Token | feels |
Token ID | 5300 |
Feature activation | +3.04e-02 |
Pos logit contributions | |
arl | +0.037 |
iren | +0.034 |
its | +0.029 |
izes | +0.028 |
irens | +0.028 |
Neg logit contributions | |
favourite | -0.038 |
companion | -0.036 |
found | -0.035 |
favorite | -0.035 |
mate | -0.034 |
Token | even |
Token ID | 772 |
Feature activation | -1.23e-03 |
Pos logit contributions | |
arl | +0.071 |
iren | +0.065 |
its | +0.058 |
irens | +0.057 |
izes | +0.056 |
Neg logit contributions | |
favourite | -0.042 |
found | -0.039 |
favorite | -0.038 |
one | -0.037 |
companion | -0.037 |
Token | happier |
Token ID | 23030 |
Feature activation | +9.36e-03 |
Pos logit contributions | |
favourite | +0.002 |
found | +0.002 |
companion | +0.002 |
favorite | +0.002 |
one | +0.002 |
Neg logit contributions | |
arl | -0.002 |
iren | -0.002 |
its | -0.002 |
irens | -0.002 |
yourselves | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | +4.88e-02 |
Pos logit contributions | |
arl | +0.022 |
iren | +0.021 |
its | +0.019 |
irens | +0.018 |
isin | +0.018 |
Neg logit contributions | |
favourite | -0.012 |
one | -0.011 |
companion | -0.011 |
found | -0.011 |
favorite | -0.011 |
Token | p |
Token ID | 279 |
Feature activation | +1.80e-02 |
Pos logit contributions | |
arl | +0.100 |
iren | +0.092 |
its | +0.082 |
izes | +0.080 |
irens | +0.078 |
Neg logit contributions | |
favourite | -0.076 |
companion | -0.071 |
favorite | -0.069 |
found | -0.068 |
mate | -0.068 |
Token | rou |
Token ID | 472 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
arl | +0.026 |
iren | +0.025 |
isin | +0.022 |
yourselves | +0.022 |
its | +0.021 |
Neg logit contributions | |
favourite | -0.036 |
favorite | -0.034 |
found | -0.033 |
companion | -0.032 |
mate | -0.032 |
Token | der |
Token ID | 1082 |
Feature activation | +6.18e-02 |
Pos logit contributions | |
arl | +0.364 |
iren | +0.346 |
irens | +0.293 |
eling | +0.283 |
izes | +0.274 |
Neg logit contributions | |
favourite | -0.320 |
favorite | -0.304 |
one | -0.296 |
Joey | -0.274 |
Dave | -0.271 |
Token | . |
Token ID | 13 |
Feature activation | +5.26e-03 |
Pos logit contributions | |
arl | +0.150 |
iren | +0.139 |
irens | +0.124 |
its | +0.123 |
bleed | +0.121 |
Neg logit contributions | |
one | -0.069 |
favourite | -0.068 |
found | -0.066 |
favorite | -0.063 |
spotted | -0.058 |
Token | He |
Token ID | 679 |
Feature activation | +5.81e-02 |
Pos logit contributions | |
arl | +0.012 |
iren | +0.011 |
its | +0.010 |
izes | +0.010 |
yourselves | +0.010 |
Neg logit contributions | |
favourite | -0.007 |
companion | -0.006 |
favorite | -0.006 |
found | -0.006 |
mate | -0.006 |
Token | displays |
Token ID | 11298 |
Feature activation | -7.54e-02 |
Pos logit contributions | |
arl | +0.125 |
iren | +0.112 |
irens | +0.098 |
isin | +0.094 |
yourselves | +0.094 |
Neg logit contributions | |
favourite | -0.094 |
found | -0.094 |
favorite | -0.088 |
one | -0.085 |
searched | -0.085 |
Token | his |
Token ID | 465 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
favourite | +0.088 |
found | +0.081 |
one | +0.080 |
favorite | +0.079 |
companion | +0.079 |
Neg logit contributions | |
arl | -0.186 |
iren | -0.173 |
its | -0.157 |
isin | -0.155 |
irens | -0.153 |
Token | 'm |
Token ID | 1101 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
favourite | +0.015 |
companion | +0.014 |
favorite | +0.013 |
mate | +0.013 |
3 | +0.013 |
Neg logit contributions | |
arl | -0.028 |
iren | -0.026 |
its | -0.023 |
izes | -0.023 |
irens | -0.022 |
Token | sorry |
Token ID | 7926 |
Feature activation | +2.04e-02 |
Pos logit contributions | |
favourite | +0.044 |
found | +0.042 |
Sally | +0.042 |
3 | +0.042 |
one | +0.041 |
Neg logit contributions | |
arl | -0.034 |
iren | -0.031 |
its | -0.028 |
izes | -0.026 |
irens | -0.025 |
Token | , |
Token ID | 11 |
Feature activation | +2.19e-02 |
Pos logit contributions | |
arl | +0.052 |
iren | +0.051 |
its | +0.046 |
izes | +0.045 |
irens | +0.045 |
Neg logit contributions | |
one | -0.021 |
favourite | -0.020 |
Sally | -0.019 |
favorite | -0.019 |
mate | -0.018 |
Token | my |
Token ID | 616 |
Feature activation | +3.00e-03 |
Pos logit contributions | |
arl | +0.051 |
iren | +0.049 |
its | +0.044 |
izes | +0.043 |
irens | +0.043 |
Neg logit contributions | |
favourite | -0.027 |
companion | -0.024 |
favorite | -0.024 |
mate | -0.024 |
found | -0.024 |
Token | d |
Token ID | 288 |
Feature activation | +2.71e-02 |
Pos logit contributions | |
arl | +0.006 |
iren | +0.006 |
izes | +0.005 |
irens | +0.005 |
its | +0.005 |
Neg logit contributions | |
favourite | -0.005 |
favorite | -0.005 |
companion | -0.005 |
3 | -0.005 |
found | -0.004 |
Token | arl |
Token ID | 7063 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
izes | +0.011 |
eling | +0.009 |
isin | +0.007 |
iren | +0.006 |
heit | +0.006 |
Neg logit contributions | |
mate | -0.075 |
companion | -0.071 |
favourite | -0.071 |
Sally | -0.070 |
favorite | -0.069 |
Token | ings |
Token ID | 654 |
Feature activation | +3.89e-02 |
Pos logit contributions | |
eling | +0.304 |
iren | +0.284 |
arl | +0.283 |
isin | +0.274 |
rou | +0.267 |
Neg logit contributions | |
companion | -0.306 |
one | -0.293 |
favourite | -0.289 |
favorite | -0.279 |
mate | -0.276 |
Token | . |
Token ID | 13 |
Feature activation | +3.97e-02 |
Pos logit contributions | |
arl | +0.096 |
iren | +0.091 |
eling | +0.088 |
its | +0.088 |
izes | +0.086 |
Neg logit contributions | |
one | -0.034 |
Sally | -0.033 |
found | -0.032 |
companion | -0.028 |
Jack | -0.028 |
Token | That |
Token ID | 1320 |
Feature activation | -3.35e-02 |
Pos logit contributions | |
arl | +0.083 |
iren | +0.078 |
its | +0.071 |
izes | +0.069 |
irens | +0.068 |
Neg logit contributions | |
favourite | -0.059 |
companion | -0.055 |
favorite | -0.054 |
Sally | -0.053 |
found | -0.052 |
Token | was |
Token ID | 373 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
favourite | +0.036 |
companion | +0.034 |
mate | +0.033 |
favorite | +0.030 |
found | +0.030 |
Neg logit contributions | |
iren | -0.079 |
arl | -0.078 |
its | -0.071 |
izes | -0.070 |
irens | -0.068 |
Token | a |
Token ID | 257 |
Feature activation | +2.21e-02 |
Pos logit contributions | |
arl | +0.051 |
iren | +0.048 |
its | +0.043 |
izes | +0.042 |
irens | +0.041 |
Neg logit contributions | |
favourite | -0.028 |
companion | -0.026 |
favorite | -0.025 |
found | -0.025 |
3 | -0.024 |
Token | 's |
Token ID | 338 |
Feature activation | -6.12e-02 |
Pos logit contributions | |
favourite | +0.033 |
companion | +0.030 |
favorite | +0.029 |
found | +0.029 |
mate | +0.028 |
Neg logit contributions | |
arl | -0.067 |
iren | -0.063 |
its | -0.057 |
izes | -0.056 |
irens | -0.055 |
Token | okay |
Token ID | 8788 |
Feature activation | -2.91e-03 |
Pos logit contributions | |
favourite | +0.065 |
favorite | +0.060 |
one | +0.060 |
Sally | +0.057 |
companion | +0.057 |
Neg logit contributions | |
arl | -0.157 |
iren | -0.156 |
its | -0.139 |
irens | -0.137 |
izes | -0.133 |
Token | , |
Token ID | 11 |
Feature activation | +2.27e-02 |
Pos logit contributions | |
favourite | +0.004 |
companion | +0.004 |
favorite | +0.004 |
found | +0.004 |
mate | +0.004 |
Neg logit contributions | |
arl | -0.007 |
iren | -0.006 |
its | -0.005 |
izes | -0.005 |
irens | -0.005 |
Token | my |
Token ID | 616 |
Feature activation | -5.82e-03 |
Pos logit contributions | |
arl | +0.050 |
iren | +0.046 |
its | +0.041 |
yourselves | +0.041 |
izes | +0.040 |
Neg logit contributions | |
favourite | -0.032 |
companion | -0.029 |
found | -0.028 |
favorite | -0.028 |
spotted | -0.027 |
Token | d |
Token ID | 288 |
Feature activation | +2.94e-02 |
Pos logit contributions | |
favourite | +0.009 |
favorite | +0.008 |
companion | +0.008 |
3 | +0.008 |
one | +0.007 |
Neg logit contributions | |
arl | -0.013 |
iren | -0.013 |
irens | -0.011 |
izes | -0.011 |
its | -0.011 |
Token | arl |
Token ID | 7063 |
Feature activation | +1.88e-01 |
Pos logit contributions | |
eling | +0.012 |
izes | +0.012 |
isin | +0.008 |
iren | +0.007 |
yourselves | +0.005 |
Neg logit contributions | |
mate | -0.084 |
companion | -0.080 |
favourite | -0.079 |
favorite | -0.078 |
3 | -0.077 |
Token | ings |
Token ID | 654 |
Feature activation | +4.35e-02 |
Pos logit contributions | |
eling | +0.322 |
isin | +0.293 |
iren | +0.290 |
rou | +0.283 |
arl | +0.283 |
Neg logit contributions | |
companion | -0.306 |
one | -0.289 |
favourite | -0.275 |
friend | -0.273 |
favorite | -0.270 |
Token | . |
Token ID | 13 |
Feature activation | +4.66e-02 |
Pos logit contributions | |
arl | +0.108 |
eling | +0.105 |
its | +0.104 |
iren | +0.104 |
izes | +0.102 |
Neg logit contributions | |
one | -0.037 |
found | -0.034 |
Sally | -0.033 |
Jack | -0.032 |
mate | -0.030 |
Token | I |
Token ID | 314 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
arl | +0.090 |
iren | +0.082 |
its | +0.072 |
izes | +0.070 |
yourselves | +0.069 |
Neg logit contributions | |
favourite | -0.078 |
companion | -0.073 |
found | -0.071 |
favorite | -0.071 |
mate | -0.069 |
Token | forgive |
Token ID | 20927 |
Feature activation | +3.52e-02 |
Pos logit contributions | |
favourite | +0.028 |
companion | +0.027 |
favorite | +0.026 |
3 | +0.026 |
mate | +0.026 |
Neg logit contributions | |
arl | -0.028 |
iren | -0.026 |
its | -0.023 |
izes | -0.022 |
yourselves | -0.021 |
Token | you |
Token ID | 345 |
Feature activation | -6.47e-03 |
Pos logit contributions | |
arl | +0.051 |
iren | +0.045 |
its | +0.038 |
izes | +0.036 |
irens | +0.035 |
Neg logit contributions | |
favourite | -0.076 |
one | -0.073 |
companion | -0.072 |
favorite | -0.072 |
found | -0.072 |
Token | you |
Token ID | 345 |
Feature activation | -3.79e-02 |
Pos logit contributions | |
favourite | +0.049 |
companion | +0.047 |
favorite | +0.044 |
mate | +0.043 |
favourites | +0.043 |
Neg logit contributions | |
arl | -0.044 |
iren | -0.042 |
yourselves | -0.038 |
izes | -0.037 |
falls | -0.036 |
Token | go |
Token ID | 467 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
favourite | +0.058 |
companion | +0.054 |
found | +0.054 |
favorite | +0.053 |
mate | +0.052 |
Neg logit contributions | |
arl | -0.079 |
iren | -0.074 |
its | -0.065 |
irens | -0.064 |
izes | -0.062 |
Token | , |
Token ID | 11 |
Feature activation | -5.32e-03 |
Pos logit contributions | |
arl | +0.042 |
iren | +0.039 |
its | +0.037 |
isin | +0.036 |
irens | +0.035 |
Neg logit contributions | |
favourite | -0.017 |
one | -0.017 |
found | -0.016 |
favorite | -0.015 |
companion | -0.015 |
Token | my |
Token ID | 616 |
Feature activation | -2.05e-02 |
Pos logit contributions | |
favourite | +0.008 |
companion | +0.007 |
favorite | +0.007 |
found | +0.007 |
mate | +0.007 |
Neg logit contributions | |
arl | -0.012 |
iren | -0.011 |
its | -0.010 |
irens | -0.009 |
izes | -0.009 |
Token | d |
Token ID | 288 |
Feature activation | +9.27e-03 |
Pos logit contributions | |
favourite | +0.025 |
favorite | +0.022 |
companion | +0.022 |
3 | +0.021 |
one | +0.020 |
Neg logit contributions | |
arl | -0.051 |
iren | -0.049 |
irens | -0.045 |
its | -0.044 |
izes | -0.043 |
Token | arl |
Token ID | 7063 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
iren | +0.007 |
izes | +0.006 |
eling | +0.006 |
arl | +0.006 |
isin | +0.005 |
Neg logit contributions | |
mate | -0.024 |
favourite | -0.024 |
companion | -0.023 |
favorite | -0.023 |
3 | -0.022 |
Token | ings |
Token ID | 654 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
iren | +0.303 |
arl | +0.293 |
eling | +0.284 |
irens | +0.271 |
isin | +0.264 |
Neg logit contributions | |
favourite | -0.322 |
companion | -0.313 |
one | -0.310 |
favorite | -0.308 |
mate | -0.297 |
Token | . |
Token ID | 13 |
Feature activation | +9.59e-03 |
Pos logit contributions | |
arl | +0.054 |
iren | +0.049 |
its | +0.047 |
isin | +0.044 |
eling | +0.044 |
Neg logit contributions | |
found | -0.041 |
one | -0.041 |
favourite | -0.041 |
companion | -0.039 |
Sally | -0.039 |
Token | Enjoy |
Token ID | 18179 |
Feature activation | +4.42e-02 |
Pos logit contributions | |
arl | +0.018 |
iren | +0.017 |
its | +0.015 |
izes | +0.014 |
irens | +0.014 |
Neg logit contributions | |
favourite | -0.016 |
companion | -0.015 |
favorite | -0.015 |
found | -0.015 |
mate | -0.015 |
Token | your |
Token ID | 534 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
iren | +0.089 |
arl | +0.089 |
bleed | +0.078 |
irens | +0.076 |
its | +0.073 |
Neg logit contributions | |
one | -0.075 |
favourite | -0.067 |
found | -0.065 |
favorite | -0.064 |
Sally | -0.060 |
Token | lunch |
Token ID | 9965 |
Feature activation | +3.06e-02 |
Pos logit contributions | |
favourite | +0.022 |
favorite | +0.020 |
companion | +0.019 |
one | +0.018 |
found | +0.018 |
Neg logit contributions | |
arl | -0.041 |
iren | -0.039 |
its | -0.034 |
irens | -0.034 |
izes | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | +3.18e-02 |
Pos logit contributions | |
favourite | +0.013 |
companion | +0.012 |
favorite | +0.012 |
found | +0.012 |
one | +0.012 |
Neg logit contributions | |
arl | -0.021 |
iren | -0.019 |
its | -0.018 |
izes | -0.017 |
yourselves | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | +2.58e-02 |
Pos logit contributions | |
arl | +0.082 |
iren | +0.076 |
its | +0.073 |
izes | +0.072 |
eling | +0.071 |
Neg logit contributions | |
favourite | -0.030 |
companion | -0.028 |
one | -0.027 |
found | -0.027 |
Sally | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | +2.57e-02 |
Pos logit contributions | |
arl | +0.080 |
iren | +0.076 |
eling | +0.075 |
izes | +0.074 |
isin | +0.072 |
Neg logit contributions | |
found | -0.008 |
one | -0.007 |
Sally | -0.007 |
One | -0.006 |
His | -0.006 |
Token | Rex |
Token ID | 47389 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
arl | +0.055 |
its | +0.049 |
iren | +0.048 |
eling | +0.048 |
ently | +0.046 |
Neg logit contributions | |
one | -0.033 |
found | -0.033 |
Sally | -0.030 |
mate | -0.029 |
companion | -0.028 |
Token | sn |
Token ID | 3013 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
arl | +0.017 |
iren | +0.016 |
its | +0.013 |
irens | +0.013 |
izes | +0.013 |
Neg logit contributions | |
favourite | -0.020 |
favorite | -0.019 |
found | -0.019 |
companion | -0.019 |
one | -0.018 |
Token | arl |
Token ID | 7063 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
isin | +0.044 |
eling | +0.042 |
arl | +0.040 |
iren | +0.039 |
fters | +0.038 |
Neg logit contributions | |
found | -0.045 |
Sally | -0.044 |
searched | -0.042 |
companion | -0.041 |
met | -0.040 |
Token | s |
Token ID | 82 |
Feature activation | -1.97e-02 |
Pos logit contributions | |
eling | +0.334 |
isin | +0.318 |
arl | +0.305 |
rou | +0.303 |
iren | +0.293 |
Neg logit contributions | |
one | -0.265 |
companion | -0.260 |
favorite | -0.252 |
friend | -0.251 |
favourite | -0.246 |
Token | . |
Token ID | 13 |
Feature activation | +3.04e-02 |
Pos logit contributions | |
one | +0.026 |
favourite | +0.026 |
found | +0.026 |
favorite | +0.025 |
companion | +0.025 |
Neg logit contributions | |
arl | -0.044 |
iren | -0.042 |
isin | -0.038 |
its | -0.037 |
irens | -0.037 |
Token | He |
Token ID | 679 |
Feature activation | +7.38e-02 |
Pos logit contributions | |
arl | +0.075 |
iren | +0.071 |
its | +0.065 |
izes | +0.065 |
isin | +0.063 |
Neg logit contributions | |
favourite | -0.033 |
companion | -0.030 |
favorite | -0.030 |
found | -0.029 |
Sally | -0.029 |
Token | does |
Token ID | 857 |
Feature activation | +7.33e-02 |
Pos logit contributions | |
arl | +0.146 |
iren | +0.132 |
irens | +0.112 |
its | +0.110 |
ister | +0.109 |
Neg logit contributions | |
favourite | -0.126 |
favorite | -0.120 |
found | -0.119 |
one | -0.117 |
companion | -0.113 |
Token | not |
Token ID | 407 |
Feature activation | +4.99e-02 |
Pos logit contributions | |
arl | +0.191 |
iren | +0.180 |
irens | +0.162 |
ymes | +0.156 |
isin | +0.154 |
Neg logit contributions | |
one | -0.086 |
favorite | -0.075 |
favourite | -0.073 |
found | -0.067 |
Sally | -0.061 |
Token | you |
Token ID | 345 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
arl | +0.016 |
iren | +0.016 |
its | +0.013 |
irens | +0.012 |
izes | +0.012 |
Neg logit contributions | |
favourite | -0.019 |
one | -0.019 |
found | -0.018 |
favorite | -0.018 |
Sally | -0.018 |
Token | okay |
Token ID | 8788 |
Feature activation | -8.29e-03 |
Pos logit contributions | |
favourite | +0.028 |
companion | +0.026 |
favorite | +0.025 |
mate | +0.024 |
3 | +0.024 |
Neg logit contributions | |
arl | -0.039 |
iren | -0.035 |
its | -0.032 |
izes | -0.031 |
yourselves | -0.031 |
Token | , |
Token ID | 11 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
favourite | +0.012 |
companion | +0.012 |
favorite | +0.011 |
found | +0.011 |
mate | +0.011 |
Neg logit contributions | |
arl | -0.018 |
iren | -0.016 |
its | -0.014 |
izes | -0.014 |
irens | -0.014 |
Token | my |
Token ID | 616 |
Feature activation | -1.26e-03 |
Pos logit contributions | |
arl | +0.040 |
iren | +0.038 |
its | +0.034 |
izes | +0.033 |
irens | +0.033 |
Neg logit contributions | |
favourite | -0.026 |
companion | -0.024 |
favorite | -0.024 |
mate | -0.024 |
found | -0.024 |
Token | d |
Token ID | 288 |
Feature activation | +2.79e-02 |
Pos logit contributions | |
favourite | +0.002 |
favorite | +0.002 |
found | +0.002 |
companion | +0.002 |
one | +0.002 |
Neg logit contributions | |
arl | -0.003 |
iren | -0.003 |
irens | -0.002 |
izes | -0.002 |
its | -0.002 |
Token | arl |
Token ID | 7063 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
izes | +0.012 |
eling | +0.010 |
isin | +0.008 |
iren | +0.007 |
yourselves | +0.005 |
Neg logit contributions | |
mate | -0.076 |
Sally | -0.074 |
companion | -0.072 |
Joey | -0.071 |
favourite | -0.071 |
Token | ings |
Token ID | 654 |
Feature activation | +3.92e-02 |
Pos logit contributions | |
eling | +0.305 |
iren | +0.287 |
isin | +0.279 |
arl | +0.278 |
rou | +0.269 |
Neg logit contributions | |
companion | -0.306 |
one | -0.290 |
found | -0.281 |
friend | -0.277 |
favourite | -0.276 |
Token | ?" |
Token ID | 1701 |
Feature activation | +5.42e-02 |
Pos logit contributions | |
arl | +0.064 |
its | +0.058 |
eling | +0.057 |
iren | +0.056 |
isin | +0.056 |
Neg logit contributions | |
found | -0.066 |
Sally | -0.066 |
one | -0.062 |
Jack | -0.061 |
Joey | -0.059 |
Token | Mom |
Token ID | 11254 |
Feature activation | +2.87e-02 |
Pos logit contributions | |
arl | +0.119 |
iren | +0.111 |
its | +0.097 |
yourselves | +0.096 |
izes | +0.096 |
Neg logit contributions | |
favourite | -0.075 |
companion | -0.072 |
mate | -0.068 |
favorite | -0.066 |
found | -0.065 |
Token | asked |
Token ID | 1965 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
arl | +0.071 |
iren | +0.069 |
its | +0.061 |
izes | +0.060 |
irens | +0.060 |
Neg logit contributions | |
favourite | -0.033 |
companion | -0.030 |
favorite | -0.028 |
mate | -0.027 |
3 | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | +2.43e-02 |
Pos logit contributions | |
favourite | +0.015 |
companion | +0.014 |
favorite | +0.014 |
found | +0.014 |
mate | +0.014 |
Neg logit contributions | |
arl | -0.022 |
iren | -0.020 |
its | -0.018 |
izes | -0.017 |
irens | -0.017 |
Token | But |
Token ID | 887 |
Feature activation | -6.34e-03 |
Pos logit contributions | |
arl | +0.061 |
iren | +0.058 |
its | +0.054 |
izes | +0.053 |
irens | +0.052 |
Neg logit contributions | |
favourite | -0.035 |
companion | -0.032 |
favorite | -0.031 |
Sally | -0.031 |
3 | -0.031 |
Token | I |
Token ID | 314 |
Feature activation | -2.47e-02 |
Pos logit contributions | |
favourite | +0.011 |
found | +0.010 |
companion | +0.010 |
favorite | +0.010 |
one | +0.010 |
Neg logit contributions | |
arl | -0.012 |
iren | -0.011 |
its | -0.010 |
irens | -0.009 |
izes | -0.009 |
Token | 'm |
Token ID | 1101 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
favourite | +0.035 |
companion | +0.033 |
favorite | +0.031 |
found | +0.031 |
mate | +0.031 |
Neg logit contributions | |
arl | -0.054 |
iren | -0.050 |
its | -0.045 |
izes | -0.044 |
irens | -0.043 |
Token | sad |
Token ID | 6507 |
Feature activation | +5.89e-02 |
Pos logit contributions | |
favourite | +0.034 |
companion | +0.032 |
favorite | +0.031 |
found | +0.031 |
mate | +0.031 |
Neg logit contributions | |
arl | -0.034 |
iren | -0.031 |
its | -0.027 |
izes | -0.026 |
irens | -0.026 |
Token | we |
Token ID | 356 |
Feature activation | -5.20e-02 |
Pos logit contributions | |
iren | +0.146 |
arl | +0.145 |
isin | +0.134 |
its | +0.134 |
irens | +0.131 |
Neg logit contributions | |
one | -0.066 |
favorite | -0.060 |
favourite | -0.059 |
Sally | -0.057 |
found | -0.055 |
Token | couldn |
Token ID | 3521 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
favourite | +0.062 |
found | +0.061 |
favorite | +0.057 |
companion | +0.055 |
one | +0.054 |
Neg logit contributions | |
arl | -0.126 |
iren | -0.121 |
irens | -0.109 |
its | -0.107 |
izes | -0.105 |
Token | 't |
Token ID | 470 |
Feature activation | +4.30e-02 |
Pos logit contributions | |
arl | +0.344 |
iren | +0.271 |
izes | +0.248 |
rou | +0.242 |
yourselves | +0.240 |
Neg logit contributions | |
one | -0.339 |
favourite | -0.329 |
found | -0.317 |
Sally | -0.309 |
favorite | -0.306 |
Token | finish |
Token ID | 5461 |
Feature activation | +6.40e-02 |
Pos logit contributions | |
arl | +0.104 |
iren | +0.097 |
its | +0.088 |
izes | +0.086 |
irens | +0.085 |
Neg logit contributions | |
favourite | -0.051 |
companion | -0.047 |
favorite | -0.045 |
found | -0.044 |
mate | -0.043 |
Token | our |
Token ID | 674 |
Feature activation | -3.72e-02 |
Pos logit contributions | |
arl | +0.158 |
iren | +0.147 |
its | +0.130 |
izes | +0.126 |
irens | +0.121 |
Neg logit contributions | |
favourite | -0.086 |
one | -0.085 |
favorite | -0.078 |
Sally | -0.073 |
found | -0.072 |
Token | show |
Token ID | 905 |
Feature activation | -6.92e-02 |
Pos logit contributions | |
favourite | +0.046 |
favorite | +0.042 |
companion | +0.040 |
one | +0.037 |
found | +0.037 |
Neg logit contributions | |
arl | -0.092 |
iren | -0.088 |
its | -0.077 |
irens | -0.077 |
izes | -0.075 |
Token | ." |
Token ID | 526 |
Feature activation | +1.80e-02 |
Pos logit contributions | |
favourite | +0.136 |
one | +0.130 |
companion | +0.129 |
favorite | +0.129 |
found | +0.128 |
Neg logit contributions | |
arl | -0.117 |
iren | -0.106 |
its | -0.093 |
irens | -0.088 |
isin | -0.087 |
Token | " |
Token ID | 366 |
Feature activation | +2.35e-02 |
Pos logit contributions | |
favourite | +0.001 |
one | +0.001 |
found | +0.001 |
3 | +0.001 |
companion | +0.001 |
Neg logit contributions | |
arl | -0.006 |
iren | -0.005 |
its | -0.005 |
izes | -0.005 |
yourselves | -0.005 |
Token | Hello |
Token ID | 15496 |
Feature activation | +6.74e-03 |
Pos logit contributions | |
arl | +0.046 |
iren | +0.042 |
its | +0.037 |
izes | +0.036 |
irens | +0.035 |
Neg logit contributions | |
favourite | -0.039 |
companion | -0.037 |
found | -0.036 |
favorite | -0.036 |
3 | -0.035 |
Token | , |
Token ID | 11 |
Feature activation | +6.44e-03 |
Pos logit contributions | |
arl | +0.016 |
iren | +0.015 |
its | +0.014 |
izes | +0.013 |
irens | +0.013 |
Neg logit contributions | |
favourite | -0.008 |
companion | -0.008 |
favorite | -0.007 |
found | -0.007 |
one | -0.007 |
Token | my |
Token ID | 616 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
arl | +0.013 |
iren | +0.012 |
its | +0.011 |
izes | +0.011 |
irens | +0.010 |
Neg logit contributions | |
favourite | -0.010 |
companion | -0.009 |
favorite | -0.009 |
mate | -0.009 |
found | -0.009 |
Token | d |
Token ID | 288 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
favourite | +0.020 |
favorite | +0.019 |
companion | +0.019 |
3 | +0.018 |
found | +0.018 |
Neg logit contributions | |
arl | -0.022 |
iren | -0.020 |
its | -0.018 |
izes | -0.017 |
irens | -0.017 |
Token | arl |
Token ID | 7063 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
izes | +0.009 |
eling | +0.009 |
isin | +0.007 |
swollen | +0.006 |
bleed | +0.006 |
Neg logit contributions | |
mate | -0.058 |
favourite | -0.055 |
3 | -0.054 |
Sally | -0.054 |
companion | -0.054 |
Token | ings |
Token ID | 654 |
Feature activation | +3.24e-02 |
Pos logit contributions | |
eling | +0.308 |
bleed | +0.286 |
isin | +0.281 |
arl | +0.280 |
rou | +0.279 |
Neg logit contributions | |
companion | -0.285 |
one | -0.276 |
favourite | -0.254 |
friend | -0.253 |
found | -0.248 |
Token | . |
Token ID | 13 |
Feature activation | +3.21e-02 |
Pos logit contributions | |
arl | +0.078 |
iren | +0.074 |
its | +0.073 |
eling | +0.071 |
izes | +0.071 |
Neg logit contributions | |
one | -0.032 |
found | -0.030 |
Sally | -0.029 |
companion | -0.028 |
Jack | -0.027 |
Token | I |
Token ID | 314 |
Feature activation | -2.52e-02 |
Pos logit contributions | |
arl | +0.060 |
iren | +0.055 |
its | +0.048 |
izes | +0.047 |
yourselves | +0.046 |
Neg logit contributions | |
favourite | -0.055 |
companion | -0.052 |
favorite | -0.051 |
found | -0.050 |
mate | -0.049 |
Token | 'm |
Token ID | 1101 |
Feature activation | -2.08e-02 |
Pos logit contributions | |
favourite | +0.034 |
companion | +0.031 |
favorite | +0.030 |
mate | +0.029 |
3 | +0.029 |
Neg logit contributions | |
arl | -0.057 |
iren | -0.053 |
its | -0.048 |
izes | -0.047 |
yourselves | -0.046 |
Token | so |
Token ID | 523 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
favourite | +0.037 |
companion | +0.035 |
found | +0.034 |
favorite | +0.034 |
mate | +0.034 |
Neg logit contributions | |
arl | -0.038 |
iren | -0.034 |
its | -0.030 |
izes | -0.029 |
irens | -0.028 |
Token | he |
Token ID | 339 |
Feature activation | +3.33e-02 |
Pos logit contributions | |
arl | +0.135 |
iren | +0.125 |
its | +0.114 |
izes | +0.111 |
irens | +0.109 |
Neg logit contributions | |
favourite | -0.071 |
companion | -0.064 |
found | -0.063 |
favorite | -0.062 |
one | -0.061 |
Token | was |
Token ID | 373 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
arl | +0.091 |
iren | +0.085 |
its | +0.077 |
izes | +0.076 |
irens | +0.076 |
Neg logit contributions | |
favourite | -0.030 |
companion | -0.027 |
found | -0.027 |
favorite | -0.025 |
one | -0.024 |
Token | feeling |
Token ID | 4203 |
Feature activation | +3.77e-02 |
Pos logit contributions | |
arl | +0.043 |
iren | +0.040 |
its | +0.036 |
izes | +0.035 |
irens | +0.034 |
Neg logit contributions | |
favourite | -0.028 |
companion | -0.026 |
found | -0.025 |
favorite | -0.025 |
mate | -0.024 |
Token | sad |
Token ID | 6507 |
Feature activation | +6.68e-02 |
Pos logit contributions | |
arl | +0.079 |
iren | +0.073 |
its | +0.065 |
irens | +0.064 |
izes | +0.061 |
Neg logit contributions | |
favourite | -0.061 |
found | -0.058 |
one | -0.057 |
favorite | -0.056 |
companion | -0.055 |
Token | and |
Token ID | 290 |
Feature activation | +7.20e-02 |
Pos logit contributions | |
arl | +0.161 |
iren | +0.156 |
its | +0.140 |
isin | +0.139 |
irens | +0.138 |
Neg logit contributions | |
favourite | -0.081 |
one | -0.077 |
favorite | -0.075 |
companion | -0.074 |
found | -0.072 |
Token | couldn |
Token ID | 3521 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
arl | +0.158 |
iren | +0.148 |
its | +0.132 |
irens | +0.130 |
isin | +0.124 |
Neg logit contributions | |
favourite | -0.103 |
found | -0.100 |
one | -0.096 |
favorite | -0.095 |
companion | -0.094 |
Token | 't |
Token ID | 470 |
Feature activation | +4.40e-02 |
Pos logit contributions | |
arl | +0.296 |
iren | +0.245 |
izes | +0.212 |
its | +0.209 |
yourselves | +0.207 |
Neg logit contributions | |
favourite | -0.372 |
one | -0.368 |
found | -0.350 |
Sally | -0.348 |
companion | -0.344 |
Token | breathe |
Token ID | 18044 |
Feature activation | +5.90e-02 |
Pos logit contributions | |
arl | +0.093 |
iren | +0.087 |
its | +0.079 |
izes | +0.077 |
yourselves | +0.075 |
Neg logit contributions | |
favourite | -0.063 |
companion | -0.058 |
favorite | -0.057 |
3 | -0.054 |
found | -0.054 |
Token | well |
Token ID | 880 |
Feature activation | +9.42e-02 |
Pos logit contributions | |
arl | +0.139 |
iren | +0.130 |
its | +0.113 |
irens | +0.113 |
isin | +0.112 |
Neg logit contributions | |
favourite | -0.076 |
one | -0.072 |
found | -0.070 |
favorite | -0.070 |
companion | -0.069 |
Token | . |
Token ID | 13 |
Feature activation | -5.12e-03 |
Pos logit contributions | |
arl | +0.223 |
iren | +0.209 |
its | +0.186 |
irens | +0.181 |
isin | +0.181 |
Neg logit contributions | |
favourite | -0.120 |
one | -0.111 |
favorite | -0.110 |
companion | -0.109 |
found | -0.108 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +9.60e-02 |
Pos logit contributions | |
favourite | +0.007 |
companion | +0.007 |
mate | +0.007 |
found | +0.006 |
favorite | +0.006 |
Neg logit contributions | |
arl | -0.011 |
iren | -0.010 |
its | -0.009 |
irens | -0.008 |
izes | -0.008 |
Token | independent |
Token ID | 4795 |
Feature activation | +7.70e-03 |
Pos logit contributions | |
favourite | +0.052 |
companion | +0.051 |
mate | +0.050 |
Joey | +0.048 |
picnic | +0.048 |
Neg logit contributions | |
arl | -0.053 |
iren | -0.051 |
yourselves | -0.046 |
izes | -0.044 |
irens | -0.043 |
Token | didn |
Token ID | 1422 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
arl | +0.019 |
iren | +0.017 |
its | +0.016 |
izes | +0.015 |
irens | +0.015 |
Neg logit contributions | |
favourite | -0.009 |
companion | -0.009 |
found | -0.008 |
favorite | -0.008 |
3 | -0.008 |
Token | 't |
Token ID | 470 |
Feature activation | +3.81e-02 |
Pos logit contributions | |
arl | +0.240 |
iren | +0.192 |
izes | +0.183 |
yourselves | +0.173 |
rou | +0.172 |
Neg logit contributions | |
one | -0.231 |
favourite | -0.217 |
Sally | -0.204 |
search | -0.203 |
found | -0.201 |
Token | mean |
Token ID | 1612 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
arl | +0.080 |
iren | +0.074 |
its | +0.067 |
izes | +0.064 |
yourselves | +0.064 |
Neg logit contributions | |
favourite | -0.058 |
companion | -0.053 |
favorite | -0.051 |
mate | -0.048 |
found | -0.048 |
Token | she |
Token ID | 673 |
Feature activation | +5.41e-02 |
Pos logit contributions | |
arl | +0.037 |
iren | +0.034 |
izes | +0.031 |
its | +0.031 |
isin | +0.030 |
Neg logit contributions | |
favourite | -0.015 |
one | -0.014 |
companion | -0.013 |
favorite | -0.013 |
Sally | -0.013 |
Token | couldn |
Token ID | 3521 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
arl | +0.134 |
iren | +0.126 |
its | +0.114 |
izes | +0.112 |
irens | +0.110 |
Neg logit contributions | |
favourite | -0.060 |
companion | -0.055 |
favorite | -0.052 |
found | -0.052 |
mate | -0.051 |
Token | 't |
Token ID | 470 |
Feature activation | +3.94e-02 |
Pos logit contributions | |
arl | +0.286 |
iren | +0.231 |
izes | +0.199 |
its | +0.192 |
yourselves | +0.189 |
Neg logit contributions | |
favourite | -0.380 |
one | -0.369 |
found | -0.362 |
Sally | -0.357 |
companion | -0.355 |
Token | ask |
Token ID | 1265 |
Feature activation | -2.06e-02 |
Pos logit contributions | |
arl | +0.076 |
iren | +0.072 |
its | +0.067 |
izes | +0.061 |
bleed | +0.061 |
Neg logit contributions | |
favourite | -0.063 |
companion | -0.059 |
favorite | -0.057 |
mate | -0.055 |
3 | -0.055 |
Token | for |
Token ID | 329 |
Feature activation | +4.60e-02 |
Pos logit contributions | |
favourite | +0.029 |
one | +0.026 |
favorite | +0.026 |
companion | +0.026 |
3 | +0.025 |
Neg logit contributions | |
arl | -0.046 |
iren | -0.045 |
its | -0.041 |
izes | -0.039 |
irens | -0.038 |
Token | help |
Token ID | 1037 |
Feature activation | +7.01e-03 |
Pos logit contributions | |
arl | +0.100 |
iren | +0.093 |
its | +0.083 |
izes | +0.081 |
irens | +0.080 |
Neg logit contributions | |
favourite | -0.066 |
companion | -0.061 |
favorite | -0.059 |
found | -0.058 |
mate | -0.058 |
Token | when |
Token ID | 618 |
Feature activation | +4.96e-02 |
Pos logit contributions | |
arl | +0.016 |
iren | +0.015 |
its | +0.013 |
izes | +0.013 |
irens | +0.013 |
Neg logit contributions | |
favourite | -0.010 |
companion | -0.009 |
found | -0.009 |
one | -0.009 |
favorite | -0.009 |
Token | you |
Token ID | 345 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
arl | +0.018 |
iren | +0.017 |
its | +0.014 |
irens | +0.014 |
izes | +0.013 |
Neg logit contributions | |
favourite | -0.021 |
one | -0.021 |
found | -0.020 |
favorite | -0.020 |
Sally | -0.020 |
Token | okay |
Token ID | 8788 |
Feature activation | -6.19e-03 |
Pos logit contributions | |
favourite | +0.029 |
companion | +0.027 |
mate | +0.026 |
favorite | +0.026 |
3 | +0.025 |
Neg logit contributions | |
arl | -0.036 |
iren | -0.033 |
izes | -0.028 |
its | -0.028 |
yourselves | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
favourite | +0.009 |
companion | +0.008 |
favorite | +0.008 |
found | +0.008 |
mate | +0.008 |
Neg logit contributions | |
arl | -0.014 |
iren | -0.013 |
its | -0.011 |
izes | -0.011 |
irens | -0.011 |
Token | my |
Token ID | 616 |
Feature activation | -2.81e-03 |
Pos logit contributions | |
arl | +0.043 |
iren | +0.039 |
its | +0.034 |
yourselves | +0.034 |
izes | +0.034 |
Neg logit contributions | |
favourite | -0.027 |
companion | -0.025 |
found | -0.024 |
spotted | -0.023 |
favorite | -0.023 |
Token | d |
Token ID | 288 |
Feature activation | +2.74e-02 |
Pos logit contributions | |
favourite | +0.005 |
companion | +0.004 |
favorite | +0.004 |
found | +0.004 |
3 | +0.004 |
Neg logit contributions | |
arl | -0.005 |
iren | -0.005 |
its | -0.004 |
izes | -0.004 |
irens | -0.004 |
Token | arl |
Token ID | 7063 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
izes | +0.008 |
eling | +0.007 |
isin | +0.006 |
iren | +0.004 |
yourselves | +0.003 |
Neg logit contributions | |
mate | -0.077 |
Sally | -0.075 |
companion | -0.075 |
favourite | -0.072 |
Joey | -0.072 |
Token | ings |
Token ID | 654 |
Feature activation | +3.94e-02 |
Pos logit contributions | |
eling | +0.277 |
arl | +0.266 |
iren | +0.264 |
isin | +0.260 |
rou | +0.250 |
Neg logit contributions | |
companion | -0.316 |
one | -0.297 |
favourite | -0.289 |
found | -0.289 |
Sally | -0.281 |
Token | ?" |
Token ID | 1701 |
Feature activation | +5.39e-02 |
Pos logit contributions | |
arl | +0.067 |
its | +0.061 |
eling | +0.060 |
isin | +0.059 |
yourselves | +0.058 |
Neg logit contributions | |
found | -0.065 |
Sally | -0.062 |
one | -0.060 |
Jack | -0.058 |
spotted | -0.057 |
Token | she |
Token ID | 673 |
Feature activation | +6.14e-02 |
Pos logit contributions | |
arl | +0.105 |
iren | +0.100 |
its | +0.086 |
irens | +0.085 |
yourselves | +0.084 |
Neg logit contributions | |
favourite | -0.087 |
companion | -0.085 |
mate | -0.080 |
favorite | -0.077 |
3 | -0.074 |
Token | asked |
Token ID | 1965 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
arl | +0.133 |
iren | +0.132 |
izes | +0.118 |
its | +0.116 |
isin | +0.115 |
Neg logit contributions | |
favourite | -0.081 |
companion | -0.075 |
favorite | -0.068 |
Joey | -0.066 |
mate | -0.065 |
Token | . |
Token ID | 13 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
favourite | +0.024 |
companion | +0.023 |
mate | +0.022 |
favorite | +0.022 |
found | +0.022 |
Neg logit contributions | |
arl | -0.028 |
iren | -0.025 |
its | -0.022 |
izes | -0.022 |
irens | -0.021 |
Token | Spot |
Token ID | 32565 |
Feature activation | +2.94e-03 |
Pos logit contributions | |
arl | +0.034 |
iren | +0.032 |
its | +0.030 |
eling | +0.029 |
irens | +0.028 |
Neg logit contributions | |
found | -0.024 |
companion | -0.024 |
one | -0.024 |
favourite | -0.023 |
mate | -0.023 |
Token | grow |
Token ID | 1663 |
Feature activation | +9.42e-03 |
Pos logit contributions | |
arl | +0.005 |
iren | +0.005 |
its | +0.004 |
irens | +0.004 |
izes | +0.004 |
Neg logit contributions | |
favourite | -0.005 |
found | -0.005 |
favorite | -0.005 |
companion | -0.005 |
one | -0.005 |
Token | ls |
Token ID | 7278 |
Feature activation | -4.70e-02 |
Pos logit contributions | |
arl | +0.020 |
iren | +0.017 |
isin | +0.016 |
aid | +0.015 |
ister | +0.015 |
Neg logit contributions | |
one | -0.015 |
favorite | -0.014 |
favourite | -0.014 |
found | -0.013 |
three | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
favourite | +0.079 |
companion | +0.073 |
favorite | +0.071 |
mate | +0.070 |
3 | +0.070 |
Neg logit contributions | |
arl | -0.089 |
iren | -0.082 |
its | -0.072 |
izes | -0.070 |
irens | -0.068 |
Token | sn |
Token ID | 3013 |
Feature activation | +2.08e-02 |
Pos logit contributions | |
arl | +0.045 |
iren | +0.042 |
its | +0.036 |
irens | +0.035 |
izes | +0.035 |
Neg logit contributions | |
favourite | -0.036 |
companion | -0.034 |
favorite | -0.034 |
found | -0.033 |
one | -0.033 |
Token | arl |
Token ID | 7063 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
isin | +0.034 |
eling | +0.033 |
swollen | +0.031 |
iren | +0.031 |
its | +0.030 |
Neg logit contributions | |
found | -0.031 |
Sally | -0.031 |
searched | -0.030 |
favorite | -0.029 |
favourite | -0.029 |
Token | s |
Token ID | 82 |
Feature activation | -2.63e-02 |
Pos logit contributions | |
eling | +0.349 |
isin | +0.323 |
iren | +0.321 |
rou | +0.320 |
arl | +0.317 |
Neg logit contributions | |
one | -0.244 |
favorite | -0.231 |
companion | -0.229 |
favourite | -0.224 |
friend | -0.218 |
Token | . |
Token ID | 13 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
favourite | +0.037 |
found | +0.035 |
one | +0.035 |
favorite | +0.035 |
companion | +0.034 |
Neg logit contributions | |
arl | -0.059 |
iren | -0.055 |
its | -0.048 |
isin | -0.048 |
irens | -0.048 |
Token | He |
Token ID | 679 |
Feature activation | +6.28e-02 |
Pos logit contributions | |
arl | +0.049 |
iren | +0.045 |
its | +0.040 |
yourselves | +0.039 |
irens | +0.039 |
Neg logit contributions | |
favourite | -0.029 |
companion | -0.027 |
found | -0.026 |
mate | -0.026 |
favorite | -0.026 |
Token | tries |
Token ID | 8404 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
arl | +0.122 |
iren | +0.110 |
irens | +0.093 |
ister | +0.090 |
isin | +0.089 |
Neg logit contributions | |
favourite | -0.113 |
favorite | -0.107 |
found | -0.107 |
one | -0.104 |
companion | -0.100 |
Token | to |
Token ID | 284 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
favourite | +0.029 |
companion | +0.026 |
favorite | +0.025 |
found | +0.025 |
mate | +0.025 |
Neg logit contributions | |
arl | -0.063 |
iren | -0.059 |
its | -0.053 |
izes | -0.053 |
irens | -0.052 |
Token | apple |
Token ID | 17180 |
Feature activation | +4.50e-02 |
Pos logit contributions | |
arl | +0.067 |
iren | +0.061 |
its | +0.054 |
izes | +0.053 |
irens | +0.052 |
Neg logit contributions | |
favourite | -0.051 |
companion | -0.047 |
favorite | -0.046 |
found | -0.046 |
3 | -0.045 |
Token | and |
Token ID | 290 |
Feature activation | +2.71e-02 |
Pos logit contributions | |
arl | +0.102 |
iren | +0.094 |
its | +0.084 |
izes | +0.082 |
irens | +0.082 |
Neg logit contributions | |
favourite | -0.062 |
found | -0.058 |
companion | -0.057 |
one | -0.057 |
favorite | -0.056 |
Token | her |
Token ID | 607 |
Feature activation | +1.39e-02 |
Pos logit contributions | |
arl | +0.071 |
iren | +0.065 |
its | +0.058 |
irens | +0.057 |
izes | +0.056 |
Neg logit contributions | |
found | -0.032 |
favourite | -0.030 |
Sally | -0.030 |
one | -0.029 |
favorite | -0.028 |
Token | tum |
Token ID | 11814 |
Feature activation | +4.31e-02 |
Pos logit contributions | |
arl | +0.035 |
iren | +0.033 |
its | +0.030 |
izes | +0.029 |
irens | +0.029 |
Neg logit contributions | |
favourite | -0.016 |
favorite | -0.015 |
companion | -0.014 |
found | -0.014 |
one | -0.013 |
Token | my |
Token ID | 1820 |
Feature activation | +8.27e-02 |
Pos logit contributions | |
arl | +0.114 |
iren | +0.105 |
izes | +0.100 |
its | +0.098 |
irens | +0.096 |
Neg logit contributions | |
Sally | -0.037 |
found | -0.036 |
one | -0.035 |
favourite | -0.035 |
favorite | -0.032 |
Token | couldn |
Token ID | 3521 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
arl | +0.224 |
iren | +0.196 |
irens | +0.181 |
yourselves | +0.174 |
ands | +0.172 |
Neg logit contributions | |
found | -0.104 |
one | -0.096 |
favourite | -0.092 |
belonged | -0.089 |
favorite | -0.089 |
Token | 't |
Token ID | 470 |
Feature activation | +3.63e-02 |
Pos logit contributions | |
arl | +0.337 |
iren | +0.276 |
izes | +0.258 |
rou | +0.254 |
yourselves | +0.249 |
Neg logit contributions | |
one | -0.319 |
favourite | -0.302 |
Sally | -0.301 |
found | -0.298 |
search | -0.285 |
Token | handle |
Token ID | 5412 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
arl | +0.081 |
iren | +0.076 |
its | +0.068 |
izes | +0.067 |
irens | +0.066 |
Neg logit contributions | |
favourite | -0.049 |
companion | -0.046 |
found | -0.045 |
favorite | -0.044 |
mate | -0.043 |
Token | it |
Token ID | 340 |
Feature activation | -3.96e-03 |
Pos logit contributions | |
arl | +0.048 |
iren | +0.043 |
izes | +0.041 |
its | +0.039 |
ands | +0.038 |
Neg logit contributions | |
one | -0.024 |
Sally | -0.022 |
found | -0.021 |
favourite | -0.019 |
favorite | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | +1.55e-03 |
Pos logit contributions | |
favourite | +0.005 |
found | +0.005 |
one | +0.005 |
Sally | +0.005 |
favorite | +0.005 |
Neg logit contributions | |
arl | -0.009 |
iren | -0.009 |
irens | -0.007 |
izes | -0.007 |
its | -0.007 |
Token | She |
Token ID | 1375 |
Feature activation | +7.12e-02 |
Pos logit contributions | |
arl | +0.004 |
iren | +0.003 |
its | +0.003 |
izes | +0.003 |
yourselves | +0.003 |
Neg logit contributions | |
favourite | -0.002 |
companion | -0.002 |
favorite | -0.002 |
found | -0.002 |
mate | -0.002 |
Token | didn |
Token ID | 1422 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
arl | +0.220 |
iren | +0.206 |
its | +0.191 |
izes | +0.188 |
yourselves | +0.182 |
Neg logit contributions | |
favourite | -0.094 |
companion | -0.088 |
mate | -0.080 |
favorite | -0.079 |
3 | -0.078 |
Token | 't |
Token ID | 470 |
Feature activation | +4.22e-02 |
Pos logit contributions | |
arl | +0.237 |
eling | +0.198 |
iren | +0.195 |
izes | +0.189 |
yourselves | +0.188 |
Neg logit contributions | |
favourite | -0.227 |
one | -0.225 |
favorite | -0.211 |
Sally | -0.207 |
3 | -0.202 |
Token | understand |
Token ID | 1833 |
Feature activation | +2.04e-02 |
Pos logit contributions | |
arl | +0.093 |
iren | +0.086 |
its | +0.077 |
izes | +0.076 |
irens | +0.074 |
Neg logit contributions | |
favourite | -0.059 |
companion | -0.055 |
favorite | -0.053 |
found | -0.052 |
3 | -0.051 |
Token | why |
Token ID | 1521 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
arl | +0.048 |
iren | +0.045 |
izes | +0.040 |
its | +0.039 |
irens | +0.039 |
Neg logit contributions | |
one | -0.027 |
Sally | -0.025 |
favourite | -0.025 |
favorite | -0.024 |
found | -0.023 |
Token | she |
Token ID | 673 |
Feature activation | +5.59e-02 |
Pos logit contributions | |
arl | +0.058 |
iren | +0.055 |
its | +0.051 |
izes | +0.050 |
irens | +0.048 |
Neg logit contributions | |
favourite | -0.023 |
Sally | -0.022 |
one | -0.021 |
favorite | -0.021 |
found | -0.020 |
Token | couldn |
Token ID | 3521 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
arl | +0.126 |
iren | +0.122 |
its | +0.116 |
izes | +0.111 |
eling | +0.108 |
Neg logit contributions | |
favourite | -0.065 |
mate | -0.062 |
companion | -0.061 |
3 | -0.056 |
Joey | -0.056 |
Token | 't |
Token ID | 470 |
Feature activation | +3.68e-02 |
Pos logit contributions | |
arl | +0.276 |
iren | +0.225 |
izes | +0.190 |
its | +0.189 |
yourselves | +0.187 |
Neg logit contributions | |
favourite | -0.382 |
one | -0.370 |
found | -0.360 |
companion | -0.358 |
favorite | -0.356 |
Token | marry |
Token ID | 12479 |
Feature activation | +7.81e-03 |
Pos logit contributions | |
arl | +0.071 |
iren | +0.068 |
its | +0.063 |
izes | +0.061 |
isin | +0.057 |
Neg logit contributions | |
favourite | -0.057 |
companion | -0.053 |
favorite | -0.051 |
3 | -0.051 |
Joey | -0.050 |
Token | Lily |
Token ID | 20037 |
Feature activation | +2.89e-02 |
Pos logit contributions | |
arl | +0.018 |
iren | +0.017 |
its | +0.015 |
izes | +0.015 |
irens | +0.015 |
Neg logit contributions | |
favourite | -0.010 |
companion | -0.009 |
favorite | -0.009 |
found | -0.009 |
mate | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
arl | +0.062 |
iren | +0.057 |
its | +0.051 |
izes | +0.050 |
irens | +0.049 |
Neg logit contributions | |
favourite | -0.042 |
companion | -0.039 |
favorite | -0.037 |
found | -0.037 |
mate | -0.037 |
Token | She |
Token ID | 1375 |
Feature activation | +8.42e-02 |
Pos logit contributions | |
arl | +0.051 |
iren | +0.050 |
its | +0.045 |
izes | +0.045 |
isin | +0.044 |
Neg logit contributions | |
favourite | -0.023 |
favorite | -0.021 |
3 | -0.020 |
Sally | -0.020 |
found | -0.020 |
Token | What |
Token ID | 2061 |
Feature activation | +4.46e-03 |
Pos logit contributions | |
arl | +0.058 |
iren | +0.053 |
its | +0.048 |
izes | +0.046 |
irens | +0.045 |
Neg logit contributions | |
favourite | -0.046 |
companion | -0.044 |
found | -0.043 |
3 | -0.042 |
one | -0.042 |
Token | happened |
Token ID | 3022 |
Feature activation | +2.34e-03 |
Pos logit contributions | |
arl | +0.009 |
iren | +0.008 |
its | +0.008 |
izes | +0.007 |
yourselves | +0.007 |
Neg logit contributions | |
favourite | -0.007 |
companion | -0.007 |
mate | -0.006 |
Joey | -0.006 |
3 | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
arl | +0.006 |
iren | +0.006 |
its | +0.005 |
isin | +0.005 |
irens | +0.005 |
Neg logit contributions | |
one | -0.003 |
found | -0.003 |
favourite | -0.003 |
3 | -0.002 |
Sally | -0.002 |
Token | my |
Token ID | 616 |
Feature activation | -6.83e-03 |
Pos logit contributions | |
iren | +0.041 |
arl | +0.041 |
irens | +0.037 |
its | +0.036 |
izes | +0.036 |
Neg logit contributions | |
favourite | -0.019 |
Sally | -0.018 |
one | -0.018 |
mate | -0.018 |
favorite | -0.017 |
Token | d |
Token ID | 288 |
Feature activation | +2.21e-02 |
Pos logit contributions | |
favourite | +0.011 |
favorite | +0.011 |
companion | +0.010 |
found | +0.010 |
3 | +0.010 |
Neg logit contributions | |
arl | -0.014 |
iren | -0.013 |
irens | -0.011 |
izes | -0.011 |
its | -0.011 |
Token | arl |
Token ID | 7063 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
izes | +0.011 |
eling | +0.009 |
isin | +0.007 |
iren | +0.005 |
heit | +0.004 |
Neg logit contributions | |
mate | -0.058 |
Sally | -0.058 |
companion | -0.056 |
favourite | -0.055 |
three | -0.055 |
Token | ings |
Token ID | 654 |
Feature activation | +3.43e-02 |
Pos logit contributions | |
eling | +0.307 |
iren | +0.277 |
isin | +0.274 |
arl | +0.273 |
rou | +0.267 |
Neg logit contributions | |
companion | -0.299 |
one | -0.281 |
friend | -0.271 |
favourite | -0.269 |
found | -0.268 |
Token | ? |
Token ID | 30 |
Feature activation | +3.53e-02 |
Pos logit contributions | |
arl | +0.062 |
its | +0.060 |
isin | +0.058 |
cks | +0.058 |
iren | +0.058 |
Neg logit contributions | |
found | -0.050 |
Sally | -0.048 |
Jack | -0.046 |
one | -0.046 |
belonged | -0.043 |
Token | Why |
Token ID | 4162 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
arl | +0.077 |
iren | +0.071 |
its | +0.067 |
izes | +0.065 |
isin | +0.063 |
Neg logit contributions | |
favourite | -0.049 |
found | -0.044 |
companion | -0.044 |
Sally | -0.044 |
3 | -0.044 |
Token | do |
Token ID | 466 |
Feature activation | +8.12e-02 |
Pos logit contributions | |
arl | +0.033 |
iren | +0.031 |
irens | +0.027 |
its | +0.027 |
izes | +0.026 |
Neg logit contributions | |
favourite | -0.019 |
found | -0.018 |
favorite | -0.018 |
companion | -0.017 |
one | -0.017 |
Token | you |
Token ID | 345 |
Feature activation | +4.33e-03 |
Pos logit contributions | |
arl | +0.121 |
iren | +0.105 |
its | +0.089 |
izes | +0.083 |
irens | +0.079 |
Neg logit contributions | |
favourite | -0.175 |
one | -0.165 |
favorite | -0.164 |
companion | -0.163 |
mate | -0.162 |
Token | and |
Token ID | 290 |
Feature activation | +7.37e-03 |
Pos logit contributions | |
arl | +0.027 |
iren | +0.025 |
its | +0.022 |
izes | +0.022 |
irens | +0.022 |
Neg logit contributions | |
favourite | -0.020 |
companion | -0.018 |
favorite | -0.018 |
found | -0.018 |
mate | -0.017 |
Token | they |
Token ID | 484 |
Feature activation | -5.81e-02 |
Pos logit contributions | |
arl | +0.017 |
iren | +0.016 |
its | +0.014 |
izes | +0.014 |
irens | +0.014 |
Neg logit contributions | |
favourite | -0.010 |
companion | -0.009 |
favorite | -0.009 |
found | -0.008 |
mate | -0.008 |
Token | played |
Token ID | 2826 |
Feature activation | +2.41e-02 |
Pos logit contributions | |
favourite | +0.061 |
companion | +0.056 |
favorite | +0.052 |
3 | +0.052 |
mate | +0.051 |
Neg logit contributions | |
arl | -0.143 |
iren | -0.135 |
its | -0.124 |
izes | -0.124 |
irens | -0.117 |
Token | make |
Token ID | 787 |
Feature activation | +1.33e-03 |
Pos logit contributions | |
arl | +0.052 |
iren | +0.049 |
its | +0.043 |
izes | +0.043 |
yourselves | +0.042 |
Neg logit contributions | |
favourite | -0.033 |
companion | -0.031 |
favorite | -0.030 |
mate | -0.030 |
found | -0.030 |
Token | - |
Token ID | 12 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
arl | +0.003 |
iren | +0.003 |
its | +0.003 |
izes | +0.003 |
irens | +0.002 |
Neg logit contributions | |
favourite | -0.002 |
favorite | -0.002 |
companion | -0.002 |
found | -0.002 |
one | -0.002 |
Token | bel |
Token ID | 6667 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
arl | +0.028 |
iren | +0.023 |
its | +0.020 |
izes | +0.020 |
yourselves | +0.019 |
Neg logit contributions | |
favourite | -0.037 |
found | -0.035 |
favorite | -0.035 |
3 | -0.034 |
companion | -0.034 |
Token | ieve |
Token ID | 12311 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
arl | +0.345 |
iren | +0.261 |
falls | +0.214 |
ocus | +0.209 |
rou | +0.208 |
Neg logit contributions | |
favourite | -0.351 |
one | -0.333 |
found | -0.329 |
favorite | -0.325 |
three | -0.318 |
Token | . |
Token ID | 13 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
favourite | +0.017 |
companion | +0.015 |
favorite | +0.015 |
found | +0.015 |
one | +0.014 |
Neg logit contributions | |
arl | -0.027 |
iren | -0.025 |
its | -0.022 |
izes | -0.021 |
irens | -0.021 |
Token | Finally |
Token ID | 9461 |
Feature activation | +8.66e-02 |
Pos logit contributions | |
favourite | +0.019 |
companion | +0.019 |
mate | +0.018 |
3 | +0.018 |
found | +0.018 |
Neg logit contributions | |
arl | -0.029 |
iren | -0.026 |
its | -0.024 |
irens | -0.023 |
izes | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | -2.28e-02 |
Pos logit contributions | |
arl | +0.194 |
iren | +0.180 |
its | +0.162 |
izes | +0.159 |
irens | +0.156 |
Neg logit contributions | |
favourite | -0.118 |
companion | -0.110 |
favorite | -0.106 |
found | -0.105 |
mate | -0.103 |
Token | they |
Token ID | 484 |
Feature activation | -4.30e-02 |
Pos logit contributions | |
favourite | +0.031 |
companion | +0.030 |
mate | +0.028 |
favorite | +0.028 |
found | +0.028 |
Neg logit contributions | |
arl | -0.050 |
iren | -0.047 |
its | -0.042 |
izes | -0.041 |
yourselves | -0.040 |
Token | that |
Token ID | 326 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
arl | +0.031 |
iren | +0.028 |
its | +0.025 |
izes | +0.025 |
isin | +0.024 |
Neg logit contributions | |
favourite | -0.019 |
companion | -0.017 |
found | -0.017 |
favorite | -0.017 |
one | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | -9.29e-03 |
Pos logit contributions | |
favourite | +0.029 |
mate | +0.028 |
companion | +0.027 |
found | +0.026 |
favorite | +0.026 |
Neg logit contributions | |
arl | -0.032 |
iren | -0.030 |
izes | -0.027 |
its | -0.026 |
falls | -0.026 |
Token | Even |
Token ID | 3412 |
Feature activation | +2.89e-02 |
Pos logit contributions | |
favourite | +0.012 |
companion | +0.011 |
mate | +0.011 |
found | +0.011 |
3 | +0.011 |
Neg logit contributions | |
arl | -0.021 |
iren | -0.019 |
its | -0.017 |
irens | -0.017 |
izes | -0.016 |
Token | clumsy |
Token ID | 39120 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
arl | +0.053 |
iren | +0.048 |
its | +0.042 |
izes | +0.041 |
irens | +0.040 |
Neg logit contributions | |
favourite | -0.052 |
companion | -0.049 |
favorite | -0.048 |
found | -0.048 |
one | -0.048 |
Token | accidents |
Token ID | 17390 |
Feature activation | +8.14e-02 |
Pos logit contributions | |
arl | +0.061 |
iren | +0.055 |
its | +0.050 |
izes | +0.049 |
yourselves | +0.048 |
Neg logit contributions | |
favourite | -0.036 |
companion | -0.035 |
found | -0.034 |
one | -0.034 |
Sally | -0.033 |
Token | couldn |
Token ID | 3521 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
arl | +0.150 |
iren | +0.137 |
irens | +0.116 |
yourselves | +0.116 |
its | +0.114 |
Neg logit contributions | |
favourite | -0.142 |
found | -0.138 |
one | -0.133 |
favorite | -0.132 |
Sally | -0.132 |
Token | 't |
Token ID | 470 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
arl | +0.282 |
iren | +0.229 |
izes | +0.196 |
yourselves | +0.188 |
isin | +0.186 |
Neg logit contributions | |
favourite | -0.370 |
one | -0.364 |
found | -0.352 |
Sally | -0.347 |
favorite | -0.347 |
Token | ruin |
Token ID | 16866 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
arl | +0.069 |
iren | +0.066 |
its | +0.062 |
izes | +0.057 |
bleed | +0.055 |
Neg logit contributions | |
favourite | -0.070 |
companion | -0.067 |
favorite | -0.063 |
spotted | -0.062 |
favourites | -0.061 |
Token | their |
Token ID | 511 |
Feature activation | -2.51e-02 |
Pos logit contributions | |
arl | +0.044 |
iren | +0.041 |
its | +0.037 |
yourselves | +0.037 |
izes | +0.036 |
Neg logit contributions | |
favourite | -0.030 |
companion | -0.028 |
found | -0.027 |
favorite | -0.027 |
3 | -0.027 |
Token | day |
Token ID | 1110 |
Feature activation | +6.25e-02 |
Pos logit contributions | |
favourite | +0.029 |
companion | +0.028 |
Sally | +0.027 |
found | +0.027 |
Joey | +0.026 |
Neg logit contributions | |
arl | -0.058 |
iren | -0.053 |
its | -0.049 |
izes | -0.048 |
yourselves | -0.047 |
Token | ! |
Token ID | 0 |
Feature activation | -3.97e-03 |
Pos logit contributions | |
arl | +0.128 |
iren | +0.119 |
its | +0.106 |
izes | +0.103 |
irens | +0.102 |
Neg logit contributions | |
favourite | -0.095 |
companion | -0.089 |
favorite | -0.086 |
found | -0.085 |
mate | -0.085 |
Token | , |
Token ID | 11 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
arl | +0.004 |
iren | +0.004 |
its | +0.004 |
irens | +0.004 |
izes | +0.004 |
Neg logit contributions | |
favourite | -0.003 |
found | -0.003 |
companion | -0.003 |
one | -0.002 |
favorite | -0.002 |
Token | I |
Token ID | 314 |
Feature activation | -3.33e-02 |
Pos logit contributions | |
favourite | +0.031 |
favorite | +0.028 |
3 | +0.028 |
one | +0.028 |
found | +0.028 |
Neg logit contributions | |
arl | -0.034 |
iren | -0.032 |
its | -0.028 |
irens | -0.027 |
izes | -0.027 |
Token | 'm |
Token ID | 1101 |
Feature activation | -2.86e-02 |
Pos logit contributions | |
favourite | +0.041 |
found | +0.039 |
companion | +0.038 |
favorite | +0.037 |
spotted | +0.036 |
Neg logit contributions | |
arl | -0.079 |
iren | -0.074 |
irens | -0.066 |
yourselves | -0.066 |
its | -0.065 |
Token | sorry |
Token ID | 7926 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
favourite | +0.051 |
found | +0.050 |
one | +0.049 |
3 | +0.049 |
Sally | +0.049 |
Neg logit contributions | |
arl | -0.052 |
iren | -0.046 |
its | -0.042 |
izes | -0.040 |
irens | -0.039 |
Token | you |
Token ID | 345 |
Feature activation | -2.61e-05 |
Pos logit contributions | |
arl | +0.039 |
iren | +0.039 |
its | +0.035 |
irens | +0.034 |
izes | +0.033 |
Neg logit contributions | |
one | -0.021 |
favourite | -0.021 |
Sally | -0.020 |
favorite | -0.020 |
Joey | -0.019 |
Token | couldn |
Token ID | 3521 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
found | +0.000 |
favourite | +0.000 |
favorite | +0.000 |
companion | +0.000 |
spotted | +0.000 |
Neg logit contributions | |
arl | -0.000 |
iren | -0.000 |
irens | -0.000 |
its | -0.000 |
isin | -0.000 |
Token | 't |
Token ID | 470 |
Feature activation | +3.37e-02 |
Pos logit contributions | |
arl | +0.362 |
iren | +0.283 |
rou | +0.266 |
izes | +0.264 |
yourselves | +0.253 |
Neg logit contributions | |
one | -0.307 |
favourite | -0.288 |
found | -0.284 |
Jack | -0.280 |
Sally | -0.280 |
Token | fix |
Token ID | 4259 |
Feature activation | +6.57e-02 |
Pos logit contributions | |
arl | +0.076 |
iren | +0.070 |
its | +0.062 |
izes | +0.062 |
irens | +0.061 |
Neg logit contributions | |
favourite | -0.045 |
found | -0.043 |
companion | -0.042 |
one | -0.041 |
3 | -0.040 |
Token | those |
Token ID | 883 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
arl | +0.163 |
iren | +0.140 |
its | +0.125 |
izes | +0.124 |
isin | +0.116 |
Neg logit contributions | |
one | -0.097 |
favourite | -0.088 |
Sally | -0.085 |
Jack | -0.084 |
favorite | -0.082 |
Token | things |
Token ID | 1243 |
Feature activation | +4.61e-02 |
Pos logit contributions | |
favourite | +0.042 |
favorite | +0.039 |
found | +0.037 |
one | +0.037 |
3 | +0.036 |
Neg logit contributions | |
arl | -0.068 |
iren | -0.062 |
its | -0.054 |
izes | -0.053 |
irens | -0.053 |
Token | with |
Token ID | 351 |
Feature activation | -7.86e-02 |
Pos logit contributions | |
arl | +0.108 |
iren | +0.103 |
irens | +0.092 |
its | +0.092 |
isin | +0.088 |
Neg logit contributions | |
one | -0.057 |
favourite | -0.055 |
found | -0.055 |
Jack | -0.052 |
favorite | -0.051 |
Token | 'm |
Token ID | 1101 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
favourite | +0.029 |
companion | +0.027 |
favorite | +0.026 |
mate | +0.025 |
3 | +0.025 |
Neg logit contributions | |
arl | -0.034 |
iren | -0.031 |
its | -0.028 |
izes | -0.027 |
ister | -0.026 |
Token | sorry |
Token ID | 7926 |
Feature activation | +1.45e-02 |
Pos logit contributions | |
favourite | +0.056 |
companion | +0.053 |
favorite | +0.052 |
found | +0.052 |
mate | +0.052 |
Neg logit contributions | |
arl | -0.041 |
iren | -0.037 |
its | -0.031 |
izes | -0.030 |
irens | -0.029 |
Token | , |
Token ID | 11 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
arl | +0.034 |
iren | +0.032 |
its | +0.029 |
izes | +0.028 |
irens | +0.028 |
Neg logit contributions | |
favourite | -0.019 |
companion | -0.017 |
favorite | -0.017 |
one | -0.017 |
mate | -0.017 |
Token | my |
Token ID | 616 |
Feature activation | -7.87e-03 |
Pos logit contributions | |
arl | +0.032 |
iren | +0.031 |
its | +0.027 |
izes | +0.027 |
irens | +0.026 |
Neg logit contributions | |
favourite | -0.022 |
companion | -0.020 |
favorite | -0.020 |
mate | -0.020 |
one | -0.020 |
Token | d |
Token ID | 288 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
favourite | +0.014 |
companion | +0.013 |
found | +0.013 |
mate | +0.013 |
favorite | +0.013 |
Neg logit contributions | |
arl | -0.014 |
iren | -0.013 |
its | -0.011 |
izes | -0.011 |
yourselves | -0.011 |
Token | arl |
Token ID | 7063 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
eling | +0.008 |
izes | +0.008 |
isin | +0.005 |
iren | +0.005 |
yourselves | +0.004 |
Neg logit contributions | |
mate | -0.064 |
favourite | -0.062 |
favorite | -0.061 |
companion | -0.061 |
Sally | -0.060 |
Token | ings |
Token ID | 654 |
Feature activation | +3.57e-02 |
Pos logit contributions | |
eling | +0.286 |
arl | +0.277 |
iren | +0.273 |
isin | +0.263 |
rou | +0.260 |
Neg logit contributions | |
companion | -0.304 |
one | -0.288 |
favourite | -0.283 |
mate | -0.274 |
found | -0.272 |
Token | . |
Token ID | 13 |
Feature activation | +3.31e-02 |
Pos logit contributions | |
arl | +0.084 |
iren | +0.080 |
its | +0.075 |
izes | +0.074 |
eling | +0.074 |
Neg logit contributions | |
one | -0.037 |
Sally | -0.035 |
found | -0.035 |
companion | -0.034 |
favourite | -0.034 |
Token | The |
Token ID | 383 |
Feature activation | +4.81e-02 |
Pos logit contributions | |
arl | +0.066 |
iren | +0.060 |
its | +0.053 |
izes | +0.052 |
yourselves | +0.051 |
Neg logit contributions | |
favourite | -0.054 |
companion | -0.050 |
found | -0.049 |
favorite | -0.049 |
mate | -0.048 |
Token | tornado |
Token ID | 33718 |
Feature activation | +4.61e-02 |
Pos logit contributions | |
arl | +0.109 |
iren | +0.102 |
its | +0.092 |
izes | +0.090 |
irens | +0.089 |
Neg logit contributions | |
favourite | -0.065 |
companion | -0.060 |
favorite | -0.059 |
found | -0.057 |
mate | -0.057 |
Token | took |
Token ID | 1718 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
arl | +0.093 |
iren | +0.083 |
irens | +0.072 |
its | +0.070 |
yourselves | +0.069 |
Neg logit contributions | |
favourite | -0.076 |
found | -0.073 |
companion | -0.071 |
favorite | -0.071 |
one | -0.070 |
Token | " |
Token ID | 366 |
Feature activation | +7.33e-03 |
Pos logit contributions | |
favourite | +0.019 |
companion | +0.017 |
favorite | +0.016 |
found | +0.016 |
mate | +0.016 |
Neg logit contributions | |
arl | -0.044 |
iren | -0.041 |
its | -0.037 |
izes | -0.037 |
irens | -0.036 |
Token | I |
Token ID | 40 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
arl | +0.012 |
iren | +0.011 |
its | +0.009 |
isin | +0.009 |
yourselves | +0.009 |
Neg logit contributions | |
favourite | -0.014 |
favorite | -0.013 |
mate | -0.013 |
companion | -0.012 |
search | -0.012 |
Token | 'm |
Token ID | 1101 |
Feature activation | -3.31e-02 |
Pos logit contributions | |
favourite | +0.034 |
companion | +0.031 |
favorite | +0.030 |
mate | +0.029 |
3 | +0.029 |
Neg logit contributions | |
arl | -0.046 |
iren | -0.043 |
its | -0.039 |
izes | -0.038 |
isin | -0.036 |
Token | sorry |
Token ID | 7926 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
favourite | +0.060 |
companion | +0.056 |
found | +0.055 |
favorite | +0.055 |
mate | +0.054 |
Neg logit contributions | |
arl | -0.060 |
iren | -0.055 |
its | -0.048 |
izes | -0.047 |
irens | -0.045 |
Token | we |
Token ID | 356 |
Feature activation | -5.81e-02 |
Pos logit contributions | |
arl | +0.032 |
iren | +0.031 |
its | +0.028 |
izes | +0.027 |
irens | +0.027 |
Neg logit contributions | |
favourite | -0.016 |
favorite | -0.015 |
one | -0.014 |
Sally | -0.014 |
companion | -0.014 |
Token | couldn |
Token ID | 3521 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
favourite | +0.070 |
found | +0.067 |
companion | +0.065 |
favorite | +0.062 |
mate | +0.060 |
Neg logit contributions | |
arl | -0.139 |
iren | -0.132 |
its | -0.118 |
irens | -0.117 |
izes | -0.116 |
Token | 't |
Token ID | 470 |
Feature activation | +3.50e-02 |
Pos logit contributions | |
arl | +0.341 |
iren | +0.270 |
rou | +0.244 |
izes | +0.243 |
its | +0.238 |
Neg logit contributions | |
one | -0.326 |
favourite | -0.316 |
found | -0.308 |
Sally | -0.298 |
search | -0.295 |
Token | finish |
Token ID | 5461 |
Feature activation | +6.07e-02 |
Pos logit contributions | |
arl | +0.078 |
iren | +0.073 |
its | +0.066 |
izes | +0.064 |
yourselves | +0.063 |
Neg logit contributions | |
favourite | -0.047 |
companion | -0.043 |
favorite | -0.042 |
3 | -0.041 |
mate | -0.040 |
Token | the |
Token ID | 262 |
Feature activation | +4.23e-02 |
Pos logit contributions | |
arl | +0.154 |
iren | +0.142 |
its | +0.128 |
izes | +0.124 |
isin | +0.120 |
Neg logit contributions | |
favourite | -0.077 |
one | -0.076 |
favorite | -0.070 |
Sally | -0.066 |
3 | -0.064 |
Token | quiz |
Token ID | 38964 |
Feature activation | +5.81e-02 |
Pos logit contributions | |
arl | +0.104 |
iren | +0.093 |
its | +0.090 |
yourselves | +0.084 |
aned | +0.083 |
Neg logit contributions | |
favourite | -0.055 |
favorite | -0.051 |
one | -0.051 |
picnic | -0.049 |
3 | -0.048 |
Token | game |
Token ID | 983 |
Feature activation | +2.27e-02 |
Pos logit contributions | |
arl | +0.116 |
iren | +0.105 |
its | +0.094 |
izes | +0.090 |
irens | +0.089 |
Neg logit contributions | |
favourite | -0.094 |
companion | -0.089 |
found | -0.088 |
favorite | -0.087 |
one | -0.086 |
Token | small |
Token ID | 1402 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
favourite | +0.019 |
companion | +0.018 |
favorite | +0.017 |
found | +0.017 |
mate | +0.017 |
Neg logit contributions | |
arl | -0.032 |
iren | -0.030 |
its | -0.027 |
izes | -0.027 |
irens | -0.026 |
Token | house |
Token ID | 2156 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
favourite | +0.035 |
companion | +0.031 |
favorite | +0.031 |
mate | +0.030 |
found | +0.030 |
Neg logit contributions | |
arl | -0.057 |
iren | -0.054 |
yourselves | -0.047 |
irens | -0.046 |
its | -0.046 |
Token | . |
Token ID | 13 |
Feature activation | -2.06e-02 |
Pos logit contributions | |
arl | +0.044 |
iren | +0.041 |
its | +0.036 |
izes | +0.035 |
irens | +0.035 |
Neg logit contributions | |
favourite | -0.034 |
companion | -0.031 |
favorite | -0.030 |
found | -0.030 |
mate | -0.030 |
Token | The |
Token ID | 383 |
Feature activation | +7.40e-03 |
Pos logit contributions | |
favourite | +0.023 |
companion | +0.022 |
favorite | +0.021 |
found | +0.021 |
spotted | +0.019 |
Neg logit contributions | |
arl | -0.048 |
iren | -0.046 |
irens | -0.040 |
yourselves | -0.040 |
its | -0.039 |
Token | family |
Token ID | 1641 |
Feature activation | -2.49e-02 |
Pos logit contributions | |
arl | +0.018 |
iren | +0.017 |
its | +0.016 |
izes | +0.015 |
yourselves | +0.015 |
Neg logit contributions | |
favourite | -0.008 |
companion | -0.007 |
found | -0.007 |
favorite | -0.007 |
mate | -0.007 |
Token | consisted |
Token ID | 19954 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
favourite | +0.028 |
companion | +0.027 |
mate | +0.025 |
favorite | +0.024 |
Sally | +0.024 |
Neg logit contributions | |
iren | -0.054 |
arl | -0.053 |
izes | -0.050 |
its | -0.050 |
falls | -0.046 |
Token | of |
Token ID | 286 |
Feature activation | -4.79e-02 |
Pos logit contributions | |
favourite | +0.288 |
companion | +0.282 |
mate | +0.268 |
found | +0.266 |
favorite | +0.261 |
Neg logit contributions | |
arl | -0.305 |
iren | -0.287 |
its | -0.246 |
irens | -0.243 |
yourselves | -0.237 |
Token | two |
Token ID | 734 |
Feature activation | -9.55e-02 |
Pos logit contributions | |
favourite | +0.065 |
companion | +0.060 |
found | +0.058 |
favorite | +0.058 |
mate | +0.056 |
Neg logit contributions | |
arl | -0.107 |
iren | -0.099 |
its | -0.089 |
izes | -0.087 |
irens | -0.086 |
Token | parents |
Token ID | 3397 |
Feature activation | -2.19e-02 |
Pos logit contributions | |
favourite | +0.132 |
companion | +0.130 |
mate | +0.124 |
Sally | +0.122 |
found | +0.122 |
Neg logit contributions | |
arl | -0.191 |
iren | -0.180 |
its | -0.166 |
irens | -0.158 |
isin | -0.157 |
Token | and |
Token ID | 290 |
Feature activation | -2.06e-02 |
Pos logit contributions | |
favourite | +0.034 |
companion | +0.033 |
favorite | +0.031 |
mate | +0.031 |
found | +0.030 |
Neg logit contributions | |
arl | -0.041 |
iren | -0.038 |
its | -0.033 |
izes | -0.033 |
isin | -0.032 |
Token | two |
Token ID | 734 |
Feature activation | -8.86e-02 |
Pos logit contributions | |
favourite | +0.027 |
companion | +0.026 |
favorite | +0.024 |
mate | +0.023 |
found | +0.022 |
Neg logit contributions | |
arl | -0.040 |
iren | -0.039 |
yourselves | -0.035 |
irens | -0.035 |
izes | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -3.33e-02 |
Pos logit contributions | |
favourite | +0.033 |
companion | +0.031 |
one | +0.030 |
found | +0.030 |
favorite | +0.030 |
Neg logit contributions | |
arl | -0.051 |
iren | -0.047 |
its | -0.042 |
izes | -0.041 |
isin | -0.040 |
Token |
|
Token ID | 198 |
Feature activation | -2.05e-02 |
Pos logit contributions | |
mate | +0.058 |
companion | +0.057 |
favourite | +0.054 |
visitor | +0.052 |
favorite | +0.050 |
Neg logit contributions | |
arl | -0.056 |
iren | -0.052 |
its | -0.047 |
irens | -0.045 |
falls | -0.044 |
Token |
|
Token ID | 198 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
favourite | +0.027 |
companion | +0.025 |
favorite | +0.024 |
mate | +0.024 |
found | +0.023 |
Neg logit contributions | |
arl | -0.047 |
iren | -0.044 |
its | -0.039 |
izes | -0.038 |
irens | -0.038 |
Token | The |
Token ID | 464 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
favourite | +0.037 |
companion | +0.036 |
favorite | +0.035 |
mate | +0.034 |
3 | +0.034 |
Neg logit contributions | |
arl | -0.030 |
iren | -0.027 |
its | -0.023 |
izes | -0.023 |
irens | -0.022 |
Token | parade |
Token ID | 16134 |
Feature activation | -3.12e-02 |
Pos logit contributions | |
arl | +0.035 |
iren | +0.032 |
its | +0.030 |
izes | +0.029 |
yourselves | +0.028 |
Neg logit contributions | |
favourite | -0.021 |
companion | -0.019 |
favorite | -0.019 |
found | -0.019 |
3 | -0.018 |
Token | consisted |
Token ID | 19954 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
favourite | +0.044 |
companion | +0.040 |
favorite | +0.040 |
mate | +0.039 |
favourites | +0.038 |
Neg logit contributions | |
iren | -0.065 |
arl | -0.058 |
izes | -0.057 |
its | -0.055 |
irens | -0.055 |
Token | of |
Token ID | 286 |
Feature activation | -4.40e-02 |
Pos logit contributions | |
favourite | +0.329 |
companion | +0.314 |
favorite | +0.303 |
found | +0.299 |
mate | +0.299 |
Neg logit contributions | |
arl | -0.257 |
iren | -0.248 |
irens | -0.204 |
its | -0.197 |
yourselves | -0.188 |
Token | people |
Token ID | 661 |
Feature activation | -3.08e-02 |
Pos logit contributions | |
favourite | +0.065 |
companion | +0.060 |
favorite | +0.058 |
mate | +0.058 |
found | +0.057 |
Neg logit contributions | |
arl | -0.091 |
iren | -0.085 |
its | -0.075 |
irens | -0.075 |
yourselves | -0.073 |
Token | , |
Token ID | 11 |
Feature activation | -4.25e-02 |
Pos logit contributions | |
favourite | +0.042 |
companion | +0.039 |
favorite | +0.037 |
found | +0.037 |
mate | +0.037 |
Neg logit contributions | |
arl | -0.068 |
iren | -0.064 |
its | -0.057 |
izes | -0.056 |
irens | -0.055 |
Token | animals |
Token ID | 4695 |
Feature activation | -2.04e-02 |
Pos logit contributions | |
favourite | +0.057 |
companion | +0.054 |
favorite | +0.050 |
mate | +0.049 |
found | +0.047 |
Neg logit contributions | |
arl | -0.090 |
iren | -0.085 |
irens | -0.076 |
yourselves | -0.075 |
its | -0.075 |
Token | and |
Token ID | 290 |
Feature activation | -5.79e-03 |
Pos logit contributions | |
favourite | +0.030 |
companion | +0.027 |
favorite | +0.027 |
mate | +0.026 |
found | +0.026 |
Neg logit contributions | |
arl | -0.043 |
iren | -0.040 |
its | -0.036 |
izes | -0.035 |
irens | -0.035 |
Token | had |
Token ID | 550 |
Feature activation | -8.14e-04 |
Pos logit contributions | |
favourite | +0.068 |
companion | +0.067 |
3 | +0.058 |
mate | +0.058 |
favorite | +0.056 |
Neg logit contributions | |
arl | -0.140 |
iren | -0.136 |
its | -0.131 |
izes | -0.124 |
eling | -0.122 |
Token | three |
Token ID | 1115 |
Feature activation | -4.65e-02 |
Pos logit contributions | |
companion | +0.001 |
favourite | +0.001 |
Joey | +0.001 |
Sally | +0.001 |
3 | +0.001 |
Neg logit contributions | |
arl | -0.002 |
iren | -0.002 |
its | -0.001 |
yourselves | -0.001 |
izes | -0.001 |
Token | sides |
Token ID | 5389 |
Feature activation | -5.18e-02 |
Pos logit contributions | |
favourite | +0.071 |
favorite | +0.066 |
found | +0.065 |
one | +0.064 |
companion | +0.063 |
Neg logit contributions | |
arl | -0.101 |
iren | -0.092 |
izes | -0.080 |
its | -0.079 |
irens | -0.078 |
Token | and |
Token ID | 290 |
Feature activation | -4.16e-03 |
Pos logit contributions | |
favourite | +0.076 |
favorite | +0.070 |
one | +0.069 |
found | +0.069 |
companion | +0.068 |
Neg logit contributions | |
arl | -0.112 |
iren | -0.103 |
izes | -0.092 |
its | -0.092 |
irens | -0.090 |
Token | three |
Token ID | 1115 |
Feature activation | -4.21e-02 |
Pos logit contributions | |
favourite | +0.005 |
companion | +0.005 |
favorite | +0.004 |
found | +0.004 |
mate | +0.004 |
Neg logit contributions | |
arl | -0.010 |
iren | -0.009 |
its | -0.008 |
izes | -0.008 |
yourselves | -0.008 |
Token | points |
Token ID | 2173 |
Feature activation | -1.58e-01 |
Pos logit contributions | |
favourite | +0.061 |
favorite | +0.056 |
found | +0.055 |
one | +0.055 |
companion | +0.053 |
Neg logit contributions | |
arl | -0.096 |
iren | -0.088 |
izes | -0.076 |
its | -0.074 |
irens | -0.074 |
Token | . |
Token ID | 13 |
Feature activation | -2.34e-02 |
Pos logit contributions | |
favourite | +0.179 |
companion | +0.161 |
favorite | +0.159 |
found | +0.155 |
one | +0.152 |
Neg logit contributions | |
arl | -0.394 |
iren | -0.370 |
its | -0.333 |
izes | -0.329 |
irens | -0.325 |
Token | Ben |
Token ID | 3932 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
favourite | +0.033 |
companion | +0.033 |
mate | +0.032 |
found | +0.031 |
spotted | +0.030 |
Neg logit contributions | |
arl | -0.051 |
iren | -0.045 |
its | -0.040 |
irens | -0.039 |
yourselves | -0.038 |
Token | liked |
Token ID | 8288 |
Feature activation | +5.69e-03 |
Pos logit contributions | |
companion | +0.014 |
favourite | +0.013 |
Sally | +0.012 |
mate | +0.012 |
Joey | +0.012 |
Neg logit contributions | |
iren | -0.021 |
izes | -0.020 |
its | -0.020 |
falls | -0.020 |
arl | -0.019 |
Token | stars |
Token ID | 5788 |
Feature activation | +8.10e-03 |
Pos logit contributions | |
arl | +0.015 |
iren | +0.014 |
its | +0.013 |
izes | +0.012 |
irens | +0.012 |
Neg logit contributions | |
favourite | -0.006 |
favorite | -0.006 |
one | -0.005 |
found | -0.005 |
companion | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
arl | +0.018 |
iren | +0.017 |
its | +0.015 |
izes | +0.015 |
irens | +0.014 |
Neg logit contributions | |
favourite | -0.011 |
companion | -0.010 |
favorite | -0.010 |
found | -0.010 |
mate | -0.010 |
Token | a |
Token ID | 257 |
Feature activation | -4.44e-03 |
Pos logit contributions | |
favourite | +0.035 |
favorite | +0.031 |
companion | +0.031 |
found | +0.031 |
one | +0.031 |
Neg logit contributions | |
arl | -0.070 |
iren | -0.065 |
its | -0.059 |
izes | -0.058 |
irens | -0.057 |
Token | special |
Token ID | 2041 |
Feature activation | -4.41e-02 |
Pos logit contributions | |
favourite | +0.006 |
companion | +0.006 |
favorite | +0.006 |
found | +0.006 |
3 | +0.005 |
Neg logit contributions | |
arl | -0.010 |
iren | -0.009 |
its | -0.008 |
izes | -0.008 |
irens | -0.008 |
Token | tool |
Token ID | 2891 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
favourite | +0.064 |
companion | +0.061 |
found | +0.058 |
mate | +0.058 |
favorite | +0.057 |
Neg logit contributions | |
arl | -0.091 |
iren | -0.085 |
its | -0.076 |
izes | -0.073 |
yourselves | -0.072 |
Token | with |
Token ID | 351 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
arl | +0.032 |
iren | +0.029 |
its | +0.026 |
izes | +0.025 |
irens | +0.025 |
Neg logit contributions | |
favourite | -0.025 |
companion | -0.023 |
favorite | -0.023 |
found | -0.023 |
mate | -0.022 |
Token | three |
Token ID | 1115 |
Feature activation | -4.67e-02 |
Pos logit contributions | |
favourite | +0.143 |
one | +0.132 |
favorite | +0.127 |
found | +0.126 |
3 | +0.122 |
Neg logit contributions | |
arl | -0.295 |
iren | -0.275 |
izes | -0.249 |
its | -0.244 |
irens | -0.241 |
Token | points |
Token ID | 2173 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
favourite | +0.076 |
one | +0.069 |
favorite | +0.069 |
found | +0.068 |
3 | +0.067 |
Neg logit contributions | |
arl | -0.101 |
iren | -0.089 |
izes | -0.079 |
irens | -0.075 |
ister | -0.075 |
Token | , |
Token ID | 11 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
favourite | +0.215 |
one | +0.208 |
favorite | +0.198 |
found | +0.195 |
3 | +0.189 |
Neg logit contributions | |
arl | -0.367 |
iren | -0.345 |
izes | -0.308 |
irens | -0.301 |
its | -0.297 |
Token | like |
Token ID | 588 |
Feature activation | -1.97e-02 |
Pos logit contributions | |
favourite | +0.025 |
one | +0.023 |
favorite | +0.022 |
3 | +0.022 |
Sally | +0.022 |
Neg logit contributions | |
arl | -0.048 |
iren | -0.044 |
izes | -0.040 |
its | -0.040 |
irens | -0.039 |
Token | a |
Token ID | 257 |
Feature activation | +6.22e-03 |
Pos logit contributions | |
favourite | +0.021 |
one | +0.019 |
favorite | +0.018 |
3 | +0.018 |
three | +0.017 |
Neg logit contributions | |
arl | -0.054 |
iren | -0.051 |
izes | -0.047 |
its | -0.046 |
irens | -0.044 |
Token | star |
Token ID | 3491 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
arl | +0.015 |
iren | +0.014 |
irens | +0.013 |
its | +0.012 |
izes | +0.012 |
Neg logit contributions | |
favourite | -0.008 |
favorite | -0.008 |
3 | -0.007 |
one | -0.007 |
companion | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
one | +0.022 |
favourite | +0.022 |
mate | +0.021 |
found | +0.021 |
favorite | +0.021 |
Neg logit contributions | |
arl | -0.028 |
iren | -0.024 |
irens | -0.022 |
its | -0.022 |
izes | -0.021 |
Token | has |
Token ID | 468 |
Feature activation | -4.55e-03 |
Pos logit contributions | |
favourite | +0.022 |
companion | +0.022 |
3 | +0.020 |
favorite | +0.020 |
Sally | +0.020 |
Neg logit contributions | |
arl | -0.016 |
iren | -0.015 |
its | -0.013 |
izes | -0.013 |
yourselves | -0.012 |
Token | three |
Token ID | 1115 |
Feature activation | -4.42e-02 |
Pos logit contributions | |
favourite | +0.006 |
companion | +0.005 |
favorite | +0.005 |
found | +0.005 |
mate | +0.005 |
Neg logit contributions | |
arl | -0.011 |
iren | -0.010 |
its | -0.009 |
izes | -0.009 |
irens | -0.009 |
Token | sides |
Token ID | 5389 |
Feature activation | -4.76e-02 |
Pos logit contributions | |
favourite | +0.063 |
one | +0.059 |
favorite | +0.058 |
found | +0.058 |
3 | +0.056 |
Neg logit contributions | |
arl | -0.103 |
iren | -0.092 |
izes | -0.083 |
ister | -0.079 |
its | -0.077 |
Token | and |
Token ID | 290 |
Feature activation | +4.93e-03 |
Pos logit contributions | |
one | +0.061 |
three | +0.054 |
favourite | +0.052 |
found | +0.051 |
mate | +0.051 |
Neg logit contributions | |
arl | -0.112 |
iren | -0.104 |
izes | -0.103 |
eling | -0.099 |
irens | -0.097 |
Token | three |
Token ID | 1115 |
Feature activation | -3.98e-02 |
Pos logit contributions | |
arl | +0.013 |
iren | +0.012 |
izes | +0.010 |
its | +0.010 |
ister | +0.010 |
Neg logit contributions | |
favourite | -0.006 |
one | -0.006 |
favorite | -0.005 |
found | -0.005 |
3 | -0.005 |
Token | points |
Token ID | 2173 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
one | +0.054 |
favourite | +0.053 |
found | +0.050 |
favorite | +0.049 |
three | +0.049 |
Neg logit contributions | |
arl | -0.097 |
iren | -0.087 |
izes | -0.078 |
ister | -0.075 |
bleed | -0.073 |
Token | . |
Token ID | 13 |
Feature activation | -7.31e-03 |
Pos logit contributions | |
one | +0.189 |
favourite | +0.183 |
favorite | +0.174 |
three | +0.166 |
mate | +0.166 |
Neg logit contributions | |
arl | -0.391 |
iren | -0.377 |
isin | -0.330 |
izes | -0.329 |
irens | -0.328 |
Token | I |
Token ID | 314 |
Feature activation | -4.75e-02 |
Pos logit contributions | |
favourite | +0.010 |
companion | +0.009 |
mate | +0.009 |
3 | +0.009 |
one | +0.009 |
Neg logit contributions | |
arl | -0.016 |
iren | -0.015 |
its | -0.014 |
izes | -0.014 |
eling | -0.013 |
Token | can |
Token ID | 460 |
Feature activation | +5.42e-02 |
Pos logit contributions | |
favourite | +0.078 |
companion | +0.074 |
favorite | +0.071 |
3 | +0.070 |
mate | +0.069 |
Neg logit contributions | |
arl | -0.092 |
iren | -0.084 |
its | -0.076 |
izes | -0.072 |
yourselves | -0.069 |
Token | make |
Token ID | 787 |
Feature activation | -6.31e-03 |
Pos logit contributions | |
arl | +0.119 |
iren | +0.111 |
its | +0.099 |
izes | +0.097 |
irens | +0.096 |
Neg logit contributions | |
favourite | -0.076 |
companion | -0.070 |
favorite | -0.068 |
found | -0.068 |
mate | -0.067 |
Token | a |
Token ID | 257 |
Feature activation | +2.55e-02 |
Pos logit contributions | |
one | +0.008 |
favourite | +0.007 |
favorite | +0.007 |
three | +0.007 |
3 | +0.006 |
Neg logit contributions | |
arl | -0.017 |
iren | -0.015 |
izes | -0.013 |
irens | -0.013 |
ister | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
favourite | +0.120 |
companion | +0.114 |
found | +0.108 |
3 | +0.107 |
mate | +0.107 |
Neg logit contributions | |
arl | -0.157 |
iren | -0.146 |
its | -0.134 |
izes | -0.127 |
yourselves | -0.125 |
Token | because |
Token ID | 780 |
Feature activation | -2.77e-02 |
Pos logit contributions | |
favourite | +0.027 |
companion | +0.026 |
mate | +0.024 |
found | +0.023 |
3 | +0.023 |
Neg logit contributions | |
arl | -0.046 |
iren | -0.045 |
its | -0.040 |
irens | -0.039 |
yourselves | -0.039 |
Token | they |
Token ID | 484 |
Feature activation | -6.05e-02 |
Pos logit contributions | |
favourite | +0.039 |
companion | +0.036 |
found | +0.034 |
favorite | +0.034 |
mate | +0.034 |
Neg logit contributions | |
arl | -0.060 |
iren | -0.056 |
its | -0.050 |
izes | -0.049 |
irens | -0.048 |
Token | had |
Token ID | 550 |
Feature activation | -1.45e-03 |
Pos logit contributions | |
favourite | +0.063 |
companion | +0.060 |
3 | +0.052 |
favorite | +0.052 |
mate | +0.052 |
Neg logit contributions | |
arl | -0.145 |
iren | -0.140 |
its | -0.133 |
izes | -0.127 |
eling | -0.124 |
Token | five |
Token ID | 1936 |
Feature activation | +8.57e-03 |
Pos logit contributions | |
favourite | +0.002 |
companion | +0.002 |
Joey | +0.002 |
3 | +0.002 |
Sally | +0.002 |
Neg logit contributions | |
arl | -0.003 |
iren | -0.003 |
its | -0.003 |
yourselves | -0.003 |
izes | -0.002 |
Token | points |
Token ID | 2173 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
arl | +0.018 |
iren | +0.017 |
izes | +0.015 |
its | +0.014 |
irens | +0.014 |
Neg logit contributions | |
favourite | -0.013 |
favorite | -0.012 |
found | -0.012 |
companion | -0.012 |
one | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | +3.03e-03 |
Pos logit contributions | |
favourite | +0.188 |
companion | +0.171 |
favorite | +0.167 |
found | +0.164 |
mate | +0.162 |
Neg logit contributions | |
arl | -0.371 |
iren | -0.348 |
its | -0.312 |
izes | -0.308 |
irens | -0.303 |
Token | looked |
Token ID | 3114 |
Feature activation | -5.47e-02 |
Pos logit contributions | |
arl | +0.007 |
iren | +0.006 |
its | +0.006 |
izes | +0.006 |
yourselves | +0.006 |
Neg logit contributions | |
favourite | -0.004 |
companion | -0.004 |
favorite | -0.003 |
found | -0.003 |
mate | -0.003 |
Token | shiny |
Token ID | 22441 |
Feature activation | +1.74e-03 |
Pos logit contributions | |
favourite | +0.065 |
favorite | +0.059 |
one | +0.056 |
found | +0.055 |
companion | +0.054 |
Neg logit contributions | |
arl | -0.139 |
iren | -0.135 |
its | -0.119 |
irens | -0.118 |
izes | -0.115 |
Token | . |
Token ID | 13 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
arl | +0.004 |
iren | +0.004 |
izes | +0.004 |
irens | +0.003 |
its | +0.003 |
Neg logit contributions | |
favourite | -0.002 |
favorite | -0.002 |
one | -0.002 |
companion | -0.002 |
found | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | -8.29e-03 |
Pos logit contributions | |
favourite | +0.026 |
companion | +0.025 |
mate | +0.024 |
found | +0.024 |
spotted | +0.023 |
Neg logit contributions | |
arl | -0.042 |
iren | -0.038 |
its | -0.034 |
yourselves | -0.032 |
irens | -0.032 |
Token | four |
Token ID | 1440 |
Feature activation | -1.95e-02 |
Pos logit contributions | |
favourite | +0.031 |
one | +0.027 |
companion | +0.027 |
3 | +0.027 |
mate | +0.027 |
Neg logit contributions | |
arl | -0.078 |
iren | -0.074 |
its | -0.068 |
izes | -0.067 |
irens | -0.065 |
Token | , |
Token ID | 11 |
Feature activation | -3.08e-02 |
Pos logit contributions | |
favourite | +0.024 |
one | +0.022 |
found | +0.022 |
3 | +0.021 |
favorite | +0.021 |
Neg logit contributions | |
arl | -0.049 |
iren | -0.045 |
its | -0.040 |
izes | -0.039 |
irens | -0.039 |
Token | five |
Token ID | 1936 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
favourite | +0.029 |
one | +0.026 |
3 | +0.026 |
companion | +0.025 |
found | +0.025 |
Neg logit contributions | |
arl | -0.083 |
iren | -0.078 |
its | -0.072 |
izes | -0.071 |
irens | -0.070 |
Token | . |
Token ID | 13 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
arl | +0.046 |
iren | +0.042 |
its | +0.036 |
izes | +0.036 |
irens | +0.036 |
Neg logit contributions | |
favourite | -0.020 |
one | -0.020 |
found | -0.019 |
favorite | -0.018 |
3 | -0.018 |
Token | Five |
Token ID | 10579 |
Feature activation | -6.06e-02 |
Pos logit contributions | |
favourite | +0.018 |
companion | +0.016 |
favorite | +0.016 |
found | +0.016 |
mate | +0.015 |
Neg logit contributions | |
arl | -0.022 |
iren | -0.020 |
its | -0.018 |
izes | -0.017 |
yourselves | -0.017 |
Token | points |
Token ID | 2173 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
favourite | +0.092 |
found | +0.084 |
favorite | +0.083 |
companion | +0.083 |
one | +0.083 |
Neg logit contributions | |
arl | -0.132 |
iren | -0.121 |
its | -0.104 |
izes | -0.104 |
irens | -0.102 |
Token | for |
Token ID | 329 |
Feature activation | +5.16e-03 |
Pos logit contributions | |
favourite | +0.220 |
one | +0.205 |
favorite | +0.201 |
companion | +0.196 |
mate | +0.195 |
Neg logit contributions | |
arl | -0.350 |
iren | -0.325 |
its | -0.284 |
izes | -0.282 |
irens | -0.280 |
Token | a |
Token ID | 257 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
arl | +0.011 |
iren | +0.010 |
its | +0.009 |
yourselves | +0.009 |
izes | +0.009 |
Neg logit contributions | |
favourite | -0.007 |
companion | -0.007 |
favorite | -0.006 |
found | -0.006 |
mate | -0.006 |
Token | star |
Token ID | 3491 |
Feature activation | -3.09e-02 |
Pos logit contributions | |
favourite | +0.019 |
companion | +0.018 |
found | +0.017 |
favorite | +0.017 |
mate | +0.017 |
Neg logit contributions | |
arl | -0.031 |
iren | -0.029 |
its | -0.026 |
izes | -0.026 |
irens | -0.025 |
Token | !" |
Token ID | 2474 |
Feature activation | -9.14e-03 |
Pos logit contributions | |
favourite | +0.050 |
one | +0.048 |
mate | +0.047 |
favorite | +0.047 |
companion | +0.047 |
Neg logit contributions | |
arl | -0.064 |
iren | -0.058 |
its | -0.052 |
irens | -0.051 |
isin | -0.049 |
Token |
|
Token ID | 198 |
Feature activation | -1.21e-04 |
Pos logit contributions | |
favourite | +0.011 |
companion | +0.010 |
favorite | +0.010 |
found | +0.009 |
mate | +0.009 |
Neg logit contributions | |
arl | -0.022 |
iren | -0.020 |
its | -0.018 |
izes | -0.018 |
irens | -0.018 |
Token | green |
Token ID | 4077 |
Feature activation | -3.79e-02 |
Pos logit contributions | |
companion | +0.021 |
favourite | +0.021 |
mate | +0.019 |
found | +0.019 |
favorite | +0.019 |
Neg logit contributions | |
arl | -0.035 |
iren | -0.032 |
its | -0.029 |
yourselves | -0.028 |
izes | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | +2.28e-02 |
Pos logit contributions | |
favourite | +0.056 |
companion | +0.051 |
favorite | +0.051 |
found | +0.050 |
one | +0.049 |
Neg logit contributions | |
arl | -0.081 |
iren | -0.075 |
its | -0.067 |
izes | -0.066 |
irens | -0.064 |
Token | had |
Token ID | 550 |
Feature activation | -2.81e-03 |
Pos logit contributions | |
arl | +0.054 |
iren | +0.051 |
its | +0.046 |
izes | +0.045 |
yourselves | +0.045 |
Neg logit contributions | |
favourite | -0.027 |
companion | -0.026 |
mate | -0.023 |
favorite | -0.023 |
found | -0.023 |
Token | many |
Token ID | 867 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
companion | +0.004 |
favourite | +0.004 |
Joey | +0.003 |
mate | +0.003 |
favorite | +0.003 |
Neg logit contributions | |
arl | -0.006 |
iren | -0.005 |
its | -0.005 |
yourselves | -0.005 |
irens | -0.005 |
Token | sharp |
Token ID | 7786 |
Feature activation | +1.04e-03 |
Pos logit contributions | |
arl | +0.035 |
iren | +0.032 |
its | +0.029 |
izes | +0.028 |
yourselves | +0.028 |
Neg logit contributions | |
favourite | -0.027 |
companion | -0.026 |
mate | -0.025 |
favorite | -0.024 |
found | -0.024 |
Token | points |
Token ID | 2173 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
arl | +0.002 |
iren | +0.002 |
izes | +0.002 |
its | +0.002 |
irens | +0.002 |
Neg logit contributions | |
favourite | -0.002 |
companion | -0.002 |
found | -0.002 |
favorite | -0.002 |
one | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
favourite | +0.219 |
companion | +0.202 |
favorite | +0.198 |
found | +0.198 |
mate | +0.194 |
Neg logit contributions | |
arl | -0.334 |
iren | -0.312 |
its | -0.275 |
izes | -0.275 |
irens | -0.268 |
Token | Mia |
Token ID | 32189 |
Feature activation | -3.27e-02 |
Pos logit contributions | |
favourite | +0.015 |
companion | +0.015 |
favorite | +0.013 |
mate | +0.013 |
found | +0.013 |
Neg logit contributions | |
arl | -0.024 |
iren | -0.023 |
its | -0.020 |
irens | -0.020 |
yourselves | -0.020 |
Token | said |
Token ID | 531 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
companion | +0.035 |
favourite | +0.035 |
mate | +0.032 |
Sally | +0.029 |
Joey | +0.029 |
Neg logit contributions | |
arl | -0.074 |
iren | -0.073 |
its | -0.071 |
izes | -0.068 |
falls | -0.067 |
Token | , |
Token ID | 11 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
favourite | +0.055 |
companion | +0.052 |
favorite | +0.049 |
found | +0.048 |
mate | +0.048 |
Neg logit contributions | |
arl | -0.085 |
iren | -0.079 |
its | -0.071 |
izes | -0.070 |
yourselves | -0.068 |
Token | " |
Token ID | 366 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
favourite | +0.018 |
companion | +0.016 |
found | +0.015 |
favorite | +0.015 |
3 | +0.015 |
Neg logit contributions | |
arl | -0.050 |
iren | -0.048 |
its | -0.044 |
izes | -0.043 |
irens | -0.042 |
Token | them |
Token ID | 606 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
favourite | +0.184 |
companion | +0.172 |
favorite | +0.166 |
found | +0.165 |
mate | +0.163 |
Neg logit contributions | |
arl | -0.249 |
iren | -0.230 |
its | -0.205 |
izes | -0.200 |
irens | -0.196 |
Token | a |
Token ID | 257 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
favourite | +0.147 |
companion | +0.144 |
mate | +0.134 |
found | +0.133 |
3 | +0.133 |
Neg logit contributions | |
arl | -0.202 |
iren | -0.191 |
its | -0.172 |
izes | -0.167 |
irens | -0.167 |
Token | map |
Token ID | 3975 |
Feature activation | +3.58e-03 |
Pos logit contributions | |
favourite | +0.029 |
companion | +0.027 |
found | +0.026 |
favorite | +0.026 |
mate | +0.025 |
Neg logit contributions | |
arl | -0.049 |
iren | -0.045 |
its | -0.041 |
izes | -0.040 |
irens | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | -3.05e-03 |
Pos logit contributions | |
arl | +0.008 |
iren | +0.007 |
its | +0.006 |
izes | +0.006 |
irens | +0.006 |
Neg logit contributions | |
favourite | -0.005 |
companion | -0.005 |
found | -0.005 |
favorite | -0.005 |
one | -0.005 |
Token | She |
Token ID | 1375 |
Feature activation | +5.96e-02 |
Pos logit contributions | |
favourite | +0.004 |
companion | +0.004 |
mate | +0.004 |
found | +0.004 |
favorite | +0.004 |
Neg logit contributions | |
arl | -0.006 |
iren | -0.006 |
its | -0.005 |
irens | -0.005 |
yourselves | -0.005 |
Token | points |
Token ID | 2173 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
arl | +0.089 |
iren | +0.083 |
its | +0.073 |
izes | +0.070 |
irens | +0.066 |
Neg logit contributions | |
favourite | -0.118 |
companion | -0.116 |
Sally | -0.109 |
Joey | -0.109 |
favorite | -0.109 |
Token | to |
Token ID | 284 |
Feature activation | -2.11e-02 |
Pos logit contributions | |
favourite | +0.187 |
companion | +0.174 |
favorite | +0.163 |
found | +0.163 |
mate | +0.163 |
Neg logit contributions | |
arl | -0.352 |
iren | -0.330 |
its | -0.299 |
izes | -0.293 |
irens | -0.288 |
Token | different |
Token ID | 1180 |
Feature activation | -6.05e-02 |
Pos logit contributions | |
favourite | +0.029 |
companion | +0.028 |
Joey | +0.026 |
favorite | +0.025 |
mate | +0.024 |
Neg logit contributions | |
arl | -0.042 |
iren | -0.039 |
yourselves | -0.037 |
its | -0.036 |
irens | -0.035 |
Token | nations |
Token ID | 7027 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
favourite | +0.105 |
companion | +0.100 |
favorite | +0.095 |
mate | +0.094 |
Sally | +0.094 |
Neg logit contributions | |
arl | -0.105 |
iren | -0.098 |
its | -0.088 |
yourselves | -0.086 |
izes | -0.086 |
Token | and |
Token ID | 290 |
Feature activation | +3.96e-03 |
Pos logit contributions | |
favourite | +0.025 |
companion | +0.023 |
found | +0.022 |
favorite | +0.022 |
mate | +0.022 |
Neg logit contributions | |
arl | -0.037 |
iren | -0.035 |
its | -0.031 |
izes | -0.030 |
irens | -0.030 |
Token | tells |
Token ID | 4952 |
Feature activation | -4.31e-02 |
Pos logit contributions | |
arl | +0.007 |
iren | +0.007 |
its | +0.006 |
izes | +0.006 |
irens | +0.006 |
Neg logit contributions | |
favourite | -0.007 |
companion | -0.006 |
favorite | -0.006 |
found | -0.006 |
mate | -0.006 |
Token | one |
Token ID | 530 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
arl | +0.029 |
iren | +0.026 |
its | +0.023 |
irens | +0.023 |
izes | +0.023 |
Neg logit contributions | |
favourite | -0.021 |
one | -0.020 |
favorite | -0.020 |
3 | -0.020 |
companion | -0.019 |
Token | who |
Token ID | 508 |
Feature activation | -6.82e-02 |
Pos logit contributions | |
favourite | +0.138 |
companion | +0.137 |
Joey | +0.132 |
Gab | +0.130 |
mate | +0.128 |
Neg logit contributions | |
iren | -0.217 |
izes | -0.207 |
arl | -0.205 |
falls | -0.200 |
its | -0.195 |
Token | got |
Token ID | 1392 |
Feature activation | -3.98e-03 |
Pos logit contributions | |
favourite | +0.084 |
companion | +0.074 |
mate | +0.073 |
favorite | +0.071 |
Joey | +0.065 |
Neg logit contributions | |
iren | -0.149 |
arl | -0.143 |
its | -0.134 |
izes | -0.130 |
ister | -0.127 |
Token | the |
Token ID | 262 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
favourite | +0.005 |
companion | +0.005 |
favorite | +0.005 |
found | +0.005 |
3 | +0.005 |
Neg logit contributions | |
arl | -0.009 |
iren | -0.009 |
its | -0.008 |
izes | -0.008 |
irens | -0.007 |
Token | most |
Token ID | 749 |
Feature activation | -6.85e-02 |
Pos logit contributions | |
arl | +0.036 |
iren | +0.034 |
its | +0.030 |
izes | +0.030 |
irens | +0.029 |
Neg logit contributions | |
favourite | -0.020 |
companion | -0.019 |
favorite | -0.018 |
found | -0.018 |
mate | -0.018 |
Token | points |
Token ID | 2173 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
favourite | +0.091 |
companion | +0.082 |
favorite | +0.081 |
found | +0.080 |
3 | +0.078 |
Neg logit contributions | |
arl | -0.163 |
iren | -0.150 |
its | -0.133 |
izes | -0.131 |
irens | -0.130 |
Token | would |
Token ID | 561 |
Feature activation | +5.48e-02 |
Pos logit contributions | |
favourite | +0.223 |
companion | +0.209 |
favorite | +0.199 |
found | +0.196 |
mate | +0.195 |
Neg logit contributions | |
arl | -0.311 |
iren | -0.291 |
its | -0.262 |
izes | -0.256 |
yourselves | -0.250 |
Token | win |
Token ID | 1592 |
Feature activation | +3.76e-03 |
Pos logit contributions | |
arl | +0.124 |
iren | +0.116 |
its | +0.104 |
izes | +0.102 |
yourselves | +0.100 |
Neg logit contributions | |
favourite | -0.072 |
companion | -0.067 |
favorite | -0.064 |
found | -0.063 |
mate | -0.062 |
Token | the |
Token ID | 262 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
arl | +0.009 |
iren | +0.008 |
its | +0.007 |
izes | +0.007 |
isin | +0.007 |
Neg logit contributions | |
favourite | -0.005 |
found | -0.005 |
one | -0.005 |
companion | -0.005 |
favorite | -0.005 |
Token | game |
Token ID | 983 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
arl | +0.028 |
iren | +0.026 |
its | +0.023 |
izes | +0.023 |
irens | +0.022 |
Neg logit contributions | |
favourite | -0.015 |
companion | -0.014 |
favorite | -0.013 |
found | -0.013 |
mate | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
favourite | +0.018 |
companion | +0.017 |
favorite | +0.016 |
3 | +0.016 |
found | +0.016 |
Neg logit contributions | |
arl | -0.023 |
iren | -0.022 |
its | -0.019 |
izes | -0.019 |
irens | -0.018 |
Token | green |
Token ID | 4077 |
Feature activation | -4.18e-02 |
Pos logit contributions | |
arl | +0.026 |
iren | +0.024 |
its | +0.022 |
izes | +0.021 |
irens | +0.021 |
Neg logit contributions | |
favourite | -0.016 |
companion | -0.015 |
favorite | -0.014 |
found | -0.014 |
mate | -0.014 |
Token | plant |
Token ID | 4618 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
favourite | +0.055 |
companion | +0.051 |
favorite | +0.049 |
found | +0.049 |
mate | +0.048 |
Neg logit contributions | |
arl | -0.095 |
iren | -0.088 |
its | -0.080 |
izes | -0.078 |
irens | -0.077 |
Token | with |
Token ID | 351 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
favourite | +0.031 |
companion | +0.030 |
favorite | +0.028 |
mate | +0.028 |
3 | +0.028 |
Neg logit contributions | |
arl | -0.037 |
iren | -0.034 |
its | -0.030 |
izes | -0.029 |
irens | -0.028 |
Token | many |
Token ID | 867 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
companion | +0.161 |
favourite | +0.156 |
mate | +0.145 |
favorite | +0.136 |
visitor | +0.136 |
Neg logit contributions | |
arl | -0.240 |
iren | -0.223 |
its | -0.210 |
yourselves | -0.203 |
irens | -0.196 |
Token | sharp |
Token ID | 7786 |
Feature activation | +1.54e-03 |
Pos logit contributions | |
arl | +0.031 |
iren | +0.028 |
its | +0.026 |
izes | +0.025 |
irens | +0.024 |
Neg logit contributions | |
favourite | -0.022 |
companion | -0.021 |
favorite | -0.020 |
mate | -0.020 |
found | -0.020 |
Token | points |
Token ID | 2173 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
arl | +0.003 |
iren | +0.003 |
its | +0.002 |
izes | +0.002 |
irens | +0.002 |
Neg logit contributions | |
favourite | -0.002 |
companion | -0.002 |
found | -0.002 |
favorite | -0.002 |
one | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -2.88e-02 |
Pos logit contributions | |
favourite | +0.239 |
companion | +0.225 |
favorite | +0.215 |
found | +0.211 |
3 | +0.211 |
Neg logit contributions | |
arl | -0.303 |
iren | -0.278 |
its | -0.253 |
yourselves | -0.239 |
izes | -0.238 |
Token | Sara |
Token ID | 24799 |
Feature activation | -2.95e-02 |
Pos logit contributions | |
companion | +0.048 |
favourite | +0.046 |
mate | +0.044 |
visitor | +0.043 |
found | +0.041 |
Neg logit contributions | |
arl | -0.054 |
iren | -0.050 |
its | -0.045 |
yourselves | -0.045 |
irens | -0.043 |
Token | wanted |
Token ID | 2227 |
Feature activation | -4.07e-03 |
Pos logit contributions | |
companion | +0.040 |
mate | +0.038 |
favourite | +0.037 |
visitor | +0.033 |
Joey | +0.032 |
Neg logit contributions | |
its | -0.056 |
arl | -0.055 |
iren | -0.055 |
falls | -0.054 |
izes | -0.051 |
Token | to |
Token ID | 284 |
Feature activation | +1.30e-03 |
Pos logit contributions | |
companion | +0.007 |
favourite | +0.007 |
mate | +0.007 |
favourites | +0.007 |
youngest | +0.006 |
Neg logit contributions | |
iren | -0.006 |
yourselves | -0.005 |
arl | -0.005 |
irens | -0.005 |
its | -0.005 |
Token | touch |
Token ID | 3638 |
Feature activation | -4.77e-02 |
Pos logit contributions | |
arl | +0.002 |
iren | +0.002 |
its | +0.002 |
bleed | +0.002 |
yourselves | +0.002 |
Neg logit contributions | |
favourite | -0.002 |
companion | -0.002 |
mate | -0.002 |
favorite | -0.002 |
3 | -0.002 |
Token | are |
Token ID | 389 |
Feature activation | -7.74e-03 |
Pos logit contributions | |
favourite | +0.119 |
companion | +0.113 |
favorite | +0.109 |
mate | +0.108 |
3 | +0.107 |
Neg logit contributions | |
arl | -0.118 |
iren | -0.110 |
its | -0.099 |
izes | -0.094 |
yourselves | -0.092 |
Token | many |
Token ID | 867 |
Feature activation | +3.35e-02 |
Pos logit contributions | |
companion | +0.013 |
favourite | +0.012 |
visitor | +0.011 |
spotted | +0.011 |
search | +0.011 |
Neg logit contributions | |
arl | -0.013 |
iren | -0.013 |
yourselves | -0.013 |
its | -0.011 |
irens | -0.011 |
Token | boats |
Token ID | 16659 |
Feature activation | -3.04e-02 |
Pos logit contributions | |
arl | +0.084 |
iren | +0.077 |
izes | +0.069 |
its | +0.068 |
irens | +0.067 |
Neg logit contributions | |
favourite | -0.040 |
favorite | -0.036 |
found | -0.035 |
one | -0.034 |
companion | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -5.80e-03 |
Pos logit contributions | |
favourite | +0.042 |
found | +0.039 |
companion | +0.038 |
favorite | +0.038 |
mate | +0.037 |
Neg logit contributions | |
arl | -0.067 |
iren | -0.062 |
its | -0.056 |
izes | -0.055 |
irens | -0.054 |
Token | She |
Token ID | 1375 |
Feature activation | +5.91e-02 |
Pos logit contributions | |
companion | +0.009 |
favourite | +0.009 |
mate | +0.008 |
found | +0.008 |
searched | +0.008 |
Neg logit contributions | |
arl | -0.012 |
iren | -0.011 |
its | -0.010 |
yourselves | -0.010 |
irens | -0.009 |
Token | points |
Token ID | 2173 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
arl | +0.100 |
iren | +0.091 |
its | +0.078 |
izes | +0.076 |
irens | +0.074 |
Neg logit contributions | |
favourite | -0.113 |
companion | -0.107 |
favorite | -0.104 |
found | -0.103 |
mate | -0.102 |
Token | to |
Token ID | 284 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
favourite | +0.193 |
companion | +0.185 |
mate | +0.171 |
found | +0.168 |
favorite | +0.166 |
Neg logit contributions | |
arl | -0.332 |
iren | -0.310 |
its | -0.279 |
yourselves | -0.273 |
irens | -0.270 |
Token | a |
Token ID | 257 |
Feature activation | -3.68e-03 |
Pos logit contributions | |
companion | +0.031 |
favourite | +0.030 |
spotted | +0.027 |
mate | +0.027 |
mattered | +0.027 |
Neg logit contributions | |
arl | -0.034 |
yourselves | -0.032 |
iren | -0.032 |
its | -0.029 |
irens | -0.028 |
Token | big |
Token ID | 1263 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
favourite | +0.005 |
companion | +0.004 |
favorite | +0.004 |
found | +0.004 |
3 | +0.004 |
Neg logit contributions | |
arl | -0.009 |
iren | -0.008 |
its | -0.007 |
izes | -0.007 |
irens | -0.007 |
Token | boat |
Token ID | 8848 |
Feature activation | -2.82e-02 |
Pos logit contributions | |
arl | +0.025 |
iren | +0.023 |
its | +0.021 |
izes | +0.020 |
irens | +0.020 |
Neg logit contributions | |
favourite | -0.013 |
companion | -0.011 |
favorite | -0.011 |
found | -0.011 |
mate | -0.011 |
Token | with |
Token ID | 351 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
favourite | +0.043 |
companion | +0.040 |
favorite | +0.038 |
found | +0.038 |
3 | +0.037 |
Neg logit contributions | |
arl | -0.059 |
iren | -0.055 |
its | -0.048 |
izes | -0.047 |
irens | -0.047 |
Token | and |
Token ID | 290 |
Feature activation | +1.67e-02 |
Pos logit contributions | |
favourite | +0.021 |
companion | +0.019 |
found | +0.019 |
favorite | +0.019 |
3 | +0.019 |
Neg logit contributions | |
arl | -0.043 |
iren | -0.040 |
its | -0.037 |
izes | -0.036 |
irens | -0.035 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
arl | +0.034 |
iren | +0.031 |
izes | +0.029 |
yourselves | +0.028 |
isin | +0.028 |
Neg logit contributions | |
favourite | -0.023 |
companion | -0.022 |
mate | -0.020 |
favorite | -0.020 |
Joey | -0.019 |
Token | scored |
Token ID | 7781 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
favourite | +0.014 |
companion | +0.013 |
mate | +0.012 |
Joey | +0.012 |
favorite | +0.012 |
Neg logit contributions | |
arl | -0.031 |
iren | -0.031 |
izes | -0.028 |
its | -0.028 |
falls | -0.027 |
Token | lots |
Token ID | 6041 |
Feature activation | +2.23e-02 |
Pos logit contributions | |
favourite | +0.032 |
favorite | +0.029 |
found | +0.029 |
companion | +0.028 |
one | +0.028 |
Neg logit contributions | |
arl | -0.077 |
iren | -0.072 |
its | -0.064 |
izes | -0.064 |
irens | -0.063 |
Token | of |
Token ID | 286 |
Feature activation | -3.29e-02 |
Pos logit contributions | |
arl | +0.059 |
izes | +0.054 |
iren | +0.054 |
bleed | +0.053 |
eling | +0.052 |
Neg logit contributions | |
one | -0.018 |
favourite | -0.016 |
favorite | -0.016 |
three | -0.014 |
companion | -0.014 |
Token | points |
Token ID | 2173 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
favourite | +0.052 |
companion | +0.048 |
favorite | +0.047 |
found | +0.046 |
3 | +0.045 |
Neg logit contributions | |
arl | -0.068 |
iren | -0.063 |
its | -0.055 |
izes | -0.055 |
irens | -0.053 |
Token | by |
Token ID | 416 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
favourite | +0.194 |
one | +0.180 |
favorite | +0.178 |
companion | +0.176 |
found | +0.172 |
Neg logit contributions | |
arl | -0.356 |
iren | -0.340 |
izes | -0.305 |
isin | -0.298 |
its | -0.297 |
Token | finding |
Token ID | 4917 |
Feature activation | -6.33e-02 |
Pos logit contributions | |
favourite | +0.024 |
companion | +0.022 |
favorite | +0.021 |
one | +0.021 |
found | +0.021 |
Neg logit contributions | |
arl | -0.040 |
iren | -0.037 |
izes | -0.033 |
its | -0.032 |
irens | -0.031 |
Token | her |
Token ID | 607 |
Feature activation | -6.73e-03 |
Pos logit contributions | |
favourite | +0.095 |
companion | +0.088 |
favorite | +0.086 |
found | +0.085 |
mate | +0.084 |
Neg logit contributions | |
arl | -0.134 |
iren | -0.124 |
its | -0.110 |
izes | -0.108 |
irens | -0.106 |
Token | mom |
Token ID | 1995 |
Feature activation | +3.90e-03 |
Pos logit contributions | |
favourite | +0.008 |
favorite | +0.007 |
companion | +0.007 |
found | +0.007 |
3 | +0.007 |
Neg logit contributions | |
arl | -0.016 |
iren | -0.016 |
its | -0.014 |
izes | -0.014 |
irens | -0.014 |
Token | in |
Token ID | 287 |
Feature activation | -2.71e-02 |
Pos logit contributions | |
arl | +0.009 |
iren | +0.009 |
its | +0.008 |
izes | +0.008 |
irens | +0.008 |
Neg logit contributions | |
favourite | -0.005 |
companion | -0.004 |
favorite | -0.004 |
mate | -0.004 |
found | -0.004 |
Token | is |
Token ID | 318 |
Feature activation | -2.51e-02 |
Pos logit contributions | |
favourite | +0.032 |
found | +0.030 |
one | +0.030 |
favorite | +0.029 |
searched | +0.028 |
Neg logit contributions | |
arl | -0.054 |
iren | -0.048 |
irens | -0.042 |
ister | -0.040 |
yourselves | -0.040 |
Token | big |
Token ID | 1263 |
Feature activation | +3.90e-02 |
Pos logit contributions | |
favourite | +0.032 |
favorite | +0.029 |
found | +0.029 |
one | +0.029 |
3 | +0.026 |
Neg logit contributions | |
arl | -0.064 |
iren | -0.060 |
irens | -0.053 |
izes | -0.052 |
its | -0.052 |
Token | and |
Token ID | 290 |
Feature activation | +2.93e-02 |
Pos logit contributions | |
arl | +0.093 |
iren | +0.085 |
its | +0.078 |
izes | +0.076 |
irens | +0.075 |
Neg logit contributions | |
favourite | -0.050 |
one | -0.048 |
companion | -0.047 |
found | -0.047 |
favorite | -0.046 |
Token | has |
Token ID | 468 |
Feature activation | +2.14e-03 |
Pos logit contributions | |
arl | +0.073 |
iren | +0.069 |
irens | +0.062 |
its | +0.061 |
ister | +0.058 |
Neg logit contributions | |
favourite | -0.037 |
favorite | -0.034 |
found | -0.034 |
one | -0.033 |
3 | -0.031 |
Token | four |
Token ID | 1440 |
Feature activation | -1.61e-02 |
Pos logit contributions | |
arl | +0.005 |
iren | +0.005 |
izes | +0.005 |
its | +0.005 |
irens | +0.004 |
Neg logit contributions | |
favourite | -0.002 |
favorite | -0.002 |
found | -0.002 |
companion | -0.002 |
one | -0.002 |
Token | points |
Token ID | 2173 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
favourite | +0.024 |
one | +0.022 |
favorite | +0.021 |
found | +0.021 |
3 | +0.020 |
Neg logit contributions | |
arl | -0.038 |
iren | -0.034 |
izes | -0.031 |
ister | -0.029 |
irens | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
favourite | +0.189 |
one | +0.177 |
favorite | +0.173 |
found | +0.172 |
companion | +0.170 |
Neg logit contributions | |
arl | -0.356 |
iren | -0.337 |
izes | -0.304 |
isin | -0.294 |
irens | -0.294 |
Token | Tom |
Token ID | 4186 |
Feature activation | -6.46e-03 |
Pos logit contributions | |
arl | +0.042 |
iren | +0.040 |
izes | +0.037 |
its | +0.036 |
eling | +0.035 |
Neg logit contributions | |
favourite | -0.021 |
Sally | -0.019 |
favorite | -0.019 |
companion | -0.019 |
one | -0.018 |
Token | 's |
Token ID | 338 |
Feature activation | -8.94e-02 |
Pos logit contributions | |
favourite | +0.010 |
found | +0.010 |
one | +0.010 |
favorite | +0.010 |
searched | +0.009 |
Neg logit contributions | |
arl | -0.014 |
iren | -0.012 |
ister | -0.011 |
irens | -0.011 |
yourselves | -0.011 |
Token | snow |
Token ID | 6729 |
Feature activation | +4.64e-02 |
Pos logit contributions | |
favourite | +0.118 |
favorite | +0.108 |
companion | +0.100 |
one | +0.100 |
3 | +0.098 |
Neg logit contributions | |
arl | -0.220 |
iren | -0.208 |
irens | -0.187 |
izes | -0.185 |
its | -0.184 |
Token | flake |
Token ID | 47597 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
arl | +0.111 |
iren | +0.101 |
irens | +0.091 |
isin | +0.090 |
izes | +0.089 |
Neg logit contributions | |
favourite | -0.062 |
favorite | -0.059 |
one | -0.056 |
found | -0.056 |
companion | -0.053 |
Token | loud |
Token ID | 7812 |
Feature activation | +5.64e-02 |
Pos logit contributions | |
arl | +0.074 |
iren | +0.070 |
its | +0.061 |
irens | +0.061 |
yourselves | +0.059 |
Neg logit contributions | |
favourite | -0.050 |
favorite | -0.046 |
found | -0.045 |
companion | -0.044 |
Sally | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | +6.68e-03 |
Pos logit contributions | |
arl | +0.146 |
iren | +0.134 |
isin | +0.125 |
its | +0.125 |
izes | +0.120 |
Neg logit contributions | |
one | -0.063 |
favourite | -0.057 |
found | -0.057 |
favorite | -0.054 |
companion | -0.054 |
Token | Sam |
Token ID | 3409 |
Feature activation | -4.70e-03 |
Pos logit contributions | |
arl | +0.015 |
iren | +0.014 |
its | +0.012 |
irens | +0.012 |
yourselves | +0.012 |
Neg logit contributions | |
favourite | -0.009 |
companion | -0.008 |
found | -0.008 |
mate | -0.008 |
favorite | -0.008 |
Token | scores |
Token ID | 8198 |
Feature activation | -2.11e-02 |
Pos logit contributions | |
found | +0.007 |
favourite | +0.007 |
one | +0.006 |
favorite | +0.006 |
spotted | +0.006 |
Neg logit contributions | |
arl | -0.011 |
iren | -0.009 |
irens | -0.008 |
its | -0.008 |
isin | -0.008 |
Token | many |
Token ID | 867 |
Feature activation | +3.03e-02 |
Pos logit contributions | |
one | +0.025 |
found | +0.024 |
favourite | +0.022 |
favorite | +0.022 |
mate | +0.021 |
Neg logit contributions | |
arl | -0.055 |
iren | -0.051 |
isin | -0.046 |
ister | -0.046 |
irens | -0.045 |
Token | points |
Token ID | 2173 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
arl | +0.060 |
iren | +0.054 |
izes | +0.045 |
ister | +0.044 |
isin | +0.044 |
Neg logit contributions | |
favourite | -0.051 |
one | -0.049 |
favorite | -0.048 |
companion | -0.048 |
found | -0.047 |
Token | . |
Token ID | 13 |
Feature activation | +3.09e-03 |
Pos logit contributions | |
favourite | +0.196 |
companion | +0.183 |
favorite | +0.180 |
one | +0.179 |
found | +0.176 |
Neg logit contributions | |
arl | -0.347 |
iren | -0.326 |
izes | -0.288 |
its | -0.288 |
isin | -0.286 |
Token | He |
Token ID | 679 |
Feature activation | +5.40e-02 |
Pos logit contributions | |
arl | +0.008 |
iren | +0.007 |
its | +0.006 |
izes | +0.006 |
irens | +0.006 |
Neg logit contributions | |
favourite | -0.004 |
companion | -0.003 |
favorite | -0.003 |
found | -0.003 |
mate | -0.003 |
Token | is |
Token ID | 318 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
arl | +0.111 |
iren | +0.096 |
irens | +0.083 |
ister | +0.081 |
yourselves | +0.080 |
Neg logit contributions | |
favourite | -0.094 |
found | -0.092 |
one | -0.091 |
favorite | -0.089 |
mate | -0.083 |
Token | very |
Token ID | 845 |
Feature activation | +3.07e-02 |
Pos logit contributions | |
favourite | +0.032 |
found | +0.029 |
favorite | +0.029 |
companion | +0.029 |
one | +0.029 |
Neg logit contributions | |
arl | -0.053 |
iren | -0.049 |
its | -0.044 |
izes | -0.042 |
irens | -0.042 |
Token | good |
Token ID | 922 |
Feature activation | -9.38e-03 |
Pos logit contributions | |
arl | +0.067 |
iren | +0.063 |
irens | +0.056 |
its | +0.056 |
yourselves | +0.054 |
Neg logit contributions | |
favourite | -0.044 |
favorite | -0.042 |
found | -0.041 |
one | -0.040 |
3 | -0.039 |
Token | on |
Token ID | 319 |
Feature activation | -3.73e-02 |
Pos logit contributions | |
favourite | +0.024 |
found | +0.024 |
one | +0.023 |
companion | +0.023 |
favorite | +0.022 |
Neg logit contributions | |
arl | -0.043 |
iren | -0.040 |
its | -0.036 |
irens | -0.034 |
izes | -0.034 |
Token | the |
Token ID | 262 |
Feature activation | +1.99e-03 |
Pos logit contributions | |
one | +0.053 |
favourite | +0.049 |
favorite | +0.045 |
found | +0.045 |
Sally | +0.043 |
Neg logit contributions | |
arl | -0.092 |
iren | -0.085 |
izes | -0.077 |
its | -0.074 |
irens | -0.073 |
Token | cover |
Token ID | 3002 |
Feature activation | -4.02e-03 |
Pos logit contributions | |
arl | +0.004 |
iren | +0.004 |
its | +0.004 |
irens | +0.004 |
izes | +0.003 |
Neg logit contributions | |
favourite | -0.003 |
favorite | -0.003 |
one | -0.003 |
companion | -0.003 |
3 | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -8.23e-03 |
Pos logit contributions | |
favourite | +0.005 |
one | +0.005 |
found | +0.004 |
favorite | +0.004 |
companion | +0.004 |
Neg logit contributions | |
arl | -0.010 |
iren | -0.009 |
its | -0.008 |
irens | -0.008 |
izes | -0.008 |
Token | She |
Token ID | 1375 |
Feature activation | +5.68e-02 |
Pos logit contributions | |
favourite | +0.011 |
companion | +0.010 |
favorite | +0.009 |
found | +0.009 |
3 | +0.009 |
Neg logit contributions | |
arl | -0.019 |
iren | -0.018 |
its | -0.016 |
izes | -0.016 |
irens | -0.015 |
Token | points |
Token ID | 2173 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
arl | +0.118 |
iren | +0.107 |
irens | +0.090 |
its | +0.089 |
ister | +0.088 |
Neg logit contributions | |
favourite | -0.096 |
found | -0.091 |
favorite | -0.090 |
one | -0.088 |
spotted | -0.085 |
Token | to |
Token ID | 284 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
favourite | +0.186 |
companion | +0.172 |
favorite | +0.164 |
found | +0.163 |
mate | +0.160 |
Neg logit contributions | |
arl | -0.354 |
iren | -0.330 |
its | -0.299 |
izes | -0.293 |
irens | -0.288 |
Token | it |
Token ID | 340 |
Feature activation | -4.75e-02 |
Pos logit contributions | |
favourite | +0.024 |
companion | +0.022 |
favorite | +0.021 |
found | +0.021 |
mate | +0.021 |
Neg logit contributions | |
arl | -0.045 |
iren | -0.042 |
its | -0.038 |
izes | -0.037 |
irens | -0.037 |
Token | and |
Token ID | 290 |
Feature activation | +6.13e-03 |
Pos logit contributions | |
favourite | +0.069 |
companion | +0.063 |
favorite | +0.062 |
found | +0.061 |
one | +0.060 |
Neg logit contributions | |
arl | -0.105 |
iren | -0.097 |
its | -0.087 |
izes | -0.084 |
irens | -0.084 |
Token | says |
Token ID | 1139 |
Feature activation | -4.98e-02 |
Pos logit contributions | |
arl | +0.012 |
iren | +0.011 |
its | +0.009 |
irens | +0.009 |
ister | +0.009 |
Neg logit contributions | |
favourite | -0.011 |
found | -0.010 |
one | -0.010 |
favorite | -0.010 |
companion | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
favourite | +0.054 |
one | +0.049 |
favorite | +0.048 |
found | +0.047 |
Sally | +0.047 |
Neg logit contributions | |
arl | -0.127 |
iren | -0.121 |
its | -0.110 |
irens | -0.106 |
eling | -0.106 |
Token | cross |
Token ID | 3272 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
arl | +0.026 |
iren | +0.024 |
its | +0.022 |
irens | +0.022 |
izes | +0.022 |
Neg logit contributions | |
favourite | -0.015 |
companion | -0.014 |
3 | -0.014 |
favorite | -0.014 |
found | -0.014 |
Token | with |
Token ID | 351 |
Feature activation | -9.50e-02 |
Pos logit contributions | |
arl | +0.045 |
iren | +0.039 |
its | +0.036 |
yourselves | +0.035 |
izes | +0.035 |
Neg logit contributions | |
favourite | -0.025 |
one | -0.024 |
found | -0.023 |
companion | -0.023 |
3 | -0.022 |
Token | its |
Token ID | 663 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
favourite | +0.116 |
companion | +0.106 |
found | +0.102 |
favorite | +0.102 |
3 | +0.099 |
Neg logit contributions | |
arl | -0.228 |
iren | -0.212 |
its | -0.192 |
izes | -0.189 |
irens | -0.185 |
Token | two |
Token ID | 734 |
Feature activation | -7.95e-02 |
Pos logit contributions | |
favourite | +0.028 |
companion | +0.026 |
found | +0.025 |
favorite | +0.025 |
mate | +0.025 |
Neg logit contributions | |
arl | -0.043 |
iren | -0.040 |
its | -0.036 |
izes | -0.035 |
irens | -0.034 |
Token | shiny |
Token ID | 22441 |
Feature activation | -1.29e-02 |
Pos logit contributions | |
favourite | +0.119 |
companion | +0.110 |
found | +0.107 |
favorite | +0.106 |
3 | +0.105 |
Neg logit contributions | |
arl | -0.172 |
iren | -0.158 |
izes | -0.140 |
its | -0.138 |
yourselves | -0.137 |
Token | points |
Token ID | 2173 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
favourite | +0.020 |
companion | +0.018 |
favorite | +0.018 |
one | +0.018 |
found | +0.017 |
Neg logit contributions | |
arl | -0.028 |
iren | -0.026 |
izes | -0.023 |
its | -0.023 |
irens | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -2.51e-02 |
Pos logit contributions | |
favourite | +0.214 |
companion | +0.199 |
favorite | +0.192 |
found | +0.191 |
mate | +0.188 |
Neg logit contributions | |
arl | -0.324 |
iren | -0.301 |
its | -0.269 |
izes | -0.263 |
irens | -0.259 |
Token | The |
Token ID | 383 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
favourite | +0.030 |
companion | +0.028 |
favorite | +0.027 |
found | +0.026 |
mate | +0.026 |
Neg logit contributions | |
arl | -0.060 |
iren | -0.056 |
its | -0.050 |
irens | -0.049 |
izes | -0.049 |
Token | bear |
Token ID | 6842 |
Feature activation | +3.82e-03 |
Pos logit contributions | |
arl | +0.036 |
iren | +0.034 |
its | +0.031 |
izes | +0.031 |
yourselves | +0.030 |
Neg logit contributions | |
favourite | -0.015 |
3 | -0.013 |
companion | -0.013 |
favorite | -0.013 |
one | -0.012 |
Token | stopped |
Token ID | 5025 |
Feature activation | +4.16e-02 |
Pos logit contributions | |
arl | +0.009 |
iren | +0.009 |
its | +0.008 |
izes | +0.008 |
irens | +0.008 |
Neg logit contributions | |
favourite | -0.004 |
companion | -0.004 |
favorite | -0.004 |
mate | -0.003 |
found | -0.003 |
Token | grow |
Token ID | 1663 |
Feature activation | +8.56e-03 |
Pos logit contributions | |
arl | +0.088 |
iren | +0.081 |
its | +0.072 |
izes | +0.071 |
irens | +0.070 |
Neg logit contributions | |
favourite | -0.062 |
companion | -0.057 |
favorite | -0.056 |
found | -0.055 |
mate | -0.055 |
Token | her |
Token ID | 607 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
favourite | +0.040 |
companion | +0.039 |
mate | +0.037 |
favorite | +0.036 |
wandered | +0.033 |
Neg logit contributions | |
arl | -0.069 |
iren | -0.068 |
yourselves | -0.060 |
its | -0.060 |
irens | -0.059 |
Token | three |
Token ID | 1115 |
Feature activation | -5.45e-02 |
Pos logit contributions | |
favourite | +0.025 |
mate | +0.024 |
companion | +0.024 |
found | +0.023 |
searched | +0.022 |
Neg logit contributions | |
arl | -0.050 |
iren | -0.047 |
its | -0.043 |
yourselves | -0.041 |
izes | -0.041 |
Token | year |
Token ID | 614 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
favourite | +0.102 |
companion | +0.099 |
mate | +0.095 |
favorite | +0.094 |
found | +0.094 |
Neg logit contributions | |
arl | -0.089 |
iren | -0.085 |
its | -0.074 |
irens | -0.070 |
izes | -0.069 |
Token | old |
Token ID | 1468 |
Feature activation | -8.26e-04 |
Pos logit contributions | |
arl | +0.025 |
iren | +0.023 |
its | +0.022 |
yourselves | +0.021 |
izes | +0.021 |
Neg logit contributions | |
favourite | -0.014 |
3 | -0.013 |
favorite | -0.013 |
one | -0.012 |
found | -0.012 |
Token | daughter |
Token ID | 4957 |
Feature activation | -2.76e-02 |
Pos logit contributions | |
favourite | +0.001 |
mate | +0.001 |
companion | +0.001 |
favorite | +0.001 |
found | +0.001 |
Neg logit contributions | |
iren | -0.002 |
arl | -0.002 |
irens | -0.002 |
yourselves | -0.001 |
its | -0.001 |
Token | pointing |
Token ID | 10609 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
favourite | +0.043 |
mate | +0.042 |
favorite | +0.041 |
companion | +0.041 |
Joey | +0.036 |
Neg logit contributions | |
arl | -0.046 |
iren | -0.046 |
its | -0.040 |
irens | -0.039 |
rou | -0.038 |
Token | at |
Token ID | 379 |
Feature activation | -7.61e-02 |
Pos logit contributions | |
favourite | +0.219 |
mate | +0.217 |
companion | +0.212 |
picnic | +0.199 |
favorite | +0.199 |
Neg logit contributions | |
arl | -0.283 |
iren | -0.265 |
its | -0.230 |
izes | -0.229 |
yourselves | -0.228 |
Token | something |
Token ID | 1223 |
Feature activation | -4.36e-02 |
Pos logit contributions | |
favourite | +0.107 |
companion | +0.102 |
mate | +0.101 |
favorite | +0.097 |
found | +0.095 |
Neg logit contributions | |
arl | -0.159 |
iren | -0.148 |
its | -0.134 |
yourselves | -0.131 |
irens | -0.129 |
Token | on |
Token ID | 319 |
Feature activation | -2.53e-02 |
Pos logit contributions | |
favourite | +0.069 |
companion | +0.068 |
mate | +0.065 |
found | +0.063 |
favorite | +0.063 |
Neg logit contributions | |
arl | -0.083 |
iren | -0.078 |
its | -0.069 |
izes | -0.067 |
isin | -0.066 |
Token | the |
Token ID | 262 |
Feature activation | +7.58e-03 |
Pos logit contributions | |
favourite | +0.034 |
companion | +0.032 |
found | +0.031 |
one | +0.031 |
3 | +0.031 |
Neg logit contributions | |
arl | -0.057 |
iren | -0.053 |
its | -0.048 |
izes | -0.047 |
irens | -0.046 |
Token | floor |
Token ID | 4314 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
arl | +0.014 |
iren | +0.013 |
its | +0.011 |
izes | +0.011 |
yourselves | +0.011 |
Neg logit contributions | |
favourite | -0.013 |
companion | -0.012 |
found | -0.012 |
mate | -0.012 |
favorite | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | +3.34e-02 |
Pos logit contributions | |
favourite | +0.008 |
companion | +0.008 |
favorite | +0.007 |
found | +0.007 |
one | +0.007 |
Neg logit contributions | |
arl | -0.018 |
iren | -0.017 |
its | -0.015 |
izes | -0.015 |
irens | -0.015 |
Token | who |
Token ID | 508 |
Feature activation | -5.07e-02 |
Pos logit contributions | |
arl | +0.072 |
iren | +0.066 |
its | +0.059 |
izes | +0.057 |
irens | +0.056 |
Neg logit contributions | |
favourite | -0.050 |
companion | -0.046 |
found | -0.046 |
favorite | -0.045 |
one | -0.044 |
Token | had |
Token ID | 550 |
Feature activation | +2.79e-03 |
Pos logit contributions | |
favourite | +0.060 |
companion | +0.055 |
favorite | +0.053 |
mate | +0.051 |
3 | +0.050 |
Neg logit contributions | |
arl | -0.119 |
iren | -0.114 |
its | -0.104 |
izes | -0.102 |
irens | -0.099 |
Token | the |
Token ID | 262 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
arl | +0.006 |
iren | +0.006 |
its | +0.005 |
yourselves | +0.005 |
izes | +0.005 |
Neg logit contributions | |
favourite | -0.003 |
companion | -0.003 |
favorite | -0.003 |
mate | -0.003 |
3 | -0.003 |
Token | most |
Token ID | 749 |
Feature activation | -6.82e-02 |
Pos logit contributions | |
arl | +0.031 |
iren | +0.029 |
its | +0.026 |
izes | +0.026 |
yourselves | +0.025 |
Neg logit contributions | |
favourite | -0.016 |
companion | -0.015 |
found | -0.014 |
Sally | -0.014 |
mate | -0.014 |
Token | points |
Token ID | 2173 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
favourite | +0.090 |
companion | +0.081 |
favorite | +0.080 |
found | +0.078 |
3 | +0.077 |
Neg logit contributions | |
arl | -0.162 |
iren | -0.150 |
its | -0.133 |
izes | -0.132 |
irens | -0.130 |
Token | . |
Token ID | 13 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
favourite | +0.225 |
companion | +0.210 |
favorite | +0.203 |
found | +0.200 |
3 | +0.198 |
Neg logit contributions | |
arl | -0.304 |
iren | -0.283 |
its | -0.252 |
izes | -0.245 |
yourselves | -0.241 |
Token | John |
Token ID | 1757 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
favourite | +0.018 |
companion | +0.016 |
favorite | +0.015 |
found | +0.015 |
mate | +0.015 |
Neg logit contributions | |
arl | -0.039 |
iren | -0.036 |
its | -0.033 |
izes | -0.032 |
irens | -0.032 |
Token | was |
Token ID | 373 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
favourite | +0.025 |
companion | +0.023 |
found | +0.022 |
favorite | +0.022 |
mate | +0.021 |
Neg logit contributions | |
arl | -0.068 |
iren | -0.064 |
its | -0.058 |
izes | -0.057 |
irens | -0.057 |
Token | very |
Token ID | 845 |
Feature activation | +2.07e-02 |
Pos logit contributions | |
favourite | +0.022 |
companion | +0.021 |
favorite | +0.020 |
mate | +0.019 |
found | +0.019 |
Neg logit contributions | |
arl | -0.036 |
iren | -0.034 |
its | -0.030 |
izes | -0.030 |
irens | -0.029 |
Token | competitive |
Token ID | 7606 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
arl | +0.044 |
iren | +0.041 |
its | +0.036 |
izes | +0.035 |
irens | +0.035 |
Neg logit contributions | |
favourite | -0.031 |
companion | -0.029 |
favorite | -0.028 |
found | -0.028 |
3 | -0.028 |
Token | bird |
Token ID | 6512 |
Feature activation | -4.56e-02 |
Pos logit contributions | |
arl | +0.006 |
iren | +0.005 |
its | +0.005 |
izes | +0.005 |
irens | +0.005 |
Neg logit contributions | |
favourite | -0.003 |
companion | -0.003 |
favorite | -0.003 |
found | -0.003 |
3 | -0.003 |
Token | in |
Token ID | 287 |
Feature activation | -3.43e-02 |
Pos logit contributions | |
favourite | +0.058 |
found | +0.058 |
one | +0.057 |
companion | +0.053 |
spotted | +0.053 |
Neg logit contributions | |
arl | -0.108 |
iren | -0.097 |
its | -0.087 |
irens | -0.086 |
izes | -0.085 |
Token | a |
Token ID | 257 |
Feature activation | -6.11e-03 |
Pos logit contributions | |
favourite | +0.047 |
companion | +0.044 |
found | +0.042 |
favorite | +0.042 |
mate | +0.041 |
Neg logit contributions | |
arl | -0.077 |
iren | -0.071 |
its | -0.064 |
izes | -0.063 |
irens | -0.062 |
Token | tree |
Token ID | 5509 |
Feature activation | +2.39e-02 |
Pos logit contributions | |
favourite | +0.008 |
companion | +0.008 |
found | +0.008 |
favorite | +0.008 |
mate | +0.007 |
Neg logit contributions | |
arl | -0.014 |
iren | -0.013 |
its | -0.011 |
izes | -0.011 |
irens | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | +7.26e-03 |
Pos logit contributions | |
arl | +0.058 |
iren | +0.053 |
izes | +0.048 |
irens | +0.048 |
its | +0.048 |
Neg logit contributions | |
one | -0.028 |
found | -0.027 |
favourite | -0.027 |
search | -0.025 |
spotted | -0.025 |
Token | points |
Token ID | 2173 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
arl | +0.018 |
iren | +0.017 |
its | +0.015 |
irens | +0.015 |
izes | +0.015 |
Neg logit contributions | |
favourite | -0.009 |
found | -0.008 |
favorite | -0.008 |
one | -0.008 |
companion | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
favourite | +0.185 |
companion | +0.171 |
favorite | +0.159 |
found | +0.159 |
3 | +0.158 |
Neg logit contributions | |
arl | -0.346 |
iren | -0.321 |
its | -0.294 |
izes | -0.286 |
yourselves | -0.283 |
Token | it |
Token ID | 340 |
Feature activation | -5.02e-02 |
Pos logit contributions | |
favourite | +0.028 |
companion | +0.026 |
favorite | +0.024 |
mate | +0.024 |
favourites | +0.024 |
Neg logit contributions | |
arl | -0.034 |
iren | -0.032 |
yourselves | -0.030 |
its | -0.029 |
izes | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -4.41e-03 |
Pos logit contributions | |
favourite | +0.075 |
companion | +0.070 |
favorite | +0.068 |
found | +0.067 |
mate | +0.066 |
Neg logit contributions | |
arl | -0.106 |
iren | -0.099 |
its | -0.087 |
izes | -0.085 |
irens | -0.084 |
Token | " |
Token ID | 366 |
Feature activation | +9.62e-03 |
Pos logit contributions | |
favourite | +0.005 |
favorite | +0.004 |
companion | +0.004 |
found | +0.004 |
one | +0.004 |
Neg logit contributions | |
arl | -0.011 |
iren | -0.011 |
its | -0.010 |
izes | -0.010 |
irens | -0.009 |
Token | Look |
Token ID | 8567 |
Feature activation | -3.82e-02 |
Pos logit contributions | |
arl | +0.021 |
iren | +0.019 |
its | +0.017 |
izes | +0.017 |
irens | +0.017 |
Neg logit contributions | |
favourite | -0.013 |
companion | -0.013 |
found | -0.012 |
favorite | -0.012 |
3 | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
favourite | +0.061 |
companion | +0.057 |
favorite | +0.055 |
found | +0.055 |
mate | +0.054 |
Neg logit contributions | |
arl | -0.083 |
iren | -0.077 |
its | -0.068 |
izes | -0.067 |
irens | -0.065 |
Token | day |
Token ID | 1110 |
Feature activation | +4.17e-02 |
Pos logit contributions | |
arl | +0.028 |
iren | +0.026 |
its | +0.024 |
yourselves | +0.024 |
izes | +0.024 |
Neg logit contributions | |
favourite | -0.013 |
companion | -0.013 |
Sally | -0.012 |
found | -0.012 |
searched | -0.012 |
Token | before |
Token ID | 878 |
Feature activation | +6.12e-02 |
Pos logit contributions | |
arl | +0.071 |
iren | +0.068 |
its | +0.059 |
isin | +0.058 |
irens | +0.058 |
Neg logit contributions | |
favourite | -0.071 |
companion | -0.067 |
mate | -0.067 |
favorite | -0.066 |
3 | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
arl | +0.139 |
iren | +0.127 |
its | +0.114 |
izes | +0.113 |
isin | +0.109 |
Neg logit contributions | |
favourite | -0.085 |
companion | -0.078 |
one | -0.077 |
found | -0.076 |
Sally | -0.076 |
Token | She |
Token ID | 1375 |
Feature activation | +5.57e-02 |
Pos logit contributions | |
favourite | +0.020 |
mate | +0.020 |
companion | +0.019 |
favorite | +0.018 |
found | +0.018 |
Neg logit contributions | |
arl | -0.031 |
iren | -0.029 |
its | -0.025 |
irens | -0.025 |
yourselves | -0.025 |
Token | couldn |
Token ID | 3521 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
iren | +0.112 |
arl | +0.111 |
izes | +0.107 |
its | +0.103 |
eling | +0.099 |
Neg logit contributions | |
favourite | -0.068 |
companion | -0.064 |
Sally | -0.062 |
favorite | -0.062 |
mate | -0.059 |
Token | 't |
Token ID | 470 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
arl | +0.269 |
iren | +0.223 |
izes | +0.197 |
its | +0.192 |
yourselves | +0.191 |
Neg logit contributions | |
favourite | -0.311 |
one | -0.304 |
found | -0.297 |
Sally | -0.296 |
3 | -0.290 |
Token | believe |
Token ID | 1975 |
Feature activation | -8.73e-02 |
Pos logit contributions | |
arl | +0.044 |
iren | +0.040 |
its | +0.039 |
bleed | +0.038 |
izes | +0.037 |
Neg logit contributions | |
favourite | -0.034 |
favorite | -0.031 |
companion | -0.031 |
Sally | -0.029 |
favourites | -0.029 |
Token | it |
Token ID | 340 |
Feature activation | -3.26e-02 |
Pos logit contributions | |
favourite | +0.123 |
favorite | +0.111 |
companion | +0.111 |
Sally | +0.111 |
one | +0.110 |
Neg logit contributions | |
arl | -0.195 |
iren | -0.187 |
its | -0.163 |
irens | -0.158 |
izes | -0.157 |
Token | - |
Token ID | 532 |
Feature activation | -3.20e-02 |
Pos logit contributions | |
favourite | +0.048 |
companion | +0.046 |
mate | +0.045 |
favorite | +0.044 |
3 | +0.043 |
Neg logit contributions | |
arl | -0.065 |
iren | -0.061 |
its | -0.056 |
izes | -0.055 |
isin | -0.053 |
Token | the |
Token ID | 262 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
favourite | +0.051 |
companion | +0.048 |
favorite | +0.046 |
mate | +0.046 |
found | +0.046 |
Neg logit contributions | |
arl | -0.063 |
iren | -0.058 |
its | -0.052 |
izes | -0.051 |
yourselves | -0.050 |
Token | of |
Token ID | 286 |
Feature activation | +6.99e-04 |
Pos logit contributions | |
arl | +0.038 |
iren | +0.035 |
its | +0.032 |
irens | +0.032 |
izes | +0.031 |
Neg logit contributions | |
favourite | -0.023 |
one | -0.023 |
Sally | -0.021 |
found | -0.021 |
mate | -0.021 |
Token | sand |
Token ID | 6450 |
Feature activation | +8.87e-03 |
Pos logit contributions | |
arl | +0.002 |
iren | +0.001 |
its | +0.001 |
izes | +0.001 |
irens | +0.001 |
Neg logit contributions | |
favourite | -0.001 |
companion | -0.001 |
favorite | -0.001 |
found | -0.001 |
one | -0.001 |
Token | ". |
Token ID | 1911 |
Feature activation | -2.35e-02 |
Pos logit contributions | |
arl | +0.018 |
iren | +0.017 |
its | +0.014 |
isin | +0.014 |
irens | +0.014 |
Neg logit contributions | |
favourite | -0.015 |
favorite | -0.014 |
one | -0.014 |
mate | -0.013 |
found | -0.013 |
Token | Tim |
Token ID | 5045 |
Feature activation | -2.99e-02 |
Pos logit contributions | |
favourite | +0.027 |
mate | +0.026 |
companion | +0.025 |
favorite | +0.024 |
found | +0.022 |
Neg logit contributions | |
arl | -0.056 |
iren | -0.054 |
its | -0.048 |
irens | -0.047 |
yourselves | -0.046 |
Token | my |
Token ID | 1820 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
favourite | +0.039 |
companion | +0.036 |
favorite | +0.035 |
found | +0.035 |
mate | +0.034 |
Neg logit contributions | |
arl | -0.069 |
iren | -0.064 |
its | -0.057 |
izes | -0.056 |
irens | -0.055 |
Token | couldn |
Token ID | 3521 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
arl | +0.055 |
iren | +0.055 |
its | +0.051 |
izes | +0.051 |
isin | +0.048 |
Neg logit contributions | |
favourite | -0.030 |
companion | -0.029 |
mate | -0.028 |
Sally | -0.027 |
favorite | -0.026 |
Token | 't |
Token ID | 470 |
Feature activation | +1.80e-02 |
Pos logit contributions | |
arl | +0.274 |
iren | +0.219 |
izes | +0.197 |
yourselves | +0.194 |
its | +0.190 |
Neg logit contributions | |
one | -0.295 |
favourite | -0.293 |
Sally | -0.280 |
found | -0.276 |
Joey | -0.275 |
Token | wait |
Token ID | 4043 |
Feature activation | -2.68e-02 |
Pos logit contributions | |
arl | +0.037 |
iren | +0.035 |
its | +0.032 |
izes | +0.031 |
isin | +0.031 |
Neg logit contributions | |
favourite | -0.025 |
companion | -0.023 |
favorite | -0.022 |
Sally | -0.021 |
spotted | -0.021 |
Token | to |
Token ID | 284 |
Feature activation | +2.21e-02 |
Pos logit contributions | |
favourite | +0.030 |
companion | +0.027 |
favorite | +0.026 |
found | +0.026 |
mate | +0.025 |
Neg logit contributions | |
arl | -0.067 |
iren | -0.063 |
its | -0.057 |
izes | -0.056 |
irens | -0.055 |
Token | try |
Token ID | 1949 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
arl | +0.050 |
iren | +0.046 |
its | +0.042 |
izes | +0.041 |
irens | +0.040 |
Neg logit contributions | |
favourite | -0.030 |
companion | -0.028 |
found | -0.027 |
favorite | -0.027 |
mate | -0.026 |
Token | ! |
Token ID | 0 |
Feature activation | -1.66e-02 |
Pos logit contributions | |
arl | +0.027 |
iren | +0.025 |
its | +0.022 |
izes | +0.022 |
irens | +0.021 |
Neg logit contributions | |
favourite | -0.018 |
companion | -0.016 |
favorite | -0.016 |
found | -0.016 |
one | -0.016 |
Token | very |
Token ID | 845 |
Feature activation | +5.36e-02 |
Pos logit contributions | |
favourite | +0.019 |
companion | +0.018 |
mate | +0.018 |
favorite | +0.018 |
Joey | +0.017 |
Neg logit contributions | |
arl | -0.021 |
iren | -0.019 |
izes | -0.017 |
yourselves | -0.017 |
its | -0.016 |
Token | careful |
Token ID | 8161 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
iren | +0.124 |
arl | +0.123 |
irens | +0.110 |
its | +0.110 |
isin | +0.101 |
Neg logit contributions | |
favourite | -0.071 |
favorite | -0.066 |
found | -0.064 |
one | -0.060 |
Sally | -0.059 |
Token | when |
Token ID | 618 |
Feature activation | +5.37e-02 |
Pos logit contributions | |
arl | +0.043 |
iren | +0.040 |
its | +0.036 |
irens | +0.034 |
izes | +0.034 |
Neg logit contributions | |
favourite | -0.023 |
one | -0.021 |
favorite | -0.020 |
found | -0.020 |
companion | -0.020 |
Token | running |
Token ID | 2491 |
Feature activation | +9.57e-04 |
Pos logit contributions | |
arl | +0.104 |
iren | +0.095 |
its | +0.083 |
izes | +0.080 |
irens | +0.079 |
Neg logit contributions | |
favourite | -0.091 |
companion | -0.085 |
found | -0.084 |
favorite | -0.083 |
one | -0.082 |
Token | ! |
Token ID | 0 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
arl | +0.002 |
iren | +0.002 |
its | +0.002 |
isin | +0.002 |
eling | +0.002 |
Neg logit contributions | |
favourite | -0.001 |
one | -0.001 |
found | -0.001 |
companion | -0.001 |
Sally | -0.001 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
favourite | +0.014 |
companion | +0.013 |
favorite | +0.013 |
found | +0.013 |
mate | +0.013 |
Neg logit contributions | |
arl | -0.025 |
iren | -0.024 |
its | -0.021 |
yourselves | -0.021 |
irens | -0.020 |
Token | Once |
Token ID | 7454 |
Feature activation | +2.54e-03 |
Pos logit contributions | |
arl | +0.240 |
iren | +0.223 |
its | +0.201 |
izes | +0.200 |
isin | +0.194 |
Neg logit contributions | |
favourite | -0.136 |
companion | -0.128 |
found | -0.124 |
favorite | -0.121 |
3 | -0.119 |
Token | upon |
Token ID | 2402 |
Feature activation | +7.69e-02 |
Pos logit contributions | |
arl | +0.006 |
iren | +0.005 |
its | +0.005 |
izes | +0.004 |
irens | +0.004 |
Neg logit contributions | |
favourite | -0.004 |
companion | -0.003 |
mate | -0.003 |
3 | -0.003 |
favorite | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | -1.87e-03 |
Pos logit contributions | |
arl | +0.194 |
iren | +0.183 |
its | +0.167 |
izes | +0.163 |
yourselves | +0.162 |
Neg logit contributions | |
favourite | -0.079 |
companion | -0.073 |
favorite | -0.068 |
found | -0.067 |
mate | -0.067 |
Token | time |
Token ID | 640 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
companion | +0.003 |
favourite | +0.003 |
Joey | +0.003 |
mate | +0.003 |
Sally | +0.003 |
Neg logit contributions | |
arl | -0.004 |
iren | -0.003 |
its | -0.003 |
isin | -0.003 |
falls | -0.003 |
Token | there |
Token ID | 612 |
Feature activation | -8.37e-02 |
Pos logit contributions | |
favourite | +0.047 |
companion | +0.047 |
nic | +0.045 |
favorite | +0.043 |
3 | +0.042 |
Neg logit contributions | |
arl | -0.044 |
iren | -0.037 |
ister | -0.033 |
falls | -0.032 |
ocus | -0.031 |
Token | for |
Token ID | 329 |
Feature activation | +2.67e-02 |
Pos logit contributions | |
arl | +0.190 |
iren | +0.177 |
izes | +0.155 |
its | +0.153 |
irens | +0.150 |
Neg logit contributions | |
favourite | -0.121 |
companion | -0.116 |
mate | -0.109 |
favorite | -0.108 |
3 | -0.105 |
Token | it |
Token ID | 340 |
Feature activation | -4.26e-02 |
Pos logit contributions | |
arl | +0.048 |
iren | +0.045 |
yourselves | +0.039 |
its | +0.039 |
irens | +0.038 |
Neg logit contributions | |
favourite | -0.044 |
companion | -0.043 |
favorite | -0.041 |
mate | -0.040 |
found | -0.039 |
Token | , |
Token ID | 11 |
Feature activation | -2.68e-02 |
Pos logit contributions | |
favourite | +0.069 |
companion | +0.066 |
mate | +0.065 |
3 | +0.063 |
favorite | +0.062 |
Neg logit contributions | |
arl | -0.075 |
iren | -0.069 |
izes | -0.060 |
its | -0.058 |
falls | -0.057 |
Token | but |
Token ID | 475 |
Feature activation | -5.34e-02 |
Pos logit contributions | |
favourite | +0.030 |
companion | +0.029 |
mate | +0.028 |
favorite | +0.025 |
3 | +0.024 |
Neg logit contributions | |
arl | -0.061 |
iren | -0.059 |
yourselves | -0.053 |
izes | -0.053 |
irens | -0.052 |
Token | he |
Token ID | 339 |
Feature activation | -8.82e-03 |
Pos logit contributions | |
companion | +0.084 |
favourite | +0.083 |
mate | +0.079 |
favorite | +0.076 |
3 | +0.075 |
Neg logit contributions | |
arl | -0.094 |
iren | -0.090 |
izes | -0.082 |
yourselves | -0.079 |
falls | -0.077 |
Token | couldn |
Token ID | 3521 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
favourite | +0.013 |
companion | +0.013 |
mate | +0.013 |
favorite | +0.012 |
picnic | +0.012 |
Neg logit contributions | |
arl | -0.015 |
iren | -0.015 |
izes | -0.014 |
its | -0.014 |
falls | -0.013 |
Token | 't |
Token ID | 470 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
arl | +0.277 |
iren | +0.224 |
izes | +0.212 |
its | +0.203 |
yourselves | +0.199 |
Neg logit contributions | |
one | -0.270 |
favourite | -0.267 |
found | -0.261 |
Sally | -0.256 |
Joey | -0.255 |
Token | find |
Token ID | 1064 |
Feature activation | -2.24e-02 |
Pos logit contributions | |
arl | +0.029 |
iren | +0.027 |
its | +0.025 |
izes | +0.025 |
yourselves | +0.024 |
Neg logit contributions | |
favourite | -0.013 |
companion | -0.011 |
favorite | -0.011 |
Sally | -0.010 |
mate | -0.010 |
Token | it |
Token ID | 340 |
Feature activation | -3.36e-02 |
Pos logit contributions | |
favourite | +0.042 |
companion | +0.040 |
favorite | +0.038 |
mate | +0.038 |
3 | +0.038 |
Neg logit contributions | |
arl | -0.036 |
iren | -0.034 |
its | -0.029 |
yourselves | -0.029 |
izes | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -1.66e-02 |
Pos logit contributions | |
favourite | +0.060 |
companion | +0.057 |
3 | +0.056 |
mate | +0.056 |
favorite | +0.054 |
Neg logit contributions | |
arl | -0.057 |
iren | -0.051 |
its | -0.045 |
izes | -0.045 |
yourselves | -0.043 |
Token | He |
Token ID | 679 |
Feature activation | +4.72e-02 |
Pos logit contributions | |
mate | +0.026 |
companion | +0.024 |
favourite | +0.024 |
picnic | +0.023 |
found | +0.022 |
Neg logit contributions | |
arl | -0.031 |
iren | -0.030 |
irens | -0.026 |
yourselves | -0.025 |
its | -0.024 |
Token | and |
Token ID | 290 |
Feature activation | +4.08e-02 |
Pos logit contributions | |
arl | +0.004 |
iren | +0.004 |
its | +0.003 |
isin | +0.003 |
irens | +0.003 |
Neg logit contributions | |
favourite | -0.002 |
one | -0.002 |
found | -0.002 |
favorite | -0.002 |
companion | -0.002 |
Token | careful |
Token ID | 8161 |
Feature activation | +2.54e-02 |
Pos logit contributions | |
arl | +0.090 |
iren | +0.085 |
its | +0.076 |
irens | +0.074 |
isin | +0.072 |
Neg logit contributions | |
favourite | -0.058 |
found | -0.054 |
favorite | -0.053 |
companion | -0.053 |
mate | -0.051 |
Token | when |
Token ID | 618 |
Feature activation | +5.78e-02 |
Pos logit contributions | |
arl | +0.063 |
iren | +0.058 |
its | +0.052 |
irens | +0.051 |
isin | +0.051 |
Neg logit contributions | |
favourite | -0.031 |
one | -0.029 |
favorite | -0.028 |
found | -0.027 |
Joey | -0.025 |
Token | balancing |
Token ID | 22486 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
arl | +0.122 |
iren | +0.113 |
its | +0.100 |
izes | +0.097 |
irens | +0.096 |
Neg logit contributions | |
favourite | -0.087 |
companion | -0.081 |
favorite | -0.078 |
found | -0.078 |
mate | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | +4.14e-03 |
Pos logit contributions | |
arl | +0.060 |
iren | +0.056 |
its | +0.050 |
irens | +0.048 |
isin | +0.048 |
Neg logit contributions | |
one | -0.035 |
favourite | -0.035 |
favorite | -0.032 |
found | -0.031 |
Sally | -0.029 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
arl | +0.010 |
iren | +0.009 |
its | +0.008 |
yourselves | +0.008 |
irens | +0.008 |
Neg logit contributions | |
favourite | -0.005 |
companion | -0.005 |
favorite | -0.005 |
found | -0.005 |
mate | -0.005 |
Token | One |
Token ID | 3198 |
Feature activation | -3.35e-02 |
Pos logit contributions | |
arl | +0.285 |
iren | +0.266 |
izes | +0.244 |
its | +0.243 |
isin | +0.237 |
Neg logit contributions | |
favourite | -0.118 |
companion | -0.108 |
found | -0.107 |
favorite | -0.102 |
one | -0.101 |
Token | day |
Token ID | 1110 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
favourite | +0.037 |
companion | +0.035 |
mate | +0.032 |
found | +0.031 |
favorite | +0.031 |
Neg logit contributions | |
arl | -0.082 |
iren | -0.077 |
its | -0.070 |
izes | -0.068 |
isin | -0.067 |
Token | , |
Token ID | 11 |
Feature activation | -5.25e-02 |
Pos logit contributions | |
arl | +0.025 |
iren | +0.022 |
its | +0.019 |
izes | +0.019 |
irens | +0.019 |
Neg logit contributions | |
favourite | -0.020 |
companion | -0.020 |
favorite | -0.018 |
3 | -0.018 |
mate | -0.017 |
Token | a |
Token ID | 257 |
Feature activation | -2.44e-03 |
Pos logit contributions | |
companion | +0.079 |
favourite | +0.073 |
visitor | +0.068 |
mate | +0.067 |
nic | +0.064 |
Neg logit contributions | |
arl | -0.107 |
iren | -0.091 |
yourselves | -0.086 |
its | -0.086 |
falls | -0.082 |
Token | little |
Token ID | 1310 |
Feature activation | -3.46e-03 |
Pos logit contributions | |
favourite | +0.003 |
companion | +0.003 |
found | +0.003 |
mate | +0.003 |
favorite | +0.003 |
Neg logit contributions | |
arl | -0.006 |
iren | -0.005 |
its | -0.005 |
izes | -0.004 |
isin | -0.004 |
Token | Yes |
Token ID | 5297 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
arl | +0.060 |
iren | +0.057 |
its | +0.049 |
izes | +0.049 |
yourselves | +0.048 |
Neg logit contributions | |
favourite | -0.045 |
companion | -0.041 |
found | -0.040 |
favorite | -0.040 |
Sally | -0.039 |
Token | , |
Token ID | 11 |
Feature activation | -6.34e-03 |
Pos logit contributions | |
favourite | +0.024 |
one | +0.022 |
mate | +0.021 |
Sally | +0.021 |
favorite | +0.021 |
Neg logit contributions | |
arl | -0.054 |
iren | -0.052 |
its | -0.048 |
irens | -0.047 |
izes | -0.046 |
Token | please |
Token ID | 3387 |
Feature activation | +4.10e-02 |
Pos logit contributions | |
favourite | +0.012 |
companion | +0.012 |
spotted | +0.011 |
found | +0.011 |
favourites | +0.011 |
Neg logit contributions | |
arl | -0.010 |
iren | -0.009 |
izes | -0.008 |
ister | -0.008 |
isin | -0.007 |
Token | !" |
Token ID | 2474 |
Feature activation | +3.11e-02 |
Pos logit contributions | |
iren | +0.101 |
arl | +0.101 |
irens | +0.092 |
isin | +0.089 |
its | +0.088 |
Neg logit contributions | |
one | -0.044 |
found | -0.042 |
favourite | -0.041 |
companion | -0.040 |
mate | -0.040 |
Token |
|
Token ID | 198 |
Feature activation | +2.62e-02 |
Pos logit contributions | |
arl | +0.084 |
iren | +0.079 |
its | +0.075 |
izes | +0.074 |
eling | +0.073 |
Neg logit contributions | |
favourite | -0.027 |
one | -0.025 |
Sally | -0.025 |
favorite | -0.024 |
companion | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
arl | +0.078 |
iren | +0.074 |
eling | +0.072 |
its | +0.070 |
ocus | +0.069 |
Neg logit contributions | |
favourite | -0.011 |
found | -0.011 |
Sally | -0.010 |
one | -0.010 |
favorite | -0.009 |
Token | The |
Token ID | 464 |
Feature activation | +4.80e-02 |
Pos logit contributions | |
arl | +0.054 |
iren | +0.047 |
its | +0.046 |
eling | +0.044 |
aned | +0.044 |
Neg logit contributions | |
found | -0.031 |
one | -0.030 |
favourite | -0.028 |
3 | -0.028 |
companion | -0.028 |
Token | lady |
Token ID | 10846 |
Feature activation | +6.51e-03 |
Pos logit contributions | |
arl | +0.099 |
iren | +0.092 |
its | +0.083 |
izes | +0.081 |
irens | +0.080 |
Neg logit contributions | |
favourite | -0.073 |
companion | -0.069 |
favorite | -0.068 |
one | -0.067 |
found | -0.067 |
Token | takes |
Token ID | 2753 |
Feature activation | -3.77e-03 |
Pos logit contributions | |
arl | +0.012 |
iren | +0.010 |
yourselves | +0.009 |
its | +0.009 |
irens | +0.009 |
Neg logit contributions | |
favourite | -0.012 |
found | -0.011 |
companion | -0.011 |
favorite | -0.011 |
one | -0.011 |
Token | them |
Token ID | 606 |
Feature activation | -6.52e-02 |
Pos logit contributions | |
favourite | +0.006 |
companion | +0.005 |
favorite | +0.005 |
mate | +0.005 |
found | +0.005 |
Neg logit contributions | |
arl | -0.008 |
iren | -0.007 |
its | -0.006 |
izes | -0.006 |
irens | -0.006 |
Token | to |
Token ID | 284 |
Feature activation | -8.86e-03 |
Pos logit contributions | |
favourite | +0.100 |
companion | +0.090 |
3 | +0.090 |
Joey | +0.088 |
favorite | +0.087 |
Neg logit contributions | |
arl | -0.130 |
iren | -0.121 |
izes | -0.103 |
irens | -0.102 |
its | -0.102 |
Token | !" |
Token ID | 2474 |
Feature activation | +2.64e-02 |
Pos logit contributions | |
arl | +0.117 |
iren | +0.108 |
irens | +0.088 |
its | +0.084 |
isin | +0.082 |
Neg logit contributions | |
favourite | -0.128 |
one | -0.127 |
found | -0.124 |
favorite | -0.123 |
three | -0.112 |
Token | Tim |
Token ID | 5045 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
arl | +0.060 |
iren | +0.056 |
its | +0.050 |
izes | +0.049 |
yourselves | +0.049 |
Neg logit contributions | |
favourite | -0.035 |
companion | -0.033 |
favorite | -0.031 |
found | -0.030 |
mate | -0.030 |
Token | says |
Token ID | 1139 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
arl | +0.025 |
iren | +0.022 |
its | +0.020 |
isin | +0.020 |
izes | +0.019 |
Neg logit contributions | |
found | -0.015 |
one | -0.014 |
favourite | -0.013 |
3 | -0.013 |
Sally | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | +2.65e-02 |
Pos logit contributions | |
favourite | +0.025 |
companion | +0.023 |
favorite | +0.022 |
found | +0.022 |
mate | +0.022 |
Neg logit contributions | |
arl | -0.031 |
iren | -0.029 |
its | -0.026 |
izes | -0.025 |
irens | -0.025 |
Token | " |
Token ID | 366 |
Feature activation | +2.15e-02 |
Pos logit contributions | |
arl | +0.064 |
iren | +0.060 |
its | +0.054 |
izes | +0.052 |
irens | +0.052 |
Neg logit contributions | |
favourite | -0.032 |
companion | -0.030 |
favorite | -0.028 |
found | -0.028 |
mate | -0.028 |
Token | Thank |
Token ID | 10449 |
Feature activation | +4.56e-02 |
Pos logit contributions | |
arl | +0.049 |
iren | +0.045 |
its | +0.040 |
izes | +0.040 |
irens | +0.040 |
Neg logit contributions | |
favourite | -0.029 |
found | -0.026 |
favorite | -0.026 |
one | -0.026 |
3 | -0.026 |
Token | you |
Token ID | 345 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
arl | +0.072 |
iren | +0.064 |
its | +0.054 |
izes | +0.051 |
irens | +0.050 |
Neg logit contributions | |
favourite | -0.092 |
one | -0.090 |
found | -0.088 |
mate | -0.086 |
favorite | -0.086 |
Token | , |
Token ID | 11 |
Feature activation | -1.83e-03 |
Pos logit contributions | |
favourite | +0.016 |
one | +0.015 |
mate | +0.015 |
3 | +0.015 |
Sally | +0.015 |
Neg logit contributions | |
arl | -0.029 |
iren | -0.027 |
its | -0.026 |
eling | -0.025 |
izes | -0.024 |
Token | Mom |
Token ID | 11254 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
favourite | +0.002 |
mate | +0.002 |
companion | +0.002 |
Sally | +0.002 |
one | +0.002 |
Neg logit contributions | |
arl | -0.005 |
iren | -0.004 |
its | -0.004 |
izes | -0.004 |
irens | -0.004 |
Token | !" |
Token ID | 2474 |
Feature activation | +3.07e-02 |
Pos logit contributions | |
arl | +0.038 |
iren | +0.033 |
its | +0.031 |
isin | +0.030 |
izes | +0.030 |
Neg logit contributions | |
one | -0.030 |
found | -0.029 |
Sally | -0.027 |
3 | -0.026 |
companion | -0.026 |
Token | Lily |
Token ID | 20037 |
Feature activation | +2.48e-02 |
Pos logit contributions | |
arl | +0.084 |
iren | +0.079 |
its | +0.075 |
eling | +0.075 |
izes | +0.074 |
Neg logit contributions | |
one | -0.025 |
favourite | -0.025 |
Sally | -0.024 |
found | -0.022 |
favorite | -0.022 |
Token | relieved |
Token ID | 22598 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
favourite | +0.022 |
companion | +0.021 |
mate | +0.021 |
favorite | +0.021 |
picnic | +0.020 |
Neg logit contributions | |
arl | -0.020 |
iren | -0.020 |
yourselves | -0.017 |
its | -0.017 |
irens | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | +2.14e-02 |
Pos logit contributions | |
favourite | +0.038 |
companion | +0.035 |
favorite | +0.035 |
mate | +0.034 |
found | +0.033 |
Neg logit contributions | |
arl | -0.045 |
iren | -0.041 |
its | -0.036 |
izes | -0.035 |
yourselves | -0.035 |
Token | felt |
Token ID | 2936 |
Feature activation | +4.54e-02 |
Pos logit contributions | |
arl | +0.037 |
iren | +0.036 |
izes | +0.034 |
eling | +0.032 |
isin | +0.031 |
Neg logit contributions | |
favourite | -0.034 |
mate | -0.032 |
companion | -0.032 |
favorite | -0.031 |
favourites | -0.029 |
Token | safe |
Token ID | 3338 |
Feature activation | +3.57e-02 |
Pos logit contributions | |
arl | +0.084 |
iren | +0.078 |
izes | +0.068 |
its | +0.068 |
yourselves | +0.066 |
Neg logit contributions | |
favourite | -0.075 |
companion | -0.072 |
mate | -0.071 |
favorite | -0.069 |
3 | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
arl | +0.067 |
iren | +0.062 |
its | +0.054 |
izes | +0.053 |
yourselves | +0.052 |
Neg logit contributions | |
favourite | -0.060 |
companion | -0.055 |
favorite | -0.054 |
mate | -0.054 |
3 | -0.052 |
Token | She |
Token ID | 1375 |
Feature activation | +5.72e-02 |
Pos logit contributions | |
favourite | +0.017 |
companion | +0.017 |
mate | +0.016 |
favorite | +0.016 |
found | +0.016 |
Neg logit contributions | |
arl | -0.025 |
iren | -0.023 |
its | -0.020 |
yourselves | -0.020 |
izes | -0.019 |
Token | got |
Token ID | 1392 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
arl | +0.109 |
iren | +0.106 |
izes | +0.099 |
its | +0.095 |
isin | +0.092 |
Neg logit contributions | |
favourite | -0.082 |
mate | -0.075 |
companion | -0.074 |
favorite | -0.074 |
3 | -0.068 |
Token | closer |
Token ID | 5699 |
Feature activation | -5.33e-02 |
Pos logit contributions | |
iren | +0.022 |
arl | +0.022 |
izes | +0.019 |
its | +0.019 |
irens | +0.018 |
Neg logit contributions | |
favourite | -0.020 |
mate | -0.020 |
companion | -0.019 |
favorite | -0.018 |
picnic | -0.017 |
Token | to |
Token ID | 284 |
Feature activation | +9.21e-03 |
Pos logit contributions | |
favourite | +0.091 |
mate | +0.088 |
companion | +0.087 |
favorite | +0.083 |
3 | +0.081 |
Neg logit contributions | |
arl | -0.091 |
iren | -0.086 |
izes | -0.073 |
its | -0.072 |
irens | -0.072 |
Token | the |
Token ID | 262 |
Feature activation | +2.55e-02 |
Pos logit contributions | |
arl | +0.018 |
iren | +0.016 |
yourselves | +0.014 |
its | +0.014 |
eling | +0.014 |
Neg logit contributions | |
favourite | -0.016 |
favorite | -0.014 |
companion | -0.014 |
mate | -0.014 |
found | -0.013 |
Token | barn |
Token ID | 25203 |
Feature activation | +3.94e-02 |
Pos logit contributions | |
arl | +0.064 |
iren | +0.058 |
its | +0.051 |
yourselves | +0.051 |
izes | +0.051 |
Neg logit contributions | |
favourite | -0.027 |
mate | -0.025 |
companion | -0.023 |
found | -0.023 |
favorite | -0.023 |
Token | flying |
Token ID | 7348 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
favourite | +0.046 |
companion | +0.044 |
favorite | +0.042 |
found | +0.040 |
mate | +0.040 |
Neg logit contributions | |
arl | -0.062 |
iren | -0.058 |
its | -0.051 |
yourselves | -0.050 |
izes | -0.050 |
Token | through |
Token ID | 832 |
Feature activation | -3.85e-03 |
Pos logit contributions | |
favourite | +0.044 |
one | +0.042 |
companion | +0.041 |
found | +0.041 |
favorite | +0.040 |
Neg logit contributions | |
arl | -0.056 |
iren | -0.053 |
its | -0.047 |
irens | -0.046 |
isin | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | +2.28e-02 |
Pos logit contributions | |
one | +0.005 |
favourite | +0.004 |
3 | +0.004 |
found | +0.004 |
Sally | +0.004 |
Neg logit contributions | |
arl | -0.010 |
iren | -0.009 |
izes | -0.009 |
its | -0.009 |
isin | -0.008 |
Token | sky |
Token ID | 6766 |
Feature activation | -2.40e-03 |
Pos logit contributions | |
arl | +0.053 |
iren | +0.047 |
izes | +0.045 |
its | +0.044 |
irens | +0.042 |
Neg logit contributions | |
favourite | -0.031 |
3 | -0.030 |
one | -0.030 |
found | -0.029 |
companion | -0.029 |
Token | like |
Token ID | 588 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
favourite | +0.004 |
companion | +0.004 |
one | +0.004 |
found | +0.004 |
favorite | +0.004 |
Neg logit contributions | |
arl | -0.004 |
iren | -0.004 |
its | -0.003 |
irens | -0.003 |
izes | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
favourite | +0.021 |
one | +0.020 |
companion | +0.020 |
found | +0.019 |
3 | +0.019 |
Neg logit contributions | |
arl | -0.028 |
iren | -0.026 |
its | -0.024 |
izes | -0.023 |
irens | -0.022 |
Token | bubble |
Token ID | 14310 |
Feature activation | -6.36e-04 |
Pos logit contributions | |
arl | +0.030 |
iren | +0.029 |
its | +0.026 |
irens | +0.026 |
izes | +0.025 |
Neg logit contributions | |
favourite | -0.015 |
companion | -0.014 |
3 | -0.013 |
favorite | -0.013 |
one | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -8.50e-03 |
Pos logit contributions | |
favourite | +0.001 |
companion | +0.001 |
one | +0.001 |
mate | +0.001 |
found | +0.001 |
Neg logit contributions | |
arl | -0.001 |
iren | -0.001 |
its | -0.001 |
irens | -0.001 |
yourselves | -0.001 |
Token | It |
Token ID | 632 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
favourite | +0.013 |
companion | +0.012 |
3 | +0.011 |
favorite | +0.011 |
found | +0.011 |
Neg logit contributions | |
arl | -0.018 |
iren | -0.016 |
its | -0.015 |
izes | -0.014 |
irens | -0.014 |
Token | 's |
Token ID | 338 |
Feature activation | -7.53e-02 |
Pos logit contributions | |
favourite | +0.016 |
companion | +0.015 |
favorite | +0.014 |
found | +0.014 |
mate | +0.014 |
Neg logit contributions | |
arl | -0.031 |
iren | -0.029 |
its | -0.026 |
izes | -0.026 |
irens | -0.026 |
Token | lots |
Token ID | 6041 |
Feature activation | +2.20e-02 |
Pos logit contributions | |
favourite | +0.097 |
companion | +0.090 |
favorite | +0.087 |
found | +0.086 |
mate | +0.084 |
Neg logit contributions | |
arl | -0.174 |
iren | -0.163 |
its | -0.146 |
izes | -0.143 |
irens | -0.142 |
Token | soon |
Token ID | 2582 |
Feature activation | +5.24e-02 |
Pos logit contributions | |
iren | +0.029 |
arl | +0.029 |
falls | +0.026 |
izes | +0.026 |
irens | +0.025 |
Neg logit contributions | |
favourite | -0.025 |
favorite | -0.023 |
companion | -0.021 |
mate | -0.021 |
newest | -0.021 |
Token | they |
Token ID | 484 |
Feature activation | -5.30e-02 |
Pos logit contributions | |
arl | +0.096 |
iren | +0.095 |
falls | +0.093 |
l | +0.089 |
izes | +0.088 |
Neg logit contributions | |
favourite | -0.080 |
3 | -0.070 |
favorite | -0.070 |
companion | -0.068 |
favourites | -0.068 |
Token | were |
Token ID | 547 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
favourite | +0.062 |
favorite | +0.054 |
3 | +0.052 |
companion | +0.051 |
mate | +0.048 |
Neg logit contributions | |
arl | -0.126 |
iren | -0.118 |
its | -0.108 |
izes | -0.107 |
falls | -0.106 |
Token | safe |
Token ID | 3338 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
favourite | +0.021 |
companion | +0.019 |
favorite | +0.019 |
3 | +0.018 |
mate | +0.018 |
Neg logit contributions | |
arl | -0.026 |
iren | -0.024 |
izes | -0.021 |
its | -0.021 |
irens | -0.021 |
Token | at |
Token ID | 379 |
Feature activation | -4.47e-02 |
Pos logit contributions | |
arl | +0.033 |
iren | +0.030 |
its | +0.027 |
izes | +0.026 |
yourselves | +0.026 |
Neg logit contributions | |
favourite | -0.024 |
companion | -0.022 |
favorite | -0.022 |
found | -0.021 |
mate | -0.021 |
Token | home |
Token ID | 1363 |
Feature activation | +2.79e-02 |
Pos logit contributions | |
favourite | +0.082 |
mate | +0.076 |
companion | +0.076 |
favorite | +0.073 |
spotted | +0.073 |
Neg logit contributions | |
arl | -0.071 |
iren | -0.069 |
yourselves | -0.060 |
irens | -0.057 |
its | -0.057 |
Token | . |
Token ID | 13 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
arl | +0.058 |
iren | +0.053 |
its | +0.047 |
izes | +0.046 |
irens | +0.046 |
Neg logit contributions | |
favourite | -0.042 |
companion | -0.040 |
favorite | -0.038 |
found | -0.038 |
mate | -0.038 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +8.78e-02 |
Pos logit contributions | |
mate | +0.036 |
companion | +0.036 |
favourite | +0.034 |
visitor | +0.033 |
favorite | +0.031 |
Neg logit contributions | |
arl | -0.047 |
iren | -0.042 |
yourselves | -0.038 |
irens | -0.036 |
its | -0.035 |
Token | Once |
Token ID | 7454 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
arl | +0.186 |
iren | +0.174 |
its | +0.153 |
irens | +0.148 |
yourselves | +0.146 |
Neg logit contributions | |
favourite | -0.130 |
companion | -0.119 |
favorite | -0.117 |
mate | -0.116 |
found | -0.111 |
Token | there |
Token ID | 612 |
Feature activation | -9.73e-02 |
Pos logit contributions | |
favourite | +0.016 |
companion | +0.016 |
mate | +0.014 |
favorite | +0.014 |
3 | +0.014 |
Neg logit contributions | |
arl | -0.027 |
iren | -0.025 |
its | -0.023 |
izes | -0.022 |
irens | -0.021 |
Token | was |
Token ID | 373 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
mate | +0.176 |
companion | +0.167 |
favourite | +0.162 |
nic | +0.149 |
visitor | +0.148 |
Neg logit contributions | |
its | -0.130 |
falls | -0.128 |
iren | -0.128 |
arl | -0.122 |
ister | -0.119 |
Token | and |
Token ID | 290 |
Feature activation | +4.98e-02 |
Pos logit contributions | |
arl | +0.048 |
iren | +0.044 |
its | +0.039 |
izes | +0.038 |
isin | +0.037 |
Neg logit contributions | |
favourite | -0.035 |
companion | -0.034 |
found | -0.033 |
favorite | -0.032 |
one | -0.032 |
Token | taller |
Token ID | 25242 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
arl | +0.098 |
iren | +0.088 |
its | +0.082 |
irens | +0.078 |
ister | +0.076 |
Neg logit contributions | |
found | -0.083 |
one | -0.083 |
favourite | -0.082 |
favorite | -0.081 |
Sally | -0.077 |
Token | than |
Token ID | 621 |
Feature activation | +7.69e-03 |
Pos logit contributions | |
arl | +0.031 |
iren | +0.029 |
isin | +0.027 |
izes | +0.026 |
its | +0.026 |
Neg logit contributions | |
one | -0.014 |
favorite | -0.013 |
favourite | -0.013 |
found | -0.012 |
companion | -0.011 |
Token | Tim |
Token ID | 5045 |
Feature activation | -5.13e-03 |
Pos logit contributions | |
arl | +0.019 |
iren | +0.018 |
its | +0.016 |
izes | +0.016 |
irens | +0.015 |
Neg logit contributions | |
favourite | -0.009 |
companion | -0.008 |
found | -0.008 |
favorite | -0.008 |
mate | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | +3.26e-02 |
Pos logit contributions | |
found | +0.006 |
favourite | +0.006 |
companion | +0.006 |
one | +0.006 |
favorite | +0.006 |
Neg logit contributions | |
arl | -0.012 |
iren | -0.011 |
its | -0.010 |
izes | -0.010 |
irens | -0.010 |
Token | Mia |
Token ID | 32189 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
arl | +0.081 |
iren | +0.075 |
its | +0.068 |
izes | +0.067 |
irens | +0.066 |
Neg logit contributions | |
favourite | -0.037 |
found | -0.034 |
companion | -0.034 |
favorite | -0.033 |
one | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | +1.79e-02 |
Pos logit contributions | |
favourite | +0.019 |
found | +0.019 |
companion | +0.018 |
one | +0.018 |
favorite | +0.017 |
Neg logit contributions | |
arl | -0.035 |
iren | -0.033 |
irens | -0.029 |
izes | -0.029 |
its | -0.029 |
Token | He |
Token ID | 679 |
Feature activation | +6.53e-02 |
Pos logit contributions | |
arl | +0.044 |
iren | +0.041 |
its | +0.038 |
izes | +0.038 |
isin | +0.037 |
Neg logit contributions | |
favourite | -0.020 |
companion | -0.018 |
favorite | -0.018 |
found | -0.017 |
Sally | -0.017 |
Token | sees |
Token ID | 7224 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
arl | +0.138 |
iren | +0.120 |
isin | +0.104 |
irens | +0.104 |
ister | +0.103 |
Neg logit contributions | |
favourite | -0.104 |
found | -0.103 |
favorite | -0.099 |
one | -0.097 |
spotted | -0.094 |
Token | the |
Token ID | 262 |
Feature activation | +2.63e-02 |
Pos logit contributions | |
favourite | +0.019 |
companion | +0.018 |
mate | +0.017 |
favorite | +0.017 |
3 | +0.016 |
Neg logit contributions | |
arl | -0.025 |
iren | -0.024 |
irens | -0.021 |
its | -0.021 |
yourselves | -0.021 |
Token | new |
Token ID | 649 |
Feature activation | -2.54e-02 |
Pos logit contributions | |
arl | +0.063 |
iren | +0.059 |
its | +0.053 |
izes | +0.052 |
yourselves | +0.051 |
Neg logit contributions | |
favourite | -0.032 |
companion | -0.029 |
found | -0.028 |
mate | -0.028 |
favorite | -0.027 |
Token | said |
Token ID | 531 |
Feature activation | -4.55e-02 |
Pos logit contributions | |
favourite | +0.028 |
mate | +0.026 |
companion | +0.026 |
favorite | +0.025 |
newest | +0.024 |
Neg logit contributions | |
iren | -0.033 |
izes | -0.030 |
falls | -0.029 |
arl | -0.028 |
l | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | -2.71e-02 |
Pos logit contributions | |
favourite | +0.075 |
companion | +0.071 |
favorite | +0.068 |
favourites | +0.067 |
mate | +0.066 |
Neg logit contributions | |
arl | -0.084 |
iren | -0.078 |
izes | -0.070 |
ister | -0.066 |
irens | -0.065 |
Token | " |
Token ID | 366 |
Feature activation | +4.62e-03 |
Pos logit contributions | |
favourite | +0.034 |
companion | +0.032 |
favorite | +0.031 |
mate | +0.030 |
found | +0.029 |
Neg logit contributions | |
arl | -0.063 |
iren | -0.058 |
izes | -0.051 |
its | -0.051 |
irens | -0.050 |
Token | Hi |
Token ID | 17250 |
Feature activation | -3.54e-02 |
Pos logit contributions | |
arl | +0.010 |
iren | +0.009 |
its | +0.008 |
yourselves | +0.008 |
irens | +0.008 |
Neg logit contributions | |
favourite | -0.007 |
favorite | -0.006 |
companion | -0.006 |
mate | -0.006 |
found | -0.006 |
Token | Toby |
Token ID | 37578 |
Feature activation | +3.22e-02 |
Pos logit contributions | |
favourite | +0.042 |
companion | +0.038 |
favorite | +0.037 |
found | +0.037 |
3 | +0.036 |
Neg logit contributions | |
arl | -0.086 |
iren | -0.081 |
its | -0.074 |
izes | -0.072 |
irens | -0.071 |
Token | ! |
Token ID | 0 |
Feature activation | -6.12e-03 |
Pos logit contributions | |
arl | +0.070 |
iren | +0.066 |
its | +0.060 |
irens | +0.058 |
izes | +0.058 |
Neg logit contributions | |
favourite | -0.046 |
companion | -0.044 |
found | -0.043 |
one | -0.043 |
Sally | -0.042 |
Token | Will |
Token ID | 2561 |
Feature activation | +5.99e-03 |
Pos logit contributions | |
favourite | +0.010 |
companion | +0.009 |
favorite | +0.009 |
found | +0.009 |
mate | +0.009 |
Neg logit contributions | |
arl | -0.012 |
iren | -0.011 |
its | -0.009 |
yourselves | -0.009 |
izes | -0.009 |
Token | you |
Token ID | 345 |
Feature activation | -3.77e-02 |
Pos logit contributions | |
arl | +0.009 |
iren | +0.008 |
its | +0.007 |
izes | +0.007 |
irens | +0.007 |
Neg logit contributions | |
favourite | -0.012 |
companion | -0.012 |
favorite | -0.011 |
found | -0.011 |
mate | -0.011 |
Token | swing |
Token ID | 9628 |
Feature activation | +4.29e-03 |
Pos logit contributions | |
favourite | +0.051 |
companion | +0.047 |
favorite | +0.046 |
found | +0.044 |
mate | +0.044 |
Neg logit contributions | |
arl | -0.084 |
iren | -0.078 |
its | -0.071 |
izes | -0.069 |
yourselves | -0.067 |
Token | with |
Token ID | 351 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
arl | +0.009 |
iren | +0.008 |
its | +0.008 |
irens | +0.007 |
isin | +0.007 |
Neg logit contributions | |
favourite | -0.007 |
favorite | -0.006 |
one | -0.006 |
found | -0.006 |
companion | -0.006 |
Token | me |
Token ID | 502 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
favourite | +0.181 |
companion | +0.169 |
found | +0.168 |
one | +0.166 |
favorite | +0.166 |
Neg logit contributions | |
arl | -0.191 |
iren | -0.174 |
its | -0.155 |
izes | -0.151 |
irens | -0.145 |
Token | his |
Token ID | 465 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
found | +0.025 |
favourite | +0.024 |
one | +0.024 |
Sally | +0.023 |
spotted | +0.023 |
Neg logit contributions | |
arl | -0.052 |
iren | -0.046 |
its | -0.042 |
izes | -0.042 |
isin | -0.041 |
Token | sock |
Token ID | 32263 |
Feature activation | +2.26e-03 |
Pos logit contributions | |
mate | +0.022 |
companion | +0.022 |
found | +0.022 |
favourite | +0.022 |
Sally | +0.021 |
Neg logit contributions | |
arl | -0.046 |
iren | -0.042 |
its | -0.039 |
izes | -0.037 |
yourselves | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
arl | +0.004 |
iren | +0.004 |
its | +0.004 |
izes | +0.004 |
irens | +0.004 |
Neg logit contributions | |
favourite | -0.004 |
companion | -0.003 |
favorite | -0.003 |
found | -0.003 |
mate | -0.003 |
Token | The |
Token ID | 383 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
favourite | +0.031 |
companion | +0.029 |
mate | +0.028 |
favorite | +0.027 |
found | +0.025 |
Neg logit contributions | |
arl | -0.047 |
iren | -0.046 |
its | -0.040 |
irens | -0.039 |
yourselves | -0.039 |
Token | naughty |
Token ID | 45128 |
Feature activation | -1.26e-03 |
Pos logit contributions | |
arl | +0.030 |
iren | +0.028 |
its | +0.026 |
izes | +0.025 |
irens | +0.025 |
Neg logit contributions | |
favourite | -0.012 |
companion | -0.010 |
favorite | -0.010 |
found | -0.010 |
3 | -0.010 |
Token | boy |
Token ID | 2933 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
favourite | +0.001 |
companion | +0.001 |
mate | +0.001 |
favorite | +0.001 |
found | +0.001 |
Neg logit contributions | |
arl | -0.003 |
iren | -0.003 |
its | -0.003 |
irens | -0.003 |
izes | -0.003 |
Token | wanted |
Token ID | 2227 |
Feature activation | +9.70e-04 |
Pos logit contributions | |
favourite | +0.022 |
companion | +0.021 |
mate | +0.021 |
favorite | +0.019 |
picnic | +0.017 |
Neg logit contributions | |
iren | -0.037 |
arl | -0.035 |
its | -0.034 |
izes | -0.033 |
falls | -0.033 |
Token | it |
Token ID | 340 |
Feature activation | -3.09e-02 |
Pos logit contributions | |
arl | +0.002 |
iren | +0.002 |
its | +0.002 |
yourselves | +0.002 |
irens | +0.002 |
Neg logit contributions | |
favourite | -0.001 |
companion | -0.001 |
mate | -0.001 |
favorite | -0.001 |
3 | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
favourite | +0.047 |
companion | +0.045 |
mate | +0.044 |
3 | +0.043 |
spotted | +0.042 |
Neg logit contributions | |
arl | -0.060 |
iren | -0.055 |
izes | -0.048 |
its | -0.048 |
yourselves | -0.047 |
Token | so |
Token ID | 523 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
favourite | +0.019 |
companion | +0.018 |
favorite | +0.017 |
found | +0.017 |
mate | +0.016 |
Neg logit contributions | |
arl | -0.045 |
iren | -0.042 |
its | -0.039 |
izes | -0.038 |
irens | -0.037 |
Token | he |
Token ID | 339 |
Feature activation | +7.32e-03 |
Pos logit contributions | |
arl | +0.026 |
iren | +0.024 |
its | +0.022 |
izes | +0.022 |
yourselves | +0.021 |
Neg logit contributions | |
favourite | -0.017 |
companion | -0.017 |
mate | -0.015 |
favorite | -0.015 |
3 | -0.015 |
Token | was |
Token ID | 373 |
Feature activation | -8.83e-03 |
Pos logit contributions | |
favourite | +0.043 |
companion | +0.040 |
mate | +0.039 |
3 | +0.038 |
favorite | +0.037 |
Neg logit contributions | |
arl | -0.068 |
iren | -0.065 |
its | -0.062 |
izes | -0.061 |
falls | -0.058 |
Token | too |
Token ID | 1165 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
favourite | +0.014 |
companion | +0.013 |
favorite | +0.013 |
found | +0.013 |
mate | +0.012 |
Neg logit contributions | |
arl | -0.018 |
iren | -0.017 |
its | -0.015 |
izes | -0.014 |
irens | -0.014 |
Token | late |
Token ID | 2739 |
Feature activation | +7.62e-02 |
Pos logit contributions | |
found | +0.018 |
one | +0.018 |
Sally | +0.017 |
favorite | +0.017 |
favourite | +0.017 |
Neg logit contributions | |
arl | -0.022 |
iren | -0.022 |
yourselves | -0.019 |
izes | -0.019 |
irens | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -7.65e-04 |
Pos logit contributions | |
arl | +0.187 |
iren | +0.178 |
its | +0.160 |
irens | +0.156 |
isin | +0.156 |
Neg logit contributions | |
favourite | -0.085 |
one | -0.084 |
found | -0.078 |
companion | -0.078 |
Sally | -0.076 |
Token | The |
Token ID | 383 |
Feature activation | +2.90e-02 |
Pos logit contributions | |
favourite | +0.001 |
companion | +0.001 |
favorite | +0.001 |
found | +0.001 |
3 | +0.001 |
Neg logit contributions | |
arl | -0.002 |
iren | -0.002 |
its | -0.002 |
izes | -0.002 |
irens | -0.002 |
Token | dog |
Token ID | 3290 |
Feature activation | +7.35e-03 |
Pos logit contributions | |
arl | +0.068 |
iren | +0.063 |
its | +0.057 |
izes | +0.056 |
yourselves | +0.056 |
Neg logit contributions | |
favourite | -0.035 |
companion | -0.033 |
mate | -0.031 |
Joey | -0.031 |
favorite | -0.030 |
Token | caught |
Token ID | 4978 |
Feature activation | -4.96e-03 |
Pos logit contributions | |
arl | +0.018 |
iren | +0.017 |
its | +0.015 |
izes | +0.015 |
irens | +0.014 |
Neg logit contributions | |
favourite | -0.009 |
companion | -0.008 |
mate | -0.007 |
favorite | -0.007 |
3 | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | +2.43e-02 |
Pos logit contributions | |
favourite | +0.006 |
one | +0.006 |
Sally | +0.006 |
found | +0.006 |
favorite | +0.006 |
Neg logit contributions | |
arl | -0.012 |
iren | -0.011 |
izes | -0.010 |
its | -0.010 |
isin | -0.009 |
Token | pigeon |
Token ID | 46387 |
Feature activation | -2.89e-02 |
Pos logit contributions | |
arl | +0.054 |
iren | +0.049 |
its | +0.045 |
yourselves | +0.045 |
izes | +0.044 |
Neg logit contributions | |
favourite | -0.031 |
companion | -0.029 |
Joey | -0.029 |
searched | -0.028 |
mate | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
favourite | +0.038 |
companion | +0.036 |
found | +0.035 |
one | +0.035 |
favorite | +0.035 |
Neg logit contributions | |
arl | -0.067 |
iren | -0.062 |
its | -0.055 |
irens | -0.055 |
izes | -0.054 |
Token | ate |
Token ID | 15063 |
Feature activation | -5.41e-03 |
Pos logit contributions | |
arl | +0.034 |
iren | +0.032 |
its | +0.029 |
izes | +0.028 |
irens | +0.028 |
Neg logit contributions | |
favourite | -0.020 |
companion | -0.018 |
favorite | -0.018 |
found | -0.018 |
mate | -0.017 |
Token | purple |
Token ID | 14032 |
Feature activation | -4.40e-02 |
Pos logit contributions | |
favourite | +0.093 |
favorite | +0.087 |
one | +0.086 |
found | +0.084 |
companion | +0.083 |
Neg logit contributions | |
arl | -0.144 |
iren | -0.137 |
irens | -0.120 |
its | -0.119 |
izes | -0.116 |
Token | or |
Token ID | 393 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
favourite | +0.073 |
companion | +0.068 |
favorite | +0.067 |
one | +0.065 |
found | +0.065 |
Neg logit contributions | |
arl | -0.088 |
iren | -0.081 |
its | -0.071 |
izes | -0.069 |
irens | -0.069 |
Token | green |
Token ID | 4077 |
Feature activation | -9.05e-03 |
Pos logit contributions | |
arl | +0.075 |
iren | +0.070 |
its | +0.062 |
irens | +0.061 |
izes | +0.058 |
Neg logit contributions | |
favourite | -0.051 |
one | -0.048 |
favorite | -0.048 |
found | -0.046 |
3 | -0.044 |
Token | . |
Token ID | 13 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
favourite | +0.014 |
one | +0.013 |
favorite | +0.013 |
companion | +0.013 |
found | +0.012 |
Neg logit contributions | |
arl | -0.019 |
iren | -0.018 |
its | -0.016 |
irens | -0.016 |
izes | -0.016 |
Token | I |
Token ID | 314 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
arl | +0.044 |
iren | +0.042 |
its | +0.038 |
izes | +0.038 |
irens | +0.037 |
Neg logit contributions | |
favourite | -0.025 |
companion | -0.023 |
favorite | -0.022 |
Sally | -0.022 |
3 | -0.022 |
Token | care |
Token ID | 1337 |
Feature activation | -3.43e-02 |
Pos logit contributions | |
favourite | +0.046 |
found | +0.043 |
companion | +0.043 |
favorite | +0.042 |
one | +0.041 |
Neg logit contributions | |
arl | -0.062 |
iren | -0.058 |
irens | -0.051 |
its | -0.050 |
yourselves | -0.050 |
Token | about |
Token ID | 546 |
Feature activation | -2.96e-02 |
Pos logit contributions | |
one | +0.026 |
favourite | +0.025 |
favorite | +0.024 |
found | +0.021 |
three | +0.020 |
Neg logit contributions | |
iren | -0.092 |
arl | -0.089 |
irens | -0.088 |
ister | -0.082 |
its | -0.082 |
Token | you |
Token ID | 345 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
favourite | +0.049 |
one | +0.048 |
favorite | +0.045 |
Sally | +0.045 |
found | +0.044 |
Neg logit contributions | |
arl | -0.061 |
iren | -0.056 |
its | -0.048 |
izes | -0.048 |
irens | -0.046 |
Token | . |
Token ID | 13 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
one | +0.035 |
favourite | +0.034 |
favorite | +0.032 |
found | +0.032 |
three | +0.029 |
Neg logit contributions | |
arl | -0.074 |
iren | -0.072 |
irens | -0.065 |
its | -0.064 |
izes | -0.059 |
Token | You |
Token ID | 921 |
Feature activation | -8.52e-03 |
Pos logit contributions | |
arl | +0.025 |
iren | +0.023 |
its | +0.020 |
izes | +0.020 |
irens | +0.019 |
Neg logit contributions | |
favourite | -0.020 |
companion | -0.019 |
favorite | -0.018 |
found | -0.018 |
mate | -0.018 |
Token | are |
Token ID | 389 |
Feature activation | -7.53e-03 |
Pos logit contributions | |
favourite | +0.013 |
found | +0.013 |
favorite | +0.012 |
companion | +0.012 |
mate | +0.012 |
Neg logit contributions | |
arl | -0.018 |
iren | -0.017 |
irens | -0.015 |
its | -0.014 |
izes | -0.014 |
Token | loved |
Token ID | 6151 |
Feature activation | +3.88e-02 |
Pos logit contributions | |
favourite | +0.016 |
found | +0.016 |
companion | +0.014 |
favorite | +0.014 |
one | +0.014 |
Neg logit contributions | |
arl | -0.047 |
iren | -0.043 |
its | -0.039 |
yourselves | -0.039 |
irens | -0.039 |
Token | going |
Token ID | 1016 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
arl | +0.098 |
iren | +0.091 |
its | +0.084 |
izes | +0.082 |
isin | +0.079 |
Neg logit contributions | |
favourite | -0.044 |
companion | -0.040 |
favorite | -0.039 |
found | -0.038 |
3 | -0.038 |
Token | to |
Token ID | 284 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
arl | +0.023 |
iren | +0.021 |
its | +0.019 |
izes | +0.019 |
yourselves | +0.018 |
Neg logit contributions | |
favourite | -0.015 |
companion | -0.014 |
favorite | -0.013 |
found | -0.013 |
one | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | +2.86e-02 |
Pos logit contributions | |
arl | +0.034 |
iren | +0.032 |
its | +0.028 |
izes | +0.028 |
irens | +0.027 |
Neg logit contributions | |
favourite | -0.025 |
companion | -0.023 |
favorite | -0.023 |
found | -0.022 |
mate | -0.022 |
Token | lab |
Token ID | 2248 |
Feature activation | +5.37e-02 |
Pos logit contributions | |
arl | +0.067 |
iren | +0.063 |
its | +0.056 |
izes | +0.056 |
irens | +0.055 |
Neg logit contributions | |
favourite | -0.035 |
companion | -0.032 |
mate | -0.031 |
found | -0.030 |
favorite | -0.030 |
Token | with |
Token ID | 351 |
Feature activation | -9.44e-02 |
Pos logit contributions | |
arl | +0.116 |
iren | +0.109 |
its | +0.097 |
izes | +0.096 |
irens | +0.094 |
Neg logit contributions | |
favourite | -0.074 |
companion | -0.069 |
favorite | -0.066 |
mate | -0.066 |
found | -0.065 |
Token | her |
Token ID | 607 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
favourite | +0.120 |
companion | +0.113 |
mate | +0.106 |
favorite | +0.106 |
found | +0.105 |
Neg logit contributions | |
arl | -0.216 |
iren | -0.202 |
its | -0.182 |
izes | -0.178 |
yourselves | -0.177 |
Token | mom |
Token ID | 1995 |
Feature activation | +8.21e-05 |
Pos logit contributions | |
favourite | +0.020 |
mate | +0.020 |
companion | +0.020 |
Joey | +0.019 |
found | +0.019 |
Neg logit contributions | |
arl | -0.036 |
iren | -0.033 |
its | -0.029 |
izes | -0.028 |
irens | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | -2.27e-02 |
Pos logit contributions | |
arl | +0.000 |
iren | +0.000 |
izes | +0.000 |
its | +0.000 |
irens | +0.000 |
Neg logit contributions | |
favourite | -0.000 |
mate | -0.000 |
companion | -0.000 |
favorite | -0.000 |
found | -0.000 |
Token | but |
Token ID | 475 |
Feature activation | -3.98e-02 |
Pos logit contributions | |
favourite | +0.032 |
mate | +0.031 |
companion | +0.030 |
Joey | +0.028 |
3 | +0.028 |
Neg logit contributions | |
arl | -0.043 |
iren | -0.040 |
yourselves | -0.037 |
rou | -0.035 |
izes | -0.035 |
Token | she |
Token ID | 673 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
favourite | +0.067 |
companion | +0.065 |
mate | +0.063 |
Joey | +0.061 |
3 | +0.061 |
Neg logit contributions | |
arl | -0.065 |
iren | -0.060 |
izes | -0.059 |
its | -0.054 |
yourselves | -0.054 |
Token | a |
Token ID | 257 |
Feature activation | +7.07e-03 |
Pos logit contributions | |
favourite | +0.003 |
companion | +0.003 |
mate | +0.003 |
spotted | +0.003 |
3 | +0.003 |
Neg logit contributions | |
iren | -0.003 |
arl | -0.003 |
yourselves | -0.003 |
irens | -0.003 |
its | -0.003 |
Token | k |
Token ID | 479 |
Feature activation | +5.93e-03 |
Pos logit contributions | |
arl | +0.014 |
iren | +0.013 |
its | +0.011 |
izes | +0.011 |
isin | +0.011 |
Neg logit contributions | |
favourite | -0.011 |
companion | -0.010 |
found | -0.010 |
Joey | -0.010 |
mate | -0.010 |
Token | ite |
Token ID | 578 |
Feature activation | -8.01e-04 |
Pos logit contributions | |
arl | +0.009 |
iren | +0.008 |
its | +0.007 |
izes | +0.007 |
irens | +0.006 |
Neg logit contributions | |
favourite | -0.012 |
companion | -0.012 |
favorite | -0.011 |
found | -0.011 |
mate | -0.011 |
Token | like |
Token ID | 588 |
Feature activation | -2.51e-02 |
Pos logit contributions | |
favourite | +0.001 |
companion | +0.001 |
3 | +0.001 |
found | +0.001 |
favorite | +0.001 |
Neg logit contributions | |
arl | -0.002 |
iren | -0.002 |
its | -0.001 |
izes | -0.001 |
yourselves | -0.001 |
Token | Tim |
Token ID | 5045 |
Feature activation | -6.86e-03 |
Pos logit contributions | |
favourite | +0.036 |
companion | +0.034 |
found | +0.033 |
favorite | +0.032 |
mate | +0.032 |
Neg logit contributions | |
arl | -0.053 |
iren | -0.050 |
its | -0.044 |
irens | -0.043 |
izes | -0.042 |
Token | 's |
Token ID | 338 |
Feature activation | -8.63e-02 |
Pos logit contributions | |
favourite | +0.009 |
companion | +0.008 |
mate | +0.008 |
3 | +0.007 |
favorite | +0.007 |
Neg logit contributions | |
arl | -0.016 |
iren | -0.015 |
its | -0.013 |
izes | -0.013 |
yourselves | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
favourite | +0.124 |
companion | +0.116 |
favorite | +0.112 |
found | +0.111 |
mate | +0.109 |
Neg logit contributions | |
arl | -0.187 |
iren | -0.173 |
its | -0.155 |
izes | -0.151 |
irens | -0.149 |
Token | Sue |
Token ID | 26089 |
Feature activation | -2.11e-02 |
Pos logit contributions | |
arl | +0.038 |
iren | +0.035 |
yourselves | +0.032 |
izes | +0.032 |
its | +0.032 |
Neg logit contributions | |
favourite | -0.017 |
companion | -0.015 |
3 | -0.015 |
mate | -0.014 |
favorite | -0.014 |
Token | 's |
Token ID | 338 |
Feature activation | -8.77e-02 |
Pos logit contributions | |
favourite | +0.031 |
companion | +0.029 |
favorite | +0.028 |
found | +0.028 |
mate | +0.028 |
Neg logit contributions | |
arl | -0.045 |
iren | -0.041 |
its | -0.037 |
izes | -0.036 |
irens | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | +3.47e-03 |
Pos logit contributions | |
favourite | +0.127 |
companion | +0.118 |
favorite | +0.115 |
found | +0.112 |
mate | +0.110 |
Neg logit contributions | |
arl | -0.190 |
iren | -0.178 |
its | -0.158 |
izes | -0.155 |
irens | -0.154 |
Token | They |
Token ID | 1119 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
arl | +0.008 |
iren | +0.007 |
its | +0.006 |
yourselves | +0.006 |
irens | +0.006 |
Neg logit contributions | |
favourite | -0.005 |
companion | -0.005 |
3 | -0.005 |
spotted | -0.004 |
mate | -0.004 |
Token | was |
Token ID | 373 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
favourite | +0.009 |
companion | +0.008 |
favorite | +0.008 |
found | +0.008 |
mate | +0.008 |
Neg logit contributions | |
arl | -0.021 |
iren | -0.019 |
its | -0.018 |
izes | -0.017 |
irens | -0.017 |
Token | so |
Token ID | 523 |
Feature activation | -2.87e-03 |
Pos logit contributions | |
favourite | +0.020 |
companion | +0.018 |
favorite | +0.018 |
found | +0.018 |
3 | +0.017 |
Neg logit contributions | |
arl | -0.035 |
iren | -0.033 |
its | -0.030 |
izes | -0.029 |
irens | -0.029 |
Token | tasty |
Token ID | 25103 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
favourite | +0.004 |
favorite | +0.003 |
found | +0.003 |
Sally | +0.003 |
3 | +0.003 |
Neg logit contributions | |
arl | -0.007 |
iren | -0.006 |
irens | -0.006 |
its | -0.006 |
izes | -0.006 |
Token | that |
Token ID | 326 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
favourite | +0.031 |
companion | +0.029 |
favorite | +0.028 |
mate | +0.028 |
3 | +0.028 |
Neg logit contributions | |
arl | -0.033 |
iren | -0.030 |
its | -0.026 |
izes | -0.026 |
irens | -0.025 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
favourite | +0.021 |
companion | +0.020 |
mate | +0.019 |
3 | +0.018 |
favorite | +0.018 |
Neg logit contributions | |
arl | -0.028 |
iren | -0.026 |
its | -0.024 |
izes | -0.024 |
yourselves | -0.023 |
Token | 's |
Token ID | 338 |
Feature activation | -8.90e-02 |
Pos logit contributions | |
favourite | +0.018 |
companion | +0.017 |
mate | +0.017 |
3 | +0.015 |
favorite | +0.015 |
Neg logit contributions | |
arl | -0.046 |
iren | -0.044 |
its | -0.041 |
izes | -0.040 |
isin | -0.039 |
Token | tum |
Token ID | 11814 |
Feature activation | +3.26e-02 |
Pos logit contributions | |
favourite | +0.105 |
companion | +0.093 |
favorite | +0.093 |
found | +0.089 |
3 | +0.088 |
Neg logit contributions | |
arl | -0.221 |
iren | -0.208 |
its | -0.187 |
izes | -0.184 |
irens | -0.183 |
Token | my |
Token ID | 1820 |
Feature activation | +6.88e-02 |
Pos logit contributions | |
arl | +0.083 |
iren | +0.077 |
izes | +0.071 |
its | +0.070 |
irens | +0.068 |
Neg logit contributions | |
favourite | -0.032 |
found | -0.030 |
Sally | -0.030 |
one | -0.029 |
favorite | -0.028 |
Token | started |
Token ID | 2067 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
arl | +0.180 |
iren | +0.160 |
irens | +0.144 |
izes | +0.143 |
its | +0.141 |
Neg logit contributions | |
found | -0.083 |
favourite | -0.082 |
one | -0.077 |
favorite | -0.077 |
spotted | -0.074 |
Token | to |
Token ID | 284 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
arl | +0.037 |
iren | +0.035 |
its | +0.032 |
izes | +0.031 |
irens | +0.031 |
Neg logit contributions | |
favourite | -0.016 |
companion | -0.014 |
favorite | -0.014 |
found | -0.014 |
3 | -0.013 |
Token | rum |
Token ID | 7440 |
Feature activation | +7.11e-02 |
Pos logit contributions | |
arl | +0.047 |
iren | +0.044 |
izes | +0.039 |
irens | +0.039 |
its | +0.038 |
Neg logit contributions | |
favourite | -0.029 |
one | -0.028 |
found | -0.028 |
Sally | -0.026 |
3 | -0.026 |
Token | she |
Token ID | 673 |
Feature activation | +1.87e-02 |
Pos logit contributions | |
favourite | +0.021 |
companion | +0.019 |
mate | +0.019 |
favorite | +0.018 |
3 | +0.018 |
Neg logit contributions | |
arl | -0.024 |
iren | -0.022 |
yourselves | -0.020 |
its | -0.020 |
izes | -0.019 |
Token | wanted |
Token ID | 2227 |
Feature activation | +2.91e-03 |
Pos logit contributions | |
arl | +0.042 |
iren | +0.040 |
its | +0.039 |
izes | +0.038 |
falls | +0.036 |
Neg logit contributions | |
favourite | -0.021 |
mate | -0.019 |
companion | -0.019 |
3 | -0.018 |
favorite | -0.018 |
Token | to |
Token ID | 284 |
Feature activation | +6.55e-03 |
Pos logit contributions | |
arl | +0.006 |
iren | +0.005 |
yourselves | +0.005 |
its | +0.005 |
irens | +0.005 |
Neg logit contributions | |
favourite | -0.004 |
companion | -0.004 |
mate | -0.004 |
3 | -0.004 |
favorite | -0.004 |
Token | be |
Token ID | 307 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
arl | +0.015 |
iren | +0.014 |
its | +0.012 |
izes | +0.012 |
yourselves | +0.012 |
Neg logit contributions | |
favourite | -0.009 |
companion | -0.008 |
favorite | -0.008 |
3 | -0.008 |
mate | -0.008 |
Token | friends |
Token ID | 2460 |
Feature activation | -8.03e-04 |
Pos logit contributions | |
favourite | +0.027 |
companion | +0.024 |
3 | +0.024 |
favorite | +0.024 |
mate | +0.024 |
Neg logit contributions | |
arl | -0.035 |
iren | -0.033 |
yourselves | -0.031 |
izes | -0.030 |
its | -0.028 |
Token | with |
Token ID | 351 |
Feature activation | -9.76e-02 |
Pos logit contributions | |
favourite | +0.001 |
companion | +0.001 |
favorite | +0.001 |
found | +0.001 |
mate | +0.001 |
Neg logit contributions | |
arl | -0.002 |
iren | -0.002 |
its | -0.001 |
izes | -0.001 |
yourselves | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
favourite | +0.118 |
one | +0.107 |
companion | +0.107 |
found | +0.106 |
favorite | +0.105 |
Neg logit contributions | |
arl | -0.241 |
iren | -0.225 |
izes | -0.200 |
its | -0.200 |
irens | -0.194 |
Token | dog |
Token ID | 3290 |
Feature activation | -3.03e-03 |
Pos logit contributions | |
arl | +0.026 |
iren | +0.023 |
its | +0.021 |
isin | +0.020 |
yourselves | +0.020 |
Neg logit contributions | |
favourite | -0.017 |
Joey | -0.017 |
searched | -0.017 |
found | -0.016 |
Sally | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -4.47e-03 |
Pos logit contributions | |
favourite | +0.005 |
companion | +0.004 |
favorite | +0.004 |
found | +0.004 |
3 | +0.004 |
Neg logit contributions | |
arl | -0.006 |
iren | -0.006 |
its | -0.005 |
izes | -0.005 |
yourselves | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +1.89e-03 |
Pos logit contributions | |
favourite | +0.006 |
companion | +0.006 |
favorite | +0.006 |
mate | +0.006 |
found | +0.005 |
Neg logit contributions | |
arl | -0.010 |
iren | -0.009 |
its | -0.008 |
yourselves | -0.008 |
irens | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | +2.76e-03 |
Pos logit contributions | |
arl | +0.005 |
iren | +0.005 |
izes | +0.005 |
its | +0.005 |
yourselves | +0.004 |
Neg logit contributions | |
favourite | -0.001 |
found | -0.001 |
Sally | -0.001 |
Joey | -0.001 |
companion | -0.001 |
Token | Lily |
Token ID | 20037 |
Feature activation | -3.24e-02 |
Pos logit contributions | |
favourite | +0.063 |
companion | +0.063 |
mate | +0.060 |
3 | +0.056 |
favorite | +0.055 |
Neg logit contributions | |
arl | -0.091 |
iren | -0.084 |
its | -0.073 |
yourselves | -0.073 |
izes | -0.073 |
Token | went |
Token ID | 1816 |
Feature activation | +3.91e-02 |
Pos logit contributions | |
favourite | +0.034 |
mate | +0.032 |
companion | +0.032 |
Joey | +0.028 |
favorite | +0.027 |
Neg logit contributions | |
arl | -0.074 |
iren | -0.073 |
falls | -0.068 |
izes | -0.068 |
its | -0.067 |
Token | to |
Token ID | 284 |
Feature activation | -7.06e-03 |
Pos logit contributions | |
arl | +0.091 |
iren | +0.086 |
izes | +0.074 |
irens | +0.074 |
its | +0.073 |
Neg logit contributions | |
favourite | -0.050 |
companion | -0.046 |
mate | -0.043 |
favorite | -0.043 |
3 | -0.043 |
Token | the |
Token ID | 262 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
favourite | +0.013 |
companion | +0.012 |
mate | +0.012 |
favorite | +0.011 |
spotted | +0.011 |
Neg logit contributions | |
arl | -0.012 |
iren | -0.011 |
yourselves | -0.010 |
irens | -0.009 |
its | -0.009 |
Token | park |
Token ID | 3952 |
Feature activation | +3.23e-02 |
Pos logit contributions | |
arl | +0.035 |
iren | +0.032 |
yourselves | +0.029 |
irens | +0.028 |
izes | +0.028 |
Neg logit contributions | |
favourite | -0.020 |
companion | -0.019 |
Joey | -0.018 |
mate | -0.018 |
found | -0.018 |
Token | with |
Token ID | 351 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
iren | +0.035 |
arl | +0.028 |
irens | +0.025 |
ister | +0.024 |
falls | +0.023 |
Neg logit contributions | |
favourite | -0.068 |
mate | -0.068 |
Joey | -0.066 |
mattered | -0.064 |
companion | -0.064 |
Token | her |
Token ID | 607 |
Feature activation | -3.22e-02 |
Pos logit contributions | |
companion | +0.210 |
favourite | +0.208 |
mate | +0.203 |
3 | +0.192 |
Joey | +0.188 |
Neg logit contributions | |
arl | -0.172 |
iren | -0.161 |
yourselves | -0.135 |
irens | -0.133 |
falls | -0.129 |
Token | mom |
Token ID | 1995 |
Feature activation | -2.00e-02 |
Pos logit contributions | |
Joey | +0.048 |
companion | +0.046 |
Frank | +0.045 |
spotted | +0.045 |
favourite | +0.044 |
Neg logit contributions | |
arl | -0.063 |
iren | -0.056 |
ister | -0.048 |
izes | -0.048 |
irens | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | -3.51e-02 |
Pos logit contributions | |
favourite | +0.036 |
companion | +0.034 |
mate | +0.033 |
favorite | +0.031 |
Joey | +0.031 |
Neg logit contributions | |
arl | -0.033 |
iren | -0.031 |
irens | -0.025 |
ister | -0.024 |
isin | -0.023 |
Token | They |
Token ID | 1119 |
Feature activation | -3.44e-02 |
Pos logit contributions | |
favourite | +0.054 |
mate | +0.053 |
companion | +0.052 |
favorite | +0.048 |
visitor | +0.046 |
Neg logit contributions | |
arl | -0.072 |
iren | -0.064 |
irens | -0.053 |
its | -0.053 |
yourselves | -0.051 |
Token | saw |
Token ID | 2497 |
Feature activation | -2.46e-02 |
Pos logit contributions | |
companion | +0.052 |
favourite | +0.049 |
mate | +0.045 |
strang | +0.044 |
newest | +0.042 |
Neg logit contributions | |
iren | -0.051 |
arl | -0.051 |
falls | -0.050 |
eling | -0.049 |
pped | -0.049 |