Token | the |
Token ID | 262 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
moment | +0.038 |
ending | +0.037 |
Sleep | +0.037 |
polar | +0.037 |
mistake | +0.037 |
Neg logit contributions | |
ay | -0.036 |
issors | -0.033 |
ong | -0.033 |
rac | -0.033 |
ars | -0.031 |
Token | forest |
Token ID | 8222 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
ay | +0.006 |
around | +0.006 |
ong | +0.006 |
high | +0.006 |
ars | +0.006 |
Neg logit contributions | |
Sweet | -0.005 |
Kids | -0.005 |
Sleep | -0.005 |
forgiveness | -0.004 |
Are | -0.004 |
Token | looking |
Token ID | 2045 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
Sleep | +0.049 |
polar | +0.049 |
ending | +0.048 |
Sweet | +0.048 |
mistake | +0.047 |
Neg logit contributions | |
ay | -0.035 |
ong | -0.032 |
rac | -0.031 |
issors | -0.030 |
around | -0.029 |
Token | for |
Token ID | 329 |
Feature activation | +2.75e+00 |
Pos logit contributions | |
ending | +0.012 |
moment | +0.012 |
polar | +0.011 |
Sleep | +0.011 |
mistake | +0.011 |
Neg logit contributions | |
ay | -0.009 |
rac | -0.009 |
issors | -0.009 |
ong | -0.009 |
ors | -0.008 |
Token | all |
Token ID | 477 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
around | +0.075 |
at | +0.070 |
ay | +0.069 |
ong | +0.066 |
high | +0.063 |
Neg logit contributions | |
department | -0.108 |
Kids | -0.107 |
fly | -0.106 |
tr | -0.105 |
divorce | -0.105 |
Token | sorts |
Token ID | 10524 |
Feature activation | +7.54e+00 |
Pos logit contributions | |
ay | +0.032 |
around | +0.031 |
ong | +0.029 |
rac | +0.027 |
bs | +0.026 |
Neg logit contributions | |
Sweet | -0.055 |
Sleep | -0.055 |
polar | -0.055 |
ending | -0.054 |
mistake | -0.054 |
Token | of |
Token ID | 286 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
rac | +0.232 |
ay | +0.227 |
issors | +0.222 |
ong | +0.220 |
ars | +0.199 |
Neg logit contributions | |
Sleep | -0.266 |
Sweet | -0.262 |
forgiveness | -0.258 |
ending | -0.254 |
polar | -0.254 |
Token | exciting |
Token ID | 7895 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
Sleep | +0.064 |
ending | +0.063 |
Sweet | +0.062 |
Kids | +0.061 |
department | +0.061 |
Neg logit contributions | |
ay | -0.056 |
around | -0.055 |
ong | -0.053 |
bs | -0.049 |
at | -0.048 |
Token | things |
Token ID | 1243 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
ay | +0.051 |
ong | +0.048 |
around | +0.048 |
rac | +0.045 |
bs | +0.044 |
Neg logit contributions | |
Sleep | -0.051 |
Sweet | -0.050 |
polar | -0.050 |
ending | -0.050 |
mistake | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | -2.71e-01 |
Pos logit contributions | |
ay | +0.031 |
ong | +0.028 |
rac | +0.028 |
issors | +0.027 |
ars | +0.026 |
Neg logit contributions | |
Sleep | -0.042 |
ending | -0.041 |
polar | -0.041 |
Sweet | -0.041 |
moment | -0.041 |
Token | She |
Token ID | 1375 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
mistake | +0.012 |
polar | +0.012 |
ending | +0.011 |
department | +0.011 |
forgiveness | +0.011 |
Neg logit contributions | |
ay | -0.007 |
around | -0.006 |
issors | -0.006 |
bs | -0.006 |
rac | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | +4.02e-01 |
Pos logit contributions | |
ay | +0.062 |
rac | +0.057 |
ong | +0.057 |
issors | +0.055 |
bs | +0.052 |
Neg logit contributions | |
ending | -0.078 |
moment | -0.077 |
polar | -0.076 |
Sleep | -0.076 |
mistake | -0.075 |
Token | thrilled |
Token ID | 20536 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
ay | +0.014 |
ong | +0.013 |
rac | +0.013 |
around | +0.012 |
issors | +0.012 |
Neg logit contributions | |
Sleep | -0.013 |
polar | -0.013 |
Sweet | -0.013 |
ending | -0.013 |
mistake | -0.012 |
Token | to |
Token ID | 284 |
Feature activation | -9.77e-02 |
Pos logit contributions | |
issors | +0.046 |
ay | +0.046 |
bs | +0.044 |
ong | +0.043 |
ars | +0.043 |
Neg logit contributions | |
ending | -0.051 |
moment | -0.051 |
mistake | -0.049 |
polar | -0.048 |
Sleep | -0.048 |
Token | discover |
Token ID | 7073 |
Feature activation | +7.94e-02 |
Pos logit contributions | |
Sleep | +0.004 |
Sweet | +0.004 |
polar | +0.003 |
department | +0.003 |
Kids | +0.003 |
Neg logit contributions | |
ay | -0.003 |
ong | -0.003 |
around | -0.003 |
rac | -0.003 |
ars | -0.003 |
Token | all |
Token ID | 477 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
ay | +0.003 |
ong | +0.002 |
around | +0.002 |
rac | +0.002 |
bs | +0.002 |
Neg logit contributions | |
Sleep | -0.003 |
Sweet | -0.003 |
ending | -0.003 |
polar | -0.003 |
Kids | -0.003 |
Token | sorts |
Token ID | 10524 |
Feature activation | +7.50e+00 |
Pos logit contributions | |
ay | +0.038 |
ong | +0.037 |
rac | +0.036 |
issors | +0.035 |
ars | +0.033 |
Neg logit contributions | |
polar | -0.044 |
Sleep | -0.044 |
moment | -0.044 |
ending | -0.043 |
mistake | -0.043 |
Token | of |
Token ID | 286 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
rac | +0.263 |
issors | +0.259 |
ay | +0.246 |
ong | +0.243 |
itching | +0.231 |
Neg logit contributions | |
forgiveness | -0.228 |
polar | -0.225 |
Sleep | -0.225 |
Sweet | -0.217 |
pand | -0.217 |
Token | new |
Token ID | 649 |
Feature activation | -6.21e-01 |
Pos logit contributions | |
Sleep | +0.064 |
Sweet | +0.063 |
polar | +0.062 |
ending | +0.062 |
mistake | +0.060 |
Neg logit contributions | |
ay | -0.061 |
ong | -0.058 |
around | -0.056 |
rac | -0.056 |
issors | -0.053 |
Token | things |
Token ID | 1243 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
polar | +0.019 |
Sleep | +0.019 |
Sweet | +0.019 |
ending | +0.019 |
mistake | +0.019 |
Neg logit contributions | |
ay | -0.022 |
ong | -0.021 |
rac | -0.021 |
issors | -0.020 |
around | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
ay | +0.031 |
rac | +0.030 |
ong | +0.029 |
issors | +0.028 |
ars | +0.026 |
Neg logit contributions | |
Sleep | -0.038 |
polar | -0.037 |
moment | -0.037 |
ending | -0.037 |
Sweet | -0.036 |
Token |
|
Token ID | 198 |
Feature activation | +3.89e-01 |
Pos logit contributions | |
mistake | +0.005 |
ending | +0.005 |
Sleep | +0.005 |
Sweet | +0.005 |
polar | +0.005 |
Neg logit contributions | |
ay | -0.004 |
around | -0.003 |
ong | -0.003 |
issors | -0.003 |
rac | -0.003 |
Token | filled |
Token ID | 5901 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
ay | +0.002 |
ong | +0.002 |
rac | +0.002 |
around | +0.002 |
issors | +0.001 |
Neg logit contributions | |
Sleep | -0.002 |
polar | -0.002 |
Sweet | -0.002 |
ending | -0.002 |
mistake | -0.002 |
Token | with |
Token ID | 351 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
ay | +0.045 |
ong | +0.042 |
around | +0.041 |
rac | +0.040 |
issors | +0.037 |
Neg logit contributions | |
Sleep | -0.076 |
ending | -0.075 |
Sweet | -0.075 |
polar | -0.075 |
mistake | -0.073 |
Token | decorations |
Token ID | 39011 |
Feature activation | +8.42e-02 |
Pos logit contributions | |
ay | +0.038 |
around | +0.036 |
ong | +0.035 |
bs | +0.032 |
rac | +0.032 |
Neg logit contributions | |
Sleep | -0.039 |
Sweet | -0.039 |
ending | -0.038 |
Kids | -0.038 |
mistake | -0.038 |
Token | and |
Token ID | 290 |
Feature activation | +5.55e-02 |
Pos logit contributions | |
ay | +0.002 |
ong | +0.002 |
rac | +0.002 |
issors | +0.002 |
around | +0.002 |
Neg logit contributions | |
Sleep | -0.003 |
polar | -0.003 |
ending | -0.003 |
Sweet | -0.003 |
mistake | -0.003 |
Token | all |
Token ID | 477 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
around | +0.002 |
ong | +0.002 |
ay | +0.002 |
bs | +0.002 |
at | +0.001 |
Neg logit contributions | |
Sleep | -0.002 |
Sweet | -0.002 |
Kids | -0.002 |
mistake | -0.002 |
divorce | -0.002 |
Token | sorts |
Token ID | 10524 |
Feature activation | +7.42e+00 |
Pos logit contributions | |
ay | +0.041 |
ong | +0.040 |
rac | +0.039 |
issors | +0.039 |
ars | +0.036 |
Neg logit contributions | |
polar | -0.052 |
Sleep | -0.052 |
ending | -0.051 |
moment | -0.051 |
Sweet | -0.051 |
Token | of |
Token ID | 286 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
rac | +0.230 |
issors | +0.230 |
ay | +0.218 |
ong | +0.216 |
itching | +0.206 |
Neg logit contributions | |
Sleep | -0.254 |
forgiveness | -0.252 |
Sweet | -0.249 |
ending | -0.245 |
polar | -0.242 |
Token | fun |
Token ID | 1257 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
Sleep | +0.066 |
ending | +0.065 |
Sweet | +0.065 |
polar | +0.065 |
mistake | +0.063 |
Neg logit contributions | |
ay | -0.067 |
ong | -0.063 |
around | -0.061 |
rac | -0.059 |
bs | -0.057 |
Token | toys |
Token ID | 14958 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
ay | +0.040 |
ong | +0.037 |
around | +0.036 |
rac | +0.035 |
bs | +0.034 |
Neg logit contributions | |
Sleep | -0.051 |
polar | -0.051 |
Sweet | -0.050 |
ending | -0.050 |
mistake | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | -2.63e-01 |
Pos logit contributions | |
ay | +0.007 |
ong | +0.006 |
rac | +0.006 |
issors | +0.006 |
around | +0.006 |
Neg logit contributions | |
Sleep | -0.010 |
polar | -0.010 |
ending | -0.010 |
Sweet | -0.010 |
mistake | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
mistake | +0.011 |
ending | +0.010 |
Sleep | +0.010 |
polar | +0.010 |
forgiveness | +0.010 |
Neg logit contributions | |
ay | -0.008 |
around | -0.007 |
issors | -0.006 |
rac | -0.006 |
ong | -0.006 |
Token | book |
Token ID | 1492 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
Sleep | +0.057 |
ending | +0.057 |
polar | +0.057 |
Sweet | +0.057 |
mistake | +0.056 |
Neg logit contributions | |
ay | -0.045 |
ong | -0.042 |
rac | -0.041 |
issors | -0.040 |
around | -0.038 |
Token | and |
Token ID | 290 |
Feature activation | +2.35e-02 |
Pos logit contributions | |
ay | +0.007 |
rac | +0.007 |
ong | +0.007 |
issors | +0.007 |
bs | +0.006 |
Neg logit contributions | |
ending | -0.009 |
Sleep | -0.009 |
moment | -0.008 |
mistake | -0.008 |
polar | -0.008 |
Token | told |
Token ID | 1297 |
Feature activation | -8.81e-01 |
Pos logit contributions | |
around | +0.001 |
ay | +0.001 |
ong | +0.001 |
bs | +0.001 |
rac | +0.001 |
Neg logit contributions | |
Sleep | -0.001 |
Sweet | -0.001 |
polar | -0.001 |
ending | -0.001 |
mistake | -0.001 |
Token | her |
Token ID | 607 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
Sleep | +0.033 |
polar | +0.033 |
ending | +0.033 |
Sweet | +0.033 |
mistake | +0.033 |
Neg logit contributions | |
ay | -0.026 |
ong | -0.024 |
rac | -0.024 |
issors | -0.023 |
ars | -0.023 |
Token | all |
Token ID | 477 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
ay | +0.031 |
around | +0.030 |
ong | +0.028 |
rac | +0.027 |
bs | +0.026 |
Neg logit contributions | |
Sleep | -0.059 |
polar | -0.058 |
Sweet | -0.058 |
Kids | -0.057 |
Are | -0.056 |
Token | sorts |
Token ID | 10524 |
Feature activation | +7.38e+00 |
Pos logit contributions | |
ay | +0.031 |
ong | +0.029 |
around | +0.028 |
rac | +0.027 |
bs | +0.026 |
Neg logit contributions | |
Sleep | -0.043 |
Sweet | -0.043 |
polar | -0.043 |
ending | -0.042 |
mistake | -0.041 |
Token | of |
Token ID | 286 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
rac | +0.232 |
issors | +0.224 |
ay | +0.223 |
ong | +0.222 |
itching | +0.204 |
Neg logit contributions | |
forgiveness | -0.258 |
Sleep | -0.254 |
ending | -0.244 |
moral | -0.241 |
Sweet | -0.241 |
Token | new |
Token ID | 649 |
Feature activation | -5.64e-01 |
Pos logit contributions | |
mistake | +0.055 |
ending | +0.054 |
moral | +0.053 |
Sweet | +0.053 |
Sleep | +0.053 |
Neg logit contributions | |
ay | -0.076 |
rac | -0.075 |
ong | -0.073 |
issors | -0.069 |
ors | -0.066 |
Token | words |
Token ID | 2456 |
Feature activation | +9.17e-01 |
Pos logit contributions | |
mistake | +0.016 |
moment | +0.016 |
ending | +0.016 |
Sweet | +0.015 |
moral | +0.015 |
Neg logit contributions | |
rac | -0.022 |
ay | -0.021 |
ong | -0.021 |
issors | -0.020 |
ars | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -2.73e-01 |
Pos logit contributions | |
ay | +0.027 |
ong | +0.025 |
rac | +0.025 |
issors | +0.024 |
ars | +0.023 |
Neg logit contributions | |
Sleep | -0.034 |
ending | -0.033 |
Sweet | -0.033 |
moment | -0.033 |
polar | -0.033 |
Token | With |
Token ID | 2080 |
Feature activation | +5.15e-01 |
Pos logit contributions | |
mistake | +0.012 |
ending | +0.012 |
polar | +0.012 |
Sleep | +0.011 |
divorce | +0.011 |
Neg logit contributions | |
ay | -0.007 |
around | -0.006 |
rac | -0.006 |
ong | -0.006 |
bs | -0.006 |
Token | lake |
Token ID | 13546 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
Sweet | +0.018 |
Sleep | +0.018 |
Kids | +0.017 |
Acc | +0.016 |
Are | +0.016 |
Neg logit contributions | |
around | -0.023 |
ay | -0.022 |
at | -0.021 |
rac | -0.019 |
high | -0.019 |
Token | was |
Token ID | 373 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
Sweet | +0.035 |
Sleep | +0.035 |
polar | +0.034 |
ending | +0.033 |
mistake | +0.033 |
Neg logit contributions | |
ay | -0.045 |
around | -0.043 |
ong | -0.043 |
rac | -0.042 |
bs | -0.040 |
Token | filled |
Token ID | 5901 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
Sleep | +0.004 |
polar | +0.004 |
Sweet | +0.004 |
ending | +0.004 |
mistake | +0.004 |
Neg logit contributions | |
ay | -0.004 |
ong | -0.004 |
rac | -0.003 |
around | -0.003 |
issors | -0.003 |
Token | with |
Token ID | 351 |
Feature activation | +9.04e-01 |
Pos logit contributions | |
ay | +0.042 |
ong | +0.041 |
rac | +0.039 |
issors | +0.038 |
bs | +0.038 |
Neg logit contributions | |
moment | -0.053 |
Sleep | -0.052 |
mistake | -0.052 |
polar | -0.052 |
ending | -0.051 |
Token | all |
Token ID | 477 |
Feature activation | +9.28e-01 |
Pos logit contributions | |
ay | +0.031 |
ong | +0.029 |
around | +0.028 |
rac | +0.027 |
bs | +0.026 |
Neg logit contributions | |
Sleep | -0.030 |
Sweet | -0.030 |
ending | -0.029 |
polar | -0.029 |
mistake | -0.029 |
Token | sorts |
Token ID | 10524 |
Feature activation | +7.36e+00 |
Pos logit contributions | |
ay | +0.027 |
ong | +0.027 |
rac | +0.026 |
issors | +0.025 |
ars | +0.025 |
Neg logit contributions | |
polar | -0.034 |
Sleep | -0.034 |
moment | -0.034 |
Sweet | -0.034 |
ending | -0.033 |
Token | of |
Token ID | 286 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
issors | +0.262 |
rac | +0.261 |
ong | +0.247 |
ay | +0.238 |
itching | +0.226 |
Neg logit contributions | |
forgiveness | -0.225 |
Sleep | -0.219 |
polar | -0.210 |
Sweet | -0.210 |
ending | -0.207 |
Token | trouble |
Token ID | 5876 |
Feature activation | +9.25e-01 |
Pos logit contributions | |
Sleep | +0.067 |
Sweet | +0.066 |
ending | +0.066 |
polar | +0.065 |
mistake | +0.064 |
Neg logit contributions | |
ay | -0.067 |
ong | -0.063 |
around | -0.061 |
rac | -0.060 |
bs | -0.057 |
Token | . |
Token ID | 13 |
Feature activation | -4.43e-01 |
Pos logit contributions | |
ay | +0.023 |
ong | +0.021 |
around | +0.021 |
rac | +0.019 |
ors | +0.019 |
Neg logit contributions | |
Sweet | -0.039 |
Sleep | -0.039 |
polar | -0.038 |
ending | -0.038 |
Kids | -0.037 |
Token | As |
Token ID | 1081 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
Sleep | +0.016 |
polar | +0.016 |
ending | +0.016 |
Sweet | +0.016 |
mistake | +0.016 |
Neg logit contributions | |
ay | -0.013 |
ong | -0.012 |
rac | -0.012 |
around | -0.012 |
issors | -0.012 |
Token | soon |
Token ID | 2582 |
Feature activation | +4.95e-01 |
Pos logit contributions | |
ay | +0.033 |
around | +0.030 |
ong | +0.028 |
bs | +0.028 |
rac | +0.028 |
Neg logit contributions | |
Sleep | -0.041 |
Sweet | -0.040 |
polar | -0.040 |
mistake | -0.039 |
Kids | -0.038 |
Token | box |
Token ID | 3091 |
Feature activation | +6.20e-01 |
Pos logit contributions | |
ay | +0.014 |
around | +0.013 |
ong | +0.013 |
ars | +0.012 |
at | +0.012 |
Neg logit contributions | |
Kids | -0.009 |
Sleep | -0.009 |
Sweet | -0.009 |
polar | -0.008 |
ending | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -3.71e-01 |
Pos logit contributions | |
ay | +0.016 |
ong | +0.015 |
rac | +0.014 |
around | +0.014 |
issors | +0.013 |
Neg logit contributions | |
Sleep | -0.026 |
Sweet | -0.025 |
polar | -0.025 |
ending | -0.025 |
mistake | -0.025 |
Token | They |
Token ID | 1119 |
Feature activation | -4.09e-01 |
Pos logit contributions | |
mistake | +0.016 |
Sleep | +0.016 |
polar | +0.015 |
ending | +0.015 |
Sweet | +0.015 |
Neg logit contributions | |
ay | -0.010 |
around | -0.009 |
ong | -0.009 |
issors | -0.009 |
rac | -0.009 |
Token | guessed |
Token ID | 25183 |
Feature activation | +3.56e-01 |
Pos logit contributions | |
polar | +0.013 |
Sleep | +0.012 |
ending | +0.012 |
Sweet | +0.012 |
mistake | +0.012 |
Neg logit contributions | |
ay | -0.015 |
ong | -0.014 |
rac | -0.014 |
issors | -0.013 |
around | -0.013 |
Token | all |
Token ID | 477 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
ay | +0.012 |
ong | +0.011 |
rac | +0.011 |
issors | +0.010 |
ars | +0.010 |
Neg logit contributions | |
polar | -0.012 |
ending | -0.012 |
Sleep | -0.012 |
Sweet | -0.012 |
mistake | -0.012 |
Token | sorts |
Token ID | 10524 |
Feature activation | +7.34e+00 |
Pos logit contributions | |
ay | +0.032 |
ong | +0.030 |
rac | +0.030 |
issors | +0.028 |
ars | +0.027 |
Neg logit contributions | |
Sleep | -0.038 |
ending | -0.038 |
polar | -0.038 |
Sweet | -0.037 |
mistake | -0.037 |
Token | of |
Token ID | 286 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
rac | +0.253 |
issors | +0.242 |
ong | +0.238 |
ay | +0.237 |
itching | +0.229 |
Neg logit contributions | |
forgiveness | -0.230 |
ending | -0.216 |
Sleep | -0.212 |
moral | -0.208 |
moment | -0.206 |
Token | things |
Token ID | 1243 |
Feature activation | +9.18e-01 |
Pos logit contributions | |
polar | +0.063 |
mistake | +0.063 |
Sleep | +0.062 |
Sweet | +0.062 |
ending | +0.062 |
Neg logit contributions | |
ay | -0.072 |
rac | -0.069 |
ong | -0.068 |
issors | -0.065 |
bs | -0.062 |
Token | like |
Token ID | 588 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
ay | +0.029 |
rac | +0.027 |
ong | +0.027 |
issors | +0.026 |
ars | +0.025 |
Neg logit contributions | |
ending | -0.032 |
Sleep | -0.032 |
moment | -0.032 |
mistake | -0.031 |
polar | -0.031 |
Token | cand |
Token ID | 2658 |
Feature activation | -2.67e+00 |
Pos logit contributions | |
ay | +0.071 |
ong | +0.067 |
around | +0.065 |
rac | +0.064 |
issors | +0.062 |
Neg logit contributions | |
Sleep | -0.068 |
Sweet | -0.067 |
polar | -0.066 |
ending | -0.065 |
Kids | -0.064 |
Token | ies |
Token ID | 444 |
Feature activation | +2.64e+00 |
Pos logit contributions | |
polar | +0.055 |
mistake | +0.055 |
moment | +0.054 |
department | +0.053 |
Sweet | +0.052 |
Neg logit contributions | |
ay | -0.116 |
rac | -0.114 |
ong | -0.107 |
issors | -0.107 |
around | -0.105 |
Token | . |
Token ID | 13 |
Feature activation | -3.09e-01 |
Pos logit contributions | |
moment | +0.007 |
ending | +0.007 |
mistake | +0.006 |
island | +0.006 |
again | +0.006 |
Neg logit contributions | |
ay | -0.007 |
rac | -0.007 |
ong | -0.006 |
issors | -0.006 |
bs | -0.006 |
Token | She |
Token ID | 1375 |
Feature activation | -6.70e-02 |
Pos logit contributions | |
Sleep | +0.010 |
Sweet | +0.010 |
ending | +0.010 |
moment | +0.010 |
polar | +0.010 |
Neg logit contributions | |
ong | -0.010 |
ay | -0.010 |
rac | -0.010 |
ars | -0.009 |
around | -0.009 |
Token | tried |
Token ID | 3088 |
Feature activation | +8.37e-01 |
Pos logit contributions | |
polar | +0.002 |
department | +0.002 |
moment | +0.002 |
Sweet | +0.002 |
mistake | +0.002 |
Neg logit contributions | |
ay | -0.003 |
ong | -0.003 |
issors | -0.003 |
rac | -0.003 |
ors | -0.002 |
Token | using |
Token ID | 1262 |
Feature activation | +5.93e-01 |
Pos logit contributions | |
ay | +0.024 |
ong | +0.023 |
rac | +0.022 |
issors | +0.021 |
ars | +0.021 |
Neg logit contributions | |
mistake | -0.032 |
moment | -0.032 |
ending | -0.032 |
Sweet | -0.032 |
Sleep | -0.031 |
Token | all |
Token ID | 477 |
Feature activation | +9.92e-01 |
Pos logit contributions | |
ay | +0.025 |
ars | +0.025 |
ong | +0.025 |
rac | +0.024 |
led | +0.024 |
Neg logit contributions | |
treasure | -0.015 |
moment | -0.015 |
mistake | -0.014 |
forgiveness | -0.014 |
pear | -0.013 |
Token | sorts |
Token ID | 10524 |
Feature activation | +7.33e+00 |
Pos logit contributions | |
ong | +0.035 |
ars | +0.034 |
rac | +0.034 |
itching | +0.033 |
issors | +0.033 |
Neg logit contributions | |
moment | -0.032 |
Sweet | -0.028 |
moral | -0.028 |
Sleep | -0.027 |
forgiveness | -0.027 |
Token | of |
Token ID | 286 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
rac | +0.281 |
ong | +0.267 |
ables | +0.265 |
issors | +0.263 |
itching | +0.262 |
Neg logit contributions | |
forgiveness | -0.203 |
moral | -0.179 |
moment | -0.178 |
Sleep | -0.175 |
Sweet | -0.173 |
Token | measures |
Token ID | 5260 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
mistake | +0.051 |
moment | +0.050 |
moral | +0.048 |
forgiveness | +0.047 |
Sweet | +0.045 |
Neg logit contributions | |
rac | -0.076 |
ay | -0.074 |
ong | -0.073 |
issors | -0.069 |
ors | -0.067 |
Token | - |
Token ID | 532 |
Feature activation | -5.43e-01 |
Pos logit contributions | |
ay | +0.001 |
rac | +0.001 |
ong | +0.001 |
issors | +0.001 |
ars | +0.001 |
Neg logit contributions | |
moment | -0.001 |
mistake | -0.001 |
Bird | -0.001 |
department | -0.001 |
ending | -0.001 |
Token | a |
Token ID | 257 |
Feature activation | +2.89e-01 |
Pos logit contributions | |
Sleep | +0.018 |
polar | +0.018 |
Sweet | +0.018 |
ending | +0.018 |
mistake | +0.017 |
Neg logit contributions | |
ay | -0.018 |
ong | -0.017 |
rac | -0.017 |
around | -0.016 |
issors | -0.016 |
Token | ruler |
Token ID | 22740 |
Feature activation | +3.83e+00 |
Pos logit contributions | |
ay | +0.009 |
ong | +0.008 |
around | +0.008 |
rac | +0.007 |
issors | +0.007 |
Neg logit contributions | |
Sleep | -0.011 |
Kids | -0.011 |
Sweet | -0.011 |
polar | -0.010 |
ending | -0.010 |
Token | Now |
Token ID | 3844 |
Feature activation | +3.67e-01 |
Pos logit contributions | |
ay | +0.071 |
rac | +0.067 |
ong | +0.065 |
around | +0.064 |
issors | +0.063 |
Neg logit contributions | |
polar | -0.053 |
mistake | -0.052 |
ending | -0.052 |
department | -0.049 |
forgiveness | -0.048 |
Token | we |
Token ID | 356 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
ay | +0.012 |
ong | +0.012 |
rac | +0.011 |
around | +0.011 |
issors | +0.011 |
Neg logit contributions | |
Sleep | -0.012 |
polar | -0.012 |
Sweet | -0.012 |
ending | -0.012 |
mistake | -0.012 |
Token | can |
Token ID | 460 |
Feature activation | +6.82e-01 |
Pos logit contributions | |
Sleep | +0.032 |
polar | +0.032 |
Sweet | +0.032 |
ending | +0.032 |
mistake | +0.031 |
Neg logit contributions | |
ay | -0.035 |
ong | -0.033 |
rac | -0.032 |
around | -0.031 |
issors | -0.031 |
Token | mix |
Token ID | 5022 |
Feature activation | -9.01e-01 |
Pos logit contributions | |
ay | +0.025 |
ong | +0.025 |
led | +0.024 |
rac | +0.024 |
issors | +0.024 |
Neg logit contributions | |
moral | -0.018 |
polar | -0.018 |
mistake | -0.018 |
Sleep | -0.017 |
Sweet | -0.017 |
Token | all |
Token ID | 477 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
moment | +0.025 |
snow | +0.025 |
Sweet | +0.024 |
mistake | +0.024 |
again | +0.024 |
Neg logit contributions | |
ay | -0.031 |
issors | -0.030 |
ong | -0.030 |
ars | -0.030 |
rac | -0.030 |
Token | sorts |
Token ID | 10524 |
Feature activation | +7.32e+00 |
Pos logit contributions | |
ong | +0.039 |
issors | +0.039 |
ars | +0.039 |
ay | +0.039 |
rac | +0.038 |
Neg logit contributions | |
moment | -0.037 |
Sweet | -0.033 |
mistake | -0.033 |
moral | -0.033 |
Sleep | -0.032 |
Token | of |
Token ID | 286 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
itching | +0.239 |
rac | +0.238 |
issors | +0.237 |
ong | +0.232 |
ables | +0.231 |
Neg logit contributions | |
forgiveness | -0.220 |
Sweet | -0.211 |
Sleep | -0.208 |
moral | -0.198 |
moment | -0.193 |
Token | cool |
Token ID | 3608 |
Feature activation | +7.67e-02 |
Pos logit contributions | |
mistake | +0.058 |
Sweet | +0.058 |
moment | +0.057 |
Sleep | +0.055 |
moral | +0.055 |
Neg logit contributions | |
ay | -0.075 |
rac | -0.073 |
ong | -0.072 |
issors | -0.071 |
ars | -0.069 |
Token | things |
Token ID | 1243 |
Feature activation | +9.70e-01 |
Pos logit contributions | |
rac | +0.003 |
ay | +0.003 |
ong | +0.003 |
issors | +0.003 |
ars | +0.003 |
Neg logit contributions | |
moment | -0.002 |
Sweet | -0.002 |
Sleep | -0.002 |
mistake | -0.002 |
Kids | -0.002 |
Token | together |
Token ID | 1978 |
Feature activation | +3.07e-02 |
Pos logit contributions | |
ay | +0.031 |
ong | +0.029 |
rac | +0.029 |
issors | +0.028 |
ars | +0.028 |
Neg logit contributions | |
Sleep | -0.032 |
moment | -0.032 |
Sweet | -0.031 |
ending | -0.031 |
polar | -0.031 |
Token | !" |
Token ID | 2474 |
Feature activation | +9.88e-02 |
Pos logit contributions | |
ay | +0.001 |
ong | +0.001 |
rac | +0.001 |
issors | +0.001 |
bs | +0.001 |
Neg logit contributions | |
moment | -0.001 |
Sleep | -0.001 |
mistake | -0.001 |
ending | -0.001 |
Sweet | -0.001 |
Token | was |
Token ID | 373 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
ay | +0.094 |
issors | +0.089 |
ong | +0.088 |
rac | +0.083 |
bs | +0.081 |
Neg logit contributions | |
polar | -0.036 |
moment | -0.034 |
Sleep | -0.034 |
Sweet | -0.034 |
mistake | -0.034 |
Token | excited |
Token ID | 6568 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
polar | +0.004 |
Are | +0.003 |
Bird | +0.003 |
moment | +0.003 |
Sweet | +0.003 |
Neg logit contributions | |
ay | -0.004 |
issors | -0.004 |
ors | -0.004 |
ong | -0.004 |
bs | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
ay | +0.057 |
rac | +0.057 |
ars | +0.056 |
issors | +0.055 |
bs | +0.053 |
Neg logit contributions | |
ending | -0.048 |
moment | -0.047 |
mistake | -0.046 |
again | -0.045 |
Sleep | -0.043 |
Token | asked |
Token ID | 1965 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
polar | +0.000 |
moment | +0.000 |
ending | +0.000 |
Sleep | +0.000 |
Sweet | +0.000 |
Neg logit contributions | |
ay | -0.001 |
issors | -0.001 |
ong | -0.001 |
ars | -0.000 |
rac | -0.000 |
Token | all |
Token ID | 477 |
Feature activation | +9.91e-01 |
Pos logit contributions | |
polar | +0.048 |
Sleep | +0.048 |
Sweet | +0.048 |
ending | +0.047 |
mistake | +0.047 |
Neg logit contributions | |
ay | -0.039 |
ong | -0.037 |
rac | -0.036 |
issors | -0.034 |
around | -0.034 |
Token | sorts |
Token ID | 10524 |
Feature activation | +7.31e+00 |
Pos logit contributions | |
ay | +0.031 |
ong | +0.030 |
rac | +0.029 |
ars | +0.028 |
issors | +0.028 |
Neg logit contributions | |
polar | -0.036 |
moment | -0.034 |
mistake | -0.034 |
ending | -0.034 |
Sleep | -0.034 |
Token | of |
Token ID | 286 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
rac | +0.276 |
issors | +0.265 |
ables | +0.261 |
ay | +0.257 |
ong | +0.256 |
Neg logit contributions | |
forgiveness | -0.222 |
polar | -0.198 |
Sleep | -0.196 |
moral | -0.190 |
Sweet | -0.186 |
Token | questions |
Token ID | 2683 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
polar | +0.063 |
mistake | +0.062 |
Sweet | +0.061 |
Sleep | +0.060 |
moral | +0.060 |
Neg logit contributions | |
ay | -0.073 |
rac | -0.071 |
ong | -0.067 |
issors | -0.067 |
bs | -0.063 |
Token | . |
Token ID | 13 |
Feature activation | -4.33e-01 |
Pos logit contributions | |
ay | +0.076 |
rac | +0.072 |
ong | +0.071 |
issors | +0.068 |
bs | +0.066 |
Neg logit contributions | |
Sleep | -0.084 |
ending | -0.082 |
polar | -0.082 |
moment | -0.082 |
Sweet | -0.081 |
Token | |
Token ID | 220 |
Feature activation | +9.50e-01 |
Pos logit contributions | |
polar | +0.014 |
Sleep | +0.014 |
Sweet | +0.014 |
ending | +0.014 |
moment | +0.014 |
Neg logit contributions | |
ay | -0.014 |
ong | -0.014 |
rac | -0.013 |
around | -0.013 |
issors | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | -3.64e-02 |
Pos logit contributions | |
ay | +0.039 |
ong | +0.039 |
rac | +0.038 |
around | +0.036 |
issors | +0.036 |
Neg logit contributions | |
Sleep | -0.022 |
Sweet | -0.021 |
polar | -0.021 |
moment | -0.021 |
moral | -0.020 |
Token | into |
Token ID | 656 |
Feature activation | +4.19e-01 |
Pos logit contributions | |
ay | +0.089 |
issors | +0.088 |
ong | +0.087 |
ars | +0.085 |
rac | +0.083 |
Neg logit contributions | |
moment | -0.084 |
moral | -0.081 |
again | -0.080 |
ending | -0.080 |
Sleep | -0.079 |
Token | the |
Token ID | 262 |
Feature activation | +3.77e-01 |
Pos logit contributions | |
ay | +0.010 |
around | +0.009 |
ong | +0.009 |
rac | +0.008 |
bs | +0.008 |
Neg logit contributions | |
Sleep | -0.018 |
ending | -0.018 |
Sweet | -0.018 |
department | -0.018 |
polar | -0.018 |
Token | stream |
Token ID | 4269 |
Feature activation | -4.14e-03 |
Pos logit contributions | |
ay | +0.017 |
around | +0.017 |
ong | +0.016 |
ars | +0.015 |
s | +0.015 |
Neg logit contributions | |
Sweet | -0.009 |
Sleep | -0.008 |
ending | -0.008 |
forgiveness | -0.008 |
mistake | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | -3.65e-02 |
Pos logit contributions | |
moment | +0.000 |
polar | +0.000 |
ending | +0.000 |
Sleep | +0.000 |
mistake | +0.000 |
Neg logit contributions | |
ay | -0.000 |
rac | -0.000 |
issors | -0.000 |
ong | -0.000 |
bs | -0.000 |
Token | all |
Token ID | 477 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
Sleep | +0.001 |
ending | +0.001 |
mistake | +0.001 |
Sweet | +0.001 |
polar | +0.001 |
Neg logit contributions | |
around | -0.001 |
ong | -0.001 |
ay | -0.001 |
rac | -0.001 |
bs | -0.001 |
Token | sorts |
Token ID | 10524 |
Feature activation | +7.30e+00 |
Pos logit contributions | |
ay | +0.032 |
issors | +0.030 |
ong | +0.030 |
rac | +0.029 |
ars | +0.028 |
Neg logit contributions | |
moment | -0.039 |
polar | -0.038 |
ending | -0.037 |
Sleep | -0.036 |
Kids | -0.036 |
Token | of |
Token ID | 286 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
rac | +0.238 |
ay | +0.238 |
issors | +0.235 |
ong | +0.225 |
ables | +0.224 |
Neg logit contributions | |
forgiveness | -0.237 |
Sleep | -0.236 |
ending | -0.232 |
Sweet | -0.225 |
moral | -0.220 |
Token | interesting |
Token ID | 3499 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
Sleep | +0.071 |
ending | +0.070 |
Sweet | +0.070 |
department | +0.069 |
divorce | +0.067 |
Neg logit contributions | |
ay | -0.063 |
around | -0.062 |
ong | -0.061 |
bs | -0.055 |
ars | -0.054 |
Token | things |
Token ID | 1243 |
Feature activation | +8.79e-01 |
Pos logit contributions | |
ay | +0.050 |
ong | +0.048 |
rac | +0.046 |
around | +0.045 |
issors | +0.044 |
Neg logit contributions | |
Sleep | -0.047 |
polar | -0.047 |
Sweet | -0.047 |
ending | -0.046 |
mistake | -0.045 |
Token | caught |
Token ID | 4978 |
Feature activation | +2.81e-01 |
Pos logit contributions | |
ay | +0.026 |
rac | +0.024 |
ong | +0.024 |
issors | +0.022 |
ars | +0.021 |
Neg logit contributions | |
Sleep | -0.033 |
ending | -0.033 |
Sweet | -0.033 |
polar | -0.032 |
moment | -0.032 |
Token | her |
Token ID | 607 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
ay | +0.007 |
around | +0.006 |
ong | +0.006 |
rac | +0.006 |
bs | +0.005 |
Neg logit contributions | |
Sleep | -0.012 |
Sweet | -0.012 |
ending | -0.012 |
polar | -0.012 |
Kids | -0.012 |
Token | packing |
Token ID | 24157 |
Feature activation | +4.90e-02 |
Pos logit contributions | |
ay | +0.051 |
ong | +0.047 |
rac | +0.046 |
issors | +0.046 |
ars | +0.045 |
Neg logit contributions | |
polar | -0.053 |
ending | -0.051 |
mistake | -0.051 |
moment | -0.051 |
Sleep | -0.050 |
Token | his |
Token ID | 465 |
Feature activation | +6.98e-01 |
Pos logit contributions | |
ay | +0.002 |
ong | +0.002 |
rac | +0.002 |
issors | +0.002 |
bs | +0.002 |
Neg logit contributions | |
moment | -0.002 |
mistake | -0.001 |
ending | -0.001 |
again | -0.001 |
polar | -0.001 |
Token | bag |
Token ID | 6131 |
Feature activation | +3.16e+00 |
Pos logit contributions | |
ay | +0.028 |
ong | +0.027 |
issors | +0.026 |
bs | +0.026 |
rac | +0.026 |
Neg logit contributions | |
polar | -0.019 |
moment | -0.018 |
mistake | -0.018 |
ending | -0.018 |
bathroom | -0.017 |
Token | with |
Token ID | 351 |
Feature activation | +9.76e-01 |
Pos logit contributions | |
rac | +0.101 |
ay | +0.101 |
ong | +0.098 |
issors | +0.095 |
bs | +0.093 |
Neg logit contributions | |
moment | -0.103 |
again | -0.097 |
ending | -0.095 |
Sleep | -0.093 |
polar | -0.092 |
Token | all |
Token ID | 477 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
ay | +0.035 |
rac | +0.034 |
ong | +0.034 |
issors | +0.033 |
ars | +0.032 |
Neg logit contributions | |
polar | -0.032 |
Sleep | -0.029 |
moment | -0.029 |
ending | -0.029 |
forgiveness | -0.029 |
Token | sorts |
Token ID | 10524 |
Feature activation | +7.29e+00 |
Pos logit contributions | |
ong | +0.042 |
issors | +0.041 |
ars | +0.040 |
rac | +0.039 |
itching | +0.038 |
Neg logit contributions | |
moment | -0.029 |
polar | -0.028 |
ig | -0.027 |
snow | -0.026 |
Sleep | -0.025 |
Token | of |
Token ID | 286 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
rac | +0.299 |
issors | +0.284 |
ong | +0.270 |
ay | +0.263 |
ables | +0.260 |
Neg logit contributions | |
forgiveness | -0.188 |
polar | -0.179 |
Sleep | -0.174 |
ending | -0.170 |
moral | -0.166 |
Token | things |
Token ID | 1243 |
Feature activation | +8.62e-01 |
Pos logit contributions | |
polar | +0.062 |
Sleep | +0.058 |
moment | +0.057 |
mistake | +0.056 |
Sweet | +0.056 |
Neg logit contributions | |
rac | -0.078 |
ay | -0.076 |
ong | -0.074 |
issors | -0.074 |
ors | -0.068 |
Token | he |
Token ID | 339 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
ay | +0.032 |
rac | +0.031 |
ong | +0.029 |
issors | +0.029 |
ars | +0.026 |
Neg logit contributions | |
moment | -0.026 |
ending | -0.025 |
Sleep | -0.025 |
polar | -0.025 |
again | -0.025 |
Token | thought |
Token ID | 1807 |
Feature activation | +9.33e-01 |
Pos logit contributions | |
ay | +0.070 |
ong | +0.067 |
rac | +0.064 |
issors | +0.063 |
ors | +0.062 |
Neg logit contributions | |
ending | -0.032 |
Sleep | -0.032 |
polar | -0.032 |
moment | -0.032 |
moral | -0.030 |
Token | he |
Token ID | 339 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
ay | +0.031 |
rac | +0.030 |
issors | +0.029 |
ong | +0.029 |
itching | +0.028 |
Neg logit contributions | |
ending | -0.031 |
moment | -0.030 |
Sleep | -0.029 |
mistake | -0.029 |
moral | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | +2.62e-01 |
Pos logit contributions | |
moment | +0.024 |
mistake | +0.024 |
ending | +0.023 |
raining | +0.023 |
bathroom | +0.023 |
Neg logit contributions | |
ay | -0.030 |
ong | -0.028 |
rac | -0.027 |
ars | -0.025 |
issors | -0.025 |
Token | getting |
Token ID | 1972 |
Feature activation | -4.76e-01 |
Pos logit contributions | |
around | +0.008 |
ay | +0.008 |
ong | +0.008 |
rac | +0.008 |
issors | +0.007 |
Neg logit contributions | |
Sleep | -0.009 |
Sweet | -0.009 |
polar | -0.009 |
ending | -0.009 |
Kids | -0.009 |
Token | up |
Token ID | 510 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
Sleep | +0.017 |
polar | +0.016 |
Sweet | +0.016 |
ending | +0.016 |
mistake | +0.016 |
Neg logit contributions | |
ay | -0.015 |
ong | -0.015 |
rac | -0.014 |
around | -0.014 |
issors | -0.013 |
Token | to |
Token ID | 284 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
ay | +0.029 |
ong | +0.027 |
rac | +0.027 |
issors | +0.025 |
ars | +0.025 |
Neg logit contributions | |
Sleep | -0.039 |
polar | -0.039 |
ending | -0.038 |
Sweet | -0.038 |
mistake | -0.038 |
Token | all |
Token ID | 477 |
Feature activation | +9.64e-01 |
Pos logit contributions | |
Sleep | +0.008 |
department | +0.008 |
Sweet | +0.008 |
fly | +0.008 |
polar | +0.007 |
Neg logit contributions | |
ay | -0.004 |
ong | -0.004 |
around | -0.004 |
rac | -0.003 |
at | -0.003 |
Token | sorts |
Token ID | 10524 |
Feature activation | +7.28e+00 |
Pos logit contributions | |
ay | +0.033 |
ong | +0.033 |
ars | +0.032 |
rac | +0.031 |
issors | +0.031 |
Neg logit contributions | |
moment | -0.031 |
Sleep | -0.030 |
Sweet | -0.030 |
mistake | -0.030 |
polar | -0.030 |
Token | of |
Token ID | 286 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
rac | +0.271 |
ay | +0.255 |
ong | +0.252 |
issors | +0.252 |
ables | +0.242 |
Neg logit contributions | |
Sleep | -0.210 |
forgiveness | -0.209 |
pand | -0.203 |
Sweet | -0.198 |
ending | -0.195 |
Token | mischief |
Token ID | 38625 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
Sleep | +0.064 |
polar | +0.063 |
Sweet | +0.063 |
ending | +0.062 |
mistake | +0.061 |
Neg logit contributions | |
ay | -0.067 |
ong | -0.064 |
rac | -0.062 |
around | -0.060 |
issors | -0.059 |
Token | . |
Token ID | 13 |
Feature activation | -3.55e-01 |
Pos logit contributions | |
ay | +0.063 |
rac | +0.060 |
ong | +0.058 |
issors | +0.055 |
bs | +0.052 |
Neg logit contributions | |
Sleep | -0.077 |
polar | -0.076 |
ending | -0.076 |
Sweet | -0.075 |
mistake | -0.074 |
Token | Her |
Token ID | 2332 |
Feature activation | -3.26e+00 |
Pos logit contributions | |
Sleep | +0.011 |
polar | +0.011 |
Sweet | +0.011 |
ending | +0.011 |
mistake | +0.010 |
Neg logit contributions | |
ay | -0.013 |
ong | -0.012 |
rac | -0.012 |
around | -0.012 |
issors | -0.011 |
Token | favorite |
Token ID | 4004 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
ending | +0.086 |
Sleep | +0.086 |
polar | +0.086 |
Sweet | +0.085 |
mistake | +0.083 |
Neg logit contributions | |
ay | -0.132 |
ong | -0.127 |
rac | -0.124 |
around | -0.120 |
issors | -0.119 |
Token | the |
Token ID | 262 |
Feature activation | +3.20e-01 |
Pos logit contributions | |
ay | +0.022 |
ong | +0.021 |
rac | +0.020 |
around | +0.020 |
issors | +0.019 |
Neg logit contributions | |
Sleep | -0.029 |
polar | -0.029 |
Sweet | -0.029 |
ending | -0.029 |
mistake | -0.028 |
Token | microscope |
Token ID | 36396 |
Feature activation | +5.37e-01 |
Pos logit contributions | |
ay | +0.011 |
ong | +0.011 |
rac | +0.011 |
issors | +0.010 |
around | +0.010 |
Neg logit contributions | |
Sleep | -0.010 |
polar | -0.010 |
ending | -0.010 |
Sweet | -0.010 |
mistake | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
rac | +0.017 |
ay | +0.017 |
issors | +0.016 |
ong | +0.015 |
bs | +0.015 |
Neg logit contributions | |
moment | -0.018 |
again | -0.018 |
ending | -0.018 |
polar | -0.017 |
department | -0.017 |
Token | discovering |
Token ID | 21611 |
Feature activation | -7.68e-01 |
Pos logit contributions | |
ay | +0.005 |
ong | +0.005 |
rac | +0.005 |
around | +0.004 |
issors | +0.004 |
Neg logit contributions | |
Sleep | -0.004 |
polar | -0.004 |
Sweet | -0.004 |
ending | -0.004 |
mistake | -0.004 |
Token | all |
Token ID | 477 |
Feature activation | +9.96e-01 |
Pos logit contributions | |
polar | +0.024 |
Sleep | +0.024 |
Sweet | +0.023 |
ending | +0.023 |
mistake | +0.023 |
Neg logit contributions | |
ay | -0.028 |
ong | -0.026 |
rac | -0.026 |
issors | -0.025 |
around | -0.024 |
Token | sorts |
Token ID | 10524 |
Feature activation | +7.28e+00 |
Pos logit contributions | |
rac | +0.034 |
ong | +0.033 |
issors | +0.033 |
ay | +0.033 |
ars | +0.032 |
Neg logit contributions | |
moment | -0.033 |
polar | -0.032 |
ending | -0.031 |
Sleep | -0.031 |
mistake | -0.030 |
Token | of |
Token ID | 286 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
rac | +0.265 |
issors | +0.258 |
ay | +0.239 |
ong | +0.236 |
ars | +0.224 |
Neg logit contributions | |
forgiveness | -0.218 |
Sleep | -0.213 |
polar | -0.211 |
ending | -0.211 |
moment | -0.206 |
Token | wonderful |
Token ID | 7932 |
Feature activation | -6.32e-01 |
Pos logit contributions | |
polar | +0.058 |
mistake | +0.058 |
Sweet | +0.058 |
Sleep | +0.058 |
ending | +0.057 |
Neg logit contributions | |
ay | -0.072 |
rac | -0.070 |
ong | -0.069 |
issors | -0.066 |
ars | -0.064 |
Token | things |
Token ID | 1243 |
Feature activation | +7.63e-01 |
Pos logit contributions | |
moment | +0.018 |
polar | +0.018 |
Sweet | +0.017 |
ending | +0.017 |
mistake | +0.017 |
Neg logit contributions | |
ay | -0.024 |
rac | -0.024 |
ong | -0.024 |
issors | -0.023 |
ars | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
rac | +0.027 |
ay | +0.026 |
issors | +0.025 |
ong | +0.024 |
ars | +0.023 |
Neg logit contributions | |
moment | -0.024 |
Sleep | -0.024 |
ending | -0.024 |
polar | -0.023 |
again | -0.023 |
Token | |
Token ID | 220 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
Sleep | +0.004 |
Sweet | +0.004 |
polar | +0.004 |
ending | +0.004 |
moment | +0.004 |
Neg logit contributions | |
ay | -0.005 |
ong | -0.005 |
rac | -0.005 |
issors | -0.004 |
around | -0.004 |
Token | mushrooms |
Token ID | 23452 |
Feature activation | +3.77e-02 |
Pos logit contributions | |
ay | +0.010 |
ong | +0.010 |
rac | +0.009 |
around | +0.009 |
issors | +0.009 |
Neg logit contributions | |
Sleep | -0.007 |
polar | -0.007 |
Sweet | -0.007 |
ending | -0.007 |
mistake | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | +4.70e-01 |
Pos logit contributions | |
ay | +0.001 |
ong | +0.001 |
rac | +0.001 |
around | +0.001 |
issors | +0.001 |
Neg logit contributions | |
Sleep | -0.001 |
polar | -0.001 |
Sweet | -0.001 |
ending | -0.001 |
mistake | -0.001 |
Token | asking |
Token ID | 4737 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
around | +0.013 |
ay | +0.013 |
ong | +0.013 |
rac | +0.012 |
at | +0.012 |
Neg logit contributions | |
Sleep | -0.018 |
Sweet | -0.018 |
polar | -0.018 |
ending | -0.017 |
Kids | -0.017 |
Token | them |
Token ID | 606 |
Feature activation | +5.29e-01 |
Pos logit contributions | |
ay | +0.035 |
ong | +0.033 |
rac | +0.032 |
issors | +0.030 |
ars | +0.030 |
Neg logit contributions | |
Sleep | -0.037 |
ending | -0.037 |
polar | -0.037 |
Sweet | -0.037 |
mistake | -0.037 |
Token | all |
Token ID | 477 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
around | +0.015 |
ay | +0.014 |
ong | +0.014 |
rac | +0.013 |
bs | +0.013 |
Neg logit contributions | |
Sleep | -0.021 |
polar | -0.020 |
Sweet | -0.020 |
Kids | -0.020 |
ending | -0.020 |
Token | sorts |
Token ID | 10524 |
Feature activation | +7.28e+00 |
Pos logit contributions | |
ay | +0.030 |
around | +0.029 |
ong | +0.029 |
rac | +0.027 |
bs | +0.026 |
Neg logit contributions | |
Sweet | -0.042 |
Sleep | -0.042 |
polar | -0.041 |
ending | -0.040 |
Kids | -0.040 |
Token | of |
Token ID | 286 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
rac | +0.226 |
ay | +0.224 |
ong | +0.215 |
issors | +0.214 |
itching | +0.196 |
Neg logit contributions | |
forgiveness | -0.252 |
Sleep | -0.248 |
polar | -0.241 |
ending | -0.240 |
moral | -0.239 |
Token | questions |
Token ID | 2683 |
Feature activation | +2.59e+00 |
Pos logit contributions | |
mistake | +0.057 |
Sleep | +0.057 |
polar | +0.057 |
moral | +0.056 |
ending | +0.056 |
Neg logit contributions | |
ay | -0.072 |
rac | -0.070 |
ong | -0.068 |
issors | -0.067 |
ors | -0.063 |
Token | . |
Token ID | 13 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
ay | +0.073 |
ong | +0.069 |
rac | +0.067 |
issors | +0.063 |
around | +0.062 |
Neg logit contributions | |
Sleep | -0.099 |
polar | -0.097 |
ending | -0.097 |
Sweet | -0.097 |
moment | -0.095 |
Token | |
Token ID | 220 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
mistake | +0.006 |
ending | +0.006 |
Sleep | +0.006 |
polar | +0.006 |
Sweet | +0.006 |
Neg logit contributions | |
ay | -0.003 |
around | -0.003 |
rac | -0.003 |
ong | -0.003 |
issors | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +3.21e-01 |
Pos logit contributions | |
ay | +0.044 |
ong | +0.042 |
rac | +0.041 |
around | +0.040 |
issors | +0.039 |
Neg logit contributions | |
Sleep | -0.032 |
Sweet | -0.031 |
polar | -0.031 |
ending | -0.030 |
mistake | -0.030 |
Token | impressed |
Token ID | 12617 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
mistake | +0.016 |
Sweet | +0.015 |
polar | +0.015 |
moment | +0.014 |
moral | +0.014 |
Neg logit contributions | |
ay | -0.024 |
ors | -0.023 |
issors | -0.023 |
rac | -0.021 |
ong | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | -3.64e-01 |
Pos logit contributions | |
ong | +0.047 |
ay | +0.047 |
rac | +0.046 |
issors | +0.046 |
ars | +0.044 |
Neg logit contributions | |
again | -0.039 |
moment | -0.038 |
forgiveness | -0.038 |
Sleep | -0.038 |
Kids | -0.038 |
Token | Inside |
Token ID | 14384 |
Feature activation | +6.59e-01 |
Pos logit contributions | |
Sleep | +0.013 |
Sweet | +0.013 |
polar | +0.013 |
ending | +0.012 |
mistake | +0.012 |
Neg logit contributions | |
ay | -0.012 |
ong | -0.011 |
rac | -0.011 |
around | -0.010 |
issors | -0.010 |
Token | were |
Token ID | 547 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
ong | +0.026 |
issors | +0.025 |
ay | +0.024 |
rac | +0.024 |
ars | +0.024 |
Neg logit contributions | |
moment | -0.019 |
ending | -0.017 |
mistake | -0.017 |
polar | -0.016 |
moral | -0.016 |
Token | all |
Token ID | 477 |
Feature activation | +9.96e-01 |
Pos logit contributions | |
Sleep | +0.045 |
mistake | +0.044 |
ending | +0.043 |
moment | +0.043 |
Sweet | +0.043 |
Neg logit contributions | |
ay | -0.038 |
ong | -0.036 |
around | -0.036 |
rac | -0.033 |
at | -0.032 |
Token | sorts |
Token ID | 10524 |
Feature activation | +7.28e+00 |
Pos logit contributions | |
ong | +0.032 |
ay | +0.032 |
rac | +0.031 |
issors | +0.031 |
ars | +0.030 |
Neg logit contributions | |
polar | -0.034 |
moment | -0.033 |
Sweet | -0.033 |
Sleep | -0.032 |
ending | -0.032 |
Token | of |
Token ID | 286 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
rac | +0.262 |
issors | +0.248 |
ong | +0.242 |
ay | +0.235 |
ables | +0.232 |
Neg logit contributions | |
forgiveness | -0.224 |
Sweet | -0.209 |
Sleep | -0.207 |
ending | -0.205 |
polar | -0.204 |
Token | snacks |
Token ID | 28487 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
Sleep | +0.070 |
polar | +0.069 |
Sweet | +0.069 |
ending | +0.068 |
mistake | +0.067 |
Neg logit contributions | |
ay | -0.072 |
ong | -0.068 |
rac | -0.067 |
around | -0.064 |
issors | -0.064 |
Token | , |
Token ID | 11 |
Feature activation | +9.29e-02 |
Pos logit contributions | |
ay | +0.052 |
rac | +0.050 |
ong | +0.049 |
issors | +0.046 |
ars | +0.044 |
Neg logit contributions | |
ending | -0.067 |
polar | -0.066 |
Sleep | -0.066 |
moment | -0.065 |
Sweet | -0.064 |
Token | treats |
Token ID | 18432 |
Feature activation | -2.89e-01 |
Pos logit contributions | |
ay | +0.003 |
around | +0.003 |
ong | +0.003 |
bs | +0.003 |
rac | +0.003 |
Neg logit contributions | |
Sleep | -0.003 |
polar | -0.003 |
Sweet | -0.003 |
ending | -0.003 |
mistake | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | -4.00e-02 |
Pos logit contributions | |
polar | +0.011 |
Sleep | +0.011 |
ending | +0.011 |
Sweet | +0.011 |
moment | +0.011 |
Neg logit contributions | |
ay | -0.008 |
ong | -0.008 |
rac | -0.007 |
issors | -0.007 |
around | -0.007 |
Token | le |
Token ID | 293 |
Feature activation | +5.27e-01 |
Pos logit contributions | |
Sleep | +0.093 |
polar | +0.092 |
ending | +0.091 |
Sweet | +0.091 |
mistake | +0.089 |
Neg logit contributions | |
ay | -0.057 |
ong | -0.054 |
around | -0.053 |
rac | -0.051 |
bs | -0.050 |
Token | ran |
Token ID | 4966 |
Feature activation | +2.60e-01 |
Pos logit contributions | |
ay | +0.021 |
ong | +0.020 |
rac | +0.020 |
around | +0.019 |
issors | +0.019 |
Neg logit contributions | |
Sleep | -0.014 |
polar | -0.014 |
Sweet | -0.014 |
ending | -0.014 |
mistake | -0.014 |
Token | to |
Token ID | 284 |
Feature activation | -1.94e-01 |
Pos logit contributions | |
rac | +0.009 |
ay | +0.009 |
ars | +0.009 |
ong | +0.008 |
issors | +0.008 |
Neg logit contributions | |
moment | -0.009 |
again | -0.008 |
department | -0.008 |
bathroom | -0.008 |
raining | -0.008 |
Token | grab |
Token ID | 5552 |
Feature activation | +8.26e-02 |
Pos logit contributions | |
moment | +0.006 |
Sweet | +0.006 |
mistake | +0.006 |
Sleep | +0.006 |
polar | +0.006 |
Neg logit contributions | |
ay | -0.007 |
ong | -0.007 |
rac | -0.006 |
ars | -0.006 |
issors | -0.006 |
Token | all |
Token ID | 477 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
ay | +0.003 |
ong | +0.003 |
rac | +0.003 |
ars | +0.003 |
issors | +0.003 |
Neg logit contributions | |
moment | -0.002 |
Sweet | -0.002 |
polar | -0.002 |
pear | -0.002 |
moral | -0.002 |
Token | sorts |
Token ID | 10524 |
Feature activation | +7.28e+00 |
Pos logit contributions | |
ars | +0.036 |
ong | +0.036 |
issors | +0.036 |
itching | +0.035 |
rac | +0.035 |
Neg logit contributions | |
moment | -0.031 |
Sweet | -0.030 |
pear | -0.029 |
Sleep | -0.029 |
Ab | -0.028 |
Token | of |
Token ID | 286 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
rac | +0.284 |
ables | +0.268 |
issors | +0.266 |
ay | +0.257 |
ong | +0.257 |
Neg logit contributions | |
forgiveness | -0.193 |
Sweet | -0.186 |
pand | -0.182 |
Sleep | -0.177 |
moral | -0.172 |
Token | decorations |
Token ID | 39011 |
Feature activation | +3.23e-02 |
Pos logit contributions | |
Sweet | +0.057 |
polar | +0.056 |
moment | +0.056 |
mistake | +0.055 |
Sleep | +0.055 |
Neg logit contributions | |
ay | -0.071 |
rac | -0.070 |
ong | -0.068 |
issors | -0.068 |
ars | -0.064 |
Token | , |
Token ID | 11 |
Feature activation | +3.02e-01 |
Pos logit contributions | |
rac | +0.001 |
ay | +0.001 |
ong | +0.001 |
issors | +0.001 |
ars | +0.001 |
Neg logit contributions | |
moment | -0.001 |
ending | -0.001 |
mistake | -0.001 |
again | -0.001 |
Sleep | -0.001 |
Token | including |
Token ID | 1390 |
Feature activation | +6.64e-01 |
Pos logit contributions | |
ay | +0.009 |
around | +0.009 |
ong | +0.009 |
rac | +0.008 |
issors | +0.008 |
Neg logit contributions | |
Sleep | -0.011 |
polar | -0.011 |
ending | -0.011 |
Sweet | -0.011 |
Kids | -0.011 |
Token | balloons |
Token ID | 35485 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
ay | +0.023 |
ong | +0.022 |
around | +0.021 |
rac | +0.020 |
bs | +0.020 |
Neg logit contributions | |
Sleep | -0.022 |
Sweet | -0.022 |
ending | -0.022 |
mistake | -0.021 |
Kids | -0.021 |
Token | coins |
Token ID | 10796 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
ay | +0.036 |
ong | +0.034 |
rac | +0.034 |
issors | +0.032 |
around | +0.031 |
Neg logit contributions | |
polar | -0.036 |
ending | -0.036 |
Sleep | -0.036 |
Sweet | -0.035 |
moment | -0.035 |
Token | and |
Token ID | 290 |
Feature activation | -3.63e-02 |
Pos logit contributions | |
ay | +0.059 |
issors | +0.054 |
rac | +0.054 |
ong | +0.053 |
ars | +0.049 |
Neg logit contributions | |
moment | -0.063 |
ending | -0.061 |
polar | -0.061 |
Sleep | -0.060 |
mistake | -0.060 |
Token | toys |
Token ID | 14958 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
Sleep | +0.001 |
Sweet | +0.001 |
mistake | +0.001 |
Kids | +0.001 |
ending | +0.001 |
Neg logit contributions | |
ay | -0.001 |
around | -0.001 |
ong | -0.001 |
rac | -0.001 |
bs | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
ay | +0.004 |
rac | +0.004 |
ong | +0.004 |
issors | +0.004 |
ars | +0.003 |
Neg logit contributions | |
ending | -0.005 |
moment | -0.005 |
Sleep | -0.005 |
polar | -0.005 |
mistake | -0.005 |
Token | all |
Token ID | 477 |
Feature activation | +9.81e-01 |
Pos logit contributions | |
ay | +0.007 |
around | +0.006 |
ong | +0.006 |
rac | +0.006 |
bs | +0.006 |
Neg logit contributions | |
Sleep | -0.008 |
Sweet | -0.008 |
ending | -0.007 |
mistake | -0.007 |
Kids | -0.007 |
Token | sorts |
Token ID | 10524 |
Feature activation | +7.27e+00 |
Pos logit contributions | |
ay | +0.029 |
ong | +0.028 |
rac | +0.028 |
issors | +0.027 |
ars | +0.026 |
Neg logit contributions | |
polar | -0.036 |
ending | -0.036 |
moment | -0.035 |
Sleep | -0.035 |
mistake | -0.034 |
Token | of |
Token ID | 286 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
rac | +0.243 |
issors | +0.241 |
ong | +0.229 |
ay | +0.229 |
itching | +0.211 |
Neg logit contributions | |
ending | -0.232 |
forgiveness | -0.231 |
Sleep | -0.228 |
polar | -0.228 |
Sweet | -0.221 |
Token | surprises |
Token ID | 24072 |
Feature activation | -6.39e-01 |
Pos logit contributions | |
Sleep | +0.068 |
Sweet | +0.068 |
polar | +0.067 |
ending | +0.067 |
mistake | +0.065 |
Neg logit contributions | |
ay | -0.068 |
ong | -0.064 |
around | -0.062 |
rac | -0.062 |
issors | -0.059 |
Token | were |
Token ID | 547 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
Sleep | +0.023 |
polar | +0.023 |
ending | +0.023 |
moment | +0.023 |
Sweet | +0.023 |
Neg logit contributions | |
ay | -0.019 |
ong | -0.018 |
rac | -0.018 |
issors | -0.017 |
ars | -0.016 |
Token | waiting |
Token ID | 4953 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
Sleep | +0.038 |
Sweet | +0.038 |
polar | +0.038 |
ending | +0.037 |
mistake | +0.037 |
Neg logit contributions | |
ay | -0.035 |
ong | -0.033 |
rac | -0.031 |
around | -0.031 |
issors | -0.030 |
Token | for |
Token ID | 329 |
Feature activation | +2.52e+00 |
Pos logit contributions | |
ending | +0.054 |
polar | +0.054 |
Sleep | +0.054 |
mistake | +0.053 |
moment | +0.053 |
Neg logit contributions | |
ay | -0.039 |
ong | -0.037 |
rac | -0.037 |
issors | -0.035 |
bs | -0.033 |
Token | a |
Token ID | 257 |
Feature activation | -9.43e-02 |
Pos logit contributions | |
around | +0.047 |
ay | +0.046 |
at | +0.042 |
ong | +0.042 |
high | +0.040 |
Neg logit contributions | |
polar | -0.078 |
Kids | -0.073 |
Sleep | -0.072 |
Sweet | -0.072 |
ending | -0.072 |
Token | strange |
Token ID | 6283 |
Feature activation | -3.59e-01 |
Pos logit contributions | |
Sleep | +0.004 |
Sweet | +0.004 |
Kids | +0.004 |
forgiveness | +0.004 |
ending | +0.004 |
Neg logit contributions | |
ay | -0.003 |
ong | -0.002 |
around | -0.002 |
high | -0.002 |
rac | -0.002 |
Token | vendor |
Token ID | 18371 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
Sweet | +0.013 |
Sleep | +0.013 |
Kids | +0.012 |
Are | +0.012 |
forgiveness | +0.011 |
Neg logit contributions | |
ay | -0.012 |
around | -0.012 |
ong | -0.011 |
at | -0.010 |
high | -0.010 |
Token | selling |
Token ID | 6301 |
Feature activation | -3.24e+00 |
Pos logit contributions | |
Sleep | +0.052 |
Sweet | +0.052 |
ending | +0.051 |
polar | +0.051 |
mistake | +0.050 |
Neg logit contributions | |
ay | -0.038 |
ong | -0.036 |
around | -0.036 |
rac | -0.033 |
issors | -0.032 |
Token | all |
Token ID | 477 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
polar | +0.084 |
Sleep | +0.082 |
Sweet | +0.081 |
mistake | +0.081 |
ending | +0.080 |
Neg logit contributions | |
ay | -0.131 |
ong | -0.126 |
rac | -0.124 |
issors | -0.120 |
around | -0.119 |
Token | sorts |
Token ID | 10524 |
Feature activation | +7.27e+00 |
Pos logit contributions | |
ay | +0.031 |
ong | +0.031 |
rac | +0.031 |
ars | +0.030 |
issors | +0.029 |
Neg logit contributions | |
moment | -0.035 |
Sleep | -0.035 |
polar | -0.035 |
Sweet | -0.034 |
ending | -0.034 |
Token | of |
Token ID | 286 |
Feature activation | -2.10e+00 |
Pos logit contributions | |
rac | +0.254 |
ay | +0.234 |
ong | +0.230 |
issors | +0.229 |
itching | +0.222 |
Neg logit contributions | |
forgiveness | -0.226 |
Sleep | -0.216 |
Sweet | -0.215 |
pand | -0.213 |
ending | -0.211 |
Token | interesting |
Token ID | 3499 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
polar | +0.061 |
Sweet | +0.060 |
Sleep | +0.060 |
mistake | +0.060 |
ending | +0.059 |
Neg logit contributions | |
ay | -0.079 |
rac | -0.075 |
ong | -0.075 |
issors | -0.071 |
around | -0.070 |
Token | things |
Token ID | 1243 |
Feature activation | +7.89e-01 |
Pos logit contributions | |
ay | +0.047 |
rac | +0.045 |
ong | +0.045 |
issors | +0.042 |
around | +0.041 |
Neg logit contributions | |
polar | -0.041 |
Sweet | -0.041 |
Sleep | -0.041 |
ending | -0.040 |
mistake | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | -5.28e-01 |
Pos logit contributions | |
ay | +0.024 |
rac | +0.022 |
ong | +0.022 |
issors | +0.020 |
ars | +0.020 |
Neg logit contributions | |
Sleep | -0.029 |
ending | -0.029 |
polar | -0.029 |
moment | -0.028 |
Sweet | -0.028 |
Token | She |
Token ID | 1375 |
Feature activation | -4.50e-02 |
Pos logit contributions | |
Sleep | +0.015 |
Sweet | +0.015 |
polar | +0.014 |
ending | +0.014 |
moment | +0.014 |
Neg logit contributions | |
ay | -0.020 |
ong | -0.020 |
rac | -0.019 |
issors | -0.018 |
ars | -0.018 |
Token | ! |
Token ID | 0 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
ay | +0.153 |
rac | +0.141 |
ong | +0.139 |
issors | +0.131 |
ars | +0.128 |
Neg logit contributions | |
Sleep | -0.187 |
polar | -0.187 |
ending | -0.185 |
mistake | -0.181 |
moment | -0.180 |
Token | He |
Token ID | 679 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
ay | +0.034 |
ong | +0.032 |
around | +0.031 |
rac | +0.031 |
issors | +0.029 |
Neg logit contributions | |
Sleep | -0.043 |
Sweet | -0.043 |
polar | -0.043 |
mistake | -0.043 |
ending | -0.042 |
Token | would |
Token ID | 561 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
ay | +0.008 |
ong | +0.007 |
rac | +0.007 |
around | +0.007 |
bs | +0.007 |
Neg logit contributions | |
Sleep | -0.007 |
Sweet | -0.007 |
polar | -0.007 |
ending | -0.007 |
mistake | -0.007 |
Token | make |
Token ID | 787 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
ay | +0.050 |
ong | +0.047 |
rac | +0.046 |
issors | +0.044 |
ors | +0.044 |
Neg logit contributions | |
polar | -0.034 |
Sleep | -0.033 |
ending | -0.032 |
mistake | -0.032 |
moment | -0.032 |
Token | all |
Token ID | 477 |
Feature activation | +9.64e-01 |
Pos logit contributions | |
ay | +0.006 |
rac | +0.006 |
issors | +0.006 |
ong | +0.006 |
bs | +0.006 |
Neg logit contributions | |
polar | -0.006 |
mistake | -0.005 |
Sweet | -0.005 |
Bird | -0.005 |
ending | -0.005 |
Token | sorts |
Token ID | 10524 |
Feature activation | +7.27e+00 |
Pos logit contributions | |
ay | +0.029 |
ong | +0.027 |
rac | +0.027 |
issors | +0.026 |
ars | +0.025 |
Neg logit contributions | |
polar | -0.036 |
Sleep | -0.035 |
Sweet | -0.035 |
ending | -0.034 |
moment | -0.034 |
Token | of |
Token ID | 286 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
rac | +0.264 |
issors | +0.256 |
ay | +0.241 |
ong | +0.231 |
ables | +0.225 |
Neg logit contributions | |
forgiveness | -0.224 |
polar | -0.213 |
Sleep | -0.211 |
Sweet | -0.205 |
ending | -0.201 |
Token | shapes |
Token ID | 15268 |
Feature activation | +2.71e+00 |
Pos logit contributions | |
polar | +0.058 |
mistake | +0.058 |
Sweet | +0.057 |
Sleep | +0.056 |
moment | +0.056 |
Neg logit contributions | |
ay | -0.076 |
rac | -0.075 |
issors | -0.071 |
ong | -0.070 |
ors | -0.066 |
Token | in |
Token ID | 287 |
Feature activation | -4.10e-01 |
Pos logit contributions | |
ay | +0.078 |
rac | +0.074 |
ong | +0.073 |
issors | +0.071 |
bs | +0.066 |
Neg logit contributions | |
polar | -0.102 |
ending | -0.100 |
Sleep | -0.100 |
moment | -0.098 |
Sweet | -0.098 |
Token | the |
Token ID | 262 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
polar | +0.015 |
mistake | +0.014 |
ending | +0.014 |
Sleep | +0.014 |
moment | +0.014 |
Neg logit contributions | |
ay | -0.013 |
ong | -0.012 |
rac | -0.012 |
issors | -0.011 |
bs | -0.011 |
Token | air |
Token ID | 1633 |
Feature activation | +2.79e+00 |
Pos logit contributions | |
rac | +0.012 |
ong | +0.012 |
ors | +0.012 |
bs | +0.011 |
issors | +0.011 |
Neg logit contributions | |
moment | -0.007 |
tunnel | -0.007 |
Sleep | -0.007 |
ending | -0.007 |
Sweet | -0.007 |
Token | he |
Token ID | 339 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
ay | +0.016 |
ong | +0.015 |
rac | +0.015 |
issors | +0.014 |
around | +0.014 |
Neg logit contributions | |
polar | -0.018 |
Sleep | -0.018 |
ending | -0.017 |
Sweet | -0.017 |
mistake | -0.017 |
Token | was |
Token ID | 373 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
ay | +0.063 |
ong | +0.060 |
issors | +0.059 |
rac | +0.059 |
bs | +0.056 |
Neg logit contributions | |
polar | -0.037 |
ending | -0.036 |
Sleep | -0.036 |
mistake | -0.036 |
Sweet | -0.035 |
Token | acting |
Token ID | 7205 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
polar | +0.005 |
Sleep | +0.005 |
ending | +0.005 |
Sweet | +0.005 |
mistake | +0.005 |
Neg logit contributions | |
ay | -0.006 |
ong | -0.005 |
rac | -0.005 |
issors | -0.005 |
bs | -0.005 |
Token | out |
Token ID | 503 |
Feature activation | +2.68e+00 |
Pos logit contributions | |
ending | +0.044 |
moment | +0.043 |
mistake | +0.043 |
polar | +0.042 |
Sleep | +0.042 |
Neg logit contributions | |
ay | -0.039 |
rac | -0.038 |
issors | -0.038 |
ong | -0.036 |
ors | -0.035 |
Token | all |
Token ID | 477 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
ay | +0.075 |
ong | +0.071 |
rac | +0.070 |
issors | +0.068 |
ars | +0.064 |
Neg logit contributions | |
polar | -0.104 |
Sleep | -0.103 |
ending | -0.103 |
mistake | -0.102 |
Sweet | -0.101 |
Token | sorts |
Token ID | 10524 |
Feature activation | +7.27e+00 |
Pos logit contributions | |
ay | +0.032 |
ong | +0.030 |
rac | +0.030 |
issors | +0.029 |
bs | +0.027 |
Neg logit contributions | |
Sleep | -0.041 |
polar | -0.041 |
ending | -0.041 |
Sweet | -0.040 |
moment | -0.040 |
Token | of |
Token ID | 286 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
rac | +0.272 |
issors | +0.263 |
ay | +0.245 |
ong | +0.242 |
itching | +0.229 |
Neg logit contributions | |
forgiveness | -0.216 |
ending | -0.213 |
Sleep | -0.212 |
again | -0.209 |
polar | -0.208 |
Token | adventures |
Token ID | 17545 |
Feature activation | +6.45e-01 |
Pos logit contributions | |
polar | +0.059 |
Sleep | +0.058 |
ending | +0.057 |
mistake | +0.057 |
Sweet | +0.057 |
Neg logit contributions | |
ay | -0.067 |
rac | -0.064 |
ong | -0.064 |
issors | -0.061 |
around | -0.059 |
Token | . |
Token ID | 13 |
Feature activation | -8.56e-02 |
Pos logit contributions | |
ay | +0.020 |
rac | +0.020 |
ong | +0.019 |
issors | +0.019 |
bs | +0.018 |
Neg logit contributions | |
polar | -0.022 |
ending | -0.022 |
moment | -0.022 |
Sleep | -0.022 |
mistake | -0.021 |
Token | Every |
Token ID | 3887 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
Sleep | +0.003 |
polar | +0.003 |
Sweet | +0.003 |
ending | +0.003 |
mistake | +0.003 |
Neg logit contributions | |
ay | -0.003 |
ong | -0.003 |
rac | -0.003 |
around | -0.003 |
issors | -0.002 |
Token | day |
Token ID | 1110 |
Feature activation | +4.22e-01 |
Pos logit contributions | |
ay | +0.046 |
around | +0.043 |
ong | +0.041 |
rac | +0.040 |
ars | +0.038 |
Neg logit contributions | |
Sweet | -0.042 |
polar | -0.041 |
department | -0.041 |
Sleep | -0.041 |
Kids | -0.039 |
Token | the |
Token ID | 262 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
mistake | +0.055 |
Sleep | +0.054 |
polar | +0.053 |
Sweet | +0.053 |
moment | +0.053 |
Neg logit contributions | |
ay | -0.048 |
ong | -0.046 |
rac | -0.045 |
issors | -0.043 |
ars | -0.042 |
Token | village |
Token ID | 7404 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
ay | +0.004 |
around | +0.004 |
ong | +0.004 |
ars | +0.004 |
at | +0.004 |
Neg logit contributions | |
polar | -0.004 |
Sleep | -0.004 |
Kids | -0.004 |
Sweet | -0.004 |
department | -0.004 |
Token | back |
Token ID | 736 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
Sleep | +0.068 |
polar | +0.067 |
Sweet | +0.067 |
ending | +0.067 |
mistake | +0.066 |
Neg logit contributions | |
ay | -0.055 |
ong | -0.051 |
rac | -0.050 |
around | -0.048 |
issors | -0.047 |
Token | to |
Token ID | 284 |
Feature activation | -2.74e-01 |
Pos logit contributions | |
Sleep | +0.076 |
polar | +0.075 |
Sweet | +0.075 |
ending | +0.074 |
mistake | +0.073 |
Neg logit contributions | |
ay | -0.054 |
ong | -0.051 |
rac | -0.049 |
around | -0.048 |
issors | -0.046 |
Token | its |
Token ID | 663 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
Sleep | +0.010 |
Sweet | +0.010 |
polar | +0.010 |
ending | +0.009 |
Kids | +0.009 |
Neg logit contributions | |
ay | -0.009 |
ong | -0.008 |
rac | -0.008 |
around | -0.008 |
issors | -0.007 |
Token | former |
Token ID | 1966 |
Feature activation | -7.31e+00 |
Pos logit contributions | |
ay | +0.006 |
around | +0.006 |
ong | +0.006 |
rac | +0.005 |
bs | +0.005 |
Neg logit contributions | |
Kids | -0.004 |
Sleep | -0.004 |
Sweet | -0.004 |
polar | -0.004 |
Are | -0.004 |
Token | glory |
Token ID | 13476 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
mistake | +0.189 |
ending | +0.184 |
former | +0.183 |
polar | +0.175 |
Sleep | +0.173 |
Neg logit contributions | |
ay | -0.304 |
ong | -0.297 |
rac | -0.286 |
issors | -0.281 |
bs | -0.279 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
ay | +0.058 |
ong | +0.055 |
rac | +0.053 |
issors | +0.052 |
around | +0.050 |
Neg logit contributions | |
Sleep | -0.069 |
polar | -0.068 |
Sweet | -0.068 |
ending | -0.068 |
mistake | -0.067 |
Token | |
Token ID | 220 |
Feature activation | +7.95e-01 |
Pos logit contributions | |
ay | +0.005 |
ong | +0.004 |
rac | +0.004 |
around | +0.004 |
issors | +0.004 |
Neg logit contributions | |
Sleep | -0.005 |
ending | -0.005 |
polar | -0.005 |
Sweet | -0.005 |
mistake | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
ay | +0.032 |
ong | +0.030 |
rac | +0.030 |
around | +0.029 |
issors | +0.029 |
Neg logit contributions | |
Sleep | -0.021 |
Sweet | -0.021 |
polar | -0.021 |
mistake | -0.020 |
ending | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | +6.07e-02 |
Pos logit contributions | |
ay | +0.004 |
ong | +0.004 |
rac | +0.004 |
around | +0.004 |
ars | +0.004 |
Neg logit contributions | |
ending | -0.005 |
polar | -0.004 |
Sleep | -0.004 |
mistake | -0.004 |
funeral | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | +6.90e-03 |
Pos logit contributions | |
polar | +0.055 |
mistake | +0.053 |
Sleep | +0.053 |
moment | +0.052 |
ending | +0.052 |
Neg logit contributions | |
ay | -0.050 |
ong | -0.048 |
rac | -0.047 |
issors | -0.045 |
bs | -0.044 |
Token | tree |
Token ID | 5509 |
Feature activation | -6.38e-01 |
Pos logit contributions | |
ay | +0.000 |
around | +0.000 |
ong | +0.000 |
ars | +0.000 |
rac | +0.000 |
Neg logit contributions | |
Kids | -0.000 |
Sleep | -0.000 |
polar | -0.000 |
Sweet | -0.000 |
department | -0.000 |
Token | to |
Token ID | 284 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
Sleep | +0.024 |
polar | +0.024 |
ending | +0.024 |
Sweet | +0.024 |
mistake | +0.024 |
Neg logit contributions | |
ay | -0.018 |
ong | -0.017 |
rac | -0.016 |
issors | -0.015 |
around | -0.015 |
Token | its |
Token ID | 663 |
Feature activation | -5.34e-02 |
Pos logit contributions | |
Sleep | +0.011 |
Sweet | +0.011 |
polar | +0.011 |
ending | +0.011 |
Kids | +0.010 |
Neg logit contributions | |
ay | -0.009 |
ong | -0.008 |
around | -0.008 |
rac | -0.008 |
issors | -0.008 |
Token | once |
Token ID | 1752 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
Sleep | +0.002 |
Sweet | +0.002 |
Kids | +0.002 |
polar | +0.001 |
ending | +0.001 |
Neg logit contributions | |
ay | -0.002 |
ong | -0.002 |
around | -0.002 |
rac | -0.002 |
issors | -0.002 |
Token | former |
Token ID | 1966 |
Feature activation | -7.21e+00 |
Pos logit contributions | |
Kids | +0.048 |
Sweet | +0.046 |
Sleep | +0.045 |
polar | +0.045 |
divorce | +0.044 |
Neg logit contributions | |
around | -0.046 |
ay | -0.046 |
ong | -0.042 |
at | -0.040 |
rac | -0.040 |
Token | beauty |
Token ID | 8737 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
former | +0.188 |
mistake | +0.180 |
ending | +0.177 |
moment | +0.158 |
moral | +0.158 |
Neg logit contributions | |
ay | -0.309 |
ong | -0.306 |
issors | -0.297 |
rac | -0.296 |
bs | -0.292 |
Token | . |
Token ID | 13 |
Feature activation | -5.87e-01 |
Pos logit contributions | |
ay | +0.035 |
rac | +0.034 |
ong | +0.033 |
issors | +0.033 |
bs | +0.030 |
Neg logit contributions | |
Sleep | -0.047 |
ending | -0.046 |
polar | -0.046 |
moment | -0.046 |
Sweet | -0.045 |
Token | He |
Token ID | 679 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
Sleep | +0.022 |
mistake | +0.022 |
ending | +0.022 |
polar | +0.022 |
Sweet | +0.022 |
Neg logit contributions | |
ay | -0.018 |
ong | -0.015 |
rac | -0.015 |
around | -0.015 |
issors | -0.014 |
Token | was |
Token ID | 373 |
Feature activation | -4.27e-01 |
Pos logit contributions | |
ay | +0.010 |
ong | +0.009 |
rac | +0.009 |
around | +0.009 |
bs | +0.009 |
Neg logit contributions | |
Sleep | -0.007 |
polar | -0.007 |
Sweet | -0.007 |
ending | -0.007 |
mistake | -0.006 |
Token | happy |
Token ID | 3772 |
Feature activation | -6.99e-02 |
Pos logit contributions | |
polar | +0.015 |
Sleep | +0.015 |
ending | +0.015 |
Sweet | +0.015 |
mistake | +0.014 |
Neg logit contributions | |
ay | -0.014 |
ong | -0.013 |
rac | -0.013 |
issors | -0.012 |
ors | -0.012 |
Token | back |
Token ID | 736 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
Sleep | +0.001 |
polar | +0.001 |
ending | +0.001 |
Sweet | +0.001 |
mistake | +0.001 |
Neg logit contributions | |
ay | -0.001 |
ong | -0.001 |
rac | -0.001 |
issors | -0.001 |
ars | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | +8.50e-02 |
Pos logit contributions | |
Sleep | +0.073 |
ending | +0.073 |
polar | +0.072 |
Sweet | +0.072 |
mistake | +0.071 |
Neg logit contributions | |
ay | -0.057 |
around | -0.053 |
ong | -0.052 |
rac | -0.051 |
bs | -0.049 |
Token | village |
Token ID | 7404 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
ay | +0.003 |
around | +0.003 |
ars | +0.002 |
ong | +0.002 |
high | +0.002 |
Neg logit contributions | |
polar | -0.003 |
Sleep | -0.003 |
Sweet | -0.003 |
Kids | -0.003 |
fly | -0.003 |
Token | to |
Token ID | 284 |
Feature activation | -2.58e-01 |
Pos logit contributions | |
Sleep | +0.067 |
polar | +0.066 |
Sweet | +0.066 |
ending | +0.066 |
mistake | +0.065 |
Neg logit contributions | |
ay | -0.050 |
ong | -0.046 |
rac | -0.045 |
around | -0.043 |
issors | -0.042 |
Token | its |
Token ID | 663 |
Feature activation | +8.79e-02 |
Pos logit contributions | |
Sleep | +0.010 |
Sweet | +0.010 |
polar | +0.010 |
ending | +0.009 |
Kids | +0.009 |
Neg logit contributions | |
ay | -0.007 |
ong | -0.007 |
rac | -0.007 |
around | -0.007 |
issors | -0.006 |
Token | former |
Token ID | 1966 |
Feature activation | -7.20e+00 |
Pos logit contributions | |
ay | +0.003 |
ong | +0.003 |
around | +0.003 |
rac | +0.003 |
issors | +0.003 |
Neg logit contributions | |
Sleep | -0.003 |
Sweet | -0.002 |
Kids | -0.002 |
polar | -0.002 |
ending | -0.002 |
Token | state |
Token ID | 1181 |
Feature activation | -7.17e-01 |
Pos logit contributions | |
mistake | +0.190 |
ending | +0.188 |
former | +0.185 |
polar | +0.180 |
Sleep | +0.178 |
Neg logit contributions | |
ay | -0.293 |
ong | -0.288 |
rac | -0.276 |
issors | -0.272 |
bs | -0.270 |
Token | . |
Token ID | 13 |
Feature activation | -4.00e-01 |
Pos logit contributions | |
Sleep | +0.027 |
ending | +0.027 |
polar | +0.026 |
mistake | +0.026 |
Sweet | +0.026 |
Neg logit contributions | |
ay | -0.021 |
ong | -0.020 |
rac | -0.019 |
issors | -0.019 |
bs | -0.018 |
Token | With |
Token ID | 2080 |
Feature activation | +2.56e-01 |
Pos logit contributions | |
mistake | +0.016 |
polar | +0.016 |
ending | +0.016 |
Sleep | +0.015 |
divorce | +0.015 |
Neg logit contributions | |
ay | -0.012 |
rac | -0.010 |
around | -0.010 |
ong | -0.010 |
issors | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | +7.21e-02 |
Pos logit contributions | |
ong | +0.010 |
ay | +0.009 |
ors | +0.009 |
issors | +0.009 |
rac | +0.009 |
Neg logit contributions | |
ending | -0.008 |
polar | -0.008 |
Sleep | -0.007 |
moral | -0.007 |
forgiveness | -0.007 |
Token | help |
Token ID | 1037 |
Feature activation | -8.78e-01 |
Pos logit contributions | |
ay | +0.002 |
ong | +0.002 |
rac | +0.002 |
around | +0.002 |
ars | +0.002 |
Neg logit contributions | |
Sleep | -0.002 |
Sweet | -0.002 |
polar | -0.002 |
Kids | -0.002 |
ending | -0.002 |
Token | to |
Token ID | 284 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
moment | +0.051 |
Sleep | +0.050 |
ending | +0.047 |
department | +0.047 |
mistake | +0.047 |
Neg logit contributions | |
ay | -0.048 |
ong | -0.047 |
rac | -0.046 |
itching | -0.045 |
iddling | -0.044 |
Token | be |
Token ID | 307 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
Sweet | +0.008 |
Sleep | +0.008 |
polar | +0.008 |
ending | +0.008 |
Kids | +0.008 |
Neg logit contributions | |
ay | -0.009 |
ong | -0.008 |
rac | -0.008 |
around | -0.008 |
issors | -0.008 |
Token | restored |
Token ID | 15032 |
Feature activation | -2.81e-01 |
Pos logit contributions | |
Sleep | +0.071 |
polar | +0.071 |
mistake | +0.070 |
ending | +0.069 |
Sweet | +0.068 |
Neg logit contributions | |
ay | -0.076 |
ong | -0.071 |
rac | -0.069 |
issors | -0.066 |
around | -0.064 |
Token | to |
Token ID | 284 |
Feature activation | -2.63e-01 |
Pos logit contributions | |
Sleep | +0.009 |
moment | +0.009 |
mistake | +0.009 |
polar | +0.009 |
ending | +0.009 |
Neg logit contributions | |
ay | -0.009 |
ong | -0.009 |
rac | -0.009 |
issors | -0.008 |
ars | -0.008 |
Token | its |
Token ID | 663 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
Sweet | +0.010 |
Sleep | +0.010 |
Kids | +0.010 |
polar | +0.009 |
department | +0.009 |
Neg logit contributions | |
ay | -0.008 |
ong | -0.008 |
around | -0.007 |
rac | -0.007 |
issors | -0.007 |
Token | former |
Token ID | 1966 |
Feature activation | -7.20e+00 |
Pos logit contributions | |
ay | +0.004 |
around | +0.004 |
ong | +0.004 |
rac | +0.004 |
at | +0.004 |
Neg logit contributions | |
Kids | -0.003 |
Sleep | -0.003 |
Sweet | -0.003 |
polar | -0.003 |
Are | -0.003 |
Token | glory |
Token ID | 13476 |
Feature activation | +1.94e+00 |
Pos logit contributions | |
former | +0.187 |
mistake | +0.185 |
ending | +0.179 |
moral | +0.164 |
moment | +0.162 |
Neg logit contributions | |
ay | -0.306 |
ong | -0.304 |
rac | -0.290 |
bs | -0.287 |
issors | -0.283 |
Token | ." |
Token ID | 526 |
Feature activation | -4.75e-01 |
Pos logit contributions | |
ay | +0.064 |
ong | +0.060 |
rac | +0.059 |
issors | +0.057 |
bs | +0.055 |
Neg logit contributions | |
Sleep | -0.065 |
moment | -0.063 |
polar | -0.063 |
Sweet | -0.063 |
mistake | -0.062 |
Token | |
Token ID | 220 |
Feature activation | +9.82e-01 |
Pos logit contributions | |
Sleep | +0.017 |
ending | +0.016 |
polar | +0.016 |
Sweet | +0.016 |
mistake | +0.016 |
Neg logit contributions | |
ay | -0.016 |
ong | -0.014 |
around | -0.014 |
rac | -0.014 |
bs | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
ay | +0.040 |
ong | +0.039 |
rac | +0.038 |
issors | +0.037 |
around | +0.036 |
Neg logit contributions | |
Sleep | -0.025 |
Sweet | -0.025 |
polar | -0.024 |
mistake | -0.024 |
moment | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | +8.21e-02 |
Pos logit contributions | |
ay | +0.005 |
ong | +0.005 |
rac | +0.005 |
around | +0.005 |
issors | +0.005 |
Neg logit contributions | |
Sleep | -0.005 |
ending | -0.005 |
polar | -0.005 |
Sweet | -0.005 |
mistake | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | +3.59e-01 |
Pos logit contributions | |
Sleep | +0.055 |
Sweet | +0.055 |
polar | +0.054 |
ending | +0.054 |
Kids | +0.053 |
Neg logit contributions | |
ay | -0.033 |
ong | -0.031 |
around | -0.029 |
rac | -0.029 |
issors | -0.027 |
Token | leaving |
Token ID | 4305 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
ay | +0.010 |
around | +0.010 |
ong | +0.010 |
rac | +0.009 |
bs | +0.009 |
Neg logit contributions | |
Sweet | -0.014 |
Sleep | -0.014 |
Kids | -0.013 |
polar | -0.013 |
ending | -0.013 |
Token | it |
Token ID | 340 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
Sleep | +0.038 |
Sweet | +0.038 |
polar | +0.038 |
ending | +0.037 |
mistake | +0.037 |
Neg logit contributions | |
ay | -0.030 |
ong | -0.028 |
rac | -0.027 |
around | -0.027 |
issors | -0.026 |
Token | to |
Token ID | 284 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
Sweet | +0.001 |
Kids | +0.001 |
Sleep | +0.001 |
polar | +0.001 |
ending | +0.001 |
Neg logit contributions | |
around | -0.001 |
ay | -0.001 |
ong | -0.001 |
rac | -0.001 |
bs | -0.000 |
Token | its |
Token ID | 663 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
Sweet | +0.007 |
Kids | +0.007 |
Sleep | +0.007 |
department | +0.006 |
Jenkins | +0.006 |
Neg logit contributions | |
around | -0.004 |
high | -0.004 |
ong | -0.004 |
ay | -0.003 |
at | -0.003 |
Token | former |
Token ID | 1966 |
Feature activation | -7.18e+00 |
Pos logit contributions | |
around | +0.009 |
ay | +0.009 |
ong | +0.009 |
at | +0.008 |
rac | +0.008 |
Neg logit contributions | |
Kids | -0.008 |
Sweet | -0.008 |
Sleep | -0.008 |
Are | -0.007 |
divorce | -0.007 |
Token | state |
Token ID | 1181 |
Feature activation | -5.98e-01 |
Pos logit contributions | |
mistake | +0.195 |
ending | +0.193 |
polar | +0.192 |
former | +0.188 |
Sleep | +0.183 |
Neg logit contributions | |
ay | -0.286 |
ong | -0.276 |
rac | -0.269 |
issors | -0.262 |
bs | -0.260 |
Token | . |
Token ID | 13 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
Sleep | +0.024 |
Sweet | +0.024 |
polar | +0.024 |
ending | +0.024 |
mistake | +0.024 |
Neg logit contributions | |
ay | -0.016 |
ong | -0.015 |
rac | -0.014 |
around | -0.014 |
issors | -0.013 |
Token | |
Token ID | 220 |
Feature activation | +9.60e-01 |
Pos logit contributions | |
ending | +0.008 |
mistake | +0.008 |
polar | +0.008 |
Sleep | +0.008 |
Sweet | +0.007 |
Neg logit contributions | |
ay | -0.006 |
rac | -0.005 |
around | -0.005 |
ong | -0.005 |
issors | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
ay | +0.037 |
ong | +0.035 |
rac | +0.034 |
around | +0.033 |
issors | +0.033 |
Neg logit contributions | |
Sleep | -0.027 |
Sweet | -0.027 |
polar | -0.027 |
ending | -0.026 |
mistake | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
ay | +0.007 |
ong | +0.006 |
rac | +0.006 |
around | +0.006 |
ars | +0.006 |
Neg logit contributions | |
ending | -0.007 |
polar | -0.007 |
Sleep | -0.007 |
mistake | -0.007 |
Sweet | -0.006 |
Token | aunt |
Token ID | 25949 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
ay | +0.038 |
ong | +0.033 |
around | +0.033 |
rac | +0.032 |
bs | +0.031 |
Neg logit contributions | |
Sleep | -0.027 |
Sweet | -0.025 |
Kids | -0.024 |
polar | -0.024 |
Are | -0.024 |
Token | told |
Token ID | 1297 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
ay | +0.007 |
around | +0.007 |
ong | +0.007 |
at | +0.006 |
bs | +0.006 |
Neg logit contributions | |
Sleep | -0.011 |
former | -0.011 |
Sweet | -0.011 |
ending | -0.011 |
mistake | -0.010 |
Token | Lucy |
Token ID | 22162 |
Feature activation | -7.73e-01 |
Pos logit contributions | |
polar | +0.057 |
Sleep | +0.056 |
Sweet | +0.056 |
former | +0.055 |
ending | +0.055 |
Neg logit contributions | |
ay | -0.025 |
around | -0.024 |
ong | -0.023 |
rac | -0.021 |
bs | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
polar | +0.036 |
Sleep | +0.035 |
Sweet | +0.035 |
Kids | +0.034 |
former | +0.034 |
Neg logit contributions | |
around | -0.016 |
ay | -0.016 |
ong | -0.015 |
bs | -0.014 |
rac | -0.014 |
Token | " |
Token ID | 366 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
around | +0.004 |
ay | +0.004 |
rac | +0.004 |
at | +0.004 |
ong | +0.004 |
Neg logit contributions | |
polar | -0.007 |
former | -0.007 |
ending | -0.007 |
Sleep | -0.007 |
mistake | -0.007 |
Token | Sweet |
Token ID | 36087 |
Feature activation | -7.16e+00 |
Pos logit contributions | |
around | +0.040 |
ay | +0.039 |
rac | +0.039 |
ized | +0.037 |
bs | +0.036 |
Neg logit contributions | |
polar | -0.048 |
ending | -0.046 |
mistake | -0.043 |
ig | -0.043 |
island | -0.043 |
Token | heart |
Token ID | 11499 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
Sleep | +0.280 |
department | +0.280 |
Are | +0.279 |
charity | +0.277 |
Sweet | +0.275 |
Neg logit contributions | |
ay | -0.197 |
around | -0.176 |
at | -0.175 |
- | -0.174 |
ors | -0.164 |
Token | , |
Token ID | 11 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
ay | +0.052 |
ong | +0.048 |
rac | +0.046 |
around | +0.044 |
issors | +0.043 |
Neg logit contributions | |
Sleep | -0.103 |
polar | -0.103 |
Sweet | -0.102 |
ending | -0.102 |
mistake | -0.101 |
Token | you |
Token ID | 345 |
Feature activation | -5.77e-01 |
Pos logit contributions | |
around | +0.006 |
ay | +0.006 |
at | +0.006 |
ong | +0.005 |
rac | +0.005 |
Neg logit contributions | |
polar | -0.008 |
Sweet | -0.008 |
former | -0.008 |
Sleep | -0.008 |
Kids | -0.008 |
Token | know |
Token ID | 760 |
Feature activation | +7.15e-01 |
Pos logit contributions | |
Sweet | +0.020 |
Sleep | +0.020 |
polar | +0.019 |
ending | +0.019 |
moment | +0.019 |
Neg logit contributions | |
ay | -0.019 |
ong | -0.018 |
around | -0.017 |
rac | -0.017 |
bs | -0.016 |
Token | that |
Token ID | 326 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
ay | +0.021 |
ong | +0.021 |
around | +0.020 |
rac | +0.020 |
ors | +0.018 |
Neg logit contributions | |
polar | -0.027 |
Sweet | -0.026 |
Sleep | -0.026 |
ending | -0.026 |
Kids | -0.025 |
Token | very |
Token ID | 845 |
Feature activation | +8.29e-01 |
Pos logit contributions | |
polar | +0.007 |
Sleep | +0.007 |
ending | +0.007 |
Sweet | +0.007 |
mistake | +0.007 |
Neg logit contributions | |
ay | -0.007 |
ong | -0.007 |
rac | -0.007 |
issors | -0.007 |
ors | -0.006 |
Token | tired |
Token ID | 10032 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
ay | +0.033 |
ors | +0.030 |
ong | +0.029 |
rac | +0.029 |
issors | +0.029 |
Neg logit contributions | |
polar | -0.026 |
ending | -0.025 |
mistake | -0.025 |
Sleep | -0.025 |
moment | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | -5.80e-01 |
Pos logit contributions | |
ay | +0.033 |
ong | +0.031 |
rac | +0.030 |
around | +0.028 |
issors | +0.028 |
Neg logit contributions | |
Sleep | -0.049 |
polar | -0.048 |
Sweet | -0.048 |
ending | -0.048 |
mistake | -0.047 |
Token | The |
Token ID | 383 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
Sleep | +0.021 |
polar | +0.021 |
Sweet | +0.021 |
ending | +0.021 |
mistake | +0.020 |
Neg logit contributions | |
ay | -0.018 |
ong | -0.017 |
rac | -0.016 |
around | -0.016 |
issors | -0.016 |
Token | ash |
Token ID | 12530 |
Feature activation | -5.11e+00 |
Pos logit contributions | |
fly | +0.070 |
Sweet | +0.067 |
former | +0.066 |
Sleep | +0.064 |
days | +0.064 |
Neg logit contributions | |
ay | -0.044 |
around | -0.042 |
ong | -0.040 |
high | -0.038 |
at | -0.037 |
Token | tr |
Token ID | 2213 |
Feature activation | -7.15e+00 |
Pos logit contributions | |
Sleep | +0.139 |
Sweet | +0.137 |
polar | +0.137 |
ending | +0.132 |
mistake | +0.132 |
Neg logit contributions | |
ay | -0.204 |
ong | -0.194 |
rac | -0.190 |
around | -0.188 |
issors | -0.181 |
Token | ay |
Token ID | 323 |
Feature activation | +4.62e+00 |
Pos logit contributions | |
department | +0.377 |
Sweet | +0.376 |
tr | +0.375 |
gen | +0.366 |
polar | +0.364 |
Neg logit contributions | |
ars | -0.050 |
issors | -0.049 |
rac | -0.049 |
ong | -0.047 |
around | -0.043 |
Token | watched |
Token ID | 7342 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
ay | +0.162 |
around | +0.161 |
ong | +0.154 |
led | +0.148 |
rac | +0.147 |
Neg logit contributions | |
charity | -0.150 |
Sweet | -0.149 |
Sleep | -0.147 |
mistake | -0.146 |
former | -0.145 |
Token | as |
Token ID | 355 |
Feature activation | +4.69e+00 |
Pos logit contributions | |
polar | +0.063 |
ending | +0.063 |
Sleep | +0.061 |
mistake | +0.061 |
moment | +0.060 |
Neg logit contributions | |
ay | -0.047 |
rac | -0.045 |
issors | -0.044 |
ars | -0.043 |
bs | -0.043 |
Token | the |
Token ID | 262 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
ay | +0.118 |
ong | +0.113 |
around | +0.107 |
rac | +0.105 |
bs | +0.097 |
Neg logit contributions | |
Sleep | -0.195 |
Sweet | -0.193 |
polar | -0.191 |
ending | -0.189 |
Kids | -0.188 |
Token | little |
Token ID | 1310 |
Feature activation | -2.30e+00 |
Pos logit contributions | |
ay | +0.005 |
ong | +0.004 |
past | +0.004 |
at | +0.004 |
around | +0.004 |
Neg logit contributions | |
sometime | -0.007 |
Sleep | -0.007 |
former | -0.007 |
fly | -0.007 |
| | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -5.59e-01 |
Pos logit contributions | |
ay | +0.012 |
ong | +0.011 |
rac | +0.011 |
around | +0.010 |
issors | +0.010 |
Neg logit contributions | |
Sleep | -0.019 |
polar | -0.019 |
Sweet | -0.019 |
ending | -0.019 |
mistake | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | -3.31e-02 |
Pos logit contributions | |
mistake | +0.022 |
ending | +0.022 |
polar | +0.022 |
Sleep | +0.022 |
Sweet | +0.022 |
Neg logit contributions | |
ay | -0.016 |
ong | -0.014 |
around | -0.014 |
rac | -0.014 |
issors | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
Sleep | +0.001 |
Sweet | +0.001 |
polar | +0.001 |
moment | +0.001 |
Kids | +0.001 |
Neg logit contributions | |
ay | -0.001 |
ong | -0.001 |
rac | -0.001 |
issors | -0.001 |
around | -0.001 |
Token | The |
Token ID | 464 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
Sleep | +0.004 |
polar | +0.004 |
Sweet | +0.004 |
ending | +0.004 |
mistake | +0.004 |
Neg logit contributions | |
ay | -0.004 |
ong | -0.004 |
rac | -0.004 |
around | -0.003 |
issors | -0.003 |
Token | ash |
Token ID | 12530 |
Feature activation | -5.14e+00 |
Pos logit contributions | |
sometime | +0.082 |
burn | +0.081 |
former | +0.081 |
Sleep | +0.081 |
fly | +0.080 |
Neg logit contributions | |
at | -0.039 |
past | -0.039 |
ay | -0.038 |
around | -0.037 |
high | -0.036 |
Token | tr |
Token ID | 2213 |
Feature activation | -7.15e+00 |
Pos logit contributions | |
Sleep | +0.076 |
Sweet | +0.074 |
polar | +0.073 |
mistake | +0.069 |
ending | +0.067 |
Neg logit contributions | |
ay | -0.271 |
ong | -0.261 |
around | -0.256 |
rac | -0.256 |
issors | -0.246 |
Token | ay |
Token ID | 323 |
Feature activation | +4.57e+00 |
Pos logit contributions | |
department | +0.397 |
Sweet | +0.395 |
tr | +0.386 |
gen | +0.386 |
polar | +0.386 |
Neg logit contributions | |
ong | -0.039 |
issors | -0.039 |
rac | -0.037 |
ars | -0.036 |
around | -0.035 |
Token | was |
Token ID | 373 |
Feature activation | -3.62e-01 |
Pos logit contributions | |
around | +0.148 |
ay | +0.147 |
ong | +0.143 |
led | +0.139 |
at | +0.137 |
Neg logit contributions | |
charity | -0.164 |
Sleep | -0.161 |
Sweet | -0.161 |
former | -0.159 |
mistake | -0.156 |
Token | ready |
Token ID | 3492 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
polar | +0.012 |
Sleep | +0.012 |
ending | +0.012 |
mistake | +0.011 |
Sweet | +0.011 |
Neg logit contributions | |
ay | -0.013 |
ong | -0.012 |
rac | -0.012 |
issors | -0.011 |
bs | -0.011 |
Token | to |
Token ID | 284 |
Feature activation | -3.80e-01 |
Pos logit contributions | |
ay | +0.036 |
rac | +0.034 |
ong | +0.034 |
issors | +0.033 |
ars | +0.032 |
Neg logit contributions | |
ending | -0.046 |
moment | -0.045 |
Sleep | -0.044 |
mistake | -0.044 |
polar | -0.043 |
Token | provide |
Token ID | 2148 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
Sweet | +0.013 |
Sleep | +0.012 |
Kids | +0.012 |
ending | +0.012 |
polar | +0.012 |
Neg logit contributions | |
ay | -0.013 |
ong | -0.013 |
rac | -0.012 |
around | -0.012 |
ars | -0.011 |
Token | backyard |
Token ID | 24296 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
ay | +0.012 |
ong | +0.011 |
rac | +0.011 |
around | +0.011 |
issors | +0.010 |
Neg logit contributions | |
Sleep | -0.006 |
Sweet | -0.006 |
polar | -0.006 |
ending | -0.006 |
mistake | -0.006 |
Token | there |
Token ID | 612 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
Sleep | +0.009 |
Sweet | +0.009 |
polar | +0.009 |
Kids | +0.009 |
former | +0.009 |
Neg logit contributions | |
ay | -0.004 |
around | -0.004 |
ong | -0.004 |
rac | -0.004 |
at | -0.003 |
Token | was |
Token ID | 373 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
Sweet | +0.003 |
Sleep | +0.003 |
polar | +0.003 |
mistake | +0.003 |
divorce | +0.003 |
Neg logit contributions | |
ay | -0.005 |
ong | -0.004 |
around | -0.004 |
rac | -0.004 |
at | -0.004 |
Token | an |
Token ID | 281 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
mistake | +0.011 |
Kids | +0.011 |
fly | +0.011 |
Sleep | +0.011 |
forgiveness | +0.010 |
Neg logit contributions | |
around | -0.005 |
high | -0.005 |
ay | -0.005 |
at | -0.004 |
ong | -0.004 |
Token | ash |
Token ID | 12530 |
Feature activation | -5.05e+00 |
Pos logit contributions | |
ay | +0.003 |
ong | +0.003 |
around | +0.003 |
rac | +0.003 |
issors | +0.003 |
Neg logit contributions | |
Sleep | -0.004 |
Sweet | -0.004 |
moment | -0.003 |
Kids | -0.003 |
ending | -0.003 |
Token | tr |
Token ID | 2213 |
Feature activation | -7.13e+00 |
Pos logit contributions | |
Sleep | +0.051 |
polar | +0.051 |
ending | +0.050 |
Sweet | +0.049 |
mistake | +0.046 |
Neg logit contributions | |
ay | -0.285 |
ong | -0.275 |
rac | -0.272 |
around | -0.266 |
issors | -0.264 |
Token | ay |
Token ID | 323 |
Feature activation | +4.64e+00 |
Pos logit contributions | |
tr | +0.378 |
department | +0.370 |
polar | +0.361 |
gen | +0.360 |
tunnel | +0.358 |
Neg logit contributions | |
ars | -0.066 |
bs | -0.061 |
rac | -0.057 |
issors | -0.055 |
around | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | -5.93e-01 |
Pos logit contributions | |
ay | +0.119 |
ong | +0.111 |
rac | +0.104 |
around | +0.104 |
issors | +0.099 |
Neg logit contributions | |
Sleep | -0.193 |
Sweet | -0.192 |
polar | -0.192 |
ending | -0.189 |
mistake | -0.187 |
Token | It |
Token ID | 632 |
Feature activation | -8.64e-01 |
Pos logit contributions | |
Sleep | +0.022 |
polar | +0.021 |
Sweet | +0.021 |
ending | +0.021 |
moment | +0.021 |
Neg logit contributions | |
ay | -0.018 |
ong | -0.017 |
rac | -0.016 |
around | -0.016 |
issors | -0.016 |
Token | was |
Token ID | 373 |
Feature activation | -2.46e-01 |
Pos logit contributions | |
Sleep | +0.021 |
Sweet | +0.021 |
polar | +0.021 |
ending | +0.021 |
mistake | +0.020 |
Neg logit contributions | |
ay | -0.037 |
ong | -0.035 |
rac | -0.035 |
around | -0.034 |
issors | -0.033 |
Token | the |
Token ID | 262 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
polar | +0.009 |
Sleep | +0.009 |
ending | +0.009 |
Sweet | +0.009 |
mistake | +0.009 |
Neg logit contributions | |
ay | -0.008 |
ong | -0.007 |
rac | -0.007 |
issors | -0.007 |
around | -0.007 |
Token | � |
Token ID | 6184 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
ay | +0.009 |
ong | +0.009 |
rac | +0.009 |
around | +0.009 |
issors | +0.008 |
Neg logit contributions | |
polar | -0.011 |
Sleep | -0.011 |
ending | -0.010 |
Sweet | -0.010 |
mistake | -0.010 |
Token | � |
Token ID | 95 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
Sleep | +0.049 |
mistake | +0.047 |
ending | +0.047 |
Sweet | +0.046 |
Kids | +0.046 |
Neg logit contributions | |
ay | -0.047 |
ong | -0.042 |
rac | -0.041 |
around | -0.040 |
issors | -0.039 |
Token | € |
Token ID | 26391 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
Sleep | +0.055 |
Sweet | +0.053 |
ending | +0.052 |
polar | +0.051 |
mistake | +0.051 |
Neg logit contributions | |
ay | -0.059 |
ong | -0.056 |
rac | -0.054 |
around | -0.053 |
bs | -0.052 |
Token | � |
Token ID | 129 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
Sweet | +0.065 |
moment | +0.064 |
Bird | +0.063 |
Sleep | +0.061 |
ending | +0.061 |
Neg logit contributions | |
ong | -0.086 |
ay | -0.081 |
issors | -0.078 |
ag | -0.077 |
around | -0.076 |
Token | � |
Token ID | 241 |
Feature activation | +2.32e+00 |
Pos logit contributions | |
mistake | +0.057 |
moral | +0.052 |
Sleep | +0.052 |
former | +0.048 |
ending | +0.048 |
Neg logit contributions | |
ay | -0.052 |
bs | -0.044 |
ong | -0.040 |
ons | -0.040 |
s | -0.040 |
Token | Sweet |
Token ID | 36087 |
Feature activation | -7.10e+00 |
Pos logit contributions | |
ay | +0.080 |
around | +0.074 |
ong | +0.072 |
bs | +0.072 |
rac | +0.070 |
Neg logit contributions | |
polar | -0.077 |
department | -0.074 |
ending | -0.074 |
mistake | -0.072 |
island | -0.071 |
Token | ie |
Token ID | 494 |
Feature activation | +8.29e-01 |
Pos logit contributions | |
department | +0.333 |
Sleep | +0.326 |
charity | +0.326 |
divorce | +0.325 |
Sweet | +0.323 |
Neg logit contributions | |
ay | -0.151 |
ong | -0.118 |
around | -0.118 |
- | -0.116 |
bs | -0.111 |
Token | , |
Token ID | 11 |
Feature activation | +2.97e-01 |
Pos logit contributions | |
ay | +0.022 |
ong | +0.021 |
rac | +0.020 |
issors | +0.019 |
around | +0.019 |
Neg logit contributions | |
polar | -0.033 |
Sleep | -0.033 |
ending | -0.033 |
Sweet | -0.033 |
mistake | -0.032 |
Token | why |
Token ID | 1521 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
ay | +0.008 |
around | +0.008 |
rac | +0.007 |
bs | +0.007 |
ong | +0.007 |
Neg logit contributions | |
polar | -0.012 |
Sleep | -0.011 |
Sweet | -0.011 |
ending | -0.011 |
department | -0.011 |
Token | don |
Token ID | 836 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
polar | +0.049 |
Sleep | +0.049 |
Sweet | +0.048 |
ending | +0.047 |
mistake | +0.046 |
Neg logit contributions | |
ay | -0.049 |
ong | -0.047 |
rac | -0.045 |
around | -0.044 |
issors | -0.043 |
Token | â |
Token ID | 22940 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
ong | +0.075 |
ay | +0.071 |
rac | +0.070 |
bs | +0.070 |
issors | +0.070 |
Neg logit contributions | |
moral | -0.090 |
Sleep | -0.089 |
Kids | -0.087 |
Sweet | -0.085 |
Are | -0.083 |
Token | room |
Token ID | 2119 |
Feature activation | -9.87e-02 |
Pos logit contributions | |
ay | +0.008 |
around | +0.008 |
ong | +0.007 |
rac | +0.007 |
at | +0.007 |
Neg logit contributions | |
Sleep | -0.004 |
ending | -0.004 |
Sweet | -0.004 |
forgiveness | -0.004 |
polar | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | -1.94e-01 |
Pos logit contributions | |
ending | +0.004 |
moment | +0.004 |
Sleep | +0.004 |
mistake | +0.004 |
Sweet | +0.004 |
Neg logit contributions | |
ay | -0.003 |
ong | -0.003 |
rac | -0.003 |
issors | -0.003 |
ars | -0.003 |
Token | saw |
Token ID | 2497 |
Feature activation | +4.16e-01 |
Pos logit contributions | |
Sleep | +0.007 |
polar | +0.006 |
Sweet | +0.006 |
ending | +0.006 |
mistake | +0.006 |
Neg logit contributions | |
ay | -0.006 |
ong | -0.006 |
rac | -0.006 |
around | -0.006 |
issors | -0.006 |
Token | an |
Token ID | 281 |
Feature activation | +7.89e-02 |
Pos logit contributions | |
ay | +0.013 |
around | +0.013 |
ong | +0.013 |
rac | +0.012 |
bs | +0.012 |
Neg logit contributions | |
Sleep | -0.015 |
ending | -0.014 |
Kids | -0.014 |
mistake | -0.014 |
polar | -0.014 |
Token | ash |
Token ID | 12530 |
Feature activation | -5.00e+00 |
Pos logit contributions | |
ay | +0.003 |
ong | +0.002 |
around | +0.002 |
rac | +0.002 |
issors | +0.002 |
Neg logit contributions | |
Sleep | -0.003 |
Sweet | -0.003 |
ending | -0.003 |
moment | -0.003 |
polar | -0.003 |
Token | tr |
Token ID | 2213 |
Feature activation | -7.07e+00 |
Pos logit contributions | |
Sleep | +0.073 |
Sweet | +0.071 |
polar | +0.065 |
Kids | +0.064 |
mistake | +0.060 |
Neg logit contributions | |
ay | -0.273 |
ong | -0.263 |
around | -0.259 |
rac | -0.250 |
issors | -0.246 |
Token | ay |
Token ID | 323 |
Feature activation | +4.66e+00 |
Pos logit contributions | |
department | +0.393 |
tr | +0.390 |
Sweet | +0.382 |
polar | +0.379 |
gen | +0.377 |
Neg logit contributions | |
ars | -0.048 |
rac | -0.048 |
bs | -0.044 |
ong | -0.040 |
issors | -0.038 |
Token | . |
Token ID | 13 |
Feature activation | -4.79e-01 |
Pos logit contributions | |
ay | +0.119 |
ong | +0.111 |
around | +0.110 |
ors | +0.098 |
rac | +0.096 |
Neg logit contributions | |
Sweet | -0.199 |
Sleep | -0.198 |
polar | -0.198 |
ending | -0.194 |
Kids | -0.192 |
Token | She |
Token ID | 1375 |
Feature activation | +5.03e-03 |
Pos logit contributions | |
Sleep | +0.017 |
Sweet | +0.017 |
polar | +0.017 |
ending | +0.017 |
mistake | +0.016 |
Neg logit contributions | |
ay | -0.015 |
ong | -0.014 |
rac | -0.014 |
around | -0.013 |
issors | -0.013 |
Token | remembered |
Token ID | 12086 |
Feature activation | -7.57e-01 |
Pos logit contributions | |
ay | +0.000 |
ong | +0.000 |
rac | +0.000 |
around | +0.000 |
issors | +0.000 |
Neg logit contributions | |
Sleep | -0.000 |
polar | -0.000 |
Sweet | -0.000 |
ending | -0.000 |
mistake | -0.000 |
Token | her |
Token ID | 607 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
Sleep | +0.029 |
polar | +0.029 |
Kids | +0.029 |
Sweet | +0.029 |
ending | +0.029 |
Neg logit contributions | |
around | -0.021 |
ong | -0.020 |
ay | -0.020 |
rac | -0.019 |
at | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | +2.58e-01 |
Pos logit contributions | |
Sleep | +0.013 |
polar | +0.013 |
Sweet | +0.013 |
ending | +0.013 |
mistake | +0.013 |
Neg logit contributions | |
ay | -0.009 |
rac | -0.008 |
ong | -0.008 |
around | -0.008 |
issors | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
ay | +0.009 |
ong | +0.009 |
rac | +0.008 |
around | +0.008 |
issors | +0.008 |
Neg logit contributions | |
Sleep | -0.008 |
ending | -0.008 |
polar | -0.008 |
Sweet | -0.008 |
mistake | -0.008 |
Token | The |
Token ID | 464 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
ay | +0.005 |
rac | +0.005 |
ong | +0.005 |
ars | +0.004 |
around | +0.004 |
Neg logit contributions | |
ending | -0.006 |
polar | -0.006 |
Sleep | -0.006 |
Sweet | -0.006 |
mistake | -0.006 |
Token | poor |
Token ID | 3595 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
Sleep | +0.062 |
Kids | +0.061 |
sometime | +0.061 |
| | +0.058 |
burn | +0.057 |
Neg logit contributions | |
ay | -0.061 |
past | -0.057 |
high | -0.055 |
around | -0.055 |
at | -0.053 |
Token | ash |
Token ID | 12530 |
Feature activation | -5.12e+00 |
Pos logit contributions | |
Sleep | +0.043 |
Kids | +0.040 |
| | +0.039 |
Sweet | +0.037 |
charity | +0.037 |
Neg logit contributions | |
- | -0.047 |
at | -0.046 |
ay | -0.046 |
around | -0.044 |
past | -0.043 |
Token | tr |
Token ID | 2213 |
Feature activation | -7.06e+00 |
Pos logit contributions | |
Sleep | +0.067 |
Sweet | +0.065 |
polar | +0.064 |
Kids | +0.058 |
ending | +0.058 |
Neg logit contributions | |
ay | -0.279 |
ong | -0.269 |
around | -0.263 |
rac | -0.263 |
issors | -0.254 |
Token | ay |
Token ID | 323 |
Feature activation | +4.57e+00 |
Pos logit contributions | |
department | +0.392 |
Sweet | +0.387 |
tr | +0.383 |
gen | +0.381 |
polar | +0.380 |
Neg logit contributions | |
issors | -0.042 |
rac | -0.041 |
around | -0.039 |
ong | -0.039 |
ars | -0.036 |
Token | was |
Token ID | 373 |
Feature activation | -4.38e-01 |
Pos logit contributions | |
ay | +0.175 |
ong | +0.165 |
around | +0.163 |
rac | +0.156 |
ars | +0.150 |
Neg logit contributions | |
Sleep | -0.141 |
Sweet | -0.139 |
former | -0.138 |
charity | -0.137 |
polar | -0.137 |
Token | broken |
Token ID | 5445 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
Sweet | +0.013 |
Sleep | +0.012 |
polar | +0.012 |
Kids | +0.012 |
ending | +0.012 |
Neg logit contributions | |
ay | -0.017 |
around | -0.016 |
ong | -0.016 |
rac | -0.016 |
issors | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | -2.75e-01 |
Pos logit contributions | |
Sleep | +0.090 |
Sweet | +0.089 |
polar | +0.089 |
ending | +0.088 |
mistake | +0.087 |
Neg logit contributions | |
ay | -0.057 |
ong | -0.053 |
rac | -0.051 |
around | -0.050 |
issors | -0.048 |
Token | it |
Token ID | 340 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
Sleep | +0.011 |
Sweet | +0.011 |
polar | +0.010 |
Kids | +0.010 |
ending | +0.010 |
Neg logit contributions | |
ay | -0.008 |
ong | -0.007 |
around | -0.007 |
rac | -0.007 |
bs | -0.007 |
Token | carefully |
Token ID | 7773 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
ay | +0.047 |
ong | +0.045 |
rac | +0.044 |
around | +0.043 |
issors | +0.042 |
Neg logit contributions | |
Sleep | -0.039 |
polar | -0.039 |
Sweet | -0.039 |
ending | -0.038 |
mistake | -0.038 |
Token | picked |
Token ID | 6497 |
Feature activation | -9.60e-01 |
Pos logit contributions | |
ay | +0.087 |
ong | +0.080 |
rac | +0.079 |
issors | +0.076 |
ars | +0.073 |
Neg logit contributions | |
polar | -0.065 |
Sleep | -0.063 |
mistake | -0.061 |
ending | -0.060 |
Sweet | -0.060 |
Token | up |
Token ID | 510 |
Feature activation | +8.16e-01 |
Pos logit contributions | |
moment | +0.030 |
polar | +0.030 |
mistake | +0.029 |
pear | +0.029 |
again | +0.028 |
Neg logit contributions | |
ay | -0.033 |
rac | -0.032 |
issors | -0.031 |
ong | -0.031 |
ars | -0.031 |
Token | the |
Token ID | 262 |
Feature activation | +2.54e-02 |
Pos logit contributions | |
ay | +0.025 |
ong | +0.024 |
rac | +0.023 |
issors | +0.022 |
ars | +0.022 |
Neg logit contributions | |
polar | -0.029 |
Sleep | -0.029 |
Sweet | -0.028 |
ending | -0.028 |
mistake | -0.028 |
Token | ash |
Token ID | 12530 |
Feature activation | -5.03e+00 |
Pos logit contributions | |
ay | +0.001 |
ong | +0.001 |
around | +0.001 |
rac | +0.001 |
issors | +0.001 |
Neg logit contributions | |
Kids | -0.001 |
Sweet | -0.001 |
Sleep | -0.001 |
polar | -0.001 |
ending | -0.001 |
Token | tr |
Token ID | 2213 |
Feature activation | -7.06e+00 |
Pos logit contributions | |
Sleep | +0.131 |
Sweet | +0.129 |
polar | +0.126 |
ending | +0.124 |
Kids | +0.122 |
Neg logit contributions | |
ay | -0.211 |
ong | -0.201 |
around | -0.196 |
rac | -0.195 |
issors | -0.189 |
Token | ay |
Token ID | 323 |
Feature activation | +4.67e+00 |
Pos logit contributions | |
department | +0.378 |
tr | +0.375 |
Sweet | +0.372 |
gen | +0.367 |
again | +0.365 |
Neg logit contributions | |
ong | -0.049 |
rac | -0.049 |
ars | -0.048 |
around | -0.042 |
bs | -0.041 |
Token | and |
Token ID | 290 |
Feature activation | -3.50e-01 |
Pos logit contributions | |
ay | +0.135 |
rac | +0.130 |
ong | +0.129 |
issors | +0.120 |
ars | +0.118 |
Neg logit contributions | |
Sleep | -0.171 |
Sweet | -0.169 |
polar | -0.169 |
moment | -0.167 |
ending | -0.164 |
Token | put |
Token ID | 1234 |
Feature activation | +5.39e-01 |
Pos logit contributions | |
polar | +0.012 |
Sleep | +0.012 |
Sweet | +0.011 |
ending | +0.011 |
mistake | +0.011 |
Neg logit contributions | |
ay | -0.012 |
ong | -0.011 |
rac | -0.011 |
issors | -0.011 |
around | -0.010 |
Token | it |
Token ID | 340 |
Feature activation | +5.15e-02 |
Pos logit contributions | |
ong | +0.019 |
rac | +0.019 |
ay | +0.019 |
issors | +0.018 |
ars | +0.017 |
Neg logit contributions | |
moment | -0.016 |
polar | -0.016 |
pear | -0.015 |
Are | -0.015 |
again | -0.015 |
Token | in |
Token ID | 287 |
Feature activation | -8.09e-01 |
Pos logit contributions | |
ay | +0.001 |
ong | +0.001 |
rac | +0.001 |
around | +0.001 |
issors | +0.001 |
Neg logit contributions | |
Sleep | -0.002 |
polar | -0.002 |
Sweet | -0.002 |
ending | -0.002 |
mistake | -0.002 |
Token | little |
Token ID | 1310 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
burn | +0.082 |
Sleep | +0.079 |
sometime | +0.078 |
former | +0.078 |
fly | +0.076 |
Neg logit contributions | |
at | -0.044 |
ay | -0.043 |
past | -0.043 |
- | -0.041 |
around | -0.040 |
Token | boy |
Token ID | 2933 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
Sleep | +0.081 |
former | +0.076 |
Kids | +0.076 |
Sweet | +0.075 |
charity | +0.074 |
Neg logit contributions | |
at | -0.082 |
around | -0.079 |
ay | -0.076 |
past | -0.075 |
- | -0.073 |
Token | squeezed |
Token ID | 29650 |
Feature activation | +2.79e-01 |
Pos logit contributions | |
Sleep | +0.051 |
Sweet | +0.051 |
ending | +0.050 |
Kids | +0.049 |
charity | +0.049 |
Neg logit contributions | |
around | -0.045 |
ay | -0.044 |
ong | -0.042 |
rac | -0.042 |
at | -0.041 |
Token | the |
Token ID | 262 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
ay | +0.009 |
ong | +0.009 |
issors | +0.009 |
rac | +0.009 |
ars | +0.008 |
Neg logit contributions | |
polar | -0.009 |
moment | -0.009 |
Bird | -0.009 |
Sweet | -0.009 |
Mole | -0.009 |
Token | ash |
Token ID | 12530 |
Feature activation | -5.13e+00 |
Pos logit contributions | |
ay | +0.001 |
ong | +0.001 |
around | +0.001 |
past | +0.001 |
at | +0.001 |
Neg logit contributions | |
Kids | -0.001 |
Sleep | -0.001 |
Sweet | -0.000 |
former | -0.000 |
sometime | -0.000 |
Token | tr |
Token ID | 2213 |
Feature activation | -7.05e+00 |
Pos logit contributions | |
ending | +0.014 |
moment | +0.014 |
department | +0.014 |
polar | +0.012 |
Bird | +0.009 |
Neg logit contributions | |
ay | -0.316 |
rac | -0.311 |
ong | -0.311 |
issors | -0.303 |
bs | -0.301 |
Token | ay |
Token ID | 323 |
Feature activation | +4.59e+00 |
Pos logit contributions | |
tr | +0.370 |
department | +0.369 |
Sweet | +0.362 |
gen | +0.357 |
again | +0.350 |
Neg logit contributions | |
ars | -0.059 |
rac | -0.056 |
around | -0.056 |
ong | -0.054 |
itching | -0.048 |
Token | close |
Token ID | 1969 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
ay | +0.129 |
ong | +0.120 |
rac | +0.118 |
issors | +0.112 |
around | +0.108 |
Neg logit contributions | |
Sleep | -0.175 |
polar | -0.172 |
Sweet | -0.172 |
ending | -0.172 |
moment | -0.171 |
Token | and |
Token ID | 290 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
ars | +0.075 |
ay | +0.075 |
rac | +0.075 |
issors | +0.075 |
ong | +0.073 |
Neg logit contributions | |
again | -0.071 |
moment | -0.069 |
ending | -0.065 |
mistake | -0.064 |
polar | -0.063 |
Token | gave |
Token ID | 2921 |
Feature activation | -6.09e-01 |
Pos logit contributions | |
Sleep | +0.004 |
polar | +0.004 |
ending | +0.004 |
Sweet | +0.004 |
mistake | +0.004 |
Neg logit contributions | |
ay | -0.005 |
ong | -0.004 |
rac | -0.004 |
issors | -0.004 |
around | -0.004 |
Token | it |
Token ID | 340 |
Feature activation | -1.87e-01 |
Pos logit contributions | |
moment | +0.017 |
again | +0.017 |
snow | +0.016 |
Bird | +0.016 |
forgiveness | +0.016 |
Neg logit contributions | |
issors | -0.023 |
ay | -0.023 |
ong | -0.022 |
itching | -0.021 |
iddling | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | +6.76e-02 |
Pos logit contributions | |
ong | +0.007 |
rac | +0.007 |
ay | +0.007 |
issors | +0.007 |
led | +0.006 |
Neg logit contributions | |
Sleep | -0.003 |
Kids | -0.003 |
Sweet | -0.003 |
moment | -0.003 |
polar | -0.003 |
Token | There |
Token ID | 1858 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
ong | +0.003 |
ay | +0.003 |
rac | +0.003 |
bs | +0.003 |
issors | +0.003 |
Neg logit contributions | |
Sleep | -0.002 |
Kids | -0.002 |
Are | -0.002 |
polar | -0.002 |
Sweet | -0.002 |
Token | was |
Token ID | 373 |
Feature activation | -2.23e-01 |
Pos logit contributions | |
ending | +0.026 |
moment | +0.025 |
polar | +0.025 |
Sleep | +0.024 |
Kids | +0.023 |
Neg logit contributions | |
issors | -0.042 |
ay | -0.042 |
ong | -0.041 |
rac | -0.040 |
bs | -0.039 |
Token | an |
Token ID | 281 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
Sweet | +0.008 |
Sleep | +0.008 |
mistake | +0.008 |
Kids | +0.008 |
polar | +0.008 |
Neg logit contributions | |
ay | -0.007 |
around | -0.007 |
ong | -0.007 |
rac | -0.006 |
at | -0.006 |
Token | ash |
Token ID | 12530 |
Feature activation | -5.00e+00 |
Pos logit contributions | |
ay | +0.004 |
ong | +0.003 |
rac | +0.003 |
around | +0.003 |
bs | +0.003 |
Neg logit contributions | |
polar | -0.004 |
ending | -0.004 |
Sleep | -0.004 |
mistake | -0.004 |
Sweet | -0.004 |
Token | tr |
Token ID | 2213 |
Feature activation | -7.05e+00 |
Pos logit contributions | |
Sleep | +0.057 |
polar | +0.055 |
Sweet | +0.055 |
ending | +0.053 |
mistake | +0.050 |
Neg logit contributions | |
ay | -0.278 |
ong | -0.269 |
rac | -0.265 |
around | -0.260 |
issors | -0.258 |
Token | ay |
Token ID | 323 |
Feature activation | +4.67e+00 |
Pos logit contributions | |
department | +0.372 |
tr | +0.367 |
polar | +0.357 |
gen | +0.356 |
ig | +0.353 |
Neg logit contributions | |
bs | -0.068 |
ars | -0.065 |
rac | -0.064 |
around | -0.057 |
issors | -0.056 |
Token | on |
Token ID | 319 |
Feature activation | +4.81e-01 |
Pos logit contributions | |
ay | +0.118 |
ong | +0.110 |
around | +0.104 |
rac | +0.100 |
issors | +0.095 |
Neg logit contributions | |
Sleep | -0.198 |
Sweet | -0.198 |
polar | -0.196 |
ending | -0.194 |
mistake | -0.191 |
Token | the |
Token ID | 262 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
ay | +0.015 |
ong | +0.014 |
rac | +0.014 |
bs | +0.014 |
ars | +0.013 |
Neg logit contributions | |
polar | -0.017 |
Sleep | -0.017 |
ending | -0.017 |
mistake | -0.016 |
moral | -0.016 |
Token | ground |
Token ID | 2323 |
Feature activation | +2.74e-02 |
Pos logit contributions | |
ay | +0.009 |
ong | +0.009 |
rac | +0.009 |
around | +0.009 |
issors | +0.008 |
Neg logit contributions | |
Sleep | -0.006 |
Sweet | -0.006 |
polar | -0.006 |
ending | -0.006 |
mistake | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -4.95e-01 |
Pos logit contributions | |
ay | +0.001 |
ong | +0.001 |
rac | +0.001 |
issors | +0.001 |
bs | +0.001 |
Neg logit contributions | |
Sleep | -0.001 |
ending | -0.001 |
polar | -0.001 |
mistake | -0.001 |
moment | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
polar | +0.060 |
Sleep | +0.060 |
Sweet | +0.059 |
ending | +0.059 |
mistake | +0.058 |
Neg logit contributions | |
ay | -0.027 |
ong | -0.025 |
rac | -0.024 |
around | -0.024 |
issors | -0.022 |
Token | create |
Token ID | 2251 |
Feature activation | -2.36e-01 |
Pos logit contributions | |
Sweet | +0.007 |
Sleep | +0.007 |
polar | +0.007 |
ending | +0.007 |
mistake | +0.006 |
Neg logit contributions | |
ay | -0.007 |
ong | -0.007 |
rac | -0.007 |
around | -0.007 |
issors | -0.006 |
Token | her |
Token ID | 607 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
mistake | +0.005 |
polar | +0.005 |
forgiveness | +0.005 |
Sleep | +0.005 |
Sweet | +0.005 |
Neg logit contributions | |
ay | -0.011 |
rac | -0.010 |
ong | -0.010 |
issors | -0.010 |
ars | -0.010 |
Token | own |
Token ID | 898 |
Feature activation | +8.69e-01 |
Pos logit contributions | |
ay | +0.033 |
ong | +0.031 |
rac | +0.030 |
around | +0.030 |
issors | +0.028 |
Neg logit contributions | |
Sleep | -0.036 |
Sweet | -0.035 |
polar | -0.034 |
Kids | -0.034 |
ending | -0.034 |
Token | ash |
Token ID | 12530 |
Feature activation | -4.99e+00 |
Pos logit contributions | |
ay | +0.033 |
ong | +0.031 |
rac | +0.030 |
around | +0.030 |
issors | +0.029 |
Neg logit contributions | |
Sleep | -0.026 |
Sweet | -0.025 |
polar | -0.025 |
ending | -0.025 |
mistake | -0.024 |
Token | tr |
Token ID | 2213 |
Feature activation | -7.04e+00 |
Pos logit contributions | |
Sleep | +0.046 |
polar | +0.045 |
ending | +0.044 |
Sweet | +0.044 |
mistake | +0.040 |
Neg logit contributions | |
ay | -0.287 |
ong | -0.277 |
rac | -0.274 |
around | -0.268 |
issors | -0.267 |
Token | ay |
Token ID | 323 |
Feature activation | +4.78e+00 |
Pos logit contributions | |
tr | +0.388 |
department | +0.381 |
Sweet | +0.373 |
gen | +0.370 |
polar | +0.368 |
Neg logit contributions | |
bs | -0.049 |
ars | -0.047 |
rac | -0.044 |
issors | -0.042 |
around | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | -5.63e-01 |
Pos logit contributions | |
ay | +0.129 |
ong | +0.120 |
rac | +0.115 |
around | +0.112 |
issors | +0.109 |
Neg logit contributions | |
Sleep | -0.192 |
polar | -0.190 |
Sweet | -0.190 |
ending | -0.188 |
mistake | -0.185 |
Token | She |
Token ID | 1375 |
Feature activation | -8.52e-02 |
Pos logit contributions | |
polar | +0.024 |
mistake | +0.024 |
ending | +0.024 |
Sleep | +0.024 |
Sweet | +0.023 |
Neg logit contributions | |
ay | -0.014 |
ong | -0.013 |
around | -0.013 |
rac | -0.012 |
issors | -0.012 |
Token | went |
Token ID | 1816 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
polar | +0.002 |
moral | +0.002 |
mistake | +0.002 |
moment | +0.002 |
Sleep | +0.002 |
Neg logit contributions | |
ay | -0.004 |
ong | -0.004 |
rac | -0.003 |
issors | -0.003 |
forward | -0.003 |
Token | to |
Token ID | 284 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
moment | +0.006 |
polar | +0.006 |
mistake | +0.006 |
Sleep | +0.006 |
department | +0.006 |
Neg logit contributions | |
ay | -0.005 |
rac | -0.005 |
ong | -0.004 |
issors | -0.004 |
ars | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | -4.49e-02 |
Pos logit contributions | |
ay | +0.003 |
issors | +0.003 |
ors | +0.003 |
ong | +0.003 |
iddling | +0.003 |
Neg logit contributions | |
ending | -0.003 |
moment | -0.003 |
Mole | -0.002 |
pear | -0.002 |
moral | -0.002 |
Token | he |
Token ID | 339 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
moment | +0.001 |
ending | +0.001 |
moral | +0.001 |
Mole | +0.001 |
Sue | +0.001 |
Neg logit contributions | |
ors | -0.002 |
ay | -0.002 |
iddling | -0.002 |
ong | -0.002 |
itching | -0.002 |
Token | saw |
Token ID | 2497 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
around | +0.035 |
at | +0.034 |
ay | +0.033 |
bs | +0.031 |
ong | +0.031 |
Neg logit contributions | |
forgiveness | -0.054 |
Sweet | -0.054 |
Sleep | -0.053 |
ending | -0.051 |
former | -0.051 |
Token | an |
Token ID | 281 |
Feature activation | +4.36e-02 |
Pos logit contributions | |
ong | +0.007 |
ay | +0.007 |
around | +0.007 |
rac | +0.007 |
at | +0.006 |
Neg logit contributions | |
Sleep | -0.009 |
ending | -0.009 |
moment | -0.009 |
former | -0.009 |
mistake | -0.009 |
Token | ash |
Token ID | 12530 |
Feature activation | -5.05e+00 |
Pos logit contributions | |
ay | +0.001 |
ong | +0.001 |
rac | +0.001 |
around | +0.001 |
issors | +0.001 |
Neg logit contributions | |
Sleep | -0.001 |
polar | -0.001 |
Sweet | -0.001 |
ending | -0.001 |
mistake | -0.001 |
Token | tr |
Token ID | 2213 |
Feature activation | -7.02e+00 |
Pos logit contributions | |
ending | +0.035 |
polar | +0.034 |
Sleep | +0.032 |
mistake | +0.029 |
Sweet | +0.029 |
Neg logit contributions | |
ay | -0.299 |
ong | -0.290 |
rac | -0.289 |
issors | -0.281 |
bs | -0.280 |
Token | ay |
Token ID | 323 |
Feature activation | +4.63e+00 |
Pos logit contributions | |
tr | +0.378 |
department | +0.367 |
gen | +0.354 |
tunnel | +0.348 |
polar | +0.348 |
Neg logit contributions | |
bs | -0.070 |
ars | -0.068 |
rac | -0.062 |
issors | -0.056 |
around | -0.055 |
Token | on |
Token ID | 319 |
Feature activation | +4.46e-01 |
Pos logit contributions | |
ay | +0.117 |
ong | +0.109 |
rac | +0.103 |
around | +0.102 |
issors | +0.097 |
Neg logit contributions | |
Sleep | -0.194 |
polar | -0.193 |
Sweet | -0.193 |
ending | -0.190 |
mistake | -0.188 |
Token | the |
Token ID | 262 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
ay | +0.014 |
ong | +0.013 |
rac | +0.013 |
around | +0.012 |
bs | +0.012 |
Neg logit contributions | |
polar | -0.016 |
Sleep | -0.016 |
ending | -0.016 |
Sweet | -0.016 |
mistake | -0.016 |
Token | table |
Token ID | 3084 |
Feature activation | +8.21e-01 |
Pos logit contributions | |
ay | +0.005 |
ong | +0.005 |
rac | +0.005 |
around | +0.005 |
issors | +0.005 |
Neg logit contributions | |
Sleep | -0.004 |
Sweet | -0.004 |
polar | -0.004 |
ending | -0.004 |
mistake | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -5.69e-01 |
Pos logit contributions | |
ay | +0.024 |
ong | +0.022 |
rac | +0.022 |
issors | +0.021 |
bs | +0.020 |
Neg logit contributions | |
Sleep | -0.031 |
ending | -0.031 |
polar | -0.030 |
mistake | -0.030 |
Sweet | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | -5.57e-01 |
Pos logit contributions | |
ay | +0.026 |
ong | +0.024 |
rac | +0.023 |
around | +0.023 |
issors | +0.022 |
Neg logit contributions | |
Sleep | -0.048 |
Sweet | -0.047 |
polar | -0.047 |
ending | -0.046 |
Kids | -0.046 |
Token | He |
Token ID | 679 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
Sleep | +0.018 |
polar | +0.018 |
Sweet | +0.018 |
ending | +0.018 |
mistake | +0.018 |
Neg logit contributions | |
ay | -0.019 |
ong | -0.018 |
rac | -0.017 |
around | -0.017 |
issors | -0.017 |
Token | said |
Token ID | 531 |
Feature activation | +8.59e-01 |
Pos logit contributions | |
ay | +0.007 |
ong | +0.007 |
rac | +0.007 |
around | +0.007 |
bs | +0.006 |
Neg logit contributions | |
Sleep | -0.009 |
Sweet | -0.009 |
polar | -0.009 |
ending | -0.009 |
mistake | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
ay | +0.019 |
ong | +0.017 |
rac | +0.017 |
around | +0.016 |
issors | +0.015 |
Neg logit contributions | |
Sleep | -0.039 |
polar | -0.038 |
Sweet | -0.038 |
ending | -0.038 |
mistake | -0.037 |
Token | " |
Token ID | 366 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
ay | +0.008 |
ong | +0.008 |
rac | +0.008 |
issors | +0.007 |
around | +0.007 |
Neg logit contributions | |
Sleep | -0.007 |
Sweet | -0.007 |
polar | -0.006 |
ending | -0.006 |
moment | -0.006 |
Token | Sweet |
Token ID | 36087 |
Feature activation | -7.01e+00 |
Pos logit contributions | |
ay | +0.053 |
ong | +0.050 |
rac | +0.050 |
around | +0.049 |
issors | +0.047 |
Neg logit contributions | |
polar | -0.044 |
ending | -0.043 |
mistake | -0.042 |
Sleep | -0.042 |
Sweet | -0.041 |
Token | heart |
Token ID | 11499 |
Feature activation | +2.45e+00 |
Pos logit contributions | |
Sleep | +0.277 |
Sweet | +0.271 |
department | +0.265 |
Are | +0.262 |
charity | +0.258 |
Neg logit contributions | |
ay | -0.201 |
rac | -0.178 |
ong | -0.176 |
around | -0.174 |
at | -0.169 |
Token | , |
Token ID | 11 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
ay | +0.063 |
ong | +0.061 |
rac | +0.059 |
issors | +0.057 |
around | +0.055 |
Neg logit contributions | |
ending | -0.098 |
mistake | -0.097 |
polar | -0.096 |
moment | -0.095 |
Sleep | -0.094 |
Token | it |
Token ID | 340 |
Feature activation | +4.51e-02 |
Pos logit contributions | |
ay | +0.005 |
around | +0.005 |
ong | +0.005 |
rac | +0.005 |
at | +0.005 |
Neg logit contributions | |
polar | -0.011 |
Sleep | -0.010 |
Sweet | -0.010 |
ending | -0.010 |
former | -0.010 |
Token | 's |
Token ID | 338 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
ay | +0.001 |
ong | +0.001 |
rac | +0.001 |
around | +0.001 |
issors | +0.001 |
Neg logit contributions | |
Sleep | -0.002 |
polar | -0.002 |
Sweet | -0.002 |
ending | -0.002 |
mistake | -0.002 |
Token | not |
Token ID | 407 |
Feature activation | -7.49e-01 |
Pos logit contributions | |
Sweet | +0.080 |
Kids | +0.078 |
polar | +0.077 |
Sleep | +0.077 |
Are | +0.074 |
Neg logit contributions | |
ay | -0.065 |
ong | -0.061 |
around | -0.061 |
issors | -0.058 |
at | -0.057 |
Token | saw |
Token ID | 2497 |
Feature activation | +3.39e-01 |
Pos logit contributions | |
ay | +0.043 |
around | +0.043 |
ong | +0.040 |
at | +0.040 |
rac | +0.039 |
Neg logit contributions | |
Sweet | -0.051 |
Sleep | -0.050 |
forgiveness | -0.049 |
ending | -0.048 |
polar | -0.047 |
Token | a |
Token ID | 257 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
ong | +0.010 |
ay | +0.009 |
around | +0.009 |
rac | +0.009 |
at | +0.008 |
Neg logit contributions | |
Sleep | -0.013 |
Sweet | -0.013 |
Kids | -0.012 |
ending | -0.012 |
mistake | -0.012 |
Token | dusty |
Token ID | 36972 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
ay | +0.005 |
ong | +0.004 |
high | +0.004 |
around | +0.004 |
at | +0.004 |
Neg logit contributions | |
Sleep | -0.009 |
Sweet | -0.008 |
Kids | -0.008 |
former | -0.008 |
days | -0.007 |
Token | old |
Token ID | 1468 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
ay | +0.042 |
ong | +0.039 |
around | +0.039 |
rac | +0.036 |
issors | +0.035 |
Neg logit contributions | |
Sweet | -0.036 |
Sleep | -0.036 |
Kids | -0.034 |
ending | -0.032 |
polar | -0.032 |
Token | ash |
Token ID | 12530 |
Feature activation | -5.04e+00 |
Pos logit contributions | |
ay | +0.006 |
around | +0.005 |
ong | +0.005 |
at | +0.005 |
high | +0.005 |
Neg logit contributions | |
Sleep | -0.006 |
Sweet | -0.006 |
Kids | -0.006 |
former | -0.005 |
moral | -0.005 |
Token | tr |
Token ID | 2213 |
Feature activation | -7.01e+00 |
Pos logit contributions | |
Sleep | +0.079 |
polar | +0.077 |
Sweet | +0.077 |
ending | +0.075 |
mistake | +0.072 |
Neg logit contributions | |
ay | -0.258 |
ong | -0.249 |
rac | -0.244 |
around | -0.240 |
issors | -0.237 |
Token | ay |
Token ID | 323 |
Feature activation | +4.69e+00 |
Pos logit contributions | |
tr | +0.375 |
department | +0.369 |
gen | +0.364 |
ig | +0.350 |
polar | +0.350 |
Neg logit contributions | |
rac | -0.064 |
bs | -0.062 |
ars | -0.061 |
issors | -0.057 |
around | -0.052 |
Token | on |
Token ID | 319 |
Feature activation | +5.41e-01 |
Pos logit contributions | |
ay | +0.120 |
ong | +0.112 |
around | +0.107 |
rac | +0.104 |
issors | +0.099 |
Neg logit contributions | |
Sweet | -0.195 |
Sleep | -0.195 |
polar | -0.194 |
ending | -0.191 |
mistake | -0.189 |
Token | the |
Token ID | 262 |
Feature activation | +2.37e-01 |
Pos logit contributions | |
ay | +0.016 |
ong | +0.015 |
rac | +0.014 |
around | +0.014 |
issors | +0.013 |
Neg logit contributions | |
Sleep | -0.021 |
polar | -0.020 |
Sweet | -0.020 |
ending | -0.020 |
mistake | -0.020 |
Token | shelf |
Token ID | 18316 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
ay | +0.009 |
ong | +0.008 |
around | +0.008 |
rac | +0.008 |
issors | +0.008 |
Neg logit contributions | |
Sweet | -0.007 |
Sleep | -0.007 |
polar | -0.007 |
Kids | -0.007 |
ending | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -4.58e-01 |
Pos logit contributions | |
ay | +0.050 |
ong | +0.047 |
rac | +0.046 |
around | +0.043 |
issors | +0.043 |
Neg logit contributions | |
Sleep | -0.069 |
polar | -0.069 |
ending | -0.069 |
Sweet | -0.068 |
mistake | -0.068 |
Token | , |
Token ID | 11 |
Feature activation | +4.50e-02 |
Pos logit contributions | |
ay | +0.004 |
ong | +0.004 |
rac | +0.004 |
issors | +0.004 |
ars | +0.004 |
Neg logit contributions | |
ending | -0.006 |
Sweet | -0.006 |
Sleep | -0.006 |
moment | -0.006 |
polar | -0.006 |
Token | Lily |
Token ID | 20037 |
Feature activation | -6.56e-01 |
Pos logit contributions | |
ay | +0.002 |
ong | +0.002 |
rac | +0.002 |
around | +0.002 |
issors | +0.002 |
Neg logit contributions | |
Sleep | -0.001 |
polar | -0.001 |
Sweet | -0.001 |
ending | -0.001 |
mistake | -0.001 |
Token | found |
Token ID | 1043 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
Sleep | +0.023 |
Sweet | +0.023 |
polar | +0.023 |
ending | +0.023 |
forgiveness | +0.022 |
Neg logit contributions | |
ay | -0.020 |
around | -0.019 |
ong | -0.019 |
rac | -0.018 |
bs | -0.018 |
Token | an |
Token ID | 281 |
Feature activation | +6.54e-02 |
Pos logit contributions | |
ay | +0.055 |
ong | +0.052 |
around | +0.051 |
rac | +0.051 |
issors | +0.048 |
Neg logit contributions | |
ending | -0.051 |
Sleep | -0.051 |
polar | -0.051 |
mistake | -0.050 |
Sweet | -0.050 |
Token | ash |
Token ID | 12530 |
Feature activation | -5.01e+00 |
Pos logit contributions | |
ay | +0.002 |
ong | +0.002 |
rac | +0.002 |
bs | +0.002 |
ars | +0.002 |
Neg logit contributions | |
polar | -0.002 |
ending | -0.002 |
mistake | -0.002 |
Sleep | -0.002 |
moral | -0.002 |
Token | tr |
Token ID | 2213 |
Feature activation | -7.00e+00 |
Pos logit contributions | |
polar | +0.033 |
Sleep | +0.033 |
ending | +0.032 |
Sweet | +0.030 |
mistake | +0.028 |
Neg logit contributions | |
ay | -0.299 |
ong | -0.290 |
rac | -0.287 |
around | -0.280 |
issors | -0.279 |
Token | ay |
Token ID | 323 |
Feature activation | +4.66e+00 |
Pos logit contributions | |
tr | +0.376 |
department | +0.371 |
gen | +0.359 |
tunnel | +0.356 |
Sweet | +0.354 |
Neg logit contributions | |
ars | -0.058 |
bs | -0.055 |
rac | -0.053 |
itching | -0.048 |
ong | -0.046 |
Token | in |
Token ID | 287 |
Feature activation | -5.66e-01 |
Pos logit contributions | |
ay | +0.116 |
ong | +0.107 |
rac | +0.102 |
around | +0.100 |
issors | +0.096 |
Neg logit contributions | |
Sleep | -0.197 |
polar | -0.197 |
Sweet | -0.196 |
ending | -0.194 |
mistake | -0.191 |
Token | the |
Token ID | 262 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
ending | +0.026 |
polar | +0.026 |
department | +0.025 |
Sweet | +0.025 |
Sleep | +0.025 |
Neg logit contributions | |
ay | -0.012 |
around | -0.010 |
ong | -0.010 |
rac | -0.010 |
issors | -0.009 |
Token | world |
Token ID | 995 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
ay | +0.010 |
around | +0.009 |
ong | +0.009 |
rac | +0.009 |
issors | +0.009 |
Neg logit contributions | |
polar | -0.006 |
Sleep | -0.005 |
Sweet | -0.005 |
department | -0.005 |
ending | -0.005 |
Token | behind |
Token ID | 2157 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
ay | +0.005 |
rac | +0.005 |
ong | +0.005 |
issors | +0.004 |
bs | +0.004 |
Neg logit contributions | |
Sleep | -0.006 |
moment | -0.006 |
ending | -0.006 |
mistake | -0.006 |
Sweet | -0.006 |
Token | was |
Token ID | 373 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
ay | +0.012 |
ong | +0.012 |
issors | +0.011 |
rac | +0.011 |
bs | +0.011 |
Neg logit contributions | |
polar | -0.006 |
ending | -0.006 |
Sleep | -0.006 |
mistake | -0.006 |
moment | -0.006 |
Token | hitting |
Token ID | 9008 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
polar | +0.007 |
ending | +0.006 |
Sleep | +0.006 |
Sweet | +0.006 |
mistake | +0.006 |
Neg logit contributions | |
ay | -0.007 |
ong | -0.006 |
rac | -0.006 |
issors | -0.006 |
ors | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
ay | +0.004 |
ong | +0.004 |
rac | +0.004 |
issors | +0.004 |
bs | +0.004 |
Neg logit contributions | |
polar | -0.004 |
mistake | -0.004 |
ending | -0.004 |
moment | -0.004 |
Sweet | -0.004 |
Token | target |
Token ID | 2496 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
ay | +0.007 |
bs | +0.007 |
rac | +0.006 |
ong | +0.006 |
ors | +0.006 |
Neg logit contributions | |
moral | -0.004 |
ending | -0.004 |
mistake | -0.004 |
moment | -0.004 |
island | -0.003 |
Token | with |
Token ID | 351 |
Feature activation | +9.48e-01 |
Pos logit contributions | |
ay | +0.037 |
rac | +0.037 |
ong | +0.035 |
issors | +0.035 |
bs | +0.032 |
Neg logit contributions | |
ending | -0.041 |
mistake | -0.040 |
moment | -0.040 |
polar | -0.039 |
moral | -0.038 |
Token | ease |
Token ID | 10152 |
Feature activation | +3.79e+00 |
Pos logit contributions | |
rac | +0.034 |
ors | +0.034 |
issors | +0.033 |
ong | +0.033 |
ay | +0.033 |
Neg logit contributions | |
polar | -0.029 |
ending | -0.027 |
forgiveness | -0.026 |
pear | -0.026 |
mistake | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -2.57e-01 |
Pos logit contributions | |
rac | +0.119 |
ay | +0.117 |
issors | +0.111 |
ong | +0.110 |
bs | +0.105 |
Neg logit contributions | |
ending | -0.132 |
polar | -0.129 |
mistake | -0.129 |
moment | -0.129 |
again | -0.124 |
Token | Everyone |
Token ID | 11075 |
Feature activation | -6.29e-01 |
Pos logit contributions | |
polar | +0.008 |
ending | +0.008 |
Sleep | +0.008 |
Sweet | +0.008 |
moment | +0.008 |
Neg logit contributions | |
ay | -0.009 |
ong | -0.008 |
rac | -0.008 |
issors | -0.008 |
around | -0.008 |
Token | was |
Token ID | 373 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
polar | +0.017 |
moment | +0.017 |
ending | +0.016 |
Sleep | +0.016 |
Sweet | +0.016 |
Neg logit contributions | |
ay | -0.025 |
issors | -0.024 |
ong | -0.024 |
rac | -0.023 |
bs | -0.022 |
Token | cheering |
Token ID | 29732 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
polar | +0.008 |
Sleep | +0.008 |
Sweet | +0.008 |
ending | +0.008 |
mistake | +0.008 |
Neg logit contributions | |
ay | -0.007 |
ong | -0.007 |
rac | -0.006 |
issors | -0.006 |
bs | -0.006 |
Token | for |
Token ID | 329 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
ay | +0.034 |
rac | +0.034 |
issors | +0.034 |
ong | +0.033 |
ars | +0.032 |
Neg logit contributions | |
ending | -0.036 |
moment | -0.035 |
polar | -0.033 |
again | -0.033 |
mistake | -0.033 |
Token | and |
Token ID | 290 |
Feature activation | +9.13e-02 |
Pos logit contributions | |
polar | +0.015 |
pear | +0.015 |
Sweet | +0.014 |
moral | +0.014 |
snow | +0.014 |
Neg logit contributions | |
rac | -0.017 |
issors | -0.016 |
ay | -0.016 |
ong | -0.016 |
bs | -0.015 |
Token | color |
Token ID | 3124 |
Feature activation | +5.54e-01 |
Pos logit contributions | |
ay | +0.003 |
ong | +0.003 |
rac | +0.003 |
around | +0.003 |
issors | +0.003 |
Neg logit contributions | |
Sleep | -0.003 |
polar | -0.003 |
Sweet | -0.003 |
ending | -0.003 |
mistake | -0.003 |
Token | with |
Token ID | 351 |
Feature activation | +8.52e-01 |
Pos logit contributions | |
rac | +0.017 |
ay | +0.016 |
ong | +0.016 |
issors | +0.016 |
bs | +0.015 |
Neg logit contributions | |
polar | -0.019 |
moment | -0.018 |
ending | -0.018 |
Sweet | -0.018 |
pear | -0.018 |
Token | her |
Token ID | 607 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
ars | +0.036 |
ors | +0.034 |
rac | +0.033 |
ong | +0.033 |
ables | +0.033 |
Neg logit contributions | |
snow | -0.022 |
pear | -0.022 |
tul | -0.021 |
moral | -0.021 |
Sweet | -0.020 |
Token | cr |
Token ID | 1067 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
ong | +0.044 |
ay | +0.043 |
rac | +0.042 |
ors | +0.042 |
ars | +0.041 |
Neg logit contributions | |
moment | -0.026 |
ending | -0.025 |
polar | -0.025 |
mistake | -0.025 |
moral | -0.025 |
Token | ay |
Token ID | 323 |
Feature activation | +4.76e+00 |
Pos logit contributions | |
forward | +0.016 |
around | +0.015 |
bs | +0.014 |
ging | +0.013 |
led | +0.013 |
Neg logit contributions | |
tr | -0.053 |
Sweet | -0.052 |
Sleep | -0.051 |
omo | -0.050 |
former | -0.050 |
Token | ons |
Token ID | 684 |
Feature activation | +2.16e+00 |
Pos logit contributions | |
forward | +0.122 |
around | +0.108 |
ized | +0.107 |
iddling | +0.107 |
ivated | +0.104 |
Neg logit contributions | |
moral | -0.182 |
Sweet | -0.181 |
moment | -0.177 |
forgiveness | -0.175 |
mistake | -0.169 |
Token | . |
Token ID | 13 |
Feature activation | -4.25e-01 |
Pos logit contributions | |
rac | +0.092 |
ay | +0.088 |
ong | +0.085 |
issors | +0.085 |
ars | +0.082 |
Neg logit contributions | |
again | -0.053 |
moment | -0.049 |
flower | -0.048 |
ending | -0.047 |
pear | -0.046 |
Token | One |
Token ID | 1881 |
Feature activation | -2.52e+00 |
Pos logit contributions | |
polar | +0.016 |
mistake | +0.015 |
Sleep | +0.015 |
Sweet | +0.015 |
forgiveness | +0.015 |
Neg logit contributions | |
ay | -0.014 |
around | -0.013 |
ong | -0.012 |
issors | -0.012 |
rac | -0.012 |
Token | day |
Token ID | 1110 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
Sleep | +0.076 |
polar | +0.076 |
former | +0.075 |
department | +0.075 |
Sweet | +0.074 |
Neg logit contributions | |
ay | -0.097 |
around | -0.087 |
ong | -0.087 |
rac | -0.084 |
bs | -0.081 |
Token | , |
Token ID | 11 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
ong | +0.006 |
rac | +0.006 |
ay | +0.006 |
issors | +0.006 |
ars | +0.006 |
Neg logit contributions | |
ending | -0.007 |
moment | -0.007 |
Sweet | -0.007 |
Are | -0.007 |
pear | -0.007 |
Token | day |
Token ID | 1110 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
ay | +0.105 |
around | +0.098 |
ong | +0.094 |
rac | +0.091 |
bs | +0.088 |
Neg logit contributions | |
Sweet | -0.097 |
polar | -0.094 |
Sleep | -0.094 |
Kids | -0.093 |
department | -0.092 |
Token | he |
Token ID | 339 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
ay | +0.004 |
ong | +0.004 |
rac | +0.004 |
issors | +0.004 |
around | +0.004 |
Neg logit contributions | |
Sleep | -0.007 |
ending | -0.007 |
polar | -0.007 |
Sweet | -0.007 |
moment | -0.007 |
Token | was |
Token ID | 373 |
Feature activation | -9.48e-02 |
Pos logit contributions | |
ay | +0.050 |
ong | +0.049 |
around | +0.049 |
rac | +0.047 |
bs | +0.047 |
Neg logit contributions | |
Sweet | -0.044 |
Sleep | -0.044 |
polar | -0.042 |
ending | -0.041 |
forgiveness | -0.041 |
Token | still |
Token ID | 991 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
Sleep | +0.003 |
polar | +0.003 |
ending | +0.003 |
mistake | +0.003 |
Sweet | +0.003 |
Neg logit contributions | |
ay | -0.004 |
rac | -0.003 |
ong | -0.003 |
issors | -0.003 |
ars | -0.003 |
Token | just |
Token ID | 655 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
ay | +0.041 |
ong | +0.039 |
rac | +0.038 |
around | +0.037 |
issors | +0.036 |
Neg logit contributions | |
Sleep | -0.034 |
polar | -0.033 |
Sweet | -0.033 |
ending | -0.033 |
mistake | -0.032 |
Token | as |
Token ID | 355 |
Feature activation | +4.81e+00 |
Pos logit contributions | |
Sleep | +0.006 |
Sweet | +0.006 |
polar | +0.006 |
Kids | +0.006 |
ending | +0.006 |
Neg logit contributions | |
ay | -0.006 |
ong | -0.006 |
around | -0.006 |
rac | -0.006 |
bs | -0.006 |
Token | troubled |
Token ID | 17840 |
Feature activation | -7.55e-01 |
Pos logit contributions | |
ay | +0.146 |
rac | +0.132 |
ong | +0.131 |
issors | +0.129 |
ars | +0.120 |
Neg logit contributions | |
Sleep | -0.183 |
polar | -0.183 |
Sweet | -0.182 |
ending | -0.179 |
moment | -0.179 |
Token | as |
Token ID | 355 |
Feature activation | +4.84e+00 |
Pos logit contributions | |
ending | +0.031 |
polar | +0.031 |
Sleep | +0.030 |
mistake | +0.030 |
moment | +0.029 |
Neg logit contributions | |
ay | -0.020 |
rac | -0.019 |
ong | -0.018 |
issors | -0.018 |
ars | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
ay | +0.127 |
ong | +0.122 |
around | +0.118 |
rac | +0.115 |
bs | +0.107 |
Neg logit contributions | |
Sleep | -0.193 |
Sweet | -0.191 |
polar | -0.189 |
Kids | -0.187 |
mistake | -0.187 |
Token | day |
Token ID | 1110 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
ay | +0.011 |
ong | +0.010 |
rac | +0.010 |
issors | +0.010 |
ors | +0.010 |
Neg logit contributions | |
ending | -0.007 |
Sleep | -0.007 |
mistake | -0.007 |
Sweet | -0.007 |
moral | -0.007 |
Token | before |
Token ID | 878 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
ay | +0.007 |
ong | +0.006 |
rac | +0.006 |
issors | +0.006 |
around | +0.006 |
Neg logit contributions | |
polar | -0.007 |
Sleep | -0.007 |
ending | -0.007 |
Sweet | -0.007 |
mistake | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
ay | +0.007 |
around | +0.006 |
ong | +0.006 |
rac | +0.006 |
bs | +0.006 |
Neg logit contributions | |
Sleep | -0.007 |
Sweet | -0.007 |
polar | -0.007 |
ending | -0.007 |
Kids | -0.007 |
Token | their |
Token ID | 511 |
Feature activation | -7.97e-01 |
Pos logit contributions | |
Sleep | +0.006 |
Sweet | +0.006 |
polar | +0.006 |
Kids | +0.006 |
mistake | +0.006 |
Neg logit contributions | |
around | -0.006 |
ay | -0.006 |
ong | -0.005 |
rac | -0.005 |
at | -0.005 |
Token | friendship |
Token ID | 14738 |
Feature activation | -5.71e-01 |
Pos logit contributions | |
Sleep | +0.023 |
polar | +0.022 |
Sweet | +0.022 |
ending | +0.022 |
Kids | +0.021 |
Neg logit contributions | |
ay | -0.031 |
ong | -0.029 |
rac | -0.029 |
around | -0.028 |
issors | -0.027 |
Token | grew |
Token ID | 6348 |
Feature activation | -6.73e-01 |
Pos logit contributions | |
Sleep | +0.017 |
polar | +0.017 |
Sweet | +0.017 |
ending | +0.017 |
mistake | +0.016 |
Neg logit contributions | |
ay | -0.021 |
ong | -0.020 |
rac | -0.019 |
around | -0.019 |
issors | -0.019 |
Token | stronger |
Token ID | 7387 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
Sleep | +0.023 |
Sweet | +0.023 |
polar | +0.023 |
Kids | +0.023 |
ending | +0.023 |
Neg logit contributions | |
ay | -0.022 |
ong | -0.021 |
around | -0.020 |
rac | -0.020 |
bs | -0.019 |
Token | than |
Token ID | 621 |
Feature activation | +5.51e+00 |
Pos logit contributions | |
ending | +0.057 |
Sleep | +0.056 |
polar | +0.056 |
moment | +0.056 |
mistake | +0.056 |
Neg logit contributions | |
ay | -0.040 |
ong | -0.038 |
rac | -0.038 |
issors | -0.035 |
bs | -0.033 |
Token | ever |
Token ID | 1683 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
around | +0.165 |
ong | +0.165 |
ay | +0.162 |
bs | +0.148 |
at | +0.147 |
Neg logit contributions | |
Sweet | -0.202 |
Sleep | -0.201 |
Kids | -0.196 |
polar | -0.191 |
moral | -0.188 |
Token | . |
Token ID | 13 |
Feature activation | -5.25e-01 |
Pos logit contributions | |
Sweet | +0.050 |
polar | +0.050 |
Sleep | +0.049 |
Kids | +0.049 |
divorce | +0.047 |
Neg logit contributions | |
around | -0.037 |
ay | -0.036 |
ong | -0.036 |
at | -0.033 |
bs | -0.032 |
Token |
|
Token ID | 198 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
polar | +0.021 |
Sleep | +0.020 |
mistake | +0.020 |
ending | +0.020 |
Sweet | +0.019 |
Neg logit contributions | |
ay | -0.016 |
around | -0.014 |
rac | -0.013 |
ong | -0.013 |
issors | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | +7.03e-02 |
Pos logit contributions | |
ay | +0.005 |
around | +0.005 |
ong | +0.005 |
rac | +0.005 |
ars | +0.004 |
Neg logit contributions | |
ending | -0.005 |
polar | -0.005 |
Sleep | -0.005 |
department | -0.005 |
mistake | -0.005 |
Token | The |
Token ID | 464 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
ay | +0.002 |
rac | +0.002 |
around | +0.002 |
ars | +0.002 |
ong | +0.002 |
Neg logit contributions | |
polar | -0.003 |
ending | -0.003 |
department | -0.002 |
mistake | -0.002 |
funeral | -0.002 |
Token | he |
Token ID | 339 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
ay | +0.026 |
ong | +0.025 |
around | +0.023 |
rac | +0.023 |
bs | +0.022 |
Neg logit contributions | |
Sleep | -0.040 |
Sweet | -0.039 |
polar | -0.039 |
ending | -0.038 |
mistake | -0.038 |
Token | felt |
Token ID | 2936 |
Feature activation | +3.94e-01 |
Pos logit contributions | |
ay | +0.053 |
around | +0.051 |
ong | +0.050 |
rac | +0.049 |
ars | +0.048 |
Neg logit contributions | |
Sleep | -0.042 |
Sweet | -0.042 |
polar | -0.041 |
ending | -0.041 |
mistake | -0.040 |
Token | the |
Token ID | 262 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
ay | +0.014 |
ong | +0.014 |
rac | +0.013 |
issors | +0.013 |
around | +0.013 |
Neg logit contributions | |
Sleep | -0.012 |
polar | -0.012 |
Sweet | -0.012 |
ending | -0.012 |
mistake | -0.012 |
Token | monster |
Token ID | 9234 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
ay | +0.007 |
ong | +0.007 |
around | +0.007 |
rac | +0.006 |
ars | +0.006 |
Neg logit contributions | |
Sweet | -0.006 |
Kids | -0.006 |
Sleep | -0.006 |
polar | -0.006 |
mistake | -0.006 |
Token | 's |
Token ID | 338 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
Sweet | +0.079 |
Sleep | +0.078 |
polar | +0.077 |
Kids | +0.077 |
ending | +0.077 |
Neg logit contributions | |
ay | -0.051 |
ong | -0.048 |
around | -0.048 |
rac | -0.045 |
bs | -0.043 |
Token | breath |
Token ID | 8033 |
Feature activation | +5.97e+00 |
Pos logit contributions | |
ending | +0.067 |
Sleep | +0.067 |
polar | +0.067 |
mistake | +0.065 |
moment | +0.064 |
Neg logit contributions | |
ay | -0.082 |
rac | -0.078 |
ong | -0.077 |
issors | -0.075 |
ors | -0.073 |
Token | on |
Token ID | 319 |
Feature activation | +5.66e-01 |
Pos logit contributions | |
rac | +0.189 |
ay | +0.188 |
ong | +0.178 |
issors | +0.178 |
ors | +0.163 |
Neg logit contributions | |
Sleep | -0.208 |
ending | -0.205 |
moment | -0.204 |
polar | -0.199 |
mistake | -0.195 |
Token | his |
Token ID | 465 |
Feature activation | +8.03e-01 |
Pos logit contributions | |
ay | +0.017 |
ong | +0.016 |
rac | +0.016 |
issors | +0.015 |
around | +0.015 |
Neg logit contributions | |
polar | -0.021 |
Sleep | -0.020 |
ending | -0.020 |
mistake | -0.020 |
Sweet | -0.020 |
Token | neck |
Token ID | 7393 |
Feature activation | +4.69e+00 |
Pos logit contributions | |
ay | +0.034 |
ized | +0.031 |
ors | +0.031 |
issors | +0.031 |
bs | +0.031 |
Neg logit contributions | |
mistake | -0.021 |
ending | -0.020 |
polar | -0.020 |
moment | -0.020 |
moral | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +3.68e-02 |
Pos logit contributions | |
ay | +0.145 |
rac | +0.137 |
ong | +0.132 |
issors | +0.128 |
bs | +0.122 |
Neg logit contributions | |
Sleep | -0.166 |
mistake | -0.166 |
ending | -0.164 |
polar | -0.163 |
moment | -0.162 |
Token |
|
Token ID | 198 |
Feature activation | +7.94e-01 |
Pos logit contributions | |
ay | +0.001 |
ong | +0.001 |
rac | +0.001 |
around | +0.001 |
issors | +0.001 |
Neg logit contributions | |
ending | -0.001 |
Sleep | -0.001 |
polar | -0.001 |
mistake | -0.001 |
Sweet | -0.001 |
Token | day |
Token ID | 1110 |
Feature activation | +1.88e-01 |
Pos logit contributions | |
former | +0.087 |
mistake | +0.083 |
forgiveness | +0.083 |
Sleep | +0.082 |
Sweet | +0.082 |
Neg logit contributions | |
ay | -0.089 |
around | -0.083 |
ong | -0.076 |
at | -0.076 |
bs | -0.075 |
Token | , |
Token ID | 11 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
ay | +0.006 |
ong | +0.006 |
issors | +0.006 |
rac | +0.006 |
ars | +0.006 |
Neg logit contributions | |
moment | -0.006 |
ending | -0.006 |
Sweet | -0.006 |
pear | -0.006 |
Bird | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
ay | +0.003 |
ong | +0.003 |
rac | +0.003 |
around | +0.003 |
bs | +0.003 |
Neg logit contributions | |
Sleep | -0.004 |
polar | -0.004 |
Sweet | -0.004 |
mistake | -0.004 |
ending | -0.004 |
Token | old |
Token ID | 1468 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
ay | +0.008 |
around | +0.007 |
at | +0.007 |
past | +0.007 |
bs | +0.007 |
Neg logit contributions | |
reality | -0.008 |
polar | -0.008 |
fly | -0.008 |
sometime | -0.008 |
waterfall | -0.008 |
Token | girl |
Token ID | 2576 |
Feature activation | -7.20e-01 |
Pos logit contributions | |
ay | +0.008 |
around | +0.008 |
at | +0.007 |
bs | +0.007 |
s | +0.007 |
Neg logit contributions | |
reality | -0.006 |
Sleep | -0.006 |
polar | -0.006 |
former | -0.006 |
forgiveness | -0.006 |
Token | went |
Token ID | 1816 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
Sleep | +0.025 |
Sweet | +0.024 |
polar | +0.024 |
ending | +0.024 |
mistake | +0.024 |
Neg logit contributions | |
ay | -0.023 |
ong | -0.022 |
around | -0.022 |
rac | -0.022 |
bs | -0.021 |
Token | outside |
Token ID | 2354 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
Sleep | +0.006 |
polar | +0.006 |
ending | +0.006 |
Sweet | +0.006 |
moment | +0.006 |
Neg logit contributions | |
ay | -0.004 |
ong | -0.004 |
rac | -0.004 |
issors | -0.003 |
ars | -0.003 |
Token | to |
Token ID | 284 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
ay | +0.005 |
ong | +0.005 |
rac | +0.005 |
around | +0.004 |
issors | +0.004 |
Neg logit contributions | |
Sleep | -0.010 |
polar | -0.010 |
Sweet | -0.010 |
ending | -0.010 |
mistake | -0.009 |
Token | take |
Token ID | 1011 |
Feature activation | -7.79e-01 |
Pos logit contributions | |
Sweet | +0.010 |
ending | +0.010 |
Sleep | +0.009 |
department | +0.009 |
mistake | +0.009 |
Neg logit contributions | |
around | -0.006 |
ay | -0.005 |
rac | -0.005 |
ong | -0.005 |
bs | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
Sleep | +0.026 |
polar | +0.026 |
ending | +0.026 |
Sweet | +0.026 |
mistake | +0.025 |
Neg logit contributions | |
ay | -0.027 |
ong | -0.025 |
around | -0.024 |
rac | -0.024 |
bs | -0.023 |
Token | walk |
Token ID | 2513 |
Feature activation | -7.06e-01 |
Pos logit contributions | |
ay | +0.009 |
ong | +0.008 |
around | +0.008 |
bs | +0.008 |
rac | +0.007 |
Neg logit contributions | |
Sleep | -0.007 |
Sweet | -0.007 |
ending | -0.007 |
former | -0.007 |
Kids | -0.007 |
Token | boy |
Token ID | 2933 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
Sleep | +0.042 |
former | +0.040 |
Sweet | +0.040 |
ending | +0.038 |
mistake | +0.038 |
Neg logit contributions | |
ay | -0.100 |
ong | -0.092 |
around | -0.091 |
bs | -0.089 |
at | -0.087 |
Token | named |
Token ID | 3706 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
polar | +0.027 |
Sleep | +0.027 |
Sweet | +0.026 |
ending | +0.025 |
mistake | +0.025 |
Neg logit contributions | |
ay | -0.065 |
ong | -0.063 |
rac | -0.062 |
around | -0.060 |
issors | -0.060 |
Token | Tim |
Token ID | 5045 |
Feature activation | +3.62e-02 |
Pos logit contributions | |
mistake | +0.042 |
Sleep | +0.042 |
Sweet | +0.042 |
ending | +0.042 |
former | +0.040 |
Neg logit contributions | |
ay | -0.068 |
ong | -0.065 |
around | -0.063 |
rac | -0.062 |
issors | -0.062 |
Token | my |
Token ID | 1820 |
Feature activation | -4.45e-01 |
Pos logit contributions | |
ay | +0.002 |
ong | +0.002 |
rac | +0.002 |
around | +0.002 |
issors | +0.002 |
Neg logit contributions | |
Sleep | -0.001 |
polar | -0.001 |
Sweet | -0.001 |
ending | -0.001 |
mistake | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -3.91e-01 |
Pos logit contributions | |
polar | +0.015 |
Sleep | +0.015 |
mistake | +0.015 |
ending | +0.015 |
Sweet | +0.015 |
Neg logit contributions | |
ay | -0.015 |
ong | -0.015 |
rac | -0.014 |
issors | -0.013 |
ars | -0.013 |
Token | Tim |
Token ID | 5045 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
Sleep | +0.010 |
Sweet | +0.010 |
mistake | +0.010 |
polar | +0.010 |
ending | +0.010 |
Neg logit contributions | |
ay | -0.016 |
ong | -0.015 |
around | -0.015 |
rac | -0.015 |
issors | -0.015 |
Token | my |
Token ID | 1820 |
Feature activation | -3.60e-01 |
Pos logit contributions | |
ay | +0.007 |
around | +0.006 |
ong | +0.006 |
rac | +0.006 |
issors | +0.006 |
Neg logit contributions | |
Sweet | -0.003 |
Sleep | -0.003 |
ending | -0.003 |
polar | -0.003 |
mistake | -0.003 |
Token | loved |
Token ID | 6151 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
Sleep | +0.013 |
polar | +0.013 |
Sweet | +0.013 |
ending | +0.013 |
mistake | +0.013 |
Neg logit contributions | |
ay | -0.011 |
ong | -0.010 |
rac | -0.010 |
around | -0.010 |
issors | -0.009 |
Token | to |
Token ID | 284 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
ay | +0.050 |
ong | +0.047 |
rac | +0.045 |
ars | +0.043 |
issors | +0.043 |
Neg logit contributions | |
polar | -0.074 |
Sleep | -0.073 |
Sweet | -0.073 |
ending | -0.071 |
mistake | -0.070 |
Token | play |
Token ID | 711 |
Feature activation | -6.58e-01 |
Pos logit contributions | |
Sleep | +0.006 |
polar | +0.006 |
Sweet | +0.006 |
ending | +0.006 |
mistake | +0.006 |
Neg logit contributions | |
ay | -0.009 |
ong | -0.008 |
rac | -0.008 |
around | -0.008 |
issors | -0.008 |
Token | with |
Token ID | 351 |
Feature activation | +8.54e-01 |
Pos logit contributions | |
polar | +0.021 |
bathroom | +0.020 |
Are | +0.020 |
snow | +0.020 |
shelter | +0.020 |
Neg logit contributions | |
ay | -0.023 |
rac | -0.022 |
ong | -0.022 |
bs | -0.022 |
issors | -0.022 |
Token | her |
Token ID | 607 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
Sweet | +0.026 |
Sleep | +0.026 |
ending | +0.025 |
Kids | +0.025 |
polar | +0.025 |
Neg logit contributions | |
ay | -0.016 |
around | -0.014 |
ong | -0.013 |
rac | -0.013 |
issors | -0.012 |
Token | pocket |
Token ID | 10000 |
Feature activation | +3.05e+00 |
Pos logit contributions | |
ay | +0.037 |
ong | +0.034 |
rac | +0.034 |
around | +0.033 |
issors | +0.032 |
Neg logit contributions | |
Sleep | -0.035 |
Sweet | -0.035 |
polar | -0.034 |
ending | -0.034 |
Kids | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
ay | +0.103 |
rac | +0.097 |
ong | +0.095 |
issors | +0.092 |
ars | +0.087 |
Neg logit contributions | |
Sleep | -0.100 |
mistake | -0.098 |
polar | -0.097 |
moment | -0.096 |
former | -0.094 |
Token | And |
Token ID | 843 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
mistake | +0.008 |
ending | +0.007 |
divorce | +0.007 |
polar | +0.007 |
moral | +0.007 |
Neg logit contributions | |
ay | -0.005 |
around | -0.004 |
rac | -0.004 |
ong | -0.004 |
bs | -0.004 |
Token | then |
Token ID | 788 |
Feature activation | -6.00e-01 |
Pos logit contributions | |
Sleep | +0.045 |
polar | +0.043 |
mistake | +0.043 |
Sweet | +0.043 |
moral | +0.042 |
Neg logit contributions | |
ay | -0.035 |
around | -0.034 |
ong | -0.031 |
rac | -0.031 |
bs | -0.031 |
Token | she |
Token ID | 673 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
Sleep | +0.030 |
Sweet | +0.029 |
polar | +0.028 |
Kids | +0.028 |
divorce | +0.028 |
Neg logit contributions | |
ay | -0.011 |
around | -0.011 |
ong | -0.010 |
bs | -0.010 |
at | -0.010 |
Token | carried |
Token ID | 5281 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
ay | +0.055 |
ong | +0.051 |
around | +0.051 |
rac | +0.051 |
bs | +0.048 |
Neg logit contributions | |
Sleep | -0.050 |
Sweet | -0.049 |
ending | -0.049 |
polar | -0.047 |
moment | -0.047 |
Token | on |
Token ID | 319 |
Feature activation | +3.66e-01 |
Pos logit contributions | |
ong | +0.056 |
rac | +0.054 |
ay | +0.054 |
issors | +0.053 |
ars | +0.050 |
Neg logit contributions | |
moment | -0.059 |
polar | -0.057 |
Sweet | -0.057 |
Sleep | -0.056 |
Bird | -0.055 |
Token | with |
Token ID | 351 |
Feature activation | +9.43e-01 |
Pos logit contributions | |
ay | +0.012 |
ong | +0.011 |
rac | +0.010 |
issors | +0.010 |
ars | +0.010 |
Neg logit contributions | |
Sleep | -0.013 |
Sweet | -0.013 |
polar | -0.013 |
mistake | -0.013 |
ending | -0.013 |
Token | what |
Token ID | 644 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
ay | +0.030 |
ong | +0.030 |
issors | +0.029 |
rac | +0.029 |
ars | +0.028 |
Neg logit contributions | |
polar | -0.033 |
Sleep | -0.032 |
forgiveness | -0.032 |
Sweet | -0.032 |
moment | -0.031 |
Token | she |
Token ID | 673 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
Sleep | +0.055 |
polar | +0.055 |
Sweet | +0.054 |
ending | +0.053 |
mistake | +0.052 |
Neg logit contributions | |
ay | -0.053 |
ong | -0.050 |
rac | -0.049 |
around | -0.049 |
bs | -0.047 |
Token | The |
Token ID | 383 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
Sleep | +0.012 |
polar | +0.012 |
Sweet | +0.012 |
Kids | +0.011 |
moment | +0.011 |
Neg logit contributions | |
ay | -0.012 |
ong | -0.011 |
rac | -0.011 |
around | -0.011 |
issors | -0.011 |
Token | boat |
Token ID | 8848 |
Feature activation | -7.67e-01 |
Pos logit contributions | |
Sweet | +0.043 |
Sleep | +0.040 |
former | +0.038 |
Kids | +0.037 |
Jenkins | +0.036 |
Neg logit contributions | |
ay | -0.043 |
ong | -0.043 |
around | -0.042 |
rac | -0.042 |
ars | -0.040 |
Token | tried |
Token ID | 3088 |
Feature activation | +8.76e-01 |
Pos logit contributions | |
Sleep | +0.024 |
polar | +0.023 |
Sweet | +0.023 |
ending | +0.023 |
mistake | +0.023 |
Neg logit contributions | |
ay | -0.028 |
ong | -0.026 |
rac | -0.026 |
around | -0.025 |
issors | -0.025 |
Token | to |
Token ID | 284 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
around | +0.012 |
ong | +0.010 |
ay | +0.010 |
at | +0.010 |
rac | +0.009 |
Neg logit contributions | |
Sleep | -0.047 |
Kids | -0.047 |
polar | -0.047 |
Sweet | -0.046 |
ending | -0.045 |
Token | drop |
Token ID | 4268 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
Sweet | +0.004 |
Sleep | +0.004 |
ending | +0.004 |
Kids | +0.004 |
polar | +0.004 |
Neg logit contributions | |
ong | -0.004 |
ay | -0.004 |
rac | -0.004 |
around | -0.004 |
ars | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | +5.14e-01 |
Pos logit contributions | |
ay | +0.042 |
ong | +0.041 |
rac | +0.039 |
issors | +0.039 |
ars | +0.038 |
Neg logit contributions | |
polar | -0.042 |
Bird | -0.040 |
moment | -0.039 |
Sleep | -0.039 |
moral | -0.039 |
Token | anchor |
Token ID | 18021 |
Feature activation | +2.50e-01 |
Pos logit contributions | |
ay | +0.022 |
ong | +0.021 |
rac | +0.021 |
issors | +0.020 |
around | +0.020 |
Neg logit contributions | |
polar | -0.013 |
Sleep | -0.012 |
ending | -0.012 |
mistake | -0.012 |
moment | -0.012 |
Token | , |
Token ID | 11 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
ay | +0.008 |
rac | +0.007 |
ong | +0.007 |
issors | +0.007 |
ars | +0.007 |
Neg logit contributions | |
polar | -0.009 |
ending | -0.009 |
mistake | -0.009 |
moment | -0.009 |
Sleep | -0.009 |
Token | but |
Token ID | 475 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
ay | +0.008 |
ong | +0.008 |
rac | +0.008 |
around | +0.007 |
issors | +0.007 |
Neg logit contributions | |
Sleep | -0.006 |
polar | -0.006 |
Sweet | -0.006 |
ending | -0.006 |
mistake | -0.006 |
Token | it |
Token ID | 340 |
Feature activation | +3.41e-01 |
Pos logit contributions | |
ay | +0.042 |
ong | +0.039 |
rac | +0.037 |
issors | +0.036 |
around | +0.035 |
Neg logit contributions | |
polar | -0.057 |
Sleep | -0.057 |
ending | -0.056 |
Sweet | -0.056 |
mistake | -0.055 |
Token | was |
Token ID | 373 |
Feature activation | +6.58e-02 |
Pos logit contributions | |
ay | +0.013 |
ong | +0.013 |
rac | +0.012 |
around | +0.012 |
issors | +0.012 |
Neg logit contributions | |
Sleep | -0.010 |
Sweet | -0.010 |
polar | -0.009 |
ending | -0.009 |
mistake | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | +3.01e-01 |
Pos logit contributions | |
polar | +0.039 |
mistake | +0.039 |
Sleep | +0.038 |
ending | +0.038 |
moment | +0.038 |
Neg logit contributions | |
ay | -0.032 |
ong | -0.029 |
rac | -0.029 |
issors | -0.027 |
around | -0.027 |
Token | fish |
Token ID | 5916 |
Feature activation | -4.36e-01 |
Pos logit contributions | |
ong | +0.011 |
around | +0.011 |
ay | +0.011 |
at | +0.010 |
rac | +0.010 |
Neg logit contributions | |
Sleep | -0.009 |
Sweet | -0.009 |
former | -0.009 |
department | -0.008 |
tr | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
Sweet | +0.018 |
Sleep | +0.018 |
Kids | +0.018 |
polar | +0.018 |
ending | +0.018 |
Neg logit contributions | |
ay | -0.011 |
ong | -0.010 |
around | -0.010 |
rac | -0.010 |
bs | -0.009 |
Token | so |
Token ID | 523 |
Feature activation | -5.91e-01 |
Pos logit contributions | |
ay | +0.004 |
ong | +0.004 |
rac | +0.004 |
around | +0.004 |
issors | +0.004 |
Neg logit contributions | |
Sleep | -0.005 |
Sweet | -0.005 |
polar | -0.005 |
ending | -0.005 |
mistake | -0.005 |
Token | it |
Token ID | 340 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
Sleep | +0.021 |
Sweet | +0.021 |
polar | +0.021 |
ending | +0.021 |
mistake | +0.020 |
Neg logit contributions | |
ay | -0.018 |
ong | -0.018 |
rac | -0.017 |
around | -0.017 |
bs | -0.016 |
Token | said |
Token ID | 531 |
Feature activation | +9.53e-01 |
Pos logit contributions | |
around | +0.004 |
rac | +0.004 |
led | +0.004 |
bs | +0.004 |
ay | +0.004 |
Neg logit contributions | |
Sweet | -0.004 |
Sleep | -0.004 |
mistake | -0.004 |
ending | -0.004 |
forgiveness | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
ay | +0.028 |
ong | +0.027 |
rac | +0.027 |
around | +0.026 |
bs | +0.025 |
Neg logit contributions | |
Sleep | -0.035 |
polar | -0.035 |
Sweet | -0.035 |
Kids | -0.035 |
former | -0.034 |
Token | " |
Token ID | 366 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
ay | +0.006 |
ong | +0.006 |
rac | +0.006 |
issors | +0.006 |
ars | +0.006 |
Neg logit contributions | |
Sleep | -0.004 |
Sweet | -0.004 |
moment | -0.004 |
polar | -0.004 |
moral | -0.004 |
Token | I |
Token ID | 40 |
Feature activation | -3.50e-01 |
Pos logit contributions | |
ay | +0.064 |
ong | +0.060 |
rac | +0.060 |
around | +0.058 |
issors | +0.057 |
Neg logit contributions | |
polar | -0.040 |
ending | -0.039 |
Sleep | -0.039 |
Sweet | -0.038 |
mistake | -0.038 |
Token | will |
Token ID | 481 |
Feature activation | +3.51e-01 |
Pos logit contributions | |
Sleep | +0.011 |
Sweet | +0.011 |
polar | +0.011 |
ending | +0.011 |
mistake | +0.011 |
Neg logit contributions | |
ay | -0.012 |
ong | -0.011 |
around | -0.011 |
rac | -0.011 |
bs | -0.011 |
Token | lead |
Token ID | 1085 |
Feature activation | +6.91e-01 |
Pos logit contributions | |
ay | +0.013 |
ong | +0.012 |
rac | +0.012 |
issors | +0.011 |
around | +0.011 |
Neg logit contributions | |
polar | -0.011 |
Sleep | -0.011 |
ending | -0.011 |
mistake | -0.011 |
moment | -0.010 |
Token | Jack |
Token ID | 3619 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
around | +0.026 |
at | +0.023 |
ay | +0.021 |
rac | +0.021 |
bs | +0.021 |
Neg logit contributions | |
polar | -0.044 |
former | -0.043 |
department | -0.042 |
Sleep | -0.042 |
mistake | -0.041 |
Token | saw |
Token ID | 2497 |
Feature activation | +2.75e-01 |
Pos logit contributions | |
ay | +0.041 |
ong | +0.039 |
rac | +0.038 |
around | +0.037 |
issors | +0.036 |
Neg logit contributions | |
Sleep | -0.029 |
polar | -0.029 |
Sweet | -0.029 |
ending | -0.029 |
mistake | -0.028 |
Token | her |
Token ID | 607 |
Feature activation | +9.45e-01 |
Pos logit contributions | |
around | +0.006 |
ay | +0.006 |
ong | +0.005 |
rac | +0.005 |
bs | +0.005 |
Neg logit contributions | |
polar | -0.012 |
Sleep | -0.012 |
ending | -0.012 |
Sweet | -0.012 |
former | -0.012 |
Token | crying |
Token ID | 13774 |
Feature activation | -3.00e+00 |
Pos logit contributions | |
around | +0.032 |
ong | +0.031 |
ay | +0.031 |
at | +0.029 |
bs | +0.028 |
Neg logit contributions | |
Kids | -0.033 |
Sleep | -0.032 |
Sweet | -0.032 |
Are | -0.029 |
polar | -0.029 |
Token | and |
Token ID | 290 |
Feature activation | -5.32e-02 |
Pos logit contributions | |
Sweet | +0.144 |
Sleep | +0.142 |
polar | +0.140 |
Kids | +0.139 |
former | +0.136 |
Neg logit contributions | |
around | -0.062 |
ay | -0.061 |
ong | -0.056 |
at | -0.054 |
led | -0.048 |
Token | helped |
Token ID | 4193 |
Feature activation | -6.71e-01 |
Pos logit contributions | |
Sleep | +0.002 |
Sweet | +0.002 |
former | +0.002 |
ending | +0.002 |
moment | +0.002 |
Neg logit contributions | |
around | -0.001 |
at | -0.001 |
rac | -0.001 |
ong | -0.001 |
bs | -0.001 |
Token | her |
Token ID | 607 |
Feature activation | +9.62e-01 |
Pos logit contributions | |
Sleep | +0.026 |
polar | +0.026 |
Sweet | +0.026 |
ending | +0.026 |
mistake | +0.026 |
Neg logit contributions | |
ay | -0.019 |
ong | -0.017 |
rac | -0.017 |
around | -0.016 |
issors | -0.016 |
Token | pick |
Token ID | 2298 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
around | +0.029 |
ay | +0.029 |
ong | +0.028 |
rac | +0.027 |
bs | +0.026 |
Neg logit contributions | |
Sleep | -0.036 |
Sweet | -0.035 |
Kids | -0.034 |
ending | -0.034 |
polar | -0.034 |
Token | up |
Token ID | 510 |
Feature activation | +9.34e-01 |
Pos logit contributions | |
Sleep | +0.051 |
Sweet | +0.050 |
ending | +0.050 |
polar | +0.050 |
mistake | +0.049 |
Neg logit contributions | |
ay | -0.022 |
ong | -0.020 |
around | -0.019 |
rac | -0.018 |
issors | -0.017 |
Token | all |
Token ID | 477 |
Feature activation | +8.79e-01 |
Pos logit contributions | |
ay | +0.023 |
around | +0.022 |
ong | +0.022 |
rac | +0.020 |
bs | +0.020 |
Neg logit contributions | |
Sleep | -0.040 |
Sweet | -0.039 |
polar | -0.039 |
ending | -0.039 |
Kids | -0.038 |
Token | the |
Token ID | 262 |
Feature activation | +9.66e-02 |
Pos logit contributions | |
ay | +0.024 |
ong | +0.022 |
rac | +0.021 |
around | +0.021 |
issors | +0.020 |
Neg logit contributions | |
Sleep | -0.035 |
polar | -0.035 |
Sweet | -0.035 |
ending | -0.035 |
mistake | -0.034 |
Token | eyel |
Token ID | 29708 |
Feature activation | +6.86e-01 |
Pos logit contributions | |
ay | +0.044 |
ong | +0.043 |
rac | +0.042 |
ors | +0.040 |
ars | +0.040 |
Neg logit contributions | |
ending | -0.029 |
Sweet | -0.029 |
mistake | -0.029 |
Sleep | -0.028 |
moment | -0.028 |
Token | ids |
Token ID | 2340 |
Feature activation | +2.75e+00 |
Pos logit contributions | |
ay | +0.019 |
rac | +0.018 |
around | +0.017 |
ong | +0.017 |
ized | +0.017 |
Neg logit contributions | |
polar | -0.025 |
Sweet | -0.025 |
ending | -0.025 |
Sleep | -0.025 |
former | -0.025 |
Token | and |
Token ID | 290 |
Feature activation | +2.50e-01 |
Pos logit contributions | |
ay | +0.099 |
rac | +0.094 |
issors | +0.094 |
ong | +0.091 |
ars | +0.089 |
Neg logit contributions | |
moment | -0.081 |
again | -0.078 |
ending | -0.076 |
raining | -0.073 |
Sleep | -0.072 |
Token | cheeks |
Token ID | 25839 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
ay | +0.009 |
ong | +0.008 |
rac | +0.008 |
issors | +0.008 |
around | +0.008 |
Neg logit contributions | |
Sweet | -0.008 |
Sleep | -0.008 |
polar | -0.008 |
ending | -0.008 |
mistake | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -5.44e-02 |
Pos logit contributions | |
rac | +0.067 |
ay | +0.066 |
issors | +0.064 |
ars | +0.060 |
ong | +0.058 |
Neg logit contributions | |
again | -0.054 |
moment | -0.053 |
ending | -0.053 |
Sleep | -0.051 |
department | -0.050 |
Token | It |
Token ID | 632 |
Feature activation | -7.24e-01 |
Pos logit contributions | |
polar | +0.002 |
mistake | +0.002 |
ending | +0.002 |
Sleep | +0.002 |
divorce | +0.002 |
Neg logit contributions | |
ay | -0.002 |
around | -0.001 |
ong | -0.001 |
rac | -0.001 |
bs | -0.001 |
Token | was |
Token ID | 373 |
Feature activation | -8.07e-02 |
Pos logit contributions | |
Sleep | +0.022 |
polar | +0.021 |
Sweet | +0.021 |
ending | +0.020 |
mistake | +0.020 |
Neg logit contributions | |
ay | -0.027 |
around | -0.026 |
ong | -0.026 |
rac | -0.025 |
bs | -0.024 |
Token | fun |
Token ID | 1257 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
Sleep | +0.003 |
polar | +0.003 |
ending | +0.003 |
Sweet | +0.003 |
mistake | +0.003 |
Neg logit contributions | |
ay | -0.003 |
ong | -0.002 |
rac | -0.002 |
around | -0.002 |
issors | -0.002 |
Token | to |
Token ID | 284 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
rac | +0.032 |
ay | +0.032 |
ong | +0.031 |
issors | +0.029 |
ars | +0.029 |
Neg logit contributions | |
ending | -0.045 |
moment | -0.045 |
mistake | -0.045 |
Sweet | -0.044 |
Sleep | -0.043 |
Token | watch |
Token ID | 2342 |
Feature activation | -9.35e-01 |
Pos logit contributions | |
Sleep | +0.005 |
polar | +0.005 |
Sweet | +0.005 |
ending | +0.005 |
mistake | +0.005 |
Neg logit contributions | |
ay | -0.005 |
ong | -0.005 |
rac | -0.005 |
around | -0.005 |
issors | -0.004 |
Token | her |
Token ID | 607 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
ending | +0.032 |
Sleep | +0.032 |
Sweet | +0.032 |
polar | +0.032 |
mistake | +0.032 |
Neg logit contributions | |
ay | -0.028 |
rac | -0.027 |
ong | -0.027 |
ars | -0.026 |
issors | -0.026 |
Token | , |
Token ID | 11 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
Sleep | +0.062 |
polar | +0.062 |
Sweet | +0.061 |
ending | +0.061 |
mistake | +0.060 |
Neg logit contributions | |
ay | -0.040 |
ong | -0.038 |
rac | -0.036 |
around | -0.035 |
issors | -0.034 |
Token | but |
Token ID | 475 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
around | +0.013 |
ong | +0.012 |
ay | +0.011 |
rac | +0.011 |
at | +0.011 |
Neg logit contributions | |
Sleep | -0.016 |
Sweet | -0.016 |
Kids | -0.015 |
ending | -0.015 |
polar | -0.015 |
Token | she |
Token ID | 673 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
ay | +0.042 |
ong | +0.039 |
rac | +0.037 |
issors | +0.036 |
around | +0.036 |
Neg logit contributions | |
Sleep | -0.047 |
ending | -0.047 |
Sweet | -0.046 |
polar | -0.046 |
moment | -0.045 |
Token | was |
Token ID | 373 |
Feature activation | -9.86e-02 |
Pos logit contributions | |
ay | +0.068 |
ong | +0.064 |
rac | +0.062 |
issors | +0.060 |
around | +0.059 |
Neg logit contributions | |
polar | -0.041 |
Sleep | -0.039 |
ending | -0.039 |
Sweet | -0.038 |
mistake | -0.038 |
Token | a |
Token ID | 257 |
Feature activation | +4.01e-01 |
Pos logit contributions | |
Sleep | +0.003 |
Sweet | +0.003 |
polar | +0.003 |
ending | +0.003 |
Kids | +0.003 |
Neg logit contributions | |
ay | -0.003 |
ong | -0.003 |
around | -0.003 |
rac | -0.003 |
issors | -0.003 |
Token | little |
Token ID | 1310 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
around | +0.011 |
ay | +0.010 |
high | +0.010 |
bs | +0.010 |
at | +0.010 |
Neg logit contributions | |
Sleep | -0.015 |
former | -0.014 |
Sweet | -0.014 |
forgiveness | -0.014 |
riage | -0.014 |
Token | scared |
Token ID | 12008 |
Feature activation | -9.04e-03 |
Pos logit contributions | |
Sleep | +0.076 |
Sweet | +0.074 |
ending | +0.072 |
Kids | +0.071 |
moral | +0.070 |
Neg logit contributions | |
around | -0.051 |
ay | -0.049 |
ong | -0.048 |
bs | -0.047 |
at | -0.046 |
Token | . |
Token ID | 13 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
former | +0.000 |
Sleep | +0.000 |
Sweet | +0.000 |
ending | +0.000 |
mistake | +0.000 |
Neg logit contributions | |
ay | -0.000 |
around | -0.000 |
at | -0.000 |
ong | -0.000 |
bs | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
Sleep | +0.004 |
polar | +0.004 |
Sweet | +0.004 |
ending | +0.004 |
mistake | +0.004 |
Neg logit contributions | |
ay | -0.005 |
ong | -0.005 |
rac | -0.004 |
around | -0.004 |
issors | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
ong | +0.008 |
ay | +0.008 |
rac | +0.008 |
issors | +0.008 |
ors | +0.007 |
Neg logit contributions | |
Sleep | -0.005 |
Sweet | -0.005 |
moment | -0.005 |
polar | -0.004 |
Kids | -0.004 |
Token | " |
Token ID | 1 |
Feature activation | +2.49e+00 |
Pos logit contributions | |
ay | +0.005 |
ong | +0.005 |
rac | +0.005 |
issors | +0.005 |
ors | +0.005 |
Neg logit contributions | |
Sleep | -0.002 |
Sweet | -0.002 |
ending | -0.002 |
moment | -0.002 |
moral | -0.002 |
Token | friends |
Token ID | 2460 |
Feature activation | +5.64e-01 |
Pos logit contributions | |
Sleep | +0.007 |
polar | +0.006 |
Sweet | +0.006 |
ending | +0.006 |
mistake | +0.006 |
Neg logit contributions | |
ay | -0.007 |
ong | -0.007 |
rac | -0.007 |
around | -0.007 |
issors | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
ay | +0.017 |
ong | +0.016 |
rac | +0.016 |
issors | +0.015 |
around | +0.014 |
Neg logit contributions | |
Sleep | -0.021 |
ending | -0.021 |
Sweet | -0.021 |
polar | -0.021 |
mistake | -0.020 |
Token | sailed |
Token ID | 33835 |
Feature activation | +3.50e-01 |
Pos logit contributions | |
Sleep | +0.004 |
polar | +0.004 |
mistake | +0.004 |
Sweet | +0.004 |
moral | +0.004 |
Neg logit contributions | |
around | -0.003 |
bs | -0.003 |
ay | -0.003 |
at | -0.003 |
rac | -0.003 |
Token | on |
Token ID | 319 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
ong | +0.011 |
ay | +0.011 |
issors | +0.011 |
rac | +0.011 |
ors | +0.010 |
Neg logit contributions | |
moment | -0.012 |
ending | -0.012 |
polar | -0.011 |
Sleep | -0.011 |
mistake | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
ay | +0.007 |
around | +0.007 |
ong | +0.007 |
rac | +0.006 |
bs | +0.006 |
Neg logit contributions | |
Sleep | -0.009 |
polar | -0.009 |
Sweet | -0.009 |
ending | -0.009 |
Kids | -0.009 |
Token | lake |
Token ID | 13546 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
ong | +0.005 |
ay | +0.005 |
around | +0.005 |
rac | +0.004 |
ars | +0.004 |
Neg logit contributions | |
Sweet | -0.003 |
Sleep | -0.003 |
Kids | -0.003 |
Are | -0.003 |
department | -0.003 |
Token | together |
Token ID | 1978 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
ending | +0.054 |
polar | +0.054 |
Sleep | +0.054 |
moment | +0.053 |
mistake | +0.053 |
Neg logit contributions | |
ay | -0.038 |
ong | -0.035 |
rac | -0.035 |
issors | -0.033 |
bs | -0.031 |
Token | every |
Token ID | 790 |
Feature activation | +3.33e+00 |
Pos logit contributions | |
ending | +0.009 |
moment | +0.009 |
again | +0.008 |
mistake | +0.008 |
moral | +0.008 |
Neg logit contributions | |
ay | -0.008 |
rac | -0.008 |
issors | -0.008 |
ong | -0.008 |
ors | -0.007 |
Token | week |
Token ID | 1285 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
around | +0.102 |
ay | +0.100 |
at | +0.090 |
forward | +0.088 |
high | +0.086 |
Neg logit contributions | |
Sweet | -0.127 |
former | -0.126 |
charity | -0.123 |
department | -0.122 |
forgiveness | -0.122 |
Token | . |
Token ID | 13 |
Feature activation | -5.35e-01 |
Pos logit contributions | |
ending | +0.005 |
Sleep | +0.005 |
moment | +0.005 |
polar | +0.004 |
Sweet | +0.004 |
Neg logit contributions | |
ay | -0.004 |
ong | -0.003 |
rac | -0.003 |
issors | -0.003 |
ars | -0.003 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
mistake | +0.026 |
department | +0.025 |
polar | +0.024 |
gen | +0.024 |
forgiveness | +0.024 |
Neg logit contributions | |
ay | -0.013 |
around | -0.009 |
s | -0.009 |
bs | -0.008 |
ables | -0.008 |
Token | mar |
Token ID | 1667 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
ay | +0.044 |
ong | +0.042 |
rac | +0.041 |
around | +0.041 |
issors | +0.040 |
Neg logit contributions | |
Sleep | -0.034 |
Sweet | -0.034 |
polar | -0.033 |
ending | -0.033 |
mistake | -0.032 |
Token | bles |
Token ID | 7689 |
Feature activation | +2.29e+00 |
Pos logit contributions | |
rac | +0.040 |
ay | +0.039 |
ong | +0.039 |
bs | +0.038 |
ned | +0.038 |
Neg logit contributions | |
polar | -0.027 |
pear | -0.027 |
tunnel | -0.026 |
ulu | -0.025 |
island | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -3.80e-01 |
Pos logit contributions | |
ay | +0.066 |
ong | +0.061 |
rac | +0.060 |
issors | +0.056 |
around | +0.055 |
Neg logit contributions | |
Sleep | -0.087 |
polar | -0.086 |
ending | -0.085 |
Sweet | -0.085 |
mistake | -0.084 |
Token | She |
Token ID | 1375 |
Feature activation | +2.93e-01 |
Pos logit contributions | |
Sleep | +0.013 |
polar | +0.013 |
Sweet | +0.013 |
ending | +0.013 |
mistake | +0.013 |
Neg logit contributions | |
ay | -0.012 |
ong | -0.012 |
rac | -0.011 |
around | -0.011 |
issors | -0.011 |
Token | only |
Token ID | 691 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
ay | +0.012 |
ong | +0.012 |
rac | +0.011 |
issors | +0.011 |
ors | +0.011 |
Neg logit contributions | |
polar | -0.007 |
ending | -0.007 |
Sleep | -0.007 |
moment | -0.007 |
Sweet | -0.007 |
Token | had |
Token ID | 550 |
Feature activation | -7.78e-01 |
Pos logit contributions | |
Sleep | +0.039 |
Sweet | +0.038 |
polar | +0.037 |
ending | +0.037 |
mistake | +0.037 |
Neg logit contributions | |
ay | -0.045 |
ong | -0.042 |
around | -0.042 |
rac | -0.041 |
bs | -0.040 |
Token | two |
Token ID | 734 |
Feature activation | -2.77e+00 |
Pos logit contributions | |
Sleep | +0.027 |
Sweet | +0.026 |
polar | +0.026 |
ending | +0.026 |
mistake | +0.026 |
Neg logit contributions | |
ay | -0.025 |
ong | -0.024 |
around | -0.024 |
rac | -0.023 |
bs | -0.022 |
Token | mar |
Token ID | 1667 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
polar | +0.084 |
moral | +0.075 |
pand | +0.075 |
pear | +0.075 |
ending | +0.074 |
Neg logit contributions | |
ay | -0.104 |
rac | -0.103 |
ong | -0.101 |
issors | -0.098 |
ors | -0.094 |
Token | bles |
Token ID | 7689 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
rac | +0.040 |
ay | +0.039 |
ong | +0.039 |
ned | +0.038 |
issors | +0.037 |
Neg logit contributions | |
polar | -0.029 |
pear | -0.027 |
pand | -0.027 |
ulu | -0.026 |
gn | -0.025 |
Token | and |
Token ID | 290 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
ay | +0.060 |
rac | +0.055 |
ong | +0.055 |
issors | +0.051 |
around | +0.049 |
Neg logit contributions | |
polar | -0.093 |
ending | -0.092 |
Sleep | -0.092 |
Sweet | -0.091 |
mistake | -0.090 |
Token | enjoyed |
Token ID | 8359 |
Feature activation | -3.14e-01 |
Pos logit contributions | |
ay | +0.005 |
ong | +0.005 |
rac | +0.005 |
around | +0.005 |
issors | +0.004 |
Neg logit contributions | |
Sleep | -0.004 |
polar | -0.004 |
Sweet | -0.004 |
ending | -0.004 |
mistake | -0.004 |
Token | teeth |
Token ID | 9941 |
Feature activation | +3.61e+00 |
Pos logit contributions | |
Sleep | +0.005 |
Sweet | +0.005 |
Kids | +0.005 |
polar | +0.004 |
ending | +0.004 |
Neg logit contributions | |
ay | -0.005 |
ong | -0.005 |
around | -0.004 |
rac | -0.004 |
bs | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
ay | +0.106 |
rac | +0.099 |
ong | +0.098 |
issors | +0.094 |
bs | +0.088 |
Neg logit contributions | |
Sleep | -0.135 |
polar | -0.132 |
ending | -0.131 |
Sweet | -0.131 |
mistake | -0.130 |
Token |
|
Token ID | 198 |
Feature activation | +6.69e-01 |
Pos logit contributions | |
ay | +0.005 |
rac | +0.005 |
issors | +0.005 |
around | +0.005 |
ong | +0.004 |
Neg logit contributions | |
ending | -0.007 |
mistake | -0.007 |
Sleep | -0.007 |
polar | -0.007 |
Sweet | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | +5.19e-01 |
Pos logit contributions | |
ay | +0.023 |
ong | +0.021 |
rac | +0.020 |
around | +0.020 |
issors | +0.020 |
Neg logit contributions | |
ending | -0.023 |
polar | -0.022 |
Sleep | -0.022 |
Sweet | -0.022 |
department | -0.022 |
Token | " |
Token ID | 1 |
Feature activation | +2.49e+00 |
Pos logit contributions | |
ay | +0.016 |
rac | +0.015 |
ong | +0.015 |
issors | +0.015 |
ars | +0.015 |
Neg logit contributions | |
ending | -0.019 |
polar | -0.018 |
mistake | -0.018 |
department | -0.018 |
Sleep | -0.018 |
Token | Hey |
Token ID | 10814 |
Feature activation | -4.71e+00 |
Pos logit contributions | |
rac | +0.076 |
ay | +0.073 |
ars | +0.072 |
issors | +0.066 |
bs | +0.066 |
Neg logit contributions | |
mistake | -0.095 |
ending | -0.089 |
polar | -0.088 |
ig | -0.086 |
former | -0.084 |
Token | , |
Token ID | 11 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
Sleep | +0.237 |
Sweet | +0.232 |
former | +0.230 |
Kids | +0.223 |
ending | +0.221 |
Neg logit contributions | |
ay | -0.089 |
ong | -0.081 |
around | -0.075 |
rac | -0.072 |
at | -0.066 |
Token | stop |
Token ID | 2245 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
ay | +0.005 |
ong | +0.005 |
around | +0.005 |
rac | +0.004 |
at | +0.004 |
Neg logit contributions | |
former | -0.006 |
Sleep | -0.006 |
Sweet | -0.006 |
ending | -0.005 |
moment | -0.005 |
Token | that |
Token ID | 326 |
Feature activation | -9.49e-01 |
Pos logit contributions | |
Sleep | +0.097 |
Sweet | +0.096 |
ending | +0.095 |
polar | +0.095 |
mistake | +0.092 |
Neg logit contributions | |
ay | -0.051 |
ong | -0.047 |
around | -0.045 |
rac | -0.045 |
issors | -0.041 |
Token | !" |
Token ID | 2474 |
Feature activation | +2.62e-03 |
Pos logit contributions | |
Sleep | +0.045 |
Sweet | +0.045 |
Kids | +0.044 |
former | +0.043 |
polar | +0.043 |
Neg logit contributions | |
ay | -0.020 |
ong | -0.019 |
around | -0.018 |
rac | -0.015 |
at | -0.015 |
Token | their |
Token ID | 511 |
Feature activation | -9.09e-01 |
Pos logit contributions | |
ay | +0.000 |
ong | +0.000 |
rac | +0.000 |
around | +0.000 |
issors | +0.000 |
Neg logit contributions | |
Sleep | -0.000 |
polar | -0.000 |
Sweet | -0.000 |
ending | -0.000 |
mistake | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | +3.63e-01 |
Pos logit contributions | |
ay | +0.017 |
ong | +0.016 |
rac | +0.016 |
around | +0.015 |
issors | +0.015 |
Neg logit contributions | |
ending | -0.015 |
Sleep | -0.015 |
polar | -0.015 |
Sweet | -0.015 |
mistake | -0.015 |
Token | M |
Token ID | 44 |
Feature activation | -8.04e-02 |
Pos logit contributions | |
ay | +0.015 |
rac | +0.014 |
issors | +0.014 |
ars | +0.014 |
around | +0.014 |
Neg logit contributions | |
ending | -0.010 |
polar | -0.009 |
funeral | -0.009 |
fly | -0.009 |
department | -0.009 |
Token | oral |
Token ID | 6864 |
Feature activation | -4.41e+00 |
Pos logit contributions | |
Sleep | +0.002 |
polar | +0.002 |
Sweet | +0.002 |
moment | +0.002 |
Are | +0.002 |
Neg logit contributions | |
ay | -0.004 |
ong | -0.003 |
rac | -0.003 |
issors | -0.003 |
ars | -0.003 |
Token | of |
Token ID | 286 |
Feature activation | -2.14e+00 |
Pos logit contributions | |
Sleep | +0.182 |
polar | +0.180 |
Sweet | +0.180 |
ending | +0.179 |
mistake | +0.176 |
Neg logit contributions | |
ay | -0.114 |
ong | -0.107 |
rac | -0.103 |
around | -0.099 |
issors | -0.096 |
Token | the |
Token ID | 262 |
Feature activation | +9.08e-02 |
Pos logit contributions | |
department | +0.100 |
former | +0.100 |
ending | +0.097 |
Sweet | +0.096 |
polar | +0.096 |
Neg logit contributions | |
ay | -0.042 |
around | -0.042 |
rac | -0.035 |
ong | -0.033 |
at | -0.029 |
Token | story |
Token ID | 1621 |
Feature activation | -4.28e+00 |
Pos logit contributions | |
ay | +0.004 |
ong | +0.004 |
around | +0.004 |
issors | +0.003 |
rac | +0.003 |
Neg logit contributions | |
polar | -0.002 |
former | -0.002 |
fly | -0.002 |
Sleep | -0.002 |
Are | -0.002 |
Token | : |
Token ID | 25 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
Sleep | +0.157 |
polar | +0.156 |
Sweet | +0.155 |
ending | +0.154 |
mistake | +0.151 |
Neg logit contributions | |
ay | -0.130 |
ong | -0.122 |
rac | -0.118 |
around | -0.115 |
issors | -0.112 |
Token | It |
Token ID | 632 |
Feature activation | -8.56e-01 |
Pos logit contributions | |
polar | +0.077 |
ending | +0.074 |
Sweet | +0.072 |
Sleep | +0.072 |
Kids | +0.071 |
Neg logit contributions | |
ay | -0.072 |
around | -0.070 |
rac | -0.068 |
issors | -0.066 |
ors | -0.063 |
Token | 's |
Token ID | 338 |
Feature activation | -2.24e+00 |
Pos logit contributions | |
polar | +0.030 |
Sleep | +0.030 |
Sweet | +0.030 |
ending | +0.030 |
mistake | +0.029 |
Neg logit contributions | |
ay | -0.027 |
ong | -0.025 |
rac | -0.025 |
around | -0.024 |
issors | -0.023 |
Token | important |
Token ID | 1593 |
Feature activation | -9.70e-01 |
Pos logit contributions | |
polar | +0.084 |
Sweet | +0.082 |
Kids | +0.079 |
department | +0.078 |
Sleep | +0.078 |
Neg logit contributions | |
ay | -0.066 |
rac | -0.063 |
around | -0.062 |
issors | -0.061 |
ong | -0.058 |
Token | to |
Token ID | 284 |
Feature activation | -2.76e-01 |
Pos logit contributions | |
Sleep | +0.044 |
polar | +0.043 |
Sweet | +0.043 |
ending | +0.043 |
mistake | +0.042 |
Neg logit contributions | |
ay | -0.022 |
ong | -0.020 |
rac | -0.019 |
around | -0.018 |
issors | -0.018 |
Token | would |
Token ID | 561 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
ay | +0.051 |
around | +0.049 |
ong | +0.047 |
rac | +0.047 |
bs | +0.047 |
Neg logit contributions | |
polar | -0.046 |
ending | -0.045 |
Sleep | -0.045 |
Sweet | -0.044 |
moment | -0.044 |
Token | not |
Token ID | 407 |
Feature activation | -7.13e-01 |
Pos logit contributions | |
ay | +0.039 |
ong | +0.037 |
around | +0.036 |
rac | +0.036 |
issors | +0.036 |
Neg logit contributions | |
Sweet | -0.036 |
polar | -0.035 |
Sleep | -0.035 |
Kids | -0.034 |
ending | -0.034 |
Token | touch |
Token ID | 3638 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
Sweet | +0.026 |
Sleep | +0.025 |
polar | +0.025 |
Kids | +0.024 |
ending | +0.024 |
Neg logit contributions | |
ay | -0.023 |
around | -0.022 |
ong | -0.022 |
bs | -0.021 |
issors | -0.021 |
Token | her |
Token ID | 607 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
Sleep | +0.044 |
Sweet | +0.043 |
ending | +0.043 |
polar | +0.043 |
mistake | +0.042 |
Neg logit contributions | |
ay | -0.037 |
ong | -0.035 |
around | -0.034 |
rac | -0.033 |
issors | -0.032 |
Token | makeup |
Token ID | 16029 |
Feature activation | -2.55e+00 |
Pos logit contributions | |
ay | +0.035 |
around | +0.033 |
ong | +0.032 |
bs | +0.029 |
at | +0.029 |
Neg logit contributions | |
Sleep | -0.051 |
Sweet | -0.050 |
polar | -0.050 |
Kids | -0.049 |
ending | -0.047 |
Token | again |
Token ID | 757 |
Feature activation | -4.94e+00 |
Pos logit contributions | |
Sleep | +0.055 |
ending | +0.055 |
mistake | +0.054 |
Sweet | +0.054 |
polar | +0.054 |
Neg logit contributions | |
ay | -0.113 |
ong | -0.108 |
rac | -0.108 |
issors | -0.104 |
around | -0.103 |
Token | . |
Token ID | 13 |
Feature activation | +2.99e-01 |
Pos logit contributions | |
Sleep | +0.204 |
polar | +0.202 |
Sweet | +0.202 |
ending | +0.199 |
mistake | +0.194 |
Neg logit contributions | |
ay | -0.128 |
ong | -0.119 |
rac | -0.114 |
around | -0.112 |
issors | -0.107 |
Token | Sara |
Token ID | 24799 |
Feature activation | +4.28e-01 |
Pos logit contributions | |
ay | +0.008 |
ong | +0.007 |
rac | +0.007 |
around | +0.007 |
issors | +0.007 |
Neg logit contributions | |
polar | -0.012 |
ending | -0.012 |
Sleep | -0.011 |
Sweet | -0.011 |
mistake | -0.011 |
Token | forg |
Token ID | 5784 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
ay | +0.013 |
around | +0.013 |
rac | +0.013 |
ong | +0.012 |
ars | +0.012 |
Neg logit contributions | |
polar | -0.015 |
ending | -0.015 |
Sleep | -0.015 |
former | -0.015 |
Sweet | -0.014 |
Token | ave |
Token ID | 1015 |
Feature activation | -3.76e-01 |
Pos logit contributions | |
Sleep | +0.078 |
Sweet | +0.076 |
Kids | +0.075 |
ending | +0.075 |
forgiveness | +0.072 |
Neg logit contributions | |
ay | -0.048 |
rac | -0.047 |
ong | -0.046 |
issors | -0.044 |
itching | -0.043 |
Token | him |
Token ID | 683 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
Sleep | +0.018 |
ending | +0.018 |
Sweet | +0.018 |
polar | +0.018 |
mistake | +0.018 |
Neg logit contributions | |
ay | -0.007 |
ong | -0.006 |
rac | -0.006 |
around | -0.006 |
issors | -0.005 |
Token | for |
Token ID | 329 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
ay | +0.032 |
around | +0.031 |
ong | +0.030 |
bs | +0.028 |
rac | +0.028 |
Neg logit contributions | |
Sleep | -0.057 |
Sweet | -0.057 |
polar | -0.056 |
Kids | -0.056 |
ending | -0.054 |
Token | selling |
Token ID | 6301 |
Feature activation | -3.40e+00 |
Pos logit contributions | |
ong | +0.071 |
around | +0.069 |
ay | +0.068 |
rac | +0.067 |
at | +0.066 |
Neg logit contributions | |
Sweet | -0.086 |
Sleep | -0.084 |
Kids | -0.083 |
polar | -0.083 |
ending | -0.080 |
Token | them |
Token ID | 606 |
Feature activation | +3.56e-01 |
Pos logit contributions | |
mistake | +0.093 |
polar | +0.092 |
forgiveness | +0.089 |
Sleep | +0.089 |
ending | +0.089 |
Neg logit contributions | |
ay | -0.134 |
rac | -0.127 |
ong | -0.126 |
issors | -0.124 |
ors | -0.120 |
Token | a |
Token ID | 257 |
Feature activation | +6.52e-02 |
Pos logit contributions | |
ay | +0.009 |
ong | +0.009 |
rac | +0.008 |
around | +0.008 |
issors | +0.008 |
Neg logit contributions | |
Sleep | -0.015 |
polar | -0.015 |
Sweet | -0.015 |
ending | -0.014 |
mistake | -0.014 |
Token | bad |
Token ID | 2089 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
ay | +0.002 |
ong | +0.002 |
around | +0.002 |
bs | +0.002 |
high | +0.002 |
Neg logit contributions | |
Kids | -0.002 |
Sweet | -0.002 |
Sleep | -0.002 |
former | -0.002 |
polar | -0.002 |
Token | movie |
Token ID | 3807 |
Feature activation | -4.38e+00 |
Pos logit contributions | |
Sweet | +0.051 |
Sleep | +0.047 |
former | +0.047 |
Kids | +0.046 |
Are | +0.045 |
Neg logit contributions | |
ay | -0.067 |
around | -0.063 |
ong | -0.061 |
at | -0.058 |
ars | -0.057 |
Token | . |
Token ID | 13 |
Feature activation | -3.09e-01 |
Pos logit contributions | |
Sleep | +0.169 |
polar | +0.168 |
ending | +0.167 |
Sweet | +0.166 |
mistake | +0.164 |
Neg logit contributions | |
ay | -0.123 |
ong | -0.115 |
rac | -0.112 |
around | -0.106 |
issors | -0.106 |
Token | The |
Token ID | 383 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
Sweet | +0.012 |
moment | +0.012 |
Sleep | +0.012 |
polar | +0.012 |
mistake | +0.012 |
Neg logit contributions | |
ay | -0.008 |
issors | -0.007 |
rac | -0.007 |
around | -0.007 |
ong | -0.007 |
Token | man |
Token ID | 582 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
Sweet | +0.050 |
Sleep | +0.048 |
former | +0.047 |
fly | +0.047 |
polar | +0.046 |
Neg logit contributions | |
ay | -0.055 |
ong | -0.052 |
around | -0.051 |
ars | -0.051 |
rac | -0.049 |
Token | felt |
Token ID | 2936 |
Feature activation | +2.76e-01 |
Pos logit contributions | |
ay | +0.042 |
around | +0.041 |
ong | +0.039 |
rac | +0.039 |
ars | +0.038 |
Neg logit contributions | |
Sleep | -0.045 |
Sweet | -0.044 |
polar | -0.043 |
ending | -0.043 |
former | -0.042 |
Token | very |
Token ID | 845 |
Feature activation | +8.03e-01 |
Pos logit contributions | |
ay | +0.010 |
ong | +0.009 |
rac | +0.009 |
issors | +0.009 |
around | +0.009 |
Neg logit contributions | |
Sleep | -0.009 |
polar | -0.009 |
ending | -0.009 |
Sweet | -0.008 |
mistake | -0.008 |
Token | drink |
Token ID | 4144 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
polar | +0.009 |
Sleep | +0.009 |
Sweet | +0.009 |
ending | +0.009 |
mistake | +0.009 |
Neg logit contributions | |
ay | -0.013 |
ong | -0.012 |
rac | -0.012 |
around | -0.012 |
issors | -0.012 |
Token | water |
Token ID | 1660 |
Feature activation | +4.20e-01 |
Pos logit contributions | |
Sleep | +0.038 |
polar | +0.038 |
Sweet | +0.037 |
ending | +0.037 |
mistake | +0.036 |
Neg logit contributions | |
ay | -0.043 |
ong | -0.040 |
around | -0.039 |
rac | -0.039 |
issors | -0.037 |
Token | from |
Token ID | 422 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
ay | +0.014 |
ong | +0.013 |
rac | +0.013 |
issors | +0.012 |
ars | +0.012 |
Neg logit contributions | |
Sleep | -0.014 |
polar | -0.014 |
ending | -0.014 |
moment | -0.013 |
mistake | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
Sleep | +0.046 |
Sweet | +0.046 |
polar | +0.045 |
ending | +0.045 |
mistake | +0.044 |
Neg logit contributions | |
ay | -0.034 |
ong | -0.032 |
around | -0.031 |
rac | -0.031 |
issors | -0.029 |
Token | f |
Token ID | 277 |
Feature activation | -7.91e-01 |
Pos logit contributions | |
ay | +0.006 |
around | +0.006 |
ong | +0.006 |
high | +0.005 |
rac | +0.005 |
Neg logit contributions | |
Sweet | -0.005 |
mistake | -0.005 |
Sleep | -0.005 |
forgiveness | -0.005 |
ending | -0.005 |
Token | auc |
Token ID | 14272 |
Feature activation | -5.06e+00 |
Pos logit contributions | |
Sleep | +0.011 |
Kids | +0.011 |
polar | +0.011 |
Sweet | +0.011 |
department | +0.010 |
Neg logit contributions | |
ay | -0.040 |
rac | -0.039 |
ong | -0.037 |
around | -0.037 |
ars | -0.037 |
Token | et |
Token ID | 316 |
Feature activation | +2.32e+00 |
Pos logit contributions | |
Sleep | +0.181 |
polar | +0.179 |
ending | +0.179 |
mistake | +0.177 |
Kids | +0.177 |
Neg logit contributions | |
ay | -0.157 |
ong | -0.147 |
rac | -0.142 |
issors | -0.127 |
around | -0.125 |
Token | but |
Token ID | 475 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
ay | +0.064 |
ong | +0.060 |
rac | +0.057 |
around | +0.056 |
issors | +0.054 |
Neg logit contributions | |
Sleep | -0.092 |
polar | -0.091 |
Sweet | -0.091 |
ending | -0.090 |
mistake | -0.089 |
Token | she |
Token ID | 673 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
ay | +0.042 |
ong | +0.041 |
rac | +0.039 |
issors | +0.038 |
around | +0.037 |
Neg logit contributions | |
Sleep | -0.036 |
ending | -0.035 |
Sweet | -0.035 |
polar | -0.035 |
mistake | -0.034 |
Token | was |
Token ID | 373 |
Feature activation | -2.23e-01 |
Pos logit contributions | |
ay | +0.051 |
ong | +0.049 |
rac | +0.048 |
around | +0.047 |
issors | +0.046 |
Neg logit contributions | |
Sleep | -0.038 |
polar | -0.038 |
Sweet | -0.037 |
ending | -0.037 |
mistake | -0.036 |
Token | too |
Token ID | 1165 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
polar | +0.008 |
Sleep | +0.008 |
Sweet | +0.008 |
ending | +0.008 |
Kids | +0.008 |
Neg logit contributions | |
ay | -0.007 |
around | -0.006 |
ong | -0.006 |
rac | -0.006 |
issors | -0.006 |