Token | more |
Token ID | 517 |
Feature activation | +6.51e-01 |
Pos logit contributions | |
itating | +0.047 |
arre | +0.047 |
melon | +0.045 |
up | +0.045 |
ling | +0.044 |
Neg logit contributions | |
year | -0.053 |
ird | -0.050 |
glass | -0.049 |
hours | -0.048 |
aster | -0.047 |
Token | careful |
Token ID | 8161 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
itating | +0.021 |
arre | +0.021 |
melon | +0.021 |
up | +0.020 |
inas | +0.020 |
Neg logit contributions | |
year | -0.013 |
hours | -0.012 |
ird | -0.012 |
minutes | -0.012 |
glass | -0.011 |
Token | when |
Token ID | 618 |
Feature activation | +3.70e-01 |
Pos logit contributions | |
arre | +0.047 |
itating | +0.044 |
rust | +0.044 |
melon | +0.043 |
ucker | +0.040 |
Neg logit contributions | |
year | -0.066 |
glass | -0.065 |
hours | -0.063 |
ird | -0.062 |
minutes | -0.060 |
Token | he |
Token ID | 339 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
itating | +0.008 |
arre | +0.008 |
up | +0.008 |
melon | +0.008 |
ling | +0.008 |
Neg logit contributions | |
year | -0.011 |
glass | -0.011 |
ird | -0.011 |
hours | -0.010 |
aster | -0.010 |
Token | picked |
Token ID | 6497 |
Feature activation | +2.96e+00 |
Pos logit contributions | |
arre | +0.003 |
up | +0.003 |
melon | +0.002 |
ling | +0.002 |
itating | +0.002 |
Neg logit contributions | |
year | -0.005 |
glass | -0.004 |
ird | -0.004 |
hours | -0.004 |
minutes | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | +4.62e+00 |
Pos logit contributions | |
itating | +0.066 |
inas | +0.063 |
arre | +0.060 |
ivering | +0.057 |
lings | +0.056 |
Neg logit contributions | |
year | -0.086 |
hours | -0.085 |
aster | -0.084 |
minutes | -0.080 |
kil | -0.079 |
Token | squash |
Token ID | 34613 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
inas | +0.108 |
ocked | +0.106 |
ouncing | +0.104 |
ucker | +0.103 |
ivering | +0.102 |
Neg logit contributions | |
hours | -0.099 |
year | -0.094 |
month | -0.093 |
minutes | -0.091 |
20 | -0.091 |
Token | and |
Token ID | 290 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
arre | +0.037 |
itating | +0.036 |
ucker | +0.032 |
rust | +0.032 |
fighting | +0.032 |
Neg logit contributions | |
year | -0.052 |
ird | -0.051 |
glass | -0.051 |
hours | -0.048 |
aster | -0.048 |
Token | make |
Token ID | 787 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
itating | +0.042 |
arre | +0.041 |
up | +0.040 |
melon | +0.039 |
ling | +0.039 |
Neg logit contributions | |
year | -0.063 |
glass | -0.061 |
ird | -0.061 |
hours | -0.057 |
aster | -0.056 |
Token | sure |
Token ID | 1654 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
itating | +0.041 |
arre | +0.039 |
melon | +0.038 |
flower | +0.038 |
inas | +0.037 |
Neg logit contributions | |
year | -0.041 |
hours | -0.040 |
glass | -0.039 |
aster | -0.039 |
ird | -0.038 |
Token | only |
Token ID | 691 |
Feature activation | +5.93e-01 |
Pos logit contributions | |
arre | +0.005 |
itating | +0.005 |
ling | +0.005 |
melon | +0.005 |
bing | +0.005 |
Neg logit contributions | |
year | -0.008 |
glass | -0.007 |
ird | -0.007 |
hours | -0.007 |
aster | -0.007 |
Token | How |
Token ID | 1374 |
Feature activation | -2.80e+00 |
Pos logit contributions | |
ling | +0.021 |
itating | +0.019 |
up | +0.019 |
arre | +0.018 |
melon | +0.018 |
Neg logit contributions | |
year | -0.027 |
ird | -0.026 |
glass | -0.025 |
ris | -0.023 |
hours | -0.023 |
Token | about |
Token ID | 546 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
year | +0.085 |
glass | +0.081 |
ird | +0.081 |
hours | +0.077 |
aster | +0.075 |
Neg logit contributions | |
arre | -0.063 |
itating | -0.063 |
up | -0.062 |
melon | -0.060 |
ling | -0.059 |
Token | we |
Token ID | 356 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
ling | +0.041 |
itating | +0.038 |
flower | +0.038 |
ivering | +0.036 |
arre | +0.036 |
Neg logit contributions | |
ird | -0.037 |
glass | -0.036 |
lan | -0.033 |
hours | -0.032 |
Grand | -0.031 |
Token | clean |
Token ID | 3424 |
Feature activation | +2.83e+00 |
Pos logit contributions | |
bing | +0.067 |
ling | +0.066 |
fl | +0.063 |
ning | +0.063 |
ping | +0.063 |
Neg logit contributions | |
ird | -0.044 |
glass | -0.034 |
year | -0.033 |
aster | -0.032 |
ris | -0.031 |
Token | your |
Token ID | 534 |
Feature activation | +3.07e+00 |
Pos logit contributions | |
itating | +0.045 |
lings | +0.040 |
flower | +0.039 |
melon | +0.036 |
arre | +0.035 |
Neg logit contributions | |
glass | -0.096 |
year | -0.093 |
hours | -0.092 |
ird | -0.090 |
minutes | -0.088 |
Token | old |
Token ID | 1468 |
Feature activation | +4.61e+00 |
Pos logit contributions | |
flower | +0.082 |
itating | +0.082 |
mill | +0.079 |
ter | +0.077 |
inas | +0.073 |
Neg logit contributions | |
hours | -0.067 |
customers | -0.066 |
allowance | -0.066 |
desks | -0.065 |
glass | -0.065 |
Token | basketball |
Token ID | 9669 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
itating | +0.121 |
ocked | +0.109 |
ucker | +0.100 |
lings | +0.100 |
inated | +0.100 |
Neg logit contributions | |
ird | -0.107 |
hours | -0.104 |
glass | -0.104 |
books | -0.102 |
desks | -0.099 |
Token | and |
Token ID | 290 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
itating | +0.035 |
arre | +0.032 |
ling | +0.032 |
rust | +0.031 |
melon | +0.031 |
Neg logit contributions | |
ird | -0.049 |
glass | -0.048 |
year | -0.048 |
hours | -0.045 |
aster | -0.044 |
Token | make |
Token ID | 787 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
itating | +0.055 |
arre | +0.052 |
ling | +0.051 |
melon | +0.050 |
flower | +0.050 |
Neg logit contributions | |
ird | -0.063 |
year | -0.062 |
glass | -0.060 |
hours | -0.059 |
minutes | -0.057 |
Token | it |
Token ID | 340 |
Feature activation | +3.44e+00 |
Pos logit contributions | |
itating | +0.031 |
melon | +0.030 |
flower | +0.029 |
arre | +0.028 |
ling | +0.028 |
Neg logit contributions | |
year | -0.040 |
ird | -0.038 |
glass | -0.038 |
hours | -0.037 |
aster | -0.036 |
Token | look |
Token ID | 804 |
Feature activation | +9.34e-01 |
Pos logit contributions | |
itating | +0.095 |
ican | +0.093 |
bing | +0.092 |
lings | +0.089 |
flower | +0.089 |
Neg logit contributions | |
ird | -0.074 |
aster | -0.067 |
year | -0.064 |
hours | -0.063 |
glass | -0.063 |
Token | make |
Token ID | 787 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
itating | +0.042 |
arre | +0.041 |
up | +0.040 |
melon | +0.039 |
ling | +0.039 |
Neg logit contributions | |
year | -0.063 |
glass | -0.061 |
ird | -0.061 |
hours | -0.057 |
aster | -0.056 |
Token | sure |
Token ID | 1654 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
itating | +0.041 |
arre | +0.039 |
melon | +0.038 |
flower | +0.038 |
inas | +0.037 |
Neg logit contributions | |
year | -0.041 |
hours | -0.040 |
glass | -0.039 |
aster | -0.039 |
ird | -0.038 |
Token | only |
Token ID | 691 |
Feature activation | +5.93e-01 |
Pos logit contributions | |
arre | +0.005 |
itating | +0.005 |
ling | +0.005 |
melon | +0.005 |
bing | +0.005 |
Neg logit contributions | |
year | -0.008 |
glass | -0.007 |
ird | -0.007 |
hours | -0.007 |
aster | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
itating | +0.012 |
arre | +0.012 |
ling | +0.012 |
up | +0.011 |
bing | +0.011 |
Neg logit contributions | |
year | -0.019 |
ird | -0.018 |
glass | -0.018 |
hours | -0.017 |
aster | -0.017 |
Token | pick |
Token ID | 2298 |
Feature activation | +3.02e+00 |
Pos logit contributions | |
arre | +0.041 |
itating | +0.040 |
ling | +0.037 |
bing | +0.036 |
sets | +0.036 |
Neg logit contributions | |
aster | -0.042 |
ird | -0.042 |
year | -0.040 |
glass | -0.038 |
hours | -0.038 |
Token | the |
Token ID | 262 |
Feature activation | +4.60e+00 |
Pos logit contributions | |
itating | +0.074 |
inas | +0.071 |
arre | +0.068 |
flower | +0.065 |
ivering | +0.064 |
Neg logit contributions | |
year | -0.082 |
hours | -0.079 |
aster | -0.077 |
minutes | -0.073 |
kil | -0.072 |
Token | ones |
Token ID | 3392 |
Feature activation | +3.53e-01 |
Pos logit contributions | |
inas | +0.106 |
ocked | +0.104 |
ucker | +0.098 |
ouncing | +0.097 |
itating | +0.096 |
Neg logit contributions | |
hours | -0.106 |
minutes | -0.103 |
month | -0.097 |
year | -0.095 |
customers | -0.095 |
Token | that |
Token ID | 326 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
arre | +0.008 |
itating | +0.008 |
inas | +0.007 |
rust | +0.007 |
lings | +0.007 |
Neg logit contributions | |
year | -0.010 |
ird | -0.010 |
glass | -0.010 |
aster | -0.009 |
kil | -0.009 |
Token | were |
Token ID | 547 |
Feature activation | +2.27e+00 |
Pos logit contributions | |
itating | +0.039 |
arre | +0.039 |
ling | +0.036 |
bing | +0.036 |
up | +0.036 |
Neg logit contributions | |
year | -0.051 |
ird | -0.050 |
glass | -0.049 |
hours | -0.047 |
aster | -0.047 |
Token | sweet |
Token ID | 6029 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
arre | +0.049 |
itating | +0.049 |
ling | +0.043 |
up | +0.043 |
bing | +0.043 |
Neg logit contributions | |
year | -0.071 |
ird | -0.066 |
hours | -0.066 |
glass | -0.065 |
minutes | -0.064 |
Token | and |
Token ID | 290 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
itating | +0.035 |
arre | +0.032 |
rust | +0.030 |
bing | +0.029 |
melon | +0.029 |
Neg logit contributions | |
year | -0.045 |
ird | -0.044 |
hours | -0.043 |
minutes | -0.043 |
glass | -0.042 |
Token | was |
Token ID | 373 |
Feature activation | +2.18e+00 |
Pos logit contributions | |
itating | +0.001 |
bing | +0.001 |
melon | +0.001 |
up | +0.001 |
sets | +0.001 |
Neg logit contributions | |
year | -0.001 |
ird | -0.001 |
glass | -0.001 |
ris | -0.001 |
hours | -0.001 |
Token | wrong |
Token ID | 2642 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
bing | +0.049 |
itating | +0.049 |
sets | +0.049 |
flower | +0.046 |
ling | +0.045 |
Neg logit contributions | |
year | -0.066 |
ird | -0.065 |
minutes | -0.061 |
aster | -0.060 |
glass | -0.059 |
Token | . |
Token ID | 13 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
itating | +0.036 |
arre | +0.036 |
melon | +0.035 |
up | +0.034 |
ling | +0.033 |
Neg logit contributions | |
year | -0.052 |
ird | -0.050 |
glass | -0.050 |
hours | -0.047 |
minutes | -0.047 |
Token | I |
Token ID | 314 |
Feature activation | -3.25e-01 |
Pos logit contributions | |
itating | +0.006 |
arre | +0.006 |
up | +0.005 |
melon | +0.005 |
ling | +0.005 |
Neg logit contributions | |
year | -0.010 |
glass | -0.010 |
ird | -0.010 |
hours | -0.009 |
aster | -0.009 |
Token | wanted |
Token ID | 2227 |
Feature activation | +5.80e-01 |
Pos logit contributions | |
year | +0.011 |
ird | +0.010 |
glass | +0.010 |
hours | +0.009 |
ris | +0.009 |
Neg logit contributions | |
itating | -0.007 |
arre | -0.006 |
up | -0.006 |
melon | -0.006 |
bing | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | +4.47e+00 |
Pos logit contributions | |
itating | +0.015 |
arre | +0.014 |
ling | +0.014 |
bing | +0.013 |
melon | +0.013 |
Neg logit contributions | |
year | -0.016 |
glass | -0.015 |
ird | -0.015 |
hours | -0.015 |
aster | -0.015 |
Token | big |
Token ID | 1263 |
Feature activation | +3.15e+00 |
Pos logit contributions | |
bing | +0.120 |
itating | +0.112 |
ocked | +0.111 |
inas | +0.111 |
en | +0.105 |
Neg logit contributions | |
passport | -0.096 |
minutes | -0.090 |
hours | -0.090 |
delivery | -0.089 |
month | -0.088 |
Token | squash |
Token ID | 34613 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
itating | +0.075 |
ocked | +0.066 |
inas | +0.066 |
bing | +0.062 |
arre | +0.059 |
Neg logit contributions | |
minutes | -0.093 |
hours | -0.084 |
aster | -0.084 |
ird | -0.081 |
hour | -0.080 |
Token | so |
Token ID | 523 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
itating | +0.042 |
arre | +0.041 |
ling | +0.038 |
bing | +0.038 |
up | +0.037 |
Neg logit contributions | |
year | -0.059 |
ird | -0.058 |
glass | -0.057 |
hours | -0.054 |
aster | -0.053 |
Token | much |
Token ID | 881 |
Feature activation | -7.09e-01 |
Pos logit contributions | |
year | +0.004 |
ird | +0.004 |
glass | +0.004 |
hours | +0.004 |
aster | +0.003 |
Neg logit contributions | |
arre | -0.007 |
itating | -0.007 |
up | -0.007 |
melon | -0.007 |
ling | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -5.92e-02 |
Pos logit contributions | |
year | +0.024 |
glass | +0.023 |
ird | +0.023 |
hours | +0.022 |
aster | +0.021 |
Neg logit contributions | |
itating | -0.014 |
arre | -0.014 |
up | -0.013 |
melon | -0.013 |
ling | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | +6.05e-01 |
Pos logit contributions | |
year | +0.002 |
ird | +0.002 |
glass | +0.002 |
hours | +0.002 |
minutes | +0.002 |
Neg logit contributions | |
itating | -0.002 |
arre | -0.002 |
rust | -0.001 |
ling | -0.001 |
bing | -0.001 |
Token | " |
Token ID | 366 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
arre | +0.018 |
rust | +0.017 |
itating | +0.016 |
ling | +0.016 |
ucker | +0.016 |
Neg logit contributions | |
year | -0.014 |
glass | -0.013 |
ird | -0.012 |
hours | -0.012 |
minutes | -0.012 |
Token | We |
Token ID | 1135 |
Feature activation | +6.34e-01 |
Pos logit contributions | |
up | +0.034 |
arre | +0.031 |
ling | +0.031 |
fighting | +0.031 |
melon | +0.030 |
Neg logit contributions | |
year | -0.047 |
glass | -0.045 |
ird | -0.043 |
Ah | -0.042 |
hours | -0.042 |
Token | can |
Token ID | 460 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
itating | +0.017 |
bing | +0.016 |
up | +0.015 |
ling | +0.015 |
mill | +0.015 |
Neg logit contributions | |
ird | -0.017 |
year | -0.016 |
glass | -0.015 |
hours | -0.015 |
ris | -0.014 |
Token | make |
Token ID | 787 |
Feature activation | +2.48e+00 |
Pos logit contributions | |
arre | +0.052 |
rust | +0.050 |
itating | +0.050 |
ling | +0.049 |
accidents | +0.048 |
Neg logit contributions | |
ird | -0.049 |
aster | -0.045 |
lock | -0.043 |
glass | -0.042 |
ris | -0.041 |
Token | the |
Token ID | 262 |
Feature activation | +4.45e+00 |
Pos logit contributions | |
itating | +0.070 |
flower | +0.062 |
arre | +0.061 |
inas | +0.060 |
mill | +0.060 |
Neg logit contributions | |
hours | -0.061 |
glass | -0.061 |
ird | -0.061 |
year | -0.060 |
aster | -0.060 |
Token | peanut |
Token ID | 26636 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
itating | +0.127 |
ocked | +0.115 |
bing | +0.115 |
mill | +0.114 |
inated | +0.107 |
Neg logit contributions | |
hours | -0.109 |
glass | -0.108 |
lan | -0.098 |
ird | -0.097 |
minutes | -0.097 |
Token | butter |
Token ID | 9215 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
itating | +0.045 |
arre | +0.040 |
Sharing | +0.039 |
Fighting | +0.039 |
Fighting | +0.039 |
Neg logit contributions | |
ird | -0.042 |
glass | -0.040 |
year | -0.040 |
aster | -0.039 |
kil | -0.039 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
itating | +0.052 |
arre | +0.050 |
rust | +0.048 |
fighting | +0.046 |
ocked | +0.046 |
Neg logit contributions | |
ird | -0.049 |
glass | -0.048 |
year | -0.047 |
aster | -0.046 |
book | -0.044 |
Token | |
Token ID | 220 |
Feature activation | +1.88e-02 |
Pos logit contributions | |
ling | +0.031 |
itating | +0.030 |
bing | +0.029 |
arre | +0.029 |
ivering | +0.028 |
Neg logit contributions | |
glass | -0.037 |
year | -0.035 |
hours | -0.033 |
ird | -0.033 |
lock | -0.032 |
Token |
|
Token ID | 198 |
Feature activation | -5.78e-01 |
Pos logit contributions | |
ling | +0.000 |
arre | +0.000 |
melon | +0.000 |
itating | +0.000 |
flower | +0.000 |
Neg logit contributions | |
year | -0.001 |
ird | -0.001 |
glass | -0.001 |
ibr | -0.000 |
aster | -0.000 |
Token | again |
Token ID | 757 |
Feature activation | +8.94e-01 |
Pos logit contributions | |
itating | +0.047 |
arre | +0.046 |
melon | +0.043 |
rust | +0.042 |
ling | +0.041 |
Neg logit contributions | |
year | -0.086 |
ird | -0.082 |
glass | -0.082 |
hours | -0.079 |
minutes | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | +3.07e-01 |
Pos logit contributions | |
up | +0.013 |
ling | +0.013 |
bing | +0.012 |
itating | +0.012 |
arre | +0.012 |
Neg logit contributions | |
year | -0.036 |
glass | -0.033 |
ird | -0.033 |
ibr | -0.032 |
ibrarian | -0.032 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +5.06e-01 |
Pos logit contributions | |
itating | +0.006 |
arre | +0.006 |
up | +0.005 |
melon | +0.005 |
ling | +0.005 |
Neg logit contributions | |
year | -0.011 |
ird | -0.010 |
glass | -0.010 |
hours | -0.010 |
aster | -0.010 |
Token | John |
Token ID | 7554 |
Feature activation | +8.66e-01 |
Pos logit contributions | |
itating | +0.009 |
arre | +0.009 |
up | +0.009 |
melon | +0.008 |
ling | +0.008 |
Neg logit contributions | |
year | -0.018 |
ird | -0.017 |
glass | -0.017 |
hours | -0.016 |
aster | -0.016 |
Token | loved |
Token ID | 6151 |
Feature activation | +3.21e-01 |
Pos logit contributions | |
arre | +0.015 |
itating | +0.014 |
up | +0.014 |
melon | +0.014 |
ling | +0.014 |
Neg logit contributions | |
year | -0.031 |
glass | -0.030 |
ird | -0.030 |
hours | -0.028 |
aster | -0.028 |
Token | baseball |
Token ID | 9283 |
Feature activation | +4.38e+00 |
Pos logit contributions | |
itating | +0.008 |
arre | +0.008 |
ling | +0.007 |
up | +0.007 |
melon | +0.007 |
Neg logit contributions | |
year | -0.009 |
glass | -0.009 |
ird | -0.009 |
hours | -0.008 |
aster | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
itating | +0.132 |
arre | +0.124 |
ucker | +0.119 |
inated | +0.115 |
aining | +0.115 |
Neg logit contributions | |
ird | -0.092 |
year | -0.091 |
glass | -0.089 |
hours | -0.088 |
aster | -0.081 |
Token | He |
Token ID | 679 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
arre | +0.004 |
itating | +0.004 |
up | +0.004 |
melon | +0.004 |
ling | +0.004 |
Neg logit contributions | |
year | -0.009 |
glass | -0.009 |
ird | -0.008 |
hours | -0.008 |
aster | -0.008 |
Token | liked |
Token ID | 8288 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
itating | +0.053 |
arre | +0.052 |
up | +0.049 |
melon | +0.048 |
ling | +0.048 |
Neg logit contributions | |
year | -0.062 |
glass | -0.059 |
ird | -0.059 |
hours | -0.058 |
minutes | -0.055 |
Token | to |
Token ID | 284 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
itating | +0.072 |
arre | +0.068 |
ling | +0.066 |
ivering | +0.066 |
flower | +0.065 |
Neg logit contributions | |
hours | -0.041 |
glass | -0.039 |
year | -0.038 |
aster | -0.038 |
minutes | -0.037 |
Token | kick |
Token ID | 4829 |
Feature activation | +2.73e+00 |
Pos logit contributions | |
itating | +0.040 |
arre | +0.039 |
ling | +0.036 |
accidents | +0.035 |
bing | +0.034 |
Neg logit contributions | |
aster | -0.043 |
ird | -0.043 |
glass | -0.042 |
year | -0.041 |
hours | -0.041 |
Token | 't |
Token ID | 470 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
arre | +0.017 |
melon | +0.016 |
itating | +0.015 |
fighting | +0.015 |
ling | +0.015 |
Neg logit contributions | |
ird | -0.032 |
year | -0.032 |
glass | -0.031 |
aster | -0.030 |
lock | -0.030 |
Token | cry |
Token ID | 3960 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
arre | +0.003 |
melon | +0.003 |
itating | +0.003 |
ucker | +0.003 |
rooms | +0.003 |
Neg logit contributions | |
ird | -0.003 |
aster | -0.002 |
lock | -0.002 |
year | -0.002 |
glass | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
arre | +0.031 |
itating | +0.031 |
rust | +0.031 |
melon | +0.030 |
rooms | +0.027 |
Neg logit contributions | |
ird | -0.030 |
year | -0.029 |
minutes | -0.028 |
glass | -0.028 |
hours | -0.027 |
Token | boys |
Token ID | 6510 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
itating | +0.019 |
arre | +0.019 |
up | +0.018 |
melon | +0.018 |
ling | +0.018 |
Neg logit contributions | |
year | -0.035 |
ird | -0.033 |
glass | -0.033 |
hours | -0.032 |
aster | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
itating | +0.061 |
arre | +0.059 |
rust | +0.054 |
melon | +0.054 |
bing | +0.054 |
Neg logit contributions | |
ird | -0.073 |
year | -0.070 |
glass | -0.068 |
minutes | -0.066 |
hours | -0.066 |
Token | Acc |
Token ID | 6366 |
Feature activation | +4.37e+00 |
Pos logit contributions | |
arre | +0.026 |
itating | +0.026 |
ling | +0.026 |
up | +0.025 |
bing | +0.024 |
Neg logit contributions | |
year | -0.034 |
ird | -0.033 |
glass | -0.033 |
aster | -0.030 |
ibr | -0.029 |
Token | idents |
Token ID | 3231 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
itating | +0.077 |
rooms | +0.073 |
melon | +0.073 |
Freddy | +0.072 |
urtle | +0.070 |
Neg logit contributions | |
glass | -0.135 |
ber | -0.127 |
bers | -0.127 |
ird | -0.126 |
lan | -0.121 |
Token | happen |
Token ID | 1645 |
Feature activation | +3.38e+00 |
Pos logit contributions | |
rust | +0.038 |
flower | +0.037 |
ican | +0.035 |
ishly | +0.035 |
ucker | +0.035 |
Neg logit contributions | |
ird | -0.031 |
book | -0.030 |
year | -0.030 |
minutes | -0.029 |
hour | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
itating | +0.099 |
arre | +0.096 |
melon | +0.094 |
ican | +0.091 |
rust | +0.090 |
Neg logit contributions | |
minutes | -0.078 |
hours | -0.074 |
year | -0.070 |
glass | -0.070 |
ird | -0.069 |
Token | I |
Token ID | 314 |
Feature activation | +4.23e-01 |
Pos logit contributions | |
arre | +0.030 |
itating | +0.030 |
ling | +0.029 |
up | +0.029 |
melon | +0.028 |
Neg logit contributions | |
year | -0.038 |
ird | -0.037 |
glass | -0.036 |
minutes | -0.033 |
hours | -0.033 |
Token | have |
Token ID | 423 |
Feature activation | +8.97e-01 |
Pos logit contributions | |
itating | +0.011 |
bing | +0.011 |
melon | +0.011 |
up | +0.011 |
arre | +0.011 |
Neg logit contributions | |
year | -0.011 |
ird | -0.010 |
glass | -0.010 |
ris | -0.009 |
hours | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
arre | +0.015 |
itating | +0.015 |
ling | +0.014 |
up | +0.014 |
rust | +0.014 |
Neg logit contributions | |
year | -0.025 |
glass | -0.024 |
ird | -0.024 |
hours | -0.023 |
minutes | -0.022 |
Token | so |
Token ID | 523 |
Feature activation | -3.61e-01 |
Pos logit contributions | |
arre | +0.047 |
rust | +0.044 |
itating | +0.044 |
rooms | +0.042 |
urtle | +0.040 |
Neg logit contributions | |
hours | -0.052 |
minutes | -0.051 |
year | -0.050 |
aster | -0.046 |
glass | -0.046 |
Token | she |
Token ID | 673 |
Feature activation | +5.77e-01 |
Pos logit contributions | |
year | +0.015 |
glass | +0.014 |
ird | +0.014 |
ris | +0.013 |
kil | +0.013 |
Neg logit contributions | |
up | -0.005 |
itating | -0.005 |
ling | -0.004 |
bing | -0.004 |
arre | -0.004 |
Token | could |
Token ID | 714 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
itating | +0.012 |
arre | +0.012 |
up | +0.012 |
melon | +0.012 |
ling | +0.012 |
Neg logit contributions | |
year | -0.018 |
glass | -0.017 |
ird | -0.017 |
hours | -0.016 |
aster | -0.016 |
Token | pick |
Token ID | 2298 |
Feature activation | +3.51e+00 |
Pos logit contributions | |
arre | +0.047 |
ling | +0.045 |
itating | +0.045 |
bing | +0.044 |
mill | +0.044 |
Neg logit contributions | |
ird | -0.042 |
aster | -0.041 |
year | -0.041 |
glass | -0.039 |
hours | -0.038 |
Token | the |
Token ID | 262 |
Feature activation | +4.34e+00 |
Pos logit contributions | |
itating | +0.098 |
arre | +0.092 |
ivering | +0.092 |
bing | +0.089 |
mill | +0.087 |
Neg logit contributions | |
year | -0.085 |
hours | -0.082 |
aster | -0.077 |
minutes | -0.074 |
ird | -0.073 |
Token | best |
Token ID | 1266 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
mill | +0.112 |
itating | +0.111 |
bing | +0.108 |
arre | +0.100 |
ling | +0.099 |
Neg logit contributions | |
hours | -0.107 |
minutes | -0.103 |
ird | -0.102 |
year | -0.097 |
ns | -0.095 |
Token | ones |
Token ID | 3392 |
Feature activation | +7.62e-01 |
Pos logit contributions | |
itating | +0.079 |
bing | +0.072 |
mill | +0.067 |
Fighting | +0.065 |
arre | +0.065 |
Neg logit contributions | |
hours | -0.062 |
minutes | -0.062 |
ird | -0.058 |
aster | -0.055 |
year | -0.055 |
Token | . |
Token ID | 13 |
Feature activation | +3.93e-01 |
Pos logit contributions | |
arre | +0.017 |
itating | +0.016 |
ling | +0.015 |
rust | +0.015 |
bing | +0.015 |
Neg logit contributions | |
year | -0.023 |
ird | -0.023 |
glass | -0.022 |
hours | -0.021 |
minutes | -0.021 |
Token | But |
Token ID | 887 |
Feature activation | +9.95e-01 |
Pos logit contributions | |
arre | +0.009 |
melon | +0.009 |
itating | +0.009 |
ling | +0.009 |
rust | +0.008 |
Neg logit contributions | |
year | -0.012 |
ird | -0.011 |
hours | -0.011 |
glass | -0.011 |
minutes | -0.010 |
Token | suddenly |
Token ID | 6451 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
up | +0.019 |
itating | +0.019 |
arre | +0.019 |
melon | +0.018 |
ling | +0.018 |
Neg logit contributions | |
year | -0.034 |
glass | -0.032 |
ird | -0.032 |
hours | -0.030 |
aster | -0.030 |
Token | so |
Token ID | 523 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
itating | +0.007 |
arre | +0.007 |
up | +0.007 |
melon | +0.007 |
ling | +0.007 |
Neg logit contributions | |
year | -0.011 |
glass | -0.011 |
ird | -0.011 |
hours | -0.010 |
aster | -0.010 |
Token | careful |
Token ID | 8161 |
Feature activation | +2.05e+00 |
Pos logit contributions | |
year | +0.004 |
glass | +0.004 |
ird | +0.004 |
hours | +0.003 |
aster | +0.003 |
Neg logit contributions | |
arre | -0.002 |
itating | -0.002 |
up | -0.002 |
melon | -0.002 |
ling | -0.002 |
Token | when |
Token ID | 618 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
arre | +0.043 |
rust | +0.042 |
itating | +0.040 |
melon | +0.039 |
Bugs | +0.037 |
Neg logit contributions | |
year | -0.064 |
glass | -0.062 |
ird | -0.060 |
aster | -0.058 |
lock | -0.058 |
Token | he |
Token ID | 339 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
up | +0.002 |
itating | +0.002 |
arre | +0.002 |
melon | +0.002 |
fighting | +0.002 |
Neg logit contributions | |
year | -0.004 |
ird | -0.003 |
glass | -0.003 |
hours | -0.003 |
aster | -0.003 |
Token | picked |
Token ID | 6497 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
up | +0.005 |
arre | +0.005 |
melon | +0.004 |
ling | +0.004 |
fighting | +0.004 |
Neg logit contributions | |
year | -0.009 |
glass | -0.009 |
ird | -0.008 |
hours | -0.008 |
ibr | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | +4.34e+00 |
Pos logit contributions | |
itating | +0.055 |
arre | +0.052 |
inas | +0.051 |
lings | +0.048 |
bing | +0.047 |
Neg logit contributions | |
year | -0.070 |
aster | -0.069 |
hours | -0.068 |
glass | -0.065 |
kil | -0.064 |
Token | squash |
Token ID | 34613 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
inas | +0.101 |
ocked | +0.100 |
ucker | +0.099 |
ouncing | +0.098 |
ivering | +0.094 |
Neg logit contributions | |
hours | -0.094 |
year | -0.090 |
20 | -0.090 |
lock | -0.087 |
month | -0.087 |
Token | but |
Token ID | 475 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
arre | +0.037 |
itating | +0.036 |
rust | +0.034 |
ucker | +0.033 |
inas | +0.033 |
Neg logit contributions | |
year | -0.048 |
ird | -0.046 |
glass | -0.046 |
aster | -0.045 |
lock | -0.044 |
Token | this |
Token ID | 428 |
Feature activation | +1.94e+00 |
Pos logit contributions | |
itating | +0.021 |
arre | +0.021 |
up | +0.020 |
ling | +0.020 |
melon | +0.020 |
Neg logit contributions | |
year | -0.034 |
ird | -0.032 |
glass | -0.032 |
hours | -0.031 |
aster | -0.030 |
Token | time |
Token ID | 640 |
Feature activation | -8.48e-01 |
Pos logit contributions | |
itating | +0.053 |
arre | +0.049 |
bing | +0.048 |
inas | +0.047 |
rust | +0.045 |
Neg logit contributions | |
hours | -0.050 |
year | -0.049 |
ird | -0.049 |
minutes | -0.049 |
aster | -0.048 |
Token | it |
Token ID | 340 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
year | +0.032 |
glass | +0.031 |
ird | +0.030 |
aster | +0.029 |
Bible | +0.029 |
Neg logit contributions | |
up | -0.014 |
fighting | -0.012 |
Freddy | -0.012 |
melon | -0.012 |
arre | -0.011 |
Token | very |
Token ID | 845 |
Feature activation | +9.92e-01 |
Pos logit contributions | |
itating | +0.042 |
arre | +0.041 |
mill | +0.037 |
ling | +0.036 |
flower | +0.036 |
Neg logit contributions | |
year | -0.044 |
hours | -0.041 |
glass | -0.039 |
minutes | -0.039 |
ird | -0.038 |
Token | enthusiastic |
Token ID | 21848 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
itating | +0.022 |
arre | +0.022 |
up | +0.021 |
melon | +0.021 |
ling | +0.020 |
Neg logit contributions | |
year | -0.031 |
glass | -0.030 |
ird | -0.030 |
hours | -0.028 |
aster | -0.028 |
Token | to |
Token ID | 284 |
Feature activation | +6.55e-01 |
Pos logit contributions | |
itating | +0.042 |
arre | +0.039 |
rust | +0.037 |
melon | +0.037 |
ling | +0.036 |
Neg logit contributions | |
year | -0.049 |
hours | -0.047 |
glass | -0.047 |
minutes | -0.045 |
ird | -0.045 |
Token | play |
Token ID | 711 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
itating | +0.015 |
arre | +0.015 |
up | +0.014 |
ling | +0.014 |
melon | +0.014 |
Neg logit contributions | |
year | -0.020 |
ird | -0.019 |
glass | -0.019 |
hours | -0.018 |
aster | -0.018 |
Token | with |
Token ID | 351 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
itating | +0.052 |
arre | +0.047 |
inas | +0.046 |
melon | +0.046 |
ucker | +0.046 |
Neg logit contributions | |
glass | -0.067 |
hours | -0.066 |
year | -0.066 |
minutes | -0.062 |
ird | -0.062 |
Token | the |
Token ID | 262 |
Feature activation | +4.33e+00 |
Pos logit contributions | |
itating | +0.035 |
arre | +0.034 |
inas | +0.029 |
melon | +0.029 |
flower | +0.029 |
Neg logit contributions | |
year | -0.062 |
glass | -0.061 |
hours | -0.060 |
ird | -0.060 |
aster | -0.057 |
Token | beans |
Token ID | 16567 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
mill | +0.137 |
itating | +0.133 |
ican | +0.126 |
bing | +0.125 |
ocked | +0.124 |
Neg logit contributions | |
hours | -0.092 |
props | -0.085 |
minutes | -0.082 |
servants | -0.070 |
chairs | -0.069 |
Token | . |
Token ID | 13 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
arre | +0.033 |
itating | +0.033 |
rust | +0.030 |
bing | +0.029 |
ucker | +0.029 |
Neg logit contributions | |
year | -0.041 |
ird | -0.041 |
glass | -0.040 |
hours | -0.039 |
minutes | -0.038 |
Token |
|
Token ID | 198 |
Feature activation | -3.64e-01 |
Pos logit contributions | |
ling | +0.048 |
itating | +0.046 |
arre | +0.045 |
flower | +0.045 |
mill | +0.044 |
Neg logit contributions | |
ird | -0.031 |
glass | -0.030 |
year | -0.029 |
ris | -0.029 |
hours | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
year | +0.013 |
glass | +0.012 |
ird | +0.012 |
hours | +0.011 |
aster | +0.011 |
Neg logit contributions | |
itating | -0.007 |
arre | -0.007 |
up | -0.006 |
melon | -0.006 |
ling | -0.006 |
Token | Tim |
Token ID | 14967 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
melon | +0.054 |
itating | +0.054 |
arre | +0.053 |
sets | +0.053 |
flower | +0.052 |
Neg logit contributions | |
year | -0.051 |
glass | -0.048 |
hours | -0.047 |
ird | -0.047 |
ris | -0.045 |
Token | !" |
Token ID | 2474 |
Feature activation | -3.59e-01 |
Pos logit contributions | |
itating | +0.011 |
arre | +0.011 |
melon | +0.011 |
up | +0.010 |
ling | +0.010 |
Neg logit contributions | |
year | -0.014 |
ird | -0.014 |
glass | -0.014 |
hours | -0.013 |
aster | -0.013 |
Token | Lily |
Token ID | 20037 |
Feature activation | -3.79e-01 |
Pos logit contributions | |
year | +0.016 |
glass | +0.015 |
ird | +0.015 |
kil | +0.015 |
ris | +0.015 |
Neg logit contributions | |
up | -0.004 |
fighting | -0.004 |
Freddy | -0.003 |
accidents | -0.003 |
Polly | -0.003 |
Token | cried |
Token ID | 16896 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
year | +0.012 |
glass | +0.011 |
ird | +0.011 |
hours | +0.011 |
aster | +0.011 |
Neg logit contributions | |
up | -0.008 |
arre | -0.008 |
itating | -0.008 |
fighting | -0.008 |
ling | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +5.43e-02 |
Pos logit contributions | |
year | +0.040 |
glass | +0.039 |
ird | +0.039 |
hours | +0.037 |
aster | +0.036 |
Neg logit contributions | |
up | -0.020 |
itating | -0.019 |
arre | -0.019 |
ling | -0.018 |
fighting | -0.018 |
Token | " |
Token ID | 366 |
Feature activation | +7.80e-01 |
Pos logit contributions | |
arre | +0.001 |
ling | +0.001 |
rust | +0.001 |
melon | +0.001 |
itating | +0.001 |
Neg logit contributions | |
year | -0.002 |
glass | -0.002 |
ird | -0.002 |
minutes | -0.002 |
lock | -0.002 |
Token | The |
Token ID | 464 |
Feature activation | +4.29e+00 |
Pos logit contributions | |
melon | +0.016 |
arre | +0.016 |
up | +0.015 |
itating | +0.015 |
ling | +0.015 |
Neg logit contributions | |
year | -0.025 |
glass | -0.025 |
aster | -0.024 |
ird | -0.024 |
hours | -0.023 |
Token | shrimp |
Token ID | 30060 |
Feature activation | +8.22e-01 |
Pos logit contributions | |
itating | +0.124 |
bing | +0.117 |
ning | +0.115 |
mill | +0.115 |
arre | +0.114 |
Neg logit contributions | |
glass | -0.094 |
passport | -0.091 |
minutes | -0.091 |
hours | -0.090 |
couple | -0.090 |
Token | hurt |
Token ID | 5938 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
itating | +0.017 |
arre | +0.017 |
rust | +0.016 |
inas | +0.015 |
bing | +0.015 |
Neg logit contributions | |
ird | -0.026 |
year | -0.025 |
glass | -0.025 |
hours | -0.024 |
minutes | -0.024 |
Token | me |
Token ID | 502 |
Feature activation | -5.61e-01 |
Pos logit contributions | |
itating | +0.021 |
arre | +0.020 |
melon | +0.019 |
ling | +0.018 |
up | +0.018 |
Neg logit contributions | |
year | -0.047 |
ird | -0.045 |
glass | -0.045 |
hours | -0.043 |
aster | -0.043 |
Token | !" |
Token ID | 2474 |
Feature activation | +5.90e-02 |
Pos logit contributions | |
year | +0.018 |
glass | +0.017 |
ird | +0.017 |
hours | +0.017 |
aster | +0.016 |
Neg logit contributions | |
itating | -0.012 |
arre | -0.011 |
up | -0.011 |
ling | -0.011 |
melon | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | -5.31e-01 |
Pos logit contributions | |
arre | +0.001 |
itating | +0.001 |
up | +0.001 |
melon | +0.001 |
ling | +0.001 |
Neg logit contributions | |
year | -0.002 |
glass | -0.002 |
ird | -0.002 |
hours | -0.002 |
aster | -0.002 |
Token | was |
Token ID | 373 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
itating | +0.024 |
bing | +0.022 |
melon | +0.022 |
arre | +0.022 |
mill | +0.021 |
Neg logit contributions | |
year | -0.026 |
ird | -0.025 |
glass | -0.024 |
hours | -0.022 |
ris | -0.022 |
Token | jealous |
Token ID | 19354 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
fl | +0.037 |
itating | +0.037 |
flower | +0.036 |
rooms | +0.036 |
bing | +0.036 |
Neg logit contributions | |
year | -0.052 |
ird | -0.051 |
hours | -0.044 |
minutes | -0.044 |
glass | -0.043 |
Token | of |
Token ID | 286 |
Feature activation | -2.46e-01 |
Pos logit contributions | |
itating | +0.059 |
bing | +0.056 |
rust | +0.056 |
melon | +0.056 |
flower | +0.056 |
Neg logit contributions | |
ird | -0.031 |
year | -0.031 |
minutes | -0.029 |
glass | -0.028 |
hours | -0.027 |
Token | Johnny |
Token ID | 15470 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
year | +0.008 |
glass | +0.007 |
ird | +0.007 |
hours | +0.007 |
aster | +0.007 |
Neg logit contributions | |
itating | -0.005 |
arre | -0.005 |
ling | -0.005 |
melon | -0.005 |
up | -0.005 |
Token | 's |
Token ID | 338 |
Feature activation | +2.87e+00 |
Pos logit contributions | |
itating | +0.030 |
arre | +0.028 |
rust | +0.027 |
melon | +0.026 |
flower | +0.026 |
Neg logit contributions | |
ird | -0.041 |
year | -0.039 |
glass | -0.039 |
aster | -0.037 |
hours | -0.036 |
Token | new |
Token ID | 649 |
Feature activation | +4.27e+00 |
Pos logit contributions | |
itating | +0.070 |
flower | +0.068 |
ter | +0.065 |
lings | +0.065 |
mill | +0.063 |
Neg logit contributions | |
hours | -0.065 |
allowance | -0.064 |
ird | -0.061 |
glass | -0.059 |
year | -0.059 |
Token | basketball |
Token ID | 9669 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
itating | +0.118 |
ping | +0.109 |
ric | +0.105 |
ucker | +0.105 |
lings | +0.103 |
Neg logit contributions | |
hours | -0.095 |
ird | -0.089 |
minutes | -0.083 |
books | -0.081 |
customers | -0.080 |
Token | . |
Token ID | 13 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
itating | +0.029 |
arre | +0.028 |
up | +0.027 |
melon | +0.027 |
ling | +0.027 |
Neg logit contributions | |
year | -0.049 |
ird | -0.047 |
glass | -0.047 |
hours | -0.045 |
aster | -0.044 |
Token | His |
Token ID | 2399 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
itating | +0.035 |
ling | +0.034 |
arre | +0.033 |
melon | +0.032 |
flower | +0.031 |
Neg logit contributions | |
year | -0.038 |
ird | -0.037 |
glass | -0.036 |
hours | -0.034 |
aster | -0.033 |
Token | mom |
Token ID | 1995 |
Feature activation | +7.08e-01 |
Pos logit contributions | |
itating | +0.034 |
arre | +0.032 |
ling | +0.032 |
melon | +0.031 |
bing | +0.031 |
Neg logit contributions | |
glass | -0.020 |
ird | -0.019 |
year | -0.019 |
hours | -0.018 |
lock | -0.016 |
Token | said |
Token ID | 531 |
Feature activation | +4.88e-01 |
Pos logit contributions | |
itating | +0.020 |
bing | +0.019 |
melon | +0.019 |
arre | +0.019 |
sets | +0.018 |
Neg logit contributions | |
year | -0.018 |
ird | -0.018 |
glass | -0.016 |
hours | -0.015 |
ibrarian | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | +8.97e-01 |
Pos logit contributions | |
itating | +0.027 |
arre | +0.025 |
bing | +0.023 |
ling | +0.023 |
melon | +0.022 |
Neg logit contributions | |
year | -0.045 |
ird | -0.044 |
glass | -0.044 |
aster | -0.042 |
hours | -0.042 |
Token | Sam |
Token ID | 3409 |
Feature activation | +2.42e+00 |
Pos logit contributions | |
itating | +0.019 |
arre | +0.019 |
up | +0.018 |
melon | +0.018 |
ling | +0.018 |
Neg logit contributions | |
year | -0.029 |
glass | -0.027 |
ird | -0.027 |
hours | -0.026 |
aster | -0.025 |
Token | love |
Token ID | 1842 |
Feature activation | +5.41e-01 |
Pos logit contributions | |
itating | +0.067 |
arre | +0.061 |
bing | +0.060 |
ocked | +0.057 |
accidents | +0.056 |
Neg logit contributions | |
ird | -0.063 |
year | -0.062 |
aster | -0.060 |
hours | -0.060 |
glass | -0.060 |
Token | to |
Token ID | 284 |
Feature activation | +7.77e-01 |
Pos logit contributions | |
up | +0.011 |
melon | +0.011 |
arre | +0.010 |
fighting | +0.010 |
itating | +0.010 |
Neg logit contributions | |
year | -0.018 |
ird | -0.018 |
glass | -0.017 |
hours | -0.016 |
minutes | -0.016 |
Token | play |
Token ID | 711 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
itating | +0.018 |
arre | +0.018 |
ling | +0.018 |
up | +0.017 |
bing | +0.016 |
Neg logit contributions | |
year | -0.022 |
ird | -0.022 |
glass | -0.022 |
aster | -0.022 |
hours | -0.021 |
Token | baseball |
Token ID | 9283 |
Feature activation | +4.27e+00 |
Pos logit contributions | |
itating | +0.036 |
arre | +0.035 |
melon | +0.033 |
ling | +0.032 |
inas | +0.031 |
Neg logit contributions | |
year | -0.057 |
glass | -0.057 |
ird | -0.056 |
hours | -0.054 |
aster | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
itating | +0.122 |
arre | +0.117 |
ucker | +0.111 |
inas | +0.107 |
accidents | +0.105 |
Neg logit contributions | |
glass | -0.102 |
year | -0.102 |
hours | -0.097 |
ird | -0.095 |
aster | -0.092 |
Token | They |
Token ID | 1119 |
Feature activation | +2.37e+00 |
Pos logit contributions | |
itating | +0.022 |
arre | +0.021 |
ling | +0.020 |
up | +0.020 |
melon | +0.019 |
Neg logit contributions | |
year | -0.043 |
ird | -0.042 |
glass | -0.042 |
hours | -0.040 |
aster | -0.040 |
Token | have |
Token ID | 423 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
itating | +0.060 |
arre | +0.059 |
bing | +0.055 |
ling | +0.055 |
up | +0.055 |
Neg logit contributions | |
year | -0.065 |
ird | -0.064 |
glass | -0.063 |
hours | -0.062 |
aster | -0.060 |
Token | a |
Token ID | 257 |
Feature activation | +8.98e-01 |
Pos logit contributions | |
arre | +0.040 |
itating | +0.039 |
up | +0.039 |
melon | +0.038 |
ling | +0.037 |
Neg logit contributions | |
year | -0.047 |
ird | -0.044 |
glass | -0.044 |
hours | -0.041 |
aster | -0.041 |
Token | big |
Token ID | 1263 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
itating | +0.022 |
arre | +0.021 |
bing | +0.020 |
ling | +0.020 |
up | +0.020 |
Neg logit contributions | |
year | -0.025 |
ird | -0.025 |
glass | -0.024 |
hours | -0.024 |
minutes | -0.023 |
Token | very |
Token ID | 845 |
Feature activation | +5.63e-02 |
Pos logit contributions | |
arre | +0.002 |
itating | +0.002 |
melon | +0.002 |
ling | +0.002 |
up | +0.002 |
Neg logit contributions | |
year | -0.002 |
ird | -0.002 |
glass | -0.002 |
hours | -0.002 |
minutes | -0.002 |
Token | careful |
Token ID | 8161 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
itating | +0.001 |
arre | +0.001 |
up | +0.001 |
melon | +0.001 |
ling | +0.001 |
Neg logit contributions | |
year | -0.002 |
ird | -0.002 |
glass | -0.002 |
hours | -0.002 |
aster | -0.002 |
Token | to |
Token ID | 284 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
rust | +0.053 |
arre | +0.052 |
melon | +0.048 |
itating | +0.048 |
rooms | +0.045 |
Neg logit contributions | |
year | -0.069 |
glass | -0.068 |
ird | -0.065 |
hours | -0.065 |
aster | -0.065 |
Token | only |
Token ID | 691 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
arre | +0.005 |
itating | +0.004 |
ling | +0.004 |
melon | +0.004 |
mill | +0.004 |
Neg logit contributions | |
ird | -0.005 |
aster | -0.005 |
year | -0.005 |
glass | -0.005 |
hours | -0.005 |
Token | pick |
Token ID | 2298 |
Feature activation | +3.21e+00 |
Pos logit contributions | |
itating | +0.023 |
arre | +0.022 |
ling | +0.021 |
up | +0.021 |
bing | +0.020 |
Neg logit contributions | |
year | -0.032 |
ird | -0.032 |
glass | -0.031 |
aster | -0.030 |
hours | -0.030 |
Token | the |
Token ID | 262 |
Feature activation | +4.26e+00 |
Pos logit contributions | |
itating | +0.074 |
arre | +0.068 |
mill | +0.065 |
ivering | +0.064 |
inas | +0.062 |
Neg logit contributions | |
hours | -0.091 |
year | -0.090 |
minutes | -0.083 |
Africa | -0.082 |
kil | -0.081 |
Token | big |
Token ID | 1263 |
Feature activation | +3.34e+00 |
Pos logit contributions | |
mill | +0.144 |
itating | +0.137 |
ocked | +0.135 |
ican | +0.125 |
ric | +0.125 |
Neg logit contributions | |
hours | -0.072 |
props | -0.071 |
minutes | -0.068 |
clues | -0.063 |
chairs | -0.059 |
Token | , |
Token ID | 11 |
Feature activation | +5.63e-01 |
Pos logit contributions | |
itating | +0.102 |
mill | +0.088 |
ocked | +0.086 |
arre | +0.085 |
Fighting | +0.083 |
Neg logit contributions | |
minutes | -0.082 |
hours | -0.078 |
seconds | -0.065 |
props | -0.065 |
kil | -0.064 |
Token | green |
Token ID | 4077 |
Feature activation | +3.16e+00 |
Pos logit contributions | |
itating | +0.011 |
arre | +0.011 |
up | +0.010 |
ling | +0.010 |
melon | +0.010 |
Neg logit contributions | |
year | -0.019 |
ird | -0.018 |
glass | -0.018 |
hours | -0.018 |
minutes | -0.018 |
Token | beans |
Token ID | 16567 |
Feature activation | +7.03e-01 |
Pos logit contributions | |
itating | +0.093 |
arre | +0.080 |
mill | +0.076 |
ucker | +0.073 |
Fighting | +0.073 |
Neg logit contributions | |
hours | -0.082 |
minutes | -0.074 |
glass | -0.069 |
year | -0.068 |
props | -0.067 |
Token | with |
Token ID | 351 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
arre | +0.016 |
itating | +0.016 |
rust | +0.016 |
ling | +0.014 |
inas | +0.014 |
Neg logit contributions | |
year | -0.020 |
ird | -0.020 |
glass | -0.020 |
hours | -0.019 |
minutes | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
arre | +0.032 |
itating | +0.032 |
up | +0.029 |
melon | +0.029 |
ling | +0.029 |
Neg logit contributions | |
year | -0.038 |
ird | -0.037 |
glass | -0.037 |
hours | -0.035 |
aster | -0.034 |
Token | They |
Token ID | 1119 |
Feature activation | +5.41e-01 |
Pos logit contributions | |
itating | +0.034 |
arre | +0.033 |
up | +0.032 |
ling | +0.031 |
melon | +0.030 |
Neg logit contributions | |
year | -0.076 |
ird | -0.073 |
glass | -0.072 |
hours | -0.069 |
aster | -0.068 |
Token | like |
Token ID | 588 |
Feature activation | -3.82e-01 |
Pos logit contributions | |
itating | +0.013 |
arre | +0.013 |
up | +0.012 |
melon | +0.012 |
ling | +0.012 |
Neg logit contributions | |
year | -0.016 |
glass | -0.015 |
ird | -0.015 |
hours | -0.014 |
aster | -0.014 |
Token | to |
Token ID | 284 |
Feature activation | +2.64e-02 |
Pos logit contributions | |
year | +0.013 |
ird | +0.012 |
glass | +0.012 |
hours | +0.011 |
minutes | +0.011 |
Neg logit contributions | |
up | -0.008 |
arre | -0.008 |
itating | -0.007 |
melon | -0.007 |
fighting | -0.007 |
Token | play |
Token ID | 711 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
up | +0.000 |
arre | +0.000 |
melon | +0.000 |
itating | +0.000 |
fighting | +0.000 |
Neg logit contributions | |
year | -0.001 |
glass | -0.001 |
ird | -0.001 |
hours | -0.001 |
minutes | -0.001 |
Token | baseball |
Token ID | 9283 |
Feature activation | +4.24e+00 |
Pos logit contributions | |
itating | +0.029 |
arre | +0.028 |
melon | +0.027 |
up | +0.027 |
ling | +0.026 |
Neg logit contributions | |
year | -0.051 |
glass | -0.049 |
ird | -0.049 |
hours | -0.047 |
aster | -0.046 |
Token | in |
Token ID | 287 |
Feature activation | +6.10e-01 |
Pos logit contributions | |
itating | +0.109 |
arre | +0.102 |
ucker | +0.098 |
inas | +0.094 |
accidents | +0.093 |
Neg logit contributions | |
glass | -0.115 |
hours | -0.110 |
year | -0.110 |
ird | -0.105 |
aster | -0.102 |
Token | the |
Token ID | 262 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
arre | +0.014 |
itating | +0.014 |
ling | +0.013 |
melon | +0.013 |
up | +0.012 |
Neg logit contributions | |
year | -0.019 |
glass | -0.018 |
ird | -0.018 |
hours | -0.018 |
aster | -0.018 |
Token | park |
Token ID | 3952 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
itating | +0.052 |
ling | +0.047 |
bing | +0.047 |
ocked | +0.047 |
arre | +0.046 |
Neg logit contributions | |
glass | -0.048 |
aster | -0.048 |
hours | -0.046 |
ird | -0.046 |
minutes | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
up | +0.029 |
itating | +0.027 |
melon | +0.027 |
ling | +0.026 |
fighting | +0.026 |
Neg logit contributions | |
year | -0.067 |
ird | -0.064 |
glass | -0.063 |
hours | -0.060 |
aster | -0.060 |
Token | They |
Token ID | 1119 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
itating | +0.024 |
arre | +0.023 |
ling | +0.023 |
up | +0.022 |
bing | +0.022 |
Neg logit contributions | |
year | -0.035 |
ird | -0.034 |
glass | -0.034 |
hours | -0.032 |
aster | -0.032 |
Token | saw |
Token ID | 2497 |
Feature activation | +4.42e-01 |
Pos logit contributions | |
arre | +0.010 |
up | +0.010 |
itating | +0.009 |
melon | +0.009 |
ling | +0.009 |
Neg logit contributions | |
year | -0.017 |
glass | -0.016 |
ird | -0.016 |
hours | -0.015 |
aster | -0.015 |
Token | some |
Token ID | 617 |
Feature activation | +6.23e-01 |
Pos logit contributions | |
arre | +0.010 |
itating | +0.010 |
up | +0.010 |
melon | +0.010 |
ling | +0.010 |
Neg logit contributions | |
year | -0.013 |
ird | -0.013 |
glass | -0.013 |
hours | -0.012 |
aster | -0.012 |
Token | other |
Token ID | 584 |
Feature activation | +7.80e-01 |
Pos logit contributions | |
itating | +0.014 |
arre | +0.014 |
melon | +0.014 |
bing | +0.013 |
up | +0.013 |
Neg logit contributions | |
year | -0.019 |
ird | -0.018 |
glass | -0.018 |
hours | -0.017 |
minutes | -0.017 |
Token | kids |
Token ID | 3988 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
itating | +0.018 |
arre | +0.017 |
melon | +0.017 |
up | +0.016 |
ling | +0.016 |
Neg logit contributions | |
year | -0.024 |
ird | -0.023 |
glass | -0.023 |
hours | -0.022 |
minutes | -0.021 |
Token | playing |
Token ID | 2712 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
itating | +0.035 |
rust | +0.035 |
arre | +0.034 |
aining | +0.034 |
adoes | +0.034 |
Neg logit contributions | |
ird | -0.050 |
year | -0.047 |
glass | -0.046 |
hours | -0.042 |
aster | -0.042 |
Token | baseball |
Token ID | 9283 |
Feature activation | +4.24e+00 |
Pos logit contributions | |
itating | +0.054 |
arre | +0.051 |
melon | +0.050 |
inas | +0.048 |
flower | +0.048 |
Neg logit contributions | |
year | -0.074 |
glass | -0.072 |
ird | -0.071 |
hours | -0.070 |
minutes | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | +5.10e-01 |
Pos logit contributions | |
itating | +0.109 |
ucker | +0.102 |
arre | +0.100 |
rust | +0.094 |
inas | +0.094 |
Neg logit contributions | |
glass | -0.115 |
year | -0.111 |
hours | -0.110 |
ird | -0.107 |
minutes | -0.101 |
Token | He |
Token ID | 679 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
itating | +0.011 |
arre | +0.011 |
ling | +0.011 |
up | +0.010 |
bing | +0.010 |
Neg logit contributions | |
year | -0.016 |
ird | -0.015 |
glass | -0.015 |
hours | -0.014 |
aster | -0.014 |
Token | begged |
Token ID | 32946 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
itating | +0.043 |
bing | +0.040 |
arre | +0.040 |
accidents | +0.039 |
melon | +0.039 |
Neg logit contributions | |
year | -0.041 |
hours | -0.040 |
ird | -0.039 |
glass | -0.039 |
minutes | -0.037 |
Token | his |
Token ID | 465 |
Feature activation | +6.51e-01 |
Pos logit contributions | |
year | +0.008 |
ird | +0.008 |
glass | +0.007 |
ris | +0.007 |
ibr | +0.007 |
Neg logit contributions | |
up | -0.002 |
fighting | -0.002 |
Freddy | -0.002 |
melon | -0.002 |
arre | -0.002 |
Token | friends |
Token ID | 2460 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
itating | +0.014 |
arre | +0.013 |
ling | +0.013 |
up | +0.013 |
melon | +0.013 |
Neg logit contributions | |
year | -0.021 |
glass | -0.020 |
ird | -0.020 |
hours | -0.019 |
minutes | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +7.86e-01 |
Pos logit contributions | |
itating | +0.015 |
arre | +0.014 |
melon | +0.014 |
rust | +0.013 |
up | +0.013 |
Neg logit contributions | |
year | -0.021 |
glass | -0.020 |
ird | -0.020 |
hours | -0.019 |
minutes | -0.019 |
Token | He |
Token ID | 679 |
Feature activation | +6.88e-01 |
Pos logit contributions | |
itating | +0.017 |
ling | +0.017 |
arre | +0.017 |
up | +0.016 |
bing | +0.016 |
Neg logit contributions | |
year | -0.025 |
ird | -0.024 |
glass | -0.024 |
aster | -0.022 |
minutes | -0.022 |
Token | saw |
Token ID | 2497 |
Feature activation | +6.07e-01 |
Pos logit contributions | |
arre | +0.013 |
up | +0.013 |
itating | +0.013 |
ling | +0.013 |
melon | +0.013 |
Neg logit contributions | |
year | -0.023 |
glass | -0.022 |
ird | -0.022 |
hours | -0.021 |
aster | -0.021 |
Token | Sara |
Token ID | 24799 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
itating | +0.013 |
arre | +0.012 |
bing | +0.011 |
ling | +0.011 |
melon | +0.011 |
Neg logit contributions | |
year | -0.020 |
glass | -0.019 |
ird | -0.019 |
hours | -0.018 |
aster | -0.018 |
Token | 's |
Token ID | 338 |
Feature activation | +7.28e-01 |
Pos logit contributions | |
rust | +0.022 |
itating | +0.020 |
Freddy | +0.019 |
melon | +0.019 |
arre | +0.019 |
Neg logit contributions | |
glass | -0.033 |
ird | -0.032 |
year | -0.032 |
aster | -0.031 |
lock | -0.029 |
Token | rubber |
Token ID | 14239 |
Feature activation | +4.22e+00 |
Pos logit contributions | |
itating | +0.017 |
arre | +0.017 |
melon | +0.016 |
up | +0.016 |
ling | +0.016 |
Neg logit contributions | |
year | -0.021 |
glass | -0.021 |
ird | -0.021 |
hours | -0.019 |
aster | -0.019 |
Token | duck |
Token ID | 22045 |
Feature activation | +2.65e+00 |
Pos logit contributions | |
itating | +0.112 |
itionally | +0.098 |
rust | +0.097 |
arre | +0.095 |
appointed | +0.094 |
Neg logit contributions | |
glass | -0.100 |
bers | -0.098 |
ird | -0.098 |
hours | -0.093 |
minutes | -0.090 |
Token | and |
Token ID | 290 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
rust | +0.060 |
itating | +0.057 |
Sharing | +0.054 |
Being | +0.054 |
arre | +0.054 |
Neg logit contributions | |
glass | -0.066 |
ird | -0.063 |
e | -0.061 |
ns | -0.060 |
kil | -0.060 |
Token | wanted |
Token ID | 2227 |
Feature activation | +7.00e-01 |
Pos logit contributions | |
itating | +0.052 |
arre | +0.049 |
up | +0.045 |
bing | +0.044 |
ling | +0.044 |
Neg logit contributions | |
year | -0.057 |
ird | -0.057 |
glass | -0.055 |
hours | -0.055 |
aster | -0.054 |
Token | to |
Token ID | 284 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
itating | +0.015 |
arre | +0.015 |
up | +0.014 |
melon | +0.014 |
ling | +0.014 |
Neg logit contributions | |
year | -0.022 |
glass | -0.021 |
ird | -0.021 |
hours | -0.020 |
aster | -0.020 |
Token | take |
Token ID | 1011 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
itating | +0.057 |
arre | +0.057 |
ling | +0.053 |
sets | +0.052 |
bing | +0.052 |
Neg logit contributions | |
aster | -0.046 |
ird | -0.043 |
year | -0.039 |
glass | -0.038 |
hours | -0.038 |
Token |
|
Token ID | 198 |
Feature activation | -5.19e-01 |
Pos logit contributions | |
year | +0.043 |
ird | +0.043 |
glass | +0.042 |
hours | +0.040 |
ibr | +0.039 |
Neg logit contributions | |
up | -0.011 |
itating | -0.010 |
ling | -0.009 |
arre | -0.009 |
fighting | -0.009 |
Token | M |
Token ID | 44 |
Feature activation | -3.04e+00 |
Pos logit contributions | |
ird | +0.020 |
year | +0.020 |
ocket | +0.019 |
glass | +0.019 |
hours | +0.019 |
Neg logit contributions | |
itating | -0.007 |
ling | -0.007 |
up | -0.006 |
bing | -0.006 |
dy | -0.006 |
Token | ira |
Token ID | 8704 |
Feature activation | -8.66e-01 |
Pos logit contributions | |
year | +0.094 |
glass | +0.090 |
ird | +0.088 |
hours | +0.088 |
ages | +0.080 |
Neg logit contributions | |
ucker | -0.070 |
ling | -0.064 |
up | -0.064 |
ies | -0.059 |
melon | -0.058 |
Token | looked |
Token ID | 3114 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
year | +0.029 |
ird | +0.027 |
glass | +0.027 |
hours | +0.026 |
aster | +0.026 |
Neg logit contributions | |
itating | -0.017 |
arre | -0.017 |
up | -0.016 |
melon | -0.016 |
ling | -0.016 |
Token | at |
Token ID | 379 |
Feature activation | -4.04e-01 |
Pos logit contributions | |
itating | +0.005 |
arre | +0.005 |
melon | +0.005 |
inas | +0.005 |
rust | +0.005 |
Neg logit contributions | |
year | -0.006 |
minutes | -0.006 |
ird | -0.006 |
glass | -0.006 |
hours | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | +4.21e+00 |
Pos logit contributions | |
year | +0.014 |
ird | +0.013 |
glass | +0.013 |
hours | +0.012 |
minutes | +0.012 |
Neg logit contributions | |
up | -0.008 |
itating | -0.008 |
arre | -0.007 |
melon | -0.007 |
ling | -0.007 |
Token | b |
Token ID | 275 |
Feature activation | -7.09e-01 |
Pos logit contributions | |
itating | +0.118 |
arre | +0.110 |
bing | +0.108 |
mill | +0.108 |
ocked | +0.102 |
Neg logit contributions | |
hours | -0.093 |
minutes | -0.091 |
year | -0.090 |
kil | -0.082 |
ird | -0.082 |
Token | erry |
Token ID | 6996 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
year | +0.015 |
ird | +0.014 |
glass | +0.013 |
aster | +0.012 |
hours | +0.012 |
Neg logit contributions | |
itating | -0.023 |
arre | -0.023 |
up | -0.022 |
fighting | -0.022 |
melon | -0.021 |
Token | in |
Token ID | 287 |
Feature activation | -7.53e-02 |
Pos logit contributions | |
arre | +0.040 |
itating | +0.039 |
rust | +0.036 |
fighting | +0.034 |
inas | +0.034 |
Neg logit contributions | |
year | -0.054 |
glass | -0.051 |
ird | -0.050 |
hours | -0.049 |
aster | -0.049 |
Token | her |
Token ID | 607 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
year | +0.003 |
glass | +0.003 |
ird | +0.003 |
hours | +0.002 |
aster | +0.002 |
Neg logit contributions | |
itating | -0.001 |
arre | -0.001 |
up | -0.001 |
melon | -0.001 |
fighting | -0.001 |
Token | hand |
Token ID | 1021 |
Feature activation | +3.34e-01 |
Pos logit contributions | |
year | +0.004 |
glass | +0.004 |
ird | +0.004 |
hours | +0.003 |
aster | +0.003 |
Neg logit contributions | |
itating | -0.003 |
arre | -0.003 |
up | -0.003 |
melon | -0.003 |
ling | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +8.52e-01 |
Pos logit contributions | |
arre | +0.022 |
itating | +0.021 |
ling | +0.020 |
up | +0.019 |
melon | +0.019 |
Neg logit contributions | |
year | -0.029 |
ird | -0.029 |
glass | -0.028 |
hours | -0.027 |
minutes | -0.026 |
Token | Every |
Token ID | 3887 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
itating | +0.017 |
arre | +0.017 |
up | +0.016 |
ling | +0.016 |
melon | +0.016 |
Neg logit contributions | |
year | -0.028 |
ird | -0.027 |
glass | -0.027 |
hours | -0.026 |
aster | -0.025 |
Token | day |
Token ID | 1110 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
itating | +0.003 |
arre | +0.003 |
up | +0.003 |
ling | +0.003 |
melon | +0.003 |
Neg logit contributions | |
year | -0.003 |
glass | -0.003 |
ird | -0.003 |
hours | -0.003 |
aster | -0.003 |
Token | he |
Token ID | 339 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
year | +0.005 |
ird | +0.005 |
ibr | +0.005 |
glass | +0.005 |
ibrarian | +0.005 |
Neg logit contributions | |
up | -0.002 |
fighting | -0.001 |
Freddy | -0.001 |
melon | -0.001 |
accidents | -0.001 |
Token | would |
Token ID | 561 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
up | +0.032 |
melon | +0.032 |
arre | +0.032 |
ling | +0.031 |
itating | +0.030 |
Neg logit contributions | |
year | -0.051 |
glass | -0.049 |
ird | -0.048 |
hours | -0.044 |
aster | -0.044 |
Token | pick |
Token ID | 2298 |
Feature activation | +4.19e+00 |
Pos logit contributions | |
itating | +0.041 |
arre | +0.041 |
ling | +0.039 |
bing | +0.039 |
up | +0.037 |
Neg logit contributions | |
ird | -0.044 |
year | -0.044 |
aster | -0.043 |
glass | -0.042 |
hours | -0.041 |
Token | one |
Token ID | 530 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
itating | +0.123 |
ivering | +0.113 |
arre | +0.112 |
bing | +0.107 |
inas | +0.106 |
Neg logit contributions | |
year | -0.097 |
hours | -0.090 |
aster | -0.082 |
Africa | -0.082 |
book | -0.080 |
Token | from |
Token ID | 422 |
Feature activation | +1.88e-01 |
Pos logit contributions | |
arre | +0.036 |
itating | +0.034 |
fighting | +0.032 |
bing | +0.032 |
inas | +0.031 |
Neg logit contributions | |
year | -0.035 |
ird | -0.033 |
hours | -0.031 |
aster | -0.031 |
minutes | -0.031 |
Token | the |
Token ID | 262 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
up | +0.003 |
melon | +0.003 |
arre | +0.003 |
itating | +0.003 |
ling | +0.003 |
Neg logit contributions | |
year | -0.007 |
ird | -0.007 |
glass | -0.006 |
hours | -0.006 |
minutes | -0.006 |
Token | bush |
Token ID | 24484 |
Feature activation | -3.38e-01 |
Pos logit contributions | |
arre | +0.059 |
itating | +0.059 |
ling | +0.059 |
bing | +0.058 |
mill | +0.057 |
Neg logit contributions | |
year | -0.070 |
ird | -0.070 |
glass | -0.069 |
hours | -0.068 |
aster | -0.065 |
Token | in |
Token ID | 287 |
Feature activation | -6.06e-01 |
Pos logit contributions | |
year | +0.012 |
ird | +0.012 |
glass | +0.012 |
hours | +0.011 |
minutes | +0.011 |
Neg logit contributions | |
up | -0.006 |
itating | -0.006 |
melon | -0.006 |
arre | -0.006 |
ling | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
year | +0.022 |
ird | +0.021 |
glass | +0.021 |
kil | +0.019 |
lock | +0.019 |
Neg logit contributions | |
up | -0.016 |
arre | -0.016 |
itating | -0.015 |
melon | -0.015 |
fighting | -0.015 |
Token | tiny |
Token ID | 7009 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
year | +0.063 |
glass | +0.062 |
ird | +0.061 |
hours | +0.057 |
ibr | +0.057 |
Neg logit contributions | |
up | -0.022 |
ling | -0.021 |
melon | -0.021 |
arre | -0.021 |
fighting | -0.021 |
Token | girl |
Token ID | 2576 |
Feature activation | +8.56e-01 |
Pos logit contributions | |
year | +0.052 |
glass | +0.051 |
ird | +0.050 |
hours | +0.047 |
aster | +0.047 |
Neg logit contributions | |
melon | -0.022 |
up | -0.022 |
ling | -0.021 |
arre | -0.020 |
fighting | -0.020 |
Token | named |
Token ID | 3706 |
Feature activation | -9.77e-01 |
Pos logit contributions | |
up | +0.023 |
ling | +0.022 |
fighting | +0.022 |
itating | +0.021 |
melon | +0.021 |
Neg logit contributions | |
year | -0.022 |
glass | -0.022 |
ird | -0.021 |
hours | -0.020 |
ibr | -0.020 |
Token | Freddy |
Token ID | 45437 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
year | +0.040 |
glass | +0.040 |
ird | +0.038 |
hours | +0.037 |
aster | +0.037 |
Neg logit contributions | |
up | -0.012 |
itating | -0.011 |
arre | -0.011 |
melon | -0.010 |
fighting | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +4.18e+00 |
Pos logit contributions | |
arre | +0.036 |
itating | +0.036 |
melon | +0.034 |
up | +0.033 |
ling | +0.032 |
Neg logit contributions | |
year | -0.050 |
ird | -0.047 |
glass | -0.047 |
hours | -0.044 |
aster | -0.044 |
Token | She |
Token ID | 1375 |
Feature activation | +5.12e-01 |
Pos logit contributions | |
ning | +0.129 |
mill | +0.129 |
itating | +0.126 |
bing | +0.125 |
izing | +0.120 |
Neg logit contributions | |
year | -0.096 |
hours | -0.083 |
book | -0.075 |
minutes | -0.075 |
ages | -0.068 |
Token | lived |
Token ID | 5615 |
Feature activation | +7.69e-01 |
Pos logit contributions | |
itating | +0.012 |
arre | +0.012 |
melon | +0.011 |
up | +0.011 |
ling | +0.011 |
Neg logit contributions | |
year | -0.016 |
glass | -0.015 |
ird | -0.015 |
hours | -0.014 |
aster | -0.014 |
Token | in |
Token ID | 287 |
Feature activation | -6.65e-01 |
Pos logit contributions | |
itating | +0.014 |
arre | +0.014 |
up | +0.014 |
melon | +0.014 |
ling | +0.013 |
Neg logit contributions | |
year | -0.026 |
glass | -0.025 |
ird | -0.025 |
hours | -0.024 |
aster | -0.023 |
Token | a |
Token ID | 257 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
year | +0.020 |
ird | +0.019 |
glass | +0.019 |
hours | +0.017 |
minutes | +0.017 |
Neg logit contributions | |
up | -0.015 |
itating | -0.015 |
arre | -0.015 |
melon | -0.015 |
ling | -0.014 |
Token | big |
Token ID | 1263 |
Feature activation | -6.69e-01 |
Pos logit contributions | |
year | +0.040 |
ird | +0.038 |
glass | +0.038 |
hours | +0.036 |
aster | +0.035 |
Neg logit contributions | |
itating | -0.027 |
arre | -0.027 |
up | -0.026 |
melon | -0.025 |
ling | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
year | +0.062 |
ird | +0.058 |
glass | +0.058 |
ibr | +0.056 |
aster | +0.056 |
Neg logit contributions | |
up | -0.027 |
itating | -0.020 |
fighting | -0.019 |
ling | -0.019 |
arre | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | -8.43e-01 |
Pos logit contributions | |
itating | +0.004 |
arre | +0.004 |
ling | +0.004 |
up | +0.004 |
melon | +0.004 |
Neg logit contributions | |
year | -0.006 |
ird | -0.006 |
glass | -0.006 |
hours | -0.006 |
aster | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +9.23e-01 |
Pos logit contributions | |
year | +0.031 |
ird | +0.030 |
glass | +0.030 |
hours | +0.029 |
minutes | +0.028 |
Neg logit contributions | |
itating | -0.013 |
arre | -0.013 |
up | -0.013 |
ling | -0.012 |
melon | -0.012 |
Token | At |
Token ID | 2953 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
itating | +0.022 |
up | +0.022 |
arre | +0.021 |
melon | +0.021 |
ling | +0.021 |
Neg logit contributions | |
year | -0.027 |
glass | -0.026 |
ird | -0.026 |
hours | -0.024 |
aster | -0.024 |
Token | the |
Token ID | 262 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
year | +0.059 |
ird | +0.057 |
glass | +0.056 |
ibr | +0.054 |
kil | +0.053 |
Neg logit contributions | |
up | -0.019 |
fighting | -0.013 |
itating | -0.013 |
melon | -0.012 |
bing | -0.011 |
Token | bar |
Token ID | 2318 |
Feature activation | -6.03e+00 |
Pos logit contributions | |
year | +0.071 |
glass | +0.065 |
ocket | +0.061 |
ird | +0.061 |
lan | +0.061 |
Neg logit contributions | |
up | -0.035 |
same | -0.029 |
itating | -0.026 |
fighting | -0.026 |
me | -0.024 |
Token | ber |
Token ID | 527 |
Feature activation | -3.53e+00 |
Pos logit contributions | |
hours | +0.131 |
year | +0.128 |
ocket | +0.126 |
Africa | +0.121 |
glass | +0.119 |
Neg logit contributions | |
up | -0.171 |
itating | -0.153 |
by | -0.151 |
ling | -0.148 |
fighting | -0.147 |
Token | , |
Token ID | 11 |
Feature activation | -5.60e-01 |
Pos logit contributions | |
year | +0.164 |
lan | +0.144 |
hours | +0.142 |
minutes | +0.141 |
glass | +0.140 |
Neg logit contributions | |
up | -0.050 |
's | -0.033 |
my | -0.032 |
fighting | -0.027 |
ning | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
year | +0.021 |
ird | +0.021 |
glass | +0.020 |
ris | +0.019 |
minutes | +0.019 |
Neg logit contributions | |
up | -0.009 |
fighting | -0.008 |
Freddy | -0.008 |
melon | -0.008 |
itating | -0.008 |
Token | bar |
Token ID | 2318 |
Feature activation | -5.63e+00 |
Pos logit contributions | |
year | +0.085 |
glass | +0.079 |
ird | +0.077 |
ris | +0.077 |
ocket | +0.076 |
Neg logit contributions | |
up | -0.033 |
same | -0.029 |
fighting | -0.025 |
hurting | -0.024 |
frown | -0.023 |
Token | ber |
Token ID | 527 |
Feature activation | -3.09e+00 |
Pos logit contributions | |
Africa | +0.133 |
hours | +0.125 |
takeoff | +0.114 |
ocket | +0.114 |
growth | +0.113 |
Neg logit contributions | |
up | -0.147 |
ling | -0.145 |
y | -0.143 |
mill | -0.141 |
ing | -0.138 |
Token | boots |
Token ID | 14412 |
Feature activation | -2.31e+00 |
Pos logit contributions | |
year | +0.076 |
ris | +0.067 |
ird | +0.067 |
ibr | +0.066 |
ibrarian | +0.066 |
Neg logit contributions | |
up | -0.034 |
qu | -0.029 |
bl | -0.028 |
swe | -0.025 |
Bugs | -0.025 |
Token | and |
Token ID | 290 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
year | +0.098 |
ird | +0.093 |
glass | +0.092 |
aster | +0.091 |
hours | +0.090 |
Neg logit contributions | |
up | -0.028 |
itating | -0.024 |
melon | -0.024 |
ling | -0.023 |
arre | -0.023 |
Token | climbed |
Token ID | 19952 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
year | +0.073 |
ae | +0.065 |
ris | +0.065 |
ibr | +0.064 |
ibrarian | +0.064 |
Neg logit contributions | |
up | -0.030 |
qu | -0.023 |
bl | -0.022 |
scatter | -0.021 |
swe | -0.021 |
Token | on |
Token ID | 319 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
ird | +0.048 |
year | +0.047 |
glass | +0.043 |
ibr | +0.043 |
aster | +0.043 |
Neg logit contributions | |
up | -0.023 |
ling | -0.014 |
melon | -0.014 |
bing | -0.014 |
mill | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
year | +0.079 |
ird | +0.076 |
glass | +0.073 |
ibr | +0.071 |
lock | +0.070 |
Neg logit contributions | |
up | -0.025 |
fighting | -0.018 |
melon | -0.017 |
Freddy | -0.016 |
bing | -0.015 |
Token | sle |
Token ID | 3133 |
Feature activation | -5.97e+00 |
Pos logit contributions | |
year | +0.040 |
ird | +0.037 |
aster | +0.033 |
ibr | +0.033 |
glass | +0.032 |
Neg logit contributions | |
up | -0.033 |
ling | -0.030 |
melon | -0.028 |
bing | -0.027 |
itating | -0.027 |
Token | igh |
Token ID | 394 |
Feature activation | -2.67e+00 |
Pos logit contributions | |
hours | +0.168 |
year | +0.164 |
minutes | +0.162 |
Ah | +0.146 |
ird | +0.145 |
Neg logit contributions | |
flower | -0.132 |
ling | -0.132 |
ocked | -0.129 |
sets | -0.127 |
my | -0.126 |
Token | . |
Token ID | 13 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
year | +0.112 |
hours | +0.104 |
ird | +0.103 |
aster | +0.101 |
minutes | +0.100 |
Neg logit contributions | |
up | -0.040 |
ling | -0.034 |
itating | -0.032 |
bing | -0.030 |
mill | -0.028 |
Token | Santa |
Token ID | 8909 |
Feature activation | -2.16e+00 |
Pos logit contributions | |
year | +0.065 |
ird | +0.065 |
glass | +0.063 |
ris | +0.060 |
aster | +0.060 |
Neg logit contributions | |
up | -0.012 |
itating | -0.009 |
Sharing | -0.008 |
fighting | -0.008 |
arre | -0.007 |
Token | gave |
Token ID | 2921 |
Feature activation | +3.74e-02 |
Pos logit contributions | |
year | +0.078 |
ird | +0.074 |
aster | +0.072 |
hours | +0.070 |
glass | +0.069 |
Neg logit contributions | |
up | -0.041 |
ling | -0.039 |
itating | -0.037 |
fighting | -0.036 |
melon | -0.036 |
Token | them |
Token ID | 606 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
up | +0.000 |
itating | +0.000 |
ling | +0.000 |
arre | +0.000 |
melon | +0.000 |
Neg logit contributions | |
year | -0.002 |
ird | -0.002 |
glass | -0.002 |
hours | -0.001 |
lock | -0.001 |
Token | instead |
Token ID | 2427 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
year | +0.045 |
glass | +0.043 |
ird | +0.042 |
hours | +0.041 |
minutes | +0.040 |
Neg logit contributions | |
up | -0.022 |
arre | -0.021 |
itating | -0.021 |
ling | -0.020 |
melon | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -6.75e-01 |
Pos logit contributions | |
year | +0.044 |
glass | +0.043 |
ird | +0.041 |
hours | +0.039 |
ibr | +0.039 |
Neg logit contributions | |
up | -0.029 |
ling | -0.028 |
itating | -0.027 |
arre | -0.026 |
fighting | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | -8.61e-01 |
Pos logit contributions | |
year | +0.025 |
glass | +0.024 |
ird | +0.024 |
hours | +0.023 |
aster | +0.022 |
Neg logit contributions | |
itating | -0.011 |
arre | -0.011 |
up | -0.011 |
melon | -0.010 |
fighting | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
year | +0.033 |
ird | +0.032 |
glass | +0.032 |
hours | +0.031 |
minutes | +0.030 |
Neg logit contributions | |
itating | -0.012 |
arre | -0.012 |
up | -0.012 |
ling | -0.011 |
melon | -0.011 |
Token | The |
Token ID | 464 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
itating | +0.000 |
arre | +0.000 |
up | +0.000 |
melon | +0.000 |
ling | +0.000 |
Neg logit contributions | |
year | -0.000 |
ird | -0.000 |
glass | -0.000 |
hours | -0.000 |
minutes | -0.000 |
Token | bar |
Token ID | 2318 |
Feature activation | -5.82e+00 |
Pos logit contributions | |
ird | +0.072 |
glass | +0.072 |
year | +0.071 |
adders | +0.068 |
lan | +0.067 |
Neg logit contributions | |
up | -0.033 |
fighting | -0.028 |
ling | -0.026 |
hurting | -0.026 |
en | -0.026 |
Token | ber |
Token ID | 527 |
Feature activation | -2.96e+00 |
Pos logit contributions | |
Africa | +0.126 |
hours | +0.123 |
ages | +0.109 |
aster | +0.106 |
ird | +0.103 |
Neg logit contributions | |
ling | -0.176 |
mill | -0.166 |
up | -0.161 |
itating | -0.161 |
fighting | -0.155 |
Token | didn |
Token ID | 1422 |
Feature activation | -2.82e-01 |
Pos logit contributions | |
year | +0.107 |
glass | +0.102 |
ird | +0.099 |
Africa | +0.096 |
hours | +0.096 |
Neg logit contributions | |
arre | -0.051 |
up | -0.050 |
ling | -0.047 |
fighting | -0.046 |
melon | -0.046 |
Token | 't |
Token ID | 470 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
year | +0.010 |
ird | +0.009 |
aster | +0.009 |
glass | +0.009 |
lock | +0.009 |
Neg logit contributions | |
ling | -0.005 |
flower | -0.005 |
arre | -0.005 |
lings | -0.005 |
ety | -0.005 |
Token | want |
Token ID | 765 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
year | +0.039 |
ird | +0.039 |
glass | +0.036 |
aster | +0.034 |
hours | +0.033 |
Neg logit contributions | |
itating | -0.054 |
bing | -0.054 |
melon | -0.053 |
arre | -0.053 |
up | -0.053 |
Token | to |
Token ID | 284 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
year | +0.061 |
ird | +0.059 |
glass | +0.058 |
lock | +0.053 |
ibr | +0.052 |
Neg logit contributions | |
up | -0.030 |
Freddy | -0.027 |
fighting | -0.026 |
melon | -0.026 |
arre | -0.025 |
Token | nice |
Token ID | 3621 |
Feature activation | -5.83e-01 |
Pos logit contributions | |
glass | +0.031 |
hours | +0.029 |
year | +0.029 |
ird | +0.028 |
aster | +0.028 |
Neg logit contributions | |
up | -0.017 |
itating | -0.016 |
melon | -0.015 |
arre | -0.015 |
ling | -0.015 |
Token | ." |
Token ID | 526 |
Feature activation | -4.52e-01 |
Pos logit contributions | |
year | +0.024 |
glass | +0.023 |
aster | +0.022 |
ird | +0.022 |
hours | +0.021 |
Neg logit contributions | |
up | -0.008 |
arre | -0.007 |
fighting | -0.007 |
mill | -0.006 |
sharing | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | -8.67e-01 |
Pos logit contributions | |
year | +0.017 |
ird | +0.016 |
glass | +0.016 |
hours | +0.015 |
aster | +0.015 |
Neg logit contributions | |
arre | -0.007 |
itating | -0.007 |
up | -0.007 |
melon | -0.007 |
fighting | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
year | +0.032 |
ird | +0.031 |
glass | +0.030 |
hours | +0.029 |
minutes | +0.029 |
Neg logit contributions | |
itating | -0.014 |
arre | -0.014 |
up | -0.014 |
melon | -0.013 |
ling | -0.013 |
Token | The |
Token ID | 464 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
itating | +0.000 |
arre | +0.000 |
up | +0.000 |
melon | +0.000 |
ling | +0.000 |
Neg logit contributions | |
year | -0.001 |
ird | -0.001 |
glass | -0.001 |
hours | -0.001 |
minutes | -0.001 |
Token | bar |
Token ID | 2318 |
Feature activation | -5.81e+00 |
Pos logit contributions | |
glass | +0.062 |
ird | +0.061 |
year | +0.060 |
ris | +0.057 |
hours | +0.056 |
Neg logit contributions | |
up | -0.028 |
fighting | -0.022 |
hurting | -0.021 |
same | -0.021 |
ling | -0.020 |
Token | ber |
Token ID | 527 |
Feature activation | -3.01e+00 |
Pos logit contributions | |
Africa | +0.124 |
hours | +0.122 |
ages | +0.107 |
aster | +0.104 |
glass | +0.102 |
Neg logit contributions | |
mill | -0.174 |
ling | -0.173 |
up | -0.167 |
fighting | -0.162 |
melon | -0.159 |
Token | gave |
Token ID | 2921 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
year | +0.110 |
glass | +0.102 |
ird | +0.101 |
Africa | +0.098 |
hours | +0.098 |
Neg logit contributions | |
up | -0.053 |
arre | -0.049 |
melon | -0.048 |
fighting | -0.047 |
ling | -0.046 |
Token | Tom |
Token ID | 4186 |
Feature activation | -4.37e-01 |
Pos logit contributions | |
year | +0.004 |
ird | +0.004 |
glass | +0.004 |
hours | +0.003 |
minutes | +0.003 |
Neg logit contributions | |
up | -0.003 |
arre | -0.003 |
itating | -0.003 |
melon | -0.003 |
ling | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | -2.14e+00 |
Pos logit contributions | |
year | +0.013 |
glass | +0.012 |
ird | +0.012 |
hours | +0.012 |
aster | +0.011 |
Neg logit contributions | |
itating | -0.010 |
arre | -0.010 |
up | -0.010 |
melon | -0.010 |
ling | -0.010 |
Token | hug |
Token ID | 16225 |
Feature activation | -9.48e-01 |
Pos logit contributions | |
year | +0.071 |
glass | +0.070 |
hours | +0.068 |
ird | +0.067 |
minutes | +0.064 |
Neg logit contributions | |
up | -0.044 |
arre | -0.042 |
melon | -0.040 |
itating | -0.040 |
ling | -0.037 |
Token | a |
Token ID | 257 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
ird | +0.074 |
glass | +0.072 |
kil | +0.072 |
year | +0.070 |
ait | +0.069 |
Neg logit contributions | |
up | -0.031 |
Freddy | -0.023 |
fighting | -0.023 |
sharing | -0.021 |
bouncing | -0.020 |
Token | bar |
Token ID | 2318 |
Feature activation | -4.50e+00 |
Pos logit contributions | |
year | +0.073 |
glass | +0.071 |
ird | +0.071 |
ibr | +0.066 |
hours | +0.066 |
Neg logit contributions | |
up | -0.027 |
fighting | -0.026 |
itating | -0.026 |
melon | -0.026 |
ling | -0.025 |
Token | ber |
Token ID | 527 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
hours | +0.104 |
aster | +0.090 |
ird | +0.089 |
isc | +0.087 |
takeoff | +0.083 |
Neg logit contributions | |
ling | -0.126 |
up | -0.125 |
fighting | -0.124 |
y | -0.122 |
melon | -0.119 |
Token | . |
Token ID | 13 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
year | +0.072 |
hours | +0.068 |
ibr | +0.067 |
Ah | +0.065 |
minutes | +0.065 |
Neg logit contributions | |
up | -0.022 |
fighting | -0.019 |
flower | -0.018 |
sharing | -0.018 |
Sharing | -0.018 |
Token | The |
Token ID | 383 |
Feature activation | -2.73e+00 |
Pos logit contributions | |
glass | +0.070 |
year | +0.070 |
ird | +0.069 |
ocket | +0.067 |
izzard | +0.066 |
Neg logit contributions | |
Freddy | -0.016 |
Sharing | -0.016 |
fighting | -0.014 |
She | -0.013 |
up | -0.013 |
Token | bar |
Token ID | 2318 |
Feature activation | -5.76e+00 |
Pos logit contributions | |
year | +0.110 |
glass | +0.104 |
ird | +0.104 |
ris | +0.103 |
ocket | +0.103 |
Neg logit contributions | |
up | -0.045 |
same | -0.039 |
fighting | -0.032 |
sad | -0.031 |
en | -0.031 |
Token | ber |
Token ID | 527 |
Feature activation | -2.42e+00 |
Pos logit contributions | |
hours | +0.133 |
aster | +0.122 |
Africa | +0.113 |
ird | +0.111 |
ages | +0.109 |
Neg logit contributions | |
ling | -0.164 |
up | -0.157 |
mill | -0.155 |
y | -0.154 |
en | -0.147 |
Token | was |
Token ID | 373 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
year | +0.108 |
aster | +0.100 |
ird | +0.098 |
ait | +0.098 |
kil | +0.097 |
Neg logit contributions | |
up | -0.023 |
fighting | -0.019 |
my | -0.018 |
sharing | -0.016 |
melon | -0.016 |
Token | a |
Token ID | 257 |
Feature activation | -3.44e+00 |
Pos logit contributions | |
ird | +0.068 |
glass | +0.067 |
year | +0.065 |
hours | +0.064 |
ris | +0.063 |
Neg logit contributions | |
up | -0.038 |
fighting | -0.033 |
melon | -0.030 |
en | -0.030 |
itating | -0.030 |
Token | funny |
Token ID | 8258 |
Feature activation | -3.13e+00 |
Pos logit contributions | |
year | +0.141 |
glass | +0.139 |
adders | +0.137 |
ird | +0.134 |
ibr | +0.131 |
Neg logit contributions | |
sad | -0.040 |
en | -0.040 |
up | -0.040 |
fighting | -0.037 |
qu | -0.034 |
Token | man |
Token ID | 582 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
year | +0.123 |
glass | +0.110 |
ird | +0.106 |
ibr | +0.105 |
adders | +0.104 |
Neg logit contributions | |
up | -0.057 |
ling | -0.048 |
fighting | -0.047 |
arre | -0.046 |
mill | -0.043 |
Token | scared |
Token ID | 12008 |
Feature activation | +3.74e-01 |
Pos logit contributions | |
glass | +0.025 |
ird | +0.025 |
year | +0.024 |
hours | +0.024 |
aster | +0.023 |
Neg logit contributions | |
up | -0.011 |
itating | -0.010 |
ling | -0.010 |
fighting | -0.010 |
melon | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
up | +0.007 |
itating | +0.007 |
arre | +0.007 |
ling | +0.006 |
fighting | +0.006 |
Neg logit contributions | |
year | -0.013 |
ird | -0.013 |
glass | -0.013 |
hours | -0.012 |
aster | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
year | +0.046 |
ird | +0.044 |
glass | +0.044 |
hours | +0.043 |
aster | +0.042 |
Neg logit contributions | |
Sharing | -0.010 |
up | -0.010 |
Freddy | -0.010 |
itating | -0.009 |
fighting | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | +4.59e-02 |
Pos logit contributions | |
ird | +0.049 |
year | +0.049 |
glass | +0.047 |
hours | +0.046 |
minutes | +0.045 |
Neg logit contributions | |
up | -0.015 |
itating | -0.014 |
arre | -0.013 |
ling | -0.012 |
fighting | -0.012 |
Token | The |
Token ID | 464 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
itating | +0.001 |
arre | +0.001 |
up | +0.001 |
ling | +0.001 |
fighting | +0.001 |
Neg logit contributions | |
ird | -0.002 |
year | -0.002 |
glass | -0.002 |
hours | -0.001 |
ocket | -0.001 |
Token | bar |
Token ID | 2318 |
Feature activation | -5.75e+00 |
Pos logit contributions | |
year | +0.048 |
ird | +0.048 |
glass | +0.045 |
ris | +0.044 |
ocket | +0.043 |
Neg logit contributions | |
up | -0.019 |
same | -0.016 |
fighting | -0.016 |
hurting | -0.016 |
mess | -0.015 |
Token | ber |
Token ID | 527 |
Feature activation | -2.91e+00 |
Pos logit contributions | |
hours | +0.131 |
Africa | +0.126 |
ages | +0.118 |
ird | +0.115 |
prise | +0.109 |
Neg logit contributions | |
ling | -0.163 |
mill | -0.154 |
up | -0.149 |
fighting | -0.148 |
y | -0.142 |
Token | started |
Token ID | 2067 |
Feature activation | -9.37e-02 |
Pos logit contributions | |
year | +0.110 |
lan | +0.096 |
Africa | +0.096 |
ird | +0.095 |
kil | +0.095 |
Neg logit contributions | |
up | -0.048 |
fighting | -0.048 |
arre | -0.043 |
ling | -0.043 |
melon | -0.041 |
Token | to |
Token ID | 284 |
Feature activation | -7.57e-01 |
Pos logit contributions | |
year | +0.003 |
ird | +0.003 |
glass | +0.003 |
hours | +0.003 |
aster | +0.003 |
Neg logit contributions | |
arre | -0.002 |
itating | -0.002 |
up | -0.002 |
melon | -0.002 |
ling | -0.002 |
Token | cut |
Token ID | 2005 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
ird | +0.022 |
aster | +0.022 |
year | +0.022 |
glass | +0.021 |
hours | +0.020 |
Neg logit contributions | |
itating | -0.017 |
arre | -0.017 |
ling | -0.016 |
melon | -0.016 |
ocked | -0.016 |
Token | Tom |
Token ID | 4186 |
Feature activation | +7.91e-01 |
Pos logit contributions | |
year | +0.008 |
ird | +0.008 |
glass | +0.008 |
Bible | +0.007 |
ibr | +0.007 |
Neg logit contributions | |
up | -0.005 |
ling | -0.004 |
fighting | -0.004 |
bing | -0.003 |
melon | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +3.42e-02 |
Pos logit contributions | |
year | +0.083 |
glass | +0.078 |
hours | +0.077 |
aster | +0.077 |
ird | +0.076 |
Neg logit contributions | |
up | -0.030 |
fighting | -0.025 |
itating | -0.024 |
melon | -0.023 |
ling | -0.023 |
Token | |
Token ID | 220 |
Feature activation | -5.95e-01 |
Pos logit contributions | |
itating | +0.001 |
arre | +0.001 |
up | +0.001 |
melon | +0.001 |
ling | +0.001 |
Neg logit contributions | |
year | -0.001 |
glass | -0.001 |
ird | -0.001 |
hours | -0.001 |
aster | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -8.00e-01 |
Pos logit contributions | |
year | +0.020 |
ird | +0.019 |
glass | +0.019 |
hours | +0.018 |
aster | +0.018 |
Neg logit contributions | |
arre | -0.011 |
itating | -0.011 |
melon | -0.011 |
up | -0.011 |
ling | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
year | +0.030 |
glass | +0.030 |
ird | +0.029 |
hours | +0.028 |
minutes | +0.028 |
Neg logit contributions | |
itating | -0.012 |
arre | -0.012 |
up | -0.011 |
ling | -0.011 |
melon | -0.010 |
Token | The |
Token ID | 464 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
itating | +0.005 |
arre | +0.005 |
ling | +0.004 |
up | +0.004 |
melon | +0.004 |
Neg logit contributions | |
year | -0.009 |
glass | -0.008 |
ird | -0.008 |
hours | -0.008 |
minutes | -0.008 |
Token | bar |
Token ID | 2318 |
Feature activation | -5.71e+00 |
Pos logit contributions | |
year | +0.071 |
ird | +0.071 |
glass | +0.069 |
adders | +0.067 |
ris | +0.067 |
Neg logit contributions | |
up | -0.030 |
same | -0.025 |
fighting | -0.024 |
hurting | -0.024 |
ling | -0.023 |
Token | ber |
Token ID | 527 |
Feature activation | -2.90e+00 |
Pos logit contributions | |
Africa | +0.131 |
hours | +0.129 |
ages | +0.115 |
aster | +0.114 |
ird | +0.113 |
Neg logit contributions | |
ling | -0.157 |
mill | -0.151 |
y | -0.144 |
up | -0.144 |
fighting | -0.134 |
Token | then |
Token ID | 788 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
year | +0.110 |
glass | +0.101 |
Africa | +0.100 |
kil | +0.099 |
aster | +0.099 |
Neg logit contributions | |
up | -0.045 |
arre | -0.042 |
fighting | -0.041 |
ling | -0.040 |
melon | -0.038 |
Token | started |
Token ID | 2067 |
Feature activation | -6.28e-01 |
Pos logit contributions | |
year | +0.046 |
glass | +0.044 |
ird | +0.043 |
Africa | +0.042 |
hours | +0.042 |
Neg logit contributions | |
up | -0.023 |
arre | -0.020 |
fighting | -0.020 |
ling | -0.019 |
melon | -0.018 |
Token | cl |
Token ID | 537 |
Feature activation | -3.35e-01 |
Pos logit contributions | |
year | +0.021 |
ird | +0.020 |
glass | +0.020 |
hours | +0.019 |
minutes | +0.018 |
Neg logit contributions | |
up | -0.013 |
arre | -0.013 |
itating | -0.013 |
melon | -0.012 |
fighting | -0.012 |
Token | apping |
Token ID | 5912 |
Feature activation | +9.19e-01 |
Pos logit contributions | |
year | +0.011 |
glass | +0.011 |
ird | +0.011 |
hours | +0.010 |
aster | +0.010 |
Neg logit contributions | |
arre | -0.007 |
itating | -0.007 |
up | -0.006 |
melon | -0.006 |
ling | -0.006 |
Token | mom |
Token ID | 1995 |
Feature activation | -4.86e-01 |
Pos logit contributions | |
year | +0.030 |
glass | +0.027 |
ird | +0.027 |
aster | +0.027 |
hours | +0.026 |
Neg logit contributions | |
up | -0.022 |
fighting | -0.020 |
arre | -0.019 |
itating | -0.019 |
melon | -0.019 |
Token | said |
Token ID | 531 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
year | +0.015 |
ird | +0.014 |
glass | +0.014 |
hours | +0.013 |
aster | +0.013 |
Neg logit contributions | |
itating | -0.011 |
arre | -0.011 |
up | -0.011 |
melon | -0.011 |
ling | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | -3.94e-01 |
Pos logit contributions | |
year | +0.039 |
glass | +0.037 |
ird | +0.036 |
hours | +0.036 |
ris | +0.035 |
Neg logit contributions | |
up | -0.016 |
melon | -0.016 |
fighting | -0.015 |
itating | -0.015 |
arre | -0.015 |
Token | " |
Token ID | 366 |
Feature activation | -2.90e-01 |
Pos logit contributions | |
year | +0.015 |
glass | +0.014 |
ird | +0.014 |
hours | +0.014 |
aster | +0.013 |
Neg logit contributions | |
itating | -0.006 |
arre | -0.006 |
up | -0.006 |
melon | -0.006 |
ling | -0.005 |
Token | A |
Token ID | 32 |
Feature activation | -3.95e+00 |
Pos logit contributions | |
year | +0.011 |
ird | +0.011 |
glass | +0.010 |
hours | +0.010 |
minutes | +0.010 |
Neg logit contributions | |
itating | -0.005 |
arre | -0.004 |
up | -0.004 |
melon | -0.004 |
ling | -0.004 |
Token | bar |
Token ID | 2318 |
Feature activation | -5.69e+00 |
Pos logit contributions | |
year | +0.132 |
glass | +0.128 |
hours | +0.124 |
adders | +0.121 |
ibr | +0.121 |
Neg logit contributions | |
up | -0.079 |
itating | -0.061 |
same | -0.060 |
sharing | -0.060 |
en | -0.060 |
Token | ber |
Token ID | 527 |
Feature activation | -3.54e+00 |
Pos logit contributions | |
hours | +0.132 |
aster | +0.103 |
year | +0.102 |
takeoff | +0.101 |
trek | +0.099 |
Neg logit contributions | |
mill | -0.160 |
up | -0.159 |
fighting | -0.157 |
y | -0.156 |
ter | -0.156 |
Token | is |
Token ID | 318 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
year | +0.141 |
hours | +0.130 |
glass | +0.127 |
minutes | +0.122 |
aster | +0.121 |
Neg logit contributions | |
up | -0.051 |
fighting | -0.048 |
melon | -0.044 |
mill | -0.044 |
Sharing | -0.043 |
Token | someone |
Token ID | 2130 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
year | +0.066 |
glass | +0.063 |
ird | +0.063 |
hours | +0.061 |
ris | +0.060 |
Neg logit contributions | |
up | -0.035 |
fighting | -0.031 |
sharing | -0.027 |
melon | -0.026 |
being | -0.026 |
Token | who |
Token ID | 508 |
Feature activation | -2.66e+00 |
Pos logit contributions | |
year | +0.065 |
glass | +0.061 |
ird | +0.059 |
hours | +0.059 |
aster | +0.058 |
Neg logit contributions | |
up | -0.040 |
itating | -0.038 |
fighting | -0.037 |
arre | -0.037 |
ling | -0.037 |
Token | cuts |
Token ID | 6630 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
year | +0.088 |
glass | +0.084 |
hours | +0.082 |
ird | +0.082 |
minutes | +0.081 |
Neg logit contributions | |
up | -0.054 |
fighting | -0.051 |
arre | -0.051 |
ling | -0.048 |
itating | -0.048 |
Token | get |
Token ID | 651 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
year | +0.079 |
minutes | +0.070 |
glass | +0.069 |
ibr | +0.068 |
lan | +0.067 |
Neg logit contributions | |
up | -0.032 |
fighting | -0.025 |
bouncing | -0.023 |
scatter | -0.023 |
sharing | -0.023 |
Token | his |
Token ID | 465 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
glass | +0.043 |
ird | +0.043 |
year | +0.042 |
hours | +0.040 |
ris | +0.040 |
Neg logit contributions | |
up | -0.024 |
fighting | -0.017 |
melon | -0.016 |
ling | -0.016 |
itating | -0.015 |
Token | haircut |
Token ID | 45548 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
year | +0.061 |
glass | +0.053 |
ris | +0.048 |
hours | +0.047 |
ird | +0.047 |
Neg logit contributions | |
up | -0.047 |
fighting | -0.037 |
same | -0.037 |
bl | -0.035 |
hurting | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -5.96e-01 |
Pos logit contributions | |
year | +0.093 |
glass | +0.084 |
hours | +0.082 |
aster | +0.082 |
ibr | +0.081 |
Neg logit contributions | |
up | -0.033 |
itating | -0.024 |
ling | -0.023 |
bing | -0.022 |
melon | -0.021 |
Token | The |
Token ID | 383 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
year | +0.022 |
ird | +0.022 |
glass | +0.021 |
hours | +0.020 |
minutes | +0.020 |
Neg logit contributions | |
arre | -0.010 |
up | -0.010 |
itating | -0.010 |
melon | -0.009 |
fighting | -0.009 |
Token | bar |
Token ID | 2318 |
Feature activation | -5.66e+00 |
Pos logit contributions | |
year | +0.071 |
glass | +0.068 |
ris | +0.066 |
ird | +0.065 |
lan | +0.065 |
Neg logit contributions | |
up | -0.031 |
same | -0.025 |
fighting | -0.022 |
hurting | -0.022 |
ling | -0.022 |
Token | ber |
Token ID | 527 |
Feature activation | -2.88e+00 |
Pos logit contributions | |
Africa | +0.128 |
hours | +0.116 |
aster | +0.106 |
takeoff | +0.104 |
ocket | +0.104 |
Neg logit contributions | |
ling | -0.165 |
up | -0.161 |
mill | -0.155 |
y | -0.155 |
itating | -0.150 |
Token | put |
Token ID | 1234 |
Feature activation | -4.80e-01 |
Pos logit contributions | |
year | +0.116 |
Africa | +0.107 |
glass | +0.107 |
lan | +0.105 |
aster | +0.105 |
Neg logit contributions | |
up | -0.043 |
ling | -0.033 |
fighting | -0.033 |
arre | -0.033 |
melon | -0.030 |
Token | a |
Token ID | 257 |
Feature activation | -2.78e+00 |
Pos logit contributions | |
year | +0.018 |
ird | +0.018 |
glass | +0.017 |
ibr | +0.016 |
ris | +0.016 |
Neg logit contributions | |
up | -0.009 |
melon | -0.007 |
ling | -0.007 |
fighting | -0.007 |
Freddy | -0.007 |
Token | cape |
Token ID | 36023 |
Feature activation | -3.00e+00 |
Pos logit contributions | |
year | +0.105 |
glass | +0.097 |
ibr | +0.093 |
lan | +0.092 |
Africa | +0.091 |
Neg logit contributions | |
up | -0.053 |
mess | -0.045 |
bl | -0.044 |
ling | -0.043 |
itating | -0.041 |
Token | around |
Token ID | 1088 |
Feature activation | -7.06e-01 |
Pos logit contributions | |
year | +0.116 |
ird | +0.112 |
glass | +0.109 |
hours | +0.107 |
aster | +0.107 |
Neg logit contributions | |
up | -0.049 |
itating | -0.041 |
ling | -0.040 |
arre | -0.038 |
melon | -0.037 |
Token | quiet |
Token ID | 5897 |
Feature activation | -8.70e-01 |
Pos logit contributions | |
glass | +0.012 |
year | +0.012 |
ird | +0.012 |
ris | +0.011 |
aster | +0.011 |
Neg logit contributions | |
up | -0.011 |
arre | -0.010 |
fighting | -0.010 |
ling | -0.010 |
itating | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | -9.94e-01 |
Pos logit contributions | |
year | +0.034 |
glass | +0.033 |
ird | +0.033 |
aster | +0.033 |
ris | +0.031 |
Neg logit contributions | |
up | -0.014 |
itating | -0.011 |
ling | -0.011 |
fighting | -0.011 |
arre | -0.011 |
Token | polite |
Token ID | 23507 |
Feature activation | -3.45e-01 |
Pos logit contributions | |
year | +0.036 |
glass | +0.035 |
ird | +0.034 |
ris | +0.033 |
hours | +0.033 |
Neg logit contributions | |
up | -0.019 |
fighting | -0.017 |
itating | -0.016 |
arre | -0.015 |
hurting | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
year | +0.013 |
ird | +0.013 |
glass | +0.013 |
aster | +0.013 |
hours | +0.012 |
Neg logit contributions | |
up | -0.006 |
fighting | -0.005 |
ling | -0.005 |
itating | -0.005 |
arre | -0.004 |
Token | The |
Token ID | 383 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
year | +0.042 |
glass | +0.041 |
hours | +0.040 |
ird | +0.039 |
aster | +0.038 |
Neg logit contributions | |
itating | -0.019 |
arre | -0.019 |
Sharing | -0.018 |
fighting | -0.018 |
up | -0.018 |
Token | bar |
Token ID | 2318 |
Feature activation | -5.65e+00 |
Pos logit contributions | |
year | +0.044 |
glass | +0.043 |
ird | +0.041 |
ris | +0.040 |
ibr | +0.039 |
Neg logit contributions | |
up | -0.024 |
same | -0.020 |
fighting | -0.019 |
en | -0.018 |
itating | -0.018 |
Token | ber |
Token ID | 527 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
hours | +0.138 |
aster | +0.119 |
year | +0.113 |
Africa | +0.113 |
growth | +0.111 |
Neg logit contributions | |
up | -0.158 |
mill | -0.152 |
ling | -0.150 |
Sharing | -0.143 |
y | -0.143 |
Token | is |
Token ID | 318 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
year | +0.089 |
glass | +0.082 |
hours | +0.079 |
aster | +0.078 |
ird | +0.077 |
Neg logit contributions | |
up | -0.041 |
arre | -0.035 |
fighting | -0.035 |
ling | -0.034 |
melon | -0.034 |
Token | busy |
Token ID | 8179 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
glass | +0.005 |
year | +0.005 |
ird | +0.005 |
hours | +0.005 |
lock | +0.005 |
Neg logit contributions | |
up | -0.003 |
fighting | -0.002 |
itating | -0.002 |
arre | -0.002 |
melon | -0.002 |
Token | working |
Token ID | 1762 |
Feature activation | -6.28e-01 |
Pos logit contributions | |
up | +0.002 |
itating | +0.002 |
arre | +0.002 |
fighting | +0.002 |
ling | +0.002 |
Neg logit contributions | |
year | -0.004 |
glass | -0.004 |
ird | -0.004 |
hours | -0.004 |
aster | -0.004 |
Token | ." |
Token ID | 526 |
Feature activation | +5.05e-01 |
Pos logit contributions | |
year | +0.023 |
ird | +0.023 |
glass | +0.022 |
aster | +0.021 |
lock | +0.021 |
Neg logit contributions | |
up | -0.013 |
ling | -0.010 |
arre | -0.010 |
itating | -0.010 |
fighting | -0.009 |
Token | be |
Token ID | 307 |
Feature activation | +4.86e-01 |
Pos logit contributions | |
itating | +0.011 |
arre | +0.011 |
up | +0.010 |
melon | +0.010 |
ling | +0.010 |
Neg logit contributions | |
year | -0.018 |
ird | -0.017 |
glass | -0.017 |
hours | -0.016 |
aster | -0.016 |
Token | back |
Token ID | 736 |
Feature activation | -8.07e-01 |
Pos logit contributions | |
up | +0.011 |
arre | +0.010 |
itating | +0.010 |
ling | +0.010 |
melon | +0.009 |
Neg logit contributions | |
year | -0.015 |
glass | -0.015 |
ird | -0.015 |
hours | -0.014 |
aster | -0.014 |
Token | again |
Token ID | 757 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
year | +0.027 |
glass | +0.026 |
ird | +0.026 |
ibr | +0.024 |
aster | +0.024 |
Neg logit contributions | |
up | -0.018 |
itating | -0.015 |
ling | -0.014 |
sharing | -0.014 |
arre | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
year | +0.009 |
glass | +0.008 |
ird | +0.008 |
ibrarian | +0.008 |
ibr | +0.008 |
Neg logit contributions | |
up | -0.005 |
bing | -0.005 |
ling | -0.005 |
arre | -0.005 |
sharing | -0.005 |
Token | The |
Token ID | 383 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
glass | +0.048 |
year | +0.047 |
ird | +0.046 |
hours | +0.044 |
minutes | +0.044 |
Neg logit contributions | |
up | -0.019 |
arre | -0.019 |
itating | -0.019 |
Sharing | -0.018 |
Freddy | -0.017 |
Token | bar |
Token ID | 2318 |
Feature activation | -5.65e+00 |
Pos logit contributions | |
glass | +0.057 |
year | +0.054 |
ird | +0.053 |
lan | +0.051 |
adders | +0.051 |
Neg logit contributions | |
up | -0.024 |
same | -0.019 |
ling | -0.018 |
hurting | -0.018 |
fighting | -0.018 |
Token | ber |
Token ID | 527 |
Feature activation | -3.19e+00 |
Pos logit contributions | |
Africa | +0.123 |
hours | +0.112 |
glass | +0.102 |
aster | +0.102 |
takeoff | +0.098 |
Neg logit contributions | |
ling | -0.165 |
mill | -0.163 |
up | -0.160 |
itating | -0.152 |
fighting | -0.149 |
Token | smiled |
Token ID | 13541 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
year | +0.126 |
glass | +0.117 |
Africa | +0.115 |
lan | +0.112 |
aster | +0.112 |
Neg logit contributions | |
up | -0.048 |
ling | -0.039 |
arre | -0.039 |
melon | -0.037 |
mill | -0.036 |
Token | and |
Token ID | 290 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
arre | +0.020 |
itating | +0.019 |
melon | +0.018 |
up | +0.018 |
ling | +0.018 |
Neg logit contributions | |
year | -0.042 |
glass | -0.040 |
ird | -0.040 |
hours | -0.038 |
minutes | -0.037 |
Token | waved |
Token ID | 26834 |
Feature activation | -7.37e-01 |
Pos logit contributions | |
glass | +0.041 |
year | +0.041 |
ird | +0.039 |
aster | +0.038 |
kil | +0.038 |
Neg logit contributions | |
up | -0.023 |
Freddy | -0.020 |
melon | -0.019 |
arre | -0.019 |
itating | -0.019 |
Token | goodbye |
Token ID | 24829 |
Feature activation | -4.12e-01 |
Pos logit contributions | |
year | +0.019 |
ird | +0.018 |
glass | +0.018 |
aster | +0.017 |
ibr | +0.016 |
Neg logit contributions | |
up | -0.022 |
ling | -0.020 |
itating | -0.019 |
arre | -0.019 |
Freddy | -0.018 |
Token | car |
Token ID | 1097 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
year | +0.028 |
ird | +0.026 |
glass | +0.025 |
ris | +0.024 |
ibr | +0.024 |
Neg logit contributions | |
up | -0.014 |
same | -0.011 |
melon | -0.011 |
mess | -0.011 |
fighting | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -7.71e-01 |
Pos logit contributions | |
year | +0.080 |
hours | +0.074 |
Ah | +0.073 |
ibr | +0.072 |
lan | +0.071 |
Neg logit contributions | |
up | -0.021 |
itating | -0.013 |
fighting | -0.012 |
ling | -0.011 |
melon | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | -9.26e-01 |
Pos logit contributions | |
year | +0.033 |
glass | +0.032 |
ird | +0.032 |
hours | +0.031 |
minutes | +0.030 |
Neg logit contributions | |
up | -0.008 |
itating | -0.008 |
arre | -0.008 |
fighting | -0.007 |
melon | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | +3.91e-01 |
Pos logit contributions | |
year | +0.037 |
ird | +0.036 |
glass | +0.035 |
hours | +0.035 |
minutes | +0.034 |
Neg logit contributions | |
up | -0.012 |
itating | -0.012 |
arre | -0.011 |
fighting | -0.010 |
melon | -0.010 |
Token | The |
Token ID | 464 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
itating | +0.008 |
arre | +0.008 |
up | +0.008 |
fighting | +0.008 |
ling | +0.008 |
Neg logit contributions | |
year | -0.012 |
ird | -0.012 |
glass | -0.012 |
hours | -0.011 |
minutes | -0.011 |
Token | bar |
Token ID | 2318 |
Feature activation | -5.65e+00 |
Pos logit contributions | |
year | +0.046 |
ird | +0.044 |
glass | +0.043 |
ris | +0.042 |
adows | +0.041 |
Neg logit contributions | |
up | -0.021 |
same | -0.018 |
mess | -0.017 |
fighting | -0.017 |
hurting | -0.016 |
Token | ber |
Token ID | 527 |
Feature activation | -3.04e+00 |
Pos logit contributions | |
Africa | +0.128 |
hours | +0.127 |
outer | +0.115 |
ages | +0.114 |
ocket | +0.111 |
Neg logit contributions | |
ling | -0.151 |
mill | -0.145 |
up | -0.143 |
fighting | -0.140 |
y | -0.137 |
Token | was |
Token ID | 373 |
Feature activation | -6.00e-01 |
Pos logit contributions | |
year | +0.127 |
lan | +0.113 |
Africa | +0.111 |
aster | +0.110 |
kil | +0.110 |
Neg logit contributions | |
up | -0.045 |
fighting | -0.039 |
ling | -0.032 |
arre | -0.032 |
qu | -0.030 |
Token | a |
Token ID | 257 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
glass | +0.023 |
ird | +0.023 |
year | +0.023 |
hours | +0.022 |
ris | +0.022 |
Neg logit contributions | |
up | -0.010 |
fighting | -0.009 |
itating | -0.008 |
hurting | -0.008 |
melon | -0.007 |
Token | big |
Token ID | 1263 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
year | +0.081 |
glass | +0.077 |
adders | +0.075 |
ird | +0.074 |
ibr | +0.073 |
Neg logit contributions | |
mess | -0.029 |
fighting | -0.028 |
up | -0.026 |
sad | -0.026 |
frown | -0.025 |
Token | man |
Token ID | 582 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
year | +0.072 |
glass | +0.065 |
ibr | +0.063 |
ird | +0.062 |
lan | +0.060 |
Neg logit contributions | |
fighting | -0.029 |
up | -0.028 |
mess | -0.027 |
frown | -0.023 |
hurting | -0.023 |
Token | hair |
Token ID | 4190 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
year | +0.045 |
glass | +0.042 |
ird | +0.041 |
hours | +0.040 |
minutes | +0.039 |
Neg logit contributions | |
up | -0.020 |
arre | -0.016 |
ling | -0.015 |
bing | -0.015 |
itating | -0.015 |
Token | ?" |
Token ID | 1701 |
Feature activation | -7.59e-01 |
Pos logit contributions | |
year | +0.065 |
glass | +0.062 |
ird | +0.061 |
hours | +0.059 |
minutes | +0.057 |
Neg logit contributions | |
up | -0.032 |
itating | -0.031 |
arre | -0.031 |
melon | -0.029 |
ling | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | -9.72e-01 |
Pos logit contributions | |
year | +0.028 |
glass | +0.026 |
ird | +0.026 |
hours | +0.026 |
minutes | +0.025 |
Neg logit contributions | |
up | -0.013 |
itating | -0.012 |
arre | -0.012 |
melon | -0.012 |
fighting | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -3.88e-01 |
Pos logit contributions | |
year | +0.041 |
ird | +0.039 |
glass | +0.039 |
hours | +0.038 |
minutes | +0.038 |
Neg logit contributions | |
up | -0.011 |
itating | -0.010 |
arre | -0.010 |
rust | -0.009 |
fighting | -0.008 |
Token | The |
Token ID | 464 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
year | +0.015 |
ird | +0.014 |
hours | +0.014 |
glass | +0.014 |
minutes | +0.013 |
Neg logit contributions | |
itating | -0.006 |
arre | -0.006 |
up | -0.005 |
rust | -0.005 |
Sharing | -0.005 |
Token | bar |
Token ID | 2318 |
Feature activation | -5.65e+00 |
Pos logit contributions | |
glass | +0.051 |
year | +0.051 |
ird | +0.048 |
ris | +0.047 |
hours | +0.046 |
Neg logit contributions | |
up | -0.025 |
fighting | -0.020 |
itating | -0.020 |
en | -0.020 |
same | -0.020 |
Token | ber |
Token ID | 527 |
Feature activation | -2.96e+00 |
Pos logit contributions | |
hours | +0.117 |
Africa | +0.114 |
ages | +0.103 |
glass | +0.100 |
ird | +0.099 |
Neg logit contributions | |
ling | -0.175 |
mill | -0.169 |
up | -0.167 |
itating | -0.163 |
fighting | -0.158 |
Token | was |
Token ID | 373 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
year | +0.104 |
glass | +0.099 |
ird | +0.097 |
hours | +0.094 |
Africa | +0.093 |
Neg logit contributions | |
up | -0.054 |
arre | -0.052 |
ling | -0.050 |
melon | -0.050 |
itating | -0.049 |
Token | surprised |
Token ID | 6655 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
glass | +0.038 |
year | +0.038 |
ird | +0.038 |
hours | +0.036 |
lan | +0.035 |
Neg logit contributions | |
up | -0.022 |
itating | -0.020 |
fighting | -0.019 |
arre | -0.019 |
melon | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
arre | +0.027 |
itating | +0.026 |
melon | +0.026 |
flower | +0.024 |
inas | +0.024 |
Neg logit contributions | |
year | -0.031 |
minutes | -0.030 |
hours | -0.029 |
glass | -0.028 |
ris | -0.028 |
Token | pleased |
Token ID | 10607 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
glass | +0.034 |
ird | +0.034 |
year | +0.034 |
hours | +0.032 |
aster | +0.031 |
Neg logit contributions | |
up | -0.021 |
itating | -0.020 |
melon | -0.019 |
fighting | -0.019 |
arre | -0.019 |
Token | brave |
Token ID | 14802 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
up | +0.005 |
itating | +0.004 |
arre | +0.004 |
fighting | +0.004 |
ling | +0.004 |
Neg logit contributions | |
glass | -0.009 |
year | -0.008 |
ird | -0.008 |
hours | -0.008 |
ibr | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
arre | +0.002 |
itating | +0.002 |
up | +0.002 |
ling | +0.002 |
melon | +0.002 |
Neg logit contributions | |
year | -0.004 |
glass | -0.004 |
ird | -0.004 |
hours | -0.004 |
aster | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -4.42e-01 |
Pos logit contributions | |
year | +0.004 |
ird | +0.004 |
glass | +0.004 |
hours | +0.004 |
aster | +0.004 |
Neg logit contributions | |
itating | -0.003 |
arre | -0.003 |
ling | -0.003 |
up | -0.003 |
melon | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
year | +0.015 |
glass | +0.014 |
ird | +0.014 |
hours | +0.014 |
aster | +0.014 |
Neg logit contributions | |
itating | -0.008 |
arre | -0.008 |
up | -0.008 |
melon | -0.008 |
ling | -0.008 |
Token | The |
Token ID | 464 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
up | +0.031 |
melon | +0.029 |
flower | +0.029 |
itating | +0.029 |
sets | +0.029 |
Neg logit contributions | |
year | -0.033 |
aster | -0.031 |
ird | -0.031 |
glass | -0.031 |
hours | -0.029 |
Token | bar |
Token ID | 2318 |
Feature activation | -5.64e+00 |
Pos logit contributions | |
year | +0.042 |
glass | +0.041 |
ird | +0.039 |
ris | +0.039 |
adders | +0.038 |
Neg logit contributions | |
up | -0.018 |
hurting | -0.015 |
same | -0.015 |
fighting | -0.015 |
mess | -0.014 |
Token | ber |
Token ID | 527 |
Feature activation | -2.73e+00 |
Pos logit contributions | |
Africa | +0.132 |
hours | +0.122 |
ages | +0.113 |
takeoff | +0.110 |
outer | +0.107 |
Neg logit contributions | |
ling | -0.160 |
mill | -0.150 |
up | -0.145 |
y | -0.141 |
fighting | -0.135 |
Token | was |
Token ID | 373 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
year | +0.103 |
glass | +0.095 |
Africa | +0.092 |
kil | +0.092 |
lan | +0.091 |
Neg logit contributions | |
up | -0.046 |
arre | -0.041 |
ling | -0.040 |
fighting | -0.038 |
mill | -0.036 |
Token | gentle |
Token ID | 10296 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
year | +0.012 |
glass | +0.012 |
ird | +0.012 |
hours | +0.011 |
ibr | +0.011 |
Neg logit contributions | |
up | -0.007 |
itating | -0.006 |
arre | -0.006 |
fighting | -0.006 |
ling | -0.006 |
Token | with |
Token ID | 351 |
Feature activation | -9.02e-01 |
Pos logit contributions | |
itating | +0.019 |
arre | +0.019 |
up | +0.018 |
melon | +0.018 |
ling | +0.017 |
Neg logit contributions | |
year | -0.048 |
glass | -0.046 |
ird | -0.046 |
hours | -0.044 |
aster | -0.043 |
Token | his |
Token ID | 465 |
Feature activation | -2.36e+00 |
Pos logit contributions | |
year | +0.033 |
glass | +0.032 |
ird | +0.032 |
lock | +0.030 |
lan | +0.030 |
Neg logit contributions | |
up | -0.016 |
fighting | -0.014 |
arre | -0.014 |
Freddy | -0.014 |
ling | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
year | +0.059 |
ird | +0.057 |
glass | +0.056 |
ibr | +0.054 |
kil | +0.053 |
Neg logit contributions | |
up | -0.019 |
fighting | -0.013 |
itating | -0.013 |
melon | -0.012 |
bing | -0.011 |
Token | bar |
Token ID | 2318 |
Feature activation | -6.03e+00 |
Pos logit contributions | |
year | +0.071 |
glass | +0.065 |
ocket | +0.061 |
ird | +0.061 |
lan | +0.061 |
Neg logit contributions | |
up | -0.035 |
same | -0.029 |
itating | -0.026 |
fighting | -0.026 |
me | -0.024 |
Token | ber |
Token ID | 527 |
Feature activation | -3.53e+00 |
Pos logit contributions | |
hours | +0.131 |
year | +0.128 |
ocket | +0.126 |
Africa | +0.121 |
glass | +0.119 |
Neg logit contributions | |
up | -0.171 |
itating | -0.153 |
by | -0.151 |
ling | -0.148 |
fighting | -0.147 |
Token | , |
Token ID | 11 |
Feature activation | -5.60e-01 |
Pos logit contributions | |
year | +0.164 |
lan | +0.144 |
hours | +0.142 |
minutes | +0.141 |
glass | +0.140 |
Neg logit contributions | |
up | -0.050 |
's | -0.033 |
my | -0.032 |
fighting | -0.027 |
ning | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
year | +0.021 |
ird | +0.021 |
glass | +0.020 |
ris | +0.019 |
minutes | +0.019 |
Neg logit contributions | |
up | -0.009 |
fighting | -0.008 |
Freddy | -0.008 |
melon | -0.008 |
itating | -0.008 |
Token | bar |
Token ID | 2318 |
Feature activation | -5.63e+00 |
Pos logit contributions | |
year | +0.085 |
glass | +0.079 |
ird | +0.077 |
ris | +0.077 |
ocket | +0.076 |
Neg logit contributions | |
up | -0.033 |
same | -0.029 |
fighting | -0.025 |
hurting | -0.024 |
frown | -0.023 |
Token | ber |
Token ID | 527 |
Feature activation | -3.09e+00 |
Pos logit contributions | |
Africa | +0.133 |
hours | +0.125 |
takeoff | +0.114 |
ocket | +0.114 |
growth | +0.113 |
Neg logit contributions | |
up | -0.147 |
ling | -0.145 |
y | -0.143 |
mill | -0.141 |
ing | -0.138 |
Token | said |
Token ID | 531 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
year | +0.128 |
glass | +0.110 |
lan | +0.110 |
aster | +0.109 |
Africa | +0.109 |
Neg logit contributions | |
up | -0.048 |
fighting | -0.038 |
ling | -0.035 |
arre | -0.034 |
my | -0.034 |
Token | that |
Token ID | 326 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
year | +0.066 |
ris | +0.063 |
ird | +0.063 |
glass | +0.063 |
hours | +0.060 |
Neg logit contributions | |
up | -0.030 |
melon | -0.029 |
Freddy | -0.029 |
fighting | -0.027 |
sharing | -0.026 |
Token | he |
Token ID | 339 |
Feature activation | -9.45e-01 |
Pos logit contributions | |
year | +0.058 |
glass | +0.055 |
ris | +0.053 |
ird | +0.053 |
minutes | +0.052 |
Neg logit contributions | |
up | -0.027 |
fighting | -0.025 |
sharing | -0.024 |
melon | -0.023 |
Freddy | -0.022 |
Token | was |
Token ID | 373 |
Feature activation | -5.14e-01 |
Pos logit contributions | |
year | +0.036 |
glass | +0.034 |
aster | +0.034 |
hours | +0.033 |
minutes | +0.033 |
Neg logit contributions | |
up | -0.014 |
arre | -0.013 |
fighting | -0.013 |
ling | -0.012 |
melon | -0.012 |
Token | off |
Token ID | 572 |
Feature activation | -8.86e-01 |
Pos logit contributions | |
year | +0.075 |
ird | +0.075 |
glass | +0.072 |
ibr | +0.068 |
kil | +0.068 |
Neg logit contributions | |
up | -0.033 |
arre | -0.024 |
ling | -0.024 |
itating | -0.023 |
melon | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -6.96e-01 |
Pos logit contributions | |
year | +0.030 |
ird | +0.029 |
glass | +0.029 |
hours | +0.027 |
aster | +0.027 |
Neg logit contributions | |
up | -0.017 |
itating | -0.017 |
arre | -0.016 |
melon | -0.016 |
fighting | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
year | +0.023 |
glass | +0.023 |
ird | +0.023 |
hours | +0.021 |
aster | +0.021 |
Neg logit contributions | |
arre | -0.013 |
itating | -0.013 |
up | -0.013 |
melon | -0.013 |
ling | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | +4.55e-01 |
Pos logit contributions | |
year | +0.040 |
ird | +0.039 |
glass | +0.038 |
hours | +0.036 |
minutes | +0.036 |
Neg logit contributions | |
itating | -0.022 |
arre | -0.022 |
up | -0.022 |
melon | -0.021 |
ling | -0.021 |
Token | The |
Token ID | 464 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
itating | +0.011 |
up | +0.011 |
arre | +0.011 |
melon | +0.010 |
ling | +0.010 |
Neg logit contributions | |
year | -0.014 |
glass | -0.013 |
ird | -0.013 |
hours | -0.012 |
aster | -0.012 |
Token | bar |
Token ID | 2318 |
Feature activation | -5.61e+00 |
Pos logit contributions | |
glass | +0.060 |
year | +0.059 |
ird | +0.058 |
adders | +0.054 |
lan | +0.054 |
Neg logit contributions | |
up | -0.026 |
ling | -0.021 |
same | -0.020 |
fighting | -0.020 |
hurting | -0.020 |
Token | ber |
Token ID | 527 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
Africa | +0.127 |
hours | +0.111 |
aster | +0.102 |
ocket | +0.102 |
glass | +0.101 |
Neg logit contributions | |
ling | -0.159 |
mill | -0.157 |
up | -0.150 |
fighting | -0.146 |
itating | -0.143 |
Token | smiled |
Token ID | 13541 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
year | +0.087 |
glass | +0.080 |
kil | +0.078 |
Africa | +0.076 |
lan | +0.076 |
Neg logit contributions | |
up | -0.037 |
arre | -0.031 |
fighting | -0.031 |
ling | -0.030 |
flower | -0.028 |
Token | as |
Token ID | 355 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
arre | +0.020 |
itating | +0.019 |
melon | +0.019 |
fighting | +0.017 |
rust | +0.017 |
Neg logit contributions | |
year | -0.038 |
glass | -0.036 |
ird | -0.036 |
hours | -0.035 |
minutes | -0.034 |
Token | he |
Token ID | 339 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
year | +0.057 |
glass | +0.056 |
ird | +0.056 |
hours | +0.051 |
minutes | +0.051 |
Neg logit contributions | |
up | -0.030 |
itating | -0.028 |
melon | -0.027 |
fighting | -0.027 |
Freddy | -0.027 |
Token | watched |
Token ID | 7342 |
Feature activation | -5.58e-01 |
Pos logit contributions | |
year | +0.053 |
glass | +0.051 |
kil | +0.047 |
Africa | +0.047 |
aster | +0.047 |
Neg logit contributions | |
arre | -0.027 |
up | -0.026 |
ling | -0.025 |
flower | -0.024 |
fighting | -0.024 |
Token | in |
Token ID | 287 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
year | +0.002 |
ird | +0.001 |
glass | +0.001 |
hours | +0.001 |
aster | +0.001 |
Neg logit contributions | |
itating | -0.001 |
up | -0.001 |
arre | -0.001 |
ling | -0.001 |
melon | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
year | +0.073 |
ird | +0.072 |
ibr | +0.071 |
glass | +0.070 |
lan | +0.066 |
Neg logit contributions | |
up | -0.023 |
fighting | -0.018 |
melon | -0.015 |
itating | -0.014 |
bing | -0.014 |
Token | chair |
Token ID | 5118 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
year | +0.060 |
glass | +0.059 |
ird | +0.055 |
ris | +0.052 |
ibr | +0.051 |
Neg logit contributions | |
up | -0.066 |
ling | -0.056 |
itating | -0.054 |
same | -0.053 |
that | -0.052 |
Token | and |
Token ID | 290 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
year | +0.066 |
ird | +0.063 |
glass | +0.062 |
ibr | +0.060 |
aster | +0.060 |
Neg logit contributions | |
up | -0.015 |
itating | -0.011 |
arre | -0.011 |
ling | -0.011 |
fighting | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
year | +0.037 |
glass | +0.035 |
ird | +0.034 |
ibr | +0.032 |
ris | +0.032 |
Neg logit contributions | |
up | -0.021 |
fighting | -0.018 |
itating | -0.017 |
ling | -0.017 |
arre | -0.017 |
Token | bar |
Token ID | 2318 |
Feature activation | -5.59e+00 |
Pos logit contributions | |
year | +0.074 |
glass | +0.073 |
ird | +0.068 |
ris | +0.068 |
lan | +0.067 |
Neg logit contributions | |
up | -0.036 |
same | -0.028 |
hurting | -0.027 |
ling | -0.027 |
itating | -0.027 |
Token | ber |
Token ID | 527 |
Feature activation | -2.63e+00 |
Pos logit contributions | |
Africa | +0.128 |
hours | +0.119 |
ages | +0.108 |
aster | +0.101 |
glass | +0.101 |
Neg logit contributions | |
ling | -0.166 |
up | -0.163 |
itating | -0.151 |
fighting | -0.151 |
mill | -0.151 |
Token | started |
Token ID | 2067 |
Feature activation | -4.45e-01 |
Pos logit contributions | |
year | +0.106 |
Africa | +0.095 |
glass | +0.092 |
aster | +0.091 |
prise | +0.090 |
Neg logit contributions | |
up | -0.045 |
ling | -0.035 |
fighting | -0.033 |
bl | -0.033 |
bing | -0.030 |
Token | to |
Token ID | 284 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
year | +0.015 |
ird | +0.014 |
glass | +0.014 |
hours | +0.013 |
aster | +0.013 |
Neg logit contributions | |
up | -0.009 |
itating | -0.009 |
arre | -0.009 |
melon | -0.008 |
fighting | -0.008 |
Token | cut |
Token ID | 2005 |
Feature activation | -3.59e-01 |
Pos logit contributions | |
year | +0.047 |
glass | +0.044 |
ird | +0.042 |
hours | +0.040 |
minutes | +0.040 |
Neg logit contributions | |
up | -0.029 |
itating | -0.028 |
arre | -0.027 |
melon | -0.027 |
fighting | -0.026 |
Token | his |
Token ID | 465 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
year | +0.014 |
glass | +0.013 |
ird | +0.013 |
Bible | +0.012 |
hours | +0.012 |
Neg logit contributions | |
up | -0.007 |
ling | -0.005 |
arre | -0.005 |
bing | -0.005 |
itating | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | -8.98e-01 |
Pos logit contributions | |
year | +0.068 |
ird | +0.066 |
glass | +0.065 |
ris | +0.062 |
hours | +0.061 |
Neg logit contributions | |
up | -0.032 |
fighting | -0.027 |
melon | -0.027 |
bing | -0.025 |
ling | -0.025 |
Token | bar |
Token ID | 2318 |
Feature activation | -5.39e+00 |
Pos logit contributions | |
year | +0.032 |
ird | +0.031 |
glass | +0.030 |
ris | +0.029 |
hours | +0.028 |
Neg logit contributions | |
up | -0.018 |
ling | -0.016 |
itating | -0.015 |
fighting | -0.015 |
arre | -0.015 |
Token | ber |
Token ID | 527 |
Feature activation | -2.26e+00 |
Pos logit contributions | |
hours | +0.119 |
ages | +0.105 |
Africa | +0.100 |
ird | +0.098 |
glass | +0.095 |
Neg logit contributions | |
up | -0.176 |
ling | -0.174 |
itating | -0.165 |
bing | -0.155 |
ing | -0.154 |
Token | . |
Token ID | 13 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
year | +0.086 |
hours | +0.079 |
glass | +0.079 |
ird | +0.079 |
minutes | +0.076 |
Neg logit contributions | |
up | -0.043 |
fighting | -0.035 |
ling | -0.034 |
itating | -0.033 |
bing | -0.032 |
Token | The |
Token ID | 383 |
Feature activation | -9.28e-01 |
Pos logit contributions | |
year | +0.057 |
ird | +0.056 |
glass | +0.056 |
hours | +0.052 |
minutes | +0.051 |
Neg logit contributions | |
up | -0.026 |
arre | -0.025 |
melon | -0.025 |
itating | -0.024 |
fighting | -0.023 |
Token | bar |
Token ID | 2318 |
Feature activation | -5.59e+00 |
Pos logit contributions | |
year | +0.036 |
glass | +0.034 |
ird | +0.033 |
ris | +0.032 |
ibr | +0.031 |
Neg logit contributions | |
up | -0.017 |
hurting | -0.013 |
fighting | -0.013 |
same | -0.013 |
ling | -0.013 |
Token | ber |
Token ID | 527 |
Feature activation | -2.71e+00 |
Pos logit contributions | |
Africa | +0.122 |
hours | +0.116 |
ages | +0.106 |
ird | +0.101 |
year | +0.100 |
Neg logit contributions | |
ling | -0.161 |
mill | -0.155 |
up | -0.153 |
y | -0.148 |
fighting | -0.146 |
Token | smiled |
Token ID | 13541 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
year | +0.106 |
glass | +0.094 |
ird | +0.093 |
Africa | +0.093 |
lan | +0.091 |
Neg logit contributions | |
up | -0.044 |
fighting | -0.039 |
ling | -0.037 |
melon | -0.035 |
mill | -0.035 |
Token | and |
Token ID | 290 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
arre | +0.021 |
itating | +0.021 |
melon | +0.020 |
ling | +0.019 |
up | +0.019 |
Neg logit contributions | |
year | -0.041 |
glass | -0.039 |
ird | -0.039 |
hours | -0.037 |
minutes | -0.037 |
Token | talked |
Token ID | 6619 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
year | +0.045 |
glass | +0.043 |
ird | +0.042 |
aster | +0.040 |
ibr | +0.039 |
Neg logit contributions | |
up | -0.023 |
melon | -0.022 |
fighting | -0.021 |
Freddy | -0.021 |
arre | -0.020 |
Token | to |
Token ID | 284 |
Feature activation | -5.68e-01 |
Pos logit contributions | |
ird | +0.053 |
year | +0.051 |
glass | +0.050 |
aster | +0.049 |
ocket | +0.047 |
Neg logit contributions | |
up | -0.023 |
ling | -0.017 |
fighting | -0.017 |
arre | -0.016 |
itating | -0.016 |
Token | help |
Token ID | 1037 |
Feature activation | -4.15e-01 |
Pos logit contributions | |
up | +0.003 |
itating | +0.003 |
arre | +0.003 |
melon | +0.002 |
fighting | +0.002 |
Neg logit contributions | |
year | -0.003 |
glass | -0.003 |
ird | -0.003 |
hours | -0.002 |
minutes | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -2.61e-01 |
Pos logit contributions | |
year | +0.016 |
glass | +0.015 |
ird | +0.015 |
hours | +0.014 |
minutes | +0.014 |
Neg logit contributions | |
up | -0.007 |
itating | -0.006 |
fighting | -0.006 |
ling | -0.006 |
bing | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
year | +0.009 |
glass | +0.009 |
ird | +0.009 |
hours | +0.008 |
aster | +0.008 |
Neg logit contributions | |
itating | -0.005 |
arre | -0.005 |
up | -0.005 |
melon | -0.005 |
ling | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | +4.28e-01 |
Pos logit contributions | |
year | +0.040 |
ird | +0.039 |
glass | +0.038 |
hours | +0.037 |
minutes | +0.036 |
Neg logit contributions | |
itating | -0.016 |
up | -0.015 |
arre | -0.015 |
ling | -0.014 |
fighting | -0.013 |
Token | The |
Token ID | 464 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
itating | +0.008 |
arre | +0.008 |
up | +0.008 |
ling | +0.008 |
melon | +0.008 |
Neg logit contributions | |
year | -0.014 |
ird | -0.014 |
glass | -0.014 |
hours | -0.013 |
aster | -0.013 |
Token | bar |
Token ID | 2318 |
Feature activation | -5.56e+00 |
Pos logit contributions | |
ird | +0.079 |
glass | +0.079 |
year | +0.078 |
ocket | +0.074 |
lan | +0.072 |
Neg logit contributions | |
up | -0.032 |
fighting | -0.026 |
ling | -0.025 |
en | -0.023 |
itating | -0.022 |
Token | ber |
Token ID | 527 |
Feature activation | -2.43e+00 |
Pos logit contributions | |
hours | +0.128 |
Africa | +0.119 |
ages | +0.113 |
ird | +0.107 |
aster | +0.106 |
Neg logit contributions | |
ling | -0.165 |
up | -0.155 |
mill | -0.154 |
itating | -0.150 |
y | -0.147 |
Token | said |
Token ID | 531 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
year | +0.089 |
glass | +0.084 |
ird | +0.081 |
kil | +0.081 |
hours | +0.080 |
Neg logit contributions | |
up | -0.041 |
arre | -0.040 |
fighting | -0.038 |
ling | -0.038 |
melon | -0.037 |
Token | to |
Token ID | 284 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
year | +0.044 |
ird | +0.043 |
glass | +0.042 |
ris | +0.041 |
hours | +0.041 |
Neg logit contributions | |
up | -0.015 |
melon | -0.014 |
fighting | -0.013 |
Freddy | -0.013 |
ling | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
year | +0.048 |
ird | +0.047 |
glass | +0.046 |
hours | +0.044 |
minutes | +0.043 |
Neg logit contributions | |
up | -0.022 |
itating | -0.021 |
arre | -0.021 |
melon | -0.021 |
fighting | -0.020 |
Token | little |
Token ID | 1310 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
year | +0.079 |
glass | +0.078 |
ird | +0.076 |
lan | +0.073 |
aster | +0.072 |
Neg logit contributions | |
up | -0.031 |
fighting | -0.024 |
ling | -0.023 |
itating | -0.023 |
same | -0.022 |
Token | happy |
Token ID | 3772 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
year | +0.027 |
glass | +0.026 |
ird | +0.026 |
hours | +0.025 |
aster | +0.024 |
Neg logit contributions | |
itating | -0.019 |
arre | -0.019 |
up | -0.019 |
melon | -0.018 |
ling | -0.018 |
Token | ! |
Token ID | 0 |
Feature activation | -7.72e-01 |
Pos logit contributions | |
arre | +0.046 |
itating | +0.046 |
melon | +0.043 |
rust | +0.042 |
ling | +0.042 |
Neg logit contributions | |
year | -0.047 |
glass | -0.046 |
hours | -0.045 |
minutes | -0.044 |
ird | -0.044 |
Token |
|
Token ID | 198 |
Feature activation | -3.62e-01 |
Pos logit contributions | |
year | +0.026 |
glass | +0.025 |
ird | +0.025 |
hours | +0.024 |
aster | +0.023 |
Neg logit contributions | |
itating | -0.015 |
arre | -0.015 |
up | -0.014 |
melon | -0.014 |
ling | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | +3.09e-01 |
Pos logit contributions | |
year | +0.013 |
ird | +0.013 |
glass | +0.012 |
hours | +0.012 |
minutes | +0.012 |
Neg logit contributions | |
itating | -0.006 |
arre | -0.006 |
up | -0.006 |
ling | -0.006 |
melon | -0.006 |
Token | The |
Token ID | 464 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
arre | +0.007 |
itating | +0.007 |
up | +0.006 |
melon | +0.006 |
ling | +0.006 |
Neg logit contributions | |
year | -0.010 |
glass | -0.009 |
ird | -0.009 |
hours | -0.009 |
aster | -0.009 |
Token | bar |
Token ID | 2318 |
Feature activation | -5.52e+00 |
Pos logit contributions | |
ird | +0.072 |
year | +0.071 |
glass | +0.070 |
ris | +0.065 |
hours | +0.064 |
Neg logit contributions | |
up | -0.037 |
fighting | -0.030 |
ling | -0.030 |
itating | -0.029 |
en | -0.028 |
Token | ber |
Token ID | 527 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
hours | +0.120 |
Africa | +0.119 |
ages | +0.105 |
aster | +0.104 |
ird | +0.100 |
Neg logit contributions | |
ling | -0.166 |
mill | -0.159 |
up | -0.156 |
itating | -0.151 |
fighting | -0.150 |
Token | worked |
Token ID | 3111 |
Feature activation | +4.04e-01 |
Pos logit contributions | |
year | +0.086 |
glass | +0.080 |
ird | +0.077 |
kil | +0.077 |
Africa | +0.076 |
Neg logit contributions | |
up | -0.038 |
arre | -0.036 |
fighting | -0.034 |
melon | -0.034 |
ling | -0.033 |
Token | very |
Token ID | 845 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
up | +0.009 |
itating | +0.009 |
arre | +0.009 |
ling | +0.008 |
melon | +0.008 |
Neg logit contributions | |
year | -0.013 |
ird | -0.012 |
glass | -0.012 |
aster | -0.011 |
hours | -0.011 |
Token | carefully |
Token ID | 7773 |
Feature activation | -4.49e-01 |
Pos logit contributions | |
year | +0.054 |
ird | +0.054 |
glass | +0.053 |
hours | +0.049 |
aster | +0.049 |
Neg logit contributions | |
up | -0.037 |
itating | -0.037 |
arre | -0.036 |
ling | -0.035 |
melon | -0.035 |
Token | and |
Token ID | 290 |
Feature activation | -7.37e-01 |
Pos logit contributions | |
year | +0.018 |
ird | +0.017 |
glass | +0.017 |
hours | +0.016 |
aster | +0.016 |
Neg logit contributions | |
up | -0.007 |
arre | -0.006 |
fighting | -0.006 |
melon | -0.006 |
itating | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
year | +0.014 |
glass | +0.014 |
ird | +0.014 |
hours | +0.013 |
aster | +0.013 |
Neg logit contributions | |
itating | -0.008 |
arre | -0.008 |
up | -0.007 |
melon | -0.007 |
ling | -0.007 |
Token | " |
Token ID | 366 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
year | +0.004 |
glass | +0.004 |
ird | +0.004 |
hours | +0.004 |
minutes | +0.004 |
Neg logit contributions | |
itating | -0.002 |
arre | -0.002 |
up | -0.002 |
ling | -0.002 |
melon | -0.002 |
Token | No |
Token ID | 2949 |
Feature activation | -4.74e-01 |
Pos logit contributions | |
itating | +0.022 |
arre | +0.022 |
melon | +0.022 |
up | +0.022 |
ling | +0.021 |
Neg logit contributions | |
year | -0.032 |
glass | -0.030 |
ird | -0.030 |
hours | -0.029 |
aster | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | -3.66e-02 |
Pos logit contributions | |
ird | +0.013 |
hours | +0.013 |
minutes | +0.013 |
glass | +0.012 |
aster | +0.012 |
Neg logit contributions | |
itating | -0.012 |
arre | -0.011 |
melon | -0.011 |
ling | -0.011 |
flower | -0.011 |
Token | mud |
Token ID | 17492 |
Feature activation | +2.27e+00 |
Pos logit contributions | |
ird | +0.001 |
year | +0.001 |
glass | +0.001 |
hours | +0.001 |
minutes | +0.001 |
Neg logit contributions | |
itating | -0.001 |
arre | -0.001 |
ling | -0.001 |
bing | -0.001 |
up | -0.001 |
Token | isn |
Token ID | 2125 |
Feature activation | +3.57e+00 |
Pos logit contributions | |
itating | +0.063 |
arre | +0.054 |
rust | +0.054 |
ling | +0.050 |
Being | +0.049 |
Neg logit contributions | |
ird | -0.058 |
glass | -0.056 |
minutes | -0.056 |
hours | -0.055 |
aster | -0.053 |
Token | 't |
Token ID | 470 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
lings | +0.103 |
rust | +0.102 |
liness | +0.098 |
arre | +0.098 |
ling | +0.098 |
Neg logit contributions | |
ird | -0.076 |
year | -0.074 |
aster | -0.073 |
ibrarian | -0.064 |
lock | -0.063 |
Token | for |
Token ID | 329 |
Feature activation | -6.83e-01 |
Pos logit contributions | |
itating | +0.038 |
arre | +0.035 |
ishly | +0.034 |
rust | +0.034 |
flower | +0.033 |
Neg logit contributions | |
minutes | -0.035 |
ird | -0.034 |
hours | -0.034 |
year | -0.033 |
aster | -0.032 |
Token | sharing |
Token ID | 7373 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
year | +0.025 |
glass | +0.024 |
ird | +0.024 |
hours | +0.022 |
aster | +0.022 |
Neg logit contributions | |
arre | -0.011 |
itating | -0.011 |
up | -0.011 |
melon | -0.011 |
ling | -0.010 |
Token | ." |
Token ID | 526 |
Feature activation | +7.28e-01 |
Pos logit contributions | |
itating | +0.030 |
arre | +0.030 |
melon | +0.028 |
ling | +0.028 |
up | +0.027 |
Neg logit contributions | |
year | -0.040 |
ird | -0.039 |
glass | -0.039 |
hours | -0.037 |
minutes | -0.036 |
Token |
|
Token ID | 198 |
Feature activation | -3.64e-01 |
Pos logit contributions | |
itating | +0.017 |
ling | +0.016 |
flower | +0.016 |
bing | +0.016 |
mill | +0.016 |
Neg logit contributions | |
year | -0.021 |
glass | -0.021 |
ird | -0.021 |
hours | -0.020 |
minutes | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
melon | +0.011 |
ling | +0.011 |
itating | +0.010 |
flower | +0.010 |
arre | +0.010 |
Neg logit contributions | |
ird | -0.012 |
year | -0.011 |
glass | -0.011 |
ibr | -0.010 |
aster | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
itating | +0.003 |
arre | +0.003 |
melon | +0.003 |
ling | +0.003 |
up | +0.003 |
Neg logit contributions | |
year | -0.004 |
ird | -0.004 |
glass | -0.004 |
hours | -0.003 |
aster | -0.003 |
Token | Tim |
Token ID | 14967 |
Feature activation | +5.60e-01 |
Pos logit contributions | |
itating | +0.047 |
melon | +0.045 |
up | +0.045 |
arre | +0.044 |
flower | +0.044 |
Neg logit contributions | |
year | -0.048 |
ird | -0.045 |
glass | -0.045 |
hours | -0.044 |
aster | -0.043 |
Token | my |
Token ID | 1820 |
Feature activation | +4.49e-01 |
Pos logit contributions | |
melon | +0.010 |
itating | +0.010 |
bing | +0.009 |
fighting | +0.009 |
izing | +0.009 |
Neg logit contributions | |
year | -0.019 |
ird | -0.019 |
ris | -0.018 |
glass | -0.018 |
hours | -0.017 |
Token | loved |
Token ID | 6151 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
melon | +0.013 |
itating | +0.013 |
rooms | +0.012 |
inas | +0.012 |
bing | +0.012 |
Neg logit contributions | |
ird | -0.010 |
year | -0.010 |
glass | -0.010 |
hours | -0.009 |
ibrarian | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | +3.67e+00 |
Pos logit contributions | |
itating | +0.038 |
ling | +0.035 |
flower | +0.034 |
arre | +0.033 |
lings | +0.032 |
Neg logit contributions | |
hours | -0.036 |
aster | -0.035 |
year | -0.034 |
glass | -0.034 |
minutes | -0.031 |
Token | idea |
Token ID | 2126 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
flies | +0.117 |
ric | +0.115 |
bing | +0.114 |
lings | +0.109 |
inated | +0.107 |
Neg logit contributions | |
hours | -0.063 |
book | -0.060 |
year | -0.059 |
month | -0.059 |
hour | -0.056 |
Token | and |
Token ID | 290 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
itating | +0.027 |
arre | +0.027 |
melon | +0.026 |
rust | +0.025 |
ling | +0.025 |
Neg logit contributions | |
year | -0.038 |
ird | -0.037 |
glass | -0.036 |
hours | -0.034 |
minutes | -0.034 |
Token | he |
Token ID | 339 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
itating | +0.031 |
arre | +0.029 |
ling | +0.028 |
bing | +0.028 |
melon | +0.027 |
Neg logit contributions | |
year | -0.032 |
ird | -0.030 |
glass | -0.030 |
hours | -0.029 |
minutes | -0.028 |
Token | and |
Token ID | 290 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
itating | +0.005 |
arre | +0.005 |
bing | +0.005 |
melon | +0.005 |
ling | +0.005 |
Neg logit contributions | |
year | -0.006 |
ird | -0.006 |
glass | -0.005 |
hours | -0.005 |
minutes | -0.005 |
Token | his |
Token ID | 465 |
Feature activation | +8.82e-01 |
Pos logit contributions | |
itating | +0.030 |
arre | +0.029 |
ling | +0.027 |
melon | +0.027 |
up | +0.027 |
Neg logit contributions | |
year | -0.032 |
glass | -0.031 |
ird | -0.030 |
hours | -0.029 |
aster | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | +3.74e-01 |
Pos logit contributions | |
up | +0.003 |
melon | +0.003 |
fighting | +0.003 |
ling | +0.003 |
itating | +0.003 |
Neg logit contributions | |
year | -0.008 |
glass | -0.007 |
ird | -0.007 |
aster | -0.007 |
ibr | -0.007 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
ling | +0.008 |
itating | +0.008 |
arre | +0.008 |
mill | +0.008 |
bing | +0.008 |
Neg logit contributions | |
year | -0.011 |
ird | -0.011 |
glass | -0.010 |
aster | -0.010 |
hours | -0.010 |
Token | Tom |
Token ID | 13787 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
itating | +0.024 |
arre | +0.023 |
up | +0.022 |
melon | +0.022 |
ling | +0.022 |
Neg logit contributions | |
year | -0.036 |
ird | -0.034 |
glass | -0.034 |
hours | -0.032 |
aster | -0.032 |
Token | my |
Token ID | 1820 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
itating | +0.013 |
arre | +0.012 |
melon | +0.011 |
ling | +0.011 |
up | +0.011 |
Neg logit contributions | |
year | -0.045 |
ird | -0.043 |
glass | -0.043 |
hours | -0.042 |
aster | -0.041 |
Token | loved |
Token ID | 6151 |
Feature activation | +6.79e-01 |
Pos logit contributions | |
itating | +0.043 |
arre | +0.040 |
melon | +0.039 |
ling | +0.038 |
bing | +0.037 |
Neg logit contributions | |
year | -0.051 |
ird | -0.050 |
glass | -0.050 |
hours | -0.048 |
aster | -0.047 |
Token | baths |
Token ID | 49076 |
Feature activation | +3.69e+00 |
Pos logit contributions | |
itating | +0.017 |
arre | +0.016 |
ling | +0.015 |
up | +0.015 |
melon | +0.015 |
Neg logit contributions | |
year | -0.020 |
glass | -0.019 |
ird | -0.019 |
hours | -0.018 |
aster | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
arre | +0.115 |
itating | +0.112 |
ucker | +0.111 |
rust | +0.110 |
ety | +0.105 |
Neg logit contributions | |
year | -0.072 |
ird | -0.065 |
hours | -0.063 |
glass | -0.059 |
book | -0.057 |
Token | Every |
Token ID | 3887 |
Feature activation | +3.01e-01 |
Pos logit contributions | |
itating | +0.024 |
arre | +0.024 |
up | +0.023 |
ling | +0.023 |
melon | +0.022 |
Neg logit contributions | |
year | -0.040 |
ird | -0.038 |
glass | -0.038 |
hours | -0.036 |
aster | -0.036 |
Token | day |
Token ID | 1110 |
Feature activation | -6.15e-02 |
Pos logit contributions | |
itating | +0.009 |
arre | +0.009 |
melon | +0.008 |
bing | +0.008 |
rust | +0.008 |
Neg logit contributions | |
year | -0.007 |
ird | -0.007 |
hours | -0.007 |
glass | -0.007 |
aster | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | +5.05e-02 |
Pos logit contributions | |
year | +0.002 |
glass | +0.002 |
ibr | +0.002 |
ibrarian | +0.002 |
ird | +0.002 |
Neg logit contributions | |
up | -0.001 |
melon | -0.001 |
fighting | -0.001 |
Freddy | -0.001 |
sharing | -0.001 |
Token | he |
Token ID | 339 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
up | +0.001 |
melon | +0.001 |
arre | +0.001 |
fighting | +0.001 |
itating | +0.001 |
Neg logit contributions | |
year | -0.002 |
glass | -0.002 |
ird | -0.002 |
ibr | -0.002 |
hours | -0.002 |
Token | that |
Token ID | 326 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
arre | +0.002 |
itating | +0.002 |
melon | +0.002 |
ling | +0.002 |
up | +0.001 |
Neg logit contributions | |
year | -0.003 |
glass | -0.003 |
ird | -0.003 |
hours | -0.003 |
aster | -0.003 |
Token | even |
Token ID | 772 |
Feature activation | -5.43e-01 |
Pos logit contributions | |
itating | +0.035 |
arre | +0.033 |
ling | +0.032 |
melon | +0.031 |
ucker | +0.030 |
Neg logit contributions | |
year | -0.043 |
glass | -0.041 |
hours | -0.041 |
ird | -0.040 |
aster | -0.039 |
Token | though |
Token ID | 996 |
Feature activation | +6.59e-01 |
Pos logit contributions | |
year | +0.017 |
ird | +0.016 |
glass | +0.016 |
hours | +0.015 |
aster | +0.015 |
Neg logit contributions | |
itating | -0.012 |
arre | -0.012 |
up | -0.011 |
melon | -0.011 |
ling | -0.011 |
Token | he |
Token ID | 339 |
Feature activation | +3.20e-01 |
Pos logit contributions | |
itating | +0.016 |
arre | +0.016 |
ling | +0.014 |
melon | +0.014 |
up | +0.014 |
Neg logit contributions | |
year | -0.019 |
ird | -0.018 |
glass | -0.018 |
hours | -0.018 |
aster | -0.017 |
Token | wanted |
Token ID | 2227 |
Feature activation | +3.73e-01 |
Pos logit contributions | |
itating | +0.006 |
arre | +0.006 |
melon | +0.006 |
up | +0.006 |
ling | +0.006 |
Neg logit contributions | |
year | -0.011 |
ird | -0.010 |
glass | -0.010 |
hours | -0.010 |
aster | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | +3.76e+00 |
Pos logit contributions | |
itating | +0.009 |
arre | +0.009 |
ling | +0.009 |
melon | +0.008 |
bing | +0.008 |
Neg logit contributions | |
year | -0.010 |
glass | -0.010 |
ird | -0.010 |
hours | -0.010 |
aster | -0.009 |
Token | mint |
Token ID | 21043 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
itating | +0.107 |
arre | +0.105 |
ocked | +0.101 |
inas | +0.100 |
ning | +0.099 |
Neg logit contributions | |
hours | -0.080 |
ird | -0.077 |
minutes | -0.071 |
year | -0.070 |
allowance | -0.070 |
Token | , |
Token ID | 11 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
itating | +0.042 |
arre | +0.040 |
inas | +0.035 |
rust | +0.035 |
bing | +0.035 |
Neg logit contributions | |
ird | -0.048 |
year | -0.048 |
hours | -0.045 |
lock | -0.043 |
glass | -0.043 |
Token | it |
Token ID | 340 |
Feature activation | +2.18e+00 |
Pos logit contributions | |
itating | +0.031 |
arre | +0.030 |
ling | +0.028 |
melon | +0.028 |
up | +0.028 |
Neg logit contributions | |
year | -0.033 |
glass | -0.031 |
ird | -0.031 |
hours | -0.030 |
aster | -0.029 |
Token | would |
Token ID | 561 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
itating | +0.054 |
bing | +0.049 |
arre | +0.049 |
melon | +0.049 |
ety | +0.048 |
Neg logit contributions | |
year | -0.054 |
ird | -0.053 |
hours | -0.051 |
glass | -0.050 |
kil | -0.049 |
Token | have |
Token ID | 423 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
arre | +0.045 |
itating | +0.042 |
ling | +0.042 |
bing | +0.041 |
ning | +0.040 |
Neg logit contributions | |
aster | -0.046 |
ird | -0.046 |
year | -0.045 |
hours | -0.044 |
glass | -0.044 |
Token | . |
Token ID | 13 |
Feature activation | +3.12e-01 |
Pos logit contributions | |
arre | +0.000 |
itating | +0.000 |
melon | +0.000 |
rust | +0.000 |
ling | +0.000 |
Neg logit contributions | |
year | -0.001 |
ird | -0.000 |
glass | -0.000 |
hours | -0.000 |
aster | -0.000 |
Token | " |
Token ID | 366 |
Feature activation | +4.32e-01 |
Pos logit contributions | |
rust | +0.007 |
ling | +0.007 |
arre | +0.007 |
itating | +0.007 |
melon | +0.007 |
Neg logit contributions | |
year | -0.009 |
glass | -0.008 |
ird | -0.008 |
minutes | -0.008 |
aster | -0.008 |
Token | You |
Token ID | 1639 |
Feature activation | -7.72e-01 |
Pos logit contributions | |
melon | +0.008 |
up | +0.008 |
itating | +0.008 |
arre | +0.008 |
ling | +0.007 |
Neg logit contributions | |
year | -0.015 |
glass | -0.014 |
aster | -0.014 |
ird | -0.014 |
hours | -0.013 |
Token | will |
Token ID | 481 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
year | +0.026 |
glass | +0.024 |
ird | +0.024 |
hours | +0.023 |
aster | +0.023 |
Neg logit contributions | |
arre | -0.015 |
itating | -0.015 |
up | -0.015 |
melon | -0.014 |
ling | -0.014 |
Token | hit |
Token ID | 2277 |
Feature activation | +1.94e+00 |
Pos logit contributions | |
arre | +0.026 |
ling | +0.026 |
itating | +0.026 |
melon | +0.024 |
bing | +0.024 |
Neg logit contributions | |
ird | -0.038 |
glass | -0.036 |
year | -0.036 |
aster | -0.036 |
hours | -0.035 |
Token | the |
Token ID | 262 |
Feature activation | +3.86e+00 |
Pos logit contributions | |
itating | +0.039 |
arre | +0.038 |
melon | +0.035 |
inas | +0.034 |
ling | +0.034 |
Neg logit contributions | |
year | -0.061 |
glass | -0.060 |
hours | -0.060 |
ird | -0.060 |
aster | -0.060 |
Token | fence |
Token ID | 13990 |
Feature activation | +2.73e-01 |
Pos logit contributions | |
itating | +0.071 |
inas | +0.070 |
ocked | +0.070 |
ucker | +0.069 |
arre | +0.069 |
Neg logit contributions | |
glass | -0.108 |
year | -0.107 |
slot | -0.106 |
hours | -0.105 |
lock | -0.104 |
Token | !" |
Token ID | 2474 |
Feature activation | +3.83e-01 |
Pos logit contributions | |
arre | +0.005 |
itating | +0.005 |
melon | +0.005 |
ling | +0.005 |
rust | +0.005 |
Neg logit contributions | |
year | -0.009 |
ird | -0.009 |
glass | -0.009 |
hours | -0.008 |
aster | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -4.07e-01 |
Pos logit contributions | |
arre | +0.008 |
itating | +0.008 |
melon | +0.008 |
ling | +0.008 |
up | +0.008 |
Neg logit contributions | |
year | -0.012 |
glass | -0.012 |
ird | -0.012 |
hours | -0.011 |
aster | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | +7.04e-01 |
Pos logit contributions | |
year | +0.014 |
ird | +0.014 |
glass | +0.014 |
hours | +0.013 |
aster | +0.013 |
Neg logit contributions | |
itating | -0.007 |
arre | -0.007 |
up | -0.007 |
melon | -0.007 |
ling | -0.007 |
Token | But |
Token ID | 1537 |
Feature activation | +9.57e-01 |
Pos logit contributions | |
itating | +0.017 |
arre | +0.017 |
melon | +0.017 |
up | +0.016 |
ling | +0.016 |
Neg logit contributions | |
year | -0.020 |
glass | -0.020 |
ird | -0.019 |
aster | -0.019 |
hours | -0.018 |
Token | day |
Token ID | 1110 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
year | +0.060 |
ird | +0.058 |
glass | +0.058 |
hours | +0.053 |
ibr | +0.053 |
Neg logit contributions | |
up | -0.041 |
arre | -0.040 |
ling | -0.039 |
itating | -0.039 |
melon | -0.039 |
Token | , |
Token ID | 11 |
Feature activation | +2.78e-01 |
Pos logit contributions | |
year | +0.005 |
glass | +0.005 |
ird | +0.005 |
hours | +0.005 |
minutes | +0.004 |
Neg logit contributions | |
itating | -0.003 |
arre | -0.003 |
up | -0.003 |
melon | -0.003 |
ling | -0.003 |
Token | they |
Token ID | 484 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
arre | +0.006 |
itating | +0.006 |
ling | +0.005 |
melon | +0.005 |
up | +0.005 |
Neg logit contributions | |
year | -0.009 |
glass | -0.009 |
ird | -0.008 |
hours | -0.008 |
minutes | -0.008 |
Token | decided |
Token ID | 3066 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
itating | +0.040 |
arre | +0.039 |
bing | +0.038 |
up | +0.038 |
melon | +0.038 |
Neg logit contributions | |
year | -0.046 |
ird | -0.045 |
glass | -0.044 |
hours | -0.042 |
minutes | -0.039 |
Token | to |
Token ID | 284 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
arre | +0.045 |
itating | +0.044 |
flower | +0.043 |
ling | +0.043 |
ucker | +0.043 |
Neg logit contributions | |
glass | -0.024 |
hours | -0.022 |
year | -0.021 |
minutes | -0.020 |
lan | -0.020 |
Token | play |
Token ID | 711 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
itating | +0.035 |
ling | +0.034 |
arre | +0.034 |
bing | +0.032 |
melon | +0.032 |
Neg logit contributions | |
ird | -0.039 |
glass | -0.037 |
year | -0.036 |
aster | -0.035 |
hours | -0.035 |
Token | a |
Token ID | 257 |
Feature activation | -2.44e-03 |
Pos logit contributions | |
up | +0.028 |
arre | +0.028 |
itating | +0.027 |
fighting | +0.026 |
ling | +0.026 |
Neg logit contributions | |
year | -0.042 |
ird | -0.040 |
glass | -0.039 |
aster | -0.037 |
hours | -0.036 |
Token | game |
Token ID | 983 |
Feature activation | +8.29e-01 |
Pos logit contributions | |
year | +0.000 |
ird | +0.000 |
glass | +0.000 |
hours | +0.000 |
aster | +0.000 |
Neg logit contributions | |
itating | -0.000 |
arre | -0.000 |
up | -0.000 |
melon | -0.000 |
ling | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
up | +0.017 |
itating | +0.016 |
arre | +0.016 |
melon | +0.015 |
fighting | +0.015 |
Neg logit contributions | |
year | -0.028 |
ird | -0.026 |
glass | -0.026 |
hours | -0.024 |
aster | -0.024 |
Token | Joe |
Token ID | 5689 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
ling | +0.025 |
arre | +0.024 |
itating | +0.024 |
bing | +0.023 |
melon | +0.023 |
Neg logit contributions | |
year | -0.031 |
glass | -0.030 |
ird | -0.029 |
hours | -0.029 |
minutes | -0.028 |
Token | wanted |
Token ID | 2227 |
Feature activation | -9.89e-01 |
Pos logit contributions | |
itating | +0.028 |
arre | +0.027 |
up | +0.026 |
melon | +0.026 |
ling | +0.026 |
Neg logit contributions | |
year | -0.041 |
ird | -0.040 |
glass | -0.039 |
hours | -0.037 |
aster | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
up | +0.014 |
itating | +0.013 |
arre | +0.013 |
ling | +0.013 |
fighting | +0.013 |
Neg logit contributions | |
year | -0.029 |
glass | -0.027 |
ird | -0.027 |
aster | -0.026 |
hours | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
itating | +0.025 |
arre | +0.025 |
ling | +0.024 |
up | +0.024 |
melon | +0.023 |
Neg logit contributions | |
year | -0.043 |
ird | -0.042 |
glass | -0.041 |
hours | -0.039 |
aster | -0.039 |
Token |
|
Token ID | 198 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
itating | +0.003 |
arre | +0.003 |
up | +0.003 |
melon | +0.003 |
ling | +0.003 |
Neg logit contributions | |
year | -0.005 |
ird | -0.004 |
glass | -0.004 |
hours | -0.004 |
aster | -0.004 |
Token | When |
Token ID | 2215 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
melon | +0.035 |
itating | +0.034 |
arre | +0.034 |
up | +0.034 |
ling | +0.033 |
Neg logit contributions | |
year | -0.044 |
ird | -0.041 |
glass | -0.041 |
hours | -0.039 |
aster | -0.039 |
Token | they |
Token ID | 484 |
Feature activation | -7.49e-01 |
Pos logit contributions | |
year | +0.006 |
glass | +0.006 |
ird | +0.006 |
kil | +0.006 |
ibr | +0.006 |
Neg logit contributions | |
up | -0.002 |
fighting | -0.002 |
melon | -0.002 |
arre | -0.002 |
Freddy | -0.002 |
Token | arrived |
Token ID | 5284 |
Feature activation | +7.39e-01 |
Pos logit contributions | |
glass | +0.026 |
year | +0.026 |
kil | +0.025 |
lan | +0.025 |
Bible | +0.024 |
Neg logit contributions | |
up | -0.013 |
arre | -0.012 |
melon | -0.012 |
fighting | -0.012 |
itating | -0.012 |
Token | at |
Token ID | 379 |
Feature activation | +5.53e-02 |
Pos logit contributions | |
itating | +0.015 |
arre | +0.015 |
up | +0.014 |
melon | +0.014 |
ling | +0.014 |
Neg logit contributions | |
year | -0.024 |
ird | -0.023 |
glass | -0.023 |
hours | -0.022 |
aster | -0.022 |
Token | grandma |
Token ID | 49890 |
Feature activation | +3.54e-01 |
Pos logit contributions | |
up | +0.001 |
fighting | +0.001 |
melon | +0.001 |
Freddy | +0.001 |
arre | +0.001 |
Neg logit contributions | |
year | -0.002 |
glass | -0.002 |
ird | -0.002 |
kil | -0.002 |
lock | -0.002 |
Token | 's |
Token ID | 338 |
Feature activation | -5.45e-01 |
Pos logit contributions | |
itating | +0.004 |
arre | +0.004 |
up | +0.004 |
melon | +0.004 |
ling | +0.004 |
Neg logit contributions | |
year | -0.015 |
glass | -0.014 |
ird | -0.014 |
hours | -0.014 |
aster | -0.013 |
Token | house |
Token ID | 2156 |
Feature activation | -5.37e-01 |
Pos logit contributions | |
year | +0.019 |
ird | +0.017 |
glass | +0.017 |
hours | +0.016 |
aster | +0.016 |
Neg logit contributions | |
up | -0.011 |
arre | -0.011 |
itating | -0.011 |
fighting | -0.010 |
melon | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | +8.03e-01 |
Pos logit contributions | |
year | +0.021 |
glass | +0.020 |
ird | +0.020 |
hours | +0.019 |
kil | +0.019 |
Neg logit contributions | |
up | -0.008 |
itating | -0.007 |
fighting | -0.007 |
arre | -0.007 |
melon | -0.007 |
Token | They |
Token ID | 1119 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
itating | +0.009 |
arre | +0.009 |
up | +0.008 |
melon | +0.008 |
ling | +0.008 |
Neg logit contributions | |
year | -0.025 |
glass | -0.024 |
ird | -0.024 |
hours | -0.023 |
aster | -0.023 |
Token | are |
Token ID | 389 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
itating | +0.003 |
arre | +0.003 |
up | +0.003 |
melon | +0.003 |
ling | +0.003 |
Neg logit contributions | |
year | -0.004 |
ird | -0.004 |
glass | -0.004 |
hours | -0.004 |
aster | -0.004 |
Token | wet |
Token ID | 9583 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
itating | +0.028 |
arre | +0.027 |
melon | +0.026 |
up | +0.026 |
ling | +0.025 |
Neg logit contributions | |
year | -0.049 |
ird | -0.046 |
glass | -0.046 |
hours | -0.044 |
minutes | -0.043 |
Token | and |
Token ID | 290 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
itating | +0.040 |
arre | +0.038 |
melon | +0.037 |
rust | +0.036 |
ucker | +0.034 |
Neg logit contributions | |
year | -0.064 |
glass | -0.062 |
ird | -0.061 |
hours | -0.060 |
minutes | -0.060 |
Token | dirty |
Token ID | 11841 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
itating | +0.030 |
arre | +0.029 |
up | +0.027 |
melon | +0.027 |
ling | +0.027 |
Neg logit contributions | |
year | -0.057 |
ird | -0.054 |
glass | -0.054 |
hours | -0.052 |
minutes | -0.051 |
Token | . |
Token ID | 13 |
Feature activation | +4.12e-01 |
Pos logit contributions | |
itating | +0.039 |
arre | +0.038 |
melon | +0.036 |
rust | +0.035 |
up | +0.035 |
Neg logit contributions | |
year | -0.053 |
ird | -0.051 |
glass | -0.051 |
hours | -0.048 |
minutes | -0.048 |
Token | They |
Token ID | 1119 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
arre | +0.006 |
itating | +0.006 |
up | +0.005 |
melon | +0.005 |
ling | +0.005 |
Neg logit contributions | |
year | -0.016 |
glass | -0.016 |
ird | -0.016 |
hours | -0.015 |
aster | -0.015 |
Token | are |
Token ID | 389 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
arre | +0.005 |
up | +0.005 |
itating | +0.005 |
melon | +0.004 |
fighting | +0.004 |
Neg logit contributions | |
year | -0.008 |
glass | -0.007 |
ird | -0.007 |
hours | -0.007 |
minutes | -0.007 |
Token | sorry |
Token ID | 7926 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
itating | +0.029 |
arre | +0.028 |
up | +0.027 |
melon | +0.026 |
ling | +0.026 |
Neg logit contributions | |
year | -0.055 |
ird | -0.053 |
glass | -0.053 |
hours | -0.051 |
minutes | -0.050 |
Token | . |
Token ID | 13 |
Feature activation | +7.40e-01 |
Pos logit contributions | |
itating | +0.055 |
arre | +0.054 |
melon | +0.052 |
rust | +0.052 |
bing | +0.050 |
Neg logit contributions | |
year | -0.051 |
minutes | -0.050 |
ird | -0.049 |
glass | -0.048 |
hours | -0.048 |
Token | They |
Token ID | 1119 |
Feature activation | +5.19e-01 |
Pos logit contributions | |
itating | +0.011 |
arre | +0.011 |
up | +0.011 |
melon | +0.011 |
ling | +0.010 |
Neg logit contributions | |
year | -0.028 |
glass | -0.027 |
ird | -0.027 |
hours | -0.026 |
aster | -0.025 |
Token | follow |
Token ID | 1061 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
year | +0.026 |
glass | +0.024 |
ird | +0.024 |
hours | +0.023 |
minutes | +0.022 |
Neg logit contributions | |
up | -0.016 |
itating | -0.016 |
arre | -0.016 |
melon | -0.015 |
fighting | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -4.80e-01 |
Pos logit contributions | |
year | +0.043 |
ird | +0.042 |
glass | +0.041 |
ibr | +0.038 |
hours | +0.038 |
Neg logit contributions | |
up | -0.018 |
ling | -0.015 |
itating | -0.014 |
fighting | -0.014 |
bing | -0.014 |
Token | As |
Token ID | 1081 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
year | +0.016 |
glass | +0.016 |
ird | +0.016 |
hours | +0.015 |
aster | +0.015 |
Neg logit contributions | |
arre | -0.009 |
itating | -0.009 |
up | -0.009 |
melon | -0.009 |
ling | -0.008 |
Token | they |
Token ID | 484 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
year | +0.047 |
ird | +0.047 |
glass | +0.045 |
ibr | +0.044 |
ibrarian | +0.043 |
Neg logit contributions | |
up | -0.017 |
Freddy | -0.016 |
fighting | -0.015 |
accidents | -0.013 |
sharing | -0.012 |
Token | got |
Token ID | 1392 |
Feature activation | -6.29e-01 |
Pos logit contributions | |
glass | +0.054 |
lan | +0.053 |
aster | +0.052 |
Bible | +0.052 |
Africa | +0.051 |
Neg logit contributions | |
ling | -0.026 |
up | -0.026 |
arre | -0.026 |
fighting | -0.024 |
qu | -0.024 |
Token | closer |
Token ID | 5699 |
Feature activation | +6.52e-01 |
Pos logit contributions | |
ird | +0.020 |
year | +0.020 |
glass | +0.020 |
aster | +0.019 |
hours | +0.018 |
Neg logit contributions | |
up | -0.014 |
itating | -0.013 |
arre | -0.012 |
melon | -0.012 |
fighting | -0.012 |
Token | , |
Token ID | 11 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
arre | +0.013 |
itating | +0.013 |
melon | +0.012 |
up | +0.012 |
ling | +0.012 |
Neg logit contributions | |
year | -0.021 |
glass | -0.020 |
ird | -0.020 |
hours | -0.020 |
minutes | -0.019 |
Token | Jack |
Token ID | 3619 |
Feature activation | -2.67e-01 |
Pos logit contributions | |
year | +0.008 |
ird | +0.008 |
glass | +0.008 |
hours | +0.008 |
ibr | +0.008 |
Neg logit contributions | |
up | -0.004 |
itating | -0.004 |
arre | -0.004 |
melon | -0.003 |
fighting | -0.003 |
Token | could |
Token ID | 714 |
Feature activation | -8.83e-01 |
Pos logit contributions | |
year | +0.009 |
ird | +0.009 |
glass | +0.009 |
aster | +0.008 |
hours | +0.008 |
Neg logit contributions | |
up | -0.005 |
arre | -0.005 |
itating | -0.005 |
ling | -0.005 |
melon | -0.005 |
Token | see |
Token ID | 766 |
Feature activation | -4.22e-01 |
Pos logit contributions | |
year | +0.022 |
glass | +0.021 |
ird | +0.021 |
hours | +0.019 |
aster | +0.019 |
Neg logit contributions | |
arre | -0.025 |
itating | -0.024 |
melon | -0.024 |
up | -0.024 |
ling | -0.023 |
Token | his |
Token ID | 465 |
Feature activation | -9.44e-01 |
Pos logit contributions | |
year | +0.015 |
ird | +0.015 |
glass | +0.014 |
aster | +0.013 |
lock | +0.013 |
Neg logit contributions | |
up | -0.008 |
itating | -0.007 |
ling | -0.007 |
arre | -0.007 |
fighting | -0.007 |
Token | She |
Token ID | 1375 |
Feature activation | +7.68e-01 |
Pos logit contributions | |
ling | +0.017 |
itating | +0.016 |
arre | +0.015 |
bing | +0.015 |
mill | +0.014 |
Neg logit contributions | |
year | -0.027 |
ird | -0.026 |
hours | -0.025 |
glass | -0.025 |
aster | -0.025 |
Token | touched |
Token ID | 12615 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
up | +0.015 |
arre | +0.014 |
itating | +0.014 |
ling | +0.014 |
melon | +0.014 |
Neg logit contributions | |
year | -0.026 |
ird | -0.025 |
glass | -0.025 |
aster | -0.023 |
ibr | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
itating | +0.030 |
arre | +0.030 |
melon | +0.027 |
rust | +0.027 |
fighting | +0.026 |
Neg logit contributions | |
year | -0.036 |
hours | -0.034 |
ird | -0.033 |
aster | -0.033 |
minutes | -0.032 |
Token | icy |
Token ID | 30284 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
arre | +0.030 |
itating | +0.029 |
melon | +0.026 |
rust | +0.026 |
up | +0.026 |
Neg logit contributions | |
year | -0.048 |
glass | -0.047 |
ird | -0.045 |
hours | -0.045 |
minutes | -0.043 |
Token | water |
Token ID | 1660 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
itating | +0.032 |
arre | +0.030 |
rust | +0.027 |
fighting | +0.026 |
inas | +0.026 |
Neg logit contributions | |
glass | -0.034 |
hours | -0.033 |
minutes | -0.033 |
year | -0.032 |
aster | -0.032 |
Token | with |
Token ID | 351 |
Feature activation | +6.24e-01 |
Pos logit contributions | |
arre | +0.050 |
rust | +0.050 |
itating | +0.047 |
ety | +0.043 |
inas | +0.042 |
Neg logit contributions | |
year | -0.054 |
glass | -0.054 |
minutes | -0.052 |
ird | -0.052 |
aster | -0.051 |
Token | her |
Token ID | 607 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
itating | +0.013 |
arre | +0.013 |
melon | +0.012 |
up | +0.012 |
ling | +0.012 |
Neg logit contributions | |
year | -0.020 |
ird | -0.019 |
glass | -0.019 |
hours | -0.019 |
aster | -0.018 |
Token | feet |
Token ID | 3625 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
itating | +0.004 |
arre | +0.004 |
up | +0.004 |
melon | +0.003 |
ling | +0.003 |
Neg logit contributions | |
year | -0.004 |
glass | -0.004 |
ird | -0.004 |
hours | -0.004 |
aster | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
rust | +0.026 |
arre | +0.025 |
itating | +0.023 |
Freddy | +0.023 |
melon | +0.022 |
Neg logit contributions | |
ird | -0.028 |
year | -0.028 |
glass | -0.027 |
lock | -0.026 |
hours | -0.025 |
Token | making |
Token ID | 1642 |
Feature activation | +6.32e-01 |
Pos logit contributions | |
arre | +0.039 |
itating | +0.039 |
rust | +0.037 |
melon | +0.034 |
rooms | +0.033 |
Neg logit contributions | |
hours | -0.049 |
minutes | -0.048 |
year | -0.046 |
aster | -0.045 |
glass | -0.045 |
Token | silly |
Token ID | 14397 |
Feature activation | +2.48e+00 |
Pos logit contributions | |
itating | +0.015 |
arre | +0.014 |
melon | +0.014 |
rust | +0.013 |
flower | +0.013 |
Neg logit contributions | |
year | -0.019 |
hours | -0.018 |
glass | -0.018 |
ird | -0.017 |
aster | -0.017 |
Token | help |
Token ID | 1037 |
Feature activation | +9.69e-02 |
Pos logit contributions | |
up | +0.003 |
itating | +0.003 |
arre | +0.003 |
melon | +0.003 |
fighting | +0.002 |
Neg logit contributions | |
year | -0.004 |
glass | -0.003 |
ird | -0.003 |
hours | -0.003 |
minutes | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +7.45e-01 |
Pos logit contributions | |
up | +0.002 |
itating | +0.002 |
arre | +0.002 |
melon | +0.001 |
ling | +0.001 |
Neg logit contributions | |
year | -0.004 |
glass | -0.003 |
ird | -0.003 |
hours | -0.003 |
minutes | -0.003 |
Token | She |
Token ID | 1375 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
itating | +0.016 |
arre | +0.015 |
ling | +0.015 |
melon | +0.014 |
up | +0.014 |
Neg logit contributions | |
year | -0.024 |
ird | -0.023 |
glass | -0.023 |
hours | -0.022 |
minutes | -0.021 |
Token | told |
Token ID | 1297 |
Feature activation | -4.82e-01 |
Pos logit contributions | |
arre | +0.003 |
up | +0.003 |
itating | +0.002 |
melon | +0.002 |
ling | +0.002 |
Neg logit contributions | |
year | -0.004 |
ird | -0.004 |
glass | -0.004 |
aster | -0.004 |
hours | -0.004 |
Token | her |
Token ID | 607 |
Feature activation | -7.08e-01 |
Pos logit contributions | |
year | +0.016 |
glass | +0.016 |
ird | +0.016 |
hours | +0.015 |
aster | +0.015 |
Neg logit contributions | |
itating | -0.009 |
arre | -0.009 |
ling | -0.009 |
melon | -0.009 |
up | -0.008 |
Token | friends |
Token ID | 2460 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
year | +0.020 |
ird | +0.019 |
glass | +0.019 |
hours | +0.018 |
minutes | +0.017 |
Neg logit contributions | |
arre | -0.018 |
up | -0.018 |
itating | -0.017 |
melon | -0.017 |
ling | -0.017 |
Token | about |
Token ID | 546 |
Feature activation | +6.89e-01 |
Pos logit contributions | |
itating | +0.005 |
arre | +0.005 |
bing | +0.005 |
flower | +0.005 |
rust | +0.005 |
Neg logit contributions | |
year | -0.005 |
ird | -0.005 |
glass | -0.005 |
hours | -0.005 |
aster | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
arre | +0.015 |
itating | +0.014 |
ling | +0.013 |
melon | +0.013 |
up | +0.013 |
Neg logit contributions | |
year | -0.022 |
glass | -0.021 |
ird | -0.021 |
hours | -0.021 |
aster | -0.020 |
Token | lost |
Token ID | 2626 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
itating | +0.031 |
arre | +0.030 |
melon | +0.029 |
ling | +0.028 |
bing | +0.028 |
Neg logit contributions | |
glass | -0.023 |
year | -0.023 |
ird | -0.022 |
hours | -0.021 |
aster | -0.021 |
Token | dog |
Token ID | 3290 |
Feature activation | +6.67e-01 |
Pos logit contributions | |
year | +0.004 |
ird | +0.004 |
glass | +0.004 |
hours | +0.004 |
minutes | +0.004 |
Neg logit contributions | |
arre | -0.003 |
up | -0.003 |
itating | -0.003 |
melon | -0.003 |
ling | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
itating | +0.012 |
arre | +0.012 |
melon | +0.010 |
rust | +0.010 |
ling | +0.010 |
Neg logit contributions | |
year | -0.023 |
ird | -0.023 |
glass | -0.023 |
hours | -0.022 |
aster | -0.021 |
Token | football |
Token ID | 4346 |
Feature activation | +6.36e-02 |
Pos logit contributions | |
bing | +0.068 |
itating | +0.065 |
inas | +0.065 |
ucker | +0.064 |
arre | +0.063 |
Neg logit contributions | |
glass | -0.050 |
passport | -0.049 |
ird | -0.048 |
hours | -0.047 |
year | -0.047 |
Token | and |
Token ID | 290 |
Feature activation | +3.18e-01 |
Pos logit contributions | |
itating | +0.001 |
arre | +0.001 |
up | +0.001 |
melon | +0.001 |
ling | +0.001 |
Neg logit contributions | |
year | -0.002 |
glass | -0.002 |
ird | -0.002 |
hours | -0.002 |
aster | -0.002 |
Token | Tim |
Token ID | 5045 |
Feature activation | -5.85e-01 |
Pos logit contributions | |
arre | +0.006 |
itating | +0.006 |
up | +0.006 |
melon | +0.006 |
ling | +0.006 |
Neg logit contributions | |
year | -0.011 |
glass | -0.010 |
ird | -0.010 |
hours | -0.010 |
aster | -0.010 |
Token | my |
Token ID | 1820 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
year | +0.027 |
glass | +0.026 |
ird | +0.026 |
hours | +0.025 |
aster | +0.025 |
Neg logit contributions | |
itating | -0.004 |
arre | -0.004 |
up | -0.004 |
melon | -0.004 |
ling | -0.004 |
Token | played |
Token ID | 2826 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
itating | +0.003 |
arre | +0.003 |
up | +0.003 |
melon | +0.003 |
ling | +0.003 |
Neg logit contributions | |
year | -0.004 |
ird | -0.004 |
glass | -0.004 |
hours | -0.004 |
aster | -0.004 |
Token | with |
Token ID | 351 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
itating | +0.027 |
arre | +0.026 |
melon | +0.025 |
ling | +0.024 |
flower | +0.023 |
Neg logit contributions | |
year | -0.048 |
glass | -0.046 |
ird | -0.045 |
hours | -0.044 |
aster | -0.043 |
Token | the |
Token ID | 262 |
Feature activation | +2.45e+00 |
Pos logit contributions | |
up | +0.002 |
arre | +0.002 |
itating | +0.002 |
melon | +0.002 |
ling | +0.002 |
Neg logit contributions | |
year | -0.005 |
ird | -0.004 |
glass | -0.004 |
hours | -0.004 |
aster | -0.004 |
Token | other |
Token ID | 584 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
bing | +0.068 |
ucker | +0.066 |
itating | +0.066 |
inas | +0.064 |
flower | +0.062 |
Neg logit contributions | |
hours | -0.057 |
year | -0.055 |
glass | -0.054 |
minutes | -0.053 |
ird | -0.050 |
Token | kids |
Token ID | 3988 |
Feature activation | +4.55e-01 |
Pos logit contributions | |
itating | +0.027 |
melon | +0.026 |
arre | +0.025 |
bing | +0.025 |
ucker | +0.025 |
Neg logit contributions | |
year | -0.028 |
ird | -0.028 |
hours | -0.028 |
minutes | -0.027 |
glass | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
itating | +0.008 |
arre | +0.008 |
up | +0.008 |
melon | +0.008 |
ling | +0.008 |
Neg logit contributions | |
year | -0.016 |
glass | -0.015 |
ird | -0.015 |
hours | -0.014 |
aster | -0.014 |
Token | |
Token ID | 220 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
itating | +0.033 |
ling | +0.033 |
arre | +0.032 |
flower | +0.031 |
mill | +0.031 |
Neg logit contributions | |
year | -0.028 |
glass | -0.026 |
hours | -0.026 |
ird | -0.025 |
Grand | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
itating | +0.008 |
arre | +0.008 |
up | +0.008 |
melon | +0.008 |
ling | +0.008 |
Neg logit contributions | |
year | -0.018 |
glass | -0.017 |
ird | -0.017 |
hours | -0.016 |
aster | -0.016 |
Token | You |
Token ID | 921 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
year | +0.066 |
glass | +0.065 |
hours | +0.064 |
minutes | +0.062 |
aster | +0.062 |
Neg logit contributions | |
Sharing | -0.027 |
up | -0.024 |
itating | -0.024 |
arre | -0.024 |
Bugs | -0.024 |
Token | must |
Token ID | 1276 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
year | +0.048 |
glass | +0.046 |
ird | +0.044 |
hours | +0.043 |
aster | +0.043 |
Neg logit contributions | |
arre | -0.031 |
itating | -0.030 |
up | -0.030 |
melon | -0.028 |
fighting | -0.028 |
Token | take |
Token ID | 1011 |
Feature activation | -7.89e-01 |
Pos logit contributions | |
itating | +0.026 |
arre | +0.025 |
ling | +0.024 |
melon | +0.024 |
bing | +0.024 |
Neg logit contributions | |
ird | -0.029 |
year | -0.027 |
glass | -0.027 |
aster | -0.027 |
ris | -0.026 |
Token | good |
Token ID | 922 |
Feature activation | -4.63e-01 |
Pos logit contributions | |
year | +0.028 |
glass | +0.027 |
ird | +0.026 |
ibr | +0.024 |
aster | +0.024 |
Neg logit contributions | |
up | -0.015 |
arre | -0.014 |
itating | -0.013 |
fighting | -0.013 |
ling | -0.013 |
Token | care |
Token ID | 1337 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
year | +0.011 |
glass | +0.011 |
ird | +0.011 |
hours | +0.010 |
aster | +0.009 |
Neg logit contributions | |
arre | -0.013 |
up | -0.013 |
itating | -0.013 |
ling | -0.013 |
melon | -0.013 |
Token | of |
Token ID | 286 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
year | +0.051 |
glass | +0.051 |
ird | +0.050 |
aster | +0.048 |
hours | +0.048 |
Neg logit contributions | |
up | -0.051 |
itating | -0.050 |
arre | -0.049 |
ling | -0.048 |
melon | -0.047 |
Token | it |
Token ID | 340 |
Feature activation | -3.24e-01 |
Pos logit contributions | |
itating | +0.003 |
arre | +0.003 |
up | +0.003 |
melon | +0.003 |
ling | +0.003 |
Neg logit contributions | |
year | -0.007 |
glass | -0.007 |
ird | -0.007 |
hours | -0.006 |
aster | -0.006 |
Token | ." |
Token ID | 526 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
year | +0.012 |
glass | +0.011 |
ird | +0.011 |
aster | +0.011 |
hours | +0.011 |
Neg logit contributions | |
up | -0.006 |
itating | -0.006 |
arre | -0.005 |
ling | -0.005 |
melon | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +4.67e-02 |
Pos logit contributions | |
year | +0.000 |
glass | +0.000 |
ird | +0.000 |
hours | +0.000 |
aster | +0.000 |
Neg logit contributions | |
arre | -0.000 |
itating | -0.000 |
up | -0.000 |
melon | -0.000 |
ling | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | +9.01e-01 |
Pos logit contributions | |
itating | +0.001 |
arre | +0.001 |
melon | +0.001 |
up | +0.001 |
ling | +0.001 |
Neg logit contributions | |
year | -0.002 |
ird | -0.001 |
glass | -0.001 |
hours | -0.001 |
aster | -0.001 |
Token | is |
Token ID | 318 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
itating | +0.007 |
arre | +0.007 |
up | +0.006 |
melon | +0.006 |
ling | +0.006 |
Neg logit contributions | |
year | -0.018 |
glass | -0.017 |
ird | -0.017 |
hours | -0.016 |
aster | -0.016 |
Token | a |
Token ID | 257 |
Feature activation | -4.46e-01 |
Pos logit contributions | |
itating | +0.025 |
arre | +0.025 |
up | +0.023 |
melon | +0.023 |
ling | +0.023 |
Neg logit contributions | |
year | -0.034 |
ird | -0.033 |
glass | -0.033 |
hours | -0.031 |
minutes | -0.030 |
Token | special |
Token ID | 2041 |
Feature activation | -5.56e-01 |
Pos logit contributions | |
year | +0.016 |
ird | +0.015 |
glass | +0.014 |
ibr | +0.014 |
aster | +0.013 |
Neg logit contributions | |
up | -0.009 |
melon | -0.008 |
fighting | -0.008 |
itating | -0.008 |
arre | -0.008 |
Token | magazine |
Token ID | 7093 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
year | +0.018 |
glass | +0.017 |
ird | +0.017 |
ibr | +0.016 |
hours | +0.016 |
Neg logit contributions | |
up | -0.012 |
melon | -0.011 |
arre | -0.011 |
fighting | -0.011 |
itating | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -4.40e-01 |
Pos logit contributions | |
year | +0.068 |
glass | +0.064 |
aster | +0.063 |
hours | +0.062 |
ird | +0.062 |
Neg logit contributions | |
up | -0.022 |
melon | -0.019 |
fighting | -0.017 |
Sharing | -0.016 |
itating | -0.016 |
Token | She |
Token ID | 1375 |
Feature activation | -6.24e-01 |
Pos logit contributions | |
year | +0.017 |
glass | +0.016 |
hours | +0.016 |
ird | +0.016 |
aster | +0.015 |
Neg logit contributions | |
itating | -0.006 |
up | -0.006 |
arre | -0.006 |
Sharing | -0.006 |
fighting | -0.006 |
Token | wants |
Token ID | 3382 |
Feature activation | -6.35e-01 |
Pos logit contributions | |
year | +0.023 |
glass | +0.022 |
ird | +0.021 |
hours | +0.020 |
aster | +0.020 |
Neg logit contributions | |
up | -0.011 |
arre | -0.011 |
itating | -0.010 |
fighting | -0.010 |
melon | -0.010 |
Token | to |
Token ID | 284 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
year | +0.023 |
ird | +0.022 |
glass | +0.022 |
ibr | +0.020 |
hours | +0.020 |
Neg logit contributions | |
up | -0.012 |
fighting | -0.011 |
melon | -0.010 |
arre | -0.010 |
itating | -0.010 |
Token | see |
Token ID | 766 |
Feature activation | -8.69e-01 |
Pos logit contributions | |
year | +0.008 |
minutes | +0.007 |
glass | +0.007 |
hours | +0.007 |
ibr | +0.006 |
Neg logit contributions | |
up | -0.004 |
itating | -0.004 |
fighting | -0.004 |
melon | -0.004 |
Freddy | -0.004 |
Token | what |
Token ID | 644 |
Feature activation | -6.95e-02 |
Pos logit contributions | |
year | +0.035 |
glass | +0.033 |
ird | +0.032 |
hours | +0.031 |
aster | +0.031 |
Neg logit contributions | |
up | -0.014 |
fighting | -0.011 |
itating | -0.011 |
ling | -0.010 |
arre | -0.010 |
Token | is |
Token ID | 318 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
year | +0.003 |
glass | +0.003 |
ird | +0.003 |
hours | +0.003 |
aster | +0.003 |
Neg logit contributions | |
up | -0.001 |
itating | -0.001 |
fighting | -0.001 |
arre | -0.001 |
Sharing | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | -4.19e-01 |
Pos logit contributions | |
year | +0.008 |
glass | +0.007 |
ird | +0.007 |
ibr | +0.007 |
ibrarian | +0.007 |
Neg logit contributions | |
up | -0.003 |
fighting | -0.003 |
melon | -0.002 |
itating | -0.002 |
sharing | -0.002 |
Token | freezer |
Token ID | 30967 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
year | +0.013 |
glass | +0.012 |
ird | +0.012 |
hours | +0.011 |
aster | +0.011 |
Neg logit contributions | |
up | -0.010 |
arre | -0.010 |
melon | -0.010 |
itating | -0.009 |
ling | -0.009 |
Token | to |
Token ID | 284 |
Feature activation | -2.58e+00 |
Pos logit contributions | |
year | +0.064 |
hours | +0.061 |
ird | +0.061 |
glass | +0.061 |
aster | +0.060 |
Neg logit contributions | |
up | -0.026 |
melon | -0.022 |
itating | -0.021 |
fighting | -0.020 |
ling | -0.020 |
Token | make |
Token ID | 787 |
Feature activation | -3.99e-01 |
Pos logit contributions | |
year | +0.099 |
glass | +0.097 |
ibr | +0.096 |
ibrarian | +0.090 |
chapters | +0.090 |
Neg logit contributions | |
up | -0.045 |
melon | -0.036 |
fighting | -0.030 |
scatter | -0.029 |
Freddy | -0.029 |
Token | it |
Token ID | 340 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
year | +0.014 |
glass | +0.014 |
ird | +0.014 |
ris | +0.013 |
hours | +0.013 |
Neg logit contributions | |
up | -0.008 |
arre | -0.007 |
ling | -0.007 |
fighting | -0.006 |
bing | -0.006 |
Token | cold |
Token ID | 4692 |
Feature activation | +5.44e-02 |
Pos logit contributions | |
itating | +0.027 |
arre | +0.027 |
melon | +0.025 |
ling | +0.024 |
bing | +0.024 |
Neg logit contributions | |
year | -0.037 |
ird | -0.035 |
glass | -0.035 |
aster | -0.033 |
hours | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | +9.26e-01 |
Pos logit contributions | |
itating | +0.001 |
arre | +0.001 |
melon | +0.001 |
up | +0.001 |
ling | +0.001 |
Neg logit contributions | |
year | -0.002 |
glass | -0.002 |
ird | -0.002 |
hours | -0.002 |
aster | -0.002 |
Token | Lily |
Token ID | 20037 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
itating | +0.022 |
ling | +0.022 |
arre | +0.022 |
melon | +0.021 |
bing | +0.021 |
Neg logit contributions | |
ird | -0.026 |
year | -0.026 |
glass | -0.024 |
hours | -0.024 |
minutes | -0.023 |
Token | asked |
Token ID | 1965 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
arre | +0.004 |
up | +0.004 |
itating | +0.003 |
ling | +0.003 |
melon | +0.003 |
Neg logit contributions | |
year | -0.008 |
glass | -0.008 |
ird | -0.008 |
hours | -0.007 |
aster | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
up | +0.002 |
fighting | +0.001 |
sharing | +0.001 |
Freddy | +0.001 |
melon | +0.001 |
Neg logit contributions | |
year | -0.010 |
glass | -0.010 |
ris | -0.010 |
kil | -0.010 |
ird | -0.009 |
Token | " |
Token ID | 366 |
Feature activation | -6.84e-02 |
Pos logit contributions | |
itating | +0.002 |
arre | +0.002 |
up | +0.002 |
melon | +0.002 |
ling | +0.002 |
Neg logit contributions | |
year | -0.005 |
glass | -0.004 |
ird | -0.004 |
hours | -0.004 |
aster | -0.004 |
Token | " |
Token ID | 1 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
itating | +0.022 |
arre | +0.022 |
ling | +0.021 |
up | +0.021 |
bing | +0.020 |
Neg logit contributions | |
year | -0.036 |
glass | -0.035 |
ird | -0.034 |
hours | -0.033 |
ibr | -0.032 |
Token | Mom |
Token ID | 29252 |
Feature activation | +3.87e-01 |
Pos logit contributions | |
year | +0.004 |
glass | +0.004 |
ird | +0.004 |
ibr | +0.004 |
aster | +0.004 |
Neg logit contributions | |
arre | -0.002 |
itating | -0.002 |
melon | -0.002 |
flower | -0.002 |
up | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
arre | +0.006 |
itating | +0.006 |
up | +0.006 |
melon | +0.006 |
ling | +0.006 |
Neg logit contributions | |
year | -0.014 |
glass | -0.014 |
ird | -0.013 |
hours | -0.013 |
aster | -0.013 |
Token | are |
Token ID | 389 |
Feature activation | +8.46e-01 |
Pos logit contributions | |
year | +0.039 |
glass | +0.036 |
ird | +0.034 |
ris | +0.034 |
hours | +0.034 |
Neg logit contributions | |
up | -0.017 |
fighting | -0.016 |
melon | -0.016 |
Freddy | -0.014 |
accidents | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
melon | +0.010 |
up | +0.010 |
fighting | +0.009 |
Freddy | +0.009 |
sharing | +0.008 |
Neg logit contributions | |
year | -0.036 |
glass | -0.036 |
ird | -0.034 |
ris | -0.034 |
hours | -0.033 |
Token | pizz |
Token ID | 38748 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
melon | +0.040 |
up | +0.039 |
arre | +0.038 |
fighting | +0.038 |
itating | +0.038 |
Neg logit contributions | |
year | -0.044 |
ird | -0.041 |
glass | -0.040 |
hours | -0.038 |
aster | -0.036 |
Token | as |
Token ID | 292 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
hours | +0.067 |
year | +0.063 |
minutes | +0.061 |
aster | +0.060 |
glass | +0.059 |
Neg logit contributions | |
itating | -0.086 |
arre | -0.086 |
melon | -0.084 |
ling | -0.080 |
up | -0.079 |
Token | ready |
Token ID | 3492 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
year | +0.007 |
ird | +0.007 |
glass | +0.007 |
aster | +0.006 |
hours | +0.006 |
Neg logit contributions | |
itating | -0.005 |
arre | -0.005 |
up | -0.005 |
melon | -0.004 |
ling | -0.004 |
Token | ?" |
Token ID | 1701 |
Feature activation | +4.54e-01 |
Pos logit contributions | |
up | +0.003 |
arre | +0.003 |
itating | +0.003 |
melon | +0.003 |
ling | +0.003 |
Neg logit contributions | |
year | -0.006 |
glass | -0.005 |
ird | -0.005 |
hours | -0.005 |
ibr | -0.005 |
Token | Lily |
Token ID | 20037 |
Feature activation | -9.72e-01 |
Pos logit contributions | |
itating | +0.007 |
arre | +0.007 |
up | +0.007 |
melon | +0.007 |
ling | +0.006 |
Neg logit contributions | |
year | -0.017 |
glass | -0.016 |
ird | -0.016 |
hours | -0.016 |
aster | -0.015 |
Token | asked |
Token ID | 1965 |
Feature activation | -5.77e-01 |
Pos logit contributions | |
year | +0.035 |
glass | +0.032 |
ird | +0.031 |
hours | +0.031 |
aster | +0.031 |
Neg logit contributions | |
up | -0.018 |
arre | -0.018 |
itating | -0.018 |
fighting | -0.017 |
ling | -0.017 |
Token | It |
Token ID | 632 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
itating | +0.009 |
arre | +0.009 |
ling | +0.009 |
melon | +0.009 |
up | +0.009 |
Neg logit contributions | |
year | -0.019 |
ird | -0.018 |
glass | -0.018 |
hours | -0.018 |
aster | -0.017 |
Token | was |
Token ID | 373 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
arre | +0.030 |
itating | +0.030 |
melon | +0.027 |
ling | +0.027 |
bing | +0.026 |
Neg logit contributions | |
year | -0.057 |
ird | -0.055 |
glass | -0.055 |
hours | -0.052 |
minutes | -0.050 |
Token | a |
Token ID | 257 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
ird | +0.009 |
glass | +0.009 |
year | +0.008 |
ibr | +0.008 |
ris | +0.008 |
Neg logit contributions | |
up | -0.005 |
melon | -0.004 |
fighting | -0.004 |
Freddy | -0.004 |
sharing | -0.004 |
Token | book |
Token ID | 1492 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
year | +0.042 |
ird | +0.039 |
glass | +0.039 |
ibr | +0.038 |
hours | +0.037 |
Neg logit contributions | |
up | -0.022 |
melon | -0.022 |
fighting | -0.020 |
itating | -0.019 |
arre | -0.018 |
Token | about |
Token ID | 546 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
year | +0.074 |
glass | +0.070 |
ird | +0.070 |
hours | +0.067 |
aster | +0.067 |
Neg logit contributions | |
up | -0.036 |
itating | -0.035 |
arre | -0.033 |
melon | -0.033 |
ling | -0.032 |
Token | history |
Token ID | 2106 |
Feature activation | -2.98e+00 |
Pos logit contributions | |
year | +0.043 |
ird | +0.043 |
glass | +0.041 |
aster | +0.040 |
ris | +0.039 |
Neg logit contributions | |
up | -0.027 |
fighting | -0.025 |
sharing | -0.024 |
melon | -0.023 |
arre | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | +6.47e-01 |
Pos logit contributions | |
year | +0.127 |
ird | +0.120 |
aster | +0.119 |
glass | +0.117 |
Ah | +0.115 |
Neg logit contributions | |
up | -0.040 |
sharing | -0.027 |
fighting | -0.025 |
ling | -0.025 |
melon | -0.024 |
Token | Sophie |
Token ID | 35331 |
Feature activation | +4.36e-01 |
Pos logit contributions | |
itating | +0.013 |
arre | +0.013 |
ling | +0.013 |
melon | +0.012 |
up | +0.012 |
Neg logit contributions | |
year | -0.021 |
ird | -0.021 |
glass | -0.020 |
hours | -0.019 |
minutes | -0.019 |
Token | was |
Token ID | 373 |
Feature activation | -7.47e-01 |
Pos logit contributions | |
up | +0.008 |
arre | +0.008 |
itating | +0.008 |
melon | +0.008 |
ling | +0.008 |
Neg logit contributions | |
year | -0.015 |
glass | -0.014 |
ird | -0.014 |
hours | -0.013 |
aster | -0.013 |
Token | so |
Token ID | 523 |
Feature activation | -8.60e-01 |
Pos logit contributions | |
year | +0.025 |
glass | +0.024 |
ird | +0.024 |
hours | +0.023 |
aster | +0.022 |
Neg logit contributions | |
up | -0.015 |
itating | -0.015 |
arre | -0.015 |
melon | -0.014 |
fighting | -0.014 |
Token | proud |
Token ID | 6613 |
Feature activation | +4.56e-01 |
Pos logit contributions | |
glass | +0.026 |
year | +0.025 |
ird | +0.025 |
hours | +0.025 |
aster | +0.024 |
Neg logit contributions | |
up | -0.020 |
itating | -0.020 |
arre | -0.019 |
fighting | -0.019 |
melon | -0.018 |
Token | They |
Token ID | 1119 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
year | +0.042 |
ird | +0.040 |
glass | +0.039 |
hours | +0.038 |
ocket | +0.037 |
Neg logit contributions | |
fighting | -0.009 |
arre | -0.009 |
Sharing | -0.009 |
Freddy | -0.009 |
up | -0.009 |
Token | had |
Token ID | 550 |
Feature activation | -4.51e-01 |
Pos logit contributions | |
year | +0.039 |
ae | +0.038 |
ird | +0.037 |
aster | +0.037 |
glass | +0.037 |
Neg logit contributions | |
fighting | -0.016 |
qu | -0.016 |
melon | -0.014 |
up | -0.014 |
en | -0.013 |
Token | h |
Token ID | 289 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
year | +0.016 |
ird | +0.015 |
glass | +0.014 |
ibr | +0.014 |
aster | +0.014 |
Neg logit contributions | |
fighting | -0.009 |
up | -0.008 |
qu | -0.008 |
melon | -0.008 |
arre | -0.007 |
Token | oses |
Token ID | 4629 |
Feature activation | -4.09e-01 |
Pos logit contributions | |
year | +0.054 |
ird | +0.051 |
glass | +0.051 |
hours | +0.049 |
aster | +0.048 |
Neg logit contributions | |
arre | -0.059 |
itating | -0.058 |
up | -0.056 |
melon | -0.055 |
fighting | -0.054 |
Token | and |
Token ID | 290 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
year | +0.016 |
ird | +0.015 |
glass | +0.015 |
hours | +0.015 |
aster | +0.014 |
Neg logit contributions | |
itating | -0.006 |
up | -0.006 |
arre | -0.006 |
melon | -0.005 |
ling | -0.005 |
Token | l |
Token ID | 300 |
Feature activation | -3.20e+00 |
Pos logit contributions | |
year | +0.036 |
ird | +0.033 |
ae | +0.032 |
ris | +0.031 |
ibrarian | +0.031 |
Neg logit contributions | |
fighting | -0.020 |
qu | -0.018 |
melon | -0.018 |
up | -0.017 |
lings | -0.017 |
Token | adders |
Token ID | 45940 |
Feature activation | -6.56e-01 |
Pos logit contributions | |
aster | +0.076 |
Bible | +0.071 |
hours | +0.065 |
ae | +0.065 |
ird | +0.064 |
Neg logit contributions | |
ters | -0.091 |
lings | -0.088 |
arre | -0.087 |
melon | -0.087 |
ling | -0.087 |
Token | . |
Token ID | 13 |
Feature activation | -8.01e-01 |
Pos logit contributions | |
year | +0.024 |
ird | +0.024 |
glass | +0.023 |
hours | +0.022 |
aster | +0.022 |
Neg logit contributions | |
itating | -0.010 |
arre | -0.010 |
up | -0.010 |
melon | -0.010 |
ling | -0.009 |
Token | They |
Token ID | 1119 |
Feature activation | -8.03e-01 |
Pos logit contributions | |
year | +0.036 |
ird | +0.034 |
glass | +0.034 |
hours | +0.032 |
lock | +0.032 |
Neg logit contributions | |
Sharing | -0.007 |
fighting | -0.007 |
itating | -0.007 |
arre | -0.007 |
Freddy | -0.007 |
Token | were |
Token ID | 547 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
year | +0.031 |
ae | +0.029 |
glass | +0.029 |
aster | +0.029 |
ird | +0.028 |
Neg logit contributions | |
fighting | -0.012 |
qu | -0.011 |
melon | -0.011 |
up | -0.010 |
en | -0.009 |
Token | firefighters |
Token ID | 22611 |
Feature activation | -9.17e-02 |
Pos logit contributions | |
ird | +0.008 |
year | +0.008 |
glass | +0.007 |
lock | +0.007 |
aster | +0.007 |
Neg logit contributions | |
up | -0.004 |
fighting | -0.004 |
melon | -0.003 |
ling | -0.003 |
en | -0.003 |
Token | something |
Token ID | 1223 |
Feature activation | -6.11e-01 |
Pos logit contributions | |
itating | +0.024 |
arre | +0.024 |
up | +0.023 |
melon | +0.023 |
ling | +0.023 |
Neg logit contributions | |
year | -0.039 |
ird | -0.038 |
glass | -0.038 |
hours | -0.036 |
aster | -0.035 |
Token | magic |
Token ID | 5536 |
Feature activation | -9.75e-01 |
Pos logit contributions | |
year | +0.018 |
glass | +0.017 |
ird | +0.017 |
hours | +0.017 |
minutes | +0.017 |
Neg logit contributions | |
itating | -0.014 |
arre | -0.014 |
bing | -0.013 |
melon | -0.013 |
up | -0.013 |
Token | happened |
Token ID | 3022 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
year | +0.032 |
glass | +0.031 |
ird | +0.031 |
hours | +0.029 |
aster | +0.029 |
Neg logit contributions | |
itating | -0.019 |
arre | -0.019 |
up | -0.018 |
melon | -0.018 |
ling | -0.018 |
Token | ! |
Token ID | 0 |
Feature activation | -8.81e-01 |
Pos logit contributions | |
itating | +0.049 |
arre | +0.049 |
melon | +0.046 |
rust | +0.046 |
ling | +0.044 |
Neg logit contributions | |
year | -0.063 |
glass | -0.062 |
ird | -0.059 |
hours | -0.058 |
minutes | -0.056 |
Token | The |
Token ID | 383 |
Feature activation | +4.19e-01 |
Pos logit contributions | |
year | +0.031 |
glass | +0.030 |
hours | +0.029 |
ird | +0.029 |
minutes | +0.029 |
Neg logit contributions | |
Freddy | -0.015 |
itating | -0.015 |
arre | -0.014 |
up | -0.014 |
melon | -0.014 |
Token | m |
Token ID | 285 |
Feature activation | -3.81e+00 |
Pos logit contributions | |
up | +0.009 |
melon | +0.008 |
ling | +0.008 |
fighting | +0.008 |
itating | +0.007 |
Neg logit contributions | |
year | -0.014 |
ird | -0.014 |
glass | -0.013 |
ris | -0.012 |
aster | -0.012 |
Token | ugs |
Token ID | 10339 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
hours | +0.106 |
year | +0.103 |
minutes | +0.101 |
glass | +0.100 |
Africa | +0.097 |
Neg logit contributions | |
ling | -0.092 |
bing | -0.089 |
flower | -0.088 |
itating | -0.087 |
up | -0.085 |
Token | split |
Token ID | 6626 |
Feature activation | +8.54e-01 |
Pos logit contributions | |
year | +0.056 |
glass | +0.054 |
ird | +0.053 |
hours | +0.051 |
aster | +0.050 |
Neg logit contributions | |
itating | -0.036 |
arre | -0.036 |
up | -0.035 |
melon | -0.034 |
ling | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
itating | +0.017 |
arre | +0.017 |
up | +0.017 |
melon | +0.016 |
ling | +0.016 |
Neg logit contributions | |
year | -0.028 |
ird | -0.027 |
glass | -0.027 |
hours | -0.025 |
aster | -0.025 |
Token | They |
Token ID | 1119 |
Feature activation | -5.86e-01 |
Pos logit contributions | |
year | +0.040 |
glass | +0.039 |
ird | +0.037 |
hours | +0.037 |
minutes | +0.036 |
Neg logit contributions | |
Freddy | -0.017 |
arre | -0.017 |
itating | -0.017 |
up | -0.016 |
melon | -0.016 |
Token | opened |
Token ID | 4721 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
year | +0.021 |
glass | +0.020 |
ird | +0.020 |
aster | +0.019 |
minutes | +0.019 |
Neg logit contributions | |
up | -0.010 |
fighting | -0.010 |
melon | -0.010 |
itating | -0.010 |
arre | -0.010 |
Token | light |
Token ID | 1657 |
Feature activation | -7.32e-01 |
Pos logit contributions | |
up | +0.013 |
ling | +0.012 |
melon | +0.012 |
arre | +0.012 |
lings | +0.012 |
Neg logit contributions | |
year | -0.033 |
ird | -0.033 |
glass | -0.031 |
aster | -0.030 |
ris | -0.030 |
Token | of |
Token ID | 286 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
year | +0.026 |
ird | +0.025 |
glass | +0.025 |
hours | +0.023 |
aster | +0.023 |
Neg logit contributions | |
up | -0.013 |
itating | -0.013 |
arre | -0.013 |
melon | -0.013 |
ling | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | -3.92e-01 |
Pos logit contributions | |
ird | +0.047 |
year | +0.046 |
glass | +0.044 |
ris | +0.042 |
lock | +0.042 |
Neg logit contributions | |
up | -0.019 |
fighting | -0.018 |
melon | -0.017 |
Freddy | -0.016 |
accidents | -0.015 |
Token | sunrise |
Token ID | 40048 |
Feature activation | -4.54e-01 |
Pos logit contributions | |
ird | +0.015 |
year | +0.015 |
glass | +0.014 |
aster | +0.014 |
ibr | +0.013 |
Neg logit contributions | |
up | -0.007 |
ling | -0.006 |
fighting | -0.006 |
melon | -0.006 |
itating | -0.006 |
Token | streaming |
Token ID | 11305 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
year | +0.018 |
ird | +0.017 |
glass | +0.017 |
hours | +0.016 |
aster | +0.016 |
Neg logit contributions | |
up | -0.007 |
melon | -0.006 |
ling | -0.006 |
itating | -0.006 |
flower | -0.006 |
Token | into |
Token ID | 656 |
Feature activation | -2.90e+00 |
Pos logit contributions | |
year | +0.070 |
ird | +0.069 |
glass | +0.068 |
hours | +0.065 |
aster | +0.065 |
Neg logit contributions | |
up | -0.022 |
itating | -0.022 |
arre | -0.021 |
ling | -0.020 |
melon | -0.019 |
Token | his |
Token ID | 465 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
ird | +0.124 |
year | +0.119 |
glass | +0.116 |
ris | +0.115 |
lock | +0.114 |
Neg logit contributions | |
up | -0.038 |
fighting | -0.034 |
ling | -0.032 |
accidents | -0.030 |
en | -0.027 |
Token | room |
Token ID | 2119 |
Feature activation | -3.90e-01 |
Pos logit contributions | |
year | +0.051 |
ird | +0.047 |
glass | +0.045 |
ris | +0.043 |
hours | +0.043 |
Neg logit contributions | |
up | -0.029 |
ling | -0.028 |
arre | -0.027 |
lings | -0.027 |
fighting | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
year | +0.015 |
ird | +0.014 |
glass | +0.014 |
hours | +0.013 |
aster | +0.013 |
Neg logit contributions | |
up | -0.006 |
arre | -0.006 |
ling | -0.006 |
itating | -0.006 |
melon | -0.006 |
Token | |
Token ID | 220 |
Feature activation | -7.56e-01 |
Pos logit contributions | |
arre | +0.003 |
itating | +0.003 |
up | +0.003 |
melon | +0.003 |
fighting | +0.002 |
Neg logit contributions | |
year | -0.006 |
glass | -0.006 |
ird | -0.006 |
hours | -0.005 |
aster | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
year | +0.025 |
ird | +0.024 |
glass | +0.024 |
hours | +0.023 |
aster | +0.022 |
Neg logit contributions | |
arre | -0.015 |
itating | -0.015 |
melon | -0.014 |
up | -0.014 |
ling | -0.014 |