Token | just |
Token ID | 655 |
Feature activation | +2.69e-01 |
Pos logit contributions | |
shark | +0.008 |
Bad | +0.008 |
Shark | +0.007 |
Fin | +0.007 |
fin | +0.007 |
Neg logit contributions | |
anna | -0.011 |
brushes | -0.010 |
raid | -0.010 |
bars | -0.010 |
types | -0.010 |
Token | like |
Token ID | 588 |
Feature activation | -4.89e-01 |
Pos logit contributions | |
anna | +0.012 |
brushes | +0.012 |
bars | +0.012 |
raid | +0.011 |
types | +0.011 |
Neg logit contributions | |
shark | -0.008 |
Bad | -0.008 |
Shark | -0.007 |
ile | -0.007 |
fin | -0.007 |
Token | in |
Token ID | 287 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
shark | +0.016 |
Bad | +0.016 |
Shark | +0.015 |
ile | +0.015 |
Fin | +0.015 |
Neg logit contributions | |
anna | -0.020 |
brushes | -0.019 |
raid | -0.019 |
bars | -0.019 |
types | -0.018 |
Token | a |
Token ID | 257 |
Feature activation | +5.10e-01 |
Pos logit contributions | |
brushes | +0.042 |
anna | +0.039 |
bars | +0.038 |
types | +0.038 |
Paula | +0.037 |
Neg logit contributions | |
shark | -0.045 |
Bad | -0.044 |
Shark | -0.043 |
ile | -0.042 |
coon | -0.042 |
Token | fair |
Token ID | 3148 |
Feature activation | +8.15e-01 |
Pos logit contributions | |
brushes | +0.017 |
anna | +0.017 |
bars | +0.016 |
raid | +0.016 |
types | +0.016 |
Neg logit contributions | |
shark | -0.020 |
Bad | -0.020 |
Shark | -0.019 |
ile | -0.019 |
coon | -0.019 |
Token | yt |
Token ID | 20760 |
Feature activation | +6.99e+00 |
Pos logit contributions | |
brushes | +0.030 |
anna | +0.029 |
bars | +0.028 |
raid | +0.028 |
types | +0.027 |
Neg logit contributions | |
shark | -0.031 |
Bad | -0.030 |
Shark | -0.029 |
ile | -0.029 |
fin | -0.028 |
Token | ale |
Token ID | 1000 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
brushes | +0.242 |
types | +0.225 |
Paula | +0.220 |
alphabet | +0.216 |
cere | +0.213 |
Neg logit contributions | |
ile | -0.278 |
shark | -0.241 |
ful | -0.237 |
Bad | -0.230 |
Fin | -0.228 |
Token | ! |
Token ID | 0 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
brushes | +0.050 |
anna | +0.050 |
bars | +0.047 |
raid | +0.047 |
types | +0.046 |
Neg logit contributions | |
shark | -0.048 |
Bad | -0.048 |
Shark | -0.046 |
ile | -0.045 |
coon | -0.044 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
brushes | +0.005 |
anna | +0.005 |
types | +0.005 |
raid | +0.005 |
bars | +0.005 |
Neg logit contributions | |
Shark | -0.003 |
Bad | -0.003 |
shark | -0.003 |
Fin | -0.003 |
Wally | -0.003 |
Token | Once |
Token ID | 7454 |
Feature activation | +3.70e-01 |
Pos logit contributions | |
Bad | +0.010 |
shark | +0.010 |
ile | +0.009 |
fin | +0.009 |
coon | +0.009 |
Neg logit contributions | |
anna | -0.007 |
brushes | -0.007 |
bars | -0.006 |
types | -0.006 |
Along | -0.006 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
brushes | +0.008 |
anna | +0.008 |
raid | +0.007 |
bars | +0.007 |
types | +0.007 |
Neg logit contributions | |
shark | -0.020 |
Bad | -0.019 |
Shark | -0.019 |
ile | -0.019 |
Fin | -0.018 |
Token | was |
Token ID | 373 |
Feature activation | -4.81e-02 |
Pos logit contributions | |
anna | +0.034 |
brushes | +0.033 |
raid | +0.031 |
bars | +0.031 |
types | +0.031 |
Neg logit contributions | |
shark | -0.035 |
Bad | -0.035 |
Shark | -0.034 |
ile | -0.032 |
coon | -0.032 |
Token | walking |
Token ID | 6155 |
Feature activation | +2.52e+00 |
Pos logit contributions | |
shark | +0.002 |
Bad | +0.002 |
Shark | +0.002 |
ile | +0.002 |
coon | +0.002 |
Neg logit contributions | |
brushes | -0.002 |
anna | -0.002 |
bars | -0.002 |
raid | -0.002 |
types | -0.002 |
Token | through |
Token ID | 832 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
anna | +0.089 |
brushes | +0.089 |
raid | +0.084 |
types | +0.083 |
bars | +0.083 |
Neg logit contributions | |
Bad | -0.098 |
shark | -0.097 |
Shark | -0.095 |
ile | -0.090 |
coon | -0.090 |
Token | a |
Token ID | 257 |
Feature activation | -4.13e-01 |
Pos logit contributions | |
brushes | +0.009 |
anna | +0.008 |
types | +0.008 |
decorations | +0.008 |
Paula | +0.008 |
Neg logit contributions | |
shark | -0.010 |
Bad | -0.010 |
Shark | -0.010 |
ile | -0.009 |
coon | -0.009 |
Token | fair |
Token ID | 3148 |
Feature activation | +4.92e-01 |
Pos logit contributions | |
Bad | +0.017 |
shark | +0.017 |
Shark | +0.017 |
ile | +0.017 |
coon | +0.016 |
Neg logit contributions | |
brushes | -0.013 |
anna | -0.013 |
types | -0.012 |
bars | -0.012 |
raid | -0.012 |
Token | yt |
Token ID | 20760 |
Feature activation | +6.57e+00 |
Pos logit contributions | |
brushes | +0.016 |
anna | +0.016 |
raid | +0.014 |
bars | +0.014 |
types | +0.014 |
Neg logit contributions | |
Bad | -0.021 |
shark | -0.021 |
Shark | -0.020 |
ile | -0.020 |
coon | -0.020 |
Token | ale |
Token ID | 1000 |
Feature activation | +6.63e-01 |
Pos logit contributions | |
brushes | +0.222 |
types | +0.207 |
Paula | +0.200 |
cere | +0.200 |
alphabet | +0.199 |
Neg logit contributions | |
ile | -0.266 |
ful | -0.234 |
Bad | -0.227 |
shark | -0.224 |
coon | -0.224 |
Token | . |
Token ID | 13 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
anna | +0.023 |
brushes | +0.023 |
raid | +0.022 |
bars | +0.021 |
types | +0.021 |
Neg logit contributions | |
Bad | -0.027 |
shark | -0.026 |
Shark | -0.026 |
Fin | -0.025 |
coon | -0.024 |
Token | She |
Token ID | 1375 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
Wally | +0.001 |
Bad | +0.001 |
Fin | +0.001 |
Shark | +0.001 |
shark | +0.001 |
Neg logit contributions | |
brushes | -0.001 |
bars | -0.001 |
types | -0.001 |
raid | -0.001 |
anna | -0.001 |
Token | fell |
Token ID | 3214 |
Feature activation | -2.84e+00 |
Pos logit contributions | |
anna | +0.048 |
brushes | +0.048 |
bars | +0.045 |
raid | +0.044 |
types | +0.044 |
Neg logit contributions | |
shark | -0.058 |
Bad | -0.057 |
Shark | -0.055 |
ile | -0.054 |
coon | -0.054 |
Token | asleep |
Token ID | 16039 |
Feature activation | -2.81e+00 |
Pos logit contributions | |
shark | +0.126 |
Bad | +0.124 |
Shark | +0.120 |
ile | +0.117 |
coon | +0.115 |
Neg logit contributions | |
anna | -0.085 |
brushes | -0.085 |
bars | -0.079 |
raid | -0.078 |
types | -0.078 |
Token | flowers |
Token ID | 12734 |
Feature activation | +3.20e+00 |
Pos logit contributions | |
brushes | +0.049 |
anna | +0.048 |
raid | +0.045 |
bars | +0.044 |
types | +0.044 |
Neg logit contributions | |
shark | -0.062 |
Bad | -0.061 |
Shark | -0.060 |
ile | -0.058 |
coon | -0.057 |
Token | and |
Token ID | 290 |
Feature activation | +7.92e-01 |
Pos logit contributions | |
brushes | +0.116 |
anna | +0.116 |
raid | +0.110 |
bars | +0.106 |
decorations | +0.103 |
Neg logit contributions | |
Bad | -0.125 |
shark | -0.121 |
Shark | -0.119 |
coon | -0.114 |
ile | -0.112 |
Token | make |
Token ID | 787 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
brushes | +0.025 |
anna | +0.025 |
bars | +0.024 |
raid | +0.023 |
types | +0.023 |
Neg logit contributions | |
shark | -0.033 |
Bad | -0.033 |
Shark | -0.032 |
ile | -0.031 |
coon | -0.031 |
Token | a |
Token ID | 257 |
Feature activation | +6.93e-01 |
Pos logit contributions | |
brushes | +0.046 |
anna | +0.045 |
types | +0.042 |
raid | +0.042 |
bars | +0.041 |
Neg logit contributions | |
shark | -0.049 |
Bad | -0.048 |
Shark | -0.047 |
ile | -0.045 |
coon | -0.045 |
Token | bou |
Token ID | 35833 |
Feature activation | +4.93e+00 |
Pos logit contributions | |
anna | +0.021 |
brushes | +0.021 |
bars | +0.019 |
raid | +0.019 |
types | +0.019 |
Neg logit contributions | |
shark | -0.031 |
Bad | -0.031 |
Shark | -0.030 |
ile | -0.029 |
coon | -0.028 |
Token | quet |
Token ID | 21108 |
Feature activation | +6.12e+00 |
Pos logit contributions | |
brushes | +0.166 |
anna | +0.165 |
types | +0.159 |
bars | +0.158 |
nie | +0.152 |
Neg logit contributions | |
coon | -0.193 |
ile | -0.186 |
Bad | -0.184 |
shark | -0.179 |
ball | -0.179 |
Token | . |
Token ID | 13 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
brushes | +0.211 |
anna | +0.207 |
bars | +0.199 |
raid | +0.198 |
types | +0.191 |
Neg logit contributions | |
Bad | -0.248 |
shark | -0.243 |
Shark | -0.235 |
coon | -0.226 |
ile | -0.224 |
Token | We |
Token ID | 775 |
Feature activation | -2.90e-01 |
Pos logit contributions | |
brushes | +0.035 |
anna | +0.034 |
bars | +0.032 |
raid | +0.032 |
types | +0.031 |
Neg logit contributions | |
shark | -0.051 |
Bad | -0.049 |
ile | -0.048 |
coon | -0.048 |
Shark | -0.048 |
Token | can |
Token ID | 460 |
Feature activation | +4.69e-01 |
Pos logit contributions | |
shark | +0.012 |
Bad | +0.012 |
Shark | +0.011 |
ile | +0.011 |
coon | +0.011 |
Neg logit contributions | |
brushes | -0.010 |
anna | -0.010 |
bars | -0.009 |
raid | -0.009 |
types | -0.009 |
Token | also |
Token ID | 635 |
Feature activation | +6.85e-01 |
Pos logit contributions | |
brushes | +0.014 |
anna | +0.014 |
bars | +0.013 |
raid | +0.013 |
types | +0.013 |
Neg logit contributions | |
shark | -0.021 |
Bad | -0.021 |
Shark | -0.020 |
ile | -0.020 |
coon | -0.019 |
Token | make |
Token ID | 787 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
brushes | +0.019 |
anna | +0.019 |
bars | +0.017 |
raid | +0.017 |
types | +0.017 |
Neg logit contributions | |
shark | -0.032 |
Bad | -0.032 |
Shark | -0.031 |
ile | -0.030 |
coon | -0.030 |
Token | something |
Token ID | 1223 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
shark | +0.015 |
Bad | +0.014 |
Shark | +0.014 |
ile | +0.014 |
coon | +0.013 |
Neg logit contributions | |
brushes | -0.019 |
anna | -0.019 |
bars | -0.018 |
raid | -0.018 |
types | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -5.95e-02 |
Pos logit contributions | |
Bad | +0.067 |
Shark | +0.065 |
shark | +0.064 |
ile | +0.064 |
fin | +0.060 |
Neg logit contributions | |
brushes | -0.052 |
anna | -0.046 |
raid | -0.044 |
sights | -0.044 |
select | -0.043 |
Token | A |
Token ID | 317 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
shark | +0.002 |
Shark | +0.002 |
Fin | +0.002 |
Bad | +0.002 |
Wally | +0.002 |
Neg logit contributions | |
bars | -0.002 |
anna | -0.002 |
brushes | -0.002 |
types | -0.002 |
raid | -0.002 |
Token | little |
Token ID | 1310 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
brushes | +0.005 |
anna | +0.005 |
types | +0.005 |
raid | +0.005 |
bars | +0.005 |
Neg logit contributions | |
Bad | -0.007 |
shark | -0.007 |
ile | -0.006 |
Shark | -0.006 |
coon | -0.006 |
Token | fair |
Token ID | 3148 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
brushes | +0.046 |
raid | +0.043 |
types | +0.043 |
sights | +0.041 |
bars | +0.040 |
Neg logit contributions | |
Bad | -0.056 |
ile | -0.053 |
coon | -0.052 |
Shark | -0.052 |
shark | -0.052 |
Token | yt |
Token ID | 20760 |
Feature activation | +6.08e+00 |
Pos logit contributions | |
brushes | +0.008 |
anna | +0.008 |
types | +0.007 |
raid | +0.007 |
bars | +0.007 |
Neg logit contributions | |
Bad | -0.007 |
Shark | -0.006 |
shark | -0.006 |
ile | -0.006 |
coon | -0.006 |
Token | ale |
Token ID | 1000 |
Feature activation | +7.34e-01 |
Pos logit contributions | |
brushes | +0.176 |
anna | +0.174 |
types | +0.158 |
bars | +0.157 |
raid | +0.156 |
Neg logit contributions | |
shark | -0.276 |
Bad | -0.272 |
ile | -0.265 |
Shark | -0.263 |
coon | -0.257 |
Token | bird |
Token ID | 6512 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
brushes | +0.026 |
anna | +0.026 |
types | +0.024 |
raid | +0.024 |
sights | +0.023 |
Neg logit contributions | |
Bad | -0.028 |
shark | -0.026 |
Shark | -0.026 |
Wally | -0.025 |
fin | -0.024 |
Token | was |
Token ID | 373 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
shark | +0.005 |
Bad | +0.005 |
Shark | +0.005 |
Wally | +0.005 |
coon | +0.005 |
Neg logit contributions | |
anna | -0.004 |
brushes | -0.004 |
raid | -0.004 |
bars | -0.004 |
types | -0.004 |
Token | singing |
Token ID | 13777 |
Feature activation | +6.04e-01 |
Pos logit contributions | |
Bad | +0.016 |
shark | +0.016 |
Shark | +0.016 |
Rex | +0.015 |
ile | +0.015 |
Neg logit contributions | |
anna | -0.008 |
brushes | -0.008 |
Paula | -0.008 |
select | -0.007 |
decorations | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | +2.81e-01 |
Pos logit contributions | |
brushes | +0.018 |
anna | +0.018 |
Paula | +0.017 |
sights | +0.017 |
types | +0.017 |
Neg logit contributions | |
Bad | -0.024 |
Shark | -0.024 |
shark | -0.024 |
fin | -0.022 |
ile | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
brushes | +0.056 |
anna | +0.056 |
bars | +0.053 |
raid | +0.052 |
types | +0.052 |
Neg logit contributions | |
shark | -0.049 |
Bad | -0.048 |
Shark | -0.046 |
ile | -0.045 |
coon | -0.044 |
Token | saw |
Token ID | 2497 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
shark | +0.006 |
Bad | +0.006 |
Shark | +0.006 |
ile | +0.006 |
coon | +0.006 |
Neg logit contributions | |
brushes | -0.006 |
anna | -0.006 |
bars | -0.006 |
raid | -0.006 |
types | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
Bad | +0.009 |
ile | +0.008 |
shark | +0.008 |
coon | +0.008 |
RR | +0.008 |
Neg logit contributions | |
types | -0.007 |
brushes | -0.007 |
sights | -0.006 |
decorations | -0.006 |
kinds | -0.006 |
Token | magical |
Token ID | 10883 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
Bad | +0.009 |
ile | +0.009 |
shark | +0.009 |
coon | +0.009 |
Shark | +0.009 |
Neg logit contributions | |
brushes | -0.006 |
anna | -0.006 |
types | -0.005 |
sights | -0.005 |
alphabet | -0.005 |
Token | fair |
Token ID | 3148 |
Feature activation | -2.60e-02 |
Pos logit contributions | |
brushes | +0.007 |
anna | +0.007 |
types | +0.007 |
sights | +0.006 |
raid | +0.006 |
Neg logit contributions | |
Bad | -0.006 |
Shark | -0.006 |
shark | -0.005 |
ile | -0.005 |
coon | -0.005 |
Token | yt |
Token ID | 20760 |
Feature activation | +6.07e+00 |
Pos logit contributions | |
Bad | +0.001 |
shark | +0.001 |
Shark | +0.001 |
ile | +0.001 |
coon | +0.001 |
Neg logit contributions | |
brushes | -0.001 |
anna | -0.001 |
bars | -0.001 |
types | -0.001 |
raid | -0.001 |
Token | ale |
Token ID | 1000 |
Feature activation | +3.96e-01 |
Pos logit contributions | |
brushes | +0.183 |
anna | +0.171 |
types | +0.162 |
alphabet | +0.155 |
bars | +0.155 |
Neg logit contributions | |
shark | -0.269 |
ile | -0.266 |
Bad | -0.259 |
Shark | -0.252 |
coon | -0.251 |
Token | creature |
Token ID | 7185 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
anna | +0.014 |
brushes | +0.014 |
bars | +0.013 |
raid | +0.013 |
types | +0.013 |
Neg logit contributions | |
Bad | -0.015 |
shark | -0.014 |
Shark | -0.014 |
Wally | -0.013 |
Fin | -0.013 |
Token | called |
Token ID | 1444 |
Feature activation | -1.88e-01 |
Pos logit contributions | |
Bad | +0.062 |
shark | +0.057 |
Shark | +0.056 |
ile | +0.055 |
coon | +0.054 |
Neg logit contributions | |
anna | -0.056 |
brushes | -0.054 |
raid | -0.052 |
bars | -0.051 |
rings | -0.050 |
Token | a |
Token ID | 257 |
Feature activation | -5.06e-01 |
Pos logit contributions | |
Bad | +0.008 |
shark | +0.007 |
ile | +0.007 |
Shark | +0.007 |
coon | +0.007 |
Neg logit contributions | |
anna | -0.006 |
brushes | -0.006 |
Paula | -0.006 |
bars | -0.006 |
types | -0.006 |
Token | n |
Token ID | 299 |
Feature activation | -7.22e-01 |
Pos logit contributions | |
Bad | +0.019 |
ile | +0.018 |
shark | +0.018 |
coon | +0.017 |
Shark | +0.017 |
Neg logit contributions | |
brushes | -0.017 |
anna | -0.017 |
types | -0.016 |
alphabet | -0.016 |
Paula | -0.016 |
Token | named |
Token ID | 3706 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
brushes | +0.059 |
anna | +0.059 |
raid | +0.056 |
bars | +0.054 |
types | +0.054 |
Neg logit contributions | |
shark | -0.031 |
Bad | -0.030 |
Shark | -0.028 |
coon | -0.026 |
ile | -0.026 |
Token | Lily |
Token ID | 20037 |
Feature activation | +3.10e-01 |
Pos logit contributions | |
bars | +0.056 |
types | +0.053 |
sights | +0.052 |
anna | +0.052 |
raid | +0.051 |
Neg logit contributions | |
Wally | -0.020 |
Bad | -0.019 |
ful | -0.019 |
nowhere | -0.019 |
Fred | -0.018 |
Token | who |
Token ID | 508 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
brushes | +0.012 |
anna | +0.012 |
raid | +0.011 |
bars | +0.011 |
types | +0.011 |
Neg logit contributions | |
shark | -0.011 |
Bad | -0.011 |
Shark | -0.011 |
ile | -0.010 |
coon | -0.010 |
Token | loved |
Token ID | 6151 |
Feature activation | -5.38e-01 |
Pos logit contributions | |
bars | +0.009 |
anna | +0.009 |
brushes | +0.008 |
raid | +0.008 |
types | +0.008 |
Neg logit contributions | |
Bad | -0.008 |
shark | -0.008 |
ile | -0.008 |
Shark | -0.008 |
coon | -0.007 |
Token | po |
Token ID | 745 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
shark | +0.020 |
Shark | +0.018 |
Bad | +0.018 |
Fin | +0.018 |
nowhere | +0.017 |
Neg logit contributions | |
anna | -0.021 |
bars | -0.020 |
raid | -0.019 |
nie | -0.018 |
brushes | -0.018 |
Token | pp |
Token ID | 381 |
Feature activation | +6.04e+00 |
Pos logit contributions | |
brushes | +0.067 |
types | +0.062 |
anna | +0.061 |
alphabet | +0.059 |
bars | +0.057 |
Neg logit contributions | |
Bad | -0.089 |
coon | -0.087 |
ball | -0.086 |
ile | -0.086 |
shark | -0.086 |
Token | ies |
Token ID | 444 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
brushes | +0.181 |
bars | +0.181 |
First | +0.174 |
four | +0.171 |
three | +0.168 |
Neg logit contributions | |
ball | -0.260 |
ile | -0.258 |
ful | -0.244 |
coon | -0.239 |
shark | -0.239 |
Token | . |
Token ID | 13 |
Feature activation | -8.11e-01 |
Pos logit contributions | |
anna | +0.070 |
brushes | +0.068 |
raid | +0.066 |
bars | +0.065 |
types | +0.064 |
Neg logit contributions | |
Bad | -0.086 |
shark | -0.082 |
Shark | -0.080 |
ile | -0.078 |
coon | -0.078 |
Token | She |
Token ID | 1375 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
Wally | +0.025 |
Spot | +0.024 |
anymore | +0.023 |
Fin | +0.023 |
Fred | +0.022 |
Neg logit contributions | |
anna | -0.032 |
stru | -0.029 |
ews | -0.029 |
ilda | -0.029 |
nie | -0.028 |
Token | had |
Token ID | 550 |
Feature activation | -8.46e-01 |
Pos logit contributions | |
bars | +0.040 |
raid | +0.039 |
brushes | +0.039 |
types | +0.038 |
subtract | +0.038 |
Neg logit contributions | |
shark | -0.034 |
Bad | -0.033 |
ile | -0.032 |
nowhere | -0.032 |
ball | -0.032 |
Token | a |
Token ID | 257 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
Bad | +0.023 |
shark | +0.022 |
Shark | +0.021 |
ile | +0.020 |
Fin | +0.020 |
Neg logit contributions | |
anna | -0.040 |
brushes | -0.039 |
bars | -0.039 |
raid | -0.038 |
types | -0.037 |
Token | them |
Token ID | 606 |
Feature activation | -7.16e-01 |
Pos logit contributions | |
brushes | +0.020 |
anna | +0.020 |
types | +0.019 |
bars | +0.018 |
raid | +0.018 |
Neg logit contributions | |
Bad | -0.025 |
shark | -0.024 |
ile | -0.023 |
Shark | -0.023 |
fin | -0.023 |
Token | over |
Token ID | 625 |
Feature activation | +3.70e-01 |
Pos logit contributions | |
shark | +0.031 |
Bad | +0.031 |
Shark | +0.030 |
ile | +0.029 |
Fin | +0.029 |
Neg logit contributions | |
anna | -0.022 |
brushes | -0.022 |
bars | -0.020 |
raid | -0.020 |
types | -0.020 |
Token | for |
Token ID | 329 |
Feature activation | -3.54e+00 |
Pos logit contributions | |
brushes | +0.012 |
anna | +0.011 |
bars | +0.011 |
raid | +0.011 |
types | +0.011 |
Neg logit contributions | |
shark | -0.016 |
Bad | -0.016 |
Shark | -0.015 |
ile | -0.015 |
coon | -0.015 |
Token | a |
Token ID | 257 |
Feature activation | -8.12e-01 |
Pos logit contributions | |
shark | +0.128 |
Bad | +0.125 |
Shark | +0.121 |
ile | +0.120 |
coon | +0.115 |
Neg logit contributions | |
brushes | -0.137 |
anna | -0.134 |
types | -0.125 |
raid | -0.125 |
bars | -0.124 |
Token | fair |
Token ID | 3148 |
Feature activation | +5.55e-02 |
Pos logit contributions | |
shark | +0.036 |
Bad | +0.033 |
Shark | +0.033 |
fin | +0.032 |
Wally | +0.032 |
Neg logit contributions | |
brushes | -0.025 |
bars | -0.025 |
anna | -0.025 |
types | -0.025 |
raid | -0.024 |
Token | yt |
Token ID | 20760 |
Feature activation | +6.04e+00 |
Pos logit contributions | |
anna | +0.002 |
bars | +0.002 |
brushes | +0.002 |
raid | +0.002 |
types | +0.002 |
Neg logit contributions | |
shark | -0.002 |
ile | -0.002 |
fin | -0.002 |
Shark | -0.002 |
Bad | -0.002 |
Token | ale |
Token ID | 1000 |
Feature activation | +2.72e-01 |
Pos logit contributions | |
brushes | +0.195 |
types | +0.182 |
Paula | +0.176 |
alphabet | +0.174 |
anna | +0.173 |
Neg logit contributions | |
ile | -0.249 |
shark | -0.224 |
ful | -0.221 |
coon | -0.218 |
Bad | -0.216 |
Token | tea |
Token ID | 8887 |
Feature activation | +3.43e-01 |
Pos logit contributions | |
brushes | +0.010 |
anna | +0.010 |
bars | +0.010 |
raid | +0.010 |
types | +0.009 |
Neg logit contributions | |
shark | -0.010 |
Bad | -0.010 |
Shark | -0.009 |
ile | -0.009 |
coon | -0.009 |
Token | party |
Token ID | 2151 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
brushes | +0.010 |
anna | +0.010 |
decorations | +0.009 |
raid | +0.009 |
types | +0.009 |
Neg logit contributions | |
Bad | -0.016 |
shark | -0.015 |
Shark | -0.015 |
fin | -0.014 |
ile | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
brushes | +0.053 |
anna | +0.052 |
raid | +0.049 |
bars | +0.048 |
types | +0.048 |
Neg logit contributions | |
Bad | -0.053 |
shark | -0.052 |
Shark | -0.049 |
ile | -0.047 |
fin | -0.047 |
Token | They |
Token ID | 1119 |
Feature activation | -6.79e-01 |
Pos logit contributions | |
shark | +0.011 |
Shark | +0.011 |
Bad | +0.011 |
Fin | +0.011 |
Wally | +0.011 |
Neg logit contributions | |
anna | -0.008 |
brushes | -0.008 |
bars | -0.008 |
types | -0.008 |
raid | -0.008 |
Token | " |
Token ID | 366 |
Feature activation | +3.66e-01 |
Pos logit contributions | |
shark | +0.050 |
Bad | +0.050 |
Shark | +0.048 |
ile | +0.048 |
coon | +0.047 |
Neg logit contributions | |
anna | -0.025 |
brushes | -0.025 |
bars | -0.023 |
raid | -0.022 |
types | -0.021 |
Token | But |
Token ID | 1537 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
brushes | +0.013 |
raid | +0.013 |
anna | +0.013 |
bars | +0.013 |
types | +0.012 |
Neg logit contributions | |
shark | -0.015 |
coon | -0.013 |
fin | -0.013 |
Shark | -0.013 |
ile | -0.013 |
Token | they |
Token ID | 484 |
Feature activation | -9.03e-01 |
Pos logit contributions | |
brushes | +0.007 |
bars | +0.006 |
anna | +0.006 |
types | +0.006 |
raid | +0.006 |
Neg logit contributions | |
Bad | -0.011 |
ile | -0.010 |
coon | -0.010 |
shark | -0.010 |
ball | -0.010 |
Token | are |
Token ID | 389 |
Feature activation | -5.68e-01 |
Pos logit contributions | |
shark | +0.031 |
Bad | +0.030 |
Shark | +0.029 |
ile | +0.028 |
coon | +0.028 |
Neg logit contributions | |
anna | -0.036 |
brushes | -0.036 |
bars | -0.035 |
raid | -0.034 |
types | -0.034 |
Token | not |
Token ID | 407 |
Feature activation | -3.75e-01 |
Pos logit contributions | |
Bad | +0.021 |
shark | +0.020 |
Shark | +0.020 |
coon | +0.019 |
ile | +0.019 |
Neg logit contributions | |
brushes | -0.022 |
anna | -0.021 |
types | -0.020 |
bars | -0.019 |
raid | -0.019 |
Token | |
Token ID | 220 |
Feature activation | +6.01e+00 |
Pos logit contributions | |
shark | +0.014 |
Bad | +0.014 |
Shark | +0.013 |
ile | +0.013 |
fin | +0.013 |
Neg logit contributions | |
anna | -0.014 |
brushes | -0.014 |
bars | -0.013 |
raid | -0.013 |
types | -0.013 |
Token | icky |
Token ID | 17479 |
Feature activation | -3.54e-02 |
Pos logit contributions | |
brushes | +0.290 |
anna | +0.281 |
raid | +0.275 |
types | +0.274 |
sights | +0.274 |
Neg logit contributions | |
Shark | -0.141 |
shark | -0.141 |
ile | -0.138 |
ful | -0.136 |
ball | -0.129 |
Token | and |
Token ID | 290 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
Bad | +0.001 |
shark | +0.001 |
Shark | +0.001 |
Fin | +0.001 |
hawk | +0.001 |
Neg logit contributions | |
anna | -0.001 |
brushes | -0.001 |
raid | -0.001 |
types | -0.001 |
bars | -0.001 |
Token | scary |
Token ID | 14343 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
brushes | +0.052 |
anna | +0.049 |
bars | +0.047 |
types | +0.047 |
raid | +0.047 |
Neg logit contributions | |
Bad | -0.050 |
shark | -0.050 |
coon | -0.048 |
Shark | -0.048 |
ile | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | +6.73e-01 |
Pos logit contributions | |
Bad | +0.080 |
shark | +0.076 |
Shark | +0.075 |
coon | +0.074 |
ile | +0.073 |
Neg logit contributions | |
anna | -0.069 |
brushes | -0.069 |
types | -0.065 |
raid | -0.065 |
bars | -0.062 |
Token | They |
Token ID | 1119 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
brushes | +0.018 |
anna | +0.018 |
bars | +0.016 |
raid | +0.016 |
types | +0.015 |
Neg logit contributions | |
shark | -0.032 |
Bad | -0.031 |
coon | -0.030 |
ile | -0.030 |
nowhere | -0.030 |
Token | bed |
Token ID | 3996 |
Feature activation | -1.99e-01 |
Pos logit contributions | |
brushes | +0.041 |
raid | +0.039 |
anna | +0.038 |
rings | +0.037 |
decorations | +0.037 |
Neg logit contributions | |
Bad | -0.047 |
shark | -0.043 |
Shark | -0.043 |
Fin | -0.040 |
coon | -0.040 |
Token | and |
Token ID | 290 |
Feature activation | +9.67e-01 |
Pos logit contributions | |
Bad | +0.008 |
shark | +0.008 |
Shark | +0.007 |
Fin | +0.007 |
fin | +0.007 |
Neg logit contributions | |
brushes | -0.007 |
anna | -0.007 |
bars | -0.007 |
raid | -0.007 |
decorations | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
anna | +0.038 |
brushes | +0.037 |
bars | +0.036 |
raid | +0.036 |
types | +0.035 |
Neg logit contributions | |
shark | -0.034 |
Bad | -0.033 |
Shark | -0.032 |
ile | -0.031 |
coon | -0.030 |
Token | perfect |
Token ID | 2818 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
brushes | +0.050 |
anna | +0.050 |
bars | +0.047 |
raid | +0.046 |
types | +0.046 |
Neg logit contributions | |
shark | -0.072 |
Bad | -0.071 |
Shark | -0.069 |
ile | -0.067 |
coon | -0.066 |
Token | bou |
Token ID | 35833 |
Feature activation | +4.96e+00 |
Pos logit contributions | |
anna | +0.045 |
brushes | +0.044 |
bars | +0.044 |
raid | +0.042 |
types | +0.041 |
Neg logit contributions | |
shark | -0.055 |
Bad | -0.053 |
ile | -0.052 |
Shark | -0.051 |
fin | -0.050 |
Token | quet |
Token ID | 21108 |
Feature activation | +5.97e+00 |
Pos logit contributions | |
brushes | +0.187 |
types | +0.185 |
bars | +0.182 |
anna | +0.182 |
nie | +0.176 |
Neg logit contributions | |
coon | -0.169 |
ile | -0.165 |
ball | -0.163 |
ful | -0.160 |
Bad | -0.152 |
Token | . |
Token ID | 13 |
Feature activation | +5.84e-01 |
Pos logit contributions | |
brushes | +0.212 |
anna | +0.209 |
bars | +0.200 |
raid | +0.199 |
types | +0.194 |
Neg logit contributions | |
Bad | -0.234 |
shark | -0.230 |
Shark | -0.221 |
ile | -0.211 |
coon | -0.210 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
anna | +0.024 |
brushes | +0.023 |
bars | +0.022 |
raid | +0.022 |
types | +0.022 |
Neg logit contributions | |
Bad | -0.020 |
shark | -0.020 |
Shark | -0.019 |
Fin | -0.018 |
Wally | -0.018 |
Token | L |
Token ID | 43 |
Feature activation | +7.11e-01 |
Pos logit contributions | |
Bad | +0.007 |
shark | +0.007 |
fin | +0.006 |
coon | +0.006 |
ile | +0.006 |
Neg logit contributions | |
brushes | -0.003 |
anna | -0.003 |
Along | -0.003 |
types | -0.003 |
bars | -0.003 |
Token | ily |
Token ID | 813 |
Feature activation | +9.48e-01 |
Pos logit contributions | |
brushes | +0.032 |
anna | +0.032 |
bars | +0.031 |
raid | +0.031 |
types | +0.030 |
Neg logit contributions | |
shark | -0.020 |
Bad | -0.019 |
ile | -0.019 |
Shark | -0.018 |
coon | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
brushes | +0.034 |
anna | +0.034 |
bars | +0.031 |
raid | +0.031 |
types | +0.031 |
Neg logit contributions | |
shark | -0.037 |
Bad | -0.036 |
Shark | -0.035 |
ile | -0.034 |
coon | -0.034 |
Token | did |
Token ID | 750 |
Feature activation | +2.99e-01 |
Pos logit contributions | |
shark | +0.003 |
Bad | +0.003 |
Shark | +0.003 |
ile | +0.003 |
coon | +0.003 |
Neg logit contributions | |
brushes | -0.002 |
anna | -0.002 |
bars | -0.002 |
types | -0.002 |
raid | -0.002 |
Token | not |
Token ID | 407 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
brushes | +0.013 |
anna | +0.013 |
bars | +0.012 |
raid | +0.012 |
types | +0.012 |
Neg logit contributions | |
shark | -0.009 |
Bad | -0.009 |
Shark | -0.009 |
ile | -0.009 |
coon | -0.008 |
Token | like |
Token ID | 588 |
Feature activation | -3.32e-01 |
Pos logit contributions | |
Bad | +0.042 |
ile | +0.041 |
shark | +0.041 |
coon | +0.039 |
Shark | +0.039 |
Neg logit contributions | |
bars | -0.041 |
anna | -0.040 |
brushes | -0.040 |
types | -0.038 |
Paula | -0.036 |
Token | colors |
Token ID | 7577 |
Feature activation | +3.43e+00 |
Pos logit contributions | |
shark | +0.014 |
Bad | +0.014 |
Shark | +0.013 |
ile | +0.013 |
coon | +0.013 |
Neg logit contributions | |
brushes | -0.011 |
anna | -0.011 |
bars | -0.010 |
raid | -0.010 |
types | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | +9.36e-01 |
Pos logit contributions | |
brushes | +0.133 |
bars | +0.131 |
anna | +0.127 |
raid | +0.122 |
types | +0.120 |
Neg logit contributions | |
Bad | -0.125 |
shark | -0.124 |
Shark | -0.123 |
fin | -0.117 |
Fin | -0.111 |
Token | brushes |
Token ID | 36377 |
Feature activation | +5.92e+00 |
Pos logit contributions | |
brushes | +0.028 |
anna | +0.026 |
bars | +0.025 |
types | +0.025 |
raid | +0.025 |
Neg logit contributions | |
shark | -0.042 |
Bad | -0.042 |
Shark | -0.041 |
ile | -0.040 |
coon | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | +5.84e-01 |
Pos logit contributions | |
bars | +0.188 |
brushes | +0.181 |
anna | +0.171 |
raid | +0.167 |
styles | +0.166 |
Neg logit contributions | |
Shark | -0.254 |
Bad | -0.250 |
shark | -0.250 |
coon | -0.243 |
ile | -0.240 |
Token |
|
Token ID | 198 |
Feature activation | -4.38e-01 |
Pos logit contributions | |
brushes | +0.015 |
anna | +0.014 |
bars | +0.013 |
raid | +0.013 |
types | +0.013 |
Neg logit contributions | |
shark | -0.029 |
ile | -0.028 |
Bad | -0.027 |
coon | -0.027 |
Shark | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | -4.83e-01 |
Pos logit contributions | |
shark | +0.018 |
Bad | +0.018 |
Shark | +0.017 |
ile | +0.017 |
fin | +0.017 |
Neg logit contributions | |
anna | -0.014 |
brushes | -0.014 |
bars | -0.013 |
types | -0.013 |
raid | -0.013 |
Token | One |
Token ID | 3198 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
shark | +0.017 |
Bad | +0.016 |
Shark | +0.016 |
ile | +0.016 |
fin | +0.016 |
Neg logit contributions | |
brushes | -0.019 |
anna | -0.019 |
bars | -0.018 |
raid | -0.018 |
types | -0.017 |
Token | day |
Token ID | 1110 |
Feature activation | -3.93e-01 |
Pos logit contributions | |
anna | +0.045 |
brushes | +0.045 |
bars | +0.042 |
raid | +0.041 |
types | +0.040 |
Neg logit contributions | |
shark | -0.084 |
Bad | -0.083 |
Shark | -0.081 |
ile | -0.080 |
coon | -0.078 |
Token | . |
Token ID | 13 |
Feature activation | +3.10e-01 |
Pos logit contributions | |
shark | +0.033 |
Bad | +0.033 |
Shark | +0.031 |
ile | +0.030 |
coon | +0.030 |
Neg logit contributions | |
anna | -0.035 |
brushes | -0.035 |
bars | -0.033 |
raid | -0.033 |
types | -0.032 |
Token | I |
Token ID | 314 |
Feature activation | -3.58e-01 |
Pos logit contributions | |
brushes | +0.011 |
anna | +0.011 |
types | +0.010 |
bars | +0.010 |
raid | +0.010 |
Neg logit contributions | |
Bad | -0.012 |
shark | -0.012 |
Shark | -0.012 |
Fin | -0.011 |
ile | -0.011 |
Token | don |
Token ID | 836 |
Feature activation | -5.94e-01 |
Pos logit contributions | |
shark | +0.015 |
Bad | +0.015 |
Shark | +0.014 |
ile | +0.014 |
coon | +0.014 |
Neg logit contributions | |
brushes | -0.012 |
anna | -0.012 |
bars | -0.011 |
raid | -0.011 |
types | -0.011 |
Token | 't |
Token ID | 470 |
Feature activation | -4.99e-01 |
Pos logit contributions | |
shark | +0.024 |
Bad | +0.023 |
Shark | +0.022 |
ile | +0.022 |
coon | +0.022 |
Neg logit contributions | |
brushes | -0.021 |
anna | -0.020 |
bars | -0.019 |
raid | -0.019 |
types | -0.019 |
Token | like |
Token ID | 588 |
Feature activation | -5.37e-02 |
Pos logit contributions | |
Bad | +0.019 |
shark | +0.018 |
ile | +0.018 |
Shark | +0.018 |
coon | +0.018 |
Neg logit contributions | |
bars | -0.018 |
brushes | -0.018 |
anna | -0.017 |
types | -0.017 |
Paula | -0.016 |
Token | brushes |
Token ID | 36377 |
Feature activation | +5.91e+00 |
Pos logit contributions | |
shark | +0.002 |
Bad | +0.002 |
Shark | +0.002 |
ile | +0.002 |
fin | +0.002 |
Neg logit contributions | |
anna | -0.002 |
brushes | -0.002 |
bars | -0.002 |
raid | -0.002 |
types | -0.002 |
Token | ." |
Token ID | 526 |
Feature activation | +9.32e-01 |
Pos logit contributions | |
anna | +0.202 |
brushes | +0.196 |
bars | +0.192 |
raid | +0.187 |
types | +0.181 |
Neg logit contributions | |
shark | -0.237 |
Bad | -0.237 |
Shark | -0.234 |
coon | -0.222 |
ile | -0.221 |
Token | |
Token ID | 220 |
Feature activation | +5.41e+00 |
Pos logit contributions | |
brushes | +0.026 |
anna | +0.026 |
bars | +0.025 |
raid | +0.024 |
types | +0.024 |
Neg logit contributions | |
shark | -0.043 |
Bad | -0.042 |
Shark | -0.042 |
Fin | -0.040 |
fin | -0.040 |
Token |
|
Token ID | 198 |
Feature activation | -2.89e-01 |
Pos logit contributions | |
brushes | +0.224 |
anna | +0.222 |
raid | +0.212 |
bars | +0.212 |
types | +0.209 |
Neg logit contributions | |
shark | -0.174 |
Bad | -0.173 |
Shark | -0.169 |
Fin | -0.160 |
ile | -0.160 |
Token |
|
Token ID | 198 |
Feature activation | -3.75e-01 |
Pos logit contributions | |
shark | +0.011 |
Bad | +0.011 |
ile | +0.011 |
Shark | +0.010 |
coon | +0.010 |
Neg logit contributions | |
anna | -0.010 |
brushes | -0.010 |
bars | -0.010 |
raid | -0.009 |
types | -0.009 |
Token | L |
Token ID | 43 |
Feature activation | +9.48e-01 |
Pos logit contributions | |
shark | +0.015 |
ile | +0.014 |
Shark | +0.013 |
Bad | +0.012 |
fin | +0.012 |
Neg logit contributions | |
anna | -0.014 |
brushes | -0.013 |
raid | -0.012 |
bars | -0.012 |
select | -0.012 |
Token | ier |
Token ID | 959 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
brushes | +0.219 |
anna | +0.217 |
bars | +0.206 |
raid | +0.202 |
types | +0.200 |
Neg logit contributions | |
shark | -0.202 |
Bad | -0.197 |
ile | -0.190 |
Shark | -0.189 |
coon | -0.183 |
Token | ? |
Token ID | 30 |
Feature activation | +8.07e-01 |
Pos logit contributions | |
anna | +0.046 |
brushes | +0.045 |
types | +0.043 |
raid | +0.043 |
bars | +0.042 |
Neg logit contributions | |
shark | -0.049 |
Bad | -0.049 |
Shark | -0.047 |
coon | -0.045 |
fin | -0.044 |
Token | You |
Token ID | 921 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
anna | +0.031 |
brushes | +0.030 |
types | +0.029 |
bars | +0.029 |
sights | +0.028 |
Neg logit contributions | |
Bad | -0.029 |
Shark | -0.028 |
shark | -0.027 |
Fin | -0.027 |
Wally | -0.026 |
Token | are |
Token ID | 389 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
shark | +0.013 |
Bad | +0.013 |
Shark | +0.012 |
ile | +0.012 |
coon | +0.012 |
Neg logit contributions | |
anna | -0.009 |
brushes | -0.009 |
bars | -0.009 |
raid | -0.008 |
types | -0.008 |
Token | y |
Token ID | 331 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
shark | +0.007 |
Bad | +0.007 |
Shark | +0.007 |
ile | +0.007 |
coon | +0.006 |
Neg logit contributions | |
anna | -0.008 |
brushes | -0.008 |
bars | -0.007 |
raid | -0.007 |
types | -0.007 |
Token | umm |
Token ID | 13929 |
Feature activation | +5.90e+00 |
Pos logit contributions | |
anna | +0.054 |
brushes | +0.054 |
bars | +0.050 |
types | +0.048 |
raid | +0.048 |
Neg logit contributions | |
shark | -0.075 |
Bad | -0.073 |
Shark | -0.072 |
ile | -0.071 |
fin | -0.069 |
Token | ier |
Token ID | 959 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
brushes | +0.233 |
anna | +0.229 |
bars | +0.218 |
types | +0.213 |
raid | +0.212 |
Neg logit contributions | |
shark | -0.208 |
Bad | -0.201 |
ile | -0.195 |
Shark | -0.194 |
coon | -0.188 |
Token | . |
Token ID | 13 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
anna | +0.049 |
brushes | +0.047 |
raid | +0.045 |
types | +0.045 |
bars | +0.043 |
Neg logit contributions | |
Bad | -0.052 |
shark | -0.052 |
Shark | -0.050 |
coon | -0.049 |
ile | -0.049 |
Token | You |
Token ID | 921 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
anna | +0.041 |
brushes | +0.040 |
bars | +0.040 |
types | +0.038 |
raid | +0.037 |
Neg logit contributions | |
Bad | -0.042 |
Shark | -0.041 |
shark | -0.040 |
Fin | -0.039 |
ile | -0.039 |
Token | are |
Token ID | 389 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
shark | +0.005 |
Bad | +0.005 |
Shark | +0.005 |
ile | +0.005 |
coon | +0.004 |
Neg logit contributions | |
anna | -0.004 |
brushes | -0.004 |
bars | -0.004 |
raid | -0.004 |
types | -0.004 |
Token | my |
Token ID | 616 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
shark | +0.001 |
Bad | +0.001 |
Shark | +0.001 |
ile | +0.001 |
coon | +0.001 |
Neg logit contributions | |
anna | -0.001 |
brushes | -0.001 |
bars | -0.001 |
raid | -0.001 |
types | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | +8.13e-02 |
Pos logit contributions | |
brushes | +0.009 |
anna | +0.009 |
bars | +0.009 |
raid | +0.009 |
types | +0.009 |
Neg logit contributions | |
Bad | -0.010 |
shark | -0.010 |
Shark | -0.009 |
coon | -0.009 |
ile | -0.009 |
Token | it |
Token ID | 340 |
Feature activation | +5.74e-01 |
Pos logit contributions | |
brushes | +0.003 |
anna | +0.003 |
bars | +0.003 |
types | +0.003 |
raid | +0.003 |
Neg logit contributions | |
shark | -0.003 |
Bad | -0.003 |
Shark | -0.003 |
ile | -0.003 |
coon | -0.003 |
Token | became |
Token ID | 2627 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
anna | +0.022 |
brushes | +0.022 |
raid | +0.021 |
bars | +0.021 |
types | +0.020 |
Neg logit contributions | |
shark | -0.021 |
Bad | -0.020 |
Shark | -0.019 |
ile | -0.019 |
coon | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
brushes | +0.004 |
anna | +0.004 |
bars | +0.004 |
raid | +0.004 |
types | +0.004 |
Neg logit contributions | |
shark | -0.004 |
Bad | -0.004 |
Shark | -0.004 |
ile | -0.004 |
coon | -0.004 |
Token | sh |
Token ID | 427 |
Feature activation | -2.97e-01 |
Pos logit contributions | |
brushes | +0.042 |
anna | +0.041 |
bars | +0.039 |
raid | +0.038 |
types | +0.038 |
Neg logit contributions | |
shark | -0.062 |
Bad | -0.061 |
Shark | -0.059 |
ile | -0.058 |
coon | -0.057 |
Token | ini |
Token ID | 5362 |
Feature activation | +5.85e+00 |
Pos logit contributions | |
ile | +0.010 |
shark | +0.010 |
Bad | +0.010 |
coon | +0.010 |
Shark | +0.010 |
Neg logit contributions | |
brushes | -0.012 |
anna | -0.011 |
bars | -0.011 |
types | -0.011 |
raid | -0.011 |
Token | est |
Token ID | 395 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
brushes | +0.160 |
anna | +0.148 |
Paula | +0.140 |
bars | +0.138 |
raid | +0.138 |
Neg logit contributions | |
ile | -0.278 |
shark | -0.275 |
Bad | -0.273 |
coon | -0.264 |
Shark | -0.255 |
Token | tower |
Token ID | 10580 |
Feature activation | +8.91e-01 |
Pos logit contributions | |
brushes | +0.050 |
anna | +0.050 |
bars | +0.047 |
raid | +0.046 |
types | +0.046 |
Neg logit contributions | |
shark | -0.050 |
Bad | -0.050 |
Shark | -0.048 |
ile | -0.047 |
coon | -0.046 |
Token | in |
Token ID | 287 |
Feature activation | +8.73e-01 |
Pos logit contributions | |
brushes | +0.036 |
anna | +0.035 |
bars | +0.034 |
raid | +0.033 |
types | +0.033 |
Neg logit contributions | |
Bad | -0.031 |
shark | -0.030 |
Shark | -0.029 |
ile | -0.028 |
coon | -0.028 |
Token | the |
Token ID | 262 |
Feature activation | +2.97e-01 |
Pos logit contributions | |
Paula | +0.027 |
brushes | +0.026 |
types | +0.025 |
decorations | +0.024 |
work | +0.024 |
Neg logit contributions | |
coon | -0.037 |
Bad | -0.036 |
ile | -0.036 |
shark | -0.034 |
Shark | -0.034 |
Token | whole |
Token ID | 2187 |
Feature activation | -2.98e-01 |
Pos logit contributions | |
anna | +0.010 |
brushes | +0.009 |
raid | +0.009 |
bars | +0.009 |
types | +0.009 |
Neg logit contributions | |
shark | -0.013 |
Bad | -0.013 |
Shark | -0.012 |
ile | -0.012 |
fin | -0.011 |
Token | spark |
Token ID | 9009 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
anna | +0.033 |
brushes | +0.032 |
raid | +0.031 |
bars | +0.030 |
types | +0.030 |
Neg logit contributions | |
shark | -0.033 |
Bad | -0.031 |
Shark | -0.030 |
fin | -0.029 |
ile | -0.029 |
Token | le |
Token ID | 293 |
Feature activation | -6.43e-01 |
Pos logit contributions | |
brushes | +0.060 |
anna | +0.059 |
bars | +0.056 |
raid | +0.056 |
types | +0.055 |
Neg logit contributions | |
shark | -0.051 |
Bad | -0.050 |
Shark | -0.048 |
ile | -0.047 |
coon | -0.046 |
Token | like |
Token ID | 588 |
Feature activation | -5.86e-01 |
Pos logit contributions | |
Bad | +0.026 |
shark | +0.025 |
Shark | +0.025 |
coon | +0.024 |
Fin | +0.023 |
Neg logit contributions | |
anna | -0.022 |
brushes | -0.021 |
types | -0.020 |
nie | -0.019 |
raid | -0.019 |
Token | theirs |
Token ID | 22021 |
Feature activation | -3.28e-01 |
Pos logit contributions | |
Bad | +0.021 |
shark | +0.021 |
Shark | +0.021 |
ile | +0.020 |
coon | +0.020 |
Neg logit contributions | |
brushes | -0.022 |
anna | -0.021 |
bars | -0.020 |
types | -0.020 |
raid | -0.019 |
Token | ." |
Token ID | 526 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
Bad | +0.013 |
shark | +0.013 |
Shark | +0.012 |
coon | +0.012 |
fin | +0.012 |
Neg logit contributions | |
anna | -0.011 |
brushes | -0.011 |
bars | -0.011 |
types | -0.010 |
raid | -0.010 |
Token | |
Token ID | 220 |
Feature activation | +5.84e+00 |
Pos logit contributions | |
anna | +0.041 |
brushes | +0.040 |
bars | +0.037 |
raid | +0.037 |
types | +0.037 |
Neg logit contributions | |
shark | -0.046 |
Bad | -0.045 |
ile | -0.044 |
coon | -0.043 |
Shark | -0.043 |
Token |
|
Token ID | 198 |
Feature activation | +2.31e-02 |
Pos logit contributions | |
brushes | +0.225 |
anna | +0.225 |
bars | +0.212 |
raid | +0.210 |
types | +0.207 |
Neg logit contributions | |
shark | -0.208 |
Bad | -0.205 |
Shark | -0.197 |
ile | -0.192 |
coon | -0.189 |
Token |
|
Token ID | 198 |
Feature activation | -7.87e-02 |
Pos logit contributions | |
anna | +0.001 |
brushes | +0.001 |
Paula | +0.001 |
bars | +0.001 |
types | +0.001 |
Neg logit contributions | |
shark | -0.001 |
Bad | -0.001 |
ile | -0.001 |
fin | -0.001 |
Shark | -0.001 |
Token | The |
Token ID | 464 |
Feature activation | +6.31e-01 |
Pos logit contributions | |
shark | +0.003 |
fin | +0.003 |
ile | +0.003 |
Shark | +0.003 |
Bad | +0.003 |
Neg logit contributions | |
anna | -0.003 |
brushes | -0.003 |
bars | -0.003 |
Paula | -0.003 |
raid | -0.003 |
Token | daughter |
Token ID | 4957 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
brushes | +0.021 |
anna | +0.020 |
types | +0.019 |
bars | +0.019 |
raid | +0.018 |
Neg logit contributions | |
shark | -0.026 |
Bad | -0.025 |
Shark | -0.025 |
ile | -0.024 |
fin | -0.024 |
Token | started |
Token ID | 2067 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
brushes | +0.056 |
anna | +0.056 |
bars | +0.054 |
raid | +0.054 |
types | +0.052 |
Neg logit contributions | |
shark | -0.055 |
Bad | -0.054 |
Shark | -0.052 |
ile | -0.051 |
coon | -0.050 |
Token | know |
Token ID | 760 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
Bad | +0.015 |
shark | +0.015 |
ile | +0.015 |
Shark | +0.014 |
fin | +0.014 |
Neg logit contributions | |
bars | -0.014 |
anna | -0.014 |
brushes | -0.014 |
types | -0.013 |
raid | -0.013 |
Token | what |
Token ID | 644 |
Feature activation | +8.41e-01 |
Pos logit contributions | |
brushes | +0.062 |
anna | +0.059 |
bars | +0.058 |
raid | +0.056 |
types | +0.055 |
Neg logit contributions | |
shark | -0.064 |
Bad | -0.063 |
Shark | -0.061 |
ile | -0.060 |
coon | -0.059 |
Token | ind |
Token ID | 773 |
Feature activation | +2.98e+00 |
Pos logit contributions | |
anna | +0.033 |
brushes | +0.032 |
bars | +0.031 |
raid | +0.030 |
types | +0.030 |
Neg logit contributions | |
shark | -0.030 |
Bad | -0.029 |
Shark | -0.028 |
ile | -0.027 |
coon | -0.026 |
Token | igo |
Token ID | 14031 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
anna | +0.126 |
brushes | +0.122 |
bars | +0.119 |
raid | +0.117 |
Along | +0.116 |
Neg logit contributions | |
Bad | -0.087 |
ile | -0.085 |
shark | -0.084 |
Shark | -0.081 |
Fin | -0.078 |
Token | and |
Token ID | 290 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
anna | +0.065 |
brushes | +0.065 |
bars | +0.062 |
raid | +0.061 |
types | +0.060 |
Neg logit contributions | |
shark | -0.059 |
Bad | -0.058 |
Shark | -0.055 |
ile | -0.055 |
coon | -0.053 |
Token | violet |
Token ID | 45725 |
Feature activation | +5.83e+00 |
Pos logit contributions | |
anna | +0.007 |
bars | +0.007 |
raid | +0.007 |
cere | +0.006 |
types | +0.006 |
Neg logit contributions | |
Bad | -0.008 |
Shark | -0.008 |
Fin | -0.007 |
shark | -0.007 |
Wally | -0.007 |
Token | are |
Token ID | 389 |
Feature activation | +3.15e-01 |
Pos logit contributions | |
anna | +0.246 |
brushes | +0.239 |
bars | +0.234 |
raid | +0.229 |
types | +0.226 |
Neg logit contributions | |
shark | -0.177 |
ile | -0.172 |
Bad | -0.169 |
Shark | -0.167 |
coon | -0.156 |
Token | ?" |
Token ID | 1701 |
Feature activation | +7.88e-01 |
Pos logit contributions | |
anna | +0.012 |
bars | +0.012 |
brushes | +0.012 |
raid | +0.012 |
types | +0.011 |
Neg logit contributions | |
Bad | -0.011 |
shark | -0.011 |
ile | -0.010 |
Shark | -0.010 |
fin | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | -8.71e-02 |
Pos logit contributions | |
anna | +0.033 |
brushes | +0.032 |
bars | +0.031 |
types | +0.030 |
raid | +0.030 |
Neg logit contributions | |
shark | -0.026 |
Bad | -0.026 |
Shark | -0.025 |
ile | -0.024 |
Fin | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
shark | +0.003 |
Bad | +0.003 |
Shark | +0.003 |
ile | +0.003 |
coon | +0.003 |
Neg logit contributions | |
brushes | -0.003 |
anna | -0.003 |
bars | -0.003 |
raid | -0.003 |
types | -0.003 |
Token | L |
Token ID | 43 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
Bad | +0.007 |
shark | +0.007 |
Shark | +0.007 |
ile | +0.006 |
coon | +0.006 |
Neg logit contributions | |
anna | -0.010 |
brushes | -0.010 |
bars | -0.010 |
raid | -0.010 |
types | -0.010 |
Token | make |
Token ID | 787 |
Feature activation | +4.65e-01 |
Pos logit contributions | |
anna | +0.063 |
bars | +0.058 |
types | +0.055 |
brushes | +0.052 |
sights | +0.052 |
Neg logit contributions | |
shark | -0.082 |
Bad | -0.078 |
fin | -0.077 |
Shark | -0.076 |
ile | -0.076 |
Token | pictures |
Token ID | 5986 |
Feature activation | +2.55e+00 |
Pos logit contributions | |
anna | +0.017 |
bars | +0.016 |
raid | +0.016 |
Along | +0.015 |
nie | +0.015 |
Neg logit contributions | |
shark | -0.017 |
Bad | -0.017 |
Shark | -0.016 |
ile | -0.016 |
fin | -0.015 |
Token | with |
Token ID | 351 |
Feature activation | -9.09e-01 |
Pos logit contributions | |
anna | +0.137 |
nie | +0.124 |
bars | +0.124 |
raid | +0.122 |
brushes | +0.120 |
Neg logit contributions | |
shark | -0.052 |
ile | -0.051 |
coon | -0.047 |
Bad | -0.046 |
ful | -0.046 |
Token | colors |
Token ID | 7577 |
Feature activation | +2.95e+00 |
Pos logit contributions | |
shark | +0.039 |
Bad | +0.038 |
Shark | +0.036 |
ile | +0.036 |
fin | +0.035 |
Neg logit contributions | |
anna | -0.030 |
bars | -0.027 |
raid | -0.026 |
brushes | -0.025 |
types | -0.025 |
Token | and |
Token ID | 290 |
Feature activation | +8.52e-01 |
Pos logit contributions | |
anna | +0.147 |
brushes | +0.136 |
raid | +0.136 |
bars | +0.135 |
types | +0.133 |
Neg logit contributions | |
shark | -0.070 |
ile | -0.066 |
Shark | -0.061 |
Bad | -0.061 |
ful | -0.060 |
Token | brushes |
Token ID | 36377 |
Feature activation | +5.82e+00 |
Pos logit contributions | |
anna | +0.030 |
annah | +0.026 |
bars | +0.025 |
Along | +0.025 |
nie | +0.024 |
Neg logit contributions | |
shark | -0.033 |
Bad | -0.033 |
fin | -0.032 |
Shark | -0.031 |
ful | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | +8.13e-02 |
Pos logit contributions | |
anna | +0.278 |
brushes | +0.265 |
bars | +0.258 |
raid | +0.257 |
types | +0.256 |
Neg logit contributions | |
shark | -0.150 |
Bad | -0.141 |
Shark | -0.135 |
ile | -0.135 |
fin | -0.131 |
Token | She |
Token ID | 1375 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
anna | +0.002 |
bars | +0.002 |
ilda | +0.002 |
types | +0.002 |
brushes | +0.002 |
Neg logit contributions | |
shark | -0.003 |
Fin | -0.003 |
Bad | -0.003 |
Wally | -0.003 |
Shark | -0.003 |
Token | had |
Token ID | 550 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
anna | +0.072 |
bars | +0.068 |
brushes | +0.068 |
raid | +0.067 |
types | +0.066 |
Neg logit contributions | |
shark | -0.058 |
Bad | -0.057 |
ile | -0.057 |
Shark | -0.055 |
coon | -0.053 |
Token | many |
Token ID | 867 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
nowhere | +0.008 |
Bad | +0.008 |
ful | +0.007 |
shark | +0.007 |
Rex | +0.006 |
Neg logit contributions | |
anna | -0.018 |
raid | -0.017 |
bars | -0.016 |
rell | -0.016 |
types | -0.015 |
Token | colors |
Token ID | 7577 |
Feature activation | +2.65e+00 |
Pos logit contributions | |
shark | +0.004 |
nowhere | +0.004 |
Bad | +0.004 |
guns | +0.004 |
ile | +0.004 |
Neg logit contributions | |
anna | -0.004 |
bars | -0.004 |
Along | -0.004 |
read | -0.003 |
rolling | -0.003 |
Token | make |
Token ID | 787 |
Feature activation | +4.53e-01 |
Pos logit contributions | |
anna | +0.062 |
bars | +0.056 |
types | +0.053 |
brushes | +0.051 |
Paula | +0.051 |
Neg logit contributions | |
shark | -0.083 |
Bad | -0.079 |
fin | -0.078 |
ile | -0.078 |
Shark | -0.077 |
Token | pictures |
Token ID | 5986 |
Feature activation | +2.46e+00 |
Pos logit contributions | |
anna | +0.017 |
bars | +0.016 |
raid | +0.015 |
Along | +0.014 |
nie | +0.014 |
Neg logit contributions | |
shark | -0.018 |
Bad | -0.017 |
Shark | -0.017 |
fin | -0.016 |
ile | -0.016 |
Token | with |
Token ID | 351 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
anna | +0.111 |
brushes | +0.105 |
bars | +0.102 |
raid | +0.102 |
types | +0.100 |
Neg logit contributions | |
shark | -0.074 |
Bad | -0.070 |
ile | -0.068 |
Shark | -0.068 |
coon | -0.066 |
Token | colors |
Token ID | 7577 |
Feature activation | +2.78e+00 |
Pos logit contributions | |
shark | +0.038 |
Bad | +0.038 |
Shark | +0.036 |
ile | +0.035 |
coon | +0.035 |
Neg logit contributions | |
anna | -0.037 |
brushes | -0.036 |
bars | -0.035 |
raid | -0.034 |
types | -0.034 |
Token | and |
Token ID | 290 |
Feature activation | +8.27e-01 |
Pos logit contributions | |
brushes | +0.118 |
anna | +0.117 |
bars | +0.112 |
raid | +0.111 |
types | +0.109 |
Neg logit contributions | |
shark | -0.089 |
Bad | -0.087 |
Shark | -0.084 |
ile | -0.081 |
coon | -0.080 |
Token | brushes |
Token ID | 36377 |
Feature activation | +5.82e+00 |
Pos logit contributions | |
anna | +0.027 |
bars | +0.022 |
Along | +0.022 |
raid | +0.022 |
annah | +0.022 |
Neg logit contributions | |
shark | -0.036 |
Bad | -0.034 |
fin | -0.034 |
Shark | -0.033 |
Fin | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
brushes | +0.229 |
anna | +0.224 |
bars | +0.217 |
raid | +0.213 |
types | +0.210 |
Neg logit contributions | |
Bad | -0.206 |
shark | -0.206 |
Shark | -0.196 |
ile | -0.192 |
coon | -0.190 |
Token | He |
Token ID | 679 |
Feature activation | -2.67e-01 |
Pos logit contributions | |
shark | +0.007 |
Fin | +0.007 |
Bad | +0.007 |
Wally | +0.007 |
Shark | +0.007 |
Neg logit contributions | |
anna | -0.006 |
raid | -0.005 |
brushes | -0.005 |
ilda | -0.005 |
rings | -0.005 |
Token | wanted |
Token ID | 2227 |
Feature activation | -4.97e-01 |
Pos logit contributions | |
shark | +0.011 |
Bad | +0.010 |
Shark | +0.010 |
ile | +0.010 |
coon | +0.010 |
Neg logit contributions | |
anna | -0.009 |
brushes | -0.009 |
bars | -0.009 |
raid | -0.009 |
types | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | +4.73e-01 |
Pos logit contributions | |
shark | +0.023 |
Bad | +0.022 |
Shark | +0.022 |
ile | +0.021 |
fin | +0.021 |
Neg logit contributions | |
anna | -0.015 |
brushes | -0.014 |
bars | -0.013 |
raid | -0.013 |
types | -0.013 |
Token | show |
Token ID | 905 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
anna | +0.016 |
bars | +0.015 |
types | +0.014 |
brushes | +0.014 |
nie | +0.014 |
Neg logit contributions | |
shark | -0.019 |
Bad | -0.019 |
Shark | -0.018 |
fin | -0.018 |
ile | -0.018 |
Token | we |
Token ID | 356 |
Feature activation | -1.58e-01 |
Pos logit contributions | |
brushes | +0.063 |
anna | +0.063 |
bars | +0.059 |
raid | +0.058 |
types | +0.058 |
Neg logit contributions | |
shark | -0.064 |
Bad | -0.064 |
Shark | -0.061 |
ile | -0.060 |
coon | -0.059 |
Token | do |
Token ID | 466 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
shark | +0.007 |
Bad | +0.007 |
Shark | +0.007 |
ile | +0.007 |
coon | +0.007 |
Neg logit contributions | |
brushes | -0.005 |
anna | -0.005 |
raid | -0.004 |
bars | -0.004 |
types | -0.004 |
Token | something |
Token ID | 1223 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
shark | +0.056 |
Bad | +0.055 |
Shark | +0.053 |
ile | +0.051 |
coon | +0.050 |
Neg logit contributions | |
brushes | -0.066 |
anna | -0.066 |
bars | -0.063 |
raid | -0.062 |
types | -0.062 |
Token | wrong |
Token ID | 2642 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
shark | +0.007 |
Bad | +0.006 |
Shark | +0.006 |
ile | +0.006 |
coon | +0.006 |
Neg logit contributions | |
brushes | -0.011 |
anna | -0.011 |
bars | -0.011 |
raid | -0.010 |
types | -0.010 |
Token | ." |
Token ID | 526 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
shark | +0.107 |
Bad | +0.106 |
Shark | +0.103 |
coon | +0.101 |
ile | +0.099 |
Neg logit contributions | |
anna | -0.076 |
brushes | -0.073 |
types | -0.070 |
raid | -0.068 |
bars | -0.068 |
Token | |
Token ID | 220 |
Feature activation | +5.82e+00 |
Pos logit contributions | |
anna | +0.043 |
brushes | +0.041 |
bars | +0.038 |
raid | +0.037 |
Along | +0.037 |
Neg logit contributions | |
shark | -0.056 |
Bad | -0.056 |
ile | -0.055 |
coon | -0.053 |
Shark | -0.051 |
Token |
|
Token ID | 198 |
Feature activation | +9.97e-02 |
Pos logit contributions | |
anna | +0.222 |
brushes | +0.222 |
bars | +0.209 |
raid | +0.207 |
types | +0.204 |
Neg logit contributions | |
shark | -0.210 |
Bad | -0.207 |
Shark | -0.198 |
ile | -0.194 |
coon | -0.190 |
Token |
|
Token ID | 198 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
anna | +0.003 |
brushes | +0.002 |
Along | +0.002 |
Paula | +0.002 |
Eve | +0.002 |
Neg logit contributions | |
shark | -0.005 |
ile | -0.005 |
Bad | -0.005 |
fin | -0.005 |
bomb | -0.005 |
Token | Sun |
Token ID | 16012 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
shark | +0.001 |
fin | +0.001 |
ile | +0.001 |
bomb | +0.001 |
sharks | +0.001 |
Neg logit contributions | |
anna | -0.001 |
nie | -0.001 |
Along | -0.001 |
three | -0.001 |
First | -0.001 |
Token | ny |
Token ID | 3281 |
Feature activation | +5.97e-01 |
Pos logit contributions | |
anna | +0.070 |
brushes | +0.066 |
raid | +0.063 |
types | +0.062 |
bars | +0.061 |
Neg logit contributions | |
shark | -0.091 |
Bad | -0.089 |
Shark | -0.085 |
ile | -0.084 |
coon | -0.084 |
Token | was |
Token ID | 373 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
brushes | +0.022 |
anna | +0.022 |
bars | +0.021 |
raid | +0.020 |
types | +0.020 |
Neg logit contributions | |
shark | -0.022 |
Bad | -0.022 |
Shark | -0.021 |
ile | -0.021 |
coon | -0.020 |
Token | make |
Token ID | 787 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
brushes | +0.040 |
anna | +0.040 |
bars | +0.038 |
types | +0.036 |
raid | +0.036 |
Neg logit contributions | |
shark | -0.056 |
Bad | -0.055 |
Shark | -0.053 |
ile | -0.052 |
fin | -0.051 |
Token | a |
Token ID | 257 |
Feature activation | +5.74e-01 |
Pos logit contributions | |
brushes | +0.037 |
anna | +0.037 |
bars | +0.035 |
raid | +0.035 |
types | +0.035 |
Neg logit contributions | |
shark | -0.037 |
Bad | -0.037 |
Shark | -0.035 |
ile | -0.035 |
coon | -0.034 |
Token | difference |
Token ID | 3580 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
brushes | +0.017 |
anna | +0.016 |
raid | +0.015 |
types | +0.015 |
bars | +0.015 |
Neg logit contributions | |
Bad | -0.026 |
shark | -0.025 |
ile | -0.025 |
Shark | -0.025 |
coon | -0.024 |
Token | today |
Token ID | 1909 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
anna | +0.031 |
nie | +0.029 |
types | +0.028 |
brushes | +0.027 |
raid | +0.026 |
Neg logit contributions | |
Bad | -0.046 |
Shark | -0.044 |
shark | -0.044 |
Fin | -0.044 |
fin | -0.043 |
Token | ." |
Token ID | 526 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
anna | +0.004 |
brushes | +0.004 |
bars | +0.004 |
raid | +0.004 |
types | +0.004 |
Neg logit contributions | |
shark | -0.005 |
Bad | -0.005 |
Shark | -0.005 |
ile | -0.005 |
coon | -0.005 |
Token | |
Token ID | 220 |
Feature activation | +5.81e+00 |
Pos logit contributions | |
brushes | +0.041 |
raid | +0.039 |
epad | +0.038 |
bars | +0.038 |
types | +0.038 |
Neg logit contributions | |
Shark | -0.036 |
Fin | -0.035 |
Bad | -0.034 |
shark | -0.034 |
Wally | -0.033 |
Token |
|
Token ID | 198 |
Feature activation | +4.68e-02 |
Pos logit contributions | |
brushes | +0.243 |
anna | +0.239 |
raid | +0.230 |
bars | +0.230 |
types | +0.226 |
Neg logit contributions | |
shark | -0.187 |
Bad | -0.185 |
Shark | -0.182 |
ile | -0.171 |
Fin | -0.171 |
Token |
|
Token ID | 198 |
Feature activation | -5.16e-02 |
Pos logit contributions | |
brushes | +0.002 |
anna | +0.002 |
bars | +0.002 |
raid | +0.002 |
types | +0.002 |
Neg logit contributions | |
Bad | -0.001 |
shark | -0.001 |
Shark | -0.001 |
ile | -0.001 |
Fin | -0.001 |
Token | E |
Token ID | 36 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
Bad | +0.002 |
shark | +0.002 |
Shark | +0.002 |
ile | +0.001 |
coon | +0.001 |
Neg logit contributions | |
brushes | -0.002 |
anna | -0.002 |
bars | -0.002 |
raid | -0.002 |
types | -0.002 |
Token | lla |
Token ID | 8466 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
anna | +0.051 |
brushes | +0.048 |
raid | +0.044 |
types | +0.044 |
nie | +0.043 |
Neg logit contributions | |
shark | -0.083 |
Shark | -0.078 |
Bad | -0.078 |
fin | -0.077 |
Fin | -0.075 |
Token | smiled |
Token ID | 13541 |
Feature activation | +1.85e+00 |
Pos logit contributions | |
brushes | +0.053 |
anna | +0.053 |
bars | +0.051 |
raid | +0.050 |
types | +0.049 |
Neg logit contributions | |
shark | -0.053 |
Bad | -0.052 |
Shark | -0.050 |
ile | -0.049 |
coon | -0.048 |
Token | seem |
Token ID | 1283 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
shark | +0.055 |
Bad | +0.053 |
Shark | +0.052 |
ile | +0.051 |
fin | +0.050 |
Neg logit contributions | |
anna | -0.046 |
brushes | -0.044 |
bars | -0.043 |
types | -0.040 |
raid | -0.040 |
Token | so |
Token ID | 523 |
Feature activation | +2.37e+00 |
Pos logit contributions | |
anna | +0.090 |
brushes | +0.090 |
bars | +0.087 |
raid | +0.084 |
types | +0.082 |
Neg logit contributions | |
shark | -0.083 |
Bad | -0.081 |
Shark | -0.078 |
fin | -0.075 |
ile | -0.074 |
Token | y |
Token ID | 331 |
Feature activation | +2.37e+00 |
Pos logit contributions | |
anna | +0.106 |
bars | +0.102 |
brushes | +0.102 |
raid | +0.098 |
Paula | +0.096 |
Neg logit contributions | |
shark | -0.076 |
Bad | -0.073 |
Shark | -0.070 |
ile | -0.068 |
Fin | -0.067 |
Token | ucky |
Token ID | 5309 |
Feature activation | -3.65e-01 |
Pos logit contributions | |
brushes | +0.069 |
anna | +0.068 |
bars | +0.066 |
raid | +0.066 |
types | +0.063 |
Neg logit contributions | |
Bad | -0.108 |
shark | -0.106 |
Shark | -0.102 |
coon | -0.101 |
ile | -0.100 |
Token | and |
Token ID | 290 |
Feature activation | +9.83e-01 |
Pos logit contributions | |
Bad | +0.017 |
Shark | +0.016 |
shark | +0.016 |
hawk | +0.015 |
fin | +0.015 |
Neg logit contributions | |
anna | -0.010 |
raid | -0.009 |
brushes | -0.009 |
types | -0.009 |
work | -0.009 |
Token | |
Token ID | 220 |
Feature activation | +5.81e+00 |
Pos logit contributions | |
brushes | +0.036 |
anna | +0.034 |
bars | +0.034 |
types | +0.033 |
raid | +0.032 |
Neg logit contributions | |
shark | -0.037 |
Bad | -0.036 |
Shark | -0.035 |
coon | -0.035 |
ile | -0.035 |
Token | icky |
Token ID | 17479 |
Feature activation | +3.21e-01 |
Pos logit contributions | |
brushes | +0.249 |
anna | +0.249 |
raid | +0.240 |
types | +0.235 |
bars | +0.226 |
Neg logit contributions | |
shark | -0.169 |
Shark | -0.169 |
Bad | -0.168 |
ile | -0.161 |
Fin | -0.160 |
Token | anymore |
Token ID | 7471 |
Feature activation | -3.55e+00 |
Pos logit contributions | |
anna | +0.011 |
brushes | +0.011 |
bars | +0.010 |
raid | +0.010 |
types | +0.010 |
Neg logit contributions | |
Bad | -0.013 |
shark | -0.013 |
Shark | -0.012 |
coon | -0.012 |
hawk | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +4.35e-01 |
Pos logit contributions | |
Bad | +0.152 |
ile | +0.144 |
shark | +0.143 |
coon | +0.142 |
Shark | +0.139 |
Neg logit contributions | |
anna | -0.109 |
brushes | -0.107 |
raid | -0.099 |
types | -0.098 |
bars | -0.097 |
Token | He |
Token ID | 679 |
Feature activation | +2.76e-01 |
Pos logit contributions | |
anna | +0.018 |
brushes | +0.017 |
bars | +0.017 |
raid | +0.017 |
types | +0.016 |
Neg logit contributions | |
shark | -0.015 |
Bad | -0.015 |
Shark | -0.014 |
Fin | -0.014 |
ile | -0.014 |
Token | said |
Token ID | 531 |
Feature activation | +6.54e-01 |
Pos logit contributions | |
brushes | +0.007 |
anna | +0.007 |
types | +0.006 |
nie | +0.006 |
bars | +0.006 |
Neg logit contributions | |
Bad | -0.013 |
shark | -0.013 |
coon | -0.012 |
afloat | -0.012 |
ile | -0.012 |
Token | grass |
Token ID | 8701 |
Feature activation | +5.20e-01 |
Pos logit contributions | |
brushes | +0.018 |
anna | +0.017 |
raid | +0.016 |
bars | +0.016 |
types | +0.015 |
Neg logit contributions | |
shark | -0.025 |
Bad | -0.024 |
Shark | -0.024 |
ile | -0.023 |
coon | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -5.55e-01 |
Pos logit contributions | |
bars | +0.016 |
brushes | +0.016 |
types | +0.015 |
raid | +0.014 |
anna | +0.014 |
Neg logit contributions | |
Bad | -0.024 |
shark | -0.022 |
Shark | -0.022 |
coon | -0.021 |
fin | -0.020 |
Token | She |
Token ID | 1375 |
Feature activation | +9.99e-01 |
Pos logit contributions | |
Fin | +0.020 |
Wally | +0.020 |
Rex | +0.020 |
shark | +0.019 |
Shark | +0.019 |
Neg logit contributions | |
anna | -0.020 |
inations | -0.020 |
nie | -0.019 |
wn | -0.019 |
uns | -0.018 |
Token | asked |
Token ID | 1965 |
Feature activation | +8.48e-01 |
Pos logit contributions | |
bars | +0.034 |
brushes | +0.034 |
anna | +0.034 |
raid | +0.034 |
types | +0.032 |
Neg logit contributions | |
shark | -0.039 |
Bad | -0.039 |
Shark | -0.038 |
Fin | -0.036 |
fin | -0.036 |
Token | her |
Token ID | 607 |
Feature activation | -4.14e-01 |
Pos logit contributions | |
brushes | +0.029 |
anna | +0.028 |
bars | +0.027 |
raid | +0.027 |
types | +0.026 |
Neg logit contributions | |
shark | -0.035 |
Bad | -0.034 |
Shark | -0.033 |
ile | -0.033 |
Fin | -0.032 |
Token | dog |
Token ID | 3290 |
Feature activation | -6.33e+00 |
Pos logit contributions | |
Bad | +0.018 |
shark | +0.017 |
ile | +0.016 |
Shark | +0.016 |
coon | +0.016 |
Neg logit contributions | |
brushes | -0.014 |
anna | -0.013 |
types | -0.012 |
raid | -0.012 |
bars | -0.012 |
Token | to |
Token ID | 284 |
Feature activation | -8.45e-01 |
Pos logit contributions | |
Bad | +0.239 |
shark | +0.231 |
ile | +0.218 |
coon | +0.214 |
fin | +0.212 |
Neg logit contributions | |
anna | -0.232 |
brushes | -0.226 |
raid | -0.215 |
types | -0.212 |
bars | -0.206 |
Token | go |
Token ID | 467 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
shark | +0.044 |
Bad | +0.044 |
ile | +0.042 |
Shark | +0.041 |
coon | +0.041 |
Neg logit contributions | |
brushes | -0.019 |
select | -0.016 |
raid | -0.016 |
anna | -0.016 |
types | -0.015 |
Token | near |
Token ID | 1474 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
shark | +0.047 |
Bad | +0.046 |
Shark | +0.044 |
ile | +0.043 |
Fin | +0.042 |
Neg logit contributions | |
brushes | -0.060 |
anna | -0.059 |
bars | -0.058 |
types | -0.056 |
raid | -0.055 |
Token | it |
Token ID | 340 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
ile | +0.067 |
shark | +0.066 |
Bad | +0.066 |
coon | +0.064 |
Shark | +0.062 |
Neg logit contributions | |
brushes | -0.035 |
work | -0.032 |
Paula | -0.031 |
select | -0.030 |
types | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
Bad | +0.049 |
shark | +0.047 |
Shark | +0.045 |
coon | +0.044 |
ile | +0.044 |
Neg logit contributions | |
brushes | -0.040 |
anna | -0.039 |
raid | -0.038 |
types | -0.038 |
bars | -0.037 |
Token | take |
Token ID | 1011 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
shark | +0.073 |
Bad | +0.072 |
Shark | +0.069 |
ile | +0.068 |
coon | +0.067 |
Neg logit contributions | |
brushes | -0.047 |
anna | -0.047 |
bars | -0.043 |
raid | -0.042 |
types | -0.042 |
Token | it |
Token ID | 340 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
anna | +0.047 |
brushes | +0.047 |
bars | +0.045 |
raid | +0.044 |
types | +0.044 |
Neg logit contributions | |
shark | -0.035 |
Bad | -0.034 |
Shark | -0.033 |
Fin | -0.031 |
ile | -0.031 |
Token | on |
Token ID | 319 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
Bad | +0.006 |
shark | +0.006 |
Shark | +0.005 |
ile | +0.005 |
coon | +0.005 |
Neg logit contributions | |
anna | -0.005 |
brushes | -0.005 |
bars | -0.005 |
raid | -0.005 |
types | -0.005 |
Token | their |
Token ID | 511 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
ile | +0.058 |
Bad | +0.057 |
shark | +0.056 |
coon | +0.056 |
Shark | +0.055 |
Neg logit contributions | |
Paula | -0.042 |
brushes | -0.040 |
decorations | -0.039 |
work | -0.039 |
anna | -0.039 |
Token | travels |
Token ID | 17781 |
Feature activation | -9.50e-01 |
Pos logit contributions | |
shark | +0.097 |
Bad | +0.097 |
Shark | +0.094 |
ile | +0.094 |
coon | +0.092 |
Neg logit contributions | |
brushes | -0.038 |
anna | -0.037 |
bars | -0.034 |
raid | -0.033 |
types | -0.032 |
Token | anymore |
Token ID | 7471 |
Feature activation | -6.29e+00 |
Pos logit contributions | |
Bad | +0.038 |
shark | +0.038 |
Shark | +0.037 |
ile | +0.035 |
fin | +0.035 |
Neg logit contributions | |
anna | -0.033 |
brushes | -0.032 |
bars | -0.032 |
raid | -0.031 |
types | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | -5.47e-01 |
Pos logit contributions | |
Bad | +0.276 |
shark | +0.263 |
Shark | +0.253 |
ile | +0.252 |
coon | +0.250 |
Neg logit contributions | |
anna | -0.207 |
brushes | -0.196 |
raid | -0.187 |
types | -0.180 |
bars | -0.179 |
Token |
|
Token ID | 198 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
shark | +0.024 |
Bad | +0.024 |
ile | +0.023 |
Shark | +0.023 |
coon | +0.022 |
Neg logit contributions | |
brushes | -0.017 |
anna | -0.017 |
bars | -0.016 |
raid | -0.015 |
types | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
shark | +0.058 |
Bad | +0.057 |
ile | +0.054 |
Shark | +0.053 |
fin | +0.053 |
Neg logit contributions | |
anna | -0.042 |
brushes | -0.040 |
Paula | -0.036 |
Along | -0.036 |
bars | -0.036 |
Token | The |
Token ID | 464 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
shark | +0.052 |
Shark | +0.048 |
ile | +0.047 |
Bad | +0.047 |
fin | +0.047 |
Neg logit contributions | |
anna | -0.046 |
brushes | -0.045 |
Along | -0.041 |
bars | -0.040 |
select | -0.040 |
Token | family |
Token ID | 1641 |
Feature activation | +3.90e-01 |
Pos logit contributions | |
Bad | +0.006 |
shark | +0.006 |
Shark | +0.006 |
ile | +0.006 |
coon | +0.006 |
Neg logit contributions | |
brushes | -0.004 |
anna | -0.004 |
bars | -0.004 |
types | -0.004 |
raid | -0.004 |
Token | 't |
Token ID | 470 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
shark | +0.021 |
Bad | +0.021 |
Shark | +0.020 |
ile | +0.020 |
coon | +0.019 |
Neg logit contributions | |
brushes | -0.017 |
anna | -0.017 |
bars | -0.016 |
raid | -0.015 |
types | -0.015 |
Token | write |
Token ID | 3551 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
shark | +0.069 |
Bad | +0.068 |
Shark | +0.066 |
ile | +0.065 |
coon | +0.064 |
Neg logit contributions | |
anna | -0.039 |
brushes | -0.039 |
bars | -0.036 |
types | -0.035 |
raid | -0.035 |
Token | in |
Token ID | 287 |
Feature activation | -7.57e-01 |
Pos logit contributions | |
anna | +0.051 |
brushes | +0.050 |
bars | +0.048 |
raid | +0.048 |
types | +0.047 |
Neg logit contributions | |
shark | -0.046 |
Bad | -0.046 |
Shark | -0.044 |
ile | -0.043 |
coon | -0.042 |
Token | her |
Token ID | 607 |
Feature activation | +6.67e-01 |
Pos logit contributions | |
shark | +0.035 |
Shark | +0.034 |
Bad | +0.034 |
ile | +0.033 |
coon | +0.033 |
Neg logit contributions | |
brushes | -0.023 |
anna | -0.021 |
types | -0.020 |
bars | -0.020 |
raid | -0.019 |
Token | journal |
Token ID | 3989 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
anna | +0.025 |
brushes | +0.025 |
bars | +0.024 |
raid | +0.024 |
types | +0.023 |
Neg logit contributions | |
shark | -0.025 |
Bad | -0.024 |
Shark | -0.023 |
ile | -0.022 |
fin | -0.022 |
Token | anymore |
Token ID | 7471 |
Feature activation | -6.29e+00 |
Pos logit contributions | |
Bad | +0.075 |
shark | +0.075 |
Shark | +0.072 |
ile | +0.069 |
coon | +0.068 |
Neg logit contributions | |
brushes | -0.084 |
anna | -0.082 |
bars | -0.079 |
raid | -0.078 |
types | -0.076 |
Token | . |
Token ID | 13 |
Feature activation | -7.22e-01 |
Pos logit contributions | |
shark | +0.231 |
Bad | +0.230 |
Shark | +0.220 |
ile | +0.214 |
coon | +0.209 |
Neg logit contributions | |
anna | -0.241 |
brushes | -0.241 |
raid | -0.226 |
bars | -0.222 |
types | -0.219 |
Token | |
Token ID | 220 |
Feature activation | +4.47e+00 |
Pos logit contributions | |
Fin | +0.019 |
Shark | +0.019 |
Bad | +0.019 |
Wally | +0.018 |
shark | +0.018 |
Neg logit contributions | |
bars | -0.034 |
brushes | -0.033 |
raid | -0.033 |
anna | -0.033 |
types | -0.032 |
Token |
|
Token ID | 198 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
brushes | +0.191 |
anna | +0.191 |
raid | +0.183 |
bars | +0.182 |
types | +0.180 |
Neg logit contributions | |
Bad | -0.134 |
shark | -0.132 |
Shark | -0.128 |
ile | -0.124 |
coon | -0.122 |
Token |
|
Token ID | 198 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
shark | +0.050 |
Bad | +0.049 |
Shark | +0.047 |
ile | +0.046 |
coon | +0.045 |
Neg logit contributions | |
brushes | -0.058 |
anna | -0.058 |
bars | -0.055 |
raid | -0.055 |
types | -0.054 |
Token | L |
Token ID | 43 |
Feature activation | +1.98e-01 |
Pos logit contributions | |
Bad | +0.050 |
shark | +0.050 |
Shark | +0.047 |
ile | +0.046 |
coon | +0.046 |
Neg logit contributions | |
brushes | -0.055 |
anna | -0.055 |
bars | -0.053 |
raid | -0.052 |
types | -0.051 |
Token | couldn |
Token ID | 3521 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
brushes | +0.006 |
anna | +0.006 |
bars | +0.006 |
raid | +0.006 |
types | +0.006 |
Neg logit contributions | |
shark | -0.007 |
Bad | -0.007 |
Shark | -0.007 |
ile | -0.007 |
coon | -0.007 |
Token | 't |
Token ID | 470 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
shark | +0.055 |
Bad | +0.055 |
Shark | +0.053 |
ile | +0.052 |
fin | +0.052 |
Neg logit contributions | |
brushes | -0.034 |
anna | -0.034 |
bars | -0.032 |
raid | -0.031 |
Paula | -0.029 |
Token | even |
Token ID | 772 |
Feature activation | -9.96e-01 |
Pos logit contributions | |
Bad | +0.096 |
shark | +0.093 |
Shark | +0.091 |
ile | +0.089 |
Wally | +0.086 |
Neg logit contributions | |
brushes | -0.054 |
anna | -0.053 |
raid | -0.051 |
select | -0.048 |
work | -0.047 |
Token | see |
Token ID | 766 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
Bad | +0.045 |
shark | +0.045 |
ile | +0.043 |
Shark | +0.043 |
coon | +0.043 |
Neg logit contributions | |
brushes | -0.029 |
anna | -0.027 |
raid | -0.027 |
types | -0.025 |
bars | -0.025 |
Token | her |
Token ID | 607 |
Feature activation | +8.67e-01 |
Pos logit contributions | |
brushes | +0.047 |
anna | +0.046 |
bars | +0.043 |
raid | +0.043 |
types | +0.043 |
Neg logit contributions | |
shark | -0.061 |
Bad | -0.060 |
Shark | -0.058 |
ile | -0.056 |
coon | -0.056 |
Token | anymore |
Token ID | 7471 |
Feature activation | -6.28e+00 |
Pos logit contributions | |
brushes | +0.030 |
anna | +0.029 |
raid | +0.028 |
decorations | +0.027 |
types | +0.027 |
Neg logit contributions | |
Bad | -0.036 |
shark | -0.034 |
Shark | -0.033 |
ile | -0.033 |
coon | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -4.22e-01 |
Pos logit contributions | |
Bad | +0.252 |
shark | +0.249 |
Shark | +0.240 |
ile | +0.234 |
coon | +0.229 |
Neg logit contributions | |
anna | -0.221 |
brushes | -0.221 |
raid | -0.208 |
bars | -0.202 |
types | -0.198 |
Token |
|
Token ID | 198 |
Feature activation | -9.99e-01 |
Pos logit contributions | |
Bad | +0.013 |
shark | +0.013 |
Shark | +0.013 |
Fin | +0.013 |
Wally | +0.013 |
Neg logit contributions | |
bars | -0.017 |
raid | -0.017 |
anna | -0.017 |
brushes | -0.017 |
types | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
shark | +0.040 |
Bad | +0.039 |
Shark | +0.038 |
ile | +0.037 |
fin | +0.036 |
Neg logit contributions | |
anna | -0.035 |
brushes | -0.035 |
bars | -0.031 |
raid | -0.031 |
types | -0.031 |
Token | L |
Token ID | 43 |
Feature activation | +4.45e-01 |
Pos logit contributions | |
Bad | +0.035 |
shark | +0.034 |
Shark | +0.032 |
ile | +0.032 |
coon | +0.032 |
Neg logit contributions | |
brushes | -0.041 |
anna | -0.041 |
bars | -0.039 |
raid | -0.039 |
types | -0.039 |
Token | ily |
Token ID | 813 |
Feature activation | +3.87e-01 |
Pos logit contributions | |
anna | +0.021 |
brushes | +0.021 |
bars | +0.020 |
raid | +0.020 |
types | +0.020 |
Neg logit contributions | |
shark | -0.013 |
Bad | -0.012 |
Shark | -0.012 |
fin | -0.011 |
ile | -0.011 |
Token | tie |
Token ID | 9839 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
Bad | +0.088 |
shark | +0.080 |
bomb | +0.079 |
ile | +0.078 |
Shark | +0.078 |
Neg logit contributions | |
raid | -0.032 |
work | -0.029 |
anna | -0.029 |
ask | -0.028 |
select | -0.028 |
Token | his |
Token ID | 465 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
brushes | +0.045 |
anna | +0.042 |
types | +0.042 |
bars | +0.041 |
raid | +0.041 |
Neg logit contributions | |
shark | -0.046 |
Bad | -0.046 |
coon | -0.044 |
Shark | -0.044 |
ile | -0.044 |
Token | sho |
Token ID | 7354 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
Bad | +0.056 |
shark | +0.056 |
Shark | +0.053 |
ile | +0.053 |
coon | +0.052 |
Neg logit contributions | |
brushes | -0.058 |
anna | -0.058 |
bars | -0.054 |
raid | -0.054 |
types | -0.054 |
Token | el |
Token ID | 417 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
shark | +0.046 |
Bad | +0.046 |
coon | +0.044 |
Shark | +0.043 |
bomb | +0.042 |
Neg logit contributions | |
raid | -0.035 |
bars | -0.034 |
anna | -0.031 |
brushes | -0.029 |
rings | -0.029 |
Token | aces |
Token ID | 2114 |
Feature activation | +9.92e-01 |
Pos logit contributions | |
shark | +0.043 |
Bad | +0.042 |
ile | +0.041 |
Shark | +0.041 |
fin | +0.040 |
Neg logit contributions | |
brushes | -0.037 |
anna | -0.036 |
bars | -0.035 |
types | -0.034 |
raid | -0.033 |
Token | anymore |
Token ID | 7471 |
Feature activation | -6.24e+00 |
Pos logit contributions | |
bars | +0.035 |
anna | +0.035 |
brushes | +0.034 |
raid | +0.034 |
rings | +0.031 |
Neg logit contributions | |
Bad | -0.040 |
shark | -0.040 |
Shark | -0.039 |
coon | -0.036 |
ile | -0.036 |
Token | ! |
Token ID | 0 |
Feature activation | -5.44e-01 |
Pos logit contributions | |
Bad | +0.293 |
shark | +0.268 |
Shark | +0.266 |
ile | +0.258 |
coon | +0.256 |
Neg logit contributions | |
anna | -0.189 |
brushes | -0.183 |
raid | -0.182 |
types | -0.169 |
bars | -0.164 |
Token | He |
Token ID | 679 |
Feature activation | -5.60e-01 |
Pos logit contributions | |
shark | +0.022 |
Shark | +0.021 |
Bad | +0.021 |
Fin | +0.020 |
Wally | +0.020 |
Neg logit contributions | |
brushes | -0.020 |
anna | -0.019 |
bars | -0.019 |
raid | -0.018 |
types | -0.018 |
Token | tried |
Token ID | 3088 |
Feature activation | +7.00e-01 |
Pos logit contributions | |
shark | +0.024 |
Bad | +0.024 |
Shark | +0.023 |
ile | +0.022 |
coon | +0.022 |
Neg logit contributions | |
anna | -0.018 |
brushes | -0.018 |
bars | -0.016 |
raid | -0.016 |
types | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
brushes | +0.023 |
anna | +0.023 |
bars | +0.021 |
raid | +0.021 |
types | +0.021 |
Neg logit contributions | |
Bad | -0.029 |
shark | -0.029 |
ile | -0.028 |
Shark | -0.028 |
coon | -0.027 |
Token | tried |
Token ID | 3088 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
shark | +0.046 |
Bad | +0.046 |
Shark | +0.044 |
ile | +0.043 |
coon | +0.042 |
Neg logit contributions | |
anna | -0.037 |
brushes | -0.037 |
bars | -0.035 |
raid | -0.034 |
types | -0.034 |
Token | flood |
Token ID | 6947 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
brushes | +0.013 |
anna | +0.012 |
bars | +0.011 |
types | +0.011 |
raid | +0.011 |
Neg logit contributions | |
Bad | -0.017 |
shark | -0.016 |
Shark | -0.016 |
ile | -0.016 |
coon | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -5.36e-02 |
Pos logit contributions | |
Bad | +0.062 |
Shark | +0.062 |
shark | +0.061 |
fin | +0.058 |
coon | +0.056 |
Neg logit contributions | |
brushes | -0.060 |
anna | -0.059 |
bars | -0.056 |
types | -0.056 |
raid | -0.054 |
Token | She |
Token ID | 1375 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
shark | +0.002 |
Bad | +0.002 |
Shark | +0.002 |
ile | +0.002 |
coon | +0.002 |
Neg logit contributions | |
brushes | -0.002 |
anna | -0.002 |
bars | -0.002 |
raid | -0.002 |
types | -0.002 |
Token | asked |
Token ID | 1965 |
Feature activation | +9.30e-01 |
Pos logit contributions | |
anna | +0.059 |
brushes | +0.059 |
bars | +0.056 |
raid | +0.056 |
types | +0.054 |
Neg logit contributions | |
shark | -0.067 |
Bad | -0.066 |
Shark | -0.063 |
ile | -0.062 |
coon | -0.061 |
Token | her |
Token ID | 607 |
Feature activation | -3.06e-01 |
Pos logit contributions | |
brushes | +0.028 |
anna | +0.028 |
bars | +0.026 |
raid | +0.025 |
types | +0.025 |
Neg logit contributions | |
shark | -0.042 |
Bad | -0.041 |
Shark | -0.040 |
ile | -0.039 |
coon | -0.038 |
Token | dog |
Token ID | 3290 |
Feature activation | -6.24e+00 |
Pos logit contributions | |
shark | +0.012 |
Bad | +0.012 |
Shark | +0.011 |
ile | +0.011 |
coon | +0.011 |
Neg logit contributions | |
brushes | -0.011 |
anna | -0.011 |
bars | -0.011 |
raid | -0.010 |
types | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | -2.41e+00 |
Pos logit contributions | |
Bad | +0.248 |
shark | +0.245 |
ile | +0.231 |
coon | +0.228 |
Shark | +0.225 |
Neg logit contributions | |
anna | -0.215 |
brushes | -0.208 |
raid | -0.197 |
types | -0.190 |
bars | -0.188 |
Token | " |
Token ID | 366 |
Feature activation | -9.04e-01 |
Pos logit contributions | |
shark | +0.129 |
Bad | +0.126 |
ile | +0.124 |
Shark | +0.120 |
coon | +0.119 |
Neg logit contributions | |
anna | -0.048 |
brushes | -0.048 |
Paula | -0.042 |
bars | -0.041 |
raid | -0.041 |
Token | shall |
Token ID | 49271 |
Feature activation | -9.45e-01 |
Pos logit contributions | |
shark | +0.033 |
fin | +0.030 |
ile | +0.029 |
Shark | +0.029 |
Fin | +0.029 |
Neg logit contributions | |
anna | -0.032 |
brushes | -0.032 |
bars | -0.032 |
raid | -0.030 |
read | -0.030 |
Token | we |
Token ID | 356 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
shark | +0.036 |
Bad | +0.036 |
Shark | +0.034 |
ile | +0.033 |
coon | +0.033 |
Neg logit contributions | |
brushes | -0.034 |
anna | -0.034 |
bars | -0.032 |
raid | -0.032 |
types | -0.031 |
Token | go |
Token ID | 467 |
Feature activation | -4.72e-01 |
Pos logit contributions | |
shark | +0.062 |
Bad | +0.062 |
Shark | +0.059 |
ile | +0.058 |
coon | +0.058 |
Neg logit contributions | |
brushes | -0.037 |
anna | -0.037 |
raid | -0.034 |
bars | -0.034 |
types | -0.033 |
Token | not |
Token ID | 407 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
brushes | +0.045 |
work | +0.042 |
anna | +0.041 |
select | +0.041 |
Paula | +0.041 |
Neg logit contributions | |
ile | -0.057 |
shark | -0.056 |
Bad | -0.055 |
coon | -0.053 |
Shark | -0.052 |
Token | put |
Token ID | 1234 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
shark | +0.079 |
Bad | +0.079 |
Shark | +0.076 |
ile | +0.073 |
Wally | +0.073 |
Neg logit contributions | |
anna | -0.045 |
brushes | -0.043 |
raid | -0.042 |
bars | -0.039 |
work | -0.038 |
Token | on |
Token ID | 319 |
Feature activation | -2.45e-01 |
Pos logit contributions | |
brushes | +0.055 |
anna | +0.050 |
bars | +0.049 |
types | +0.049 |
decorations | +0.048 |
Neg logit contributions | |
shark | -0.057 |
Bad | -0.056 |
coon | -0.055 |
Shark | -0.055 |
ile | -0.054 |
Token | their |
Token ID | 511 |
Feature activation | -8.97e-01 |
Pos logit contributions | |
shark | +0.008 |
coon | +0.008 |
ile | +0.008 |
Bad | +0.008 |
Shark | +0.008 |
Neg logit contributions | |
brushes | -0.010 |
anna | -0.009 |
types | -0.009 |
bars | -0.009 |
decorations | -0.009 |
Token | show |
Token ID | 905 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
Bad | +0.039 |
shark | +0.039 |
Shark | +0.038 |
ile | +0.037 |
coon | +0.037 |
Neg logit contributions | |
brushes | -0.028 |
anna | -0.027 |
raid | -0.025 |
bars | -0.024 |
types | -0.024 |
Token | anymore |
Token ID | 7471 |
Feature activation | -6.21e+00 |
Pos logit contributions | |
Bad | +0.005 |
shark | +0.005 |
Shark | +0.005 |
coon | +0.005 |
Fin | +0.005 |
Neg logit contributions | |
anna | -0.004 |
brushes | -0.004 |
bars | -0.004 |
raid | -0.004 |
rings | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
Bad | +0.285 |
shark | +0.273 |
Shark | +0.268 |
coon | +0.266 |
ile | +0.258 |
Neg logit contributions | |
anna | -0.185 |
brushes | -0.184 |
raid | -0.167 |
bars | -0.162 |
styles | -0.161 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -6.24e-01 |
Pos logit contributions | |
shark | +0.009 |
Bad | +0.008 |
ile | +0.008 |
coon | +0.008 |
fin | +0.008 |
Neg logit contributions | |
anna | -0.006 |
brushes | -0.006 |
types | -0.006 |
bars | -0.006 |
Along | -0.006 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
Bad | +0.032 |
shark | +0.031 |
fin | +0.030 |
coon | +0.030 |
ile | +0.030 |
Neg logit contributions | |
Along | -0.013 |
anna | -0.013 |
brushes | -0.012 |
stairs | -0.012 |
Paula | -0.012 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
brushes | +0.003 |
anna | +0.003 |
bars | +0.003 |
raid | +0.002 |
types | +0.002 |
Neg logit contributions | |
Bad | -0.008 |
shark | -0.008 |
Shark | -0.008 |
ile | -0.008 |
coon | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | +3.67e-02 |
Pos logit contributions | |
brushes | +0.051 |
anna | +0.048 |
types | +0.045 |
Along | +0.044 |
decorations | +0.043 |
Neg logit contributions | |
shark | -0.062 |
Bad | -0.059 |
Shark | -0.059 |
ile | -0.059 |
coon | -0.057 |
Token | idea |
Token ID | 2126 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
brushes | +0.021 |
anna | +0.021 |
bars | +0.019 |
raid | +0.018 |
types | +0.018 |
Neg logit contributions | |
shark | -0.049 |
Bad | -0.048 |
Shark | -0.047 |
ile | -0.046 |
coon | -0.046 |
Token | . |
Token ID | 13 |
Feature activation | -2.38e-01 |
Pos logit contributions | |
brushes | +0.050 |
anna | +0.049 |
raid | +0.047 |
types | +0.046 |
bars | +0.045 |
Neg logit contributions | |
Bad | -0.058 |
shark | -0.057 |
Shark | -0.056 |
ile | -0.053 |
fin | -0.052 |
Token | He |
Token ID | 679 |
Feature activation | -5.01e-01 |
Pos logit contributions | |
Fin | +0.006 |
shark | +0.006 |
Shark | +0.006 |
Rex | +0.006 |
Wally | +0.006 |
Neg logit contributions | |
anna | -0.012 |
brushes | -0.011 |
raid | -0.011 |
Paula | -0.011 |
decorations | -0.011 |
Token | asked |
Token ID | 1965 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
shark | +0.020 |
Shark | +0.018 |
Bad | +0.018 |
ile | +0.018 |
fin | +0.018 |
Neg logit contributions | |
anna | -0.018 |
bars | -0.017 |
brushes | -0.017 |
raid | -0.017 |
types | -0.016 |
Token | his |
Token ID | 465 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
brushes | +0.045 |
anna | +0.045 |
bars | +0.043 |
types | +0.043 |
raid | +0.043 |
Neg logit contributions | |
shark | -0.033 |
Bad | -0.033 |
Shark | -0.031 |
ile | -0.030 |
coon | -0.029 |
Token | dog |
Token ID | 3290 |
Feature activation | -6.18e+00 |
Pos logit contributions | |
Bad | +0.083 |
shark | +0.080 |
ile | +0.077 |
Shark | +0.077 |
coon | +0.077 |
Neg logit contributions | |
brushes | -0.063 |
anna | -0.060 |
raid | -0.057 |
types | -0.057 |
bars | -0.056 |
Token | , |
Token ID | 11 |
Feature activation | -2.39e+00 |
Pos logit contributions | |
Bad | +0.251 |
shark | +0.241 |
ile | +0.230 |
coon | +0.227 |
fin | +0.225 |
Neg logit contributions | |
anna | -0.205 |
brushes | -0.204 |
raid | -0.192 |
types | -0.189 |
bars | -0.181 |
Token | Spot |
Token ID | 15899 |
Feature activation | -3.66e+00 |
Pos logit contributions | |
shark | +0.123 |
Bad | +0.122 |
ile | +0.119 |
Shark | +0.116 |
coon | +0.115 |
Neg logit contributions | |
brushes | -0.052 |
anna | -0.050 |
bars | -0.045 |
types | -0.044 |
raid | -0.044 |
Token | , |
Token ID | 11 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
Bad | +0.134 |
shark | +0.130 |
ile | +0.123 |
coon | +0.121 |
Shark | +0.120 |
Neg logit contributions | |
anna | -0.137 |
brushes | -0.134 |
raid | -0.126 |
types | -0.126 |
bars | -0.123 |
Token | to |
Token ID | 284 |
Feature activation | +7.49e-01 |
Pos logit contributions | |
shark | +0.067 |
Bad | +0.066 |
Shark | +0.064 |
ile | +0.063 |
coon | +0.061 |
Neg logit contributions | |
brushes | -0.040 |
anna | -0.040 |
bars | -0.037 |
types | -0.036 |
raid | -0.036 |
Token | help |
Token ID | 1037 |
Feature activation | -3.53e-01 |
Pos logit contributions | |
brushes | +0.017 |
anna | +0.016 |
raid | +0.015 |
select | +0.015 |
types | +0.015 |
Neg logit contributions | |
shark | -0.038 |
Bad | -0.038 |
ile | -0.037 |
Shark | -0.036 |
coon | -0.035 |
Token | want |
Token ID | 765 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
Bad | +0.045 |
shark | +0.045 |
ile | +0.043 |
Shark | +0.043 |
coon | +0.042 |
Neg logit contributions | |
bars | -0.043 |
anna | -0.043 |
brushes | -0.043 |
types | -0.040 |
three | -0.038 |
Token | to |
Token ID | 284 |
Feature activation | +9.43e-01 |
Pos logit contributions | |
Bad | +0.063 |
shark | +0.063 |
ile | +0.061 |
Shark | +0.060 |
coon | +0.060 |
Neg logit contributions | |
brushes | -0.031 |
anna | -0.028 |
raid | -0.027 |
bars | -0.026 |
types | -0.026 |
Token | play |
Token ID | 711 |
Feature activation | -4.79e-01 |
Pos logit contributions | |
brushes | +0.029 |
anna | +0.029 |
bars | +0.027 |
raid | +0.027 |
types | +0.026 |
Neg logit contributions | |
shark | -0.041 |
Bad | -0.041 |
Shark | -0.039 |
ile | -0.039 |
coon | -0.038 |
Token | with |
Token ID | 351 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
Shark | +0.022 |
fin | +0.021 |
shark | +0.021 |
Bad | +0.021 |
ile | +0.020 |
Neg logit contributions | |
decorations | -0.012 |
work | -0.012 |
brushes | -0.012 |
shopping | -0.012 |
sights | -0.012 |
Token | her |
Token ID | 607 |
Feature activation | +7.45e-01 |
Pos logit contributions | |
Shark | +0.077 |
Bad | +0.076 |
shark | +0.076 |
ile | +0.076 |
fin | +0.074 |
Neg logit contributions | |
brushes | -0.034 |
anna | -0.028 |
types | -0.026 |
raid | -0.025 |
cere | -0.025 |
Token | anymore |
Token ID | 7471 |
Feature activation | -6.18e+00 |
Pos logit contributions | |
brushes | +0.024 |
anna | +0.023 |
raid | +0.022 |
types | +0.022 |
decorations | +0.022 |
Neg logit contributions | |
Bad | -0.032 |
Shark | -0.031 |
shark | -0.030 |
ile | -0.030 |
coon | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | -3.50e-01 |
Pos logit contributions | |
Bad | +0.269 |
shark | +0.260 |
Shark | +0.255 |
ile | +0.253 |
coon | +0.244 |
Neg logit contributions | |
anna | -0.197 |
brushes | -0.192 |
raid | -0.180 |
Paula | -0.173 |
types | -0.173 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
Bad | +0.011 |
shark | +0.011 |
Fin | +0.011 |
Shark | +0.011 |
Wally | +0.011 |
Neg logit contributions | |
bars | -0.014 |
brushes | -0.014 |
raid | -0.014 |
anna | -0.014 |
types | -0.014 |
Token | tried |
Token ID | 3088 |
Feature activation | +8.09e-01 |
Pos logit contributions | |
shark | +0.005 |
Bad | +0.005 |
Shark | +0.005 |
ile | +0.005 |
coon | +0.004 |
Neg logit contributions | |
anna | -0.004 |
brushes | -0.004 |
bars | -0.004 |
raid | -0.004 |
types | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | -4.75e-01 |
Pos logit contributions | |
brushes | +0.022 |
anna | +0.021 |
bars | +0.020 |
raid | +0.020 |
types | +0.019 |
Neg logit contributions | |
shark | -0.038 |
Bad | -0.038 |
ile | -0.037 |
Shark | -0.037 |
coon | -0.036 |
Token | say |
Token ID | 910 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
shark | +0.024 |
Shark | +0.023 |
Bad | +0.023 |
ile | +0.023 |
fin | +0.021 |
Neg logit contributions | |
brushes | -0.012 |
raid | -0.011 |
anna | -0.011 |
work | -0.011 |
select | -0.011 |
Token | draw |
Token ID | 3197 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
shark | +0.069 |
Bad | +0.069 |
Shark | +0.066 |
ile | +0.066 |
coon | +0.066 |
Neg logit contributions | |
anna | -0.045 |
brushes | -0.043 |
types | -0.042 |
bars | -0.041 |
raid | -0.038 |
Token | on |
Token ID | 319 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
brushes | +0.062 |
bars | +0.058 |
anna | +0.057 |
types | +0.056 |
alphabet | +0.055 |
Neg logit contributions | |
shark | -0.063 |
Bad | -0.063 |
Shark | -0.062 |
ile | -0.059 |
coon | -0.058 |
Token | his |
Token ID | 465 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
Bad | +0.044 |
shark | +0.044 |
ile | +0.043 |
Shark | +0.043 |
coon | +0.042 |
Neg logit contributions | |
brushes | -0.048 |
anna | -0.046 |
bars | -0.044 |
types | -0.044 |
raid | -0.043 |
Token | black |
Token ID | 2042 |
Feature activation | -2.74e+00 |
Pos logit contributions | |
shark | +0.077 |
Bad | +0.075 |
Shark | +0.073 |
ile | +0.072 |
coon | +0.071 |
Neg logit contributions | |
anna | -0.048 |
brushes | -0.048 |
bars | -0.045 |
raid | -0.044 |
types | -0.043 |
Token | board |
Token ID | 3526 |
Feature activation | -5.45e-01 |
Pos logit contributions | |
shark | +0.115 |
Bad | +0.111 |
Shark | +0.108 |
ile | +0.106 |
fin | +0.105 |
Neg logit contributions | |
anna | -0.090 |
brushes | -0.089 |
bars | -0.084 |
raid | -0.082 |
types | -0.082 |
Token | anymore |
Token ID | 7471 |
Feature activation | -6.18e+00 |
Pos logit contributions | |
Bad | +0.020 |
shark | +0.020 |
Shark | +0.020 |
ile | +0.019 |
Fin | +0.018 |
Neg logit contributions | |
brushes | -0.021 |
bars | -0.020 |
anna | -0.020 |
raid | -0.019 |
types | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -6.77e-01 |
Pos logit contributions | |
Bad | +0.275 |
shark | +0.259 |
ile | +0.253 |
Shark | +0.252 |
hawk | +0.245 |
Neg logit contributions | |
brushes | -0.191 |
anna | -0.191 |
raid | -0.183 |
bars | -0.174 |
types | -0.173 |
Token |
|
Token ID | 198 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
Fin | +0.019 |
Shark | +0.019 |
shark | +0.019 |
Bad | +0.019 |
Wally | +0.018 |
Neg logit contributions | |
brushes | -0.031 |
raid | -0.031 |
anna | -0.030 |
bars | -0.030 |
types | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
shark | +0.057 |
Bad | +0.056 |
Shark | +0.053 |
ile | +0.053 |
fin | +0.052 |
Neg logit contributions | |
anna | -0.046 |
brushes | -0.045 |
bars | -0.041 |
types | -0.041 |
raid | -0.041 |
Token | Tim |
Token ID | 14967 |
Feature activation | -6.24e-01 |
Pos logit contributions | |
shark | +0.050 |
Bad | +0.047 |
Shark | +0.047 |
ile | +0.046 |
fin | +0.044 |
Neg logit contributions | |
anna | -0.055 |
brushes | -0.054 |
bars | -0.050 |
raid | -0.050 |
types | -0.049 |
Token | my |
Token ID | 1820 |
Feature activation | +9.90e-01 |
Pos logit contributions | |
Bad | +0.031 |
shark | +0.031 |
ile | +0.031 |
Shark | +0.030 |
coon | +0.030 |
Neg logit contributions | |
brushes | -0.014 |
anna | -0.013 |
bars | -0.013 |
raid | -0.013 |
types | -0.012 |
Token | couldn |
Token ID | 3521 |
Feature activation | -7.99e-01 |
Pos logit contributions | |
Bad | +0.110 |
shark | +0.110 |
ile | +0.107 |
coon | +0.106 |
bomb | +0.104 |
Neg logit contributions | |
anna | -0.049 |
brushes | -0.044 |
paid | -0.041 |
climbs | -0.041 |
Along | -0.039 |
Token | 't |
Token ID | 470 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
shark | +0.037 |
Bad | +0.037 |
Shark | +0.035 |
ile | +0.035 |
fin | +0.035 |
Neg logit contributions | |
anna | -0.022 |
brushes | -0.021 |
bars | -0.021 |
raid | -0.019 |
rell | -0.019 |
Token | see |
Token ID | 766 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
Bad | +0.098 |
shark | +0.092 |
ile | +0.092 |
bomb | +0.091 |
Shark | +0.089 |
Neg logit contributions | |
anna | -0.041 |
work | -0.039 |
raid | -0.039 |
select | -0.037 |
ask | -0.036 |
Token | his |
Token ID | 465 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
bars | +0.044 |
brushes | +0.043 |
anna | +0.041 |
types | +0.041 |
rings | +0.041 |
Neg logit contributions | |
coon | -0.065 |
Bad | -0.064 |
shark | -0.064 |
ile | -0.063 |
Shark | -0.062 |
Token | yard |
Token ID | 12699 |
Feature activation | -6.57e-02 |
Pos logit contributions | |
shark | +0.045 |
Bad | +0.044 |
ile | +0.044 |
coon | +0.043 |
Shark | +0.042 |
Neg logit contributions | |
brushes | -0.034 |
raid | -0.033 |
anna | -0.031 |
sights | -0.031 |
bars | -0.030 |
Token | anymore |
Token ID | 7471 |
Feature activation | -6.15e+00 |
Pos logit contributions | |
Bad | +0.003 |
Shark | +0.002 |
shark | +0.002 |
ile | +0.002 |
fin | +0.002 |
Neg logit contributions | |
bars | -0.002 |
brushes | -0.002 |
types | -0.002 |
raid | -0.002 |
stairs | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -4.75e-01 |
Pos logit contributions | |
Bad | +0.317 |
shark | +0.293 |
Shark | +0.292 |
ile | +0.292 |
coon | +0.283 |
Neg logit contributions | |
anna | -0.143 |
raid | -0.142 |
brushes | -0.139 |
bars | -0.136 |
work | -0.127 |
Token | Max |
Token ID | 5436 |
Feature activation | -2.16e+00 |
Pos logit contributions | |
shark | +0.016 |
Bad | +0.016 |
Shark | +0.016 |
Fin | +0.016 |
Rex | +0.015 |
Neg logit contributions | |
anna | -0.019 |
brushes | -0.019 |
raid | -0.019 |
types | -0.018 |
bars | -0.018 |
Token | was |
Token ID | 373 |
Feature activation | -7.08e-01 |
Pos logit contributions | |
shark | +0.096 |
ile | +0.095 |
Bad | +0.094 |
Shark | +0.090 |
coon | +0.089 |
Neg logit contributions | |
anna | -0.064 |
brushes | -0.059 |
nie | -0.052 |
types | -0.052 |
select | -0.051 |
Token | lost |
Token ID | 2626 |
Feature activation | -3.21e+00 |
Pos logit contributions | |
shark | +0.032 |
ile | +0.031 |
Bad | +0.031 |
Shark | +0.030 |
coon | +0.030 |
Neg logit contributions | |
anna | -0.019 |
brushes | -0.018 |
Paula | -0.017 |
select | -0.017 |
Along | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | -3.79e-01 |
Pos logit contributions | |
Shark | +0.164 |
ile | +0.164 |
Bad | +0.159 |
coon | +0.157 |
shark | +0.151 |
Neg logit contributions | |
work | -0.057 |
com | -0.056 |
few | -0.052 |
sights | -0.051 |
nie | -0.050 |
Token | want |
Token ID | 765 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
shark | +0.049 |
Shark | +0.046 |
ile | +0.045 |
fin | +0.045 |
Bad | +0.045 |
Neg logit contributions | |
brushes | -0.046 |
bars | -0.045 |
anna | -0.042 |
types | -0.042 |
three | -0.041 |
Token | to |
Token ID | 284 |
Feature activation | +9.67e-01 |
Pos logit contributions | |
shark | +0.051 |
Bad | +0.051 |
Shark | +0.049 |
ile | +0.048 |
coon | +0.048 |
Neg logit contributions | |
brushes | -0.030 |
anna | -0.030 |
bars | -0.027 |
raid | -0.027 |
types | -0.026 |
Token | play |
Token ID | 711 |
Feature activation | -4.59e-01 |
Pos logit contributions | |
anna | +0.025 |
brushes | +0.024 |
raid | +0.023 |
bars | +0.021 |
types | +0.021 |
Neg logit contributions | |
Bad | -0.048 |
shark | -0.047 |
ile | -0.045 |
coon | -0.045 |
Shark | -0.045 |
Token | with |
Token ID | 351 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
Bad | +0.021 |
hawk | +0.020 |
Shark | +0.020 |
ile | +0.020 |
fin | +0.020 |
Neg logit contributions | |
decorations | -0.012 |
work | -0.012 |
shopping | -0.011 |
styles | -0.011 |
anna | -0.011 |
Token | her |
Token ID | 607 |
Feature activation | +8.23e-01 |
Pos logit contributions | |
shark | +0.075 |
Bad | +0.074 |
ile | +0.074 |
Shark | +0.074 |
fin | +0.072 |
Neg logit contributions | |
brushes | -0.030 |
anna | -0.027 |
raid | -0.024 |
Paula | -0.023 |
decorations | -0.022 |
Token | anymore |
Token ID | 7471 |
Feature activation | -6.15e+00 |
Pos logit contributions | |
brushes | +0.024 |
raid | +0.022 |
anna | +0.022 |
styles | +0.022 |
decorations | +0.021 |
Neg logit contributions | |
Bad | -0.038 |
Shark | -0.037 |
coon | -0.036 |
ile | -0.036 |
hawk | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -6.07e-01 |
Pos logit contributions | |
Bad | +0.253 |
shark | +0.246 |
Shark | +0.237 |
ile | +0.235 |
coon | +0.230 |
Neg logit contributions | |
anna | -0.213 |
brushes | -0.212 |
raid | -0.199 |
bars | -0.190 |
types | -0.190 |
Token | Lily |
Token ID | 20037 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
Bad | +0.022 |
shark | +0.022 |
Shark | +0.022 |
Fin | +0.021 |
Wally | +0.021 |
Neg logit contributions | |
anna | -0.022 |
brushes | -0.022 |
bars | -0.022 |
raid | -0.021 |
types | -0.021 |
Token | felt |
Token ID | 2936 |
Feature activation | -2.87e-02 |
Pos logit contributions | |
shark | +0.011 |
Bad | +0.011 |
ile | +0.011 |
Shark | +0.011 |
coon | +0.011 |
Neg logit contributions | |
anna | -0.008 |
brushes | -0.008 |
bars | -0.007 |
types | -0.007 |
raid | -0.007 |
Token | sad |
Token ID | 6507 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
ile | +0.001 |
coon | +0.001 |
Rex | +0.001 |
Shark | +0.001 |
shark | +0.001 |
Neg logit contributions | |
anna | -0.001 |
brushes | -0.001 |
work | -0.001 |
com | -0.001 |
necessary | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
Bad | +0.058 |
ile | +0.056 |
shark | +0.055 |
Shark | +0.055 |
coon | +0.055 |
Neg logit contributions | |
anna | -0.047 |
brushes | -0.045 |
raid | -0.044 |
bars | -0.043 |
types | -0.041 |
Token | 't |
Token ID | 470 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
shark | +0.034 |
Bad | +0.034 |
Shark | +0.032 |
ile | +0.032 |
fin | +0.032 |
Neg logit contributions | |
anna | -0.022 |
brushes | -0.022 |
bars | -0.021 |
raid | -0.020 |
types | -0.019 |
Token | play |
Token ID | 711 |
Feature activation | -9.12e-02 |
Pos logit contributions | |
Bad | +0.087 |
shark | +0.087 |
Shark | +0.085 |
ile | +0.083 |
Wally | +0.082 |
Neg logit contributions | |
anna | -0.039 |
raid | -0.038 |
brushes | -0.038 |
bars | -0.035 |
work | -0.034 |
Token | with |
Token ID | 351 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
Shark | +0.004 |
fin | +0.004 |
ile | +0.004 |
shark | +0.004 |
hawk | +0.004 |
Neg logit contributions | |
work | -0.003 |
brushes | -0.002 |
decorations | -0.002 |
select | -0.002 |
shopping | -0.002 |
Token | his |
Token ID | 465 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
shark | +0.045 |
ile | +0.044 |
Bad | +0.044 |
Shark | +0.044 |
coon | +0.042 |
Neg logit contributions | |
brushes | -0.050 |
anna | -0.048 |
bars | -0.047 |
raid | -0.046 |
types | -0.045 |
Token | car |
Token ID | 1097 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
Shark | +0.051 |
ile | +0.051 |
shark | +0.051 |
Bad | +0.051 |
coon | +0.050 |
Neg logit contributions | |
brushes | -0.030 |
bars | -0.027 |
anna | -0.026 |
types | -0.026 |
raid | -0.025 |
Token | anymore |
Token ID | 7471 |
Feature activation | -6.14e+00 |
Pos logit contributions | |
Shark | +0.076 |
shark | +0.076 |
Bad | +0.074 |
fin | +0.070 |
ile | +0.069 |
Neg logit contributions | |
bars | -0.095 |
brushes | -0.091 |
raid | -0.090 |
anna | -0.088 |
work | -0.085 |
Token | . |
Token ID | 13 |
Feature activation | -7.47e-01 |
Pos logit contributions | |
Bad | +0.281 |
shark | +0.269 |
Shark | +0.267 |
ile | +0.266 |
coon | +0.256 |
Neg logit contributions | |
anna | -0.179 |
brushes | -0.171 |
raid | -0.167 |
bars | -0.162 |
types | -0.155 |
Token |
|
Token ID | 198 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
shark | +0.025 |
Shark | +0.025 |
Bad | +0.025 |
Fin | +0.025 |
Wally | +0.023 |
Neg logit contributions | |
brushes | -0.030 |
anna | -0.030 |
raid | -0.030 |
bars | -0.029 |
types | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
shark | +0.066 |
Bad | +0.066 |
ile | +0.062 |
Shark | +0.062 |
fin | +0.061 |
Neg logit contributions | |
anna | -0.052 |
brushes | -0.050 |
bars | -0.046 |
types | -0.046 |
raid | -0.046 |
Token | Tim |
Token ID | 14967 |
Feature activation | -8.04e-01 |
Pos logit contributions | |
shark | +0.056 |
Shark | +0.052 |
Bad | +0.052 |
ile | +0.051 |
fin | +0.050 |
Neg logit contributions | |
anna | -0.062 |
brushes | -0.060 |
bars | -0.055 |
raid | -0.055 |
types | -0.055 |
Token | my |
Token ID | 1820 |
Feature activation | +7.96e-01 |
Pos logit contributions | |
Bad | +0.042 |
shark | +0.042 |
ile | +0.041 |
Shark | +0.040 |
coon | +0.040 |
Neg logit contributions | |
brushes | -0.017 |
anna | -0.016 |
bars | -0.015 |
raid | -0.015 |
types | -0.014 |
Token | play |
Token ID | 711 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
Bad | +0.097 |
shark | +0.095 |
Shark | +0.093 |
ile | +0.091 |
hawk | +0.088 |
Neg logit contributions | |
brushes | -0.036 |
anna | -0.035 |
raid | -0.034 |
select | -0.030 |
bars | -0.030 |
Token | with |
Token ID | 351 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
Shark | +0.004 |
shark | +0.004 |
Bad | +0.004 |
ile | +0.004 |
fin | +0.004 |
Neg logit contributions | |
brushes | -0.003 |
decorations | -0.003 |
work | -0.003 |
alphabet | -0.003 |
select | -0.003 |
Token | his |
Token ID | 465 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
shark | +0.044 |
Bad | +0.044 |
Shark | +0.042 |
ile | +0.042 |
coon | +0.041 |
Neg logit contributions | |
brushes | -0.053 |
anna | -0.050 |
bars | -0.050 |
types | -0.048 |
raid | -0.048 |
Token | favorite |
Token ID | 4004 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
shark | +0.048 |
Bad | +0.048 |
Shark | +0.046 |
ile | +0.045 |
coon | +0.044 |
Neg logit contributions | |
brushes | -0.034 |
anna | -0.034 |
bars | -0.032 |
raid | -0.031 |
types | -0.031 |
Token | car |
Token ID | 1097 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
shark | +0.081 |
Bad | +0.078 |
Shark | +0.074 |
ile | +0.073 |
coon | +0.072 |
Neg logit contributions | |
anna | -0.092 |
brushes | -0.090 |
bars | -0.086 |
raid | -0.085 |
types | -0.082 |
Token | anymore |
Token ID | 7471 |
Feature activation | -6.10e+00 |
Pos logit contributions | |
shark | +0.090 |
Shark | +0.090 |
Bad | +0.090 |
fin | +0.084 |
ile | +0.082 |
Neg logit contributions | |
bars | -0.071 |
brushes | -0.069 |
raid | -0.068 |
anna | -0.067 |
types | -0.063 |
Token | . |
Token ID | 13 |
Feature activation | -3.47e-01 |
Pos logit contributions | |
Bad | +0.301 |
shark | +0.279 |
Shark | +0.278 |
ile | +0.277 |
hawk | +0.270 |
Neg logit contributions | |
anna | -0.158 |
raid | -0.149 |
brushes | -0.147 |
types | -0.138 |
bars | -0.136 |
Token | He |
Token ID | 679 |
Feature activation | -7.35e-01 |
Pos logit contributions | |
Fin | +0.011 |
shark | +0.011 |
Shark | +0.011 |
Bad | +0.011 |
Wally | +0.010 |
Neg logit contributions | |
brushes | -0.014 |
anna | -0.014 |
raid | -0.014 |
bars | -0.014 |
types | -0.013 |
Token | asked |
Token ID | 1965 |
Feature activation | +9.33e-01 |
Pos logit contributions | |
shark | +0.031 |
Bad | +0.031 |
Shark | +0.030 |
ile | +0.029 |
coon | +0.029 |
Neg logit contributions | |
anna | -0.024 |
brushes | -0.023 |
bars | -0.022 |
raid | -0.021 |
types | -0.021 |
Token | his |
Token ID | 465 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
anna | +0.038 |
brushes | +0.038 |
types | +0.036 |
raid | +0.035 |
bars | +0.035 |
Neg logit contributions | |
Bad | -0.031 |
shark | -0.031 |
Shark | -0.029 |
ile | -0.029 |
coon | -0.028 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
Bad | +0.090 |
shark | +0.085 |
ile | +0.084 |
coon | +0.084 |
Shark | +0.083 |
Neg logit contributions | |
brushes | -0.068 |
anna | -0.063 |
types | -0.062 |
raid | -0.061 |
bars | -0.060 |
Token | 't |
Token ID | 470 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
shark | +0.006 |
Bad | +0.006 |
Shark | +0.006 |
ile | +0.006 |
fin | +0.006 |
Neg logit contributions | |
anna | -0.004 |
brushes | -0.004 |
bars | -0.003 |
raid | -0.003 |
Paula | -0.003 |
Token | want |
Token ID | 765 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
shark | +0.053 |
Bad | +0.052 |
Shark | +0.051 |
ile | +0.050 |
coon | +0.049 |
Neg logit contributions | |
brushes | -0.032 |
anna | -0.032 |
bars | -0.030 |
types | -0.029 |
raid | -0.029 |
Token | to |
Token ID | 284 |
Feature activation | +8.73e-01 |
Pos logit contributions | |
Bad | +0.056 |
shark | +0.056 |
Shark | +0.054 |
ile | +0.053 |
coon | +0.053 |
Neg logit contributions | |
brushes | -0.029 |
anna | -0.028 |
raid | -0.026 |
bars | -0.026 |
types | -0.025 |
Token | face |
Token ID | 1986 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
anna | +0.020 |
raid | +0.019 |
brushes | +0.019 |
work | +0.019 |
Paula | +0.017 |
Neg logit contributions | |
Bad | -0.044 |
shark | -0.043 |
ile | -0.042 |
coon | -0.041 |
Shark | -0.041 |
Token | her |
Token ID | 607 |
Feature activation | +9.10e-01 |
Pos logit contributions | |
Bad | +0.009 |
shark | +0.009 |
ile | +0.008 |
Shark | +0.008 |
coon | +0.008 |
Neg logit contributions | |
anna | -0.007 |
brushes | -0.007 |
raid | -0.006 |
Paula | -0.006 |
bars | -0.006 |
Token | anymore |
Token ID | 7471 |
Feature activation | -6.08e+00 |
Pos logit contributions | |
anna | +0.027 |
brushes | +0.027 |
raid | +0.026 |
attention | +0.025 |
sights | +0.025 |
Neg logit contributions | |
Bad | -0.041 |
coon | -0.039 |
ile | -0.039 |
Shark | -0.038 |
shark | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | -3.11e-01 |
Pos logit contributions | |
Bad | +0.265 |
shark | +0.255 |
Shark | +0.247 |
ile | +0.244 |
coon | +0.240 |
Neg logit contributions | |
anna | -0.198 |
brushes | -0.188 |
raid | -0.179 |
bars | -0.175 |
types | -0.175 |
Token | Unfortunately |
Token ID | 8989 |
Feature activation | -5.91e-01 |
Pos logit contributions | |
shark | +0.013 |
Bad | +0.013 |
Shark | +0.013 |
Fin | +0.012 |
Wally | +0.012 |
Neg logit contributions | |
brushes | -0.010 |
anna | -0.010 |
bars | -0.009 |
raid | -0.009 |
types | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | -6.09e-01 |
Pos logit contributions | |
shark | +0.020 |
Bad | +0.020 |
Shark | +0.019 |
ile | +0.019 |
coon | +0.019 |
Neg logit contributions | |
anna | -0.023 |
brushes | -0.023 |
bars | -0.022 |
raid | -0.022 |
types | -0.022 |
Token | he |
Token ID | 339 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
ile | +0.027 |
shark | +0.026 |
Bad | +0.026 |
coon | +0.026 |
Shark | +0.025 |
Neg logit contributions | |
brushes | -0.018 |
anna | -0.017 |
Paula | -0.016 |
bars | -0.016 |
types | -0.016 |
Token | was |
Token ID | 373 |
Feature activation | -2.95e-01 |
Pos logit contributions | |
Bad | +0.074 |
shark | +0.071 |
Shark | +0.067 |
coon | +0.067 |
bomb | +0.066 |
Neg logit contributions | |
anna | -0.042 |
brushes | -0.035 |
nie | -0.035 |
raid | -0.033 |
bars | -0.033 |
Token | couldn |
Token ID | 3521 |
Feature activation | -8.14e-01 |
Pos logit contributions | |
Bad | +0.112 |
shark | +0.110 |
coon | +0.108 |
ile | +0.108 |
bomb | +0.106 |
Neg logit contributions | |
anna | -0.051 |
brushes | -0.046 |
paid | -0.040 |
climbs | -0.040 |
select | -0.040 |
Token | 't |
Token ID | 470 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
shark | +0.038 |
Bad | +0.038 |
ile | +0.037 |
Shark | +0.036 |
fin | +0.036 |
Neg logit contributions | |
anna | -0.021 |
brushes | -0.021 |
bars | -0.020 |
raid | -0.019 |
Along | -0.019 |
Token | see |
Token ID | 766 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
Bad | +0.096 |
shark | +0.090 |
ile | +0.089 |
bomb | +0.088 |
Shark | +0.087 |
Neg logit contributions | |
anna | -0.042 |
raid | -0.039 |
work | -0.038 |
select | -0.037 |
brushes | -0.035 |
Token | his |
Token ID | 465 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
bars | +0.032 |
brushes | +0.032 |
types | +0.032 |
rings | +0.031 |
anna | +0.030 |
Neg logit contributions | |
coon | -0.053 |
Bad | -0.053 |
ile | -0.052 |
kin | -0.051 |
shark | -0.050 |
Token | family |
Token ID | 1641 |
Feature activation | -6.75e-01 |
Pos logit contributions | |
Bad | +0.070 |
coon | +0.068 |
ile | +0.067 |
Shark | +0.062 |
Fred | +0.061 |
Neg logit contributions | |
brushes | -0.030 |
raid | -0.029 |
bars | -0.028 |
sights | -0.027 |
work | -0.027 |
Token | anymore |
Token ID | 7471 |
Feature activation | -6.08e+00 |
Pos logit contributions | |
Bad | +0.028 |
shark | +0.026 |
Shark | +0.026 |
ile | +0.025 |
fin | +0.025 |
Neg logit contributions | |
bars | -0.022 |
raid | -0.022 |
anna | -0.022 |
brushes | -0.021 |
select | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | -7.28e-01 |
Pos logit contributions | |
Bad | +0.312 |
ile | +0.289 |
shark | +0.283 |
Shark | +0.281 |
coon | +0.279 |
Neg logit contributions | |
anna | -0.141 |
brushes | -0.136 |
raid | -0.135 |
work | -0.129 |
types | -0.124 |
Token | Max |
Token ID | 5436 |
Feature activation | -2.44e+00 |
Pos logit contributions | |
shark | +0.030 |
Bad | +0.030 |
Shark | +0.029 |
ile | +0.028 |
coon | +0.027 |
Neg logit contributions | |
anna | -0.024 |
brushes | -0.024 |
bars | -0.023 |
raid | -0.023 |
types | -0.022 |
Token | was |
Token ID | 373 |
Feature activation | -9.30e-01 |
Pos logit contributions | |
ile | +0.110 |
shark | +0.109 |
Bad | +0.108 |
coon | +0.104 |
bomb | +0.102 |
Neg logit contributions | |
anna | -0.072 |
brushes | -0.064 |
nie | -0.057 |
types | -0.056 |
Paula | -0.056 |
Token | lost |
Token ID | 2626 |
Feature activation | -3.28e+00 |
Pos logit contributions | |
shark | +0.041 |
ile | +0.041 |
Bad | +0.040 |
coon | +0.040 |
Shark | +0.039 |
Neg logit contributions | |
anna | -0.026 |
Paula | -0.024 |
brushes | -0.024 |
select | -0.023 |
nie | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | -5.64e-01 |
Pos logit contributions | |
ile | +0.170 |
Shark | +0.169 |
Bad | +0.167 |
coon | +0.163 |
Rex | +0.154 |
Neg logit contributions | |
com | -0.056 |
work | -0.053 |
few | -0.051 |
nie | -0.050 |
stairs | -0.048 |
Token | nos |
Token ID | 43630 |
Feature activation | -9.72e-01 |
Pos logit contributions | |
Bad | +0.024 |
ile | +0.024 |
Shark | +0.023 |
shark | +0.023 |
coon | +0.023 |
Neg logit contributions | |
brushes | -0.021 |
anna | -0.020 |
raid | -0.019 |
bars | -0.019 |
types | -0.019 |
Token | y |
Token ID | 88 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
Bad | +0.042 |
shark | +0.040 |
Shark | +0.039 |
coon | +0.038 |
Fin | +0.038 |
Neg logit contributions | |
anna | -0.032 |
bars | -0.032 |
raid | -0.031 |
brushes | -0.031 |
types | -0.030 |
Token | sister |
Token ID | 6621 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
Bad | +0.052 |
shark | +0.051 |
Shark | +0.050 |
ile | +0.049 |
Fin | +0.047 |
Neg logit contributions | |
anna | -0.059 |
brushes | -0.059 |
raid | -0.055 |
bars | -0.055 |
types | -0.055 |
Token | bothering |
Token ID | 41656 |
Feature activation | -3.80e+00 |
Pos logit contributions | |
shark | +0.095 |
Bad | +0.094 |
Shark | +0.093 |
hawk | +0.090 |
ile | +0.088 |
Neg logit contributions | |
anna | -0.067 |
raid | -0.061 |
brushes | -0.060 |
bars | -0.056 |
alphabet | -0.055 |
Token | him |
Token ID | 683 |
Feature activation | -3.04e+00 |
Pos logit contributions | |
shark | +0.188 |
Bad | +0.187 |
Shark | +0.183 |
ile | +0.182 |
fin | +0.177 |
Neg logit contributions | |
anna | -0.090 |
brushes | -0.087 |
raid | -0.082 |
Paula | -0.079 |
bars | -0.079 |
Token | anymore |
Token ID | 7471 |
Feature activation | -6.08e+00 |
Pos logit contributions | |
Bad | +0.148 |
Shark | +0.140 |
shark | +0.140 |
hawk | +0.135 |
ile | +0.132 |
Neg logit contributions | |
anna | -0.085 |
raid | -0.079 |
bars | -0.076 |
brushes | -0.073 |
types | -0.072 |
Token | . |
Token ID | 13 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
Bad | +0.292 |
shark | +0.279 |
Shark | +0.276 |
ile | +0.271 |
hawk | +0.270 |
Neg logit contributions | |
anna | -0.166 |
brushes | -0.151 |
raid | -0.147 |
types | -0.139 |
styles | -0.138 |
Token | The |
Token ID | 383 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
shark | +0.008 |
Fin | +0.008 |
Shark | +0.008 |
Bad | +0.008 |
Wally | +0.008 |
Neg logit contributions | |
brushes | -0.013 |
raid | -0.012 |
bars | -0.012 |
anna | -0.012 |
types | -0.012 |
Token | end |
Token ID | 886 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
brushes | +0.007 |
alphabet | +0.007 |
work | +0.006 |
bars | +0.006 |
types | +0.006 |
Neg logit contributions | |
Shark | -0.009 |
ile | -0.009 |
shark | -0.009 |
Fin | -0.009 |
Bad | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -2.61e-01 |
Pos logit contributions | |
shark | +0.059 |
Bad | +0.059 |
Shark | +0.057 |
fin | +0.055 |
Fin | +0.054 |
Neg logit contributions | |
brushes | -0.049 |
anna | -0.049 |
raid | -0.046 |
types | -0.043 |
bars | -0.043 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -6.91e-01 |
Pos logit contributions | |
shark | +0.011 |
Bad | +0.011 |
ile | +0.010 |
Shark | +0.010 |
fin | +0.010 |
Neg logit contributions | |
brushes | -0.008 |
anna | -0.008 |
bars | -0.008 |
types | -0.008 |
raid | -0.008 |
Token | couldn |
Token ID | 3521 |
Feature activation | -6.20e-01 |
Pos logit contributions | |
shark | +0.084 |
Bad | +0.082 |
Shark | +0.080 |
ile | +0.079 |
coon | +0.078 |
Neg logit contributions | |
anna | -0.063 |
brushes | -0.062 |
bars | -0.056 |
raid | -0.056 |
types | -0.055 |
Token | 't |
Token ID | 470 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
shark | +0.030 |
Bad | +0.030 |
ile | +0.029 |
fin | +0.029 |
Shark | +0.028 |
Neg logit contributions | |
anna | -0.015 |
brushes | -0.015 |
bars | -0.015 |
Paula | -0.013 |
rell | -0.013 |
Token | use |
Token ID | 779 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
Bad | +0.081 |
shark | +0.076 |
Shark | +0.075 |
Wally | +0.073 |
ile | +0.073 |
Neg logit contributions | |
anna | -0.039 |
work | -0.039 |
raid | -0.038 |
brushes | -0.037 |
select | -0.036 |
Token | his |
Token ID | 465 |
Feature activation | -9.23e-01 |
Pos logit contributions | |
shark | +0.001 |
Bad | +0.001 |
Shark | +0.001 |
ile | +0.001 |
coon | +0.001 |
Neg logit contributions | |
brushes | -0.002 |
anna | -0.002 |
bars | -0.002 |
raid | -0.002 |
types | -0.002 |
Token | finger |
Token ID | 7660 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
Bad | +0.042 |
shark | +0.041 |
ile | +0.040 |
coon | +0.040 |
Shark | +0.040 |
Neg logit contributions | |
brushes | -0.027 |
anna | -0.026 |
raid | -0.024 |
bars | -0.024 |
types | -0.024 |
Token | anymore |
Token ID | 7471 |
Feature activation | -6.07e+00 |
Pos logit contributions | |
Bad | +0.078 |
shark | +0.077 |
Shark | +0.075 |
coon | +0.073 |
ile | +0.073 |
Neg logit contributions | |
brushes | -0.061 |
anna | -0.060 |
bars | -0.058 |
raid | -0.058 |
types | -0.053 |
Token | . |
Token ID | 13 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
Bad | +0.251 |
shark | +0.241 |
Shark | +0.234 |
ile | +0.232 |
coon | +0.230 |
Neg logit contributions | |
anna | -0.207 |
brushes | -0.201 |
raid | -0.193 |
bars | -0.186 |
types | -0.185 |
Token | He |
Token ID | 679 |
Feature activation | -6.83e-01 |
Pos logit contributions | |
Fin | +0.007 |
Bad | +0.007 |
Shark | +0.007 |
shark | +0.006 |
Wally | +0.006 |
Neg logit contributions | |
brushes | -0.008 |
raid | -0.008 |
anna | -0.008 |
bars | -0.008 |
types | -0.008 |
Token | learned |
Token ID | 4499 |
Feature activation | +7.39e-01 |
Pos logit contributions | |
shark | +0.028 |
Bad | +0.028 |
Shark | +0.027 |
ile | +0.027 |
coon | +0.027 |
Neg logit contributions | |
anna | -0.023 |
brushes | -0.022 |
bars | -0.020 |
types | -0.020 |
raid | -0.020 |
Token | a |
Token ID | 257 |
Feature activation | -8.89e-02 |
Pos logit contributions | |
brushes | +0.031 |
anna | +0.030 |
Paula | +0.029 |
ilda | +0.029 |
bars | +0.029 |
Neg logit contributions | |
ile | -0.023 |
shark | -0.023 |
Bad | -0.022 |
Shark | -0.021 |
coon | -0.021 |
Token | lesson |
Token ID | 11483 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
shark | +0.004 |
Bad | +0.004 |
Shark | +0.004 |
fin | +0.004 |
bomb | +0.003 |
Neg logit contributions | |
anna | -0.003 |
brushes | -0.003 |
bars | -0.003 |
raid | -0.003 |
types | -0.003 |
Token | couldn |
Token ID | 3521 |
Feature activation | -3.86e-01 |
Pos logit contributions | |
Bad | +0.074 |
shark | +0.074 |
Shark | +0.072 |
ile | +0.071 |
coon | +0.071 |
Neg logit contributions | |
anna | -0.056 |
brushes | -0.056 |
bars | -0.051 |
raid | -0.050 |
types | -0.049 |
Token | 't |
Token ID | 470 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
shark | +0.017 |
Bad | +0.016 |
Shark | +0.016 |
ile | +0.016 |
fin | +0.015 |
Neg logit contributions | |
brushes | -0.012 |
anna | -0.012 |
bars | -0.011 |
raid | -0.011 |
types | -0.011 |
Token | watch |
Token ID | 2342 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
shark | +0.064 |
Bad | +0.063 |
Shark | +0.061 |
ile | +0.060 |
coon | +0.059 |
Neg logit contributions | |
brushes | -0.044 |
anna | -0.044 |
bars | -0.041 |
raid | -0.041 |
types | -0.039 |
Token | his |
Token ID | 465 |
Feature activation | -5.75e-01 |
Pos logit contributions | |
anna | +0.042 |
brushes | +0.042 |
bars | +0.040 |
raid | +0.039 |
types | +0.038 |
Neg logit contributions | |
Bad | -0.040 |
shark | -0.040 |
Shark | -0.038 |
ile | -0.038 |
coon | -0.037 |
Token | show |
Token ID | 905 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
Bad | +0.025 |
shark | +0.025 |
Shark | +0.024 |
ile | +0.024 |
coon | +0.024 |
Neg logit contributions | |
brushes | -0.017 |
anna | -0.017 |
bars | -0.016 |
raid | -0.016 |
types | -0.016 |
Token | anymore |
Token ID | 7471 |
Feature activation | -6.05e+00 |
Pos logit contributions | |
anna | +0.005 |
brushes | +0.005 |
bars | +0.005 |
raid | +0.005 |
Paula | +0.005 |
Neg logit contributions | |
Bad | -0.006 |
shark | -0.006 |
Shark | -0.005 |
ile | -0.005 |
coon | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | -7.35e-01 |
Pos logit contributions | |
Bad | +0.228 |
shark | +0.223 |
Shark | +0.215 |
ile | +0.211 |
coon | +0.206 |
Neg logit contributions | |
brushes | -0.224 |
anna | -0.224 |
bars | -0.211 |
raid | -0.211 |
types | -0.205 |
Token | Tim |
Token ID | 5045 |
Feature activation | -8.22e-01 |
Pos logit contributions | |
shark | +0.024 |
Bad | +0.023 |
Shark | +0.023 |
Fin | +0.023 |
Wally | +0.022 |
Neg logit contributions | |
anna | -0.031 |
brushes | -0.031 |
raid | -0.030 |
bars | -0.029 |
types | -0.029 |
Token | my |
Token ID | 1820 |
Feature activation | +7.32e-01 |
Pos logit contributions | |
shark | +0.051 |
Bad | +0.051 |
Shark | +0.050 |
ile | +0.049 |
coon | +0.048 |
Neg logit contributions | |
anna | -0.010 |
brushes | -0.010 |
bars | -0.008 |
raid | -0.008 |
types | -0.008 |
Token | was |
Token ID | 373 |
Feature activation | -6.80e-01 |
Pos logit contributions | |
anna | +0.025 |
brushes | +0.025 |
bars | +0.023 |
raid | +0.023 |
types | +0.023 |
Neg logit contributions | |
shark | -0.029 |
Bad | -0.029 |
Shark | -0.028 |
ile | -0.027 |
coon | -0.027 |
Token | sad |
Token ID | 6507 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
shark | +0.029 |
Bad | +0.028 |
ile | +0.028 |
Shark | +0.028 |
coon | +0.027 |
Neg logit contributions | |
anna | -0.021 |
brushes | -0.021 |
bars | -0.019 |
Paula | -0.019 |
raid | -0.019 |
Token | couldn |
Token ID | 3521 |
Feature activation | -6.27e-01 |
Pos logit contributions | |
Bad | +0.006 |
shark | +0.006 |
Shark | +0.005 |
coon | +0.005 |
fin | +0.005 |
Neg logit contributions | |
bars | -0.007 |
brushes | -0.007 |
anna | -0.007 |
types | -0.006 |
raid | -0.006 |
Token | 't |
Token ID | 470 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
shark | +0.026 |
Bad | +0.025 |
Shark | +0.024 |
ile | +0.024 |
coon | +0.023 |
Neg logit contributions | |
brushes | -0.021 |
anna | -0.021 |
bars | -0.020 |
raid | -0.019 |
types | -0.019 |
Token | watch |
Token ID | 2342 |
Feature activation | +8.63e-01 |
Pos logit contributions | |
shark | +0.067 |
Bad | +0.066 |
Shark | +0.064 |
ile | +0.063 |
coon | +0.062 |
Neg logit contributions | |
anna | -0.048 |
brushes | -0.048 |
bars | -0.045 |
types | -0.044 |
raid | -0.043 |
Token | her |
Token ID | 607 |
Feature activation | +9.87e-01 |
Pos logit contributions | |
anna | +0.033 |
brushes | +0.033 |
bars | +0.031 |
raid | +0.031 |
types | +0.031 |
Neg logit contributions | |
shark | -0.031 |
Bad | -0.030 |
Shark | -0.029 |
ile | -0.028 |
coon | -0.028 |
Token | show |
Token ID | 905 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
anna | +0.035 |
brushes | +0.034 |
bars | +0.032 |
raid | +0.032 |
types | +0.031 |
Neg logit contributions | |
shark | -0.039 |
Bad | -0.038 |
Shark | -0.037 |
ile | -0.036 |
coon | -0.035 |
Token | anymore |
Token ID | 7471 |
Feature activation | -6.02e+00 |
Pos logit contributions | |
shark | +0.001 |
Bad | +0.001 |
Shark | +0.000 |
ile | +0.000 |
fin | +0.000 |
Neg logit contributions | |
brushes | -0.001 |
anna | -0.001 |
bars | -0.000 |
raid | -0.000 |
types | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | -4.25e-01 |
Pos logit contributions | |
shark | +0.202 |
Bad | +0.199 |
Shark | +0.191 |
ile | +0.186 |
coon | +0.182 |
Neg logit contributions | |
brushes | -0.247 |
anna | -0.246 |
bars | -0.233 |
raid | -0.232 |
types | -0.228 |
Token | She |
Token ID | 1375 |
Feature activation | +9.81e-01 |
Pos logit contributions | |
nowhere | +0.016 |
Bad | +0.016 |
Wally | +0.016 |
Fin | +0.016 |
Shark | +0.016 |
Neg logit contributions | |
bars | -0.012 |
wn | -0.012 |
types | -0.012 |
uns | -0.012 |
raid | -0.012 |
Token | started |
Token ID | 2067 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
brushes | +0.033 |
anna | +0.033 |
bars | +0.032 |
raid | +0.031 |
types | +0.031 |
Neg logit contributions | |
shark | -0.039 |
Bad | -0.039 |
Shark | -0.038 |
ile | -0.037 |
fin | -0.036 |
Token | to |
Token ID | 284 |
Feature activation | +6.68e-01 |
Pos logit contributions | |
anna | +0.040 |
bars | +0.038 |
types | +0.037 |
brushes | +0.037 |
three | +0.036 |
Neg logit contributions | |
shark | -0.036 |
Bad | -0.035 |
attacking | -0.035 |
Shark | -0.034 |
nowhere | -0.034 |
Token | think |
Token ID | 892 |
Feature activation | +6.84e-01 |
Pos logit contributions | |
anna | +0.028 |
brushes | +0.026 |
bars | +0.026 |
types | +0.026 |
raid | +0.025 |
Neg logit contributions | |
Bad | -0.022 |
shark | -0.022 |
Shark | -0.020 |
fin | -0.020 |
coon | -0.019 |
Token | my |
Token ID | 1820 |
Feature activation | +3.75e-01 |
Pos logit contributions | |
shark | +0.069 |
Bad | +0.069 |
Shark | +0.067 |
ile | +0.067 |
coon | +0.065 |
Neg logit contributions | |
anna | -0.019 |
brushes | -0.019 |
bars | -0.016 |
raid | -0.016 |
types | -0.015 |
Token | his |
Token ID | 465 |
Feature activation | -9.05e-01 |
Pos logit contributions | |
brushes | +0.016 |
anna | +0.016 |
bars | +0.015 |
raid | +0.015 |
types | +0.015 |
Neg logit contributions | |
shark | -0.012 |
Bad | -0.011 |
Shark | -0.011 |
ile | -0.011 |
coon | -0.010 |
Token | toy |
Token ID | 13373 |
Feature activation | -3.20e+00 |
Pos logit contributions | |
Bad | +0.039 |
shark | +0.039 |
Shark | +0.038 |
ile | +0.038 |
coon | +0.037 |
Neg logit contributions | |
brushes | -0.028 |
anna | -0.027 |
bars | -0.025 |
raid | -0.025 |
types | -0.025 |
Token | back |
Token ID | 736 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
Bad | +0.138 |
shark | +0.133 |
Shark | +0.131 |
ile | +0.131 |
coon | +0.128 |
Neg logit contributions | |
brushes | -0.101 |
anna | -0.096 |
bars | -0.095 |
raid | -0.093 |
types | -0.088 |
Token | . |
Token ID | 13 |
Feature activation | -3.87e-01 |
Pos logit contributions | |
shark | +0.040 |
Bad | +0.040 |
Shark | +0.038 |
ile | +0.037 |
coon | +0.037 |
Neg logit contributions | |
brushes | -0.043 |
anna | -0.043 |
bars | -0.040 |
raid | -0.040 |
types | -0.039 |
Token | |
Token ID | 220 |
Feature activation | +4.63e+00 |
Pos logit contributions | |
Fin | +0.013 |
Wally | +0.013 |
Shark | +0.012 |
Bad | +0.012 |
shark | +0.012 |
Neg logit contributions | |
raid | -0.015 |
anna | -0.015 |
brushes | -0.015 |
bars | -0.014 |
wn | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | -9.86e-01 |
Pos logit contributions | |
anna | +0.196 |
brushes | +0.196 |
raid | +0.189 |
bars | +0.187 |
types | +0.185 |
Neg logit contributions | |
Bad | -0.142 |
shark | -0.140 |
Shark | -0.138 |
Fin | -0.132 |
ile | -0.129 |
Token |
|
Token ID | 198 |
Feature activation | -9.49e-01 |
Pos logit contributions | |
shark | +0.037 |
Bad | +0.036 |
Shark | +0.035 |
ile | +0.034 |
coon | +0.033 |
Neg logit contributions | |
anna | -0.037 |
brushes | -0.037 |
bars | -0.034 |
raid | -0.034 |
types | -0.034 |
Token | Tim |
Token ID | 14967 |
Feature activation | -4.48e-01 |
Pos logit contributions | |
shark | +0.033 |
Shark | +0.031 |
Bad | +0.030 |
fin | +0.030 |
ile | +0.030 |
Neg logit contributions | |
brushes | -0.039 |
anna | -0.038 |
bars | -0.036 |
types | -0.035 |
raid | -0.035 |
Token | my |
Token ID | 1820 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
Bad | +0.024 |
shark | +0.023 |
ile | +0.023 |
Shark | +0.023 |
coon | +0.022 |
Neg logit contributions | |
brushes | -0.009 |
anna | -0.009 |
bars | -0.008 |
raid | -0.008 |
types | -0.008 |
Token | felt |
Token ID | 2936 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
anna | +0.038 |
brushes | +0.037 |
bars | +0.033 |
types | +0.033 |
raid | +0.033 |
Neg logit contributions | |
Bad | -0.049 |
shark | -0.049 |
Shark | -0.047 |
ile | -0.047 |
coon | -0.046 |
Token | wasn |
Token ID | 2492 |
Feature activation | +2.85e-01 |
Pos logit contributions | |
shark | +0.031 |
Bad | +0.030 |
Shark | +0.029 |
ile | +0.028 |
coon | +0.028 |
Neg logit contributions | |
anna | -0.027 |
brushes | -0.026 |
bars | -0.025 |
raid | -0.025 |
types | -0.024 |
Token | 't |
Token ID | 470 |
Feature activation | -2.21e-04 |
Pos logit contributions | |
anna | +0.007 |
bars | +0.006 |
Paula | +0.006 |
brushes | +0.006 |
select | +0.006 |
Neg logit contributions | |
shark | -0.014 |
Bad | -0.014 |
ile | -0.013 |
Shark | -0.013 |
fin | -0.013 |
Token | in |
Token ID | 287 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
shark | +0.000 |
Bad | +0.000 |
Shark | +0.000 |
ile | +0.000 |
coon | +0.000 |
Neg logit contributions | |
brushes | -0.000 |
anna | -0.000 |
bars | -0.000 |
raid | -0.000 |
types | -0.000 |
Token | trouble |
Token ID | 5876 |
Feature activation | -2.78e+00 |
Pos logit contributions | |
Paula | +0.050 |
work | +0.048 |
types | +0.046 |
alphabet | +0.044 |
attention | +0.043 |
Neg logit contributions | |
ile | -0.066 |
coon | -0.063 |
Bad | -0.060 |
Fred | -0.058 |
shark | -0.058 |
Token | . |
Token ID | 13 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
Bad | +0.107 |
shark | +0.104 |
ile | +0.103 |
Shark | +0.101 |
coon | +0.098 |
Neg logit contributions | |
anna | -0.101 |
raid | -0.091 |
brushes | -0.091 |
bars | -0.089 |
Paula | -0.088 |
Token | |
Token ID | 220 |
Feature activation | +5.14e+00 |
Pos logit contributions | |
shark | +0.001 |
Shark | +0.001 |
Bad | +0.001 |
Fin | +0.001 |
fin | +0.001 |
Neg logit contributions | |
brushes | -0.000 |
anna | -0.000 |
raid | -0.000 |
bars | -0.000 |
types | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | -6.14e-01 |
Pos logit contributions | |
brushes | +0.202 |
anna | +0.202 |
bars | +0.191 |
raid | +0.189 |
types | +0.187 |
Neg logit contributions | |
shark | -0.179 |
Bad | -0.177 |
Shark | -0.170 |
ile | -0.165 |
coon | -0.162 |
Token |
|
Token ID | 198 |
Feature activation | -6.45e-01 |
Pos logit contributions | |
shark | +0.026 |
Bad | +0.025 |
ile | +0.025 |
coon | +0.024 |
Shark | +0.023 |
Neg logit contributions | |
anna | -0.020 |
brushes | -0.019 |
bars | -0.018 |
Paula | -0.017 |
raid | -0.017 |
Token | The |
Token ID | 464 |
Feature activation | +9.90e-02 |
Pos logit contributions | |
ile | +0.024 |
shark | +0.024 |
Bad | +0.023 |
Wally | +0.022 |
Shark | +0.022 |
Neg logit contributions | |
anna | -0.022 |
brushes | -0.022 |
bars | -0.021 |
Along | -0.020 |
select | -0.020 |
Token | owner |
Token ID | 4870 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
brushes | +0.003 |
sights | +0.003 |
bars | +0.003 |
alphabet | +0.003 |
work | +0.003 |
Neg logit contributions | |
ile | -0.004 |
shark | -0.004 |
Shark | -0.004 |
fin | -0.004 |
ful | -0.004 |
Token | then |
Token ID | 788 |
Feature activation | -8.39e-01 |
Pos logit contributions | |
shark | +0.047 |
Bad | +0.047 |
Shark | +0.044 |
ile | +0.043 |
fin | +0.043 |
Neg logit contributions | |
brushes | -0.038 |
anna | -0.038 |
bars | -0.035 |
raid | -0.035 |
types | -0.034 |
Token | beautiful |
Token ID | 4950 |
Feature activation | +7.64e-01 |
Pos logit contributions | |
shark | +0.051 |
Bad | +0.051 |
Shark | +0.049 |
ile | +0.048 |
coon | +0.047 |
Neg logit contributions | |
brushes | -0.051 |
anna | -0.051 |
bars | -0.048 |
raid | -0.047 |
types | -0.047 |
Token | ," |
Token ID | 553 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
brushes | +0.026 |
anna | +0.026 |
bars | +0.025 |
raid | +0.025 |
types | +0.024 |
Neg logit contributions | |
Bad | -0.030 |
shark | -0.030 |
Shark | -0.029 |
ile | -0.028 |
coon | -0.028 |
Token | said |
Token ID | 531 |
Feature activation | +7.49e-01 |
Pos logit contributions | |
shark | +0.052 |
Bad | +0.052 |
Shark | +0.050 |
ile | +0.048 |
coon | +0.048 |
Neg logit contributions | |
anna | -0.050 |
brushes | -0.049 |
bars | -0.048 |
raid | -0.047 |
types | -0.046 |
Token | mom |
Token ID | 1995 |
Feature activation | +6.53e-01 |
Pos logit contributions | |
brushes | +0.027 |
anna | +0.027 |
bars | +0.025 |
raid | +0.025 |
types | +0.025 |
Neg logit contributions | |
shark | -0.029 |
Bad | -0.028 |
Shark | -0.027 |
ile | -0.027 |
coon | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
brushes | +0.032 |
anna | +0.031 |
types | +0.030 |
sights | +0.030 |
bars | +0.030 |
Neg logit contributions | |
shark | -0.016 |
ful | -0.015 |
Shark | -0.015 |
ile | -0.015 |
fin | -0.014 |
Token | |
Token ID | 220 |
Feature activation | +4.51e+00 |
Pos logit contributions | |
Shark | +0.011 |
Fin | +0.011 |
shark | +0.011 |
Wally | +0.011 |
Bad | +0.011 |
Neg logit contributions | |
raid | -0.013 |
bars | -0.013 |
anna | -0.013 |
types | -0.012 |
wn | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
brushes | +0.205 |
anna | +0.205 |
raid | +0.200 |
bars | +0.197 |
types | +0.197 |
Neg logit contributions | |
Bad | -0.119 |
Shark | -0.117 |
Fin | -0.116 |
shark | -0.116 |
Wally | -0.111 |
Token |
|
Token ID | 198 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
Bad | +0.034 |
shark | +0.033 |
Shark | +0.032 |
coon | +0.031 |
ile | +0.031 |
Neg logit contributions | |
brushes | -0.049 |
bars | -0.048 |
anna | -0.048 |
raid | -0.048 |
types | -0.046 |
Token | " |
Token ID | 1 |
Feature activation | +6.58e-01 |
Pos logit contributions | |
Bad | +0.045 |
shark | +0.044 |
Shark | +0.042 |
coon | +0.042 |
ile | +0.042 |
Neg logit contributions | |
brushes | -0.037 |
anna | -0.037 |
bars | -0.036 |
raid | -0.035 |
types | -0.034 |
Token | The |
Token ID | 464 |
Feature activation | -1.91e-01 |
Pos logit contributions | |
brushes | +0.027 |
anna | +0.027 |
bars | +0.025 |
raid | +0.025 |
types | +0.025 |
Neg logit contributions | |
shark | -0.022 |
Bad | -0.022 |
Shark | -0.021 |
ile | -0.021 |
coon | -0.020 |
Token | bird |
Token ID | 6512 |
Feature activation | -3.22e-01 |
Pos logit contributions | |
shark | +0.007 |
Shark | +0.007 |
Bad | +0.007 |
Wally | +0.007 |
coon | +0.007 |
Neg logit contributions | |
brushes | -0.007 |
anna | -0.006 |
types | -0.006 |
alphabet | -0.006 |
bars | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | -8.53e-01 |
Pos logit contributions | |
shark | +0.059 |
Bad | +0.055 |
Shark | +0.053 |
Fin | +0.051 |
Wally | +0.049 |
Neg logit contributions | |
anna | -0.126 |
brushes | -0.125 |
bars | -0.124 |
Along | -0.119 |
raid | -0.118 |
Token | it |
Token ID | 340 |
Feature activation | +3.70e-01 |
Pos logit contributions | |
ile | +0.037 |
shark | +0.037 |
Bad | +0.036 |
Shark | +0.036 |
coon | +0.035 |
Neg logit contributions | |
brushes | -0.026 |
anna | -0.024 |
bars | -0.023 |
raid | -0.023 |
rings | -0.022 |
Token | 's |
Token ID | 338 |
Feature activation | -4.22e-01 |
Pos logit contributions | |
anna | +0.015 |
brushes | +0.015 |
raid | +0.015 |
bars | +0.014 |
Paula | +0.014 |
Neg logit contributions | |
shark | -0.012 |
Bad | -0.012 |
ile | -0.011 |
Shark | -0.011 |
ball | -0.011 |
Token | mine |
Token ID | 6164 |
Feature activation | -2.40e+00 |
Pos logit contributions | |
Bad | +0.014 |
shark | +0.014 |
Shark | +0.013 |
ile | +0.013 |
fin | +0.012 |
Neg logit contributions | |
anna | -0.017 |
brushes | -0.017 |
bars | -0.017 |
raid | -0.017 |
types | -0.016 |
Token | ." |
Token ID | 526 |
Feature activation | +3.28e-01 |
Pos logit contributions | |
shark | +0.086 |
Bad | +0.085 |
Shark | +0.082 |
ile | +0.079 |
coon | +0.078 |
Neg logit contributions | |
brushes | -0.093 |
anna | -0.092 |
bars | -0.088 |
raid | -0.086 |
types | -0.085 |
Token | |
Token ID | 220 |
Feature activation | +4.68e+00 |
Pos logit contributions | |
bars | +0.010 |
anna | +0.010 |
types | +0.010 |
raid | +0.010 |
brushes | +0.010 |
Neg logit contributions | |
Bad | -0.013 |
Fin | -0.013 |
Shark | -0.013 |
Wally | -0.013 |
Rex | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
brushes | +0.199 |
anna | +0.199 |
raid | +0.189 |
bars | +0.189 |
types | +0.187 |
Neg logit contributions | |
Bad | -0.143 |
shark | -0.141 |
Shark | -0.138 |
ile | -0.131 |
Fin | -0.131 |
Token |
|
Token ID | 198 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
Bad | +0.031 |
shark | +0.030 |
Shark | +0.030 |
Fin | +0.028 |
ile | +0.028 |
Neg logit contributions | |
brushes | -0.046 |
anna | -0.045 |
bars | -0.044 |
raid | -0.044 |
types | -0.043 |
Token | The |
Token ID | 464 |
Feature activation | +2.78e-02 |
Pos logit contributions | |
Bad | +0.033 |
shark | +0.033 |
Shark | +0.031 |
ile | +0.030 |
coon | +0.030 |
Neg logit contributions | |
anna | -0.047 |
brushes | -0.047 |
bars | -0.045 |
raid | -0.045 |
types | -0.044 |
Token | other |
Token ID | 584 |
Feature activation | -7.40e-01 |
Pos logit contributions | |
brushes | +0.001 |
anna | +0.001 |
bars | +0.001 |
raid | +0.001 |
types | +0.001 |
Neg logit contributions | |
shark | -0.001 |
Bad | -0.001 |
Shark | -0.001 |
ile | -0.001 |
coon | -0.001 |
Token | girl |
Token ID | 2576 |
Feature activation | +5.29e-01 |
Pos logit contributions | |
Bad | +0.033 |
shark | +0.033 |
Shark | +0.031 |
ile | +0.031 |
coon | +0.030 |
Neg logit contributions | |
anna | -0.023 |
brushes | -0.022 |
bars | -0.020 |
raid | -0.020 |
Paula | -0.019 |
Token | arm |
Token ID | 3211 |
Feature activation | -2.41e+00 |
Pos logit contributions | |
Bad | +0.030 |
shark | +0.029 |
Shark | +0.029 |
coon | +0.028 |
ile | +0.028 |
Neg logit contributions | |
brushes | -0.017 |
anna | -0.016 |
raid | -0.015 |
types | -0.015 |
bars | -0.015 |
Token | was |
Token ID | 373 |
Feature activation | -9.85e-01 |
Pos logit contributions | |
shark | +0.090 |
Bad | +0.088 |
Shark | +0.085 |
ile | +0.083 |
coon | +0.080 |
Neg logit contributions | |
brushes | -0.091 |
anna | -0.090 |
bars | -0.087 |
raid | -0.085 |
types | -0.082 |
Token | too |
Token ID | 1165 |
Feature activation | +3.40e-01 |
Pos logit contributions | |
shark | +0.034 |
Bad | +0.033 |
Shark | +0.033 |
ile | +0.031 |
coon | +0.031 |
Neg logit contributions | |
brushes | -0.039 |
anna | -0.038 |
bars | -0.037 |
types | -0.036 |
raid | -0.036 |
Token | short |
Token ID | 1790 |
Feature activation | -3.68e-01 |
Pos logit contributions | |
brushes | +0.011 |
anna | +0.011 |
bars | +0.010 |
raid | +0.010 |
types | +0.010 |
Neg logit contributions | |
shark | -0.014 |
Bad | -0.014 |
Shark | -0.014 |
ile | -0.014 |
coon | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -4.62e-01 |
Pos logit contributions | |
shark | +0.013 |
Bad | +0.013 |
Shark | +0.012 |
ile | +0.012 |
coon | +0.011 |
Neg logit contributions | |
brushes | -0.015 |
anna | -0.015 |
bars | -0.014 |
raid | -0.014 |
types | -0.014 |
Token | |
Token ID | 220 |
Feature activation | +4.44e+00 |
Pos logit contributions | |
shark | +0.020 |
Bad | +0.019 |
Shark | +0.019 |
ile | +0.018 |
coon | +0.018 |
Neg logit contributions | |
anna | -0.015 |
brushes | -0.015 |
bars | -0.014 |
raid | -0.014 |
types | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
brushes | +0.176 |
anna | +0.175 |
bars | +0.166 |
raid | +0.165 |
types | +0.163 |
Neg logit contributions | |
shark | -0.152 |
Bad | -0.150 |
Shark | -0.145 |
ile | -0.141 |
coon | -0.138 |
Token |
|
Token ID | 198 |
Feature activation | -9.25e-01 |
Pos logit contributions | |
shark | +0.038 |
Bad | +0.037 |
Shark | +0.035 |
ile | +0.035 |
coon | +0.034 |
Neg logit contributions | |
anna | -0.038 |
brushes | -0.038 |
bars | -0.035 |
raid | -0.035 |
types | -0.034 |
Token | Tim |
Token ID | 14967 |
Feature activation | -3.75e-01 |
Pos logit contributions | |
shark | +0.035 |
Bad | +0.033 |
Shark | +0.032 |
ile | +0.032 |
fin | +0.031 |
Neg logit contributions | |
anna | -0.035 |
brushes | -0.034 |
bars | -0.032 |
raid | -0.032 |
types | -0.031 |
Token | my |
Token ID | 1820 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
shark | +0.021 |
Bad | +0.021 |
Shark | +0.021 |
ile | +0.021 |
coon | +0.020 |
Neg logit contributions | |
brushes | -0.006 |
anna | -0.006 |
bars | -0.006 |
raid | -0.005 |
types | -0.005 |
Token | said |
Token ID | 531 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
anna | +0.040 |
brushes | +0.038 |
bars | +0.034 |
raid | +0.034 |
types | +0.034 |
Neg logit contributions | |
shark | -0.059 |
Bad | -0.058 |
Shark | -0.056 |
ile | -0.055 |
coon | -0.054 |
Token | someday |
Token ID | 25580 |
Feature activation | -7.72e-01 |
Pos logit contributions | |
shark | +0.073 |
Bad | +0.072 |
Shark | +0.069 |
coon | +0.067 |
ile | +0.067 |
Neg logit contributions | |
anna | -0.074 |
brushes | -0.073 |
raid | -0.070 |
types | -0.069 |
bars | -0.069 |
Token | . |
Token ID | 13 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
shark | +0.027 |
Bad | +0.027 |
Shark | +0.025 |
ile | +0.025 |
coon | +0.024 |
Neg logit contributions | |
anna | -0.030 |
brushes | -0.030 |
bars | -0.029 |
raid | -0.028 |
types | -0.028 |
Token | |
Token ID | 220 |
Feature activation | +4.99e+00 |
Pos logit contributions | |
brushes | +0.006 |
anna | +0.006 |
bars | +0.006 |
raid | +0.006 |
types | +0.006 |
Neg logit contributions | |
shark | -0.008 |
Bad | -0.008 |
Shark | -0.007 |
Fin | -0.007 |
ile | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | -4.61e-01 |
Pos logit contributions | |
brushes | +0.201 |
anna | +0.199 |
bars | +0.190 |
raid | +0.190 |
types | +0.186 |
Neg logit contributions | |
shark | -0.168 |
Bad | -0.166 |
Shark | -0.161 |
ile | -0.155 |
coon | -0.152 |
Token |
|
Token ID | 198 |
Feature activation | -4.64e-01 |
Pos logit contributions | |
shark | +0.017 |
Bad | +0.016 |
Shark | +0.016 |
ile | +0.015 |
fin | +0.015 |
Neg logit contributions | |
anna | -0.018 |
brushes | -0.017 |
bars | -0.016 |
types | -0.016 |
raid | -0.016 |
Token | From |
Token ID | 4863 |
Feature activation | +4.61e-01 |
Pos logit contributions | |
shark | +0.017 |
Bad | +0.016 |
Shark | +0.016 |
ile | +0.016 |
fin | +0.016 |
Neg logit contributions | |
anna | -0.018 |
brushes | -0.017 |
bars | -0.016 |
types | -0.016 |
raid | -0.016 |
Token | that |
Token ID | 326 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
brushes | +0.016 |
anna | +0.016 |
work | +0.015 |
Paula | +0.015 |
sights | +0.015 |
Neg logit contributions | |
shark | -0.016 |
Shark | -0.016 |
fin | -0.016 |
Bad | -0.016 |
ile | -0.016 |
Token | day |
Token ID | 1110 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
shark | +0.053 |
Bad | +0.053 |
Shark | +0.051 |
ile | +0.050 |
coon | +0.050 |
Neg logit contributions | |
anna | -0.023 |
brushes | -0.023 |
bars | -0.021 |
types | -0.021 |
raid | -0.020 |
Token | on |
Token ID | 319 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
anna | +0.009 |
brushes | +0.009 |
bars | +0.009 |
raid | +0.008 |
types | +0.008 |
Neg logit contributions | |
shark | -0.006 |
Bad | -0.006 |
Shark | -0.006 |
ile | -0.005 |
coon | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
Bad | +0.006 |
shark | +0.006 |
Shark | +0.005 |
ile | +0.005 |
coon | +0.005 |
Neg logit contributions | |
anna | -0.006 |
brushes | -0.006 |
bars | -0.005 |
Paula | -0.005 |
raid | -0.005 |
Token | whenever |
Token ID | 8797 |
Feature activation | +9.00e-01 |
Pos logit contributions | |
shark | +0.003 |
Bad | +0.003 |
Shark | +0.003 |
Fin | +0.003 |
ile | +0.003 |
Neg logit contributions | |
anna | -0.005 |
brushes | -0.005 |
bars | -0.004 |
raid | -0.004 |
types | -0.004 |
Token | that |
Token ID | 326 |
Feature activation | -5.01e-01 |
Pos logit contributions | |
shark | +0.030 |
Bad | +0.030 |
Shark | +0.029 |
ile | +0.028 |
fin | +0.028 |
Neg logit contributions | |
anna | -0.026 |
brushes | -0.025 |
bars | -0.024 |
raid | -0.024 |
types | -0.023 |
Token | ?" |
Token ID | 1701 |
Feature activation | +3.56e-01 |
Pos logit contributions | |
shark | +0.016 |
Bad | +0.015 |
Shark | +0.014 |
fin | +0.014 |
ile | +0.013 |
Neg logit contributions | |
brushes | -0.022 |
anna | -0.022 |
raid | -0.021 |
bars | -0.021 |
types | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
raid | +0.018 |
anna | +0.017 |
brushes | +0.017 |
bars | +0.017 |
types | +0.017 |
Neg logit contributions | |
shark | -0.009 |
Shark | -0.009 |
Bad | -0.008 |
nowhere | -0.008 |
Wally | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -3.71e-01 |
Pos logit contributions | |
shark | +0.010 |
Bad | +0.010 |
Shark | +0.009 |
ile | +0.009 |
coon | +0.009 |
Neg logit contributions | |
anna | -0.010 |
brushes | -0.010 |
bars | -0.009 |
types | -0.009 |
raid | -0.009 |
Token | James |
Token ID | 14731 |
Feature activation | -1.94e-01 |
Pos logit contributions | |
Bad | +0.013 |
shark | +0.013 |
Shark | +0.012 |
ile | +0.012 |
coon | +0.011 |
Neg logit contributions | |
brushes | -0.015 |
anna | -0.015 |
raid | -0.014 |
bars | -0.014 |
types | -0.014 |
Token | smiled |
Token ID | 13541 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
shark | +0.008 |
Bad | +0.008 |
Shark | +0.008 |
ile | +0.007 |
coon | +0.007 |
Neg logit contributions | |
brushes | -0.006 |
anna | -0.006 |
bars | -0.006 |
raid | -0.006 |
types | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | +2.63e-01 |
Pos logit contributions | |
brushes | +0.053 |
anna | +0.053 |
bars | +0.051 |
types | +0.050 |
raid | +0.049 |
Neg logit contributions | |
Bad | -0.056 |
ile | -0.054 |
shark | -0.053 |
coon | -0.053 |
Shark | -0.051 |
Token | and |
Token ID | 290 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
brushes | +0.011 |
raid | +0.011 |
anna | +0.011 |
types | +0.010 |
bars | +0.010 |
Neg logit contributions | |
shark | -0.009 |
Bad | -0.009 |
Shark | -0.008 |
fin | -0.008 |
nowhere | -0.008 |
Token | nodded |
Token ID | 14464 |
Feature activation | +2.69e+00 |
Pos logit contributions | |
anna | +0.041 |
brushes | +0.040 |
raid | +0.038 |
bars | +0.038 |
types | +0.037 |
Neg logit contributions | |
shark | -0.043 |
Bad | -0.042 |
Shark | -0.041 |
Fin | -0.039 |
fin | -0.039 |
Token | his |
Token ID | 465 |
Feature activation | +8.68e-01 |
Pos logit contributions | |
anna | +0.094 |
brushes | +0.093 |
bars | +0.088 |
Paula | +0.086 |
types | +0.086 |
Neg logit contributions | |
Bad | -0.106 |
ile | -0.103 |
shark | -0.102 |
coon | -0.100 |
Shark | -0.097 |
Token | head |
Token ID | 1182 |
Feature activation | +6.91e-01 |
Pos logit contributions | |
raid | +0.034 |
anna | +0.034 |
brushes | +0.033 |
bars | +0.033 |
types | +0.033 |
Neg logit contributions | |
shark | -0.030 |
fin | -0.029 |
Shark | -0.028 |
Bad | -0.027 |
Fin | -0.025 |
Token | |
Token ID | 220 |
Feature activation | +5.34e+00 |
Pos logit contributions | |
shark | +0.012 |
Bad | +0.012 |
fin | +0.011 |
Shark | +0.011 |
ile | +0.011 |
Neg logit contributions | |
anna | -0.013 |
brushes | -0.012 |
Paula | -0.011 |
Along | -0.011 |
ilda | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | -3.33e-01 |
Pos logit contributions | |
brushes | +0.234 |
anna | +0.232 |
bars | +0.228 |
raid | +0.227 |
types | +0.227 |
Neg logit contributions | |
Bad | -0.153 |
Shark | -0.149 |
shark | -0.147 |
Fin | -0.146 |
Wally | -0.141 |
Token | The |
Token ID | 464 |
Feature activation | +3.28e-01 |
Pos logit contributions | |
shark | +0.013 |
Shark | +0.012 |
fin | +0.012 |
Bad | +0.011 |
ile | +0.011 |
Neg logit contributions | |
anna | -0.013 |
brushes | -0.012 |
bars | -0.011 |
raid | -0.011 |
Paula | -0.011 |
Token | pupil |
Token ID | 34047 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
brushes | +0.010 |
anna | +0.010 |
types | +0.009 |
alphabet | +0.009 |
raid | +0.009 |
Neg logit contributions | |
shark | -0.014 |
Bad | -0.014 |
Shark | -0.014 |
ile | -0.013 |
fin | -0.013 |
Token | was |
Token ID | 373 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
shark | +0.078 |
Bad | +0.077 |
Shark | +0.075 |
ile | +0.073 |
coon | +0.072 |
Neg logit contributions | |
anna | -0.071 |
brushes | -0.071 |
bars | -0.065 |
raid | -0.065 |
types | -0.064 |
Token | so |
Token ID | 523 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
shark | +0.007 |
Bad | +0.007 |
Shark | +0.006 |
coon | +0.006 |
ile | +0.006 |
Neg logit contributions | |
anna | -0.004 |
brushes | -0.004 |
Paula | -0.004 |
bars | -0.004 |
raid | -0.004 |
Token | relieved |
Token ID | 22598 |
Feature activation | +8.09e-01 |
Pos logit contributions | |
anna | +0.075 |
brushes | +0.075 |
bars | +0.068 |
raid | +0.068 |
types | +0.067 |
Neg logit contributions | |
shark | -0.117 |
Bad | -0.115 |
Shark | -0.111 |
ile | -0.110 |
coon | -0.109 |
Token | and |
Token ID | 290 |
Feature activation | +5.85e-01 |
Pos logit contributions | |
anna | +0.034 |
brushes | +0.033 |
raid | +0.032 |
bars | +0.031 |
types | +0.031 |
Neg logit contributions | |
Bad | -0.026 |
shark | -0.026 |
ile | -0.025 |
Shark | -0.025 |
coon | -0.024 |
Token | grateful |
Token ID | 14066 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
brushes | +0.018 |
anna | +0.018 |
bars | +0.017 |
raid | +0.016 |
types | +0.016 |
Neg logit contributions | |
shark | -0.025 |
Bad | -0.025 |
Shark | -0.024 |
ile | -0.024 |
coon | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +6.28e-01 |
Pos logit contributions | |
anna | +0.038 |
brushes | +0.038 |
raid | +0.036 |
bars | +0.035 |
types | +0.035 |
Neg logit contributions | |
Bad | -0.042 |
shark | -0.042 |
Shark | -0.040 |
ile | -0.040 |
coon | -0.039 |
Token | She |
Token ID | 1375 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
brushes | +0.018 |
anna | +0.017 |
raid | +0.017 |
types | +0.016 |
bars | +0.016 |
Neg logit contributions | |
Fin | -0.028 |
Wally | -0.028 |
Shark | -0.028 |
shark | -0.028 |
Bad | -0.027 |
Token | add |
Token ID | 751 |
Feature activation | +8.99e-01 |
Pos logit contributions | |
anna | +0.046 |
brushes | +0.045 |
bars | +0.043 |
raid | +0.042 |
types | +0.041 |
Neg logit contributions | |
shark | -0.062 |
Bad | -0.061 |
Shark | -0.059 |
fin | -0.057 |
ile | -0.056 |
Token | numbers |
Token ID | 3146 |
Feature activation | +9.28e-01 |
Pos logit contributions | |
anna | +0.036 |
brushes | +0.034 |
bars | +0.033 |
raid | +0.033 |
types | +0.031 |
Neg logit contributions | |
shark | -0.031 |
Bad | -0.030 |
Shark | -0.029 |
ile | -0.028 |
fin | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
anna | +0.039 |
brushes | +0.038 |
raid | +0.036 |
types | +0.035 |
bars | +0.035 |
Neg logit contributions | |
shark | -0.029 |
Bad | -0.028 |
Shark | -0.027 |
ile | -0.027 |
ful | -0.026 |
Token | she |
Token ID | 673 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
shark | +0.008 |
Bad | +0.008 |
Shark | +0.007 |
fin | +0.007 |
Fin | +0.007 |
Neg logit contributions | |
anna | -0.009 |
raid | -0.009 |
brushes | -0.009 |
bars | -0.008 |
types | -0.008 |
Token | said |
Token ID | 531 |
Feature activation | +5.49e-01 |
Pos logit contributions | |
anna | +0.008 |
brushes | +0.007 |
raid | +0.007 |
bars | +0.007 |
types | +0.007 |
Neg logit contributions | |
shark | -0.008 |
Bad | -0.008 |
Shark | -0.008 |
ile | -0.008 |
fin | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | +5.15e-01 |
Pos logit contributions | |
brushes | +0.022 |
anna | +0.022 |
raid | +0.022 |
bars | +0.021 |
types | +0.021 |
Neg logit contributions | |
shark | -0.018 |
ful | -0.017 |
Bad | -0.017 |
Shark | -0.017 |
fin | -0.016 |
Token | answer |
Token ID | 3280 |
Feature activation | +3.75e-01 |
Pos logit contributions | |
anna | +0.018 |
brushes | +0.017 |
raid | +0.016 |
bars | +0.016 |
types | +0.015 |
Neg logit contributions | |
shark | -0.022 |
Bad | -0.021 |
Shark | -0.020 |
ile | -0.020 |
fin | -0.019 |
Token | out |
Token ID | 503 |
Feature activation | +5.08e-01 |
Pos logit contributions | |
anna | +0.015 |
brushes | +0.015 |
bars | +0.014 |
raid | +0.014 |
types | +0.014 |
Neg logit contributions | |
shark | -0.012 |
Bad | -0.012 |
Shark | -0.011 |
ile | -0.011 |
coon | -0.011 |
Token | loud |
Token ID | 7812 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
anna | +0.022 |
brushes | +0.021 |
bars | +0.021 |
raid | +0.020 |
types | +0.020 |
Neg logit contributions | |
shark | -0.016 |
Bad | -0.016 |
Shark | -0.015 |
ile | -0.014 |
coon | -0.014 |
Token | when |
Token ID | 618 |
Feature activation | +4.54e-01 |
Pos logit contributions | |
shark | +0.081 |
Bad | +0.079 |
Shark | +0.077 |
ile | +0.074 |
coon | +0.073 |
Neg logit contributions | |
brushes | -0.091 |
anna | -0.091 |
bars | -0.086 |
raid | -0.085 |
types | -0.084 |
Token | she |
Token ID | 673 |
Feature activation | -1.76e-02 |
Pos logit contributions | |
anna | +0.014 |
brushes | +0.012 |
raid | +0.012 |
bars | +0.012 |
types | +0.012 |
Neg logit contributions | |
shark | -0.021 |
Bad | -0.020 |
Shark | -0.020 |
Fin | -0.018 |
nowhere | -0.018 |
Token | said |
Token ID | 531 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
anna | +0.038 |
brushes | +0.037 |
bars | +0.036 |
raid | +0.035 |
types | +0.034 |
Neg logit contributions | |
shark | -0.041 |
Bad | -0.040 |
Shark | -0.039 |
ile | -0.038 |
coon | -0.037 |
Token | , |
Token ID | 11 |
Feature activation | -4.49e-01 |
Pos logit contributions | |
brushes | +0.066 |
anna | +0.065 |
types | +0.065 |
bars | +0.062 |
raid | +0.062 |
Neg logit contributions | |
shark | -0.028 |
Shark | -0.026 |
ful | -0.026 |
Bad | -0.025 |
Wally | -0.024 |
Token | " |
Token ID | 366 |
Feature activation | +6.12e-01 |
Pos logit contributions | |
shark | +0.019 |
Bad | +0.019 |
Shark | +0.019 |
Wally | +0.018 |
coon | +0.018 |
Neg logit contributions | |
brushes | -0.014 |
anna | -0.014 |
types | -0.013 |
raid | -0.013 |
bars | -0.013 |
Token | Okay |
Token ID | 16454 |
Feature activation | +3.81e-01 |
Pos logit contributions | |
anna | +0.027 |
brushes | +0.027 |
types | +0.026 |
bars | +0.026 |
raid | +0.025 |
Neg logit contributions | |
Bad | -0.021 |
Rex | -0.017 |
Shark | -0.017 |
shark | -0.017 |
coon | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
brushes | +0.019 |
anna | +0.019 |
bars | +0.019 |
types | +0.019 |
raid | +0.018 |
Neg logit contributions | |
shark | -0.009 |
Bad | -0.009 |
Shark | -0.008 |
Wally | -0.008 |
Fin | -0.008 |
Token | mom |
Token ID | 1995 |
Feature activation | +2.25e+00 |
Pos logit contributions | |
shark | +0.001 |
Bad | +0.001 |
Shark | +0.001 |
ile | +0.001 |
coon | +0.001 |
Neg logit contributions | |
anna | -0.001 |
brushes | -0.001 |
bars | -0.001 |
raid | -0.001 |
types | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +9.11e-01 |
Pos logit contributions | |
anna | +0.121 |
brushes | +0.118 |
bars | +0.115 |
rolling | +0.115 |
select | +0.114 |
Neg logit contributions | |
ile | -0.040 |
ful | -0.040 |
Bad | -0.037 |
shark | -0.037 |
Rex | -0.036 |
Token | That |
Token ID | 1320 |
Feature activation | +5.40e-02 |
Pos logit contributions | |
anna | +0.036 |
brushes | +0.034 |
types | +0.034 |
bars | +0.033 |
raid | +0.032 |
Neg logit contributions | |
Bad | -0.033 |
Shark | -0.032 |
shark | -0.031 |
Rex | -0.030 |
Fin | -0.030 |
Token | sounds |
Token ID | 5238 |
Feature activation | -2.67e-01 |
Pos logit contributions | |
anna | +0.002 |
raid | +0.002 |
brushes | +0.002 |
bars | +0.002 |
Paula | +0.002 |
Neg logit contributions | |
shark | -0.002 |
Bad | -0.002 |
Shark | -0.002 |
ile | -0.002 |
ball | -0.002 |
Token | fun |
Token ID | 1257 |
Feature activation | -5.06e-01 |
Pos logit contributions | |
shark | +0.011 |
Bad | +0.011 |
Shark | +0.011 |
ile | +0.011 |
coon | +0.010 |
Neg logit contributions | |
anna | -0.009 |
brushes | -0.009 |
bars | -0.008 |
raid | -0.008 |
types | -0.008 |
Token | ." |
Token ID | 526 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
shark | +0.018 |
Bad | +0.017 |
Shark | +0.017 |
ile | +0.017 |
coon | +0.016 |
Neg logit contributions | |
brushes | -0.020 |
anna | -0.019 |
bars | -0.019 |
raid | -0.019 |
types | -0.018 |
Token | The |
Token ID | 383 |
Feature activation | +3.51e-01 |
Pos logit contributions | |
Fin | +0.029 |
Shark | +0.028 |
shark | +0.027 |
Wally | +0.027 |
Bad | +0.026 |
Neg logit contributions | |
brushes | -0.027 |
bars | -0.025 |
raid | -0.025 |
types | -0.025 |
anna | -0.024 |
Token | airport |
Token ID | 9003 |
Feature activation | +9.26e-01 |
Pos logit contributions | |
brushes | +0.013 |
anna | +0.013 |
raid | +0.013 |
bars | +0.012 |
types | +0.012 |
Neg logit contributions | |
shark | -0.013 |
Bad | -0.013 |
Shark | -0.012 |
coon | -0.012 |
ile | -0.012 |
Token | was |
Token ID | 373 |
Feature activation | -6.79e-01 |
Pos logit contributions | |
brushes | +0.037 |
anna | +0.037 |
bars | +0.035 |
raid | +0.035 |
types | +0.034 |
Neg logit contributions | |
shark | -0.032 |
Bad | -0.032 |
Shark | -0.030 |
ile | -0.030 |
coon | -0.029 |
Token | big |
Token ID | 1263 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
Bad | +0.023 |
shark | +0.023 |
Shark | +0.021 |
ile | +0.021 |
nowhere | +0.021 |
Neg logit contributions | |
anna | -0.028 |
brushes | -0.027 |
raid | -0.027 |
bars | -0.026 |
types | -0.025 |
Token | and |
Token ID | 290 |
Feature activation | +5.10e-01 |
Pos logit contributions | |
anna | +0.007 |
brushes | +0.007 |
bars | +0.006 |
raid | +0.006 |
types | +0.006 |
Neg logit contributions | |
shark | -0.004 |
Bad | -0.004 |
Shark | -0.004 |
ile | -0.004 |
coon | -0.004 |
Token | gray |
Token ID | 12768 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
anna | +0.021 |
raid | +0.020 |
brushes | +0.019 |
bars | +0.018 |
types | +0.018 |
Neg logit contributions | |
Bad | -0.018 |
shark | -0.018 |
Shark | -0.017 |
nowhere | -0.017 |
Fin | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | -2.74e-01 |
Pos logit contributions | |
shark | +0.006 |
Bad | +0.006 |
Shark | +0.006 |
ile | +0.006 |
coon | +0.006 |
Neg logit contributions | |
brushes | -0.008 |
anna | -0.008 |
bars | -0.008 |
raid | -0.008 |
types | -0.008 |
Token | with |
Token ID | 351 |
Feature activation | -8.84e-01 |
Pos logit contributions | |
shark | +0.010 |
Bad | +0.010 |
Shark | +0.010 |
ile | +0.010 |
coon | +0.010 |
Neg logit contributions | |
brushes | -0.010 |
anna | -0.010 |
bars | -0.009 |
types | -0.009 |
raid | -0.009 |
Token | lots |
Token ID | 6041 |
Feature activation | +9.62e-01 |
Pos logit contributions | |
shark | +0.033 |
Bad | +0.033 |
Shark | +0.031 |
ile | +0.030 |
fin | +0.030 |
Neg logit contributions | |
anna | -0.033 |
brushes | -0.032 |
raid | -0.031 |
bars | -0.030 |
types | -0.029 |
Token | of |
Token ID | 286 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
anna | +0.041 |
brushes | +0.040 |
raid | +0.039 |
nie | +0.039 |
bars | +0.039 |
Neg logit contributions | |
shark | -0.030 |
Bad | -0.027 |
ile | -0.026 |
coon | -0.026 |
Fin | -0.025 |
Token | people |
Token ID | 661 |
Feature activation | +2.79e-02 |
Pos logit contributions | |
anna | +0.043 |
raid | +0.041 |
bars | +0.041 |
read | +0.039 |
ilda | +0.039 |
Neg logit contributions | |
shark | -0.046 |
Bad | -0.044 |
sharks | -0.041 |
hawk | -0.041 |
fin | -0.040 |
Token | puppies |
Token ID | 37793 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
shark | +0.025 |
Bad | +0.024 |
Shark | +0.023 |
ile | +0.023 |
coon | +0.023 |
Neg logit contributions | |
anna | -0.026 |
brushes | -0.026 |
bars | -0.024 |
types | -0.024 |
raid | -0.024 |
Token | barking |
Token ID | 48961 |
Feature activation | -3.70e+00 |
Pos logit contributions | |
Bad | +0.075 |
shark | +0.074 |
Shark | +0.070 |
ile | +0.070 |
coon | +0.067 |
Neg logit contributions | |
anna | -0.076 |
brushes | -0.075 |
bars | -0.072 |
raid | -0.071 |
types | -0.070 |
Token | , |
Token ID | 11 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
Bad | +0.144 |
shark | +0.138 |
ile | +0.135 |
Shark | +0.132 |
coon | +0.131 |
Neg logit contributions | |
anna | -0.131 |
brushes | -0.122 |
types | -0.119 |
sights | -0.118 |
Paula | -0.114 |
Token | people |
Token ID | 661 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
shark | +0.038 |
Bad | +0.038 |
ile | +0.037 |
Shark | +0.036 |
coon | +0.036 |
Neg logit contributions | |
brushes | -0.037 |
anna | -0.037 |
types | -0.034 |
bars | -0.034 |
Paula | -0.033 |
Token | talking |
Token ID | 3375 |
Feature activation | +4.73e-01 |
Pos logit contributions | |
anna | +0.007 |
brushes | +0.007 |
bars | +0.007 |
raid | +0.007 |
types | +0.007 |
Neg logit contributions | |
Bad | -0.007 |
shark | -0.007 |
Shark | -0.007 |
ile | -0.007 |
coon | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | -5.90e-01 |
Pos logit contributions | |
anna | +0.016 |
brushes | +0.016 |
bars | +0.015 |
types | +0.015 |
raid | +0.014 |
Neg logit contributions | |
Bad | -0.019 |
shark | -0.019 |
ile | -0.018 |
Shark | -0.018 |
coon | -0.018 |
Token | even |
Token ID | 772 |
Feature activation | -4.63e-01 |
Pos logit contributions | |
shark | +0.024 |
Bad | +0.024 |
ile | +0.023 |
Shark | +0.023 |
coon | +0.022 |
Neg logit contributions | |
brushes | -0.020 |
anna | -0.019 |
types | -0.018 |
bars | -0.018 |
raid | -0.018 |
Token | learned |
Token ID | 4499 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
shark | +0.018 |
Bad | +0.018 |
Shark | +0.017 |
ile | +0.017 |
coon | +0.017 |
Neg logit contributions | |
brushes | -0.017 |
anna | -0.016 |
types | -0.015 |
bars | -0.015 |
raid | -0.015 |
Token | to |
Token ID | 284 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
brushes | +0.052 |
anna | +0.049 |
types | +0.049 |
sights | +0.047 |
bars | +0.046 |
Neg logit contributions | |
shark | -0.057 |
Bad | -0.056 |
coon | -0.055 |
ile | -0.054 |
Shark | -0.054 |
Token | understand |
Token ID | 1833 |
Feature activation | -2.24e-02 |
Pos logit contributions | |
brushes | +0.030 |
anna | +0.029 |
raid | +0.027 |
types | +0.027 |
select | +0.026 |
Neg logit contributions | |
shark | -0.054 |
Bad | -0.053 |
Shark | -0.051 |
ile | -0.051 |
coon | -0.050 |
Token | what |
Token ID | 644 |
Feature activation | +5.33e-01 |
Pos logit contributions | |
Bad | +0.001 |
shark | +0.001 |
Shark | +0.001 |
coon | +0.001 |
ile | +0.001 |
Neg logit contributions | |
brushes | -0.001 |
anna | -0.001 |
types | -0.001 |
bars | -0.001 |
raid | -0.001 |
Token | birds |
Token ID | 10087 |
Feature activation | +2.93e-01 |
Pos logit contributions | |
brushes | +0.012 |
anna | +0.011 |
bars | +0.011 |
types | +0.011 |
raid | +0.011 |
Neg logit contributions | |
shark | -0.015 |
Bad | -0.015 |
Shark | -0.015 |
Fin | -0.015 |
ile | -0.015 |
Token | flying |
Token ID | 7348 |
Feature activation | +5.36e-01 |
Pos logit contributions | |
anna | +0.011 |
brushes | +0.011 |
bars | +0.011 |
raid | +0.011 |
types | +0.010 |
Neg logit contributions | |
shark | -0.010 |
Bad | -0.010 |
Shark | -0.010 |
ile | -0.010 |
coon | -0.009 |
Token | away |
Token ID | 1497 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
brushes | +0.018 |
bars | +0.016 |
types | +0.016 |
anna | +0.016 |
sights | +0.015 |
Neg logit contributions | |
shark | -0.022 |
Bad | -0.022 |
Shark | -0.022 |
coon | -0.021 |
fin | -0.020 |
Token | without |
Token ID | 1231 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
Bad | +0.054 |
shark | +0.052 |
Shark | +0.051 |
ile | +0.050 |
coon | +0.050 |
Neg logit contributions | |
brushes | -0.051 |
anna | -0.049 |
bars | -0.047 |
types | -0.047 |
raid | -0.045 |
Token | her |
Token ID | 607 |
Feature activation | +1.85e+00 |
Pos logit contributions | |
shark | +0.079 |
Bad | +0.077 |
Shark | +0.075 |
ile | +0.073 |
coon | +0.072 |
Neg logit contributions | |
brushes | -0.066 |
anna | -0.065 |
bars | -0.061 |
raid | -0.061 |
types | -0.060 |
Token | , |
Token ID | 11 |
Feature activation | -5.05e-01 |
Pos logit contributions | |
brushes | +0.058 |
raid | +0.056 |
anna | +0.056 |
bars | +0.053 |
alphabet | +0.050 |
Neg logit contributions | |
Bad | -0.081 |
shark | -0.079 |
Shark | -0.078 |
coon | -0.077 |
ile | -0.077 |
Token | the |
Token ID | 262 |
Feature activation | +9.23e-01 |
Pos logit contributions | |
shark | +0.026 |
Bad | +0.026 |
ile | +0.025 |
Shark | +0.025 |
coon | +0.025 |
Neg logit contributions | |
brushes | -0.012 |
anna | -0.011 |
bars | -0.010 |
types | -0.010 |
raid | -0.010 |
Token | loneliness |
Token ID | 37378 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
brushes | +0.032 |
anna | +0.030 |
bars | +0.028 |
types | +0.028 |
raid | +0.028 |
Neg logit contributions | |
shark | -0.037 |
Bad | -0.036 |
Shark | -0.036 |
ile | -0.034 |
coon | -0.034 |
Token | crept |
Token ID | 47092 |
Feature activation | +8.28e-01 |
Pos logit contributions | |
Bad | +0.077 |
shark | +0.077 |
Shark | +0.075 |
ile | +0.072 |
coon | +0.071 |
Neg logit contributions | |
brushes | -0.066 |
anna | -0.065 |
bars | -0.062 |
raid | -0.060 |
types | -0.060 |
Token | back |
Token ID | 736 |
Feature activation | -5.63e-01 |
Pos logit contributions | |
anna | +0.028 |
brushes | +0.027 |
bars | +0.027 |
Paula | +0.025 |
types | +0.025 |
Neg logit contributions | |
shark | -0.033 |
Bad | -0.032 |
Shark | -0.032 |
ile | -0.031 |
coon | -0.030 |
Token | in |
Token ID | 287 |
Feature activation | -5.33e-01 |
Pos logit contributions | |
shark | +0.023 |
Bad | +0.023 |
Shark | +0.022 |
ile | +0.021 |
coon | +0.021 |
Neg logit contributions | |
brushes | -0.019 |
anna | -0.019 |
bars | -0.018 |
raid | -0.018 |
types | -0.017 |
Token | my |
Token ID | 1820 |
Feature activation | +8.07e-01 |
Pos logit contributions | |
anna | +0.017 |
brushes | +0.017 |
bars | +0.016 |
raid | +0.015 |
types | +0.015 |
Neg logit contributions | |
shark | -0.035 |
Bad | -0.035 |
Shark | -0.034 |
ile | -0.033 |
coon | -0.033 |
Token | and |
Token ID | 290 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
brushes | +0.032 |
anna | +0.032 |
bars | +0.030 |
raid | +0.030 |
types | +0.029 |
Neg logit contributions | |
Bad | -0.029 |
shark | -0.028 |
Shark | -0.027 |
ile | -0.026 |
coon | -0.026 |
Token | daddy |
Token ID | 35822 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
shark | +0.007 |
Bad | +0.007 |
Shark | +0.007 |
Fin | +0.006 |
ful | +0.006 |
Neg logit contributions | |
anna | -0.006 |
raid | -0.005 |
bars | -0.005 |
brushes | -0.005 |
types | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | -7.90e-01 |
Pos logit contributions | |
brushes | +0.040 |
anna | +0.040 |
bars | +0.038 |
raid | +0.038 |
types | +0.037 |
Neg logit contributions | |
shark | -0.035 |
Bad | -0.034 |
Shark | -0.033 |
ile | -0.032 |
coon | -0.031 |
Token | One |
Token ID | 1881 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
shark | +0.031 |
Bad | +0.030 |
Shark | +0.030 |
Wally | +0.030 |
Fin | +0.029 |
Neg logit contributions | |
anna | -0.027 |
bars | -0.026 |
raid | -0.026 |
brushes | -0.025 |
types | -0.025 |
Token | day |
Token ID | 1110 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
brushes | +0.027 |
anna | +0.027 |
raid | +0.024 |
types | +0.024 |
bars | +0.024 |
Neg logit contributions | |
Bad | -0.060 |
shark | -0.060 |
Shark | -0.058 |
coon | -0.057 |
ile | -0.057 |
Token | , |
Token ID | 11 |
Feature activation | -8.87e-01 |
Pos logit contributions | |
shark | +0.031 |
Bad | +0.030 |
Shark | +0.029 |
ile | +0.027 |
Fin | +0.026 |
Neg logit contributions | |
anna | -0.049 |
brushes | -0.049 |
bars | -0.048 |
types | -0.047 |
raid | -0.046 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
shark | +0.033 |
Bad | +0.032 |
Shark | +0.031 |
ile | +0.031 |
coon | +0.030 |
Neg logit contributions | |
brushes | -0.034 |
anna | -0.033 |
bars | -0.031 |
raid | -0.031 |
types | -0.030 |
Token | and |
Token ID | 290 |
Feature activation | +3.32e-01 |
Pos logit contributions | |
shark | +0.005 |
Bad | +0.005 |
Shark | +0.005 |
ile | +0.005 |
coon | +0.005 |
Neg logit contributions | |
anna | -0.004 |
brushes | -0.004 |
bars | -0.003 |
raid | -0.003 |
types | -0.003 |
Token | her |
Token ID | 607 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
anna | +0.011 |
brushes | +0.010 |
bars | +0.010 |
raid | +0.010 |
types | +0.010 |
Neg logit contributions | |
shark | -0.014 |
Bad | -0.014 |
Shark | -0.013 |
Wally | -0.013 |
Fin | -0.013 |
Token | mom |
Token ID | 1995 |
Feature activation | +3.65e-01 |
Pos logit contributions | |
brushes | +0.073 |
anna | +0.071 |
bars | +0.067 |
raid | +0.066 |
types | +0.066 |
Neg logit contributions | |
Bad | -0.080 |
shark | -0.080 |
Shark | -0.077 |
ile | -0.076 |
coon | -0.075 |
Token | brought |
Token ID | 3181 |
Feature activation | +5.69e-01 |
Pos logit contributions | |
shark | +0.025 |
Bad | +0.025 |
Shark | +0.023 |
nowhere | +0.023 |
ile | +0.023 |
Neg logit contributions | |
anna | -0.033 |
raid | -0.032 |
bars | -0.032 |
brushes | -0.032 |
types | -0.031 |
Token | along |
Token ID | 1863 |
Feature activation | +6.42e-01 |
Pos logit contributions | |
anna | +0.026 |
brushes | +0.025 |
bars | +0.024 |
raid | +0.024 |
types | +0.023 |
Neg logit contributions | |
shark | -0.017 |
Bad | -0.017 |
Shark | -0.016 |
ile | -0.015 |
Fin | -0.015 |
Token | a |
Token ID | 257 |
Feature activation | -9.00e-01 |
Pos logit contributions | |
brushes | +0.028 |
anna | +0.028 |
bars | +0.027 |
raid | +0.026 |
types | +0.026 |
Neg logit contributions | |
shark | -0.020 |
Bad | -0.020 |
Shark | -0.019 |
ile | -0.018 |
coon | -0.018 |
Token | stuffed |
Token ID | 22259 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
shark | +0.035 |
Bad | +0.033 |
Shark | +0.032 |
fin | +0.031 |
ile | +0.031 |
Neg logit contributions | |
anna | -0.033 |
brushes | -0.032 |
bars | -0.031 |
raid | -0.030 |
types | -0.030 |
Token | animal |
Token ID | 5044 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
shark | +0.046 |
Bad | +0.044 |
Shark | +0.042 |
ile | +0.041 |
coon | +0.040 |
Neg logit contributions | |
brushes | -0.074 |
anna | -0.074 |
bars | -0.072 |
raid | -0.070 |
types | -0.069 |
Token | and |
Token ID | 290 |
Feature activation | +8.41e-02 |
Pos logit contributions | |
Bad | +0.111 |
shark | +0.106 |
Shark | +0.103 |
ile | +0.098 |
coon | +0.097 |
Neg logit contributions | |
brushes | -0.113 |
anna | -0.106 |
raid | -0.105 |
bars | -0.100 |
types | -0.099 |
Token | some |
Token ID | 617 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
brushes | +0.003 |
anna | +0.003 |
bars | +0.003 |
raid | +0.003 |
types | +0.003 |
Neg logit contributions | |
shark | -0.003 |
Bad | -0.003 |
Shark | -0.003 |
ile | -0.003 |
coon | -0.003 |
Token | snacks |
Token ID | 28487 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
brushes | +0.033 |
anna | +0.031 |
bars | +0.029 |
raid | +0.029 |
types | +0.029 |
Neg logit contributions | |
shark | -0.044 |
Bad | -0.043 |
Shark | -0.042 |
ile | -0.041 |
coon | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | -4.15e-01 |
Pos logit contributions | |
brushes | +0.007 |
bars | +0.006 |
anna | +0.006 |
types | +0.006 |
raid | +0.006 |
Neg logit contributions | |
Bad | -0.009 |
shark | -0.008 |
Shark | -0.008 |
ile | -0.008 |
coon | -0.008 |
Token | He |
Token ID | 679 |
Feature activation | -4.45e-01 |
Pos logit contributions | |
Fin | +0.012 |
Shark | +0.011 |
shark | +0.010 |
Rex | +0.010 |
Wally | +0.010 |
Neg logit contributions | |
anna | -0.021 |
raid | -0.020 |
nie | -0.020 |
ilda | -0.020 |
brushes | -0.019 |
Token | had |
Token ID | 550 |
Feature activation | -6.24e-01 |
Pos logit contributions | |
shark | +0.017 |
Bad | +0.017 |
Shark | +0.016 |
ile | +0.016 |
fin | +0.015 |
Neg logit contributions | |
anna | -0.016 |
brushes | -0.016 |
bars | -0.015 |
raid | -0.015 |
types | -0.014 |
Token | felt |
Token ID | 2936 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
anna | +0.042 |
brushes | +0.041 |
types | +0.037 |
bars | +0.037 |
raid | +0.037 |
Neg logit contributions | |
Bad | -0.054 |
shark | -0.053 |
ile | -0.051 |
Shark | -0.051 |
coon | -0.050 |
Token | better |
Token ID | 1365 |
Feature activation | -8.30e-01 |
Pos logit contributions | |
com | +0.003 |
anna | +0.003 |
brushes | +0.003 |
sights | +0.003 |
necessary | +0.003 |
Neg logit contributions | |
ile | -0.006 |
shark | -0.006 |
coon | -0.006 |
Shark | -0.006 |
Bad | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | +8.38e-01 |
Pos logit contributions | |
ile | +0.038 |
shark | +0.037 |
Bad | +0.036 |
coon | +0.036 |
Fin | +0.035 |
Neg logit contributions | |
anna | -0.024 |
raid | -0.022 |
work | -0.020 |
select | -0.019 |
read | -0.019 |
Token | asked |
Token ID | 1965 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
brushes | +0.029 |
anna | +0.029 |
bars | +0.027 |
types | +0.026 |
raid | +0.026 |
Neg logit contributions | |
shark | -0.033 |
Bad | -0.033 |
Shark | -0.032 |
ile | -0.031 |
coon | -0.031 |
Token | the |
Token ID | 262 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
anna | +0.066 |
brushes | +0.066 |
types | +0.063 |
bars | +0.062 |
raid | +0.062 |
Neg logit contributions | |
Bad | -0.062 |
shark | -0.062 |
ile | -0.058 |
Shark | -0.058 |
coon | -0.056 |
Token | cop |
Token ID | 2243 |
Feature activation | -3.44e+00 |
Pos logit contributions | |
shark | +0.006 |
Bad | +0.006 |
Shark | +0.006 |
ile | +0.006 |
fin | +0.006 |
Neg logit contributions | |
brushes | -0.006 |
anna | -0.006 |
bars | -0.005 |
types | -0.005 |
raid | -0.005 |
Token | if |
Token ID | 611 |
Feature activation | -8.75e-01 |
Pos logit contributions | |
shark | +0.124 |
Shark | +0.118 |
Bad | +0.118 |
ile | +0.113 |
Fin | +0.111 |
Neg logit contributions | |
brushes | -0.134 |
anna | -0.133 |
bars | -0.126 |
raid | -0.123 |
types | -0.121 |
Token | he |
Token ID | 339 |
Feature activation | -2.30e+00 |
Pos logit contributions | |
shark | +0.031 |
Bad | +0.031 |
Shark | +0.030 |
ile | +0.029 |
coon | +0.028 |
Neg logit contributions | |
anna | -0.034 |
brushes | -0.034 |
bars | -0.032 |
raid | -0.032 |
types | -0.031 |
Token | wanted |
Token ID | 2227 |
Feature activation | -3.55e-01 |
Pos logit contributions | |
Bad | +0.098 |
shark | +0.097 |
Shark | +0.091 |
coon | +0.090 |
ile | +0.090 |
Neg logit contributions | |
anna | -0.073 |
brushes | -0.072 |
raid | -0.069 |
bars | -0.067 |
types | -0.065 |
Token | to |
Token ID | 284 |
Feature activation | +6.23e-01 |
Pos logit contributions | |
Bad | +0.017 |
shark | +0.017 |
ile | +0.017 |
Shark | +0.017 |
coon | +0.016 |
Neg logit contributions | |
brushes | -0.009 |
anna | -0.009 |
raid | -0.008 |
bars | -0.008 |
types | -0.008 |
Token | play |
Token ID | 711 |
Feature activation | -5.16e-01 |
Pos logit contributions | |
brushes | +0.017 |
anna | +0.017 |
raid | +0.016 |
bars | +0.015 |
types | +0.015 |
Neg logit contributions | |
shark | -0.029 |
Bad | -0.029 |
Shark | -0.028 |
ile | -0.028 |
coon | -0.027 |
Token | the |
Token ID | 262 |
Feature activation | +6.89e-01 |
Pos logit contributions | |
Bad | +0.061 |
shark | +0.059 |
ile | +0.059 |
Shark | +0.058 |
coon | +0.058 |
Neg logit contributions | |
brushes | -0.051 |
anna | -0.046 |
bars | -0.046 |
types | -0.044 |
raid | -0.044 |
Token | volcano |
Token ID | 32410 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
anna | +0.035 |
brushes | +0.034 |
raid | +0.033 |
bars | +0.033 |
types | +0.032 |
Neg logit contributions | |
shark | -0.018 |
Bad | -0.017 |
Shark | -0.016 |
ile | -0.015 |
coon | -0.015 |
Token | is |
Token ID | 318 |
Feature activation | -9.06e-01 |
Pos logit contributions | |
shark | +0.053 |
Bad | +0.053 |
Shark | +0.051 |
ile | +0.049 |
coon | +0.049 |
Neg logit contributions | |
brushes | -0.043 |
anna | -0.043 |
bars | -0.040 |
types | -0.039 |
raid | -0.039 |
Token | going |
Token ID | 1016 |
Feature activation | +2.57e+00 |
Pos logit contributions | |
shark | +0.034 |
Bad | +0.034 |
Shark | +0.033 |
coon | +0.032 |
ile | +0.032 |
Neg logit contributions | |
brushes | -0.030 |
anna | -0.029 |
types | -0.028 |
decorations | -0.027 |
work | -0.027 |
Token | to |
Token ID | 284 |
Feature activation | +3.03e-01 |
Pos logit contributions | |
brushes | +0.089 |
anna | +0.087 |
bars | +0.084 |
raid | +0.083 |
types | +0.081 |
Neg logit contributions | |
shark | -0.104 |
Bad | -0.103 |
Shark | -0.099 |
ile | -0.096 |
Fin | -0.095 |
Token | explode |
Token ID | 22818 |
Feature activation | -3.82e+00 |
Pos logit contributions | |
brushes | +0.009 |
work | +0.008 |
types | +0.008 |
anna | +0.008 |
select | +0.008 |
Neg logit contributions | |
shark | -0.013 |
Bad | -0.013 |
Shark | -0.012 |
ile | -0.012 |
coon | -0.012 |
Token | ! |
Token ID | 0 |
Feature activation | -7.55e-01 |
Pos logit contributions | |
Bad | +0.177 |
Shark | +0.176 |
shark | +0.175 |
coon | +0.163 |
ile | +0.163 |
Neg logit contributions | |
anna | -0.112 |
brushes | -0.102 |
nie | -0.100 |
bars | -0.096 |
types | -0.095 |
Token | We |
Token ID | 775 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
shark | +0.034 |
Bad | +0.034 |
ile | +0.033 |
coon | +0.032 |
Shark | +0.032 |
Neg logit contributions | |
anna | -0.021 |
brushes | -0.021 |
bars | -0.020 |
types | -0.018 |
raid | -0.018 |
Token | need |
Token ID | 761 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
shark | +0.070 |
Bad | +0.070 |
Shark | +0.067 |
ile | +0.066 |
coon | +0.065 |
Neg logit contributions | |
brushes | -0.049 |
anna | -0.049 |
raid | -0.045 |
bars | -0.045 |
types | -0.044 |
Token | to |
Token ID | 284 |
Feature activation | +8.74e-01 |
Pos logit contributions | |
shark | +0.074 |
Bad | +0.074 |
Shark | +0.071 |
ile | +0.070 |
coon | +0.070 |
Neg logit contributions | |
brushes | -0.040 |
anna | -0.039 |
bars | -0.036 |
raid | -0.036 |
types | -0.035 |
Token | save |
Token ID | 3613 |
Feature activation | -8.76e-01 |
Pos logit contributions | |
brushes | +0.020 |
anna | +0.019 |
work | +0.018 |
select | +0.018 |
raid | +0.018 |
Neg logit contributions | |
shark | -0.044 |
Bad | -0.044 |
ile | -0.043 |
Shark | -0.042 |
coon | -0.041 |
Token |
|
Token ID | 198 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
shark | +0.041 |
Shark | +0.041 |
Fin | +0.040 |
Wally | +0.039 |
Bad | +0.037 |
Neg logit contributions | |
anna | -0.053 |
raid | -0.052 |
bars | -0.051 |
brushes | -0.051 |
types | -0.049 |
Token |
|
Token ID | 198 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
shark | +0.062 |
Bad | +0.061 |
Shark | +0.059 |
ile | +0.057 |
coon | +0.056 |
Neg logit contributions | |
brushes | -0.076 |
anna | -0.075 |
bars | -0.072 |
raid | -0.072 |
types | -0.070 |
Token | His |
Token ID | 6653 |
Feature activation | -6.59e-01 |
Pos logit contributions | |
shark | +0.065 |
Bad | +0.064 |
Shark | +0.062 |
ile | +0.060 |
coon | +0.059 |
Neg logit contributions | |
brushes | -0.065 |
anna | -0.065 |
bars | -0.062 |
raid | -0.061 |
types | -0.060 |
Token | mum |
Token ID | 25682 |
Feature activation | +7.82e-01 |
Pos logit contributions | |
Bad | +0.034 |
coon | +0.033 |
ile | +0.032 |
Rex | +0.029 |
nowhere | +0.029 |
Neg logit contributions | |
brushes | -0.018 |
anna | -0.015 |
types | -0.015 |
raid | -0.015 |
decorations | -0.014 |
Token | said |
Token ID | 531 |
Feature activation | -3.65e-01 |
Pos logit contributions | |
brushes | +0.026 |
anna | +0.026 |
bars | +0.025 |
raid | +0.025 |
types | +0.024 |
Neg logit contributions | |
shark | -0.032 |
Bad | -0.031 |
Shark | -0.030 |
ile | -0.030 |
coon | -0.029 |
Token | no |
Token ID | 645 |
Feature activation | -3.89e+00 |
Pos logit contributions | |
Bad | +0.013 |
shark | +0.013 |
ile | +0.012 |
Shark | +0.012 |
coon | +0.012 |
Neg logit contributions | |
anna | -0.014 |
brushes | -0.014 |
bars | -0.013 |
raid | -0.013 |
types | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | +9.10e-02 |
Pos logit contributions | |
shark | +0.098 |
Bad | +0.091 |
Shark | +0.090 |
Fin | +0.086 |
ful | +0.084 |
Neg logit contributions | |
brushes | -0.190 |
anna | -0.182 |
bars | -0.178 |
cere | -0.174 |
select | -0.174 |
Token | told |
Token ID | 1297 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
brushes | +0.003 |
anna | +0.003 |
bars | +0.003 |
types | +0.003 |
Paula | +0.003 |
Neg logit contributions | |
shark | -0.003 |
Bad | -0.003 |
ile | -0.003 |
Shark | -0.003 |
coon | -0.003 |
Token | him |
Token ID | 683 |
Feature activation | -2.97e+00 |
Pos logit contributions | |
brushes | +0.008 |
anna | +0.008 |
types | +0.007 |
work | +0.007 |
Paula | +0.007 |
Neg logit contributions | |
Bad | -0.007 |
shark | -0.007 |
ile | -0.007 |
Shark | -0.007 |
hawk | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | -6.12e-01 |
Pos logit contributions | |
shark | +0.110 |
Bad | +0.106 |
Shark | +0.104 |
ile | +0.099 |
Fin | +0.099 |
Neg logit contributions | |
anna | -0.111 |
brushes | -0.111 |
bars | -0.105 |
raid | -0.104 |
types | -0.103 |
Token | stay |
Token ID | 2652 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
shark | +0.029 |
Bad | +0.029 |
ile | +0.028 |
Shark | +0.028 |
coon | +0.027 |
Neg logit contributions | |
brushes | -0.017 |
anna | -0.016 |
raid | -0.015 |
bars | -0.015 |
types | -0.014 |
Token | wheel |
Token ID | 7825 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
anna | +0.092 |
brushes | +0.091 |
raid | +0.086 |
bars | +0.085 |
types | +0.083 |
Neg logit contributions | |
shark | -0.085 |
Bad | -0.081 |
Shark | -0.079 |
ile | -0.079 |
coon | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | +4.63e-02 |
Pos logit contributions | |
anna | +0.061 |
brushes | +0.060 |
raid | +0.056 |
bars | +0.056 |
types | +0.055 |
Neg logit contributions | |
shark | -0.062 |
Bad | -0.061 |
Shark | -0.059 |
ile | -0.058 |
coon | -0.057 |
Token | Not |
Token ID | 1892 |
Feature activation | -7.94e-01 |
Pos logit contributions | |
brushes | +0.002 |
anna | +0.002 |
types | +0.002 |
bars | +0.002 |
raid | +0.002 |
Neg logit contributions | |
Shark | -0.002 |
Fin | -0.002 |
Bad | -0.002 |
Wally | -0.002 |
shark | -0.002 |
Token | wanting |
Token ID | 10291 |
Feature activation | +3.54e-01 |
Pos logit contributions | |
shark | +0.028 |
Bad | +0.027 |
Shark | +0.026 |
ile | +0.026 |
Fin | +0.025 |
Neg logit contributions | |
anna | -0.031 |
brushes | -0.031 |
bars | -0.029 |
raid | -0.029 |
types | -0.028 |
Token | her |
Token ID | 607 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
anna | +0.011 |
brushes | +0.011 |
bars | +0.010 |
raid | +0.010 |
types | +0.010 |
Neg logit contributions | |
shark | -0.015 |
Bad | -0.015 |
Shark | -0.015 |
ile | -0.014 |
Fin | -0.014 |
Token | furry |
Token ID | 40857 |
Feature activation | -3.14e+00 |
Pos logit contributions | |
anna | +0.056 |
bars | +0.051 |
brushes | +0.050 |
types | +0.050 |
nie | +0.048 |
Neg logit contributions | |
shark | -0.065 |
Bad | -0.063 |
Shark | -0.062 |
fin | -0.061 |
Fin | -0.059 |
Token | friend |
Token ID | 1545 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
shark | +0.125 |
Bad | +0.124 |
Shark | +0.119 |
ile | +0.117 |
coon | +0.115 |
Neg logit contributions | |
brushes | -0.108 |
anna | -0.107 |
bars | -0.101 |
raid | -0.100 |
types | -0.098 |
Token | to |
Token ID | 284 |
Feature activation | +7.59e-01 |
Pos logit contributions | |
shark | +0.049 |
Bad | +0.048 |
Shark | +0.046 |
ile | +0.045 |
coon | +0.045 |
Neg logit contributions | |
anna | -0.043 |
brushes | -0.043 |
bars | -0.040 |
raid | -0.040 |
types | -0.039 |
Token | get |
Token ID | 651 |
Feature activation | +7.82e-02 |
Pos logit contributions | |
brushes | +0.021 |
anna | +0.020 |
raid | +0.019 |
bars | +0.019 |
types | +0.019 |
Neg logit contributions | |
shark | -0.035 |
Bad | -0.035 |
Shark | -0.034 |
ile | -0.034 |
coon | -0.033 |
Token | hurt |
Token ID | 5938 |
Feature activation | -2.99e+00 |
Pos logit contributions | |
anna | +0.003 |
brushes | +0.003 |
bars | +0.003 |
raid | +0.003 |
types | +0.003 |
Neg logit contributions | |
shark | -0.003 |
Bad | -0.003 |
Shark | -0.003 |
ile | -0.002 |
fin | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -4.34e-01 |
Pos logit contributions | |
shark | +0.118 |
Bad | +0.117 |
Shark | +0.113 |
ile | +0.108 |
fin | +0.107 |
Neg logit contributions | |
anna | -0.102 |
brushes | -0.100 |
bars | -0.098 |
raid | -0.096 |
types | -0.095 |
Token | She |
Token ID | 1375 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
Shark | +0.013 |
Fin | +0.012 |
shark | +0.012 |
Wally | +0.012 |
Bad | +0.012 |
Neg logit contributions | |
brushes | -0.011 |
anna | -0.011 |
types | -0.011 |
bars | -0.011 |
raid | -0.010 |
Token | started |
Token ID | 2067 |
Feature activation | +9.68e-01 |
Pos logit contributions | |
brushes | +0.051 |
anna | +0.050 |
raid | +0.050 |
bars | +0.050 |
types | +0.047 |
Neg logit contributions | |
shark | -0.051 |
Bad | -0.049 |
Shark | -0.048 |
ile | -0.046 |
fin | -0.046 |
Token | to |
Token ID | 284 |
Feature activation | +6.69e-01 |
Pos logit contributions | |
anna | +0.037 |
raid | +0.035 |
brushes | +0.034 |
bars | +0.034 |
types | +0.033 |
Neg logit contributions | |
shark | -0.035 |
Shark | -0.032 |
fin | -0.031 |
Bad | -0.031 |
ful | -0.031 |
Token | chew |
Token ID | 34722 |
Feature activation | -3.11e+00 |
Pos logit contributions | |
anna | +0.021 |
brushes | +0.021 |
bars | +0.020 |
raid | +0.020 |
types | +0.019 |
Neg logit contributions | |
shark | -0.028 |
Bad | -0.028 |
Shark | -0.027 |
fin | -0.026 |
ile | -0.025 |
Token | the |
Token ID | 262 |
Feature activation | +6.29e-01 |
Pos logit contributions | |
shark | +0.118 |
Bad | +0.116 |
Shark | +0.114 |
ile | +0.111 |
fin | +0.106 |
Neg logit contributions | |
brushes | -0.117 |
anna | -0.110 |
bars | -0.108 |
raid | -0.105 |
types | -0.105 |
Token | tuna |
Token ID | 38883 |
Feature activation | -3.37e+00 |
Pos logit contributions | |
brushes | +0.021 |
anna | +0.020 |
bars | +0.019 |
raid | +0.019 |
types | +0.019 |
Neg logit contributions | |
shark | -0.026 |
Bad | -0.026 |
Shark | -0.025 |
ile | -0.024 |
coon | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
Bad | +0.136 |
shark | +0.125 |
Shark | +0.124 |
coon | +0.121 |
ile | +0.119 |
Neg logit contributions | |
brushes | -0.120 |
anna | -0.114 |
types | -0.112 |
bars | -0.111 |
raid | -0.109 |
Token | and |
Token ID | 290 |
Feature activation | -2.29e-01 |
Pos logit contributions | |
shark | +0.040 |
Bad | +0.039 |
Shark | +0.038 |
ile | +0.038 |
coon | +0.037 |
Neg logit contributions | |
brushes | -0.039 |
anna | -0.038 |
bars | -0.036 |
raid | -0.035 |
types | -0.035 |
Token | wow |
Token ID | 28796 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
shark | +0.009 |
Bad | +0.009 |
Shark | +0.009 |
ile | +0.009 |
coon | +0.009 |
Neg logit contributions | |
brushes | -0.008 |
anna | -0.008 |
bars | -0.007 |
raid | -0.007 |
types | -0.007 |
Token | - |
Token ID | 532 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
anna | +0.064 |
brushes | +0.064 |
bars | +0.061 |
raid | +0.061 |
types | +0.060 |
Neg logit contributions | |
shark | -0.047 |
Bad | -0.046 |
Shark | -0.044 |
ile | -0.043 |
coon | -0.042 |
Token | it |
Token ID | 340 |
Feature activation | +5.14e-01 |
Pos logit contributions | |
brushes | +0.004 |
anna | +0.004 |
bars | +0.004 |
raid | +0.004 |
types | +0.004 |
Neg logit contributions | |
shark | -0.004 |
Bad | -0.004 |
Shark | -0.004 |
ile | -0.004 |
coon | -0.004 |