Token | belonged |
Token ID | 19611 |
Feature activation | +2.65e+00 |
Pos logit contributions | |
different | +0.015 |
fast | +0.014 |
slow | +0.013 |
very | +0.013 |
really | +0.013 |
Neg logit contributions | |
lor | -0.026 |
ocket | -0.024 |
urger | -0.023 |
oll | -0.021 |
card | -0.021 |
Token | to |
Token ID | 284 |
Feature activation | +6.88e-01 |
Pos logit contributions | |
lor | +0.165 |
ocket | +0.154 |
urger | +0.149 |
oll | +0.142 |
card | +0.141 |
Neg logit contributions | |
different | -0.029 |
fast | -0.028 |
slow | -0.023 |
very | -0.021 |
really | -0.019 |
Token | your |
Token ID | 534 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
lor | +0.035 |
ocket | +0.032 |
urger | +0.030 |
oll | +0.029 |
card | +0.028 |
Neg logit contributions | |
fast | -0.013 |
different | -0.013 |
slow | -0.012 |
late | -0.012 |
very | -0.011 |
Token | great |
Token ID | 1049 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
lor | +0.082 |
ocket | +0.072 |
urger | +0.069 |
card | +0.065 |
oll | +0.065 |
Neg logit contributions | |
different | -0.056 |
fast | -0.055 |
very | -0.051 |
slow | -0.051 |
really | -0.048 |
Token | - |
Token ID | 12 |
Feature activation | -3.56e+00 |
Pos logit contributions | |
different | +0.039 |
fast | +0.038 |
slow | +0.035 |
very | +0.034 |
really | +0.032 |
Neg logit contributions | |
lor | -0.075 |
ocket | -0.066 |
urger | -0.062 |
card | -0.060 |
oll | -0.059 |
Token | grand |
Token ID | 23936 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
fast | +0.144 |
different | +0.144 |
very | +0.134 |
slow | +0.133 |
really | +0.131 |
Neg logit contributions | |
lor | -0.117 |
ocket | -0.102 |
urger | -0.099 |
oll | -0.088 |
anger | -0.084 |
Token | mother |
Token ID | 13552 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
lor | +0.367 |
ocket | +0.323 |
urger | +0.316 |
anger | +0.312 |
isc | +0.299 |
Neg logit contributions | |
different | -0.362 |
very | -0.334 |
fast | -0.324 |
the | -0.321 |
slow | -0.317 |
Token | . |
Token ID | 13 |
Feature activation | +6.49e-01 |
Pos logit contributions | |
lor | +0.135 |
ocket | +0.125 |
urger | +0.122 |
oll | +0.115 |
card | +0.114 |
Neg logit contributions | |
different | -0.029 |
fast | -0.028 |
very | -0.023 |
slow | -0.023 |
really | -0.021 |
Token | She |
Token ID | 1375 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
lor | +0.034 |
ocket | +0.031 |
urger | +0.030 |
card | +0.028 |
ster | +0.028 |
Neg logit contributions | |
different | -0.014 |
fast | -0.013 |
slow | -0.012 |
very | -0.012 |
late | -0.012 |
Token | gave |
Token ID | 2921 |
Feature activation | +3.42e+00 |
Pos logit contributions | |
lor | +0.086 |
ocket | +0.077 |
urger | +0.076 |
anger | +0.071 |
card | +0.070 |
Neg logit contributions | |
different | -0.038 |
fast | -0.037 |
very | -0.035 |
really | -0.035 |
slow | -0.034 |
Token | it |
Token ID | 340 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
lor | +0.194 |
ocket | +0.180 |
urger | +0.173 |
oll | +0.164 |
card | +0.163 |
Neg logit contributions | |
different | -0.057 |
fast | -0.056 |
slow | -0.049 |
very | -0.047 |
really | -0.044 |
Token | belong |
Token ID | 5594 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
lor | +0.100 |
urger | +0.089 |
oll | +0.086 |
ocket | +0.085 |
ogo | +0.082 |
Neg logit contributions | |
different | -0.035 |
fast | -0.033 |
very | -0.032 |
to | -0.028 |
slow | -0.028 |
Token | to |
Token ID | 284 |
Feature activation | -6.03e-02 |
Pos logit contributions | |
lor | +0.077 |
ocket | +0.071 |
urger | +0.070 |
oll | +0.066 |
anger | +0.065 |
Neg logit contributions | |
different | -0.014 |
fast | -0.014 |
very | -0.011 |
slow | -0.011 |
really | -0.009 |
Token | their |
Token ID | 511 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
different | +0.001 |
fast | +0.001 |
slow | +0.001 |
very | +0.001 |
really | +0.001 |
Neg logit contributions | |
lor | -0.003 |
ocket | -0.003 |
urger | -0.003 |
oll | -0.003 |
card | -0.003 |
Token | great |
Token ID | 1049 |
Feature activation | -2.36e+00 |
Pos logit contributions | |
lor | +0.086 |
ocket | +0.075 |
urger | +0.071 |
oll | +0.068 |
anger | +0.066 |
Neg logit contributions | |
different | -0.057 |
fast | -0.056 |
very | -0.052 |
slow | -0.051 |
really | -0.048 |
Token | - |
Token ID | 12 |
Feature activation | -4.34e+00 |
Pos logit contributions | |
different | +0.064 |
fast | +0.060 |
very | +0.055 |
fun | +0.055 |
slow | +0.054 |
Neg logit contributions | |
lor | -0.122 |
ocket | -0.099 |
urger | -0.098 |
ogo | -0.096 |
oll | -0.096 |
Token | grand |
Token ID | 23936 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
fast | +0.178 |
different | +0.176 |
very | +0.165 |
late | +0.160 |
slow | +0.160 |
Neg logit contributions | |
lor | -0.141 |
ocket | -0.121 |
urger | -0.118 |
oll | -0.108 |
anger | -0.102 |
Token | pa |
Token ID | 8957 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
lor | +0.384 |
ocket | +0.348 |
urger | +0.337 |
anger | +0.325 |
oll | +0.321 |
Neg logit contributions | |
different | -0.336 |
very | -0.307 |
fast | -0.296 |
the | -0.294 |
really | -0.285 |
Token | . |
Token ID | 13 |
Feature activation | +7.27e-01 |
Pos logit contributions | |
lor | +0.156 |
ocket | +0.143 |
urger | +0.139 |
anger | +0.135 |
oll | +0.134 |
Neg logit contributions | |
different | -0.023 |
fast | -0.021 |
very | -0.018 |
slow | -0.016 |
really | -0.015 |
Token | He |
Token ID | 679 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
lor | +0.044 |
ocket | +0.041 |
urger | +0.039 |
oll | +0.037 |
card | +0.037 |
Neg logit contributions | |
different | -0.009 |
fast | -0.009 |
slow | -0.008 |
very | -0.007 |
really | -0.007 |
Token | was |
Token ID | 373 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
lor | +0.085 |
urger | +0.074 |
ocket | +0.072 |
anger | +0.070 |
immer | +0.068 |
Neg logit contributions | |
different | -0.042 |
very | -0.039 |
really | -0.039 |
fast | -0.038 |
slow | -0.035 |
Token | an |
Token ID | 281 |
Feature activation | +9.28e-01 |
Pos logit contributions | |
different | +0.002 |
fast | +0.002 |
slow | +0.001 |
very | +0.001 |
really | +0.001 |
Neg logit contributions | |
lor | -0.009 |
ocket | -0.008 |
urger | -0.008 |
oll | -0.007 |
card | -0.007 |
Token | chest |
Token ID | 7721 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
lor | +0.052 |
ocket | +0.050 |
urger | +0.047 |
oll | +0.045 |
card | +0.044 |
Neg logit contributions | |
different | -0.020 |
fast | -0.019 |
very | -0.018 |
slow | -0.017 |
really | -0.017 |
Token | was |
Token ID | 373 |
Feature activation | +6.33e-01 |
Pos logit contributions | |
lor | +0.103 |
ocket | +0.093 |
urger | +0.090 |
oll | +0.085 |
card | +0.083 |
Neg logit contributions | |
different | -0.044 |
fast | -0.044 |
very | -0.042 |
slow | -0.039 |
really | -0.039 |
Token | your |
Token ID | 534 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
lor | +0.040 |
ocket | +0.036 |
urger | +0.035 |
anger | +0.034 |
oll | +0.033 |
Neg logit contributions | |
different | -0.009 |
fast | -0.009 |
very | -0.008 |
slow | -0.007 |
really | -0.007 |
Token | great |
Token ID | 1049 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
lor | +0.058 |
ocket | +0.046 |
urger | +0.046 |
oll | +0.044 |
bridge | +0.044 |
Neg logit contributions | |
different | -0.048 |
fast | -0.046 |
very | -0.045 |
fun | -0.042 |
slow | -0.041 |
Token | - |
Token ID | 12 |
Feature activation | -4.15e+00 |
Pos logit contributions | |
different | +0.055 |
fast | +0.052 |
fun | +0.049 |
very | +0.048 |
slow | +0.046 |
Neg logit contributions | |
lor | -0.109 |
ocket | -0.097 |
urger | -0.089 |
icking | -0.087 |
oll | -0.086 |
Token | grand |
Token ID | 23936 |
Feature activation | +1.01e+01 |
Pos logit contributions | |
fast | +0.165 |
different | +0.160 |
very | +0.152 |
slow | +0.147 |
really | +0.147 |
Neg logit contributions | |
lor | -0.139 |
ocket | -0.117 |
urger | -0.117 |
oll | -0.105 |
icking | -0.103 |
Token | ma |
Token ID | 2611 |
Feature activation | -6.44e-01 |
Pos logit contributions | |
lor | +0.379 |
ocket | +0.345 |
anger | +0.330 |
urger | +0.327 |
isc | +0.316 |
Neg logit contributions | |
different | -0.285 |
the | -0.270 |
very | -0.263 |
her | -0.262 |
care | -0.252 |
Token | 's |
Token ID | 338 |
Feature activation | +2.05e+00 |
Pos logit contributions | |
different | +0.009 |
fast | +0.009 |
very | +0.009 |
slow | +0.008 |
really | +0.008 |
Neg logit contributions | |
lor | -0.038 |
ocket | -0.035 |
urger | -0.034 |
anger | -0.033 |
oll | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | +4.02e-01 |
Pos logit contributions | |
lor | +0.095 |
ocket | +0.078 |
oll | +0.078 |
urger | +0.077 |
icking | +0.076 |
Neg logit contributions | |
different | -0.064 |
very | -0.061 |
fast | -0.061 |
fun | -0.056 |
really | -0.054 |
Token | She |
Token ID | 1375 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
lor | +0.022 |
ocket | +0.021 |
urger | +0.021 |
ogo | +0.019 |
icking | +0.019 |
Neg logit contributions | |
fast | -0.006 |
She | -0.006 |
They | -0.006 |
different | -0.006 |
late | -0.006 |
Token | left |
Token ID | 1364 |
Feature activation | -6.64e-01 |
Pos logit contributions | |
lor | +0.086 |
ocket | +0.076 |
urger | +0.075 |
anger | +0.073 |
uit | +0.071 |
Neg logit contributions | |
different | -0.029 |
very | -0.028 |
really | -0.028 |
fast | -0.027 |
slow | -0.025 |
Token | That |
Token ID | 2504 |
Feature activation | +7.77e-01 |
Pos logit contributions | |
different | +0.052 |
fast | +0.051 |
slow | +0.045 |
very | +0.043 |
really | +0.040 |
Neg logit contributions | |
lor | -0.168 |
ocket | -0.153 |
urger | -0.144 |
card | -0.142 |
oll | -0.139 |
Token | 's |
Token ID | 338 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
lor | +0.040 |
ocket | +0.038 |
urger | +0.036 |
oll | +0.035 |
card | +0.033 |
Neg logit contributions | |
different | -0.017 |
fast | -0.016 |
slow | -0.015 |
very | -0.015 |
really | -0.014 |
Token | your |
Token ID | 534 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
lor | +0.106 |
ocket | +0.099 |
urger | +0.095 |
oll | +0.090 |
card | +0.090 |
Neg logit contributions | |
different | -0.027 |
fast | -0.027 |
slow | -0.023 |
very | -0.022 |
really | -0.020 |
Token | great |
Token ID | 1049 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
lor | +0.089 |
ocket | +0.078 |
urger | +0.075 |
oll | +0.073 |
bridge | +0.071 |
Neg logit contributions | |
different | -0.057 |
fast | -0.051 |
very | -0.050 |
slow | -0.048 |
really | -0.046 |
Token | - |
Token ID | 12 |
Feature activation | -3.99e+00 |
Pos logit contributions | |
different | +0.051 |
fast | +0.047 |
very | +0.045 |
slow | +0.043 |
really | +0.042 |
Neg logit contributions | |
lor | -0.107 |
ocket | -0.099 |
urger | -0.091 |
oll | -0.090 |
ster | -0.085 |
Token | grand |
Token ID | 23936 |
Feature activation | +9.90e+00 |
Pos logit contributions | |
different | +0.173 |
fast | +0.173 |
very | +0.162 |
slow | +0.160 |
really | +0.158 |
Neg logit contributions | |
lor | -0.121 |
ocket | -0.104 |
urger | -0.099 |
oll | -0.087 |
ster | -0.082 |
Token | ma |
Token ID | 2611 |
Feature activation | -7.24e-01 |
Pos logit contributions | |
lor | +0.387 |
urger | +0.347 |
ocket | +0.346 |
anger | +0.326 |
oll | +0.317 |
Neg logit contributions | |
different | -0.301 |
very | -0.275 |
the | -0.268 |
really | -0.262 |
her | -0.261 |
Token | ," |
Token ID | 553 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
different | +0.012 |
fast | +0.011 |
very | +0.010 |
slow | +0.010 |
really | +0.009 |
Neg logit contributions | |
lor | -0.042 |
ocket | -0.039 |
urger | -0.037 |
anger | -0.036 |
oll | -0.036 |
Token | her |
Token ID | 607 |
Feature activation | +5.16e-01 |
Pos logit contributions | |
different | +0.025 |
fast | +0.023 |
slow | +0.019 |
very | +0.018 |
really | +0.016 |
Neg logit contributions | |
lor | -0.138 |
ocket | -0.128 |
urger | -0.124 |
oll | -0.118 |
anger | -0.116 |
Token | mom |
Token ID | 1995 |
Feature activation | -2.41e+00 |
Pos logit contributions | |
lor | +0.022 |
ocket | +0.020 |
urger | +0.019 |
oll | +0.017 |
card | +0.017 |
Neg logit contributions | |
different | -0.017 |
fast | -0.016 |
slow | -0.015 |
very | -0.015 |
really | -0.015 |
Token | replied |
Token ID | 8712 |
Feature activation | +4.63e+00 |
Pos logit contributions | |
different | +0.071 |
fast | +0.068 |
very | +0.065 |
slow | +0.064 |
really | +0.063 |
Neg logit contributions | |
lor | -0.110 |
ocket | -0.100 |
urger | -0.094 |
anger | -0.088 |
oll | -0.086 |
Token | parents |
Token ID | 3397 |
Feature activation | -3.53e-01 |
Pos logit contributions | |
lor | +0.052 |
oll | +0.042 |
ocket | +0.042 |
urger | +0.040 |
immer | +0.040 |
Neg logit contributions | |
different | -0.037 |
fast | -0.036 |
very | -0.034 |
slow | -0.032 |
fun | -0.031 |
Token | and |
Token ID | 290 |
Feature activation | +8.30e-01 |
Pos logit contributions | |
different | +0.005 |
fast | +0.005 |
slow | +0.004 |
very | +0.004 |
really | +0.003 |
Neg logit contributions | |
lor | -0.021 |
ocket | -0.020 |
urger | -0.019 |
oll | -0.018 |
card | -0.018 |
Token | their |
Token ID | 511 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
lor | +0.041 |
ocket | +0.036 |
urger | +0.035 |
anger | +0.033 |
oll | +0.033 |
Neg logit contributions | |
different | -0.023 |
fast | -0.023 |
slow | -0.021 |
very | -0.021 |
really | -0.019 |
Token | great |
Token ID | 1049 |
Feature activation | -2.36e+00 |
Pos logit contributions | |
lor | +0.068 |
ocket | +0.055 |
urger | +0.053 |
oll | +0.052 |
anger | +0.050 |
Neg logit contributions | |
different | -0.051 |
fast | -0.049 |
very | -0.046 |
slow | -0.045 |
fun | -0.043 |
Token | - |
Token ID | 12 |
Feature activation | -4.62e+00 |
Pos logit contributions | |
different | +0.069 |
fast | +0.061 |
fun | +0.061 |
to | +0.058 |
very | +0.058 |
Neg logit contributions | |
lor | -0.118 |
ocket | -0.094 |
ancing | -0.092 |
oll | -0.092 |
uit | -0.091 |
Token | grand |
Token ID | 23936 |
Feature activation | +9.90e+00 |
Pos logit contributions | |
different | +0.191 |
fast | +0.191 |
very | +0.179 |
slow | +0.175 |
late | +0.173 |
Neg logit contributions | |
lor | -0.149 |
ocket | -0.128 |
urger | -0.120 |
oll | -0.111 |
anger | -0.107 |
Token | pa |
Token ID | 8957 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
lor | +0.335 |
ocket | +0.301 |
urger | +0.278 |
oll | +0.268 |
anger | +0.264 |
Neg logit contributions | |
different | -0.383 |
fast | -0.352 |
very | -0.352 |
slow | -0.338 |
really | -0.333 |
Token | 's |
Token ID | 338 |
Feature activation | +2.16e+00 |
Pos logit contributions | |
lor | +0.131 |
ocket | +0.119 |
urger | +0.116 |
anger | +0.114 |
oll | +0.111 |
Neg logit contributions | |
different | -0.014 |
fast | -0.014 |
very | -0.012 |
slow | -0.008 |
really | -0.008 |
Token | picture |
Token ID | 4286 |
Feature activation | +3.72e+00 |
Pos logit contributions | |
lor | +0.107 |
oll | +0.089 |
ogo | +0.087 |
anger | +0.085 |
urger | +0.084 |
Neg logit contributions | |
different | -0.067 |
very | -0.064 |
fast | -0.063 |
fun | -0.061 |
happy | -0.058 |
Token | . |
Token ID | 13 |
Feature activation | +4.44e-01 |
Pos logit contributions | |
lor | +0.241 |
ocket | +0.223 |
urger | +0.216 |
oll | +0.203 |
card | +0.202 |
Neg logit contributions | |
fast | -0.038 |
different | -0.038 |
very | -0.031 |
slow | -0.028 |
really | -0.024 |
Token | They |
Token ID | 1119 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
lor | +0.026 |
ocket | +0.024 |
urger | +0.023 |
oll | +0.022 |
card | +0.022 |
Neg logit contributions | |
different | -0.006 |
fast | -0.006 |
slow | -0.005 |
very | -0.005 |
really | -0.004 |
Token | helpful |
Token ID | 7613 |
Feature activation | +4.24e-02 |
Pos logit contributions | |
lor | +0.202 |
ocket | +0.186 |
urger | +0.179 |
oll | +0.168 |
card | +0.167 |
Neg logit contributions | |
different | -0.083 |
fast | -0.082 |
slow | -0.074 |
very | -0.071 |
really | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
lor | +0.002 |
ocket | +0.002 |
urger | +0.002 |
oll | +0.002 |
card | +0.002 |
Neg logit contributions | |
different | -0.001 |
fast | -0.001 |
slow | -0.001 |
very | -0.001 |
really | -0.001 |
Token | She |
Token ID | 1375 |
Feature activation | +2.21e+00 |
Pos logit contributions | |
lor | +0.094 |
ocket | +0.088 |
urger | +0.085 |
oll | +0.080 |
card | +0.080 |
Neg logit contributions | |
different | -0.024 |
fast | -0.023 |
slow | -0.020 |
very | -0.019 |
really | -0.017 |
Token | took |
Token ID | 1718 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
lor | +0.116 |
ocket | +0.108 |
urger | +0.104 |
card | +0.097 |
oll | +0.097 |
Neg logit contributions | |
different | -0.047 |
fast | -0.045 |
slow | -0.041 |
very | -0.041 |
really | -0.039 |
Token | a |
Token ID | 257 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
lor | +0.138 |
ocket | +0.128 |
urger | +0.124 |
oll | +0.117 |
card | +0.117 |
Neg logit contributions | |
different | -0.033 |
fast | -0.032 |
slow | -0.027 |
very | -0.025 |
really | -0.023 |
Token | bisc |
Token ID | 36228 |
Feature activation | +9.90e+00 |
Pos logit contributions | |
lor | +0.066 |
ocket | +0.062 |
urger | +0.059 |
card | +0.056 |
oll | +0.056 |
Neg logit contributions | |
different | -0.024 |
fast | -0.023 |
slow | -0.021 |
very | -0.020 |
really | -0.019 |
Token | uit |
Token ID | 5013 |
Feature activation | +3.78e+00 |
Pos logit contributions | |
lor | +0.208 |
ocket | +0.204 |
Majesty | +0.188 |
urger | +0.167 |
passports | +0.146 |
Neg logit contributions | |
to | -0.427 |
very | -0.426 |
B | -0.423 |
fast | -0.418 |
| -0.418 |
Token | from |
Token ID | 422 |
Feature activation | +3.29e+00 |
Pos logit contributions | |
lor | +0.244 |
ocket | +0.229 |
urger | +0.216 |
oll | +0.212 |
Majesty | +0.211 |
Neg logit contributions | |
fast | -0.039 |
very | -0.034 |
to | -0.031 |
different | -0.030 |
slow | -0.028 |
Token | her |
Token ID | 607 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
lor | +0.190 |
ocket | +0.183 |
urger | +0.173 |
card | +0.165 |
oll | +0.165 |
Neg logit contributions | |
different | -0.042 |
fast | -0.040 |
slow | -0.035 |
very | -0.032 |
really | -0.030 |
Token | pocket |
Token ID | 10000 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
lor | +0.098 |
ocket | +0.087 |
urger | +0.082 |
oll | +0.078 |
card | +0.076 |
Neg logit contributions | |
different | -0.074 |
fast | -0.073 |
very | -0.070 |
slow | -0.069 |
really | -0.066 |
Token | and |
Token ID | 290 |
Feature activation | +1.86e+00 |
Pos logit contributions | |
lor | +0.088 |
ocket | +0.084 |
urger | +0.082 |
oll | +0.077 |
card | +0.075 |
Neg logit contributions | |
different | -0.017 |
fast | -0.017 |
very | -0.013 |
slow | -0.013 |
really | -0.011 |
Token | cl |
Token ID | 537 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
different | +0.012 |
fast | +0.011 |
slow | +0.011 |
Stop | +0.011 |
competitive | +0.011 |
Neg logit contributions | |
lor | -0.029 |
urger | -0.027 |
ocket | -0.027 |
ster | -0.026 |
oll | -0.026 |
Token | ink |
Token ID | 676 |
Feature activation | +5.97e+00 |
Pos logit contributions | |
lor | +0.052 |
ocket | +0.047 |
urger | +0.044 |
oll | +0.040 |
card | +0.040 |
Neg logit contributions | |
different | -0.051 |
fast | -0.050 |
slow | -0.047 |
very | -0.046 |
really | -0.045 |
Token | their |
Token ID | 511 |
Feature activation | +4.40e+00 |
Pos logit contributions | |
lor | +0.267 |
ocket | +0.248 |
urger | +0.234 |
oll | +0.217 |
card | +0.214 |
Neg logit contributions | |
fast | -0.172 |
different | -0.172 |
slow | -0.159 |
very | -0.158 |
really | -0.149 |
Token | cups |
Token ID | 14180 |
Feature activation | +5.00e+00 |
Pos logit contributions | |
lor | +0.198 |
ocket | +0.174 |
urger | +0.163 |
oll | +0.159 |
anger | +0.152 |
Neg logit contributions | |
different | -0.137 |
fast | -0.134 |
slow | -0.124 |
very | -0.124 |
really | -0.117 |
Token | and |
Token ID | 290 |
Feature activation | +3.08e+00 |
Pos logit contributions | |
lor | +0.317 |
ocket | +0.297 |
urger | +0.286 |
card | +0.269 |
oll | +0.269 |
Neg logit contributions | |
fast | -0.052 |
different | -0.052 |
very | -0.044 |
slow | -0.041 |
really | -0.036 |
Token | say |
Token ID | 910 |
Feature activation | +9.87e+00 |
Pos logit contributions | |
lor | +0.162 |
ocket | +0.144 |
urger | +0.136 |
anger | +0.131 |
oll | +0.130 |
Neg logit contributions | |
different | -0.075 |
fast | -0.074 |
slow | -0.067 |
very | -0.067 |
really | -0.062 |
Token | " |
Token ID | 366 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
lor | +0.545 |
ocket | +0.511 |
urger | +0.489 |
card | +0.463 |
oll | +0.462 |
Neg logit contributions | |
different | -0.187 |
fast | -0.183 |
very | -0.163 |
slow | -0.161 |
really | -0.149 |
Token | che |
Token ID | 2395 |
Feature activation | +6.74e+00 |
Pos logit contributions | |
different | +0.039 |
fast | +0.038 |
slow | +0.034 |
very | +0.033 |
really | +0.032 |
Neg logit contributions | |
lor | -0.099 |
ocket | -0.092 |
urger | -0.088 |
oll | -0.083 |
card | -0.082 |
Token | ers |
Token ID | 364 |
Feature activation | +3.70e-01 |
Pos logit contributions | |
lor | +0.272 |
ocket | +0.268 |
isc | +0.234 |
oll | +0.232 |
anger | +0.230 |
Neg logit contributions | |
fast | -0.187 |
slow | -0.183 |
different | -0.182 |
very | -0.182 |
really | -0.172 |
Token | ". |
Token ID | 1911 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
lor | +0.019 |
ocket | +0.018 |
urger | +0.017 |
oll | +0.017 |
card | +0.016 |
Neg logit contributions | |
different | -0.007 |
fast | -0.007 |
very | -0.007 |
slow | -0.007 |
really | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | -5.86e-02 |
Pos logit contributions | |
different | +0.016 |
fast | +0.016 |
slow | +0.014 |
very | +0.013 |
late | +0.012 |
Neg logit contributions | |
lor | -0.071 |
ocket | -0.068 |
urger | -0.065 |
oll | -0.062 |
card | -0.061 |
Token | belonged |
Token ID | 19611 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
different | +0.015 |
fast | +0.015 |
slow | +0.013 |
very | +0.012 |
really | +0.012 |
Neg logit contributions | |
lor | -0.041 |
ocket | -0.038 |
urger | -0.036 |
oll | -0.034 |
card | -0.034 |
Token | to |
Token ID | 284 |
Feature activation | +1.00e-02 |
Pos logit contributions | |
lor | +0.095 |
ocket | +0.089 |
urger | +0.086 |
oll | +0.082 |
card | +0.082 |
Neg logit contributions | |
different | -0.013 |
fast | -0.012 |
slow | -0.009 |
very | -0.008 |
really | -0.007 |
Token | his |
Token ID | 465 |
Feature activation | +2.28e+00 |
Pos logit contributions | |
lor | +0.001 |
ocket | +0.001 |
urger | +0.000 |
oll | +0.000 |
card | +0.000 |
Neg logit contributions | |
different | -0.000 |
fast | -0.000 |
slow | -0.000 |
very | -0.000 |
really | -0.000 |
Token | great |
Token ID | 1049 |
Feature activation | -2.39e+00 |
Pos logit contributions | |
lor | +0.100 |
ocket | +0.091 |
urger | +0.086 |
oll | +0.080 |
card | +0.080 |
Neg logit contributions | |
different | -0.067 |
fast | -0.066 |
slow | -0.062 |
very | -0.060 |
really | -0.058 |
Token | - |
Token ID | 12 |
Feature activation | -4.52e+00 |
Pos logit contributions | |
different | +0.074 |
fast | +0.072 |
slow | +0.067 |
very | +0.066 |
really | +0.064 |
Neg logit contributions | |
lor | -0.104 |
ocket | -0.094 |
urger | -0.089 |
card | -0.082 |
oll | -0.082 |
Token | grand |
Token ID | 23936 |
Feature activation | +9.87e+00 |
Pos logit contributions | |
different | +0.141 |
fast | +0.140 |
slow | +0.129 |
very | +0.127 |
really | +0.124 |
Neg logit contributions | |
lor | -0.191 |
ocket | -0.173 |
urger | -0.165 |
oll | -0.152 |
card | -0.149 |
Token | father |
Token ID | 11358 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
lor | +0.347 |
ocket | +0.314 |
urger | +0.296 |
anger | +0.294 |
oll | +0.279 |
Neg logit contributions | |
different | -0.349 |
very | -0.317 |
really | -0.311 |
fast | -0.302 |
the | -0.300 |
Token | , |
Token ID | 11 |
Feature activation | -9.55e-01 |
Pos logit contributions | |
lor | +0.132 |
ocket | +0.122 |
urger | +0.118 |
anger | +0.114 |
oll | +0.112 |
Neg logit contributions | |
different | -0.027 |
fast | -0.024 |
very | -0.023 |
really | -0.020 |
slow | -0.019 |
Token | and |
Token ID | 290 |
Feature activation | +8.03e-01 |
Pos logit contributions | |
different | +0.018 |
fast | +0.017 |
very | +0.015 |
slow | +0.015 |
really | +0.014 |
Neg logit contributions | |
lor | -0.054 |
ocket | -0.050 |
urger | -0.048 |
anger | -0.045 |
card | -0.045 |
Token | he |
Token ID | 339 |
Feature activation | +5.34e-01 |
Pos logit contributions | |
lor | +0.045 |
ocket | +0.041 |
urger | +0.039 |
anger | +0.038 |
oll | +0.037 |
Neg logit contributions | |
different | -0.017 |
fast | -0.016 |
very | -0.015 |
really | -0.014 |
slow | -0.014 |
Token | used |
Token ID | 973 |
Feature activation | +6.67e-02 |
Pos logit contributions | |
lor | +0.029 |
ocket | +0.027 |
urger | +0.026 |
anger | +0.025 |
oll | +0.025 |
Neg logit contributions | |
different | -0.010 |
fast | -0.010 |
very | -0.009 |
slow | -0.009 |
really | -0.009 |
Token | imagine |
Token ID | 5967 |
Feature activation | +3.53e+00 |
Pos logit contributions | |
fast | +0.023 |
different | +0.022 |
late | +0.022 |
slow | +0.021 |
very | +0.021 |
Neg logit contributions | |
lor | -0.033 |
ocket | -0.031 |
urger | -0.030 |
card | -0.028 |
ster | -0.026 |
Token | what |
Token ID | 644 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
lor | +0.199 |
ocket | +0.183 |
urger | +0.181 |
oll | +0.168 |
ster | +0.165 |
Neg logit contributions | |
different | -0.070 |
fast | -0.066 |
very | -0.060 |
slow | -0.056 |
really | -0.053 |
Token | her |
Token ID | 607 |
Feature activation | +2.47e+00 |
Pos logit contributions | |
lor | +0.052 |
ocket | +0.049 |
urger | +0.045 |
card | +0.045 |
oll | +0.044 |
Neg logit contributions | |
fast | -0.021 |
different | -0.019 |
slow | -0.019 |
late | -0.018 |
very | -0.016 |
Token | great |
Token ID | 1049 |
Feature activation | -2.69e+00 |
Pos logit contributions | |
lor | +0.123 |
ocket | +0.105 |
urger | +0.102 |
oll | +0.095 |
bridge | +0.093 |
Neg logit contributions | |
different | -0.072 |
fast | -0.069 |
very | -0.066 |
slow | -0.063 |
really | -0.060 |
Token | - |
Token ID | 12 |
Feature activation | -4.28e+00 |
Pos logit contributions | |
different | +0.067 |
fast | +0.060 |
very | +0.057 |
slow | +0.055 |
really | +0.055 |
Neg logit contributions | |
lor | -0.143 |
ocket | -0.129 |
urger | -0.122 |
oll | -0.119 |
ogo | -0.113 |
Token | grand |
Token ID | 23936 |
Feature activation | +9.84e+00 |
Pos logit contributions | |
different | +0.175 |
fast | +0.174 |
very | +0.163 |
slow | +0.162 |
really | +0.159 |
Neg logit contributions | |
lor | -0.140 |
ocket | -0.123 |
urger | -0.118 |
oll | -0.105 |
ster | -0.100 |
Token | ma |
Token ID | 2611 |
Feature activation | -6.57e-01 |
Pos logit contributions | |
lor | +0.342 |
urger | +0.308 |
ocket | +0.303 |
oll | +0.279 |
anger | +0.277 |
Neg logit contributions | |
different | -0.348 |
very | -0.316 |
the | -0.306 |
her | -0.306 |
really | -0.303 |
Token | was |
Token ID | 373 |
Feature activation | +7.98e-02 |
Pos logit contributions | |
different | +0.011 |
fast | +0.010 |
very | +0.010 |
really | +0.010 |
slow | +0.009 |
Neg logit contributions | |
lor | -0.039 |
ocket | -0.034 |
urger | -0.034 |
anger | -0.032 |
oll | -0.032 |
Token | like |
Token ID | 588 |
Feature activation | +5.97e-02 |
Pos logit contributions | |
lor | +0.004 |
ocket | +0.004 |
urger | +0.004 |
oll | +0.004 |
anger | +0.004 |
Neg logit contributions | |
different | -0.002 |
fast | -0.002 |
very | -0.001 |
slow | -0.001 |
really | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -4.64e-01 |
Pos logit contributions | |
lor | +0.004 |
ocket | +0.003 |
urger | +0.003 |
oll | +0.003 |
ster | +0.003 |
Neg logit contributions | |
different | -0.001 |
fast | -0.001 |
very | -0.001 |
slow | -0.001 |
really | -0.001 |
Token | She |
Token ID | 1375 |
Feature activation | +8.15e-01 |
Pos logit contributions | |
different | +0.007 |
fast | +0.006 |
slow | +0.005 |
very | +0.005 |
really | +0.005 |
Neg logit contributions | |
lor | -0.028 |
ocket | -0.026 |
urger | -0.025 |
oll | -0.024 |
card | -0.023 |
Token | She |
Token ID | 1375 |
Feature activation | +9.42e-01 |
Pos logit contributions | |
different | +0.007 |
fast | +0.007 |
slow | +0.006 |
very | +0.005 |
really | +0.005 |
Neg logit contributions | |
lor | -0.029 |
ocket | -0.027 |
urger | -0.026 |
oll | -0.025 |
card | -0.025 |
Token | imagined |
Token ID | 15758 |
Feature activation | +4.49e+00 |
Pos logit contributions | |
lor | +0.049 |
ocket | +0.044 |
urger | +0.043 |
card | +0.040 |
anger | +0.040 |
Neg logit contributions | |
different | -0.022 |
fast | -0.021 |
very | -0.019 |
slow | -0.019 |
really | -0.019 |
Token | her |
Token ID | 607 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
lor | +0.248 |
urger | +0.230 |
ocket | +0.227 |
oll | +0.209 |
anger | +0.205 |
Neg logit contributions | |
different | -0.096 |
fast | -0.094 |
very | -0.086 |
slow | -0.076 |
really | -0.076 |
Token | great |
Token ID | 1049 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
lor | +0.146 |
ocket | +0.124 |
urger | +0.123 |
oll | +0.117 |
bridge | +0.112 |
Neg logit contributions | |
different | -0.070 |
fast | -0.065 |
very | -0.064 |
slow | -0.059 |
fun | -0.056 |
Token | - |
Token ID | 12 |
Feature activation | -3.82e+00 |
Pos logit contributions | |
different | +0.055 |
fast | +0.051 |
very | +0.050 |
slow | +0.047 |
really | +0.047 |
Neg logit contributions | |
lor | -0.086 |
ocket | -0.077 |
urger | -0.071 |
oll | -0.069 |
cher | -0.067 |
Token | grand |
Token ID | 23936 |
Feature activation | +9.81e+00 |
Pos logit contributions | |
fast | +0.131 |
different | +0.130 |
very | +0.120 |
slow | +0.120 |
really | +0.117 |
Neg logit contributions | |
lor | -0.149 |
ocket | -0.134 |
urger | -0.131 |
oll | -0.119 |
ster | -0.115 |
Token | ma |
Token ID | 2611 |
Feature activation | -5.52e-01 |
Pos logit contributions | |
lor | +0.341 |
ocket | +0.309 |
urger | +0.304 |
oll | +0.275 |
anger | +0.268 |
Neg logit contributions | |
different | -0.354 |
very | -0.322 |
fast | -0.316 |
really | -0.313 |
fun | -0.307 |
Token | wearing |
Token ID | 5762 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
different | +0.009 |
fast | +0.009 |
very | +0.009 |
really | +0.008 |
slow | +0.008 |
Neg logit contributions | |
lor | -0.032 |
ocket | -0.029 |
urger | -0.029 |
anger | -0.027 |
oll | -0.027 |
Token | the |
Token ID | 262 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
lor | +0.064 |
ocket | +0.059 |
urger | +0.058 |
oll | +0.055 |
card | +0.054 |
Neg logit contributions | |
different | -0.012 |
fast | -0.011 |
slow | -0.009 |
very | -0.009 |
really | -0.008 |
Token | same |
Token ID | 976 |
Feature activation | -5.37e-01 |
Pos logit contributions | |
lor | +0.055 |
urger | +0.051 |
ocket | +0.049 |
oll | +0.047 |
icking | +0.045 |
Neg logit contributions | |
different | -0.026 |
fast | -0.023 |
very | -0.021 |
slow | -0.021 |
really | -0.019 |
Token | dress |
Token ID | 6576 |
Feature activation | +3.05e+00 |
Pos logit contributions | |
different | +0.020 |
fast | +0.019 |
very | +0.018 |
slow | +0.018 |
really | +0.017 |
Neg logit contributions | |
lor | -0.021 |
ocket | -0.019 |
urger | -0.018 |
oll | -0.017 |
icking | -0.016 |
Token | water |
Token ID | 1660 |
Feature activation | +4.71e+00 |
Pos logit contributions | |
ocket | +0.043 |
lor | +0.041 |
urger | +0.039 |
card | +0.039 |
ster | +0.037 |
Neg logit contributions | |
different | -0.027 |
B | -0.027 |
fast | -0.027 |
Da | -0.026 |
late | -0.026 |
Token | into |
Token ID | 656 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
lor | +0.256 |
ocket | +0.238 |
urger | +0.229 |
card | +0.216 |
oll | +0.216 |
Neg logit contributions | |
different | -0.083 |
fast | -0.081 |
slow | -0.072 |
very | -0.067 |
really | -0.064 |
Token | the |
Token ID | 262 |
Feature activation | +9.93e-01 |
Pos logit contributions | |
fast | +0.015 |
different | +0.014 |
slow | +0.014 |
very | +0.013 |
really | +0.013 |
Neg logit contributions | |
lor | -0.067 |
ocket | -0.063 |
urger | -0.059 |
bridge | -0.056 |
anger | -0.056 |
Token | cups |
Token ID | 14180 |
Feature activation | +3.42e+00 |
Pos logit contributions | |
lor | +0.042 |
ocket | +0.039 |
urger | +0.037 |
card | +0.034 |
oll | +0.034 |
Neg logit contributions | |
different | -0.029 |
fast | -0.029 |
slow | -0.027 |
very | -0.026 |
really | -0.025 |
Token | and |
Token ID | 290 |
Feature activation | +2.77e+00 |
Pos logit contributions | |
lor | +0.210 |
ocket | +0.199 |
urger | +0.191 |
oll | +0.181 |
card | +0.180 |
Neg logit contributions | |
different | -0.040 |
fast | -0.039 |
very | -0.033 |
slow | -0.032 |
really | -0.029 |
Token | say |
Token ID | 910 |
Feature activation | +9.70e+00 |
Pos logit contributions | |
lor | +0.144 |
ocket | +0.129 |
urger | +0.122 |
anger | +0.115 |
oll | +0.115 |
Neg logit contributions | |
different | -0.068 |
fast | -0.067 |
slow | -0.062 |
very | -0.060 |
really | -0.057 |
Token | " |
Token ID | 366 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
lor | +0.525 |
ocket | +0.489 |
urger | +0.467 |
card | +0.442 |
oll | +0.440 |
Neg logit contributions | |
different | -0.194 |
fast | -0.189 |
slow | -0.168 |
very | -0.168 |
really | -0.156 |
Token | che |
Token ID | 2395 |
Feature activation | +6.59e+00 |
Pos logit contributions | |
different | +0.040 |
fast | +0.039 |
slow | +0.035 |
very | +0.034 |
really | +0.032 |
Neg logit contributions | |
lor | -0.107 |
ocket | -0.099 |
urger | -0.094 |
card | -0.089 |
oll | -0.089 |
Token | ers |
Token ID | 364 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
lor | +0.258 |
ocket | +0.250 |
isc | +0.214 |
anger | +0.212 |
oll | +0.208 |
Neg logit contributions | |
fast | -0.193 |
slow | -0.191 |
different | -0.191 |
very | -0.189 |
really | -0.183 |
Token | " |
Token ID | 1 |
Feature activation | -8.67e-01 |
Pos logit contributions | |
lor | +0.013 |
ocket | +0.012 |
urger | +0.011 |
oll | +0.011 |
card | +0.011 |
Neg logit contributions | |
different | -0.005 |
fast | -0.005 |
very | -0.004 |
slow | -0.004 |
really | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | +5.52e-01 |
Pos logit contributions | |
different | +0.017 |
fast | +0.017 |
slow | +0.015 |
very | +0.015 |
really | +0.014 |
Neg logit contributions | |
lor | -0.046 |
ocket | -0.042 |
urger | -0.041 |
oll | -0.038 |
card | -0.038 |
Token | was |
Token ID | 373 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
lor | +0.205 |
ocket | +0.189 |
urger | +0.182 |
card | +0.172 |
oll | +0.172 |
Neg logit contributions | |
different | -0.072 |
fast | -0.071 |
slow | -0.063 |
very | -0.062 |
really | -0.059 |
Token | going |
Token ID | 1016 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
lor | +0.074 |
ocket | +0.068 |
urger | +0.065 |
oll | +0.060 |
anger | +0.060 |
Neg logit contributions | |
different | -0.035 |
fast | -0.035 |
very | -0.032 |
slow | -0.032 |
really | -0.030 |
Token | on |
Token ID | 319 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
lor | +0.126 |
ocket | +0.120 |
urger | +0.115 |
oll | +0.111 |
ster | +0.108 |
Neg logit contributions | |
fast | -0.013 |
different | -0.013 |
very | -0.010 |
slow | -0.009 |
really | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | +9.91e-01 |
Pos logit contributions | |
fast | +0.023 |
different | +0.023 |
slow | +0.021 |
very | +0.018 |
B | +0.018 |
Neg logit contributions | |
lor | -0.076 |
ocket | -0.068 |
urger | -0.066 |
oll | -0.062 |
card | -0.061 |
Token | special |
Token ID | 2041 |
Feature activation | +7.28e-01 |
Pos logit contributions | |
lor | +0.041 |
ocket | +0.037 |
urger | +0.035 |
card | +0.032 |
oll | +0.032 |
Neg logit contributions | |
different | -0.029 |
fast | -0.029 |
slow | -0.027 |
late | -0.027 |
very | -0.026 |
Token | saf |
Token ID | 1932 |
Feature activation | +9.57e+00 |
Pos logit contributions | |
lor | +0.028 |
ocket | +0.026 |
urger | +0.024 |
oll | +0.023 |
card | +0.022 |
Neg logit contributions | |
different | -0.026 |
fast | -0.025 |
slow | -0.024 |
very | -0.024 |
really | -0.023 |
Token | ari |
Token ID | 2743 |
Feature activation | +5.37e+00 |
Pos logit contributions | |
ocket | +0.192 |
lor | +0.189 |
urger | +0.176 |
oll | +0.133 |
card | +0.126 |
Neg logit contributions | |
fast | -0.491 |
different | -0.469 |
very | -0.458 |
slow | -0.454 |
really | -0.451 |
Token | . |
Token ID | 13 |
Feature activation | +3.00e-01 |
Pos logit contributions | |
lor | +0.329 |
ocket | +0.317 |
urger | +0.306 |
oll | +0.294 |
anger | +0.282 |
Neg logit contributions | |
different | -0.067 |
fast | -0.066 |
very | -0.056 |
really | -0.049 |
slow | -0.049 |
Token | He |
Token ID | 679 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
lor | +0.018 |
ocket | +0.017 |
urger | +0.016 |
card | +0.015 |
ster | +0.015 |
Neg logit contributions | |
different | -0.004 |
fast | -0.004 |
slow | -0.003 |
really | -0.003 |
very | -0.003 |
Token | got |
Token ID | 1392 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
lor | +0.096 |
ocket | +0.089 |
urger | +0.085 |
card | +0.081 |
oll | +0.081 |
Neg logit contributions | |
different | -0.034 |
fast | -0.033 |
slow | -0.029 |
very | -0.029 |
really | -0.029 |
Token | to |
Token ID | 284 |
Feature activation | -3.03e-01 |
Pos logit contributions | |
different | +0.030 |
fast | +0.029 |
very | +0.026 |
slow | +0.025 |
really | +0.024 |
Neg logit contributions | |
lor | -0.114 |
ocket | -0.104 |
urger | -0.100 |
oll | -0.094 |
anger | -0.094 |
Token | , |
Token ID | 11 |
Feature activation | +2.16e+00 |
Pos logit contributions | |
different | +0.014 |
fast | +0.014 |
slow | +0.013 |
very | +0.012 |
really | +0.011 |
Neg logit contributions | |
lor | -0.041 |
ocket | -0.038 |
urger | -0.037 |
oll | -0.035 |
card | -0.034 |
Token | what |
Token ID | 644 |
Feature activation | +3.59e+00 |
Pos logit contributions | |
lor | +0.118 |
ocket | +0.109 |
urger | +0.104 |
oll | +0.099 |
card | +0.098 |
Neg logit contributions | |
different | -0.042 |
fast | -0.042 |
slow | -0.037 |
very | -0.036 |
really | -0.034 |
Token | does |
Token ID | 857 |
Feature activation | +2.73e+00 |
Pos logit contributions | |
lor | +0.164 |
ocket | +0.153 |
urger | +0.144 |
oll | +0.137 |
card | +0.132 |
Neg logit contributions | |
fast | -0.092 |
different | -0.089 |
slow | -0.084 |
late | -0.078 |
very | -0.078 |
Token | that |
Token ID | 326 |
Feature activation | +2.99e+00 |
Pos logit contributions | |
lor | +0.142 |
ocket | +0.133 |
urger | +0.127 |
oll | +0.121 |
card | +0.119 |
Neg logit contributions | |
different | -0.062 |
fast | -0.060 |
slow | -0.054 |
very | -0.053 |
really | -0.050 |
Token | sign |
Token ID | 1051 |
Feature activation | +2.59e+00 |
Pos logit contributions | |
lor | +0.136 |
ocket | +0.125 |
urger | +0.119 |
oll | +0.111 |
card | +0.110 |
Neg logit contributions | |
different | -0.082 |
fast | -0.081 |
slow | -0.075 |
very | -0.073 |
really | -0.071 |
Token | say |
Token ID | 910 |
Feature activation | +9.55e+00 |
Pos logit contributions | |
lor | +0.122 |
ocket | +0.112 |
urger | +0.107 |
oll | +0.100 |
card | +0.099 |
Neg logit contributions | |
different | -0.067 |
fast | -0.066 |
slow | -0.060 |
very | -0.059 |
really | -0.057 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
lor | +0.529 |
ocket | +0.491 |
urger | +0.478 |
oll | +0.451 |
card | +0.443 |
Neg logit contributions | |
different | -0.177 |
fast | -0.174 |
very | -0.157 |
slow | -0.152 |
really | -0.143 |
Token | she |
Token ID | 673 |
Feature activation | -8.96e-02 |
Pos logit contributions | |
different | +0.026 |
fast | +0.025 |
slow | +0.022 |
very | +0.021 |
really | +0.020 |
Neg logit contributions | |
lor | -0.072 |
ocket | -0.067 |
urger | -0.063 |
oll | -0.060 |
ster | -0.060 |
Token | asked |
Token ID | 1965 |
Feature activation | +3.14e+00 |
Pos logit contributions | |
different | +0.002 |
fast | +0.002 |
slow | +0.002 |
very | +0.002 |
really | +0.002 |
Neg logit contributions | |
lor | -0.004 |
ocket | -0.004 |
urger | -0.004 |
oll | -0.004 |
card | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
lor | +0.175 |
ocket | +0.166 |
urger | +0.159 |
oll | +0.151 |
card | +0.149 |
Neg logit contributions | |
different | -0.050 |
fast | -0.049 |
slow | -0.042 |
late | -0.039 |
very | -0.037 |
Token | " |
Token ID | 366 |
Feature activation | -3.52e+00 |
Pos logit contributions | |
lor | +0.055 |
ocket | +0.050 |
urger | +0.048 |
oll | +0.046 |
anger | +0.046 |
Neg logit contributions | |
different | -0.016 |
fast | -0.015 |
slow | -0.013 |
very | -0.013 |
really | -0.012 |
Token | frown |
Token ID | 22002 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
lor | +0.168 |
ocket | +0.160 |
urger | +0.159 |
card | +0.152 |
ster | +0.144 |
Neg logit contributions | |
fast | -0.041 |
late | -0.039 |
different | -0.036 |
slow | -0.036 |
Stop | -0.035 |
Token | s |
Token ID | 82 |
Feature activation | +4.93e+00 |
Pos logit contributions | |
lor | +0.097 |
ocket | +0.091 |
urger | +0.088 |
oll | +0.085 |
ster | +0.082 |
Neg logit contributions | |
fast | -0.056 |
different | -0.055 |
slow | -0.052 |
busy | -0.047 |
late | -0.046 |
Token | . |
Token ID | 13 |
Feature activation | +2.29e+00 |
Pos logit contributions | |
lor | +0.209 |
ocket | +0.188 |
urger | +0.177 |
oll | +0.165 |
card | +0.165 |
Neg logit contributions | |
different | -0.155 |
fast | -0.152 |
slow | -0.143 |
very | -0.142 |
really | -0.137 |
Token | Their |
Token ID | 5334 |
Feature activation | +4.94e+00 |
Pos logit contributions | |
lor | +0.130 |
ocket | +0.121 |
urger | +0.115 |
oll | +0.110 |
card | +0.109 |
Neg logit contributions | |
different | -0.037 |
fast | -0.036 |
slow | -0.031 |
very | -0.029 |
really | -0.028 |
Token | moms |
Token ID | 34015 |
Feature activation | +6.81e-01 |
Pos logit contributions | |
lor | +0.203 |
ocket | +0.179 |
urger | +0.167 |
oll | +0.156 |
card | +0.155 |
Neg logit contributions | |
different | -0.167 |
fast | -0.164 |
slow | -0.154 |
very | -0.153 |
really | -0.147 |
Token | say |
Token ID | 910 |
Feature activation | +9.46e+00 |
Pos logit contributions | |
lor | +0.033 |
ocket | +0.031 |
urger | +0.029 |
oll | +0.027 |
card | +0.027 |
Neg logit contributions | |
different | -0.017 |
fast | -0.017 |
slow | -0.015 |
very | -0.015 |
really | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
lor | +0.519 |
ocket | +0.482 |
urger | +0.462 |
oll | +0.435 |
card | +0.434 |
Neg logit contributions | |
different | -0.190 |
fast | -0.185 |
slow | -0.165 |
very | -0.165 |
really | -0.152 |
Token | " |
Token ID | 366 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
lor | +0.056 |
ocket | +0.052 |
urger | +0.049 |
oll | +0.047 |
card | +0.046 |
Neg logit contributions | |
different | -0.022 |
fast | -0.022 |
slow | -0.020 |
very | -0.019 |
really | -0.018 |
Token | Hello |
Token ID | 15496 |
Feature activation | +2.77e+00 |
Pos logit contributions | |
different | +0.053 |
fast | +0.051 |
very | +0.049 |
slow | +0.046 |
really | +0.044 |
Neg logit contributions | |
lor | -0.124 |
ocket | -0.106 |
urger | -0.101 |
card | -0.100 |
ster | -0.099 |
Token | . |
Token ID | 13 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
lor | +0.159 |
ocket | +0.148 |
urger | +0.142 |
oll | +0.135 |
card | +0.134 |
Neg logit contributions | |
different | -0.045 |
fast | -0.044 |
slow | -0.038 |
very | -0.036 |
really | -0.034 |
Token | We |
Token ID | 775 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
lor | +0.056 |
ocket | +0.052 |
urger | +0.050 |
oll | +0.046 |
card | +0.046 |
Neg logit contributions | |
different | -0.031 |
fast | -0.031 |
slow | -0.028 |
very | -0.027 |
really | -0.027 |
Token | special |
Token ID | 2041 |
Feature activation | +2.91e+00 |
Pos logit contributions | |
lor | +0.167 |
ocket | +0.154 |
urger | +0.147 |
oll | +0.139 |
card | +0.138 |
Neg logit contributions | |
different | -0.068 |
fast | -0.067 |
slow | -0.060 |
very | -0.058 |
really | -0.055 |
Token | place |
Token ID | 1295 |
Feature activation | +3.05e+00 |
Pos logit contributions | |
lor | +0.125 |
ocket | +0.112 |
urger | +0.107 |
oll | +0.104 |
ogo | +0.096 |
Neg logit contributions | |
different | -0.093 |
fast | -0.090 |
very | -0.085 |
slow | -0.083 |
really | -0.081 |
Token | in |
Token ID | 287 |
Feature activation | -2.65e+00 |
Pos logit contributions | |
lor | +0.179 |
ocket | +0.167 |
urger | +0.161 |
oll | +0.154 |
card | +0.151 |
Neg logit contributions | |
different | -0.046 |
fast | -0.045 |
slow | -0.038 |
very | -0.037 |
really | -0.034 |
Token | Santa |
Token ID | 8909 |
Feature activation | +3.49e+00 |
Pos logit contributions | |
different | +0.027 |
fast | +0.025 |
slow | +0.020 |
very | +0.018 |
really | +0.016 |
Neg logit contributions | |
lor | -0.168 |
ocket | -0.158 |
urger | -0.153 |
oll | -0.146 |
card | -0.145 |
Token | 's |
Token ID | 338 |
Feature activation | +3.05e+00 |
Pos logit contributions | |
lor | +0.201 |
ocket | +0.189 |
oll | +0.179 |
urger | +0.177 |
anger | +0.174 |
Neg logit contributions | |
different | -0.056 |
fast | -0.056 |
very | -0.052 |
slow | -0.050 |
really | -0.047 |
Token | sle |
Token ID | 3133 |
Feature activation | +9.43e+00 |
Pos logit contributions | |
lor | +0.138 |
ocket | +0.126 |
urger | +0.119 |
oll | +0.117 |
anger | +0.113 |
Neg logit contributions | |
different | -0.092 |
fast | -0.088 |
very | -0.087 |
slow | -0.082 |
quiet | -0.078 |
Token | igh |
Token ID | 394 |
Feature activation | +4.71e+00 |
Pos logit contributions | |
lor | +0.216 |
ocket | +0.171 |
urger | +0.163 |
card | +0.155 |
anger | +0.154 |
Neg logit contributions | |
fast | -0.410 |
very | -0.398 |
different | -0.396 |
to | -0.381 |
slow | -0.380 |
Token | . |
Token ID | 13 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
lor | +0.279 |
ocket | +0.264 |
urger | +0.252 |
oll | +0.242 |
card | +0.239 |
Neg logit contributions | |
fast | -0.064 |
different | -0.064 |
slow | -0.053 |
very | -0.053 |
really | -0.047 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +4.91e+00 |
Pos logit contributions | |
lor | +0.061 |
ocket | +0.060 |
urger | +0.057 |
card | +0.054 |
ster | +0.053 |
Neg logit contributions | |
different | -0.014 |
fast | -0.014 |
slow | -0.012 |
late | -0.012 |
very | -0.011 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.95e+00 |
Pos logit contributions | |
lor | +0.275 |
ocket | +0.250 |
urger | +0.235 |
oll | +0.227 |
anger | +0.224 |
Neg logit contributions | |
different | -0.084 |
fast | -0.081 |
slow | -0.073 |
very | -0.068 |
really | -0.063 |
Token | upon |
Token ID | 2402 |
Feature activation | +2.91e+00 |
Pos logit contributions | |
lor | +0.210 |
ocket | +0.192 |
urger | +0.181 |
oll | +0.169 |
card | +0.167 |
Neg logit contributions | |
different | -0.147 |
fast | -0.146 |
slow | -0.137 |
very | -0.132 |
really | -0.129 |
Token | share |
Token ID | 2648 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
lor | +0.140 |
ocket | +0.129 |
urger | +0.123 |
oll | +0.115 |
card | +0.115 |
Neg logit contributions | |
different | -0.070 |
fast | -0.069 |
slow | -0.063 |
very | -0.062 |
really | -0.059 |
Token | and |
Token ID | 290 |
Feature activation | +8.21e-01 |
Pos logit contributions | |
fast | +0.034 |
late | +0.034 |
Da | +0.033 |
B | +0.033 |
slow | +0.031 |
Neg logit contributions | |
lor | -0.079 |
ocket | -0.077 |
urger | -0.074 |
card | -0.069 |
ster | -0.068 |
Token | be |
Token ID | 307 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
lor | +0.034 |
ocket | +0.032 |
urger | +0.031 |
card | +0.030 |
ster | +0.028 |
Neg logit contributions | |
fast | -0.020 |
different | -0.019 |
late | -0.018 |
slow | -0.018 |
very | -0.017 |
Token | kind |
Token ID | 1611 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
heavily | +0.038 |
wildly | +0.033 |
different | +0.033 |
panic | +0.032 |
Da | +0.032 |
Neg logit contributions | |
card | -0.066 |
lor | -0.065 |
urger | -0.062 |
ocket | -0.062 |
ster | -0.058 |
Token | . |
Token ID | 13 |
Feature activation | +8.91e-01 |
Pos logit contributions | |
different | +0.039 |
fast | +0.036 |
slow | +0.033 |
really | +0.030 |
very | +0.030 |
Neg logit contributions | |
lor | -0.107 |
ocket | -0.100 |
urger | -0.097 |
card | -0.091 |
ster | -0.090 |
Token | Say |
Token ID | 13816 |
Feature activation | +9.42e+00 |
Pos logit contributions | |
lor | +0.039 |
ocket | +0.035 |
urger | +0.033 |
oll | +0.031 |
card | +0.031 |
Neg logit contributions | |
different | -0.026 |
fast | -0.026 |
slow | -0.024 |
very | -0.023 |
really | -0.023 |
Token | sorry |
Token ID | 7926 |
Feature activation | -7.22e-01 |
Pos logit contributions | |
lor | +0.416 |
ocket | +0.376 |
urger | +0.353 |
oll | +0.329 |
card | +0.326 |
Neg logit contributions | |
different | -0.288 |
fast | -0.287 |
slow | -0.268 |
very | -0.262 |
really | -0.250 |
Token | to |
Token ID | 284 |
Feature activation | +7.39e-01 |
Pos logit contributions | |
different | +0.012 |
fast | +0.012 |
slow | +0.010 |
very | +0.009 |
busy | +0.009 |
Neg logit contributions | |
lor | -0.038 |
ocket | -0.035 |
urger | -0.035 |
ster | -0.033 |
oll | -0.033 |
Token | Tom |
Token ID | 4186 |
Feature activation | -6.42e-02 |
Pos logit contributions | |
lor | +0.037 |
ocket | +0.037 |
urger | +0.036 |
card | +0.033 |
ster | +0.033 |
Neg logit contributions | |
fast | -0.009 |
different | -0.009 |
late | -0.009 |
B | -0.008 |
very | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
different | +0.001 |
fast | +0.001 |
slow | +0.001 |
very | +0.001 |
really | +0.001 |
Neg logit contributions | |
lor | -0.004 |
ocket | -0.003 |
urger | -0.003 |
oll | -0.003 |
card | -0.003 |
Token | Lily |
Token ID | 20037 |
Feature activation | -7.15e-01 |
Pos logit contributions | |
lor | +0.090 |
ocket | +0.089 |
urger | +0.085 |
oll | +0.080 |
card | +0.080 |
Neg logit contributions | |
different | -0.029 |
fast | -0.029 |
very | -0.025 |
slow | -0.025 |
really | -0.025 |
Token | a |
Token ID | 257 |
Feature activation | -2.54e-01 |
Pos logit contributions | |
different | +0.001 |
fast | +0.001 |
slow | +0.001 |
very | +0.001 |
really | +0.001 |
Neg logit contributions | |
lor | -0.003 |
ocket | -0.003 |
urger | -0.003 |
oll | -0.003 |
card | -0.003 |
Token | voice |
Token ID | 3809 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
different | +0.006 |
fast | +0.006 |
late | +0.006 |
slow | +0.005 |
each | +0.005 |
Neg logit contributions | |
lor | -0.011 |
ocket | -0.010 |
urger | -0.010 |
ster | -0.009 |
card | -0.009 |
Token | inside |
Token ID | 2641 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
lor | +0.012 |
ocket | +0.011 |
urger | +0.011 |
oll | +0.010 |
anger | +0.010 |
Neg logit contributions | |
different | -0.005 |
fast | -0.005 |
very | -0.005 |
slow | -0.005 |
really | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | +6.81e-01 |
Pos logit contributions | |
different | +0.016 |
fast | +0.016 |
slow | +0.013 |
very | +0.012 |
late | +0.011 |
Neg logit contributions | |
lor | -0.086 |
ocket | -0.081 |
urger | -0.077 |
card | -0.074 |
oll | -0.073 |
Token | sack |
Token ID | 23704 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
lor | +0.023 |
ocket | +0.021 |
urger | +0.019 |
card | +0.017 |
oll | +0.017 |
Neg logit contributions | |
different | -0.027 |
fast | -0.026 |
slow | -0.025 |
very | -0.024 |
really | -0.024 |
Token | say |
Token ID | 910 |
Feature activation | +9.42e+00 |
Pos logit contributions | |
lor | +0.170 |
urger | +0.169 |
ocket | +0.162 |
oll | +0.156 |
Majesty | +0.154 |
Neg logit contributions | |
different | -0.027 |
fast | -0.024 |
very | -0.023 |
slow | -0.018 |
really | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | +8.45e-02 |
Pos logit contributions | |
lor | +0.518 |
ocket | +0.486 |
urger | +0.472 |
oll | +0.443 |
card | +0.433 |
Neg logit contributions | |
different | -0.172 |
fast | -0.169 |
very | -0.153 |
slow | -0.148 |
really | -0.138 |
Token | " |
Token ID | 366 |
Feature activation | -3.75e+00 |
Pos logit contributions | |
lor | +0.004 |
ocket | +0.004 |
urger | +0.004 |
card | +0.004 |
oll | +0.003 |
Neg logit contributions | |
different | -0.002 |
fast | -0.002 |
slow | -0.002 |
very | -0.002 |
late | -0.002 |
Token | Who |
Token ID | 8241 |
Feature activation | +3.95e+00 |
Pos logit contributions | |
different | +0.066 |
fast | +0.065 |
slow | +0.057 |
very | +0.056 |
really | +0.051 |
Neg logit contributions | |
lor | -0.210 |
ocket | -0.193 |
urger | -0.185 |
card | -0.176 |
oll | -0.174 |
Token | 's |
Token ID | 338 |
Feature activation | +2.96e+00 |
Pos logit contributions | |
lor | +0.212 |
urger | +0.194 |
ocket | +0.193 |
card | +0.184 |
anger | +0.177 |
Neg logit contributions | |
different | -0.080 |
fast | -0.072 |
really | -0.068 |
slow | -0.067 |
very | -0.067 |
Token | there |
Token ID | 612 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
lor | +0.175 |
urger | +0.162 |
ocket | +0.162 |
oll | +0.149 |
isc | +0.146 |
Neg logit contributions | |
different | -0.055 |
fast | -0.055 |
the | -0.052 |
very | -0.051 |
slow | -0.050 |
Token | , |
Token ID | 11 |
Feature activation | -4.80e-01 |
Pos logit contributions | |
lor | +0.102 |
ocket | +0.096 |
urger | +0.090 |
oll | +0.087 |
card | +0.086 |
Neg logit contributions | |
different | -0.035 |
fast | -0.033 |
slow | -0.030 |
B | -0.029 |
panic | -0.028 |
Token | not |
Token ID | 407 |
Feature activation | -8.37e-01 |
Pos logit contributions | |
different | +0.010 |
fast | +0.010 |
slow | +0.009 |
very | +0.009 |
really | +0.008 |
Neg logit contributions | |
lor | -0.026 |
ocket | -0.023 |
urger | -0.022 |
oll | -0.021 |
card | -0.021 |
Token | your |
Token ID | 534 |
Feature activation | +2.70e+00 |
Pos logit contributions | |
fast | +0.016 |
different | +0.016 |
slow | +0.014 |
very | +0.013 |
late | +0.013 |
Neg logit contributions | |
lor | -0.042 |
ocket | -0.040 |
urger | -0.039 |
oll | -0.036 |
card | -0.036 |
Token | hands |
Token ID | 2832 |
Feature activation | -4.52e-02 |
Pos logit contributions | |
lor | +0.115 |
ocket | +0.104 |
urger | +0.096 |
oll | +0.091 |
card | +0.090 |
Neg logit contributions | |
different | -0.086 |
fast | -0.085 |
slow | -0.079 |
very | -0.078 |
really | -0.075 |
Token | . |
Token ID | 13 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
different | +0.001 |
fast | +0.001 |
slow | +0.001 |
very | +0.001 |
late | +0.001 |
Neg logit contributions | |
lor | -0.002 |
ocket | -0.002 |
urger | -0.002 |
oll | -0.002 |
card | -0.002 |
Token | Say |
Token ID | 13816 |
Feature activation | +9.40e+00 |
Pos logit contributions | |
lor | +0.053 |
ocket | +0.049 |
urger | +0.046 |
card | +0.043 |
oll | +0.043 |
Neg logit contributions | |
different | -0.029 |
fast | -0.028 |
slow | -0.026 |
very | -0.025 |
really | -0.025 |
Token | sorry |
Token ID | 7926 |
Feature activation | -7.16e-01 |
Pos logit contributions | |
lor | +0.447 |
ocket | +0.405 |
urger | +0.379 |
oll | +0.357 |
card | +0.356 |
Neg logit contributions | |
fast | -0.261 |
different | -0.261 |
slow | -0.242 |
very | -0.237 |
really | -0.223 |
Token | to |
Token ID | 284 |
Feature activation | +7.17e-01 |
Pos logit contributions | |
different | +0.010 |
fast | +0.010 |
slow | +0.008 |
very | +0.007 |
late | +0.007 |
Neg logit contributions | |
lor | -0.041 |
ocket | -0.038 |
urger | -0.038 |
oll | -0.035 |
ster | -0.035 |
Token | each |
Token ID | 1123 |
Feature activation | -4.03e+00 |
Pos logit contributions | |
lor | +0.034 |
ocket | +0.034 |
urger | +0.033 |
card | +0.030 |
oll | +0.030 |
Neg logit contributions | |
fast | -0.011 |
different | -0.011 |
late | -0.011 |
very | -0.010 |
B | -0.010 |
Token | other |
Token ID | 584 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
fast | +0.062 |
different | +0.061 |
slow | +0.055 |
late | +0.053 |
very | +0.049 |
Neg logit contributions | |
lor | -0.216 |
ocket | -0.203 |
urger | -0.197 |
card | -0.185 |
ster | -0.183 |
Token | and |
Token ID | 290 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
different | +0.021 |
fast | +0.020 |
slow | +0.017 |
very | +0.016 |
really | +0.016 |
Neg logit contributions | |
lor | -0.061 |
ocket | -0.056 |
urger | -0.055 |
card | -0.052 |
ster | -0.051 |
Token | Ben |
Token ID | 3932 |
Feature activation | +3.21e+00 |
Pos logit contributions | |
lor | +0.260 |
ocket | +0.238 |
urger | +0.228 |
oll | +0.215 |
card | +0.214 |
Neg logit contributions | |
different | -0.105 |
fast | -0.103 |
slow | -0.093 |
very | -0.090 |
really | -0.086 |
Token | shake |
Token ID | 13279 |
Feature activation | -2.10e+00 |
Pos logit contributions | |
lor | +0.167 |
ocket | +0.154 |
urger | +0.147 |
card | +0.139 |
oll | +0.138 |
Neg logit contributions | |
different | -0.069 |
fast | -0.068 |
slow | -0.062 |
very | -0.060 |
really | -0.058 |
Token | their |
Token ID | 511 |
Feature activation | +3.28e+00 |
Pos logit contributions | |
different | +0.034 |
fast | +0.033 |
late | +0.031 |
Ste | +0.030 |
slow | +0.030 |
Neg logit contributions | |
lor | -0.100 |
urger | -0.094 |
ocket | -0.093 |
uit | -0.086 |
ster | -0.086 |
Token | heads |
Token ID | 6665 |
Feature activation | -3.25e-01 |
Pos logit contributions | |
lor | +0.121 |
ocket | +0.112 |
urger | +0.107 |
ster | +0.096 |
oll | +0.095 |
Neg logit contributions | |
different | -0.112 |
fast | -0.110 |
slow | -0.103 |
late | -0.102 |
very | -0.100 |
Token | and |
Token ID | 290 |
Feature activation | +2.65e+00 |
Pos logit contributions | |
different | +0.005 |
fast | +0.005 |
slow | +0.004 |
late | +0.004 |
very | +0.004 |
Neg logit contributions | |
lor | -0.017 |
ocket | -0.016 |
urger | -0.015 |
oll | -0.015 |
card | -0.015 |
Token | say |
Token ID | 910 |
Feature activation | +9.40e+00 |
Pos logit contributions | |
lor | +0.132 |
ocket | +0.121 |
urger | +0.115 |
oll | +0.108 |
card | +0.107 |
Neg logit contributions | |
different | -0.065 |
fast | -0.064 |
slow | -0.059 |
very | -0.057 |
really | -0.054 |
Token | , |
Token ID | 11 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
lor | +0.475 |
ocket | +0.437 |
urger | +0.419 |
oll | +0.395 |
card | +0.392 |
Neg logit contributions | |
different | -0.199 |
fast | -0.194 |
slow | -0.176 |
late | -0.166 |
very | -0.164 |
Token | " |
Token ID | 366 |
Feature activation | -2.56e+00 |
Pos logit contributions | |
lor | +0.065 |
ocket | +0.060 |
urger | +0.057 |
oll | +0.054 |
card | +0.054 |
Neg logit contributions | |
different | -0.026 |
fast | -0.026 |
slow | -0.023 |
very | -0.022 |
really | -0.022 |
Token | No |
Token ID | 2949 |
Feature activation | +7.09e-01 |
Pos logit contributions | |
different | +0.055 |
fast | +0.054 |
very | +0.053 |
slow | +0.048 |
hard | +0.045 |
Neg logit contributions | |
lor | -0.130 |
ocket | -0.111 |
urger | -0.108 |
bridge | -0.103 |
oll | -0.102 |
Token | , |
Token ID | 11 |
Feature activation | -3.56e-01 |
Pos logit contributions | |
lor | +0.037 |
ocket | +0.033 |
urger | +0.033 |
oll | +0.030 |
anger | +0.030 |
Neg logit contributions | |
fast | -0.014 |
different | -0.014 |
slow | -0.012 |
late | -0.012 |
very | -0.011 |
Token | thank |
Token ID | 5875 |
Feature activation | +2.90e+00 |
Pos logit contributions | |
different | +0.008 |
fast | +0.008 |
slow | +0.007 |
very | +0.007 |
the | +0.007 |
Neg logit contributions | |
lor | -0.020 |
ocket | -0.018 |
urger | -0.017 |
card | -0.017 |
anger | -0.016 |
Token | olded |
Token ID | 48959 |
Feature activation | +3.63e+00 |
Pos logit contributions | |
lor | +0.059 |
ocket | +0.050 |
urger | +0.048 |
ster | +0.041 |
bridge | +0.041 |
Neg logit contributions | |
fast | -0.071 |
different | -0.071 |
very | -0.067 |
slow | -0.066 |
really | -0.065 |
Token | Tom |
Token ID | 4186 |
Feature activation | +5.00e-01 |
Pos logit contributions | |
lor | +0.179 |
ocket | +0.166 |
urger | +0.160 |
oll | +0.152 |
anger | +0.148 |
Neg logit contributions | |
different | -0.079 |
fast | -0.077 |
slow | -0.070 |
very | -0.068 |
really | -0.066 |
Token | and |
Token ID | 290 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
lor | +0.030 |
ocket | +0.028 |
urger | +0.027 |
oll | +0.026 |
card | +0.026 |
Neg logit contributions | |
different | -0.007 |
fast | -0.006 |
slow | -0.005 |
very | -0.005 |
really | -0.005 |
Token | made |
Token ID | 925 |
Feature activation | +5.56e-01 |
Pos logit contributions | |
lor | +0.098 |
ocket | +0.090 |
urger | +0.087 |
card | +0.082 |
anger | +0.081 |
Neg logit contributions | |
different | -0.036 |
fast | -0.035 |
slow | -0.032 |
very | -0.031 |
really | -0.029 |
Token | him |
Token ID | 683 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
lor | +0.032 |
ocket | +0.029 |
urger | +0.028 |
card | +0.026 |
ster | +0.026 |
Neg logit contributions | |
different | -0.011 |
fast | -0.011 |
slow | -0.009 |
very | -0.009 |
really | -0.009 |
Token | say |
Token ID | 910 |
Feature activation | +9.38e+00 |
Pos logit contributions | |
lor | +0.075 |
ocket | +0.070 |
urger | +0.068 |
oll | +0.064 |
card | +0.064 |
Neg logit contributions | |
different | -0.025 |
fast | -0.025 |
slow | -0.022 |
very | -0.021 |
really | -0.020 |
Token | sorry |
Token ID | 7926 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
lor | +0.413 |
ocket | +0.377 |
urger | +0.357 |
oll | +0.332 |
card | +0.330 |
Neg logit contributions | |
different | -0.276 |
fast | -0.273 |
slow | -0.254 |
very | -0.248 |
really | -0.240 |
Token | . |
Token ID | 13 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
lor | +0.001 |
urger | +0.001 |
ocket | +0.001 |
oll | +0.001 |
ster | +0.001 |
Neg logit contributions | |
fast | -0.000 |
different | -0.000 |
slow | -0.000 |
late | -0.000 |
busy | -0.000 |
Token | She |
Token ID | 1375 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
lor | +0.049 |
ocket | +0.042 |
oll | +0.041 |
anger | +0.041 |
ster | +0.040 |
Neg logit contributions | |
different | -0.028 |
fast | -0.024 |
really | -0.022 |
B | -0.022 |
fun | -0.022 |
Token | gave |
Token ID | 2921 |
Feature activation | +3.71e+00 |
Pos logit contributions | |
lor | +0.106 |
ocket | +0.097 |
urger | +0.092 |
anger | +0.085 |
card | +0.085 |
Neg logit contributions | |
different | -0.055 |
fast | -0.053 |
very | -0.049 |
slow | -0.049 |
really | -0.047 |
Token | Lily |
Token ID | 20037 |
Feature activation | +3.73e-01 |
Pos logit contributions | |
lor | +0.203 |
ocket | +0.188 |
urger | +0.180 |
card | +0.169 |
oll | +0.169 |
Neg logit contributions | |
different | -0.073 |
fast | -0.071 |
slow | -0.063 |
very | -0.061 |
really | -0.058 |
Token |
|
Token ID | 198 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
fast | +0.032 |
different | +0.030 |
slow | +0.030 |
very | +0.028 |
really | +0.028 |
Neg logit contributions | |
lor | -0.060 |
ocket | -0.058 |
urger | -0.054 |
oll | -0.052 |
card | -0.051 |
Token |
|
Token ID | 198 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
different | +0.013 |
fast | +0.012 |
slow | +0.010 |
very | +0.010 |
really | +0.008 |
Neg logit contributions | |
lor | -0.071 |
ocket | -0.065 |
urger | -0.062 |
oll | -0.060 |
card | -0.059 |
Token | His |
Token ID | 6653 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
different | +0.026 |
fast | +0.025 |
slow | +0.022 |
very | +0.022 |
really | +0.020 |
Neg logit contributions | |
lor | -0.074 |
ocket | -0.065 |
urger | -0.063 |
oll | -0.061 |
card | -0.060 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
lor | +0.074 |
ocket | +0.066 |
urger | +0.059 |
bridge | +0.057 |
card | +0.057 |
Neg logit contributions | |
fast | -0.041 |
different | -0.040 |
slow | -0.038 |
very | -0.038 |
really | -0.035 |
Token | went |
Token ID | 1816 |
Feature activation | -5.26e-01 |
Pos logit contributions | |
different | +0.037 |
fast | +0.036 |
very | +0.033 |
slow | +0.033 |
really | +0.033 |
Neg logit contributions | |
lor | -0.096 |
ocket | -0.090 |
urger | -0.082 |
anger | -0.079 |
card | -0.078 |
Token | over |
Token ID | 625 |
Feature activation | -8.27e+00 |
Pos logit contributions | |
different | +0.005 |
fast | +0.005 |
slow | +0.004 |
very | +0.004 |
really | +0.003 |
Neg logit contributions | |
lor | -0.034 |
ocket | -0.032 |
urger | -0.031 |
oll | -0.029 |
ster | -0.029 |
Token | to |
Token ID | 284 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
different | +0.107 |
fast | +0.104 |
slow | +0.088 |
very | +0.081 |
really | +0.075 |
Neg logit contributions | |
lor | -0.490 |
ocket | -0.458 |
urger | -0.440 |
oll | -0.420 |
card | -0.416 |
Token | the |
Token ID | 262 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
different | +0.036 |
fast | +0.035 |
slow | +0.032 |
very | +0.029 |
really | +0.029 |
Neg logit contributions | |
lor | -0.113 |
ocket | -0.105 |
urger | -0.101 |
oll | -0.097 |
card | -0.096 |
Token | net |
Token ID | 2010 |
Feature activation | -5.63e-01 |
Pos logit contributions | |
different | +0.049 |
fast | +0.049 |
slow | +0.045 |
very | +0.044 |
really | +0.043 |
Neg logit contributions | |
lor | -0.078 |
ocket | -0.071 |
urger | -0.067 |
oll | -0.064 |
card | -0.063 |
Token | and |
Token ID | 290 |
Feature activation | +4.04e-01 |
Pos logit contributions | |
different | +0.007 |
fast | +0.007 |
slow | +0.006 |
very | +0.005 |
really | +0.005 |
Neg logit contributions | |
lor | -0.034 |
ocket | -0.032 |
urger | -0.031 |
oll | -0.030 |
card | -0.029 |
Token | tried |
Token ID | 3088 |
Feature activation | -6.29e-01 |
Pos logit contributions | |
lor | +0.020 |
ocket | +0.019 |
urger | +0.018 |
card | +0.017 |
oll | +0.017 |
Neg logit contributions | |
different | -0.010 |
fast | -0.010 |
slow | -0.009 |
very | -0.009 |
really | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -7.23e-01 |
Pos logit contributions | |
different | +0.012 |
fast | +0.010 |
slow | +0.010 |
competitive | +0.010 |
very | +0.009 |
Neg logit contributions | |
lor | -0.035 |
ocket | -0.031 |
oll | -0.030 |
urger | -0.029 |
ogo | -0.029 |
Token | S |
Token ID | 50 |
Feature activation | -4.92e-01 |
Pos logit contributions | |
different | +0.016 |
fast | +0.014 |
slow | +0.013 |
very | +0.013 |
really | +0.013 |
Neg logit contributions | |
lor | -0.037 |
ocket | -0.032 |
urger | -0.031 |
oll | -0.031 |
ster | -0.030 |
Token | ara |
Token ID | 3301 |
Feature activation | -3.10e+00 |
Pos logit contributions | |
different | +0.015 |
fast | +0.015 |
slow | +0.014 |
very | +0.013 |
late | +0.013 |
Neg logit contributions | |
lor | -0.021 |
ocket | -0.020 |
urger | -0.018 |
card | -0.017 |
oll | -0.016 |
Token | felt |
Token ID | 2936 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
different | +0.067 |
fast | +0.065 |
slow | +0.058 |
late | +0.056 |
very | +0.055 |
Neg logit contributions | |
lor | -0.157 |
ocket | -0.146 |
urger | -0.139 |
oll | -0.132 |
ster | -0.130 |
Token | a |
Token ID | 257 |
Feature activation | -6.41e-01 |
Pos logit contributions | |
different | +0.026 |
B | +0.024 |
Lu | +0.024 |
fast | +0.024 |
Stop | +0.022 |
Neg logit contributions | |
ocket | -0.072 |
lor | -0.070 |
oll | -0.065 |
urger | -0.062 |
ster | -0.060 |
Token | gust |
Token ID | 35253 |
Feature activation | -8.08e+00 |
Pos logit contributions | |
different | +0.013 |
late | +0.013 |
fast | +0.012 |
slow | +0.011 |
to | +0.011 |
Neg logit contributions | |
lor | -0.028 |
ocket | -0.027 |
ster | -0.025 |
urger | -0.025 |
oll | -0.025 |
Token | of |
Token ID | 286 |
Feature activation | -6.93e-01 |
Pos logit contributions | |
different | +0.248 |
fast | +0.236 |
slow | +0.230 |
very | +0.212 |
really | +0.211 |
Neg logit contributions | |
lor | -0.318 |
ocket | -0.300 |
oll | -0.277 |
urger | -0.268 |
card | -0.261 |
Token | wind |
Token ID | 2344 |
Feature activation | -4.19e+00 |
Pos logit contributions | |
different | +0.013 |
fast | +0.011 |
really | +0.011 |
very | +0.010 |
slow | +0.010 |
Neg logit contributions | |
lor | -0.034 |
ocket | -0.033 |
urger | -0.031 |
oll | -0.031 |
card | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | -4.27e-02 |
Pos logit contributions | |
different | +0.088 |
fast | +0.084 |
slow | +0.076 |
very | +0.072 |
really | +0.070 |
Neg logit contributions | |
lor | -0.216 |
ocket | -0.200 |
urger | -0.191 |
card | -0.181 |
oll | -0.181 |
Token | She |
Token ID | 1375 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
different | +0.001 |
fast | +0.001 |
slow | +0.001 |
very | +0.001 |
B | +0.001 |
Neg logit contributions | |
lor | -0.002 |
ocket | -0.002 |
oll | -0.002 |
anger | -0.002 |
urger | -0.002 |
Token | saw |
Token ID | 2497 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
lor | +0.062 |
ocket | +0.057 |
urger | +0.055 |
oll | +0.052 |
card | +0.051 |
Neg logit contributions | |
different | -0.026 |
fast | -0.025 |
slow | -0.023 |
very | -0.022 |
really | -0.021 |
Token | I |
Token ID | 314 |
Feature activation | -3.29e+00 |
Pos logit contributions | |
lor | +0.067 |
ocket | +0.065 |
urger | +0.062 |
card | +0.059 |
oll | +0.057 |
Neg logit contributions | |
fast | -0.019 |
different | -0.019 |
slow | -0.017 |
very | -0.016 |
the | -0.016 |
Token | have |
Token ID | 423 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
different | +0.067 |
fast | +0.066 |
really | +0.062 |
slow | +0.062 |
very | +0.060 |
Neg logit contributions | |
lor | -0.179 |
ocket | -0.164 |
urger | -0.159 |
card | -0.151 |
ster | -0.147 |
Token | to |
Token ID | 284 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
different | +0.027 |
fast | +0.026 |
slow | +0.022 |
very | +0.021 |
really | +0.020 |
Neg logit contributions | |
lor | -0.100 |
ocket | -0.093 |
urger | -0.089 |
oll | -0.085 |
card | -0.084 |
Token | start |
Token ID | 923 |
Feature activation | -4.47e+00 |
Pos logit contributions | |
different | +0.001 |
fast | +0.001 |
slow | +0.001 |
really | +0.001 |
very | +0.001 |
Neg logit contributions | |
lor | -0.001 |
ocket | -0.001 |
urger | -0.001 |
anger | -0.001 |
card | -0.001 |
Token | all |
Token ID | 477 |
Feature activation | -5.88e+00 |
Pos logit contributions | |
different | +0.080 |
fast | +0.078 |
slow | +0.070 |
very | +0.066 |
really | +0.062 |
Neg logit contributions | |
lor | -0.243 |
ocket | -0.225 |
urger | -0.216 |
oll | -0.204 |
card | -0.204 |
Token | over |
Token ID | 625 |
Feature activation | -8.07e+00 |
Pos logit contributions | |
fast | +0.097 |
different | +0.091 |
slow | +0.089 |
low | +0.085 |
B | +0.085 |
Neg logit contributions | |
lor | -0.286 |
ocket | -0.250 |
card | -0.247 |
urger | -0.240 |
icking | -0.232 |
Token | . |
Token ID | 13 |
Feature activation | -7.30e-01 |
Pos logit contributions | |
different | +0.144 |
fast | +0.141 |
slow | +0.126 |
very | +0.120 |
really | +0.113 |
Neg logit contributions | |
lor | -0.438 |
ocket | -0.407 |
urger | -0.389 |
oll | -0.368 |
card | -0.366 |
Token | â |
Token ID | 22940 |
Feature activation | +4.86e-01 |
Pos logit contributions | |
different | +0.017 |
fast | +0.017 |
slow | +0.016 |
very | +0.015 |
really | +0.015 |
Neg logit contributions | |
lor | -0.036 |
ocket | -0.033 |
urger | -0.032 |
card | -0.030 |
ster | -0.030 |
Token | € |
Token ID | 26391 |
Feature activation | -9.38e-01 |
Pos logit contributions | |
lor | +0.023 |
ocket | +0.021 |
urger | +0.020 |
oll | +0.019 |
card | +0.019 |
Neg logit contributions | |
fast | -0.012 |
different | -0.012 |
slow | -0.011 |
very | -0.011 |
really | -0.011 |
Token | Lily |
Token ID | 20037 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
fast | +0.025 |
different | +0.024 |
slow | +0.023 |
very | +0.022 |
really | +0.022 |
Neg logit contributions | |
lor | -0.043 |
ocket | -0.041 |
urger | -0.039 |
oll | -0.036 |
card | -0.036 |
Token | quickly |
Token ID | 2952 |
Feature activation | -3.62e+00 |
Pos logit contributions | |
different | +0.052 |
fast | +0.051 |
slow | +0.047 |
very | +0.046 |
really | +0.044 |
Neg logit contributions | |
lor | -0.107 |
ocket | -0.099 |
urger | -0.094 |
oll | -0.088 |
card | -0.088 |
Token | . |
Token ID | 13 |
Feature activation | -7.88e-01 |
Pos logit contributions | |
fast | +0.026 |
different | +0.024 |
slow | +0.023 |
very | +0.022 |
really | +0.022 |
Neg logit contributions | |
lor | -0.055 |
ocket | -0.053 |
urger | -0.050 |
card | -0.047 |
oll | -0.047 |
Token |
|
Token ID | 198 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
different | +0.018 |
fast | +0.018 |
slow | +0.016 |
very | +0.016 |
really | +0.015 |
Neg logit contributions | |
lor | -0.039 |
ocket | -0.037 |
urger | -0.035 |
card | -0.033 |
ster | -0.032 |
Token |
|
Token ID | 198 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
different | +0.014 |
fast | +0.013 |
slow | +0.011 |
very | +0.011 |
really | +0.009 |
Neg logit contributions | |
lor | -0.070 |
ocket | -0.064 |
urger | -0.061 |
oll | -0.060 |
anger | -0.058 |
Token | Tom |
Token ID | 13787 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
different | +0.026 |
fast | +0.025 |
very | +0.022 |
slow | +0.022 |
really | +0.020 |
Neg logit contributions | |
lor | -0.071 |
ocket | -0.063 |
urger | -0.060 |
oll | -0.059 |
card | -0.057 |
Token | rushed |
Token ID | 13999 |
Feature activation | -2.40e+00 |
Pos logit contributions | |
different | +0.029 |
fast | +0.029 |
slow | +0.026 |
very | +0.026 |
really | +0.025 |
Neg logit contributions | |
lor | -0.079 |
ocket | -0.074 |
urger | -0.069 |
card | -0.066 |
oll | -0.066 |
Token | over |
Token ID | 625 |
Feature activation | -8.06e+00 |
Pos logit contributions | |
different | +0.035 |
fast | +0.033 |
slow | +0.029 |
very | +0.027 |
really | +0.025 |
Neg logit contributions | |
lor | -0.139 |
ocket | -0.130 |
urger | -0.125 |
card | -0.119 |
oll | -0.118 |
Token | and |
Token ID | 290 |
Feature activation | +3.68e-02 |
Pos logit contributions | |
different | +0.094 |
fast | +0.091 |
slow | +0.074 |
very | +0.069 |
really | +0.063 |
Neg logit contributions | |
lor | -0.498 |
ocket | -0.467 |
urger | -0.451 |
oll | -0.428 |
card | -0.427 |
Token | opened |
Token ID | 4721 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
lor | +0.002 |
ocket | +0.002 |
urger | +0.002 |
oll | +0.002 |
card | +0.002 |
Neg logit contributions | |
fast | -0.001 |
different | -0.001 |
slow | -0.001 |
very | -0.001 |
really | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
lor | +0.110 |
ocket | +0.104 |
urger | +0.101 |
oll | +0.096 |
card | +0.095 |
Neg logit contributions | |
different | -0.014 |
fast | -0.013 |
slow | -0.009 |
very | -0.008 |
really | -0.007 |
Token | buckle |
Token ID | 44983 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
lor | +0.008 |
ocket | +0.007 |
urger | +0.007 |
oll | +0.006 |
card | +0.006 |
Neg logit contributions | |
different | -0.008 |
fast | -0.007 |
slow | -0.007 |
very | -0.007 |
really | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -8.66e-01 |
Pos logit contributions | |
lor | +0.125 |
ocket | +0.119 |
urger | +0.116 |
oll | +0.109 |
card | +0.105 |
Neg logit contributions | |
fast | -0.017 |
different | -0.017 |
very | -0.013 |
slow | -0.012 |
really | -0.009 |
Token | head |
Token ID | 1182 |
Feature activation | -3.63e-01 |
Pos logit contributions | |
lor | +0.096 |
ocket | +0.086 |
urger | +0.079 |
oll | +0.075 |
card | +0.072 |
Neg logit contributions | |
fast | -0.074 |
different | -0.073 |
slow | -0.070 |
very | -0.069 |
really | -0.064 |
Token | low |
Token ID | 1877 |
Feature activation | -4.63e+00 |
Pos logit contributions | |
fast | +0.006 |
different | +0.006 |
slow | +0.005 |
very | +0.005 |
really | +0.005 |
Neg logit contributions | |
lor | -0.021 |
ocket | -0.020 |
urger | -0.019 |
oll | -0.018 |
card | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -9.83e-01 |
Pos logit contributions | |
different | +0.059 |
fast | +0.058 |
slow | +0.049 |
very | +0.046 |
really | +0.042 |
Neg logit contributions | |
lor | -0.280 |
ocket | -0.263 |
urger | -0.252 |
oll | -0.240 |
card | -0.239 |
Token | He |
Token ID | 679 |
Feature activation | +9.38e-01 |
Pos logit contributions | |
different | +0.019 |
fast | +0.018 |
slow | +0.016 |
very | +0.016 |
really | +0.015 |
Neg logit contributions | |
lor | -0.053 |
ocket | -0.049 |
urger | -0.047 |
oll | -0.045 |
card | -0.045 |
Token | went |
Token ID | 1816 |
Feature activation | -2.79e-01 |
Pos logit contributions | |
lor | +0.049 |
ocket | +0.045 |
urger | +0.043 |
card | +0.040 |
oll | +0.040 |
Neg logit contributions | |
different | -0.021 |
fast | -0.021 |
slow | -0.019 |
very | -0.018 |
really | -0.018 |
Token | over |
Token ID | 625 |
Feature activation | -8.06e+00 |
Pos logit contributions | |
different | +0.004 |
fast | +0.004 |
slow | +0.003 |
very | +0.003 |
really | +0.003 |
Neg logit contributions | |
lor | -0.017 |
ocket | -0.016 |
urger | -0.015 |
oll | -0.014 |
card | -0.014 |
Token | to |
Token ID | 284 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
different | +0.091 |
fast | +0.087 |
slow | +0.072 |
very | +0.065 |
late | +0.059 |
Neg logit contributions | |
lor | -0.486 |
ocket | -0.458 |
urger | -0.437 |
oll | -0.420 |
card | -0.416 |
Token | his |
Token ID | 465 |
Feature activation | +9.51e-01 |
Pos logit contributions | |
different | +0.027 |
fast | +0.026 |
slow | +0.023 |
very | +0.021 |
really | +0.020 |
Neg logit contributions | |
lor | -0.107 |
ocket | -0.100 |
urger | -0.097 |
oll | -0.092 |
card | -0.092 |
Token | owner |
Token ID | 4870 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
lor | +0.043 |
ocket | +0.039 |
urger | +0.036 |
oll | +0.034 |
card | +0.034 |
Neg logit contributions | |
different | -0.028 |
fast | -0.028 |
slow | -0.026 |
very | -0.026 |
really | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
different | +0.015 |
fast | +0.013 |
very | +0.012 |
slow | +0.011 |
really | +0.010 |
Neg logit contributions | |
lor | -0.103 |
ocket | -0.094 |
urger | -0.091 |
anger | -0.087 |
bridge | -0.086 |
Token | who |
Token ID | 508 |
Feature activation | +9.22e-01 |
Pos logit contributions | |
different | +0.042 |
fast | +0.041 |
slow | +0.037 |
very | +0.035 |
really | +0.034 |
Neg logit contributions | |
lor | -0.107 |
ocket | -0.099 |
urger | -0.095 |
card | -0.089 |
oll | -0.089 |
Token | |
Token ID | 220 |
Feature activation | -9.34e-01 |
Pos logit contributions | |
fast | +0.024 |
different | +0.024 |
slow | +0.022 |
very | +0.021 |
really | +0.021 |
Neg logit contributions | |
lor | -0.042 |
ocket | -0.039 |
urger | -0.037 |
oll | -0.035 |
card | -0.034 |
Token |
|
Token ID | 198 |
Feature activation | -7.12e-01 |
Pos logit contributions | |
different | +0.010 |
fast | +0.010 |
slow | +0.008 |
very | +0.007 |
really | +0.006 |
Neg logit contributions | |
lor | -0.058 |
ocket | -0.055 |
urger | -0.053 |
oll | -0.050 |
card | -0.050 |
Token |
|
Token ID | 198 |
Feature activation | -9.58e-01 |
Pos logit contributions | |
different | +0.009 |
fast | +0.009 |
slow | +0.008 |
very | +0.007 |
really | +0.006 |
Neg logit contributions | |
lor | -0.042 |
ocket | -0.038 |
urger | -0.036 |
oll | -0.035 |
anger | -0.034 |
Token | John |
Token ID | 7554 |
Feature activation | -3.45e+00 |
Pos logit contributions | |
different | +0.022 |
fast | +0.022 |
very | +0.020 |
slow | +0.020 |
scared | +0.019 |
Neg logit contributions | |
lor | -0.047 |
ocket | -0.038 |
urger | -0.037 |
bridge | -0.037 |
uit | -0.037 |
Token | ran |
Token ID | 4966 |
Feature activation | -2.64e-01 |
Pos logit contributions | |
different | +0.068 |
fast | +0.068 |
slow | +0.062 |
very | +0.060 |
really | +0.059 |
Neg logit contributions | |
lor | -0.186 |
ocket | -0.173 |
urger | -0.163 |
card | -0.155 |
oll | -0.154 |
Token | over |
Token ID | 625 |
Feature activation | -8.01e+00 |
Pos logit contributions | |
different | +0.005 |
fast | +0.005 |
slow | +0.004 |
late | +0.004 |
panic | +0.004 |
Neg logit contributions | |
lor | -0.013 |
ocket | -0.012 |
urger | -0.011 |
anger | -0.011 |
card | -0.011 |
Token | to |
Token ID | 284 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
different | +0.096 |
fast | +0.093 |
slow | +0.081 |
late | +0.073 |
very | +0.072 |
Neg logit contributions | |
lor | -0.457 |
ocket | -0.432 |
urger | -0.409 |
anger | -0.395 |
oll | -0.394 |
Token | the |
Token ID | 262 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
different | +0.022 |
fast | +0.020 |
slow | +0.017 |
very | +0.015 |
really | +0.014 |
Neg logit contributions | |
lor | -0.125 |
ocket | -0.116 |
urger | -0.113 |
card | -0.108 |
oll | -0.107 |
Token | rock |
Token ID | 3881 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
different | +0.039 |
fast | +0.038 |
slow | +0.035 |
late | +0.034 |
very | +0.034 |
Neg logit contributions | |
lor | -0.058 |
ocket | -0.053 |
urger | -0.050 |
ster | -0.047 |
card | -0.047 |
Token | and |
Token ID | 290 |
Feature activation | +3.93e-01 |
Pos logit contributions | |
fast | +0.022 |
different | +0.021 |
very | +0.018 |
slow | +0.017 |
really | +0.015 |
Neg logit contributions | |
lor | -0.114 |
ocket | -0.108 |
urger | -0.105 |
oll | -0.099 |
card | -0.097 |
Token | started |
Token ID | 2067 |
Feature activation | -3.89e+00 |
Pos logit contributions | |
lor | +0.019 |
ocket | +0.018 |
urger | +0.017 |
oll | +0.016 |
card | +0.016 |
Neg logit contributions | |
different | -0.010 |
fast | -0.009 |
slow | -0.009 |
very | -0.008 |
really | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
different | +0.103 |
fast | +0.100 |
slow | +0.089 |
very | +0.085 |
really | +0.081 |
Neg logit contributions | |
lor | -0.308 |
ocket | -0.286 |
urger | -0.274 |
oll | -0.260 |
card | -0.258 |
Token | He |
Token ID | 679 |
Feature activation | +8.47e-01 |
Pos logit contributions | |
different | +0.020 |
fast | +0.020 |
slow | +0.018 |
very | +0.017 |
really | +0.016 |
Neg logit contributions | |
lor | -0.058 |
ocket | -0.054 |
urger | -0.052 |
oll | -0.049 |
card | -0.049 |
Token | jumped |
Token ID | 11687 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
lor | +0.043 |
ocket | +0.040 |
urger | +0.038 |
oll | +0.036 |
card | +0.035 |
Neg logit contributions | |
different | -0.019 |
fast | -0.019 |
slow | -0.017 |
very | -0.017 |
really | -0.016 |
Token | around |
Token ID | 1088 |
Feature activation | -6.23e+00 |
Pos logit contributions | |
lor | +0.069 |
ocket | +0.064 |
urger | +0.061 |
oll | +0.058 |
card | +0.058 |
Neg logit contributions | |
different | -0.024 |
fast | -0.023 |
slow | -0.021 |
very | -0.020 |
late | -0.019 |
Token | with |
Token ID | 351 |
Feature activation | -8.75e-01 |
Pos logit contributions | |
different | +0.093 |
fast | +0.091 |
slow | +0.080 |
very | +0.073 |
late | +0.073 |
Neg logit contributions | |
lor | -0.346 |
ocket | -0.326 |
urger | -0.310 |
card | -0.295 |
oll | -0.293 |
Token | gust |
Token ID | 35253 |
Feature activation | -7.98e+00 |
Pos logit contributions | |
different | +0.019 |
fast | +0.019 |
slow | +0.017 |
very | +0.016 |
really | +0.016 |
Neg logit contributions | |
lor | -0.045 |
ocket | -0.041 |
urger | -0.040 |
oll | -0.037 |
card | -0.037 |
Token | o |
Token ID | 78 |
Feature activation | -7.11e-02 |
Pos logit contributions | |
different | +0.213 |
fast | +0.205 |
slow | +0.192 |
very | +0.183 |
late | +0.180 |
Neg logit contributions | |
lor | -0.360 |
ocket | -0.325 |
urger | -0.314 |
oll | -0.299 |
ster | -0.296 |
Token | and |
Token ID | 290 |
Feature activation | +8.01e-01 |
Pos logit contributions | |
different | +0.001 |
fast | +0.001 |
slow | +0.001 |
very | +0.001 |
really | +0.001 |
Neg logit contributions | |
lor | -0.004 |
ocket | -0.003 |
urger | -0.003 |
oll | -0.003 |
card | -0.003 |
Token | spun |
Token ID | 26843 |
Feature activation | -8.54e-01 |
Pos logit contributions | |
lor | +0.036 |
ocket | +0.034 |
urger | +0.032 |
card | +0.031 |
ster | +0.030 |
Neg logit contributions | |
different | -0.018 |
fast | -0.017 |
late | -0.016 |
slow | -0.016 |
really | -0.015 |
Token | in |
Token ID | 287 |
Feature activation | -4.11e+00 |
Pos logit contributions | |
different | +0.015 |
late | +0.014 |
fast | +0.014 |
slow | +0.013 |
upset | +0.013 |
Neg logit contributions | |
lor | -0.041 |
ocket | -0.037 |
urger | -0.036 |
bridge | -0.035 |
anger | -0.035 |
Token | circles |
Token ID | 13332 |
Feature activation | -2.38e-03 |
Pos logit contributions | |
fast | +0.065 |
different | +0.062 |
late | +0.061 |
slow | +0.059 |
very | +0.055 |
Neg logit contributions | |
lor | -0.206 |
ocket | -0.195 |
urger | -0.187 |
card | -0.176 |
uit | -0.174 |
Token | the |
Token ID | 262 |
Feature activation | -9.22e-01 |
Pos logit contributions | |
fast | +0.056 |
different | +0.053 |
slow | +0.046 |
late | +0.045 |
very | +0.042 |
Neg logit contributions | |
lor | -0.291 |
ocket | -0.277 |
urger | -0.261 |
card | -0.252 |
oll | -0.248 |
Token | park |
Token ID | 3952 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
fast | +0.025 |
different | +0.024 |
late | +0.023 |
slow | +0.023 |
B | +0.022 |
Neg logit contributions | |
lor | -0.037 |
ocket | -0.033 |
card | -0.031 |
urger | -0.031 |
suitcase | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | -5.66e-01 |
Pos logit contributions | |
different | +0.002 |
fast | +0.002 |
slow | +0.002 |
late | +0.002 |
very | +0.002 |
Neg logit contributions | |
lor | -0.009 |
ocket | -0.009 |
urger | -0.008 |
card | -0.008 |
oll | -0.008 |
Token | They |
Token ID | 1119 |
Feature activation | -6.55e-01 |
Pos logit contributions | |
different | +0.012 |
fast | +0.011 |
slow | +0.010 |
very | +0.010 |
hurt | +0.009 |
Neg logit contributions | |
lor | -0.029 |
ocket | -0.027 |
urger | -0.025 |
oll | -0.024 |
card | -0.024 |
Token | went |
Token ID | 1816 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
different | +0.015 |
fast | +0.015 |
late | +0.014 |
slow | +0.013 |
very | +0.013 |
Neg logit contributions | |
lor | -0.031 |
ocket | -0.029 |
urger | -0.027 |
oll | -0.026 |
card | -0.026 |
Token | over |
Token ID | 625 |
Feature activation | -7.96e+00 |
Pos logit contributions | |
different | +0.005 |
fast | +0.005 |
slow | +0.004 |
late | +0.004 |
very | +0.004 |
Neg logit contributions | |
lor | -0.014 |
ocket | -0.013 |
urger | -0.012 |
card | -0.012 |
anger | -0.012 |
Token | to |
Token ID | 284 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
different | +0.103 |
fast | +0.102 |
late | +0.095 |
slow | +0.092 |
upset | +0.084 |
Neg logit contributions | |
lor | -0.418 |
ocket | -0.401 |
urger | -0.375 |
card | -0.362 |
anger | -0.359 |
Token | a |
Token ID | 257 |
Feature activation | -6.17e-01 |
Pos logit contributions | |
fast | +0.021 |
different | +0.020 |
late | +0.017 |
slow | +0.017 |
very | +0.016 |
Neg logit contributions | |
lor | -0.097 |
ocket | -0.092 |
urger | -0.088 |
card | -0.083 |
oll | -0.082 |
Token | tree |
Token ID | 5509 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
fast | +0.012 |
late | +0.012 |
different | +0.012 |
slow | +0.011 |
Stop | +0.011 |
Neg logit contributions | |
lor | -0.029 |
ocket | -0.027 |
urger | -0.025 |
card | -0.025 |
ster | -0.024 |
Token | with |
Token ID | 351 |
Feature activation | -5.14e-01 |
Pos logit contributions | |
different | +0.027 |
fast | +0.026 |
slow | +0.023 |
very | +0.022 |
really | +0.020 |
Neg logit contributions | |
lor | -0.109 |
ocket | -0.102 |
urger | -0.098 |
oll | -0.093 |
card | -0.092 |
Token | a |
Token ID | 257 |
Feature activation | +6.33e-01 |
Pos logit contributions | |
different | +0.008 |
fast | +0.008 |
slow | +0.007 |
very | +0.006 |
really | +0.006 |
Neg logit contributions | |
lor | -0.030 |
ocket | -0.028 |
urger | -0.027 |
oll | -0.025 |
card | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
different | +0.007 |
fast | +0.007 |
slow | +0.006 |
very | +0.005 |
really | +0.004 |
Neg logit contributions | |
lor | -0.054 |
ocket | -0.052 |
urger | -0.050 |
oll | -0.048 |
card | -0.047 |
Token | The |
Token ID | 464 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
different | +0.020 |
fast | +0.019 |
slow | +0.017 |
very | +0.017 |
really | +0.016 |
Neg logit contributions | |
lor | -0.058 |
ocket | -0.052 |
urger | -0.050 |
oll | -0.048 |
card | -0.048 |
Token | two |
Token ID | 734 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
lor | +0.007 |
ocket | +0.006 |
urger | +0.006 |
oll | +0.006 |
card | +0.006 |
Neg logit contributions | |
different | -0.003 |
fast | -0.003 |
slow | -0.003 |
very | -0.003 |
really | -0.003 |
Token | friends |
Token ID | 2460 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
different | +0.041 |
fast | +0.040 |
slow | +0.036 |
very | +0.036 |
really | +0.035 |
Neg logit contributions | |
lor | -0.086 |
ocket | -0.079 |
urger | -0.075 |
card | -0.071 |
anger | -0.068 |
Token | ran |
Token ID | 4966 |
Feature activation | -3.06e-01 |
Pos logit contributions | |
different | +0.044 |
fast | +0.041 |
slow | +0.038 |
really | +0.038 |
very | +0.037 |
Neg logit contributions | |
lor | -0.113 |
ocket | -0.101 |
urger | -0.098 |
card | -0.092 |
ster | -0.091 |
Token | over |
Token ID | 625 |
Feature activation | -7.96e+00 |
Pos logit contributions | |
different | +0.004 |
fast | +0.004 |
slow | +0.004 |
late | +0.003 |
very | +0.003 |
Neg logit contributions | |
lor | -0.016 |
ocket | -0.015 |
urger | -0.015 |
card | -0.014 |
anger | -0.014 |
Token | to |
Token ID | 284 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
different | +0.094 |
fast | +0.093 |
slow | +0.080 |
late | +0.072 |
very | +0.070 |
Neg logit contributions | |
lor | -0.465 |
ocket | -0.436 |
urger | -0.418 |
oll | -0.398 |
card | -0.397 |
Token | the |
Token ID | 262 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
different | +0.025 |
fast | +0.025 |
slow | +0.021 |
very | +0.020 |
really | +0.018 |
Neg logit contributions | |
lor | -0.108 |
ocket | -0.101 |
urger | -0.098 |
oll | -0.093 |
card | -0.092 |
Token | wealthy |
Token ID | 11574 |
Feature activation | -2.63e+00 |
Pos logit contributions | |
fast | +0.041 |
different | +0.041 |
slow | +0.038 |
late | +0.037 |
very | +0.036 |
Neg logit contributions | |
lor | -0.061 |
ocket | -0.056 |
urger | -0.053 |
card | -0.050 |
ster | -0.049 |
Token | game |
Token ID | 983 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
different | +0.076 |
fast | +0.076 |
slow | +0.070 |
very | +0.069 |
late | +0.068 |
Neg logit contributions | |
lor | -0.109 |
ocket | -0.100 |
urger | -0.095 |
ster | -0.089 |
card | -0.088 |
Token | and |
Token ID | 290 |
Feature activation | +5.73e-01 |
Pos logit contributions | |
different | +0.002 |
fast | +0.002 |
very | +0.001 |
slow | +0.001 |
really | +0.001 |
Neg logit contributions | |
lor | -0.008 |
ocket | -0.008 |
urger | -0.008 |
oll | -0.007 |
card | -0.007 |
Token | ." |
Token ID | 526 |
Feature activation | -3.07e+00 |
Pos logit contributions | |
different | +0.040 |
fast | +0.039 |
slow | +0.035 |
very | +0.034 |
really | +0.032 |
Neg logit contributions | |
lor | -0.101 |
ocket | -0.093 |
urger | -0.089 |
oll | -0.084 |
card | -0.084 |
Token |
|
Token ID | 198 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
fast | +0.066 |
different | +0.065 |
slow | +0.059 |
very | +0.054 |
quiet | +0.051 |
Neg logit contributions | |
lor | -0.154 |
ocket | -0.140 |
urger | -0.133 |
oll | -0.127 |
card | -0.126 |
Token |
|
Token ID | 198 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
fast | +0.024 |
different | +0.023 |
slow | +0.021 |
very | +0.019 |
late | +0.016 |
Neg logit contributions | |
lor | -0.089 |
ocket | -0.081 |
urger | -0.077 |
oll | -0.074 |
card | -0.072 |
Token | Jack |
Token ID | 14295 |
Feature activation | -3.21e+00 |
Pos logit contributions | |
fast | +0.033 |
different | +0.031 |
slow | +0.030 |
very | +0.029 |
hurt | +0.027 |
Neg logit contributions | |
lor | -0.073 |
ocket | -0.063 |
urger | -0.060 |
ster | -0.057 |
oll | -0.057 |
Token | went |
Token ID | 1816 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
different | +0.070 |
fast | +0.069 |
slow | +0.062 |
very | +0.060 |
really | +0.058 |
Neg logit contributions | |
lor | -0.165 |
ocket | -0.153 |
urger | -0.145 |
oll | -0.137 |
card | -0.136 |
Token | over |
Token ID | 625 |
Feature activation | -7.95e+00 |
Pos logit contributions | |
fast | +0.002 |
different | +0.002 |
slow | +0.002 |
very | +0.002 |
really | +0.001 |
Neg logit contributions | |
lor | -0.013 |
ocket | -0.012 |
urger | -0.012 |
oll | -0.011 |
card | -0.011 |
Token | to |
Token ID | 284 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
different | +0.101 |
fast | +0.096 |
slow | +0.083 |
late | +0.074 |
very | +0.072 |
Neg logit contributions | |
lor | -0.436 |
ocket | -0.418 |
urger | -0.392 |
card | -0.380 |
anger | -0.378 |
Token | the |
Token ID | 262 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
different | +0.021 |
fast | +0.021 |
slow | +0.017 |
very | +0.016 |
really | +0.014 |
Neg logit contributions | |
lor | -0.109 |
ocket | -0.103 |
urger | -0.099 |
oll | -0.094 |
card | -0.094 |
Token | tap |
Token ID | 9814 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
different | +0.032 |
fast | +0.032 |
slow | +0.029 |
very | +0.028 |
late | +0.028 |
Neg logit contributions | |
lor | -0.047 |
ocket | -0.044 |
urger | -0.041 |
card | -0.039 |
ster | -0.039 |
Token | and |
Token ID | 290 |
Feature activation | +7.69e-01 |
Pos logit contributions | |
different | +0.019 |
fast | +0.019 |
slow | +0.016 |
very | +0.016 |
really | +0.014 |
Neg logit contributions | |
lor | -0.093 |
ocket | -0.088 |
urger | -0.084 |
oll | -0.080 |
card | -0.079 |
Token | pulled |
Token ID | 5954 |
Feature activation | -3.03e-01 |
Pos logit contributions | |
lor | +0.036 |
ocket | +0.033 |
urger | +0.032 |
card | +0.030 |
oll | +0.030 |
Neg logit contributions | |
different | -0.019 |
fast | -0.018 |
slow | -0.017 |
very | -0.016 |
late | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
different | +0.016 |
fast | +0.016 |
slow | +0.013 |
very | +0.012 |
really | +0.010 |
Neg logit contributions | |
lor | -0.094 |
ocket | -0.087 |
urger | -0.084 |
oll | -0.081 |
card | -0.080 |
Token |
|
Token ID | 198 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
different | +0.015 |
fast | +0.014 |
slow | +0.013 |
very | +0.012 |
late | +0.011 |
Neg logit contributions | |
lor | -0.059 |
ocket | -0.051 |
urger | -0.049 |
anger | -0.048 |
card | -0.048 |
Token | They |
Token ID | 2990 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
different | +0.025 |
fast | +0.024 |
slow | +0.022 |
very | +0.022 |
late | +0.021 |
Neg logit contributions | |
lor | -0.052 |
ocket | -0.044 |
urger | -0.043 |
card | -0.041 |
ster | -0.040 |
Token | both |
Token ID | 1111 |
Feature activation | -4.71e+00 |
Pos logit contributions | |
different | +0.003 |
fast | +0.003 |
very | +0.003 |
really | +0.003 |
slow | +0.003 |
Neg logit contributions | |
lor | -0.008 |
ocket | -0.007 |
urger | -0.007 |
anger | -0.007 |
card | -0.007 |
Token | went |
Token ID | 1816 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
different | +0.120 |
fast | +0.120 |
slow | +0.110 |
late | +0.107 |
very | +0.105 |
Neg logit contributions | |
lor | -0.217 |
ocket | -0.203 |
urger | -0.192 |
oll | -0.182 |
card | -0.177 |
Token | over |
Token ID | 625 |
Feature activation | -7.95e+00 |
Pos logit contributions | |
fast | +0.002 |
different | +0.002 |
very | +0.002 |
slow | +0.002 |
really | +0.002 |
Neg logit contributions | |
lor | -0.007 |
ocket | -0.007 |
urger | -0.007 |
oll | -0.006 |
ster | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | +3.43e-01 |
Pos logit contributions | |
fast | +0.091 |
different | +0.090 |
slow | +0.082 |
late | +0.078 |
quiet | +0.065 |
Neg logit contributions | |
lor | -0.442 |
ocket | -0.426 |
urger | -0.400 |
card | -0.384 |
oll | -0.380 |
Token | tried |
Token ID | 3088 |
Feature activation | -7.02e-01 |
Pos logit contributions | |
lor | +0.018 |
ocket | +0.016 |
urger | +0.015 |
oll | +0.015 |
card | +0.014 |
Neg logit contributions | |
different | -0.008 |
fast | -0.007 |
slow | -0.007 |
very | -0.007 |
really | -0.006 |
Token | to |
Token ID | 284 |
Feature activation | +2.78e-01 |
Pos logit contributions | |
fast | +0.010 |
late | +0.009 |
B | +0.009 |
Stop | +0.009 |
different | +0.009 |
Neg logit contributions | |
lor | -0.037 |
urger | -0.034 |
ocket | -0.034 |
card | -0.032 |
uit | -0.031 |
Token | decide |
Token ID | 5409 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
lor | +0.012 |
ocket | +0.012 |
urger | +0.011 |
card | +0.010 |
oll | +0.010 |
Neg logit contributions | |
different | -0.007 |
fast | -0.007 |
late | -0.007 |
slow | -0.007 |
very | -0.007 |
Token | which |
Token ID | 543 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
lor | +0.073 |
ocket | +0.068 |
urger | +0.065 |
oll | +0.061 |
card | +0.060 |
Neg logit contributions | |
different | -0.033 |
fast | -0.032 |
slow | -0.029 |
very | -0.028 |
really | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | -3.97e-01 |
Pos logit contributions | |
fast | +0.008 |
different | +0.008 |
slow | +0.007 |
very | +0.007 |
really | +0.006 |
Neg logit contributions | |
lor | -0.023 |
ocket | -0.020 |
urger | -0.020 |
card | -0.019 |
oll | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | -5.49e-01 |
Pos logit contributions | |
different | +0.005 |
fast | +0.005 |
slow | +0.004 |
very | +0.004 |
late | +0.004 |
Neg logit contributions | |
lor | -0.023 |
ocket | -0.021 |
urger | -0.020 |
card | -0.019 |
anger | -0.019 |
Token | Tim |
Token ID | 14967 |
Feature activation | -2.16e+00 |
Pos logit contributions | |
different | +0.011 |
fast | +0.010 |
very | +0.009 |
slow | +0.009 |
really | +0.008 |
Neg logit contributions | |
lor | -0.029 |
ocket | -0.025 |
urger | -0.025 |
card | -0.024 |
oll | -0.023 |
Token | my |
Token ID | 1820 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
different | +0.045 |
fast | +0.044 |
slow | +0.040 |
very | +0.038 |
really | +0.037 |
Neg logit contributions | |
lor | -0.114 |
ocket | -0.105 |
urger | -0.100 |
oll | -0.095 |
card | -0.094 |
Token | went |
Token ID | 1816 |
Feature activation | -9.56e-03 |
Pos logit contributions | |
different | +0.041 |
fast | +0.040 |
slow | +0.036 |
late | +0.036 |
very | +0.034 |
Neg logit contributions | |
lor | -0.086 |
ocket | -0.078 |
urger | -0.077 |
ster | -0.071 |
oll | -0.071 |
Token | over |
Token ID | 625 |
Feature activation | -7.94e+00 |
Pos logit contributions | |
different | +0.000 |
fast | +0.000 |
slow | +0.000 |
very | +0.000 |
late | +0.000 |
Neg logit contributions | |
lor | -0.001 |
ocket | -0.001 |
urger | -0.000 |
card | -0.000 |
anger | -0.000 |
Token | to |
Token ID | 284 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
different | +0.108 |
fast | +0.098 |
slow | +0.088 |
late | +0.086 |
very | +0.080 |
Neg logit contributions | |
lor | -0.413 |
ocket | -0.396 |
urger | -0.372 |
anger | -0.364 |
uit | -0.361 |
Token | the |
Token ID | 262 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
different | +0.020 |
fast | +0.020 |
slow | +0.016 |
very | +0.016 |
late | +0.015 |
Neg logit contributions | |
lor | -0.098 |
ocket | -0.092 |
urger | -0.088 |
card | -0.084 |
oll | -0.083 |
Token | kids |
Token ID | 3988 |
Feature activation | -3.26e+00 |
Pos logit contributions | |
different | +0.035 |
fast | +0.034 |
late | +0.033 |
slow | +0.032 |
very | +0.031 |
Neg logit contributions | |
lor | -0.061 |
ocket | -0.056 |
urger | -0.053 |
card | -0.051 |
anger | -0.050 |
Token | and |
Token ID | 290 |
Feature activation | +3.04e-01 |
Pos logit contributions | |
different | +0.044 |
fast | +0.042 |
slow | +0.036 |
very | +0.036 |
really | +0.033 |
Neg logit contributions | |
lor | -0.201 |
ocket | -0.186 |
urger | -0.179 |
oll | -0.169 |
card | -0.169 |
Token | tried |
Token ID | 3088 |
Feature activation | -2.73e-01 |
Pos logit contributions | |
lor | +0.014 |
ocket | +0.013 |
urger | +0.013 |
card | +0.012 |
oll | +0.012 |
Neg logit contributions | |
different | -0.007 |
fast | -0.007 |
slow | -0.007 |
very | -0.006 |
late | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | -9.80e-01 |
Pos logit contributions | |
different | +0.016 |
gloomy | +0.014 |
very | +0.014 |
fast | +0.014 |
muddy | +0.013 |
Neg logit contributions | |
lor | -0.045 |
ocket | -0.037 |
urger | -0.036 |
card | -0.035 |
glass | -0.035 |
Token | John |
Token ID | 7554 |
Feature activation | -3.50e+00 |
Pos logit contributions | |
different | +0.022 |
fast | +0.019 |
very | +0.019 |
slow | +0.018 |
fun | +0.018 |
Neg logit contributions | |
lor | -0.048 |
ocket | -0.040 |
urger | -0.039 |
card | -0.039 |
glass | -0.038 |
Token | decided |
Token ID | 3066 |
Feature activation | +3.20e+00 |
Pos logit contributions | |
different | +0.071 |
fast | +0.070 |
slow | +0.063 |
very | +0.060 |
really | +0.057 |
Neg logit contributions | |
lor | -0.184 |
ocket | -0.171 |
urger | -0.163 |
oll | -0.154 |
card | -0.153 |
Token | to |
Token ID | 284 |
Feature activation | -4.40e-01 |
Pos logit contributions | |
lor | +0.207 |
ocket | +0.195 |
urger | +0.188 |
oll | +0.180 |
card | +0.178 |
Neg logit contributions | |
different | -0.029 |
fast | -0.029 |
slow | -0.022 |
very | -0.020 |
really | -0.018 |
Token | go |
Token ID | 467 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
different | +0.011 |
fast | +0.011 |
slow | +0.011 |
very | +0.010 |
really | +0.010 |
Neg logit contributions | |
lor | -0.021 |
ocket | -0.020 |
urger | -0.019 |
oll | -0.018 |
card | -0.017 |
Token | over |
Token ID | 625 |
Feature activation | -7.94e+00 |
Pos logit contributions | |
fast | +0.028 |
different | +0.025 |
slow | +0.023 |
very | +0.022 |
really | +0.020 |
Neg logit contributions | |
lor | -0.117 |
ocket | -0.109 |
urger | -0.107 |
oll | -0.101 |
bridge | -0.098 |
Token | to |
Token ID | 284 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
different | +0.097 |
fast | +0.092 |
slow | +0.077 |
very | +0.071 |
really | +0.065 |
Neg logit contributions | |
lor | -0.478 |
ocket | -0.449 |
urger | -0.430 |
card | -0.410 |
oll | -0.409 |
Token | the |
Token ID | 262 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
different | +0.021 |
fast | +0.021 |
slow | +0.017 |
very | +0.016 |
really | +0.015 |
Neg logit contributions | |
lor | -0.097 |
ocket | -0.091 |
urger | -0.087 |
card | -0.083 |
oll | -0.082 |
Token | child |
Token ID | 1200 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
different | +0.046 |
fast | +0.046 |
slow | +0.043 |
late | +0.041 |
very | +0.041 |
Neg logit contributions | |
lor | -0.063 |
ocket | -0.057 |
urger | -0.054 |
card | -0.052 |
ster | -0.050 |
Token | and |
Token ID | 290 |
Feature activation | +6.14e-01 |
Pos logit contributions | |
different | +0.013 |
fast | +0.012 |
very | +0.011 |
slow | +0.010 |
really | +0.010 |
Neg logit contributions | |
lor | -0.067 |
ocket | -0.063 |
urger | -0.060 |
oll | -0.057 |
anger | -0.057 |
Token | ask |
Token ID | 1265 |
Feature activation | +3.62e+00 |
Pos logit contributions | |
lor | +0.033 |
ocket | +0.030 |
urger | +0.029 |
oll | +0.027 |
card | +0.027 |
Neg logit contributions | |
different | -0.013 |
fast | -0.013 |
slow | -0.012 |
very | -0.011 |
really | -0.011 |
Token | in |
Token ID | 287 |
Feature activation | -4.61e+00 |
Pos logit contributions | |
different | +0.007 |
fast | +0.007 |
slow | +0.006 |
very | +0.006 |
really | +0.006 |
Neg logit contributions | |
lor | -0.019 |
ocket | -0.018 |
urger | -0.017 |
oll | -0.016 |
card | -0.016 |
Token | time |
Token ID | 640 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
different | +0.067 |
fast | +0.065 |
slow | +0.056 |
very | +0.053 |
really | +0.050 |
Neg logit contributions | |
lor | -0.262 |
ocket | -0.245 |
urger | -0.234 |
card | -0.223 |
oll | -0.222 |
Token | . |
Token ID | 13 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
different | +0.024 |
fast | +0.023 |
slow | +0.019 |
very | +0.018 |
really | +0.017 |
Neg logit contributions | |
lor | -0.103 |
ocket | -0.096 |
urger | -0.092 |
oll | -0.088 |
card | -0.088 |
Token | They |
Token ID | 1119 |
Feature activation | -4.80e-01 |
Pos logit contributions | |
different | +0.006 |
fast | +0.006 |
slow | +0.006 |
very | +0.005 |
really | +0.005 |
Neg logit contributions | |
lor | -0.019 |
ocket | -0.017 |
urger | -0.017 |
oll | -0.016 |
card | -0.016 |
Token | went |
Token ID | 1816 |
Feature activation | -2.39e-01 |
Pos logit contributions | |
different | +0.012 |
fast | +0.012 |
slow | +0.011 |
very | +0.011 |
really | +0.010 |
Neg logit contributions | |
lor | -0.023 |
ocket | -0.021 |
urger | -0.020 |
oll | -0.019 |
card | -0.019 |
Token | over |
Token ID | 625 |
Feature activation | -7.93e+00 |
Pos logit contributions | |
different | +0.004 |
fast | +0.004 |
slow | +0.004 |
very | +0.004 |
really | +0.003 |
Neg logit contributions | |
lor | -0.013 |
ocket | -0.012 |
urger | -0.012 |
card | -0.011 |
oll | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | -6.94e-01 |
Pos logit contributions | |
different | +0.107 |
fast | +0.097 |
late | +0.088 |
slow | +0.088 |
upset | +0.086 |
Neg logit contributions | |
lor | -0.397 |
ocket | -0.380 |
urger | -0.359 |
anger | -0.354 |
glass | -0.352 |
Token | big |
Token ID | 1263 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
fast | +0.014 |
late | +0.014 |
different | +0.014 |
Stop | +0.014 |
each | +0.013 |
Neg logit contributions | |
lor | -0.031 |
ocket | -0.028 |
urger | -0.027 |
card | -0.026 |
suitcase | -0.026 |
Token | hill |
Token ID | 12788 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
different | +0.055 |
fast | +0.054 |
late | +0.052 |
very | +0.049 |
really | +0.049 |
Neg logit contributions | |
lor | -0.073 |
ocket | -0.067 |
urger | -0.065 |
card | -0.061 |
suitcase | -0.059 |
Token | and |
Token ID | 290 |
Feature activation | +8.32e-01 |
Pos logit contributions | |
different | +0.033 |
fast | +0.030 |
Stop | +0.028 |
slow | +0.028 |
B | +0.027 |
Neg logit contributions | |
lor | -0.097 |
ocket | -0.087 |
urger | -0.084 |
card | -0.084 |
anger | -0.081 |
Token | down |
Token ID | 866 |
Feature activation | -3.87e+00 |
Pos logit contributions | |
lor | +0.039 |
ocket | +0.037 |
urger | +0.035 |
card | +0.034 |
oll | +0.033 |
Neg logit contributions | |
different | -0.018 |
fast | -0.017 |
slow | -0.015 |
very | -0.015 |
late | -0.014 |
Token | cry |
Token ID | 3960 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
fast | +0.017 |
different | +0.017 |
late | +0.017 |
B | +0.016 |
very | +0.015 |
Neg logit contributions | |
lor | -0.031 |
ocket | -0.029 |
urger | -0.029 |
icking | -0.026 |
card | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -5.82e-01 |
Pos logit contributions | |
different | +0.020 |
fast | +0.018 |
slow | +0.017 |
Stop | +0.017 |
late | +0.016 |
Neg logit contributions | |
lor | -0.056 |
ocket | -0.052 |
urger | -0.051 |
oll | -0.049 |
ster | -0.048 |
Token | Mom |
Token ID | 11254 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
fast | +0.014 |
different | +0.014 |
slow | +0.012 |
B | +0.012 |
competitive | +0.012 |
Neg logit contributions | |
lor | -0.027 |
ocket | -0.023 |
urger | -0.022 |
card | -0.021 |
anger | -0.021 |
Token | my |
Token ID | 1820 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
different | +0.042 |
fast | +0.041 |
slow | +0.037 |
very | +0.036 |
really | +0.034 |
Neg logit contributions | |
lor | -0.102 |
ocket | -0.095 |
urger | -0.091 |
oll | -0.085 |
card | -0.085 |
Token | went |
Token ID | 1816 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
different | +0.033 |
fast | +0.032 |
slow | +0.029 |
very | +0.028 |
really | +0.027 |
Neg logit contributions | |
lor | -0.077 |
ocket | -0.071 |
urger | -0.068 |
oll | -0.064 |
card | -0.064 |
Token | over |
Token ID | 625 |
Feature activation | -7.93e+00 |
Pos logit contributions | |
different | +0.002 |
fast | +0.001 |
slow | +0.001 |
very | +0.001 |
really | +0.001 |
Neg logit contributions | |
lor | -0.007 |
ocket | -0.006 |
urger | -0.006 |
card | -0.006 |
oll | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | +7.93e-01 |
Pos logit contributions | |
different | +0.126 |
fast | +0.120 |
slow | +0.111 |
very | +0.099 |
late | +0.097 |
Neg logit contributions | |
lor | -0.405 |
ocket | -0.387 |
urger | -0.359 |
uit | -0.349 |
card | -0.349 |
Token | told |
Token ID | 1297 |
Feature activation | +4.18e+00 |
Pos logit contributions | |
lor | +0.037 |
ocket | +0.034 |
urger | +0.032 |
card | +0.031 |
ster | +0.030 |
Neg logit contributions | |
different | -0.019 |
fast | -0.019 |
slow | -0.017 |
very | -0.016 |
late | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
lor | +0.232 |
ocket | +0.218 |
urger | +0.209 |
oll | +0.199 |
card | +0.197 |
Neg logit contributions | |
fast | -0.067 |
different | -0.067 |
slow | -0.059 |
late | -0.054 |
very | -0.054 |
Token | other |
Token ID | 584 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
different | +0.002 |
fast | +0.002 |
slow | +0.002 |
very | +0.002 |
really | +0.002 |
Neg logit contributions | |
lor | -0.007 |
ocket | -0.007 |
urger | -0.006 |
card | -0.006 |
oll | -0.006 |
Token | girl |
Token ID | 2576 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
different | +0.052 |
fast | +0.048 |
very | +0.045 |
slow | +0.044 |
really | +0.043 |
Neg logit contributions | |
lor | -0.105 |
ocket | -0.093 |
urger | -0.088 |
card | -0.082 |
oll | -0.081 |
Token | ! |
Token ID | 0 |
Feature activation | -9.58e-01 |
Pos logit contributions | |
fast | +0.001 |
different | +0.001 |
slow | +0.001 |
late | +0.001 |
B | +0.001 |
Neg logit contributions | |
lor | -0.003 |
ocket | -0.003 |
urger | -0.003 |
oll | -0.002 |
ster | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | -5.51e-01 |
Pos logit contributions | |
different | +0.018 |
fast | +0.017 |
very | +0.014 |
slow | +0.014 |
busy | +0.013 |
Neg logit contributions | |
lor | -0.051 |
ocket | -0.044 |
urger | -0.044 |
oll | -0.042 |
ster | -0.041 |
Token |
|
Token ID | 198 |
Feature activation | -6.64e-01 |
Pos logit contributions | |
different | +0.008 |
fast | +0.007 |
slow | +0.007 |
very | +0.007 |
competitive | +0.006 |
Neg logit contributions | |
lor | -0.031 |
ocket | -0.027 |
urger | -0.026 |
oll | -0.025 |
ogo | -0.025 |
Token | He |
Token ID | 1544 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
different | +0.013 |
fast | +0.012 |
very | +0.012 |
upset | +0.011 |
slow | +0.011 |
Neg logit contributions | |
lor | -0.034 |
urger | -0.029 |
ocket | -0.028 |
uit | -0.027 |
ster | -0.027 |
Token | went |
Token ID | 1816 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
lor | +0.068 |
ocket | +0.064 |
urger | +0.061 |
ster | +0.056 |
card | +0.056 |
Neg logit contributions | |
different | -0.035 |
fast | -0.034 |
late | -0.032 |
slow | -0.031 |
very | -0.030 |
Token | over |
Token ID | 625 |
Feature activation | -7.91e+00 |
Pos logit contributions | |
different | +0.001 |
fast | +0.001 |
slow | +0.001 |
late | +0.001 |
Stop | +0.001 |
Neg logit contributions | |
lor | -0.006 |
ocket | -0.006 |
urger | -0.005 |
card | -0.005 |
anger | -0.005 |
Token | to |
Token ID | 284 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
different | +0.115 |
fast | +0.109 |
upset | +0.105 |
slow | +0.103 |
Ste | +0.101 |
Neg logit contributions | |
ocket | -0.368 |
lor | -0.366 |
urger | -0.335 |
ster | -0.330 |
glass | -0.326 |
Token | the |
Token ID | 262 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
fast | +0.020 |
different | +0.020 |
late | +0.016 |
slow | +0.016 |
very | +0.015 |
Neg logit contributions | |
lor | -0.104 |
ocket | -0.100 |
urger | -0.096 |
ster | -0.090 |
card | -0.090 |
Token | frog |
Token ID | 21264 |
Feature activation | -2.36e+00 |
Pos logit contributions | |
different | +0.038 |
fast | +0.037 |
late | +0.035 |
slow | +0.034 |
very | +0.033 |
Neg logit contributions | |
lor | -0.060 |
ocket | -0.055 |
ster | -0.051 |
urger | -0.051 |
oll | -0.048 |
Token | , |
Token ID | 11 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
different | +0.029 |
fast | +0.027 |
very | +0.023 |
slow | +0.022 |
really | +0.020 |
Neg logit contributions | |
lor | -0.148 |
ocket | -0.138 |
urger | -0.133 |
oll | -0.127 |
card | -0.125 |
Token | but |
Token ID | 475 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
different | +0.037 |
fast | +0.036 |
slow | +0.032 |
late | +0.032 |
Stop | +0.031 |
Neg logit contributions | |
lor | -0.088 |
ocket | -0.084 |
urger | -0.080 |
ster | -0.075 |
card | -0.075 |
Token |
|
Token ID | 198 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
different | +0.012 |
fast | +0.012 |
slow | +0.009 |
very | +0.008 |
really | +0.007 |
Neg logit contributions | |
lor | -0.092 |
ocket | -0.087 |
urger | -0.084 |
oll | -0.080 |
card | -0.079 |
Token |
|
Token ID | 198 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
different | +0.014 |
fast | +0.014 |
slow | +0.012 |
very | +0.012 |
hurt | +0.010 |
Neg logit contributions | |
lor | -0.058 |
ocket | -0.052 |
urger | -0.048 |
card | -0.047 |
oll | -0.047 |
Token | Emily |
Token ID | 48640 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
different | +0.025 |
fast | +0.024 |
slow | +0.022 |
very | +0.021 |
late | +0.020 |
Neg logit contributions | |
lor | -0.057 |
ocket | -0.051 |
urger | -0.047 |
card | -0.046 |
ster | -0.045 |
Token | happily |
Token ID | 18177 |
Feature activation | -4.47e+00 |
Pos logit contributions | |
different | +0.022 |
fast | +0.021 |
slow | +0.019 |
very | +0.019 |
really | +0.018 |
Neg logit contributions | |
lor | -0.053 |
ocket | -0.049 |
urger | -0.047 |
oll | -0.044 |
card | -0.044 |
Token | ran |
Token ID | 4966 |
Feature activation | -3.77e-01 |
Pos logit contributions | |
different | +0.108 |
fast | +0.106 |
slow | +0.099 |
very | +0.097 |
really | +0.095 |
Neg logit contributions | |
lor | -0.226 |
ocket | -0.207 |
urger | -0.196 |
oll | -0.182 |
anger | -0.182 |
Token | over |
Token ID | 625 |
Feature activation | -7.91e+00 |
Pos logit contributions | |
different | +0.006 |
fast | +0.006 |
slow | +0.005 |
late | +0.005 |
Stop | +0.005 |
Neg logit contributions | |
lor | -0.019 |
ocket | -0.018 |
urger | -0.017 |
card | -0.017 |
anger | -0.017 |
Token | start |
Token ID | 923 |
Feature activation | -5.22e+00 |
Pos logit contributions | |
different | +0.085 |
fast | +0.081 |
slow | +0.069 |
late | +0.064 |
very | +0.060 |
Neg logit contributions | |
lor | -0.464 |
ocket | -0.441 |
urger | -0.416 |
card | -0.404 |
anger | -0.401 |
Token | playing |
Token ID | 2712 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
different | +0.073 |
fast | +0.071 |
slow | +0.061 |
very | +0.057 |
really | +0.053 |
Neg logit contributions | |
lor | -0.306 |
ocket | -0.286 |
urger | -0.275 |
card | -0.261 |
oll | -0.261 |
Token | with |
Token ID | 351 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
different | +0.026 |
fast | +0.025 |
slow | +0.022 |
very | +0.020 |
late | +0.020 |
Neg logit contributions | |
lor | -0.087 |
ocket | -0.082 |
urger | -0.078 |
card | -0.074 |
oll | -0.074 |
Token | everyone |
Token ID | 2506 |
Feature activation | -5.06e+00 |
Pos logit contributions | |
different | +0.012 |
fast | +0.012 |
slow | +0.010 |
very | +0.009 |
really | +0.008 |
Neg logit contributions | |
lor | -0.072 |
ocket | -0.068 |
urger | -0.065 |
oll | -0.062 |
card | -0.062 |
Token | else |
Token ID | 2073 |
Feature activation | -4.18e+00 |
Pos logit contributions | |
different | +0.092 |
fast | +0.092 |
slow | +0.084 |
late | +0.079 |
very | +0.074 |
Neg logit contributions | |
lor | -0.264 |
ocket | -0.247 |
urger | -0.235 |
card | -0.226 |
oll | -0.226 |
Token | . |
Token ID | 13 |
Feature activation | -4.38e-01 |
Pos logit contributions | |
lor | +0.127 |
ocket | +0.120 |
urger | +0.115 |
oll | +0.109 |
card | +0.107 |
Neg logit contributions | |
different | -0.026 |
fast | -0.026 |
slow | -0.021 |
very | -0.021 |
really | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | -6.18e-01 |
Pos logit contributions | |
different | +0.008 |
fast | +0.008 |
slow | +0.007 |
very | +0.007 |
really | +0.006 |
Neg logit contributions | |
lor | -0.024 |
ocket | -0.022 |
urger | -0.021 |
oll | -0.020 |
card | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -7.72e-01 |
Pos logit contributions | |
different | +0.007 |
fast | +0.006 |
slow | +0.005 |
very | +0.005 |
really | +0.004 |
Neg logit contributions | |
lor | -0.038 |
ocket | -0.035 |
urger | -0.034 |
oll | -0.032 |
card | -0.032 |
Token | Tom |
Token ID | 13787 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
different | +0.015 |
fast | +0.014 |
slow | +0.012 |
very | +0.012 |
really | +0.011 |
Neg logit contributions | |
lor | -0.042 |
ocket | -0.038 |
urger | -0.037 |
oll | -0.035 |
card | -0.035 |
Token | went |
Token ID | 1816 |
Feature activation | -3.24e-02 |
Pos logit contributions | |
different | +0.037 |
fast | +0.036 |
slow | +0.033 |
very | +0.032 |
really | +0.030 |
Neg logit contributions | |
lor | -0.089 |
ocket | -0.083 |
urger | -0.079 |
oll | -0.075 |
card | -0.074 |
Token | over |
Token ID | 625 |
Feature activation | -7.91e+00 |
Pos logit contributions | |
different | +0.000 |
fast | +0.000 |
slow | +0.000 |
very | +0.000 |
really | +0.000 |
Neg logit contributions | |
lor | -0.002 |
ocket | -0.002 |
urger | -0.002 |
oll | -0.002 |
card | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | +4.06e-01 |
Pos logit contributions | |
different | +0.093 |
fast | +0.086 |
slow | +0.075 |
very | +0.067 |
late | +0.065 |
Neg logit contributions | |
lor | -0.451 |
ocket | -0.431 |
urger | -0.409 |
card | -0.391 |
anger | -0.390 |
Token | carefully |
Token ID | 7773 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
lor | +0.020 |
ocket | +0.019 |
urger | +0.018 |
card | +0.017 |
ster | +0.017 |
Neg logit contributions | |
different | -0.009 |
fast | -0.008 |
slow | -0.008 |
very | -0.007 |
late | -0.007 |
Token | looked |
Token ID | 3114 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
different | +0.044 |
fast | +0.043 |
slow | +0.040 |
very | +0.039 |
really | +0.037 |
Neg logit contributions | |
lor | -0.090 |
ocket | -0.082 |
urger | -0.079 |
oll | -0.074 |
card | -0.072 |
Token | around |
Token ID | 1088 |
Feature activation | -5.20e+00 |
Pos logit contributions | |
lor | +0.064 |
ocket | +0.062 |
urger | +0.058 |
card | +0.054 |
oll | +0.054 |
Neg logit contributions | |
different | -0.029 |
fast | -0.029 |
slow | -0.027 |
late | -0.025 |
very | -0.024 |
Token | until |
Token ID | 1566 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
different | +0.082 |
fast | +0.082 |
slow | +0.075 |
B | +0.068 |
late | +0.066 |
Neg logit contributions | |
lor | -0.258 |
ocket | -0.251 |
urger | -0.237 |
card | -0.223 |
ster | -0.223 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | +4.60e-01 |
Pos logit contributions | |
fast | +0.007 |
different | +0.007 |
slow | +0.006 |
very | +0.006 |
really | +0.005 |
Neg logit contributions | |
lor | -0.022 |
ocket | -0.021 |
urger | -0.020 |
card | -0.019 |
oll | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
lor | +0.025 |
ocket | +0.024 |
urger | +0.023 |
oll | +0.021 |
card | +0.021 |
Neg logit contributions | |
different | -0.008 |
fast | -0.008 |
slow | -0.007 |
very | -0.007 |
really | -0.006 |
Token | he |
Token ID | 339 |
Feature activation | +5.75e-01 |
Pos logit contributions | |
different | +0.032 |
fast | +0.031 |
slow | +0.028 |
very | +0.027 |
really | +0.025 |
Neg logit contributions | |
lor | -0.087 |
ocket | -0.081 |
urger | -0.078 |
oll | -0.073 |
card | -0.073 |
Token | felt |
Token ID | 2936 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
lor | +0.032 |
ocket | +0.030 |
urger | +0.028 |
card | +0.028 |
oll | +0.027 |
Neg logit contributions | |
different | -0.010 |
fast | -0.010 |
slow | -0.009 |
very | -0.009 |
really | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | -6.34e-01 |
Pos logit contributions | |
fast | +0.031 |
different | +0.030 |
slow | +0.027 |
B | +0.026 |
Stop | +0.025 |
Neg logit contributions | |
lor | -0.092 |
ocket | -0.090 |
urger | -0.083 |
oll | -0.081 |
ster | -0.078 |
Token | gust |
Token ID | 35253 |
Feature activation | -7.90e+00 |
Pos logit contributions | |
different | +0.013 |
fast | +0.012 |
late | +0.012 |
slow | +0.011 |
B | +0.010 |
Neg logit contributions | |
lor | -0.030 |
ocket | -0.029 |
urger | -0.028 |
ster | -0.026 |
oll | -0.026 |
Token | of |
Token ID | 286 |
Feature activation | -7.75e-01 |
Pos logit contributions | |
different | +0.216 |
fast | +0.210 |
slow | +0.196 |
very | +0.187 |
really | +0.182 |
Neg logit contributions | |
lor | -0.356 |
ocket | -0.329 |
urger | -0.308 |
oll | -0.296 |
card | -0.291 |
Token | wind |
Token ID | 2344 |
Feature activation | -4.01e+00 |
Pos logit contributions | |
different | +0.014 |
fast | +0.013 |
slow | +0.012 |
really | +0.011 |
very | +0.011 |
Neg logit contributions | |
lor | -0.037 |
ocket | -0.037 |
urger | -0.035 |
oll | -0.034 |
card | -0.033 |
Token | ! |
Token ID | 0 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
different | +0.090 |
fast | +0.089 |
slow | +0.081 |
very | +0.079 |
really | +0.075 |
Neg logit contributions | |
lor | -0.203 |
ocket | -0.188 |
urger | -0.179 |
oll | -0.168 |
card | -0.167 |
Token | He |
Token ID | 679 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
different | +0.020 |
fast | +0.019 |
slow | +0.017 |
very | +0.017 |
really | +0.015 |
Neg logit contributions | |
lor | -0.054 |
ocket | -0.048 |
urger | -0.046 |
oll | -0.045 |
anger | -0.044 |
Token | was |
Token ID | 373 |
Feature activation | -5.20e-01 |
Pos logit contributions | |
lor | +0.049 |
ocket | +0.046 |
urger | +0.044 |
oll | +0.041 |
ster | +0.041 |
Neg logit contributions | |
different | -0.024 |
fast | -0.023 |
slow | -0.021 |
late | -0.021 |
very | -0.020 |
Token | the |
Token ID | 262 |
Feature activation | +2.22e-02 |
Pos logit contributions | |
different | +0.004 |
fast | +0.004 |
slow | +0.004 |
very | +0.004 |
really | +0.004 |
Neg logit contributions | |
lor | -0.014 |
ocket | -0.013 |
urger | -0.012 |
oll | -0.012 |
anger | -0.011 |
Token | angel |
Token ID | 18304 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
lor | +0.001 |
ocket | +0.001 |
urger | +0.001 |
card | +0.001 |
oll | +0.001 |
Neg logit contributions | |
different | -0.001 |
fast | -0.001 |
slow | -0.001 |
late | -0.001 |
really | -0.001 |
Token | felt |
Token ID | 2936 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
different | +0.048 |
fast | +0.047 |
slow | +0.043 |
very | +0.042 |
really | +0.041 |
Neg logit contributions | |
lor | -0.108 |
ocket | -0.099 |
urger | -0.095 |
oll | -0.091 |
card | -0.088 |
Token | a |
Token ID | 257 |
Feature activation | -5.91e-01 |
Pos logit contributions | |
fast | +0.028 |
different | +0.027 |
slow | +0.024 |
B | +0.023 |
Stop | +0.022 |
Neg logit contributions | |
lor | -0.093 |
ocket | -0.091 |
urger | -0.082 |
oll | -0.082 |
bridge | -0.078 |
Token | great |
Token ID | 1049 |
Feature activation | -3.14e+00 |
Pos logit contributions | |
different | +0.012 |
fast | +0.012 |
late | +0.012 |
slow | +0.010 |
B | +0.010 |
Neg logit contributions | |
lor | -0.027 |
ocket | -0.026 |
urger | -0.024 |
ster | -0.023 |
oll | -0.023 |
Token | gust |
Token ID | 35253 |
Feature activation | -7.90e+00 |
Pos logit contributions | |
different | +0.079 |
fast | +0.079 |
slow | +0.072 |
very | +0.069 |
late | +0.068 |
Neg logit contributions | |
lor | -0.144 |
ocket | -0.133 |
urger | -0.126 |
oll | -0.119 |
anger | -0.117 |
Token | of |
Token ID | 286 |
Feature activation | -9.40e-01 |
Pos logit contributions | |
different | +0.220 |
fast | +0.213 |
slow | +0.201 |
very | +0.191 |
really | +0.186 |
Neg logit contributions | |
lor | -0.353 |
ocket | -0.326 |
urger | -0.301 |
oll | -0.293 |
card | -0.288 |
Token | wind |
Token ID | 2344 |
Feature activation | -3.96e+00 |
Pos logit contributions | |
different | +0.017 |
fast | +0.016 |
really | +0.014 |
very | +0.014 |
slow | +0.014 |
Neg logit contributions | |
lor | -0.045 |
ocket | -0.045 |
urger | -0.041 |
oll | -0.041 |
card | -0.039 |
Token | and |
Token ID | 290 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
different | +0.086 |
fast | +0.085 |
slow | +0.077 |
very | +0.075 |
really | +0.071 |
Neg logit contributions | |
lor | -0.204 |
ocket | -0.189 |
urger | -0.180 |
oll | -0.170 |
card | -0.168 |
Token | her |
Token ID | 607 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
lor | +0.064 |
ocket | +0.059 |
urger | +0.057 |
oll | +0.054 |
card | +0.054 |
Neg logit contributions | |
different | -0.020 |
fast | -0.020 |
slow | -0.018 |
very | -0.017 |
really | -0.016 |
Token | wings |
Token ID | 12098 |
Feature activation | -3.69e-02 |
Pos logit contributions | |
lor | +0.081 |
ocket | +0.073 |
urger | +0.069 |
oll | +0.064 |
card | +0.063 |
Neg logit contributions | |
different | -0.065 |
fast | -0.064 |
slow | -0.060 |
very | -0.059 |
really | -0.057 |
Token |
|
Token ID | 198 |
Feature activation | -5.13e-01 |
Pos logit contributions | |
different | +0.006 |
fast | +0.006 |
slow | +0.005 |
very | +0.005 |
late | +0.005 |
Neg logit contributions | |
lor | -0.027 |
ocket | -0.024 |
urger | -0.023 |
anger | -0.022 |
card | -0.022 |
Token | Mar |
Token ID | 7676 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
different | +0.010 |
fast | +0.010 |
slow | +0.009 |
very | +0.008 |
late | +0.008 |
Neg logit contributions | |
lor | -0.028 |
ocket | -0.025 |
urger | -0.024 |
card | -0.023 |
oll | -0.023 |
Token | ina |
Token ID | 1437 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
lor | +0.050 |
ocket | +0.045 |
urger | +0.043 |
oll | +0.040 |
bridge | +0.038 |
Neg logit contributions | |
different | -0.059 |
fast | -0.059 |
slow | -0.056 |
very | -0.055 |
really | -0.053 |
Token | 's |
Token ID | 338 |
Feature activation | +2.39e+00 |
Pos logit contributions | |
lor | +0.065 |
ocket | +0.061 |
urger | +0.058 |
oll | +0.055 |
card | +0.055 |
Neg logit contributions | |
fast | -0.023 |
different | -0.023 |
slow | -0.021 |
very | -0.021 |
really | -0.020 |
Token | mom |
Token ID | 1995 |
Feature activation | -9.47e-01 |
Pos logit contributions | |
lor | +0.126 |
ocket | +0.108 |
urger | +0.104 |
oll | +0.097 |
card | +0.096 |
Neg logit contributions | |
fast | -0.066 |
different | -0.061 |
very | -0.060 |
slow | -0.060 |
hard | -0.053 |
Token | said |
Token ID | 531 |
Feature activation | +8.09e+00 |
Pos logit contributions | |
fast | +0.018 |
different | +0.018 |
slow | +0.016 |
very | +0.016 |
really | +0.016 |
Neg logit contributions | |
lor | -0.052 |
ocket | -0.049 |
urger | -0.046 |
oll | -0.044 |
anger | -0.044 |
Token | , |
Token ID | 11 |
Feature activation | +3.64e-02 |
Pos logit contributions | |
lor | +0.434 |
ocket | +0.402 |
urger | +0.383 |
card | +0.364 |
oll | +0.364 |
Neg logit contributions | |
different | -0.154 |
fast | -0.150 |
slow | -0.134 |
very | -0.125 |
late | -0.122 |
Token | " |
Token ID | 366 |
Feature activation | -3.51e+00 |
Pos logit contributions | |
lor | +0.002 |
ocket | +0.002 |
urger | +0.002 |
card | +0.001 |
oll | +0.001 |
Neg logit contributions | |
different | -0.001 |
fast | -0.001 |
slow | -0.001 |
very | -0.001 |
late | -0.001 |
Token | Oh |
Token ID | 5812 |
Feature activation | +3.01e+00 |
Pos logit contributions | |
different | +0.065 |
fast | +0.063 |
slow | +0.056 |
very | +0.055 |
really | +0.050 |
Neg logit contributions | |
lor | -0.193 |
ocket | -0.174 |
urger | -0.167 |
card | -0.161 |
oll | -0.157 |
Token | , |
Token ID | 11 |
Feature activation | -4.02e-01 |
Pos logit contributions | |
lor | +0.169 |
ocket | +0.162 |
urger | +0.156 |
oll | +0.146 |
card | +0.146 |
Neg logit contributions | |
different | -0.051 |
fast | -0.049 |
very | -0.046 |
slow | -0.043 |
really | -0.041 |
Token | it |
Token ID | 340 |
Feature activation | -9.90e-01 |
Pos logit contributions | |
different | +0.010 |
the | +0.009 |
fast | +0.009 |
very | +0.009 |
slow | +0.009 |
Neg logit contributions | |
lor | -0.020 |
ocket | -0.020 |
urger | -0.019 |
bridge | -0.018 |
oll | -0.018 |
Token | rice |
Token ID | 11464 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
lor | +0.016 |
ocket | +0.014 |
urger | +0.013 |
oll | +0.013 |
ogo | +0.013 |
Neg logit contributions | |
different | -0.011 |
fast | -0.010 |
very | -0.010 |
slow | -0.010 |
really | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -5.60e-01 |
Pos logit contributions | |
lor | +0.165 |
ocket | +0.154 |
urger | +0.148 |
oll | +0.140 |
card | +0.140 |
Neg logit contributions | |
different | -0.031 |
fast | -0.030 |
very | -0.025 |
slow | -0.024 |
really | -0.022 |
Token | The |
Token ID | 383 |
Feature activation | +7.60e-01 |
Pos logit contributions | |
different | +0.012 |
fast | +0.012 |
slow | +0.011 |
very | +0.010 |
B | +0.010 |
Neg logit contributions | |
lor | -0.026 |
ocket | -0.024 |
urger | -0.024 |
anger | -0.022 |
oll | -0.022 |
Token | shop |
Token ID | 6128 |
Feature activation | +4.97e-01 |
Pos logit contributions | |
lor | +0.043 |
ocket | +0.040 |
urger | +0.038 |
oll | +0.036 |
card | +0.036 |
Neg logit contributions | |
different | -0.014 |
fast | -0.014 |
slow | -0.012 |
very | -0.012 |
really | -0.011 |
Token | keeper |
Token ID | 13884 |
Feature activation | -2.50e+00 |
Pos logit contributions | |
lor | +0.018 |
ocket | +0.016 |
urger | +0.015 |
oll | +0.014 |
anger | +0.013 |
Neg logit contributions | |
different | -0.019 |
fast | -0.018 |
slow | -0.017 |
very | -0.017 |
really | -0.017 |
Token | said |
Token ID | 531 |
Feature activation | +7.55e+00 |
Pos logit contributions | |
different | +0.048 |
fast | +0.046 |
slow | +0.041 |
very | +0.041 |
really | +0.040 |
Neg logit contributions | |
lor | -0.138 |
ocket | -0.128 |
urger | -0.122 |
oll | -0.116 |
anger | -0.115 |
Token | he |
Token ID | 339 |
Feature activation | +2.55e+00 |
Pos logit contributions | |
lor | +0.405 |
ocket | +0.376 |
urger | +0.360 |
oll | +0.340 |
card | +0.338 |
Neg logit contributions | |
different | -0.147 |
fast | -0.145 |
slow | -0.129 |
very | -0.123 |
really | -0.118 |
Token | was |
Token ID | 373 |
Feature activation | +3.69e-01 |
Pos logit contributions | |
lor | +0.145 |
ocket | +0.131 |
urger | +0.125 |
anger | +0.120 |
card | +0.120 |
Neg logit contributions | |
different | -0.046 |
fast | -0.046 |
very | -0.044 |
really | -0.043 |
slow | -0.041 |
Token | too |
Token ID | 1165 |
Feature activation | -4.68e+00 |
Pos logit contributions | |
lor | +0.021 |
ocket | +0.019 |
urger | +0.018 |
oll | +0.017 |
card | +0.017 |
Neg logit contributions | |
different | -0.006 |
fast | -0.006 |
slow | -0.005 |
very | -0.005 |
really | -0.005 |
Token | young |
Token ID | 1862 |
Feature activation | -3.66e+00 |
Pos logit contributions | |
Stop | +0.080 |
B | +0.080 |
different | +0.079 |
fast | +0.070 |
opposite | +0.069 |
Neg logit contributions | |
lor | -0.220 |
ocket | -0.214 |
urger | -0.212 |
ster | -0.197 |
oll | -0.196 |
Token | and |
Token ID | 290 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
different | +0.037 |
fast | +0.036 |
slow | +0.029 |
very | +0.026 |
really | +0.023 |
Neg logit contributions | |
lor | -0.230 |
ocket | -0.216 |
urger | -0.208 |
oll | -0.198 |
card | -0.198 |
Token | . |
Token ID | 13 |
Feature activation | +6.05e-01 |
Pos logit contributions | |
lor | +0.033 |
ocket | +0.031 |
urger | +0.030 |
card | +0.029 |
oll | +0.027 |
Neg logit contributions | |
different | -0.005 |
fast | -0.004 |
very | -0.004 |
slow | -0.003 |
really | -0.003 |
Token | But |
Token ID | 887 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
lor | +0.031 |
ocket | +0.029 |
oll | +0.027 |
urger | +0.026 |
ster | +0.025 |
Neg logit contributions | |
different | -0.012 |
fast | -0.012 |
slow | -0.011 |
very | -0.010 |
really | -0.010 |
Token | he |
Token ID | 339 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
lor | +0.107 |
ocket | +0.096 |
urger | +0.094 |
card | +0.088 |
anger | +0.087 |
Neg logit contributions | |
different | -0.047 |
fast | -0.046 |
very | -0.042 |
slow | -0.042 |
really | -0.039 |
Token | did |
Token ID | 750 |
Feature activation | +2.44e+00 |
Pos logit contributions | |
lor | +0.081 |
ocket | +0.076 |
urger | +0.072 |
oll | +0.068 |
card | +0.068 |
Neg logit contributions | |
different | -0.031 |
fast | -0.031 |
slow | -0.027 |
very | -0.026 |
late | -0.025 |
Token | not |
Token ID | 407 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
lor | +0.126 |
ocket | +0.115 |
urger | +0.110 |
card | +0.104 |
oll | +0.103 |
Neg logit contributions | |
different | -0.058 |
fast | -0.057 |
slow | -0.052 |
very | -0.051 |
really | -0.048 |
Token | say |
Token ID | 910 |
Feature activation | +8.35e+00 |
Pos logit contributions | |
fast | +0.029 |
different | +0.029 |
slow | +0.025 |
late | +0.025 |
very | +0.025 |
Neg logit contributions | |
lor | -0.057 |
ocket | -0.056 |
urger | -0.053 |
oll | -0.050 |
ster | -0.048 |
Token | anything |
Token ID | 1997 |
Feature activation | -3.92e-01 |
Pos logit contributions | |
lor | +0.401 |
ocket | +0.367 |
urger | +0.352 |
card | +0.327 |
oll | +0.323 |
Neg logit contributions | |
different | -0.227 |
fast | -0.220 |
slow | -0.203 |
very | -0.202 |
really | -0.188 |
Token | . |
Token ID | 13 |
Feature activation | +7.63e-01 |
Pos logit contributions | |
fast | +0.007 |
different | +0.007 |
slow | +0.006 |
late | +0.006 |
panic | +0.006 |
Neg logit contributions | |
lor | -0.020 |
ocket | -0.019 |
oll | -0.018 |
urger | -0.017 |
ster | -0.017 |
Token | He |
Token ID | 679 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
lor | +0.040 |
ocket | +0.037 |
oll | +0.035 |
urger | +0.035 |
ster | +0.033 |
Neg logit contributions | |
different | -0.014 |
fast | -0.014 |
slow | -0.012 |
very | -0.012 |
really | -0.011 |
Token | ate |
Token ID | 15063 |
Feature activation | +2.99e+00 |
Pos logit contributions | |
lor | +0.114 |
ocket | +0.104 |
urger | +0.100 |
card | +0.095 |
anger | +0.094 |
Neg logit contributions | |
different | -0.044 |
fast | -0.043 |
slow | -0.039 |
very | -0.038 |
really | -0.037 |
Token | his |
Token ID | 465 |
Feature activation | +2.92e+00 |
Pos logit contributions | |
lor | +0.163 |
ocket | +0.155 |
urger | +0.147 |
oll | +0.142 |
ster | +0.138 |
Neg logit contributions | |
different | -0.046 |
fast | -0.046 |
slow | -0.041 |
late | -0.037 |
very | -0.036 |
Token | song |
Token ID | 3496 |
Feature activation | +9.54e-01 |
Pos logit contributions | |
lor | +0.003 |
ocket | +0.002 |
urger | +0.002 |
oll | +0.002 |
ogo | +0.002 |
Neg logit contributions | |
fast | -0.001 |
different | -0.001 |
slow | -0.001 |
very | -0.001 |
fun | -0.001 |
Token | was |
Token ID | 373 |
Feature activation | -4.69e-01 |
Pos logit contributions | |
lor | +0.047 |
ocket | +0.043 |
urger | +0.041 |
card | +0.039 |
oll | +0.039 |
Neg logit contributions | |
different | -0.021 |
fast | -0.021 |
slow | -0.019 |
very | -0.018 |
late | -0.018 |
Token | over |
Token ID | 625 |
Feature activation | -7.02e+00 |
Pos logit contributions | |
different | +0.008 |
fast | +0.008 |
B | +0.008 |
slow | +0.007 |
Stop | +0.007 |
Neg logit contributions | |
lor | -0.022 |
ocket | -0.020 |
urger | -0.020 |
card | -0.020 |
icking | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
different | +0.140 |
fast | +0.139 |
slow | +0.124 |
very | +0.119 |
really | +0.114 |
Neg logit contributions | |
lor | -0.345 |
ocket | -0.320 |
urger | -0.309 |
anger | -0.292 |
ster | -0.291 |
Token | Bobby |
Token ID | 17155 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
different | +0.027 |
fast | +0.026 |
slow | +0.023 |
very | +0.021 |
really | +0.020 |
Neg logit contributions | |
lor | -0.095 |
ocket | -0.089 |
urger | -0.085 |
oll | -0.081 |
card | -0.080 |
Token | said |
Token ID | 531 |
Feature activation | +7.35e+00 |
Pos logit contributions | |
different | +0.031 |
fast | +0.030 |
slow | +0.027 |
very | +0.026 |
late | +0.026 |
Neg logit contributions | |
lor | -0.069 |
ocket | -0.063 |
urger | -0.061 |
oll | -0.057 |
card | -0.057 |
Token | , |
Token ID | 11 |
Feature activation | -4.91e-01 |
Pos logit contributions | |
lor | +0.385 |
ocket | +0.356 |
urger | +0.341 |
card | +0.322 |
oll | +0.322 |
Neg logit contributions | |
different | -0.147 |
fast | -0.143 |
slow | -0.128 |
late | -0.120 |
very | -0.120 |
Token | " |
Token ID | 366 |
Feature activation | -3.91e+00 |
Pos logit contributions | |
different | +0.010 |
fast | +0.010 |
slow | +0.009 |
very | +0.009 |
really | +0.009 |
Neg logit contributions | |
lor | -0.025 |
ocket | -0.023 |
urger | -0.022 |
oll | -0.021 |
card | -0.021 |
Token | Mom |
Token ID | 29252 |
Feature activation | -7.16e-01 |
Pos logit contributions | |
different | +0.068 |
fast | +0.067 |
slow | +0.058 |
very | +0.056 |
really | +0.052 |
Neg logit contributions | |
lor | -0.219 |
ocket | -0.202 |
urger | -0.195 |
card | -0.184 |
oll | -0.184 |
Token | my |
Token ID | 1820 |
Feature activation | -6.81e-01 |
Pos logit contributions | |
different | +0.012 |
fast | +0.011 |
slow | +0.010 |
very | +0.010 |
really | +0.010 |
Neg logit contributions | |
lor | -0.042 |
ocket | -0.040 |
urger | -0.036 |
oll | -0.035 |
anger | -0.035 |
Token | , |
Token ID | 11 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
different | +0.009 |
fast | +0.009 |
slow | +0.008 |
very | +0.008 |
really | +0.007 |
Neg logit contributions | |
lor | -0.042 |
ocket | -0.039 |
urger | -0.037 |
oll | -0.035 |
card | -0.035 |
Token |
|
Token ID | 198 |
Feature activation | -3.64e-01 |
Pos logit contributions | |
different | +0.004 |
fast | +0.004 |
very | +0.003 |
slow | +0.003 |
late | +0.003 |
Neg logit contributions | |
lor | -0.019 |
ocket | -0.017 |
urger | -0.016 |
oll | -0.016 |
anger | -0.015 |
Token | Danny |
Token ID | 45478 |
Feature activation | -6.41e-01 |
Pos logit contributions | |
different | +0.009 |
fast | +0.009 |
very | +0.008 |
slow | +0.008 |
really | +0.007 |
Neg logit contributions | |
lor | -0.017 |
ocket | -0.015 |
urger | -0.015 |
oll | -0.014 |
ster | -0.013 |
Token | was |
Token ID | 373 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
different | +0.014 |
fast | +0.014 |
late | +0.013 |
slow | +0.013 |
very | +0.012 |
Neg logit contributions | |
lor | -0.031 |
ocket | -0.029 |
urger | -0.028 |
ster | -0.026 |
oll | -0.026 |
Token | surprised |
Token ID | 6655 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
lor | +0.007 |
ocket | +0.007 |
urger | +0.007 |
card | +0.006 |
oll | +0.006 |
Neg logit contributions | |
different | -0.002 |
B | -0.002 |
fast | -0.002 |
Stop | -0.002 |
late | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | +1.75e+00 |
Pos logit contributions | |
fast | +0.036 |
different | +0.036 |
slow | +0.032 |
very | +0.029 |
late | +0.028 |
Neg logit contributions | |
lor | -0.118 |
ocket | -0.110 |
urger | -0.107 |
oll | -0.101 |
ster | -0.100 |
Token | said |
Token ID | 531 |
Feature activation | +7.82e+00 |
Pos logit contributions | |
lor | +0.080 |
ocket | +0.076 |
urger | +0.072 |
ster | +0.069 |
card | +0.068 |
Neg logit contributions | |
different | -0.038 |
fast | -0.036 |
slow | -0.033 |
late | -0.032 |
very | -0.031 |
Token | � |
Token ID | 6184 |
Feature activation | +3.42e+00 |
Pos logit contributions | |
lor | +0.393 |
ocket | +0.364 |
urger | +0.344 |
ster | +0.334 |
card | +0.329 |
Neg logit contributions | |
different | -0.157 |
fast | -0.152 |
late | -0.138 |
slow | -0.138 |
busy | -0.124 |
Token | � |
Token ID | 95 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
lor | +0.127 |
ocket | +0.115 |
urger | +0.107 |
oll | +0.098 |
ster | +0.097 |
Neg logit contributions | |
different | -0.123 |
fast | -0.119 |
slow | -0.112 |
very | -0.111 |
late | -0.108 |
Token | € |
Token ID | 26391 |
Feature activation | +3.49e-01 |
Pos logit contributions | |
lor | +0.077 |
ocket | +0.069 |
urger | +0.065 |
card | +0.060 |
oll | +0.059 |
Neg logit contributions | |
different | -0.065 |
fast | -0.063 |
slow | -0.058 |
very | -0.058 |
really | -0.055 |
Token | � |
Token ID | 129 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
lor | +0.016 |
ocket | +0.015 |
urger | +0.014 |
oll | +0.013 |
card | +0.013 |
Neg logit contributions | |
fast | -0.010 |
different | -0.009 |
slow | -0.009 |
very | -0.009 |
really | -0.008 |
Token | � |
Token ID | 241 |
Feature activation | -3.04e+00 |
Pos logit contributions | |
lor | +0.055 |
ocket | +0.050 |
urger | +0.048 |
card | +0.044 |
oll | +0.043 |
Neg logit contributions | |
different | -0.041 |
fast | -0.040 |
slow | -0.037 |
very | -0.036 |
late | -0.036 |
Token | she |
Token ID | 673 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
lor | +0.109 |
ocket | +0.101 |
urger | +0.098 |
oll | +0.092 |
card | +0.092 |
Neg logit contributions | |
different | -0.035 |
fast | -0.035 |
slow | -0.031 |
very | -0.030 |
really | -0.028 |
Token | decided |
Token ID | 3066 |
Feature activation | +4.78e+00 |
Pos logit contributions | |
lor | +0.060 |
ocket | +0.055 |
urger | +0.052 |
oll | +0.049 |
ster | +0.048 |
Neg logit contributions | |
different | -0.034 |
fast | -0.033 |
slow | -0.030 |
very | -0.029 |
late | -0.028 |
Token | to |
Token ID | 284 |
Feature activation | +9.91e-01 |
Pos logit contributions | |
lor | +0.296 |
ocket | +0.275 |
urger | +0.265 |
card | +0.252 |
oll | +0.251 |
Neg logit contributions | |
different | -0.048 |
fast | -0.046 |
slow | -0.036 |
very | -0.031 |
late | -0.029 |
Token | take |
Token ID | 1011 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
lor | +0.047 |
ocket | +0.044 |
urger | +0.042 |
card | +0.040 |
oll | +0.039 |
Neg logit contributions | |
different | -0.026 |
fast | -0.026 |
slow | -0.025 |
very | -0.024 |
really | -0.023 |
Token | a |
Token ID | 257 |
Feature activation | +7.31e-01 |
Pos logit contributions | |
lor | +0.134 |
ocket | +0.125 |
urger | +0.116 |
oll | +0.116 |
anger | +0.108 |
Neg logit contributions | |
different | -0.043 |
fast | -0.041 |
slow | -0.037 |
late | -0.033 |
competitive | -0.032 |
Token | nap |
Token ID | 25422 |
Feature activation | +2.60e+00 |
Pos logit contributions | |
lor | +0.034 |
ocket | +0.031 |
urger | +0.030 |
card | +0.028 |
oll | +0.028 |
Neg logit contributions | |
different | -0.021 |
fast | -0.021 |
slow | -0.019 |
very | -0.019 |
really | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
lor | +0.155 |
ocket | +0.146 |
urger | +0.142 |
card | +0.134 |
oll | +0.132 |
Neg logit contributions | |
fast | -0.036 |
different | -0.036 |
very | -0.031 |
slow | -0.030 |
really | -0.027 |
Token | She |
Token ID | 1375 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
fast | +0.003 |
different | +0.003 |
slow | +0.003 |
competitive | +0.003 |
late | +0.003 |
Neg logit contributions | |
lor | -0.007 |
ocket | -0.006 |
urger | -0.006 |
anger | -0.006 |
oll | -0.006 |
Token | found |
Token ID | 1043 |
Feature activation | +1.94e+00 |
Pos logit contributions | |
lor | +0.080 |
ocket | +0.073 |
urger | +0.069 |
card | +0.066 |
anger | +0.065 |
Neg logit contributions | |
different | -0.046 |
fast | -0.045 |
slow | -0.042 |
very | -0.042 |
really | -0.041 |
Token | a |
Token ID | 257 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
lor | +0.111 |
ocket | +0.104 |
urger | +0.099 |
oll | +0.096 |
card | +0.094 |
Neg logit contributions | |
different | -0.028 |
fast | -0.028 |
slow | -0.025 |
very | -0.022 |
really | -0.021 |
Token | cozy |
Token ID | 37438 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
lor | +0.083 |
ocket | +0.077 |
urger | +0.074 |
oll | +0.069 |
ster | +0.069 |
Neg logit contributions | |
different | -0.035 |
fast | -0.034 |
slow | -0.031 |
late | -0.030 |
very | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | -5.40e-01 |
Pos logit contributions | |
different | +0.009 |
fast | +0.009 |
slow | +0.008 |
late | +0.008 |
competitive | +0.007 |
Neg logit contributions | |
lor | -0.016 |
ocket | -0.014 |
anger | -0.013 |
urger | -0.013 |
oll | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | -5.61e-01 |
Pos logit contributions | |
different | +0.009 |
fast | +0.008 |
slow | +0.008 |
very | +0.007 |
competitive | +0.007 |
Neg logit contributions | |
lor | -0.029 |
ocket | -0.026 |
oll | -0.023 |
anger | -0.023 |
urger | -0.023 |
Token | Soon |
Token ID | 28093 |
Feature activation | +6.47e-01 |
Pos logit contributions | |
different | +0.014 |
fast | +0.013 |
slow | +0.011 |
very | +0.011 |
each | +0.011 |
Neg logit contributions | |
lor | -0.026 |
ocket | -0.022 |
card | -0.020 |
oll | -0.020 |
ster | -0.020 |
Token | she |
Token ID | 673 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
lor | +0.036 |
ocket | +0.033 |
urger | +0.032 |
oll | +0.030 |
card | +0.030 |
Neg logit contributions | |
different | -0.011 |
fast | -0.011 |
slow | -0.010 |
very | -0.009 |
really | -0.009 |
Token | realized |
Token ID | 6939 |
Feature activation | +4.02e+00 |
Pos logit contributions | |
lor | +0.057 |
ocket | +0.053 |
urger | +0.051 |
card | +0.048 |
oll | +0.048 |
Neg logit contributions | |
different | -0.021 |
fast | -0.020 |
slow | -0.018 |
very | -0.018 |
really | -0.017 |
Token | that |
Token ID | 326 |
Feature activation | +2.70e+00 |
Pos logit contributions | |
lor | +0.196 |
ocket | +0.180 |
urger | +0.172 |
oll | +0.161 |
card | +0.160 |
Neg logit contributions | |
different | -0.099 |
fast | -0.098 |
slow | -0.089 |
very | -0.087 |
really | -0.084 |
Token | she |
Token ID | 673 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
lor | +0.143 |
ocket | +0.136 |
urger | +0.125 |
oll | +0.123 |
card | +0.123 |
Neg logit contributions | |
fast | -0.043 |
different | -0.042 |
slow | -0.037 |
really | -0.034 |
very | -0.034 |
Token | would |
Token ID | 561 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
lor | +0.101 |
ocket | +0.094 |
urger | +0.090 |
oll | +0.085 |
card | +0.085 |
Neg logit contributions | |
different | -0.037 |
fast | -0.036 |
slow | -0.032 |
very | -0.031 |
really | -0.029 |
Token | have |
Token ID | 423 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
lor | +0.085 |
ocket | +0.078 |
urger | +0.074 |
oll | +0.070 |
card | +0.069 |
Neg logit contributions | |
different | -0.042 |
fast | -0.041 |
slow | -0.038 |
very | -0.037 |
really | -0.035 |
Token | to |
Token ID | 284 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
fast | +0.029 |
different | +0.026 |
slow | +0.025 |
low | +0.024 |
wildly | +0.024 |
Neg logit contributions | |
lor | -0.063 |
ocket | -0.055 |
urger | -0.053 |
oll | -0.050 |
icking | -0.048 |
Token | suffer |
Token ID | 8659 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
lor | +0.063 |
ocket | +0.059 |
urger | +0.055 |
oll | +0.053 |
card | +0.052 |
Neg logit contributions | |
different | -0.032 |
fast | -0.031 |
slow | -0.030 |
very | -0.028 |
really | -0.027 |
Token | continued |
Token ID | 3767 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
different | +0.001 |
fast | +0.001 |
very | +0.001 |
slow | +0.001 |
really | +0.001 |
Neg logit contributions | |
lor | -0.004 |
ocket | -0.004 |
urger | -0.003 |
anger | -0.003 |
card | -0.003 |
Token | their |
Token ID | 511 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
different | +0.036 |
fast | +0.036 |
Stop | +0.032 |
slow | +0.032 |
really | +0.030 |
Neg logit contributions | |
lor | -0.113 |
ocket | -0.104 |
urger | -0.103 |
card | -0.098 |
oll | -0.095 |
Token | walk |
Token ID | 2513 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
lor | +0.058 |
ocket | +0.053 |
urger | +0.050 |
oll | +0.046 |
card | +0.046 |
Neg logit contributions | |
different | -0.041 |
fast | -0.040 |
slow | -0.037 |
very | -0.036 |
really | -0.035 |
Token | around |
Token ID | 1088 |
Feature activation | -5.83e+00 |
Pos logit contributions | |
lor | +0.058 |
card | +0.052 |
ocket | +0.052 |
anger | +0.050 |
urger | +0.049 |
Neg logit contributions | |
Stop | -0.030 |
Enough | -0.027 |
different | -0.026 |
panic | -0.024 |
Acc | -0.024 |
Token | the |
Token ID | 262 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
Stop | +0.106 |
umpy | +0.100 |
Enough | +0.099 |
Girls | +0.096 |
late | +0.095 |
Neg logit contributions | |
ocket | -0.228 |
lor | -0.227 |
glass | -0.209 |
ster | -0.208 |
suitcase | -0.201 |
Token | harbor |
Token ID | 25451 |
Feature activation | +2.30e+00 |
Pos logit contributions | |
different | +0.009 |
fast | +0.009 |
slow | +0.008 |
very | +0.008 |
really | +0.008 |
Neg logit contributions | |
lor | -0.009 |
ocket | -0.008 |
urger | -0.008 |
card | -0.007 |
ster | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +2.88e-02 |
Pos logit contributions | |
lor | +0.131 |
ocket | +0.121 |
urger | +0.117 |
card | +0.112 |
oll | +0.110 |
Neg logit contributions | |
different | -0.035 |
fast | -0.034 |
slow | -0.029 |
very | -0.027 |
really | -0.027 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +4.62e+00 |
Pos logit contributions | |
lor | +0.002 |
ocket | +0.002 |
urger | +0.001 |
oll | +0.001 |
card | +0.001 |
Neg logit contributions | |
different | -0.000 |
fast | -0.000 |
slow | -0.000 |
very | -0.000 |
really | -0.000 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.62e+00 |
Pos logit contributions | |
lor | +0.261 |
ocket | +0.235 |
urger | +0.217 |
oll | +0.217 |
anger | +0.213 |
Neg logit contributions | |
different | -0.069 |
fast | -0.065 |
slow | -0.060 |
very | -0.052 |
competitive | -0.051 |
Token | upon |
Token ID | 2402 |
Feature activation | +3.06e+00 |
Pos logit contributions | |
lor | +0.209 |
ocket | +0.191 |
urger | +0.181 |
oll | +0.169 |
card | +0.168 |
Neg logit contributions | |
different | -0.128 |
fast | -0.126 |
slow | -0.117 |
very | -0.114 |
really | -0.110 |
Token | a |
Token ID | 257 |
Feature activation | +7.00e-01 |
Pos logit contributions | |
lor | +0.156 |
ocket | +0.147 |
urger | +0.139 |
oll | +0.134 |
bridge | +0.131 |
Neg logit contributions | |
different | -0.051 |
fast | -0.050 |
slow | -0.047 |
Stop | -0.041 |
late | -0.039 |
Token | searching |
Token ID | 10342 |
Feature activation | +5.22e+00 |
Pos logit contributions | |
different | +0.009 |
fast | +0.008 |
slow | +0.008 |
very | +0.008 |
really | +0.007 |
Neg logit contributions | |
lor | -0.012 |
ocket | -0.011 |
urger | -0.011 |
oll | -0.010 |
card | -0.010 |
Token | for |
Token ID | 329 |
Feature activation | +2.39e+00 |
Pos logit contributions | |
lor | +0.264 |
ocket | +0.248 |
urger | +0.235 |
oll | +0.221 |
card | +0.220 |
Neg logit contributions | |
fast | -0.107 |
different | -0.107 |
slow | -0.098 |
very | -0.091 |
late | -0.088 |
Token | his |
Token ID | 465 |
Feature activation | +4.35e+00 |
Pos logit contributions | |
lor | +0.128 |
ocket | +0.118 |
urger | +0.114 |
card | +0.108 |
oll | +0.107 |
Neg logit contributions | |
different | -0.050 |
fast | -0.048 |
slow | -0.043 |
very | -0.042 |
really | -0.040 |
Token | favourite |
Token ID | 12507 |
Feature activation | +2.89e+00 |
Pos logit contributions | |
lor | +0.210 |
ocket | +0.180 |
urger | +0.170 |
oll | +0.163 |
anger | +0.161 |
Neg logit contributions | |
different | -0.131 |
fast | -0.125 |
very | -0.122 |
slow | -0.116 |
really | -0.114 |
Token | snack |
Token ID | 26906 |
Feature activation | +3.31e+00 |
Pos logit contributions | |
lor | +0.134 |
ocket | +0.121 |
urger | +0.113 |
oll | +0.110 |
card | +0.107 |
Neg logit contributions | |
different | -0.087 |
fast | -0.082 |
very | -0.078 |
slow | -0.076 |
really | -0.074 |
Token | when |
Token ID | 618 |
Feature activation | +3.35e+00 |
Pos logit contributions | |
lor | +0.202 |
ocket | +0.189 |
urger | +0.182 |
card | +0.173 |
oll | +0.172 |
Neg logit contributions | |
different | -0.041 |
fast | -0.040 |
very | -0.035 |
slow | -0.032 |
really | -0.031 |
Token | he |
Token ID | 339 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
lor | +0.175 |
ocket | +0.163 |
urger | +0.155 |
oll | +0.147 |
card | +0.146 |
Neg logit contributions | |
different | -0.075 |
fast | -0.074 |
very | -0.067 |
slow | -0.066 |
really | -0.063 |
Token | suddenly |
Token ID | 6451 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
lor | +0.098 |
ocket | +0.088 |
urger | +0.087 |
card | +0.085 |
anger | +0.085 |
Neg logit contributions | |
different | -0.030 |
fast | -0.028 |
really | -0.028 |
very | -0.028 |
slow | -0.026 |
Token | came |
Token ID | 1625 |
Feature activation | -2.83e-01 |
Pos logit contributions | |
different | +0.030 |
fast | +0.030 |
slow | +0.027 |
very | +0.026 |
really | +0.025 |
Neg logit contributions | |
lor | -0.071 |
ocket | -0.066 |
urger | -0.063 |
oll | -0.060 |
card | -0.060 |
Token | across |
Token ID | 1973 |
Feature activation | -4.21e+00 |
Pos logit contributions | |
different | +0.005 |
fast | +0.004 |
slow | +0.004 |
very | +0.003 |
late | +0.003 |
Neg logit contributions | |
lor | -0.016 |
ocket | -0.015 |
urger | -0.014 |
card | -0.013 |
ster | -0.013 |
Token | a |
Token ID | 257 |
Feature activation | +2.85e-01 |
Pos logit contributions | |
different | +0.060 |
fast | +0.059 |
slow | +0.051 |
very | +0.047 |
really | +0.044 |
Neg logit contributions | |
lor | -0.243 |
ocket | -0.228 |
urger | -0.218 |
card | -0.206 |
oll | -0.206 |
Token | the |
Token ID | 262 |
Feature activation | +7.39e-01 |
Pos logit contributions | |
fast | +0.030 |
different | +0.029 |
slow | +0.024 |
very | +0.022 |
late | +0.022 |
Neg logit contributions | |
lor | -0.169 |
ocket | -0.162 |
urger | -0.153 |
card | -0.146 |
oll | -0.145 |
Token | mirror |
Token ID | 10162 |
Feature activation | -5.38e-01 |
Pos logit contributions | |
lor | +0.033 |
ocket | +0.029 |
urger | +0.029 |
card | +0.026 |
oll | +0.026 |
Neg logit contributions | |
different | -0.023 |
fast | -0.022 |
slow | -0.021 |
very | -0.021 |
really | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
different | +0.006 |
fast | +0.006 |
very | +0.005 |
slow | +0.004 |
really | +0.004 |
Neg logit contributions | |
lor | -0.034 |
ocket | -0.032 |
urger | -0.031 |
card | -0.029 |
oll | -0.029 |
Token | She |
Token ID | 1375 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
different | +0.004 |
fast | +0.004 |
slow | +0.003 |
very | +0.003 |
competitive | +0.003 |
Neg logit contributions | |
lor | -0.012 |
ocket | -0.010 |
anger | -0.010 |
card | -0.010 |
oll | -0.010 |
Token | thought |
Token ID | 1807 |
Feature activation | +6.44e+00 |
Pos logit contributions | |
lor | +0.054 |
ocket | +0.053 |
urger | +0.049 |
ster | +0.047 |
oll | +0.047 |
Neg logit contributions | |
late | -0.028 |
fast | -0.027 |
different | -0.027 |
slow | -0.024 |
busy | -0.023 |
Token | she |
Token ID | 673 |
Feature activation | +2.50e+00 |
Pos logit contributions | |
lor | +0.366 |
ocket | +0.337 |
urger | +0.327 |
card | +0.309 |
oll | +0.304 |
Neg logit contributions | |
different | -0.114 |
fast | -0.112 |
slow | -0.098 |
very | -0.094 |
really | -0.088 |
Token | looked |
Token ID | 3114 |
Feature activation | +2.81e+00 |
Pos logit contributions | |
lor | +0.123 |
ocket | +0.113 |
urger | +0.108 |
oll | +0.101 |
card | +0.101 |
Neg logit contributions | |
different | -0.060 |
fast | -0.059 |
slow | -0.054 |
very | -0.052 |
late | -0.050 |
Token | very |
Token ID | 845 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
ocket | +0.142 |
lor | +0.140 |
urger | +0.135 |
oll | +0.126 |
card | +0.125 |
Neg logit contributions | |
fast | -0.034 |
late | -0.032 |
different | -0.031 |
slow | -0.030 |
B | -0.030 |
Token | cool |
Token ID | 3608 |
Feature activation | -6.26e-01 |
Pos logit contributions | |
Stop | +0.023 |
opposite | +0.023 |
Da | +0.022 |
late | +0.022 |
B | +0.022 |
Neg logit contributions | |
ocket | -0.065 |
lor | -0.063 |
urger | -0.060 |
oll | -0.059 |
ster | -0.057 |
Token | . |
Token ID | 13 |
Feature activation | -2.71e-01 |
Pos logit contributions | |
different | +0.009 |
fast | +0.009 |
slow | +0.008 |
very | +0.007 |
late | +0.007 |
Neg logit contributions | |
lor | -0.036 |
ocket | -0.033 |
urger | -0.032 |
oll | -0.031 |
card | -0.030 |
Token |
|
Token ID | 198 |
Feature activation | -4.10e-01 |
Pos logit contributions | |
different | +0.005 |
fast | +0.005 |
slow | +0.004 |
late | +0.004 |
competitive | +0.004 |
Neg logit contributions | |
lor | -0.014 |
ocket | -0.012 |
oll | -0.012 |
anger | -0.012 |
urger | -0.011 |
Token | His |
Token ID | 2399 |
Feature activation | +3.20e+00 |
Pos logit contributions | |
different | +0.017 |
fast | +0.017 |
slow | +0.015 |
very | +0.015 |
really | +0.014 |
Neg logit contributions | |
lor | -0.035 |
ocket | -0.032 |
urger | -0.030 |
oll | -0.029 |
anger | -0.028 |
Token | friends |
Token ID | 2460 |
Feature activation | -5.90e-01 |
Pos logit contributions | |
lor | +0.156 |
ocket | +0.142 |
urger | +0.133 |
card | +0.125 |
oll | +0.124 |
Neg logit contributions | |
different | -0.085 |
fast | -0.084 |
slow | -0.078 |
very | -0.076 |
really | -0.072 |
Token | laughed |
Token ID | 13818 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
different | +0.012 |
fast | +0.012 |
slow | +0.011 |
very | +0.010 |
late | +0.010 |
Neg logit contributions | |
lor | -0.030 |
ocket | -0.028 |
urger | -0.027 |
card | -0.025 |
oll | -0.025 |
Token | and |
Token ID | 290 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
lor | +0.057 |
ocket | +0.053 |
urger | +0.051 |
oll | +0.049 |
card | +0.048 |
Neg logit contributions | |
different | -0.016 |
fast | -0.016 |
slow | -0.014 |
very | -0.013 |
late | -0.012 |
Token | cl |
Token ID | 537 |
Feature activation | -9.51e-01 |
Pos logit contributions | |
lor | +0.075 |
ocket | +0.069 |
urger | +0.066 |
card | +0.062 |
oll | +0.062 |
Neg logit contributions | |
different | -0.044 |
fast | -0.042 |
late | -0.039 |
slow | -0.039 |
very | -0.037 |
Token | apped |
Token ID | 6320 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
different | +0.041 |
fast | +0.041 |
slow | +0.039 |
very | +0.038 |
really | +0.037 |
Neg logit contributions | |
lor | -0.029 |
ocket | -0.025 |
urger | -0.023 |
oll | -0.021 |
card | -0.020 |
Token | , |
Token ID | 11 |
Feature activation | -8.34e-01 |
Pos logit contributions | |
different | +0.025 |
fast | +0.025 |
slow | +0.022 |
late | +0.020 |
very | +0.020 |
Neg logit contributions | |
lor | -0.083 |
ocket | -0.078 |
urger | -0.077 |
card | -0.071 |
oll | -0.071 |
Token | they |
Token ID | 484 |
Feature activation | -8.83e-02 |
Pos logit contributions | |
different | +0.017 |
fast | +0.016 |
slow | +0.015 |
very | +0.014 |
really | +0.014 |
Neg logit contributions | |
lor | -0.041 |
ocket | -0.038 |
urger | -0.037 |
card | -0.035 |
oll | -0.034 |
Token | all |
Token ID | 477 |
Feature activation | -4.47e+00 |
Pos logit contributions | |
different | +0.002 |
fast | +0.002 |
slow | +0.002 |
very | +0.002 |
really | +0.002 |
Neg logit contributions | |
lor | -0.005 |
ocket | -0.004 |
urger | -0.004 |
oll | -0.004 |
card | -0.004 |
Token | thought |
Token ID | 1807 |
Feature activation | +6.86e+00 |
Pos logit contributions | |
fast | +0.115 |
late | +0.114 |
different | +0.113 |
slow | +0.105 |
Stop | +0.102 |
Neg logit contributions | |
lor | -0.165 |
ocket | -0.157 |
urger | -0.150 |
card | -0.148 |
oll | -0.144 |
Token | it |
Token ID | 340 |
Feature activation | +6.44e-01 |
Pos logit contributions | |
lor | +0.352 |
ocket | +0.327 |
urger | +0.312 |
oll | +0.295 |
card | +0.292 |
Neg logit contributions | |
different | -0.147 |
fast | -0.145 |
slow | -0.131 |
very | -0.126 |
really | -0.121 |
Token | boy |
Token ID | 2933 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
lor | +0.026 |
ocket | +0.024 |
urger | +0.023 |
oll | +0.021 |
card | +0.021 |
Neg logit contributions | |
different | -0.009 |
fast | -0.009 |
slow | -0.008 |
very | -0.008 |
really | -0.007 |
Token | was |
Token ID | 373 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
fast | +0.048 |
different | +0.048 |
slow | +0.043 |
late | +0.042 |
very | +0.040 |
Neg logit contributions | |
lor | -0.107 |
ocket | -0.099 |
urger | -0.097 |
oll | -0.091 |
ster | -0.090 |
Token | so |
Token ID | 523 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
B | +0.002 |
Stop | +0.002 |
different | +0.002 |
Da | +0.002 |
Good | +0.002 |
Neg logit contributions | |
lor | -0.006 |
ocket | -0.006 |
urger | -0.006 |
oll | -0.006 |
card | -0.006 |
Token | happy |
Token ID | 3772 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
opposite | +0.021 |
Da | +0.021 |
B | +0.020 |
Stop | +0.020 |
different | +0.019 |
Neg logit contributions | |
oll | -0.054 |
lor | -0.053 |
ocket | -0.053 |
urger | -0.051 |
icking | -0.050 |
Token | that |
Token ID | 326 |
Feature activation | +1.86e+00 |
Pos logit contributions | |
different | +0.025 |
fast | +0.024 |
slow | +0.022 |
very | +0.019 |
really | +0.018 |
Neg logit contributions | |
lor | -0.082 |
ocket | -0.076 |
urger | -0.074 |
oll | -0.070 |
ster | -0.069 |
Token | he |
Token ID | 339 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
lor | +0.088 |
ocket | +0.081 |
urger | +0.077 |
oll | +0.072 |
card | +0.072 |
Neg logit contributions | |
different | -0.048 |
fast | -0.047 |
slow | -0.044 |
very | -0.042 |
really | -0.041 |
Token | forgot |
Token ID | 16453 |
Feature activation | -9.96e-01 |
Pos logit contributions | |
lor | +0.062 |
ocket | +0.057 |
urger | +0.055 |
oll | +0.051 |
ster | +0.050 |
Neg logit contributions | |
different | -0.033 |
fast | -0.031 |
slow | -0.029 |
late | -0.028 |
very | -0.027 |
Token | to |
Token ID | 284 |
Feature activation | +9.46e-01 |
Pos logit contributions | |
fast | +0.020 |
different | +0.020 |
slow | +0.019 |
late | +0.017 |
B | +0.017 |
Neg logit contributions | |
lor | -0.049 |
ocket | -0.045 |
urger | -0.044 |
oll | -0.041 |
card | -0.040 |
Token | be |
Token ID | 307 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
lor | +0.047 |
ocket | +0.044 |
urger | +0.042 |
oll | +0.039 |
card | +0.039 |
Neg logit contributions | |
different | -0.022 |
fast | -0.022 |
slow | -0.020 |
very | -0.019 |
really | -0.019 |
Token | careful |
Token ID | 8161 |
Feature activation | +2.74e-01 |
Pos logit contributions | |
different | +0.023 |
Stop | +0.023 |
fast | +0.022 |
opposite | +0.021 |
B | +0.021 |
Neg logit contributions | |
lor | -0.051 |
ocket | -0.049 |
urger | -0.049 |
card | -0.048 |
oll | -0.046 |
Token | . |
Token ID | 13 |
Feature activation | -2.70e-01 |
Pos logit contributions | |
lor | +0.016 |
ocket | +0.015 |
urger | +0.014 |
oll | +0.014 |
card | +0.014 |
Neg logit contributions | |
different | -0.004 |
fast | -0.004 |
slow | -0.003 |
very | -0.003 |
really | -0.003 |
Token | Tom |
Token ID | 13787 |
Feature activation | -4.76e-01 |
Pos logit contributions | |
lor | +0.218 |
ocket | +0.196 |
urger | +0.178 |
oll | +0.177 |
anger | +0.174 |
Neg logit contributions | |
different | -0.111 |
fast | -0.108 |
slow | -0.101 |
very | -0.095 |
really | -0.092 |
Token | my |
Token ID | 1820 |
Feature activation | -7.73e-01 |
Pos logit contributions | |
different | +0.010 |
fast | +0.010 |
slow | +0.008 |
very | +0.008 |
late | +0.008 |
Neg logit contributions | |
lor | -0.024 |
ocket | -0.023 |
urger | -0.022 |
oll | -0.021 |
ster | -0.020 |
Token | was |
Token ID | 373 |
Feature activation | -7.53e-01 |
Pos logit contributions | |
different | +0.015 |
fast | +0.015 |
slow | +0.013 |
very | +0.013 |
really | +0.012 |
Neg logit contributions | |
lor | -0.042 |
ocket | -0.039 |
urger | -0.037 |
card | -0.035 |
oll | -0.035 |
Token | a |
Token ID | 257 |
Feature activation | -5.04e-01 |
Pos logit contributions | |
different | +0.014 |
fast | +0.014 |
very | +0.012 |
slow | +0.012 |
really | +0.011 |
Neg logit contributions | |
lor | -0.043 |
ocket | -0.039 |
urger | -0.038 |
card | -0.035 |
anger | -0.035 |
Token | big |
Token ID | 1263 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
different | +0.009 |
fast | +0.009 |
late | +0.009 |
Stop | +0.009 |
B | +0.008 |
Neg logit contributions | |
lor | -0.022 |
ocket | -0.022 |
urger | -0.021 |
ster | -0.021 |
oll | -0.020 |
Token | boy |
Token ID | 2933 |
Feature activation | -9.94e-01 |
Pos logit contributions | |
different | +0.041 |
fast | +0.040 |
slow | +0.036 |
late | +0.036 |
B | +0.035 |
Neg logit contributions | |
lor | -0.077 |
ocket | -0.075 |
urger | -0.072 |
ster | -0.068 |
oll | -0.068 |
Token | who |
Token ID | 508 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
different | +0.017 |
fast | +0.016 |
slow | +0.014 |
late | +0.013 |
very | +0.012 |
Neg logit contributions | |
lor | -0.053 |
ocket | -0.050 |
urger | -0.048 |
oll | -0.046 |
card | -0.046 |
Token | liked |
Token ID | 8288 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
lor | +0.089 |
ocket | +0.085 |
urger | +0.080 |
oll | +0.079 |
ster | +0.075 |
Neg logit contributions | |
different | -0.028 |
fast | -0.028 |
slow | -0.024 |
late | -0.023 |
very | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | -2.98e-01 |
Pos logit contributions | |
fast | +0.002 |
different | +0.002 |
slow | +0.002 |
late | +0.002 |
really | +0.002 |
Neg logit contributions | |
lor | -0.011 |
ocket | -0.010 |
urger | -0.010 |
card | -0.009 |
oll | -0.009 |
Token | play |
Token ID | 711 |
Feature activation | -5.25e-04 |
Pos logit contributions | |
different | +0.007 |
fast | +0.007 |
slow | +0.006 |
very | +0.006 |
really | +0.006 |
Neg logit contributions | |
lor | -0.015 |
ocket | -0.014 |
urger | -0.013 |
oll | -0.012 |
card | -0.012 |
Token | with |
Token ID | 351 |
Feature activation | -5.44e-01 |
Pos logit contributions | |
fast | +0.000 |
different | +0.000 |
slow | +0.000 |
late | +0.000 |
very | +0.000 |
Neg logit contributions | |
lor | -0.000 |
ocket | -0.000 |
urger | -0.000 |
oll | -0.000 |
anger | -0.000 |
Token | your |
Token ID | 534 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
different | +0.006 |
fast | +0.006 |
slow | +0.005 |
very | +0.005 |
really | +0.005 |
Neg logit contributions | |
lor | -0.014 |
ocket | -0.013 |
urger | -0.012 |
oll | -0.012 |
card | -0.012 |
Token | daddy |
Token ID | 35822 |
Feature activation | +2.30e-01 |
Pos logit contributions | |
lor | +0.066 |
ocket | +0.059 |
urger | +0.055 |
oll | +0.053 |
card | +0.053 |
Neg logit contributions | |
different | -0.038 |
fast | -0.038 |
slow | -0.035 |
very | -0.034 |
really | -0.032 |
Token | 's |
Token ID | 338 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
lor | +0.013 |
ocket | +0.012 |
urger | +0.012 |
oll | +0.011 |
anger | +0.011 |
Neg logit contributions | |
different | -0.003 |
fast | -0.003 |
slow | -0.003 |
very | -0.003 |
really | -0.003 |
Token | birthday |
Token ID | 10955 |
Feature activation | +4.85e+00 |
Pos logit contributions | |
lor | +0.100 |
ocket | +0.088 |
urger | +0.082 |
oll | +0.081 |
anger | +0.079 |
Neg logit contributions | |
different | -0.060 |
fast | -0.059 |
very | -0.055 |
slow | -0.055 |
really | -0.051 |
Token | !" |
Token ID | 2474 |
Feature activation | -3.14e+00 |
Pos logit contributions | |
lor | +0.276 |
ocket | +0.260 |
urger | +0.247 |
oll | +0.236 |
card | +0.234 |
Neg logit contributions | |
different | -0.079 |
fast | -0.077 |
very | -0.067 |
slow | -0.067 |
really | -0.062 |
Token |
|
Token ID | 198 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
different | +0.038 |
fast | +0.036 |
slow | +0.030 |
very | +0.027 |
really | +0.025 |
Neg logit contributions | |
lor | -0.192 |
ocket | -0.180 |
urger | -0.173 |
oll | -0.165 |
card | -0.164 |
Token |
|
Token ID | 198 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
different | +0.016 |
fast | +0.015 |
slow | +0.012 |
very | +0.011 |
really | +0.010 |
Neg logit contributions | |
lor | -0.083 |
ocket | -0.077 |
urger | -0.074 |
oll | -0.071 |
card | -0.070 |
Token | M |
Token ID | 44 |
Feature activation | -3.41e-01 |
Pos logit contributions | |
different | +0.022 |
fast | +0.022 |
slow | +0.019 |
very | +0.018 |
really | +0.017 |
Neg logit contributions | |
lor | -0.068 |
ocket | -0.062 |
urger | -0.059 |
oll | -0.056 |
card | -0.056 |
Token | andy |
Token ID | 10757 |
Feature activation | -9.69e-01 |
Pos logit contributions | |
different | +0.013 |
fast | +0.012 |
really | +0.012 |
very | +0.012 |
slow | +0.012 |
Neg logit contributions | |
lor | -0.012 |
ocket | -0.011 |
urger | -0.010 |
card | -0.010 |
suitcase | -0.009 |
Token | felt |
Token ID | 2936 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
different | +0.020 |
fast | +0.020 |
slow | +0.018 |
very | +0.017 |
really | +0.017 |
Neg logit contributions | |
lor | -0.051 |
ocket | -0.048 |
urger | -0.046 |
oll | -0.043 |
card | -0.043 |
Token | much |
Token ID | 881 |
Feature activation | -5.31e+00 |
Pos logit contributions | |
different | +0.030 |
fast | +0.030 |
slow | +0.026 |
Stop | +0.024 |
B | +0.024 |
Neg logit contributions | |
lor | -0.096 |
ocket | -0.092 |
oll | -0.083 |
urger | -0.082 |
card | -0.080 |
Token | Daisy |
Token ID | 40355 |
Feature activation | +4.70e-01 |
Pos logit contributions | |
lor | +0.213 |
ocket | +0.204 |
urger | +0.188 |
oll | +0.184 |
card | +0.180 |
Neg logit contributions | |
different | -0.064 |
fast | -0.064 |
slow | -0.057 |
very | -0.052 |
late | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | -6.01e-01 |
Pos logit contributions | |
lor | +0.024 |
ocket | +0.023 |
urger | +0.022 |
ster | +0.021 |
oll | +0.021 |
Neg logit contributions | |
fast | -0.008 |
different | -0.008 |
slow | -0.007 |
late | -0.007 |
Stop | -0.007 |
Token | Daisy |
Token ID | 40355 |
Feature activation | -8.16e-01 |
Pos logit contributions | |
different | +0.014 |
fast | +0.014 |
slow | +0.012 |
very | +0.012 |
late | +0.011 |
Neg logit contributions | |
lor | -0.028 |
ocket | -0.026 |
oll | -0.023 |
urger | -0.023 |
ster | -0.022 |
Token | lived |
Token ID | 5615 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
fast | +0.018 |
different | +0.018 |
late | +0.017 |
slow | +0.016 |
opposite | +0.016 |
Neg logit contributions | |
lor | -0.037 |
ocket | -0.035 |
urger | -0.032 |
oll | -0.031 |
ster | -0.031 |
Token | in |
Token ID | 287 |
Feature activation | -4.74e+00 |
Pos logit contributions | |
lor | +0.055 |
ocket | +0.053 |
ster | +0.049 |
oll | +0.048 |
card | +0.047 |
Neg logit contributions | |
different | -0.024 |
fast | -0.021 |
Stop | -0.021 |
slow | -0.021 |
opposite | -0.020 |
Token | a |
Token ID | 257 |
Feature activation | -4.00e-01 |
Pos logit contributions | |
different | +0.061 |
fast | +0.060 |
slow | +0.051 |
very | +0.047 |
really | +0.044 |
Neg logit contributions | |
lor | -0.279 |
ocket | -0.263 |
urger | -0.249 |
oll | -0.239 |
card | -0.238 |
Token | very |
Token ID | 845 |
Feature activation | -3.49e+00 |
Pos logit contributions | |
Stop | +0.006 |
fast | +0.006 |
wildly | +0.006 |
late | +0.006 |
different | +0.006 |
Neg logit contributions | |
lor | -0.019 |
ocket | -0.018 |
ster | -0.018 |
suitcase | -0.017 |
urger | -0.016 |
Token | colorful |
Token ID | 20239 |
Feature activation | -5.58e-01 |
Pos logit contributions | |
different | +0.050 |
fast | +0.047 |
Stop | +0.047 |
B | +0.044 |
each | +0.044 |
Neg logit contributions | |
ocket | -0.175 |
lor | -0.175 |
urger | -0.162 |
ster | -0.159 |
oll | -0.158 |
Token | room |
Token ID | 2119 |
Feature activation | +6.91e-01 |
Pos logit contributions | |
different | +0.017 |
fast | +0.017 |
slow | +0.016 |
very | +0.016 |
really | +0.015 |
Neg logit contributions | |
lor | -0.023 |
ocket | -0.021 |
urger | -0.020 |
oll | -0.018 |
card | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -7.82e-01 |
Pos logit contributions | |
lor | +0.042 |
ocket | +0.040 |
urger | +0.039 |
oll | +0.036 |
anger | +0.036 |
Neg logit contributions | |
different | -0.009 |
fast | -0.009 |
very | -0.007 |
slow | -0.007 |
really | -0.006 |
Token | One |
Token ID | 1881 |
Feature activation | +9.44e-01 |
Pos logit contributions | |
fast | +0.016 |
different | +0.016 |
slow | +0.014 |
very | +0.014 |
late | +0.013 |
Neg logit contributions | |
lor | -0.038 |
ocket | -0.036 |
oll | -0.032 |
urger | -0.031 |
anger | -0.031 |
Token | into |
Token ID | 656 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
different | +0.035 |
fast | +0.033 |
slow | +0.030 |
very | +0.026 |
really | +0.026 |
Neg logit contributions | |
lor | -0.102 |
ocket | -0.099 |
urger | -0.096 |
oll | -0.091 |
card | -0.089 |
Token | trouble |
Token ID | 5876 |
Feature activation | -9.68e-01 |
Pos logit contributions | |
fast | +0.044 |
different | +0.043 |
slow | +0.041 |
B | +0.039 |
really | +0.038 |
Neg logit contributions | |
lor | -0.094 |
ocket | -0.090 |
urger | -0.085 |
oll | -0.079 |
suitcase | -0.078 |
Token | . |
Token ID | 13 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
different | +0.014 |
fast | +0.014 |
slow | +0.012 |
very | +0.010 |
really | +0.010 |
Neg logit contributions | |
lor | -0.055 |
ocket | -0.052 |
urger | -0.050 |
oll | -0.047 |
card | -0.047 |
Token | He |
Token ID | 679 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
fast | +0.006 |
different | +0.005 |
slow | +0.005 |
very | +0.005 |
busy | +0.004 |
Neg logit contributions | |
lor | -0.013 |
ocket | -0.011 |
urger | -0.011 |
oll | -0.011 |
anger | -0.010 |
Token | went |
Token ID | 1816 |
Feature activation | +6.60e-01 |
Pos logit contributions | |
lor | +0.070 |
ocket | +0.067 |
urger | +0.064 |
oll | +0.062 |
ster | +0.058 |
Neg logit contributions | |
fast | -0.038 |
different | -0.038 |
late | -0.036 |
slow | -0.033 |
busy | -0.030 |
Token | outside |
Token ID | 2354 |
Feature activation | -5.08e+00 |
Pos logit contributions | |
lor | +0.036 |
ocket | +0.034 |
urger | +0.032 |
card | +0.031 |
anger | +0.031 |
Neg logit contributions | |
different | -0.009 |
fast | -0.008 |
Stop | -0.007 |
slow | -0.007 |
B | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | -6.48e-01 |
Pos logit contributions | |
Stop | +0.093 |
B | +0.087 |
different | +0.087 |
fast | +0.085 |
Blue | +0.080 |
Neg logit contributions | |
lor | -0.221 |
ocket | -0.209 |
urger | -0.200 |
ster | -0.197 |
card | -0.195 |
Token | play |
Token ID | 711 |
Feature activation | +3.73e-01 |
Pos logit contributions | |
fast | +0.016 |
different | +0.015 |
late | +0.015 |
B | +0.015 |
very | +0.014 |
Neg logit contributions | |
lor | -0.026 |
ocket | -0.025 |
urger | -0.024 |
card | -0.022 |
suitcase | -0.021 |
Token | and |
Token ID | 290 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
lor | +0.018 |
ocket | +0.017 |
urger | +0.015 |
oll | +0.015 |
anger | +0.015 |
Neg logit contributions | |
different | -0.007 |
fast | -0.007 |
slow | -0.006 |
late | -0.006 |
panic | -0.006 |
Token | saw |
Token ID | 2497 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
lor | +0.042 |
ocket | +0.042 |
urger | +0.040 |
card | +0.037 |
ster | +0.036 |
Neg logit contributions | |
different | -0.026 |
fast | -0.026 |
late | -0.025 |
B | -0.025 |
slow | -0.023 |
Token | a |
Token ID | 257 |
Feature activation | +8.23e-01 |
Pos logit contributions | |
lor | +0.089 |
ocket | +0.083 |
urger | +0.080 |
oll | +0.078 |
ster | +0.076 |
Neg logit contributions | |
fast | -0.032 |
different | -0.030 |
slow | -0.030 |
late | -0.027 |
really | -0.026 |
Token | hug |
Token ID | 16225 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
fast | +0.020 |
different | +0.020 |
slow | +0.019 |
very | +0.018 |
really | +0.017 |
Neg logit contributions | |
lor | -0.042 |
ocket | -0.038 |
urger | -0.036 |
oll | -0.034 |
card | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
lor | +0.083 |
ocket | +0.079 |
urger | +0.077 |
oll | +0.072 |
Majesty | +0.071 |
Neg logit contributions | |
fast | -0.010 |
different | -0.009 |
very | -0.009 |
slow | -0.006 |
really | -0.006 |
Token | Then |
Token ID | 3244 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
different | +0.003 |
fast | +0.003 |
slow | +0.002 |
very | +0.002 |
really | +0.002 |
Neg logit contributions | |
lor | -0.008 |
ocket | -0.008 |
urger | -0.007 |
oll | -0.007 |
card | -0.007 |
Token | she |
Token ID | 673 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
different | +0.002 |
fast | +0.002 |
very | +0.001 |
slow | +0.001 |
the | +0.001 |
Neg logit contributions | |
lor | -0.007 |
ocket | -0.007 |
urger | -0.007 |
oll | -0.006 |
card | -0.006 |
Token | got |
Token ID | 1392 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
lor | +0.010 |
ocket | +0.009 |
urger | +0.009 |
anger | +0.009 |
card | +0.008 |
Neg logit contributions | |
fast | -0.004 |
different | -0.004 |
really | -0.004 |
slow | -0.004 |
very | -0.004 |
Token | back |
Token ID | 736 |
Feature activation | -5.40e+00 |
Pos logit contributions | |
different | +0.041 |
fast | +0.041 |
very | +0.036 |
slow | +0.036 |
really | +0.034 |
Neg logit contributions | |
lor | -0.123 |
ocket | -0.113 |
urger | -0.109 |
anger | -0.102 |
oll | -0.102 |
Token | on |
Token ID | 319 |
Feature activation | -2.95e+00 |
Pos logit contributions | |
different | +0.094 |
fast | +0.091 |
slow | +0.080 |
very | +0.076 |
really | +0.072 |
Neg logit contributions | |
lor | -0.300 |
ocket | -0.280 |
urger | -0.268 |
oll | -0.255 |
card | -0.254 |
Token | her |
Token ID | 607 |
Feature activation | +9.39e-01 |
Pos logit contributions | |
different | +0.030 |
fast | +0.028 |
slow | +0.023 |
very | +0.020 |
really | +0.019 |
Neg logit contributions | |
lor | -0.182 |
ocket | -0.172 |
urger | -0.164 |
card | -0.157 |
oll | -0.156 |
Token | seat |
Token ID | 5852 |
Feature activation | +3.09e-01 |
Pos logit contributions | |
lor | +0.040 |
ocket | +0.036 |
urger | +0.034 |
oll | +0.032 |
card | +0.031 |
Neg logit contributions | |
different | -0.030 |
fast | -0.030 |
slow | -0.028 |
very | -0.027 |
really | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
lor | +0.019 |
ocket | +0.018 |
urger | +0.018 |
oll | +0.016 |
ster | +0.016 |
Neg logit contributions | |
fast | -0.004 |
different | -0.003 |
very | -0.003 |
slow | -0.003 |
really | -0.003 |
Token | continued |
Token ID | 3767 |
Feature activation | -2.85e+00 |
Pos logit contributions | |
lor | +0.063 |
ocket | +0.059 |
urger | +0.056 |
anger | +0.053 |
oll | +0.052 |
Neg logit contributions | |
fast | -0.025 |
different | -0.023 |
slow | -0.022 |
the | -0.022 |
very | -0.021 |
Token | ?" |
Token ID | 1701 |
Feature activation | -3.41e+00 |
Pos logit contributions | |
lor | +0.142 |
ocket | +0.130 |
urger | +0.124 |
oll | +0.116 |
card | +0.115 |
Neg logit contributions | |
different | -0.081 |
fast | -0.080 |
slow | -0.074 |
very | -0.072 |
really | -0.069 |
Token | Her |
Token ID | 2332 |
Feature activation | +2.72e+00 |
Pos logit contributions | |
fast | +0.097 |
different | +0.094 |
slow | +0.088 |
very | +0.082 |
hard | +0.080 |
Neg logit contributions | |
lor | -0.148 |
ocket | -0.134 |
urger | -0.129 |
card | -0.118 |
ster | -0.118 |
Token | dad |
Token ID | 9955 |
Feature activation | -4.81e-01 |
Pos logit contributions | |
lor | +0.117 |
ocket | +0.102 |
urger | +0.094 |
card | +0.089 |
anger | +0.088 |
Neg logit contributions | |
different | -0.093 |
fast | -0.091 |
very | -0.087 |
slow | -0.086 |
really | -0.083 |
Token | replied |
Token ID | 8712 |
Feature activation | +6.17e+00 |
Pos logit contributions | |
different | +0.015 |
fast | +0.015 |
very | +0.014 |
slow | +0.014 |
really | +0.014 |
Neg logit contributions | |
lor | -0.021 |
ocket | -0.019 |
urger | -0.018 |
anger | -0.017 |
oll | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | -4.92e-01 |
Pos logit contributions | |
lor | +0.300 |
ocket | +0.285 |
urger | +0.275 |
ster | +0.261 |
oll | +0.261 |
Neg logit contributions | |
different | -0.129 |
fast | -0.124 |
slow | -0.114 |
late | -0.109 |
Stop | -0.102 |
Token | " |
Token ID | 366 |
Feature activation | -3.81e+00 |
Pos logit contributions | |
different | +0.011 |
fast | +0.011 |
slow | +0.010 |
very | +0.009 |
late | +0.009 |
Neg logit contributions | |
lor | -0.024 |
ocket | -0.022 |
urger | -0.021 |
card | -0.020 |
ster | -0.020 |
Token | I |
Token ID | 40 |
Feature activation | -2.52e+00 |
Pos logit contributions | |
different | +0.067 |
fast | +0.067 |
slow | +0.061 |
very | +0.056 |
busy | +0.051 |
Neg logit contributions | |
lor | -0.208 |
ocket | -0.189 |
urger | -0.177 |
card | -0.176 |
oll | -0.173 |
Token | 'm |
Token ID | 1101 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
different | +0.047 |
fast | +0.045 |
slow | +0.040 |
very | +0.039 |
really | +0.038 |
Neg logit contributions | |
lor | -0.140 |
ocket | -0.130 |
urger | -0.125 |
oll | -0.118 |
card | -0.117 |
Token | printing |
Token ID | 13570 |
Feature activation | +2.83e+00 |
Pos logit contributions | |
lor | +0.006 |
ocket | +0.005 |
urger | +0.005 |
oll | +0.005 |
card | +0.004 |
Neg logit contributions | |
different | -0.004 |
fast | -0.003 |
slow | -0.003 |
very | -0.003 |
really | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
lor | +0.151 |
ocket | +0.141 |
urger | +0.135 |
card | +0.128 |
oll | +0.127 |
Neg logit contributions | |
fast | -0.051 |
different | -0.051 |
slow | -0.046 |
very | -0.043 |
late | -0.041 |
Token | recipe |
Token ID | 8364 |
Feature activation | +5.28e+00 |
Pos logit contributions | |
lor | +0.048 |
ocket | +0.044 |
urger | +0.042 |
card | +0.040 |
ster | +0.038 |
Neg logit contributions | |
fast | -0.044 |
different | -0.043 |
slow | -0.042 |
late | -0.040 |
very | -0.039 |
Token |
|
Token ID | 198 |
Feature activation | -6.09e-01 |
Pos logit contributions | |
different | +0.009 |
fast | +0.008 |
slow | +0.007 |
very | +0.007 |
hurt | +0.007 |
Neg logit contributions | |
lor | -0.028 |
ocket | -0.024 |
anger | -0.023 |
card | -0.023 |
uit | -0.022 |
Token | Suddenly |
Token ID | 38582 |
Feature activation | +3.56e-01 |
Pos logit contributions | |
different | +0.014 |
fast | +0.014 |
slow | +0.013 |
very | +0.013 |
hard | +0.012 |
Neg logit contributions | |
lor | -0.028 |
ocket | -0.024 |
card | -0.023 |
bridge | -0.022 |
urger | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
lor | +0.019 |
ocket | +0.018 |
urger | +0.017 |
card | +0.017 |
oll | +0.016 |
Neg logit contributions | |
different | -0.006 |
fast | -0.006 |
slow | -0.005 |
very | -0.005 |
really | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | +8.67e-01 |
Pos logit contributions | |
different | +0.022 |
fast | +0.021 |
slow | +0.019 |
very | +0.018 |
really | +0.017 |
Neg logit contributions | |
lor | -0.060 |
ocket | -0.056 |
urger | -0.053 |
card | -0.050 |
oll | -0.050 |
Token | strange |
Token ID | 6283 |
Feature activation | -2.82e-01 |
Pos logit contributions | |
lor | +0.042 |
ocket | +0.040 |
urger | +0.037 |
ster | +0.036 |
card | +0.036 |
Neg logit contributions | |
different | -0.018 |
fast | -0.017 |
slow | -0.015 |
late | -0.015 |
each | -0.014 |
Token | dizz |
Token ID | 29527 |
Feature activation | -3.19e+00 |
Pos logit contributions | |
different | +0.008 |
fast | +0.008 |
slow | +0.007 |
very | +0.007 |
really | +0.007 |
Neg logit contributions | |
lor | -0.013 |
ocket | -0.012 |
urger | -0.011 |
card | -0.010 |
oll | -0.010 |
Token | y |
Token ID | 88 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
different | +0.050 |
fast | +0.048 |
slow | +0.042 |
very | +0.040 |
really | +0.036 |
Neg logit contributions | |
lor | -0.182 |
ocket | -0.171 |
urger | -0.162 |
oll | -0.157 |
anger | -0.155 |
Token | feeling |
Token ID | 4203 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
different | +0.047 |
fast | +0.047 |
slow | +0.043 |
very | +0.042 |
really | +0.040 |
Neg logit contributions | |
lor | -0.085 |
ocket | -0.079 |
urger | -0.075 |
oll | -0.070 |
card | -0.069 |
Token | came |
Token ID | 1625 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
different | +0.036 |
fast | +0.035 |
slow | +0.032 |
very | +0.030 |
late | +0.029 |
Neg logit contributions | |
lor | -0.088 |
ocket | -0.083 |
urger | -0.076 |
oll | -0.075 |
card | -0.074 |
Token | over |
Token ID | 625 |
Feature activation | -6.02e+00 |
Pos logit contributions | |
different | +0.003 |
fast | +0.003 |
slow | +0.003 |
fun | +0.003 |
Stop | +0.003 |
Neg logit contributions | |
lor | -0.005 |
ocket | -0.005 |
urger | -0.005 |
icking | -0.005 |
card | -0.004 |
Token | her |
Token ID | 607 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
different | +0.088 |
fast | +0.087 |
slow | +0.075 |
very | +0.070 |
really | +0.065 |
Neg logit contributions | |
lor | -0.346 |
ocket | -0.326 |
urger | -0.310 |
oll | -0.295 |
card | -0.295 |
Token | She |
Token ID | 1375 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
lor | +0.018 |
ocket | +0.017 |
oll | +0.016 |
urger | +0.015 |
anger | +0.015 |
Neg logit contributions | |
different | -0.006 |
fast | -0.005 |
slow | -0.005 |
very | -0.005 |
really | -0.004 |
Token | learned |
Token ID | 4499 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
lor | +0.079 |
ocket | +0.073 |
urger | +0.070 |
oll | +0.066 |
card | +0.065 |
Neg logit contributions | |
different | -0.032 |
fast | -0.032 |
slow | -0.028 |
very | -0.027 |
late | -0.026 |
Token | how |
Token ID | 703 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
lor | +0.100 |
ocket | +0.093 |
oll | +0.084 |
card | +0.084 |
urger | +0.083 |
Neg logit contributions | |
fast | -0.046 |
different | -0.045 |
slow | -0.044 |
late | -0.040 |
panic | -0.040 |
Token | to |
Token ID | 284 |
Feature activation | +8.50e-01 |
Pos logit contributions | |
opposite | +0.035 |
panic | +0.034 |
Uh | +0.034 |
ny | +0.033 |
Stop | +0.033 |
Neg logit contributions | |
ocket | -0.068 |
lor | -0.068 |
anger | -0.063 |
oll | -0.063 |
ster | -0.063 |
Token | be |
Token ID | 307 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
lor | +0.033 |
urger | +0.029 |
ocket | +0.029 |
uit | +0.027 |
oll | +0.027 |
Neg logit contributions | |
late | -0.024 |
different | -0.023 |
fast | -0.023 |
Da | -0.022 |
very | -0.021 |
Token | less |
Token ID | 1342 |
Feature activation | -4.78e+00 |
Pos logit contributions | |
heavily | +0.042 |
Stop | +0.041 |
B | +0.040 |
wildly | +0.039 |
| | +0.038 |
Neg logit contributions | |
card | -0.063 |
ocket | -0.061 |
lor | -0.060 |
oll | -0.059 |
urger | -0.059 |
Token | shy |
Token ID | 15800 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
B | +0.101 |
Amy | +0.097 |
Da | +0.096 |
Lu | +0.094 |
| | +0.094 |
Neg logit contributions | |
ocket | -0.181 |
oll | -0.177 |
ster | -0.172 |
sters | -0.170 |
lor | -0.170 |
Token | . |
Token ID | 13 |
Feature activation | +5.03e-01 |
Pos logit contributions | |
different | +0.036 |
fast | +0.035 |
slow | +0.031 |
late | +0.029 |
very | +0.028 |
Neg logit contributions | |
lor | -0.101 |
ocket | -0.097 |
oll | -0.090 |
urger | -0.090 |
ster | -0.087 |
Token | She |
Token ID | 1375 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
lor | +0.028 |
ocket | +0.026 |
oll | +0.024 |
urger | +0.024 |
card | +0.023 |
Neg logit contributions | |
different | -0.008 |
fast | -0.008 |
slow | -0.007 |
very | -0.006 |
really | -0.006 |
Token | had |
Token ID | 550 |
Feature activation | -4.54e-01 |
Pos logit contributions | |
lor | +0.083 |
ocket | +0.077 |
urger | +0.074 |
oll | +0.070 |
card | +0.069 |
Neg logit contributions | |
different | -0.033 |
fast | -0.032 |
slow | -0.029 |
very | -0.028 |
really | -0.026 |
Token | the |
Token ID | 262 |
Feature activation | +8.12e-01 |
Pos logit contributions | |
fast | +0.008 |
heavily | +0.008 |
late | +0.007 |
B | +0.007 |
wildly | +0.007 |
Neg logit contributions | |
lor | -0.020 |
ocket | -0.019 |
oll | -0.018 |
urger | -0.017 |
inch | -0.017 |