Token | amazed |
Token ID | 24562 |
Feature activation | -4.81e-01 |
Pos logit contributions | |
agging | +0.004 |
five | +0.004 |
mine | +0.004 |
cheek | +0.004 |
liking | +0.003 |
Neg logit contributions | |
Pig | -0.004 |
wood | -0.004 |
Ally | -0.004 |
Bert | -0.004 |
Beth | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -3.23e-01 |
Pos logit contributions | |
agging | +0.020 |
five | +0.019 |
cheek | +0.018 |
mine | +0.017 |
goodness | +0.017 |
Neg logit contributions | |
Pig | -0.020 |
Tweet | -0.017 |
known | -0.017 |
Beth | -0.017 |
gy | -0.017 |
Token | He |
Token ID | 679 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
five | +0.012 |
arl | +0.010 |
agging | +0.010 |
cheek | +0.010 |
EEP | +0.010 |
Neg logit contributions | |
Pig | -0.014 |
known | -0.013 |
Madd | -0.013 |
Tweet | -0.013 |
house | -0.013 |
Token | watched |
Token ID | 7342 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
Pig | +0.005 |
Tweet | +0.005 |
Beth | +0.005 |
Bunny | +0.004 |
Ally | +0.004 |
Neg logit contributions | |
five | -0.007 |
agging | -0.007 |
cheek | -0.007 |
mine | -0.006 |
fourteen | -0.006 |
Token | as |
Token ID | 355 |
Feature activation | -6.91e-01 |
Pos logit contributions | |
five | +0.055 |
agging | +0.053 |
cheek | +0.051 |
fourteen | +0.049 |
mine | +0.047 |
Neg logit contributions | |
Pig | -0.043 |
gy | -0.039 |
wood | -0.039 |
known | -0.039 |
Beth | -0.038 |
Token | Per |
Token ID | 2448 |
Feature activation | +6.02e+00 |
Pos logit contributions | |
five | +0.027 |
agging | +0.024 |
fourteen | +0.023 |
cheek | +0.023 |
arl | +0.023 |
Neg logit contributions | |
wood | -0.027 |
house | -0.027 |
gy | -0.026 |
dam | -0.026 |
Pig | -0.026 |
Token | c |
Token ID | 66 |
Feature activation | +5.39e+00 |
Pos logit contributions | |
Pig | +0.280 |
Tweet | +0.257 |
Bunny | +0.253 |
Beth | +0.252 |
Ally | +0.249 |
Neg logit contributions | |
five | -0.223 |
agging | -0.203 |
cheek | -0.194 |
fourteen | -0.182 |
mine | -0.177 |
Token | ival |
Token ID | 2473 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
Pig | +0.213 |
Beth | +0.191 |
Bunny | +0.188 |
Ally | +0.186 |
Tweet | +0.186 |
Neg logit contributions | |
five | -0.233 |
agging | -0.222 |
cheek | -0.206 |
fourteen | -0.203 |
goodness | -0.190 |
Token | cut |
Token ID | 2005 |
Feature activation | +3.94e-01 |
Pos logit contributions | |
Pig | +0.063 |
Suz | +0.057 |
known | +0.057 |
Bunny | +0.056 |
Madd | +0.056 |
Neg logit contributions | |
five | -0.069 |
agging | -0.062 |
fourteen | -0.060 |
mine | -0.059 |
cheek | -0.059 |
Token | out |
Token ID | 503 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
known | +0.015 |
Pig | +0.014 |
Ally | +0.013 |
dam | +0.013 |
gy | +0.013 |
Neg logit contributions | |
five | -0.016 |
agging | -0.015 |
fourteen | -0.014 |
goodness | -0.014 |
mine | -0.013 |
Token | two |
Token ID | 734 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
Pig | +0.012 |
Tweet | +0.011 |
known | +0.011 |
gy | +0.011 |
Bunny | +0.011 |
Neg logit contributions | |
five | -0.012 |
agging | -0.011 |
cheek | -0.010 |
fourteen | -0.010 |
mine | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -3.23e-01 |
Pos logit contributions | |
agging | +0.020 |
five | +0.019 |
cheek | +0.018 |
mine | +0.017 |
goodness | +0.017 |
Neg logit contributions | |
Pig | -0.020 |
Tweet | -0.017 |
known | -0.017 |
Beth | -0.017 |
gy | -0.017 |
Token | He |
Token ID | 679 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
five | +0.012 |
arl | +0.010 |
agging | +0.010 |
cheek | +0.010 |
EEP | +0.010 |
Neg logit contributions | |
Pig | -0.014 |
known | -0.013 |
Madd | -0.013 |
Tweet | -0.013 |
house | -0.013 |
Token | watched |
Token ID | 7342 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
Pig | +0.005 |
Tweet | +0.005 |
Beth | +0.005 |
Bunny | +0.004 |
Ally | +0.004 |
Neg logit contributions | |
five | -0.007 |
agging | -0.007 |
cheek | -0.007 |
mine | -0.006 |
fourteen | -0.006 |
Token | as |
Token ID | 355 |
Feature activation | -6.91e-01 |
Pos logit contributions | |
five | +0.055 |
agging | +0.053 |
cheek | +0.051 |
fourteen | +0.049 |
mine | +0.047 |
Neg logit contributions | |
Pig | -0.043 |
gy | -0.039 |
wood | -0.039 |
known | -0.039 |
Beth | -0.038 |
Token | Per |
Token ID | 2448 |
Feature activation | +6.02e+00 |
Pos logit contributions | |
five | +0.027 |
agging | +0.024 |
fourteen | +0.023 |
cheek | +0.023 |
arl | +0.023 |
Neg logit contributions | |
wood | -0.027 |
house | -0.027 |
gy | -0.026 |
dam | -0.026 |
Pig | -0.026 |
Token | c |
Token ID | 66 |
Feature activation | +5.39e+00 |
Pos logit contributions | |
Pig | +0.280 |
Tweet | +0.257 |
Bunny | +0.253 |
Beth | +0.252 |
Ally | +0.249 |
Neg logit contributions | |
five | -0.223 |
agging | -0.203 |
cheek | -0.194 |
fourteen | -0.182 |
mine | -0.177 |
Token | ival |
Token ID | 2473 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
Pig | +0.213 |
Beth | +0.191 |
Bunny | +0.188 |
Ally | +0.186 |
Tweet | +0.186 |
Neg logit contributions | |
five | -0.233 |
agging | -0.222 |
cheek | -0.206 |
fourteen | -0.203 |
goodness | -0.190 |
Token | cut |
Token ID | 2005 |
Feature activation | +3.94e-01 |
Pos logit contributions | |
Pig | +0.063 |
Suz | +0.057 |
known | +0.057 |
Bunny | +0.056 |
Madd | +0.056 |
Neg logit contributions | |
five | -0.069 |
agging | -0.062 |
fourteen | -0.060 |
mine | -0.059 |
cheek | -0.059 |
Token | out |
Token ID | 503 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
known | +0.015 |
Pig | +0.014 |
Ally | +0.013 |
dam | +0.013 |
gy | +0.013 |
Neg logit contributions | |
five | -0.016 |
agging | -0.015 |
fourteen | -0.014 |
goodness | -0.014 |
mine | -0.013 |
Token | two |
Token ID | 734 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
Pig | +0.012 |
Tweet | +0.011 |
known | +0.011 |
gy | +0.011 |
Bunny | +0.011 |
Neg logit contributions | |
five | -0.012 |
agging | -0.011 |
cheek | -0.010 |
fourteen | -0.010 |
mine | -0.010 |
Token | shapes |
Token ID | 15268 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
five | +0.006 |
agging | +0.006 |
cheek | +0.006 |
arl | +0.005 |
fourteen | +0.005 |
Neg logit contributions | |
Pig | -0.006 |
Tweet | -0.005 |
Beth | -0.005 |
Bunny | -0.005 |
Bert | -0.005 |
Token | They |
Token ID | 1119 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
five | +0.005 |
cheek | +0.005 |
agging | +0.005 |
fourteen | +0.005 |
arl | +0.005 |
Neg logit contributions | |
Pig | -0.004 |
gy | -0.004 |
known | -0.004 |
wood | -0.004 |
Beth | -0.003 |
Token | played |
Token ID | 2826 |
Feature activation | +1.97e-01 |
Pos logit contributions | |
Pig | +0.008 |
wood | +0.007 |
Beth | +0.007 |
Ally | +0.007 |
Tweet | +0.007 |
Neg logit contributions | |
five | -0.008 |
agging | -0.008 |
cheek | -0.008 |
fourteen | -0.007 |
ige | -0.007 |
Token | catch |
Token ID | 4929 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
Pig | +0.006 |
Tweet | +0.006 |
Bunny | +0.005 |
Beth | +0.005 |
Ally | +0.005 |
Neg logit contributions | |
five | -0.010 |
agging | -0.010 |
cheek | -0.009 |
fourteen | -0.009 |
Deal | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
five | +0.045 |
agging | +0.040 |
cheek | +0.039 |
mine | +0.037 |
goodness | +0.037 |
Neg logit contributions | |
Pig | -0.042 |
known | -0.037 |
wood | -0.036 |
Beth | -0.036 |
gy | -0.036 |
Token | tug |
Token ID | 27762 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
five | +0.005 |
cheek | +0.004 |
agging | +0.004 |
mine | +0.004 |
fourteen | +0.004 |
Neg logit contributions | |
Pig | -0.005 |
Beth | -0.004 |
wood | -0.004 |
Ally | -0.004 |
Tweet | -0.004 |
Token | - |
Token ID | 12 |
Feature activation | +5.35e+00 |
Pos logit contributions | |
five | +0.010 |
agging | +0.009 |
cheek | +0.009 |
Deal | +0.008 |
sidelines | +0.008 |
Neg logit contributions | |
Pig | -0.009 |
Tweet | -0.008 |
Bunny | -0.008 |
Beth | -0.008 |
gy | -0.008 |
Token | of |
Token ID | 1659 |
Feature activation | +6.56e-01 |
Pos logit contributions | |
Pig | +0.275 |
Beth | +0.250 |
Tweet | +0.249 |
Bunny | +0.248 |
Ally | +0.246 |
Neg logit contributions | |
five | -0.180 |
agging | -0.156 |
cheek | -0.148 |
fourteen | -0.142 |
sidelines | -0.131 |
Token | - |
Token ID | 12 |
Feature activation | +4.99e+00 |
Pos logit contributions | |
Pig | +0.027 |
Tweet | +0.024 |
Beth | +0.023 |
Bunny | +0.023 |
Ally | +0.023 |
Neg logit contributions | |
five | -0.027 |
agging | -0.026 |
cheek | -0.026 |
fourteen | -0.024 |
mine | -0.023 |
Token | war |
Token ID | 5767 |
Feature activation | +3.43e-01 |
Pos logit contributions | |
Pig | +0.256 |
Tweet | +0.234 |
Beth | +0.232 |
Bunny | +0.231 |
Ally | +0.229 |
Neg logit contributions | |
five | -0.163 |
agging | -0.147 |
cheek | -0.140 |
fourteen | -0.130 |
mine | -0.123 |
Token | . |
Token ID | 13 |
Feature activation | -2.72e-02 |
Pos logit contributions | |
Pig | +0.015 |
known | +0.014 |
Ally | +0.013 |
Beth | +0.013 |
Bunny | +0.013 |
Neg logit contributions | |
five | -0.014 |
agging | -0.014 |
cheek | -0.014 |
fourteen | -0.012 |
ater | -0.011 |
Token | The |
Token ID | 383 |
Feature activation | +3.72e-01 |
Pos logit contributions | |
five | +0.001 |
fourteen | +0.001 |
Deal | +0.001 |
arl | +0.001 |
agging | +0.001 |
Neg logit contributions | |
Pig | -0.001 |
gy | -0.001 |
wood | -0.001 |
known | -0.001 |
Beth | -0.001 |
Token | hug |
Token ID | 16225 |
Feature activation | -6.61e-01 |
Pos logit contributions | |
Pig | +0.012 |
Tweet | +0.011 |
Beth | +0.011 |
Bunny | +0.011 |
Ally | +0.011 |
Neg logit contributions | |
five | -0.008 |
agging | -0.008 |
cheek | -0.007 |
fourteen | -0.007 |
mine | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | +5.81e-01 |
Pos logit contributions | |
five | +0.026 |
cheek | +0.025 |
agging | +0.025 |
fourteen | +0.022 |
mine | +0.022 |
Neg logit contributions | |
Pig | -0.027 |
Tweet | -0.026 |
Ally | -0.025 |
known | -0.025 |
Beth | -0.025 |
Token | lots |
Token ID | 6041 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
Pig | +0.022 |
gy | +0.022 |
wood | +0.022 |
dam | +0.021 |
Tweet | +0.020 |
Neg logit contributions | |
five | -0.023 |
cheek | -0.021 |
mine | -0.019 |
fourteen | -0.019 |
goodness | -0.019 |
Token | of |
Token ID | 286 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
five | +0.046 |
agging | +0.045 |
cheek | +0.044 |
Deal | +0.041 |
fourteen | +0.041 |
Neg logit contributions | |
Pig | -0.039 |
Tweet | -0.035 |
Bunny | -0.034 |
Beth | -0.034 |
Ally | -0.033 |
Token | up |
Token ID | 510 |
Feature activation | +8.83e-01 |
Pos logit contributions | |
Pig | +0.081 |
gy | +0.073 |
Tweet | +0.073 |
wood | +0.072 |
Beth | +0.072 |
Neg logit contributions | |
five | -0.069 |
agging | -0.066 |
cheek | -0.064 |
mine | -0.057 |
fourteen | -0.057 |
Token | - |
Token ID | 12 |
Feature activation | +5.22e+00 |
Pos logit contributions | |
Pig | +0.036 |
Tweet | +0.032 |
gy | +0.032 |
Ally | +0.032 |
wood | +0.032 |
Neg logit contributions | |
agging | -0.036 |
five | -0.036 |
cheek | -0.033 |
mine | -0.032 |
fourteen | -0.031 |
Token | downs |
Token ID | 30371 |
Feature activation | +2.00e+00 |
Pos logit contributions | |
Pig | +0.248 |
Bunny | +0.223 |
Tweet | +0.223 |
Beth | +0.221 |
Bert | +0.220 |
Neg logit contributions | |
five | -0.202 |
agging | -0.172 |
cheek | -0.167 |
fourteen | -0.155 |
leaf | -0.152 |
Token | to |
Token ID | 284 |
Feature activation | +6.43e-01 |
Pos logit contributions | |
Pig | +0.074 |
gy | +0.068 |
known | +0.068 |
Tweet | +0.067 |
Ally | +0.066 |
Neg logit contributions | |
five | -0.089 |
cheek | -0.083 |
agging | -0.081 |
mine | -0.075 |
goodness | -0.075 |
Token | make |
Token ID | 787 |
Feature activation | +7.39e-01 |
Pos logit contributions | |
Pig | +0.027 |
Tweet | +0.025 |
Beth | +0.024 |
Bunny | +0.024 |
Ally | +0.024 |
Neg logit contributions | |
five | -0.026 |
agging | -0.025 |
cheek | -0.024 |
fourteen | -0.022 |
mine | -0.022 |
Token | it |
Token ID | 340 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
gy | +0.030 |
Pig | +0.029 |
wood | +0.028 |
known | +0.027 |
house | +0.027 |
Neg logit contributions | |
five | -0.029 |
goodness | -0.027 |
cheek | -0.027 |
agging | -0.027 |
mine | -0.025 |
Token | feel |
Token ID | 1254 |
Feature activation | +2.43e-03 |
Pos logit contributions | |
Pig | +0.008 |
Ally | +0.007 |
Beth | +0.007 |
Bert | +0.007 |
wood | +0.007 |
Neg logit contributions | |
cheek | -0.009 |
five | -0.009 |
agging | -0.008 |
mine | -0.008 |
liking | -0.007 |
Token | playing |
Token ID | 2712 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
mine | +0.006 |
five | +0.006 |
agging | +0.006 |
arrest | +0.005 |
liking | +0.005 |
Neg logit contributions | |
Pig | -0.005 |
wood | -0.005 |
gy | -0.005 |
Ally | -0.005 |
Tweet | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | +4.92e-01 |
Pos logit contributions | |
five | +0.007 |
agging | +0.007 |
cheek | +0.006 |
mine | +0.006 |
goodness | +0.006 |
Neg logit contributions | |
Pig | -0.006 |
known | -0.005 |
wood | -0.005 |
Ally | -0.005 |
Beth | -0.005 |
Token | game |
Token ID | 983 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
Pig | +0.021 |
Tweet | +0.019 |
Bunny | +0.019 |
Beth | +0.019 |
Ally | +0.019 |
Neg logit contributions | |
five | -0.019 |
agging | -0.019 |
cheek | -0.018 |
fourteen | -0.017 |
Deal | -0.017 |
Token | of |
Token ID | 286 |
Feature activation | +9.03e-01 |
Pos logit contributions | |
Pig | +0.009 |
Tweet | +0.007 |
known | +0.007 |
Ally | +0.007 |
gy | +0.007 |
Neg logit contributions | |
five | -0.010 |
agging | -0.010 |
cheek | -0.009 |
mine | -0.009 |
fourteen | -0.009 |
Token | kick |
Token ID | 4829 |
Feature activation | +2.56e-01 |
Pos logit contributions | |
Pig | +0.038 |
Tweet | +0.036 |
Beth | +0.035 |
Bunny | +0.034 |
Ally | +0.034 |
Neg logit contributions | |
five | -0.035 |
agging | -0.034 |
cheek | -0.033 |
arl | -0.031 |
fourteen | -0.031 |
Token | - |
Token ID | 12 |
Feature activation | +5.05e+00 |
Pos logit contributions | |
Pig | +0.010 |
known | +0.009 |
gy | +0.009 |
Beth | +0.009 |
Ally | +0.009 |
Neg logit contributions | |
five | -0.011 |
agging | -0.011 |
mine | -0.010 |
cheek | -0.010 |
la | -0.009 |
Token | ball |
Token ID | 1894 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
Pig | +0.235 |
Tweet | +0.218 |
Beth | +0.212 |
known | +0.212 |
wood | +0.211 |
Neg logit contributions | |
agging | -0.169 |
five | -0.166 |
cheek | -0.166 |
Deal | -0.151 |
fourteen | -0.146 |
Token | . |
Token ID | 13 |
Feature activation | +9.69e-01 |
Pos logit contributions | |
Pig | +0.044 |
Tweet | +0.039 |
known | +0.038 |
Beth | +0.038 |
Ally | +0.038 |
Neg logit contributions | |
five | -0.058 |
agging | -0.056 |
cheek | -0.054 |
mine | -0.051 |
fourteen | -0.050 |
Token | Lily |
Token ID | 20037 |
Feature activation | +4.75e-01 |
Pos logit contributions | |
Pig | +0.033 |
known | +0.029 |
Tweet | +0.028 |
wood | +0.028 |
Ally | +0.028 |
Neg logit contributions | |
five | -0.047 |
cheek | -0.045 |
agging | -0.044 |
mine | -0.042 |
fourteen | -0.041 |
Token | was |
Token ID | 373 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
Pig | +0.017 |
wood | +0.016 |
Tweet | +0.015 |
darker | +0.015 |
Ally | +0.015 |
Neg logit contributions | |
five | -0.021 |
cheek | -0.020 |
agging | -0.019 |
mine | -0.019 |
ige | -0.018 |
Token | so |
Token ID | 523 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
wood | +0.009 |
Pig | +0.008 |
gy | +0.007 |
bul | +0.007 |
Bert | +0.007 |
Neg logit contributions | |
liking | -0.007 |
mine | -0.007 |
agging | -0.006 |
five | -0.006 |
cheek | -0.006 |
Token | which |
Token ID | 543 |
Feature activation | -6.27e-03 |
Pos logit contributions | |
Pig | +0.000 |
Tweet | +0.000 |
Bunny | +0.000 |
Beth | +0.000 |
Ally | +0.000 |
Neg logit contributions | |
five | -0.000 |
agging | -0.000 |
cheek | -0.000 |
fourteen | -0.000 |
mine | -0.000 |
Token | tw |
Token ID | 665 |
Feature activation | +6.57e-01 |
Pos logit contributions | |
five | +0.000 |
mine | +0.000 |
cheek | +0.000 |
agging | +0.000 |
fourteen | +0.000 |
Neg logit contributions | |
Pig | -0.000 |
gy | -0.000 |
wood | -0.000 |
Tweet | -0.000 |
Bert | -0.000 |
Token | irled |
Token ID | 49376 |
Feature activation | +8.92e-01 |
Pos logit contributions | |
Pig | +0.024 |
Tweet | +0.022 |
Bunny | +0.021 |
Beth | +0.021 |
Ally | +0.021 |
Neg logit contributions | |
five | -0.029 |
agging | -0.029 |
cheek | -0.028 |
fourteen | -0.026 |
mine | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | +5.83e-01 |
Pos logit contributions | |
Pig | +0.033 |
known | +0.029 |
Tweet | +0.027 |
Bunny | +0.027 |
wood | +0.027 |
Neg logit contributions | |
five | -0.041 |
agging | -0.041 |
cheek | -0.039 |
mine | -0.038 |
la | -0.036 |
Token | t |
Token ID | 256 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
Pig | +0.023 |
wood | +0.020 |
gy | +0.019 |
Bunny | +0.019 |
Tweet | +0.019 |
Neg logit contributions | |
five | -0.025 |
cheek | -0.024 |
agging | -0.023 |
la | -0.023 |
mine | -0.023 |
Token | - |
Token ID | 12 |
Feature activation | +5.04e+00 |
Pos logit contributions | |
Pig | +0.048 |
Tweet | +0.043 |
Beth | +0.043 |
Bunny | +0.042 |
gy | +0.042 |
Neg logit contributions | |
five | -0.052 |
agging | -0.050 |
cheek | -0.048 |
fourteen | -0.045 |
Deal | -0.044 |
Token | shirts |
Token ID | 23231 |
Feature activation | +5.96e-01 |
Pos logit contributions | |
Pig | +0.225 |
Tweet | +0.206 |
Beth | +0.202 |
Bunny | +0.200 |
Ally | +0.199 |
Neg logit contributions | |
five | -0.186 |
agging | -0.181 |
cheek | -0.175 |
fourteen | -0.159 |
Deal | -0.158 |
Token | that |
Token ID | 326 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
Pig | +0.020 |
Tweet | +0.018 |
Bunny | +0.017 |
Ally | +0.017 |
Beth | +0.017 |
Neg logit contributions | |
five | -0.029 |
agging | -0.028 |
cheek | -0.027 |
fourteen | -0.025 |
mine | -0.025 |
Token | jumped |
Token ID | 11687 |
Feature activation | -4.21e-01 |
Pos logit contributions | |
gy | +0.054 |
Pig | +0.053 |
dam | +0.052 |
wood | +0.052 |
Ally | +0.049 |
Neg logit contributions | |
five | -0.055 |
cheek | -0.054 |
mine | -0.054 |
la | -0.050 |
goodness | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
agging | +0.019 |
five | +0.018 |
cheek | +0.018 |
mine | +0.018 |
fourteen | +0.017 |
Neg logit contributions | |
Pig | -0.015 |
known | -0.014 |
Tweet | -0.013 |
Ally | -0.013 |
wood | -0.013 |
Token | He |
Token ID | 679 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
Pig | +0.006 |
known | +0.005 |
wood | +0.005 |
gy | +0.005 |
Tweet | +0.005 |
Neg logit contributions | |
five | -0.008 |
cheek | -0.007 |
agging | -0.007 |
Deal | -0.007 |
la | -0.007 |
Token | she |
Token ID | 673 |
Feature activation | +8.62e-01 |
Pos logit contributions | |
cheek | +0.034 |
agging | +0.031 |
five | +0.031 |
goodness | +0.029 |
mine | +0.028 |
Neg logit contributions | |
Pig | -0.035 |
wood | -0.034 |
gy | -0.033 |
dam | -0.032 |
house | -0.031 |
Token | stood |
Token ID | 6204 |
Feature activation | +4.07e-01 |
Pos logit contributions | |
Pig | +0.030 |
Tweet | +0.027 |
Ally | +0.027 |
Beth | +0.026 |
wood | +0.026 |
Neg logit contributions | |
five | -0.040 |
agging | -0.037 |
cheek | -0.037 |
fourteen | -0.034 |
la | -0.034 |
Token | up |
Token ID | 510 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
Pig | +0.015 |
Tweet | +0.014 |
wood | +0.014 |
known | +0.014 |
Ally | +0.013 |
Neg logit contributions | |
agging | -0.018 |
cheek | -0.017 |
mine | -0.015 |
goodness | -0.015 |
ater | -0.014 |
Token | and |
Token ID | 290 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
Pig | +0.013 |
wood | +0.012 |
known | +0.012 |
Tweet | +0.012 |
gy | +0.012 |
Neg logit contributions | |
agging | -0.010 |
cheek | -0.010 |
five | -0.009 |
mine | -0.009 |
fourteen | -0.009 |
Token | tip |
Token ID | 8171 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
Pig | +0.009 |
Tweet | +0.008 |
wood | +0.008 |
Ally | +0.008 |
dam | +0.008 |
Neg logit contributions | |
cheek | -0.011 |
five | -0.011 |
mine | -0.010 |
la | -0.010 |
agging | -0.009 |
Token | - |
Token ID | 12 |
Feature activation | +5.01e+00 |
Pos logit contributions | |
Pig | +0.010 |
known | +0.009 |
Tweet | +0.009 |
Ally | +0.009 |
wood | +0.009 |
Neg logit contributions | |
agging | -0.010 |
five | -0.010 |
cheek | -0.010 |
goodness | -0.009 |
fourteen | -0.009 |
Token | to |
Token ID | 1462 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
Pig | +0.209 |
Tweet | +0.189 |
Bunny | +0.185 |
Beth | +0.184 |
Ally | +0.182 |
Neg logit contributions | |
five | -0.207 |
agging | -0.191 |
cheek | -0.188 |
fourteen | -0.176 |
mine | -0.169 |
Token | ed |
Token ID | 276 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
five | +0.011 |
agging | +0.010 |
cheek | +0.010 |
fourteen | +0.009 |
mine | +0.009 |
Neg logit contributions | |
Pig | -0.010 |
Tweet | -0.009 |
Bunny | -0.008 |
Beth | -0.008 |
Ally | -0.008 |
Token | out |
Token ID | 503 |
Feature activation | +5.67e-01 |
Pos logit contributions | |
Pig | +0.056 |
known | +0.051 |
Tweet | +0.049 |
dam | +0.049 |
gy | +0.049 |
Neg logit contributions | |
five | -0.059 |
cheek | -0.055 |
agging | -0.054 |
goodness | -0.053 |
mine | -0.052 |
Token | of |
Token ID | 286 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
Pig | +0.024 |
gy | +0.022 |
wood | +0.022 |
Tweet | +0.022 |
known | +0.021 |
Neg logit contributions | |
five | -0.022 |
agging | -0.022 |
cheek | -0.021 |
fourteen | -0.019 |
mine | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | +4.08e-01 |
Pos logit contributions | |
Pig | +0.061 |
gy | +0.057 |
wood | +0.055 |
Tweet | +0.053 |
house | +0.052 |
Neg logit contributions | |
five | -0.075 |
mine | -0.072 |
cheek | -0.072 |
agging | -0.071 |
fourteen | -0.069 |
Token | play |
Token ID | 711 |
Feature activation | +4.48e-01 |
Pos logit contributions | |
Pig | +0.029 |
Tweet | +0.026 |
Beth | +0.026 |
Bunny | +0.026 |
Ally | +0.026 |
Neg logit contributions | |
five | -0.028 |
agging | -0.028 |
cheek | -0.027 |
fourteen | -0.024 |
mine | -0.024 |
Token | a |
Token ID | 257 |
Feature activation | +8.90e-01 |
Pos logit contributions | |
Pig | +0.017 |
wood | +0.015 |
Bunny | +0.014 |
known | +0.014 |
Ally | +0.014 |
Neg logit contributions | |
five | -0.021 |
agging | -0.020 |
cheek | -0.019 |
mine | -0.019 |
goodness | -0.018 |
Token | game |
Token ID | 983 |
Feature activation | +7.89e-02 |
Pos logit contributions | |
Pig | +0.040 |
Tweet | +0.037 |
Beth | +0.036 |
Bunny | +0.036 |
Ally | +0.035 |
Neg logit contributions | |
five | -0.033 |
agging | -0.032 |
cheek | -0.030 |
fourteen | -0.028 |
Deal | -0.028 |
Token | of |
Token ID | 286 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
Pig | +0.003 |
Tweet | +0.003 |
Bunny | +0.002 |
Beth | +0.002 |
Ally | +0.002 |
Neg logit contributions | |
five | -0.004 |
agging | -0.004 |
cheek | -0.003 |
fourteen | -0.003 |
mine | -0.003 |
Token | tug |
Token ID | 27762 |
Feature activation | -6.41e-01 |
Pos logit contributions | |
Pig | +0.049 |
Tweet | +0.044 |
Beth | +0.043 |
Bunny | +0.043 |
Ally | +0.042 |
Neg logit contributions | |
five | -0.048 |
agging | -0.046 |
cheek | -0.045 |
fourteen | -0.041 |
mine | -0.041 |
Token | - |
Token ID | 12 |
Feature activation | +5.00e+00 |
Pos logit contributions | |
five | +0.029 |
agging | +0.028 |
cheek | +0.027 |
fourteen | +0.025 |
mine | +0.025 |
Neg logit contributions | |
Pig | -0.024 |
Tweet | -0.022 |
Beth | -0.021 |
Bunny | -0.021 |
Ally | -0.021 |
Token | o |
Token ID | 78 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
Pig | +0.242 |
Tweet | +0.222 |
Beth | +0.218 |
Bunny | +0.217 |
Ally | +0.215 |
Neg logit contributions | |
five | -0.169 |
agging | -0.163 |
cheek | -0.156 |
fourteen | -0.141 |
mine | -0.138 |
Token | - |
Token ID | 12 |
Feature activation | +4.54e+00 |
Pos logit contributions | |
Pig | +0.054 |
Tweet | +0.048 |
Beth | +0.047 |
Bunny | +0.047 |
gy | +0.047 |
Neg logit contributions | |
five | -0.065 |
agging | -0.063 |
cheek | -0.061 |
fourteen | -0.057 |
Deal | -0.057 |
Token | war |
Token ID | 5767 |
Feature activation | +7.21e-02 |
Pos logit contributions | |
Pig | +0.209 |
Tweet | +0.191 |
Beth | +0.187 |
Bunny | +0.186 |
gy | +0.185 |
Neg logit contributions | |
five | -0.159 |
agging | -0.158 |
cheek | -0.152 |
Deal | -0.138 |
fourteen | -0.136 |
Token | . |
Token ID | 13 |
Feature activation | +2.72e-01 |
Pos logit contributions | |
Pig | +0.003 |
known | +0.003 |
Tweet | +0.003 |
Ally | +0.003 |
Beth | +0.003 |
Neg logit contributions | |
five | -0.003 |
agging | -0.003 |
cheek | -0.003 |
mine | -0.003 |
fourteen | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -2.77e-01 |
Pos logit contributions | |
known | +0.011 |
Pig | +0.011 |
darker | +0.010 |
gy | +0.010 |
rou | +0.010 |
Neg logit contributions | |
five | -0.011 |
fourteen | -0.009 |
EEP | -0.009 |
cheek | -0.009 |
goodness | -0.008 |
Token | catch |
Token ID | 4929 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
Pig | +0.006 |
Tweet | +0.006 |
Bunny | +0.005 |
Beth | +0.005 |
Ally | +0.005 |
Neg logit contributions | |
five | -0.010 |
agging | -0.010 |
cheek | -0.009 |
fourteen | -0.009 |
Deal | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
five | +0.045 |
agging | +0.040 |
cheek | +0.039 |
mine | +0.037 |
goodness | +0.037 |
Neg logit contributions | |
Pig | -0.042 |
known | -0.037 |
wood | -0.036 |
Beth | -0.036 |
gy | -0.036 |
Token | tug |
Token ID | 27762 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
five | +0.005 |
cheek | +0.004 |
agging | +0.004 |
mine | +0.004 |
fourteen | +0.004 |
Neg logit contributions | |
Pig | -0.005 |
Beth | -0.004 |
wood | -0.004 |
Ally | -0.004 |
Tweet | -0.004 |
Token | - |
Token ID | 12 |
Feature activation | +5.35e+00 |
Pos logit contributions | |
five | +0.010 |
agging | +0.009 |
cheek | +0.009 |
Deal | +0.008 |
sidelines | +0.008 |
Neg logit contributions | |
Pig | -0.009 |
Tweet | -0.008 |
Bunny | -0.008 |
Beth | -0.008 |
gy | -0.008 |
Token | of |
Token ID | 1659 |
Feature activation | +6.56e-01 |
Pos logit contributions | |
Pig | +0.275 |
Beth | +0.250 |
Tweet | +0.249 |
Bunny | +0.248 |
Ally | +0.246 |
Neg logit contributions | |
five | -0.180 |
agging | -0.156 |
cheek | -0.148 |
fourteen | -0.142 |
sidelines | -0.131 |
Token | - |
Token ID | 12 |
Feature activation | +4.99e+00 |
Pos logit contributions | |
Pig | +0.027 |
Tweet | +0.024 |
Beth | +0.023 |
Bunny | +0.023 |
Ally | +0.023 |
Neg logit contributions | |
five | -0.027 |
agging | -0.026 |
cheek | -0.026 |
fourteen | -0.024 |
mine | -0.023 |
Token | war |
Token ID | 5767 |
Feature activation | +3.43e-01 |
Pos logit contributions | |
Pig | +0.256 |
Tweet | +0.234 |
Beth | +0.232 |
Bunny | +0.231 |
Ally | +0.229 |
Neg logit contributions | |
five | -0.163 |
agging | -0.147 |
cheek | -0.140 |
fourteen | -0.130 |
mine | -0.123 |
Token | . |
Token ID | 13 |
Feature activation | -2.72e-02 |
Pos logit contributions | |
Pig | +0.015 |
known | +0.014 |
Ally | +0.013 |
Beth | +0.013 |
Bunny | +0.013 |
Neg logit contributions | |
five | -0.014 |
agging | -0.014 |
cheek | -0.014 |
fourteen | -0.012 |
ater | -0.011 |
Token | The |
Token ID | 383 |
Feature activation | +3.72e-01 |
Pos logit contributions | |
five | +0.001 |
fourteen | +0.001 |
Deal | +0.001 |
arl | +0.001 |
agging | +0.001 |
Neg logit contributions | |
Pig | -0.001 |
gy | -0.001 |
wood | -0.001 |
known | -0.001 |
Beth | -0.001 |
Token | fox |
Token ID | 21831 |
Feature activation | -4.37e-01 |
Pos logit contributions | |
Pig | +0.014 |
Tweet | +0.012 |
Bunny | +0.012 |
Beth | +0.012 |
Ally | +0.012 |
Neg logit contributions | |
five | -0.017 |
agging | -0.016 |
cheek | -0.016 |
fourteen | -0.015 |
Deal | -0.015 |
Token | was |
Token ID | 373 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
five | +0.019 |
cheek | +0.017 |
agging | +0.017 |
fourteen | +0.016 |
la | +0.016 |
Neg logit contributions | |
Pig | -0.018 |
wood | -0.016 |
Beth | -0.016 |
Ally | -0.016 |
Tweet | -0.016 |
Token | would |
Token ID | 561 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
five | +0.002 |
cheek | +0.001 |
agging | +0.001 |
fourteen | +0.001 |
mine | +0.001 |
Neg logit contributions | |
Pig | -0.001 |
Tweet | -0.001 |
Beth | -0.001 |
Ally | -0.001 |
wood | -0.001 |
Token | always |
Token ID | 1464 |
Feature activation | +8.63e-02 |
Pos logit contributions | |
five | +0.010 |
agging | +0.010 |
cheek | +0.010 |
arl | +0.009 |
Deal | +0.009 |
Neg logit contributions | |
Pig | -0.010 |
Tweet | -0.009 |
Bunny | -0.008 |
Beth | -0.008 |
known | -0.008 |
Token | say |
Token ID | 910 |
Feature activation | -3.33e-01 |
Pos logit contributions | |
Pig | +0.004 |
Tweet | +0.003 |
known | +0.003 |
Bunny | +0.003 |
Beth | +0.003 |
Neg logit contributions | |
five | -0.004 |
agging | -0.003 |
cheek | -0.003 |
fourteen | -0.003 |
arl | -0.003 |
Token | ' |
Token ID | 705 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
five | +0.015 |
agging | +0.014 |
cheek | +0.014 |
sidelines | +0.013 |
fourteen | +0.013 |
Neg logit contributions | |
Pig | -0.013 |
Tweet | -0.011 |
Bunny | -0.011 |
Beth | -0.011 |
Ally | -0.011 |
Token | thank |
Token ID | 40716 |
Feature activation | +4.23e-01 |
Pos logit contributions | |
Pig | +0.090 |
Tweet | +0.082 |
Bunny | +0.081 |
Beth | +0.081 |
Ally | +0.080 |
Neg logit contributions | |
five | -0.078 |
agging | -0.070 |
cheek | -0.069 |
fourteen | -0.064 |
goodness | -0.063 |
Token | - |
Token ID | 12 |
Feature activation | +4.94e+00 |
Pos logit contributions | |
Pig | +0.025 |
Tweet | +0.023 |
Bunny | +0.022 |
Beth | +0.022 |
Ally | +0.022 |
Neg logit contributions | |
five | -0.010 |
agging | -0.010 |
cheek | -0.009 |
fourteen | -0.008 |
arl | -0.008 |
Token | you |
Token ID | 5832 |
Feature activation | +3.12e-01 |
Pos logit contributions | |
Pig | +0.251 |
known | +0.232 |
Tweet | +0.231 |
wood | +0.225 |
Ally | +0.225 |
Neg logit contributions | |
agging | -0.157 |
cheek | -0.142 |
Deal | -0.134 |
five | -0.133 |
arl | -0.128 |
Token | ' |
Token ID | 6 |
Feature activation | +9.50e-01 |
Pos logit contributions | |
Pig | +0.014 |
Ally | +0.012 |
Bunny | +0.012 |
Beth | +0.012 |
wood | +0.012 |
Neg logit contributions | |
agging | -0.012 |
five | -0.011 |
cheek | -0.011 |
sidelines | -0.011 |
arl | -0.010 |
Token | to |
Token ID | 284 |
Feature activation | +3.33e-01 |
Pos logit contributions | |
Pig | +0.036 |
Tweet | +0.033 |
Beth | +0.032 |
Bunny | +0.031 |
known | +0.031 |
Neg logit contributions | |
five | -0.042 |
agging | -0.040 |
cheek | -0.040 |
fourteen | -0.036 |
mine | -0.036 |
Token | his |
Token ID | 465 |
Feature activation | -3.54e-01 |
Pos logit contributions | |
Pig | +0.011 |
Tweet | +0.010 |
Beth | +0.010 |
Bunny | +0.009 |
Ally | +0.009 |
Neg logit contributions | |
five | -0.017 |
agging | -0.017 |
cheek | -0.016 |
arl | -0.015 |
fourteen | -0.015 |
Token | toast |
Token ID | 27805 |
Feature activation | -7.80e-01 |
Pos logit contributions | |
five | +0.016 |
agging | +0.015 |
cheek | +0.015 |
mine | +0.014 |
Deal | +0.013 |
Neg logit contributions | |
Pig | -0.013 |
Tweet | -0.012 |
Beth | -0.012 |
Ally | -0.012 |
known | -0.012 |
Token | push |
Token ID | 4574 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
Pig | +0.027 |
gy | +0.023 |
Tweet | +0.023 |
Bunny | +0.023 |
wood | +0.023 |
Neg logit contributions | |
five | -0.026 |
agging | -0.024 |
cheek | -0.023 |
mine | -0.022 |
goodness | -0.021 |
Token | - |
Token ID | 12 |
Feature activation | +4.27e+00 |
Pos logit contributions | |
five | +0.006 |
agging | +0.005 |
mine | +0.005 |
goodness | +0.005 |
cheek | +0.005 |
Neg logit contributions | |
Pig | -0.006 |
known | -0.006 |
Ally | -0.006 |
Tweet | -0.005 |
Bunny | -0.005 |
Token | ups |
Token ID | 4739 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
Pig | +0.188 |
Tweet | +0.170 |
Bunny | +0.168 |
Beth | +0.167 |
Ally | +0.165 |
Neg logit contributions | |
five | -0.171 |
agging | -0.157 |
cheek | -0.150 |
fourteen | -0.142 |
mine | -0.136 |
Token | and |
Token ID | 290 |
Feature activation | -6.35e-02 |
Pos logit contributions | |
Pig | +0.060 |
Tweet | +0.055 |
Ally | +0.052 |
Bunny | +0.052 |
known | +0.051 |
Neg logit contributions | |
five | -0.075 |
agging | -0.072 |
cheek | -0.070 |
fourteen | -0.065 |
mine | -0.065 |
Token | pull |
Token ID | 2834 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
five | +0.002 |
cheek | +0.002 |
agging | +0.002 |
mine | +0.002 |
fourteen | +0.002 |
Neg logit contributions | |
Pig | -0.003 |
Tweet | -0.003 |
Ally | -0.003 |
Bunny | -0.003 |
Beth | -0.002 |
Token | - |
Token ID | 12 |
Feature activation | +4.93e+00 |
Pos logit contributions | |
Pig | +0.005 |
known | +0.005 |
Ally | +0.004 |
Tweet | +0.004 |
gy | +0.004 |
Neg logit contributions | |
five | -0.004 |
agging | -0.004 |
mine | -0.004 |
fourteen | -0.004 |
goodness | -0.004 |
Token | ups |
Token ID | 4739 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
Pig | +0.210 |
Tweet | +0.188 |
Bunny | +0.186 |
Beth | +0.185 |
Ally | +0.183 |
Neg logit contributions | |
five | -0.207 |
agging | -0.186 |
cheek | -0.179 |
fourteen | -0.174 |
mine | -0.165 |
Token | . |
Token ID | 13 |
Feature activation | +4.15e-01 |
Pos logit contributions | |
Pig | +0.092 |
Tweet | +0.083 |
Ally | +0.080 |
known | +0.079 |
Bunny | +0.079 |
Neg logit contributions | |
five | -0.121 |
agging | -0.116 |
cheek | -0.114 |
fourteen | -0.106 |
mine | -0.106 |
Token | He |
Token ID | 679 |
Feature activation | +2.62e-01 |
Pos logit contributions | |
Pig | +0.015 |
known | +0.014 |
darker | +0.014 |
wood | +0.014 |
gy | +0.013 |
Neg logit contributions | |
five | -0.018 |
Deal | -0.016 |
fourteen | -0.016 |
sidelines | -0.015 |
agging | -0.015 |
Token | worked |
Token ID | 3111 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
Pig | +0.010 |
Ally | +0.010 |
Tweet | +0.009 |
wood | +0.009 |
Bunny | +0.009 |
Neg logit contributions | |
five | -0.012 |
fourteen | -0.010 |
mine | -0.010 |
cheek | -0.010 |
agging | -0.009 |
Token | really |
Token ID | 1107 |
Feature activation | -3.81e-01 |
Pos logit contributions | |
Pig | +0.006 |
Bunny | +0.005 |
Ally | +0.005 |
Tweet | +0.005 |
gy | +0.005 |
Neg logit contributions | |
agging | -0.005 |
five | -0.005 |
fourteen | -0.005 |
mine | -0.005 |
cheek | -0.005 |
Token | soft |
Token ID | 2705 |
Feature activation | -4.27e-01 |
Pos logit contributions | |
Pig | +0.007 |
Beth | +0.006 |
Bunny | +0.006 |
Ch | +0.006 |
Bert | +0.006 |
Neg logit contributions | |
five | -0.007 |
fourteen | -0.007 |
agging | -0.007 |
Deal | -0.006 |
arl | -0.006 |
Token | rug |
Token ID | 14477 |
Feature activation | +6.20e-01 |
Pos logit contributions | |
five | +0.017 |
agging | +0.017 |
Deal | +0.017 |
fourteen | +0.016 |
arl | +0.015 |
Neg logit contributions | |
Pig | -0.017 |
Beth | -0.016 |
Bunny | -0.016 |
gy | -0.016 |
Tweet | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | -7.70e-02 |
Pos logit contributions | |
Pig | +0.026 |
known | +0.024 |
Tweet | +0.022 |
Beth | +0.022 |
Ally | +0.022 |
Neg logit contributions | |
five | -0.026 |
agging | -0.025 |
cheek | -0.025 |
mine | -0.022 |
fourteen | -0.022 |
Token | even |
Token ID | 772 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
five | +0.003 |
cheek | +0.003 |
agging | +0.003 |
mine | +0.003 |
la | +0.003 |
Neg logit contributions | |
Pig | -0.003 |
wood | -0.003 |
Tweet | -0.003 |
known | -0.003 |
gy | -0.003 |
Token | hand |
Token ID | 1021 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
Pig | +0.005 |
Bert | +0.004 |
gy | +0.004 |
wood | +0.004 |
Ot | +0.004 |
Neg logit contributions | |
five | -0.005 |
cheek | -0.005 |
la | -0.005 |
mine | -0.004 |
agging | -0.004 |
Token | - |
Token ID | 12 |
Feature activation | +4.92e+00 |
Pos logit contributions | |
Pig | +0.047 |
known | +0.044 |
Tweet | +0.043 |
Ally | +0.042 |
Bert | +0.041 |
Neg logit contributions | |
five | -0.053 |
cheek | -0.052 |
agging | -0.052 |
mine | -0.047 |
fourteen | -0.045 |
Token | fed |
Token ID | 19082 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
Pig | +0.213 |
Tweet | +0.192 |
Bunny | +0.189 |
Beth | +0.187 |
Ally | +0.187 |
Neg logit contributions | |
five | -0.200 |
agging | -0.178 |
cheek | -0.175 |
fourteen | -0.167 |
goodness | -0.158 |
Token | the |
Token ID | 262 |
Feature activation | +4.94e-01 |
Pos logit contributions | |
Pig | +0.041 |
known | +0.039 |
Beth | +0.036 |
Bert | +0.035 |
Ally | +0.035 |
Neg logit contributions | |
five | -0.043 |
cheek | -0.042 |
agging | -0.040 |
mine | -0.039 |
goodness | -0.037 |
Token | chickens |
Token ID | 25972 |
Feature activation | +8.83e-01 |
Pos logit contributions | |
Pig | +0.021 |
Tweet | +0.019 |
Beth | +0.019 |
Bunny | +0.019 |
known | +0.019 |
Neg logit contributions | |
five | -0.020 |
agging | -0.019 |
cheek | -0.019 |
sidelines | -0.017 |
fourteen | -0.017 |
Token | her |
Token ID | 607 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
Pig | +0.030 |
Tweet | +0.026 |
Beth | +0.026 |
Bunny | +0.025 |
Ally | +0.025 |
Neg logit contributions | |
five | -0.042 |
agging | -0.041 |
cheek | -0.040 |
fourteen | -0.037 |
mine | -0.037 |
Token | kept |
Token ID | 4030 |
Feature activation | -2.46e-01 |
Pos logit contributions | |
Pig | +0.010 |
Tweet | +0.009 |
known | +0.009 |
Beth | +0.009 |
Ally | +0.009 |
Neg logit contributions | |
five | -0.006 |
cheek | -0.006 |
agging | -0.006 |
mine | -0.005 |
Deal | -0.005 |
Token | jumping |
Token ID | 14284 |
Feature activation | -8.92e-01 |
Pos logit contributions | |
five | +0.029 |
agging | +0.029 |
cheek | +0.028 |
mine | +0.025 |
goodness | +0.025 |
Neg logit contributions | |
Pig | -0.035 |
gy | -0.034 |
wood | -0.032 |
Ally | -0.031 |
Tweet | -0.031 |
Token | j |
Token ID | 474 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
five | +0.039 |
agging | +0.037 |
cheek | +0.036 |
mine | +0.035 |
fourteen | +0.035 |
Neg logit contributions | |
Pig | -0.035 |
known | -0.033 |
Beth | -0.030 |
Ally | -0.030 |
gy | -0.029 |
Token | acks |
Token ID | 4595 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
Pig | +0.010 |
Tweet | +0.009 |
Beth | +0.009 |
Bunny | +0.009 |
Ally | +0.009 |
Neg logit contributions | |
five | -0.006 |
agging | -0.005 |
cheek | -0.005 |
fourteen | -0.005 |
Deal | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | -3.25e-01 |
Pos logit contributions | |
five | +0.001 |
agging | +0.001 |
cheek | +0.001 |
fourteen | +0.001 |
mine | +0.001 |
Neg logit contributions | |
Pig | -0.001 |
known | -0.001 |
Tweet | -0.001 |
Ally | -0.001 |
gy | -0.001 |
Token | push |
Token ID | 4574 |
Feature activation | +8.04e-01 |
Pos logit contributions | |
five | +0.014 |
cheek | +0.013 |
agging | +0.012 |
mine | +0.012 |
la | +0.012 |
Neg logit contributions | |
Pig | -0.013 |
wood | -0.011 |
Tweet | -0.011 |
gy | -0.011 |
Ally | -0.011 |
Token | - |
Token ID | 12 |
Feature activation | +4.86e+00 |
Pos logit contributions | |
Pig | +0.033 |
known | +0.032 |
gy | +0.030 |
wood | +0.029 |
dam | +0.029 |
Neg logit contributions | |
five | -0.034 |
agging | -0.031 |
goodness | -0.030 |
cheek | -0.030 |
mine | -0.030 |
Token | ups |
Token ID | 4739 |
Feature activation | +2.47e+00 |
Pos logit contributions | |
Pig | +0.212 |
Tweet | +0.192 |
Beth | +0.188 |
Bunny | +0.188 |
Ally | +0.186 |
Neg logit contributions | |
five | -0.191 |
agging | -0.182 |
cheek | -0.175 |
fourteen | -0.162 |
mine | -0.158 |
Token | . |
Token ID | 13 |
Feature activation | +8.60e-02 |
Pos logit contributions | |
Pig | +0.082 |
Tweet | +0.074 |
Ally | +0.070 |
Bunny | +0.070 |
known | +0.070 |
Neg logit contributions | |
five | -0.121 |
agging | -0.117 |
cheek | -0.114 |
fourteen | -0.106 |
mine | -0.105 |
Token |
|
Token ID | 198 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
Pig | +0.003 |
known | +0.003 |
gy | +0.003 |
wood | +0.003 |
darker | +0.003 |
Neg logit contributions | |
five | -0.004 |
Deal | -0.003 |
cheek | -0.003 |
goodness | -0.003 |
agging | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
five | +0.008 |
agging | +0.007 |
cheek | +0.007 |
fourteen | +0.007 |
Deal | +0.007 |
Neg logit contributions | |
Pig | -0.005 |
Tweet | -0.004 |
gy | -0.004 |
wood | -0.004 |
Beth | -0.004 |
Token | Finally |
Token ID | 11158 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
cheek | +0.008 |
five | +0.008 |
agging | +0.008 |
mine | +0.007 |
fourteen | +0.007 |
Neg logit contributions | |
Pig | -0.008 |
Tweet | -0.007 |
Beth | -0.007 |
Ally | -0.007 |
known | -0.007 |
Token | over |
Token ID | 625 |
Feature activation | +2.72e-01 |
Pos logit contributions | |
Pig | +0.043 |
Tweet | +0.038 |
Beth | +0.037 |
Bunny | +0.037 |
Ally | +0.036 |
Neg logit contributions | |
five | -0.064 |
agging | -0.062 |
cheek | -0.061 |
fourteen | -0.057 |
Deal | -0.056 |
Token | and |
Token ID | 290 |
Feature activation | -6.98e-02 |
Pos logit contributions | |
Pig | +0.008 |
Tweet | +0.007 |
Beth | +0.007 |
Bunny | +0.007 |
Ally | +0.007 |
Neg logit contributions | |
five | -0.014 |
agging | -0.014 |
cheek | -0.013 |
Deal | -0.012 |
sidelines | -0.012 |
Token | get |
Token ID | 651 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
five | +0.003 |
agging | +0.003 |
cheek | +0.002 |
fourteen | +0.002 |
arl | +0.002 |
Neg logit contributions | |
Pig | -0.003 |
Tweet | -0.003 |
Beth | -0.003 |
Bunny | -0.003 |
Ally | -0.003 |
Token | some |
Token ID | 617 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
Pig | +0.043 |
Tweet | +0.038 |
gy | +0.038 |
Ally | +0.038 |
Beth | +0.038 |
Neg logit contributions | |
five | -0.060 |
cheek | -0.056 |
agging | -0.055 |
mine | -0.054 |
fourteen | -0.052 |
Token | ice |
Token ID | 4771 |
Feature activation | +7.87e-01 |
Pos logit contributions | |
Pig | +0.007 |
Tweet | +0.006 |
Bunny | +0.006 |
wood | +0.006 |
Beth | +0.006 |
Neg logit contributions | |
agging | -0.006 |
five | -0.006 |
cheek | -0.006 |
la | -0.005 |
Deal | -0.005 |
Token | - |
Token ID | 12 |
Feature activation | +4.80e+00 |
Pos logit contributions | |
Pig | +0.031 |
wood | +0.028 |
gy | +0.028 |
Tweet | +0.028 |
Bunny | +0.027 |
Neg logit contributions | |
five | -0.033 |
agging | -0.033 |
cheek | -0.031 |
sidelines | -0.030 |
Deal | -0.030 |
Token | cream |
Token ID | 36277 |
Feature activation | -4.39e-01 |
Pos logit contributions | |
Pig | +0.212 |
Tweet | +0.192 |
Beth | +0.190 |
Ally | +0.189 |
Bunny | +0.189 |
Neg logit contributions | |
five | -0.187 |
agging | -0.173 |
cheek | -0.169 |
fourteen | -0.158 |
mine | -0.152 |
Token | ! |
Token ID | 0 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
agging | +0.019 |
five | +0.018 |
cheek | +0.017 |
Deal | +0.016 |
sidelines | +0.016 |
Neg logit contributions | |
Pig | -0.017 |
wood | -0.016 |
Tweet | -0.015 |
Bunny | -0.015 |
Beth | -0.015 |
Token | â |
Token ID | 22940 |
Feature activation | +7.19e-01 |
Pos logit contributions | |
Pig | +0.004 |
known | +0.004 |
gy | +0.004 |
Tweet | +0.004 |
Ally | +0.003 |
Neg logit contributions | |
five | -0.006 |
cheek | -0.006 |
agging | -0.006 |
Deal | -0.005 |
mine | -0.005 |
Token | € |
Token ID | 26391 |
Feature activation | +9.15e-01 |
Pos logit contributions | |
Pig | +0.029 |
Tweet | +0.026 |
gy | +0.026 |
wood | +0.026 |
Beth | +0.025 |
Neg logit contributions | |
five | -0.029 |
agging | -0.028 |
cheek | -0.027 |
fourteen | -0.026 |
Deal | -0.026 |
Token | So |
Token ID | 1406 |
Feature activation | +6.94e-01 |
Pos logit contributions | |
Pig | +0.038 |
Tweet | +0.034 |
Beth | +0.034 |
Bunny | +0.034 |
Ally | +0.033 |
Neg logit contributions | |
five | -0.037 |
agging | -0.036 |
cheek | -0.035 |
fourteen | -0.032 |
mine | -0.031 |
Token | out |
Token ID | 503 |
Feature activation | -3.54e-01 |
Pos logit contributions | |
Pig | +0.037 |
Tweet | +0.033 |
Beth | +0.032 |
Bunny | +0.032 |
Ally | +0.032 |
Neg logit contributions | |
five | -0.047 |
agging | -0.046 |
cheek | -0.045 |
fourteen | -0.042 |
mine | -0.041 |
Token | and |
Token ID | 290 |
Feature activation | -4.39e-01 |
Pos logit contributions | |
five | +0.014 |
agging | +0.014 |
cheek | +0.013 |
mine | +0.012 |
liking | +0.012 |
Neg logit contributions | |
Pig | -0.015 |
Tweet | -0.014 |
Ally | -0.013 |
Beth | -0.013 |
gy | -0.013 |
Token | collect |
Token ID | 2824 |
Feature activation | -3.57e-02 |
Pos logit contributions | |
five | +0.017 |
cheek | +0.015 |
agging | +0.015 |
mine | +0.014 |
sidelines | +0.014 |
Neg logit contributions | |
Pig | -0.019 |
Tweet | -0.017 |
Beth | -0.017 |
Bunny | -0.017 |
Ally | -0.017 |
Token | some |
Token ID | 617 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
five | +0.001 |
agging | +0.001 |
mine | +0.001 |
cheek | +0.001 |
goodness | +0.001 |
Neg logit contributions | |
Pig | -0.002 |
gy | -0.002 |
Beth | -0.001 |
Ally | -0.001 |
known | -0.001 |
Token | ice |
Token ID | 4771 |
Feature activation | +8.36e-01 |
Pos logit contributions | |
Pig | +0.005 |
Tweet | +0.005 |
Beth | +0.005 |
gy | +0.005 |
Bunny | +0.005 |
Neg logit contributions | |
five | -0.005 |
agging | -0.005 |
cheek | -0.005 |
mine | -0.004 |
fourteen | -0.004 |
Token | - |
Token ID | 12 |
Feature activation | +4.79e+00 |
Pos logit contributions | |
Pig | +0.043 |
Tweet | +0.039 |
Beth | +0.039 |
Ally | +0.038 |
Bunny | +0.038 |
Neg logit contributions | |
five | -0.027 |
agging | -0.025 |
cheek | -0.024 |
fourteen | -0.022 |
mine | -0.021 |
Token | cream |
Token ID | 36277 |
Feature activation | -3.20e-01 |
Pos logit contributions | |
Pig | +0.214 |
Beth | +0.192 |
Tweet | +0.192 |
Ally | +0.191 |
Bunny | +0.191 |
Neg logit contributions | |
five | -0.185 |
agging | -0.168 |
cheek | -0.164 |
fourteen | -0.156 |
mine | -0.150 |
Token | . |
Token ID | 13 |
Feature activation | +4.96e-01 |
Pos logit contributions | |
five | +0.015 |
agging | +0.015 |
cheek | +0.015 |
fourteen | +0.014 |
mine | +0.013 |
Neg logit contributions | |
Pig | -0.011 |
Tweet | -0.010 |
Beth | -0.010 |
Ally | -0.009 |
Bunny | -0.009 |
Token | He |
Token ID | 679 |
Feature activation | +4.16e-01 |
Pos logit contributions | |
Pig | +0.015 |
Tweet | +0.014 |
Beth | +0.013 |
Bunny | +0.013 |
gy | +0.013 |
Neg logit contributions | |
five | -0.025 |
agging | -0.025 |
cheek | -0.024 |
fourteen | -0.022 |
Deal | -0.022 |
Token | went |
Token ID | 1816 |
Feature activation | +5.07e-01 |
Pos logit contributions | |
Pig | +0.017 |
Ally | +0.016 |
Tweet | +0.016 |
Beth | +0.016 |
wood | +0.016 |
Neg logit contributions | |
five | -0.015 |
mine | -0.014 |
cheek | -0.013 |
ige | -0.013 |
picks | -0.013 |
Token | to |
Token ID | 284 |
Feature activation | -5.16e-01 |
Pos logit contributions | |
Pig | +0.018 |
Tweet | +0.016 |
Beth | +0.015 |
Bunny | +0.015 |
Ally | +0.015 |
Neg logit contributions | |
five | -0.024 |
agging | -0.023 |
cheek | -0.023 |
fourteen | -0.021 |
mine | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | -3.36e-01 |
Pos logit contributions | |
five | +0.029 |
agging | +0.028 |
cheek | +0.028 |
mine | +0.026 |
sidelines | +0.026 |
Neg logit contributions | |
Pig | -0.020 |
Tweet | -0.018 |
gy | -0.018 |
wood | -0.017 |
known | -0.017 |
Token | yard |
Token ID | 12699 |
Feature activation | -7.63e-01 |
Pos logit contributions | |
five | +0.015 |
agging | +0.014 |
cheek | +0.014 |
fourteen | +0.013 |
Deal | +0.013 |
Neg logit contributions | |
Pig | -0.013 |
Tweet | -0.012 |
Bunny | -0.011 |
Beth | -0.011 |
Ally | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | -4.53e-02 |
Pos logit contributions | |
five | +0.035 |
agging | +0.034 |
cheek | +0.033 |
fourteen | +0.030 |
sidelines | +0.030 |
Neg logit contributions | |
Pig | -0.028 |
Tweet | -0.024 |
wood | -0.024 |
Ally | -0.024 |
Beth | -0.024 |
Token | even |
Token ID | 772 |
Feature activation | -7.25e-02 |
Pos logit contributions | |
five | +0.002 |
cheek | +0.002 |
agging | +0.002 |
la | +0.002 |
sidelines | +0.002 |
Neg logit contributions | |
Pig | -0.002 |
wood | -0.002 |
Tweet | -0.002 |
Ally | -0.001 |
gy | -0.001 |
Token | zip |
Token ID | 19974 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
five | +0.003 |
cheek | +0.003 |
la | +0.003 |
agging | +0.003 |
liking | +0.003 |
Neg logit contributions | |
Pig | -0.003 |
wood | -0.003 |
Bert | -0.003 |
Ally | -0.002 |
Tweet | -0.002 |
Token | - |
Token ID | 12 |
Feature activation | +4.78e+00 |
Pos logit contributions | |
Pig | +0.089 |
known | +0.082 |
Ally | +0.079 |
dam | +0.079 |
Tweet | +0.077 |
Neg logit contributions | |
five | -0.058 |
agging | -0.055 |
cheek | -0.051 |
goodness | -0.045 |
mine | -0.045 |
Token | lined |
Token ID | 10837 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
Pig | +0.201 |
Ally | +0.178 |
Tweet | +0.177 |
Beth | +0.177 |
Bunny | +0.176 |
Neg logit contributions | |
five | -0.201 |
agging | -0.177 |
cheek | -0.172 |
fourteen | -0.167 |
leaf | -0.159 |
Token | across |
Token ID | 1973 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
Pig | +0.043 |
known | +0.038 |
Tweet | +0.038 |
Beth | +0.038 |
Ally | +0.037 |
Neg logit contributions | |
five | -0.045 |
agging | -0.042 |
cheek | -0.041 |
mine | -0.039 |
fourteen | -0.037 |
Token | the |
Token ID | 262 |
Feature activation | -5.00e-01 |
Pos logit contributions | |
Pig | +0.004 |
gy | +0.004 |
wood | +0.004 |
Ally | +0.003 |
Bunny | +0.003 |
Neg logit contributions | |
five | -0.004 |
mine | -0.004 |
agging | -0.004 |
liking | -0.004 |
goodness | -0.004 |
Token | driveway |
Token ID | 33515 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
five | +0.022 |
agging | +0.021 |
cheek | +0.020 |
Deal | +0.019 |
fourteen | +0.019 |
Neg logit contributions | |
Pig | -0.020 |
Tweet | -0.017 |
Bunny | -0.017 |
gy | -0.017 |
wood | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +3.51e-01 |
Pos logit contributions | |
five | +0.083 |
agging | +0.080 |
cheek | +0.079 |
mine | +0.072 |
fourteen | +0.072 |
Neg logit contributions | |
Pig | -0.074 |
Tweet | -0.066 |
Bunny | -0.065 |
Ally | -0.065 |
Beth | -0.063 |
Token | " |
Token ID | 366 |
Feature activation | +8.14e-01 |
Pos logit contributions | |
five | +0.005 |
cheek | +0.005 |
wo | +0.004 |
fourteen | +0.004 |
EEP | +0.004 |
Neg logit contributions | |
known | -0.007 |
Pig | -0.007 |
wood | -0.007 |
increasingly | -0.006 |
shelters | -0.006 |
Token | Let |
Token ID | 5756 |
Feature activation | -5.82e-01 |
Pos logit contributions | |
Pig | +0.030 |
Tweet | +0.029 |
known | +0.027 |
Ally | +0.027 |
wood | +0.026 |
Neg logit contributions | |
agging | -0.037 |
cheek | -0.034 |
five | -0.033 |
Deal | -0.031 |
mine | -0.031 |
Token | 's |
Token ID | 338 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
five | +0.026 |
agging | +0.026 |
cheek | +0.025 |
fourteen | +0.023 |
arl | +0.023 |
Neg logit contributions | |
Pig | -0.022 |
Tweet | -0.020 |
Bunny | -0.019 |
Beth | -0.019 |
Ally | -0.019 |
Token | play |
Token ID | 711 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
Pig | +0.072 |
known | +0.065 |
Tweet | +0.065 |
Beth | +0.065 |
Bunny | +0.064 |
Neg logit contributions | |
five | -0.062 |
agging | -0.058 |
cheek | -0.057 |
mine | -0.054 |
fourteen | -0.052 |
Token | tug |
Token ID | 27762 |
Feature activation | -8.43e-01 |
Pos logit contributions | |
Pig | +0.009 |
Tweet | +0.008 |
Bunny | +0.008 |
Beth | +0.008 |
Ally | +0.008 |
Neg logit contributions | |
five | -0.011 |
agging | -0.011 |
cheek | -0.011 |
fourteen | -0.010 |
Deal | -0.010 |
Token | - |
Token ID | 12 |
Feature activation | +4.77e+00 |
Pos logit contributions | |
five | +0.041 |
cheek | +0.039 |
agging | +0.038 |
Deal | +0.038 |
sidelines | +0.037 |
Neg logit contributions | |
Pig | -0.027 |
Tweet | -0.024 |
Bert | -0.024 |
Bunny | -0.023 |
Ch | -0.023 |
Token | of |
Token ID | 1659 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
Pig | +0.233 |
Tweet | +0.214 |
Beth | +0.210 |
Bunny | +0.210 |
Ally | +0.208 |
Neg logit contributions | |
five | -0.158 |
agging | -0.153 |
cheek | -0.147 |
fourteen | -0.132 |
Deal | -0.130 |
Token | - |
Token ID | 12 |
Feature activation | +4.58e+00 |
Pos logit contributions | |
Pig | +0.008 |
Tweet | +0.007 |
Bunny | +0.007 |
Beth | +0.006 |
Ch | +0.006 |
Neg logit contributions | |
five | -0.009 |
agging | -0.009 |
cheek | -0.009 |
Deal | -0.009 |
sidelines | -0.009 |
Token | war |
Token ID | 5767 |
Feature activation | -9.48e-02 |
Pos logit contributions | |
Pig | +0.226 |
Tweet | +0.209 |
Beth | +0.204 |
Bunny | +0.204 |
gy | +0.202 |
Neg logit contributions | |
five | -0.147 |
agging | -0.144 |
cheek | -0.138 |
Deal | -0.123 |
fourteen | -0.122 |
Token | !" |
Token ID | 2474 |
Feature activation | -4.99e-01 |
Pos logit contributions | |
five | +0.004 |
agging | +0.004 |
cheek | +0.004 |
fourteen | +0.003 |
mine | +0.003 |
Neg logit contributions | |
Pig | -0.004 |
known | -0.004 |
Beth | -0.004 |
Ally | -0.003 |
Tweet | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +4.92e-01 |
Pos logit contributions | |
five | +0.021 |
cheek | +0.019 |
agging | +0.017 |
EEP | +0.016 |
fourteen | +0.016 |
Neg logit contributions | |
Pig | -0.020 |
known | -0.020 |
wood | -0.019 |
darker | -0.018 |
gy | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | -4.31e-01 |
Pos logit contributions | |
five | +0.063 |
agging | +0.062 |
cheek | +0.060 |
fourteen | +0.056 |
mine | +0.055 |
Neg logit contributions | |
Pig | -0.061 |
Tweet | -0.054 |
Ally | -0.053 |
Beth | -0.053 |
Bunny | -0.053 |
Token | could |
Token ID | 714 |
Feature activation | +8.67e-01 |
Pos logit contributions | |
five | +0.019 |
agging | +0.017 |
cheek | +0.017 |
mine | +0.017 |
fourteen | +0.016 |
Neg logit contributions | |
Pig | -0.016 |
wood | -0.014 |
Tweet | -0.014 |
gy | -0.014 |
Bunny | -0.014 |
Token | do |
Token ID | 466 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
Pig | +0.037 |
Tweet | +0.033 |
Beth | +0.033 |
Bunny | +0.033 |
Ally | +0.032 |
Neg logit contributions | |
five | -0.034 |
agging | -0.033 |
cheek | -0.032 |
fourteen | -0.029 |
mine | -0.029 |
Token | more |
Token ID | 517 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
Pig | +0.045 |
wood | +0.040 |
Bunny | +0.039 |
gy | +0.038 |
Tweet | +0.038 |
Neg logit contributions | |
five | -0.045 |
agging | -0.044 |
mine | -0.042 |
cheek | -0.041 |
goodness | -0.040 |
Token | push |
Token ID | 4574 |
Feature activation | +5.14e-01 |
Pos logit contributions | |
Pig | +0.068 |
Tweet | +0.060 |
gy | +0.059 |
Beth | +0.059 |
known | +0.059 |
Neg logit contributions | |
five | -0.074 |
agging | -0.074 |
cheek | -0.069 |
mine | -0.065 |
goodness | -0.063 |
Token | - |
Token ID | 12 |
Feature activation | +4.77e+00 |
Pos logit contributions | |
Pig | +0.021 |
known | +0.020 |
dam | +0.019 |
gy | +0.018 |
Tweet | +0.018 |
Neg logit contributions | |
five | -0.020 |
agging | -0.020 |
mine | -0.019 |
cheek | -0.019 |
goodness | -0.018 |
Token | ups |
Token ID | 4739 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
Pig | +0.209 |
Bunny | +0.189 |
Beth | +0.188 |
Tweet | +0.188 |
Ally | +0.184 |
Neg logit contributions | |
five | -0.200 |
agging | -0.177 |
cheek | -0.165 |
fourteen | -0.161 |
leaf | -0.154 |
Token | than |
Token ID | 621 |
Feature activation | +2.90e-01 |
Pos logit contributions | |
Pig | +0.078 |
Tweet | +0.070 |
Bunny | +0.067 |
Beth | +0.067 |
Ally | +0.067 |
Neg logit contributions | |
five | -0.106 |
agging | -0.103 |
cheek | -0.099 |
mine | -0.093 |
fourteen | -0.092 |
Token | most |
Token ID | 749 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
Pig | +0.011 |
Tweet | +0.009 |
gy | +0.009 |
Ally | +0.009 |
Bunny | +0.009 |
Neg logit contributions | |
five | -0.013 |
agging | -0.013 |
cheek | -0.012 |
fourteen | -0.012 |
mine | -0.012 |
Token | adults |
Token ID | 6490 |
Feature activation | +6.72e-01 |
Pos logit contributions | |
Pig | +0.043 |
Tweet | +0.038 |
Beth | +0.037 |
Bunny | +0.037 |
Ally | +0.037 |
Neg logit contributions | |
five | -0.053 |
agging | -0.051 |
cheek | -0.050 |
fourteen | -0.046 |
mine | -0.046 |
Token | . |
Token ID | 13 |
Feature activation | +3.21e-01 |
Pos logit contributions | |
Pig | +0.024 |
Tweet | +0.022 |
Bunny | +0.021 |
gy | +0.021 |
Beth | +0.021 |
Neg logit contributions | |
five | -0.031 |
agging | -0.030 |
cheek | -0.029 |
mine | -0.027 |
fourteen | -0.027 |
Token | to |
Token ID | 284 |
Feature activation | +3.56e-01 |
Pos logit contributions | |
Pig | +0.056 |
known | +0.049 |
Bunny | +0.048 |
Bert | +0.047 |
Tweet | +0.047 |
Neg logit contributions | |
five | -0.058 |
agging | -0.055 |
cheek | -0.054 |
mine | -0.051 |
arl | -0.048 |
Token | read |
Token ID | 1100 |
Feature activation | +8.35e-01 |
Pos logit contributions | |
Pig | +0.012 |
Tweet | +0.011 |
Beth | +0.010 |
known | +0.010 |
Ally | +0.010 |
Neg logit contributions | |
five | -0.017 |
agging | -0.017 |
cheek | -0.016 |
fourteen | -0.015 |
arl | -0.015 |
Token | in |
Token ID | 287 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
Pig | +0.033 |
wood | +0.029 |
gy | +0.029 |
Bunny | +0.028 |
known | +0.028 |
Neg logit contributions | |
five | -0.037 |
agging | -0.034 |
cheek | -0.033 |
mine | -0.033 |
fourteen | -0.031 |
Token | her |
Token ID | 607 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
Pig | +0.037 |
Tweet | +0.032 |
Bunny | +0.032 |
gy | +0.031 |
Beth | +0.031 |
Neg logit contributions | |
five | -0.055 |
agging | -0.054 |
cheek | -0.052 |
mine | -0.049 |
fourteen | -0.049 |
Token | tree |
Token ID | 5509 |
Feature activation | +3.76e-01 |
Pos logit contributions | |
Pig | +0.054 |
Tweet | +0.049 |
Bunny | +0.048 |
Beth | +0.048 |
gy | +0.047 |
Neg logit contributions | |
five | -0.045 |
agging | -0.044 |
cheek | -0.042 |
fourteen | -0.039 |
sidelines | -0.038 |
Token | - |
Token ID | 12 |
Feature activation | +4.77e+00 |
Pos logit contributions | |
Pig | +0.016 |
known | +0.014 |
Bert | +0.014 |
Beth | +0.014 |
wood | +0.014 |
Neg logit contributions | |
five | -0.014 |
agging | -0.014 |
cheek | -0.014 |
mine | -0.013 |
la | -0.012 |
Token | house |
Token ID | 4803 |
Feature activation | +2.85e+00 |
Pos logit contributions | |
Pig | +0.195 |
Tweet | +0.175 |
Beth | +0.172 |
Bunny | +0.171 |
Ally | +0.169 |
Neg logit contributions | |
five | -0.198 |
agging | -0.190 |
cheek | -0.184 |
fourteen | -0.170 |
mine | -0.168 |
Token | . |
Token ID | 13 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
Pig | +0.129 |
known | +0.115 |
Bert | +0.114 |
Ally | +0.113 |
Beth | +0.112 |
Neg logit contributions | |
five | -0.103 |
mine | -0.096 |
cheek | -0.096 |
agging | -0.095 |
fourteen | -0.086 |
Token |
|
Token ID | 198 |
Feature activation | -3.97e-02 |
Pos logit contributions | |
Pig | +0.005 |
Tweet | +0.004 |
Bunny | +0.004 |
Ally | +0.004 |
Beth | +0.004 |
Neg logit contributions | |
agging | -0.010 |
five | -0.010 |
cheek | -0.010 |
sidelines | -0.009 |
fourteen | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
five | +0.002 |
agging | +0.002 |
cheek | +0.002 |
fourteen | +0.002 |
Deal | +0.002 |
Neg logit contributions | |
Pig | -0.001 |
Tweet | -0.001 |
Beth | -0.001 |
wood | -0.001 |
Bunny | -0.001 |
Token | One |
Token ID | 3198 |
Feature activation | -2.60e-01 |
Pos logit contributions | |
cheek | +0.007 |
agging | +0.007 |
arl | +0.007 |
five | +0.007 |
sidelines | +0.007 |
Neg logit contributions | |
Tweet | -0.004 |
Ally | -0.004 |
known | -0.004 |
Pig | -0.004 |
Bunny | -0.004 |
Token | They |
Token ID | 1119 |
Feature activation | -2.96e-02 |
Pos logit contributions | |
known | +0.004 |
Pig | +0.004 |
wood | +0.004 |
gy | +0.004 |
Bert | +0.004 |
Neg logit contributions | |
five | -0.004 |
cheek | -0.004 |
mine | -0.004 |
agging | -0.004 |
arl | -0.004 |
Token | decided |
Token ID | 3066 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
five | +0.001 |
agging | +0.001 |
cheek | +0.001 |
fourteen | +0.001 |
mine | +0.001 |
Neg logit contributions | |
Pig | -0.001 |
Tweet | -0.001 |
Beth | -0.001 |
Bunny | -0.001 |
wood | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | -4.89e-01 |
Pos logit contributions | |
five | +0.049 |
agging | +0.047 |
cheek | +0.046 |
mine | +0.043 |
fourteen | +0.043 |
Neg logit contributions | |
Pig | -0.037 |
Tweet | -0.033 |
Bunny | -0.032 |
Beth | -0.032 |
wood | -0.032 |
Token | play |
Token ID | 711 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
five | +0.018 |
cheek | +0.016 |
agging | +0.016 |
mine | +0.015 |
sidelines | +0.014 |
Neg logit contributions | |
Pig | -0.023 |
known | -0.020 |
Tweet | -0.020 |
Bunny | -0.020 |
Beth | -0.020 |
Token | tug |
Token ID | 27762 |
Feature activation | -8.35e-01 |
Pos logit contributions | |
Pig | +0.007 |
Tweet | +0.006 |
Beth | +0.006 |
Bunny | +0.006 |
wood | +0.006 |
Neg logit contributions | |
five | -0.010 |
agging | -0.010 |
cheek | -0.009 |
fourteen | -0.009 |
mine | -0.009 |
Token | - |
Token ID | 12 |
Feature activation | +4.77e+00 |
Pos logit contributions | |
five | +0.040 |
agging | +0.039 |
cheek | +0.038 |
fourteen | +0.035 |
Deal | +0.035 |
Neg logit contributions | |
Pig | -0.029 |
Tweet | -0.025 |
Beth | -0.025 |
Bunny | -0.025 |
Ally | -0.024 |
Token | of |
Token ID | 1659 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
Pig | +0.225 |
Tweet | +0.205 |
Bunny | +0.202 |
Beth | +0.202 |
Ally | +0.200 |
Neg logit contributions | |
five | -0.171 |
agging | -0.160 |
cheek | -0.153 |
fourteen | -0.141 |
mine | -0.137 |
Token | - |
Token ID | 12 |
Feature activation | +4.58e+00 |
Pos logit contributions | |
Pig | +0.007 |
Tweet | +0.006 |
Beth | +0.006 |
Bunny | +0.006 |
Ally | +0.006 |
Neg logit contributions | |
five | -0.010 |
agging | -0.010 |
cheek | -0.009 |
fourteen | -0.009 |
Deal | -0.009 |
Token | war |
Token ID | 5767 |
Feature activation | -2.17e-01 |
Pos logit contributions | |
Pig | +0.216 |
Tweet | +0.196 |
Beth | +0.193 |
Bunny | +0.193 |
Ally | +0.191 |
Neg logit contributions | |
five | -0.164 |
agging | -0.155 |
cheek | -0.149 |
fourteen | -0.136 |
mine | -0.133 |
Token | with |
Token ID | 351 |
Feature activation | +9.71e-01 |
Pos logit contributions | |
five | +0.009 |
agging | +0.009 |
cheek | +0.009 |
mine | +0.007 |
ater | +0.007 |
Neg logit contributions | |
Pig | -0.009 |
known | -0.009 |
Ally | -0.008 |
Beth | -0.008 |
Bunny | -0.008 |
Token | it |
Token ID | 340 |
Feature activation | +7.24e-01 |
Pos logit contributions | |
Pig | +0.038 |
wood | +0.036 |
known | +0.036 |
gy | +0.034 |
Bunny | +0.032 |
Neg logit contributions | |
five | -0.041 |
agging | -0.037 |
mine | -0.037 |
cheek | -0.037 |
goodness | -0.036 |
Token | had |
Token ID | 550 |
Feature activation | -3.62e-01 |
Pos logit contributions | |
mine | +0.030 |
five | +0.030 |
cheek | +0.030 |
la | +0.028 |
liking | +0.028 |
Neg logit contributions | |
Pig | -0.033 |
wood | -0.032 |
shelters | -0.029 |
dam | -0.028 |
Ally | -0.028 |
Token | found |
Token ID | 1043 |
Feature activation | -7.41e-01 |
Pos logit contributions | |
five | +0.016 |
agging | +0.015 |
mine | +0.015 |
cheek | +0.015 |
fourteen | +0.014 |
Neg logit contributions | |
Pig | -0.013 |
Beth | -0.012 |
Ally | -0.012 |
wood | -0.012 |
Tweet | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | -4.25e-01 |
Pos logit contributions | |
five | +0.028 |
goodness | +0.028 |
mine | +0.028 |
arrest | +0.026 |
yours | +0.025 |
Neg logit contributions | |
known | -0.026 |
Pig | -0.026 |
rou | -0.025 |
house | -0.025 |
gy | -0.024 |
Token | l |
Token ID | 300 |
Feature activation | -2.14e+00 |
Pos logit contributions | |
five | +0.016 |
agging | +0.016 |
cheek | +0.015 |
fourteen | +0.014 |
Deal | +0.014 |
Neg logit contributions | |
Pig | -0.019 |
Tweet | -0.017 |
Bunny | -0.017 |
known | -0.017 |
Beth | -0.017 |
Token | ily |
Token ID | 813 |
Feature activation | -5.40e-01 |
Pos logit contributions | |
cheek | +0.083 |
five | +0.080 |
sidelines | +0.073 |
Deal | +0.073 |
fourteen | +0.071 |
Neg logit contributions | |
Tweet | -0.086 |
Pig | -0.083 |
Bunny | -0.082 |
Madd | -0.077 |
known | -0.076 |
Token | pad |
Token ID | 14841 |
Feature activation | -4.69e+00 |
Pos logit contributions | |
five | +0.022 |
agging | +0.021 |
cheek | +0.021 |
Deal | +0.019 |
fourteen | +0.019 |
Neg logit contributions | |
Pig | -0.022 |
Tweet | -0.020 |
Bunny | -0.020 |
gy | -0.019 |
Beth | -0.019 |
Token | again |
Token ID | 757 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
five | +0.233 |
agging | +0.228 |
Deal | +0.218 |
sidelines | +0.216 |
cheek | +0.214 |
Neg logit contributions | |
Pig | -0.147 |
gy | -0.139 |
Tweet | -0.134 |
house | -0.128 |
wood | -0.128 |
Token | . |
Token ID | 13 |
Feature activation | -4.33e-01 |
Pos logit contributions | |
five | +0.048 |
agging | +0.047 |
cheek | +0.046 |
fourteen | +0.043 |
mine | +0.042 |
Neg logit contributions | |
Pig | -0.036 |
Tweet | -0.032 |
Beth | -0.031 |
Bunny | -0.031 |
Ally | -0.031 |
Token | Bobby |
Token ID | 17155 |
Feature activation | -3.58e-01 |
Pos logit contributions | |
cheek | +0.018 |
five | +0.018 |
agging | +0.018 |
mine | +0.017 |
goodness | +0.017 |
Neg logit contributions | |
known | -0.015 |
Pig | -0.014 |
wood | -0.013 |
gy | -0.013 |
darker | -0.013 |
Token | told |
Token ID | 1297 |
Feature activation | -7.53e-01 |
Pos logit contributions | |
cheek | +0.015 |
five | +0.015 |
agging | +0.014 |
mine | +0.013 |
fourteen | +0.013 |
Neg logit contributions | |
Pig | -0.012 |
wood | -0.011 |
known | -0.011 |
Beth | -0.011 |
darker | -0.011 |
Token | Freddy |
Token ID | 45437 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
agging | +0.025 |
mine | +0.025 |
goodness | +0.024 |
cheek | +0.024 |
five | +0.024 |
Neg logit contributions | |
known | -0.032 |
house | -0.031 |
wood | -0.030 |
darker | -0.029 |
attendants | -0.029 |
Token | to |
Token ID | 284 |
Feature activation | -9.79e-01 |
Pos logit contributions | |
Pig | +0.003 |
Tweet | +0.002 |
known | +0.002 |
Beth | +0.002 |
Ally | +0.002 |
Neg logit contributions | |
five | -0.003 |
agging | -0.003 |
cheek | -0.003 |
mine | -0.003 |
arl | -0.002 |
Token | find |
Token ID | 1064 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
five | +0.038 |
cheek | +0.035 |
mine | +0.035 |
arrest | +0.034 |
sidelines | +0.034 |
Neg logit contributions | |
Pig | -0.036 |
wood | -0.035 |
darker | -0.035 |
rou | -0.034 |
known | -0.034 |
Token | the |
Token ID | 262 |
Feature activation | -3.60e-01 |
Pos logit contributions | |
five | +0.063 |
agging | +0.057 |
cheek | +0.055 |
goodness | +0.055 |
mine | +0.054 |
Neg logit contributions | |
gy | -0.044 |
wood | -0.043 |
Pig | -0.043 |
known | -0.042 |
rou | -0.042 |
Token | l |
Token ID | 300 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
five | +0.014 |
agging | +0.014 |
cheek | +0.014 |
mine | +0.013 |
fourteen | +0.012 |
Neg logit contributions | |
Pig | -0.015 |
Beth | -0.014 |
gy | -0.013 |
wood | -0.013 |
Ally | -0.013 |
Token | ily |
Token ID | 813 |
Feature activation | -5.61e-01 |
Pos logit contributions | |
cheek | +0.075 |
five | +0.074 |
fourteen | +0.067 |
sidelines | +0.067 |
Deal | +0.066 |
Neg logit contributions | |
Tweet | -0.082 |
Pig | -0.080 |
Bunny | -0.078 |
known | -0.074 |
Madd | -0.074 |
Token | pad |
Token ID | 14841 |
Feature activation | -4.67e+00 |
Pos logit contributions | |
five | +0.024 |
agging | +0.023 |
cheek | +0.023 |
mine | +0.021 |
fourteen | +0.020 |
Neg logit contributions | |
Pig | -0.022 |
Tweet | -0.020 |
Beth | -0.020 |
Ally | -0.019 |
known | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -4.75e-01 |
Pos logit contributions | |
five | +0.233 |
agging | +0.227 |
cheek | +0.218 |
fourteen | +0.211 |
Deal | +0.211 |
Neg logit contributions | |
Pig | -0.147 |
Tweet | -0.132 |
gy | -0.129 |
Bunny | -0.126 |
wood | -0.124 |
Token | He |
Token ID | 679 |
Feature activation | +9.00e-02 |
Pos logit contributions | |
five | +0.018 |
cheek | +0.018 |
mine | +0.017 |
agging | +0.016 |
sidelines | +0.016 |
Neg logit contributions | |
known | -0.019 |
Pig | -0.019 |
Ally | -0.016 |
darker | -0.016 |
gy | -0.016 |
Token | searched |
Token ID | 16499 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
wood | +0.004 |
ained | +0.004 |
Pig | +0.004 |
darker | +0.004 |
bul | +0.004 |
Neg logit contributions | |
mine | -0.002 |
cheek | -0.002 |
five | -0.002 |
liking | -0.002 |
wo | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
agging | +0.004 |
five | +0.004 |
cheek | +0.004 |
mine | +0.004 |
fourteen | +0.004 |
Neg logit contributions | |
Pig | -0.005 |
wood | -0.004 |
Beth | -0.004 |
known | -0.004 |
Tweet | -0.004 |
Token | searched |
Token ID | 16499 |
Feature activation | +7.91e-01 |
Pos logit contributions | |
Pig | +0.006 |
wood | +0.005 |
known | +0.005 |
attendants | +0.005 |
Kang | +0.005 |
Neg logit contributions | |
five | -0.010 |
cheek | -0.009 |
mine | -0.009 |
agging | -0.009 |
sidelines | -0.009 |
Token | big |
Token ID | 1263 |
Feature activation | +2.73e-01 |
Pos logit contributions | |
cheek | +0.009 |
mine | +0.009 |
five | +0.009 |
agging | +0.009 |
liking | +0.008 |
Neg logit contributions | |
Pig | -0.011 |
wood | -0.010 |
Beth | -0.010 |
known | -0.010 |
gy | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
Pig | +0.010 |
known | +0.009 |
Beth | +0.009 |
wood | +0.008 |
Ally | +0.008 |
Neg logit contributions | |
five | -0.012 |
agging | -0.012 |
cheek | -0.012 |
mine | -0.012 |
sidelines | -0.011 |
Token | beautiful |
Token ID | 4950 |
Feature activation | +7.85e-02 |
Pos logit contributions | |
cheek | +0.056 |
five | +0.055 |
agging | +0.055 |
mine | +0.052 |
la | +0.049 |
Neg logit contributions | |
Pig | -0.055 |
wood | -0.051 |
known | -0.051 |
gy | -0.050 |
Beth | -0.050 |
Token | l |
Token ID | 300 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
Pig | +0.003 |
known | +0.003 |
wood | +0.003 |
Beth | +0.003 |
gy | +0.002 |
Neg logit contributions | |
agging | -0.003 |
cheek | -0.003 |
five | -0.003 |
mine | -0.003 |
la | -0.003 |
Token | ily |
Token ID | 813 |
Feature activation | -4.13e-01 |
Pos logit contributions | |
five | +0.071 |
cheek | +0.071 |
sidelines | +0.065 |
fourteen | +0.063 |
Deal | +0.063 |
Neg logit contributions | |
Pig | -0.078 |
Tweet | -0.076 |
Bunny | -0.073 |
known | -0.072 |
Bert | -0.070 |
Token | pad |
Token ID | 14841 |
Feature activation | -4.63e+00 |
Pos logit contributions | |
five | +0.019 |
agging | +0.018 |
cheek | +0.018 |
fourteen | +0.016 |
mine | +0.016 |
Neg logit contributions | |
Pig | -0.015 |
Tweet | -0.014 |
Beth | -0.013 |
Bunny | -0.013 |
Ally | -0.013 |
Token | that |
Token ID | 326 |
Feature activation | +5.30e-01 |
Pos logit contributions | |
five | +0.229 |
Deal | +0.221 |
sidelines | +0.221 |
agging | +0.220 |
fourteen | +0.212 |
Neg logit contributions | |
Pig | -0.143 |
gy | -0.140 |
Tweet | -0.130 |
house | -0.129 |
Bunny | -0.125 |
Token | he |
Token ID | 339 |
Feature activation | -6.30e-01 |
Pos logit contributions | |
wood | +0.021 |
Pig | +0.021 |
gy | +0.020 |
dam | +0.019 |
increasingly | +0.019 |
Neg logit contributions | |
mine | -0.019 |
five | -0.018 |
cheek | -0.018 |
wo | -0.017 |
la | -0.016 |
Token | found |
Token ID | 1043 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
five | +0.025 |
agging | +0.024 |
cheek | +0.023 |
mine | +0.023 |
la | +0.022 |
Neg logit contributions | |
Pig | -0.026 |
wood | -0.024 |
Beth | -0.023 |
Ally | -0.023 |
Tweet | -0.023 |
Token | in |
Token ID | 287 |
Feature activation | -6.01e-01 |
Pos logit contributions | |
five | +0.060 |
agging | +0.059 |
cheek | +0.057 |
mine | +0.054 |
goodness | +0.053 |
Neg logit contributions | |
Pig | -0.045 |
Beth | -0.039 |
Ally | -0.039 |
gy | -0.039 |
wood | -0.038 |
Token | the |
Token ID | 262 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
mine | +0.024 |
five | +0.024 |
arrest | +0.024 |
agging | +0.024 |
cheek | +0.023 |
Neg logit contributions | |
Pig | -0.021 |
wood | -0.020 |
Ally | -0.019 |
gy | -0.019 |
Beth | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | -3.09e-01 |
Pos logit contributions | |
Pig | +0.019 |
Tweet | +0.017 |
gy | +0.017 |
Beth | +0.017 |
Ally | +0.017 |
Neg logit contributions | |
five | -0.022 |
agging | -0.021 |
cheek | -0.021 |
mine | -0.020 |
fourteen | -0.019 |
Token | you |
Token ID | 345 |
Feature activation | -9.62e-01 |
Pos logit contributions | |
five | +0.013 |
agging | +0.012 |
cheek | +0.012 |
mine | +0.011 |
fourteen | +0.011 |
Neg logit contributions | |
Pig | -0.013 |
Tweet | -0.011 |
Beth | -0.011 |
gy | -0.011 |
Bunny | -0.011 |
Token | â |
Token ID | 22940 |
Feature activation | -8.07e-02 |
Pos logit contributions | |
five | +0.037 |
agging | +0.034 |
cheek | +0.034 |
mine | +0.031 |
fourteen | +0.031 |
Neg logit contributions | |
Pig | -0.042 |
Tweet | -0.039 |
Beth | -0.038 |
Ally | -0.037 |
gy | -0.037 |
Token | € |
Token ID | 26391 |
Feature activation | +4.95e-01 |
Pos logit contributions | |
five | +0.003 |
agging | +0.003 |
cheek | +0.003 |
fourteen | +0.003 |
Deal | +0.002 |
Neg logit contributions | |
Pig | -0.004 |
Tweet | -0.003 |
Beth | -0.003 |
Bunny | -0.003 |
Ally | -0.003 |
Token | ™ |
Token ID | 8151 |
Feature activation | +7.84e-01 |
Pos logit contributions | |
Pig | +0.023 |
Tweet | +0.021 |
wood | +0.020 |
Ally | +0.020 |
Bunny | +0.020 |
Neg logit contributions | |
five | -0.018 |
agging | -0.016 |
cheek | -0.016 |
la | -0.015 |
fourteen | -0.015 |
Token | ll |
Token ID | 297 |
Feature activation | -4.59e+00 |
Pos logit contributions | |
Pig | +0.034 |
Tweet | +0.030 |
Beth | +0.030 |
Bunny | +0.030 |
Ally | +0.030 |
Neg logit contributions | |
five | -0.031 |
agging | -0.029 |
cheek | -0.028 |
fourteen | -0.026 |
mine | -0.026 |
Token | always |
Token ID | 1464 |
Feature activation | -2.18e-01 |
Pos logit contributions | |
five | +0.189 |
agging | +0.183 |
cheek | +0.176 |
fourteen | +0.164 |
Deal | +0.162 |
Neg logit contributions | |
Pig | -0.187 |
Tweet | -0.168 |
Beth | -0.165 |
Bunny | -0.164 |
gy | -0.164 |
Token | have |
Token ID | 423 |
Feature activation | -6.66e-01 |
Pos logit contributions | |
five | +0.009 |
agging | +0.008 |
cheek | +0.008 |
mine | +0.008 |
fourteen | +0.008 |
Neg logit contributions | |
Pig | -0.009 |
Tweet | -0.008 |
Beth | -0.008 |
Ally | -0.008 |
Bunny | -0.008 |
Token | your |
Token ID | 534 |
Feature activation | -4.78e-01 |
Pos logit contributions | |
five | +0.032 |
agging | +0.031 |
cheek | +0.030 |
mine | +0.028 |
fourteen | +0.028 |
Neg logit contributions | |
Pig | -0.023 |
Tweet | -0.020 |
Beth | -0.020 |
Bunny | -0.019 |
gy | -0.019 |
Token | memories |
Token ID | 9846 |
Feature activation | -9.87e-01 |
Pos logit contributions | |
five | +0.018 |
agging | +0.018 |
cheek | +0.017 |
mine | +0.016 |
fourteen | +0.016 |
Neg logit contributions | |
Pig | -0.021 |
Tweet | -0.019 |
Beth | -0.018 |
Ally | -0.018 |
Bunny | -0.018 |
Token | of |
Token ID | 286 |
Feature activation | +8.52e-01 |
Pos logit contributions | |
five | +0.046 |
agging | +0.045 |
cheek | +0.043 |
sidelines | +0.041 |
Deal | +0.041 |
Neg logit contributions | |
Pig | -0.034 |
Tweet | -0.030 |
Beth | -0.030 |
Bunny | -0.030 |
wood | -0.030 |
Token | It |
Token ID | 632 |
Feature activation | +7.44e-01 |
Pos logit contributions | |
Pig | +0.010 |
gy | +0.009 |
wood | +0.009 |
darker | +0.009 |
Tweet | +0.009 |
Neg logit contributions | |
five | -0.011 |
Deal | -0.009 |
cheek | -0.009 |
goodness | -0.009 |
agging | -0.009 |
Token | became |
Token ID | 2627 |
Feature activation | +8.37e-02 |
Pos logit contributions | |
Pig | +0.025 |
Tweet | +0.023 |
wood | +0.023 |
Beth | +0.022 |
gy | +0.022 |
Neg logit contributions | |
five | -0.034 |
cheek | -0.032 |
agging | -0.031 |
mine | -0.029 |
la | -0.029 |
Token | the |
Token ID | 262 |
Feature activation | -8.41e-01 |
Pos logit contributions | |
bank | +0.004 |
gy | +0.004 |
wood | +0.003 |
stal | +0.003 |
storms | +0.003 |
Neg logit contributions | |
mine | -0.003 |
yours | -0.002 |
my | -0.002 |
ours | -0.002 |
love | -0.002 |
Token | happiest |
Token ID | 49414 |
Feature activation | -7.88e-01 |
Pos logit contributions | |
cheek | +0.042 |
five | +0.041 |
agging | +0.041 |
mine | +0.039 |
Deal | +0.038 |
Neg logit contributions | |
Pig | -0.026 |
Tweet | -0.024 |
gy | -0.024 |
Beth | -0.023 |
Ally | -0.022 |
Token | books |
Token ID | 3835 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
five | +0.035 |
agging | +0.034 |
cheek | +0.033 |
mine | +0.030 |
fourteen | +0.030 |
Neg logit contributions | |
Pig | -0.030 |
Tweet | -0.027 |
gy | -0.026 |
wood | -0.026 |
Bunny | -0.026 |
Token | he |
Token ID | 258 |
Feature activation | -4.55e+00 |
Pos logit contributions | |
five | +0.082 |
agging | +0.080 |
cheek | +0.077 |
Deal | +0.073 |
fourteen | +0.073 |
Neg logit contributions | |
Pig | -0.068 |
Tweet | -0.060 |
Beth | -0.059 |
Bunny | -0.059 |
Ally | -0.058 |
Token | lf |
Token ID | 1652 |
Feature activation | -6.99e-01 |
Pos logit contributions | |
five | +0.160 |
agging | +0.145 |
Deal | +0.142 |
arl | +0.140 |
cheek | +0.138 |
Neg logit contributions | |
Pig | -0.217 |
gy | -0.199 |
Tweet | -0.198 |
Bunny | -0.197 |
Beth | -0.196 |
Token | in |
Token ID | 287 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
five | +0.028 |
agging | +0.027 |
cheek | +0.027 |
mine | +0.024 |
fourteen | +0.023 |
Neg logit contributions | |
Pig | -0.029 |
Tweet | -0.026 |
Bunny | -0.026 |
Beth | -0.026 |
wood | -0.026 |
Token | the |
Token ID | 262 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
agging | +0.073 |
five | +0.073 |
mine | +0.071 |
cheek | +0.070 |
fourteen | +0.069 |
Neg logit contributions | |
Pig | -0.051 |
wood | -0.047 |
Tweet | -0.046 |
gy | -0.045 |
Beth | -0.045 |
Token | world |
Token ID | 995 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
five | +0.067 |
agging | +0.066 |
cheek | +0.063 |
fourteen | +0.060 |
arl | +0.059 |
Neg logit contributions | |
Pig | -0.039 |
Tweet | -0.034 |
Beth | -0.034 |
Bunny | -0.033 |
gy | -0.033 |
Token | ! |
Token ID | 0 |
Feature activation | -5.11e-01 |
Pos logit contributions | |
Pig | +0.009 |
Tweet | +0.008 |
Bunny | +0.008 |
known | +0.008 |
Beth | +0.008 |
Neg logit contributions | |
five | -0.010 |
agging | -0.009 |
cheek | -0.009 |
mine | -0.008 |
fourteen | -0.008 |
Token | before |
Token ID | 878 |
Feature activation | +4.27e-01 |
Pos logit contributions | |
five | +0.031 |
agging | +0.030 |
cheek | +0.029 |
fourteen | +0.027 |
Deal | +0.026 |
Neg logit contributions | |
Pig | -0.033 |
Tweet | -0.030 |
Beth | -0.029 |
Bunny | -0.029 |
Ally | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
Pig | +0.016 |
Tweet | +0.014 |
Beth | +0.014 |
Bunny | +0.014 |
Ally | +0.013 |
Neg logit contributions | |
five | -0.019 |
agging | -0.019 |
cheek | -0.018 |
fourteen | -0.017 |
Deal | -0.017 |
Token | This |
Token ID | 770 |
Feature activation | +8.63e-02 |
Pos logit contributions | |
five | +0.005 |
agging | +0.004 |
cheek | +0.004 |
fourteen | +0.004 |
Deal | +0.004 |
Neg logit contributions | |
Pig | -0.006 |
Tweet | -0.005 |
Beth | -0.005 |
Bunny | -0.005 |
gy | -0.005 |
Token | is |
Token ID | 318 |
Feature activation | +3.41e-01 |
Pos logit contributions | |
Pig | +0.004 |
Tweet | +0.004 |
Beth | +0.004 |
Bunny | +0.004 |
gy | +0.004 |
Neg logit contributions | |
five | -0.003 |
cheek | -0.003 |
agging | -0.003 |
mine | -0.002 |
la | -0.002 |
Token | O |
Token ID | 440 |
Feature activation | -6.73e-01 |
Pos logit contributions | |
Pig | +0.013 |
gy | +0.013 |
wood | +0.012 |
Beth | +0.012 |
Tweet | +0.012 |
Neg logit contributions | |
mine | -0.014 |
five | -0.013 |
fourteen | -0.013 |
cheek | -0.012 |
agging | -0.012 |
Token | ll |
Token ID | 297 |
Feature activation | -4.54e+00 |
Pos logit contributions | |
five | +0.028 |
agging | +0.027 |
cheek | +0.026 |
fourteen | +0.024 |
Deal | +0.024 |
Neg logit contributions | |
Pig | -0.027 |
Tweet | -0.024 |
Bunny | -0.024 |
Beth | -0.024 |
gy | -0.024 |
Token | ie |
Token ID | 494 |
Feature activation | -7.44e-01 |
Pos logit contributions | |
five | +0.195 |
agging | +0.182 |
Deal | +0.177 |
cheek | +0.173 |
sidelines | +0.172 |
Neg logit contributions | |
Pig | -0.182 |
Ally | -0.164 |
gy | -0.162 |
Tweet | -0.162 |
Beth | -0.157 |
Token | , |
Token ID | 11 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
five | +0.029 |
agging | +0.028 |
cheek | +0.028 |
mine | +0.026 |
la | +0.025 |
Neg logit contributions | |
Pig | -0.031 |
Tweet | -0.029 |
Beth | -0.028 |
known | -0.028 |
wood | -0.028 |
Token | the |
Token ID | 262 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
five | +0.059 |
agging | +0.057 |
cheek | +0.057 |
mine | +0.053 |
la | +0.052 |
Neg logit contributions | |
Pig | -0.053 |
Tweet | -0.047 |
gy | -0.047 |
wood | -0.047 |
Beth | -0.047 |
Token | ostr |
Token ID | 42387 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
five | +0.046 |
agging | +0.044 |
cheek | +0.043 |
fourteen | +0.040 |
mine | +0.039 |
Neg logit contributions | |
Pig | -0.046 |
Tweet | -0.041 |
Beth | -0.041 |
Bunny | -0.040 |
gy | -0.040 |
Token | ich |
Token ID | 488 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
five | +0.090 |
mine | +0.079 |
sidelines | +0.078 |
cheek | +0.077 |
Deal | +0.076 |
Neg logit contributions | |
Pig | -0.094 |
gy | -0.090 |
Tweet | -0.084 |
wood | -0.081 |
Ch | -0.081 |
Token | , |
Token ID | 11 |
Feature activation | -6.70e-01 |
Pos logit contributions | |
cheek | +0.035 |
five | +0.034 |
agging | +0.032 |
mine | +0.032 |
goodness | +0.031 |
Neg logit contributions | |
Pig | -0.029 |
known | -0.026 |
Tweet | -0.026 |
Ally | -0.026 |
Bert | -0.025 |
Token | we |
Token ID | 356 |
Feature activation | -7.88e-01 |
Pos logit contributions | |
five | +0.028 |
cheek | +0.026 |
agging | +0.026 |
mine | +0.025 |
fourteen | +0.024 |
Neg logit contributions | |
Pig | -0.026 |
gy | -0.024 |
wood | -0.024 |
known | -0.023 |
Tweet | -0.023 |
Token | â |
Token ID | 22940 |
Feature activation | +5.93e-02 |
Pos logit contributions | |
five | +0.029 |
agging | +0.027 |
cheek | +0.027 |
mine | +0.026 |
Deal | +0.025 |
Neg logit contributions | |
Pig | -0.034 |
Tweet | -0.031 |
gy | -0.031 |
wood | -0.030 |
Ally | -0.030 |
Token | € |
Token ID | 26391 |
Feature activation | +5.75e-01 |
Pos logit contributions | |
Pig | +0.003 |
Tweet | +0.002 |
Bunny | +0.002 |
Beth | +0.002 |
Ally | +0.002 |
Neg logit contributions | |
five | -0.002 |
agging | -0.002 |
cheek | -0.002 |
fourteen | -0.002 |
sidelines | -0.002 |
Token | ™ |
Token ID | 8151 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
Pig | +0.025 |
Tweet | +0.023 |
wood | +0.022 |
Bunny | +0.022 |
Beth | +0.022 |
Neg logit contributions | |
five | -0.022 |
agging | -0.021 |
cheek | -0.021 |
la | -0.019 |
fourteen | -0.019 |
Token | ll |
Token ID | 297 |
Feature activation | -4.53e+00 |
Pos logit contributions | |
Pig | +0.046 |
Tweet | +0.041 |
Beth | +0.040 |
Bunny | +0.040 |
Ally | +0.040 |
Neg logit contributions | |
five | -0.043 |
agging | -0.041 |
cheek | -0.040 |
fourteen | -0.037 |
mine | -0.036 |
Token | mention |
Token ID | 3068 |
Feature activation | +9.59e-02 |
Pos logit contributions | |
five | +0.186 |
agging | +0.179 |
cheek | +0.172 |
fourteen | +0.160 |
Deal | +0.158 |
Neg logit contributions | |
Pig | -0.186 |
Tweet | -0.168 |
Beth | -0.165 |
Bunny | -0.165 |
gy | -0.163 |
Token | it |
Token ID | 340 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
Pig | +0.004 |
Tweet | +0.003 |
gy | +0.003 |
Beth | +0.003 |
Bunny | +0.003 |
Neg logit contributions | |
five | -0.004 |
agging | -0.004 |
cheek | -0.004 |
fourteen | -0.004 |
mine | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | +7.96e-02 |
Pos logit contributions | |
Pig | +0.004 |
Tweet | +0.004 |
known | +0.004 |
Beth | +0.004 |
Ally | +0.004 |
Neg logit contributions | |
five | -0.005 |
cheek | -0.005 |
agging | -0.005 |
mine | -0.005 |
fourteen | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | +8.36e-02 |
Pos logit contributions | |
Pig | +0.003 |
Tweet | +0.003 |
gy | +0.003 |
wood | +0.003 |
Beth | +0.003 |
Neg logit contributions | |
five | -0.003 |
agging | -0.003 |
cheek | -0.003 |
mine | -0.003 |
fourteen | -0.003 |
Token | grown |
Token ID | 7334 |
Feature activation | +2.57e-01 |
Pos logit contributions | |
Pig | +0.004 |
Tweet | +0.003 |
Beth | +0.003 |
gy | +0.003 |
Ally | +0.003 |
Neg logit contributions | |
five | -0.003 |
agging | -0.003 |
cheek | -0.003 |
mine | -0.003 |
fourteen | -0.003 |
Token | key |
Token ID | 1994 |
Feature activation | -2.97e-01 |
Pos logit contributions | |
Pig | +0.011 |
Tweet | +0.010 |
Bunny | +0.009 |
Ally | +0.009 |
Beth | +0.009 |
Neg logit contributions | |
five | -0.014 |
agging | -0.013 |
fourteen | -0.012 |
la | -0.012 |
cheek | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
five | +0.015 |
agging | +0.014 |
cheek | +0.014 |
fourteen | +0.013 |
Deal | +0.013 |
Neg logit contributions | |
Pig | -0.010 |
Tweet | -0.009 |
Beth | -0.008 |
Bunny | -0.008 |
Ally | -0.008 |
Token | She |
Token ID | 1375 |
Feature activation | +3.32e-01 |
Pos logit contributions | |
Pig | +0.003 |
Tweet | +0.003 |
Bunny | +0.003 |
gy | +0.003 |
Beth | +0.003 |
Neg logit contributions | |
five | -0.005 |
agging | -0.005 |
cheek | -0.005 |
fourteen | -0.005 |
Deal | -0.005 |
Token | unlocked |
Token ID | 14838 |
Feature activation | -6.05e-02 |
Pos logit contributions | |
Pig | +0.012 |
wood | +0.012 |
Tweet | +0.011 |
Ally | +0.011 |
Beth | +0.011 |
Neg logit contributions | |
five | -0.014 |
cheek | -0.013 |
mine | -0.012 |
agging | -0.012 |
ige | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | -6.86e-02 |
Pos logit contributions | |
five | +0.003 |
agging | +0.003 |
cheek | +0.003 |
mine | +0.003 |
fourteen | +0.003 |
Neg logit contributions | |
Pig | -0.002 |
wood | -0.002 |
gy | -0.002 |
Bunny | -0.002 |
known | -0.002 |
Token | pad |
Token ID | 14841 |
Feature activation | -4.52e+00 |
Pos logit contributions | |
five | +0.004 |
agging | +0.003 |
la | +0.003 |
leaf | +0.003 |
fourteen | +0.003 |
Neg logit contributions | |
Pig | -0.002 |
shelter | -0.002 |
Tweet | -0.002 |
gy | -0.002 |
Bunny | -0.002 |
Token | lock |
Token ID | 5354 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
agging | +0.233 |
five | +0.229 |
Deal | +0.218 |
sidelines | +0.217 |
la | +0.215 |
Neg logit contributions | |
Pig | -0.132 |
gy | -0.123 |
house | -0.118 |
Tweet | -0.117 |
wood | -0.113 |
Token | and |
Token ID | 290 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
five | +0.108 |
agging | +0.105 |
cheek | +0.102 |
Deal | +0.098 |
fourteen | +0.095 |
Neg logit contributions | |
Pig | -0.074 |
Tweet | -0.065 |
Beth | -0.063 |
gy | -0.062 |
Bunny | -0.062 |
Token | opened |
Token ID | 4721 |
Feature activation | +8.00e-01 |
Pos logit contributions | |
Pig | +0.007 |
Tweet | +0.006 |
wood | +0.006 |
Bunny | +0.006 |
gy | +0.006 |
Neg logit contributions | |
five | -0.010 |
cheek | -0.009 |
agging | -0.009 |
fourteen | -0.009 |
mine | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | +6.72e-01 |
Pos logit contributions | |
Pig | +0.027 |
known | +0.023 |
wood | +0.023 |
Tweet | +0.023 |
Bunny | +0.022 |
Neg logit contributions | |
five | -0.039 |
agging | -0.036 |
cheek | -0.036 |
mine | -0.035 |
fourteen | -0.035 |
Token | door |
Token ID | 3420 |
Feature activation | +5.25e-01 |
Pos logit contributions | |
Pig | +0.022 |
shelter | +0.020 |
wood | +0.019 |
Tweet | +0.019 |
gy | +0.018 |
Neg logit contributions | |
five | -0.034 |
agging | -0.034 |
la | -0.031 |
sidelines | -0.030 |
leaf | -0.030 |
Token | back |
Token ID | 736 |
Feature activation | -3.30e-01 |
Pos logit contributions | |
five | +0.024 |
agging | +0.023 |
cheek | +0.022 |
fourteen | +0.021 |
mine | +0.021 |
Neg logit contributions | |
Pig | -0.017 |
Tweet | -0.015 |
Beth | -0.014 |
known | -0.014 |
gy | -0.014 |
Token | onto |
Token ID | 4291 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
five | +0.015 |
agging | +0.014 |
cheek | +0.014 |
mine | +0.013 |
arl | +0.012 |
Neg logit contributions | |
Pig | -0.013 |
known | -0.011 |
Tweet | -0.011 |
Beth | -0.011 |
Ally | -0.011 |
Token | his |
Token ID | 465 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
five | +0.053 |
cheek | +0.050 |
fourteen | +0.048 |
mine | +0.048 |
sidelines | +0.048 |
Neg logit contributions | |
Pig | -0.040 |
wood | -0.038 |
gy | -0.037 |
known | -0.037 |
house | -0.037 |
Token | l |
Token ID | 300 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
five | +0.005 |
fourteen | +0.005 |
leaf | +0.004 |
agging | +0.004 |
Deal | +0.004 |
Neg logit contributions | |
Pig | -0.004 |
Bunny | -0.004 |
Tweet | -0.004 |
gy | -0.004 |
Ch | -0.004 |
Token | ily |
Token ID | 813 |
Feature activation | -4.14e-01 |
Pos logit contributions | |
cheek | +0.071 |
five | +0.070 |
mine | +0.065 |
sidelines | +0.065 |
fourteen | +0.064 |
Neg logit contributions | |
Pig | -0.078 |
Tweet | -0.074 |
Bunny | -0.072 |
known | -0.071 |
Ch | -0.070 |
Token | pad |
Token ID | 14841 |
Feature activation | -4.49e+00 |
Pos logit contributions | |
five | +0.016 |
agging | +0.016 |
cheek | +0.015 |
fourteen | +0.014 |
Deal | +0.014 |
Neg logit contributions | |
Pig | -0.018 |
Tweet | -0.016 |
Bunny | -0.016 |
gy | -0.016 |
Beth | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
five | +0.232 |
agging | +0.224 |
sidelines | +0.214 |
Deal | +0.213 |
fourteen | +0.213 |
Neg logit contributions | |
Pig | -0.132 |
gy | -0.119 |
Tweet | -0.117 |
Bunny | -0.112 |
wood | -0.111 |
Token | stood |
Token ID | 6204 |
Feature activation | +4.09e-01 |
Pos logit contributions | |
wood | +0.007 |
Pig | +0.007 |
Ally | +0.006 |
Bert | +0.006 |
Beth | +0.006 |
Neg logit contributions | |
five | -0.007 |
cheek | -0.007 |
mine | -0.006 |
forehead | -0.006 |
la | -0.006 |
Token | there |
Token ID | 612 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
Pig | +0.014 |
Tweet | +0.013 |
Beth | +0.013 |
Ally | +0.012 |
wood | +0.012 |
Neg logit contributions | |
agging | -0.019 |
five | -0.019 |
cheek | -0.018 |
mine | -0.017 |
fourteen | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | -2.96e-01 |
Pos logit contributions | |
Pig | +0.052 |
Beth | +0.047 |
Tweet | +0.046 |
Ally | +0.046 |
wood | +0.045 |
Neg logit contributions | |
five | -0.055 |
agging | -0.054 |
cheek | -0.053 |
mine | -0.047 |
fourteen | -0.047 |
Token | feeling |
Token ID | 4203 |
Feature activation | -7.17e-01 |
Pos logit contributions | |
five | +0.016 |
agging | +0.015 |
cheek | +0.014 |
arl | +0.014 |
Deal | +0.014 |
Neg logit contributions | |
Pig | -0.009 |
Tweet | -0.008 |
Bunny | -0.007 |
known | -0.007 |
Beth | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -4.24e-01 |
Pos logit contributions | |
five | +0.050 |
agging | +0.050 |
Deal | +0.047 |
cheek | +0.047 |
sidelines | +0.046 |
Neg logit contributions | |
Pig | -0.036 |
known | -0.034 |
Tweet | -0.033 |
gy | -0.032 |
Bunny | -0.031 |
Token | I |
Token ID | 314 |
Feature activation | -4.51e-02 |
Pos logit contributions | |
five | +0.019 |
cheek | +0.018 |
agging | +0.018 |
Deal | +0.017 |
fourteen | +0.017 |
Neg logit contributions | |
Pig | -0.015 |
wood | -0.013 |
known | -0.013 |
gy | -0.013 |
Tweet | -0.013 |
Token | â |
Token ID | 22940 |
Feature activation | +7.50e-02 |
Pos logit contributions | |
five | +0.002 |
cheek | +0.001 |
agging | +0.001 |
mine | +0.001 |
goodness | +0.001 |
Neg logit contributions | |
Pig | -0.002 |
Tweet | -0.002 |
wood | -0.002 |
Ally | -0.002 |
Beth | -0.002 |
Token | € |
Token ID | 26391 |
Feature activation | +6.41e-01 |
Pos logit contributions | |
Pig | +0.004 |
Tweet | +0.003 |
Bunny | +0.003 |
Beth | +0.003 |
Ally | +0.003 |
Neg logit contributions | |
five | -0.003 |
agging | -0.003 |
cheek | -0.002 |
fourteen | -0.002 |
sidelines | -0.002 |
Token | ™ |
Token ID | 8151 |
Feature activation | +8.52e-01 |
Pos logit contributions | |
Pig | +0.030 |
Tweet | +0.028 |
wood | +0.028 |
Bunny | +0.027 |
Beth | +0.027 |
Neg logit contributions | |
five | -0.022 |
agging | -0.020 |
cheek | -0.020 |
la | -0.019 |
goodness | -0.018 |
Token | ll |
Token ID | 297 |
Feature activation | -4.48e+00 |
Pos logit contributions | |
Pig | +0.038 |
Tweet | +0.035 |
Beth | +0.034 |
Bunny | +0.034 |
known | +0.034 |
Neg logit contributions | |
five | -0.031 |
agging | -0.030 |
cheek | -0.029 |
Deal | -0.026 |
sidelines | -0.026 |
Token | keep |
Token ID | 1394 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
five | +0.186 |
agging | +0.178 |
cheek | +0.171 |
fourteen | +0.162 |
Deal | +0.160 |
Neg logit contributions | |
Pig | -0.182 |
Tweet | -0.164 |
Beth | -0.162 |
Bunny | -0.160 |
gy | -0.160 |
Token | her |
Token ID | 607 |
Feature activation | -7.01e-01 |
Pos logit contributions | |
five | +0.014 |
agging | +0.013 |
cheek | +0.013 |
fourteen | +0.012 |
mine | +0.012 |
Neg logit contributions | |
Pig | -0.015 |
Tweet | -0.013 |
Beth | -0.013 |
Bunny | -0.013 |
gy | -0.013 |
Token | obedient |
Token ID | 41652 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
cheek | +0.029 |
five | +0.028 |
agging | +0.028 |
mine | +0.027 |
Deal | +0.025 |
Neg logit contributions | |
Pig | -0.028 |
Tweet | -0.025 |
known | -0.025 |
wood | -0.025 |
Ally | -0.025 |
Token | and |
Token ID | 290 |
Feature activation | -4.72e-01 |
Pos logit contributions | |
five | +0.073 |
cheek | +0.071 |
agging | +0.070 |
mine | +0.064 |
la | +0.062 |
Neg logit contributions | |
Pig | -0.068 |
Tweet | -0.063 |
known | -0.062 |
Bert | -0.060 |
Beth | -0.060 |
Token | make |
Token ID | 787 |
Feature activation | +7.75e-01 |
Pos logit contributions | |
five | +0.017 |
agging | +0.016 |
cheek | +0.015 |
mine | +0.014 |
fourteen | +0.014 |
Neg logit contributions | |
Pig | -0.022 |
Tweet | -0.020 |
Beth | -0.020 |
Bunny | -0.020 |
gy | -0.020 |
Token | stayed |
Token ID | 9658 |
Feature activation | -5.00e-02 |
Pos logit contributions | |
five | +0.020 |
cheek | +0.020 |
agging | +0.019 |
mine | +0.018 |
fourteen | +0.018 |
Neg logit contributions | |
Pig | -0.019 |
wood | -0.018 |
Beth | -0.018 |
Tweet | -0.017 |
Ally | -0.017 |
Token | on |
Token ID | 319 |
Feature activation | -4.04e-01 |
Pos logit contributions | |
agging | +0.002 |
cheek | +0.002 |
five | +0.002 |
mine | +0.002 |
goodness | +0.002 |
Neg logit contributions | |
Pig | -0.002 |
Tweet | -0.002 |
wood | -0.002 |
Beth | -0.002 |
Bert | -0.002 |
Token | the |
Token ID | 262 |
Feature activation | -3.92e-01 |
Pos logit contributions | |
mine | +0.016 |
five | +0.016 |
agging | +0.015 |
cheek | +0.015 |
arrest | +0.015 |
Neg logit contributions | |
Pig | -0.015 |
wood | -0.015 |
gy | -0.013 |
Beth | -0.013 |
known | -0.013 |
Token | l |
Token ID | 300 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
five | +0.018 |
Deal | +0.018 |
agging | +0.017 |
arl | +0.017 |
la | +0.016 |
Neg logit contributions | |
Pig | -0.014 |
Tweet | -0.013 |
Bunny | -0.013 |
dam | -0.012 |
shelter | -0.012 |
Token | ily |
Token ID | 813 |
Feature activation | -4.96e-01 |
Pos logit contributions | |
cheek | +0.074 |
five | +0.073 |
Deal | +0.068 |
mine | +0.066 |
agging | +0.066 |
Neg logit contributions | |
Pig | -0.089 |
Tweet | -0.085 |
Bunny | -0.081 |
house | -0.080 |
dam | -0.079 |
Token | pad |
Token ID | 14841 |
Feature activation | -4.46e+00 |
Pos logit contributions | |
five | +0.017 |
agging | +0.016 |
cheek | +0.016 |
mine | +0.014 |
fourteen | +0.014 |
Neg logit contributions | |
Pig | -0.023 |
Tweet | -0.022 |
Beth | -0.021 |
known | -0.021 |
Ally | -0.021 |
Token | and |
Token ID | 290 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
five | +0.213 |
agging | +0.207 |
cheek | +0.200 |
fourteen | +0.188 |
Deal | +0.187 |
Neg logit contributions | |
Pig | -0.152 |
Tweet | -0.135 |
Bunny | -0.131 |
Beth | -0.130 |
gy | -0.130 |
Token | watched |
Token ID | 7342 |
Feature activation | -9.30e-01 |
Pos logit contributions | |
Pig | +0.000 |
wood | +0.000 |
Beth | +0.000 |
Bert | +0.000 |
Ally | +0.000 |
Neg logit contributions | |
five | -0.001 |
cheek | -0.001 |
mine | -0.000 |
fourteen | -0.000 |
agging | -0.000 |
Token | the |
Token ID | 262 |
Feature activation | +2.65e-03 |
Pos logit contributions | |
five | +0.044 |
cheek | +0.042 |
agging | +0.042 |
mine | +0.039 |
fourteen | +0.038 |
Neg logit contributions | |
Pig | -0.032 |
known | -0.029 |
gy | -0.029 |
wood | -0.029 |
Beth | -0.027 |
Token | butterfly |
Token ID | 35113 |
Feature activation | +5.50e-01 |
Pos logit contributions | |
Pig | +0.000 |
Tweet | +0.000 |
Bunny | +0.000 |
Beth | +0.000 |
Ally | +0.000 |
Neg logit contributions | |
five | -0.000 |
agging | -0.000 |
cheek | -0.000 |
fourteen | -0.000 |
Deal | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
Pig | +0.022 |
known | +0.020 |
Beth | +0.020 |
wood | +0.019 |
Tweet | +0.019 |
Neg logit contributions | |
five | -0.023 |
cheek | -0.022 |
agging | -0.022 |
mine | -0.020 |
la | -0.019 |
Token | Bang |
Token ID | 9801 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
five | +0.021 |
agging | +0.020 |
cheek | +0.019 |
fourteen | +0.017 |
mine | +0.017 |
Neg logit contributions | |
Pig | -0.033 |
Tweet | -0.030 |
Beth | -0.030 |
Bunny | -0.029 |
Ally | -0.029 |
Token | ! |
Token ID | 0 |
Feature activation | -7.39e-01 |
Pos logit contributions | |
five | +0.045 |
agging | +0.043 |
cheek | +0.043 |
Deal | +0.039 |
sidelines | +0.038 |
Neg logit contributions | |
Pig | -0.049 |
Tweet | -0.045 |
Beth | -0.044 |
Bunny | -0.044 |
Bert | -0.043 |
Token | You |
Token ID | 921 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
five | +0.025 |
agging | +0.025 |
cheek | +0.023 |
fourteen | +0.021 |
sidelines | +0.021 |
Neg logit contributions | |
Pig | -0.036 |
Tweet | -0.033 |
Beth | -0.032 |
Bunny | -0.032 |
gy | -0.032 |
Token | 're |
Token ID | 821 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
Pig | +0.010 |
Tweet | +0.009 |
gy | +0.009 |
Bunny | +0.009 |
Beth | +0.009 |
Neg logit contributions | |
five | -0.009 |
agging | -0.009 |
cheek | -0.009 |
sidelines | -0.008 |
Deal | -0.008 |
Token | under |
Token ID | 739 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
agging | +0.065 |
five | +0.065 |
mine | +0.062 |
cheek | +0.062 |
fourteen | +0.059 |
Neg logit contributions | |
Pig | -0.059 |
Beth | -0.052 |
Ally | -0.052 |
wood | -0.052 |
Tweet | -0.051 |
Token | arrest |
Token ID | 3251 |
Feature activation | -4.46e+00 |
Pos logit contributions | |
five | +0.031 |
agging | +0.029 |
cheek | +0.028 |
Deal | +0.026 |
arl | +0.024 |
Neg logit contributions | |
Pig | -0.062 |
Tweet | -0.060 |
Bunny | -0.058 |
Ch | -0.057 |
Beth | -0.057 |
Token | !" |
Token ID | 2474 |
Feature activation | -5.61e-01 |
Pos logit contributions | |
agging | +0.135 |
five | +0.133 |
cheek | +0.128 |
Deal | +0.124 |
leaf | +0.117 |
Neg logit contributions | |
Pig | -0.218 |
Tweet | -0.210 |
Beth | -0.200 |
Bunny | -0.199 |
gy | -0.199 |
Token |
|
Token ID | 198 |
Feature activation | +4.68e-01 |
Pos logit contributions | |
five | +0.024 |
cheek | +0.023 |
agging | +0.022 |
fourteen | +0.021 |
la | +0.020 |
Neg logit contributions | |
Pig | -0.021 |
known | -0.020 |
wood | -0.020 |
gy | -0.019 |
increasingly | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | +2.30e-01 |
Pos logit contributions | |
Pig | +0.016 |
wood | +0.014 |
gy | +0.014 |
Tweet | +0.013 |
known | +0.013 |
Neg logit contributions | |
five | -0.022 |
agging | -0.022 |
cheek | -0.021 |
fourteen | -0.020 |
arl | -0.020 |
Token | Their |
Token ID | 14574 |
Feature activation | -6.05e-01 |
Pos logit contributions | |
Pig | +0.010 |
Tweet | +0.009 |
Beth | +0.009 |
Bunny | +0.009 |
Ally | +0.009 |
Neg logit contributions | |
five | -0.009 |
agging | -0.009 |
cheek | -0.009 |
fourteen | -0.008 |
mine | -0.008 |
Token | mom |
Token ID | 1995 |
Feature activation | +5.20e-01 |
Pos logit contributions | |
five | +0.025 |
agging | +0.025 |
cheek | +0.024 |
mine | +0.023 |
fourteen | +0.022 |
Neg logit contributions | |
Pig | -0.024 |
Tweet | -0.021 |
Beth | -0.021 |
Ally | -0.021 |
known | -0.021 |
Token | found |
Token ID | 1043 |
Feature activation | -9.37e-01 |
Pos logit contributions | |
five | +0.021 |
fourteen | +0.018 |
agging | +0.018 |
mine | +0.018 |
cheek | +0.018 |
Neg logit contributions | |
Pig | -0.020 |
Beth | -0.018 |
wood | -0.018 |
Ally | -0.018 |
Tweet | -0.018 |
Token | a |
Token ID | 257 |
Feature activation | -3.79e-01 |
Pos logit contributions | |
five | +0.042 |
mine | +0.038 |
agging | +0.038 |
cheek | +0.038 |
goodness | +0.037 |
Neg logit contributions | |
Pig | -0.033 |
gy | -0.030 |
Ally | -0.029 |
wood | -0.029 |
Beth | -0.029 |
Token | big |
Token ID | 1263 |
Feature activation | +2.67e-02 |
Pos logit contributions | |
five | +0.015 |
cheek | +0.015 |
agging | +0.014 |
mine | +0.014 |
sidelines | +0.013 |
Neg logit contributions | |
Pig | -0.016 |
Tweet | -0.014 |
Beth | -0.014 |
wood | -0.014 |
Ally | -0.014 |
Token | l |
Token ID | 300 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
Pig | +0.001 |
Tweet | +0.001 |
Bunny | +0.001 |
Beth | +0.001 |
Ally | +0.001 |
Neg logit contributions | |
five | -0.001 |
agging | -0.001 |
cheek | -0.001 |
fourteen | -0.001 |
Deal | -0.001 |
Token | ily |
Token ID | 813 |
Feature activation | -4.21e-01 |
Pos logit contributions | |
cheek | +0.076 |
five | +0.075 |
Deal | +0.070 |
sidelines | +0.067 |
mine | +0.067 |
Neg logit contributions | |
Pig | -0.070 |
called | -0.069 |
known | -0.066 |
Tweet | -0.066 |
dam | -0.065 |
Token | pad |
Token ID | 14841 |
Feature activation | -4.46e+00 |
Pos logit contributions | |
five | +0.012 |
agging | +0.011 |
cheek | +0.010 |
Deal | +0.009 |
arl | +0.009 |
Neg logit contributions | |
Pig | -0.023 |
Tweet | -0.021 |
gy | -0.021 |
Bunny | -0.021 |
Beth | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | -6.03e-01 |
Pos logit contributions | |
five | +0.231 |
agging | +0.223 |
Deal | +0.219 |
sidelines | +0.217 |
cheek | +0.212 |
Neg logit contributions | |
Pig | -0.127 |
gy | -0.123 |
Tweet | -0.117 |
wood | -0.113 |
house | -0.109 |
Token | He |
Token ID | 679 |
Feature activation | -2.55e-01 |
Pos logit contributions | |
five | +0.027 |
cheek | +0.025 |
agging | +0.025 |
Deal | +0.024 |
mine | +0.024 |
Neg logit contributions | |
Pig | -0.021 |
known | -0.019 |
wood | -0.019 |
gy | -0.019 |
darker | -0.018 |
Token | quickly |
Token ID | 2952 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
five | +0.010 |
mine | +0.010 |
cheek | +0.010 |
fourteen | +0.009 |
ige | +0.009 |
Neg logit contributions | |
Pig | -0.010 |
wood | -0.009 |
Beth | -0.009 |
Ally | -0.009 |
gy | -0.008 |
Token | hopped |
Token ID | 45230 |
Feature activation | -2.58e-01 |
Pos logit contributions | |
five | +0.083 |
agging | +0.074 |
cheek | +0.073 |
mine | +0.070 |
fourteen | +0.069 |
Neg logit contributions | |
Pig | -0.096 |
wood | -0.089 |
Beth | -0.088 |
known | -0.087 |
Bert | -0.086 |
Token | on |
Token ID | 319 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
five | +0.012 |
agging | +0.012 |
cheek | +0.011 |
mine | +0.011 |
fourteen | +0.011 |
Neg logit contributions | |
Pig | -0.009 |
known | -0.008 |
Beth | -0.008 |
gy | -0.008 |
Tweet | -0.008 |
Token | He |
Token ID | 679 |
Feature activation | +6.01e-02 |
Pos logit contributions | |
Pig | +0.002 |
known | +0.002 |
gy | +0.001 |
darker | +0.001 |
wood | +0.001 |
Neg logit contributions | |
five | -0.002 |
Deal | -0.002 |
cheek | -0.002 |
agging | -0.002 |
goodness | -0.002 |
Token | found |
Token ID | 1043 |
Feature activation | -6.48e-01 |
Pos logit contributions | |
Pig | +0.002 |
wood | +0.002 |
Ally | +0.002 |
Beth | +0.002 |
Tweet | +0.002 |
Neg logit contributions | |
five | -0.002 |
cheek | -0.002 |
mine | -0.002 |
agging | -0.002 |
fourteen | -0.002 |
Token | a |
Token ID | 257 |
Feature activation | -1.87e-01 |
Pos logit contributions | |
five | +0.028 |
fourteen | +0.024 |
goodness | +0.024 |
mine | +0.024 |
agging | +0.024 |
Neg logit contributions | |
Pig | -0.023 |
gy | -0.022 |
house | -0.021 |
Beth | -0.021 |
Ally | -0.020 |
Token | l |
Token ID | 300 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
five | +0.008 |
agging | +0.007 |
cheek | +0.007 |
mine | +0.007 |
sidelines | +0.007 |
Neg logit contributions | |
Pig | -0.008 |
Tweet | -0.007 |
Beth | -0.007 |
gy | -0.007 |
Ally | -0.007 |
Token | ily |
Token ID | 813 |
Feature activation | -3.31e-01 |
Pos logit contributions | |
cheek | +0.066 |
five | +0.062 |
mine | +0.058 |
sidelines | +0.056 |
Deal | +0.056 |
Neg logit contributions | |
Pig | -0.068 |
known | -0.065 |
Tweet | -0.064 |
Bunny | -0.063 |
Bert | -0.061 |
Token | pad |
Token ID | 14841 |
Feature activation | -4.45e+00 |
Pos logit contributions | |
five | +0.012 |
agging | +0.012 |
cheek | +0.011 |
fourteen | +0.010 |
Deal | +0.010 |
Neg logit contributions | |
Pig | -0.015 |
Tweet | -0.014 |
Bunny | -0.013 |
Beth | -0.013 |
gy | -0.013 |
Token | in |
Token ID | 287 |
Feature activation | -5.70e-01 |
Pos logit contributions | |
five | +0.225 |
agging | +0.222 |
cheek | +0.212 |
Deal | +0.210 |
sidelines | +0.206 |
Neg logit contributions | |
Pig | -0.134 |
Tweet | -0.121 |
gy | -0.121 |
wood | -0.115 |
Bunny | -0.115 |
Token | the |
Token ID | 262 |
Feature activation | -2.78e-01 |
Pos logit contributions | |
five | +0.026 |
cheek | +0.025 |
agging | +0.024 |
mine | +0.024 |
fourteen | +0.024 |
Neg logit contributions | |
Pig | -0.020 |
wood | -0.017 |
gy | -0.017 |
Ally | -0.017 |
Tweet | -0.017 |
Token | pond |
Token ID | 16723 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
five | +0.012 |
agging | +0.012 |
cheek | +0.012 |
fourteen | +0.011 |
Deal | +0.011 |
Neg logit contributions | |
Pig | -0.011 |
Tweet | -0.010 |
Bunny | -0.009 |
Beth | -0.009 |
Ally | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | +4.03e-01 |
Pos logit contributions | |
five | +0.054 |
agging | +0.052 |
cheek | +0.051 |
mine | +0.047 |
fourteen | +0.047 |
Neg logit contributions | |
Pig | -0.046 |
Beth | -0.041 |
Tweet | -0.041 |
Ally | -0.040 |
Bert | -0.040 |
Token | rested |
Token ID | 29263 |
Feature activation | +8.73e-01 |
Pos logit contributions | |
Pig | +0.015 |
wood | +0.014 |
gy | +0.013 |
Ally | +0.013 |
Beth | +0.013 |
Neg logit contributions | |
five | -0.017 |
cheek | -0.016 |
mine | -0.015 |
fourteen | -0.014 |
sidelines | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | +5.45e-02 |
Pos logit contributions | |
five | +0.017 |
agging | +0.016 |
cheek | +0.016 |
fourteen | +0.015 |
mine | +0.015 |
Neg logit contributions | |
Pig | -0.013 |
wood | -0.012 |
gy | -0.012 |
Beth | -0.012 |
Ally | -0.012 |
Token | " |
Token ID | 366 |
Feature activation | +5.86e-01 |
Pos logit contributions | |
Pig | +0.002 |
Tweet | +0.002 |
Beth | +0.002 |
Bunny | +0.002 |
Ally | +0.002 |
Neg logit contributions | |
five | -0.002 |
agging | -0.002 |
cheek | -0.002 |
fourteen | -0.002 |
mine | -0.002 |
Token | You |
Token ID | 1639 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
Pig | +0.026 |
known | +0.023 |
Tweet | +0.023 |
Bunny | +0.023 |
Beth | +0.022 |
Neg logit contributions | |
agging | -0.023 |
five | -0.022 |
cheek | -0.021 |
mine | -0.019 |
fourteen | -0.019 |
Token | are |
Token ID | 389 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
five | +0.009 |
agging | +0.009 |
cheek | +0.009 |
fourteen | +0.008 |
mine | +0.008 |
Neg logit contributions | |
Pig | -0.011 |
Tweet | -0.010 |
Beth | -0.010 |
Bunny | -0.010 |
Ally | -0.010 |
Token | under |
Token ID | 739 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
Pig | +0.005 |
Tweet | +0.004 |
Bunny | +0.004 |
known | +0.004 |
Beth | +0.004 |
Neg logit contributions | |
five | -0.008 |
agging | -0.008 |
cheek | -0.008 |
fourteen | -0.007 |
arl | -0.007 |
Token | arrest |
Token ID | 3251 |
Feature activation | -4.45e+00 |
Pos logit contributions | |
five | +0.034 |
agging | +0.032 |
cheek | +0.031 |
Deal | +0.026 |
fourteen | +0.026 |
Neg logit contributions | |
Pig | -0.077 |
Tweet | -0.071 |
Bunny | -0.070 |
Beth | -0.070 |
gy | -0.069 |
Token | for |
Token ID | 329 |
Feature activation | -7.40e-01 |
Pos logit contributions | |
five | +0.191 |
agging | +0.187 |
cheek | +0.182 |
Deal | +0.172 |
fourteen | +0.168 |
Neg logit contributions | |
Pig | -0.169 |
Tweet | -0.155 |
Bunny | -0.149 |
Beth | -0.147 |
gy | -0.146 |
Token | harassing |
Token ID | 34110 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
five | +0.037 |
agging | +0.035 |
cheek | +0.034 |
mine | +0.033 |
fourteen | +0.033 |
Neg logit contributions | |
Pig | -0.024 |
Tweet | -0.021 |
wood | -0.021 |
Ally | -0.021 |
Beth | -0.021 |
Token | and |
Token ID | 290 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
Pig | +0.051 |
Tweet | +0.046 |
Beth | +0.045 |
Ally | +0.045 |
Bunny | +0.044 |
Neg logit contributions | |
five | -0.056 |
agging | -0.056 |
cheek | -0.052 |
arl | -0.050 |
mine | -0.049 |
Token | stealing |
Token ID | 16568 |
Feature activation | -5.10e-01 |
Pos logit contributions | |
five | +0.073 |
cheek | +0.071 |
agging | +0.070 |
mine | +0.066 |
sidelines | +0.064 |
Neg logit contributions | |
Pig | -0.089 |
Ally | -0.080 |
Tweet | -0.080 |
Beth | -0.079 |
gy | -0.079 |
Token | from |
Token ID | 422 |
Feature activation | +9.23e-01 |
Pos logit contributions | |
five | +0.022 |
agging | +0.021 |
cheek | +0.020 |
sidelines | +0.019 |
Deal | +0.019 |
Neg logit contributions | |
Pig | -0.020 |
Tweet | -0.018 |
Bunny | -0.018 |
Beth | -0.018 |
wood | -0.017 |
Token | hopped |
Token ID | 45230 |
Feature activation | -5.38e-01 |
Pos logit contributions | |
Pig | +0.011 |
wood | +0.011 |
Beth | +0.011 |
Tweet | +0.010 |
ained | +0.010 |
Neg logit contributions | |
five | -0.011 |
cheek | -0.011 |
mine | -0.010 |
wo | -0.010 |
agging | -0.010 |
Token | off |
Token ID | 572 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
five | +0.024 |
agging | +0.023 |
cheek | +0.023 |
mine | +0.021 |
fourteen | +0.021 |
Neg logit contributions | |
Pig | -0.020 |
known | -0.018 |
gy | -0.017 |
wood | -0.017 |
Tweet | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | -3.87e-01 |
Pos logit contributions | |
five | +0.068 |
cheek | +0.064 |
agging | +0.064 |
fourteen | +0.059 |
mine | +0.059 |
Neg logit contributions | |
Pig | -0.061 |
wood | -0.056 |
known | -0.054 |
gy | -0.054 |
Beth | -0.052 |
Token | l |
Token ID | 300 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
five | +0.018 |
agging | +0.017 |
Deal | +0.017 |
ige | +0.016 |
fourteen | +0.016 |
Neg logit contributions | |
Pig | -0.014 |
Bunny | -0.013 |
Tweet | -0.013 |
shelter | -0.012 |
Beth | -0.012 |
Token | ily |
Token ID | 813 |
Feature activation | -4.84e-01 |
Pos logit contributions | |
cheek | +0.076 |
five | +0.074 |
Deal | +0.069 |
mine | +0.068 |
agging | +0.067 |
Neg logit contributions | |
Pig | -0.092 |
Tweet | -0.087 |
Bunny | -0.083 |
Ally | -0.082 |
Bert | -0.081 |
Token | pad |
Token ID | 14841 |
Feature activation | -4.43e+00 |
Pos logit contributions | |
five | +0.014 |
agging | +0.013 |
cheek | +0.013 |
mine | +0.011 |
fourteen | +0.011 |
Neg logit contributions | |
Pig | -0.026 |
Tweet | -0.024 |
Beth | -0.023 |
known | -0.023 |
Bunny | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | -7.94e-01 |
Pos logit contributions | |
five | +0.213 |
agging | +0.206 |
cheek | +0.199 |
fourteen | +0.189 |
Deal | +0.189 |
Neg logit contributions | |
Pig | -0.150 |
Tweet | -0.133 |
gy | -0.129 |
Bunny | -0.128 |
Beth | -0.127 |
Token | looked |
Token ID | 3114 |
Feature activation | -6.96e-01 |
Pos logit contributions | |
five | +0.036 |
agging | +0.035 |
cheek | +0.034 |
fourteen | +0.032 |
mine | +0.031 |
Neg logit contributions | |
Pig | -0.029 |
Tweet | -0.026 |
Beth | -0.025 |
Bunny | -0.025 |
wood | -0.025 |
Token | around |
Token ID | 1088 |
Feature activation | -6.83e-01 |
Pos logit contributions | |
five | +0.033 |
agging | +0.033 |
cheek | +0.032 |
mine | +0.030 |
fourteen | +0.029 |
Neg logit contributions | |
Pig | -0.024 |
Tweet | -0.021 |
Beth | -0.020 |
gy | -0.020 |
wood | -0.020 |
Token | , |
Token ID | 11 |
Feature activation | -2.72e-01 |
Pos logit contributions | |
five | +0.030 |
agging | +0.030 |
cheek | +0.029 |
mine | +0.027 |
fourteen | +0.026 |
Neg logit contributions | |
Pig | -0.025 |
Tweet | -0.022 |
wood | -0.022 |
gy | -0.022 |
Beth | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
five | +0.012 |
agging | +0.012 |
cheek | +0.012 |
fourteen | +0.011 |
mine | +0.011 |
Neg logit contributions | |
Pig | -0.010 |
Tweet | -0.009 |
Beth | -0.009 |
Bunny | -0.009 |
wood | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -4.13e-01 |
Pos logit contributions | |
Pig | +0.027 |
Tweet | +0.024 |
Bunny | +0.023 |
gy | +0.023 |
Beth | +0.023 |
Neg logit contributions | |
five | -0.041 |
agging | -0.039 |
cheek | -0.038 |
Deal | -0.036 |
fourteen | -0.036 |
Token | The |
Token ID | 383 |
Feature activation | +8.11e-02 |
Pos logit contributions | |
five | +0.019 |
cheek | +0.018 |
agging | +0.017 |
la | +0.017 |
fourteen | +0.017 |
Neg logit contributions | |
known | -0.013 |
Pig | -0.012 |
gy | -0.012 |
wood | -0.012 |
darker | -0.011 |
Token | bird |
Token ID | 6512 |
Feature activation | +4.25e-01 |
Pos logit contributions | |
Pig | +0.003 |
Tweet | +0.003 |
Bunny | +0.002 |
Beth | +0.002 |
Ally | +0.002 |
Neg logit contributions | |
five | -0.004 |
agging | -0.004 |
cheek | -0.004 |
fourteen | -0.003 |
Deal | -0.003 |
Token | saw |
Token ID | 2497 |
Feature activation | -8.34e-01 |
Pos logit contributions | |
Pig | +0.013 |
wood | +0.011 |
Beth | +0.011 |
known | +0.011 |
Tweet | +0.011 |
Neg logit contributions | |
five | -0.021 |
agging | -0.021 |
cheek | -0.021 |
mine | -0.019 |
fourteen | -0.019 |
Token | O |
Token ID | 440 |
Feature activation | -8.22e-01 |
Pos logit contributions | |
five | +0.039 |
goodness | +0.038 |
fourteen | +0.038 |
mine | +0.037 |
agging | +0.036 |
Neg logit contributions | |
Pig | -0.024 |
known | -0.023 |
wood | -0.023 |
gy | -0.023 |
house | -0.023 |
Token | ll |
Token ID | 297 |
Feature activation | -4.40e+00 |
Pos logit contributions | |
five | +0.025 |
agging | +0.022 |
cheek | +0.022 |
Deal | +0.019 |
fourteen | +0.019 |
Neg logit contributions | |
Pig | -0.044 |
Tweet | -0.040 |
gy | -0.039 |
Bunny | -0.039 |
Beth | -0.039 |
Token | ie |
Token ID | 494 |
Feature activation | -4.36e-01 |
Pos logit contributions | |
five | +0.287 |
agging | +0.274 |
Deal | +0.274 |
fourteen | +0.266 |
arl | +0.265 |
Neg logit contributions | |
Pig | -0.075 |
Tweet | -0.069 |
gy | -0.063 |
Ally | -0.058 |
Bunny | -0.057 |
Token | and |
Token ID | 290 |
Feature activation | -3.22e-01 |
Pos logit contributions | |
cheek | +0.018 |
agging | +0.017 |
five | +0.016 |
mine | +0.016 |
liking | +0.015 |
Neg logit contributions | |
Pig | -0.017 |
known | -0.017 |
Beth | -0.016 |
wood | -0.016 |
Ally | -0.016 |
Token | wanted |
Token ID | 2227 |
Feature activation | -8.74e-01 |
Pos logit contributions | |
five | +0.013 |
cheek | +0.013 |
wo | +0.012 |
mine | +0.012 |
la | +0.012 |
Neg logit contributions | |
wood | -0.011 |
Pig | -0.011 |
darker | -0.010 |
house | -0.010 |
gy | -0.010 |
Token | to |
Token ID | 284 |
Feature activation | +8.94e-02 |
Pos logit contributions | |
five | +0.032 |
mine | +0.030 |
cheek | +0.029 |
agging | +0.027 |
fourteen | +0.027 |
Neg logit contributions | |
Pig | -0.036 |
gy | -0.035 |
house | -0.034 |
Ally | -0.034 |
wood | -0.033 |
Token | eat |
Token ID | 4483 |
Feature activation | -6.61e-01 |
Pos logit contributions | |
Pig | +0.004 |
Tweet | +0.004 |
Bunny | +0.003 |
Beth | +0.003 |
gy | +0.003 |
Neg logit contributions | |
agging | -0.004 |
five | -0.004 |
cheek | -0.003 |
fourteen | -0.003 |
Deal | -0.003 |
Token | him |
Token ID | 683 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
Pig | +0.009 |
known | +0.009 |
gy | +0.008 |
wood | +0.008 |
Tweet | +0.008 |
Neg logit contributions | |
agging | -0.010 |
five | -0.009 |
cheek | -0.009 |
mine | -0.009 |
goodness | -0.008 |
Token | on |
Token ID | 319 |
Feature activation | -4.17e-01 |
Pos logit contributions | |
Pig | +0.001 |
known | +0.001 |
Ally | +0.001 |
Beth | +0.001 |
Tweet | +0.001 |
Neg logit contributions | |
five | -0.001 |
cheek | -0.001 |
agging | -0.001 |
mine | -0.001 |
la | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | -4.81e-01 |
Pos logit contributions | |
mine | +0.017 |
agging | +0.016 |
five | +0.015 |
fourteen | +0.015 |
arrest | +0.015 |
Neg logit contributions | |
Pig | -0.015 |
known | -0.014 |
wood | -0.014 |
gy | -0.014 |
Ally | -0.014 |
Token | l |
Token ID | 300 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
five | +0.022 |
agging | +0.021 |
Deal | +0.020 |
cheek | +0.019 |
arl | +0.019 |
Neg logit contributions | |
Pig | -0.018 |
Tweet | -0.016 |
Bunny | -0.016 |
Ally | -0.015 |
Beth | -0.015 |
Token | ily |
Token ID | 813 |
Feature activation | -4.62e-01 |
Pos logit contributions | |
cheek | +0.071 |
five | +0.067 |
mine | +0.060 |
Deal | +0.059 |
agging | +0.059 |
Neg logit contributions | |
Pig | -0.093 |
Tweet | -0.088 |
Bunny | -0.086 |
known | -0.084 |
dam | -0.083 |
Token | pad |
Token ID | 14841 |
Feature activation | -4.37e+00 |
Pos logit contributions | |
five | +0.018 |
cheek | +0.017 |
agging | +0.017 |
mine | +0.015 |
la | +0.015 |
Neg logit contributions | |
Pig | -0.020 |
known | -0.018 |
Tweet | -0.018 |
Beth | -0.018 |
Ally | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
five | +0.212 |
agging | +0.206 |
cheek | +0.200 |
fourteen | +0.188 |
Deal | +0.186 |
Neg logit contributions | |
Pig | -0.146 |
Tweet | -0.129 |
Beth | -0.125 |
Bunny | -0.125 |
gy | -0.123 |
Token | They |
Token ID | 1119 |
Feature activation | -7.63e-02 |
Pos logit contributions | |
five | +0.011 |
agging | +0.010 |
cheek | +0.010 |
mine | +0.010 |
fourteen | +0.010 |
Neg logit contributions | |
Pig | -0.008 |
known | -0.008 |
gy | -0.008 |
wood | -0.007 |
Beth | -0.007 |
Token | all |
Token ID | 477 |
Feature activation | -2.30e+00 |
Pos logit contributions | |
five | +0.003 |
agging | +0.003 |
cheek | +0.003 |
fourteen | +0.003 |
mine | +0.003 |
Neg logit contributions | |
Pig | -0.003 |
wood | -0.003 |
gy | -0.003 |
Tweet | -0.003 |
house | -0.002 |
Token | rested |
Token ID | 29263 |
Feature activation | -4.05e-02 |
Pos logit contributions | |
five | +0.088 |
agging | +0.081 |
cheek | +0.080 |
mine | +0.073 |
fourteen | +0.073 |
Neg logit contributions | |
Pig | -0.098 |
wood | -0.089 |
Tweet | -0.087 |
gy | -0.086 |
Ally | -0.084 |
Token | together |
Token ID | 1978 |
Feature activation | -4.28e-01 |
Pos logit contributions | |
agging | +0.002 |
five | +0.002 |
cheek | +0.002 |
fourteen | +0.002 |
mine | +0.002 |
Neg logit contributions | |
Pig | -0.001 |
Tweet | -0.001 |
Beth | -0.001 |
wood | -0.001 |
Ally | -0.001 |
Token | giant |
Token ID | 6175 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
agging | +0.044 |
five | +0.040 |
cheek | +0.038 |
ater | +0.037 |
arest | +0.036 |
Neg logit contributions | |
Pig | -0.035 |
Tweet | -0.035 |
Bunny | -0.033 |
Beth | -0.033 |
Bert | -0.032 |
Token | ! |
Token ID | 0 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
five | +0.057 |
agging | +0.055 |
cheek | +0.054 |
fourteen | +0.050 |
mine | +0.049 |
Neg logit contributions | |
Pig | -0.045 |
Tweet | -0.040 |
Beth | -0.039 |
Bunny | -0.039 |
Ally | -0.038 |
Token | You |
Token ID | 921 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
agging | +0.067 |
five | +0.063 |
cheek | +0.061 |
sidelines | +0.056 |
ige | +0.054 |
Neg logit contributions | |
Pig | -0.088 |
Tweet | -0.080 |
Bunny | -0.079 |
Beth | -0.078 |
gy | -0.077 |
Token | are |
Token ID | 389 |
Feature activation | -2.64e-01 |
Pos logit contributions | |
five | +0.005 |
agging | +0.005 |
cheek | +0.005 |
fourteen | +0.004 |
sidelines | +0.004 |
Neg logit contributions | |
Pig | -0.007 |
Tweet | -0.006 |
Bunny | -0.006 |
Beth | -0.006 |
gy | -0.006 |
Token | under |
Token ID | 739 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
five | +0.013 |
agging | +0.012 |
cheek | +0.012 |
arl | +0.011 |
fourteen | +0.011 |
Neg logit contributions | |
Pig | -0.010 |
Tweet | -0.009 |
Bunny | -0.008 |
Beth | -0.008 |
known | -0.008 |
Token | arrest |
Token ID | 3251 |
Feature activation | -4.36e+00 |
Pos logit contributions | |
five | +0.036 |
agging | +0.033 |
cheek | +0.032 |
Deal | +0.028 |
ige | +0.025 |
Neg logit contributions | |
Pig | -0.100 |
Tweet | -0.095 |
Bunny | -0.092 |
Beth | -0.091 |
gy | -0.090 |
Token | ! |
Token ID | 0 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
five | +0.170 |
agging | +0.168 |
cheek | +0.163 |
Deal | +0.156 |
leaf | +0.151 |
Neg logit contributions | |
Pig | -0.179 |
Tweet | -0.168 |
Bunny | -0.160 |
Beth | -0.158 |
gy | -0.157 |
Token | You |
Token ID | 921 |
Feature activation | -6.24e-02 |
Pos logit contributions | |
agging | +0.046 |
five | +0.045 |
cheek | +0.044 |
sidelines | +0.038 |
fourteen | +0.037 |
Neg logit contributions | |
Pig | -0.078 |
Tweet | -0.072 |
Beth | -0.071 |
Bunny | -0.070 |
Bert | -0.069 |
Token | can |
Token ID | 460 |
Feature activation | +5.42e-01 |
Pos logit contributions | |
five | +0.002 |
agging | +0.002 |
cheek | +0.002 |
fourteen | +0.002 |
sidelines | +0.002 |
Neg logit contributions | |
Pig | -0.003 |
Tweet | -0.003 |
Bunny | -0.003 |
Beth | -0.002 |
gy | -0.002 |
Token | 't |
Token ID | 470 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
Pig | +0.024 |
Tweet | +0.023 |
gy | +0.022 |
Bunny | +0.022 |
Beth | +0.022 |
Neg logit contributions | |
five | -0.020 |
agging | -0.020 |
cheek | -0.019 |
Deal | -0.017 |
arl | -0.017 |
Token | take |
Token ID | 1011 |
Feature activation | -6.68e-01 |
Pos logit contributions | |
five | +0.049 |
agging | +0.049 |
cheek | +0.047 |
Deal | +0.044 |
fourteen | +0.044 |
Neg logit contributions | |
Pig | -0.044 |
Tweet | -0.041 |
gy | -0.039 |
Bunny | -0.039 |
Beth | -0.038 |
Token | , |
Token ID | 11 |
Feature activation | -1.93e-01 |
Pos logit contributions | |
five | +0.023 |
agging | +0.022 |
cheek | +0.022 |
fourteen | +0.020 |
mine | +0.020 |
Neg logit contributions | |
Pig | -0.019 |
Tweet | -0.017 |
Beth | -0.017 |
Bunny | -0.017 |
Ally | -0.016 |
Token | I |
Token ID | 314 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
five | +0.008 |
agging | +0.008 |
cheek | +0.007 |
mine | +0.007 |
fourteen | +0.007 |
Neg logit contributions | |
Pig | -0.008 |
gy | -0.007 |
Tweet | -0.007 |
known | -0.007 |
wood | -0.007 |
Token | â |
Token ID | 22940 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
five | +0.007 |
agging | +0.006 |
cheek | +0.006 |
la | +0.006 |
Deal | +0.005 |
Neg logit contributions | |
Pig | -0.007 |
Tweet | -0.007 |
Beth | -0.007 |
Ally | -0.007 |
Bunny | -0.007 |
Token | € |
Token ID | 26391 |
Feature activation | +5.84e-01 |
Pos logit contributions | |
Pig | +0.006 |
Tweet | +0.006 |
Bunny | +0.006 |
Beth | +0.006 |
Ally | +0.006 |
Neg logit contributions | |
five | -0.005 |
agging | -0.005 |
cheek | -0.005 |
fourteen | -0.004 |
sidelines | -0.004 |
Token | ™ |
Token ID | 8151 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
Pig | +0.024 |
Tweet | +0.022 |
Bunny | +0.021 |
Beth | +0.021 |
wood | +0.021 |
Neg logit contributions | |
five | -0.024 |
agging | -0.023 |
cheek | -0.022 |
la | -0.021 |
fourteen | -0.021 |
Token | ll |
Token ID | 297 |
Feature activation | -4.35e+00 |
Pos logit contributions | |
Pig | +0.041 |
Tweet | +0.038 |
Beth | +0.037 |
Bunny | +0.037 |
gy | +0.036 |
Neg logit contributions | |
five | -0.041 |
agging | -0.039 |
cheek | -0.038 |
fourteen | -0.035 |
sidelines | -0.035 |
Token | allow |
Token ID | 1249 |
Feature activation | -6.84e-01 |
Pos logit contributions | |
five | +0.178 |
agging | +0.171 |
cheek | +0.165 |
fourteen | +0.155 |
Deal | +0.153 |
Neg logit contributions | |
Pig | -0.180 |
Tweet | -0.160 |
Beth | -0.160 |
gy | -0.158 |
Ally | -0.158 |
Token | you |
Token ID | 345 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
five | +0.018 |
agging | +0.017 |
cheek | +0.016 |
mine | +0.015 |
fourteen | +0.015 |
Neg logit contributions | |
Pig | -0.037 |
Tweet | -0.035 |
gy | -0.035 |
Beth | -0.034 |
wood | -0.034 |
Token | to |
Token ID | 284 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
Pig | +0.004 |
Tweet | +0.004 |
Bunny | +0.004 |
Beth | +0.004 |
wood | +0.004 |
Neg logit contributions | |
five | -0.007 |
agging | -0.007 |
cheek | -0.007 |
sidelines | -0.006 |
fourteen | -0.006 |
Token | take |
Token ID | 1011 |
Feature activation | -3.86e-01 |
Pos logit contributions | |
five | +0.008 |
agging | +0.007 |
cheek | +0.007 |
mine | +0.007 |
fourteen | +0.006 |
Neg logit contributions | |
Pig | -0.009 |
Tweet | -0.008 |
Beth | -0.008 |
Bunny | -0.008 |
Ally | -0.008 |
Token | it |
Token ID | 340 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
five | +0.018 |
agging | +0.017 |
cheek | +0.016 |
mine | +0.016 |
fourteen | +0.015 |
Neg logit contributions | |
Pig | -0.014 |
gy | -0.013 |
Tweet | -0.012 |
wood | -0.012 |
known | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
Pig | +0.009 |
Tweet | +0.008 |
Bunny | +0.008 |
Beth | +0.008 |
gy | +0.008 |
Neg logit contributions | |
five | -0.008 |
agging | -0.008 |
cheek | -0.007 |
fourteen | -0.007 |
sidelines | -0.007 |
Token | We |
Token ID | 775 |
Feature activation | -4.98e-02 |
Pos logit contributions | |
five | +0.052 |
agging | +0.049 |
cheek | +0.048 |
fourteen | +0.044 |
Deal | +0.043 |
Neg logit contributions | |
Pig | -0.059 |
Tweet | -0.054 |
Beth | -0.053 |
Bunny | -0.053 |
known | -0.053 |
Token | need |
Token ID | 761 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
five | +0.002 |
agging | +0.002 |
cheek | +0.002 |
fourteen | +0.002 |
mine | +0.002 |
Neg logit contributions | |
Pig | -0.002 |
Tweet | -0.002 |
Beth | -0.002 |
Bunny | -0.002 |
Ally | -0.002 |
Token | a |
Token ID | 257 |
Feature activation | +5.51e-01 |
Pos logit contributions | |
Pig | +0.007 |
Tweet | +0.006 |
Bunny | +0.006 |
Beth | +0.006 |
Ally | +0.005 |
Neg logit contributions | |
agging | -0.011 |
five | -0.011 |
cheek | -0.011 |
Deal | -0.010 |
arl | -0.010 |
Token | grown |
Token ID | 7334 |
Feature activation | +4.72e-01 |
Pos logit contributions | |
Pig | +0.025 |
Tweet | +0.023 |
Bunny | +0.023 |
Ally | +0.022 |
Beth | +0.022 |
Neg logit contributions | |
five | -0.020 |
agging | -0.020 |
cheek | -0.018 |
fourteen | -0.017 |
Deal | -0.017 |
Token | - |
Token ID | 12 |
Feature activation | +3.30e+00 |
Pos logit contributions | |
Pig | +0.020 |
Tweet | +0.019 |
Bunny | +0.019 |
Beth | +0.019 |
Ally | +0.018 |
Neg logit contributions | |
five | -0.020 |
agging | -0.019 |
cheek | -0.018 |
ater | -0.018 |
mine | -0.017 |
Token | up |
Token ID | 929 |
Feature activation | +8.65e-01 |
Pos logit contributions | |
Pig | +0.128 |
Tweet | +0.115 |
Bunny | +0.114 |
Beth | +0.113 |
Ally | +0.110 |
Neg logit contributions | |
five | -0.150 |
agging | -0.137 |
cheek | -0.134 |
fourteen | -0.127 |
mine | -0.124 |
Token | to |
Token ID | 284 |
Feature activation | -7.64e-01 |
Pos logit contributions | |
Pig | +0.033 |
Tweet | +0.030 |
Beth | +0.029 |
Bunny | +0.029 |
Ally | +0.028 |
Neg logit contributions | |
five | -0.038 |
agging | -0.037 |
cheek | -0.036 |
fourteen | -0.033 |
mine | -0.033 |
Token | help |
Token ID | 1037 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
five | +0.030 |
agging | +0.028 |
cheek | +0.028 |
mine | +0.025 |
fourteen | +0.025 |
Neg logit contributions | |
Pig | -0.033 |
Tweet | -0.030 |
Beth | -0.029 |
Bunny | -0.029 |
wood | -0.029 |
Token | us |
Token ID | 514 |
Feature activation | +5.92e-01 |
Pos logit contributions | |
Pig | +0.007 |
Tweet | +0.006 |
Beth | +0.006 |
Bunny | +0.006 |
known | +0.006 |
Neg logit contributions | |
five | -0.006 |
agging | -0.005 |
cheek | -0.005 |
mine | -0.005 |
fourteen | -0.005 |
Token | read |
Token ID | 1100 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
Pig | +0.025 |
Beth | +0.023 |
Tweet | +0.023 |
Ally | +0.023 |
known | +0.023 |
Neg logit contributions | |
five | -0.023 |
cheek | -0.022 |
agging | -0.021 |
mine | -0.020 |
fourteen | -0.019 |
Token | dress |
Token ID | 6576 |
Feature activation | +7.94e-01 |
Pos logit contributions | |
five | +0.008 |
cheek | +0.007 |
agging | +0.007 |
fourteen | +0.007 |
Deal | +0.007 |
Neg logit contributions | |
Pig | -0.004 |
Tweet | -0.004 |
Bunny | -0.003 |
Beth | -0.003 |
Ally | -0.003 |
Token | - |
Token ID | 12 |
Feature activation | +3.88e+00 |
Pos logit contributions | |
Pig | +0.026 |
Tweet | +0.022 |
known | +0.022 |
Bunny | +0.022 |
Beth | +0.022 |
Neg logit contributions | |
five | -0.040 |
agging | -0.039 |
cheek | -0.038 |
mine | -0.035 |
fourteen | -0.035 |
Token | up |
Token ID | 929 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
Pig | +0.146 |
Tweet | +0.128 |
Bunny | +0.128 |
Beth | +0.125 |
Ally | +0.124 |
Neg logit contributions | |
five | -0.176 |
agging | -0.167 |
cheek | -0.163 |
fourteen | -0.149 |
mine | -0.149 |
Token | and |
Token ID | 290 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
Pig | +0.045 |
Tweet | +0.040 |
Bunny | +0.039 |
known | +0.039 |
Beth | +0.039 |
Neg logit contributions | |
five | -0.057 |
agging | -0.057 |
cheek | -0.055 |
mine | -0.051 |
la | -0.049 |
Token | make |
Token ID | 787 |
Feature activation | +7.82e-01 |
Pos logit contributions | |
five | +0.002 |
agging | +0.002 |
cheek | +0.002 |
fourteen | +0.002 |
Deal | +0.002 |
Neg logit contributions | |
Pig | -0.002 |
Tweet | -0.002 |
Beth | -0.002 |
Ally | -0.002 |
Bunny | -0.002 |
Token | - |
Token ID | 12 |
Feature activation | +3.85e+00 |
Pos logit contributions | |
Pig | +0.036 |
wood | +0.034 |
gy | +0.034 |
Bunny | +0.033 |
known | +0.032 |
Neg logit contributions | |
agging | -0.027 |
five | -0.026 |
cheek | -0.025 |
mine | -0.024 |
goodness | -0.024 |
Token | bel |
Token ID | 6667 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
Pig | +0.128 |
Tweet | +0.112 |
Bunny | +0.111 |
Beth | +0.108 |
Ally | +0.107 |
Neg logit contributions | |
five | -0.191 |
agging | -0.182 |
cheek | -0.179 |
fourteen | -0.165 |
mine | -0.164 |
Token | ieve |
Token ID | 12311 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
Pig | +0.074 |
Tweet | +0.068 |
known | +0.067 |
Beth | +0.066 |
Bert | +0.064 |
Neg logit contributions | |
five | -0.077 |
agging | -0.074 |
cheek | -0.073 |
fourteen | -0.068 |
arl | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | +2.35e-01 |
Pos logit contributions | |
Pig | +0.062 |
Tweet | +0.054 |
Bunny | +0.053 |
known | +0.053 |
gy | +0.053 |
Neg logit contributions | |
five | -0.072 |
agging | -0.071 |
cheek | -0.069 |
mine | -0.065 |
la | -0.063 |
Token | One |
Token ID | 1881 |
Feature activation | -6.94e-02 |
Pos logit contributions | |
Pig | +0.006 |
Tweet | +0.005 |
Beth | +0.005 |
Bunny | +0.005 |
Ally | +0.005 |
Neg logit contributions | |
five | -0.014 |
agging | -0.013 |
cheek | -0.013 |
fourteen | -0.012 |
sidelines | -0.012 |
Token | day |
Token ID | 1110 |
Feature activation | +2.83e-01 |
Pos logit contributions | |
five | +0.004 |
agging | +0.004 |
cheek | +0.004 |
fourteen | +0.003 |
mine | +0.003 |
Neg logit contributions | |
Pig | -0.002 |
Tweet | -0.002 |
Beth | -0.002 |
Bunny | -0.002 |
Ally | -0.002 |
Token | Do |
Token ID | 2141 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
five | +0.007 |
cheek | +0.006 |
agging | +0.006 |
Deal | +0.005 |
fourteen | +0.005 |
Neg logit contributions | |
Pig | -0.012 |
known | -0.011 |
Tweet | -0.011 |
gy | -0.011 |
Beth | -0.011 |
Token | you |
Token ID | 345 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
Pig | +0.013 |
Tweet | +0.013 |
Bunny | +0.012 |
Ally | +0.012 |
Beth | +0.012 |
Neg logit contributions | |
five | -0.008 |
agging | -0.007 |
cheek | -0.007 |
arl | -0.006 |
sidelines | -0.006 |
Token | know |
Token ID | 760 |
Feature activation | -9.63e-01 |
Pos logit contributions | |
five | +0.043 |
agging | +0.043 |
cheek | +0.041 |
sidelines | +0.040 |
fourteen | +0.039 |
Neg logit contributions | |
Pig | -0.045 |
known | -0.040 |
Tweet | -0.040 |
Bunny | -0.039 |
gy | -0.039 |
Token | how |
Token ID | 703 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
five | +0.046 |
agging | +0.044 |
cheek | +0.043 |
sidelines | +0.040 |
fourteen | +0.040 |
Neg logit contributions | |
Pig | -0.034 |
Tweet | -0.030 |
Bunny | -0.029 |
Beth | -0.029 |
Ally | -0.029 |
Token | to |
Token ID | 284 |
Feature activation | +2.38e-01 |
Pos logit contributions | |
Pig | +0.005 |
Tweet | +0.004 |
Bunny | +0.004 |
Beth | +0.004 |
Ally | +0.004 |
Neg logit contributions | |
five | -0.007 |
agging | -0.007 |
cheek | -0.006 |
fourteen | -0.006 |
Deal | -0.006 |
Token | stitch |
Token ID | 24695 |
Feature activation | +3.44e+00 |
Pos logit contributions | |
Pig | +0.009 |
Tweet | +0.008 |
Bunny | +0.008 |
gy | +0.008 |
Beth | +0.008 |
Neg logit contributions | |
five | -0.010 |
agging | -0.010 |
cheek | -0.010 |
fourteen | -0.009 |
Deal | -0.009 |
Token | ?" |
Token ID | 1701 |
Feature activation | +6.22e-01 |
Pos logit contributions | |
Pig | +0.146 |
Tweet | +0.133 |
known | +0.132 |
Beth | +0.131 |
gy | +0.130 |
Neg logit contributions | |
five | -0.130 |
agging | -0.128 |
cheek | -0.127 |
mine | -0.117 |
fourteen | -0.115 |
Token |
|
Token ID | 198 |
Feature activation | +6.69e-01 |
Pos logit contributions | |
Pig | +0.023 |
wood | +0.022 |
gy | +0.022 |
known | +0.022 |
darker | +0.021 |
Neg logit contributions | |
five | -0.028 |
cheek | -0.024 |
goodness | -0.023 |
agging | -0.023 |
la | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | +4.06e-01 |
Pos logit contributions | |
Pig | +0.020 |
Tweet | +0.017 |
Beth | +0.016 |
Bunny | +0.016 |
Ally | +0.016 |
Neg logit contributions | |
five | -0.035 |
agging | -0.034 |
cheek | -0.034 |
fourteen | -0.032 |
mine | -0.031 |
Token | The |
Token ID | 464 |
Feature activation | +6.29e-01 |
Pos logit contributions | |
Pig | +0.017 |
Tweet | +0.015 |
Ally | +0.015 |
Bunny | +0.015 |
Beth | +0.015 |
Neg logit contributions | |
agging | -0.017 |
five | -0.017 |
cheek | -0.016 |
fourteen | -0.015 |
mine | -0.015 |
Token | tree |
Token ID | 5509 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
Pig | +0.020 |
Tweet | +0.019 |
Bunny | +0.018 |
house | +0.017 |
Ally | +0.017 |
Neg logit contributions | |
five | -0.032 |
agging | -0.031 |
fourteen | -0.030 |
ige | -0.028 |
leaf | -0.028 |
Token | and |
Token ID | 290 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
five | +0.030 |
agging | +0.029 |
cheek | +0.029 |
sidelines | +0.027 |
Deal | +0.026 |
Neg logit contributions | |
Pig | -0.021 |
Tweet | -0.019 |
Bunny | -0.018 |
gy | -0.018 |
wood | -0.018 |
Token | Mia |
Token ID | 32189 |
Feature activation | -3.61e-02 |
Pos logit contributions | |
five | +0.046 |
cheek | +0.041 |
agging | +0.038 |
mine | +0.038 |
la | +0.038 |
Neg logit contributions | |
Pig | -0.050 |
wood | -0.044 |
known | -0.044 |
gy | -0.043 |
Tweet | -0.042 |
Token | lick |
Token ID | 36728 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
agging | +0.001 |
cheek | +0.001 |
five | +0.001 |
arl | +0.001 |
mine | +0.001 |
Neg logit contributions | |
Pig | -0.002 |
Bunny | -0.001 |
known | -0.001 |
gy | -0.001 |
Tweet | -0.001 |
Token | their |
Token ID | 511 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
five | +0.049 |
agging | +0.045 |
cheek | +0.043 |
fourteen | +0.042 |
mine | +0.041 |
Neg logit contributions | |
Pig | -0.037 |
gy | -0.034 |
known | -0.033 |
wood | -0.033 |
Bert | -0.033 |
Token | ice |
Token ID | 4771 |
Feature activation | +3.15e-01 |
Pos logit contributions | |
Pig | +0.063 |
Bunny | +0.058 |
Tweet | +0.057 |
wood | +0.057 |
gy | +0.056 |
Neg logit contributions | |
agging | -0.060 |
five | -0.058 |
arl | -0.054 |
la | -0.054 |
fourteen | -0.053 |
Token | - |
Token ID | 12 |
Feature activation | +4.10e+00 |
Pos logit contributions | |
Pig | +0.012 |
Tweet | +0.011 |
Beth | +0.011 |
Ally | +0.010 |
Bunny | +0.010 |
Neg logit contributions | |
five | -0.014 |
agging | -0.013 |
cheek | -0.013 |
fourteen | -0.012 |
mine | -0.012 |
Token | cream |
Token ID | 36277 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
Pig | +0.182 |
Tweet | +0.164 |
Beth | +0.163 |
Ally | +0.162 |
Bunny | +0.162 |
Neg logit contributions | |
five | -0.160 |
agging | -0.149 |
cheek | -0.146 |
fourteen | -0.135 |
mine | -0.129 |
Token | . |
Token ID | 13 |
Feature activation | -4.81e-01 |
Pos logit contributions | |
five | +0.051 |
agging | +0.050 |
cheek | +0.048 |
fourteen | +0.045 |
mine | +0.044 |
Neg logit contributions | |
Pig | -0.043 |
Tweet | -0.038 |
Beth | -0.037 |
Bunny | -0.037 |
Ally | -0.037 |
Token | It |
Token ID | 632 |
Feature activation | +9.96e-01 |
Pos logit contributions | |
five | +0.023 |
fourteen | +0.021 |
cheek | +0.020 |
agging | +0.020 |
mine | +0.020 |
Neg logit contributions | |
Pig | -0.018 |
known | -0.016 |
Beth | -0.015 |
gy | -0.015 |
Ally | -0.015 |
Token | is |
Token ID | 318 |
Feature activation | +4.94e-01 |
Pos logit contributions | |
Pig | +0.048 |
Tweet | +0.044 |
Beth | +0.043 |
Ally | +0.043 |
Bunny | +0.043 |
Neg logit contributions | |
five | -0.035 |
agging | -0.033 |
cheek | -0.032 |
fourteen | -0.029 |
mine | -0.029 |
Token | cold |
Token ID | 4692 |
Feature activation | +3.49e-01 |
Pos logit contributions | |
Pig | +0.021 |
Ally | +0.019 |
gy | +0.019 |
Beth | +0.019 |
Bert | +0.019 |
Neg logit contributions | |
mine | -0.019 |
cheek | -0.019 |
agging | -0.019 |
fourteen | -0.018 |
five | -0.018 |
Token | They |
Token ID | 1119 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
five | +0.017 |
Deal | +0.016 |
agging | +0.016 |
cheek | +0.016 |
arl | +0.016 |
Neg logit contributions | |
Pig | -0.012 |
gy | -0.011 |
wood | -0.011 |
known | -0.011 |
Tweet | -0.011 |
Token | like |
Token ID | 588 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
agging | +0.007 |
five | +0.007 |
cheek | +0.006 |
fourteen | +0.006 |
sidelines | +0.006 |
Neg logit contributions | |
Pig | -0.009 |
Tweet | -0.008 |
Bunny | -0.008 |
known | -0.008 |
Beth | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
five | +0.007 |
cheek | +0.007 |
mine | +0.007 |
agging | +0.007 |
liking | +0.007 |
Neg logit contributions | |
Pig | -0.007 |
gy | -0.007 |
Ally | -0.007 |
wood | -0.007 |
Beth | -0.007 |
Token | play |
Token ID | 711 |
Feature activation | +3.52e-01 |
Pos logit contributions | |
five | +0.011 |
agging | +0.011 |
cheek | +0.011 |
fourteen | +0.010 |
Deal | +0.010 |
Neg logit contributions | |
Pig | -0.012 |
Tweet | -0.011 |
Beth | -0.011 |
Bunny | -0.011 |
gy | -0.011 |
Token | dress |
Token ID | 6576 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
Pig | +0.013 |
Tweet | +0.011 |
Beth | +0.011 |
known | +0.011 |
wood | +0.011 |
Neg logit contributions | |
five | -0.016 |
agging | -0.016 |
cheek | -0.015 |
mine | -0.014 |
fourteen | -0.014 |
Token | - |
Token ID | 12 |
Feature activation | +4.24e+00 |
Pos logit contributions | |
Pig | +0.054 |
known | +0.053 |
Tweet | +0.047 |
wood | +0.047 |
Bunny | +0.047 |
Neg logit contributions | |
agging | -0.052 |
five | -0.051 |
cheek | -0.049 |
mine | -0.045 |
goodness | -0.043 |
Token | up |
Token ID | 929 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
Pig | +0.178 |
Bunny | +0.158 |
Tweet | +0.155 |
Ally | +0.154 |
Bert | +0.154 |
Neg logit contributions | |
five | -0.185 |
agging | -0.162 |
cheek | -0.159 |
fourteen | -0.147 |
leaf | -0.146 |
Token | . |
Token ID | 13 |
Feature activation | -3.46e-01 |
Pos logit contributions | |
Pig | +0.064 |
Tweet | +0.057 |
Beth | +0.056 |
known | +0.056 |
Ally | +0.056 |
Neg logit contributions | |
five | -0.073 |
agging | -0.070 |
cheek | -0.069 |
mine | -0.063 |
fourteen | -0.061 |
Token | They |
Token ID | 1119 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
five | +0.016 |
Deal | +0.015 |
arl | +0.015 |
agging | +0.015 |
cheek | +0.015 |
Neg logit contributions | |
Pig | -0.011 |
gy | -0.010 |
wood | -0.010 |
known | -0.010 |
Tweet | -0.010 |
Token | like |
Token ID | 588 |
Feature activation | -2.39e-01 |
Pos logit contributions | |
agging | +0.010 |
five | +0.010 |
cheek | +0.009 |
fourteen | +0.008 |
sidelines | +0.008 |
Neg logit contributions | |
Pig | -0.012 |
Tweet | -0.011 |
known | -0.011 |
Bunny | -0.011 |
Beth | -0.010 |
Token | to |
Token ID | 284 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
five | +0.010 |
cheek | +0.009 |
agging | +0.009 |
mine | +0.009 |
liking | +0.009 |
Neg logit contributions | |
Pig | -0.009 |
gy | -0.009 |
wood | -0.009 |
Ally | -0.009 |
Tweet | -0.008 |
Token | big |
Token ID | 1263 |
Feature activation | -7.75e-02 |
Pos logit contributions | |
five | +0.008 |
agging | +0.008 |
cheek | +0.007 |
mine | +0.007 |
sidelines | +0.007 |
Neg logit contributions | |
Pig | -0.007 |
Beth | -0.006 |
Tweet | -0.006 |
gy | -0.006 |
Bunny | -0.006 |
Token | tree |
Token ID | 5509 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
five | +0.003 |
mine | +0.003 |
agging | +0.003 |
cheek | +0.003 |
sidelines | +0.003 |
Neg logit contributions | |
Pig | -0.003 |
known | -0.003 |
Beth | -0.003 |
gy | -0.003 |
wood | -0.003 |
Token | ." |
Token ID | 526 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
five | +0.010 |
agging | +0.009 |
cheek | +0.009 |
mine | +0.008 |
fourteen | +0.008 |
Neg logit contributions | |
Pig | -0.016 |
Beth | -0.015 |
Tweet | -0.015 |
Bunny | -0.015 |
known | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | +3.20e-01 |
Pos logit contributions | |
five | +0.009 |
cheek | +0.008 |
agging | +0.008 |
fourteen | +0.007 |
Deal | +0.007 |
Neg logit contributions | |
Pig | -0.006 |
known | -0.006 |
Beth | -0.005 |
Ally | -0.005 |
Tweet | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
Pig | +0.010 |
Tweet | +0.009 |
Beth | +0.009 |
wood | +0.009 |
gy | +0.008 |
Neg logit contributions | |
five | -0.016 |
agging | -0.016 |
cheek | -0.015 |
fourteen | -0.014 |
arl | -0.014 |
Token | The |
Token ID | 464 |
Feature activation | +6.54e-01 |
Pos logit contributions | |
Pig | +0.005 |
Tweet | +0.004 |
Bunny | +0.004 |
gy | +0.004 |
known | +0.004 |
Neg logit contributions | |
five | -0.005 |
cheek | -0.005 |
agging | -0.005 |
Deal | -0.005 |
sidelines | -0.005 |
Token | race |
Token ID | 3234 |
Feature activation | -6.62e-01 |
Pos logit contributions | |
Pig | +0.027 |
Tweet | +0.024 |
Bunny | +0.023 |
Ally | +0.023 |
known | +0.023 |
Neg logit contributions | |
five | -0.028 |
agging | -0.027 |
cheek | -0.026 |
fourteen | -0.025 |
Deal | -0.024 |
Token | began |
Token ID | 2540 |
Feature activation | +3.82e-01 |
Pos logit contributions | |
five | +0.029 |
affe | +0.025 |
mine | +0.024 |
agging | +0.024 |
cheek | +0.023 |
Neg logit contributions | |
Pig | -0.025 |
Chick | -0.024 |
Ang | -0.022 |
darker | -0.022 |
known | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | -2.88e-01 |
Pos logit contributions | |
Pig | +0.019 |
Chick | +0.017 |
Bunny | +0.017 |
Ally | +0.017 |
Beth | +0.016 |
Neg logit contributions | |
five | -0.012 |
agging | -0.010 |
liking | -0.010 |
arrest | -0.010 |
fourteen | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
five | +0.011 |
cheek | +0.010 |
agging | +0.010 |
fourteen | +0.010 |
mine | +0.010 |
Neg logit contributions | |
Pig | -0.012 |
Beth | -0.011 |
known | -0.011 |
wood | -0.011 |
gy | -0.011 |
Token | Remy |
Token ID | 32912 |
Feature activation | -3.83e-01 |
Pos logit contributions | |
fourteen | +0.006 |
five | +0.005 |
mine | +0.005 |
cheek | +0.005 |
sidelines | +0.005 |
Neg logit contributions | |
Beth | -0.007 |
bul | -0.007 |
Pig | -0.006 |
known | -0.006 |
house | -0.006 |
Token | said |
Token ID | 531 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
agging | +0.023 |
five | +0.022 |
cheek | +0.022 |
sidelines | +0.021 |
Deal | +0.020 |
Neg logit contributions | |
Pig | -0.023 |
Tweet | -0.020 |
known | -0.020 |
Beth | -0.020 |
gy | -0.020 |
Token | that |
Token ID | 326 |
Feature activation | +6.62e-01 |
Pos logit contributions | |
five | +0.009 |
cheek | +0.009 |
agging | +0.009 |
mine | +0.009 |
la | +0.008 |
Neg logit contributions | |
Pig | -0.009 |
known | -0.008 |
wood | -0.008 |
Tweet | -0.008 |
gy | -0.008 |
Token | ?" |
Token ID | 1701 |
Feature activation | +6.47e-01 |
Pos logit contributions | |
Pig | +0.030 |
Tweet | +0.027 |
Beth | +0.027 |
Bunny | +0.027 |
gy | +0.026 |
Neg logit contributions | |
five | -0.024 |
cheek | -0.023 |
agging | -0.023 |
mine | -0.021 |
fourteen | -0.020 |
Token | he |
Token ID | 339 |
Feature activation | +3.75e-01 |
Pos logit contributions | |
gy | +0.027 |
Pig | +0.027 |
wood | +0.026 |
dam | +0.026 |
Tweet | +0.025 |
Neg logit contributions | |
five | -0.025 |
cheek | -0.021 |
goodness | -0.020 |
Deal | -0.019 |
Tom | -0.019 |
Token | asked |
Token ID | 1965 |
Feature activation | +9.66e-01 |
Pos logit contributions | |
wood | +0.015 |
Pig | +0.014 |
dam | +0.014 |
Tweet | +0.014 |
shelters | +0.014 |
Neg logit contributions | |
cheek | -0.014 |
agging | -0.013 |
la | -0.013 |
mine | -0.012 |
wig | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +9.84e-01 |
Pos logit contributions | |
Pig | +0.042 |
dam | +0.042 |
wood | +0.039 |
known | +0.038 |
Tweet | +0.038 |
Neg logit contributions | |
five | -0.034 |
cheek | -0.034 |
agging | -0.031 |
goodness | -0.027 |
mine | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | +7.26e-01 |
Pos logit contributions | |
Pig | +0.036 |
gy | +0.034 |
wood | +0.034 |
Tweet | +0.033 |
Beth | +0.033 |
Neg logit contributions | |
five | -0.042 |
cheek | -0.039 |
agging | -0.037 |
Deal | -0.036 |
fourteen | -0.036 |
Token |
|
Token ID | 198 |
Feature activation | +5.66e-01 |
Pos logit contributions | |
Pig | +0.023 |
Tweet | +0.021 |
wood | +0.020 |
gy | +0.020 |
Beth | +0.020 |
Neg logit contributions | |
five | -0.036 |
agging | -0.035 |
cheek | -0.034 |
fourteen | -0.032 |
Deal | -0.031 |
Token | There |
Token ID | 1858 |
Feature activation | +7.11e-01 |
Pos logit contributions | |
Pig | +0.027 |
Tweet | +0.025 |
known | +0.024 |
Beth | +0.024 |
Bunny | +0.024 |
Neg logit contributions | |
agging | -0.019 |
five | -0.019 |
cheek | -0.018 |
mine | -0.016 |
sidelines | -0.016 |
Token | was |
Token ID | 373 |
Feature activation | +7.19e-01 |
Pos logit contributions | |
wood | +0.026 |
Pig | +0.025 |
dam | +0.025 |
gy | +0.025 |
Tweet | +0.023 |
Neg logit contributions | |
agging | -0.028 |
five | -0.027 |
cheek | -0.026 |
goodness | -0.026 |
Tom | -0.025 |
Token | a |
Token ID | 257 |
Feature activation | +3.46e-02 |
Pos logit contributions | |
Pig | +0.027 |
wood | +0.025 |
gy | +0.024 |
Ally | +0.023 |
Beth | +0.022 |
Neg logit contributions | |
five | -0.030 |
fourteen | -0.030 |
mine | -0.028 |
agging | -0.027 |
arrest | -0.027 |
Token | They |
Token ID | 1119 |
Feature activation | -5.96e-02 |
Pos logit contributions | |
Pig | +0.000 |
known | +0.000 |
wood | +0.000 |
darker | +0.000 |
gy | +0.000 |
Neg logit contributions | |
five | -0.000 |
cheek | -0.000 |
Deal | -0.000 |
fourteen | -0.000 |
la | -0.000 |
Token | hugged |
Token ID | 41130 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
five | +0.002 |
ige | +0.002 |
cheek | +0.002 |
mine | +0.002 |
agging | +0.002 |
Neg logit contributions | |
Pig | -0.002 |
wood | -0.002 |
Tweet | -0.002 |
Bert | -0.002 |
dam | -0.002 |
Token | each |
Token ID | 1123 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
five | +0.007 |
cheek | +0.007 |
agging | +0.007 |
mine | +0.006 |
goodness | +0.006 |
Neg logit contributions | |
Pig | -0.007 |
known | -0.007 |
wood | -0.007 |
Tweet | -0.006 |
gy | -0.006 |
Token | other |
Token ID | 584 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
agging | +0.050 |
ider | +0.048 |
cheek | +0.048 |
fourteen | +0.048 |
arl | +0.048 |
Neg logit contributions | |
Pig | -0.034 |
Bunny | -0.031 |
Ally | -0.030 |
Beth | -0.028 |
called | -0.028 |
Token | and |
Token ID | 290 |
Feature activation | +4.89e-01 |
Pos logit contributions | |
five | +0.044 |
cheek | +0.042 |
agging | +0.041 |
mine | +0.037 |
fourteen | +0.036 |
Neg logit contributions | |
Pig | -0.045 |
Tweet | -0.041 |
known | -0.041 |
Ally | -0.040 |
Beth | -0.040 |
Token | played |
Token ID | 2826 |
Feature activation | +7.84e-01 |
Pos logit contributions | |
wood | +0.021 |
Pig | +0.021 |
gy | +0.020 |
dam | +0.020 |
Bert | +0.019 |
Neg logit contributions | |
five | -0.017 |
cheek | -0.016 |
mine | -0.014 |
ige | -0.013 |
sidelines | -0.013 |
Token | together |
Token ID | 1978 |
Feature activation | +1.53e-02 |
Pos logit contributions | |
Pig | +0.024 |
wood | +0.021 |
Tweet | +0.020 |
known | +0.020 |
Bunny | +0.020 |
Neg logit contributions | |
five | -0.041 |
agging | -0.040 |
cheek | -0.039 |
mine | -0.036 |
sidelines | -0.036 |
Token | happily |
Token ID | 18177 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
Pig | +0.001 |
Tweet | +0.000 |
Beth | +0.000 |
Bunny | +0.000 |
wood | +0.000 |
Neg logit contributions | |
five | -0.001 |
agging | -0.001 |
cheek | -0.001 |
mine | -0.001 |
fourteen | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
five | +0.050 |
agging | +0.048 |
cheek | +0.047 |
fourteen | +0.043 |
mine | +0.043 |
Neg logit contributions | |
Pig | -0.039 |
Tweet | -0.034 |
Beth | -0.034 |
Bunny | -0.033 |
Ally | -0.033 |
Token | From |
Token ID | 3574 |
Feature activation | -5.44e-01 |
Pos logit contributions | |
Pig | +0.008 |
wood | +0.007 |
known | +0.007 |
gy | +0.007 |
darker | +0.007 |
Neg logit contributions | |
five | -0.010 |
cheek | -0.009 |
agging | -0.009 |
Deal | -0.009 |
fourteen | -0.009 |
Token | that |
Token ID | 326 |
Feature activation | +9.48e-01 |
Pos logit contributions | |
five | +0.020 |
agging | +0.020 |
cheek | +0.019 |
mine | +0.018 |
fourteen | +0.018 |
Neg logit contributions | |
Pig | -0.024 |
wood | -0.022 |
known | -0.022 |
Tweet | -0.022 |
gy | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | +5.75e-03 |
Pos logit contributions | |
Pig | +0.061 |
Tweet | +0.053 |
Beth | +0.052 |
Bunny | +0.052 |
Ally | +0.052 |
Neg logit contributions | |
five | -0.083 |
agging | -0.080 |
cheek | -0.078 |
fourteen | -0.074 |
mine | -0.073 |
Token | They |
Token ID | 1119 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
Pig | +0.000 |
wood | +0.000 |
known | +0.000 |
gy | +0.000 |
Ally | +0.000 |
Neg logit contributions | |
five | -0.000 |
Deal | -0.000 |
cheek | -0.000 |
agging | -0.000 |
fourteen | -0.000 |
Token | have |
Token ID | 423 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
five | +0.004 |
agging | +0.004 |
cheek | +0.004 |
fourteen | +0.003 |
mine | +0.003 |
Neg logit contributions | |
Pig | -0.006 |
Tweet | -0.006 |
Beth | -0.005 |
Bunny | -0.005 |
Ally | -0.005 |
Token | fun |
Token ID | 1257 |
Feature activation | +8.95e-01 |
Pos logit contributions | |
five | +0.008 |
agging | +0.008 |
mine | +0.008 |
fourteen | +0.007 |
cheek | +0.007 |
Neg logit contributions | |
Pig | -0.007 |
Ally | -0.006 |
Bert | -0.006 |
Beth | -0.006 |
Bunny | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | -2.57e-01 |
Pos logit contributions | |
Pig | +0.024 |
Tweet | +0.021 |
Beth | +0.020 |
gy | +0.020 |
Ch | +0.020 |
Neg logit contributions | |
five | -0.048 |
cheek | -0.046 |
agging | -0.046 |
Deal | -0.045 |
ige | -0.044 |
Token | are |
Token ID | 389 |
Feature activation | +8.18e-01 |
Pos logit contributions | |
five | +0.010 |
cheek | +0.009 |
agging | +0.009 |
mine | +0.008 |
sidelines | +0.008 |
Neg logit contributions | |
Pig | -0.012 |
Tweet | -0.011 |
Bunny | -0.011 |
Ally | -0.011 |
Beth | -0.011 |
Token | happy |
Token ID | 3772 |
Feature activation | -6.59e-01 |
Pos logit contributions | |
Pig | +0.037 |
wood | +0.035 |
Ally | +0.034 |
Bert | +0.033 |
Beth | +0.033 |
Neg logit contributions | |
agging | -0.031 |
mine | -0.030 |
cheek | -0.028 |
five | -0.028 |
fourteen | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -2.92e-01 |
Pos logit contributions | |
five | +0.031 |
agging | +0.030 |
cheek | +0.029 |
fourteen | +0.027 |
mine | +0.027 |
Neg logit contributions | |
Pig | -0.024 |
Tweet | -0.020 |
Ally | -0.020 |
Beth | -0.020 |
known | -0.020 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -3.82e-01 |
Pos logit contributions | |
five | +0.014 |
agging | +0.013 |
cheek | +0.013 |
Deal | +0.013 |
fourteen | +0.013 |
Neg logit contributions | |
Pig | -0.010 |
wood | -0.008 |
gy | -0.008 |
Tweet | -0.008 |
known | -0.008 |
Token | L |
Token ID | 43 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
five | +0.017 |
agging | +0.016 |
cheek | +0.015 |
arl | +0.015 |
fourteen | +0.015 |
Neg logit contributions | |
Pig | -0.014 |
Beth | -0.013 |
Ally | -0.013 |
Tweet | -0.013 |
gy | -0.012 |
Token | ily |
Token ID | 813 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
Pig | +0.059 |
Bunny | +0.051 |
Beth | +0.051 |
called | +0.051 |
Tweet | +0.050 |
Neg logit contributions | |
five | -0.061 |
cheek | -0.060 |
sidelines | -0.058 |
agging | -0.057 |
Deal | -0.054 |
Token | , |
Token ID | 11 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
Pig | +0.027 |
known | +0.024 |
Ally | +0.024 |
Beth | +0.024 |
Bert | +0.023 |
Neg logit contributions | |
five | -0.030 |
cheek | -0.030 |
agging | -0.029 |
mine | -0.028 |
arl | -0.026 |
Token | we |
Token ID | 356 |
Feature activation | -4.38e-01 |
Pos logit contributions | |
Pig | +0.006 |
gy | +0.006 |
wood | +0.006 |
known | +0.005 |
increasingly | +0.005 |
Neg logit contributions | |
cheek | -0.002 |
mine | -0.002 |
five | -0.002 |
agging | -0.002 |
goodness | -0.002 |
Token | will |
Token ID | 481 |
Feature activation | +3.72e-01 |
Pos logit contributions | |
five | +0.018 |
agging | +0.017 |
cheek | +0.016 |
mine | +0.015 |
fourteen | +0.015 |
Neg logit contributions | |
Pig | -0.018 |
Tweet | -0.017 |
Beth | -0.016 |
Ally | -0.016 |
Bunny | -0.016 |
Token | have |
Token ID | 423 |
Feature activation | -5.04e-01 |
Pos logit contributions | |
Pig | +0.016 |
Tweet | +0.014 |
Beth | +0.014 |
Ally | +0.014 |
Bunny | +0.014 |
Neg logit contributions | |
five | -0.015 |
agging | -0.014 |
cheek | -0.014 |
fourteen | -0.013 |
mine | -0.013 |
Token | to |
Token ID | 284 |
Feature activation | -3.65e-01 |
Pos logit contributions | |
five | +0.022 |
agging | +0.021 |
cheek | +0.021 |
mine | +0.021 |
fourteen | +0.020 |
Neg logit contributions | |
Pig | -0.019 |
gy | -0.017 |
Ally | -0.016 |
wood | -0.016 |
Bert | -0.016 |
Token | punish |
Token ID | 6878 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
five | +0.013 |
agging | +0.013 |
cheek | +0.013 |
mine | +0.011 |
fourteen | +0.011 |
Neg logit contributions | |
Pig | -0.016 |
Tweet | -0.015 |
Beth | -0.015 |
Ally | -0.015 |
Bunny | -0.015 |
Token | you |
Token ID | 345 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
Pig | +0.088 |
known | +0.081 |
wood | +0.081 |
Tweet | +0.081 |
gy | +0.081 |
Neg logit contributions | |
five | -0.029 |
cheek | -0.028 |
agging | -0.027 |
mine | -0.023 |
la | -0.021 |
Token | ." |
Token ID | 526 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
Pig | +0.005 |
Tweet | +0.005 |
Bunny | +0.004 |
gy | +0.004 |
Beth | +0.004 |
Neg logit contributions | |
five | -0.004 |
agging | -0.004 |
cheek | -0.004 |
sidelines | -0.003 |
fourteen | -0.003 |
Token | |
Token ID | 220 |
Feature activation | +6.45e-01 |
Pos logit contributions | |
known | +0.003 |
Pig | +0.003 |
darker | +0.003 |
increasingly | +0.003 |
rou | +0.003 |
Neg logit contributions | |
cheek | -0.005 |
five | -0.005 |
goodness | -0.004 |
agging | -0.004 |
Deal | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | +3.13e-01 |
Pos logit contributions | |
Pig | +0.023 |
wood | +0.021 |
known | +0.020 |
gy | +0.020 |
Tweet | +0.020 |
Neg logit contributions | |
five | -0.028 |
agging | -0.027 |
cheek | -0.027 |
Deal | -0.026 |
arl | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | +7.26e-02 |
Pos logit contributions | |
Pig | +0.011 |
wood | +0.009 |
Tweet | +0.009 |
gy | +0.009 |
Bunny | +0.009 |
Neg logit contributions | |
five | -0.015 |
agging | -0.014 |
cheek | -0.014 |
arl | -0.013 |
Deal | -0.013 |
Token | her |
Token ID | 607 |
Feature activation | -7.05e-01 |
Pos logit contributions | |
five | +0.030 |
agging | +0.029 |
cheek | +0.029 |
mine | +0.027 |
fourteen | +0.027 |
Neg logit contributions | |
Pig | -0.021 |
Tweet | -0.018 |
wood | -0.017 |
gy | -0.017 |
Bunny | -0.017 |
Token | fingers |
Token ID | 9353 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
five | +0.029 |
agging | +0.027 |
fourteen | +0.026 |
cheek | +0.026 |
arl | +0.025 |
Neg logit contributions | |
Pig | -0.029 |
Tweet | -0.026 |
Beth | -0.026 |
Bunny | -0.026 |
gy | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | +3.28e-01 |
Pos logit contributions | |
five | +0.078 |
agging | +0.076 |
cheek | +0.074 |
fourteen | +0.069 |
mine | +0.068 |
Neg logit contributions | |
Pig | -0.057 |
Tweet | -0.050 |
Beth | -0.048 |
Bunny | -0.048 |
Ally | -0.048 |
Token | before |
Token ID | 878 |
Feature activation | +7.30e-01 |
Pos logit contributions | |
Pig | +0.013 |
wood | +0.012 |
gy | +0.011 |
Bunny | +0.011 |
Tweet | +0.011 |
Neg logit contributions | |
five | -0.014 |
cheek | -0.013 |
mine | -0.013 |
la | -0.012 |
goodness | -0.012 |
Token | long |
Token ID | 890 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
wood | +0.032 |
gy | +0.031 |
Pig | +0.030 |
known | +0.028 |
house | +0.028 |
Neg logit contributions | |
cheek | -0.025 |
agging | -0.024 |
sidelines | -0.024 |
five | -0.023 |
liking | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
Pig | +0.043 |
wood | +0.040 |
known | +0.039 |
gy | +0.038 |
Tweet | +0.037 |
Neg logit contributions | |
five | -0.055 |
agging | -0.053 |
cheek | -0.052 |
fourteen | -0.050 |
mine | -0.049 |
Token | the |
Token ID | 262 |
Feature activation | -1.87e-01 |
Pos logit contributions | |
gy | +0.008 |
wood | +0.007 |
Pig | +0.007 |
dam | +0.006 |
bank | +0.006 |
Neg logit contributions | |
five | -0.006 |
cheek | -0.005 |
goodness | -0.005 |
mine | -0.005 |
la | -0.005 |
Token | magic |
Token ID | 5536 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
five | +0.009 |
agging | +0.008 |
fourteen | +0.008 |
cheek | +0.008 |
Deal | +0.008 |
Neg logit contributions | |
Pig | -0.007 |
Tweet | -0.006 |
Beth | -0.006 |
Bunny | -0.006 |
Bert | -0.006 |
Token | had |
Token ID | 550 |
Feature activation | -2.14e-01 |
Pos logit contributions | |
Pig | +0.010 |
known | +0.009 |
Tweet | +0.009 |
gy | +0.009 |
wood | +0.009 |
Neg logit contributions | |
five | -0.010 |
cheek | -0.010 |
agging | -0.009 |
mine | -0.008 |
fourteen | -0.008 |
Token | made |
Token ID | 925 |
Feature activation | +6.98e-01 |
Pos logit contributions | |
five | +0.009 |
agging | +0.009 |
cheek | +0.009 |
fourteen | +0.009 |
mine | +0.009 |
Neg logit contributions | |
Pig | -0.008 |
gy | -0.007 |
wood | -0.007 |
Bunny | -0.007 |
Tweet | -0.007 |
Token | something |
Token ID | 1223 |
Feature activation | -2.38e-01 |
Pos logit contributions | |
gy | +0.033 |
Pig | +0.032 |
wood | +0.031 |
dam | +0.029 |
Kang | +0.028 |
Neg logit contributions | |
five | -0.023 |
goodness | -0.020 |
cheek | -0.020 |
mine | -0.019 |
icing | -0.018 |
Token | brave |
Token ID | 14802 |
Feature activation | -7.40e-01 |
Pos logit contributions | |
Pig | +0.022 |
wood | +0.021 |
Tweet | +0.020 |
Beth | +0.019 |
Bert | +0.019 |
Neg logit contributions | |
five | -0.022 |
cheek | -0.019 |
agging | -0.018 |
la | -0.018 |
mine | -0.018 |
Token | ly |
Token ID | 306 |
Feature activation | -9.77e-01 |
Pos logit contributions | |
five | +0.015 |
agging | +0.014 |
cheek | +0.013 |
fourteen | +0.011 |
mine | +0.010 |
Neg logit contributions | |
Pig | -0.046 |
Tweet | -0.043 |
Beth | -0.042 |
Bunny | -0.042 |
Ally | -0.042 |
Token | reached |
Token ID | 4251 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
five | +0.041 |
ige | +0.036 |
cheek | +0.036 |
agging | +0.036 |
fourteen | +0.035 |
Neg logit contributions | |
Pig | -0.037 |
Tweet | -0.036 |
wood | -0.035 |
Bert | -0.034 |
Beth | -0.034 |
Token | out |
Token ID | 503 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
five | +0.066 |
cheek | +0.060 |
agging | +0.059 |
mine | +0.058 |
fourteen | +0.056 |
Neg logit contributions | |
Pig | -0.051 |
known | -0.046 |
gy | -0.044 |
Tweet | -0.044 |
wood | -0.043 |
Token | and |
Token ID | 290 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
five | +0.001 |
agging | +0.001 |
cheek | +0.001 |
fourteen | +0.001 |
liking | +0.001 |
Neg logit contributions | |
Pig | -0.001 |
wood | -0.001 |
known | -0.001 |
gy | -0.001 |
Tweet | -0.001 |
Token | touched |
Token ID | 12615 |
Feature activation | +3.74e-01 |
Pos logit contributions | |
five | +0.006 |
cheek | +0.005 |
leaf | +0.005 |
mine | +0.005 |
fourteen | +0.004 |
Neg logit contributions | |
Pig | -0.006 |
wood | -0.005 |
known | -0.005 |
gy | -0.005 |
Tweet | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | +6.09e-01 |
Pos logit contributions | |
Pig | +0.015 |
gy | +0.015 |
wood | +0.015 |
dam | +0.014 |
known | +0.014 |
Neg logit contributions | |
five | -0.014 |
goodness | -0.012 |
mine | -0.012 |
fourteen | -0.012 |
cheek | -0.012 |
Token | ancient |
Token ID | 6156 |
Feature activation | +4.72e-01 |
Pos logit contributions | |
Tweet | +0.022 |
Pig | +0.022 |
H | +0.021 |
Bunny | +0.021 |
gy | +0.020 |
Neg logit contributions | |
agging | -0.025 |
five | -0.025 |
fourteen | -0.024 |
arl | -0.023 |
Deal | -0.022 |
Token | skeleton |
Token ID | 18328 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
Pig | +0.019 |
Tweet | +0.019 |
gy | +0.018 |
Bunny | +0.018 |
Bert | +0.017 |
Neg logit contributions | |
agging | -0.019 |
five | -0.018 |
Deal | -0.018 |
fourteen | -0.017 |
arl | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
five | +0.001 |
cheek | +0.001 |
agging | +0.001 |
mine | +0.001 |
fourteen | +0.001 |
Neg logit contributions | |
Pig | -0.001 |
known | -0.001 |
Tweet | -0.001 |
Beth | -0.001 |
Ally | -0.001 |
Token | She |
Token ID | 1375 |
Feature activation | +5.42e-01 |
Pos logit contributions | |
Pig | +0.006 |
Tweet | +0.005 |
Beth | +0.005 |
Bunny | +0.005 |
Ally | +0.005 |
Neg logit contributions | |
five | -0.008 |
agging | -0.008 |
cheek | -0.008 |
fourteen | -0.007 |
mine | -0.007 |
Token | an |
Token ID | 281 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
five | +0.040 |
agging | +0.039 |
cheek | +0.037 |
mine | +0.037 |
fourteen | +0.037 |
Neg logit contributions | |
Pig | -0.018 |
wood | -0.015 |
Ally | -0.015 |
Beth | -0.015 |
gy | -0.015 |
Token | idea |
Token ID | 2126 |
Feature activation | -6.23e-01 |
Pos logit contributions | |
five | +0.061 |
agging | +0.061 |
cheek | +0.059 |
mine | +0.056 |
fourteen | +0.055 |
Neg logit contributions | |
Pig | -0.022 |
Tweet | -0.018 |
Beth | -0.018 |
Bunny | -0.017 |
gy | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | -2.91e-01 |
Pos logit contributions | |
five | +0.033 |
agging | +0.032 |
cheek | +0.031 |
fourteen | +0.030 |
Deal | +0.030 |
Neg logit contributions | |
Pig | -0.019 |
Tweet | -0.016 |
Beth | -0.016 |
Bunny | -0.016 |
Bert | -0.015 |
Token | He |
Token ID | 679 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
five | +0.011 |
cheek | +0.011 |
arl | +0.010 |
agging | +0.010 |
Deal | +0.010 |
Neg logit contributions | |
known | -0.012 |
Pig | -0.012 |
wood | -0.011 |
gy | -0.011 |
Ally | -0.010 |
Token | ran |
Token ID | 4966 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
Pig | +0.008 |
wood | +0.007 |
Beth | +0.007 |
Tweet | +0.007 |
known | +0.007 |
Neg logit contributions | |
five | -0.009 |
cheek | -0.009 |
agging | -0.009 |
mine | -0.008 |
fourteen | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | -4.61e-01 |
Pos logit contributions | |
five | +0.047 |
agging | +0.042 |
mine | +0.041 |
cheek | +0.040 |
sidelines | +0.039 |
Neg logit contributions | |
Pig | -0.044 |
known | -0.038 |
wood | -0.036 |
Bunny | -0.036 |
Ally | -0.036 |
Token | get |
Token ID | 651 |
Feature activation | +4.15e-01 |
Pos logit contributions | |
five | +0.023 |
cheek | +0.022 |
agging | +0.022 |
mine | +0.020 |
sidelines | +0.020 |
Neg logit contributions | |
Pig | -0.015 |
wood | -0.013 |
Tweet | -0.013 |
known | -0.012 |
gy | -0.012 |
Token | his |
Token ID | 465 |
Feature activation | +8.23e-02 |
Pos logit contributions | |
Pig | +0.016 |
gy | +0.016 |
wood | +0.015 |
dam | +0.015 |
house | +0.015 |
Neg logit contributions | |
five | -0.016 |
mine | -0.015 |
cheek | -0.014 |
leaf | -0.014 |
Deal | -0.013 |
Token | dad |
Token ID | 9955 |
Feature activation | +6.10e-01 |
Pos logit contributions | |
Pig | +0.004 |
Tweet | +0.003 |
Beth | +0.003 |
Bunny | +0.003 |
Ally | +0.003 |
Neg logit contributions | |
five | -0.003 |
agging | -0.003 |
cheek | -0.003 |
fourteen | -0.003 |
mine | -0.003 |
Token | 's |
Token ID | 338 |
Feature activation | +2.59e-01 |
Pos logit contributions | |
Pig | +0.024 |
Tweet | +0.021 |
known | +0.021 |
Ally | +0.021 |
Beth | +0.021 |
Neg logit contributions | |
five | -0.026 |
agging | -0.025 |
cheek | -0.025 |
mine | -0.023 |
fourteen | -0.022 |
Token | ladder |
Token ID | 18002 |
Feature activation | -8.01e-01 |
Pos logit contributions | |
Pig | +0.012 |
Tweet | +0.011 |
Beth | +0.011 |
Bunny | +0.011 |
gy | +0.011 |
Neg logit contributions | |
five | -0.009 |
agging | -0.009 |
cheek | -0.008 |
fourteen | -0.008 |
Deal | -0.007 |
Token | Jack |
Token ID | 3619 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
Pig | +0.015 |
known | +0.013 |
gy | +0.013 |
wood | +0.013 |
Tweet | +0.013 |
Neg logit contributions | |
five | -0.021 |
agging | -0.019 |
cheek | -0.019 |
fourteen | -0.019 |
Deal | -0.018 |
Token | found |
Token ID | 1043 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
five | +0.084 |
agging | +0.079 |
cheek | +0.079 |
fourteen | +0.074 |
mine | +0.073 |
Neg logit contributions | |
Pig | -0.061 |
Tweet | -0.054 |
wood | -0.052 |
Beth | -0.052 |
Ally | -0.052 |
Token | an |
Token ID | 281 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
five | +0.047 |
fourteen | +0.041 |
mine | +0.041 |
agging | +0.041 |
goodness | +0.040 |
Neg logit contributions | |
Pig | -0.036 |
gy | -0.034 |
wood | -0.031 |
Tweet | -0.031 |
Ally | -0.030 |
Token | old |
Token ID | 1468 |
Feature activation | +5.69e-01 |
Pos logit contributions | |
five | +0.062 |
agging | +0.060 |
cheek | +0.058 |
fourteen | +0.054 |
mine | +0.053 |
Neg logit contributions | |
Pig | -0.049 |
Tweet | -0.043 |
Beth | -0.042 |
Bunny | -0.042 |
Ally | -0.042 |
Token | anchor |
Token ID | 18021 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
Pig | +0.026 |
Tweet | +0.024 |
Beth | +0.024 |
Bunny | +0.024 |
Ally | +0.023 |
Neg logit contributions | |
five | -0.020 |
agging | -0.019 |
cheek | -0.019 |
mine | -0.017 |
fourteen | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
five | +0.071 |
agging | +0.069 |
cheek | +0.068 |
mine | +0.063 |
fourteen | +0.062 |
Neg logit contributions | |
Pig | -0.069 |
Tweet | -0.062 |
Beth | -0.061 |
known | -0.061 |
Ally | -0.061 |
Token | said |
Token ID | 531 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
Pig | +0.007 |
gy | +0.006 |
wood | +0.006 |
Tweet | +0.006 |
known | +0.006 |
Neg logit contributions | |
five | -0.008 |
mine | -0.007 |
cheek | -0.007 |
fourteen | -0.007 |
agging | -0.007 |
Token | : |
Token ID | 25 |
Feature activation | +6.95e-01 |
Pos logit contributions | |
Pig | +0.006 |
gy | +0.005 |
wood | +0.005 |
Tweet | +0.005 |
Beth | +0.005 |
Neg logit contributions | |
five | -0.008 |
agging | -0.008 |
cheek | -0.008 |
mine | -0.007 |
fourteen | -0.007 |
Token | " |
Token ID | 366 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
Pig | +0.026 |
Tweet | +0.023 |
Bunny | +0.022 |
Beth | +0.022 |
Ally | +0.022 |
Neg logit contributions | |
five | -0.031 |
agging | -0.031 |
cheek | -0.029 |
fourteen | -0.027 |
mine | -0.027 |
Token | This |
Token ID | 1212 |
Feature activation | -9.69e-02 |
Pos logit contributions | |
Pig | +0.056 |
Tweet | +0.051 |
Beth | +0.050 |
Bunny | +0.050 |
Ally | +0.050 |
Neg logit contributions | |
five | -0.039 |
agging | -0.038 |
cheek | -0.037 |
fourteen | -0.033 |
mine | -0.032 |
Token | will |
Token ID | 481 |
Feature activation | +6.31e-01 |
Pos logit contributions | |
five | +0.004 |
agging | +0.004 |
cheek | +0.004 |
fourteen | +0.003 |
mine | +0.003 |
Neg logit contributions | |
Pig | -0.004 |
Tweet | -0.004 |
Beth | -0.004 |
Bunny | -0.004 |
Ally | -0.004 |
Token | castle |
Token ID | 16669 |
Feature activation | -6.20e-01 |
Pos logit contributions | |
Pig | +0.019 |
Tweet | +0.018 |
gy | +0.017 |
Bunny | +0.017 |
Beth | +0.017 |
Neg logit contributions | |
five | -0.020 |
agging | -0.019 |
cheek | -0.018 |
fourteen | -0.018 |
Deal | -0.017 |
Token | was |
Token ID | 373 |
Feature activation | +4.44e-01 |
Pos logit contributions | |
cheek | +0.028 |
five | +0.027 |
agging | +0.026 |
mine | +0.024 |
EEP | +0.024 |
Neg logit contributions | |
Pig | -0.024 |
Ally | -0.022 |
wood | -0.021 |
Bunny | -0.021 |
Beth | -0.021 |
Token | safe |
Token ID | 3338 |
Feature activation | -4.84e-01 |
Pos logit contributions | |
Pig | +0.018 |
wood | +0.017 |
Ally | +0.016 |
Bert | +0.016 |
Beth | +0.016 |
Neg logit contributions | |
agging | -0.018 |
cheek | -0.017 |
mine | -0.017 |
five | -0.016 |
liking | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
agging | +0.018 |
cheek | +0.018 |
five | +0.017 |
mine | +0.016 |
fourteen | +0.015 |
Neg logit contributions | |
Pig | -0.021 |
Ally | -0.020 |
known | -0.019 |
Beth | -0.019 |
Tweet | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | -4.15e-01 |
Pos logit contributions | |
five | +0.009 |
cheek | +0.008 |
agging | +0.008 |
mine | +0.008 |
EEP | +0.008 |
Neg logit contributions | |
Pig | -0.007 |
known | -0.006 |
wood | -0.006 |
Ally | -0.006 |
Beth | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | -4.86e-01 |
Pos logit contributions | |
five | +0.023 |
agging | +0.022 |
cheek | +0.022 |
fourteen | +0.021 |
mine | +0.021 |
Neg logit contributions | |
Pig | -0.011 |
Tweet | -0.010 |
Bunny | -0.009 |
Beth | -0.009 |
Ally | -0.009 |
Token | Later |
Token ID | 18602 |
Feature activation | -2.61e-01 |
Pos logit contributions | |
five | +0.021 |
agging | +0.021 |
cheek | +0.021 |
Deal | +0.020 |
la | +0.019 |
Neg logit contributions | |
Pig | -0.019 |
Tweet | -0.018 |
house | -0.017 |
gy | -0.017 |
Beth | -0.017 |
Token | that |
Token ID | 326 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
five | +0.011 |
agging | +0.011 |
cheek | +0.010 |
fourteen | +0.009 |
mine | +0.009 |
Neg logit contributions | |
Pig | -0.011 |
Tweet | -0.010 |
Beth | -0.009 |
Bunny | -0.009 |
Ally | -0.009 |
Token | day |
Token ID | 1110 |
Feature activation | +3.20e-01 |
Pos logit contributions | |
Pig | +0.038 |
Tweet | +0.032 |
Beth | +0.031 |
Bunny | +0.031 |
Ally | +0.030 |
Neg logit contributions | |
five | -0.072 |
agging | -0.070 |
cheek | -0.069 |
fourteen | -0.064 |
mine | -0.064 |
Token | , |
Token ID | 11 |
Feature activation | +2.67e-01 |
Pos logit contributions | |
Pig | +0.009 |
wood | +0.008 |
Tweet | +0.008 |
known | +0.008 |
Ally | +0.008 |
Neg logit contributions | |
five | -0.017 |
agging | -0.017 |
cheek | -0.016 |
fourteen | -0.015 |
mine | -0.015 |
Token | Tim |
Token ID | 5045 |
Feature activation | -9.73e-01 |
Pos logit contributions | |
Pig | +0.009 |
wood | +0.008 |
known | +0.007 |
Ally | +0.007 |
gy | +0.007 |
Neg logit contributions | |
five | -0.013 |
mine | -0.012 |
cheek | -0.012 |
fourteen | -0.012 |
agging | -0.012 |
Token | nice |
Token ID | 3621 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
Pig | +0.017 |
Beth | +0.016 |
Tweet | +0.015 |
Ally | +0.015 |
Bunny | +0.015 |
Neg logit contributions | |
five | -0.021 |
agging | -0.021 |
Deal | -0.020 |
fourteen | -0.019 |
cheek | -0.019 |
Token | tune |
Token ID | 14009 |
Feature activation | +6.24e-01 |
Pos logit contributions | |
Pig | +0.006 |
Beth | +0.006 |
Ally | +0.006 |
Tweet | +0.006 |
Bunny | +0.006 |
Neg logit contributions | |
five | -0.007 |
agging | -0.006 |
cheek | -0.006 |
Deal | -0.006 |
fourteen | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
Pig | +0.023 |
Tweet | +0.020 |
known | +0.020 |
Bunny | +0.020 |
wood | +0.020 |
Neg logit contributions | |
five | -0.029 |
agging | -0.028 |
cheek | -0.027 |
fourteen | -0.025 |
mine | -0.025 |
Token | she |
Token ID | 673 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
gy | +0.001 |
Pig | +0.001 |
wood | +0.001 |
known | +0.001 |
Tweet | +0.001 |
Neg logit contributions | |
five | -0.001 |
cheek | -0.001 |
goodness | -0.001 |
fourteen | -0.001 |
agging | -0.001 |
Token | made |
Token ID | 925 |
Feature activation | +6.76e-01 |
Pos logit contributions | |
five | +0.008 |
cheek | +0.008 |
agging | +0.008 |
la | +0.008 |
fourteen | +0.007 |
Neg logit contributions | |
Pig | -0.008 |
wood | -0.007 |
Tweet | -0.007 |
Ally | -0.007 |
Bunny | -0.007 |
Token | lots |
Token ID | 6041 |
Feature activation | -2.71e+00 |
Pos logit contributions | |
Pig | +0.027 |
gy | +0.026 |
wood | +0.025 |
Tweet | +0.024 |
Ally | +0.023 |
Neg logit contributions | |
five | -0.029 |
agging | -0.026 |
cheek | -0.026 |
goodness | -0.025 |
sidelines | -0.024 |
Token | of |
Token ID | 286 |
Feature activation | +5.42e-01 |
Pos logit contributions | |
five | +0.098 |
agging | +0.091 |
cheek | +0.086 |
sidelines | +0.078 |
fourteen | +0.077 |
Neg logit contributions | |
Pig | -0.134 |
gy | -0.126 |
wood | -0.121 |
known | -0.120 |
Ally | -0.119 |
Token | strange |
Token ID | 6283 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
Pig | +0.022 |
Tweet | +0.020 |
gy | +0.019 |
Bunny | +0.019 |
wood | +0.019 |
Neg logit contributions | |
five | -0.023 |
agging | -0.022 |
cheek | -0.022 |
sidelines | -0.020 |
fourteen | -0.019 |
Token | noises |
Token ID | 26782 |
Feature activation | +8.36e-01 |
Pos logit contributions | |
Pig | +0.068 |
Tweet | +0.062 |
Bunny | +0.060 |
known | +0.059 |
gy | +0.059 |
Neg logit contributions | |
five | -0.069 |
agging | -0.066 |
cheek | -0.064 |
mine | -0.057 |
goodness | -0.057 |
Token | . |
Token ID | 13 |
Feature activation | +3.58e-01 |
Pos logit contributions | |
Pig | +0.033 |
Tweet | +0.029 |
gy | +0.029 |
known | +0.029 |
Ally | +0.029 |
Neg logit contributions | |
five | -0.036 |
agging | -0.035 |
cheek | -0.033 |
fourteen | -0.032 |
mine | -0.031 |
Token |
|
Token ID | 198 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
Pig | +0.012 |
gy | +0.011 |
Tweet | +0.011 |
wood | +0.011 |
known | +0.011 |
Neg logit contributions | |
five | -0.016 |
cheek | -0.015 |
agging | -0.015 |
Deal | -0.015 |
goodness | -0.014 |
Token | of |
Token ID | 286 |
Feature activation | +9.04e-01 |
Pos logit contributions | |
Pig | +0.016 |
known | +0.016 |
wood | +0.016 |
gy | +0.016 |
increasingly | +0.015 |
Neg logit contributions | |
five | -0.007 |
agging | -0.006 |
mine | -0.006 |
cheek | -0.005 |
la | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | -7.72e-03 |
Pos logit contributions | |
wood | +0.034 |
Pig | +0.033 |
increasingly | +0.029 |
known | +0.029 |
rou | +0.028 |
Neg logit contributions | |
five | -0.035 |
fourteen | -0.035 |
agging | -0.035 |
mine | -0.034 |
cheek | -0.033 |
Token | day |
Token ID | 1110 |
Feature activation | -8.01e-01 |
Pos logit contributions | |
five | +0.000 |
agging | +0.000 |
cheek | +0.000 |
Deal | +0.000 |
arl | +0.000 |
Neg logit contributions | |
Pig | -0.000 |
Tweet | -0.000 |
Ally | -0.000 |
Bunny | -0.000 |
Beth | -0.000 |
Token | , |
Token ID | 11 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
five | +0.032 |
agging | +0.029 |
goodness | +0.028 |
la | +0.028 |
mine | +0.028 |
Neg logit contributions | |
Pig | -0.032 |
wood | -0.031 |
gy | -0.031 |
known | -0.030 |
increasingly | -0.028 |
Token | they |
Token ID | 484 |
Feature activation | -6.78e-01 |
Pos logit contributions | |
five | +0.006 |
mine | +0.006 |
cheek | +0.005 |
la | +0.005 |
fourteen | +0.005 |
Neg logit contributions | |
wood | -0.006 |
Pig | -0.006 |
known | -0.005 |
gy | -0.005 |
dam | -0.005 |
Token | both |
Token ID | 1111 |
Feature activation | -3.35e+00 |
Pos logit contributions | |
five | +0.029 |
agging | +0.027 |
cheek | +0.027 |
ige | +0.026 |
mine | +0.025 |
Neg logit contributions | |
Pig | -0.025 |
wood | -0.024 |
gy | -0.023 |
Tweet | -0.022 |
Beth | -0.022 |
Token | knew |
Token ID | 2993 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
five | +0.134 |
cheek | +0.119 |
agging | +0.114 |
ige | +0.113 |
la | +0.113 |
Neg logit contributions | |
Pig | -0.140 |
wood | -0.135 |
Tweet | -0.127 |
Bert | -0.125 |
gy | -0.124 |
Token | they |
Token ID | 484 |
Feature activation | -5.40e-01 |
Pos logit contributions | |
mine | +0.046 |
five | +0.043 |
agging | +0.043 |
cheek | +0.043 |
fourteen | +0.041 |
Neg logit contributions | |
Pig | -0.070 |
gy | -0.068 |
wood | -0.067 |
known | -0.065 |
increasingly | -0.063 |
Token | would |
Token ID | 561 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
five | +0.025 |
agging | +0.024 |
cheek | +0.023 |
mine | +0.022 |
fourteen | +0.021 |
Neg logit contributions | |
Pig | -0.020 |
Tweet | -0.017 |
wood | -0.017 |
Beth | -0.017 |
Bunny | -0.017 |
Token | be |
Token ID | 307 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
five | +0.004 |
agging | +0.004 |
cheek | +0.004 |
mine | +0.004 |
fourteen | +0.004 |
Neg logit contributions | |
Pig | -0.005 |
Beth | -0.004 |
wood | -0.004 |
Tweet | -0.004 |
Bunny | -0.004 |
Token | best |
Token ID | 1266 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
agging | +0.004 |
mine | +0.004 |
five | +0.004 |
cheek | +0.004 |
ige | +0.004 |
Neg logit contributions | |
wood | -0.005 |
Pig | -0.005 |
Bert | -0.004 |
Beth | -0.004 |
stal | -0.004 |
Token | he |
Token ID | 339 |
Feature activation | -3.13e-01 |
Pos logit contributions | |
five | +0.014 |
cheek | +0.014 |
agging | +0.014 |
mine | +0.013 |
fourteen | +0.013 |
Neg logit contributions | |
Pig | -0.014 |
gy | -0.013 |
wood | -0.013 |
known | -0.013 |
Beth | -0.013 |
Token | had |
Token ID | 550 |
Feature activation | -6.00e-01 |
Pos logit contributions | |
five | +0.014 |
agging | +0.013 |
cheek | +0.013 |
mine | +0.012 |
fourteen | +0.012 |
Neg logit contributions | |
Pig | -0.012 |
Tweet | -0.010 |
Beth | -0.010 |
wood | -0.010 |
gy | -0.010 |
Token | never |
Token ID | 1239 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
five | +0.031 |
agging | +0.030 |
cheek | +0.029 |
arl | +0.028 |
Deal | +0.027 |
Neg logit contributions | |
Pig | -0.019 |
Tweet | -0.016 |
known | -0.016 |
Bunny | -0.016 |
Beth | -0.016 |
Token | seen |
Token ID | 1775 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
five | +0.047 |
agging | +0.045 |
cheek | +0.044 |
mine | +0.040 |
fourteen | +0.040 |
Neg logit contributions | |
Pig | -0.055 |
Beth | -0.050 |
Tweet | -0.050 |
Ally | -0.049 |
Bunny | -0.049 |
Token | this |
Token ID | 428 |
Feature activation | -8.25e-01 |
Pos logit contributions | |
five | +0.043 |
fourteen | +0.042 |
mine | +0.041 |
agging | +0.041 |
cheek | +0.039 |
Neg logit contributions | |
Pig | -0.035 |
gy | -0.034 |
wood | -0.032 |
house | -0.032 |
Ally | -0.031 |
Token | many |
Token ID | 867 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
five | +0.033 |
cheek | +0.032 |
agging | +0.032 |
mine | +0.032 |
fourteen | +0.030 |
Neg logit contributions | |
Pig | -0.032 |
Beth | -0.030 |
known | -0.029 |
gy | -0.029 |
wood | -0.029 |
Token | fish |
Token ID | 5916 |
Feature activation | -5.90e-01 |
Pos logit contributions | |
five | +0.105 |
agging | +0.103 |
cheek | +0.098 |
mine | +0.092 |
fourteen | +0.090 |
Neg logit contributions | |
Pig | -0.112 |
Tweet | -0.100 |
gy | -0.100 |
Beth | -0.099 |
wood | -0.098 |
Token | before |
Token ID | 878 |
Feature activation | +3.41e-01 |
Pos logit contributions | |
five | +0.028 |
agging | +0.027 |
cheek | +0.026 |
fourteen | +0.025 |
Deal | +0.025 |
Neg logit contributions | |
Pig | -0.021 |
Tweet | -0.018 |
Bunny | -0.018 |
Beth | -0.018 |
gy | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +2.76e-01 |
Pos logit contributions | |
Pig | +0.013 |
Beth | +0.012 |
Tweet | +0.012 |
known | +0.012 |
gy | +0.012 |
Neg logit contributions | |
agging | -0.014 |
five | -0.014 |
cheek | -0.014 |
mine | -0.013 |
fourteen | -0.012 |
Token | He |
Token ID | 679 |
Feature activation | +3.51e-01 |
Pos logit contributions | |
Pig | +0.010 |
known | +0.010 |
gy | +0.009 |
Ally | +0.009 |
Beth | +0.009 |
Neg logit contributions | |
five | -0.012 |
cheek | -0.011 |
agging | -0.011 |
Deal | -0.010 |
fourteen | -0.010 |
Token | was |
Token ID | 373 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
Pig | +0.013 |
Beth | +0.012 |
wood | +0.012 |
Tweet | +0.012 |
Ally | +0.011 |
Neg logit contributions | |
five | -0.015 |
cheek | -0.014 |
agging | -0.014 |
mine | -0.013 |
fourteen | -0.013 |
Token | little |
Token ID | 1310 |
Feature activation | +7.05e-01 |
Pos logit contributions | |
Pig | +0.006 |
Ally | +0.005 |
known | +0.005 |
Tweet | +0.005 |
Beth | +0.005 |
Neg logit contributions | |
cheek | -0.007 |
five | -0.006 |
agging | -0.006 |
mine | -0.006 |
goodness | -0.006 |
Token | boy |
Token ID | 2933 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
known | +0.033 |
ained | +0.032 |
Ally | +0.031 |
Days | +0.030 |
Pig | +0.030 |
Neg logit contributions | |
cheek | -0.025 |
five | -0.022 |
mine | -0.019 |
fork | -0.019 |
goodness | -0.019 |
Token | some |
Token ID | 617 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
Pig | +0.065 |
Beth | +0.056 |
Ally | +0.056 |
Tweet | +0.055 |
Bert | +0.053 |
Neg logit contributions | |
five | -0.099 |
agging | -0.087 |
cheek | -0.086 |
mine | -0.083 |
fourteen | -0.083 |
Token | of |
Token ID | 286 |
Feature activation | +6.46e-01 |
Pos logit contributions | |
five | +0.040 |
agging | +0.039 |
cheek | +0.038 |
fourteen | +0.035 |
Deal | +0.034 |
Neg logit contributions | |
Pig | -0.046 |
Tweet | -0.042 |
Bunny | -0.041 |
Beth | -0.040 |
Ally | -0.040 |
Token | his |
Token ID | 465 |
Feature activation | -2.91e-01 |
Pos logit contributions | |
gy | +0.022 |
Pig | +0.022 |
house | +0.022 |
wood | +0.021 |
darker | +0.020 |
Neg logit contributions | |
five | -0.028 |
agging | -0.026 |
cheek | -0.025 |
mine | -0.025 |
fourteen | -0.025 |
Token | lunch |
Token ID | 9965 |
Feature activation | -2.60e+00 |
Pos logit contributions | |
five | +0.011 |
cheek | +0.011 |
agging | +0.011 |
mine | +0.010 |
goodness | +0.010 |
Neg logit contributions | |
Pig | -0.012 |
Beth | -0.011 |
Ally | -0.011 |
known | -0.011 |
Tweet | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
five | +0.120 |
agging | +0.116 |
cheek | +0.114 |
fourteen | +0.104 |
mine | +0.104 |
Neg logit contributions | |
Pig | -0.093 |
Tweet | -0.083 |
Beth | -0.081 |
Ally | -0.081 |
known | -0.080 |
Token | they |
Token ID | 484 |
Feature activation | +4.40e-01 |
Pos logit contributions | |
Pig | +0.015 |
ained | +0.015 |
wood | +0.014 |
darker | +0.014 |
gy | +0.014 |
Neg logit contributions | |
five | -0.017 |
cheek | -0.015 |
fourteen | -0.015 |
mine | -0.015 |
slice | -0.014 |
Token | had |
Token ID | 550 |
Feature activation | -4.22e-02 |
Pos logit contributions | |
Pig | +0.018 |
wood | +0.016 |
Tweet | +0.016 |
Beth | +0.016 |
Ally | +0.015 |
Neg logit contributions | |
five | -0.018 |
agging | -0.017 |
cheek | -0.016 |
fourteen | -0.016 |
mine | -0.016 |
Token | a |
Token ID | 257 |
Feature activation | +5.37e-01 |
Pos logit contributions | |
five | +0.002 |
agging | +0.002 |
mine | +0.002 |
cheek | +0.002 |
fourteen | +0.002 |
Neg logit contributions | |
Pig | -0.001 |
Beth | -0.001 |
Ally | -0.001 |
gy | -0.001 |
Tweet | -0.001 |
Token | happy |
Token ID | 3772 |
Feature activation | -7.75e-01 |
Pos logit contributions | |
Pig | +0.023 |
wood | +0.021 |
Tweet | +0.020 |
Beth | +0.020 |
gy | +0.020 |
Neg logit contributions | |
five | -0.022 |
agging | -0.020 |
cheek | -0.020 |
mine | -0.019 |
fourteen | -0.018 |
Token | notebook |
Token ID | 20922 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
five | +0.008 |
agging | +0.008 |
cheek | +0.008 |
mine | +0.007 |
fourteen | +0.007 |
Neg logit contributions | |
Pig | -0.008 |
Tweet | -0.008 |
Beth | -0.007 |
Bunny | -0.007 |
known | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +8.32e-01 |
Pos logit contributions | |
five | +0.052 |
agging | +0.050 |
cheek | +0.050 |
mine | +0.047 |
fourteen | +0.046 |
Neg logit contributions | |
Pig | -0.042 |
Tweet | -0.037 |
known | -0.036 |
Bunny | -0.036 |
Ally | -0.036 |
Token |
|
Token ID | 198 |
Feature activation | +3.68e-01 |
Pos logit contributions | |
Pig | +0.032 |
known | +0.029 |
wood | +0.028 |
gy | +0.028 |
Tweet | +0.027 |
Neg logit contributions | |
five | -0.035 |
mine | -0.032 |
cheek | -0.031 |
agging | -0.031 |
fourteen | -0.031 |
Token |
|
Token ID | 198 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
Pig | +0.013 |
Tweet | +0.011 |
wood | +0.011 |
gy | +0.011 |
Bunny | +0.011 |
Neg logit contributions | |
five | -0.017 |
agging | -0.016 |
cheek | -0.016 |
fourteen | -0.015 |
arl | -0.015 |
Token | They |
Token ID | 2990 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
Pig | +0.010 |
Tweet | +0.009 |
Beth | +0.009 |
Ally | +0.009 |
Bunny | +0.009 |
Neg logit contributions | |
five | -0.005 |
agging | -0.005 |
cheek | -0.005 |
fourteen | -0.004 |
Deal | -0.004 |
Token | both |
Token ID | 1111 |
Feature activation | -2.86e+00 |
Pos logit contributions | |
five | +0.005 |
agging | +0.004 |
mine | +0.004 |
cheek | +0.004 |
fourteen | +0.004 |
Neg logit contributions | |
Pig | -0.004 |
Tweet | -0.004 |
wood | -0.004 |
Bunny | -0.003 |
gy | -0.003 |
Token | argued |
Token ID | 7189 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
five | +0.121 |
agging | +0.112 |
cheek | +0.109 |
fourteen | +0.103 |
mine | +0.103 |
Neg logit contributions | |
Pig | -0.118 |
Tweet | -0.106 |
Bunny | -0.103 |
wood | -0.103 |
Ally | -0.101 |
Token | , |
Token ID | 11 |
Feature activation | +4.84e-01 |
Pos logit contributions | |
five | +0.058 |
agging | +0.056 |
cheek | +0.054 |
fourteen | +0.051 |
mine | +0.050 |
Neg logit contributions | |
Pig | -0.049 |
Tweet | -0.043 |
Bunny | -0.042 |
Beth | -0.042 |
Ally | -0.041 |
Token | but |
Token ID | 475 |
Feature activation | +5.22e-01 |
Pos logit contributions | |
Pig | +0.020 |
Tweet | +0.017 |
wood | +0.017 |
gy | +0.017 |
Bunny | +0.017 |
Neg logit contributions | |
five | -0.020 |
agging | -0.019 |
cheek | -0.019 |
fourteen | -0.018 |
mine | -0.018 |
Token | neither |
Token ID | 6159 |
Feature activation | -5.20e-01 |
Pos logit contributions | |
Pig | +0.021 |
wood | +0.019 |
gy | +0.018 |
Tweet | +0.018 |
dam | +0.018 |
Neg logit contributions | |
mine | -0.021 |
five | -0.020 |
fourteen | -0.019 |
cheek | -0.019 |
agging | -0.019 |
Token | one |
Token ID | 530 |
Feature activation | +3.79e-01 |
Pos logit contributions | |
five | +0.022 |
agging | +0.020 |
cheek | +0.020 |
mine | +0.019 |
fourteen | +0.019 |
Neg logit contributions | |
Pig | -0.021 |
Tweet | -0.019 |
gy | -0.019 |
wood | -0.019 |
Bunny | -0.018 |