Token | " |
Token ID | 366 |
Feature activation | +3.92e-03 |
Pos logit contributions | |
enough | +0.007 |
once | +0.007 |
best | +0.006 |
again | +0.006 |
more | +0.006 |
Neg logit contributions | |
agons | -0.011 |
ramids | -0.010 |
alks | -0.010 |
disl | -0.010 |
pier | -0.010 |
Token | That |
Token ID | 2504 |
Feature activation | -1.25e-02 |
Pos logit contributions | |
disl | +0.006 |
agons | +0.006 |
ramids | +0.006 |
igator | +0.006 |
curb | +0.006 |
Neg logit contributions | |
enough | -0.007 |
once | -0.006 |
best | -0.006 |
again | -0.005 |
forever | -0.005 |
Token | 's |
Token ID | 338 |
Feature activation | -9.05e-03 |
Pos logit contributions | |
enough | +0.012 |
once | +0.011 |
moments | +0.010 |
again | +0.009 |
best | +0.009 |
Neg logit contributions | |
agons | -0.027 |
disl | -0.026 |
ramids | -0.026 |
igator | -0.025 |
pier | -0.025 |
Token | how |
Token ID | 703 |
Feature activation | -7.42e-03 |
Pos logit contributions | |
enough | +0.009 |
once | +0.009 |
again | +0.008 |
moments | +0.007 |
best | +0.007 |
Neg logit contributions | |
agons | -0.020 |
ramids | -0.018 |
disl | -0.018 |
issors | -0.018 |
ants | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | +2.84e-02 |
Pos logit contributions | |
enough | +0.009 |
once | +0.009 |
best | +0.008 |
again | +0.008 |
more | +0.008 |
Neg logit contributions | |
agons | -0.015 |
ramids | -0.014 |
curb | -0.014 |
pier | -0.014 |
alks | -0.014 |
Token | oct |
Token ID | 19318 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
agons | +0.051 |
disl | +0.048 |
ramids | +0.047 |
curb | +0.047 |
igator | +0.047 |
Neg logit contributions | |
enough | -0.039 |
once | -0.037 |
again | -0.033 |
best | -0.032 |
moments | -0.031 |
Token | opus |
Token ID | 25790 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
ramids | +0.119 |
agons | +0.119 |
curb | +0.113 |
issors | +0.106 |
alks | +0.106 |
Neg logit contributions | |
once | -0.206 |
enough | -0.197 |
best | -0.192 |
again | -0.190 |
moments | -0.175 |
Token | feels |
Token ID | 5300 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
agons | +0.079 |
ramids | +0.075 |
disl | +0.073 |
alks | +0.073 |
ants | +0.072 |
Neg logit contributions | |
enough | -0.042 |
once | -0.039 |
best | -0.035 |
again | -0.034 |
moments | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | +8.93e-03 |
Pos logit contributions | |
agons | +0.049 |
ramids | +0.047 |
alks | +0.046 |
ants | +0.045 |
rolling | +0.045 |
Neg logit contributions | |
enough | -0.023 |
once | -0.021 |
best | -0.019 |
again | -0.019 |
more | -0.017 |
Token | And |
Token ID | 843 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
agons | +0.017 |
ramids | +0.016 |
rolling | +0.016 |
ants | +0.015 |
pier | +0.015 |
Neg logit contributions | |
enough | -0.012 |
once | -0.012 |
best | -0.010 |
again | -0.010 |
moments | -0.010 |
Token | that |
Token ID | 326 |
Feature activation | +2.82e-03 |
Pos logit contributions | |
agons | +0.038 |
disl | +0.038 |
ramids | +0.036 |
pier | +0.036 |
aro | +0.036 |
Neg logit contributions | |
enough | -0.035 |
once | -0.035 |
again | -0.031 |
best | -0.030 |
moments | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | +4.95e-03 |
Pos logit contributions | |
agons | +0.007 |
ramids | +0.006 |
disl | +0.006 |
alks | +0.006 |
rolling | +0.006 |
Neg logit contributions | |
enough | -0.002 |
once | -0.002 |
again | -0.002 |
best | -0.002 |
more | -0.001 |
Token | we |
Token ID | 356 |
Feature activation | -3.09e-02 |
Pos logit contributions | |
aro | +0.006 |
pier | +0.006 |
igator | +0.005 |
yacht | +0.005 |
agons | +0.005 |
Neg logit contributions | |
once | -0.009 |
enough | -0.009 |
quality | -0.008 |
best | -0.008 |
moments | -0.008 |
Token | can |
Token ID | 460 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
enough | +0.038 |
once | +0.035 |
best | +0.033 |
again | +0.031 |
more | +0.029 |
Neg logit contributions | |
agons | -0.066 |
ramids | -0.064 |
ants | -0.062 |
curb | -0.061 |
aro | -0.061 |
Token | ," |
Token ID | 553 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
once | +0.013 |
again | +0.013 |
enough | +0.013 |
more | +0.011 |
best | +0.011 |
Neg logit contributions | |
agons | -0.027 |
alks | -0.026 |
ramids | -0.025 |
ants | -0.025 |
pier | -0.025 |
Token | the |
Token ID | 262 |
Feature activation | +2.83e-02 |
Pos logit contributions | |
enough | +0.020 |
once | +0.019 |
best | +0.018 |
moments | +0.016 |
again | +0.016 |
Neg logit contributions | |
agons | -0.023 |
ramids | -0.023 |
alks | -0.022 |
pier | -0.022 |
igator | -0.022 |
Token | oct |
Token ID | 19318 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
agons | +0.041 |
ramids | +0.039 |
issors | +0.037 |
ants | +0.037 |
rolling | +0.036 |
Neg logit contributions | |
once | -0.051 |
enough | -0.051 |
best | -0.050 |
again | -0.047 |
more | -0.046 |
Token | opus |
Token ID | 25790 |
Feature activation | +3.16e-02 |
Pos logit contributions | |
agons | +0.140 |
ramids | +0.130 |
issors | +0.121 |
curb | +0.120 |
alks | +0.119 |
Neg logit contributions | |
once | -0.190 |
again | -0.175 |
enough | -0.168 |
best | -0.165 |
ever | -0.152 |
Token | said |
Token ID | 531 |
Feature activation | +4.23e-03 |
Pos logit contributions | |
agons | +0.059 |
ramids | +0.056 |
alks | +0.054 |
disl | +0.054 |
ants | +0.053 |
Neg logit contributions | |
enough | -0.043 |
once | -0.041 |
best | -0.038 |
again | -0.037 |
moments | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -6.33e-03 |
Pos logit contributions | |
agons | +0.007 |
ramids | +0.006 |
rus | +0.006 |
alks | +0.006 |
pier | +0.006 |
Neg logit contributions | |
enough | -0.006 |
once | -0.006 |
best | -0.005 |
moments | -0.005 |
quality | -0.005 |
Token | " |
Token ID | 366 |
Feature activation | +4.47e-03 |
Pos logit contributions | |
enough | +0.009 |
once | +0.009 |
best | +0.008 |
moments | +0.008 |
again | +0.008 |
Neg logit contributions | |
agons | -0.010 |
ramids | -0.010 |
alks | -0.010 |
pier | -0.010 |
curb | -0.010 |
Token | We |
Token ID | 1135 |
Feature activation | +3.28e-03 |
Pos logit contributions | |
ramids | +0.007 |
agons | +0.007 |
issors | +0.007 |
curb | +0.007 |
disl | +0.007 |
Neg logit contributions | |
enough | -0.007 |
best | -0.006 |
once | -0.006 |
moments | -0.005 |
forever | -0.005 |
Token | selfish |
Token ID | 20363 |
Feature activation | +6.33e-03 |
Pos logit contributions | |
enough | +0.009 |
once | +0.008 |
best | +0.008 |
again | +0.007 |
more | +0.007 |
Neg logit contributions | |
agons | -0.012 |
ramids | -0.011 |
alks | -0.011 |
ants | -0.011 |
disl | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | +2.41e-03 |
Pos logit contributions | |
agons | +0.011 |
cases | +0.010 |
issors | +0.010 |
alks | +0.010 |
lee | +0.010 |
Neg logit contributions | |
once | -0.006 |
moments | -0.006 |
Sweet | -0.006 |
chapter | -0.006 |
moment | -0.006 |
Token | not |
Token ID | 407 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
agons | +0.005 |
ramids | +0.004 |
curb | +0.004 |
alks | +0.004 |
disl | +0.004 |
Neg logit contributions | |
once | -0.003 |
enough | -0.002 |
best | -0.002 |
again | -0.002 |
moments | -0.002 |
Token | letting |
Token ID | 9616 |
Feature activation | -9.49e-03 |
Pos logit contributions | |
once | +0.018 |
moments | +0.016 |
enough | +0.016 |
best | +0.016 |
again | +0.015 |
Neg logit contributions | |
disl | -0.024 |
ramids | -0.023 |
agons | -0.023 |
alks | -0.023 |
pier | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
enough | +0.013 |
once | +0.012 |
best | +0.011 |
again | +0.011 |
more | +0.010 |
Neg logit contributions | |
agons | -0.018 |
ramids | -0.017 |
disl | -0.017 |
alks | -0.017 |
curb | -0.017 |
Token | oct |
Token ID | 19318 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
agons | +0.032 |
ramids | +0.030 |
ants | +0.029 |
alks | +0.028 |
disl | +0.028 |
Neg logit contributions | |
enough | -0.034 |
once | -0.033 |
best | -0.031 |
again | -0.030 |
more | -0.029 |
Token | opus |
Token ID | 25790 |
Feature activation | +3.48e-02 |
Pos logit contributions | |
agons | +0.129 |
ramids | +0.125 |
curb | +0.121 |
disl | +0.119 |
issors | +0.113 |
Neg logit contributions | |
once | -0.193 |
enough | -0.183 |
again | -0.179 |
best | -0.176 |
more | -0.160 |
Token | swim |
Token ID | 9422 |
Feature activation | +9.23e-03 |
Pos logit contributions | |
ramids | +0.070 |
agons | +0.069 |
alks | +0.065 |
ants | +0.065 |
issors | +0.064 |
Neg logit contributions | |
enough | -0.049 |
once | -0.046 |
again | -0.045 |
best | -0.043 |
more | -0.041 |
Token | free |
Token ID | 1479 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
agons | +0.020 |
ramids | +0.020 |
ants | +0.019 |
alks | +0.019 |
disl | +0.019 |
Neg logit contributions | |
enough | -0.011 |
once | -0.010 |
again | -0.010 |
best | -0.009 |
more | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -2.29e-03 |
Pos logit contributions | |
agons | +0.038 |
ramids | +0.036 |
alks | +0.035 |
disl | +0.035 |
pier | +0.035 |
Neg logit contributions | |
enough | -0.024 |
once | -0.023 |
best | -0.021 |
moments | -0.019 |
again | -0.019 |
Token | So |
Token ID | 1406 |
Feature activation | +2.28e-02 |
Pos logit contributions | |
enough | +0.003 |
once | +0.003 |
best | +0.003 |
again | +0.003 |
more | +0.002 |
Neg logit contributions | |
agons | -0.005 |
ramids | -0.004 |
alks | -0.004 |
disl | -0.004 |
ants | -0.004 |
Token | - |
Token ID | 12 |
Feature activation | +3.33e-02 |
Pos logit contributions | |
agons | +0.035 |
ramids | +0.033 |
alks | +0.032 |
disl | +0.032 |
curb | +0.032 |
Neg logit contributions | |
enough | -0.022 |
once | -0.021 |
best | -0.018 |
again | -0.018 |
more | -0.017 |
Token | war |
Token ID | 5767 |
Feature activation | +5.45e-03 |
Pos logit contributions | |
agons | +0.055 |
ramids | +0.052 |
alks | +0.050 |
disl | +0.049 |
ants | +0.049 |
Neg logit contributions | |
enough | -0.054 |
once | -0.051 |
best | -0.048 |
again | -0.047 |
moments | -0.044 |
Token | ," |
Token ID | 553 |
Feature activation | -8.08e-03 |
Pos logit contributions | |
agons | +0.010 |
ramids | +0.010 |
ants | +0.010 |
rus | +0.009 |
issors | +0.009 |
Neg logit contributions | |
enough | -0.007 |
once | -0.006 |
best | -0.006 |
moments | -0.006 |
friendship | -0.005 |
Token | said |
Token ID | 531 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
enough | +0.014 |
once | +0.013 |
best | +0.013 |
forever | +0.012 |
again | +0.012 |
Neg logit contributions | |
igator | -0.011 |
ramids | -0.011 |
pans | -0.011 |
disl | -0.011 |
agons | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | +2.10e-02 |
Pos logit contributions | |
ramids | +0.020 |
agons | +0.020 |
igator | +0.020 |
alks | +0.019 |
aro | +0.019 |
Neg logit contributions | |
enough | -0.016 |
once | -0.015 |
best | -0.015 |
moments | -0.013 |
again | -0.013 |
Token | oct |
Token ID | 19318 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
agons | +0.030 |
ramids | +0.029 |
ants | +0.028 |
issors | +0.027 |
disl | +0.027 |
Neg logit contributions | |
enough | -0.037 |
once | -0.037 |
best | -0.036 |
again | -0.034 |
more | -0.033 |
Token | opus |
Token ID | 25790 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
agons | +0.127 |
ramids | +0.119 |
curb | +0.112 |
disl | +0.112 |
alks | +0.110 |
Neg logit contributions | |
once | -0.194 |
again | -0.179 |
enough | -0.179 |
best | -0.171 |
moments | -0.160 |
Token | . |
Token ID | 13 |
Feature activation | +6.41e-04 |
Pos logit contributions | |
agons | +0.055 |
ramids | +0.052 |
pier | +0.050 |
alks | +0.050 |
disl | +0.050 |
Neg logit contributions | |
enough | -0.027 |
once | -0.026 |
again | -0.023 |
best | -0.022 |
more | -0.021 |
Token | " |
Token ID | 366 |
Feature activation | +3.06e-03 |
Pos logit contributions | |
ramids | +0.001 |
agons | +0.001 |
curb | +0.001 |
alks | +0.001 |
pier | +0.001 |
Neg logit contributions | |
enough | -0.001 |
best | -0.001 |
once | -0.001 |
forever | -0.001 |
moments | -0.001 |
Token | I |
Token ID | 40 |
Feature activation | +6.76e-03 |
Pos logit contributions | |
agons | +0.006 |
ramids | +0.006 |
issors | +0.005 |
ants | +0.005 |
curb | +0.005 |
Neg logit contributions | |
enough | -0.004 |
once | -0.004 |
best | -0.004 |
again | -0.003 |
moments | -0.003 |
Token | have |
Token ID | 423 |
Feature activation | -4.45e-03 |
Pos logit contributions | |
agons | +0.012 |
ramids | +0.012 |
alks | +0.011 |
ants | +0.011 |
curb | +0.011 |
Neg logit contributions | |
enough | -0.010 |
once | -0.010 |
best | -0.009 |
again | -0.009 |
more | -0.008 |
Token | he |
Token ID | 339 |
Feature activation | -5.31e-03 |
Pos logit contributions | |
enough | +0.014 |
once | +0.014 |
best | +0.012 |
moments | +0.011 |
again | +0.010 |
Neg logit contributions | |
agons | -0.036 |
ramids | -0.035 |
disl | -0.035 |
curb | -0.035 |
pier | -0.035 |
Token | wanted |
Token ID | 2227 |
Feature activation | +9.93e-03 |
Pos logit contributions | |
enough | +0.007 |
once | +0.006 |
best | +0.006 |
again | +0.006 |
more | +0.005 |
Neg logit contributions | |
agons | -0.011 |
ramids | -0.010 |
alks | -0.010 |
disl | -0.010 |
ants | -0.010 |
Token | to |
Token ID | 284 |
Feature activation | +3.24e-03 |
Pos logit contributions | |
alks | +0.020 |
agons | +0.020 |
ramids | +0.019 |
disl | +0.019 |
pier | +0.019 |
Neg logit contributions | |
once | -0.008 |
enough | -0.008 |
best | -0.007 |
moments | -0.007 |
moment | -0.007 |
Token | keep |
Token ID | 1394 |
Feature activation | +2.25e-03 |
Pos logit contributions | |
agons | +0.006 |
ramids | +0.006 |
ants | +0.006 |
alks | +0.006 |
rus | +0.005 |
Neg logit contributions | |
enough | -0.005 |
once | -0.005 |
again | -0.004 |
best | -0.004 |
more | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
agons | +0.004 |
ramids | +0.004 |
alks | +0.004 |
disl | +0.004 |
pier | +0.004 |
Neg logit contributions | |
enough | -0.003 |
once | -0.003 |
best | -0.002 |
again | -0.002 |
moments | -0.002 |
Token | oct |
Token ID | 19318 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
agons | +0.027 |
pier | +0.025 |
ramids | +0.024 |
alks | +0.024 |
disl | +0.023 |
Neg logit contributions | |
enough | -0.045 |
once | -0.043 |
again | -0.040 |
moments | -0.039 |
best | -0.039 |
Token | opus |
Token ID | 25790 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
agons | +0.136 |
ramids | +0.131 |
curb | +0.131 |
disl | +0.127 |
stret | +0.123 |
Neg logit contributions | |
once | -0.183 |
again | -0.167 |
best | -0.166 |
enough | -0.160 |
ever | -0.149 |
Token | anyway |
Token ID | 6949 |
Feature activation | +5.83e-03 |
Pos logit contributions | |
agons | +0.049 |
ramids | +0.046 |
alks | +0.045 |
disl | +0.044 |
ants | +0.044 |
Neg logit contributions | |
enough | -0.028 |
once | -0.026 |
best | -0.023 |
again | -0.023 |
moments | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | +2.75e-03 |
Pos logit contributions | |
agons | +0.011 |
ramids | +0.010 |
alks | +0.010 |
ants | +0.010 |
issors | +0.010 |
Neg logit contributions | |
enough | -0.007 |
once | -0.007 |
best | -0.006 |
moments | -0.006 |
again | -0.006 |
Token | " |
Token ID | 366 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
agons | +0.005 |
ramids | +0.005 |
pier | +0.005 |
ants | +0.005 |
alks | +0.005 |
Neg logit contributions | |
enough | -0.004 |
once | -0.004 |
again | -0.003 |
best | -0.003 |
more | -0.003 |
Token | I |
Token ID | 40 |
Feature activation | +6.07e-03 |
Pos logit contributions | |
agons | +0.019 |
ramids | +0.019 |
disl | +0.018 |
issors | +0.018 |
curb | +0.018 |
Neg logit contributions | |
enough | -0.017 |
once | -0.016 |
best | -0.015 |
again | -0.014 |
moments | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | +2.99e-02 |
Pos logit contributions | |
agons | +0.039 |
ramids | +0.037 |
alks | +0.036 |
disl | +0.036 |
ants | +0.036 |
Neg logit contributions | |
enough | -0.017 |
once | -0.016 |
best | -0.014 |
again | -0.014 |
moments | -0.012 |
Token | duck |
Token ID | 22045 |
Feature activation | -1.36e-03 |
Pos logit contributions | |
agons | +0.058 |
aned | +0.054 |
issors | +0.053 |
rolling | +0.053 |
ramids | +0.053 |
Neg logit contributions | |
enough | -0.042 |
once | -0.041 |
best | -0.040 |
more | -0.036 |
again | -0.036 |
Token | ," |
Token ID | 553 |
Feature activation | -3.70e-03 |
Pos logit contributions | |
enough | +0.002 |
once | +0.002 |
again | +0.002 |
best | +0.002 |
more | +0.002 |
Neg logit contributions | |
agons | -0.003 |
ramids | -0.002 |
alks | -0.002 |
curb | -0.002 |
issors | -0.002 |
Token | said |
Token ID | 531 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
enough | +0.005 |
once | +0.005 |
best | +0.004 |
again | +0.004 |
forever | +0.004 |
Neg logit contributions | |
agons | -0.006 |
igator | -0.006 |
ramids | -0.006 |
disl | -0.006 |
pans | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
ramids | +0.024 |
agons | +0.024 |
igator | +0.024 |
ants | +0.023 |
alks | +0.023 |
Neg logit contributions | |
enough | -0.018 |
once | -0.016 |
best | -0.015 |
again | -0.014 |
moments | -0.014 |
Token | oct |
Token ID | 19318 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
agons | +0.047 |
ramids | +0.045 |
ants | +0.043 |
alks | +0.043 |
issors | +0.043 |
Neg logit contributions | |
enough | -0.035 |
once | -0.034 |
best | -0.032 |
again | -0.030 |
more | -0.029 |
Token | opus |
Token ID | 25790 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
agons | +0.124 |
ramids | +0.114 |
curb | +0.109 |
disl | +0.106 |
alks | +0.105 |
Neg logit contributions | |
once | -0.197 |
enough | -0.185 |
again | -0.184 |
best | -0.178 |
more | -0.166 |
Token | . |
Token ID | 13 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
agons | +0.064 |
ramids | +0.060 |
alks | +0.058 |
issors | +0.058 |
curb | +0.058 |
Neg logit contributions | |
once | -0.027 |
enough | -0.027 |
again | -0.025 |
best | -0.023 |
more | -0.022 |
Token | " |
Token ID | 366 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
agons | +0.017 |
ramids | +0.017 |
alks | +0.016 |
curb | +0.016 |
disl | +0.016 |
Neg logit contributions | |
enough | -0.015 |
once | -0.014 |
best | -0.013 |
again | -0.012 |
moments | -0.012 |
Token | Do |
Token ID | 5211 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
agons | +0.024 |
ramids | +0.023 |
alks | +0.022 |
disl | +0.022 |
ants | +0.022 |
Neg logit contributions | |
enough | -0.016 |
once | -0.015 |
best | -0.013 |
again | -0.013 |
moments | -0.012 |
Token | you |
Token ID | 345 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
enough | +0.016 |
once | +0.015 |
best | +0.014 |
again | +0.013 |
moments | +0.013 |
Neg logit contributions | |
agons | -0.022 |
ramids | -0.021 |
disl | -0.020 |
alks | -0.020 |
ants | -0.020 |
Token | you |
Token ID | 345 |
Feature activation | +1.27e-02 |
Pos logit contributions | |
enough | +0.004 |
once | +0.004 |
best | +0.004 |
again | +0.004 |
more | +0.004 |
Neg logit contributions | |
agons | -0.007 |
ramids | -0.007 |
alks | -0.006 |
curb | -0.006 |
pier | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | +6.38e-03 |
Pos logit contributions | |
agons | +0.025 |
ramids | +0.023 |
alks | +0.023 |
disl | +0.022 |
ants | +0.022 |
Neg logit contributions | |
enough | -0.017 |
once | -0.016 |
best | -0.014 |
again | -0.014 |
moments | -0.013 |
Token | we |
Token ID | 356 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
agons | +0.011 |
ramids | +0.011 |
alks | +0.010 |
ants | +0.010 |
disl | +0.010 |
Neg logit contributions | |
enough | -0.010 |
once | -0.009 |
best | -0.008 |
again | -0.008 |
more | -0.008 |
Token | love |
Token ID | 1842 |
Feature activation | -7.48e-04 |
Pos logit contributions | |
enough | +0.010 |
once | +0.009 |
best | +0.008 |
again | +0.008 |
more | +0.007 |
Neg logit contributions | |
agons | -0.026 |
alks | -0.026 |
ramids | -0.025 |
ants | -0.025 |
pier | -0.025 |
Token | the |
Token ID | 262 |
Feature activation | +2.23e-02 |
Pos logit contributions | |
enough | +0.001 |
once | +0.001 |
best | +0.001 |
again | +0.001 |
moments | +0.001 |
Neg logit contributions | |
agons | -0.001 |
ramids | -0.001 |
alks | -0.001 |
disl | -0.001 |
ants | -0.001 |
Token | oct |
Token ID | 19318 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
agons | +0.029 |
ramids | +0.027 |
ants | +0.026 |
alks | +0.026 |
disl | +0.025 |
Neg logit contributions | |
enough | -0.044 |
once | -0.043 |
best | -0.041 |
again | -0.039 |
more | -0.038 |
Token | opus |
Token ID | 25790 |
Feature activation | +2.62e-02 |
Pos logit contributions | |
curb | +0.131 |
agons | +0.131 |
ramids | +0.128 |
issors | +0.126 |
disl | +0.124 |
Neg logit contributions | |
once | -0.176 |
again | -0.161 |
best | -0.158 |
enough | -0.152 |
ever | -0.143 |
Token | ." |
Token ID | 526 |
Feature activation | -4.72e-03 |
Pos logit contributions | |
agons | +0.044 |
ramids | +0.042 |
igator | +0.041 |
disl | +0.041 |
ants | +0.040 |
Neg logit contributions | |
enough | -0.038 |
once | -0.036 |
best | -0.033 |
moments | -0.032 |
again | -0.031 |
Token |
|
Token ID | 198 |
Feature activation | +4.97e-03 |
Pos logit contributions | |
enough | +0.005 |
once | +0.004 |
again | +0.004 |
best | +0.004 |
more | +0.003 |
Neg logit contributions | |
agons | -0.011 |
ramids | -0.010 |
ants | -0.010 |
disl | -0.010 |
pier | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | +7.47e-03 |
Pos logit contributions | |
agons | +0.013 |
ramids | +0.013 |
igator | +0.013 |
issors | +0.012 |
curb | +0.012 |
Neg logit contributions | |
once | -0.003 |
again | -0.003 |
enough | -0.002 |
best | -0.002 |
more | -0.002 |
Token | Mom |
Token ID | 29252 |
Feature activation | +8.20e-03 |
Pos logit contributions | |
agons | +0.014 |
ramids | +0.014 |
alks | +0.013 |
cases | +0.013 |
curb | +0.013 |
Neg logit contributions | |
once | -0.009 |
enough | -0.009 |
again | -0.009 |
more | -0.009 |
best | -0.008 |
Token | ily |
Token ID | 813 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
agons | +0.028 |
ants | +0.026 |
ramids | +0.026 |
issors | +0.025 |
disl | +0.025 |
Neg logit contributions | |
enough | -0.024 |
once | -0.023 |
best | -0.022 |
again | -0.021 |
moments | -0.020 |
Token | wanted |
Token ID | 2227 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
enough | +0.023 |
once | +0.022 |
best | +0.019 |
again | +0.019 |
more | +0.018 |
Neg logit contributions | |
agons | -0.038 |
ramids | -0.036 |
alks | -0.035 |
disl | -0.035 |
ants | -0.034 |
Token | to |
Token ID | 284 |
Feature activation | +2.36e-03 |
Pos logit contributions | |
ramids | +0.040 |
alks | +0.039 |
pier | +0.039 |
aro | +0.038 |
agons | +0.038 |
Neg logit contributions | |
enough | -0.018 |
once | -0.017 |
moments | -0.016 |
forever | -0.016 |
best | -0.016 |
Token | touch |
Token ID | 3638 |
Feature activation | +4.05e-03 |
Pos logit contributions | |
agons | +0.006 |
cases | +0.006 |
uffed | +0.006 |
angers | +0.006 |
aned | +0.005 |
Neg logit contributions | |
to | -0.002 |
more | -0.002 |
one | -0.002 |
once | -0.002 |
enough | -0.002 |
Token | the |
Token ID | 262 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
agons | +0.007 |
ramids | +0.007 |
alks | +0.007 |
aro | +0.007 |
ants | +0.007 |
Neg logit contributions | |
enough | -0.005 |
once | -0.005 |
best | -0.005 |
moments | -0.005 |
again | -0.004 |
Token | oct |
Token ID | 19318 |
Feature activation | +9.99e-02 |
Pos logit contributions | |
agons | +0.028 |
ramids | +0.024 |
ants | +0.024 |
disl | +0.024 |
issors | +0.024 |
Neg logit contributions | |
once | -0.046 |
enough | -0.046 |
best | -0.043 |
again | -0.042 |
more | -0.041 |
Token | opus |
Token ID | 25790 |
Feature activation | +3.15e-02 |
Pos logit contributions | |
agons | +0.132 |
curb | +0.120 |
ramids | +0.120 |
disl | +0.119 |
stret | +0.117 |
Neg logit contributions | |
once | -0.186 |
again | -0.172 |
enough | -0.169 |
best | -0.165 |
more | -0.151 |
Token | , |
Token ID | 11 |
Feature activation | +1.53e-02 |
Pos logit contributions | |
agons | +0.064 |
igator | +0.061 |
ants | +0.061 |
ramids | +0.061 |
uses | +0.060 |
Neg logit contributions | |
enough | -0.032 |
once | -0.030 |
moments | -0.028 |
best | -0.027 |
quality | -0.026 |
Token | but |
Token ID | 475 |
Feature activation | -8.21e-03 |
Pos logit contributions | |
agons | +0.028 |
ramids | +0.028 |
curb | +0.028 |
issors | +0.027 |
pier | +0.027 |
Neg logit contributions | |
once | -0.017 |
enough | -0.017 |
moments | -0.015 |
best | -0.015 |
again | -0.015 |
Token | her |
Token ID | 607 |
Feature activation | +3.73e-04 |
Pos logit contributions | |
enough | +0.009 |
once | +0.009 |
best | +0.008 |
again | +0.008 |
moments | +0.008 |
Neg logit contributions | |
agons | -0.016 |
ramids | -0.016 |
ants | -0.015 |
disl | -0.015 |
alks | -0.015 |
Token | mom |
Token ID | 1995 |
Feature activation | +5.95e-03 |
Pos logit contributions | |
agons | +0.001 |
ramids | +0.001 |
alks | +0.001 |
disl | +0.001 |
pier | +0.001 |
Neg logit contributions | |
enough | -0.001 |
once | -0.001 |
again | -0.001 |
best | -0.001 |
moments | -0.001 |
Token | Can |
Token ID | 6090 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
agons | +0.017 |
alks | +0.015 |
ramids | +0.015 |
disl | +0.015 |
ants | +0.015 |
Neg logit contributions | |
enough | -0.012 |
once | -0.012 |
again | -0.011 |
best | -0.011 |
more | -0.010 |
Token | we |
Token ID | 356 |
Feature activation | +2.55e-03 |
Pos logit contributions | |
once | +0.012 |
enough | +0.012 |
again | +0.011 |
best | +0.011 |
more | +0.010 |
Neg logit contributions | |
agons | -0.025 |
alks | -0.023 |
ramids | -0.022 |
curb | -0.022 |
ants | -0.022 |
Token | go |
Token ID | 467 |
Feature activation | -9.84e-03 |
Pos logit contributions | |
aned | +0.006 |
agons | +0.006 |
angers | +0.006 |
ants | +0.006 |
uffed | +0.006 |
Neg logit contributions | |
enough | -0.003 |
once | -0.003 |
best | -0.003 |
more | -0.002 |
to | -0.002 |
Token | see |
Token ID | 766 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
enough | +0.012 |
once | +0.012 |
again | +0.010 |
best | +0.010 |
more | +0.010 |
Neg logit contributions | |
agons | -0.020 |
ramids | -0.019 |
disl | -0.019 |
alks | -0.019 |
ants | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | +2.12e-02 |
Pos logit contributions | |
agons | +0.033 |
ramids | +0.033 |
pier | +0.031 |
ants | +0.031 |
disl | +0.031 |
Neg logit contributions | |
enough | -0.025 |
once | -0.025 |
best | -0.022 |
moments | -0.022 |
again | -0.021 |
Token | oct |
Token ID | 19318 |
Feature activation | +9.98e-02 |
Pos logit contributions | |
agons | +0.026 |
ramids | +0.023 |
ants | +0.022 |
disl | +0.022 |
alks | +0.022 |
Neg logit contributions | |
enough | -0.044 |
once | -0.043 |
best | -0.041 |
again | -0.040 |
more | -0.038 |
Token | opus |
Token ID | 25790 |
Feature activation | +2.39e-02 |
Pos logit contributions | |
agons | +0.132 |
ramids | +0.121 |
disl | +0.119 |
curb | +0.118 |
igator | +0.117 |
Neg logit contributions | |
once | -0.182 |
again | -0.170 |
enough | -0.168 |
best | -0.165 |
more | -0.150 |
Token | ?" |
Token ID | 1701 |
Feature activation | +7.97e-03 |
Pos logit contributions | |
agons | +0.034 |
ramids | +0.031 |
alks | +0.030 |
disl | +0.030 |
ants | +0.029 |
Neg logit contributions | |
enough | -0.044 |
once | -0.043 |
best | -0.040 |
again | -0.039 |
more | -0.037 |
Token | His |
Token ID | 2399 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
agons | +0.013 |
ramids | +0.012 |
pier | +0.012 |
rus | +0.011 |
alks | +0.011 |
Neg logit contributions | |
once | -0.013 |
enough | -0.013 |
again | -0.012 |
best | -0.012 |
more | -0.011 |
Token | mom |
Token ID | 1995 |
Feature activation | +6.06e-03 |
Pos logit contributions | |
agons | +0.037 |
ramids | +0.035 |
ants | +0.034 |
disl | +0.034 |
alks | +0.034 |
Neg logit contributions | |
enough | -0.030 |
once | -0.029 |
best | -0.028 |
again | -0.026 |
more | -0.025 |
Token | answered |
Token ID | 9373 |
Feature activation | +2.87e-03 |
Pos logit contributions | |
agons | +0.010 |
ramids | +0.009 |
rus | +0.009 |
ants | +0.009 |
disl | +0.009 |
Neg logit contributions | |
enough | -0.010 |
once | -0.009 |
best | -0.008 |
moments | -0.008 |
again | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | +6.05e-03 |
Pos logit contributions | |
agons | +0.005 |
ramids | +0.005 |
disl | +0.005 |
alks | +0.005 |
rolling | +0.005 |
Neg logit contributions | |
enough | -0.002 |
once | -0.002 |
again | -0.001 |
best | -0.001 |
more | -0.001 |
Token | I |
Token ID | 314 |
Feature activation | -2.33e-02 |
Pos logit contributions | |
agons | +0.010 |
ramids | +0.009 |
pier | +0.009 |
alks | +0.009 |
disl | +0.009 |
Neg logit contributions | |
enough | -0.010 |
once | -0.009 |
best | -0.009 |
again | -0.009 |
moments | -0.008 |
Token | do |
Token ID | 466 |
Feature activation | -7.89e-03 |
Pos logit contributions | |
enough | +0.034 |
once | +0.032 |
best | +0.030 |
again | +0.029 |
moments | +0.029 |
Neg logit contributions | |
agons | -0.041 |
disl | -0.038 |
ramids | -0.038 |
pier | -0.037 |
alks | -0.037 |
Token | !" |
Token ID | 2474 |
Feature activation | -7.71e-03 |
Pos logit contributions | |
enough | +0.011 |
once | +0.010 |
best | +0.009 |
again | +0.009 |
more | +0.009 |
Neg logit contributions | |
agons | -0.015 |
ramids | -0.014 |
alks | -0.014 |
curb | -0.014 |
ants | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | +2.41e-02 |
Pos logit contributions | |
enough | +0.010 |
once | +0.009 |
best | +0.008 |
again | +0.008 |
moments | +0.007 |
Neg logit contributions | |
agons | -0.015 |
ramids | -0.015 |
alks | -0.014 |
disl | -0.014 |
pier | -0.014 |
Token | oct |
Token ID | 19318 |
Feature activation | +9.93e-02 |
Pos logit contributions | |
agons | +0.033 |
ramids | +0.032 |
ants | +0.031 |
disl | +0.031 |
issors | +0.030 |
Neg logit contributions | |
once | -0.044 |
best | -0.044 |
enough | -0.043 |
more | -0.039 |
again | -0.039 |
Token | opus |
Token ID | 25790 |
Feature activation | +2.65e-02 |
Pos logit contributions | |
agons | +0.135 |
ramids | +0.130 |
curb | +0.124 |
disl | +0.122 |
issors | +0.120 |
Neg logit contributions | |
once | -0.173 |
again | -0.159 |
enough | -0.156 |
best | -0.153 |
ever | -0.138 |
Token | says |
Token ID | 1139 |
Feature activation | -7.03e-03 |
Pos logit contributions | |
agons | +0.055 |
ramids | +0.052 |
alks | +0.050 |
issors | +0.050 |
rolling | +0.050 |
Neg logit contributions | |
enough | -0.033 |
once | -0.032 |
again | -0.030 |
best | -0.029 |
more | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -2.56e-03 |
Pos logit contributions | |
enough | +0.010 |
once | +0.009 |
best | +0.009 |
moments | +0.008 |
quality | +0.008 |
Neg logit contributions | |
agons | -0.012 |
ramids | -0.012 |
alks | -0.012 |
ants | -0.011 |
curb | -0.011 |
Token | " |
Token ID | 366 |
Feature activation | +7.62e-03 |
Pos logit contributions | |
enough | +0.004 |
once | +0.003 |
best | +0.003 |
again | +0.003 |
moments | +0.003 |
Neg logit contributions | |
agons | -0.005 |
ramids | -0.004 |
alks | -0.004 |
pier | -0.004 |
disl | -0.004 |
Token | I |
Token ID | 40 |
Feature activation | +5.86e-03 |
Pos logit contributions | |
agons | +0.014 |
ramids | +0.014 |
disl | +0.013 |
curb | +0.013 |
pier | +0.013 |
Neg logit contributions | |
enough | -0.011 |
once | -0.010 |
best | -0.009 |
again | -0.009 |
moments | -0.008 |
Token | who |
Token ID | 508 |
Feature activation | +3.05e-02 |
Pos logit contributions | |
enough | +0.018 |
once | +0.018 |
best | +0.016 |
again | +0.016 |
moments | +0.016 |
Neg logit contributions | |
agons | -0.016 |
ramids | -0.015 |
pier | -0.015 |
disl | -0.014 |
curb | -0.014 |
Token | did |
Token ID | 750 |
Feature activation | +2.71e-02 |
Pos logit contributions | |
agons | +0.057 |
ramids | +0.054 |
alks | +0.052 |
disl | +0.052 |
pier | +0.052 |
Neg logit contributions | |
enough | -0.042 |
once | -0.039 |
best | -0.036 |
again | -0.035 |
moments | -0.033 |
Token | not |
Token ID | 407 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
agons | +0.057 |
ramids | +0.054 |
alks | +0.052 |
ants | +0.052 |
curb | +0.051 |
Neg logit contributions | |
enough | -0.034 |
once | -0.032 |
again | -0.029 |
best | -0.028 |
more | -0.027 |
Token | like |
Token ID | 588 |
Feature activation | +2.99e-02 |
Pos logit contributions | |
agons | +0.025 |
ramids | +0.023 |
alks | +0.022 |
disl | +0.022 |
ants | +0.022 |
Neg logit contributions | |
enough | -0.027 |
once | -0.026 |
best | -0.024 |
again | -0.024 |
more | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | +2.35e-02 |
Pos logit contributions | |
agons | +0.061 |
ramids | +0.057 |
rolling | +0.055 |
curb | +0.055 |
alks | +0.055 |
Neg logit contributions | |
enough | -0.038 |
once | -0.036 |
again | -0.032 |
best | -0.032 |
more | -0.030 |
Token | oct |
Token ID | 19318 |
Feature activation | +9.92e-02 |
Pos logit contributions | |
agons | +0.031 |
ramids | +0.029 |
alks | +0.028 |
pier | +0.027 |
disl | +0.027 |
Neg logit contributions | |
enough | -0.045 |
once | -0.043 |
best | -0.040 |
again | -0.040 |
moments | -0.038 |
Token | opus |
Token ID | 25790 |
Feature activation | +3.05e-02 |
Pos logit contributions | |
agons | +0.126 |
ramids | +0.120 |
curb | +0.119 |
issors | +0.115 |
disl | +0.114 |
Neg logit contributions | |
once | -0.185 |
enough | -0.172 |
again | -0.171 |
best | -0.166 |
moments | -0.151 |
Token | . |
Token ID | 13 |
Feature activation | -1.18e-04 |
Pos logit contributions | |
agons | +0.060 |
ramids | +0.057 |
disl | +0.055 |
alks | +0.055 |
ants | +0.055 |
Neg logit contributions | |
enough | -0.038 |
once | -0.036 |
best | -0.032 |
again | -0.031 |
moments | -0.030 |
Token | It |
Token ID | 632 |
Feature activation | +2.25e-03 |
Pos logit contributions | |
enough | +0.000 |
once | +0.000 |
again | +0.000 |
best | +0.000 |
more | +0.000 |
Neg logit contributions | |
agons | -0.000 |
ramids | -0.000 |
ants | -0.000 |
rolling | -0.000 |
rus | -0.000 |
Token | was |
Token ID | 373 |
Feature activation | -3.87e-03 |
Pos logit contributions | |
agons | +0.005 |
ramids | +0.005 |
alks | +0.005 |
ants | +0.004 |
disl | +0.004 |
Neg logit contributions | |
enough | -0.003 |
once | -0.002 |
best | -0.002 |
again | -0.002 |
more | -0.002 |
Token | a |
Token ID | 257 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
enough | +0.005 |
once | +0.005 |
again | +0.005 |
best | +0.005 |
moments | +0.005 |
Neg logit contributions | |
agons | -0.007 |
ramids | -0.007 |
alks | -0.006 |
pier | -0.006 |
disl | -0.006 |
Token | okay |
Token ID | 8788 |
Feature activation | -3.36e-02 |
Pos logit contributions | |
enough | +0.018 |
once | +0.017 |
best | +0.015 |
again | +0.014 |
more | +0.013 |
Neg logit contributions | |
agons | -0.042 |
ramids | -0.041 |
alks | -0.039 |
disl | -0.039 |
pier | -0.039 |
Token | , |
Token ID | 11 |
Feature activation | +7.10e-03 |
Pos logit contributions | |
enough | +0.045 |
once | +0.044 |
best | +0.041 |
moments | +0.039 |
again | +0.036 |
Neg logit contributions | |
agons | -0.058 |
disl | -0.056 |
ramids | -0.056 |
ants | -0.055 |
alks | -0.054 |
Token | crab |
Token ID | 32202 |
Feature activation | -1.68e-02 |
Pos logit contributions | |
agons | +0.012 |
ramids | +0.011 |
disl | +0.011 |
pier | +0.011 |
alks | +0.011 |
Neg logit contributions | |
enough | -0.011 |
once | -0.010 |
best | -0.009 |
again | -0.009 |
moments | -0.009 |
Token | ," |
Token ID | 553 |
Feature activation | -9.98e-03 |
Pos logit contributions | |
enough | +0.023 |
once | +0.023 |
best | +0.021 |
moments | +0.020 |
again | +0.020 |
Neg logit contributions | |
agons | -0.030 |
ramids | -0.028 |
disl | -0.028 |
ants | -0.027 |
igator | -0.027 |
Token | the |
Token ID | 262 |
Feature activation | +2.87e-02 |
Pos logit contributions | |
enough | +0.013 |
once | +0.013 |
best | +0.012 |
again | +0.011 |
moments | +0.011 |
Neg logit contributions | |
agons | -0.018 |
ramids | -0.018 |
alks | -0.017 |
disl | -0.017 |
pier | -0.017 |
Token | oct |
Token ID | 19318 |
Feature activation | +9.92e-02 |
Pos logit contributions | |
agons | +0.045 |
ramids | +0.043 |
ants | +0.041 |
issors | +0.040 |
alks | +0.040 |
Neg logit contributions | |
enough | -0.049 |
once | -0.047 |
best | -0.045 |
again | -0.044 |
more | -0.042 |
Token | opus |
Token ID | 25790 |
Feature activation | +2.99e-02 |
Pos logit contributions | |
agons | +0.132 |
ramids | +0.123 |
alks | +0.112 |
issors | +0.112 |
curb | +0.111 |
Neg logit contributions | |
once | -0.186 |
again | -0.174 |
enough | -0.172 |
best | -0.164 |
moments | -0.152 |
Token | said |
Token ID | 531 |
Feature activation | +3.55e-03 |
Pos logit contributions | |
agons | +0.050 |
ramids | +0.048 |
disl | +0.047 |
igator | +0.046 |
ants | +0.046 |
Neg logit contributions | |
enough | -0.044 |
once | -0.041 |
best | -0.039 |
moments | -0.036 |
again | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | -4.48e-03 |
Pos logit contributions | |
agons | +0.006 |
ramids | +0.006 |
alks | +0.006 |
rus | +0.006 |
ants | +0.005 |
Neg logit contributions | |
enough | -0.005 |
once | -0.005 |
best | -0.004 |
moments | -0.004 |
quality | -0.004 |
Token | It |
Token ID | 632 |
Feature activation | +5.13e-03 |
Pos logit contributions | |
enough | +0.007 |
once | +0.006 |
best | +0.006 |
again | +0.006 |
moments | +0.005 |
Neg logit contributions | |
agons | -0.008 |
ramids | -0.007 |
alks | -0.007 |
pier | -0.007 |
curb | -0.007 |
Token | was |
Token ID | 373 |
Feature activation | -5.33e-03 |
Pos logit contributions | |
agons | +0.009 |
ramids | +0.009 |
disl | +0.009 |
alks | +0.008 |
pier | +0.008 |
Neg logit contributions | |
enough | -0.007 |
once | -0.007 |
best | -0.006 |
again | -0.006 |
moments | -0.006 |
Token | this |
Token ID | 428 |
Feature activation | -8.57e-03 |
Pos logit contributions | |
enough | +0.002 |
once | +0.002 |
best | +0.002 |
again | +0.002 |
more | +0.001 |
Neg logit contributions | |
agons | -0.003 |
ramids | -0.003 |
alks | -0.003 |
disl | -0.003 |
ants | -0.003 |
Token | mess |
Token ID | 2085 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
enough | +0.013 |
once | +0.013 |
again | +0.012 |
best | +0.012 |
more | +0.011 |
Neg logit contributions | |
agons | -0.015 |
ramids | -0.015 |
rolling | -0.014 |
ants | -0.014 |
rolled | -0.014 |
Token | and |
Token ID | 290 |
Feature activation | +6.85e-03 |
Pos logit contributions | |
agons | +0.036 |
ramids | +0.034 |
ants | +0.033 |
disl | +0.033 |
alks | +0.033 |
Neg logit contributions | |
enough | -0.015 |
once | -0.014 |
best | -0.012 |
again | -0.012 |
moments | -0.011 |
Token | fix |
Token ID | 4259 |
Feature activation | +1.79e-02 |
Pos logit contributions | |
agons | +0.011 |
ramids | +0.010 |
disl | +0.010 |
curb | +0.010 |
pier | +0.009 |
Neg logit contributions | |
enough | -0.011 |
once | -0.011 |
best | -0.010 |
again | -0.010 |
moments | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | +9.70e-03 |
Pos logit contributions | |
agons | +0.034 |
ramids | +0.032 |
ants | +0.031 |
alks | +0.031 |
disl | +0.031 |
Neg logit contributions | |
enough | -0.023 |
once | -0.022 |
best | -0.019 |
moments | -0.019 |
again | -0.019 |
Token | oct |
Token ID | 19318 |
Feature activation | +9.76e-02 |
Pos logit contributions | |
agons | +0.014 |
ramids | +0.013 |
ants | +0.013 |
rolling | +0.013 |
rolled | +0.013 |
Neg logit contributions | |
once | -0.017 |
enough | -0.017 |
best | -0.017 |
again | -0.016 |
more | -0.015 |
Token | opus |
Token ID | 25790 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
agons | +0.126 |
ramids | +0.124 |
curb | +0.123 |
disl | +0.120 |
rolled | +0.118 |
Neg logit contributions | |
once | -0.173 |
again | -0.162 |
best | -0.156 |
enough | -0.153 |
forever | -0.137 |
Token | . |
Token ID | 13 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
agons | +0.057 |
ramids | +0.055 |
alks | +0.053 |
rolling | +0.053 |
pier | +0.052 |
Neg logit contributions | |
enough | -0.027 |
once | -0.026 |
again | -0.024 |
best | -0.023 |
more | -0.022 |
Token | Maybe |
Token ID | 6674 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
enough | +0.014 |
once | +0.014 |
best | +0.012 |
again | +0.012 |
moments | +0.011 |
Neg logit contributions | |
agons | -0.020 |
ramids | -0.019 |
alks | -0.018 |
ants | -0.018 |
pier | -0.018 |
Token | you |
Token ID | 345 |
Feature activation | +7.04e-03 |
Pos logit contributions | |
enough | +0.017 |
once | +0.017 |
moments | +0.016 |
best | +0.015 |
osa | +0.015 |
Neg logit contributions | |
ramids | -0.019 |
agons | -0.018 |
alks | -0.018 |
ants | -0.018 |
pier | -0.018 |
Token | can |
Token ID | 460 |
Feature activation | +8.97e-03 |
Pos logit contributions | |
agons | +0.015 |
ramids | +0.014 |
alks | +0.014 |
disl | +0.014 |
ants | +0.014 |
Neg logit contributions | |
enough | -0.008 |
once | -0.008 |
again | -0.007 |
best | -0.007 |
more | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
agons | +0.025 |
ramids | +0.023 |
ants | +0.023 |
disl | +0.022 |
alks | +0.022 |
Neg logit contributions | |
enough | -0.015 |
once | -0.015 |
moments | -0.013 |
best | -0.013 |
quality | -0.012 |
Token | duck |
Token ID | 22045 |
Feature activation | -8.41e-03 |
Pos logit contributions | |
agons | +0.034 |
ramids | +0.032 |
alks | +0.031 |
disl | +0.031 |
pier | +0.031 |
Neg logit contributions | |
enough | -0.029 |
once | -0.028 |
best | -0.026 |
again | -0.025 |
moments | -0.024 |
Token | ," |
Token ID | 553 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
enough | +0.011 |
once | +0.010 |
best | +0.009 |
again | +0.009 |
more | +0.009 |
Neg logit contributions | |
agons | -0.017 |
ramids | -0.016 |
alks | -0.015 |
disl | -0.015 |
ants | -0.015 |
Token | said |
Token ID | 531 |
Feature activation | +8.25e-03 |
Pos logit contributions | |
enough | +0.019 |
once | +0.018 |
best | +0.017 |
again | +0.016 |
moments | +0.015 |
Neg logit contributions | |
agons | -0.021 |
ramids | -0.020 |
disl | -0.020 |
pier | -0.020 |
alks | -0.020 |
Token | the |
Token ID | 262 |
Feature activation | +1.86e-02 |
Pos logit contributions | |
agons | +0.016 |
ramids | +0.015 |
alks | +0.014 |
disl | +0.014 |
ants | +0.014 |
Neg logit contributions | |
enough | -0.011 |
once | -0.010 |
best | -0.010 |
again | -0.009 |
moments | -0.009 |
Token | oct |
Token ID | 19318 |
Feature activation | +9.75e-02 |
Pos logit contributions | |
agons | +0.037 |
ramids | +0.036 |
ants | +0.035 |
issors | +0.035 |
cases | +0.035 |
Neg logit contributions | |
once | -0.023 |
enough | -0.023 |
best | -0.022 |
again | -0.020 |
more | -0.019 |
Token | opus |
Token ID | 25790 |
Feature activation | +2.43e-02 |
Pos logit contributions | |
agons | +0.124 |
ramids | +0.118 |
disl | +0.110 |
curb | +0.110 |
alks | +0.108 |
Neg logit contributions | |
once | -0.185 |
enough | -0.173 |
again | -0.172 |
best | -0.166 |
more | -0.152 |
Token | . |
Token ID | 13 |
Feature activation | -3.47e-03 |
Pos logit contributions | |
agons | +0.049 |
ramids | +0.046 |
alks | +0.045 |
disl | +0.045 |
ants | +0.045 |
Neg logit contributions | |
enough | -0.030 |
once | -0.028 |
best | -0.026 |
again | -0.025 |
moments | -0.023 |
Token | " |
Token ID | 366 |
Feature activation | +5.72e-03 |
Pos logit contributions | |
enough | +0.005 |
once | +0.004 |
best | +0.004 |
again | +0.004 |
more | +0.004 |
Neg logit contributions | |
agons | -0.007 |
ramids | -0.006 |
alks | -0.006 |
disl | -0.006 |
ants | -0.006 |
Token | I |
Token ID | 40 |
Feature activation | +7.69e-03 |
Pos logit contributions | |
agons | +0.013 |
ramids | +0.012 |
alks | +0.012 |
rolling | +0.012 |
pier | +0.012 |
Neg logit contributions | |
once | -0.006 |
enough | -0.005 |
again | -0.005 |
best | -0.005 |
more | -0.004 |
Token | didn |
Token ID | 1422 |
Feature activation | +3.48e-02 |
Pos logit contributions | |
agons | +0.015 |
ramids | +0.015 |
alks | +0.014 |
ants | +0.014 |
pier | +0.014 |
Neg logit contributions | |
enough | -0.010 |
once | -0.010 |
best | -0.008 |
again | -0.008 |
more | -0.008 |
Token | felt |
Token ID | 2936 |
Feature activation | +9.20e-03 |
Pos logit contributions | |
agons | +0.007 |
ramids | +0.007 |
alks | +0.006 |
ants | +0.006 |
issors | +0.006 |
Neg logit contributions | |
enough | -0.007 |
once | -0.006 |
again | -0.006 |
best | -0.006 |
more | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | +2.35e-02 |
Pos logit contributions | |
agons | +0.018 |
ramids | +0.017 |
ants | +0.017 |
alks | +0.016 |
issors | +0.016 |
Neg logit contributions | |
enough | -0.013 |
once | -0.012 |
best | -0.011 |
again | -0.011 |
more | -0.011 |
Token | familiar |
Token ID | 5385 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
agons | +0.034 |
curb | +0.033 |
pier | +0.031 |
igator | +0.031 |
disl | +0.031 |
Neg logit contributions | |
enough | -0.039 |
once | -0.036 |
again | -0.034 |
best | -0.032 |
forever | -0.031 |
Token | feeling |
Token ID | 4203 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
disl | +0.028 |
agons | +0.028 |
curb | +0.027 |
alks | +0.027 |
pier | +0.026 |
Neg logit contributions | |
enough | -0.026 |
once | -0.025 |
best | -0.023 |
again | -0.021 |
more | -0.020 |
Token | of |
Token ID | 286 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
agons | +0.024 |
ramids | +0.023 |
alks | +0.022 |
ants | +0.022 |
disl | +0.022 |
Neg logit contributions | |
enough | -0.012 |
once | -0.012 |
best | -0.010 |
again | -0.010 |
more | -0.010 |
Token | impat |
Token ID | 28189 |
Feature activation | +9.68e-02 |
Pos logit contributions | |
agons | +0.021 |
disl | +0.021 |
curb | +0.020 |
alks | +0.020 |
aro | +0.019 |
Neg logit contributions | |
enough | -0.016 |
once | -0.015 |
best | -0.014 |
again | -0.014 |
chapter | -0.012 |
Token | ience |
Token ID | 1240 |
Feature activation | -9.43e-03 |
Pos logit contributions | |
agons | +0.162 |
ramids | +0.154 |
curb | +0.149 |
pier | +0.148 |
alks | +0.147 |
Neg logit contributions | |
enough | -0.158 |
once | -0.149 |
again | -0.140 |
best | -0.135 |
moments | -0.129 |
Token | . |
Token ID | 13 |
Feature activation | -1.14e-02 |
Pos logit contributions | |
enough | +0.012 |
once | +0.012 |
best | +0.011 |
moments | +0.010 |
more | +0.010 |
Neg logit contributions | |
agons | -0.016 |
curb | -0.015 |
ramids | -0.015 |
alks | -0.014 |
pier | -0.014 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | +6.97e-03 |
Pos logit contributions | |
enough | +0.018 |
once | +0.016 |
best | +0.016 |
again | +0.015 |
more | +0.015 |
Neg logit contributions | |
agons | -0.018 |
ramids | -0.018 |
alks | -0.018 |
disl | -0.018 |
curb | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | -5.33e-03 |
Pos logit contributions | |
alks | +0.012 |
ramids | +0.012 |
agons | +0.012 |
aro | +0.012 |
pier | +0.011 |
Neg logit contributions | |
once | -0.008 |
enough | -0.007 |
best | -0.007 |
quality | -0.007 |
chapter | -0.006 |
Token | she |
Token ID | 673 |
Feature activation | -5.20e-03 |
Pos logit contributions | |
enough | +0.007 |
once | +0.006 |
best | +0.006 |
again | +0.005 |
moments | +0.005 |
Neg logit contributions | |
agons | -0.009 |
disl | -0.009 |
curb | -0.009 |
igator | -0.009 |
ramids | -0.009 |
Token | 't |
Token ID | 470 |
Feature activation | +2.81e-02 |
Pos logit contributions | |
agons | +0.060 |
ount | +0.056 |
stret | +0.054 |
alks | +0.054 |
ramids | +0.053 |
Neg logit contributions | |
once | -0.034 |
again | -0.032 |
enough | -0.031 |
one | -0.030 |
best | -0.028 |
Token | I |
Token ID | 314 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
agons | +0.067 |
ramids | +0.062 |
ric | +0.061 |
alks | +0.061 |
ants | +0.061 |
Neg logit contributions | |
enough | -0.028 |
again | -0.028 |
once | -0.028 |
more | -0.024 |
one | -0.022 |
Token | play |
Token ID | 711 |
Feature activation | +2.81e-02 |
Pos logit contributions | |
agons | +0.050 |
ants | +0.048 |
ramids | +0.048 |
cks | +0.048 |
pans | +0.047 |
Neg logit contributions | |
again | -0.031 |
once | -0.030 |
enough | -0.028 |
more | -0.027 |
best | -0.027 |
Token | with |
Token ID | 351 |
Feature activation | +2.60e-02 |
Pos logit contributions | |
agons | +0.068 |
ramids | +0.065 |
alks | +0.063 |
ants | +0.063 |
disl | +0.062 |
Neg logit contributions | |
enough | -0.024 |
once | -0.023 |
best | -0.020 |
again | -0.020 |
more | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
agons | +0.062 |
rolling | +0.061 |
curb | +0.058 |
ramids | +0.058 |
rolled | +0.057 |
Neg logit contributions | |
enough | -0.026 |
once | -0.025 |
again | -0.025 |
more | -0.024 |
best | -0.021 |
Token | oct |
Token ID | 19318 |
Feature activation | +9.66e-02 |
Pos logit contributions | |
agons | +0.015 |
ramids | +0.014 |
alks | +0.013 |
ants | +0.013 |
disl | +0.013 |
Neg logit contributions | |
enough | -0.028 |
once | -0.027 |
best | -0.026 |
again | -0.026 |
more | -0.025 |
Token | opus |
Token ID | 25790 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
agons | +0.121 |
ramids | +0.117 |
curb | +0.115 |
disl | +0.110 |
rolled | +0.108 |
Neg logit contributions | |
once | -0.180 |
again | -0.168 |
enough | -0.167 |
best | -0.164 |
more | -0.149 |
Token | , |
Token ID | 11 |
Feature activation | +2.27e-02 |
Pos logit contributions | |
agons | +0.032 |
ramids | +0.030 |
alks | +0.028 |
ants | +0.028 |
curb | +0.027 |
Neg logit contributions | |
enough | -0.047 |
once | -0.046 |
again | -0.043 |
best | -0.043 |
more | -0.041 |
Token | too |
Token ID | 1165 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
agons | +0.045 |
ramids | +0.043 |
alks | +0.041 |
disl | +0.041 |
ants | +0.041 |
Neg logit contributions | |
enough | -0.029 |
once | -0.028 |
best | -0.025 |
again | -0.024 |
more | -0.023 |
Token | ?" |
Token ID | 1701 |
Feature activation | -9.82e-04 |
Pos logit contributions | |
enough | +0.024 |
once | +0.023 |
best | +0.021 |
again | +0.021 |
more | +0.021 |
Neg logit contributions | |
agons | -0.016 |
ramids | -0.015 |
alks | -0.014 |
rus | -0.014 |
ants | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
enough | +0.001 |
once | +0.001 |
best | +0.001 |
again | +0.001 |
more | +0.001 |
Neg logit contributions | |
agons | -0.002 |
ramids | -0.002 |
alks | -0.002 |
disl | -0.002 |
ants | -0.002 |
Token | " |
Token ID | 1 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
agons | +0.020 |
ramids | +0.019 |
alks | +0.019 |
rus | +0.019 |
disl | +0.018 |
Neg logit contributions | |
enough | -0.010 |
once | -0.010 |
again | -0.008 |
more | -0.008 |
best | -0.008 |
Token | O |
Token ID | 46 |
Feature activation | -2.50e-04 |
Pos logit contributions | |
agons | +0.023 |
ramids | +0.022 |
disl | +0.021 |
alks | +0.021 |
pier | +0.020 |
Neg logit contributions | |
enough | -0.018 |
once | -0.017 |
best | -0.016 |
again | -0.016 |
more | -0.015 |
Token | uch |
Token ID | 794 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
enough | +0.000 |
once | +0.000 |
best | +0.000 |
again | +0.000 |
moments | +0.000 |
Neg logit contributions | |
agons | -0.000 |
ramids | -0.000 |
disl | -0.000 |
alks | -0.000 |
ants | -0.000 |
Token | !" |
Token ID | 2474 |
Feature activation | -2.48e-03 |
Pos logit contributions | |
enough | +0.019 |
once | +0.018 |
best | +0.016 |
again | +0.016 |
moments | +0.016 |
Neg logit contributions | |
agons | -0.023 |
ramids | -0.022 |
alks | -0.021 |
ants | -0.021 |
disl | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | +2.65e-02 |
Pos logit contributions | |
enough | +0.003 |
once | +0.003 |
again | +0.002 |
best | +0.002 |
more | +0.002 |
Neg logit contributions | |
agons | -0.006 |
ramids | -0.005 |
rolling | -0.005 |
ants | -0.005 |
alks | -0.005 |
Token | oct |
Token ID | 19318 |
Feature activation | +9.60e-02 |
Pos logit contributions | |
agons | +0.036 |
ramids | +0.033 |
issors | +0.032 |
ants | +0.032 |
disl | +0.032 |
Neg logit contributions | |
once | -0.051 |
enough | -0.050 |
best | -0.049 |
again | -0.046 |
more | -0.045 |
Token | opus |
Token ID | 25790 |
Feature activation | +2.80e-02 |
Pos logit contributions | |
agons | +0.126 |
ramids | +0.117 |
issors | +0.108 |
curb | +0.108 |
alks | +0.106 |
Neg logit contributions | |
once | -0.183 |
enough | -0.170 |
again | -0.170 |
best | -0.163 |
moments | -0.151 |
Token | said |
Token ID | 531 |
Feature activation | +4.06e-03 |
Pos logit contributions | |
agons | +0.046 |
ramids | +0.043 |
alks | +0.041 |
ants | +0.041 |
disl | +0.041 |
Neg logit contributions | |
enough | -0.045 |
once | -0.043 |
best | -0.040 |
again | -0.038 |
moments | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | -5.80e-04 |
Pos logit contributions | |
agons | +0.007 |
ramids | +0.007 |
alks | +0.007 |
ants | +0.007 |
curb | +0.007 |
Neg logit contributions | |
enough | -0.005 |
once | -0.005 |
best | -0.005 |
moments | -0.004 |
again | -0.004 |
Token | It |
Token ID | 632 |
Feature activation | +3.07e-03 |
Pos logit contributions | |
enough | +0.001 |
once | +0.001 |
best | +0.001 |
again | +0.001 |
more | +0.001 |
Neg logit contributions | |
agons | -0.001 |
ramids | -0.001 |
disl | -0.001 |
alks | -0.001 |
ants | -0.001 |
Token | spoke |
Token ID | 5158 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
agons | +0.006 |
ramids | +0.005 |
alks | +0.005 |
issors | +0.005 |
ants | +0.005 |
Neg logit contributions | |
enough | -0.004 |
once | -0.004 |
again | -0.004 |
best | -0.004 |
more | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | +7.53e-03 |
Pos logit contributions | |
agons | +0.020 |
ramids | +0.019 |
disl | +0.019 |
ants | +0.019 |
curb | +0.019 |
Neg logit contributions | |
enough | -0.010 |
once | -0.010 |
best | -0.009 |
moments | -0.008 |
again | -0.008 |
Token | strange |
Token ID | 6283 |
Feature activation | +5.99e-03 |
Pos logit contributions | |
agons | +0.014 |
ramids | +0.013 |
alks | +0.012 |
disl | +0.012 |
ants | +0.012 |
Neg logit contributions | |
enough | -0.011 |
once | -0.010 |
best | -0.009 |
again | -0.009 |
more | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | +2.68e-02 |
Pos logit contributions | |
agons | +0.010 |
curb | +0.009 |
pier | +0.009 |
disl | +0.009 |
ramids | +0.009 |
Neg logit contributions | |
enough | -0.009 |
once | -0.009 |
best | -0.008 |
again | -0.008 |
moments | -0.007 |
Token | raspberry |
Token ID | 38973 |
Feature activation | -4.33e-04 |
Pos logit contributions | |
agons | +0.037 |
curb | +0.036 |
disl | +0.036 |
igator | +0.035 |
ramids | +0.035 |
Neg logit contributions | |
enough | -0.047 |
once | -0.044 |
best | -0.041 |
again | -0.040 |
moments | -0.039 |
Token | - |
Token ID | 12 |
Feature activation | +3.87e-02 |
Pos logit contributions | |
enough | +0.001 |
once | +0.001 |
best | +0.001 |
again | +0.000 |
moments | +0.000 |
Neg logit contributions | |
agons | -0.001 |
ramids | -0.001 |
alks | -0.001 |
ants | -0.001 |
curb | -0.001 |
Token | col |
Token ID | 4033 |
Feature activation | +9.54e-02 |
Pos logit contributions | |
agons | +0.057 |
ramids | +0.051 |
pier | +0.050 |
rolling | +0.050 |
alks | +0.049 |
Neg logit contributions | |
enough | -0.070 |
once | -0.068 |
again | -0.062 |
best | -0.062 |
more | -0.060 |
Token | oured |
Token ID | 8167 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
igator | +0.140 |
agons | +0.135 |
pier | +0.131 |
disl | +0.124 |
aro | +0.123 |
Neg logit contributions | |
enough | -0.185 |
once | -0.161 |
again | -0.160 |
eter | -0.137 |
joy | -0.133 |
Token | object |
Token ID | 2134 |
Feature activation | -1.22e-03 |
Pos logit contributions | |
agons | +0.024 |
alks | +0.023 |
ramids | +0.022 |
ants | +0.022 |
igator | +0.022 |
Neg logit contributions | |
enough | -0.023 |
once | -0.022 |
best | -0.020 |
again | -0.019 |
moments | -0.019 |
Token | hiding |
Token ID | 11816 |
Feature activation | -2.76e-02 |
Pos logit contributions | |
enough | +0.001 |
once | +0.001 |
best | +0.001 |
again | +0.001 |
moments | +0.001 |
Neg logit contributions | |
agons | -0.003 |
ramids | -0.002 |
alks | -0.002 |
ants | -0.002 |
disl | -0.002 |
Token | under |
Token ID | 739 |
Feature activation | +4.96e-03 |
Pos logit contributions | |
enough | +0.044 |
once | +0.042 |
best | +0.040 |
again | +0.038 |
moments | +0.037 |
Neg logit contributions | |
agons | -0.043 |
ramids | -0.041 |
alks | -0.040 |
ants | -0.039 |
pier | -0.039 |
Token | a |
Token ID | 257 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
agons | +0.011 |
ramids | +0.011 |
alks | +0.010 |
disl | +0.010 |
ants | +0.010 |
Neg logit contributions | |
enough | -0.005 |
once | -0.005 |
best | -0.004 |
again | -0.004 |
moments | -0.004 |
Token | each |
Token ID | 1123 |
Feature activation | +4.74e-03 |
Pos logit contributions | |
agons | +0.011 |
ramids | +0.010 |
curb | +0.010 |
issors | +0.010 |
stret | +0.010 |
Neg logit contributions | |
enough | -0.010 |
best | -0.009 |
once | -0.009 |
moments | -0.008 |
moment | -0.008 |
Token | other |
Token ID | 584 |
Feature activation | -7.20e-03 |
Pos logit contributions | |
agons | +0.008 |
ramids | +0.007 |
igator | +0.007 |
curb | +0.007 |
ants | +0.007 |
Neg logit contributions | |
enough | -0.007 |
once | -0.007 |
forever | -0.006 |
best | -0.006 |
again | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | +5.91e-03 |
Pos logit contributions | |
enough | +0.008 |
once | +0.008 |
best | +0.007 |
moments | +0.007 |
quality | +0.006 |
Neg logit contributions | |
agons | -0.014 |
ramids | -0.013 |
curb | -0.013 |
pier | -0.013 |
ants | -0.013 |
Token | thanked |
Token ID | 26280 |
Feature activation | +1.80e-02 |
Pos logit contributions | |
disl | +0.007 |
pier | +0.007 |
agons | +0.007 |
blamed | +0.007 |
lee | +0.007 |
Neg logit contributions | |
once | -0.011 |
enough | -0.010 |
best | -0.010 |
chapter | -0.009 |
again | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
agons | +0.030 |
ramids | +0.029 |
alks | +0.027 |
curb | +0.027 |
issors | +0.027 |
Neg logit contributions | |
enough | -0.027 |
once | -0.025 |
best | -0.024 |
moments | -0.022 |
again | -0.022 |
Token | oct |
Token ID | 19318 |
Feature activation | +9.52e-02 |
Pos logit contributions | |
agons | +0.025 |
ramids | +0.024 |
ants | +0.024 |
alks | +0.023 |
issors | +0.023 |
Neg logit contributions | |
enough | -0.015 |
once | -0.014 |
best | -0.013 |
again | -0.013 |
more | -0.012 |
Token | opus |
Token ID | 25790 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
ramids | +0.120 |
agons | +0.120 |
curb | +0.114 |
disl | +0.112 |
alks | +0.111 |
Neg logit contributions | |
once | -0.170 |
again | -0.165 |
enough | -0.162 |
best | -0.159 |
more | -0.146 |
Token | . |
Token ID | 13 |
Feature activation | -9.28e-03 |
Pos logit contributions | |
agons | +0.039 |
ramids | +0.037 |
alks | +0.036 |
ants | +0.035 |
disl | +0.035 |
Neg logit contributions | |
enough | -0.024 |
once | -0.022 |
best | -0.020 |
moments | -0.019 |
again | -0.019 |
Token | They |
Token ID | 1119 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
enough | +0.012 |
once | +0.012 |
again | +0.011 |
best | +0.010 |
more | +0.010 |
Neg logit contributions | |
agons | -0.019 |
ants | -0.017 |
igator | -0.017 |
rus | -0.017 |
rolling | -0.017 |
Token | learned |
Token ID | 4499 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
agons | +0.036 |
ramids | +0.034 |
alks | +0.033 |
ants | +0.033 |
disl | +0.032 |
Neg logit contributions | |
enough | -0.027 |
once | -0.025 |
best | -0.023 |
again | -0.023 |
more | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
enough | +0.013 |
once | +0.012 |
best | +0.011 |
again | +0.011 |
moments | +0.010 |
Neg logit contributions | |
agons | -0.025 |
ramids | -0.024 |
disl | -0.023 |
alks | -0.023 |
pier | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | -3.89e-04 |
Pos logit contributions | |
agons | +0.001 |
ramids | +0.001 |
alks | +0.001 |
curb | +0.001 |
disl | +0.001 |
Neg logit contributions | |
enough | -0.001 |
once | -0.001 |
best | -0.001 |
moments | -0.001 |
again | -0.000 |
Token | duck |
Token ID | 22045 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
once | +0.001 |
enough | +0.001 |
quality | +0.001 |
moments | +0.001 |
best | +0.001 |
Neg logit contributions | |
pier | -0.000 |
aro | -0.000 |
disl | -0.000 |
agons | -0.000 |
dock | -0.000 |
Token | ," |
Token ID | 553 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
enough | +0.016 |
once | +0.016 |
best | +0.014 |
again | +0.014 |
moments | +0.013 |
Neg logit contributions | |
agons | -0.025 |
ramids | -0.023 |
disl | -0.022 |
ants | -0.022 |
alks | -0.022 |
Token | said |
Token ID | 531 |
Feature activation | +3.18e-03 |
Pos logit contributions | |
enough | +0.027 |
once | +0.025 |
best | +0.024 |
again | +0.022 |
moments | +0.022 |
Neg logit contributions | |
agons | -0.026 |
ramids | -0.025 |
disl | -0.025 |
pier | -0.025 |
aro | -0.025 |
Token | the |
Token ID | 262 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
agons | +0.006 |
ramids | +0.006 |
alks | +0.006 |
pier | +0.005 |
ants | +0.005 |
Neg logit contributions | |
enough | -0.004 |
once | -0.004 |
best | -0.004 |
again | -0.004 |
moments | -0.003 |
Token | oct |
Token ID | 19318 |
Feature activation | +9.51e-02 |
Pos logit contributions | |
agons | +0.027 |
ramids | +0.026 |
ants | +0.025 |
issors | +0.025 |
cases | +0.025 |
Neg logit contributions | |
once | -0.026 |
enough | -0.025 |
best | -0.025 |
again | -0.023 |
more | -0.023 |
Token | opus |
Token ID | 25790 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
agons | +0.123 |
ramids | +0.120 |
disl | +0.112 |
curb | +0.111 |
alks | +0.110 |
Neg logit contributions | |
once | -0.176 |
again | -0.162 |
enough | -0.160 |
best | -0.155 |
more | -0.142 |
Token | . |
Token ID | 13 |
Feature activation | -8.86e-03 |
Pos logit contributions | |
agons | +0.039 |
ramids | +0.038 |
ants | +0.037 |
igator | +0.037 |
disl | +0.037 |
Neg logit contributions | |
enough | -0.029 |
once | -0.027 |
best | -0.025 |
moments | -0.024 |
again | -0.023 |
Token | " |
Token ID | 366 |
Feature activation | +3.19e-03 |
Pos logit contributions | |
enough | +0.014 |
once | +0.013 |
best | +0.013 |
moments | +0.012 |
again | +0.012 |
Neg logit contributions | |
ramids | -0.014 |
agons | -0.014 |
alks | -0.013 |
curb | -0.013 |
pier | -0.013 |
Token | Next |
Token ID | 10019 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
agons | +0.006 |
ramids | +0.005 |
disl | +0.005 |
ants | +0.005 |
curb | +0.005 |
Neg logit contributions | |
enough | -0.005 |
once | -0.004 |
best | -0.004 |
moments | -0.004 |
again | -0.004 |
Token | time |
Token ID | 640 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
knowing | +0.022 |
enough | +0.022 |
once | +0.022 |
listener | +0.020 |
joy | +0.020 |
Neg logit contributions | |
agons | -0.045 |
alks | -0.039 |
cks | -0.038 |
pier | -0.038 |
issors | -0.038 |
Token | bird |
Token ID | 6512 |
Feature activation | -3.69e-02 |
Pos logit contributions | |
enough | +0.003 |
once | +0.003 |
again | +0.002 |
best | +0.002 |
moments | +0.002 |
Neg logit contributions | |
agons | -0.004 |
ramids | -0.004 |
curb | -0.004 |
pier | -0.004 |
disl | -0.004 |
Token | fl |
Token ID | 781 |
Feature activation | +7.95e-03 |
Pos logit contributions | |
enough | +0.054 |
best | +0.047 |
once | +0.047 |
quality | +0.045 |
moments | +0.044 |
Neg logit contributions | |
igator | -0.056 |
agons | -0.056 |
ramids | -0.054 |
disl | -0.054 |
uses | -0.054 |
Token | apped |
Token ID | 6320 |
Feature activation | -1.66e-02 |
Pos logit contributions | |
agons | +0.014 |
ramids | +0.013 |
curb | +0.013 |
disl | +0.013 |
pier | +0.013 |
Neg logit contributions | |
once | -0.012 |
enough | -0.012 |
best | -0.011 |
again | -0.011 |
forever | -0.010 |
Token | its |
Token ID | 663 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
enough | +0.021 |
once | +0.019 |
best | +0.019 |
chapter | +0.019 |
moments | +0.018 |
Neg logit contributions | |
lee | -0.024 |
igator | -0.024 |
agons | -0.023 |
alks | -0.023 |
bell | -0.023 |
Token | wings |
Token ID | 12098 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
enough | +0.027 |
once | +0.024 |
again | +0.023 |
best | +0.022 |
moments | +0.021 |
Neg logit contributions | |
agons | -0.023 |
rolling | -0.022 |
rolled | -0.021 |
alks | -0.021 |
ramids | -0.021 |
Token | one |
Token ID | 530 |
Feature activation | -9.69e-02 |
Pos logit contributions | |
enough | +0.037 |
moments | +0.034 |
best | +0.033 |
once | +0.032 |
osa | +0.032 |
Neg logit contributions | |
rolling | -0.063 |
rolled | -0.062 |
agons | -0.061 |
rus | -0.060 |
ramids | -0.059 |
Token | more |
Token ID | 517 |
Feature activation | -4.45e-02 |
Pos logit contributions | |
enough | +0.110 |
quality | +0.102 |
osa | +0.094 |
spending | +0.091 |
Remember | +0.091 |
Neg logit contributions | |
bell | -0.154 |
igator | -0.149 |
ifying | -0.141 |
lights | -0.141 |
rus | -0.139 |
Token | time |
Token ID | 640 |
Feature activation | +9.35e-03 |
Pos logit contributions | |
enough | +0.041 |
best | +0.036 |
once | +0.036 |
again | +0.032 |
moments | +0.028 |
Neg logit contributions | |
agons | -0.091 |
disl | -0.089 |
alks | -0.087 |
lee | -0.086 |
bell | -0.084 |
Token | and |
Token ID | 290 |
Feature activation | -1.90e-03 |
Pos logit contributions | |
alks | +0.017 |
ramids | +0.017 |
rolling | +0.017 |
agons | +0.016 |
disl | +0.016 |
Neg logit contributions | |
enough | -0.010 |
best | -0.009 |
once | -0.009 |
moments | -0.008 |
quality | -0.008 |
Token | flew |
Token ID | 13112 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
best | +0.003 |
quality | +0.003 |
osa | +0.003 |
enough | +0.003 |
once | +0.003 |
Neg logit contributions | |
aro | -0.002 |
arrested | -0.002 |
igator | -0.002 |
rolled | -0.002 |
uses | -0.002 |
Token | away |
Token ID | 1497 |
Feature activation | +2.83e-02 |
Pos logit contributions | |
enough | +0.029 |
best | +0.027 |
once | +0.027 |
moments | +0.027 |
chapter | +0.026 |
Neg logit contributions | |
ramids | -0.023 |
agons | -0.023 |
alks | -0.022 |
rolled | -0.022 |
igator | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
Remember | +0.015 |
best | +0.015 |
osa | +0.015 |
once | +0.015 |
enough | +0.015 |
Neg logit contributions | |
cases | -0.015 |
bell | -0.015 |
curb | -0.015 |
disl | -0.015 |
knock | -0.014 |
Token | She |
Token ID | 1375 |
Feature activation | -4.87e-04 |
Pos logit contributions | |
enough | +0.018 |
once | +0.017 |
best | +0.016 |
again | +0.016 |
more | +0.015 |
Neg logit contributions | |
agons | -0.022 |
ramids | -0.021 |
alks | -0.020 |
disl | -0.020 |
curb | -0.020 |
Token | snapped |
Token ID | 20821 |
Feature activation | -8.91e-03 |
Pos logit contributions | |
enough | +0.001 |
once | +0.001 |
best | +0.001 |
again | +0.000 |
moments | +0.000 |
Neg logit contributions | |
agons | -0.001 |
ramids | -0.001 |
disl | -0.001 |
rolled | -0.001 |
alks | -0.001 |
Token | it |
Token ID | 340 |
Feature activation | -2.08e-02 |
Pos logit contributions | |
enough | +0.010 |
best | +0.010 |
once | +0.010 |
bond | +0.009 |
chapter | +0.009 |
Neg logit contributions | |
ramids | -0.015 |
alks | -0.015 |
lee | -0.014 |
ants | -0.014 |
rolled | -0.014 |
Token | shut |
Token ID | 4423 |
Feature activation | +3.97e-03 |
Pos logit contributions | |
enough | +0.026 |
moments | +0.025 |
once | +0.024 |
best | +0.023 |
quality | +0.022 |
Neg logit contributions | |
agons | -0.035 |
ramids | -0.035 |
rolling | -0.034 |
bell | -0.033 |
rolled | -0.033 |
Token | one |
Token ID | 530 |
Feature activation | -9.51e-02 |
Pos logit contributions | |
ramids | +0.008 |
alks | +0.007 |
curb | +0.007 |
agons | +0.007 |
lee | +0.007 |
Neg logit contributions | |
best | -0.004 |
enough | -0.004 |
once | -0.004 |
moments | -0.004 |
moment | -0.003 |
Token | last |
Token ID | 938 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
enough | +0.126 |
moments | +0.121 |
quality | +0.115 |
once | +0.114 |
Remember | +0.111 |
Neg logit contributions | |
bell | -0.140 |
issors | -0.140 |
agons | -0.140 |
alks | -0.139 |
ifying | -0.139 |
Token | time |
Token ID | 640 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
enough | +0.020 |
once | +0.018 |
best | +0.018 |
moments | +0.017 |
Sweet | +0.017 |
Neg logit contributions | |
alks | -0.036 |
bell | -0.036 |
agons | -0.036 |
issors | -0.035 |
igator | -0.035 |
Token | , |
Token ID | 11 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
alks | +0.021 |
ramids | +0.021 |
lee | +0.020 |
ants | +0.020 |
rolling | +0.019 |
Neg logit contributions | |
enough | -0.013 |
moments | -0.012 |
once | -0.012 |
best | -0.012 |
bond | -0.011 |
Token | hugged |
Token ID | 41130 |
Feature activation | -3.18e-03 |
Pos logit contributions | |
enough | +0.014 |
best | +0.013 |
osa | +0.012 |
quality | +0.012 |
once | +0.012 |
Neg logit contributions | |
curb | -0.023 |
ramids | -0.023 |
disl | -0.023 |
dump | -0.022 |
knock | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | +5.54e-03 |
Pos logit contributions | |
best | +0.004 |
enough | +0.004 |
quality | +0.004 |
moments | +0.004 |
chapter | +0.004 |
Neg logit contributions | |
ramids | -0.005 |
ants | -0.004 |
curb | -0.004 |
lee | -0.004 |
alks | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
enough | +0.016 |
once | +0.015 |
best | +0.012 |
again | +0.012 |
more | +0.012 |
Neg logit contributions | |
agons | -0.032 |
alks | -0.031 |
ramids | -0.030 |
pier | -0.030 |
disl | -0.030 |
Token | the |
Token ID | 262 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
enough | +0.042 |
once | +0.037 |
best | +0.034 |
again | +0.033 |
moments | +0.033 |
Neg logit contributions | |
agons | -0.050 |
ramids | -0.049 |
alks | -0.048 |
disl | -0.048 |
aro | -0.047 |
Token | ordinary |
Token ID | 8850 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
enough | +0.025 |
once | +0.025 |
best | +0.023 |
again | +0.022 |
more | +0.021 |
Neg logit contributions | |
agons | -0.039 |
ramids | -0.037 |
ants | -0.036 |
alks | -0.036 |
disl | -0.036 |
Token | band |
Token ID | 4097 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
enough | +0.017 |
once | +0.015 |
again | +0.014 |
best | +0.014 |
moments | +0.013 |
Neg logit contributions | |
agons | -0.020 |
ramids | -0.019 |
pier | -0.018 |
rolling | -0.018 |
curb | -0.018 |
Token | became |
Token ID | 2627 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
enough | +0.017 |
once | +0.016 |
best | +0.014 |
again | +0.014 |
moments | +0.013 |
Neg logit contributions | |
agons | -0.027 |
ramids | -0.026 |
disl | -0.025 |
alks | -0.025 |
ants | -0.025 |
Token | one |
Token ID | 530 |
Feature activation | -9.37e-02 |
Pos logit contributions | |
enough | +0.030 |
once | +0.030 |
moments | +0.026 |
again | +0.026 |
chapter | +0.024 |
Neg logit contributions | |
agons | -0.055 |
ramids | -0.053 |
alks | -0.052 |
rolled | -0.052 |
rolling | -0.051 |
Token | of |
Token ID | 286 |
Feature activation | -3.17e-03 |
Pos logit contributions | |
enough | +0.115 |
once | +0.099 |
quality | +0.095 |
osa | +0.094 |
moments | +0.092 |
Neg logit contributions | |
igator | -0.149 |
agons | -0.138 |
washer | -0.138 |
aro | -0.134 |
ifying | -0.134 |
Token | the |
Token ID | 262 |
Feature activation | -1.00e-02 |
Pos logit contributions | |
enough | +0.004 |
once | +0.004 |
best | +0.003 |
again | +0.003 |
moments | +0.003 |
Neg logit contributions | |
agons | -0.006 |
ramids | -0.006 |
alks | -0.006 |
disl | -0.006 |
ants | -0.006 |
Token | best |
Token ID | 1266 |
Feature activation | -3.27e-02 |
Pos logit contributions | |
enough | +0.012 |
once | +0.011 |
again | +0.010 |
moments | +0.010 |
best | +0.009 |
Neg logit contributions | |
agons | -0.020 |
ramids | -0.018 |
pier | -0.018 |
disl | -0.018 |
rolling | -0.018 |
Token | bands |
Token ID | 11760 |
Feature activation | -2.26e-02 |
Pos logit contributions | |
enough | +0.036 |
once | +0.033 |
best | +0.029 |
again | +0.028 |
moments | +0.026 |
Neg logit contributions | |
agons | -0.067 |
alks | -0.062 |
rolling | -0.060 |
disl | -0.060 |
pier | -0.060 |
Token | around |
Token ID | 1088 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
enough | +0.024 |
once | +0.022 |
best | +0.020 |
again | +0.019 |
more | +0.017 |
Neg logit contributions | |
agons | -0.050 |
ramids | -0.047 |
alks | -0.046 |
disl | -0.046 |
ants | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | -7.07e-03 |
Pos logit contributions | |
enough | +0.010 |
once | +0.010 |
again | +0.009 |
best | +0.009 |
more | +0.008 |
Neg logit contributions | |
agons | -0.020 |
ramids | -0.020 |
disl | -0.019 |
curb | -0.019 |
alks | -0.019 |
Token | It |
Token ID | 632 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
enough | +0.009 |
once | +0.009 |
again | +0.008 |
best | +0.008 |
moments | +0.007 |
Neg logit contributions | |
agons | -0.014 |
ramids | -0.013 |
ants | -0.013 |
disl | -0.013 |
pier | -0.013 |
Token | was |
Token ID | 373 |
Feature activation | -2.04e-02 |
Pos logit contributions | |
enough | +0.013 |
once | +0.012 |
best | +0.011 |
again | +0.010 |
moments | +0.010 |
Neg logit contributions | |
agons | -0.025 |
ramids | -0.024 |
pier | -0.023 |
alks | -0.023 |
disl | -0.023 |
Token | a |
Token ID | 257 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
enough | +0.021 |
once | +0.020 |
again | +0.018 |
best | +0.018 |
moments | +0.017 |
Neg logit contributions | |
agons | -0.044 |
ramids | -0.042 |
alks | -0.041 |
disl | -0.040 |
ants | -0.040 |
Token | big |
Token ID | 1263 |
Feature activation | -4.63e-04 |
Pos logit contributions | |
enough | +0.033 |
once | +0.031 |
again | +0.028 |
best | +0.028 |
moments | +0.027 |
Neg logit contributions | |
agons | -0.041 |
ramids | -0.039 |
pier | -0.038 |
alks | -0.038 |
curb | -0.038 |
Token | one |
Token ID | 530 |
Feature activation | -9.26e-02 |
Pos logit contributions | |
enough | +0.001 |
once | +0.001 |
best | +0.001 |
moments | +0.001 |
again | +0.001 |
Neg logit contributions | |
agons | -0.001 |
pier | -0.001 |
melon | -0.001 |
dock | -0.001 |
alks | -0.001 |
Token | ! |
Token ID | 0 |
Feature activation | -3.55e-03 |
Pos logit contributions | |
enough | +0.109 |
once | +0.100 |
moments | +0.091 |
best | +0.088 |
again | +0.088 |
Neg logit contributions | |
agons | -0.174 |
igator | -0.167 |
alks | -0.165 |
ramids | -0.165 |
ants | -0.163 |
Token |
|
Token ID | 198 |
Feature activation | -1.10e-03 |
Pos logit contributions | |
enough | +0.004 |
once | +0.004 |
again | +0.003 |
best | +0.003 |
moments | +0.003 |
Neg logit contributions | |
agons | -0.008 |
ramids | -0.007 |
disl | -0.007 |
pier | -0.007 |
ants | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | +4.71e-03 |
Pos logit contributions | |
once | +0.001 |
enough | +0.001 |
again | +0.001 |
best | +0.001 |
more | +0.000 |
Neg logit contributions | |
agons | -0.003 |
ramids | -0.003 |
issors | -0.003 |
alks | -0.003 |
disl | -0.003 |
Token | But |
Token ID | 1537 |
Feature activation | -6.65e-03 |
Pos logit contributions | |
agons | +0.010 |
ramids | +0.010 |
alks | +0.010 |
issors | +0.010 |
pier | +0.009 |
Neg logit contributions | |
once | -0.005 |
enough | -0.005 |
again | -0.004 |
more | -0.004 |
best | -0.004 |
Token | Bob |
Token ID | 5811 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
enough | +0.009 |
best | +0.008 |
once | +0.008 |
quality | +0.008 |
moments | +0.008 |
Neg logit contributions | |
aro | -0.010 |
igator | -0.010 |
ramids | -0.010 |
melon | -0.010 |
rolled | -0.010 |
Token | adventure |
Token ID | 8855 |
Feature activation | -6.60e-03 |
Pos logit contributions | |
enough | +0.018 |
once | +0.016 |
again | +0.015 |
best | +0.014 |
more | +0.014 |
Neg logit contributions | |
agons | -0.023 |
pier | -0.021 |
rolling | -0.021 |
alks | -0.021 |
ramids | -0.021 |
Token | had |
Token ID | 550 |
Feature activation | -3.60e-02 |
Pos logit contributions | |
enough | +0.008 |
osa | +0.007 |
once | +0.007 |
best | +0.007 |
chapter | +0.007 |
Neg logit contributions | |
agons | -0.011 |
rolled | -0.011 |
aro | -0.010 |
igator | -0.010 |
iren | -0.010 |
Token | been |
Token ID | 587 |
Feature activation | -3.36e-02 |
Pos logit contributions | |
enough | +0.048 |
once | +0.046 |
best | +0.045 |
moments | +0.043 |
again | +0.042 |
Neg logit contributions | |
rolled | -0.054 |
ramids | -0.052 |
rolling | -0.052 |
alks | -0.052 |
agons | -0.051 |
Token | a |
Token ID | 257 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
enough | +0.037 |
once | +0.035 |
moments | +0.033 |
best | +0.033 |
osa | +0.030 |
Neg logit contributions | |
ramids | -0.058 |
agons | -0.057 |
rolled | -0.057 |
rolling | -0.057 |
alks | -0.056 |
Token | great |
Token ID | 1049 |
Feature activation | -2.05e-02 |
Pos logit contributions | |
enough | +0.040 |
once | +0.037 |
moments | +0.032 |
again | +0.032 |
best | +0.031 |
Neg logit contributions | |
pier | -0.051 |
curb | -0.050 |
agons | -0.050 |
rolling | -0.050 |
ramids | -0.050 |
Token | one |
Token ID | 530 |
Feature activation | -9.24e-02 |
Pos logit contributions | |
enough | +0.028 |
osa | +0.025 |
Sweet | +0.025 |
una | +0.024 |
ing | +0.023 |
Neg logit contributions | |
yacht | -0.029 |
flood | -0.028 |
dock | -0.028 |
wreck | -0.027 |
hurricane | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | -4.84e-03 |
Pos logit contributions | |
enough | +0.104 |
quality | +0.092 |
osa | +0.090 |
moments | +0.088 |
consequences | +0.086 |
Neg logit contributions | |
igator | -0.156 |
bell | -0.138 |
ifying | -0.136 |
yacht | -0.133 |
aro | -0.131 |
Token | he |
Token ID | 339 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
enough | +0.007 |
once | +0.006 |
osa | +0.006 |
best | +0.006 |
again | +0.006 |
Neg logit contributions | |
disl | -0.007 |
igator | -0.007 |
pier | -0.007 |
rolled | -0.007 |
cases | -0.006 |
Token | couldn |
Token ID | 3521 |
Feature activation | +8.70e-03 |
Pos logit contributions | |
enough | +0.019 |
once | +0.017 |
best | +0.016 |
again | +0.015 |
moments | +0.015 |
Neg logit contributions | |
agons | -0.027 |
ramids | -0.025 |
pier | -0.025 |
rolled | -0.025 |
disl | -0.025 |
Token | 't |
Token ID | 470 |
Feature activation | -2.89e-03 |
Pos logit contributions | |
agons | +0.011 |
ramids | +0.010 |
disl | +0.010 |
pier | +0.009 |
alks | +0.009 |
Neg logit contributions | |
once | -0.018 |
enough | -0.018 |
again | -0.017 |
best | -0.016 |
more | -0.015 |
Token | wait |
Token ID | 4043 |
Feature activation | +1.22e-02 |
Pos logit contributions | |
enough | +0.004 |
once | +0.004 |
again | +0.004 |
best | +0.004 |
more | +0.003 |
Neg logit contributions | |
agons | -0.005 |
ramids | -0.005 |
alks | -0.005 |
ants | -0.005 |
issors | -0.005 |
Token | coin |
Token ID | 10752 |
Feature activation | -2.60e-02 |
Pos logit contributions | |
enough | +0.035 |
once | +0.033 |
again | +0.030 |
best | +0.029 |
moments | +0.028 |
Neg logit contributions | |
agons | -0.049 |
ramids | -0.046 |
pier | -0.043 |
alks | -0.043 |
igator | -0.043 |
Token | was |
Token ID | 373 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
enough | +0.026 |
once | +0.024 |
again | +0.021 |
moments | +0.021 |
best | +0.020 |
Neg logit contributions | |
agons | -0.056 |
ramids | -0.053 |
alks | -0.052 |
igator | -0.052 |
rus | -0.051 |
Token | a |
Token ID | 257 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
enough | +0.021 |
once | +0.020 |
again | +0.018 |
moments | +0.018 |
best | +0.018 |
Neg logit contributions | |
agons | -0.036 |
ramids | -0.035 |
alks | -0.034 |
ants | -0.033 |
pier | -0.033 |
Token | shiny |
Token ID | 22441 |
Feature activation | -7.39e-03 |
Pos logit contributions | |
enough | +0.023 |
once | +0.022 |
again | +0.020 |
best | +0.020 |
moments | +0.019 |
Neg logit contributions | |
agons | -0.027 |
ramids | -0.026 |
pier | -0.025 |
curb | -0.024 |
alks | -0.024 |
Token | gold |
Token ID | 3869 |
Feature activation | -2.27e-02 |
Pos logit contributions | |
enough | +0.011 |
once | +0.011 |
best | +0.010 |
moments | +0.010 |
again | +0.010 |
Neg logit contributions | |
agons | -0.012 |
ramids | -0.011 |
alks | -0.011 |
curb | -0.011 |
pier | -0.011 |
Token | one |
Token ID | 530 |
Feature activation | -9.24e-02 |
Pos logit contributions | |
enough | +0.035 |
once | +0.034 |
moments | +0.031 |
best | +0.031 |
again | +0.030 |
Neg logit contributions | |
agons | -0.037 |
ramids | -0.034 |
alks | -0.033 |
rus | -0.033 |
curb | -0.032 |
Token | that |
Token ID | 326 |
Feature activation | +7.03e-03 |
Pos logit contributions | |
enough | +0.110 |
moments | +0.106 |
again | +0.094 |
joy | +0.094 |
once | +0.094 |
Neg logit contributions | |
igator | -0.165 |
agons | -0.155 |
issors | -0.154 |
ramids | -0.153 |
melon | -0.152 |
Token | her |
Token ID | 607 |
Feature activation | +5.39e-04 |
Pos logit contributions | |
agons | +0.012 |
ramids | +0.012 |
igator | +0.011 |
alks | +0.011 |
issors | +0.011 |
Neg logit contributions | |
enough | -0.010 |
once | -0.009 |
again | -0.008 |
moments | -0.008 |
best | -0.008 |
Token | grandma |
Token ID | 49890 |
Feature activation | +8.68e-03 |
Pos logit contributions | |
agons | +0.001 |
ramids | +0.001 |
alks | +0.001 |
disl | +0.001 |
ants | +0.001 |
Neg logit contributions | |
enough | -0.001 |
once | -0.001 |
again | -0.001 |
best | -0.001 |
moments | -0.001 |
Token | had |
Token ID | 550 |
Feature activation | -3.54e-03 |
Pos logit contributions | |
agons | +0.015 |
ramids | +0.015 |
alks | +0.014 |
disl | +0.014 |
ants | +0.014 |
Neg logit contributions | |
enough | -0.013 |
once | -0.012 |
best | -0.011 |
again | -0.011 |
moments | -0.011 |
Token | given |
Token ID | 1813 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
enough | +0.004 |
once | +0.004 |
best | +0.004 |
again | +0.004 |
more | +0.003 |
Neg logit contributions | |
agons | -0.007 |
ramids | -0.007 |
disl | -0.007 |
ants | -0.007 |
alks | -0.007 |
Token | had |
Token ID | 550 |
Feature activation | -1.68e-02 |
Pos logit contributions | |
once | +0.045 |
enough | +0.045 |
best | +0.041 |
forever | +0.039 |
quality | +0.038 |
Neg logit contributions | |
igator | -0.038 |
disl | -0.034 |
agons | -0.033 |
bel | -0.032 |
lee | -0.032 |
Token | a |
Token ID | 257 |
Feature activation | -2.80e-02 |
Pos logit contributions | |
enough | +0.017 |
once | +0.016 |
again | +0.015 |
best | +0.015 |
moments | +0.014 |
Neg logit contributions | |
ramids | -0.032 |
agons | -0.031 |
disl | -0.031 |
rolling | -0.030 |
alks | -0.030 |
Token | mild |
Token ID | 11607 |
Feature activation | -2.52e-02 |
Pos logit contributions | |
enough | +0.036 |
once | +0.035 |
again | +0.031 |
moments | +0.029 |
best | +0.028 |
Neg logit contributions | |
agons | -0.053 |
ramids | -0.050 |
rolling | -0.048 |
pier | -0.048 |
curb | -0.048 |
Token | ice |
Token ID | 4771 |
Feature activation | -4.64e-02 |
Pos logit contributions | |
enough | +0.027 |
once | +0.023 |
again | +0.022 |
best | +0.021 |
more | +0.019 |
Neg logit contributions | |
agons | -0.052 |
curb | -0.048 |
stret | -0.047 |
rolling | -0.047 |
disl | -0.047 |
Token | cream |
Token ID | 8566 |
Feature activation | -3.41e-02 |
Pos logit contributions | |
enough | +0.083 |
once | +0.078 |
again | +0.071 |
best | +0.071 |
moments | +0.071 |
Neg logit contributions | |
agons | -0.067 |
igator | -0.058 |
curb | -0.057 |
aro | -0.056 |
ramids | -0.056 |
Token | one |
Token ID | 530 |
Feature activation | -9.24e-02 |
Pos logit contributions | |
enough | +0.050 |
once | +0.048 |
best | +0.043 |
moments | +0.043 |
again | +0.041 |
Neg logit contributions | |
agons | -0.056 |
ramids | -0.052 |
curb | -0.052 |
pier | -0.050 |
igator | -0.050 |
Token | time |
Token ID | 640 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
enough | +0.089 |
once | +0.079 |
moments | +0.072 |
again | +0.071 |
joy | +0.066 |
Neg logit contributions | |
igator | -0.178 |
agons | -0.175 |
issors | -0.161 |
ants | -0.160 |
aro | -0.160 |
Token | ! |
Token ID | 0 |
Feature activation | -7.45e-03 |
Pos logit contributions | |
agons | +0.020 |
alks | +0.020 |
ramids | +0.020 |
disl | +0.020 |
ants | +0.019 |
Neg logit contributions | |
enough | -0.011 |
once | -0.011 |
best | -0.010 |
again | -0.009 |
moments | -0.009 |
Token | Even |
Token ID | 3412 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
enough | +0.010 |
once | +0.009 |
best | +0.008 |
again | +0.008 |
more | +0.008 |
Neg logit contributions | |
agons | -0.014 |
ramids | -0.014 |
alks | -0.013 |
ants | -0.013 |
disl | -0.013 |
Token | though |
Token ID | 996 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
once | +0.027 |
enough | +0.026 |
moments | +0.023 |
best | +0.023 |
again | +0.023 |
Neg logit contributions | |
igator | -0.037 |
agons | -0.036 |
ramids | -0.035 |
pans | -0.035 |
disl | -0.034 |
Token | Molly |
Token ID | 30236 |
Feature activation | +5.20e-03 |
Pos logit contributions | |
enough | +0.028 |
once | +0.027 |
best | +0.024 |
moments | +0.024 |
again | +0.024 |
Neg logit contributions | |
agons | -0.035 |
alks | -0.035 |
ramids | -0.034 |
ants | -0.033 |
aro | -0.033 |
Token | a |
Token ID | 257 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
agons | +0.014 |
ramids | +0.013 |
alks | +0.013 |
disl | +0.013 |
ants | +0.012 |
Neg logit contributions | |
enough | -0.005 |
once | -0.004 |
best | -0.004 |
again | -0.004 |
more | -0.003 |
Token | surprise |
Token ID | 5975 |
Feature activation | -4.21e-02 |
Pos logit contributions | |
enough | +0.027 |
best | +0.024 |
once | +0.023 |
more | +0.023 |
again | +0.021 |
Neg logit contributions | |
agons | -0.051 |
ants | -0.050 |
ramids | -0.050 |
cks | -0.049 |
aro | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
enough | +0.047 |
once | +0.044 |
best | +0.041 |
again | +0.038 |
moments | +0.037 |
Neg logit contributions | |
agons | -0.087 |
ramids | -0.083 |
alks | -0.081 |
curb | -0.080 |
pier | -0.080 |
Token | He |
Token ID | 679 |
Feature activation | +9.81e-03 |
Pos logit contributions | |
once | +0.016 |
enough | +0.013 |
again | +0.013 |
for | +0.011 |
more | +0.010 |
Neg logit contributions | |
agons | -0.033 |
cks | -0.031 |
eling | -0.030 |
ants | -0.030 |
aro | -0.030 |
Token | opened |
Token ID | 4721 |
Feature activation | -6.31e-04 |
Pos logit contributions | |
agons | +0.020 |
ramids | +0.019 |
disl | +0.018 |
alks | +0.018 |
ants | +0.018 |
Neg logit contributions | |
enough | -0.012 |
once | -0.012 |
best | -0.011 |
again | -0.010 |
more | -0.010 |
Token | one |
Token ID | 530 |
Feature activation | -9.23e-02 |
Pos logit contributions | |
enough | +0.001 |
once | +0.001 |
best | +0.001 |
again | +0.001 |
moments | +0.001 |
Neg logit contributions | |
agons | -0.001 |
ramids | -0.001 |
alks | -0.001 |
disl | -0.001 |
ants | -0.001 |
Token | of |
Token ID | 286 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
enough | +0.130 |
once | +0.124 |
best | +0.112 |
moments | +0.112 |
again | +0.109 |
Neg logit contributions | |
agons | -0.163 |
ramids | -0.155 |
alks | -0.149 |
curb | -0.148 |
issors | -0.147 |
Token | the |
Token ID | 262 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
agons | +0.036 |
rolling | +0.034 |
disl | +0.034 |
ramids | +0.034 |
ants | +0.034 |
Neg logit contributions | |
enough | -0.015 |
once | -0.014 |
again | -0.012 |
more | -0.011 |
best | -0.011 |
Token | boxes |
Token ID | 10559 |
Feature activation | +1.69e-03 |
Pos logit contributions | |
enough | +0.023 |
once | +0.022 |
best | +0.021 |
again | +0.020 |
more | +0.020 |
Neg logit contributions | |
disl | -0.020 |
agons | -0.020 |
ants | -0.019 |
aro | -0.019 |
ramids | -0.019 |
Token | and |
Token ID | 290 |
Feature activation | +6.26e-03 |
Pos logit contributions | |
ramids | +0.004 |
agons | +0.004 |
alks | +0.004 |
pier | +0.004 |
curb | +0.004 |
Neg logit contributions | |
enough | -0.001 |
moments | -0.001 |
best | -0.001 |
once | -0.001 |
quality | -0.001 |
Token | inside |
Token ID | 2641 |
Feature activation | +5.88e-03 |
Pos logit contributions | |
agons | +0.010 |
ramids | +0.010 |
pier | +0.010 |
alks | +0.010 |
curb | +0.009 |
Neg logit contributions | |
enough | -0.010 |
once | -0.009 |
best | -0.009 |
again | -0.008 |
moments | -0.008 |
Token | She |
Token ID | 1375 |
Feature activation | +4.08e-03 |
Pos logit contributions | |
enough | +0.016 |
once | +0.016 |
best | +0.015 |
again | +0.014 |
more | +0.014 |
Neg logit contributions | |
agons | -0.018 |
ramids | -0.017 |
alks | -0.017 |
disl | -0.017 |
curb | -0.017 |
Token | blinked |
Token ID | 40360 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
agons | +0.007 |
ramids | +0.007 |
pier | +0.007 |
alks | +0.007 |
disl | +0.007 |
Neg logit contributions | |
enough | -0.006 |
once | -0.005 |
best | -0.005 |
again | -0.005 |
moments | -0.005 |
Token | her |
Token ID | 607 |
Feature activation | -6.65e-03 |
Pos logit contributions | |
enough | +0.019 |
best | +0.018 |
once | +0.017 |
moments | +0.016 |
chapter | +0.016 |
Neg logit contributions | |
ramids | -0.030 |
lee | -0.030 |
rolling | -0.028 |
bell | -0.028 |
ants | -0.028 |
Token | eyes |
Token ID | 2951 |
Feature activation | -2.30e-03 |
Pos logit contributions | |
enough | +0.010 |
once | +0.010 |
best | +0.009 |
again | +0.009 |
more | +0.008 |
Neg logit contributions | |
agons | -0.012 |
ramids | -0.011 |
disl | -0.011 |
ants | -0.011 |
pier | -0.011 |
Token | open |
Token ID | 1280 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
enough | +0.002 |
best | +0.002 |
moments | +0.002 |
friendship | +0.002 |
once | +0.002 |
Neg logit contributions | |
bell | -0.005 |
ramids | -0.004 |
rolling | -0.004 |
curb | -0.004 |
cases | -0.004 |
Token | one |
Token ID | 530 |
Feature activation | -9.23e-02 |
Pos logit contributions | |
best | +0.011 |
moments | +0.011 |
Remember | +0.011 |
Sweet | +0.010 |
enough | +0.010 |
Neg logit contributions | |
ramids | -0.020 |
bell | -0.020 |
ifying | -0.019 |
alks | -0.018 |
lee | -0.018 |
Token | last |
Token ID | 938 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
enough | +0.124 |
quality | +0.116 |
moments | +0.115 |
once | +0.113 |
Remember | +0.111 |
Neg logit contributions | |
bell | -0.135 |
ifying | -0.129 |
agons | -0.128 |
alks | -0.128 |
ramids | -0.121 |
Token | time |
Token ID | 640 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
enough | +0.013 |
once | +0.012 |
best | +0.011 |
moments | +0.010 |
Sweet | +0.010 |
Neg logit contributions | |
agons | -0.025 |
bell | -0.025 |
alks | -0.025 |
ramids | -0.024 |
ifying | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | -8.03e-03 |
Pos logit contributions | |
alks | +0.029 |
ramids | +0.029 |
lee | +0.028 |
rolling | +0.027 |
ants | +0.027 |
Neg logit contributions | |
enough | -0.020 |
best | -0.019 |
moments | -0.019 |
once | -0.018 |
bond | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
enough | +0.010 |
best | +0.009 |
once | +0.008 |
moments | +0.008 |
quality | +0.008 |
Neg logit contributions | |
curb | -0.014 |
ramids | -0.014 |
disl | -0.013 |
rolling | -0.013 |
ric | -0.013 |
Token | smiled |
Token ID | 13541 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
best | +0.022 |
enough | +0.021 |
quality | +0.020 |
once | +0.020 |
chapter | +0.019 |
Neg logit contributions | |
bel | -0.019 |
bell | -0.019 |
ramids | -0.019 |
rolled | -0.019 |
lee | -0.019 |
Token | said |
Token ID | 531 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
enough | +0.010 |
once | +0.010 |
best | +0.009 |
again | +0.008 |
moments | +0.008 |
Neg logit contributions | |
agons | -0.017 |
ramids | -0.016 |
disl | -0.015 |
alks | -0.015 |
pier | -0.015 |
Token | it |
Token ID | 340 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
enough | +0.014 |
once | +0.014 |
best | +0.012 |
again | +0.012 |
moments | +0.011 |
Neg logit contributions | |
agons | -0.022 |
ramids | -0.021 |
disl | -0.021 |
alks | -0.021 |
ants | -0.021 |
Token | was |
Token ID | 373 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
enough | +0.027 |
once | +0.025 |
best | +0.022 |
again | +0.021 |
moments | +0.020 |
Neg logit contributions | |
agons | -0.048 |
ramids | -0.047 |
disl | -0.046 |
rolled | -0.045 |
ants | -0.045 |
Token | the |
Token ID | 262 |
Feature activation | +1.59e-03 |
Pos logit contributions | |
enough | +0.014 |
once | +0.013 |
best | +0.011 |
again | +0.011 |
more | +0.010 |
Neg logit contributions | |
agons | -0.027 |
ramids | -0.026 |
alks | -0.025 |
disl | -0.025 |
ants | -0.025 |
Token | best |
Token ID | 1266 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
agons | +0.004 |
igator | +0.003 |
pier | +0.003 |
curb | +0.003 |
dock | +0.003 |
Neg logit contributions | |
enough | -0.001 |
once | -0.001 |
again | -0.001 |
moments | -0.001 |
quality | -0.001 |
Token | one |
Token ID | 530 |
Feature activation | -9.22e-02 |
Pos logit contributions | |
enough | +0.036 |
once | +0.032 |
again | +0.029 |
best | +0.027 |
more | +0.026 |
Neg logit contributions | |
agons | -0.052 |
disl | -0.049 |
alks | -0.049 |
pier | -0.049 |
curb | -0.048 |
Token | yet |
Token ID | 1865 |
Feature activation | -1.66e-02 |
Pos logit contributions | |
enough | +0.103 |
once | +0.086 |
quality | +0.078 |
best | +0.074 |
again | +0.073 |
Neg logit contributions | |
igator | -0.180 |
aro | -0.168 |
agons | -0.160 |
disl | -0.157 |
uses | -0.155 |
Token | ! |
Token ID | 0 |
Feature activation | -6.71e-03 |
Pos logit contributions | |
enough | +0.019 |
once | +0.017 |
best | +0.015 |
again | +0.015 |
moments | +0.014 |
Neg logit contributions | |
agons | -0.033 |
disl | -0.032 |
alks | -0.031 |
ramids | -0.031 |
aro | -0.030 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -5.34e-03 |
Pos logit contributions | |
once | +0.008 |
enough | +0.007 |
again | +0.006 |
best | +0.006 |
moments | +0.006 |
Neg logit contributions | |
agons | -0.015 |
ramids | -0.014 |
disl | -0.013 |
pier | -0.013 |
curb | -0.013 |
Token | Joe |
Token ID | 19585 |
Feature activation | +2.56e-03 |
Pos logit contributions | |
enough | +0.009 |
once | +0.008 |
best | +0.008 |
more | +0.008 |
moments | +0.008 |
Neg logit contributions | |
agons | -0.008 |
ramids | -0.007 |
alks | -0.007 |
pier | -0.007 |
rolled | -0.007 |
Token | y |
Token ID | 88 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
agons | +0.007 |
ramids | +0.006 |
issors | +0.006 |
alks | +0.006 |
curb | +0.006 |
Neg logit contributions | |
once | -0.002 |
enough | -0.002 |
again | -0.002 |
best | -0.001 |
more | -0.001 |
Token | The |
Token ID | 464 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
enough | +0.010 |
once | +0.009 |
best | +0.009 |
again | +0.009 |
moments | +0.008 |
Neg logit contributions | |
agons | -0.011 |
ramids | -0.010 |
disl | -0.010 |
ants | -0.010 |
curb | -0.010 |
Token | purple |
Token ID | 14032 |
Feature activation | -3.57e-03 |
Pos logit contributions | |
enough | +0.019 |
once | +0.017 |
again | +0.016 |
best | +0.015 |
forever | +0.015 |
Neg logit contributions | |
agons | -0.022 |
ramids | -0.020 |
disl | -0.020 |
curb | -0.020 |
alks | -0.020 |
Token | perfume |
Token ID | 40094 |
Feature activation | -2.70e-03 |
Pos logit contributions | |
enough | +0.006 |
once | +0.006 |
best | +0.005 |
again | +0.005 |
forever | +0.005 |
Neg logit contributions | |
agons | -0.005 |
ramids | -0.005 |
alks | -0.005 |
disl | -0.004 |
issors | -0.004 |
Token | bottle |
Token ID | 9294 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
enough | +0.003 |
once | +0.003 |
best | +0.003 |
again | +0.003 |
moments | +0.002 |
Neg logit contributions | |
agons | -0.005 |
ramids | -0.005 |
igator | -0.005 |
disl | -0.005 |
alks | -0.005 |
Token | was |
Token ID | 373 |
Feature activation | -3.12e-02 |
Pos logit contributions | |
agons | +0.022 |
ramids | +0.021 |
alks | +0.020 |
disl | +0.020 |
igator | +0.020 |
Neg logit contributions | |
enough | -0.011 |
once | -0.010 |
best | -0.009 |
again | -0.009 |
moments | -0.009 |
Token | one |
Token ID | 530 |
Feature activation | -9.22e-02 |
Pos logit contributions | |
enough | +0.034 |
once | +0.032 |
best | +0.029 |
again | +0.028 |
moments | +0.027 |
Neg logit contributions | |
agons | -0.064 |
ramids | -0.061 |
alks | -0.060 |
disl | -0.060 |
ants | -0.059 |
Token | of |
Token ID | 286 |
Feature activation | +1.23e-03 |
Pos logit contributions | |
enough | +0.119 |
once | +0.106 |
moments | +0.100 |
quality | +0.099 |
again | +0.096 |
Neg logit contributions | |
igator | -0.147 |
aro | -0.129 |
melon | -0.127 |
bell | -0.127 |
ifying | -0.122 |
Token | Lucy |
Token ID | 22162 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
agons | +0.003 |
ramids | +0.002 |
alks | +0.002 |
disl | +0.002 |
ants | +0.002 |
Neg logit contributions | |
enough | -0.001 |
once | -0.001 |
best | -0.001 |
again | -0.001 |
moments | -0.001 |
Token | 's |
Token ID | 338 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
enough | +0.016 |
once | +0.014 |
moments | +0.013 |
best | +0.013 |
quality | +0.012 |
Neg logit contributions | |
agons | -0.026 |
lee | -0.026 |
ramids | -0.026 |
alks | -0.025 |
rolling | -0.025 |
Token | favorite |
Token ID | 4004 |
Feature activation | -2.83e-02 |
Pos logit contributions | |
enough | +0.028 |
once | +0.026 |
again | +0.024 |
best | +0.022 |
moments | +0.022 |
Neg logit contributions | |
agons | -0.032 |
alks | -0.031 |
disl | -0.030 |
ramids | -0.030 |
issors | -0.029 |
Token | things |
Token ID | 1243 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
enough | +0.040 |
once | +0.035 |
best | +0.032 |
again | +0.031 |
moments | +0.029 |
Neg logit contributions | |
agons | -0.046 |
disl | -0.044 |
alks | -0.044 |
ramids | -0.043 |
igator | -0.042 |
Token | She |
Token ID | 1375 |
Feature activation | +7.13e-04 |
Pos logit contributions | |
enough | +0.024 |
once | +0.022 |
best | +0.021 |
again | +0.020 |
more | +0.020 |
Neg logit contributions | |
agons | -0.030 |
ramids | -0.029 |
alks | -0.028 |
disl | -0.027 |
rolling | -0.027 |
Token | hugged |
Token ID | 41130 |
Feature activation | +4.19e-04 |
Pos logit contributions | |
agons | +0.001 |
ramids | +0.001 |
pier | +0.001 |
rolling | +0.001 |
disl | +0.001 |
Neg logit contributions | |
enough | -0.001 |
once | -0.001 |
best | -0.001 |
again | -0.001 |
moments | -0.001 |
Token | her |
Token ID | 607 |
Feature activation | +9.07e-03 |
Pos logit contributions | |
ramids | +0.001 |
agons | +0.001 |
ants | +0.001 |
alks | +0.001 |
lee | +0.001 |
Neg logit contributions | |
enough | -0.001 |
best | -0.001 |
once | -0.001 |
quality | -0.000 |
moments | -0.000 |
Token | fierce |
Token ID | 14800 |
Feature activation | +5.31e-03 |
Pos logit contributions | |
agons | +0.018 |
ramids | +0.017 |
alks | +0.016 |
disl | +0.016 |
ants | +0.016 |
Neg logit contributions | |
enough | -0.012 |
once | -0.011 |
best | -0.010 |
again | -0.010 |
more | -0.009 |
Token | folder |
Token ID | 9483 |
Feature activation | +2.68e-04 |
Pos logit contributions | |
agons | +0.010 |
alks | +0.009 |
ramids | +0.009 |
disl | +0.009 |
issors | +0.008 |
Neg logit contributions | |
enough | -0.008 |
once | -0.007 |
again | -0.006 |
best | -0.006 |
more | -0.006 |
Token | one |
Token ID | 530 |
Feature activation | -9.22e-02 |
Pos logit contributions | |
agons | +0.001 |
curb | +0.000 |
bell | +0.000 |
cases | +0.000 |
igator | +0.000 |
Neg logit contributions | |
enough | -0.000 |
once | -0.000 |
moments | -0.000 |
best | -0.000 |
quality | -0.000 |
Token | last |
Token ID | 938 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
enough | +0.128 |
quality | +0.116 |
Remember | +0.114 |
moments | +0.114 |
Sweet | +0.113 |
Neg logit contributions | |
igator | -0.114 |
bell | -0.108 |
ifying | -0.105 |
agons | -0.094 |
yacht | -0.093 |
Token | time |
Token ID | 640 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
enough | +0.025 |
best | +0.022 |
once | +0.022 |
again | +0.020 |
moments | +0.019 |
Neg logit contributions | |
agons | -0.050 |
bell | -0.049 |
igator | -0.049 |
pier | -0.048 |
curb | -0.048 |
Token | and |
Token ID | 290 |
Feature activation | -8.34e-03 |
Pos logit contributions | |
alks | +0.020 |
ramids | +0.020 |
rolling | +0.020 |
agons | +0.020 |
lee | +0.020 |
Neg logit contributions | |
enough | -0.011 |
once | -0.010 |
best | -0.010 |
moments | -0.009 |
quality | -0.008 |
Token | promised |
Token ID | 8072 |
Feature activation | -2.05e-02 |
Pos logit contributions | |
once | +0.012 |
enough | +0.012 |
best | +0.012 |
quality | +0.012 |
moments | +0.011 |
Neg logit contributions | |
rolled | -0.011 |
pier | -0.011 |
lee | -0.010 |
cases | -0.010 |
disl | -0.010 |
Token | to |
Token ID | 284 |
Feature activation | -7.05e-03 |
Pos logit contributions | |
enough | +0.021 |
best | +0.019 |
once | +0.019 |
moments | +0.017 |
chapter | +0.016 |
Neg logit contributions | |
disl | -0.040 |
ramids | -0.039 |
rolling | -0.039 |
alks | -0.039 |
rolled | -0.038 |
Token | with |
Token ID | 351 |
Feature activation | +3.58e-03 |
Pos logit contributions | |
enough | +0.005 |
once | +0.005 |
best | +0.004 |
moments | +0.004 |
listener | +0.004 |
Neg logit contributions | |
agons | -0.007 |
disl | -0.007 |
ramids | -0.007 |
pier | -0.007 |
pans | -0.007 |
Token | rice |
Token ID | 11464 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
agons | +0.007 |
disl | +0.006 |
alks | +0.006 |
ramids | +0.006 |
pier | +0.006 |
Neg logit contributions | |
enough | -0.004 |
once | -0.004 |
again | -0.004 |
best | -0.004 |
moments | -0.003 |
Token | was |
Token ID | 373 |
Feature activation | -2.70e-02 |
Pos logit contributions | |
enough | +0.020 |
once | +0.019 |
best | +0.016 |
again | +0.016 |
moments | +0.015 |
Neg logit contributions | |
agons | -0.048 |
ramids | -0.045 |
alks | -0.044 |
disl | -0.044 |
pier | -0.044 |
Token | a |
Token ID | 257 |
Feature activation | -2.82e-02 |
Pos logit contributions | |
enough | +0.029 |
once | +0.028 |
best | +0.024 |
again | +0.024 |
moments | +0.024 |
Neg logit contributions | |
agons | -0.054 |
ramids | -0.052 |
disl | -0.050 |
alks | -0.050 |
ants | -0.049 |
Token | special |
Token ID | 2041 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
enough | +0.035 |
once | +0.034 |
again | +0.028 |
moments | +0.028 |
best | +0.028 |
Neg logit contributions | |
agons | -0.050 |
pier | -0.049 |
disl | -0.048 |
ramids | -0.048 |
curb | -0.048 |
Token | one |
Token ID | 530 |
Feature activation | -9.20e-02 |
Pos logit contributions | |
enough | +0.017 |
once | +0.016 |
best | +0.014 |
again | +0.014 |
moments | +0.014 |
Neg logit contributions | |
agons | -0.024 |
yacht | -0.024 |
melon | -0.024 |
igator | -0.023 |
disl | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | +2.38e-03 |
Pos logit contributions | |
enough | +0.102 |
once | +0.092 |
moments | +0.082 |
again | +0.080 |
best | +0.079 |
Neg logit contributions | |
igator | -0.169 |
agons | -0.164 |
disl | -0.160 |
melon | -0.158 |
ramids | -0.156 |
Token | would |
Token ID | 561 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
disl | +0.004 |
agons | +0.004 |
igator | +0.004 |
ramids | +0.004 |
pans | +0.004 |
Neg logit contributions | |
once | -0.003 |
enough | -0.003 |
again | -0.003 |
best | -0.002 |
moments | -0.002 |
Token | always |
Token ID | 1464 |
Feature activation | -4.22e-03 |
Pos logit contributions | |
enough | +0.013 |
once | +0.012 |
best | +0.011 |
again | +0.011 |
more | +0.010 |
Neg logit contributions | |
agons | -0.022 |
ramids | -0.021 |
alks | -0.020 |
ants | -0.020 |
disl | -0.020 |
Token | remember |
Token ID | 3505 |
Feature activation | -2.68e-02 |
Pos logit contributions | |
enough | +0.007 |
once | +0.006 |
best | +0.006 |
again | +0.006 |
moments | +0.006 |
Neg logit contributions | |
agons | -0.006 |
disl | -0.006 |
pier | -0.006 |
alks | -0.006 |
igator | -0.006 |
Token | it |
Token ID | 340 |
Feature activation | -2.85e-02 |
Pos logit contributions | |
enough | +0.032 |
once | +0.031 |
best | +0.028 |
again | +0.026 |
moments | +0.026 |
Neg logit contributions | |
agons | -0.052 |
ramids | -0.050 |
disl | -0.049 |
alks | -0.049 |
pier | -0.048 |
Token | a |
Token ID | 257 |
Feature activation | -2.61e-02 |
Pos logit contributions | |
agons | +0.008 |
ramids | +0.007 |
alks | +0.007 |
disl | +0.007 |
pier | +0.007 |
Neg logit contributions | |
enough | -0.003 |
once | -0.002 |
best | -0.002 |
again | -0.002 |
more | -0.002 |
Token | surprise |
Token ID | 5975 |
Feature activation | -4.60e-02 |
Pos logit contributions | |
enough | +0.030 |
best | +0.028 |
once | +0.027 |
more | +0.027 |
again | +0.025 |
Neg logit contributions | |
agons | -0.057 |
ramids | -0.055 |
ants | -0.055 |
cks | -0.055 |
aro | -0.053 |
Token | . |
Token ID | 13 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
enough | +0.048 |
once | +0.046 |
best | +0.041 |
again | +0.038 |
moments | +0.037 |
Neg logit contributions | |
agons | -0.098 |
ramids | -0.094 |
alks | -0.091 |
curb | -0.091 |
pier | -0.091 |
Token | He |
Token ID | 679 |
Feature activation | +9.41e-03 |
Pos logit contributions | |
once | +0.012 |
enough | +0.010 |
again | +0.009 |
best | +0.007 |
more | +0.007 |
Neg logit contributions | |
agons | -0.029 |
cks | -0.027 |
eling | -0.026 |
els | -0.026 |
ants | -0.026 |
Token | opened |
Token ID | 4721 |
Feature activation | -5.28e-03 |
Pos logit contributions | |
agons | +0.019 |
ramids | +0.018 |
disl | +0.017 |
alks | +0.017 |
ants | +0.017 |
Neg logit contributions | |
enough | -0.012 |
once | -0.012 |
best | -0.010 |
again | -0.010 |
more | -0.010 |
Token | one |
Token ID | 530 |
Feature activation | -9.18e-02 |
Pos logit contributions | |
enough | +0.006 |
once | +0.006 |
best | +0.005 |
again | +0.005 |
more | +0.005 |
Neg logit contributions | |
agons | -0.011 |
ramids | -0.010 |
disl | -0.010 |
alks | -0.010 |
ants | -0.010 |
Token | of |
Token ID | 286 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
enough | +0.128 |
once | +0.122 |
best | +0.111 |
moments | +0.110 |
again | +0.107 |
Neg logit contributions | |
agons | -0.161 |
ramids | -0.154 |
alks | -0.148 |
curb | -0.147 |
pier | -0.145 |
Token | the |
Token ID | 262 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
agons | +0.029 |
ramids | +0.028 |
ants | +0.027 |
disl | +0.027 |
rolling | +0.027 |
Neg logit contributions | |
enough | -0.015 |
once | -0.014 |
again | -0.012 |
best | -0.012 |
more | -0.012 |
Token | boxes |
Token ID | 10559 |
Feature activation | +2.23e-04 |
Pos logit contributions | |
enough | +0.027 |
once | +0.026 |
best | +0.025 |
again | +0.024 |
more | +0.023 |
Neg logit contributions | |
agons | -0.024 |
disl | -0.022 |
ants | -0.022 |
ramids | -0.022 |
alks | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | +5.85e-03 |
Pos logit contributions | |
ramids | +0.000 |
agons | +0.000 |
alks | +0.000 |
pier | +0.000 |
curb | +0.000 |
Neg logit contributions | |
enough | -0.000 |
once | -0.000 |
best | -0.000 |
moments | -0.000 |
quality | -0.000 |
Token | inside |
Token ID | 2641 |
Feature activation | +4.63e-03 |
Pos logit contributions | |
agons | +0.010 |
ramids | +0.009 |
alks | +0.009 |
pier | +0.009 |
curb | +0.009 |
Neg logit contributions | |
enough | -0.009 |
once | -0.009 |
best | -0.008 |
again | -0.008 |
moments | -0.007 |
Token | She |
Token ID | 1375 |
Feature activation | +5.43e-03 |
Pos logit contributions | |
enough | +0.012 |
once | +0.011 |
best | +0.010 |
again | +0.010 |
moments | +0.010 |
Neg logit contributions | |
agons | -0.015 |
alks | -0.015 |
ramids | -0.015 |
issors | -0.014 |
disl | -0.014 |
Token | chose |
Token ID | 7690 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
agons | +0.010 |
ramids | +0.010 |
pier | +0.009 |
alks | +0.009 |
disl | +0.009 |
Neg logit contributions | |
enough | -0.007 |
once | -0.007 |
best | -0.006 |
again | -0.006 |
moments | -0.006 |
Token | a |
Token ID | 257 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
enough | +0.013 |
once | +0.013 |
moments | +0.011 |
again | +0.011 |
best | +0.011 |
Neg logit contributions | |
ramids | -0.029 |
alks | -0.028 |
agons | -0.028 |
ants | -0.027 |
rolled | -0.027 |
Token | pretty |
Token ID | 2495 |
Feature activation | -9.24e-03 |
Pos logit contributions | |
enough | +0.027 |
once | +0.026 |
again | +0.023 |
moments | +0.022 |
best | +0.022 |
Neg logit contributions | |
agons | -0.028 |
ramids | -0.027 |
alks | -0.026 |
pier | -0.026 |
curb | -0.026 |
Token | pink |
Token ID | 11398 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
enough | +0.014 |
once | +0.014 |
best | +0.012 |
again | +0.012 |
moments | +0.012 |
Neg logit contributions | |
agons | -0.015 |
ramids | -0.014 |
alks | -0.014 |
disl | -0.014 |
pier | -0.014 |
Token | one |
Token ID | 530 |
Feature activation | -9.17e-02 |
Pos logit contributions | |
enough | +0.024 |
once | +0.022 |
best | +0.021 |
again | +0.019 |
moments | +0.019 |
Neg logit contributions | |
agons | -0.025 |
ramids | -0.024 |
alks | -0.023 |
curb | -0.023 |
pier | -0.023 |
Token | that |
Token ID | 326 |
Feature activation | +9.99e-03 |
Pos logit contributions | |
enough | +0.115 |
moments | +0.107 |
joy | +0.102 |
quality | +0.100 |
once | +0.099 |
Neg logit contributions | |
igator | -0.142 |
alks | -0.139 |
agons | -0.135 |
ramids | -0.134 |
bell | -0.133 |
Token | looked |
Token ID | 3114 |
Feature activation | -4.81e-05 |
Pos logit contributions | |
agons | +0.016 |
igator | +0.016 |
ramids | +0.016 |
disl | +0.016 |
alks | +0.016 |
Neg logit contributions | |
enough | -0.015 |
once | -0.013 |
again | -0.012 |
moments | -0.012 |
best | -0.011 |
Token | just |
Token ID | 655 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
enough | +0.000 |
once | +0.000 |
best | +0.000 |
moments | +0.000 |
again | +0.000 |
Neg logit contributions | |
agons | -0.000 |
ramids | -0.000 |
disl | -0.000 |
alks | -0.000 |
pier | -0.000 |
Token | her |
Token ID | 607 |
Feature activation | +2.88e-03 |
Pos logit contributions | |
enough | +0.035 |
once | +0.034 |
best | +0.030 |
again | +0.030 |
moments | +0.029 |
Neg logit contributions | |
disl | -0.061 |
ramids | -0.061 |
agons | -0.060 |
alks | -0.060 |
pier | -0.057 |
Token | size |
Token ID | 2546 |
Feature activation | -4.50e-03 |
Pos logit contributions | |
agons | +0.005 |
ramids | +0.005 |
alks | +0.005 |
rolling | +0.005 |
issors | +0.005 |
Neg logit contributions | |
enough | -0.004 |
once | -0.003 |
again | -0.003 |
moments | -0.003 |
more | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | -4.60e-04 |
Pos logit contributions | |
enough | +0.002 |
once | +0.002 |
best | +0.002 |
again | +0.002 |
moments | +0.001 |
Neg logit contributions | |
agons | -0.004 |
ramids | -0.004 |
alks | -0.004 |
ants | -0.004 |
disl | -0.004 |
Token | green |
Token ID | 4077 |
Feature activation | +2.86e-03 |
Pos logit contributions | |
enough | +0.001 |
once | +0.001 |
again | +0.001 |
best | +0.001 |
moments | +0.001 |
Neg logit contributions | |
ramids | -0.001 |
agons | -0.001 |
disl | -0.001 |
alks | -0.001 |
pans | -0.001 |
Token | with |
Token ID | 351 |
Feature activation | +4.14e-03 |
Pos logit contributions | |
agons | +0.006 |
ramids | +0.006 |
alks | +0.005 |
disl | +0.005 |
ants | +0.005 |
Neg logit contributions | |
enough | -0.003 |
once | -0.003 |
best | -0.003 |
again | -0.003 |
moments | -0.003 |
Token | one |
Token ID | 530 |
Feature activation | -7.80e-02 |
Pos logit contributions | |
agons | +0.008 |
ramids | +0.007 |
alks | +0.007 |
disl | +0.007 |
ants | +0.007 |
Neg logit contributions | |
enough | -0.006 |
once | -0.005 |
best | -0.005 |
again | -0.005 |
moments | -0.005 |
Token | purple |
Token ID | 14032 |
Feature activation | -2.35e-03 |
Pos logit contributions | |
enough | +0.092 |
once | +0.087 |
best | +0.077 |
again | +0.076 |
moments | +0.071 |
Neg logit contributions | |
agons | -0.160 |
ramids | -0.153 |
alks | -0.148 |
ants | -0.146 |
disl | -0.146 |
Token | one |
Token ID | 530 |
Feature activation | -9.16e-02 |
Pos logit contributions | |
enough | +0.004 |
once | +0.003 |
again | +0.003 |
best | +0.003 |
more | +0.003 |
Neg logit contributions | |
agons | -0.004 |
pier | -0.004 |
ramids | -0.004 |
disl | -0.004 |
rolling | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -7.21e-03 |
Pos logit contributions | |
enough | +0.110 |
once | +0.104 |
best | +0.091 |
again | +0.090 |
moments | +0.089 |
Neg logit contributions | |
agons | -0.178 |
ramids | -0.170 |
alks | -0.168 |
ants | -0.163 |
igator | -0.161 |
Token | She |
Token ID | 1375 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
enough | +0.010 |
once | +0.009 |
best | +0.009 |
again | +0.008 |
more | +0.008 |
Neg logit contributions | |
agons | -0.013 |
ramids | -0.013 |
alks | -0.012 |
rolling | -0.012 |
ants | -0.012 |
Token | put |
Token ID | 1234 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
agons | +0.019 |
ramids | +0.018 |
alks | +0.017 |
ants | +0.017 |
pier | +0.017 |
Neg logit contributions | |
enough | -0.016 |
once | -0.016 |
best | -0.014 |
again | -0.014 |
moments | -0.013 |
Token | them |
Token ID | 606 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
enough | +0.021 |
once | +0.020 |
best | +0.018 |
again | +0.017 |
moments | +0.017 |
Neg logit contributions | |
agons | -0.028 |
ramids | -0.027 |
alks | -0.026 |
ants | -0.026 |
issors | -0.025 |
Token | all |
Token ID | 477 |
Feature activation | -9.47e-03 |
Pos logit contributions | |
agons | +0.027 |
ramids | +0.025 |
disl | +0.025 |
alks | +0.024 |
curb | +0.024 |
Neg logit contributions | |
enough | -0.013 |
once | -0.012 |
best | -0.011 |
again | -0.011 |
more | -0.010 |
Token | her |
Token ID | 607 |
Feature activation | +3.28e-03 |
Pos logit contributions | |
enough | +0.004 |
once | +0.004 |
est | +0.004 |
quality | +0.004 |
osa | +0.004 |
Neg logit contributions | |
disl | -0.004 |
ramids | -0.003 |
uses | -0.003 |
aro | -0.003 |
knock | -0.003 |
Token | drawing |
Token ID | 8263 |
Feature activation | -4.91e-03 |
Pos logit contributions | |
agons | +0.006 |
ramids | +0.006 |
ants | +0.005 |
pier | +0.005 |
curb | +0.005 |
Neg logit contributions | |
enough | -0.005 |
once | -0.005 |
best | -0.005 |
again | -0.005 |
more | -0.004 |
Token | was |
Token ID | 373 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
enough | +0.005 |
once | +0.005 |
best | +0.004 |
again | +0.004 |
moments | +0.004 |
Neg logit contributions | |
agons | -0.010 |
ramids | -0.010 |
disl | -0.010 |
alks | -0.010 |
issors | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | +1.72e-03 |
Pos logit contributions | |
enough | +0.020 |
once | +0.019 |
best | +0.016 |
again | +0.016 |
more | +0.015 |
Neg logit contributions | |
agons | -0.040 |
ramids | -0.038 |
alks | -0.037 |
disl | -0.037 |
ants | -0.037 |
Token | best |
Token ID | 1266 |
Feature activation | -3.01e-02 |
Pos logit contributions | |
agons | +0.004 |
disl | +0.004 |
ramids | +0.004 |
curb | +0.004 |
igator | +0.004 |
Neg logit contributions | |
enough | -0.001 |
once | -0.001 |
again | -0.001 |
moments | -0.001 |
forever | -0.001 |
Token | one |
Token ID | 530 |
Feature activation | -9.16e-02 |
Pos logit contributions | |
enough | +0.040 |
once | +0.035 |
osa | +0.033 |
best | +0.031 |
moments | +0.031 |
Neg logit contributions | |
alks | -0.048 |
agons | -0.047 |
disl | -0.046 |
bell | -0.045 |
curb | -0.043 |
Token | in |
Token ID | 287 |
Feature activation | +9.34e-03 |
Pos logit contributions | |
enough | +0.109 |
once | +0.090 |
quality | +0.083 |
osa | +0.082 |
moments | +0.081 |
Neg logit contributions | |
igator | -0.165 |
aro | -0.151 |
disl | -0.143 |
washer | -0.141 |
uses | -0.139 |
Token | the |
Token ID | 262 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
agons | +0.018 |
ramids | +0.017 |
alks | +0.017 |
disl | +0.016 |
pier | +0.016 |
Neg logit contributions | |
enough | -0.012 |
once | -0.011 |
best | -0.010 |
again | -0.010 |
moments | -0.010 |
Token | drawer |
Token ID | 33451 |
Feature activation | -5.16e-03 |
Pos logit contributions | |
enough | +0.020 |
once | +0.019 |
again | +0.016 |
best | +0.016 |
moments | +0.015 |
Neg logit contributions | |
agons | -0.031 |
ramids | -0.029 |
alks | -0.029 |
disl | -0.029 |
pier | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | -8.72e-03 |
Pos logit contributions | |
enough | +0.006 |
once | +0.005 |
best | +0.005 |
moments | +0.005 |
again | +0.005 |
Neg logit contributions | |
agons | -0.010 |
ramids | -0.010 |
alks | -0.009 |
pier | -0.009 |
curb | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | -2.24e-02 |
Pos logit contributions | |
enough | +0.008 |
once | +0.008 |
best | +0.007 |
moments | +0.006 |
again | +0.006 |
Neg logit contributions | |
ramids | -0.018 |
agons | -0.018 |
disl | -0.018 |
alks | -0.018 |
curb | -0.018 |
Token | gave |
Token ID | 2921 |
Feature activation | -2.12e-02 |
Pos logit contributions | |
enough | +0.016 |
once | +0.016 |
best | +0.015 |
again | +0.014 |
moments | +0.014 |
Neg logit contributions | |
agons | -0.019 |
ramids | -0.018 |
pier | -0.017 |
ants | -0.017 |
curb | -0.017 |
Token | him |
Token ID | 683 |
Feature activation | +9.98e-03 |
Pos logit contributions | |
enough | +0.035 |
once | +0.034 |
best | +0.032 |
moments | +0.032 |
again | +0.031 |
Neg logit contributions | |
agons | -0.031 |
issors | -0.028 |
alks | -0.028 |
ramids | -0.027 |
uses | -0.027 |
Token | a |
Token ID | 257 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
agons | +0.025 |
ramids | +0.024 |
alks | +0.023 |
curb | +0.023 |
disl | +0.023 |
Neg logit contributions | |
enough | -0.008 |
once | -0.007 |
best | -0.006 |
again | -0.006 |
more | -0.005 |
Token | bright |
Token ID | 6016 |
Feature activation | -1.59e-02 |
Pos logit contributions | |
enough | +0.020 |
once | +0.020 |
best | +0.019 |
more | +0.018 |
again | +0.018 |
Neg logit contributions | |
agons | -0.033 |
ramids | -0.032 |
ants | -0.031 |
alks | -0.031 |
cks | -0.031 |
Token | orange |
Token ID | 10912 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
enough | +0.023 |
once | +0.023 |
best | +0.021 |
again | +0.020 |
more | +0.019 |
Neg logit contributions | |
agons | -0.029 |
ramids | -0.027 |
alks | -0.026 |
ants | -0.026 |
disl | -0.026 |
Token | one |
Token ID | 530 |
Feature activation | -9.16e-02 |
Pos logit contributions | |
enough | +0.036 |
once | +0.036 |
again | +0.033 |
best | +0.033 |
more | +0.031 |
Neg logit contributions | |
agons | -0.037 |
ramids | -0.035 |
disl | -0.033 |
alks | -0.033 |
ants | -0.033 |
Token | with |
Token ID | 351 |
Feature activation | +2.41e-03 |
Pos logit contributions | |
enough | +0.112 |
once | +0.098 |
moments | +0.096 |
joy | +0.091 |
quality | +0.089 |
Neg logit contributions | |
agons | -0.156 |
igator | -0.153 |
alks | -0.145 |
rus | -0.142 |
issors | -0.142 |
Token | yellow |
Token ID | 7872 |
Feature activation | -7.84e-03 |
Pos logit contributions | |
agons | +0.005 |
ramids | +0.005 |
disl | +0.005 |
alks | +0.005 |
pier | +0.005 |
Neg logit contributions | |
enough | -0.003 |
once | -0.003 |
best | -0.002 |
again | -0.002 |
more | -0.002 |
Token | dots |
Token ID | 22969 |
Feature activation | -3.34e-02 |
Pos logit contributions | |
enough | +0.011 |
once | +0.011 |
again | +0.010 |
best | +0.010 |
more | +0.009 |
Neg logit contributions | |
agons | -0.015 |
ramids | -0.014 |
disl | -0.013 |
pier | -0.013 |
issors | -0.013 |
Token | all |
Token ID | 477 |
Feature activation | -1.97e-02 |
Pos logit contributions | |
enough | +0.036 |
once | +0.034 |
best | +0.031 |
moments | +0.029 |
again | +0.029 |
Neg logit contributions | |
agons | -0.070 |
ramids | -0.067 |
alks | -0.066 |
ants | -0.065 |
pier | -0.064 |
Token | over |
Token ID | 625 |
Feature activation | -2.77e-03 |
Pos logit contributions | |
enough | +0.029 |
once | +0.027 |
best | +0.025 |
again | +0.024 |
moments | +0.024 |
Neg logit contributions | |
agons | -0.034 |
ramids | -0.032 |
disl | -0.031 |
alks | -0.031 |
igator | -0.031 |
Token | , |
Token ID | 11 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
enough | +0.007 |
once | +0.007 |
best | +0.006 |
again | +0.006 |
moments | +0.005 |
Neg logit contributions | |
agons | -0.016 |
ramids | -0.015 |
alks | -0.015 |
pier | -0.015 |
disl | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
enough | +0.032 |
once | +0.029 |
best | +0.029 |
moments | +0.029 |
est | +0.027 |
Neg logit contributions | |
ramids | -0.036 |
alks | -0.035 |
disl | -0.035 |
rolled | -0.034 |
curb | -0.034 |
Token | ig |
Token ID | 45329 |
Feature activation | +3.15e-02 |
Pos logit contributions | |
enough | +0.029 |
once | +0.028 |
best | +0.026 |
again | +0.025 |
more | +0.024 |
Neg logit contributions | |
agons | -0.024 |
ramids | -0.023 |
alks | -0.022 |
ants | -0.022 |
disl | -0.022 |
Token | loo |
Token ID | 29680 |
Feature activation | -9.93e-03 |
Pos logit contributions | |
agons | +0.035 |
ramids | +0.032 |
disl | +0.030 |
alks | +0.030 |
pier | +0.030 |
Neg logit contributions | |
once | -0.067 |
enough | -0.067 |
best | -0.063 |
again | -0.062 |
more | -0.059 |
Token | became |
Token ID | 2627 |
Feature activation | -2.78e-02 |
Pos logit contributions | |
enough | +0.011 |
once | +0.010 |
best | +0.009 |
again | +0.008 |
moments | +0.008 |
Neg logit contributions | |
agons | -0.021 |
ramids | -0.020 |
curb | -0.019 |
alks | -0.019 |
pier | -0.019 |
Token | one |
Token ID | 530 |
Feature activation | -9.16e-02 |
Pos logit contributions | |
enough | +0.027 |
once | +0.025 |
best | +0.022 |
again | +0.021 |
moments | +0.020 |
Neg logit contributions | |
agons | -0.063 |
ramids | -0.060 |
alks | -0.059 |
disl | -0.058 |
ants | -0.058 |
Token | of |
Token ID | 286 |
Feature activation | -3.37e-04 |
Pos logit contributions | |
enough | +0.111 |
moments | +0.106 |
once | +0.102 |
quality | +0.102 |
osa | +0.096 |
Neg logit contributions | |
igator | -0.146 |
bell | -0.139 |
agons | -0.136 |
glers | -0.131 |
wreck | -0.131 |
Token | her |
Token ID | 607 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
enough | +0.000 |
once | +0.000 |
best | +0.000 |
again | +0.000 |
more | +0.000 |
Neg logit contributions | |
agons | -0.001 |
ramids | -0.001 |
alks | -0.001 |
disl | -0.001 |
ants | -0.001 |
Token | most |
Token ID | 749 |
Feature activation | -1.61e-02 |
Pos logit contributions | |
enough | +0.017 |
once | +0.015 |
again | +0.014 |
best | +0.013 |
moments | +0.013 |
Neg logit contributions | |
agons | -0.032 |
ramids | -0.030 |
alks | -0.030 |
disl | -0.029 |
rolling | -0.029 |
Token | popular |
Token ID | 2968 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
enough | +0.021 |
once | +0.020 |
again | +0.018 |
best | +0.017 |
moments | +0.017 |
Neg logit contributions | |
agons | -0.029 |
disl | -0.028 |
alks | -0.027 |
ramids | -0.026 |
lee | -0.026 |
Token | memories |
Token ID | 9846 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
agons | +0.025 |
ramids | +0.023 |
alks | +0.023 |
disl | +0.022 |
curb | +0.022 |
Neg logit contributions | |
enough | -0.016 |
once | -0.015 |
best | -0.014 |
again | -0.013 |
more | -0.012 |
Token | And |
Token ID | 843 |
Feature activation | +6.96e-03 |
Pos logit contributions | |
enough | +0.023 |
once | +0.022 |
best | +0.019 |
again | +0.019 |
more | +0.017 |
Neg logit contributions | |
agons | -0.045 |
ramids | -0.043 |
pier | -0.041 |
alks | -0.041 |
ants | -0.041 |
Token | Lily |
Token ID | 20037 |
Feature activation | -3.55e-02 |
Pos logit contributions | |
disl | +0.007 |
nam | +0.007 |
lee | +0.006 |
un | +0.006 |
ols | +0.006 |
Neg logit contributions | |
est | -0.011 |
moments | -0.011 |
joy | -0.011 |
consequences | -0.010 |
quality | -0.010 |
Token | played |
Token ID | 2826 |
Feature activation | -8.63e-03 |
Pos logit contributions | |
enough | +0.047 |
moments | +0.043 |
once | +0.041 |
quality | +0.040 |
chapter | +0.039 |
Neg logit contributions | |
agons | -0.058 |
disl | -0.054 |
lee | -0.053 |
ramids | -0.052 |
rolled | -0.052 |
Token | beautiful |
Token ID | 4950 |
Feature activation | +2.15e-03 |
Pos logit contributions | |
enough | +0.011 |
once | +0.009 |
moment | +0.009 |
moments | +0.009 |
best | +0.009 |
Neg logit contributions | |
agons | -0.015 |
stret | -0.014 |
ts | -0.013 |
ols | -0.013 |
curb | -0.013 |
Token | music |
Token ID | 2647 |
Feature activation | +3.97e-05 |
Pos logit contributions | |
agons | +0.003 |
ramids | +0.003 |
disl | +0.003 |
pier | +0.003 |
stret | +0.003 |
Neg logit contributions | |
enough | -0.004 |
once | -0.003 |
best | -0.003 |
forever | -0.003 |
again | -0.003 |
Token | again |
Token ID | 757 |
Feature activation | -9.15e-02 |
Pos logit contributions | |
agons | +0.000 |
igator | +0.000 |
ols | +0.000 |
curb | +0.000 |
ramids | +0.000 |
Neg logit contributions | |
enough | -0.000 |
moments | -0.000 |
moment | -0.000 |
osa | -0.000 |
once | -0.000 |
Token | , |
Token ID | 11 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
enough | +0.124 |
best | +0.109 |
once | +0.107 |
moments | +0.104 |
luck | +0.104 |
Neg logit contributions | |
agons | -0.135 |
lf | -0.127 |
ants | -0.126 |
alks | -0.126 |
issors | -0.125 |
Token | always |
Token ID | 1464 |
Feature activation | -4.34e-02 |
Pos logit contributions | |
enough | +0.030 |
moments | +0.028 |
osa | +0.028 |
heartedly | +0.028 |
est | +0.025 |
Neg logit contributions | |
curb | -0.036 |
ramids | -0.036 |
disl | -0.035 |
dump | -0.035 |
parked | -0.034 |
Token | remembering |
Token ID | 24865 |
Feature activation | -4.61e-02 |
Pos logit contributions | |
una | +0.059 |
enough | +0.057 |
once | +0.055 |
oph | +0.053 |
osa | +0.053 |
Neg logit contributions | |
un | -0.061 |
nam | -0.061 |
dump | -0.059 |
pans | -0.057 |
knock | -0.054 |
Token | to |
Token ID | 284 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
enough | +0.052 |
once | +0.050 |
best | +0.045 |
moments | +0.042 |
again | +0.041 |
Neg logit contributions | |
agons | -0.090 |
ramids | -0.088 |
disl | -0.086 |
alks | -0.085 |
ants | -0.085 |
Token | be |
Token ID | 307 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
enough | +0.029 |
once | +0.028 |
again | +0.026 |
best | +0.025 |
moments | +0.025 |
Neg logit contributions | |
agons | -0.025 |
disl | -0.023 |
pier | -0.021 |
ramids | -0.021 |
alks | -0.021 |
Token | race |
Token ID | 3234 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
enough | +0.002 |
once | +0.001 |
best | +0.001 |
again | +0.001 |
moments | +0.001 |
Neg logit contributions | |
agons | -0.002 |
ramids | -0.002 |
alks | -0.002 |
disl | -0.002 |
pier | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -4.90e-03 |
Pos logit contributions | |
agons | +0.024 |
ramids | +0.023 |
alks | +0.022 |
pier | +0.022 |
disl | +0.022 |
Neg logit contributions | |
enough | -0.014 |
once | -0.014 |
best | -0.012 |
again | -0.012 |
moments | -0.011 |
Token | no |
Token ID | 645 |
Feature activation | -4.67e-02 |
Pos logit contributions | |
enough | +0.005 |
once | +0.004 |
best | +0.004 |
moments | +0.004 |
again | +0.004 |
Neg logit contributions | |
agons | -0.010 |
disl | -0.010 |
ramids | -0.010 |
curb | -0.010 |
pier | -0.009 |
Token | matter |
Token ID | 2300 |
Feature activation | -2.71e-02 |
Pos logit contributions | |
once | +0.055 |
enough | +0.048 |
best | +0.047 |
osa | +0.043 |
Sweet | +0.042 |
Neg logit contributions | |
agons | -0.076 |
wreck | -0.072 |
dump | -0.072 |
disl | -0.071 |
curb | -0.069 |
Token | how |
Token ID | 703 |
Feature activation | -1.78e-02 |
Pos logit contributions | |
enough | +0.043 |
once | +0.041 |
best | +0.038 |
again | +0.037 |
moments | +0.036 |
Neg logit contributions | |
agons | -0.044 |
ramids | -0.042 |
ants | -0.040 |
alks | -0.040 |
disl | -0.039 |
Token | competitive |
Token ID | 7606 |
Feature activation | +5.50e-02 |
Pos logit contributions | |
enough | +0.019 |
once | +0.017 |
again | +0.015 |
best | +0.015 |
more | +0.014 |
Neg logit contributions | |
agons | -0.042 |
ramids | -0.041 |
curb | -0.039 |
issors | -0.039 |
pier | -0.039 |
Token | it |
Token ID | 340 |
Feature activation | +5.75e-03 |
Pos logit contributions | |
agons | +0.105 |
ramids | +0.100 |
alks | +0.096 |
disl | +0.096 |
ants | +0.095 |
Neg logit contributions | |
enough | -0.073 |
once | -0.070 |
best | -0.064 |
again | -0.062 |
more | -0.058 |
Token | was |
Token ID | 373 |
Feature activation | -9.94e-05 |
Pos logit contributions | |
agons | +0.012 |
ramids | +0.011 |
alks | +0.011 |
ants | +0.011 |
pier | +0.011 |
Neg logit contributions | |
enough | -0.007 |
once | -0.007 |
best | -0.006 |
again | -0.006 |
more | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -9.10e-03 |
Pos logit contributions | |
enough | +0.000 |
once | +0.000 |
best | +0.000 |
again | +0.000 |
moments | +0.000 |
Neg logit contributions | |
agons | -0.000 |
ramids | -0.000 |
alks | -0.000 |
disl | -0.000 |
ants | -0.000 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.75e-03 |
Pos logit contributions | |
enough | +0.011 |
once | +0.011 |
best | +0.010 |
again | +0.009 |
more | +0.009 |
Neg logit contributions | |
agons | -0.018 |
ramids | -0.017 |
alks | -0.017 |
disl | -0.017 |
ants | -0.017 |
Token | The |
Token ID | 464 |
Feature activation | +2.71e-03 |
Pos logit contributions | |
enough | +0.007 |
best | +0.006 |
once | +0.006 |
moments | +0.006 |
more | +0.006 |
Neg logit contributions | |
agons | -0.008 |
ramids | -0.008 |
alks | -0.007 |
rolled | -0.007 |
pier | -0.007 |
Token | didn |
Token ID | 1422 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
enough | +0.013 |
once | +0.012 |
best | +0.011 |
moments | +0.011 |
again | +0.010 |
Neg logit contributions | |
agons | -0.013 |
ramids | -0.012 |
pier | -0.012 |
ants | -0.012 |
rus | -0.012 |
Token | 't |
Token ID | 470 |
Feature activation | -9.16e-04 |
Pos logit contributions | |
agons | +0.036 |
disl | +0.035 |
ramids | +0.034 |
pier | +0.034 |
cient | +0.034 |
Neg logit contributions | |
once | -0.036 |
enough | -0.033 |
again | -0.031 |
best | -0.029 |
more | -0.027 |
Token | anticipate |
Token ID | 23794 |
Feature activation | +1.54e-03 |
Pos logit contributions | |
enough | +0.002 |
once | +0.001 |
best | +0.001 |
again | +0.001 |
more | +0.001 |
Neg logit contributions | |
agons | -0.002 |
ramids | -0.001 |
alks | -0.001 |
ants | -0.001 |
disl | -0.001 |
Token | just |
Token ID | 655 |
Feature activation | -9.36e-03 |
Pos logit contributions | |
agons | +0.003 |
ramids | +0.003 |
alks | +0.003 |
issors | +0.003 |
pier | +0.003 |
Neg logit contributions | |
enough | -0.002 |
once | -0.002 |
again | -0.002 |
best | -0.002 |
more | -0.001 |
Token | how |
Token ID | 703 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
enough | +0.011 |
once | +0.011 |
best | +0.010 |
again | +0.009 |
moments | +0.009 |
Neg logit contributions | |
disl | -0.017 |
agons | -0.017 |
ramids | -0.016 |
curb | -0.016 |
alks | -0.016 |
Token | deep |
Token ID | 2769 |
Feature activation | +5.81e-02 |
Pos logit contributions | |
once | +0.019 |
enough | +0.019 |
again | +0.018 |
best | +0.017 |
more | +0.016 |
Neg logit contributions | |
agons | -0.032 |
ramids | -0.030 |
issors | -0.028 |
curb | -0.028 |
pier | -0.028 |
Token | it |
Token ID | 340 |
Feature activation | +4.73e-03 |
Pos logit contributions | |
agons | +0.109 |
ramids | +0.101 |
alks | +0.100 |
disl | +0.100 |
pier | +0.099 |
Neg logit contributions | |
enough | -0.074 |
once | -0.072 |
best | -0.065 |
again | -0.061 |
moments | -0.059 |
Token | was |
Token ID | 373 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
agons | +0.011 |
ramids | +0.010 |
alks | +0.010 |
disl | +0.010 |
ants | +0.010 |
Neg logit contributions | |
enough | -0.005 |
once | -0.004 |
best | -0.004 |
again | -0.004 |
moments | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
agons | +0.026 |
ramids | +0.024 |
disl | +0.024 |
alks | +0.024 |
curb | +0.024 |
Neg logit contributions | |
enough | -0.013 |
once | -0.012 |
best | -0.011 |
moments | -0.011 |
again | -0.010 |
Token | ended |
Token ID | 4444 |
Feature activation | -8.42e-03 |
Pos logit contributions | |
agons | +0.019 |
ramids | +0.018 |
alks | +0.017 |
ants | +0.017 |
pier | +0.017 |
Neg logit contributions | |
enough | -0.014 |
once | -0.013 |
best | -0.012 |
again | -0.012 |
more | -0.011 |
Token | up |
Token ID | 510 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
enough | +0.011 |
once | +0.010 |
again | +0.009 |
best | +0.009 |
more | +0.008 |
Neg logit contributions | |
agons | -0.018 |
ramids | -0.017 |
issors | -0.016 |
alks | -0.016 |
pier | -0.016 |
Token | time |
Token ID | 640 |
Feature activation | +4.92e-03 |
Pos logit contributions | |
enough | +0.034 |
once | +0.031 |
again | +0.028 |
best | +0.027 |
more | +0.025 |
Neg logit contributions | |
agons | -0.055 |
disl | -0.051 |
alks | -0.050 |
curb | -0.049 |
rolling | -0.049 |
Token | playing |
Token ID | 2712 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
agons | +0.009 |
ramids | +0.009 |
alks | +0.009 |
rolling | +0.009 |
disl | +0.009 |
Neg logit contributions | |
enough | -0.006 |
once | -0.005 |
best | -0.005 |
again | -0.005 |
moments | -0.005 |
Token | with |
Token ID | 351 |
Feature activation | +9.21e-03 |
Pos logit contributions | |
agons | +0.035 |
igator | +0.033 |
ramids | +0.033 |
disl | +0.033 |
pier | +0.032 |
Neg logit contributions | |
enough | -0.020 |
once | -0.016 |
best | -0.016 |
moments | -0.014 |
again | -0.013 |
Token | her |
Token ID | 607 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
agons | +0.019 |
ramids | +0.018 |
alks | +0.018 |
disl | +0.017 |
ants | +0.017 |
Neg logit contributions | |
enough | -0.010 |
once | -0.010 |
best | -0.009 |
again | -0.008 |
moments | -0.008 |
Token | or |
Token ID | 393 |
Feature activation | -1.41e-03 |
Pos logit contributions | |
agons | +0.027 |
ramids | +0.026 |
ants | +0.025 |
alks | +0.025 |
issors | +0.025 |
Neg logit contributions | |
once | -0.017 |
enough | -0.017 |
best | -0.016 |
again | -0.015 |
more | -0.014 |
Token | nam |
Token ID | 7402 |
Feature activation | +6.30e-02 |
Pos logit contributions | |
enough | +0.002 |
once | +0.002 |
best | +0.002 |
moments | +0.002 |
again | +0.002 |
Neg logit contributions | |
agons | -0.002 |
ramids | -0.002 |
disl | -0.002 |
curb | -0.002 |
pier | -0.002 |
Token | ents |
Token ID | 658 |
Feature activation | +2.35e-03 |
Pos logit contributions | |
agons | +0.061 |
disl | +0.052 |
pier | +0.051 |
ramids | +0.050 |
curb | +0.050 |
Neg logit contributions | |
enough | -0.145 |
once | -0.143 |
best | -0.133 |
again | -0.132 |
more | -0.128 |
Token | and |
Token ID | 290 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
agons | +0.005 |
ramids | +0.004 |
alks | +0.004 |
ants | +0.004 |
igator | +0.004 |
Neg logit contributions | |
enough | -0.003 |
once | -0.003 |
best | -0.002 |
moments | -0.002 |
again | -0.002 |
Token | making |
Token ID | 1642 |
Feature activation | +7.90e-03 |
Pos logit contributions | |
agons | +0.020 |
ramids | +0.019 |
disl | +0.019 |
rolled | +0.018 |
curb | +0.018 |
Neg logit contributions | |
once | -0.018 |
enough | -0.017 |
best | -0.015 |
again | -0.015 |
moments | -0.015 |
Token | her |
Token ID | 607 |
Feature activation | +2.58e-02 |
Pos logit contributions | |
agons | +0.016 |
ramids | +0.015 |
disl | +0.015 |
alks | +0.014 |
ants | +0.014 |
Neg logit contributions | |
enough | -0.009 |
once | -0.009 |
best | -0.008 |
again | -0.008 |
moments | -0.007 |
Token | special |
Token ID | 2041 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
agons | +0.049 |
ramids | +0.046 |
alks | +0.044 |
pier | +0.044 |
ants | +0.044 |
Neg logit contributions | |
enough | -0.036 |
once | -0.035 |
best | -0.032 |
again | -0.031 |
more | -0.029 |
Token | saw |
Token ID | 2497 |
Feature activation | +1.47e-02 |
Pos logit contributions | |
enough | +0.006 |
once | +0.006 |
best | +0.005 |
again | +0.005 |
moments | +0.005 |
Neg logit contributions | |
agons | -0.008 |
ramids | -0.008 |
pier | -0.007 |
alks | -0.007 |
ants | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | -4.35e-03 |
Pos logit contributions | |
agons | +0.032 |
ramids | +0.031 |
disl | +0.031 |
pier | +0.030 |
alks | +0.030 |
Neg logit contributions | |
enough | -0.014 |
once | -0.013 |
best | -0.012 |
again | -0.011 |
moments | -0.010 |
Token | big |
Token ID | 1263 |
Feature activation | +1.03e-03 |
Pos logit contributions | |
enough | +0.007 |
once | +0.006 |
best | +0.006 |
again | +0.006 |
moments | +0.005 |
Neg logit contributions | |
agons | -0.008 |
ramids | -0.007 |
pier | -0.007 |
curb | -0.007 |
alks | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
agons | +0.002 |
pier | +0.001 |
ramids | +0.001 |
curb | +0.001 |
igator | +0.001 |
Neg logit contributions | |
enough | -0.002 |
once | -0.002 |
best | -0.002 |
again | -0.001 |
moments | -0.001 |
Token | troubled |
Token ID | 17840 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
agons | +0.023 |
ramids | +0.022 |
pier | +0.021 |
curb | +0.021 |
disl | +0.021 |
Neg logit contributions | |
enough | -0.021 |
once | -0.020 |
again | -0.019 |
best | -0.019 |
moments | -0.017 |
Token | oct |
Token ID | 19318 |
Feature activation | +6.77e-02 |
Pos logit contributions | |
agons | +0.041 |
ramids | +0.038 |
alks | +0.038 |
pier | +0.038 |
curb | +0.038 |
Neg logit contributions | |
enough | -0.033 |
once | -0.032 |
best | -0.029 |
again | -0.028 |
more | -0.026 |
Token | opus |
Token ID | 25790 |
Feature activation | -6.09e-03 |
Pos logit contributions | |
agons | +0.073 |
ramids | +0.070 |
disl | +0.066 |
alks | +0.064 |
curb | +0.063 |
Neg logit contributions | |
enough | -0.141 |
once | -0.140 |
again | -0.131 |
best | -0.131 |
more | -0.123 |
Token | . |
Token ID | 13 |
Feature activation | +4.18e-03 |
Pos logit contributions | |
enough | +0.006 |
once | +0.006 |
again | +0.005 |
best | +0.005 |
more | +0.005 |
Neg logit contributions | |
agons | -0.014 |
ramids | -0.014 |
disl | -0.013 |
alks | -0.013 |
issors | -0.013 |
Token | The |
Token ID | 383 |
Feature activation | +1.15e-03 |
Pos logit contributions | |
agons | +0.008 |
ramids | +0.007 |
ants | +0.007 |
disl | +0.007 |
issors | +0.007 |
Neg logit contributions | |
once | -0.006 |
enough | -0.006 |
again | -0.005 |
best | -0.005 |
more | -0.004 |
Token | oct |
Token ID | 19318 |
Feature activation | +7.76e-02 |
Pos logit contributions | |
agons | +0.002 |
ants | +0.002 |
ramids | +0.002 |
rolling | +0.001 |
cks | +0.001 |
Neg logit contributions | |
once | -0.002 |
best | -0.002 |
enough | -0.002 |
more | -0.002 |
again | -0.002 |
Token | opus |
Token ID | 25790 |
Feature activation | +5.76e-03 |
Pos logit contributions | |
agons | +0.092 |
ramids | +0.088 |
curb | +0.083 |
disl | +0.081 |
alks | +0.080 |
Neg logit contributions | |
once | -0.152 |
enough | -0.146 |
again | -0.140 |
best | -0.140 |
moments | -0.129 |
Token | the |
Token ID | 262 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
enough | +0.028 |
once | +0.028 |
best | +0.027 |
moments | +0.026 |
quality | +0.026 |
Neg logit contributions | |
igator | -0.022 |
agons | -0.022 |
ramids | -0.022 |
aro | -0.021 |
ants | -0.021 |
Token | bird |
Token ID | 6512 |
Feature activation | -2.19e-02 |
Pos logit contributions | |
agons | +0.021 |
ramids | +0.020 |
curb | +0.020 |
disl | +0.020 |
rus | +0.020 |
Neg logit contributions | |
enough | -0.015 |
once | -0.013 |
again | -0.012 |
moments | -0.012 |
best | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | -9.24e-03 |
Pos logit contributions | |
enough | +0.028 |
once | +0.027 |
moments | +0.024 |
best | +0.024 |
again | +0.023 |
Neg logit contributions | |
agons | -0.041 |
ramids | -0.040 |
ants | -0.039 |
igator | -0.038 |
alks | -0.038 |
Token | " |
Token ID | 366 |
Feature activation | +7.35e-03 |
Pos logit contributions | |
enough | +0.013 |
once | +0.012 |
best | +0.012 |
again | +0.011 |
moments | +0.011 |
Neg logit contributions | |
agons | -0.016 |
ramids | -0.016 |
alks | -0.015 |
ants | -0.015 |
pier | -0.015 |
Token | Where |
Token ID | 8496 |
Feature activation | +2.80e-03 |
Pos logit contributions | |
agons | +0.013 |
ramids | +0.012 |
alks | +0.012 |
disl | +0.012 |
ants | +0.012 |
Neg logit contributions | |
enough | -0.011 |
once | -0.010 |
again | -0.009 |
best | -0.009 |
more | -0.009 |
Token | did |
Token ID | 750 |
Feature activation | +6.00e-02 |
Pos logit contributions | |
igator | +0.004 |
agons | +0.004 |
disl | +0.004 |
cient | +0.004 |
aro | +0.004 |
Neg logit contributions | |
enough | -0.004 |
once | -0.004 |
moments | -0.004 |
quality | -0.004 |
knowing | -0.004 |
Token | you |
Token ID | 345 |
Feature activation | +3.03e-02 |
Pos logit contributions | |
agons | +0.133 |
ramids | +0.126 |
alks | +0.123 |
curb | +0.123 |
rolling | +0.122 |
Neg logit contributions | |
enough | -0.069 |
once | -0.062 |
best | -0.056 |
again | -0.054 |
more | -0.053 |
Token | come |
Token ID | 1282 |
Feature activation | +5.41e-03 |
Pos logit contributions | |
agons | +0.058 |
ramids | +0.055 |
alks | +0.053 |
issors | +0.052 |
curb | +0.052 |
Neg logit contributions | |
enough | -0.045 |
once | -0.044 |
best | -0.039 |
again | -0.039 |
more | -0.037 |
Token | from |
Token ID | 422 |
Feature activation | +3.44e-02 |
Pos logit contributions | |
agons | +0.011 |
disl | +0.010 |
ramids | +0.010 |
alks | +0.010 |
issors | +0.010 |
Neg logit contributions | |
once | -0.007 |
enough | -0.007 |
again | -0.006 |
best | -0.006 |
more | -0.006 |
Token | ?" |
Token ID | 1701 |
Feature activation | -7.19e-04 |
Pos logit contributions | |
agons | +0.048 |
ramids | +0.042 |
alks | +0.042 |
issors | +0.042 |
ants | +0.041 |
Neg logit contributions | |
enough | -0.069 |
once | -0.068 |
again | -0.063 |
best | -0.060 |
more | -0.059 |
Token | The |
Token ID | 383 |
Feature activation | +2.28e-02 |
Pos logit contributions | |
enough | +0.001 |
once | +0.001 |
again | +0.001 |
best | +0.001 |
more | +0.001 |
Neg logit contributions | |
agons | -0.002 |
ramids | -0.001 |
alks | -0.001 |
disl | -0.001 |
ants | -0.001 |
Token | you |
Token ID | 345 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
agons | +0.064 |
ramids | +0.060 |
alks | +0.058 |
iren | +0.057 |
issors | +0.057 |
Neg logit contributions | |
enough | -0.030 |
once | -0.028 |
again | -0.026 |
best | -0.024 |
more | -0.023 |
Token | tell |
Token ID | 1560 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
agons | +0.050 |
ramids | +0.050 |
alks | +0.048 |
iren | +0.046 |
ants | +0.046 |
Neg logit contributions | |
enough | -0.037 |
once | -0.035 |
again | -0.032 |
best | -0.030 |
more | -0.030 |
Token | them |
Token ID | 606 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
agons | +0.020 |
ramids | +0.018 |
alks | +0.018 |
pier | +0.018 |
disl | +0.018 |
Neg logit contributions | |
enough | -0.014 |
once | -0.013 |
best | -0.012 |
again | -0.012 |
more | -0.012 |
Token | not |
Token ID | 407 |
Feature activation | +6.37e-03 |
Pos logit contributions | |
agons | +0.050 |
ramids | +0.046 |
alks | +0.046 |
curb | +0.045 |
rolling | +0.045 |
Neg logit contributions | |
enough | -0.018 |
once | -0.016 |
again | -0.015 |
best | -0.014 |
more | -0.013 |
Token | to |
Token ID | 284 |
Feature activation | -7.53e-03 |
Pos logit contributions | |
agons | +0.016 |
ramids | +0.015 |
disl | +0.015 |
alks | +0.015 |
ants | +0.014 |
Neg logit contributions | |
enough | -0.005 |
once | -0.005 |
best | -0.004 |
again | -0.004 |
moments | -0.003 |
Token | be |
Token ID | 307 |
Feature activation | +4.02e-03 |
Pos logit contributions | |
enough | +0.012 |
once | +0.011 |
again | +0.010 |
best | +0.010 |
more | +0.010 |
Neg logit contributions | |
agons | -0.013 |
ramids | -0.013 |
alks | -0.012 |
rolling | -0.012 |
ants | -0.012 |
Token | scared |
Token ID | 12008 |
Feature activation | +3.00e-02 |
Pos logit contributions | |
agons | +0.008 |
ramids | +0.007 |
curb | +0.007 |
pier | +0.007 |
alks | +0.007 |
Neg logit contributions | |
enough | -0.006 |
once | -0.005 |
best | -0.005 |
again | -0.005 |
more | -0.005 |
Token | ?" |
Token ID | 1701 |
Feature activation | -4.12e-03 |
Pos logit contributions | |
agons | +0.054 |
ramids | +0.051 |
ants | +0.049 |
alks | +0.049 |
igator | +0.049 |
Neg logit contributions | |
enough | -0.041 |
once | -0.040 |
best | -0.037 |
moments | -0.035 |
again | -0.035 |
Token | said |
Token ID | 531 |
Feature activation | +6.94e-03 |
Pos logit contributions | |
enough | +0.005 |
once | +0.005 |
best | +0.004 |
again | +0.004 |
more | +0.004 |
Neg logit contributions | |
agons | -0.009 |
ramids | -0.008 |
alks | -0.008 |
disl | -0.008 |
ants | -0.008 |
Token | her |
Token ID | 607 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
aro | +0.011 |
alks | +0.011 |
ramids | +0.011 |
igator | +0.011 |
rolled | +0.010 |
Neg logit contributions | |
enough | -0.009 |
once | -0.009 |
best | -0.009 |
quality | -0.008 |
moments | -0.008 |
Token | mom |
Token ID | 1995 |
Feature activation | +4.83e-03 |
Pos logit contributions | |
agons | +0.025 |
ramids | +0.024 |
alks | +0.023 |
ants | +0.022 |
rolling | +0.022 |
Neg logit contributions | |
enough | -0.026 |
once | -0.025 |
again | -0.022 |
moments | -0.022 |
best | -0.022 |
Token | wanted |
Token ID | 2227 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
enough | +0.030 |
once | +0.029 |
again | +0.027 |
best | +0.026 |
moments | +0.026 |
Neg logit contributions | |
agons | -0.024 |
disl | -0.024 |
ramids | -0.022 |
rolled | -0.022 |
pier | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | -4.58e-04 |
Pos logit contributions | |
ramids | +0.021 |
disl | +0.021 |
ships | +0.021 |
pier | +0.021 |
alks | +0.021 |
Neg logit contributions | |
once | -0.011 |
enough | -0.011 |
moments | -0.010 |
osa | -0.010 |
moment | -0.009 |
Token | perform |
Token ID | 1620 |
Feature activation | +2.35e-02 |
Pos logit contributions | |
once | +0.001 |
best | +0.001 |
enough | +0.001 |
more | +0.001 |
again | +0.001 |
Neg logit contributions | |
ramids | -0.001 |
agons | -0.001 |
ants | -0.001 |
uffed | -0.001 |
rus | -0.001 |
Token | for |
Token ID | 329 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
agons | +0.054 |
ramids | +0.051 |
alks | +0.050 |
disl | +0.049 |
pier | +0.049 |
Neg logit contributions | |
enough | -0.022 |
once | -0.021 |
best | -0.017 |
again | -0.017 |
moments | -0.016 |
Token | Susan |
Token ID | 14919 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
enough | +0.023 |
once | +0.023 |
again | +0.020 |
best | +0.019 |
moment | +0.018 |
Neg logit contributions | |
agons | -0.034 |
disl | -0.031 |
alks | -0.031 |
ramids | -0.030 |
pier | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | +7.43e-03 |
Pos logit contributions | |
agons | +0.030 |
ramids | +0.029 |
ants | +0.028 |
igator | +0.028 |
alks | +0.028 |
Neg logit contributions | |
enough | -0.017 |
once | -0.016 |
moments | -0.015 |
best | -0.014 |
again | -0.014 |
Token | One |
Token ID | 1881 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
agons | +0.013 |
ramids | +0.012 |
alks | +0.011 |
disl | +0.011 |
curb | +0.011 |
Neg logit contributions | |
enough | -0.012 |
once | -0.011 |
best | -0.010 |
again | -0.010 |
moments | -0.010 |
Token | day |
Token ID | 1110 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
enough | +0.028 |
again | +0.021 |
moments | +0.021 |
once | +0.021 |
forever | +0.020 |
Neg logit contributions | |
igator | -0.045 |
agons | -0.043 |
bell | -0.041 |
pier | -0.040 |
curb | -0.040 |
Token | , |
Token ID | 11 |
Feature activation | -8.08e-03 |
Pos logit contributions | |
agons | +0.025 |
ramids | +0.024 |
rolled | +0.024 |
alks | +0.024 |
ants | +0.024 |
Neg logit contributions | |
once | -0.013 |
enough | -0.013 |
again | -0.012 |
best | -0.011 |
moments | -0.011 |
Token | he |
Token ID | 339 |
Feature activation | -3.01e-03 |
Pos logit contributions | |
enough | +0.013 |
once | +0.012 |
again | +0.012 |
best | +0.011 |
moments | +0.011 |
Neg logit contributions | |
agons | -0.012 |
ramids | -0.012 |
curb | -0.011 |
ants | -0.011 |
rolled | -0.011 |
Token | decided |
Token ID | 3066 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
enough | +0.004 |
once | +0.004 |
best | +0.003 |
again | +0.003 |
more | +0.003 |
Neg logit contributions | |
agons | -0.006 |
ramids | -0.006 |
alks | -0.005 |
disl | -0.005 |
ants | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | -2.69e-03 |
Pos logit contributions | |
enough | +0.016 |
once | +0.015 |
best | +0.013 |
again | +0.013 |
more | +0.012 |
Neg logit contributions | |
agons | -0.030 |
ramids | -0.029 |
disl | -0.028 |
alks | -0.028 |
ants | -0.028 |
Token | Thank |
Token ID | 6952 |
Feature activation | -3.42e-02 |
Pos logit contributions | |
enough | +0.004 |
once | +0.004 |
best | +0.003 |
again | +0.003 |
moments | +0.003 |
Neg logit contributions | |
agons | -0.005 |
ramids | -0.005 |
alks | -0.005 |
ants | -0.005 |
pier | -0.005 |
Token | you |
Token ID | 345 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
enough | +0.053 |
once | +0.050 |
again | +0.046 |
best | +0.045 |
more | +0.042 |
Neg logit contributions | |
agons | -0.058 |
ramids | -0.054 |
aro | -0.052 |
ants | -0.052 |
disl | -0.051 |
Token | for |
Token ID | 329 |
Feature activation | -4.19e-02 |
Pos logit contributions | |
enough | +0.009 |
once | +0.009 |
moments | +0.008 |
best | +0.007 |
again | +0.007 |
Neg logit contributions | |
agons | -0.027 |
ramids | -0.026 |
alks | -0.026 |
ants | -0.025 |
pier | -0.025 |
Token | joining |
Token ID | 9679 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
enough | +0.055 |
once | +0.053 |
best | +0.048 |
again | +0.047 |
moments | +0.044 |
Neg logit contributions | |
agons | -0.080 |
ramids | -0.076 |
disl | -0.073 |
alks | -0.073 |
pier | -0.073 |
Token | me |
Token ID | 502 |
Feature activation | +8.36e-03 |
Pos logit contributions | |
enough | +0.035 |
once | +0.034 |
best | +0.031 |
again | +0.030 |
more | +0.028 |
Neg logit contributions | |
agons | -0.044 |
ramids | -0.042 |
alks | -0.040 |
disl | -0.040 |
ants | -0.040 |
Token | !" |
Token ID | 2474 |
Feature activation | -9.85e-03 |
Pos logit contributions | |
agons | +0.017 |
ramids | +0.016 |
alks | +0.015 |
disl | +0.015 |
curb | +0.015 |
Neg logit contributions | |
enough | -0.010 |
once | -0.010 |
best | -0.009 |
again | -0.009 |
more | -0.008 |
Token | Lucy |
Token ID | 22162 |
Feature activation | -6.24e-03 |
Pos logit contributions | |
enough | +0.013 |
once | +0.012 |
best | +0.011 |
again | +0.011 |
moments | +0.010 |
Neg logit contributions | |
agons | -0.019 |
ramids | -0.018 |
disl | -0.017 |
curb | -0.017 |
alks | -0.017 |
Token | thanked |
Token ID | 26280 |
Feature activation | +8.61e-03 |
Pos logit contributions | |
enough | +0.009 |
once | +0.009 |
best | +0.008 |
again | +0.008 |
moments | +0.008 |
Neg logit contributions | |
agons | -0.011 |
ramids | -0.010 |
alks | -0.010 |
pier | -0.010 |
rolling | -0.010 |
Token | him |
Token ID | 683 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
agons | +0.014 |
ramids | +0.013 |
ants | +0.013 |
aro | +0.013 |
curb | +0.013 |
Neg logit contributions | |
enough | -0.013 |
best | -0.012 |
once | -0.012 |
moments | -0.011 |
quality | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
agons | +0.020 |
ramids | +0.020 |
cases | +0.019 |
ants | +0.019 |
curb | +0.018 |
Neg logit contributions | |
enough | -0.010 |
moments | -0.010 |
quality | -0.009 |
once | -0.009 |
best | -0.008 |
Token | ics |
Token ID | 873 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
once | +0.026 |
enough | +0.025 |
best | +0.024 |
again | +0.023 |
more | +0.022 |
Neg logit contributions | |
agons | -0.014 |
pier | -0.013 |
ramids | -0.013 |
disl | -0.013 |
curb | -0.012 |
Token | with |
Token ID | 351 |
Feature activation | +2.74e-02 |
Pos logit contributions | |
enough | +0.031 |
moments | +0.030 |
Sweet | +0.029 |
once | +0.028 |
Good | +0.027 |
Neg logit contributions | |
agons | -0.027 |
igator | -0.025 |
glers | -0.024 |
bel | -0.024 |
ards | -0.023 |
Token | her |
Token ID | 607 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
agons | +0.061 |
ramids | +0.059 |
pier | +0.057 |
alks | +0.057 |
disl | +0.056 |
Neg logit contributions | |
enough | -0.028 |
once | -0.027 |
best | -0.024 |
again | -0.023 |
more | -0.021 |
Token | family |
Token ID | 1641 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
agons | +0.040 |
ramids | +0.039 |
alks | +0.038 |
disl | +0.037 |
ants | +0.037 |
Neg logit contributions | |
enough | -0.020 |
once | -0.019 |
best | -0.017 |
again | -0.016 |
more | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -5.36e-03 |
Pos logit contributions | |
agons | +0.039 |
ramids | +0.037 |
ants | +0.035 |
alks | +0.035 |
pier | +0.035 |
Neg logit contributions | |
enough | -0.027 |
once | -0.025 |
best | -0.022 |
again | -0.022 |
moments | -0.022 |
Token | In |
Token ID | 554 |
Feature activation | +9.09e-03 |
Pos logit contributions | |
enough | +0.008 |
once | +0.007 |
best | +0.007 |
again | +0.006 |
more | +0.006 |
Neg logit contributions | |
agons | -0.010 |
ramids | -0.009 |
alks | -0.009 |
rolling | -0.009 |
ants | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
agons | +0.016 |
alks | +0.015 |
ramids | +0.015 |
pier | +0.014 |
disl | +0.014 |
Neg logit contributions | |
enough | -0.012 |
once | -0.012 |
again | -0.011 |
best | -0.011 |
moments | -0.010 |
Token | autumn |
Token ID | 23608 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
enough | +0.022 |
once | +0.020 |
again | +0.018 |
moments | +0.016 |
knowing | +0.016 |
Neg logit contributions | |
agons | -0.035 |
pier | -0.031 |
curb | -0.031 |
ramids | -0.030 |
igator | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | -5.05e-03 |
Pos logit contributions | |
enough | +0.012 |
once | +0.011 |
best | +0.009 |
again | +0.009 |
moments | +0.009 |
Neg logit contributions | |
agons | -0.024 |
ramids | -0.023 |
alks | -0.022 |
pier | -0.022 |
igator | -0.022 |
Token | Daisy |
Token ID | 40355 |
Feature activation | -2.30e-02 |
Pos logit contributions | |
enough | +0.007 |
once | +0.007 |
again | +0.006 |
moments | +0.006 |
quality | +0.006 |
Neg logit contributions | |
igator | -0.008 |
agons | -0.008 |
ramids | -0.008 |
cient | -0.007 |
pans | -0.007 |
Token | liked |
Token ID | 8288 |
Feature activation | +2.77e-03 |
Pos logit contributions | |
enough | +0.027 |
once | +0.026 |
best | +0.023 |
again | +0.023 |
moments | +0.021 |
Neg logit contributions | |
agons | -0.047 |
ramids | -0.045 |
alks | -0.043 |
disl | -0.043 |
pier | -0.043 |
Token | a |
Token ID | 257 |
Feature activation | -5.60e-03 |
Pos logit contributions | |
agons | +0.005 |
pier | +0.005 |
ramids | +0.005 |
alks | +0.005 |
rolling | +0.005 |
Neg logit contributions | |
enough | -0.004 |
once | -0.004 |
est | -0.004 |
best | -0.004 |
moments | -0.004 |
Token | journey |
Token ID | 7002 |
Feature activation | -3.86e-04 |
Pos logit contributions | |
enough | +0.008 |
once | +0.008 |
best | +0.007 |
again | +0.007 |
moments | +0.006 |
Neg logit contributions | |
agons | -0.010 |
ramids | -0.010 |
pier | -0.009 |
curb | -0.009 |
rolling | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +3.94e-03 |
Pos logit contributions | |
enough | +0.001 |
once | +0.000 |
moments | +0.000 |
best | +0.000 |
Sweet | +0.000 |
Neg logit contributions | |
agons | -0.001 |
rolling | -0.001 |
pier | -0.001 |
curb | -0.001 |
ramids | -0.001 |
Token | Even |
Token ID | 3412 |
Feature activation | -9.42e-03 |
Pos logit contributions | |
agons | +0.006 |
ramids | +0.006 |
alks | +0.005 |
pier | +0.005 |
rolling | +0.005 |
Neg logit contributions | |
enough | -0.007 |
once | -0.006 |
best | -0.006 |
again | -0.006 |
moments | -0.006 |
Token | though |
Token ID | 996 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
once | +0.012 |
enough | +0.012 |
best | +0.010 |
again | +0.010 |
moments | +0.010 |
Neg logit contributions | |
agons | -0.017 |
ramids | -0.016 |
igator | -0.016 |
pier | -0.016 |
rolling | -0.016 |
Token | they |
Token ID | 484 |
Feature activation | +2.76e-02 |
Pos logit contributions | |
enough | +0.023 |
once | +0.022 |
est | +0.021 |
moments | +0.021 |
best | +0.020 |
Neg logit contributions | |
ramids | -0.020 |
agons | -0.020 |
pier | -0.020 |
rolling | -0.019 |
aro | -0.019 |
Token | were |
Token ID | 547 |
Feature activation | +7.30e-03 |
Pos logit contributions | |
agons | +0.058 |
ramids | +0.053 |
alks | +0.053 |
ants | +0.052 |
disl | +0.052 |
Neg logit contributions | |
enough | -0.036 |
once | -0.035 |
best | -0.030 |
again | -0.030 |
more | -0.028 |
Token | anxious |
Token ID | 18116 |
Feature activation | +3.12e-02 |
Pos logit contributions | |
agons | +0.013 |
ramids | +0.012 |
rolling | +0.012 |
alks | +0.012 |
pier | +0.012 |
Neg logit contributions | |
enough | -0.010 |
once | -0.010 |
moments | -0.009 |
again | -0.009 |
best | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | -4.46e-03 |
Pos logit contributions | |
agons | +0.060 |
ramids | +0.058 |
disl | +0.058 |
rolling | +0.058 |
alks | +0.057 |
Neg logit contributions | |
once | -0.031 |
enough | -0.031 |
best | -0.029 |
moments | -0.027 |
more | -0.025 |
Token | they |
Token ID | 484 |
Feature activation | +6.05e-03 |
Pos logit contributions | |
once | +0.007 |
enough | +0.007 |
est | +0.006 |
again | +0.006 |
best | +0.006 |
Neg logit contributions | |
ramids | -0.005 |
pier | -0.005 |
agons | -0.005 |
cases | -0.005 |
dock | -0.005 |
Token | kept |
Token ID | 4030 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
agons | +0.013 |
alks | +0.012 |
disl | +0.012 |
ramids | +0.012 |
ants | +0.012 |
Neg logit contributions | |
enough | -0.007 |
once | -0.007 |
best | -0.006 |
again | -0.006 |
more | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | +3.16e-03 |
Pos logit contributions | |
agons | +0.023 |
ramids | +0.022 |
curb | +0.022 |
alks | +0.021 |
pier | +0.021 |
Neg logit contributions | |
enough | -0.014 |
once | -0.013 |
best | -0.012 |
moments | -0.011 |
again | -0.011 |
Token | never |
Token ID | 1239 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
agons | +0.006 |
ants | +0.006 |
ramids | +0.005 |
alks | +0.005 |
issors | +0.005 |
Neg logit contributions | |
enough | -0.005 |
once | -0.005 |
again | -0.004 |
best | -0.004 |
more | -0.004 |
Token | came |
Token ID | 1625 |
Feature activation | +5.82e-03 |
Pos logit contributions | |
enough | +0.030 |
once | +0.029 |
best | +0.027 |
again | +0.026 |
moments | +0.025 |
Neg logit contributions | |
agons | -0.028 |
ramids | -0.027 |
disl | -0.026 |
pier | -0.026 |
curb | -0.026 |
Token | back |
Token ID | 736 |
Feature activation | -2.52e-02 |
Pos logit contributions | |
agons | +0.012 |
ramids | +0.012 |
alks | +0.011 |
disl | +0.011 |
ants | +0.011 |
Neg logit contributions | |
enough | -0.007 |
once | -0.006 |
best | -0.006 |
again | -0.005 |
moments | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +3.21e-03 |
Pos logit contributions | |
enough | +0.029 |
once | +0.027 |
best | +0.025 |
again | +0.024 |
moments | +0.022 |
Neg logit contributions | |
agons | -0.052 |
ramids | -0.050 |
alks | -0.048 |
disl | -0.048 |
pier | -0.048 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.06e-03 |
Pos logit contributions | |
agons | +0.007 |
ants | +0.007 |
ramids | +0.007 |
rus | +0.007 |
alks | +0.007 |
Neg logit contributions | |
once | -0.003 |
enough | -0.003 |
again | -0.003 |
best | -0.003 |
moments | -0.002 |
Token | L |
Token ID | 43 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
enough | +0.001 |
once | +0.001 |
best | +0.001 |
again | +0.001 |
moments | +0.001 |
Neg logit contributions | |
agons | -0.002 |
ramids | -0.002 |
alks | -0.002 |
pier | -0.002 |
curb | -0.002 |
Token | ily |
Token ID | 813 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
agons | +0.045 |
ramids | +0.043 |
disl | +0.041 |
alks | +0.041 |
curb | +0.041 |
Neg logit contributions | |
enough | -0.033 |
once | -0.032 |
best | -0.029 |
again | -0.028 |
more | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | -8.72e-03 |
Pos logit contributions | |
enough | +0.016 |
once | +0.014 |
best | +0.013 |
again | +0.012 |
moments | +0.011 |
Neg logit contributions | |
agons | -0.041 |
ramids | -0.039 |
alks | -0.038 |
disl | -0.038 |
pier | -0.037 |
Token | Max |
Token ID | 5436 |
Feature activation | +1.11e-03 |
Pos logit contributions | |
enough | +0.011 |
once | +0.010 |
best | +0.009 |
again | +0.009 |
more | +0.008 |
Neg logit contributions | |
agons | -0.019 |
ramids | -0.017 |
alks | -0.017 |
ants | -0.017 |
pier | -0.017 |
Token | are |
Token ID | 389 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
agons | +0.002 |
alks | +0.002 |
ramids | +0.002 |
ants | +0.002 |
issors | +0.002 |
Neg logit contributions | |
enough | -0.001 |
once | -0.001 |
best | -0.001 |
again | -0.001 |
more | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +3.73e-03 |
Pos logit contributions | |
agons | +0.100 |
ramids | +0.095 |
alks | +0.092 |
disl | +0.092 |
ants | +0.091 |
Neg logit contributions | |
enough | -0.053 |
once | -0.050 |
best | -0.045 |
again | -0.044 |
more | -0.040 |
Token | |
Token ID | 220 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
agons | +0.008 |
ramids | +0.007 |
alks | +0.007 |
ants | +0.007 |
pier | +0.007 |
Neg logit contributions | |
enough | -0.004 |
once | -0.004 |
again | -0.004 |
best | -0.004 |
more | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +7.42e-03 |
Pos logit contributions | |
agons | +0.052 |
ramids | +0.049 |
pier | +0.048 |
disl | +0.048 |
alks | +0.048 |
Neg logit contributions | |
once | -0.019 |
enough | -0.019 |
best | -0.016 |
again | -0.016 |
more | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
agons | +0.013 |
ramids | +0.013 |
rolling | +0.012 |
rolled | +0.012 |
ants | +0.012 |
Neg logit contributions | |
enough | -0.011 |
once | -0.009 |
best | -0.009 |
moments | -0.009 |
again | -0.008 |
Token | L |
Token ID | 43 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
agons | +0.030 |
alks | +0.030 |
ramids | +0.028 |
ric | +0.028 |
pier | +0.028 |
Neg logit contributions | |
enough | -0.013 |
once | -0.013 |
again | -0.011 |
more | -0.011 |
best | -0.011 |
Token | ily |
Token ID | 813 |
Feature activation | -2.00e-02 |
Pos logit contributions | |
agons | +0.032 |
ramids | +0.030 |
alks | +0.029 |
disl | +0.029 |
ants | +0.029 |
Neg logit contributions | |
enough | -0.022 |
once | -0.021 |
best | -0.019 |
again | -0.019 |
moments | -0.018 |
Token | said |
Token ID | 531 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
enough | +0.025 |
once | +0.024 |
again | +0.022 |
best | +0.020 |
more | +0.020 |
Neg logit contributions | |
agons | -0.042 |
alks | -0.040 |
ramids | -0.039 |
issors | -0.039 |
ants | -0.038 |
Token | , |
Token ID | 11 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
enough | +0.011 |
once | +0.011 |
best | +0.009 |
moments | +0.009 |
quality | +0.008 |
Neg logit contributions | |
ramids | -0.021 |
agons | -0.020 |
disl | -0.020 |
ants | -0.020 |
rus | -0.020 |
Token | " |
Token ID | 366 |
Feature activation | +4.75e-03 |
Pos logit contributions | |
enough | +0.024 |
once | +0.023 |
best | +0.021 |
again | +0.020 |
more | +0.019 |
Neg logit contributions | |
agons | -0.035 |
ramids | -0.033 |
alks | -0.032 |
disl | -0.032 |
ants | -0.032 |
Token | Hi |
Token ID | 17250 |
Feature activation | +4.42e-03 |
Pos logit contributions | |
ramids | +0.007 |
curb | +0.007 |
disl | +0.007 |
agons | +0.007 |
pier | +0.007 |
Neg logit contributions | |
enough | -0.008 |
best | -0.007 |
once | -0.007 |
more | -0.006 |
again | -0.006 |
Token | Max |
Token ID | 5436 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
agons | +0.009 |
ramids | +0.008 |
disl | +0.008 |
pier | +0.008 |
ants | +0.008 |
Neg logit contributions | |
enough | -0.005 |
once | -0.005 |
best | -0.005 |
moments | -0.004 |
again | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | +6.38e-03 |
Pos logit contributions | |
igator | +0.007 |
agons | +0.007 |
ramids | +0.007 |
ric | +0.007 |
cks | +0.007 |
Neg logit contributions | |
enough | -0.005 |
once | -0.005 |
moments | -0.005 |
moment | -0.005 |
quality | -0.005 |
Token | whole |
Token ID | 2187 |
Feature activation | +3.67e-03 |
Pos logit contributions | |
agons | +0.010 |
ramids | +0.010 |
rus | +0.010 |
curb | +0.010 |
ric | +0.010 |
Neg logit contributions | |
enough | -0.009 |
once | -0.008 |
moments | -0.008 |
again | -0.008 |
quality | -0.007 |
Token | jungle |
Token ID | 20712 |
Feature activation | +2.60e-02 |
Pos logit contributions | |
flood | +0.005 |
wreck | +0.005 |
alks | +0.005 |
ramids | +0.005 |
igator | +0.005 |
Neg logit contributions | |
enough | -0.005 |
knowing | -0.005 |
quality | -0.005 |
once | -0.005 |
moments | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | -3.97e-03 |
Pos logit contributions | |
agons | +0.046 |
ramids | +0.046 |
igator | +0.046 |
curb | +0.045 |
pier | +0.045 |
Neg logit contributions | |
enough | -0.030 |
once | -0.027 |
best | -0.026 |
moments | -0.025 |
quality | -0.025 |
Token | He |
Token ID | 679 |
Feature activation | +9.68e-03 |
Pos logit contributions | |
enough | +0.006 |
once | +0.006 |
best | +0.006 |
again | +0.006 |
moments | +0.005 |
Neg logit contributions | |
agons | -0.007 |
ramids | -0.006 |
alks | -0.006 |
disl | -0.006 |
ants | -0.006 |
Token | felt |
Token ID | 2936 |
Feature activation | +2.82e-03 |
Pos logit contributions | |
agons | +0.017 |
ramids | +0.017 |
pier | +0.016 |
rolling | +0.016 |
alks | +0.016 |
Neg logit contributions | |
enough | -0.013 |
once | -0.013 |
best | -0.012 |
again | -0.011 |
moments | -0.011 |
Token | very |
Token ID | 845 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
agons | +0.006 |
ramids | +0.006 |
alks | +0.006 |
disl | +0.005 |
ants | +0.005 |
Neg logit contributions | |
enough | -0.003 |
once | -0.003 |
best | -0.003 |
again | -0.003 |
more | -0.002 |
Token | proud |
Token ID | 6613 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
agons | +0.034 |
ramids | +0.033 |
ants | +0.031 |
issors | +0.031 |
alks | +0.031 |
Neg logit contributions | |
enough | -0.024 |
once | -0.023 |
best | -0.022 |
again | -0.021 |
more | -0.020 |
Token | of |
Token ID | 286 |
Feature activation | -4.83e-03 |
Pos logit contributions | |
lee | +0.015 |
alks | +0.014 |
uses | +0.014 |
agons | +0.014 |
aro | +0.013 |
Neg logit contributions | |
best | -0.012 |
enough | -0.012 |
moment | -0.012 |
once | -0.012 |
bond | -0.011 |
Token | himself |
Token ID | 2241 |
Feature activation | -1.88e-03 |
Pos logit contributions | |
enough | +0.009 |
once | +0.009 |
best | +0.008 |
eter | +0.008 |
again | +0.008 |
Neg logit contributions | |
agons | -0.006 |
disl | -0.005 |
aro | -0.005 |
cks | -0.005 |
igator | -0.005 |
Token | for |
Token ID | 329 |
Feature activation | -4.02e-02 |
Pos logit contributions | |
moments | +0.002 |
enough | +0.002 |
once | +0.002 |
quality | +0.001 |
best | +0.001 |
Neg logit contributions | |
agons | -0.004 |
ants | -0.004 |
alks | -0.004 |
igator | -0.003 |
cases | -0.003 |
Token | dry |
Token ID | 5894 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
enough | +0.028 |
once | +0.027 |
best | +0.024 |
again | +0.023 |
more | +0.022 |
Neg logit contributions | |
agons | -0.042 |
ramids | -0.040 |
disl | -0.038 |
pier | -0.038 |
alks | -0.038 |
Token | for |
Token ID | 329 |
Feature activation | -4.04e-02 |
Pos logit contributions | |
moments | +0.012 |
enough | +0.011 |
best | +0.011 |
osa | +0.010 |
friendship | +0.010 |
Neg logit contributions | |
agons | -0.021 |
ramids | -0.021 |
cases | -0.020 |
curb | -0.020 |
rus | -0.020 |
Token | the |
Token ID | 262 |
Feature activation | -5.48e-03 |
Pos logit contributions | |
enough | +0.049 |
once | +0.047 |
best | +0.043 |
again | +0.041 |
moments | +0.039 |
Neg logit contributions | |
agons | -0.080 |
ramids | -0.076 |
alks | -0.075 |
ants | -0.073 |
pier | -0.072 |
Token | little |
Token ID | 1310 |
Feature activation | +3.23e-02 |
Pos logit contributions | |
enough | +0.008 |
once | +0.007 |
again | +0.007 |
best | +0.006 |
moments | +0.006 |
Neg logit contributions | |
agons | -0.009 |
curb | -0.009 |
ramids | -0.009 |
igator | -0.009 |
rus | -0.008 |
Token | bird |
Token ID | 6512 |
Feature activation | -3.27e-02 |
Pos logit contributions | |
igator | +0.056 |
yacht | +0.049 |
leak | +0.049 |
flood | +0.049 |
curb | +0.048 |
Neg logit contributions | |
once | -0.039 |
enough | -0.037 |
best | -0.034 |
osa | -0.033 |
Remember | -0.032 |
Token | to |
Token ID | 284 |
Feature activation | -2.76e-02 |
Pos logit contributions | |
enough | +0.033 |
once | +0.031 |
best | +0.028 |
moments | +0.027 |
again | +0.026 |
Neg logit contributions | |
agons | -0.068 |
ramids | -0.065 |
disl | -0.064 |
ants | -0.064 |
alks | -0.064 |
Token | live |
Token ID | 2107 |
Feature activation | +5.77e-03 |
Pos logit contributions | |
once | +0.041 |
enough | +0.041 |
again | +0.038 |
more | +0.036 |
best | +0.035 |
Neg logit contributions | |
agons | -0.051 |
ramids | -0.049 |
ants | -0.046 |
alks | -0.046 |
issors | -0.045 |
Token | in |
Token ID | 287 |
Feature activation | +3.64e-03 |
Pos logit contributions | |
agons | +0.011 |
ramids | +0.010 |
pier | +0.010 |
alks | +0.010 |
igator | +0.010 |
Neg logit contributions | |
enough | -0.007 |
once | -0.007 |
best | -0.006 |
moments | -0.006 |
again | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -7.12e-03 |
Pos logit contributions | |
agons | +0.007 |
ramids | +0.007 |
alks | +0.007 |
disl | +0.006 |
pier | +0.006 |
Neg logit contributions | |
enough | -0.004 |
once | -0.004 |
best | -0.004 |
moments | -0.004 |
again | -0.004 |
Token | The |
Token ID | 383 |
Feature activation | -4.02e-03 |
Pos logit contributions | |
enough | +0.011 |
once | +0.010 |
best | +0.009 |
again | +0.009 |
more | +0.009 |
Neg logit contributions | |
agons | -0.013 |
ramids | -0.012 |
alks | -0.012 |
disl | -0.011 |
ants | -0.011 |
Token | queen |
Token ID | 16599 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
enough | +0.005 |
once | +0.005 |
best | +0.005 |
again | +0.004 |
more | +0.004 |
Neg logit contributions | |
agons | -0.008 |
ramids | -0.008 |
disl | -0.008 |
alks | -0.008 |
ants | -0.008 |
Token | perfect |
Token ID | 2818 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
enough | +0.037 |
once | +0.035 |
again | +0.032 |
moments | +0.031 |
quality | +0.027 |
Neg logit contributions | |
agons | -0.034 |
stret | -0.033 |
curb | -0.032 |
pier | -0.031 |
rolling | -0.031 |
Token | nap |
Token ID | 25422 |
Feature activation | -3.32e-02 |
Pos logit contributions | |
enough | +0.016 |
once | +0.013 |
best | +0.013 |
again | +0.012 |
moments | +0.012 |
Neg logit contributions | |
agons | -0.020 |
curb | -0.019 |
stret | -0.019 |
igator | -0.018 |
rolling | -0.018 |
Token | in |
Token ID | 287 |
Feature activation | +3.87e-03 |
Pos logit contributions | |
enough | +0.046 |
once | +0.042 |
moments | +0.040 |
friendship | +0.039 |
best | +0.038 |
Neg logit contributions | |
agons | -0.054 |
igator | -0.048 |
pans | -0.046 |
kins | -0.046 |
uses | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
agons | +0.007 |
ramids | +0.007 |
alks | +0.006 |
curb | +0.006 |
disl | +0.006 |
Neg logit contributions | |
enough | -0.005 |
best | -0.004 |
once | -0.004 |
moments | -0.004 |
again | -0.004 |
Token | arm |
Token ID | 3211 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
enough | +0.024 |
once | +0.022 |
again | +0.020 |
best | +0.019 |
moments | +0.019 |
Neg logit contributions | |
agons | -0.027 |
curb | -0.026 |
ramids | -0.025 |
pier | -0.024 |
igator | -0.024 |
Token | chair |
Token ID | 16337 |
Feature activation | +9.27e-04 |
Pos logit contributions | |
agons | +0.018 |
curb | +0.016 |
ramids | +0.016 |
igator | +0.015 |
iren | +0.015 |
Neg logit contributions | |
enough | -0.033 |
once | -0.031 |
best | -0.031 |
moments | -0.029 |
again | -0.029 |
Token | , |
Token ID | 11 |
Feature activation | -6.62e-03 |
Pos logit contributions | |
igator | +0.002 |
agons | +0.002 |
curb | +0.002 |
ramids | +0.002 |
cient | +0.002 |
Neg logit contributions | |
enough | -0.001 |
moments | -0.001 |
best | -0.001 |
once | -0.001 |
quality | -0.001 |
Token | feeling |
Token ID | 4203 |
Feature activation | +1.45e-04 |
Pos logit contributions | |
enough | +0.008 |
best | +0.007 |
once | +0.007 |
moments | +0.006 |
quality | +0.006 |
Neg logit contributions | |
curb | -0.012 |
ramids | -0.012 |
agons | -0.011 |
pier | -0.011 |
disl | -0.011 |
Token | so |
Token ID | 523 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
agons | +0.000 |
disl | +0.000 |
ramids | +0.000 |
igator | +0.000 |
curb | +0.000 |
Neg logit contributions | |
enough | -0.000 |
once | -0.000 |
best | -0.000 |
again | -0.000 |
moments | -0.000 |
Token | cozy |
Token ID | 37438 |
Feature activation | -2.31e-03 |
Pos logit contributions | |
agons | +0.036 |
ramids | +0.034 |
disl | +0.034 |
alks | +0.033 |
curb | +0.033 |
Neg logit contributions | |
enough | -0.023 |
once | -0.021 |
best | -0.019 |
again | -0.019 |
moments | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | -9.80e-04 |
Pos logit contributions | |
best | +0.003 |
friendship | +0.003 |
ability | +0.003 |
moments | +0.003 |
faith | +0.003 |
Neg logit contributions | |
curb | -0.003 |
stret | -0.003 |
ramids | -0.003 |
agons | -0.003 |
pier | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | +1.29e-02 |
Pos logit contributions | |
agons | +0.043 |
disl | +0.039 |
ramids | +0.039 |
pier | +0.038 |
ants | +0.037 |
Neg logit contributions | |
enough | -0.037 |
once | -0.035 |
again | -0.031 |
best | -0.031 |
moments | -0.031 |
Token | pet |
Token ID | 4273 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
agons | +0.021 |
pier | +0.020 |
curb | +0.020 |
disl | +0.020 |
stret | +0.020 |
Neg logit contributions | |
enough | -0.019 |
once | -0.018 |
moments | -0.016 |
again | -0.015 |
best | -0.015 |
Token | before |
Token ID | 878 |
Feature activation | -7.70e-03 |
Pos logit contributions | |
agons | +0.021 |
ramids | +0.020 |
igator | +0.020 |
curb | +0.020 |
pier | +0.019 |
Neg logit contributions | |
enough | -0.013 |
once | -0.013 |
best | -0.012 |
moments | -0.011 |
again | -0.011 |
Token | he |
Token ID | 339 |
Feature activation | +1.93e-02 |
Pos logit contributions | |
enough | +0.009 |
once | +0.009 |
best | +0.008 |
again | +0.008 |
more | +0.007 |
Neg logit contributions | |
agons | -0.016 |
ramids | -0.015 |
alks | -0.015 |
disl | -0.015 |
ants | -0.014 |
Token | gets |
Token ID | 3011 |
Feature activation | +7.74e-03 |
Pos logit contributions | |
agons | +0.039 |
ramids | +0.037 |
alks | +0.036 |
curb | +0.036 |
ants | +0.036 |
Neg logit contributions | |
enough | -0.025 |
once | -0.025 |
again | -0.022 |
best | -0.022 |
more | -0.020 |
Token | one |
Token ID | 530 |
Feature activation | -5.26e-02 |
Pos logit contributions | |
agons | +0.016 |
ramids | +0.015 |
curb | +0.014 |
ants | +0.014 |
issors | +0.014 |
Neg logit contributions | |
enough | -0.012 |
once | -0.011 |
again | -0.010 |
more | -0.010 |
best | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -1.50e-02 |
Pos logit contributions | |
enough | +0.064 |
once | +0.060 |
moments | +0.054 |
best | +0.053 |
again | +0.052 |
Neg logit contributions | |
agons | -0.100 |
igator | -0.096 |
ramids | -0.092 |
alks | -0.091 |
disl | -0.091 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -5.61e-03 |
Pos logit contributions | |
enough | +0.017 |
once | +0.016 |
best | +0.014 |
again | +0.014 |
more | +0.013 |
Neg logit contributions | |
agons | -0.033 |
ramids | -0.031 |
pier | -0.030 |
ants | -0.030 |
alks | -0.030 |
Token | One |
Token ID | 3198 |
Feature activation | -3.34e-02 |
Pos logit contributions | |
enough | +0.009 |
moments | +0.007 |
best | +0.007 |
quality | +0.007 |
knowing | +0.007 |
Neg logit contributions | |
agons | -0.009 |
ramids | -0.008 |
alks | -0.008 |
rolled | -0.008 |
iren | -0.008 |
Token | day |
Token ID | 1110 |
Feature activation | -4.07e-03 |
Pos logit contributions | |
enough | +0.027 |
once | +0.022 |
again | +0.020 |
best | +0.019 |
moments | +0.018 |
Neg logit contributions | |
agons | -0.080 |
igator | -0.076 |
disl | -0.075 |
ramids | -0.074 |
alks | -0.074 |
Token | , |
Token ID | 11 |
Feature activation | -1.76e-02 |
Pos logit contributions | |
enough | +0.003 |
once | +0.003 |
best | +0.002 |
again | +0.002 |
more | +0.002 |
Neg logit contributions | |
agons | -0.011 |
ramids | -0.010 |
curb | -0.010 |
disl | -0.010 |
alks | -0.010 |
Token | seat |
Token ID | 5852 |
Feature activation | +3.57e-03 |
Pos logit contributions | |
agons | +0.009 |
ramids | +0.009 |
ants | +0.008 |
alks | +0.008 |
issors | +0.008 |
Neg logit contributions | |
once | -0.007 |
enough | -0.007 |
again | -0.006 |
best | -0.006 |
more | -0.005 |
Token | on |
Token ID | 319 |
Feature activation | +6.40e-05 |
Pos logit contributions | |
agons | +0.008 |
ramids | +0.007 |
alks | +0.007 |
pier | +0.007 |
disl | +0.007 |
Neg logit contributions | |
enough | -0.004 |
once | -0.004 |
again | -0.004 |
best | -0.003 |
more | -0.003 |
Token | her |
Token ID | 607 |
Feature activation | -2.80e-03 |
Pos logit contributions | |
agons | +0.000 |
ramids | +0.000 |
alks | +0.000 |
curb | +0.000 |
disl | +0.000 |
Neg logit contributions | |
enough | -0.000 |
once | -0.000 |
best | -0.000 |
again | -0.000 |
moments | -0.000 |
Token | clean |
Token ID | 3424 |
Feature activation | +1.29e-02 |
Pos logit contributions | |
enough | +0.004 |
once | +0.004 |
best | +0.004 |
again | +0.004 |
more | +0.004 |
Neg logit contributions | |
agons | -0.005 |
ramids | -0.005 |
alks | -0.004 |
disl | -0.004 |
ants | -0.004 |
Token | chair |
Token ID | 5118 |
Feature activation | -6.04e-03 |
Pos logit contributions | |
agons | +0.022 |
ramids | +0.021 |
alks | +0.020 |
disl | +0.020 |
curb | +0.020 |
Neg logit contributions | |
enough | -0.020 |
once | -0.019 |
best | -0.018 |
again | -0.017 |
moments | -0.017 |
Token | again |
Token ID | 757 |
Feature activation | -7.47e-02 |
Pos logit contributions | |
enough | +0.006 |
once | +0.006 |
best | +0.005 |
again | +0.005 |
moments | +0.005 |
Neg logit contributions | |
agons | -0.013 |
ramids | -0.012 |
curb | -0.012 |
alks | -0.012 |
pier | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
enough | +0.096 |
once | +0.089 |
best | +0.086 |
moments | +0.083 |
friendship | +0.076 |
Neg logit contributions | |
agons | -0.132 |
ramids | -0.124 |
alks | -0.122 |
ants | -0.121 |
curb | -0.121 |
Token | She |
Token ID | 1375 |
Feature activation | +9.65e-03 |
Pos logit contributions | |
enough | +0.014 |
once | +0.013 |
best | +0.012 |
again | +0.012 |
moments | +0.011 |
Neg logit contributions | |
agons | -0.022 |
ramids | -0.020 |
alks | -0.020 |
disl | -0.020 |
ants | -0.020 |
Token | was |
Token ID | 373 |
Feature activation | -1.53e-03 |
Pos logit contributions | |
agons | +0.020 |
ramids | +0.019 |
alks | +0.019 |
disl | +0.018 |
ants | +0.018 |
Neg logit contributions | |
enough | -0.011 |
once | -0.011 |
best | -0.010 |
again | -0.010 |
more | -0.009 |
Token | very |
Token ID | 845 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
enough | +0.002 |
once | +0.002 |
best | +0.001 |
again | +0.001 |
moments | +0.001 |
Neg logit contributions | |
agons | -0.003 |
ramids | -0.003 |
alks | -0.003 |
disl | -0.003 |
curb | -0.003 |
Token | happy |
Token ID | 3772 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
agons | +0.032 |
ramids | +0.031 |
ants | +0.030 |
alks | +0.030 |
issors | +0.030 |
Neg logit contributions | |
enough | -0.013 |
once | -0.013 |
best | -0.013 |
more | -0.011 |
again | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
agons | +0.037 |
ramids | +0.036 |
pier | +0.035 |
alks | +0.034 |
curb | +0.034 |
Neg logit contributions | |
enough | -0.026 |
once | -0.024 |
best | -0.023 |
again | -0.021 |
moments | -0.021 |
Token | tree |
Token ID | 5509 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
enough | +0.024 |
once | +0.023 |
best | +0.021 |
again | +0.021 |
moments | +0.020 |
Neg logit contributions | |
agons | -0.027 |
ramids | -0.025 |
curb | -0.024 |
pier | -0.024 |
alks | -0.024 |
Token | and |
Token ID | 290 |
Feature activation | -2.61e-03 |
Pos logit contributions | |
enough | +0.024 |
once | +0.022 |
moments | +0.021 |
best | +0.020 |
quality | +0.018 |
Neg logit contributions | |
agons | -0.041 |
ramids | -0.040 |
alks | -0.040 |
rolling | -0.039 |
ants | -0.039 |
Token | felt |
Token ID | 2936 |
Feature activation | +7.05e-03 |
Pos logit contributions | |
enough | +0.004 |
once | +0.004 |
best | +0.004 |
moments | +0.004 |
quality | +0.004 |
Neg logit contributions | |
ramids | -0.003 |
pier | -0.003 |
curb | -0.003 |
uses | -0.003 |
disl | -0.003 |
Token | full |
Token ID | 1336 |
Feature activation | -4.97e-03 |
Pos logit contributions | |
agons | +0.016 |
ramids | +0.015 |
ants | +0.014 |
alks | +0.014 |
issors | +0.014 |
Neg logit contributions | |
enough | -0.008 |
once | -0.007 |
best | -0.006 |
again | -0.006 |
more | -0.006 |
Token | again |
Token ID | 757 |
Feature activation | -6.76e-02 |
Pos logit contributions | |
enough | +0.006 |
best | +0.006 |
once | +0.006 |
quality | +0.005 |
moment | +0.005 |
Neg logit contributions | |
agons | -0.008 |
igator | -0.008 |
melon | -0.007 |
alks | -0.007 |
ramids | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
enough | +0.098 |
once | +0.092 |
best | +0.090 |
moments | +0.084 |
again | +0.079 |
Neg logit contributions | |
agons | -0.107 |
alks | -0.104 |
ramids | -0.101 |
ants | -0.098 |
pier | -0.098 |
Token | |
Token ID | 220 |
Feature activation | +1.11e-03 |
Pos logit contributions | |
once | +0.018 |
enough | +0.017 |
again | +0.015 |
best | +0.015 |
moments | +0.013 |
Neg logit contributions | |
agons | -0.034 |
ramids | -0.031 |
ants | -0.030 |
disl | -0.030 |
pier | -0.030 |
Token |
|
Token ID | 198 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
agons | +0.003 |
ramids | +0.003 |
pier | +0.003 |
igator | +0.003 |
disl | +0.003 |
Neg logit contributions | |
once | -0.001 |
enough | -0.001 |
best | -0.001 |
again | -0.001 |
more | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | -4.44e-03 |
Pos logit contributions | |
once | +0.007 |
again | +0.005 |
enough | +0.005 |
best | +0.004 |
more | +0.004 |
Neg logit contributions | |
agons | -0.027 |
ramids | -0.025 |
issors | -0.025 |
alks | -0.025 |
rus | -0.025 |
Token | Finally |
Token ID | 11158 |
Feature activation | -6.82e-03 |
Pos logit contributions | |
once | +0.005 |
enough | +0.004 |
again | +0.004 |
more | +0.004 |
best | +0.004 |
Neg logit contributions | |
agons | -0.010 |
alks | -0.009 |
ramids | -0.009 |
issors | -0.009 |
cient | -0.009 |
Token | his |
Token ID | 465 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
best | +0.020 |
enough | +0.020 |
once | +0.019 |
chapter | +0.018 |
moments | +0.018 |
Neg logit contributions | |
pier | -0.030 |
ramids | -0.030 |
rolled | -0.030 |
agons | -0.029 |
alks | -0.029 |
Token | mum |
Token ID | 25682 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
enough | +0.035 |
once | +0.034 |
best | +0.031 |
again | +0.031 |
more | +0.029 |
Neg logit contributions | |
agons | -0.034 |
ramids | -0.032 |
alks | -0.031 |
disl | -0.030 |
pier | -0.030 |
Token | came |
Token ID | 1625 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
enough | +0.032 |
once | +0.030 |
best | +0.030 |
moments | +0.028 |
quality | +0.028 |
Neg logit contributions | |
agons | -0.034 |
ramids | -0.033 |
pier | -0.032 |
rolled | -0.032 |
rolling | -0.032 |
Token | back |
Token ID | 736 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
enough | +0.012 |
once | +0.011 |
best | +0.011 |
moments | +0.010 |
quality | +0.010 |
Neg logit contributions | |
agons | -0.019 |
rolling | -0.019 |
ramids | -0.019 |
rolled | -0.018 |
pier | -0.018 |
Token | with |
Token ID | 351 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
enough | +0.032 |
once | +0.028 |
best | +0.027 |
moments | +0.024 |
again | +0.022 |
Neg logit contributions | |
agons | -0.077 |
ramids | -0.076 |
rolling | -0.075 |
alks | -0.074 |
rolled | -0.074 |
Token | some |
Token ID | 617 |
Feature activation | -5.03e-02 |
Pos logit contributions | |
agons | +0.026 |
alks | +0.024 |
pier | +0.024 |
ramids | +0.024 |
disl | +0.024 |
Neg logit contributions | |
enough | -0.012 |
once | -0.012 |
again | -0.011 |
best | -0.011 |
moments | -0.010 |
Token | y |
Token ID | 331 |
Feature activation | -8.80e-03 |
Pos logit contributions | |
enough | +0.062 |
once | +0.061 |
again | +0.054 |
best | +0.052 |
forever | +0.049 |
Neg logit contributions | |
agons | -0.098 |
alks | -0.087 |
pier | -0.086 |
disl | -0.085 |
igator | -0.085 |
Token | ummy |
Token ID | 13513 |
Feature activation | -3.02e-03 |
Pos logit contributions | |
enough | +0.013 |
once | +0.013 |
best | +0.012 |
again | +0.012 |
moments | +0.011 |
Neg logit contributions | |
agons | -0.015 |
alks | -0.014 |
ramids | -0.014 |
ants | -0.013 |
pier | -0.013 |
Token | chicken |
Token ID | 9015 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
enough | +0.004 |
once | +0.004 |
best | +0.003 |
again | +0.003 |
forever | +0.003 |
Neg logit contributions | |
agons | -0.005 |
pans | -0.005 |
igator | -0.005 |
alks | -0.005 |
disl | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | +4.17e-03 |
Pos logit contributions | |
enough | +0.021 |
once | +0.020 |
best | +0.017 |
again | +0.017 |
moments | +0.016 |
Neg logit contributions | |
agons | -0.042 |
ramids | -0.040 |
alks | -0.039 |
disl | -0.038 |
pier | -0.038 |
Token | chips |
Token ID | 12014 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
agons | +0.007 |
ramids | +0.007 |
disl | +0.007 |
alks | +0.006 |
pier | +0.006 |
Neg logit contributions | |
enough | -0.006 |
once | -0.006 |
best | -0.005 |
again | -0.005 |
moments | -0.005 |
Token | asked |
Token ID | 1965 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
disl | +0.002 |
agons | +0.002 |
ramids | +0.002 |
curb | +0.002 |
pier | +0.002 |
Neg logit contributions | |
enough | -0.002 |
once | -0.002 |
best | -0.002 |
again | -0.001 |
moments | -0.001 |
Token | her |
Token ID | 607 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
enough | +0.016 |
once | +0.016 |
best | +0.014 |
moments | +0.013 |
again | +0.013 |
Neg logit contributions | |
agons | -0.025 |
ramids | -0.024 |
ants | -0.024 |
aro | -0.023 |
disl | -0.023 |
Token | to |
Token ID | 284 |
Feature activation | -5.13e-03 |
Pos logit contributions | |
agons | +0.022 |
ramids | +0.021 |
alks | +0.020 |
disl | +0.020 |
ants | +0.020 |
Neg logit contributions | |
enough | -0.014 |
once | -0.013 |
best | -0.011 |
again | -0.011 |
moments | -0.011 |
Token | read |
Token ID | 1100 |
Feature activation | -1.00e-02 |
Pos logit contributions | |
enough | +0.008 |
once | +0.008 |
best | +0.007 |
again | +0.007 |
moments | +0.007 |
Neg logit contributions | |
agons | -0.008 |
ramids | -0.008 |
alks | -0.008 |
disl | -0.007 |
ants | -0.007 |
Token | it |
Token ID | 340 |
Feature activation | -6.46e-03 |
Pos logit contributions | |
enough | +0.012 |
once | +0.012 |
best | +0.011 |
moments | +0.011 |
quality | +0.010 |
Neg logit contributions | |
ramids | -0.017 |
agons | -0.016 |
ants | -0.016 |
bell | -0.015 |
alks | -0.015 |
Token | again |
Token ID | 757 |
Feature activation | -6.29e-02 |
Pos logit contributions | |
enough | +0.007 |
moments | +0.006 |
once | +0.006 |
best | +0.006 |
joy | +0.006 |
Neg logit contributions | |
agons | -0.012 |
ramids | -0.011 |
ants | -0.011 |
cases | -0.011 |
issors | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -1.75e-03 |
Pos logit contributions | |
enough | +0.091 |
once | +0.083 |
best | +0.083 |
moments | +0.080 |
quality | +0.071 |
Neg logit contributions | |
ramids | -0.089 |
ants | -0.088 |
agons | -0.088 |
alks | -0.087 |
bell | -0.085 |
Token | |
Token ID | 220 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
once | +0.002 |
enough | +0.002 |
again | +0.002 |
best | +0.002 |
moments | +0.002 |
Neg logit contributions | |
agons | -0.004 |
ramids | -0.003 |
pier | -0.003 |
rus | -0.003 |
ants | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +1.46e-03 |
Pos logit contributions | |
agons | +0.043 |
pier | +0.041 |
disl | +0.040 |
curb | +0.040 |
ramids | +0.040 |
Neg logit contributions | |
once | -0.013 |
enough | -0.011 |
best | -0.009 |
again | -0.009 |
more | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +6.00e-03 |
Pos logit contributions | |
agons | +0.003 |
ramids | +0.003 |
alks | +0.003 |
disl | +0.003 |
pier | +0.003 |
Neg logit contributions | |
enough | -0.001 |
once | -0.001 |
best | -0.001 |
again | -0.001 |
more | -0.001 |
Token | P |
Token ID | 47 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
agons | +0.012 |
ramids | +0.011 |
alks | +0.011 |
pier | +0.010 |
disl | +0.010 |
Neg logit contributions | |
once | -0.008 |
enough | -0.008 |
again | -0.007 |
best | -0.007 |
more | -0.007 |