007 |
frying | -0.007 |
piping | -0.007 |
Token | Sam |
Token ID | 16305 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
ide | +0.002 |
winning | +0.002 |
seem | +0.002 |
icated | +0.002 |
shooting | +0.001 |
Neg logit contributions | |
rolley | -0.002 |
butcher | -0.002 |
microwave | -0.002 |
piping | -0.002 |
frying | -0.002 |
Token | felt |
Token ID | 2936 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
ide | +0.029 |
winning | +0.027 |
seem | +0.026 |
shooting | +0.025 |
icated | +0.025 |
Neg logit contributions | |
rolley | -0.038 |
butcher | -0.036 |
microwave | -0.035 |
piping | -0.035 |
frying | -0.034 |