Token | pay |
Token ID | 1414 |
Feature activation | -3.74e+00 |
Pos logit contributions | |
ectar | +0.173 |
ilies | +0.172 |
ppy | +0.169 |
per | +0.168 |
opsy | +0.166 |
Neg logit contributions | |
Paul | -0.154 |
Paula | -0.150 |
John | -0.146 |
irens | -0.145 |
Sarah | -0.138 |
Token | me |
Token ID | 502 |
Feature activation | +4.68e+00 |
Pos logit contributions | |
irens | +0.295 |
Paul | +0.287 |
Paula | +0.286 |
John | +0.277 |
Sarah | +0.269 |
Neg logit contributions | |
opsy | -0.237 |
comb | -0.234 |
ectar | -0.232 |
ppy | -0.232 |
ilies | -0.230 |
Token | with |
Token ID | 351 |
Feature activation | -2.91e-01 |
Pos logit contributions | |
ilies | +0.296 |
ppy | +0.295 |
opsy | +0.295 |
ectar | +0.293 |
per | +0.290 |
Neg logit contributions | |
irens | -0.359 |
Paula | -0.349 |
Paul | -0.348 |
John | -0.337 |
iren | -0.325 |
Token | some |
Token ID | 617 |
Feature activation | +3.46e+00 |
Pos logit contributions | |
irens | +0.024 |
Paula | +0.023 |
Paul | +0.022 |
iren | +0.021 |
John | +0.021 |
Neg logit contributions | |
per | -0.018 |
ppy | -0.018 |
pper | -0.017 |
ilies | -0.017 |
opsy | -0.017 |
Token | honey |
Token ID | 12498 |
Feature activation | +5.91e+00 |
Pos logit contributions | |
ppy | +0.245 |
per | +0.243 |
pper | +0.243 |
opsy | +0.241 |
ilies | +0.234 |
Neg logit contributions | |
irens | -0.250 |
Paula | -0.238 |
Paul | -0.232 |
iren | -0.226 |
John | -0.222 |
Token | comb |
Token ID | 24011 |
Feature activation | +1.16e+01 |
Pos logit contributions | |
opsy | +0.411 |
ilies | +0.405 |
per | +0.389 |
ectar | +0.386 |
ppy | +0.386 |
Neg logit contributions | |
irens | -0.399 |
Paula | -0.398 |
Paul | -0.392 |
John | -0.385 |
Sarah | -0.374 |
Token | . |
Token ID | 13 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
opsy | +0.757 |
ppy | +0.742 |
ilies | +0.742 |
per | +0.739 |
pper | +0.738 |
Neg logit contributions | |
irens | -0.838 |
Paula | -0.830 |
Paul | -0.810 |
John | -0.805 |
Sarah | -0.783 |
Token | â |
Token ID | 22940 |
Feature activation | +2.93e+00 |
Pos logit contributions | |
irens | +0.019 |
Paula | +0.018 |
Paul | +0.018 |
John | +0.017 |
iren | +0.017 |
Neg logit contributions | |
per | -0.017 |
ppy | -0.017 |
pper | -0.017 |
opsy | -0.017 |
ilies | -0.017 |
Token | € |
Token ID | 26391 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
per | +0.196 |
ilies | +0.194 |
opsy | +0.194 |
ppy | +0.194 |
pper | +0.190 |
Neg logit contributions | |
irens | -0.222 |
Paula | -0.215 |
Paul | -0.207 |
iren | -0.203 |
John | -0.198 |
Token | John |
Token ID | 1757 |
Feature activation | -8.20e+00 |
Pos logit contributions | |
per | +0.139 |
pper | +0.136 |
ppy | +0.135 |
opsy | +0.132 |
ilies | +0.128 |
Neg logit contributions | |
irens | -0.164 |
Paula | -0.160 |
iren | -0.149 |
spe | -0.147 |
Paul | -0.147 |
Token | agreed |
Token ID | 4987 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
Paula | +0.492 |
Paul | +0.486 |
irens | +0.484 |
John | +0.470 |
spe | +0.456 |
Neg logit contributions | |
ilies | -0.640 |
ectar | -0.632 |
per | -0.625 |
opsy | -0.621 |
ppy | -0.607 |
Token | She |
Token ID | 3347 |
Feature activation | +3.57e+00 |
Pos logit contributions | |
irens | +0.036 |
iren | +0.033 |
Sarah | +0.032 |
Paula | +0.031 |
spe | +0.031 |
Neg logit contributions | |
comb | -0.030 |
opsy | -0.029 |
ilies | -0.029 |
ppy | -0.028 |
thirst | -0.028 |
Token | gave |
Token ID | 2921 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
per | +0.252 |
opsy | +0.251 |
ppy | +0.251 |
ilies | +0.250 |
pper | +0.249 |
Neg logit contributions | |
irens | -0.251 |
Paula | -0.242 |
Paul | -0.232 |
iren | -0.225 |
John | -0.221 |
Token | them |
Token ID | 606 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
irens | +0.147 |
Paula | +0.142 |
Paul | +0.135 |
iren | +0.131 |
spe | +0.129 |
Neg logit contributions | |
per | -0.157 |
ppy | -0.155 |
ilies | -0.154 |
pper | -0.153 |
opsy | -0.153 |
Token | a |
Token ID | 257 |
Feature activation | +9.13e-01 |
Pos logit contributions | |
irens | +0.094 |
Paula | +0.092 |
John | +0.091 |
Paul | +0.091 |
Sarah | +0.090 |
Neg logit contributions | |
per | -0.089 |
pper | -0.088 |
comb | -0.087 |
opsy | -0.086 |
ppy | -0.084 |
Token | honey |
Token ID | 12498 |
Feature activation | +5.86e+00 |
Pos logit contributions | |
per | +0.064 |
ppy | +0.063 |
pper | +0.063 |
ilies | +0.062 |
opsy | +0.061 |
Neg logit contributions | |
irens | -0.066 |
Paula | -0.064 |
Paul | -0.061 |
iren | -0.061 |
flashing | -0.059 |
Token | comb |
Token ID | 24011 |
Feature activation | +1.16e+01 |
Pos logit contributions | |
opsy | +0.437 |
ilies | +0.436 |
izards | +0.408 |
ars | +0.402 |
per | +0.398 |
Neg logit contributions | |
Sarah | -0.315 |
Smith | -0.312 |
John | -0.300 |
bul | -0.297 |
Paula | -0.293 |
Token | . |
Token ID | 13 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
opsy | +0.846 |
ilies | +0.826 |
ppy | +0.823 |
pper | +0.816 |
per | +0.794 |
Neg logit contributions | |
Sarah | -0.666 |
John | -0.665 |
irens | -0.655 |
flashing | -0.648 |
Paula | -0.644 |
Token | It |
Token ID | 632 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
irens | +0.014 |
Paula | +0.013 |
Paul | +0.013 |
Sarah | +0.012 |
John | +0.012 |
Neg logit contributions | |
pper | -0.017 |
ppy | -0.017 |
per | -0.016 |
opsy | -0.016 |
ilies | -0.016 |
Token | was |
Token ID | 373 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
per | +0.227 |
ilies | +0.218 |
pper | +0.216 |
opsy | +0.214 |
ppy | +0.214 |
Neg logit contributions | |
irens | -0.074 |
Paula | -0.073 |
Sarah | -0.069 |
John | -0.068 |
flashing | -0.067 |
Token | sweet |
Token ID | 6029 |
Feature activation | +5.69e-01 |
Pos logit contributions | |
Sarah | +0.103 |
John | +0.101 |
Paula | +0.101 |
Paul | +0.097 |
flashing | +0.096 |
Neg logit contributions | |
per | -0.111 |
ppy | -0.105 |
pper | -0.105 |
ive | -0.104 |
ilies | -0.101 |
Token | and |
Token ID | 290 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
opsy | +0.040 |
pper | +0.040 |
ilies | +0.040 |
per | +0.040 |
ppy | +0.039 |
Neg logit contributions | |
irens | -0.039 |
John | -0.037 |
Paula | -0.036 |
Sarah | -0.036 |
Paul | -0.036 |
Token | crawled |
Token ID | 45668 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
ectar | +0.062 |
ilies | +0.061 |
pper | +0.061 |
per | +0.059 |
opsy | +0.057 |
Neg logit contributions | |
Paula | -0.034 |
Paul | -0.033 |
John | -0.032 |
Sarah | -0.032 |
messed | -0.031 |
Token | into |
Token ID | 656 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
pper | +0.191 |
opsy | +0.183 |
per | +0.179 |
ppy | +0.179 |
ilies | +0.177 |
Neg logit contributions | |
irens | -0.138 |
Paula | -0.136 |
Sarah | -0.135 |
Paul | -0.128 |
John | -0.128 |
Token | her |
Token ID | 607 |
Feature activation | +4.05e-01 |
Pos logit contributions | |
per | +0.010 |
opsy | +0.009 |
ilies | +0.009 |
ppy | +0.009 |
pper | +0.009 |
Neg logit contributions | |
irens | -0.011 |
Paula | -0.011 |
Paul | -0.010 |
iren | -0.010 |
spe | -0.010 |
Token | little |
Token ID | 1310 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
pper | +0.027 |
ppy | +0.026 |
per | +0.026 |
ilies | +0.025 |
ily | +0.024 |
Neg logit contributions | |
Paula | -0.029 |
irens | -0.028 |
Sarah | -0.028 |
John | -0.027 |
flashing | -0.027 |
Token | honey |
Token ID | 12498 |
Feature activation | +5.93e+00 |
Pos logit contributions | |
irens | +0.156 |
Paula | +0.153 |
Paul | +0.144 |
Sarah | +0.144 |
John | +0.142 |
Neg logit contributions | |
pper | -0.146 |
ppy | -0.134 |
ilies | -0.134 |
per | -0.133 |
ily | -0.128 |
Token | comb |
Token ID | 24011 |
Feature activation | +1.15e+01 |
Pos logit contributions | |
ilies | +0.411 |
ily | +0.383 |
ppy | +0.371 |
pper | +0.363 |
izards | +0.360 |
Neg logit contributions | |
irens | -0.401 |
Paula | -0.391 |
flashing | -0.382 |
Paul | -0.370 |
Sarah | -0.367 |
Token | bed |
Token ID | 3996 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
ilies | +0.808 |
ppy | +0.802 |
pper | +0.791 |
opsy | +0.772 |
ily | +0.753 |
Neg logit contributions | |
irens | -0.770 |
Paula | -0.739 |
flashing | -0.736 |
Sarah | -0.721 |
John | -0.719 |
Token | and |
Token ID | 290 |
Feature activation | -6.78e-02 |
Pos logit contributions | |
ilies | +0.102 |
ppy | +0.102 |
per | +0.102 |
opsy | +0.102 |
pper | +0.102 |
Neg logit contributions | |
irens | -0.148 |
Paula | -0.144 |
Paul | -0.139 |
iren | -0.135 |
John | -0.134 |
Token | fell |
Token ID | 3214 |
Feature activation | -5.15e-01 |
Pos logit contributions | |
Sarah | +0.004 |
John | +0.004 |
Paula | +0.004 |
Paul | +0.004 |
James | +0.004 |
Neg logit contributions | |
pper | -0.005 |
ilies | -0.005 |
per | -0.005 |
ectar | -0.005 |
opard | -0.005 |
Token | fast |
Token ID | 3049 |
Feature activation | -3.18e+00 |
Pos logit contributions | |
irens | +0.027 |
Paula | +0.025 |
Paul | +0.024 |
iren | +0.023 |
flashing | +0.022 |
Neg logit contributions | |
per | -0.047 |
pper | -0.046 |
ppy | -0.046 |
ilies | -0.046 |
opsy | -0.046 |
Token | asleep |
Token ID | 16039 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
irens | +0.190 |
Paula | +0.182 |
Paul | +0.172 |
iren | +0.166 |
spe | +0.162 |
Neg logit contributions | |
per | -0.272 |
ppy | -0.263 |
ilies | -0.258 |
pper | -0.256 |
opsy | -0.256 |
Token | and |
Token ID | 290 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
per | +0.072 |
ppy | +0.070 |
pper | +0.070 |
ilies | +0.069 |
opsy | +0.069 |
Neg logit contributions | |
irens | -0.122 |
Paula | -0.119 |
Paul | -0.114 |
iren | -0.112 |
spe | -0.110 |
Token | saw |
Token ID | 2497 |
Feature activation | +8.42e-01 |
Pos logit contributions | |
John | +0.011 |
Paula | +0.011 |
Paul | +0.011 |
Sarah | +0.011 |
James | +0.010 |
Neg logit contributions | |
pper | -0.020 |
ilies | -0.020 |
ectar | -0.019 |
per | -0.019 |
ppy | -0.019 |
Token | a |
Token ID | 257 |
Feature activation | +9.14e-01 |
Pos logit contributions | |
per | +0.051 |
pper | +0.050 |
ppy | +0.050 |
opsy | +0.048 |
ilies | +0.048 |
Neg logit contributions | |
irens | -0.070 |
Paula | -0.068 |
Paul | -0.067 |
John | -0.065 |
flashing | -0.064 |
Token | brown |
Token ID | 7586 |
Feature activation | -3.28e-01 |
Pos logit contributions | |
per | +0.066 |
ppy | +0.065 |
pper | +0.064 |
ilies | +0.064 |
ily | +0.060 |
Neg logit contributions | |
irens | -0.063 |
Paula | -0.063 |
Paul | -0.060 |
flashing | -0.058 |
iren | -0.058 |
Token | honey |
Token ID | 12498 |
Feature activation | +5.89e+00 |
Pos logit contributions | |
Smith | +0.022 |
Sarah | +0.022 |
irens | +0.021 |
Paula | +0.021 |
flashing | +0.021 |
Neg logit contributions | |
ilies | -0.022 |
ppy | -0.021 |
per | -0.021 |
pper | -0.021 |
ends | -0.021 |
Token | comb |
Token ID | 24011 |
Feature activation | +1.15e+01 |
Pos logit contributions | |
ilies | +0.325 |
ppy | +0.314 |
per | +0.288 |
careful | +0.288 |
ars | +0.288 |
Neg logit contributions | |
flashing | -0.437 |
irens | -0.422 |
Smith | -0.416 |
Paula | -0.405 |
bul | -0.403 |
Token | . |
Token ID | 13 |
Feature activation | -2.69e-01 |
Pos logit contributions | |
ppy | +0.816 |
ilies | +0.786 |
per | +0.765 |
opsy | +0.765 |
pper | +0.758 |
Neg logit contributions | |
irens | -0.780 |
flashing | -0.760 |
Paula | -0.735 |
John | -0.729 |
iren | -0.717 |
Token | He |
Token ID | 679 |
Feature activation | -2.97e-01 |
Pos logit contributions | |
irens | +0.020 |
Paula | +0.019 |
iren | +0.018 |
Paul | +0.018 |
spe | +0.017 |
Neg logit contributions | |
per | -0.019 |
opsy | -0.018 |
ppy | -0.018 |
ilies | -0.018 |
pper | -0.018 |
Token | was |
Token ID | 373 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
John | +0.013 |
Paul | +0.013 |
Paula | +0.013 |
Sarah | +0.013 |
spe | +0.013 |
Neg logit contributions | |
ilies | -0.027 |
ectar | -0.026 |
per | -0.026 |
seeker | -0.026 |
pper | -0.025 |
Token | very |
Token ID | 845 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
Paula | +0.086 |
John | +0.085 |
Paul | +0.083 |
flashing | +0.082 |
Sarah | +0.082 |
Neg logit contributions | |
ppy | -0.114 |
pper | -0.114 |
opsy | -0.108 |
ilies | -0.107 |
per | -0.107 |
Token | happy |
Token ID | 3772 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
pper | +0.019 |
ppy | +0.019 |
per | +0.018 |
ilies | +0.017 |
ive | +0.017 |
Neg logit contributions | |
John | -0.013 |
Sarah | -0.012 |
irens | -0.012 |
Paula | -0.012 |
Paul | -0.012 |
Token | It |
Token ID | 632 |
Feature activation | +2.30e+00 |
Pos logit contributions | |
irens | +0.012 |
Paula | +0.012 |
iren | +0.011 |
spe | +0.011 |
Paul | +0.011 |
Neg logit contributions | |
per | -0.012 |
ilies | -0.011 |
opsy | -0.011 |
pper | -0.011 |
ppy | -0.011 |
Token | was |
Token ID | 373 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
per | +0.195 |
ppy | +0.193 |
pper | +0.192 |
ilies | +0.192 |
opsy | +0.191 |
Neg logit contributions | |
irens | -0.131 |
Paula | -0.127 |
Paul | -0.120 |
iren | -0.114 |
John | -0.113 |
Token | full |
Token ID | 1336 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
irens | +0.121 |
Paula | +0.116 |
Paul | +0.111 |
iren | +0.110 |
spe | +0.108 |
Neg logit contributions | |
per | -0.083 |
ppy | -0.081 |
pper | -0.081 |
opsy | -0.081 |
ilies | -0.081 |
Token | of |
Token ID | 286 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
per | +0.117 |
ppy | +0.116 |
pper | +0.116 |
opsy | +0.115 |
ilies | +0.114 |
Neg logit contributions | |
irens | -0.112 |
Paula | -0.108 |
Paul | -0.104 |
iren | -0.100 |
John | -0.100 |
Token | honey |
Token ID | 12498 |
Feature activation | +5.86e+00 |
Pos logit contributions | |
irens | +0.120 |
Paula | +0.117 |
Paul | +0.113 |
John | +0.109 |
iren | +0.109 |
Neg logit contributions | |
per | -0.088 |
ppy | -0.087 |
pper | -0.087 |
opsy | -0.085 |
ilies | -0.084 |
Token | comb |
Token ID | 24011 |
Feature activation | +1.15e+01 |
Pos logit contributions | |
ilies | +0.387 |
per | +0.387 |
ppy | +0.385 |
opsy | +0.383 |
pper | +0.378 |
Neg logit contributions | |
irens | -0.428 |
Paula | -0.419 |
Paul | -0.408 |
John | -0.395 |
flashing | -0.389 |
Token | and |
Token ID | 290 |
Feature activation | -1.61e-01 |
Pos logit contributions | |
per | +0.708 |
ppy | +0.706 |
ilies | +0.696 |
pper | +0.696 |
opsy | +0.693 |
Neg logit contributions | |
irens | -0.917 |
Paula | -0.895 |
Paul | -0.866 |
John | -0.839 |
iren | -0.829 |
Token | it |
Token ID | 340 |
Feature activation | +2.99e-01 |
Pos logit contributions | |
irens | +0.013 |
Paula | +0.012 |
iren | +0.012 |
Paul | +0.011 |
spe | +0.011 |
Neg logit contributions | |
per | -0.010 |
ilies | -0.010 |
opsy | -0.010 |
ppy | -0.010 |
pper | -0.010 |
Token | smelled |
Token ID | 37609 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
per | +0.023 |
ppy | +0.023 |
pper | +0.023 |
opsy | +0.023 |
ilies | +0.023 |
Neg logit contributions | |
irens | -0.020 |
Paula | -0.019 |
Paul | -0.018 |
iren | -0.017 |
John | -0.016 |
Token | so |
Token ID | 523 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
irens | +0.156 |
Paula | +0.152 |
Paul | +0.146 |
iren | +0.141 |
John | +0.141 |
Neg logit contributions | |
per | -0.114 |
ppy | -0.113 |
pper | -0.112 |
opsy | -0.111 |
ilies | -0.111 |
Token | sweet |
Token ID | 6029 |
Feature activation | +5.33e-01 |
Pos logit contributions | |
irens | +0.081 |
Paula | +0.080 |
Paul | +0.078 |
John | +0.077 |
Sarah | +0.074 |
Neg logit contributions | |
per | -0.063 |
pper | -0.063 |
ppy | -0.062 |
ilies | -0.060 |
opsy | -0.060 |
Token | and |
Token ID | 290 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
ilies | +0.195 |
opsy | +0.192 |
per | +0.191 |
ppy | +0.188 |
ectar | +0.186 |
Neg logit contributions | |
irens | -0.258 |
Paula | -0.253 |
Paul | -0.248 |
John | -0.239 |
iren | -0.235 |
Token | he |
Token ID | 339 |
Feature activation | +7.59e-02 |
Pos logit contributions | |
irens | +0.017 |
Paula | +0.016 |
Paul | +0.016 |
John | +0.016 |
Sarah | +0.016 |
Neg logit contributions | |
per | -0.018 |
ppy | -0.017 |
opsy | -0.017 |
ilies | -0.017 |
pper | -0.017 |
Token | split |
Token ID | 6626 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
per | +0.006 |
ilies | +0.006 |
ectar | +0.006 |
opsy | +0.005 |
ppy | +0.005 |
Neg logit contributions | |
Paul | -0.004 |
Paula | -0.004 |
John | -0.004 |
spe | -0.004 |
Sarah | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | -7.81e-01 |
Pos logit contributions | |
per | +0.103 |
ppy | +0.102 |
pper | +0.102 |
opsy | +0.101 |
ilies | +0.101 |
Neg logit contributions | |
irens | -0.080 |
Paula | -0.078 |
Paul | -0.076 |
John | -0.075 |
Sarah | -0.072 |
Token | honey |
Token ID | 12498 |
Feature activation | +5.86e+00 |
Pos logit contributions | |
million | +0.052 |
chaos | +0.051 |
stairs | +0.049 |
mess | +0.047 |
flashing | +0.046 |
Neg logit contributions | |
pper | -0.055 |
per | -0.054 |
ilies | -0.052 |
ppy | -0.050 |
amb | -0.048 |
Token | comb |
Token ID | 24011 |
Feature activation | +1.15e+01 |
Pos logit contributions | |
opsy | +0.407 |
ilies | +0.390 |
ppy | +0.383 |
pper | +0.381 |
ani | +0.377 |
Neg logit contributions | |
spe | -0.349 |
Sarah | -0.346 |
irens | -0.343 |
Smith | -0.341 |
John | -0.341 |
Token | in |
Token ID | 287 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
opsy | +0.780 |
ppy | +0.773 |
pper | +0.761 |
ilies | +0.751 |
per | +0.720 |
Neg logit contributions | |
irens | -0.768 |
John | -0.756 |
spe | -0.752 |
Paula | -0.746 |
Sarah | -0.740 |
Token | two |
Token ID | 734 |
Feature activation | +7.36e-01 |
Pos logit contributions | |
per | +0.176 |
ppy | +0.171 |
pper | +0.171 |
opsy | +0.168 |
ilies | +0.167 |
Neg logit contributions | |
irens | -0.226 |
Paula | -0.217 |
Paul | -0.205 |
iren | -0.204 |
spe | -0.199 |
Token | with |
Token ID | 351 |
Feature activation | -3.17e-01 |
Pos logit contributions | |
pper | +0.052 |
opsy | +0.051 |
ppy | +0.050 |
per | +0.050 |
ilies | +0.049 |
Neg logit contributions | |
irens | -0.051 |
Paula | -0.049 |
Paul | -0.047 |
John | -0.046 |
flashing | -0.046 |
Token | his |
Token ID | 465 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
irens | +0.025 |
Paula | +0.024 |
Paul | +0.024 |
John | +0.023 |
flashing | +0.023 |
Neg logit contributions | |
per | -0.020 |
ppy | -0.020 |
pper | -0.020 |
opsy | -0.019 |
ilies | -0.019 |
Token | tiny |
Token ID | 7009 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
John | +0.098 |
Sarah | +0.096 |
Paula | +0.095 |
Paul | +0.093 |
million | +0.092 |
Neg logit contributions | |
per | -0.127 |
pper | -0.120 |
ilies | -0.118 |
ppy | -0.117 |
ily | -0.110 |
Token | one |
Token ID | 530 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
per | +0.381 |
pper | +0.372 |
opsy | +0.369 |
ilies | +0.366 |
ppy | +0.364 |
Neg logit contributions | |
irens | -0.456 |
Paula | -0.436 |
Paul | -0.427 |
Sarah | -0.426 |
John | -0.426 |
Token | half |
Token ID | 2063 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
per | +0.122 |
ppy | +0.117 |
ilies | +0.116 |
pper | +0.113 |
opsy | +0.110 |
Neg logit contributions | |
John | -0.104 |
Sarah | -0.102 |
Paula | -0.100 |
irens | -0.100 |
Paul | -0.099 |
Token | of |
Token ID | 286 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
John | +0.109 |
spe | +0.101 |
Sarah | +0.101 |
Paul | +0.100 |
irens | +0.099 |
Neg logit contributions | |
opsy | -0.168 |
ppy | -0.163 |
per | -0.163 |
seeker | -0.161 |
ilies | -0.159 |
Token | the |
Token ID | 262 |
Feature activation | -8.05e-01 |
Pos logit contributions | |
John | +0.075 |
Paula | +0.074 |
irens | +0.073 |
Sarah | +0.072 |
Paul | +0.071 |
Neg logit contributions | |
per | -0.133 |
ppy | -0.132 |
pper | -0.128 |
ilies | -0.126 |
ider | -0.126 |
Token | honey |
Token ID | 12498 |
Feature activation | +5.85e+00 |
Pos logit contributions | |
million | +0.052 |
chaos | +0.052 |
Sarah | +0.049 |
stairs | +0.049 |
Paula | +0.048 |
Neg logit contributions | |
per | -0.057 |
pper | -0.054 |
ilies | -0.051 |
ppy | -0.050 |
izards | -0.047 |
Token | comb |
Token ID | 24011 |
Feature activation | +1.15e+01 |
Pos logit contributions | |
opsy | +0.394 |
ilies | +0.392 |
ppy | +0.392 |
per | +0.388 |
pper | +0.375 |
Neg logit contributions | |
John | -0.383 |
Paula | -0.374 |
irens | -0.372 |
Paul | -0.372 |
Sarah | -0.372 |
Token | and |
Token ID | 290 |
Feature activation | -2.54e-01 |
Pos logit contributions | |
ppy | +0.773 |
opsy | +0.758 |
ilies | +0.752 |
per | +0.749 |
pper | +0.745 |
Neg logit contributions | |
irens | -0.784 |
John | -0.775 |
Paula | -0.768 |
Paul | -0.759 |
spe | -0.750 |
Token | kept |
Token ID | 4030 |
Feature activation | +3.50e+00 |
Pos logit contributions | |
Paula | +0.018 |
irens | +0.017 |
John | +0.017 |
Paul | +0.017 |
Sarah | +0.017 |
Neg logit contributions | |
per | -0.018 |
ppy | -0.018 |
opsy | -0.017 |
ilies | -0.017 |
pper | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | -8.12e-01 |
Pos logit contributions | |
per | +0.248 |
ppy | +0.245 |
pper | +0.243 |
ilies | +0.242 |
opsy | +0.242 |
Neg logit contributions | |
irens | -0.252 |
Paula | -0.244 |
Paul | -0.234 |
iren | -0.225 |
John | -0.223 |
Token | other |
Token ID | 584 |
Feature activation | +3.66e+00 |
Pos logit contributions | |
chaos | +0.051 |
million | +0.051 |
stairs | +0.048 |
mess | +0.047 |
ticking | +0.047 |
Neg logit contributions | |
per | -0.060 |
pper | -0.057 |
ppy | -0.056 |
ilies | -0.055 |
izards | -0.052 |
Token | one |
Token ID | 530 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
per | +0.218 |
pper | +0.215 |
ute | +0.203 |
ppy | +0.202 |
ilies | +0.195 |
Neg logit contributions | |
million | -0.259 |
irens | -0.254 |
Paula | -0.253 |
chaos | -0.252 |
Sarah | -0.251 |
Token | he |
Token ID | 339 |
Feature activation | +3.07e-02 |
Pos logit contributions | |
per | +0.021 |
opsy | +0.021 |
ppy | +0.021 |
ilies | +0.021 |
pper | +0.021 |
Neg logit contributions | |
irens | -0.022 |
Paula | -0.021 |
iren | -0.019 |
Paul | -0.019 |
spe | -0.019 |
Token | saw |
Token ID | 2497 |
Feature activation | +7.71e-01 |
Pos logit contributions | |
ilies | +0.002 |
per | +0.002 |
ectar | +0.002 |
ppy | +0.002 |
pper | +0.002 |
Neg logit contributions | |
irens | -0.002 |
Paula | -0.002 |
Paul | -0.002 |
John | -0.002 |
Sarah | -0.002 |
Token | that |
Token ID | 326 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
per | +0.047 |
ilies | +0.046 |
ppy | +0.046 |
opsy | +0.045 |
pper | +0.045 |
Neg logit contributions | |
irens | -0.064 |
Paula | -0.062 |
Paul | -0.059 |
iren | -0.058 |
spe | -0.057 |
Token | the |
Token ID | 262 |
Feature activation | -8.44e-01 |
Pos logit contributions | |
irens | +0.084 |
Paula | +0.083 |
Paul | +0.079 |
John | +0.077 |
Sarah | +0.074 |
Neg logit contributions | |
per | -0.116 |
ppy | -0.115 |
pper | -0.115 |
ilies | -0.113 |
opsy | -0.113 |
Token | honey |
Token ID | 12498 |
Feature activation | +5.80e+00 |
Pos logit contributions | |
chaos | +0.057 |
million | +0.054 |
stairs | +0.053 |
Paula | +0.051 |
irens | +0.050 |
Neg logit contributions | |
per | -0.059 |
pper | -0.059 |
ppy | -0.056 |
ilies | -0.054 |
amb | -0.052 |
Token | comb |
Token ID | 24011 |
Feature activation | +1.14e+01 |
Pos logit contributions | |
per | +0.517 |
ppy | +0.512 |
pper | +0.506 |
ilies | +0.505 |
opsy | +0.503 |
Neg logit contributions | |
irens | -0.298 |
Paula | -0.288 |
Paul | -0.272 |
John | -0.261 |
iren | -0.258 |
Token | was |
Token ID | 373 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
per | +0.968 |
ppy | +0.961 |
pper | +0.952 |
ilies | +0.948 |
opsy | +0.948 |
Neg logit contributions | |
irens | -0.650 |
Paula | -0.624 |
Paul | -0.592 |
iren | -0.567 |
John | -0.562 |
Token | filled |
Token ID | 5901 |
Feature activation | +4.67e+00 |
Pos logit contributions | |
irens | +0.113 |
Paula | +0.112 |
John | +0.108 |
Paul | +0.108 |
flashing | +0.108 |
Neg logit contributions | |
ppy | -0.101 |
per | -0.099 |
pper | -0.098 |
opsy | -0.097 |
ilies | -0.097 |
Token | with |
Token ID | 351 |
Feature activation | -3.87e-01 |
Pos logit contributions | |
per | +0.262 |
ppy | +0.257 |
pper | +0.255 |
opsy | +0.253 |
ilies | +0.253 |
Neg logit contributions | |
irens | -0.406 |
Paula | -0.395 |
Paul | -0.380 |
iren | -0.370 |
John | -0.363 |
Token | brown |
Token ID | 7586 |
Feature activation | -4.04e-01 |
Pos logit contributions | |
irens | +0.031 |
Paula | +0.030 |
Paul | +0.029 |
John | +0.028 |
flashing | +0.028 |
Neg logit contributions | |
per | -0.024 |
pper | -0.024 |
ppy | -0.024 |
opsy | -0.023 |
ilies | -0.023 |
Token | honey |
Token ID | 12498 |
Feature activation | +5.79e+00 |
Pos logit contributions | |
irens | +0.034 |
Paula | +0.032 |
John | +0.031 |
Paul | +0.031 |
flashing | +0.031 |
Neg logit contributions | |
pper | -0.023 |
per | -0.023 |
opsy | -0.022 |
ppy | -0.022 |
ilies | -0.022 |
Token | they |
Token ID | 484 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
per | +0.024 |
ilies | +0.024 |
opsy | +0.024 |
pper | +0.024 |
ppy | +0.024 |
Neg logit contributions | |
irens | -0.026 |
Paula | -0.024 |
iren | -0.024 |
Paul | -0.022 |
spe | -0.022 |
Token | found |
Token ID | 1043 |
Feature activation | +3.04e+00 |
Pos logit contributions | |
Paul | +0.130 |
Paula | +0.129 |
John | +0.129 |
Sarah | +0.127 |
irens | +0.115 |
Neg logit contributions | |
ectar | -0.176 |
ider | -0.175 |
per | -0.167 |
ilies | -0.167 |
opsy | -0.165 |
Token | lots |
Token ID | 6041 |
Feature activation | -9.28e-01 |
Pos logit contributions | |
per | +0.194 |
ppy | +0.192 |
pper | +0.188 |
opsy | +0.182 |
ilies | +0.177 |
Neg logit contributions | |
irens | -0.234 |
Paula | -0.232 |
Paul | -0.230 |
John | -0.229 |
Sarah | -0.221 |
Token | of |
Token ID | 286 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
irens | +0.062 |
Paula | +0.059 |
Paul | +0.056 |
iren | +0.055 |
John | +0.054 |
Neg logit contributions | |
per | -0.071 |
ppy | -0.070 |
pper | -0.070 |
opsy | -0.069 |
ilies | -0.068 |
Token | honey |
Token ID | 12498 |
Feature activation | +5.86e+00 |
Pos logit contributions | |
irens | +0.118 |
Paula | +0.115 |
Paul | +0.110 |
flashing | +0.110 |
Sarah | +0.109 |
Neg logit contributions | |
pper | -0.100 |
per | -0.100 |
ppy | -0.095 |
opsy | -0.090 |
comb | -0.089 |
Token | comb |
Token ID | 24011 |
Feature activation | +1.14e+01 |
Pos logit contributions | |
opsy | +0.371 |
ppy | +0.362 |
per | +0.354 |
thirst | +0.353 |
ilies | +0.347 |
Neg logit contributions | |
irens | -0.412 |
Paula | -0.399 |
Sarah | -0.386 |
John | -0.381 |
flashing | -0.381 |
Token | and |
Token ID | 290 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
ppy | +0.698 |
per | +0.685 |
opsy | +0.683 |
pper | +0.677 |
ilies | +0.665 |
Neg logit contributions | |
irens | -0.889 |
Paula | -0.870 |
Paul | -0.842 |
John | -0.839 |
Sarah | -0.824 |
Token | honey |
Token ID | 12498 |
Feature activation | +5.85e+00 |
Pos logit contributions | |
irens | +0.015 |
Paula | +0.015 |
Paul | +0.015 |
John | +0.014 |
Sarah | +0.014 |
Neg logit contributions | |
per | -0.015 |
ppy | -0.014 |
pper | -0.014 |
opsy | -0.014 |
ilies | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
opsy | +0.377 |
per | +0.373 |
ppy | +0.366 |
pper | +0.363 |
ilies | +0.360 |
Neg logit contributions | |
Paula | -0.399 |
irens | -0.395 |
Sarah | -0.381 |
John | -0.380 |
Paul | -0.380 |
Token | They |
Token ID | 1119 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
irens | +0.019 |
Paula | +0.019 |
iren | +0.018 |
spe | +0.017 |
flashing | +0.017 |
Neg logit contributions | |
ilies | -0.016 |
per | -0.016 |
opsy | -0.015 |
ily | -0.015 |
ppy | -0.015 |
Token | shared |
Token ID | 4888 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
John | +0.069 |
Paula | +0.067 |
Sarah | +0.066 |
Paul | +0.066 |
James | +0.063 |
Neg logit contributions | |
ider | -0.109 |
per | -0.108 |
ectar | -0.106 |
ute | -0.105 |
ilies | -0.102 |
Token | and |
Token ID | 290 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
irens | +0.131 |
Paula | +0.125 |
Paul | +0.120 |
iren | +0.119 |
John | +0.115 |
Neg logit contributions | |
per | -0.085 |
ppy | -0.082 |
pper | -0.081 |
ilies | -0.080 |
comb | -0.077 |
Token | twist |
Token ID | 14528 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
irens | +0.011 |
Paula | +0.010 |
iren | +0.010 |
Paul | +0.010 |
spe | +0.009 |
Neg logit contributions | |
per | -0.007 |
ilies | -0.007 |
ppy | -0.007 |
pper | -0.007 |
comb | -0.007 |
Token | it |
Token ID | 340 |
Feature activation | +3.74e-01 |
Pos logit contributions | |
per | +0.074 |
ppy | +0.072 |
ilies | +0.071 |
pper | +0.069 |
opsy | +0.068 |
Neg logit contributions | |
irens | -0.097 |
Paula | -0.090 |
iren | -0.088 |
Paul | -0.087 |
spe | -0.084 |
Token | until |
Token ID | 1566 |
Feature activation | +4.86e+00 |
Pos logit contributions | |
per | +0.021 |
ppy | +0.021 |
pper | +0.021 |
opsy | +0.021 |
ilies | +0.021 |
Neg logit contributions | |
irens | -0.032 |
Paula | -0.031 |
Paul | -0.030 |
iren | -0.029 |
John | -0.028 |
Token | the |
Token ID | 262 |
Feature activation | -7.12e-01 |
Pos logit contributions | |
per | +0.287 |
ilies | +0.284 |
opsy | +0.279 |
ppy | +0.277 |
pper | +0.272 |
Neg logit contributions | |
irens | -0.405 |
Paula | -0.380 |
iren | -0.374 |
Paul | -0.365 |
spe | -0.357 |
Token | pad |
Token ID | 14841 |
Feature activation | +1.13e+01 |
Pos logit contributions | |
irens | +0.053 |
Paula | +0.051 |
Paul | +0.049 |
iren | +0.047 |
John | +0.047 |
Neg logit contributions | |
per | -0.049 |
ppy | -0.048 |
pper | -0.048 |
opsy | -0.048 |
ilies | -0.048 |
Token | lock |
Token ID | 5354 |
Feature activation | +3.29e+00 |
Pos logit contributions | |
opsy | +0.777 |
ppy | +0.774 |
per | +0.768 |
ilies | +0.762 |
pper | +0.759 |
Neg logit contributions | |
irens | -0.823 |
Paula | -0.799 |
Paul | -0.774 |
iren | -0.746 |
spe | -0.743 |
Token | clicked |
Token ID | 28384 |
Feature activation | -2.80e+00 |
Pos logit contributions | |
per | +0.234 |
ppy | +0.230 |
pper | +0.229 |
ilies | +0.228 |
opsy | +0.225 |
Neg logit contributions | |
irens | -0.235 |
Paula | -0.227 |
Paul | -0.216 |
iren | -0.210 |
John | -0.205 |
Token | open |
Token ID | 1280 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
irens | +0.249 |
Paula | +0.237 |
iren | +0.229 |
Paul | +0.227 |
spe | +0.220 |
Neg logit contributions | |
per | -0.156 |
ppy | -0.149 |
pper | -0.145 |
ilies | -0.144 |
opsy | -0.141 |
Token | . |
Token ID | 13 |
Feature activation | -9.17e-02 |
Pos logit contributions | |
irens | +0.096 |
Paula | +0.090 |
iren | +0.089 |
Paul | +0.086 |
spe | +0.083 |
Neg logit contributions | |
per | -0.051 |
ilies | -0.046 |
ppy | -0.046 |
comb | -0.045 |
ectar | -0.044 |
Token |
|
Token ID | 198 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
irens | +0.007 |
Paula | +0.006 |
iren | +0.006 |
Paul | +0.006 |
spe | +0.006 |
Neg logit contributions | |
per | -0.006 |
ilies | -0.006 |
opsy | -0.006 |
ppy | -0.006 |
ectar | -0.006 |
Token | hopped |
Token ID | 45230 |
Feature activation | +3.07e+00 |
Pos logit contributions | |
irens | +0.016 |
Paula | +0.015 |
Paul | +0.015 |
John | +0.014 |
spe | +0.014 |
Neg logit contributions | |
per | -0.018 |
pper | -0.018 |
ppy | -0.018 |
ilies | -0.018 |
opsy | -0.018 |
Token | from |
Token ID | 422 |
Feature activation | +2.71e-02 |
Pos logit contributions | |
pper | +0.259 |
opsy | +0.258 |
ppy | +0.257 |
per | +0.243 |
comb | +0.242 |
Neg logit contributions | |
irens | -0.175 |
Paula | -0.172 |
Paul | -0.172 |
John | -0.167 |
Sarah | -0.165 |
Token | one |
Token ID | 530 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
ppy | +0.002 |
pper | +0.002 |
per | +0.002 |
ilies | +0.002 |
opsy | +0.002 |
Neg logit contributions | |
irens | -0.002 |
Paula | -0.002 |
Paul | -0.002 |
John | -0.002 |
Sarah | -0.002 |
Token | l |
Token ID | 300 |
Feature activation | +4.25e+00 |
Pos logit contributions | |
ppy | +0.112 |
pper | +0.111 |
per | +0.110 |
opsy | +0.108 |
ilies | +0.108 |
Neg logit contributions | |
irens | -0.122 |
Paula | -0.117 |
Paul | -0.113 |
iren | -0.111 |
John | -0.110 |
Token | ily |
Token ID | 813 |
Feature activation | +7.44e+00 |
Pos logit contributions | |
wait | +0.188 |
opsy | +0.178 |
pper | +0.169 |
prett | +0.163 |
remember | +0.155 |
Neg logit contributions | |
irens | -0.391 |
iren | -0.383 |
seat | -0.326 |
Sarah | -0.324 |
doorway | -0.324 |
Token | pad |
Token ID | 14841 |
Feature activation | +1.13e+01 |
Pos logit contributions | |
ilies | +0.328 |
opsy | +0.325 |
ectar | +0.316 |
per | +0.304 |
pper | +0.301 |
Neg logit contributions | |
Paula | -0.703 |
Paul | -0.694 |
irens | -0.685 |
John | -0.676 |
Sarah | -0.668 |
Token | to |
Token ID | 284 |
Feature activation | -6.25e-01 |
Pos logit contributions | |
opsy | +0.843 |
ilies | +0.819 |
ppy | +0.810 |
thirst | +0.772 |
per | +0.766 |
Neg logit contributions | |
irens | -0.718 |
Paul | -0.710 |
Paula | -0.695 |
John | -0.692 |
spe | -0.674 |
Token | the |
Token ID | 262 |
Feature activation | -8.68e-01 |
Pos logit contributions | |
Paul | +0.042 |
Paula | +0.042 |
John | +0.042 |
Sarah | +0.042 |
irens | +0.040 |
Neg logit contributions | |
ppy | -0.046 |
pper | -0.044 |
opsy | -0.044 |
ectar | -0.043 |
ilies | -0.043 |
Token | other |
Token ID | 584 |
Feature activation | +3.67e+00 |
Pos logit contributions | |
chaos | +0.058 |
stairs | +0.054 |
flashing | +0.054 |
irens | +0.053 |
Smith | +0.053 |
Neg logit contributions | |
ppy | -0.066 |
pper | -0.059 |
oll | -0.058 |
ectar | -0.058 |
int | -0.058 |
Token | . |
Token ID | 13 |
Feature activation | -6.34e-02 |
Pos logit contributions | |
ppy | +0.268 |
pper | +0.267 |
opsy | +0.256 |
ilies | +0.250 |
ectar | +0.246 |
Neg logit contributions | |
John | -0.229 |
Paul | -0.225 |
Sarah | -0.223 |
irens | -0.221 |
Paula | -0.218 |
Token | It |
Token ID | 632 |
Feature activation | +2.28e+00 |
Pos logit contributions | |
irens | +0.003 |
Paul | +0.003 |
Paula | +0.003 |
John | +0.003 |
Sarah | +0.003 |
Neg logit contributions | |
pper | -0.006 |
ppy | -0.006 |
per | -0.005 |
opsy | -0.005 |
ilies | -0.005 |
Token | him |
Token ID | 683 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
per | +0.009 |
ppy | +0.009 |
ilies | +0.009 |
pper | +0.009 |
opsy | +0.009 |
Neg logit contributions | |
irens | -0.017 |
Paula | -0.017 |
Paul | -0.016 |
iren | -0.016 |
spe | -0.015 |
Token | on |
Token ID | 319 |
Feature activation | -3.72e-01 |
Pos logit contributions | |
opsy | +0.009 |
pper | +0.009 |
per | +0.009 |
ilies | +0.009 |
ppy | +0.009 |
Neg logit contributions | |
irens | -0.012 |
Paula | -0.012 |
Paul | -0.012 |
John | -0.011 |
spe | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | -6.48e-01 |
Pos logit contributions | |
irens | +0.032 |
Paula | +0.031 |
iren | +0.030 |
spe | +0.029 |
Paul | +0.028 |
Neg logit contributions | |
per | -0.021 |
opsy | -0.020 |
ilies | -0.020 |
ppy | -0.019 |
ectar | -0.019 |
Token | l |
Token ID | 300 |
Feature activation | +4.31e+00 |
Pos logit contributions | |
irens | +0.042 |
stairs | +0.041 |
flashing | +0.040 |
chaos | +0.039 |
Paul | +0.039 |
Neg logit contributions | |
ppy | -0.050 |
ilies | -0.046 |
ectar | -0.046 |
pper | -0.046 |
per | -0.045 |
Token | ily |
Token ID | 813 |
Feature activation | +7.47e+00 |
Pos logit contributions | |
wait | +0.180 |
prett | +0.176 |
okay | +0.171 |
ppy | +0.169 |
remember | +0.165 |
Neg logit contributions | |
iren | -0.384 |
irens | -0.380 |
chairs | -0.324 |
igsaw | -0.318 |
doorway | -0.314 |
Token | pad |
Token ID | 14841 |
Feature activation | +1.13e+01 |
Pos logit contributions | |
opsy | +0.429 |
per | +0.425 |
ilies | +0.421 |
pper | +0.410 |
ectar | +0.404 |
Neg logit contributions | |
Paula | -0.607 |
irens | -0.596 |
Paul | -0.593 |
John | -0.575 |
Sarah | -0.568 |
Token | . |
Token ID | 13 |
Feature activation | -7.01e-02 |
Pos logit contributions | |
opsy | +0.792 |
ilies | +0.763 |
ppy | +0.757 |
per | +0.746 |
pper | +0.728 |
Neg logit contributions | |
irens | -0.770 |
Paul | -0.741 |
Paula | -0.738 |
spe | -0.727 |
John | -0.721 |
Token | They |
Token ID | 1119 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
irens | +0.004 |
Paula | +0.004 |
Paul | +0.004 |
John | +0.004 |
Sarah | +0.004 |
Neg logit contributions | |
pper | -0.006 |
ppy | -0.006 |
per | -0.006 |
opsy | -0.006 |
ilies | -0.005 |
Token | all |
Token ID | 477 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
Paula | +0.088 |
irens | +0.087 |
Paul | +0.085 |
John | +0.082 |
Sarah | +0.080 |
Neg logit contributions | |
per | -0.079 |
pper | -0.076 |
opsy | -0.076 |
ppy | -0.076 |
ilies | -0.076 |
Token | rested |
Token ID | 29263 |
Feature activation | +5.58e+00 |
Pos logit contributions | |
per | +0.119 |
ppy | +0.119 |
ilies | +0.117 |
pper | +0.117 |
opsy | +0.116 |
Neg logit contributions | |
irens | -0.133 |
Paula | -0.128 |
Paul | -0.122 |
iren | -0.120 |
John | -0.115 |
Token | together |
Token ID | 1978 |
Feature activation | -4.74e-01 |
Pos logit contributions | |
ilies | +0.344 |
opsy | +0.342 |
ppy | +0.337 |
pper | +0.336 |
per | +0.330 |
Neg logit contributions | |
irens | -0.432 |
Paul | -0.427 |
John | -0.425 |
Paula | -0.423 |
Sarah | -0.417 |
Token | back |
Token ID | 736 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
opsy | +0.281 |
pper | +0.273 |
ilies | +0.270 |
ppy | +0.265 |
ectar | +0.263 |
Neg logit contributions | |
irens | -0.163 |
Sarah | -0.161 |
John | -0.160 |
Paul | -0.159 |
iren | -0.158 |
Token | onto |
Token ID | 4291 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
per | +0.089 |
ppy | +0.088 |
opsy | +0.087 |
pper | +0.087 |
ilies | +0.086 |
Neg logit contributions | |
irens | -0.097 |
Paula | -0.094 |
Paul | -0.089 |
iren | -0.087 |
spe | -0.086 |
Token | his |
Token ID | 465 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
irens | +0.085 |
Paula | +0.082 |
Paul | +0.078 |
iren | +0.075 |
John | +0.074 |
Neg logit contributions | |
per | -0.107 |
ppy | -0.107 |
pper | -0.105 |
ilies | -0.105 |
opsy | -0.105 |
Token | l |
Token ID | 300 |
Feature activation | +4.36e+00 |
Pos logit contributions | |
flashing | +0.097 |
irens | +0.096 |
stairs | +0.093 |
iren | +0.093 |
Paul | +0.092 |
Neg logit contributions | |
ilies | -0.109 |
ppy | -0.100 |
ectar | -0.099 |
izards | -0.097 |
ily | -0.095 |
Token | ily |
Token ID | 813 |
Feature activation | +7.45e+00 |
Pos logit contributions | |
wait | +0.177 |
prett | +0.162 |
opsy | +0.155 |
ppy | +0.154 |
ender | +0.152 |
Neg logit contributions | |
iren | -0.405 |
irens | -0.392 |
doorway | -0.337 |
seat | -0.325 |
ons | -0.321 |
Token | pad |
Token ID | 14841 |
Feature activation | +1.12e+01 |
Pos logit contributions | |
ilies | +0.496 |
opsy | +0.462 |
ectar | +0.461 |
ily | +0.450 |
per | +0.440 |
Neg logit contributions | |
Paula | -0.540 |
Paul | -0.528 |
John | -0.520 |
Sarah | -0.519 |
James | -0.504 |
Token | and |
Token ID | 290 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
ilies | +0.871 |
opsy | +0.850 |
ppy | +0.809 |
per | +0.786 |
izards | +0.783 |
Neg logit contributions | |
irens | -0.656 |
Paul | -0.631 |
Paula | -0.625 |
John | -0.624 |
iren | -0.622 |
Token | stood |
Token ID | 6204 |
Feature activation | +2.73e+00 |
Pos logit contributions | |
Paula | +0.007 |
Paul | +0.007 |
irens | +0.007 |
John | +0.007 |
Sarah | +0.007 |
Neg logit contributions | |
ilies | -0.008 |
per | -0.008 |
ppy | -0.008 |
ectar | -0.008 |
pper | -0.008 |
Token | there |
Token ID | 612 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
ppy | +0.238 |
ilies | +0.236 |
pper | +0.226 |
izards | +0.215 |
opard | +0.214 |
Neg logit contributions | |
irens | -0.151 |
Paula | -0.150 |
Paul | -0.148 |
John | -0.148 |
Sarah | -0.147 |
Token | , |
Token ID | 11 |
Feature activation | +4.13e-01 |
Pos logit contributions | |
Paula | +0.100 |
Paul | +0.100 |
John | +0.099 |
irens | +0.097 |
Sarah | +0.095 |
Neg logit contributions | |
ilies | -0.074 |
opsy | -0.074 |
ppy | -0.074 |
pper | -0.073 |
per | -0.071 |
Token | feeling |
Token ID | 4203 |
Feature activation | +6.21e-01 |
Pos logit contributions | |
ppy | +0.029 |
ilies | +0.029 |
per | +0.029 |
pper | +0.029 |
opsy | +0.029 |
Neg logit contributions | |
irens | -0.029 |
Paula | -0.028 |
Paul | -0.027 |
John | -0.026 |
iren | -0.026 |
Token | He |
Token ID | 679 |
Feature activation | -1.94e-01 |
Pos logit contributions | |
irens | +0.005 |
Paula | +0.005 |
Paul | +0.005 |
John | +0.005 |
Sarah | +0.005 |
Neg logit contributions | |
per | -0.007 |
pper | -0.007 |
ppy | -0.007 |
ilies | -0.007 |
opsy | -0.007 |
Token | found |
Token ID | 1043 |
Feature activation | +3.12e+00 |
Pos logit contributions | |
Paul | +0.012 |
Paula | +0.012 |
irens | +0.012 |
John | +0.012 |
spe | +0.012 |
Neg logit contributions | |
ectar | -0.014 |
ilies | -0.014 |
per | -0.014 |
opsy | -0.014 |
pper | -0.014 |
Token | a |
Token ID | 257 |
Feature activation | +9.38e-01 |
Pos logit contributions | |
per | +0.180 |
ppy | +0.178 |
pper | +0.176 |
opsy | +0.174 |
ilies | +0.173 |
Neg logit contributions | |
irens | -0.266 |
Paula | -0.257 |
Paul | -0.252 |
John | -0.244 |
iren | -0.242 |
Token | l |
Token ID | 300 |
Feature activation | +4.27e+00 |
Pos logit contributions | |
per | +0.060 |
ppy | +0.059 |
pper | +0.059 |
ilies | +0.059 |
opsy | +0.058 |
Neg logit contributions | |
irens | -0.073 |
Paula | -0.071 |
Paul | -0.069 |
iren | -0.067 |
John | -0.066 |
Token | ily |
Token ID | 813 |
Feature activation | +7.38e+00 |
Pos logit contributions | |
wait | +0.189 |
opsy | +0.170 |
prett | +0.159 |
ourselves | +0.158 |
decide | +0.158 |
Neg logit contributions | |
iren | -0.337 |
irens | -0.321 |
doorway | -0.278 |
smart | -0.277 |
run | -0.265 |
Token | pad |
Token ID | 14841 |
Feature activation | +1.12e+01 |
Pos logit contributions | |
ilies | +0.390 |
ectar | +0.380 |
opsy | +0.379 |
per | +0.376 |
ppy | +0.367 |
Neg logit contributions | |
Paula | -0.617 |
irens | -0.606 |
Paul | -0.605 |
John | -0.595 |
Sarah | -0.588 |
Token | in |
Token ID | 287 |
Feature activation | +2.77e+00 |
Pos logit contributions | |
opsy | +0.795 |
ppy | +0.771 |
ilies | +0.766 |
per | +0.751 |
ectar | +0.725 |
Neg logit contributions | |
irens | -0.778 |
Paul | -0.753 |
Paula | -0.744 |
John | -0.741 |
spe | -0.727 |
Token | the |
Token ID | 262 |
Feature activation | -8.61e-01 |
Pos logit contributions | |
per | +0.145 |
worm | +0.144 |
like | +0.140 |
pper | +0.134 |
opsy | +0.134 |
Neg logit contributions | |
irens | -0.254 |
iren | -0.236 |
Paula | -0.234 |
spe | -0.231 |
telev | -0.220 |
Token | pond |
Token ID | 16723 |
Feature activation | +6.98e+00 |
Pos logit contributions | |
irens | +0.058 |
iren | +0.053 |
chaos | +0.053 |
Paula | +0.053 |
Paul | +0.052 |
Neg logit contributions | |
ppy | -0.064 |
ilies | -0.063 |
pper | -0.063 |
ectar | -0.063 |
per | -0.060 |
Token | and |
Token ID | 290 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
ppy | +0.447 |
opsy | +0.445 |
pper | +0.440 |
ilies | +0.434 |
per | +0.431 |
Neg logit contributions | |
irens | -0.534 |
Paula | -0.511 |
Paul | -0.502 |
John | -0.498 |
spe | -0.490 |
Token | rested |
Token ID | 29263 |
Feature activation | +5.50e+00 |
Pos logit contributions | |
irens | +0.009 |
Paula | +0.009 |
Paul | +0.009 |
John | +0.008 |
Sarah | +0.008 |
Neg logit contributions | |
per | -0.009 |
ppy | -0.009 |
pper | -0.009 |
opsy | -0.009 |
ilies | -0.009 |
Token | paints |
Token ID | 32281 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
irens | +0.111 |
Paula | +0.107 |
Paul | +0.102 |
iren | +0.100 |
spe | +0.099 |
Neg logit contributions | |
per | -0.097 |
ilies | -0.097 |
ppy | -0.096 |
opsy | -0.094 |
pper | -0.094 |
Token | , |
Token ID | 11 |
Feature activation | +3.11e-01 |
Pos logit contributions | |
irens | +0.109 |
Paula | +0.106 |
Paul | +0.102 |
iren | +0.100 |
John | +0.098 |
Neg logit contributions | |
per | -0.071 |
ppy | -0.070 |
opsy | -0.070 |
pper | -0.070 |
ilies | -0.069 |
Token | and |
Token ID | 290 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
per | +0.018 |
ppy | +0.017 |
pper | +0.017 |
ilies | +0.017 |
opsy | +0.017 |
Neg logit contributions | |
irens | -0.027 |
Paula | -0.026 |
Paul | -0.025 |
iren | -0.024 |
John | -0.024 |
Token | a |
Token ID | 257 |
Feature activation | +9.76e-01 |
Pos logit contributions | |
irens | +0.015 |
Paula | +0.014 |
iren | +0.014 |
Paul | +0.014 |
spe | +0.013 |
Neg logit contributions | |
per | -0.010 |
ilies | -0.010 |
ppy | -0.010 |
pper | -0.010 |
opsy | -0.010 |
Token | huge |
Token ID | 3236 |
Feature activation | -2.76e+00 |
Pos logit contributions | |
per | +0.071 |
ppy | +0.070 |
pper | +0.069 |
opsy | +0.069 |
ilies | +0.069 |
Neg logit contributions | |
irens | -0.068 |
Paula | -0.066 |
Paul | -0.063 |
iren | -0.061 |
John | -0.059 |
Token | pad |
Token ID | 14841 |
Feature activation | +1.12e+01 |
Pos logit contributions | |
irens | +0.208 |
Paula | +0.199 |
Paul | +0.190 |
iren | +0.186 |
spe | +0.182 |
Neg logit contributions | |
per | -0.186 |
ppy | -0.183 |
pper | -0.181 |
opsy | -0.181 |
ilies | -0.181 |
Token | of |
Token ID | 286 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
opsy | +0.701 |
ppy | +0.677 |
ilies | +0.671 |
per | +0.670 |
pper | +0.654 |
Neg logit contributions | |
irens | -0.887 |
Paula | -0.868 |
Paul | -0.843 |
iren | -0.825 |
flashing | -0.819 |
Token | paper |
Token ID | 3348 |
Feature activation | -5.32e-01 |
Pos logit contributions | |
irens | +0.115 |
Paula | +0.110 |
Paul | +0.105 |
iren | +0.104 |
spe | +0.102 |
Neg logit contributions | |
ilies | -0.092 |
per | -0.092 |
ppy | -0.090 |
pper | -0.088 |
opsy | -0.088 |
Token | . |
Token ID | 13 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
irens | +0.042 |
Paula | +0.041 |
Paul | +0.039 |
iren | +0.038 |
flashing | +0.038 |
Neg logit contributions | |
opsy | -0.034 |
ppy | -0.033 |
pper | -0.033 |
per | -0.033 |
ilies | -0.032 |
Token | She |
Token ID | 1375 |
Feature activation | +6.62e-01 |
Pos logit contributions | |
irens | +0.022 |
Paula | +0.021 |
iren | +0.020 |
spe | +0.020 |
Paul | +0.020 |
Neg logit contributions | |
per | -0.016 |
ilies | -0.016 |
opsy | -0.016 |
ppy | -0.015 |
ectar | -0.015 |
Token | couldn |
Token ID | 3521 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
per | +0.051 |
pper | +0.051 |
ppy | +0.050 |
ilies | +0.050 |
opsy | +0.050 |
Neg logit contributions | |
irens | -0.043 |
Paula | -0.042 |
Paul | -0.040 |
spe | -0.038 |
John | -0.038 |
Token | brush |
Token ID | 14093 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
per | +0.005 |
ppy | +0.005 |
pper | +0.005 |
ilies | +0.004 |
izards | +0.004 |
Neg logit contributions | |
Sarah | -0.004 |
Paula | -0.004 |
Paul | -0.004 |
John | -0.004 |
irens | -0.004 |
Token | under |
Token ID | 739 |
Feature activation | +7.84e-01 |
Pos logit contributions | |
opsy | +0.106 |
ilies | +0.099 |
pper | +0.098 |
ppy | +0.097 |
per | +0.096 |
Neg logit contributions | |
Paula | -0.110 |
irens | -0.109 |
Paul | -0.108 |
John | -0.106 |
Sarah | -0.104 |
Token | the |
Token ID | 262 |
Feature activation | -8.14e-01 |
Pos logit contributions | |
ppy | +0.053 |
opsy | +0.053 |
per | +0.053 |
ilies | +0.053 |
pper | +0.052 |
Neg logit contributions | |
irens | -0.059 |
Paula | -0.056 |
Paul | -0.055 |
John | -0.053 |
iren | -0.053 |
Token | l |
Token ID | 300 |
Feature activation | +4.26e+00 |
Pos logit contributions | |
irens | +0.059 |
iren | +0.057 |
flashing | +0.056 |
Paul | +0.053 |
Paula | +0.052 |
Neg logit contributions | |
ppy | -0.059 |
pper | -0.055 |
per | -0.054 |
ilies | -0.053 |
opsy | -0.053 |
Token | ily |
Token ID | 813 |
Feature activation | +7.43e+00 |
Pos logit contributions | |
thirst | +0.204 |
opsy | +0.200 |
pper | +0.189 |
ppy | +0.184 |
wait | +0.181 |
Neg logit contributions | |
iren | -0.380 |
irens | -0.376 |
stairs | -0.315 |
chairs | -0.310 |
Paula | -0.308 |
Token | pad |
Token ID | 14841 |
Feature activation | +1.12e+01 |
Pos logit contributions | |
opsy | +0.471 |
thirst | +0.446 |
ilies | +0.435 |
ectar | +0.423 |
per | +0.416 |
Neg logit contributions | |
Sarah | -0.518 |
Paula | -0.516 |
irens | -0.508 |
Smith | -0.500 |
John | -0.500 |
Token | in |
Token ID | 287 |
Feature activation | +2.80e+00 |
Pos logit contributions | |
opsy | +0.876 |
thirst | +0.822 |
ilies | +0.811 |
pollen | +0.775 |
ppy | +0.774 |
Neg logit contributions | |
irens | -0.674 |
iren | -0.653 |
Smith | -0.635 |
Paula | -0.635 |
Paul | -0.633 |
Token | my |
Token ID | 616 |
Feature activation | +4.39e-02 |
Pos logit contributions | |
per | +0.165 |
pper | +0.151 |
ppy | +0.149 |
ilies | +0.148 |
worm | +0.147 |
Neg logit contributions | |
irens | -0.242 |
Paula | -0.236 |
spe | -0.224 |
iren | -0.221 |
Paul | -0.220 |
Token | nest |
Token ID | 16343 |
Feature activation | +7.75e+00 |
Pos logit contributions | |
pper | +0.003 |
ppy | +0.003 |
ilies | +0.003 |
ectar | +0.003 |
per | +0.003 |
Neg logit contributions | |
Sarah | -0.003 |
Paula | -0.003 |
irens | -0.003 |
Paul | -0.003 |
John | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
opsy | +0.470 |
pper | +0.459 |
ppy | +0.459 |
per | +0.455 |
ilies | +0.455 |
Neg logit contributions | |
irens | -0.619 |
Paula | -0.600 |
Paul | -0.580 |
iren | -0.572 |
spe | -0.566 |
Token | You |
Token ID | 921 |
Feature activation | -2.89e-01 |
Pos logit contributions | |
Paul | +0.011 |
Paula | +0.011 |
irens | +0.011 |
John | +0.011 |
Sarah | +0.011 |
Neg logit contributions | |
opsy | -0.014 |
pper | -0.014 |
per | -0.014 |
ilies | -0.014 |
ppy | -0.014 |
Token | stayed |
Token ID | 9658 |
Feature activation | +4.07e+00 |
Pos logit contributions | |
ectar | +0.012 |
ilies | +0.012 |
per | +0.012 |
seeker | +0.011 |
ider | +0.011 |
Neg logit contributions | |
Paul | -0.009 |
John | -0.009 |
Paula | -0.009 |
spe | -0.009 |
Sarah | -0.009 |
Token | on |
Token ID | 319 |
Feature activation | -3.58e-01 |
Pos logit contributions | |
ppy | +0.275 |
pper | +0.272 |
ilies | +0.270 |
ily | +0.252 |
opsy | +0.248 |
Neg logit contributions | |
irens | -0.293 |
Paula | -0.291 |
Sarah | -0.281 |
Paul | -0.280 |
John | -0.277 |
Token | the |
Token ID | 262 |
Feature activation | -8.29e-01 |
Pos logit contributions | |
irens | +0.027 |
Paula | +0.026 |
iren | +0.025 |
Paul | +0.025 |
spe | +0.024 |
Neg logit contributions | |
per | -0.024 |
opsy | -0.023 |
ilies | -0.023 |
pper | -0.023 |
ppy | -0.023 |
Token | l |
Token ID | 300 |
Feature activation | +4.26e+00 |
Pos logit contributions | |
stairs | +0.050 |
flashing | +0.048 |
chaos | +0.046 |
irens | +0.046 |
hallway | +0.046 |
Neg logit contributions | |
ppy | -0.069 |
ilies | -0.061 |
ectar | -0.061 |
pper | -0.061 |
per | -0.060 |
Token | ily |
Token ID | 813 |
Feature activation | +7.40e+00 |
Pos logit contributions | |
wait | +0.183 |
ppy | +0.171 |
okay | +0.168 |
prett | +0.166 |
esty | +0.165 |
Neg logit contributions | |
iren | -0.369 |
irens | -0.362 |
doorway | -0.328 |
ons | -0.308 |
igsaw | -0.307 |
Token | pad |
Token ID | 14841 |
Feature activation | +1.12e+01 |
Pos logit contributions | |
ilies | +0.381 |
opsy | +0.375 |
ectar | +0.363 |
ily | +0.363 |
per | +0.362 |
Neg logit contributions | |
Paula | -0.631 |
Paul | -0.615 |
John | -0.609 |
Sarah | -0.603 |
spe | -0.589 |
Token | and |
Token ID | 290 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
opsy | +0.768 |
ilies | +0.752 |
ppy | +0.734 |
ily | +0.705 |
pper | +0.696 |
Neg logit contributions | |
John | -0.740 |
Paul | -0.738 |
Paula | -0.734 |
irens | -0.732 |
spe | -0.720 |
Token | watched |
Token ID | 7342 |
Feature activation | +9.62e-01 |
Pos logit contributions | |
Paula | +0.010 |
John | +0.010 |
Paul | +0.010 |
Sarah | +0.009 |
irens | +0.009 |
Neg logit contributions | |
ilies | -0.010 |
ppy | -0.010 |
per | -0.010 |
pper | -0.009 |
ectar | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | -8.22e-01 |
Pos logit contributions | |
pper | +0.070 |
ppy | +0.070 |
opsy | +0.068 |
per | +0.067 |
ilies | +0.066 |
Neg logit contributions | |
Paula | -0.066 |
irens | -0.065 |
Paul | -0.065 |
John | -0.065 |
Sarah | -0.064 |
Token | butterfly |
Token ID | 35113 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
irens | +0.059 |
chaos | +0.057 |
Paula | +0.057 |
flashing | +0.056 |
spe | +0.055 |
Neg logit contributions | |
ppy | -0.057 |
per | -0.056 |
pper | -0.056 |
ilies | -0.050 |
opsy | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
pper | +0.080 |
opsy | +0.076 |
ppy | +0.076 |
per | +0.076 |
seeker | +0.075 |
Neg logit contributions | |
Paula | -0.098 |
Sarah | -0.097 |
Paul | -0.093 |
flashing | -0.093 |
irens | -0.092 |
Token | sitting |
Token ID | 5586 |
Feature activation | +2.30e+00 |
Pos logit contributions | |
ppy | +0.437 |
opsy | +0.426 |
ilies | +0.425 |
pper | +0.423 |
per | +0.418 |
Neg logit contributions | |
irens | -0.423 |
Paul | -0.394 |
Paula | -0.388 |
flashing | -0.381 |
iren | -0.381 |
Token | on |
Token ID | 319 |
Feature activation | -3.45e-01 |
Pos logit contributions | |
ppy | +0.174 |
wait | +0.174 |
ilies | +0.171 |
pper | +0.170 |
opsy | +0.167 |
Neg logit contributions | |
irens | -0.142 |
iren | -0.127 |
Sarah | -0.121 |
Paul | -0.119 |
John | -0.119 |
Token | a |
Token ID | 257 |
Feature activation | +9.20e-01 |
Pos logit contributions | |
irens | +0.025 |
Paula | +0.025 |
Paul | +0.024 |
John | +0.024 |
Sarah | +0.023 |
Neg logit contributions | |
ppy | -0.024 |
ilies | -0.023 |
pper | -0.023 |
per | -0.023 |
opsy | -0.023 |
Token | l |
Token ID | 300 |
Feature activation | +4.32e+00 |
Pos logit contributions | |
ilies | +0.062 |
ppy | +0.061 |
ectar | +0.060 |
per | +0.059 |
ider | +0.059 |
Neg logit contributions | |
irens | -0.059 |
flashing | -0.058 |
iren | -0.058 |
million | -0.058 |
Paula | -0.057 |
Token | ily |
Token ID | 813 |
Feature activation | +7.44e+00 |
Pos logit contributions | |
wait | +0.184 |
prett | +0.168 |
anuts | +0.167 |
Deal | +0.166 |
esty | +0.166 |
Neg logit contributions | |
iren | -0.371 |
irens | -0.354 |
doorway | -0.323 |
igsaw | -0.293 |
seat | -0.286 |
Token | pad |
Token ID | 14841 |
Feature activation | +1.12e+01 |
Pos logit contributions | |
ectar | +0.296 |
opsy | +0.295 |
ilies | +0.293 |
per | +0.270 |
ily | +0.270 |
Neg logit contributions | |
Paula | -0.687 |
Paul | -0.681 |
irens | -0.674 |
John | -0.668 |
flashing | -0.666 |
Token | . |
Token ID | 13 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
opsy | +0.823 |
ilies | +0.788 |
thirst | +0.784 |
ppy | +0.783 |
pollen | +0.762 |
Neg logit contributions | |
irens | -0.664 |
Paul | -0.629 |
Smith | -0.622 |
spe | -0.621 |
iren | -0.619 |
Token | |
Token ID | 220 |
Feature activation | -9.36e-01 |
Pos logit contributions | |
irens | +0.011 |
Paula | +0.011 |
Paul | +0.011 |
iren | +0.010 |
John | +0.010 |
Neg logit contributions | |
per | -0.011 |
ppy | -0.011 |
pper | -0.011 |
ilies | -0.011 |
opsy | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | -5.11e-01 |
Pos logit contributions | |
irens | +0.059 |
Paula | +0.058 |
Paul | +0.055 |
John | +0.053 |
Sarah | +0.053 |
Neg logit contributions | |
pper | -0.073 |
per | -0.073 |
ppy | -0.072 |
opsy | -0.071 |
ilies | -0.070 |
Token |
|
Token ID | 198 |
Feature activation | -5.07e-01 |
Pos logit contributions | |
irens | +0.036 |
Paula | +0.035 |
Paul | +0.033 |
iren | +0.032 |
John | +0.032 |
Neg logit contributions | |
per | -0.037 |
ppy | -0.037 |
pper | -0.037 |
ilies | -0.036 |
opsy | -0.036 |
Token | " |
Token ID | 1 |
Feature activation | +3.25e+00 |
Pos logit contributions | |
irens | +0.042 |
Paula | +0.041 |
Sarah | +0.039 |
Paul | +0.039 |
iren | +0.039 |
Neg logit contributions | |
ilies | -0.031 |
opsy | -0.030 |
pper | -0.030 |
per | -0.029 |
ectar | -0.029 |
Token | , |
Token ID | 11 |
Feature activation | +3.54e-01 |
Pos logit contributions | |
irens | +0.049 |
Paula | +0.049 |
John | +0.048 |
Paul | +0.048 |
Sarah | +0.047 |
Neg logit contributions | |
opsy | -0.043 |
pper | -0.042 |
ppy | -0.041 |
per | -0.041 |
ilies | -0.039 |
Token | hopping |
Token ID | 47153 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
per | +0.023 |
pper | +0.023 |
ppy | +0.022 |
opsy | +0.022 |
ilies | +0.022 |
Neg logit contributions | |
Paula | -0.027 |
Paul | -0.027 |
John | -0.027 |
Sarah | -0.026 |
James | -0.025 |
Token | from |
Token ID | 422 |
Feature activation | +3.78e-02 |
Pos logit contributions | |
opsy | +0.155 |
pper | +0.152 |
comb | +0.149 |
ilies | +0.147 |
ppy | +0.146 |
Neg logit contributions | |
irens | -0.153 |
spe | -0.147 |
Paula | -0.147 |
Paul | -0.143 |
iren | -0.143 |
Token | l |
Token ID | 300 |
Feature activation | +4.28e+00 |
Pos logit contributions | |
ppy | +0.003 |
pper | +0.003 |
per | +0.003 |
opsy | +0.003 |
ilies | +0.003 |
Neg logit contributions | |
irens | -0.003 |
Paula | -0.003 |
Paul | -0.003 |
John | -0.003 |
Sarah | -0.003 |
Token | ily |
Token ID | 813 |
Feature activation | +7.44e+00 |
Pos logit contributions | |
wait | +0.189 |
pper | +0.182 |
comb | +0.179 |
opsy | +0.170 |
ppy | +0.161 |
Neg logit contributions | |
iren | -0.382 |
irens | -0.380 |
smart | -0.326 |
seat | -0.303 |
Paula | -0.301 |
Token | pad |
Token ID | 14841 |
Feature activation | +1.12e+01 |
Pos logit contributions | |
opsy | +0.470 |
ilies | +0.470 |
pper | +0.456 |
per | +0.456 |
ppy | +0.446 |
Neg logit contributions | |
Paula | -0.558 |
irens | -0.549 |
Paul | -0.535 |
John | -0.524 |
Sarah | -0.516 |
Token | to |
Token ID | 284 |
Feature activation | -5.95e-01 |
Pos logit contributions | |
opsy | +0.930 |
ppy | +0.854 |
ilies | +0.851 |
thirst | +0.835 |
pper | +0.826 |
Neg logit contributions | |
irens | -0.628 |
spe | -0.598 |
Paul | -0.590 |
iren | -0.588 |
John | -0.586 |
Token | l |
Token ID | 300 |
Feature activation | +4.26e+00 |
Pos logit contributions | |
Sarah | +0.040 |
John | +0.039 |
Paula | +0.039 |
Paul | +0.039 |
irens | +0.038 |
Neg logit contributions | |
ppy | -0.043 |
pper | -0.042 |
opsy | -0.041 |
per | -0.041 |
ilies | -0.041 |
Token | ily |
Token ID | 813 |
Feature activation | +7.43e+00 |
Pos logit contributions | |
wait | +0.183 |
pper | +0.181 |
comb | +0.176 |
opsy | +0.166 |
ppy | +0.163 |
Neg logit contributions | |
iren | -0.363 |
irens | -0.363 |
smart | -0.320 |
Paula | -0.312 |
John | -0.308 |
Token | pad |
Token ID | 14841 |
Feature activation | +1.12e+01 |
Pos logit contributions | |
opsy | +0.480 |
ilies | +0.475 |
pper | +0.464 |
per | +0.459 |
ily | +0.451 |
Neg logit contributions | |
Paula | -0.551 |
irens | -0.541 |
Paul | -0.528 |
John | -0.517 |
Sarah | -0.511 |
Token | . |
Token ID | 13 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
opsy | +0.886 |
ilies | +0.811 |
ppy | +0.806 |
thirst | +0.805 |
pper | +0.776 |
Neg logit contributions | |
irens | -0.668 |
Paul | -0.629 |
Paula | -0.629 |
spe | -0.628 |
iren | -0.627 |
Token | ily |
Token ID | 813 |
Feature activation | +7.44e+00 |
Pos logit contributions | |
wait | +0.189 |
pper | +0.182 |
comb | +0.179 |
opsy | +0.170 |
ppy | +0.161 |
Neg logit contributions | |
iren | -0.382 |
irens | -0.380 |
smart | -0.326 |
seat | -0.303 |
Paula | -0.301 |
Token | pad |
Token ID | 14841 |
Feature activation | +1.12e+01 |
Pos logit contributions | |
opsy | +0.470 |
ilies | +0.470 |
pper | +0.456 |
per | +0.456 |
ppy | +0.446 |
Neg logit contributions | |
Paula | -0.558 |
irens | -0.549 |
Paul | -0.535 |
John | -0.524 |
Sarah | -0.516 |
Token | to |
Token ID | 284 |
Feature activation | -5.95e-01 |
Pos logit contributions | |
opsy | +0.930 |
ppy | +0.854 |
ilies | +0.851 |
thirst | +0.835 |
pper | +0.826 |
Neg logit contributions | |
irens | -0.628 |
spe | -0.598 |
Paul | -0.590 |
iren | -0.588 |
John | -0.586 |
Token | l |
Token ID | 300 |
Feature activation | +4.26e+00 |
Pos logit contributions | |
Sarah | +0.040 |
John | +0.039 |
Paula | +0.039 |
Paul | +0.039 |
irens | +0.038 |
Neg logit contributions | |
ppy | -0.043 |
pper | -0.042 |
opsy | -0.041 |
per | -0.041 |
ilies | -0.041 |
Token | ily |
Token ID | 813 |
Feature activation | +7.43e+00 |
Pos logit contributions | |
wait | +0.183 |
pper | +0.181 |
comb | +0.176 |
opsy | +0.166 |
ppy | +0.163 |
Neg logit contributions | |
iren | -0.363 |
irens | -0.363 |
smart | -0.320 |
Paula | -0.312 |
John | -0.308 |
Token | pad |
Token ID | 14841 |
Feature activation | +1.12e+01 |
Pos logit contributions | |
opsy | +0.480 |
ilies | +0.475 |
pper | +0.464 |
per | +0.459 |
ily | +0.451 |
Neg logit contributions | |
Paula | -0.551 |
irens | -0.541 |
Paul | -0.528 |
John | -0.517 |
Sarah | -0.511 |
Token | . |
Token ID | 13 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
opsy | +0.886 |
ilies | +0.811 |
ppy | +0.806 |
thirst | +0.805 |
pper | +0.776 |
Neg logit contributions | |
irens | -0.668 |
Paul | -0.629 |
Paula | -0.629 |
spe | -0.628 |
iren | -0.627 |
Token | Would |
Token ID | 10928 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
Paula | +0.011 |
Paul | +0.011 |
irens | +0.010 |
John | +0.010 |
Sarah | +0.010 |
Neg logit contributions | |
pper | -0.013 |
per | -0.013 |
ppy | -0.012 |
opsy | -0.012 |
ilies | -0.012 |
Token | you |
Token ID | 345 |
Feature activation | +2.26e+00 |
Pos logit contributions | |
Paula | +0.138 |
irens | +0.138 |
Paul | +0.136 |
Sarah | +0.133 |
John | +0.133 |
Neg logit contributions | |
per | -0.099 |
ppy | -0.094 |
pper | -0.092 |
opsy | -0.091 |
comb | -0.089 |
Token | like |
Token ID | 588 |
Feature activation | +5.99e+00 |
Pos logit contributions | |
opard | +0.130 |
per | +0.125 |
opsy | +0.120 |
ster | +0.113 |
pper | +0.112 |
Neg logit contributions | |
Sarah | -0.182 |
John | -0.174 |
Paul | -0.168 |
Paula | -0.167 |
bell | -0.159 |
Token | to |
Token ID | 284 |
Feature activation | -6.13e-01 |
Pos logit contributions | |
per | +0.515 |
pper | +0.514 |
ppy | +0.507 |
opsy | +0.498 |
comb | +0.493 |
Neg logit contributions | |
irens | -0.342 |
Paula | -0.331 |
Paul | -0.322 |
Sarah | -0.311 |
John | -0.309 |
Token | when |
Token ID | 618 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
opsy | +0.170 |
ilies | +0.169 |
pper | +0.167 |
ppy | +0.166 |
per | +0.163 |
Neg logit contributions | |
irens | -0.206 |
Paula | -0.205 |
Paul | -0.202 |
John | -0.194 |
Sarah | -0.189 |
Token | he |
Token ID | 339 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
ilies | +0.139 |
per | +0.138 |
opsy | +0.136 |
ily | +0.135 |
dove | +0.132 |
Neg logit contributions | |
irens | -0.187 |
iren | -0.175 |
Paula | -0.173 |
spe | -0.161 |
Paul | -0.160 |
Token | heard |
Token ID | 2982 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
per | +0.009 |
ppy | +0.009 |
pper | +0.009 |
ilies | +0.009 |
opsy | +0.009 |
Neg logit contributions | |
irens | -0.010 |
Paula | -0.010 |
Paul | -0.009 |
iren | -0.009 |
John | -0.009 |
Token | loud |
Token ID | 7812 |
Feature activation | -3.12e+00 |
Pos logit contributions | |
irens | +0.120 |
Paula | +0.116 |
spe | +0.116 |
iren | +0.115 |
Smith | +0.109 |
Neg logit contributions | |
like | -0.043 |
ily | -0.043 |
per | -0.041 |
ilies | -0.039 |
dove | -0.037 |
Token | s |
Token ID | 264 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
spe | +0.189 |
Paula | +0.188 |
irens | +0.186 |
Paul | +0.171 |
Sarah | +0.171 |
Neg logit contributions | |
per | -0.236 |
like | -0.229 |
ilies | -0.219 |
ppy | -0.212 |
ily | -0.211 |
Token | irens |
Token ID | 42917 |
Feature activation | -1.36e+01 |
Pos logit contributions | |
Paula | +0.025 |
Paul | +0.024 |
spe | +0.022 |
John | +0.021 |
flashing | +0.019 |
Neg logit contributions | |
per | -0.158 |
ppy | -0.149 |
ilies | -0.149 |
pper | -0.149 |
ily | -0.145 |
Token | . |
Token ID | 13 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
irens | +1.249 |
Paula | +1.244 |
spe | +1.191 |
Paul | +1.178 |
Sarah | +1.166 |
Neg logit contributions | |
per | -0.659 |
like | -0.602 |
pper | -0.574 |
ppy | -0.566 |
ilies | -0.520 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
irens | +0.022 |
bul | +0.020 |
iren | +0.020 |
spe | +0.020 |
flashing | +0.019 |
Neg logit contributions | |
ily | -0.012 |
per | -0.011 |
like | -0.011 |
middle | -0.010 |
center | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | +3.21e-01 |
Pos logit contributions | |
per | +0.055 |
ppy | +0.052 |
pper | +0.050 |
ily | +0.050 |
ilies | +0.050 |
Neg logit contributions | |
irens | -0.132 |
Paula | -0.126 |
iren | -0.125 |
spe | -0.122 |
Paul | -0.117 |
Token | an |
Token ID | 281 |
Feature activation | +5.24e-01 |
Pos logit contributions | |
ily | +0.018 |
like | +0.018 |
wait | +0.017 |
per | +0.017 |
ilies | +0.017 |
Neg logit contributions | |
irens | -0.026 |
iren | -0.025 |
spe | -0.023 |
Paula | -0.023 |
parent | -0.022 |
Token | ambulance |
Token ID | 22536 |
Feature activation | -2.44e+00 |
Pos logit contributions | |
per | +0.039 |
ily | +0.037 |
ppy | +0.036 |
ilies | +0.036 |
pper | +0.036 |
Neg logit contributions | |
Paula | -0.034 |
irens | -0.034 |
spe | -0.032 |
Paul | -0.032 |
James | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | +3.46e-01 |
Pos logit contributions | |
irens | +0.069 |
Paula | +0.067 |
iren | +0.064 |
spe | +0.064 |
Paul | +0.063 |
Neg logit contributions | |
per | -0.030 |
ilies | -0.030 |
ppy | -0.030 |
opsy | -0.028 |
pper | -0.028 |
Token | they |
Token ID | 484 |
Feature activation | -2.23e+00 |
Pos logit contributions | |
ppy | +0.025 |
ilies | +0.025 |
per | +0.025 |
opsy | +0.025 |
ily | +0.024 |
Neg logit contributions | |
irens | -0.023 |
iren | -0.022 |
spe | -0.021 |
Paula | -0.021 |
flashing | -0.019 |
Token | both |
Token ID | 1111 |
Feature activation | +4.34e+00 |
Pos logit contributions | |
Paula | +0.145 |
Paul | +0.142 |
irens | +0.138 |
Sarah | +0.138 |
John | +0.137 |
Neg logit contributions | |
ectar | -0.165 |
opsy | -0.162 |
ilies | -0.160 |
per | -0.159 |
pper | -0.159 |
Token | heard |
Token ID | 2982 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
per | +0.305 |
pper | +0.302 |
opsy | +0.302 |
ilies | +0.300 |
ectar | +0.297 |
Neg logit contributions | |
Paula | -0.303 |
irens | -0.302 |
Paul | -0.292 |
spe | -0.282 |
John | -0.282 |
Token | s |
Token ID | 264 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
irens | +0.127 |
iren | +0.123 |
Paula | +0.123 |
spe | +0.122 |
parent | +0.115 |
Neg logit contributions | |
ily | -0.035 |
like | -0.032 |
ilies | -0.030 |
dove | -0.029 |
per | -0.028 |
Token | irens |
Token ID | 42917 |
Feature activation | -1.36e+01 |
Pos logit contributions | |
Paula | +0.042 |
Paul | +0.040 |
spe | +0.038 |
John | +0.036 |
flashing | +0.034 |
Neg logit contributions | |
per | -0.139 |
ppy | -0.135 |
pper | -0.133 |
ilies | -0.132 |
opsy | -0.130 |
Token | . |
Token ID | 13 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
Paula | +1.142 |
irens | +1.120 |
spe | +1.097 |
Sarah | +1.065 |
Paul | +1.063 |
Neg logit contributions | |
like | -0.746 |
per | -0.733 |
pper | -0.630 |
wait | -0.623 |
ppy | -0.617 |
Token | The |
Token ID | 383 |
Feature activation | +4.12e-01 |
Pos logit contributions | |
irens | +0.014 |
spe | +0.014 |
iren | +0.013 |
Paula | +0.013 |
flashing | +0.012 |
Neg logit contributions | |
per | -0.011 |
ppy | -0.011 |
ilies | -0.011 |
opsy | -0.010 |
ily | -0.010 |
Token | police |
Token ID | 1644 |
Feature activation | -5.43e+00 |
Pos logit contributions | |
careful | +0.023 |
ily | +0.023 |
like | +0.022 |
wait | +0.022 |
per | +0.022 |
Neg logit contributions | |
Paula | -0.034 |
James | -0.030 |
irens | -0.030 |
Paul | -0.030 |
John | -0.030 |
Token | had |
Token ID | 550 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
igsaw | +0.443 |
irens | +0.436 |
Paula | +0.398 |
telev | +0.395 |
parent | +0.389 |
Neg logit contributions | |
per | -0.298 |
like | -0.291 |
ppy | -0.286 |
dove | -0.277 |
pper | -0.276 |
Token | come |
Token ID | 1282 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
like | +0.009 |
per | +0.009 |
careful | +0.009 |
mate | +0.008 |
dove | +0.008 |
Neg logit contributions | |
irens | -0.019 |
iren | -0.018 |
Paula | -0.018 |
igsaw | -0.017 |
telev | -0.017 |
Token | a |
Token ID | 257 |
Feature activation | +9.44e-01 |
Pos logit contributions | |
irens | +0.139 |
Paula | +0.137 |
spe | +0.132 |
iren | +0.130 |
Paul | +0.129 |
Neg logit contributions | |
per | -0.064 |
like | -0.062 |
ppy | -0.060 |
ilies | -0.059 |
ily | -0.059 |
Token | fire |
Token ID | 2046 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
opsy | +0.057 |
worm | +0.057 |
per | +0.055 |
dove | +0.054 |
ppy | +0.054 |
Neg logit contributions | |
irens | -0.078 |
Paula | -0.075 |
Paul | -0.071 |
iren | -0.069 |
John | -0.068 |
Token | truck |
Token ID | 7779 |
Feature activation | -3.52e+00 |
Pos logit contributions | |
irens | +0.018 |
Paula | +0.017 |
Paul | +0.016 |
iren | +0.015 |
spe | +0.015 |
Neg logit contributions | |
per | -0.017 |
pper | -0.017 |
ppy | -0.016 |
comb | -0.016 |
opsy | -0.016 |
Token | with |
Token ID | 351 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
irens | +0.327 |
Paula | +0.327 |
telev | +0.307 |
spe | +0.294 |
iren | +0.294 |
Neg logit contributions | |
per | -0.175 |
pper | -0.173 |
like | -0.170 |
ppy | -0.154 |
wait | -0.145 |
Token | s |
Token ID | 264 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
Paula | +0.031 |
irens | +0.031 |
spe | +0.029 |
iren | +0.029 |
Paul | +0.029 |
Neg logit contributions | |
per | -0.014 |
ilies | -0.013 |
worm | -0.013 |
ily | -0.013 |
ppy | -0.013 |
Token | irens |
Token ID | 42917 |
Feature activation | -1.35e+01 |
Pos logit contributions | |
Paula | +0.033 |
Paul | +0.029 |
spe | +0.028 |
John | +0.026 |
James | +0.025 |
Neg logit contributions | |
per | -0.147 |
ppy | -0.146 |
pper | -0.143 |
ilies | -0.141 |
opsy | -0.140 |
Token | that |
Token ID | 326 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
Paula | +1.312 |
irens | +1.272 |
spe | +1.256 |
Paul | +1.224 |
Sarah | +1.217 |
Neg logit contributions | |
per | -0.633 |
pper | -0.564 |
like | -0.547 |
ppy | -0.538 |
ilies | -0.502 |
Token | went |
Token ID | 1816 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
Paula | +0.111 |
irens | +0.110 |
spe | +0.100 |
Paul | +0.100 |
iren | +0.099 |
Neg logit contributions | |
per | -0.076 |
pper | -0.074 |
ppy | -0.073 |
ilies | -0.073 |
dove | -0.070 |
Token | " |
Token ID | 366 |
Feature activation | +8.82e-01 |
Pos logit contributions | |
Paula | +0.206 |
irens | +0.200 |
Sarah | +0.188 |
Paul | +0.187 |
spe | +0.187 |
Neg logit contributions | |
per | -0.116 |
like | -0.114 |
pper | -0.098 |
ppy | -0.097 |
ilies | -0.091 |
Token | we |
Token ID | 732 |
Feature activation | -3.84e+00 |
Pos logit contributions | |
ppy | +0.057 |
pper | +0.057 |
per | +0.056 |
ilies | +0.054 |
ectar | +0.053 |
Neg logit contributions | |
Paula | -0.066 |
irens | -0.065 |
Paul | -0.062 |
James | -0.060 |
Sarah | -0.060 |
Token | e |
Token ID | 68 |
Feature activation | +4.20e+00 |
Pos logit contributions | |
irens | +0.339 |
Paula | +0.330 |
Paul | +0.321 |
iren | +0.315 |
John | +0.310 |
Neg logit contributions | |
opsy | -0.206 |
per | -0.205 |
ilies | -0.198 |
pper | -0.197 |
ectar | -0.196 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
irens | +0.020 |
Paula | +0.019 |
iren | +0.018 |
spe | +0.018 |
Paul | +0.018 |
Neg logit contributions | |
opsy | -0.016 |
per | -0.016 |
ppy | -0.016 |
ilies | -0.016 |
pper | -0.016 |
Token | , |
Token ID | 11 |
Feature activation | +3.48e-01 |
Pos logit contributions | |
per | +0.088 |
pper | +0.087 |
ppy | +0.086 |
ilies | +0.085 |
opsy | +0.084 |
Neg logit contributions | |
irens | -0.115 |
Paula | -0.112 |
Paul | -0.109 |
John | -0.105 |
Sarah | -0.105 |
Token | they |
Token ID | 484 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
per | +0.025 |
pper | +0.025 |
ppy | +0.025 |
ilies | +0.024 |
opsy | +0.024 |
Neg logit contributions | |
irens | -0.025 |
Paula | -0.025 |
Paul | -0.024 |
John | -0.023 |
Sarah | -0.023 |
Token | heard |
Token ID | 2982 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
irens | +0.147 |
Paula | +0.142 |
Paul | +0.136 |
iren | +0.129 |
John | +0.129 |
Neg logit contributions | |
per | -0.166 |
pper | -0.163 |
opsy | -0.162 |
ilies | -0.162 |
ppy | -0.161 |
Token | s |
Token ID | 264 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
irens | +0.119 |
spe | +0.117 |
Paula | +0.116 |
iren | +0.114 |
Paul | +0.110 |
Neg logit contributions | |
ily | -0.037 |
ilies | -0.033 |
per | -0.032 |
like | -0.031 |
ppy | -0.030 |
Token | irens |
Token ID | 42917 |
Feature activation | -1.35e+01 |
Pos logit contributions | |
Paula | +0.023 |
Paul | +0.021 |
spe | +0.019 |
irens | +0.018 |
John | +0.017 |
Neg logit contributions | |
per | -0.150 |
ppy | -0.147 |
pper | -0.146 |
ilies | -0.145 |
opsy | -0.144 |
Token | and |
Token ID | 290 |
Feature activation | -1.99e-01 |
Pos logit contributions | |
Paula | +1.189 |
irens | +1.165 |
spe | +1.156 |
Paul | +1.116 |
Sarah | +1.100 |
Neg logit contributions | |
per | -0.706 |
like | -0.676 |
pper | -0.620 |
ppy | -0.609 |
wait | -0.568 |
Token | a |
Token ID | 257 |
Feature activation | +9.72e-01 |
Pos logit contributions | |
irens | +0.016 |
igsaw | +0.015 |
iren | +0.015 |
Paula | +0.015 |
parent | +0.014 |
Neg logit contributions | |
like | -0.011 |
dove | -0.011 |
wait | -0.011 |
per | -0.010 |
ppy | -0.010 |
Token | cop |
Token ID | 2243 |
Feature activation | -6.08e+00 |
Pos logit contributions | |
dove | +0.053 |
worm | +0.052 |
opsy | +0.048 |
per | +0.048 |
hop | +0.047 |
Neg logit contributions | |
irens | -0.080 |
Paula | -0.078 |
John | -0.073 |
Paul | -0.073 |
James | -0.071 |
Token | car |
Token ID | 1097 |
Feature activation | -2.60e+00 |
Pos logit contributions | |
irens | +0.509 |
Paula | +0.469 |
iren | +0.454 |
lights | +0.453 |
telev | +0.438 |
Neg logit contributions | |
per | -0.350 |
like | -0.328 |
dove | -0.325 |
wait | -0.324 |
pper | -0.321 |
Token | pulled |
Token ID | 5954 |
Feature activation | -3.00e+00 |
Pos logit contributions | |
Paula | +0.233 |
igsaw | +0.227 |
irens | +0.218 |
telev | +0.216 |
James | +0.213 |
Neg logit contributions | |
like | -0.131 |
per | -0.128 |
wait | -0.117 |
pper | -0.115 |
dove | -0.110 |
Token | flashing |
Token ID | 25293 |
Feature activation | -8.88e+00 |
Pos logit contributions | |
irens | +0.029 |
Paula | +0.029 |
spe | +0.028 |
iren | +0.028 |
Smith | +0.027 |
Neg logit contributions | |
ily | -0.011 |
like | -0.010 |
per | -0.010 |
careful | -0.010 |
ilies | -0.010 |
Token | lights |
Token ID | 7588 |
Feature activation | -5.19e+00 |
Pos logit contributions | |
Paula | +0.632 |
spe | +0.597 |
James | +0.587 |
Paul | +0.570 |
Smith | +0.567 |
Neg logit contributions | |
like | -0.616 |
per | -0.568 |
ily | -0.564 |
wings | -0.561 |
aw | -0.552 |
Token | and |
Token ID | 290 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
Paula | +0.523 |
irens | +0.521 |
Paul | +0.498 |
spe | +0.498 |
James | +0.492 |
Neg logit contributions | |
per | -0.203 |
like | -0.203 |
pper | -0.172 |
ilies | -0.162 |
ppy | -0.159 |
Token | loud |
Token ID | 7812 |
Feature activation | -3.10e+00 |
Pos logit contributions | |
irens | +0.015 |
Paula | +0.014 |
iren | +0.014 |
spe | +0.013 |
Paul | +0.013 |
Neg logit contributions | |
like | -0.010 |
dove | -0.010 |
ilies | -0.009 |
worm | -0.009 |
ily | -0.009 |
Token | s |
Token ID | 264 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
Paula | +0.197 |
spe | +0.186 |
irens | +0.184 |
Sarah | +0.175 |
Paul | +0.175 |
Neg logit contributions | |
per | -0.225 |
like | -0.224 |
wings | -0.210 |
feathers | -0.205 |
ilies | -0.204 |
Token | irens |
Token ID | 42917 |
Feature activation | -1.35e+01 |
Pos logit contributions | |
Paula | +0.028 |
Paul | +0.026 |
spe | +0.026 |
James | +0.024 |
John | +0.023 |
Neg logit contributions | |
per | -0.148 |
pper | -0.143 |
ilies | -0.142 |
ppy | -0.142 |
ily | -0.139 |
Token | . |
Token ID | 13 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
Paula | +1.280 |
irens | +1.249 |
Paul | +1.209 |
Sarah | +1.204 |
spe | +1.195 |
Neg logit contributions | |
per | -0.643 |
like | -0.557 |
pper | -0.549 |
ppy | -0.534 |
ilies | -0.520 |
Token | She |
Token ID | 1375 |
Feature activation | +6.85e-01 |
Pos logit contributions | |
irens | +0.016 |
Paula | +0.015 |
Paul | +0.014 |
iren | +0.014 |
spe | +0.014 |
Neg logit contributions | |
per | -0.013 |
ppy | -0.013 |
opsy | -0.013 |
ilies | -0.013 |
pper | -0.012 |
Token | asked |
Token ID | 1965 |
Feature activation | -4.67e-01 |
Pos logit contributions | |
per | +0.043 |
ppy | +0.043 |
opsy | +0.042 |
pper | +0.042 |
ilies | +0.040 |
Neg logit contributions | |
irens | -0.055 |
Paula | -0.052 |
iren | -0.050 |
Paul | -0.049 |
John | -0.048 |
Token | her |
Token ID | 607 |
Feature activation | +3.74e-01 |
Pos logit contributions | |
irens | +0.038 |
Paula | +0.036 |
iren | +0.035 |
Paul | +0.034 |
spe | +0.034 |
Neg logit contributions | |
per | -0.029 |
ppy | -0.029 |
ilies | -0.028 |
pper | -0.028 |
opsy | -0.028 |
Token | mom |
Token ID | 1995 |
Feature activation | +9.06e-02 |
Pos logit contributions | |
ppy | +0.019 |
opsy | +0.019 |
per | +0.018 |
ilies | +0.018 |
ily | +0.018 |
Neg logit contributions | |
irens | -0.035 |
Paula | -0.032 |
iren | -0.032 |
spe | -0.031 |
Paul | -0.031 |
Token | heard |
Token ID | 2982 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
irens | +0.092 |
Paula | +0.089 |
Paul | +0.086 |
iren | +0.082 |
John | +0.081 |
Neg logit contributions | |
per | -0.093 |
ppy | -0.091 |
ilies | -0.091 |
pper | -0.091 |
opsy | -0.091 |
Token | people |
Token ID | 661 |
Feature activation | -6.26e-01 |
Pos logit contributions | |
irens | +0.121 |
spe | +0.119 |
Paula | +0.118 |
iren | +0.116 |
parent | +0.112 |
Neg logit contributions | |
like | -0.034 |
ily | -0.034 |
dove | -0.033 |
worm | -0.032 |
per | -0.030 |
Token | shouting |
Token ID | 19642 |
Feature activation | -6.43e+00 |
Pos logit contributions | |
irens | +0.041 |
Paula | +0.040 |
Paul | +0.037 |
spe | +0.036 |
James | +0.035 |
Neg logit contributions | |
per | -0.047 |
like | -0.045 |
ppy | -0.045 |
pper | -0.044 |
ilies | -0.044 |
Token | and |
Token ID | 290 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
irens | +0.583 |
Paula | +0.567 |
Paul | +0.545 |
iren | +0.535 |
spe | +0.526 |
Neg logit contributions | |
per | -0.336 |
ppy | -0.330 |
pper | -0.324 |
ilies | -0.323 |
opsy | -0.320 |
Token | s |
Token ID | 264 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
irens | +0.016 |
Paula | +0.015 |
iren | +0.015 |
igsaw | +0.014 |
spe | +0.014 |
Neg logit contributions | |
per | -0.014 |
like | -0.014 |
ppy | -0.014 |
wait | -0.014 |
dove | -0.013 |
Token | irens |
Token ID | 42917 |
Feature activation | -1.35e+01 |
Pos logit contributions | |
spe | +0.026 |
Paula | +0.026 |
Paul | +0.022 |
James | +0.020 |
John | +0.019 |
Neg logit contributions | |
per | -0.151 |
pper | -0.148 |
ppy | -0.147 |
ilies | -0.143 |
opsy | -0.140 |
Token | w |
Token ID | 266 |
Feature activation | +2.73e+00 |
Pos logit contributions | |
Paula | +1.246 |
irens | +1.225 |
spe | +1.163 |
Paul | +1.162 |
Sarah | +1.151 |
Neg logit contributions | |
per | -0.675 |
pper | -0.610 |
ppy | -0.604 |
like | -0.594 |
ilies | -0.573 |
Token | ailing |
Token ID | 11608 |
Feature activation | -8.34e-01 |
Pos logit contributions | |
per | +0.117 |
ppy | +0.115 |
pper | +0.115 |
opsy | +0.115 |
ilies | +0.112 |
Neg logit contributions | |
irens | -0.277 |
Paula | -0.261 |
iren | -0.255 |
Paul | -0.253 |
flashing | -0.245 |
Token | . |
Token ID | 13 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
irens | +0.081 |
Paula | +0.080 |
Sarah | +0.077 |
iren | +0.076 |
igsaw | +0.074 |
Neg logit contributions | |
per | -0.035 |
like | -0.031 |
pper | -0.026 |
ilies | -0.025 |
dove | -0.025 |
Token | They |
Token ID | 1119 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
irens | +0.013 |
Paula | +0.013 |
iren | +0.012 |
Paul | +0.012 |
spe | +0.012 |
Neg logit contributions | |
per | -0.011 |
ilies | -0.011 |
opsy | -0.011 |
ily | -0.011 |
ppy | -0.011 |
Token | were |
Token ID | 547 |
Feature activation | -6.65e-01 |
Pos logit contributions | |
irens | +0.103 |
Paula | +0.097 |
iren | +0.094 |
Paul | +0.090 |
spe | +0.090 |
Neg logit contributions | |
ppy | -0.084 |
per | -0.083 |
pper | -0.080 |
opsy | -0.078 |
ilies | -0.078 |
Token | Tim |
Token ID | 5045 |
Feature activation | -2.10e+00 |
Pos logit contributions | |
per | +0.005 |
ilies | +0.005 |
ppy | +0.005 |
opsy | +0.004 |
pper | +0.004 |
Neg logit contributions | |
irens | -0.006 |
Paula | -0.005 |
Paul | -0.005 |
spe | -0.005 |
iren | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | +5.09e-02 |
Pos logit contributions | |
irens | +0.173 |
Paula | +0.167 |
Paul | +0.161 |
iren | +0.158 |
John | +0.156 |
Neg logit contributions | |
ilies | -0.124 |
opsy | -0.123 |
per | -0.122 |
pper | -0.121 |
ppy | -0.120 |
Token | Sam |
Token ID | 3409 |
Feature activation | -3.23e-01 |
Pos logit contributions | |
per | +0.003 |
ppy | +0.003 |
ilies | +0.003 |
opsy | +0.003 |
pper | +0.003 |
Neg logit contributions | |
irens | -0.004 |
Paula | -0.004 |
Paul | -0.004 |
iren | -0.004 |
spe | -0.003 |
Token | heard |
Token ID | 2982 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
irens | +0.018 |
Paula | +0.018 |
flashing | +0.018 |
Paul | +0.017 |
John | +0.017 |
Neg logit contributions | |
ectar | -0.027 |
opsy | -0.027 |
ilies | -0.026 |
pper | -0.025 |
per | -0.025 |
Token | s |
Token ID | 264 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
irens | +0.099 |
Paula | +0.098 |
iren | +0.093 |
spe | +0.093 |
Paul | +0.088 |
Neg logit contributions | |
ily | -0.058 |
ilies | -0.058 |
per | -0.057 |
worm | -0.055 |
like | -0.055 |
Token | irens |
Token ID | 42917 |
Feature activation | -1.35e+01 |
Pos logit contributions | |
Paula | +0.019 |
Paul | +0.017 |
spe | +0.014 |
John | +0.013 |
James | +0.011 |
Neg logit contributions | |
per | -0.154 |
ppy | -0.150 |
ilies | -0.149 |
pper | -0.149 |
ily | -0.147 |
Token | and |
Token ID | 290 |
Feature activation | +4.96e-02 |
Pos logit contributions | |
Paula | +1.316 |
irens | +1.281 |
Paul | +1.260 |
spe | +1.236 |
Sarah | +1.215 |
Neg logit contributions | |
per | -0.636 |
ppy | -0.548 |
like | -0.543 |
pper | -0.541 |
ilies | -0.503 |
Token | saw |
Token ID | 2497 |
Feature activation | +8.81e-01 |
Pos logit contributions | |
per | +0.003 |
ppy | +0.003 |
ilies | +0.003 |
worm | +0.003 |
dove | +0.003 |
Neg logit contributions | |
irens | -0.004 |
Paula | -0.004 |
iren | -0.004 |
Paul | -0.004 |
spe | -0.003 |
Token | fire |
Token ID | 2046 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
ilies | +0.049 |
per | +0.049 |
ily | +0.049 |
like | +0.048 |
worm | +0.047 |
Neg logit contributions | |
Paula | -0.075 |
irens | -0.071 |
spe | -0.068 |
Paul | -0.067 |
iren | -0.067 |
Token | trucks |
Token ID | 13960 |
Feature activation | -6.47e+00 |
Pos logit contributions | |
Paula | +0.021 |
irens | +0.020 |
Sarah | +0.020 |
Sarah | +0.020 |
Paul | +0.020 |
Neg logit contributions | |
per | -0.014 |
pper | -0.013 |
ppy | -0.012 |
like | -0.012 |
comb | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +7.85e-02 |
Pos logit contributions | |
Paula | +0.619 |
irens | +0.598 |
Paul | +0.584 |
Sarah | +0.562 |
spe | +0.554 |
Neg logit contributions | |
per | -0.328 |
pper | -0.298 |
like | -0.286 |
ppy | -0.284 |
ilies | -0.280 |
Token | some |
Token ID | 617 |
Feature activation | +3.47e+00 |
Pos logit contributions | |
ilies | +0.037 |
ily | +0.036 |
per | +0.036 |
worm | +0.033 |
like | +0.033 |
Neg logit contributions | |
irens | -0.085 |
Paula | -0.083 |
iren | -0.081 |
spe | -0.080 |
Paul | -0.077 |
Token | flashing |
Token ID | 25293 |
Feature activation | -8.87e+00 |
Pos logit contributions | |
ilies | +0.205 |
per | +0.201 |
ily | +0.192 |
worm | +0.190 |
dove | +0.189 |
Neg logit contributions | |
Paula | -0.278 |
irens | -0.274 |
Paul | -0.263 |
John | -0.253 |
James | -0.249 |
Token | lights |
Token ID | 7588 |
Feature activation | -5.21e+00 |
Pos logit contributions | |
Paula | +0.658 |
Mechan | +0.609 |
Paul | +0.601 |
spe | +0.598 |
irens | +0.598 |
Neg logit contributions | |
like | -0.616 |
per | -0.526 |
ily | -0.508 |
ilies | -0.486 |
at | -0.482 |
Token | and |
Token ID | 290 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
irens | +0.523 |
Paula | +0.515 |
Paul | +0.492 |
spe | +0.490 |
iren | +0.483 |
Neg logit contributions | |
per | -0.209 |
like | -0.207 |
pper | -0.180 |
ilies | -0.178 |
comb | -0.170 |
Token | s |
Token ID | 264 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
irens | +0.013 |
Paula | +0.012 |
iren | +0.012 |
Paul | +0.011 |
spe | +0.011 |
Neg logit contributions | |
like | -0.012 |
ilies | -0.012 |
per | -0.012 |
dove | -0.012 |
ily | -0.012 |
Token | irens |
Token ID | 42917 |
Feature activation | -1.34e+01 |
Pos logit contributions | |
spe | +0.029 |
Paula | +0.026 |
Paul | +0.022 |
John | +0.021 |
James | +0.021 |
Neg logit contributions | |
per | -0.140 |
ppy | -0.135 |
pper | -0.132 |
ilies | -0.131 |
ars | -0.130 |
Token | . |
Token ID | 13 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
Paula | +1.229 |
irens | +1.223 |
Paul | +1.161 |
Sarah | +1.154 |
spe | +1.152 |
Neg logit contributions | |
per | -0.672 |
like | -0.602 |
pper | -0.586 |
ppy | -0.569 |
ilies | -0.557 |
Token | They |
Token ID | 1119 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
irens | +0.007 |
Paula | +0.006 |
Paul | +0.006 |
John | +0.006 |
Sarah | +0.006 |
Neg logit contributions | |
pper | -0.008 |
ppy | -0.008 |
per | -0.008 |
opsy | -0.008 |
comb | -0.008 |
Token | were |
Token ID | 547 |
Feature activation | -5.93e-01 |
Pos logit contributions | |
irens | +0.097 |
Paula | +0.092 |
iren | +0.089 |
Paul | +0.087 |
John | +0.083 |
Neg logit contributions | |
per | -0.078 |
ppy | -0.077 |
pper | -0.074 |
ilies | -0.074 |
opsy | -0.072 |
Token | scared |
Token ID | 12008 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
irens | +0.048 |
Paula | +0.046 |
Paul | +0.044 |
iren | +0.044 |
spe | +0.043 |
Neg logit contributions | |
per | -0.036 |
ilies | -0.035 |
ppy | -0.034 |
pper | -0.034 |
opsy | -0.033 |
Token | and |
Token ID | 290 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
Paula | +0.234 |
irens | +0.234 |
spe | +0.227 |
Mechan | +0.223 |
iren | +0.222 |
Neg logit contributions | |
like | -0.092 |
per | -0.088 |
ppy | -0.085 |
ily | -0.084 |
ilies | -0.078 |
Token | car |
Token ID | 1097 |
Feature activation | -2.62e+00 |
Pos logit contributions | |
Paula | +0.398 |
irens | +0.395 |
Paul | +0.358 |
Sarah | +0.357 |
spe | +0.354 |
Neg logit contributions | |
per | -0.373 |
pper | -0.343 |
dove | -0.336 |
like | -0.335 |
ilies | -0.330 |
Token | . |
Token ID | 13 |
Feature activation | -1.68e-02 |
Pos logit contributions | |
Paula | +0.259 |
irens | +0.246 |
Paul | +0.239 |
Sarah | +0.235 |
Sarah | +0.233 |
Neg logit contributions | |
per | -0.126 |
pper | -0.111 |
like | -0.108 |
ilies | -0.097 |
dove | -0.096 |
Token | They |
Token ID | 1119 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
irens | +0.001 |
Paula | +0.001 |
Paul | +0.001 |
iren | +0.001 |
spe | +0.001 |
Neg logit contributions | |
per | -0.001 |
ppy | -0.001 |
ilies | -0.001 |
opsy | -0.001 |
pper | -0.001 |
Token | had |
Token ID | 550 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
irens | +0.098 |
Paula | +0.094 |
iren | +0.089 |
Paul | +0.086 |
Sarah | +0.084 |
Neg logit contributions | |
per | -0.071 |
ppy | -0.069 |
pper | -0.067 |
dove | -0.065 |
ilies | -0.065 |
Token | s |
Token ID | 264 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
per | +0.011 |
careful | +0.010 |
like | +0.010 |
pper | +0.010 |
dove | +0.010 |
Neg logit contributions | |
irens | -0.018 |
Paula | -0.018 |
iren | -0.017 |
Paul | -0.016 |
spe | -0.016 |
Token | irens |
Token ID | 42917 |
Feature activation | -1.34e+01 |
Pos logit contributions | |
Paula | +0.023 |
spe | +0.019 |
Paul | +0.018 |
John | +0.015 |
James | +0.015 |
Neg logit contributions | |
per | -0.145 |
ppy | -0.142 |
pper | -0.141 |
ilies | -0.141 |
ars | -0.138 |
Token | and |
Token ID | 290 |
Feature activation | -5.93e-02 |
Pos logit contributions | |
Paula | +1.400 |
irens | +1.360 |
Sarah | +1.309 |
Paul | +1.301 |
spe | +1.284 |
Neg logit contributions | |
per | -0.543 |
like | -0.480 |
pper | -0.451 |
ppy | -0.421 |
ilies | -0.387 |
Token | lights |
Token ID | 7588 |
Feature activation | -5.21e+00 |
Pos logit contributions | |
Paula | +0.004 |
irens | +0.004 |
iren | +0.004 |
Paul | +0.004 |
spe | +0.004 |
Neg logit contributions | |
per | -0.004 |
dove | -0.004 |
like | -0.004 |
ilies | -0.004 |
ppy | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
Paula | +0.526 |
irens | +0.525 |
Paul | +0.489 |
Sarah | +0.485 |
iren | +0.482 |
Neg logit contributions | |
per | -0.210 |
like | -0.209 |
pper | -0.184 |
ilies | -0.180 |
ppy | -0.169 |
Token | They |
Token ID | 1119 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
per | +0.001 |
ppy | +0.001 |
pper | +0.001 |
opsy | +0.001 |
ilies | +0.001 |
Neg logit contributions | |
irens | -0.001 |
Paula | -0.001 |
Paul | -0.001 |
iren | -0.001 |
John | -0.001 |
Token | were |
Token ID | 547 |
Feature activation | -5.89e-01 |
Pos logit contributions | |
irens | +0.098 |
Paula | +0.092 |
iren | +0.088 |
Paul | +0.085 |
Sarah | +0.082 |
Neg logit contributions | |
per | -0.071 |
ppy | -0.069 |
pper | -0.067 |
ilies | -0.066 |
dove | -0.065 |
Token | drove |
Token ID | 10357 |
Feature activation | -6.49e+00 |
Pos logit contributions | |
Paula | +0.189 |
irens | +0.178 |
igsaw | +0.169 |
telev | +0.168 |
channels | +0.164 |
Neg logit contributions | |
per | -0.193 |
pper | -0.187 |
dove | -0.178 |
like | -0.172 |
ppy | -0.168 |
Token | by |
Token ID | 416 |
Feature activation | +2.20e+00 |
Pos logit contributions | |
irens | +0.598 |
Paula | +0.563 |
iren | +0.547 |
Paul | +0.528 |
telev | +0.527 |
Neg logit contributions | |
per | -0.335 |
pper | -0.303 |
ppy | -0.303 |
like | -0.290 |
ily | -0.284 |
Token | with |
Token ID | 351 |
Feature activation | -2.92e-01 |
Pos logit contributions | |
per | +0.110 |
ppy | +0.095 |
pper | +0.095 |
ilies | +0.093 |
ily | +0.093 |
Neg logit contributions | |
irens | -0.210 |
Paula | -0.201 |
iren | -0.197 |
Paul | -0.188 |
flashing | -0.186 |
Token | its |
Token ID | 663 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
irens | +0.026 |
iren | +0.024 |
spe | +0.024 |
Paula | +0.024 |
telev | +0.023 |
Neg logit contributions | |
per | -0.015 |
ily | -0.015 |
ilies | -0.015 |
ppy | -0.014 |
pper | -0.014 |
Token | s |
Token ID | 264 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
per | +0.096 |
pper | +0.093 |
opsy | +0.092 |
ilies | +0.091 |
ppy | +0.090 |
Neg logit contributions | |
irens | -0.129 |
Paula | -0.124 |
iren | -0.115 |
Paul | -0.114 |
John | -0.110 |
Token | irens |
Token ID | 42917 |
Feature activation | -1.34e+01 |
Pos logit contributions | |
Paula | +0.030 |
Paul | +0.024 |
James | +0.023 |
John | +0.023 |
Sarah | +0.020 |
Neg logit contributions | |
per | -0.147 |
ppy | -0.143 |
pper | -0.143 |
ilies | -0.138 |
ars | -0.137 |
Token | bl |
Token ID | 698 |
Feature activation | +4.53e-01 |
Pos logit contributions | |
irens | +1.276 |
Paula | +1.268 |
Sarah | +1.172 |
Paul | +1.172 |
telev | +1.166 |
Neg logit contributions | |
per | -0.649 |
pper | -0.581 |
ppy | -0.571 |
like | -0.554 |
ilies | -0.511 |
Token | aring |
Token ID | 1723 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
ilies | +0.030 |
opsy | +0.030 |
per | +0.029 |
ectar | +0.029 |
comb | +0.029 |
Neg logit contributions | |
irens | -0.033 |
spe | -0.031 |
Paula | -0.031 |
Paul | -0.031 |
iren | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | +5.84e-03 |
Pos logit contributions | |
irens | +0.144 |
Paula | +0.138 |
iren | +0.131 |
Paul | +0.127 |
spe | +0.125 |
Neg logit contributions | |
per | -0.106 |
ppy | -0.102 |
ily | -0.101 |
ilies | -0.100 |
like | -0.098 |
Token | The |
Token ID | 383 |
Feature activation | +2.79e-01 |
Pos logit contributions | |
per | +0.000 |
ilies | +0.000 |
ppy | +0.000 |
ily | +0.000 |
opsy | +0.000 |
Neg logit contributions | |
irens | -0.000 |
Paula | -0.000 |
iren | -0.000 |
flashing | -0.000 |
spe | -0.000 |
Token | loud |
Token ID | 7812 |
Feature activation | -3.18e+00 |
Pos logit contributions | |
per | +0.020 |
ily | +0.019 |
ilies | +0.019 |
pper | +0.019 |
worm | +0.019 |
Neg logit contributions | |
irens | -0.019 |
Paula | -0.019 |
Paul | -0.018 |
John | -0.017 |
iren | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
Paula | +0.365 |
irens | +0.362 |
Paul | +0.347 |
spe | +0.340 |
John | +0.331 |
Neg logit contributions | |
per | -0.195 |
ppy | -0.187 |
pper | -0.183 |
ilies | -0.181 |
ily | -0.174 |
Token | They |
Token ID | 1119 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
irens | +0.001 |
Paula | +0.001 |
Paul | +0.001 |
iren | +0.001 |
John | +0.001 |
Neg logit contributions | |
per | -0.001 |
ppy | -0.001 |
pper | -0.001 |
opsy | -0.001 |
ilies | -0.001 |
Token | heard |
Token ID | 2982 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
irens | +0.090 |
Paula | +0.088 |
Paul | +0.083 |
iren | +0.080 |
spe | +0.079 |
Neg logit contributions | |
per | -0.077 |
ppy | -0.076 |
pper | -0.074 |
ilies | -0.074 |
opsy | -0.072 |
Token | the |
Token ID | 262 |
Feature activation | -8.73e-01 |
Pos logit contributions | |
irens | +0.091 |
Paula | +0.088 |
spe | +0.086 |
iren | +0.084 |
Paul | +0.084 |
Neg logit contributions | |
per | -0.072 |
ilies | -0.072 |
ily | -0.070 |
ppy | -0.067 |
opsy | -0.067 |
Token | s |
Token ID | 264 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
Paula | +0.055 |
irens | +0.055 |
Paul | +0.051 |
spe | +0.048 |
Sarah | +0.048 |
Neg logit contributions | |
per | -0.068 |
ilies | -0.065 |
pper | -0.064 |
ppy | -0.064 |
opsy | -0.064 |
Token | irens |
Token ID | 42917 |
Feature activation | -1.34e+01 |
Pos logit contributions | |
Paula | +0.027 |
Paul | +0.025 |
spe | +0.022 |
John | +0.021 |
Sarah | +0.018 |
Neg logit contributions | |
per | -0.145 |
ppy | -0.137 |
pper | -0.137 |
ily | -0.136 |
ars | -0.134 |
Token | and |
Token ID | 290 |
Feature activation | -2.68e-02 |
Pos logit contributions | |
Paula | +1.405 |
irens | +1.350 |
Paul | +1.347 |
spe | +1.326 |
Sarah | +1.301 |
Neg logit contributions | |
per | -0.559 |
like | -0.462 |
ppy | -0.448 |
pper | -0.442 |
ilies | -0.395 |
Token | the |
Token ID | 262 |
Feature activation | -8.65e-01 |
Pos logit contributions | |
irens | +0.002 |
Paula | +0.002 |
Paul | +0.002 |
iren | +0.002 |
spe | +0.002 |
Neg logit contributions | |
per | -0.002 |
dove | -0.001 |
like | -0.001 |
worm | -0.001 |
ppy | -0.001 |
Token | shouts |
Token ID | 34757 |
Feature activation | -4.57e+00 |
Pos logit contributions | |
Paula | +0.058 |
irens | +0.058 |
Paul | +0.054 |
Sarah | +0.051 |
spe | +0.051 |
Neg logit contributions | |
per | -0.063 |
ilies | -0.060 |
worm | -0.059 |
ppy | -0.059 |
pper | -0.059 |
Token | and |
Token ID | 290 |
Feature activation | -4.81e-02 |
Pos logit contributions | |
irens | +0.370 |
Paula | +0.357 |
Paul | +0.346 |
iren | +0.335 |
John | +0.332 |
Neg logit contributions | |
opsy | -0.282 |
pper | -0.280 |
ppy | -0.276 |
per | -0.276 |
ilies | -0.274 |
Token | the |
Token ID | 262 |
Feature activation | -8.69e-01 |
Pos logit contributions | |
irens | +0.004 |
Paula | +0.004 |
Paul | +0.004 |
iren | +0.004 |
spe | +0.004 |
Neg logit contributions | |
per | -0.003 |
ilies | -0.003 |
dove | -0.003 |
ppy | -0.003 |
like | -0.003 |
Token | coming |
Token ID | 2406 |
Feature activation | -7.59e+00 |
Pos logit contributions | |
irens | +0.581 |
Paula | +0.574 |
Paul | +0.550 |
iren | +0.532 |
spe | +0.527 |
Neg logit contributions | |
per | -0.337 |
pper | -0.324 |
ppy | -0.324 |
ilies | -0.320 |
opsy | -0.313 |
Token | . |
Token ID | 13 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
irens | +0.654 |
Paula | +0.638 |
Paul | +0.602 |
iren | +0.595 |
spe | +0.576 |
Neg logit contributions | |
per | -0.439 |
ppy | -0.418 |
pper | -0.413 |
ilies | -0.408 |
ily | -0.390 |
Token | They |
Token ID | 1119 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
per | +0.001 |
ilies | +0.001 |
ppy | +0.001 |
ily | +0.001 |
pper | +0.001 |
Neg logit contributions | |
irens | -0.001 |
Paula | -0.001 |
spe | -0.001 |
Paul | -0.001 |
iren | -0.001 |
Token | had |
Token ID | 550 |
Feature activation | +3.21e-01 |
Pos logit contributions | |
irens | +0.078 |
Paula | +0.076 |
Paul | +0.072 |
iren | +0.070 |
John | +0.069 |
Neg logit contributions | |
per | -0.073 |
ppy | -0.072 |
pper | -0.071 |
ilies | -0.071 |
opsy | -0.071 |
Token | s |
Token ID | 264 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
per | +0.016 |
ilies | +0.015 |
ppy | +0.015 |
pper | +0.015 |
worm | +0.014 |
Neg logit contributions | |
irens | -0.030 |
Paula | -0.029 |
iren | -0.028 |
Paul | -0.027 |
spe | -0.027 |
Token | irens |
Token ID | 42917 |
Feature activation | -1.34e+01 |
Pos logit contributions | |
Paula | +0.023 |
spe | +0.020 |
Paul | +0.017 |
John | +0.013 |
Sarah | +0.012 |
Neg logit contributions | |
per | -0.141 |
ppy | -0.138 |
pper | -0.137 |
ilies | -0.136 |
ily | -0.132 |
Token | and |
Token ID | 290 |
Feature activation | +7.04e-02 |
Pos logit contributions | |
Paula | +1.376 |
irens | +1.321 |
Paul | +1.295 |
spe | +1.277 |
Sarah | +1.267 |
Neg logit contributions | |
per | -0.575 |
like | -0.505 |
pper | -0.479 |
ppy | -0.465 |
ilies | -0.425 |
Token | lights |
Token ID | 7588 |
Feature activation | -5.11e+00 |
Pos logit contributions | |
per | +0.005 |
ilies | +0.005 |
ppy | +0.005 |
like | +0.005 |
worm | +0.005 |
Neg logit contributions | |
Paula | -0.005 |
irens | -0.004 |
Paul | -0.004 |
iren | -0.004 |
spe | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
Paula | +0.509 |
irens | +0.502 |
Paul | +0.476 |
spe | +0.468 |
Sarah | +0.462 |
Neg logit contributions | |
per | -0.222 |
like | -0.209 |
pper | -0.195 |
ilies | -0.195 |
ppy | -0.184 |
Token | They |
Token ID | 1119 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
per | +0.008 |
ilies | +0.007 |
ppy | +0.007 |
opsy | +0.007 |
pper | +0.007 |
Neg logit contributions | |
irens | -0.008 |
Paula | -0.008 |
Paul | -0.007 |
spe | -0.007 |
iren | -0.007 |
Token | were |
Token ID | 547 |
Feature activation | -4.56e-01 |
Pos logit contributions | |
irens | +0.081 |
Paula | +0.078 |
Paul | +0.074 |
iren | +0.073 |
John | +0.070 |
Neg logit contributions | |
per | -0.066 |
ppy | -0.065 |
pper | -0.063 |
ilies | -0.063 |
opsy | -0.061 |
Token | do |
Token ID | 466 |
Feature activation | +6.10e+00 |
Pos logit contributions | |
irens | +0.067 |
Paula | +0.065 |
Paul | +0.064 |
iren | +0.062 |
John | +0.062 |
Neg logit contributions | |
per | -0.019 |
ppy | -0.019 |
ilies | -0.018 |
pper | -0.018 |
opsy | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
per | +0.345 |
ppy | +0.337 |
pper | +0.333 |
ilies | +0.331 |
opsy | +0.327 |
Neg logit contributions | |
irens | -0.529 |
Paula | -0.516 |
Paul | -0.495 |
iren | -0.483 |
spe | -0.473 |
Token | He |
Token ID | 679 |
Feature activation | -1.58e-01 |
Pos logit contributions | |
irens | +0.008 |
Paula | +0.008 |
Paul | +0.008 |
spe | +0.008 |
flashing | +0.007 |
Neg logit contributions | |
per | -0.008 |
ppy | -0.008 |
ily | -0.008 |
ilies | -0.008 |
opsy | -0.008 |
Token | hears |
Token ID | 23180 |
Feature activation | -5.59e+00 |
Pos logit contributions | |
irens | +0.010 |
Paula | +0.009 |
iren | +0.009 |
chaos | +0.008 |
Paul | +0.008 |
Neg logit contributions | |
ectar | -0.013 |
opsy | -0.013 |
comb | -0.012 |
ilies | -0.012 |
pper | -0.012 |
Token | s |
Token ID | 264 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
Paula | +0.480 |
irens | +0.479 |
spe | +0.462 |
Paul | +0.449 |
iren | +0.440 |
Neg logit contributions | |
per | -0.310 |
ily | -0.302 |
ilies | -0.297 |
ppy | -0.294 |
pper | -0.282 |
Token | irens |
Token ID | 42917 |
Feature activation | -1.34e+01 |
Pos logit contributions | |
Paula | +0.017 |
Paul | +0.013 |
spe | +0.010 |
John | +0.010 |
flashing | +0.007 |
Neg logit contributions | |
per | -0.159 |
ppy | -0.155 |
pper | -0.154 |
ilies | -0.152 |
opsy | -0.150 |
Token | . |
Token ID | 13 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
Paula | +1.256 |
irens | +1.215 |
Paul | +1.205 |
spe | +1.172 |
Sarah | +1.152 |
Neg logit contributions | |
per | -0.713 |
ppy | -0.635 |
pper | -0.614 |
like | -0.586 |
ilies | -0.560 |
Token | He |
Token ID | 679 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
Paula | +0.009 |
irens | +0.009 |
spe | +0.008 |
Paul | +0.008 |
flashing | +0.008 |
Neg logit contributions | |
per | -0.008 |
ily | -0.008 |
ilies | -0.008 |
ppy | -0.008 |
ectar | -0.008 |
Token | sees |
Token ID | 7224 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
irens | +0.010 |
Paula | +0.009 |
Paul | +0.009 |
iren | +0.009 |
John | +0.009 |
Neg logit contributions | |
per | -0.012 |
ppy | -0.012 |
pper | -0.011 |
opsy | -0.011 |
ilies | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | +9.77e-01 |
Pos logit contributions | |
Paula | +0.165 |
spe | +0.159 |
irens | +0.156 |
Paul | +0.151 |
iren | +0.145 |
Neg logit contributions | |
ily | -0.098 |
per | -0.093 |
ilies | -0.092 |
worm | -0.089 |
ppy | -0.086 |
Token | police |
Token ID | 1644 |
Feature activation | -5.45e+00 |
Pos logit contributions | |
worm | +0.058 |
dove | +0.055 |
per | +0.054 |
opsy | +0.053 |
ster | +0.050 |
Neg logit contributions | |
irens | -0.081 |
Paula | -0.081 |
Paul | -0.076 |
John | -0.075 |
James | -0.074 |
Token |
|
Token ID | 198 |
Feature activation | -2.92e-01 |
Pos logit contributions | |
per | +0.040 |
ily | +0.039 |
ilies | +0.039 |
opsy | +0.038 |
ectar | +0.038 |
Neg logit contributions | |
irens | -0.058 |
Paula | -0.057 |
flashing | -0.053 |
spe | -0.053 |
Paul | -0.053 |
Token |
|
Token ID | 198 |
Feature activation | -2.97e-01 |
Pos logit contributions | |
irens | +0.024 |
Paula | +0.024 |
spe | +0.022 |
Paul | +0.022 |
flashing | +0.022 |
Neg logit contributions | |
per | -0.017 |
opsy | -0.017 |
ily | -0.017 |
ppy | -0.017 |
ilies | -0.016 |
Token | They |
Token ID | 2990 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
irens | +0.027 |
Paula | +0.026 |
Paul | +0.025 |
iren | +0.024 |
spe | +0.024 |
Neg logit contributions | |
per | -0.016 |
ppy | -0.016 |
ilies | -0.016 |
pper | -0.016 |
opsy | -0.016 |
Token | heard |
Token ID | 2982 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
ectar | +0.012 |
opsy | +0.012 |
ilies | +0.012 |
pper | +0.011 |
per | +0.011 |
Neg logit contributions | |
Paula | -0.009 |
irens | -0.009 |
Paul | -0.008 |
John | -0.008 |
spe | -0.008 |
Token | s |
Token ID | 264 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
irens | +0.107 |
Paula | +0.103 |
iren | +0.100 |
spe | +0.100 |
Paul | +0.097 |
Neg logit contributions | |
ilies | -0.054 |
per | -0.052 |
ily | -0.052 |
ppy | -0.049 |
worm | -0.048 |
Token | irens |
Token ID | 42917 |
Feature activation | -1.34e+01 |
Pos logit contributions | |
Paula | +0.016 |
Paul | +0.013 |
spe | +0.011 |
John | +0.009 |
flashing | +0.007 |
Neg logit contributions | |
per | -0.153 |
ppy | -0.150 |
pper | -0.149 |
ilies | -0.148 |
opsy | -0.145 |
Token | and |
Token ID | 290 |
Feature activation | -5.05e-03 |
Pos logit contributions | |
Paula | +1.300 |
irens | +1.273 |
Paul | +1.244 |
spe | +1.228 |
Sarah | +1.199 |
Neg logit contributions | |
per | -0.622 |
like | -0.533 |
ppy | -0.531 |
pper | -0.520 |
ilies | -0.485 |
Token | saw |
Token ID | 2497 |
Feature activation | +8.49e-01 |
Pos logit contributions | |
irens | +0.000 |
Paula | +0.000 |
iren | +0.000 |
Paul | +0.000 |
Smith | +0.000 |
Neg logit contributions | |
per | -0.000 |
dove | -0.000 |
ppy | -0.000 |
ilies | -0.000 |
like | -0.000 |
Token | red |
Token ID | 2266 |
Feature activation | -2.54e+00 |
Pos logit contributions | |
ilies | +0.046 |
per | +0.046 |
ily | +0.045 |
worm | +0.043 |
like | +0.043 |
Neg logit contributions | |
Paula | -0.072 |
irens | -0.071 |
iren | -0.067 |
Paul | -0.067 |
spe | -0.066 |
Token | trucks |
Token ID | 13960 |
Feature activation | -6.42e+00 |
Pos logit contributions | |
irens | +0.189 |
Paula | +0.186 |
Paul | +0.177 |
iren | +0.170 |
spe | +0.169 |
Neg logit contributions | |
per | -0.173 |
ppy | -0.168 |
ilies | -0.167 |
pper | -0.166 |
opsy | -0.164 |
Token | coming |
Token ID | 2406 |
Feature activation | -7.58e+00 |
Pos logit contributions | |
Paula | +0.595 |
irens | +0.587 |
Paul | +0.563 |
Sarah | +0.540 |
spe | +0.536 |
Neg logit contributions | |
per | -0.334 |
pper | -0.304 |
ilies | -0.293 |
ppy | -0.292 |
like | -0.292 |
Token | They |
Token ID | 1119 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
per | +0.008 |
pper | +0.008 |
ppy | +0.008 |
opsy | +0.008 |
ilies | +0.008 |
Neg logit contributions | |
irens | -0.007 |
Paula | -0.006 |
Paul | -0.006 |
iren | -0.006 |
John | -0.006 |
Token | hear |
Token ID | 3285 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
irens | +0.088 |
Paula | +0.087 |
Paul | +0.084 |
John | +0.081 |
iren | +0.080 |
Neg logit contributions | |
opsy | -0.059 |
pper | -0.058 |
ectar | -0.058 |
ilies | -0.057 |
ppy | -0.057 |
Token | horns |
Token ID | 26771 |
Feature activation | -1.58e-01 |
Pos logit contributions | |
irens | +0.088 |
Paula | +0.087 |
iren | +0.083 |
spe | +0.083 |
Paul | +0.081 |
Neg logit contributions | |
per | -0.062 |
ilies | -0.060 |
ily | -0.058 |
ppy | -0.056 |
like | -0.056 |
Token | and |
Token ID | 290 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
Paula | +0.015 |
irens | +0.015 |
spe | +0.015 |
Paul | +0.014 |
John | +0.014 |
Neg logit contributions | |
per | -0.007 |
ppy | -0.007 |
pper | -0.006 |
like | -0.006 |
ilies | -0.006 |
Token | s |
Token ID | 264 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
per | +0.008 |
ppy | +0.008 |
pper | +0.008 |
ilies | +0.008 |
opsy | +0.008 |
Neg logit contributions | |
irens | -0.010 |
Paula | -0.010 |
Paul | -0.010 |
iren | -0.009 |
John | -0.009 |
Token | irens |
Token ID | 42917 |
Feature activation | -1.34e+01 |
Pos logit contributions | |
Paula | +0.016 |
spe | +0.015 |
Paul | +0.012 |
John | +0.010 |
James | +0.008 |
Neg logit contributions | |
per | -0.143 |
ppy | -0.137 |
pper | -0.136 |
ilies | -0.136 |
ectar | -0.132 |
Token | . |
Token ID | 13 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
Paula | +1.263 |
irens | +1.244 |
Paul | +1.215 |
spe | +1.189 |
Sarah | +1.176 |
Neg logit contributions | |
per | -0.667 |
ppy | -0.582 |
pper | -0.572 |
ilies | -0.558 |
like | -0.555 |
Token | They |
Token ID | 1119 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
per | +0.011 |
ppy | +0.011 |
ilies | +0.011 |
opsy | +0.011 |
pper | +0.011 |
Neg logit contributions | |
irens | -0.010 |
Paula | -0.010 |
Paul | -0.009 |
iren | -0.009 |
spe | -0.009 |
Token | are |
Token ID | 389 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
irens | +0.087 |
Paula | +0.085 |
Paul | +0.083 |
iren | +0.079 |
John | +0.079 |
Neg logit contributions | |
opsy | -0.057 |
per | -0.056 |
pper | -0.056 |
ppy | -0.056 |
ilies | -0.056 |
Token | scared |
Token ID | 12008 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
ppy | +0.129 |
per | +0.129 |
pper | +0.128 |
opsy | +0.128 |
ilies | +0.126 |
Neg logit contributions | |
irens | -0.164 |
Paula | -0.159 |
Paul | -0.153 |
iren | -0.148 |
John | -0.146 |
Token | . |
Token ID | 13 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
Paula | +0.218 |
irens | +0.211 |
spe | +0.207 |
Paul | +0.204 |
iren | +0.198 |
Neg logit contributions | |
per | -0.087 |
ily | -0.083 |
like | -0.082 |
ppy | -0.082 |
ilies | -0.077 |
Token | did |
Token ID | 750 |
Feature activation | +3.94e+00 |
Pos logit contributions | |
Paula | +0.067 |
irens | +0.067 |
Paul | +0.065 |
John | +0.061 |
spe | +0.061 |
Neg logit contributions | |
ectar | -0.068 |
per | -0.067 |
opsy | -0.067 |
ilies | -0.066 |
ppy | -0.066 |
Token | not |
Token ID | 407 |
Feature activation | +8.41e-01 |
Pos logit contributions | |
per | +0.178 |
ppy | +0.175 |
pper | +0.173 |
ilies | +0.172 |
opsy | +0.171 |
Neg logit contributions | |
irens | -0.387 |
Paula | -0.377 |
Paul | -0.365 |
iren | -0.357 |
John | -0.352 |
Token | hear |
Token ID | 3285 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
opsy | +0.049 |
ilies | +0.048 |
per | +0.048 |
pper | +0.047 |
ectar | +0.047 |
Neg logit contributions | |
irens | -0.070 |
Paula | -0.069 |
Paul | -0.066 |
iren | -0.065 |
John | -0.064 |
Token | the |
Token ID | 262 |
Feature activation | -6.91e-01 |
Pos logit contributions | |
irens | +0.086 |
Paula | +0.083 |
spe | +0.082 |
iren | +0.081 |
Paul | +0.079 |
Neg logit contributions | |
per | -0.067 |
ilies | -0.064 |
ily | -0.062 |
like | -0.062 |
ppy | -0.062 |
Token | s |
Token ID | 264 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
irens | +0.046 |
Paula | +0.044 |
Paul | +0.042 |
iren | +0.040 |
John | +0.039 |
Neg logit contributions | |
per | -0.053 |
ppy | -0.053 |
pper | -0.052 |
opsy | -0.052 |
ilies | -0.052 |
Token | irens |
Token ID | 42917 |
Feature activation | -1.33e+01 |
Pos logit contributions | |
Paula | +0.029 |
Paul | +0.029 |
John | +0.027 |
spe | +0.025 |
James | +0.024 |
Neg logit contributions | |
per | -0.127 |
pper | -0.120 |
ilies | -0.120 |
ily | -0.119 |
ppy | -0.119 |
Token | of |
Token ID | 286 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
Paula | +1.289 |
irens | +1.272 |
Paul | +1.234 |
spe | +1.211 |
Sarah | +1.181 |
Neg logit contributions | |
per | -0.655 |
ppy | -0.577 |
pper | -0.576 |
like | -0.555 |
ilies | -0.509 |
Token | the |
Token ID | 262 |
Feature activation | -6.78e-01 |
Pos logit contributions | |
irens | +0.106 |
Paula | +0.103 |
spe | +0.099 |
iren | +0.098 |
Paul | +0.096 |
Neg logit contributions | |
per | -0.090 |
ilies | -0.086 |
ppy | -0.083 |
pper | -0.082 |
opsy | -0.082 |
Token | ambulance |
Token ID | 22536 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
irens | +0.047 |
Paula | +0.046 |
Paul | +0.043 |
iren | +0.042 |
spe | +0.041 |
Neg logit contributions | |
per | -0.050 |
ppy | -0.049 |
pper | -0.049 |
opsy | -0.049 |
ilies | -0.049 |
Token | that |
Token ID | 326 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
Paula | +0.199 |
irens | +0.199 |
Paul | +0.190 |
telev | +0.183 |
John | +0.183 |
Neg logit contributions | |
per | -0.112 |
pper | -0.103 |
ppy | -0.097 |
like | -0.095 |
wait | -0.093 |
Token | came |
Token ID | 1625 |
Feature activation | -3.44e+00 |
Pos logit contributions | |
irens | +0.098 |
Paula | +0.095 |
iren | +0.087 |
Paul | +0.087 |
spe | +0.086 |
Neg logit contributions | |
per | -0.073 |
ilies | -0.073 |
ppy | -0.072 |
pper | -0.071 |
ily | -0.071 |
Token |
|
Token ID | 198 |
Feature activation | -3.47e-01 |
Pos logit contributions | |
per | +0.046 |
ppy | +0.045 |
ilies | +0.045 |
opsy | +0.044 |
pper | +0.044 |
Neg logit contributions | |
Paula | -0.048 |
irens | -0.048 |
Paul | -0.044 |
spe | -0.044 |
iren | -0.043 |
Token |
|
Token ID | 198 |
Feature activation | -3.48e-01 |
Pos logit contributions | |
irens | +0.025 |
Paula | +0.024 |
Paul | +0.023 |
iren | +0.023 |
John | +0.022 |
Neg logit contributions | |
per | -0.024 |
ppy | -0.024 |
pper | -0.024 |
ilies | -0.024 |
opsy | -0.024 |
Token | They |
Token ID | 2990 |
Feature activation | +9.31e-02 |
Pos logit contributions | |
irens | +0.025 |
Paula | +0.023 |
iren | +0.022 |
Paul | +0.022 |
John | +0.021 |
Neg logit contributions | |
ilies | -0.025 |
per | -0.025 |
ppy | -0.025 |
opsy | -0.025 |
pper | -0.025 |
Token | heard |
Token ID | 2982 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
ectar | +0.008 |
ilies | +0.007 |
ider | +0.007 |
per | +0.007 |
opsy | +0.007 |
Neg logit contributions | |
Paul | -0.005 |
Paula | -0.005 |
John | -0.005 |
Sarah | -0.005 |
irens | -0.005 |
Token | s |
Token ID | 264 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
irens | +0.086 |
Paula | +0.085 |
spe | +0.082 |
iren | +0.079 |
Paul | +0.075 |
Neg logit contributions | |
ilies | -0.073 |
ily | -0.072 |
worm | -0.070 |
per | -0.069 |
ppy | -0.068 |
Token | irens |
Token ID | 42917 |
Feature activation | -1.33e+01 |
Pos logit contributions | |
Paula | +0.035 |
Paul | +0.031 |
spe | +0.029 |
John | +0.026 |
flashing | +0.025 |
Neg logit contributions | |
per | -0.133 |
ppy | -0.130 |
pper | -0.129 |
ilies | -0.128 |
ily | -0.127 |
Token | and |
Token ID | 290 |
Feature activation | -5.29e-02 |
Pos logit contributions | |
Paula | +1.330 |
irens | +1.285 |
Paul | +1.263 |
spe | +1.250 |
Sarah | +1.197 |
Neg logit contributions | |
per | -0.585 |
like | -0.520 |
ppy | -0.518 |
pper | -0.495 |
ilies | -0.462 |
Token | saw |
Token ID | 2497 |
Feature activation | +8.71e-01 |
Pos logit contributions | |
irens | +0.004 |
Paula | +0.004 |
iren | +0.004 |
Paul | +0.004 |
parent | +0.004 |
Neg logit contributions | |
dove | -0.003 |
wait | -0.003 |
per | -0.003 |
like | -0.003 |
careful | -0.003 |
Token | lights |
Token ID | 7588 |
Feature activation | -5.17e+00 |
Pos logit contributions | |
ily | +0.045 |
ilies | +0.044 |
like | +0.044 |
per | +0.042 |
worm | +0.042 |
Neg logit contributions | |
Paula | -0.079 |
irens | -0.073 |
spe | -0.072 |
iren | -0.070 |
Paul | -0.069 |
Token | . |
Token ID | 13 |
Feature activation | -4.48e-02 |
Pos logit contributions | |
Paula | +0.522 |
irens | +0.508 |
Paul | +0.492 |
spe | +0.485 |
Sarah | +0.476 |
Neg logit contributions | |
like | -0.219 |
per | -0.214 |
pper | -0.173 |
ilies | -0.173 |
ppy | -0.169 |
Token | They |
Token ID | 1119 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
Paula | +0.003 |
irens | +0.003 |
Paul | +0.003 |
spe | +0.003 |
iren | +0.003 |
Neg logit contributions | |
ilies | -0.003 |
per | -0.003 |
ily | -0.003 |
ppy | -0.003 |
ectar | -0.003 |
Token | purse |
Token ID | 27517 |
Feature activation | -2.24e+00 |
Pos logit contributions | |
irens | +0.057 |
Paula | +0.055 |
Paul | +0.053 |
John | +0.051 |
iren | +0.050 |
Neg logit contributions | |
per | -0.050 |
ilies | -0.049 |
pper | -0.048 |
ppy | -0.048 |
opsy | -0.047 |
Token | . |
Token ID | 13 |
Feature activation | +3.49e-02 |
Pos logit contributions | |
irens | +0.195 |
Paula | +0.190 |
Paul | +0.183 |
iren | +0.177 |
John | +0.175 |
Neg logit contributions | |
per | -0.126 |
ppy | -0.123 |
pper | -0.122 |
ilies | -0.121 |
opsy | -0.120 |
Token | She |
Token ID | 1375 |
Feature activation | +8.80e-01 |
Pos logit contributions | |
per | +0.002 |
ppy | +0.002 |
pper | +0.002 |
ilies | +0.002 |
opsy | +0.002 |
Neg logit contributions | |
irens | -0.003 |
Paula | -0.003 |
Paul | -0.002 |
iren | -0.002 |
John | -0.002 |
Token | heard |
Token ID | 2982 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
per | +0.063 |
ppy | +0.062 |
ilies | +0.062 |
pper | +0.062 |
opsy | +0.062 |
Neg logit contributions | |
irens | -0.062 |
Paula | -0.060 |
Paul | -0.058 |
iren | -0.056 |
John | -0.055 |
Token | s |
Token ID | 264 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
irens | +0.095 |
spe | +0.092 |
Paula | +0.091 |
iren | +0.088 |
Paul | +0.086 |
Neg logit contributions | |
ily | -0.056 |
per | -0.055 |
ilies | -0.055 |
ppy | -0.053 |
opsy | -0.050 |
Token | irens |
Token ID | 42917 |
Feature activation | -1.33e+01 |
Pos logit contributions | |
Paula | +0.021 |
Paul | +0.020 |
spe | +0.019 |
John | +0.016 |
flashing | +0.013 |
Neg logit contributions | |
per | -0.135 |
ppy | -0.131 |
pper | -0.130 |
ily | -0.129 |
ilies | -0.129 |
Token | outside |
Token ID | 2354 |
Feature activation | -5.23e-01 |
Pos logit contributions | |
Paula | +1.294 |
irens | +1.253 |
Paul | +1.252 |
spe | +1.243 |
Sarah | +1.198 |
Neg logit contributions | |
per | -0.631 |
ppy | -0.536 |
like | -0.531 |
pper | -0.518 |
ilies | -0.463 |
Token | . |
Token ID | 13 |
Feature activation | +7.66e-02 |
Pos logit contributions | |
Paula | +0.053 |
irens | +0.053 |
Paul | +0.051 |
spe | +0.050 |
iren | +0.049 |
Neg logit contributions | |
per | -0.019 |
ily | -0.018 |
like | -0.018 |
ppy | -0.016 |
ectar | -0.015 |
Token | She |
Token ID | 1375 |
Feature activation | +9.01e-01 |
Pos logit contributions | |
per | +0.005 |
ppy | +0.005 |
ilies | +0.005 |
opsy | +0.005 |
ectar | +0.004 |
Neg logit contributions | |
irens | -0.006 |
Paula | -0.006 |
Paul | -0.006 |
spe | -0.006 |
iren | -0.006 |
Token | ran |
Token ID | 4966 |
Feature activation | -2.76e+00 |
Pos logit contributions | |
per | +0.063 |
ppy | +0.062 |
pper | +0.061 |
opsy | +0.060 |
ilies | +0.060 |
Neg logit contributions | |
irens | -0.067 |
Paula | -0.065 |
Paul | -0.062 |
iren | -0.060 |
John | -0.059 |
Token | to |
Token ID | 284 |
Feature activation | -4.43e-01 |
Pos logit contributions | |
irens | +0.222 |
Paula | +0.218 |
Paul | +0.209 |
spe | +0.207 |
iren | +0.199 |
Neg logit contributions | |
per | -0.170 |
ppy | -0.160 |
ily | -0.157 |
ilies | -0.156 |
like | -0.155 |
Token | tower |
Token ID | 10580 |
Feature activation | -2.74e-01 |
Pos logit contributions | |
irens | +0.045 |
Paula | +0.043 |
Paul | +0.041 |
iren | +0.040 |
John | +0.039 |
Neg logit contributions | |
per | -0.046 |
ppy | -0.045 |
ilies | -0.045 |
pper | -0.045 |
opsy | -0.044 |
Token | . |
Token ID | 13 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
irens | +0.024 |
Paula | +0.024 |
Paul | +0.023 |
iren | +0.022 |
John | +0.022 |
Neg logit contributions | |
per | -0.015 |
ppy | -0.015 |
pper | -0.015 |
ilies | -0.015 |
opsy | -0.014 |
Token | They |
Token ID | 1119 |
Feature activation | -9.76e-01 |
Pos logit contributions | |
per | +0.012 |
ilies | +0.012 |
ppy | +0.012 |
opsy | +0.012 |
pper | +0.012 |
Neg logit contributions | |
irens | -0.012 |
Paula | -0.011 |
Paul | -0.011 |
iren | -0.011 |
spe | -0.010 |
Token | heard |
Token ID | 2982 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
irens | +0.078 |
Paula | +0.075 |
Paul | +0.071 |
iren | +0.070 |
spe | +0.068 |
Neg logit contributions | |
per | -0.063 |
ppy | -0.062 |
pper | -0.061 |
ilies | -0.061 |
opsy | -0.060 |
Token | s |
Token ID | 264 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
irens | +0.084 |
Paula | +0.082 |
spe | +0.080 |
iren | +0.079 |
Paul | +0.077 |
Neg logit contributions | |
ilies | -0.058 |
ily | -0.057 |
per | -0.057 |
like | -0.053 |
ppy | -0.053 |
Token | irens |
Token ID | 42917 |
Feature activation | -1.33e+01 |
Pos logit contributions | |
Paula | +0.015 |
Paul | +0.013 |
spe | +0.012 |
John | +0.010 |
Sarah | +0.007 |
Neg logit contributions | |
per | -0.138 |
ppy | -0.134 |
pper | -0.133 |
ilies | -0.132 |
ily | -0.130 |
Token | and |
Token ID | 290 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
Paula | +1.357 |
Paul | +1.312 |
spe | +1.301 |
irens | +1.298 |
Sarah | +1.277 |
Neg logit contributions | |
per | -0.580 |
like | -0.475 |
pper | -0.470 |
ppy | -0.466 |
ilies | -0.406 |
Token | saw |
Token ID | 2497 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
per | +0.009 |
dove | +0.009 |
like | +0.009 |
wait | +0.009 |
ppy | +0.009 |
Neg logit contributions | |
irens | -0.010 |
Paula | -0.010 |
Paul | -0.010 |
iren | -0.010 |
Smith | -0.009 |
Token | fire |
Token ID | 2046 |
Feature activation | -8.39e-02 |
Pos logit contributions | |
ilies | +0.059 |
per | +0.059 |
ily | +0.058 |
like | +0.055 |
ectar | +0.053 |
Neg logit contributions | |
Paula | -0.084 |
irens | -0.080 |
spe | -0.078 |
Paul | -0.077 |
iren | -0.076 |
Token | trucks |
Token ID | 13960 |
Feature activation | -6.30e+00 |
Pos logit contributions | |
Paula | +0.007 |
irens | +0.007 |
Sarah | +0.007 |
Sarah | +0.007 |
Paul | +0.007 |
Neg logit contributions | |
per | -0.005 |
pper | -0.005 |
ppy | -0.004 |
like | -0.004 |
comb | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
Paula | +0.610 |
irens | +0.589 |
Paul | +0.578 |
Sarah | +0.557 |
spe | +0.552 |
Neg logit contributions | |
per | -0.313 |
pper | -0.282 |
ppy | -0.269 |
like | -0.266 |
ilies | -0.262 |
Token | did |
Token ID | 750 |
Feature activation | +3.99e+00 |
Pos logit contributions | |
irens | +0.062 |
Paula | +0.061 |
Paul | +0.060 |
John | +0.058 |
flashing | +0.057 |
Neg logit contributions | |
ectar | -0.078 |
opsy | -0.075 |
ilies | -0.074 |
per | -0.073 |
pper | -0.073 |
Token | not |
Token ID | 407 |
Feature activation | +9.00e-01 |
Pos logit contributions | |
per | +0.175 |
ppy | +0.171 |
pper | +0.169 |
ilies | +0.168 |
opsy | +0.168 |
Neg logit contributions | |
irens | -0.396 |
Paula | -0.387 |
Paul | -0.374 |
iren | -0.366 |
John | -0.360 |
Token | hear |
Token ID | 3285 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
opsy | +0.059 |
ilies | +0.056 |
pper | +0.056 |
per | +0.054 |
ectar | +0.054 |
Neg logit contributions | |
irens | -0.069 |
Paula | -0.068 |
Paul | -0.065 |
iren | -0.064 |
John | -0.064 |
Token | the |
Token ID | 262 |
Feature activation | -7.53e-01 |
Pos logit contributions | |
irens | +0.086 |
Paula | +0.083 |
iren | +0.082 |
spe | +0.080 |
Paul | +0.077 |
Neg logit contributions | |
per | -0.059 |
ilies | -0.058 |
ily | -0.056 |
like | -0.056 |
thirst | -0.055 |
Token | s |
Token ID | 264 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
irens | +0.046 |
Paula | +0.042 |
iren | +0.041 |
Paul | +0.041 |
flashing | +0.041 |
Neg logit contributions | |
ppy | -0.063 |
per | -0.063 |
pper | -0.062 |
ilies | -0.062 |
opsy | -0.061 |
Token | irens |
Token ID | 42917 |
Feature activation | -1.33e+01 |
Pos logit contributions | |
Paula | +0.025 |
Paul | +0.020 |
spe | +0.017 |
John | +0.017 |
James | +0.015 |
Neg logit contributions | |
per | -0.133 |
pper | -0.129 |
ppy | -0.128 |
ilies | -0.126 |
ily | -0.124 |
Token | and |
Token ID | 290 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
Paula | +1.224 |
irens | +1.188 |
Paul | +1.152 |
spe | +1.127 |
telev | +1.109 |
Neg logit contributions | |
per | -0.734 |
pper | -0.655 |
ppy | -0.648 |
like | -0.632 |
ilies | -0.572 |
Token | the |
Token ID | 262 |
Feature activation | -7.28e-01 |
Pos logit contributions | |
per | +0.008 |
ppy | +0.008 |
ilies | +0.008 |
worm | +0.008 |
dove | +0.008 |
Neg logit contributions | |
irens | -0.010 |
Paula | -0.010 |
iren | -0.010 |
Paul | -0.010 |
spe | -0.009 |
Token | shouts |
Token ID | 34757 |
Feature activation | -4.45e+00 |
Pos logit contributions | |
irens | +0.053 |
Paula | +0.050 |
Paul | +0.048 |
iren | +0.048 |
flashing | +0.047 |
Neg logit contributions | |
ppy | -0.052 |
per | -0.052 |
pper | -0.052 |
opsy | -0.051 |
ilies | -0.051 |
Token | . |
Token ID | 13 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
irens | +0.355 |
Paula | +0.340 |
Paul | +0.332 |
John | +0.320 |
iren | +0.319 |
Neg logit contributions | |
opsy | -0.280 |
pper | -0.274 |
ppy | -0.271 |
per | -0.268 |
ilies | -0.267 |
Token |
|
Token ID | 198 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
per | +0.009 |
ilies | +0.009 |
ily | +0.008 |
opsy | +0.008 |
ectar | +0.008 |
Neg logit contributions | |
irens | -0.012 |
Paula | -0.012 |
iren | -0.011 |
Paul | -0.011 |
spe | -0.011 |
Token | Tom |
Token ID | 4186 |
Feature activation | -9.88e-01 |
Pos logit contributions | |
Paula | +0.023 |
irens | +0.023 |
Paul | +0.022 |
John | +0.021 |
Sarah | +0.021 |
Neg logit contributions | |
per | -0.027 |
ppy | -0.027 |
ilies | -0.027 |
pper | -0.027 |
opsy | -0.026 |
Token | are |
Token ID | 389 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
irens | +0.072 |
Paula | +0.072 |
Paul | +0.069 |
John | +0.067 |
Sarah | +0.065 |
Neg logit contributions | |
per | -0.066 |
ilies | -0.066 |
opsy | -0.065 |
ectar | -0.065 |
pper | -0.065 |
Token | friends |
Token ID | 2460 |
Feature activation | +2.60e+00 |
Pos logit contributions | |
pper | +0.134 |
comb | +0.129 |
per | +0.127 |
ppy | +0.125 |
ilies | +0.119 |
Neg logit contributions | |
irens | -0.132 |
Paula | -0.129 |
John | -0.126 |
Paul | -0.125 |
Sarah | -0.124 |
Token | . |
Token ID | 13 |
Feature activation | -3.04e-01 |
Pos logit contributions | |
per | +0.154 |
ppy | +0.153 |
ilies | +0.152 |
pper | +0.151 |
opsy | +0.151 |
Neg logit contributions | |
irens | -0.216 |
Paula | -0.210 |
Paul | -0.203 |
iren | -0.196 |
John | -0.195 |
Token | They |
Token ID | 1119 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
irens | +0.020 |
Paula | +0.019 |
Paul | +0.018 |
iren | +0.017 |
John | +0.017 |
Neg logit contributions | |
per | -0.024 |
pper | -0.024 |
ppy | -0.024 |
opsy | -0.023 |
ilies | -0.023 |
Token | like |
Token ID | 588 |
Feature activation | +6.05e+00 |
Pos logit contributions | |
Paula | +0.113 |
John | +0.111 |
Paul | +0.110 |
Sarah | +0.110 |
James | +0.104 |
Neg logit contributions | |
comb | -0.063 |
ectar | -0.063 |
opsy | -0.062 |
pper | -0.061 |
per | -0.059 |
Token | to |
Token ID | 284 |
Feature activation | -5.38e-01 |
Pos logit contributions | |
ilies | +0.372 |
per | +0.372 |
ppy | +0.367 |
ectar | +0.365 |
opsy | +0.365 |
Neg logit contributions | |
irens | -0.481 |
Paula | -0.461 |
iren | -0.438 |
Paul | -0.435 |
spe | -0.423 |
Token | play |
Token ID | 711 |
Feature activation | +2.70e-01 |
Pos logit contributions | |
irens | +0.036 |
Paula | +0.036 |
Paul | +0.035 |
Sarah | +0.034 |
John | +0.034 |
Neg logit contributions | |
per | -0.040 |
pper | -0.039 |
ppy | -0.039 |
ilies | -0.039 |
opsy | -0.038 |
Token | on |
Token ID | 319 |
Feature activation | -4.06e-01 |
Pos logit contributions | |
pper | +0.024 |
ppy | +0.022 |
opsy | +0.022 |
comb | +0.022 |
ilies | +0.022 |
Neg logit contributions | |
irens | -0.015 |
Sarah | -0.014 |
telev | -0.014 |
Paul | -0.014 |
James | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | -8.18e-01 |
Pos logit contributions | |
irens | +0.033 |
Paula | +0.032 |
iren | +0.032 |
spe | +0.030 |
Paul | +0.030 |
Neg logit contributions | |
per | -0.024 |
opsy | -0.024 |
ilies | -0.023 |
worm | -0.023 |
dove | -0.023 |
Token | farm |
Token ID | 5318 |
Feature activation | +3.89e+00 |
Pos logit contributions | |
irens | +0.058 |
Paula | +0.055 |
Paul | +0.054 |
flashing | +0.052 |
iren | +0.051 |
Neg logit contributions | |
ppy | -0.060 |
per | -0.058 |
pper | -0.058 |
ilies | -0.057 |
ectar | -0.057 |
Token | was |
Token ID | 373 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
irens | +0.010 |
Paula | +0.010 |
iren | +0.010 |
Paul | +0.009 |
flashing | +0.009 |
Neg logit contributions | |
per | -0.008 |
ppy | -0.008 |
ilies | -0.007 |
pper | -0.007 |
ily | -0.007 |
Token | able |
Token ID | 1498 |
Feature activation | -5.28e+00 |
Pos logit contributions | |
irens | +0.120 |
Paula | +0.117 |
Paul | +0.113 |
iren | +0.109 |
John | +0.108 |
Neg logit contributions | |
per | -0.074 |
ppy | -0.073 |
pper | -0.072 |
ilies | -0.072 |
opsy | -0.072 |
Token | to |
Token ID | 284 |
Feature activation | -5.60e-01 |
Pos logit contributions | |
irens | +0.433 |
Paula | +0.411 |
iren | +0.398 |
Paul | +0.384 |
telev | +0.374 |
Neg logit contributions | |
per | -0.317 |
ilies | -0.312 |
opsy | -0.309 |
ectar | -0.307 |
ppy | -0.307 |
Token | come |
Token ID | 1282 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
irens | +0.047 |
Paula | +0.047 |
Paul | +0.046 |
John | +0.044 |
iren | +0.043 |
Neg logit contributions | |
ppy | -0.032 |
ilies | -0.032 |
opsy | -0.032 |
pper | -0.031 |
per | -0.031 |
Token | home |
Token ID | 1363 |
Feature activation | +3.66e+00 |
Pos logit contributions | |
irens | +0.098 |
Paula | +0.096 |
Paul | +0.092 |
iren | +0.089 |
John | +0.088 |
Neg logit contributions | |
per | -0.067 |
ppy | -0.067 |
pper | -0.066 |
opsy | -0.066 |
ilies | -0.066 |
Token | for |
Token ID | 329 |
Feature activation | +6.26e+00 |
Pos logit contributions | |
per | +0.205 |
ppy | +0.191 |
pper | +0.188 |
ilies | +0.188 |
opsy | +0.186 |
Neg logit contributions | |
irens | -0.324 |
Paula | -0.308 |
iren | -0.298 |
Paul | -0.295 |
flashing | -0.286 |
Token | good |
Token ID | 922 |
Feature activation | -9.64e-03 |
Pos logit contributions | |
per | +0.390 |
ilies | +0.374 |
ppy | +0.368 |
pper | +0.368 |
opsy | +0.366 |
Neg logit contributions | |
irens | -0.499 |
Paula | -0.476 |
spe | -0.461 |
iren | -0.460 |
Paul | -0.447 |
Token | . |
Token ID | 13 |
Feature activation | -1.58e-01 |
Pos logit contributions | |
Paula | +0.001 |
irens | +0.001 |
iren | +0.001 |
spe | +0.001 |
flashing | +0.001 |
Neg logit contributions | |
per | -0.001 |
like | -0.001 |
wait | -0.001 |
ilies | -0.001 |
ectar | -0.001 |
Token | |
Token ID | 220 |
Feature activation | -9.04e-01 |
Pos logit contributions | |
irens | +0.012 |
Paula | +0.011 |
flashing | +0.011 |
iren | +0.011 |
spe | +0.010 |
Neg logit contributions | |
per | -0.010 |
opsy | -0.010 |
ppy | -0.010 |
ilies | -0.010 |
pper | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | -4.15e-01 |
Pos logit contributions | |
irens | +0.069 |
Paula | +0.066 |
Paul | +0.063 |
iren | +0.062 |
spe | +0.062 |
Neg logit contributions | |
per | -0.060 |
ppy | -0.058 |
ilies | -0.058 |
opsy | -0.058 |
pper | -0.058 |
Token |
|
Token ID | 198 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
irens | +0.029 |
Paula | +0.029 |
Paul | +0.027 |
iren | +0.026 |
John | +0.026 |
Neg logit contributions | |
per | -0.030 |
ppy | -0.030 |
pper | -0.029 |
ilies | -0.029 |
opsy | -0.029 |
Token | a |
Token ID | 257 |
Feature activation | +9.49e-01 |
Pos logit contributions | |
per | +0.001 |
ppy | +0.001 |
ilies | +0.001 |
opsy | +0.001 |
ily | +0.001 |
Neg logit contributions | |
irens | -0.001 |
Paula | -0.001 |
iren | -0.001 |
Paul | -0.001 |
spe | -0.001 |
Token | drum |
Token ID | 13026 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
opsy | +0.059 |
per | +0.056 |
ppy | +0.056 |
ilies | +0.056 |
worm | +0.055 |
Neg logit contributions | |
irens | -0.079 |
Paula | -0.074 |
Paul | -0.071 |
iren | -0.070 |
John | -0.068 |
Token | , |
Token ID | 11 |
Feature activation | +3.61e-01 |
Pos logit contributions | |
irens | +0.138 |
Paula | +0.134 |
Paul | +0.129 |
iren | +0.126 |
John | +0.124 |
Neg logit contributions | |
per | -0.086 |
ppy | -0.085 |
pper | -0.084 |
ilies | -0.084 |
opsy | -0.083 |
Token | a |
Token ID | 257 |
Feature activation | +9.42e-01 |
Pos logit contributions | |
per | +0.020 |
opsy | +0.019 |
ppy | +0.019 |
pper | +0.019 |
ilies | +0.019 |
Neg logit contributions | |
irens | -0.032 |
Paula | -0.031 |
Paul | -0.030 |
iren | -0.029 |
John | -0.029 |
Token | fl |
Token ID | 781 |
Feature activation | +5.15e+00 |
Pos logit contributions | |
per | +0.067 |
ppy | +0.066 |
pper | +0.066 |
ilies | +0.066 |
opsy | +0.065 |
Neg logit contributions | |
irens | -0.067 |
Paula | -0.065 |
Paul | -0.062 |
iren | -0.060 |
John | -0.059 |
Token | ute |
Token ID | 1133 |
Feature activation | +5.63e+00 |
Pos logit contributions | |
ppy | +0.169 |
ilies | +0.168 |
pper | +0.162 |
per | +0.160 |
comb | +0.155 |
Neg logit contributions | |
irens | -0.543 |
Paula | -0.529 |
iren | -0.529 |
Paul | -0.522 |
spe | -0.514 |
Token | , |
Token ID | 11 |
Feature activation | +3.46e-01 |
Pos logit contributions | |
per | +0.329 |
opsy | +0.329 |
ppy | +0.329 |
pper | +0.325 |
ilies | +0.323 |
Neg logit contributions | |
irens | -0.468 |
Paula | -0.455 |
Paul | -0.440 |
iren | -0.428 |
flashing | -0.420 |
Token | a |
Token ID | 257 |
Feature activation | +9.35e-01 |
Pos logit contributions | |
opsy | +0.021 |
per | +0.021 |
ppy | +0.021 |
pper | +0.021 |
ilies | +0.020 |
Neg logit contributions | |
irens | -0.028 |
Paula | -0.028 |
Paul | -0.027 |
flashing | -0.026 |
John | -0.026 |
Token | guitar |
Token ID | 10047 |
Feature activation | -4.44e-02 |
Pos logit contributions | |
per | +0.068 |
pper | +0.068 |
ppy | +0.068 |
ilies | +0.067 |
opsy | +0.066 |
Neg logit contributions | |
irens | -0.064 |
Paula | -0.064 |
Paul | -0.061 |
flashing | -0.059 |
iren | -0.059 |
Token | and |
Token ID | 290 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
irens | +0.003 |
Paula | +0.003 |
Paul | +0.003 |
iren | +0.003 |
flashing | +0.003 |
Neg logit contributions | |
opsy | -0.003 |
ppy | -0.003 |
ilies | -0.003 |
pper | -0.003 |
per | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | +9.30e-01 |
Pos logit contributions | |
irens | +0.021 |
Paula | +0.021 |
Paul | +0.020 |
iren | +0.019 |
John | +0.019 |
Neg logit contributions | |
per | -0.012 |
ppy | -0.012 |
pper | -0.012 |
opsy | -0.012 |
ilies | -0.012 |
Token | happy |
Token ID | 3772 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
per | +0.065 |
ppy | +0.064 |
pper | +0.063 |
ilies | +0.063 |
opsy | +0.063 |
Neg logit contributions | |
irens | -0.102 |
Paula | -0.100 |
Paul | -0.096 |
iren | -0.094 |
John | -0.092 |
Token | family |
Token ID | 1641 |
Feature activation | +3.42e-01 |
Pos logit contributions | |
opsy | +0.116 |
per | +0.115 |
ectar | +0.111 |
ilies | +0.111 |
ppy | +0.111 |
Neg logit contributions | |
irens | -0.176 |
Paula | -0.167 |
iren | -0.160 |
Paul | -0.158 |
spe | -0.152 |
Token | . |
Token ID | 13 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
per | +0.021 |
ppy | +0.021 |
opsy | +0.020 |
pper | +0.020 |
ilies | +0.020 |
Neg logit contributions | |
irens | -0.028 |
Paula | -0.027 |
Paul | -0.026 |
John | -0.025 |
iren | -0.025 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +4.11e+00 |
Pos logit contributions | |
per | +0.009 |
ilies | +0.009 |
opsy | +0.009 |
ppy | +0.009 |
ily | +0.009 |
Neg logit contributions | |
irens | -0.012 |
Paula | -0.011 |
iren | -0.011 |
spe | -0.011 |
Paul | -0.011 |
Token | L |
Token ID | 43 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
per | +0.282 |
ilies | +0.275 |
ppy | +0.274 |
opsy | +0.271 |
ectar | +0.267 |
Neg logit contributions | |
irens | -0.299 |
Paula | -0.297 |
Paul | -0.282 |
spe | -0.276 |
flashing | -0.268 |
Token | ily |
Token ID | 813 |
Feature activation | +7.56e+00 |
Pos logit contributions | |
per | +0.089 |
pper | +0.088 |
ppy | +0.088 |
opsy | +0.087 |
ilies | +0.086 |
Neg logit contributions | |
irens | -0.129 |
Paula | -0.126 |
Paul | -0.120 |
iren | -0.117 |
John | -0.116 |
Token | and |
Token ID | 290 |
Feature activation | -6.51e-02 |
Pos logit contributions | |
ilies | +0.490 |
pper | +0.487 |
per | +0.485 |
opsy | +0.483 |
ppy | +0.483 |
Neg logit contributions | |
irens | -0.573 |
Paula | -0.557 |
Paul | -0.539 |
John | -0.527 |
iren | -0.520 |
Token | Ben |
Token ID | 3932 |
Feature activation | +4.84e+00 |
Pos logit contributions | |
irens | +0.004 |
Paula | +0.004 |
Paul | +0.004 |
John | +0.004 |
Sarah | +0.004 |
Neg logit contributions | |
per | -0.005 |
ppy | -0.005 |
pper | -0.005 |
ilies | -0.005 |
opsy | -0.005 |
Token | were |
Token ID | 547 |
Feature activation | -5.58e-01 |
Pos logit contributions | |
ilies | +0.298 |
ectar | +0.298 |
opsy | +0.297 |
per | +0.291 |
pper | +0.291 |
Neg logit contributions | |
Paula | -0.369 |
irens | -0.369 |
Paul | -0.363 |
John | -0.361 |
Sarah | -0.347 |
Token | best |
Token ID | 1266 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
irens | +0.042 |
Paula | +0.041 |
John | +0.040 |
Sarah | +0.040 |
Paul | +0.040 |
Neg logit contributions | |
pper | -0.040 |
comb | -0.038 |
per | -0.037 |
ppy | -0.037 |
ilies | -0.034 |
Token | friends |
Token ID | 2460 |
Feature activation | +2.78e+00 |
Pos logit contributions | |
pper | +0.084 |
per | +0.084 |
comb | +0.081 |
ppy | +0.081 |
opsy | +0.072 |
Neg logit contributions | |
irens | -0.117 |
Paul | -0.108 |
Paula | -0.108 |
John | -0.106 |
Sarah | -0.104 |
Token | ma |
Token ID | 2611 |
Feature activation | -6.70e+00 |
Pos logit contributions | |
irens | +0.134 |
flashing | +0.130 |
Paula | +0.124 |
spe | +0.122 |
chaos | +0.120 |
Neg logit contributions | |
pper | -0.196 |
ppy | -0.196 |
ily | -0.195 |
per | -0.185 |
opsy | -0.184 |
Token | . |
Token ID | 13 |
Feature activation | +5.97e-02 |
Pos logit contributions | |
irens | +0.569 |
Paula | +0.552 |
Paul | +0.531 |
iren | +0.518 |
John | +0.509 |
Neg logit contributions | |
per | -0.387 |
ppy | -0.382 |
pper | -0.378 |
ilies | -0.376 |
opsy | -0.376 |
Token |
|
Token ID | 198 |
Feature activation | -2.77e-01 |
Pos logit contributions | |
per | +0.003 |
ppy | +0.003 |
ily | +0.003 |
ilies | +0.003 |
opsy | +0.003 |
Neg logit contributions | |
irens | -0.005 |
Paula | -0.005 |
spe | -0.005 |
iren | -0.005 |
flashing | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | -2.91e-01 |
Pos logit contributions | |
irens | +0.022 |
Paula | +0.021 |
Paul | +0.020 |
spe | +0.020 |
flashing | +0.020 |
Neg logit contributions | |
per | -0.017 |
ppy | -0.017 |
opsy | -0.017 |
ilies | -0.017 |
pper | -0.017 |
Token | L |
Token ID | 43 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
irens | +0.024 |
Paula | +0.023 |
Paul | +0.022 |
spe | +0.021 |
iren | +0.021 |
Neg logit contributions | |
ilies | -0.018 |
opsy | -0.018 |
per | -0.018 |
pper | -0.018 |
ppy | -0.018 |
Token | ily |
Token ID | 813 |
Feature activation | +7.55e+00 |
Pos logit contributions | |
opsy | +0.023 |
per | +0.022 |
pper | +0.022 |
ilies | +0.018 |
ppy | +0.018 |
Neg logit contributions | |
irens | -0.205 |
Paula | -0.200 |
iren | -0.195 |
Paul | -0.194 |
Sarah | -0.194 |
Token | was |
Token ID | 373 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
ilies | +0.594 |
ectar | +0.581 |
per | +0.571 |
opsy | +0.564 |
pper | +0.563 |
Neg logit contributions | |
Paula | -0.462 |
irens | -0.460 |
Paul | -0.444 |
John | -0.434 |
Sarah | -0.426 |
Token | happy |
Token ID | 3772 |
Feature activation | +1.85e+00 |
Pos logit contributions | |
irens | +0.112 |
Paula | +0.109 |
Paul | +0.105 |
John | +0.101 |
iren | +0.101 |
Neg logit contributions | |
per | -0.079 |
ppy | -0.079 |
pper | -0.078 |
opsy | -0.078 |
ilies | -0.078 |
Token | and |
Token ID | 290 |
Feature activation | -6.16e-03 |
Pos logit contributions | |
per | +0.098 |
ppy | +0.095 |
ectar | +0.093 |
opsy | +0.093 |
ilies | +0.093 |
Neg logit contributions | |
irens | -0.168 |
Paula | -0.161 |
iren | -0.155 |
Paul | -0.153 |
spe | -0.152 |
Token | said |
Token ID | 531 |
Feature activation | +2.28e+00 |
Pos logit contributions | |
irens | +0.000 |
Paula | +0.000 |
Paul | +0.000 |
John | +0.000 |
Sarah | +0.000 |
Neg logit contributions | |
per | -0.000 |
ppy | -0.000 |
ilies | -0.000 |
pper | -0.000 |
opsy | -0.000 |
Token | , |
Token ID | 11 |
Feature activation | +5.13e-01 |
Pos logit contributions | |
pper | +0.153 |
per | +0.152 |
opsy | +0.151 |
ppy | +0.149 |
ilies | +0.145 |
Neg logit contributions | |
Paula | -0.173 |
Paul | -0.168 |
irens | -0.168 |
John | -0.167 |
Sarah | -0.163 |
Token | books |
Token ID | 3835 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
irens | +0.024 |
Paula | +0.023 |
iren | +0.022 |
Paul | +0.022 |
spe | +0.021 |
Neg logit contributions | |
per | -0.020 |
ilies | -0.020 |
ppy | -0.019 |
like | -0.019 |
worm | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -3.55e-01 |
Pos logit contributions | |
irens | +0.166 |
Paula | +0.162 |
Paul | +0.155 |
iren | +0.151 |
John | +0.150 |
Neg logit contributions | |
opsy | -0.125 |
ppy | -0.124 |
per | -0.123 |
pper | -0.122 |
ilies | -0.121 |
Token | They |
Token ID | 1119 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
irens | +0.025 |
Paula | +0.024 |
Paul | +0.023 |
iren | +0.023 |
spe | +0.022 |
Neg logit contributions | |
per | -0.025 |
ilies | -0.025 |
ppy | -0.025 |
pper | -0.025 |
opsy | -0.025 |
Token | are |
Token ID | 389 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
Paula | +0.114 |
Paul | +0.111 |
irens | +0.110 |
John | +0.107 |
spe | +0.106 |
Neg logit contributions | |
opsy | -0.086 |
ectar | -0.083 |
pper | -0.081 |
ilies | -0.080 |
per | -0.079 |
Token | happy |
Token ID | 3772 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
per | +0.096 |
ppy | +0.094 |
ilies | +0.094 |
pper | +0.093 |
opsy | +0.092 |
Neg logit contributions | |
irens | -0.165 |
Paula | -0.159 |
Paul | -0.153 |
iren | -0.151 |
spe | -0.147 |
Token | . |
Token ID | 13 |
Feature activation | -3.70e-01 |
Pos logit contributions | |
per | +0.077 |
ilies | +0.076 |
ectar | +0.076 |
like | +0.074 |
ily | +0.071 |
Neg logit contributions | |
irens | -0.175 |
Paula | -0.171 |
iren | -0.166 |
Paul | -0.160 |
spe | -0.156 |
Token |
|
Token ID | 198 |
Feature activation | -6.27e-01 |
Pos logit contributions | |
irens | +0.028 |
Paula | +0.027 |
Paul | +0.026 |
iren | +0.026 |
spe | +0.025 |
Neg logit contributions | |
per | -0.024 |
ilies | -0.024 |
opsy | -0.024 |
ppy | -0.024 |
pper | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | -6.41e-01 |
Pos logit contributions | |
irens | +0.049 |
Paula | +0.048 |
Paul | +0.045 |
spe | +0.045 |
iren | +0.044 |
Neg logit contributions | |
per | -0.039 |
ilies | -0.039 |
opsy | -0.039 |
ppy | -0.038 |
pper | -0.038 |
Token | " |
Token ID | 1 |
Feature activation | +3.14e+00 |
Pos logit contributions | |
irens | +0.060 |
Paula | +0.056 |
iren | +0.054 |
Paul | +0.054 |
spe | +0.054 |
Neg logit contributions | |
ppy | -0.035 |
opsy | -0.034 |
ilies | -0.034 |
ectar | -0.033 |
per | -0.033 |
Token | Wow |
Token ID | 22017 |
Feature activation | -3.06e+00 |
Pos logit contributions | |
per | +0.235 |
ppy | +0.232 |
pper | +0.231 |
ilies | +0.230 |
opsy | +0.230 |
Neg logit contributions | |
irens | -0.215 |
Paula | -0.208 |
Paul | -0.198 |
iren | -0.191 |
John | -0.187 |
Token | , |
Token ID | 11 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
irens | +0.236 |
Paula | +0.233 |
Paul | +0.229 |
John | +0.223 |
Sarah | +0.218 |
Neg logit contributions | |
per | -0.195 |
opsy | -0.191 |
ilies | -0.191 |
pper | -0.191 |
ppy | -0.190 |
Token | they |
Token ID | 484 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
per | +0.017 |
ilies | +0.017 |
pper | +0.017 |
opsy | +0.017 |
ppy | +0.017 |
Neg logit contributions | |
irens | -0.024 |
Paula | -0.023 |
iren | -0.023 |
spe | -0.021 |
Paul | -0.021 |
Token | would |
Token ID | 561 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
irens | +0.204 |
Paula | +0.193 |
iren | +0.186 |
Paul | +0.183 |
spe | +0.177 |
Neg logit contributions | |
ppy | -0.126 |
per | -0.125 |
pper | -0.122 |
ilies | -0.121 |
opsy | -0.120 |
Token | need |
Token ID | 761 |
Feature activation | -4.13e+00 |
Pos logit contributions | |
opsy | +0.079 |
ppy | +0.079 |
ilies | +0.078 |
pper | +0.077 |
per | +0.075 |
Neg logit contributions | |
Paula | -0.084 |
Paul | -0.084 |
irens | -0.080 |
Sarah | -0.079 |
John | -0.079 |
Token | it |
Token ID | 340 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
irens | +0.330 |
Paula | +0.321 |
spe | +0.301 |
iren | +0.299 |
Paul | +0.298 |
Neg logit contributions | |
per | -0.260 |
ilies | -0.255 |
ppy | -0.250 |
pper | -0.245 |
ectar | -0.245 |
Token | later |
Token ID | 1568 |
Feature activation | +4.23e+00 |
Pos logit contributions | |
per | +0.017 |
opsy | +0.017 |
ppy | +0.016 |
pper | +0.016 |
ilies | +0.016 |
Neg logit contributions | |
irens | -0.018 |
Paula | -0.017 |
Paul | -0.017 |
iren | -0.016 |
John | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -2.67e-01 |
Pos logit contributions | |
per | +0.232 |
ppy | +0.224 |
pper | +0.220 |
ilies | +0.219 |
opsy | +0.215 |
Neg logit contributions | |
irens | -0.376 |
Paula | -0.363 |
iren | -0.346 |
Paul | -0.345 |
spe | -0.335 |
Token | They |
Token ID | 1119 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
irens | +0.022 |
Paula | +0.022 |
iren | +0.020 |
Paul | +0.020 |
spe | +0.019 |
Neg logit contributions | |
per | -0.015 |
ilies | -0.015 |
ily | -0.015 |
ppy | -0.015 |
opsy | -0.014 |
Token | took |
Token ID | 1718 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
Paula | +0.088 |
Paul | +0.086 |
irens | +0.086 |
John | +0.082 |
Sarah | +0.081 |
Neg logit contributions | |
opsy | -0.097 |
per | -0.096 |
pper | -0.096 |
ilies | -0.096 |
ectar | -0.096 |
Token | the |
Token ID | 262 |
Feature activation | -9.15e-01 |
Pos logit contributions | |
irens | +0.156 |
Paula | +0.150 |
iren | +0.145 |
spe | +0.144 |
Paul | +0.134 |
Neg logit contributions | |
per | -0.110 |
ilies | -0.108 |
ily | -0.105 |
like | -0.105 |
ppy | -0.102 |
Token | box |
Token ID | 3091 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
irens | +0.072 |
Paula | +0.068 |
iren | +0.064 |
Paul | +0.064 |
spe | +0.062 |
Neg logit contributions | |
per | -0.060 |
ilies | -0.059 |
ppy | -0.059 |
pper | -0.058 |
opsy | -0.058 |
Token | to |
Token ID | 284 |
Feature activation | -6.00e-01 |
Pos logit contributions | |
irens | +0.144 |
Paula | +0.137 |
iren | +0.131 |
Paul | +0.130 |
telev | +0.126 |
Neg logit contributions | |
per | -0.087 |
ppy | -0.083 |
ilies | -0.081 |
pper | -0.080 |
ectar | -0.078 |
Token | All |
Token ID | 1439 |
Feature activation | -2.47e+00 |
Pos logit contributions | |
irens | +0.015 |
Paul | +0.015 |
John | +0.015 |
Paula | +0.014 |
Sarah | +0.014 |
Neg logit contributions | |
ppy | -0.017 |
pper | -0.016 |
per | -0.016 |
ilies | -0.016 |
comb | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | -7.56e-01 |
Pos logit contributions | |
irens | +0.206 |
Paula | +0.203 |
Paul | +0.187 |
iren | +0.183 |
spe | +0.178 |
Neg logit contributions | |
ilies | -0.151 |
ppy | -0.150 |
worm | -0.149 |
opsy | -0.148 |
per | -0.146 |
Token | fish |
Token ID | 5916 |
Feature activation | +5.28e+00 |
Pos logit contributions | |
irens | +0.069 |
Paula | +0.068 |
Paul | +0.065 |
iren | +0.064 |
John | +0.063 |
Neg logit contributions | |
per | -0.039 |
ppy | -0.039 |
pper | -0.038 |
ilies | -0.038 |
opsy | -0.038 |
Token | were |
Token ID | 547 |
Feature activation | -6.28e-01 |
Pos logit contributions | |
per | +0.389 |
pper | +0.380 |
ppy | +0.380 |
ilies | +0.379 |
opsy | +0.378 |
Neg logit contributions | |
irens | -0.358 |
Paula | -0.344 |
Paul | -0.336 |
John | -0.322 |
iren | -0.321 |
Token | scared |
Token ID | 12008 |
Feature activation | -2.30e+00 |
Pos logit contributions | |
irens | +0.042 |
Paula | +0.040 |
Paul | +0.039 |
flashing | +0.038 |
John | +0.038 |
Neg logit contributions | |
ppy | -0.047 |
per | -0.047 |
pper | -0.047 |
opsy | -0.046 |
ilies | -0.046 |
Token | and |
Token ID | 290 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
irens | +0.220 |
Paula | +0.220 |
spe | +0.208 |
iren | +0.205 |
Paul | +0.204 |
Neg logit contributions | |
per | -0.110 |
ppy | -0.108 |
ilies | -0.103 |
ectar | -0.101 |
opsy | -0.100 |
Token | quickly |
Token ID | 2952 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
irens | +0.011 |
Paula | +0.010 |
Paul | +0.010 |
John | +0.010 |
Sarah | +0.010 |
Neg logit contributions | |
per | -0.011 |
pper | -0.011 |
ppy | -0.011 |
opsy | -0.011 |
ilies | -0.011 |
Token | sw |
Token ID | 1509 |
Feature activation | +8.26e+00 |
Pos logit contributions | |
irens | +0.169 |
Paul | +0.162 |
Paula | +0.161 |
John | +0.156 |
Sarah | +0.155 |
Neg logit contributions | |
opsy | -0.172 |
ilies | -0.170 |
ectar | -0.170 |
pper | -0.168 |
per | -0.168 |
Token | am |
Token ID | 321 |
Feature activation | -3.89e-01 |
Pos logit contributions | |
ilies | +0.584 |
ily | +0.579 |
comb | +0.575 |
ectar | +0.572 |
ppy | +0.555 |
Neg logit contributions | |
irens | -0.566 |
Paul | -0.505 |
iren | -0.484 |
Paula | -0.473 |
John | -0.472 |
Token | away |
Token ID | 1497 |
Feature activation | +3.63e-01 |
Pos logit contributions | |
irens | +0.031 |
Paula | +0.031 |
Paul | +0.029 |
iren | +0.028 |
John | +0.028 |
Neg logit contributions | |
per | -0.024 |
ppy | -0.024 |
opsy | -0.024 |
pper | -0.024 |
ilies | -0.024 |
Token | to |
Token ID | 284 |
Feature activation | -5.72e-01 |
Pos logit contributions | |
per | +0.022 |
ppy | +0.021 |
pper | +0.021 |
ilies | +0.021 |
opsy | +0.021 |
Neg logit contributions | |
irens | -0.030 |
Paula | -0.029 |
Paul | -0.028 |
iren | -0.027 |
John | -0.027 |
Token | a |
Token ID | 257 |
Feature activation | +8.07e-01 |
Pos logit contributions | |
per | +0.294 |
pper | +0.290 |
ilies | +0.286 |
ppy | +0.286 |
opsy | +0.283 |
Neg logit contributions | |
irens | -0.385 |
Paula | -0.381 |
Paul | -0.369 |
John | -0.363 |
Sarah | -0.357 |
Token | time |
Token ID | 640 |
Feature activation | -2.88e+00 |
Pos logit contributions | |
per | +0.067 |
ilies | +0.066 |
pper | +0.065 |
ectar | +0.064 |
ppy | +0.063 |
Neg logit contributions | |
irens | -0.046 |
Paula | -0.045 |
Paul | -0.043 |
John | -0.043 |
Sarah | -0.042 |
Token | , |
Token ID | 11 |
Feature activation | +2.62e-01 |
Pos logit contributions | |
irens | +0.279 |
Paula | +0.269 |
iren | +0.257 |
Paul | +0.255 |
spe | +0.252 |
Neg logit contributions | |
ppy | -0.131 |
per | -0.131 |
opsy | -0.129 |
ilies | -0.128 |
pper | -0.126 |
Token | there |
Token ID | 612 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
per | +0.018 |
pper | +0.018 |
ilies | +0.018 |
ppy | +0.018 |
opsy | +0.017 |
Neg logit contributions | |
Paula | -0.019 |
irens | -0.019 |
Paul | -0.018 |
John | -0.018 |
Sarah | -0.017 |
Token | was |
Token ID | 373 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
irens | +0.116 |
Paula | +0.106 |
iren | +0.104 |
flashing | +0.095 |
spe | +0.095 |
Neg logit contributions | |
opsy | -0.094 |
ily | -0.093 |
worm | -0.089 |
ectar | -0.088 |
per | -0.088 |
Token | a |
Token ID | 257 |
Feature activation | +8.38e-01 |
Pos logit contributions | |
irens | +0.111 |
Paula | +0.110 |
John | +0.108 |
Paul | +0.107 |
Sarah | +0.106 |
Neg logit contributions | |
per | -0.096 |
pper | -0.096 |
ppy | -0.094 |
ilies | -0.090 |
comb | -0.088 |
Token | little |
Token ID | 1310 |
Feature activation | -2.25e+00 |
Pos logit contributions | |
ilies | +0.065 |
per | +0.064 |
pper | +0.064 |
ppy | +0.063 |
ectar | +0.061 |
Neg logit contributions | |
Paula | -0.052 |
irens | -0.052 |
Paul | -0.050 |
John | -0.048 |
Sarah | -0.048 |
Token | girl |
Token ID | 2576 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
irens | +0.135 |
Paula | +0.132 |
Paul | +0.128 |
iren | +0.123 |
John | +0.122 |
Neg logit contributions | |
per | -0.180 |
ilies | -0.180 |
pper | -0.179 |
ppy | -0.176 |
comb | -0.171 |
Token | named |
Token ID | 3706 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
irens | +0.127 |
Paula | +0.124 |
Paul | +0.120 |
Sarah | +0.117 |
iren | +0.116 |
Neg logit contributions | |
ilies | -0.090 |
per | -0.089 |
pper | -0.089 |
ppy | -0.088 |
opsy | -0.087 |
Token | Lily |
Token ID | 20037 |
Feature activation | -9.08e-01 |
Pos logit contributions | |
per | +0.021 |
pper | +0.021 |
ppy | +0.021 |
opsy | +0.021 |
ilies | +0.021 |
Neg logit contributions | |
Paula | -0.015 |
irens | -0.015 |
Paul | -0.014 |
Sarah | -0.014 |
John | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
irens | +0.080 |
Paula | +0.077 |
Paul | +0.074 |
iren | +0.073 |
spe | +0.071 |
Neg logit contributions | |
per | -0.051 |
ppy | -0.050 |
pper | -0.049 |
opsy | -0.049 |
ilies | -0.048 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +3.95e+00 |
Pos logit contributions | |
irens | +0.012 |
Paula | +0.012 |
Paul | +0.011 |
iren | +0.011 |
flashing | +0.011 |
Neg logit contributions | |
per | -0.010 |
ppy | -0.010 |
opsy | -0.010 |
ilies | -0.010 |
pper | -0.010 |
Token | Once |
Token ID | 7454 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
per | +0.206 |
ily | +0.206 |
ilies | +0.197 |
ppy | +0.191 |
ectar | +0.191 |
Neg logit contributions | |
irens | -0.351 |
Paula | -0.342 |
spe | -0.333 |
flashing | -0.326 |
Paul | -0.320 |
Token | upon |
Token ID | 2402 |
Feature activation | +4.89e+00 |
Pos logit contributions | |
irens | +0.175 |
Paula | +0.165 |
iren | +0.160 |
Paul | +0.158 |
flashing | +0.156 |
Neg logit contributions | |
opsy | -0.084 |
ppy | -0.084 |
ilies | -0.083 |
per | -0.081 |
ectar | -0.079 |
Token | a |
Token ID | 257 |
Feature activation | +7.93e-01 |
Pos logit contributions | |
per | +0.312 |
pper | +0.309 |
ilies | +0.304 |
ppy | +0.303 |
opsy | +0.301 |
Neg logit contributions | |
Paula | -0.379 |
irens | -0.379 |
Paul | -0.366 |
John | -0.362 |
Sarah | -0.355 |
Token | time |
Token ID | 640 |
Feature activation | -2.83e+00 |
Pos logit contributions | |
per | +0.064 |
ilies | +0.063 |
pper | +0.062 |
ppy | +0.062 |
ectar | +0.061 |
Neg logit contributions | |
irens | -0.047 |
Paula | -0.046 |
Paul | -0.044 |
John | -0.043 |
Sarah | -0.042 |
Token | , |
Token ID | 11 |
Feature activation | +2.80e-01 |
Pos logit contributions | |
irens | +0.269 |
Paula | +0.260 |
iren | +0.248 |
Paul | +0.248 |
spe | +0.243 |
Neg logit contributions | |
per | -0.134 |
ppy | -0.134 |
opsy | -0.131 |
ilies | -0.131 |
pper | -0.130 |
Token | there |
Token ID | 612 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
per | +0.020 |
pper | +0.020 |
ilies | +0.020 |
ppy | +0.020 |
opsy | +0.020 |
Neg logit contributions | |
Paula | -0.019 |
irens | -0.019 |
Paul | -0.018 |
John | -0.018 |
Sarah | -0.018 |
Token | was |
Token ID | 373 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
irens | +0.111 |
Paula | +0.100 |
iren | +0.100 |
flashing | +0.091 |
spe | +0.090 |
Neg logit contributions | |
opsy | -0.091 |
ily | -0.090 |
worm | -0.086 |
per | -0.086 |
ectar | -0.086 |
Token | a |
Token ID | 257 |
Feature activation | +8.25e-01 |
Pos logit contributions | |
irens | +0.106 |
Paula | +0.106 |
John | +0.105 |
Sarah | +0.104 |
Paul | +0.103 |
Neg logit contributions | |
per | -0.098 |
pper | -0.097 |
ppy | -0.095 |
ilies | -0.090 |
comb | -0.090 |
Token | boy |
Token ID | 2933 |
Feature activation | -2.92e+00 |
Pos logit contributions | |
ilies | +0.064 |
per | +0.063 |
pper | +0.063 |
ppy | +0.063 |
ectar | +0.060 |
Neg logit contributions | |
Paula | -0.051 |
irens | -0.051 |
Paul | -0.049 |
Sarah | -0.047 |
John | -0.047 |
Token | named |
Token ID | 3706 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
irens | +0.258 |
Paula | +0.250 |
Paul | +0.243 |
iren | +0.237 |
John | +0.234 |
Neg logit contributions | |
per | -0.157 |
ilies | -0.156 |
ppy | -0.154 |
pper | -0.153 |
opsy | -0.153 |
Token | He |
Token ID | 679 |
Feature activation | -3.43e-01 |
Pos logit contributions | |
irens | +0.205 |
Paula | +0.196 |
spe | +0.186 |
iren | +0.181 |
flashing | +0.177 |
Neg logit contributions | |
per | -0.178 |
opsy | -0.172 |
ppy | -0.171 |
pper | -0.165 |
ily | -0.163 |
Token | raise |
Token ID | 5298 |
Feature activation | -9.29e-01 |
Pos logit contributions | |
irens | +0.024 |
Paula | +0.023 |
Paul | +0.022 |
iren | +0.021 |
John | +0.021 |
Neg logit contributions | |
per | -0.025 |
ilies | -0.025 |
pper | -0.025 |
ppy | -0.025 |
opsy | -0.025 |
Token | ed |
Token ID | 276 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
irens | +0.072 |
Paula | +0.068 |
spe | +0.067 |
iren | +0.066 |
Paul | +0.061 |
Neg logit contributions | |
per | -0.061 |
ppy | -0.057 |
like | -0.055 |
pper | -0.054 |
thirst | -0.053 |
Token | the |
Token ID | 262 |
Feature activation | -8.29e-01 |
Pos logit contributions | |
per | +0.001 |
ppy | +0.001 |
ilies | +0.001 |
like | +0.001 |
pper | +0.001 |
Neg logit contributions | |
irens | -0.001 |
Paula | -0.001 |
iren | -0.001 |
spe | -0.001 |
Paul | -0.001 |
Token | flashlight |
Token ID | 38371 |
Feature activation | -2.37e+00 |
Pos logit contributions | |
irens | +0.064 |
Paula | +0.061 |
Paul | +0.059 |
iren | +0.057 |
John | +0.056 |
Neg logit contributions | |
per | -0.056 |
ppy | -0.055 |
opsy | -0.053 |
pper | -0.053 |
worm | -0.052 |
Token | up |
Token ID | 510 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
irens | +0.222 |
Paula | +0.215 |
Paul | +0.205 |
iren | +0.203 |
John | +0.196 |
Neg logit contributions | |
per | -0.121 |
ppy | -0.115 |
pper | -0.112 |
ilies | -0.109 |
comb | -0.108 |
Token | higher |
Token ID | 2440 |
Feature activation | +9.69e-01 |
Pos logit contributions | |
irens | +0.210 |
Paula | +0.204 |
Paul | +0.195 |
iren | +0.191 |
spe | +0.187 |
Neg logit contributions | |
per | -0.159 |
ppy | -0.156 |
pper | -0.154 |
opsy | -0.154 |
ilies | -0.153 |
Token | and |
Token ID | 290 |
Feature activation | -2.57e-01 |
Pos logit contributions | |
ilies | +0.062 |
ppy | +0.061 |
pper | +0.060 |
opsy | +0.060 |
per | +0.059 |
Neg logit contributions | |
irens | -0.077 |
Paula | -0.074 |
Paul | -0.072 |
John | -0.070 |
iren | -0.068 |
Token | saw |
Token ID | 2497 |
Feature activation | +7.94e-01 |
Pos logit contributions | |
irens | +0.019 |
Paula | +0.019 |
Paul | +0.018 |
iren | +0.017 |
John | +0.017 |
Neg logit contributions | |
per | -0.017 |
ppy | -0.017 |
pper | -0.017 |
ilies | -0.017 |
opsy | -0.017 |
Token | it |
Token ID | 340 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
per | +0.038 |
ilies | +0.037 |
worm | +0.036 |
ily | +0.036 |
ppy | +0.036 |
Neg logit contributions | |
irens | -0.073 |
Paula | -0.072 |
spe | -0.069 |
iren | -0.068 |
Paul | -0.065 |
Token | was |
Token ID | 373 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
per | +0.016 |
ppy | +0.016 |
pper | +0.016 |
ilies | +0.016 |
opsy | +0.016 |
Neg logit contributions | |
irens | -0.013 |
Paula | -0.013 |
Paul | -0.012 |
iren | -0.012 |
John | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
irens | +0.225 |
Paula | +0.216 |
Paul | +0.210 |
iren | +0.203 |
John | +0.202 |
Neg logit contributions | |
ppy | -0.174 |
opsy | -0.173 |
pper | -0.172 |
per | -0.172 |
ilies | -0.169 |
Token | They |
Token ID | 1119 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
irens | +0.019 |
Paula | +0.019 |
iren | +0.018 |
spe | +0.018 |
flashing | +0.017 |
Neg logit contributions | |
per | -0.016 |
ppy | -0.016 |
ilies | -0.016 |
opsy | -0.016 |
ily | -0.016 |
Token | laughed |
Token ID | 13818 |
Feature activation | -8.88e-01 |
Pos logit contributions | |
irens | +0.103 |
Paula | +0.103 |
Paul | +0.100 |
John | +0.096 |
Sarah | +0.095 |
Neg logit contributions | |
per | -0.082 |
opsy | -0.079 |
ppy | -0.079 |
pper | -0.079 |
ilies | -0.079 |
Token | , |
Token ID | 11 |
Feature activation | +3.51e-01 |
Pos logit contributions | |
Paul | +0.066 |
Paula | +0.065 |
irens | +0.064 |
John | +0.064 |
Sarah | +0.063 |
Neg logit contributions | |
opsy | -0.062 |
pper | -0.061 |
ilies | -0.060 |
ppy | -0.059 |
per | -0.059 |
Token | and |
Token ID | 290 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
per | +0.020 |
ppy | +0.020 |
pper | +0.020 |
ilies | +0.020 |
opsy | +0.019 |
Neg logit contributions | |
irens | -0.030 |
Paula | -0.029 |
Paul | -0.028 |
iren | -0.027 |
spe | -0.027 |
Token | ran |
Token ID | 4966 |
Feature activation | -2.87e+00 |
Pos logit contributions | |
irens | +0.012 |
Paula | +0.011 |
iren | +0.011 |
Paul | +0.010 |
spe | +0.010 |
Neg logit contributions | |
ppy | -0.010 |
per | -0.010 |
ilies | -0.010 |
pper | -0.010 |
opsy | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | +3.74e-01 |
Pos logit contributions | |
irens | +0.264 |
Paula | +0.255 |
Paul | +0.243 |
iren | +0.242 |
spe | +0.237 |
Neg logit contributions | |
per | -0.148 |
ppy | -0.140 |
ilies | -0.140 |
pper | -0.137 |
opsy | -0.134 |
Token | and |
Token ID | 290 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
per | +0.023 |
ppy | +0.023 |
pper | +0.023 |
ilies | +0.023 |
opsy | +0.023 |
Neg logit contributions | |
irens | -0.030 |
Paula | -0.029 |
Paul | -0.028 |
iren | -0.027 |
spe | -0.027 |
Token | even |
Token ID | 772 |
Feature activation | +2.85e+00 |
Pos logit contributions | |
irens | +0.011 |
Paula | +0.010 |
iren | +0.010 |
Paul | +0.010 |
flashing | +0.009 |
Neg logit contributions | |
per | -0.010 |
ppy | -0.010 |
ilies | -0.010 |
pper | -0.010 |
opsy | -0.010 |
Token | fell |
Token ID | 3214 |
Feature activation | -4.98e-01 |
Pos logit contributions | |
ppy | +0.162 |
per | +0.158 |
ilies | +0.152 |
pper | +0.149 |
ily | +0.145 |
Neg logit contributions | |
irens | -0.253 |
Paula | -0.244 |
iren | -0.233 |
Paul | -0.230 |
flashing | -0.223 |
Token | down |
Token ID | 866 |
Feature activation | -3.03e+00 |
Pos logit contributions | |
irens | +0.040 |
Paula | +0.039 |
Paul | +0.037 |
iren | +0.036 |
spe | +0.035 |
Neg logit contributions | |
per | -0.031 |
ppy | -0.029 |
ilies | -0.029 |
ectar | -0.028 |
pper | -0.028 |
Token | pants |
Token ID | 12581 |
Feature activation | +7.51e-01 |
Pos logit contributions | |
irens | +0.118 |
Paula | +0.113 |
Paul | +0.110 |
iren | +0.108 |
flashing | +0.107 |
Neg logit contributions | |
per | -0.084 |
ppy | -0.084 |
pper | -0.084 |
ilies | -0.083 |
opsy | -0.083 |
Token | . |
Token ID | 13 |
Feature activation | -8.92e-02 |
Pos logit contributions | |
opsy | +0.050 |
ilies | +0.049 |
ppy | +0.048 |
ectar | +0.048 |
pper | +0.047 |
Neg logit contributions | |
irens | -0.058 |
Paula | -0.055 |
Paul | -0.054 |
iren | -0.053 |
flashing | -0.052 |
Token | He |
Token ID | 679 |
Feature activation | -6.54e-02 |
Pos logit contributions | |
Paula | +0.007 |
irens | +0.007 |
spe | +0.007 |
flashing | +0.007 |
Paul | +0.007 |
Neg logit contributions | |
ily | -0.006 |
per | -0.005 |
ilies | -0.005 |
opsy | -0.005 |
ppy | -0.005 |
Token | saw |
Token ID | 2497 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
irens | +0.005 |
Paula | +0.004 |
Paul | +0.004 |
iren | +0.004 |
John | +0.004 |
Neg logit contributions | |
per | -0.005 |
ppy | -0.005 |
pper | -0.005 |
ilies | -0.005 |
opsy | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
ily | +0.047 |
worm | +0.046 |
ilies | +0.042 |
per | +0.042 |
like | +0.041 |
Neg logit contributions | |
Paula | -0.100 |
irens | -0.096 |
spe | -0.095 |
Paul | -0.091 |
iren | -0.090 |
Token | big |
Token ID | 1263 |
Feature activation | -2.31e+00 |
Pos logit contributions | |
worm | +0.069 |
opsy | +0.069 |
per | +0.068 |
pper | +0.067 |
dove | +0.066 |
Neg logit contributions | |
irens | -0.097 |
Paula | -0.094 |
Paul | -0.089 |
John | -0.086 |
iren | -0.086 |
Token | hole |
Token ID | 7604 |
Feature activation | +3.25e+00 |
Pos logit contributions | |
irens | +0.184 |
Paula | +0.180 |
Paul | +0.170 |
spe | +0.165 |
iren | +0.165 |
Neg logit contributions | |
per | -0.147 |
pper | -0.142 |
ppy | -0.141 |
opsy | -0.141 |
ilies | -0.138 |
Token | on |
Token ID | 319 |
Feature activation | -1.99e-01 |
Pos logit contributions | |
per | +0.135 |
ppy | +0.129 |
pper | +0.129 |
ilies | +0.126 |
opsy | +0.124 |
Neg logit contributions | |
irens | -0.331 |
Paula | -0.326 |
Paul | -0.315 |
iren | -0.305 |
John | -0.302 |
Token | his |
Token ID | 465 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
irens | +0.017 |
Paula | +0.016 |
spe | +0.016 |
iren | +0.016 |
Paul | +0.015 |
Neg logit contributions | |
per | -0.012 |
worm | -0.011 |
pper | -0.011 |
ily | -0.010 |
like | -0.010 |
Token | knee |
Token ID | 10329 |
Feature activation | +2.75e+00 |
Pos logit contributions | |
irens | +0.102 |
Paula | +0.100 |
Paul | +0.098 |
John | +0.095 |
flashing | +0.094 |
Neg logit contributions | |
ilies | -0.096 |
ectar | -0.093 |
ppy | -0.092 |
pper | -0.091 |
opsy | -0.091 |
Token | and |
Token ID | 290 |
Feature activation | -6.40e-02 |
Pos logit contributions | |
per | +0.147 |
pper | +0.142 |
ppy | +0.141 |
ilies | +0.135 |
opsy | +0.133 |
Neg logit contributions | |
irens | -0.246 |
Paula | -0.243 |
Paul | -0.232 |
spe | -0.226 |
John | -0.224 |
Token | make |
Token ID | 787 |
Feature activation | -3.75e-01 |
Pos logit contributions | |
irens | +0.010 |
Paula | +0.010 |
Paul | +0.010 |
John | +0.009 |
iren | +0.009 |
Neg logit contributions | |
per | -0.008 |
opsy | -0.008 |
ppy | -0.008 |
pper | -0.008 |
ilies | -0.008 |
Token | loud |
Token ID | 7812 |
Feature activation | -3.00e+00 |
Pos logit contributions | |
irens | +0.029 |
Paula | +0.028 |
Paul | +0.027 |
John | +0.026 |
iren | +0.026 |
Neg logit contributions | |
per | -0.025 |
ppy | -0.025 |
pper | -0.025 |
opsy | -0.025 |
ilies | -0.024 |
Token | noises |
Token ID | 26782 |
Feature activation | -4.10e+00 |
Pos logit contributions | |
irens | +0.182 |
Paula | +0.181 |
Paul | +0.167 |
iren | +0.162 |
spe | +0.157 |
Neg logit contributions | |
per | -0.245 |
ppy | -0.235 |
ilies | -0.232 |
pper | -0.231 |
ectar | -0.225 |
Token | . |
Token ID | 13 |
Feature activation | -7.65e-02 |
Pos logit contributions | |
irens | +0.346 |
Paula | +0.336 |
Paul | +0.326 |
iren | +0.317 |
John | +0.316 |
Neg logit contributions | |
opsy | -0.244 |
ilies | -0.234 |
ppy | -0.231 |
pper | -0.230 |
per | -0.229 |
Token | V |
Token ID | 569 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
irens | +0.005 |
Paul | +0.005 |
Paula | +0.005 |
John | +0.005 |
Sarah | +0.005 |
Neg logit contributions | |
opsy | -0.006 |
pper | -0.006 |
ilies | -0.006 |
ppy | -0.006 |
per | -0.006 |
Token | room |
Token ID | 3823 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
per | +0.092 |
ppy | +0.091 |
pper | +0.090 |
ilies | +0.090 |
opsy | +0.089 |
Neg logit contributions | |
irens | -0.090 |
Paula | -0.089 |
Paul | -0.084 |
iren | -0.080 |
spe | -0.080 |
Token | , |
Token ID | 11 |
Feature activation | +4.36e-01 |
Pos logit contributions | |
irens | +0.243 |
Paula | +0.242 |
Paul | +0.227 |
spe | +0.221 |
iren | +0.220 |
Neg logit contributions | |
per | -0.134 |
pper | -0.130 |
ppy | -0.129 |
ilies | -0.122 |
ily | -0.118 |
Token | v |
Token ID | 410 |
Feature activation | +9.89e-01 |
Pos logit contributions | |
opsy | +0.032 |
ppy | +0.032 |
per | +0.032 |
pper | +0.032 |
ilies | +0.031 |
Neg logit contributions | |
irens | -0.030 |
Paula | -0.029 |
Paul | -0.028 |
iren | -0.027 |
flashing | -0.027 |
Token | room |
Token ID | 3823 |
Feature activation | -2.62e+00 |
Pos logit contributions | |
opsy | +0.081 |
ppy | +0.081 |
per | +0.080 |
ilies | +0.080 |
pper | +0.079 |
Neg logit contributions | |
irens | -0.062 |
iren | -0.055 |
Paula | -0.054 |
Paul | -0.053 |
spe | -0.051 |
Token | !" |
Token ID | 2474 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
irens | +0.235 |
Paula | +0.232 |
Paul | +0.220 |
iren | +0.214 |
spe | +0.214 |
Neg logit contributions | |
per | -0.139 |
pper | -0.135 |
ppy | -0.135 |
ilies | -0.129 |
opsy | -0.126 |
Token |
|
Token ID | 198 |
Feature activation | -4.52e-01 |
Pos logit contributions | |
ily | +0.001 |
ilies | +0.001 |
per | +0.001 |
like | +0.001 |
ectar | +0.001 |
Neg logit contributions | |
irens | -0.002 |
Paula | -0.002 |
spe | -0.002 |
flashing | -0.002 |
telev | -0.002 |
Token | asking |
Token ID | 4737 |
Feature activation | -4.39e+00 |
Pos logit contributions | |
ppy | +0.014 |
pper | +0.014 |
opsy | +0.014 |
ilies | +0.014 |
per | +0.013 |
Neg logit contributions | |
Paula | -0.014 |
irens | -0.014 |
Paul | -0.013 |
John | -0.013 |
flashing | -0.013 |
Token | her |
Token ID | 607 |
Feature activation | +2.26e-01 |
Pos logit contributions | |
Paula | +0.338 |
Paul | +0.334 |
Sarah | +0.325 |
John | +0.318 |
irens | +0.312 |
Neg logit contributions | |
pper | -0.293 |
opsy | -0.290 |
ppy | -0.282 |
per | -0.280 |
ilies | -0.275 |
Token | mum |
Token ID | 25682 |
Feature activation | +3.52e+00 |
Pos logit contributions | |
pper | +0.015 |
per | +0.014 |
ppy | +0.014 |
ider | +0.014 |
ars | +0.014 |
Neg logit contributions | |
Paul | -0.017 |
Paula | -0.017 |
Sarah | -0.017 |
John | -0.017 |
spe | -0.016 |
Token | if |
Token ID | 611 |
Feature activation | -2.62e-01 |
Pos logit contributions | |
per | +0.206 |
ppy | +0.202 |
pper | +0.201 |
opsy | +0.201 |
ilies | +0.199 |
Neg logit contributions | |
irens | -0.293 |
Paula | -0.289 |
Paul | -0.278 |
John | -0.267 |
iren | -0.267 |
Token | it |
Token ID | 340 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
Paula | +0.018 |
irens | +0.018 |
Paul | +0.017 |
Sarah | +0.017 |
John | +0.017 |
Neg logit contributions | |
per | -0.019 |
ppy | -0.019 |
pper | -0.019 |
opsy | -0.019 |
ilies | -0.018 |
Token | was |
Token ID | 373 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
per | +0.014 |
ppy | +0.014 |
pper | +0.014 |
opsy | +0.014 |
ilies | +0.014 |
Neg logit contributions | |
irens | -0.011 |
Paula | -0.010 |
Paul | -0.010 |
iren | -0.009 |
John | -0.009 |
Token | like |
Token ID | 588 |
Feature activation | +5.84e+00 |
Pos logit contributions | |
Paula | +0.121 |
irens | +0.119 |
Paul | +0.117 |
John | +0.114 |
Sarah | +0.114 |
Neg logit contributions | |
per | -0.091 |
ppy | -0.090 |
pper | -0.090 |
opsy | -0.089 |
ilies | -0.089 |
Token | her |
Token ID | 607 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
per | +0.328 |
ppy | +0.323 |
ilies | +0.321 |
pper | +0.319 |
opsy | +0.318 |
Neg logit contributions | |
irens | -0.503 |
Paula | -0.487 |
Paul | -0.468 |
iren | -0.460 |
spe | -0.450 |
Token | jewelry |
Token ID | 22634 |
Feature activation | -2.39e+00 |
Pos logit contributions | |
per | +0.013 |
ppy | +0.013 |
pper | +0.013 |
opsy | +0.013 |
ilies | +0.013 |
Neg logit contributions | |
Paula | -0.016 |
irens | -0.016 |
Paul | -0.016 |
John | -0.015 |
spe | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -4.22e-01 |
Pos logit contributions | |
irens | +0.188 |
Paula | +0.187 |
Paul | +0.180 |
John | +0.172 |
spe | +0.172 |
Neg logit contributions | |
per | -0.147 |
ppy | -0.145 |
opsy | -0.145 |
ilies | -0.144 |
pper | -0.143 |
Token |
|
Token ID | 198 |
Feature activation | -5.75e-01 |
Pos logit contributions | |
irens | +0.034 |
Paula | +0.032 |
spe | +0.031 |
iren | +0.031 |
flashing | +0.031 |
Neg logit contributions | |
opsy | -0.026 |
per | -0.025 |
ilies | -0.025 |
ppy | -0.025 |
ectar | -0.025 |
Token | He |
Token ID | 679 |
Feature activation | -2.29e-01 |
Pos logit contributions | |
irens | +0.018 |
Paula | +0.018 |
spe | +0.017 |
iren | +0.017 |
Paul | +0.016 |
Neg logit contributions | |
per | -0.016 |
ppy | -0.016 |
opsy | -0.015 |
ilies | -0.015 |
comb | -0.015 |
Token | saw |
Token ID | 2497 |
Feature activation | +9.00e-01 |
Pos logit contributions | |
irens | +0.015 |
Paula | +0.014 |
Paul | +0.014 |
iren | +0.013 |
John | +0.013 |
Neg logit contributions | |
per | -0.018 |
ppy | -0.018 |
pper | -0.017 |
ilies | -0.017 |
opsy | -0.017 |
Token | a |
Token ID | 257 |
Feature activation | +9.51e-01 |
Pos logit contributions | |
ily | +0.038 |
worm | +0.038 |
ilies | +0.037 |
dove | +0.035 |
per | +0.034 |
Neg logit contributions | |
irens | -0.086 |
Paula | -0.086 |
spe | -0.085 |
iren | -0.081 |
Paul | -0.078 |
Token | big |
Token ID | 1263 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
worm | +0.055 |
opsy | +0.053 |
dove | +0.052 |
per | +0.051 |
pper | +0.049 |
Neg logit contributions | |
irens | -0.083 |
Paula | -0.080 |
Paul | -0.075 |
iren | -0.073 |
John | -0.073 |
Token | foot |
Token ID | 2366 |
Feature activation | +4.22e+00 |
Pos logit contributions | |
irens | +0.214 |
Paula | +0.212 |
Paul | +0.198 |
spe | +0.198 |
iren | +0.192 |
Neg logit contributions | |
per | -0.142 |
worm | -0.136 |
opsy | -0.134 |
pper | -0.134 |
ppy | -0.132 |
Token | coming |
Token ID | 2406 |
Feature activation | -7.64e+00 |
Pos logit contributions | |
per | +0.212 |
pper | +0.204 |
ppy | +0.199 |
ilies | +0.188 |
ily | +0.186 |
Neg logit contributions | |
Paula | -0.393 |
irens | -0.387 |
Paul | -0.369 |
spe | -0.355 |
telev | -0.355 |
Token | down |
Token ID | 866 |
Feature activation | -3.01e+00 |
Pos logit contributions | |
irens | +0.654 |
Paula | +0.639 |
Paul | +0.597 |
iren | +0.590 |
spe | +0.585 |
Neg logit contributions | |
per | -0.445 |
ppy | -0.429 |
pper | -0.426 |
ilies | -0.409 |
opsy | -0.405 |
Token | to |
Token ID | 284 |
Feature activation | -5.80e-01 |
Pos logit contributions | |
Paula | +0.295 |
parent | +0.273 |
spe | +0.273 |
irens | +0.269 |
telev | +0.269 |
Neg logit contributions | |
like | -0.142 |
ily | -0.125 |
per | -0.124 |
still | -0.110 |
dove | -0.105 |
Token | destroy |
Token ID | 4117 |
Feature activation | -5.47e+00 |
Pos logit contributions | |
irens | +0.051 |
Paula | +0.049 |
spe | +0.047 |
Paul | +0.047 |
iren | +0.046 |
Neg logit contributions | |
per | -0.032 |
pper | -0.032 |
ppy | -0.032 |
opsy | -0.031 |
ilies | -0.031 |
Token | the |
Token ID | 262 |
Feature activation | -6.96e-01 |
Pos logit contributions | |
irens | +0.414 |
Paula | +0.409 |
spe | +0.385 |
iren | +0.383 |
Paul | +0.374 |
Neg logit contributions | |
per | -0.365 |
ppy | -0.351 |
ilies | -0.351 |
pper | -0.347 |
ily | -0.340 |
Token | hill |
Token ID | 12788 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
irens | +0.051 |
Paula | +0.050 |
Paul | +0.048 |
iren | +0.046 |
John | +0.045 |
Neg logit contributions | |
per | -0.048 |
ppy | -0.048 |
pper | -0.047 |
opsy | -0.047 |
ilies | -0.047 |
Token | . |
Token ID | 13 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
per | +0.259 |
ppy | +0.255 |
pper | +0.252 |
opsy | +0.244 |
ilies | +0.238 |
Neg logit contributions | |
irens | -0.446 |
Paula | -0.432 |
Paul | -0.411 |
iren | -0.405 |
spe | -0.398 |
Token | I |
Token ID | 314 |
Feature activation | +9.76e-01 |
Pos logit contributions | |
per | +0.002 |
pper | +0.002 |
ppy | +0.002 |
ilies | +0.002 |
opsy | +0.002 |
Neg logit contributions | |
irens | -0.002 |
Paula | -0.002 |
Paul | -0.002 |
iren | -0.002 |
John | -0.002 |
Token | should |
Token ID | 815 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
per | +0.072 |
pper | +0.070 |
ilies | +0.070 |
ppy | +0.070 |
opsy | +0.069 |
Neg logit contributions | |
irens | -0.067 |
Paula | -0.065 |
Paul | -0.063 |
spe | -0.061 |
iren | -0.060 |
Token | not |
Token ID | 407 |
Feature activation | +8.10e-01 |
Pos logit contributions | |
per | +0.084 |
ppy | +0.083 |
pper | +0.082 |
opsy | +0.081 |
ilies | +0.081 |
Neg logit contributions | |
irens | -0.138 |
Paula | -0.135 |
Paul | -0.130 |
iren | -0.127 |
spe | -0.124 |
Token | have |
Token ID | 423 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
per | +0.046 |
ppy | +0.046 |
pper | +0.045 |
opsy | +0.045 |
ilies | +0.045 |
Neg logit contributions | |
irens | -0.070 |
Paula | -0.068 |
Paul | -0.065 |
iren | -0.063 |
John | -0.062 |
Token | taken |
Token ID | 2077 |
Feature activation | -7.99e+00 |
Pos logit contributions | |
per | +0.001 |
ppy | +0.001 |
pper | +0.001 |
ily | +0.001 |
ilies | +0.001 |
Neg logit contributions | |
irens | -0.001 |
Paula | -0.001 |
iren | -0.001 |
Paul | -0.001 |
spe | -0.001 |
Token | your |
Token ID | 534 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
irens | +0.669 |
Paula | +0.653 |
Paul | +0.621 |
spe | +0.612 |
iren | +0.611 |
Neg logit contributions | |
per | -0.469 |
ppy | -0.465 |
ilies | -0.457 |
pper | -0.449 |
opsy | -0.441 |
Token | car |
Token ID | 1097 |
Feature activation | -2.56e+00 |
Pos logit contributions | |
per | +0.104 |
ilies | +0.103 |
opsy | +0.102 |
ppy | +0.101 |
pper | +0.100 |
Neg logit contributions | |
irens | -0.122 |
Paula | -0.117 |
Paul | -0.111 |
iren | -0.109 |
John | -0.106 |
Token | . |
Token ID | 13 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
irens | +0.215 |
Paula | +0.209 |
Paul | +0.201 |
iren | +0.196 |
John | +0.193 |
Neg logit contributions | |
per | -0.149 |
ppy | -0.147 |
opsy | -0.146 |
pper | -0.146 |
ilies | -0.146 |
Token | Can |
Token ID | 1680 |
Feature activation | -2.60e+00 |
Pos logit contributions | |
pper | +0.001 |
ilies | +0.001 |
per | +0.001 |
ppy | +0.001 |
opsy | +0.001 |
Neg logit contributions | |
irens | -0.001 |
Paula | -0.001 |
Paul | -0.001 |
John | -0.001 |
Sarah | -0.001 |
Token | I |
Token ID | 314 |
Feature activation | +9.90e-01 |
Pos logit contributions | |
irens | +0.224 |
Paula | +0.216 |
Paul | +0.207 |
iren | +0.204 |
spe | +0.200 |
Neg logit contributions | |
per | -0.147 |
ppy | -0.145 |
pper | -0.144 |
opsy | -0.143 |
ilies | -0.142 |
Token | then |
Token ID | 788 |
Feature activation | +5.46e+00 |
Pos logit contributions | |
irens | +0.112 |
Paula | +0.111 |
Paul | +0.108 |
John | +0.105 |
iren | +0.103 |
Neg logit contributions | |
per | -0.052 |
ppy | -0.052 |
ilies | -0.050 |
pper | -0.050 |
opsy | -0.049 |
Token | on |
Token ID | 319 |
Feature activation | -4.31e-01 |
Pos logit contributions | |
per | +0.323 |
ppy | +0.319 |
pper | +0.316 |
ilies | +0.315 |
opsy | +0.315 |
Neg logit contributions | |
irens | -0.459 |
Paula | -0.446 |
Paul | -0.431 |
iren | -0.418 |
John | -0.411 |
Token | , |
Token ID | 11 |
Feature activation | +3.18e-01 |
Pos logit contributions | |
irens | +0.042 |
iren | +0.040 |
Paula | +0.039 |
spe | +0.038 |
flashing | +0.038 |
Neg logit contributions | |
per | -0.019 |
opsy | -0.018 |
ppy | -0.018 |
pper | -0.018 |
ilies | -0.017 |
Token | Emma |
Token ID | 18966 |
Feature activation | -6.07e+00 |
Pos logit contributions | |
like | +0.024 |
ilies | +0.024 |
ppy | +0.024 |
opsy | +0.024 |
ily | +0.024 |
Neg logit contributions | |
irens | -0.021 |
iren | -0.020 |
flashing | -0.017 |
Mechan | -0.015 |
telev | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
Paula | +0.450 |
irens | +0.450 |
Paul | +0.435 |
John | +0.417 |
spe | +0.411 |
Neg logit contributions | |
ilies | -0.396 |
per | -0.392 |
pper | -0.386 |
opsy | -0.385 |
ppy | -0.385 |
Token | Sarah |
Token ID | 10490 |
Feature activation | -8.67e+00 |
Pos logit contributions | |
irens | +0.012 |
iren | +0.012 |
flashing | +0.010 |
spe | +0.009 |
lights | +0.008 |
Neg logit contributions | |
ily | -0.015 |
ppy | -0.015 |
per | -0.014 |
like | -0.014 |
opsy | -0.014 |
Token | ate |
Token ID | 15063 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
Paula | +0.654 |
irens | +0.642 |
Paul | +0.630 |
spe | +0.611 |
John | +0.608 |
Neg logit contributions | |
per | -0.562 |
ilies | -0.553 |
opsy | -0.546 |
ectar | -0.543 |
pper | -0.539 |
Token | muff |
Token ID | 27563 |
Feature activation | +8.54e-01 |
Pos logit contributions | |
per | +0.074 |
ppy | +0.073 |
pper | +0.072 |
ilies | +0.072 |
opsy | +0.072 |
Neg logit contributions | |
irens | -0.088 |
Paula | -0.085 |
Paul | -0.082 |
iren | -0.079 |
John | -0.078 |
Token | ins |
Token ID | 1040 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
per | +0.063 |
ppy | +0.061 |
pper | +0.061 |
opsy | +0.060 |
ilies | +0.060 |
Neg logit contributions | |
irens | -0.060 |
Paula | -0.057 |
Paul | -0.054 |
iren | -0.053 |
spe | -0.052 |
Token | together |
Token ID | 1978 |
Feature activation | -5.28e-01 |
Pos logit contributions | |
irens | +0.088 |
Paula | +0.085 |
Paul | +0.082 |
John | +0.079 |
iren | +0.079 |
Neg logit contributions | |
per | -0.074 |
opsy | -0.072 |
ppy | -0.072 |
pper | -0.072 |
ilies | -0.071 |
Token | every |
Token ID | 790 |
Feature activation | +9.17e-01 |
Pos logit contributions | |
irens | +0.043 |
Paula | +0.043 |
Paul | +0.041 |
John | +0.040 |
spe | +0.039 |
Neg logit contributions | |
per | -0.031 |
ppy | -0.031 |
pper | -0.031 |
opsy | -0.030 |
ilies | -0.030 |
Token | 's |
Token ID | 338 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
irens | +0.343 |
Paula | +0.338 |
Paul | +0.331 |
John | +0.319 |
Sarah | +0.314 |
Neg logit contributions | |
ilies | -0.400 |
per | -0.394 |
ectar | -0.392 |
ppy | -0.388 |
pper | -0.387 |
Token | mom |
Token ID | 1995 |
Feature activation | +3.90e-03 |
Pos logit contributions | |
opsy | +0.129 |
per | +0.127 |
ppy | +0.125 |
pper | +0.124 |
ilies | +0.121 |
Neg logit contributions | |
irens | -0.258 |
Paula | -0.247 |
iren | -0.237 |
Paul | -0.233 |
spe | -0.226 |
Token | gave |
Token ID | 2921 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
ppy | +0.000 |
pper | +0.000 |
per | +0.000 |
opsy | +0.000 |
ily | +0.000 |
Neg logit contributions | |
irens | -0.000 |
Paula | -0.000 |
iren | -0.000 |
Paul | -0.000 |
flashing | -0.000 |
Token | them |
Token ID | 606 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
irens | +0.199 |
spe | +0.191 |
iren | +0.186 |
Paula | +0.183 |
million | +0.180 |
Neg logit contributions | |
ily | -0.102 |
ppy | -0.098 |
like | -0.098 |
per | -0.096 |
ilies | -0.093 |
Token | a |
Token ID | 257 |
Feature activation | +8.63e-01 |
Pos logit contributions | |
irens | +0.110 |
Paula | +0.107 |
Paul | +0.106 |
John | +0.105 |
Sarah | +0.104 |
Neg logit contributions | |
per | -0.075 |
opsy | -0.074 |
pper | -0.073 |
ilies | -0.072 |
ectar | -0.072 |
Token | brand |
Token ID | 4508 |
Feature activation | -7.32e+00 |
Pos logit contributions | |
opsy | +0.050 |
worm | +0.050 |
ppy | +0.048 |
ilies | +0.047 |
per | +0.047 |
Neg logit contributions | |
irens | -0.076 |
Paula | -0.071 |
Paul | -0.068 |
iren | -0.065 |
spe | -0.065 |
Token | new |
Token ID | 649 |
Feature activation | -3.73e+00 |
Pos logit contributions | |
irens | +0.464 |
Paula | +0.448 |
Paul | +0.421 |
iren | +0.404 |
John | +0.396 |
Neg logit contributions | |
per | -0.577 |
opsy | -0.571 |
ilies | -0.570 |
ppy | -0.569 |
pper | -0.566 |
Token | basketball |
Token ID | 9669 |
Feature activation | -7.80e-01 |
Pos logit contributions | |
irens | +0.319 |
Paula | +0.308 |
Paul | +0.294 |
iren | +0.289 |
spe | +0.281 |
Neg logit contributions | |
per | -0.215 |
opsy | -0.211 |
ppy | -0.211 |
ilies | -0.209 |
pper | -0.208 |
Token | and |
Token ID | 290 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
irens | +0.055 |
Paula | +0.052 |
Paul | +0.052 |
iren | +0.051 |
John | +0.050 |
Neg logit contributions | |
opsy | -0.054 |
ppy | -0.054 |
pper | -0.053 |
per | -0.053 |
ilies | -0.053 |
Token | they |
Token ID | 484 |
Feature activation | -2.26e+00 |
Pos logit contributions | |
irens | +0.019 |
Paula | +0.018 |
iren | +0.018 |
Paul | +0.016 |
spe | +0.016 |
Neg logit contributions | |
per | -0.015 |
ppy | -0.015 |
opsy | -0.015 |
ilies | -0.015 |
pper | -0.015 |
Token | played |
Token ID | 2826 |
Feature activation | +7.16e-01 |
Pos logit contributions | |
Paula | +0.164 |
Paul | +0.163 |
irens | +0.158 |
John | +0.157 |
Sarah | +0.156 |
Neg logit contributions | |
per | -0.153 |
ectar | -0.151 |
ilies | -0.149 |
pper | -0.147 |
opsy | -0.147 |
Token | of |
Token ID | 286 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
ily | +0.152 |
ilies | +0.144 |
ectar | +0.142 |
per | +0.129 |
ap | +0.128 |
Neg logit contributions | |
irens | -0.328 |
Paula | -0.315 |
flashing | -0.305 |
telev | -0.302 |
Mechan | -0.300 |
Token | the |
Token ID | 262 |
Feature activation | -7.44e-01 |
Pos logit contributions | |
irens | +0.130 |
spe | +0.124 |
iren | +0.120 |
flashing | +0.118 |
Paula | +0.117 |
Neg logit contributions | |
opsy | -0.075 |
per | -0.074 |
ily | -0.070 |
pper | -0.070 |
ilies | -0.070 |
Token | day |
Token ID | 1110 |
Feature activation | -6.28e-01 |
Pos logit contributions | |
irens | +0.057 |
Paula | +0.055 |
Paul | +0.053 |
iren | +0.052 |
John | +0.050 |
Neg logit contributions | |
per | -0.049 |
opsy | -0.048 |
ilies | -0.048 |
ppy | -0.048 |
pper | -0.048 |
Token | , |
Token ID | 11 |
Feature activation | +3.81e-01 |
Pos logit contributions | |
irens | +0.063 |
Paula | +0.059 |
iren | +0.058 |
flashing | +0.058 |
Paul | +0.057 |
Neg logit contributions | |
per | -0.026 |
ilies | -0.026 |
ily | -0.024 |
ectar | -0.024 |
ppy | -0.023 |
Token | Miss |
Token ID | 4544 |
Feature activation | -5.59e+00 |
Pos logit contributions | |
like | +0.031 |
ily | +0.031 |
ilies | +0.030 |
per | +0.030 |
wait | +0.029 |
Neg logit contributions | |
irens | -0.022 |
iren | -0.021 |
spe | -0.018 |
Paula | -0.017 |
Mechan | -0.017 |
Token | Sarah |
Token ID | 10490 |
Feature activation | -8.67e+00 |
Pos logit contributions | |
irens | +0.279 |
iren | +0.249 |
flashing | +0.238 |
chaos | +0.215 |
Paula | +0.198 |
Neg logit contributions | |
ppy | -0.546 |
ily | -0.545 |
per | -0.532 |
ilies | -0.522 |
ectar | -0.491 |
Token | gave |
Token ID | 2921 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
irens | +0.647 |
Paula | +0.626 |
Paul | +0.599 |
iren | +0.582 |
John | +0.570 |
Neg logit contributions | |
per | -0.588 |
ppy | -0.581 |
pper | -0.576 |
ilies | -0.573 |
opsy | -0.573 |
Token | all |
Token ID | 477 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
irens | +0.182 |
iren | +0.174 |
spe | +0.171 |
flashing | +0.163 |
Mechan | +0.161 |
Neg logit contributions | |
ily | -0.107 |
ilies | -0.099 |
per | -0.098 |
ppy | -0.095 |
like | -0.095 |
Token | the |
Token ID | 262 |
Feature activation | -7.48e-01 |
Pos logit contributions | |
ilies | +0.089 |
ily | +0.089 |
opsy | +0.082 |
ectar | +0.082 |
ppy | +0.082 |
Neg logit contributions | |
irens | -0.154 |
spe | -0.142 |
Paula | -0.140 |
iren | -0.140 |
Paul | -0.137 |
Token | pupils |
Token ID | 25613 |
Feature activation | -2.30e+00 |
Pos logit contributions | |
irens | +0.056 |
Paula | +0.054 |
Paul | +0.051 |
iren | +0.050 |
John | +0.049 |
Neg logit contributions | |
per | -0.052 |
ppy | -0.051 |
pper | -0.051 |
ilies | -0.051 |
opsy | -0.050 |
Token | a |
Token ID | 257 |
Feature activation | +9.90e-01 |
Pos logit contributions | |
irens | +0.186 |
Paula | +0.184 |
Paul | +0.176 |
John | +0.172 |
Sarah | +0.170 |
Neg logit contributions | |
pper | -0.140 |
per | -0.139 |
ppy | -0.139 |
opsy | -0.138 |
ilies | -0.131 |