Token | do |
Token ID | 466 |
Feature activation | -5.82e+00 |
Pos logit contributions | |
Paula | +0.081 |
irens | +0.079 |
Paul | +0.078 |
John | +0.075 |
Sarah | +0.075 |
Neg logit contributions | |
ppy | -0.026 |
ilies | -0.025 |
per | -0.024 |
opsy | -0.024 |
pper | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
ppy | +0.345 |
pper | +0.332 |
per | +0.330 |
ilies | +0.330 |
opsy | +0.326 |
Neg logit contributions | |
Paula | -0.507 |
irens | -0.492 |
Paul | -0.481 |
spe | -0.455 |
John | -0.452 |
Token | He |
Token ID | 679 |
Feature activation | +3.23e-01 |
Pos logit contributions | |
Paula | +0.014 |
irens | +0.013 |
Paul | +0.012 |
spe | +0.012 |
flashing | +0.012 |
Neg logit contributions | |
ppy | -0.013 |
per | -0.013 |
ily | -0.013 |
ilies | -0.013 |
pper | -0.012 |
Token | hears |
Token ID | 23180 |
Feature activation | +6.05e+00 |
Pos logit contributions | |
irens | +0.020 |
Paula | +0.019 |
Paul | +0.018 |
flashing | +0.018 |
James | +0.017 |
Neg logit contributions | |
ectar | -0.027 |
ppy | -0.026 |
opsy | -0.025 |
pper | -0.025 |
ilies | -0.025 |
Token | s |
Token ID | 264 |
Feature activation | +9.77e-01 |
Pos logit contributions | |
Paula | +0.548 |
spe | +0.518 |
irens | +0.517 |
Paul | +0.505 |
Ellen | +0.483 |
Neg logit contributions | |
ily | -0.338 |
ppy | -0.329 |
per | -0.326 |
ilies | -0.322 |
pper | -0.305 |
Token | irens |
Token ID | 42917 |
Feature activation | +1.34e+01 |
Pos logit contributions | |
Paula | +0.016 |
Paul | +0.012 |
spe | +0.008 |
John | +0.007 |
James | +0.006 |
Neg logit contributions | |
ppy | -0.130 |
per | -0.129 |
pper | -0.128 |
ilies | -0.126 |
ily | -0.125 |
Token | . |
Token ID | 13 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
Paula | +1.296 |
Paul | +1.230 |
irens | +1.195 |
spe | +1.183 |
Sarah | +1.169 |
Neg logit contributions | |
per | -0.710 |
ppy | -0.683 |
pper | -0.641 |
like | -0.609 |
ilies | -0.589 |
Token | He |
Token ID | 679 |
Feature activation | +3.27e-01 |
Pos logit contributions | |
Paula | +0.015 |
irens | +0.014 |
spe | +0.014 |
Paul | +0.014 |
flashing | +0.013 |
Neg logit contributions | |
ily | -0.013 |
per | -0.013 |
ppy | -0.013 |
ilies | -0.013 |
ectar | -0.012 |
Token | sees |
Token ID | 7224 |
Feature activation | +1.89e+00 |
Pos logit contributions | |
Paula | +0.022 |
irens | +0.022 |
Paul | +0.021 |
spe | +0.019 |
John | +0.019 |
Neg logit contributions | |
ppy | -0.025 |
pper | -0.025 |
ilies | -0.025 |
per | -0.025 |
opsy | -0.024 |
Token | a |
Token ID | 257 |
Feature activation | -6.69e-01 |
Pos logit contributions | |
Paula | +0.178 |
spe | +0.168 |
Paul | +0.162 |
irens | +0.161 |
Ellen | +0.156 |
Neg logit contributions | |
ily | -0.107 |
ilies | -0.100 |
per | -0.098 |
ppy | -0.096 |
worm | -0.095 |
Token | police |
Token ID | 1644 |
Feature activation | +5.66e+00 |
Pos logit contributions | |
worm | +0.040 |
dove | +0.037 |
opsy | +0.037 |
per | +0.036 |
ppy | +0.035 |
Neg logit contributions | |
Paula | -0.058 |
irens | -0.055 |
Paul | -0.054 |
James | -0.053 |
John | -0.052 |
Token | heard |
Token ID | 2982 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
Paula | +0.105 |
irens | +0.101 |
Paul | +0.100 |
John | +0.093 |
spe | +0.092 |
Neg logit contributions | |
ppy | -0.103 |
ilies | -0.101 |
pper | -0.101 |
per | -0.100 |
opsy | -0.100 |
Token | people |
Token ID | 661 |
Feature activation | +5.83e-01 |
Pos logit contributions | |
Paula | +0.182 |
spe | +0.180 |
irens | +0.177 |
parent | +0.171 |
iren | +0.169 |
Neg logit contributions | |
like | -0.053 |
ily | -0.053 |
worm | -0.047 |
dove | -0.045 |
ilies | -0.045 |
Token | shouting |
Token ID | 19642 |
Feature activation | +6.70e+00 |
Pos logit contributions | |
Paula | +0.038 |
irens | +0.036 |
Paul | +0.035 |
spe | +0.034 |
James | +0.033 |
Neg logit contributions | |
per | -0.045 |
ppy | -0.044 |
like | -0.044 |
pper | -0.043 |
ilies | -0.043 |
Token | and |
Token ID | 290 |
Feature activation | +1.97e-01 |
Pos logit contributions | |
Paula | +0.625 |
irens | +0.609 |
Paul | +0.592 |
spe | +0.568 |
John | +0.557 |
Neg logit contributions | |
ppy | -0.355 |
per | -0.339 |
ilies | -0.334 |
pper | -0.333 |
opsy | -0.325 |
Token | s |
Token ID | 264 |
Feature activation | +9.75e-01 |
Pos logit contributions | |
irens | +0.015 |
Paula | +0.014 |
igsaw | +0.013 |
iren | +0.013 |
parent | +0.013 |
Neg logit contributions | |
ppy | -0.013 |
like | -0.013 |
per | -0.013 |
wait | -0.013 |
careful | -0.012 |
Token | irens |
Token ID | 42917 |
Feature activation | +1.34e+01 |
Pos logit contributions | |
Paula | +0.023 |
spe | +0.021 |
Paul | +0.020 |
James | +0.017 |
John | +0.015 |
Neg logit contributions | |
ppy | -0.119 |
per | -0.119 |
pper | -0.119 |
ilies | -0.115 |
ars | -0.112 |
Token | w |
Token ID | 266 |
Feature activation | -2.72e+00 |
Pos logit contributions | |
Paula | +1.293 |
irens | +1.207 |
Paul | +1.193 |
spe | +1.181 |
Sarah | +1.163 |
Neg logit contributions | |
per | -0.658 |
ppy | -0.634 |
pper | -0.617 |
like | -0.606 |
ilies | -0.578 |
Token | ailing |
Token ID | 11608 |
Feature activation | +8.95e-01 |
Pos logit contributions | |
ppy | +0.124 |
pper | +0.120 |
opsy | +0.118 |
ilies | +0.116 |
per | +0.116 |
Neg logit contributions | |
irens | -0.272 |
Paula | -0.267 |
Paul | -0.256 |
iren | -0.246 |
flashing | -0.243 |
Token | . |
Token ID | 13 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
Paula | +0.091 |
irens | +0.087 |
Sarah | +0.086 |
parent | +0.084 |
James | +0.084 |
Neg logit contributions | |
per | -0.035 |
like | -0.033 |
pper | -0.027 |
ilies | -0.026 |
ppy | -0.026 |
Token | They |
Token ID | 1119 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
Paula | +0.015 |
irens | +0.014 |
Paul | +0.014 |
spe | +0.014 |
iren | +0.013 |
Neg logit contributions | |
per | -0.013 |
ilies | -0.013 |
ily | -0.012 |
ppy | -0.012 |
opsy | -0.012 |
Token | were |
Token ID | 547 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
Paula | +0.116 |
irens | +0.116 |
Paul | +0.107 |
spe | +0.105 |
iren | +0.103 |
Neg logit contributions | |
ppy | -0.096 |
per | -0.090 |
pper | -0.090 |
ilies | -0.087 |
opsy | -0.087 |
Token | drove |
Token ID | 10357 |
Feature activation | +6.76e+00 |
Pos logit contributions | |
Paula | +0.212 |
irens | +0.186 |
igsaw | +0.183 |
telev | +0.180 |
channels | +0.179 |
Neg logit contributions | |
per | -0.202 |
pper | -0.200 |
like | -0.186 |
ppy | -0.184 |
dove | -0.183 |
Token | by |
Token ID | 416 |
Feature activation | -2.30e+00 |
Pos logit contributions | |
irens | +0.629 |
Paula | +0.624 |
Paul | +0.578 |
spe | +0.570 |
iren | +0.567 |
Neg logit contributions | |
per | -0.331 |
ppy | -0.318 |
pper | -0.307 |
like | -0.301 |
ily | -0.296 |
Token | with |
Token ID | 351 |
Feature activation | +3.01e-01 |
Pos logit contributions | |
per | +0.110 |
ppy | +0.100 |
ily | +0.098 |
pper | +0.096 |
like | +0.096 |
Neg logit contributions | |
Paula | -0.222 |
irens | -0.222 |
Paul | -0.206 |
iren | -0.205 |
spe | -0.200 |
Token | its |
Token ID | 663 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
irens | +0.026 |
Paula | +0.026 |
spe | +0.026 |
iren | +0.025 |
telev | +0.024 |
Neg logit contributions | |
ily | -0.016 |
per | -0.015 |
ppy | -0.015 |
ilies | -0.015 |
like | -0.015 |
Token | s |
Token ID | 264 |
Feature activation | +9.67e-01 |
Pos logit contributions | |
pper | +0.075 |
per | +0.075 |
ppy | +0.074 |
ilies | +0.074 |
opsy | +0.073 |
Neg logit contributions | |
Paula | -0.103 |
irens | -0.101 |
Paul | -0.093 |
James | -0.090 |
spe | -0.088 |
Token | irens |
Token ID | 42917 |
Feature activation | +1.34e+01 |
Pos logit contributions | |
Paula | +0.028 |
Paul | +0.022 |
James | +0.021 |
John | +0.020 |
Ellen | +0.019 |
Neg logit contributions | |
ppy | -0.115 |
per | -0.114 |
pper | -0.113 |
ilies | -0.110 |
ily | -0.109 |
Token | bl |
Token ID | 698 |
Feature activation | -5.47e-01 |
Pos logit contributions | |
Paula | +1.314 |
irens | +1.259 |
Paul | +1.204 |
Sarah | +1.183 |
Sarah | +1.180 |
Neg logit contributions | |
per | -0.637 |
ppy | -0.607 |
pper | -0.595 |
like | -0.570 |
ilies | -0.524 |
Token | aring |
Token ID | 1723 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
ilies | +0.036 |
opsy | +0.036 |
ppy | +0.035 |
ily | +0.035 |
ectar | +0.035 |
Neg logit contributions | |
irens | -0.041 |
Paula | -0.040 |
spe | -0.040 |
Paul | -0.040 |
flashing | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
Paula | +0.139 |
irens | +0.137 |
Paul | +0.127 |
spe | +0.123 |
iren | +0.121 |
Neg logit contributions | |
ppy | -0.102 |
ily | -0.101 |
per | -0.100 |
ilies | -0.097 |
like | -0.096 |
Token | The |
Token ID | 383 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
irens | +0.011 |
Paula | +0.011 |
flashing | +0.010 |
spe | +0.010 |
Paul | +0.010 |
Neg logit contributions | |
per | -0.011 |
ppy | -0.011 |
ilies | -0.010 |
ily | -0.010 |
opsy | -0.010 |
Token | loud |
Token ID | 7812 |
Feature activation | +3.58e+00 |
Pos logit contributions | |
per | +0.017 |
ily | +0.017 |
ilies | +0.016 |
ppy | +0.016 |
worm | +0.016 |
Neg logit contributions | |
Paula | -0.017 |
irens | -0.016 |
Paul | -0.015 |
John | -0.015 |
James | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
Paula | +0.323 |
irens | +0.306 |
Paul | +0.305 |
spe | +0.295 |
John | +0.286 |
Neg logit contributions | |
ppy | -0.172 |
per | -0.166 |
pper | -0.163 |
ilies | -0.160 |
ily | -0.157 |
Token | They |
Token ID | 1119 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
Paula | +0.010 |
irens | +0.010 |
Paul | +0.010 |
John | +0.009 |
spe | +0.009 |
Neg logit contributions | |
ppy | -0.013 |
pper | -0.013 |
opsy | -0.012 |
per | -0.012 |
ilies | -0.012 |
Token | heard |
Token ID | 2982 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
Paula | +0.111 |
irens | +0.107 |
Paul | +0.105 |
spe | +0.098 |
John | +0.097 |
Neg logit contributions | |
ppy | -0.096 |
per | -0.093 |
pper | -0.092 |
ilies | -0.092 |
opsy | -0.090 |
Token | the |
Token ID | 262 |
Feature activation | +8.70e-01 |
Pos logit contributions | |
Paula | +0.136 |
irens | +0.132 |
spe | +0.128 |
Paul | +0.127 |
Ellen | +0.121 |
Neg logit contributions | |
ilies | -0.108 |
ily | -0.107 |
per | -0.106 |
ppy | -0.105 |
opsy | -0.100 |
Token | s |
Token ID | 264 |
Feature activation | +9.53e-01 |
Pos logit contributions | |
Paula | +0.058 |
irens | +0.054 |
Paul | +0.053 |
Sarah | +0.050 |
spe | +0.050 |
Neg logit contributions | |
per | -0.066 |
ppy | -0.065 |
ilies | -0.065 |
pper | -0.064 |
ily | -0.064 |
Token | irens |
Token ID | 42917 |
Feature activation | +1.34e+01 |
Pos logit contributions | |
Paula | +0.025 |
Paul | +0.022 |
spe | +0.018 |
John | +0.017 |
Sarah | +0.016 |
Neg logit contributions | |
per | -0.119 |
ppy | -0.116 |
pper | -0.114 |
ily | -0.114 |
ilies | -0.112 |
Token | and |
Token ID | 290 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
Paula | +1.439 |
Paul | +1.367 |
spe | +1.331 |
irens | +1.324 |
Sarah | +1.312 |
Neg logit contributions | |
per | -0.558 |
ppy | -0.500 |
like | -0.487 |
pper | -0.471 |
ilies | -0.425 |
Token | the |
Token ID | 262 |
Feature activation | +8.75e-01 |
Pos logit contributions | |
Paula | +0.015 |
irens | +0.014 |
Paul | +0.014 |
spe | +0.013 |
parent | +0.013 |
Neg logit contributions | |
per | -0.010 |
like | -0.009 |
ppy | -0.009 |
careful | -0.009 |
worm | -0.009 |
Token | shouts |
Token ID | 34757 |
Feature activation | +4.97e+00 |
Pos logit contributions | |
Paula | +0.062 |
irens | +0.057 |
Paul | +0.057 |
Sarah | +0.054 |
Ellen | +0.053 |
Neg logit contributions | |
per | -0.063 |
ilies | -0.061 |
ppy | -0.061 |
worm | -0.060 |
ily | -0.060 |
Token | and |
Token ID | 290 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
Paula | +0.411 |
irens | +0.402 |
Paul | +0.393 |
John | +0.370 |
spe | +0.369 |
Neg logit contributions | |
ppy | -0.311 |
pper | -0.306 |
opsy | -0.304 |
ilies | -0.299 |
per | -0.293 |
Token | the |
Token ID | 262 |
Feature activation | +8.77e-01 |
Pos logit contributions | |
Paula | +0.015 |
irens | +0.015 |
Paul | +0.014 |
spe | +0.014 |
Smith | +0.013 |
Neg logit contributions | |
per | -0.011 |
ppy | -0.011 |
ilies | -0.011 |
like | -0.011 |
careful | -0.011 |
Token | some |
Token ID | 617 |
Feature activation | -3.53e+00 |
Pos logit contributions | |
ilies | +0.028 |
ily | +0.028 |
per | +0.026 |
like | +0.026 |
ppy | +0.026 |
Neg logit contributions | |
Paula | -0.064 |
irens | -0.063 |
spe | -0.060 |
iren | -0.059 |
Paul | -0.059 |
Token | flashing |
Token ID | 25293 |
Feature activation | +8.67e+00 |
Pos logit contributions | |
ilies | +0.213 |
ily | +0.204 |
per | +0.202 |
ppy | +0.200 |
worm | +0.196 |
Neg logit contributions | |
Paula | -0.297 |
Paul | -0.278 |
irens | -0.275 |
James | -0.262 |
John | -0.262 |
Token | lights |
Token ID | 7588 |
Feature activation | +5.24e+00 |
Pos logit contributions | |
Paula | +0.670 |
Paul | +0.603 |
Ellen | +0.600 |
Smith | +0.592 |
spe | +0.592 |
Neg logit contributions | |
like | -0.624 |
ily | -0.522 |
per | -0.514 |
at | -0.491 |
ilies | -0.490 |
Token | and |
Token ID | 290 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
Paula | +0.540 |
irens | +0.524 |
Paul | +0.511 |
spe | +0.504 |
Ellen | +0.496 |
Neg logit contributions | |
like | -0.215 |
per | -0.204 |
pper | -0.183 |
ppy | -0.181 |
ilies | -0.180 |
Token | s |
Token ID | 264 |
Feature activation | +9.02e-01 |
Pos logit contributions | |
Paula | +0.009 |
irens | +0.008 |
Paul | +0.008 |
Smith | +0.008 |
iren | +0.008 |
Neg logit contributions | |
like | -0.009 |
ilies | -0.008 |
ily | -0.008 |
per | -0.008 |
ppy | -0.008 |
Token | irens |
Token ID | 42917 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
Paula | +0.023 |
spe | +0.022 |
Paul | +0.018 |
Ellen | +0.018 |
James | +0.017 |
Neg logit contributions | |
per | -0.106 |
ppy | -0.106 |
pper | -0.102 |
ilies | -0.102 |
ars | -0.101 |
Token | . |
Token ID | 13 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
Paula | +1.271 |
irens | +1.201 |
Paul | +1.188 |
spe | +1.167 |
Sarah | +1.159 |
Neg logit contributions | |
per | -0.654 |
like | -0.613 |
ppy | -0.598 |
pper | -0.593 |
ilies | -0.563 |
Token | They |
Token ID | 1119 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
Paula | +0.008 |
irens | +0.008 |
Paul | +0.007 |
Sarah | +0.007 |
John | +0.007 |
Neg logit contributions | |
ppy | -0.009 |
pper | -0.009 |
opsy | -0.009 |
per | -0.009 |
ilies | -0.009 |
Token | were |
Token ID | 547 |
Feature activation | +5.21e-02 |
Pos logit contributions | |
Paula | +0.111 |
irens | +0.110 |
Paul | +0.104 |
iren | +0.099 |
James | +0.098 |
Neg logit contributions | |
ppy | -0.088 |
per | -0.084 |
pper | -0.083 |
ilies | -0.082 |
opsy | -0.079 |
Token | scared |
Token ID | 12008 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
Paula | +0.004 |
irens | +0.004 |
Paul | +0.004 |
spe | +0.004 |
iren | +0.004 |
Neg logit contributions | |
per | -0.003 |
ilies | -0.003 |
ppy | -0.003 |
pper | -0.003 |
ectar | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
Paula | +0.226 |
irens | +0.215 |
spe | +0.214 |
Paul | +0.206 |
Mechan | +0.206 |
Neg logit contributions | |
like | -0.087 |
ppy | -0.084 |
ily | -0.082 |
per | -0.080 |
ilies | -0.074 |
Token |
|
Token ID | 198 |
Feature activation | +3.39e-01 |
Pos logit contributions | |
ppy | +0.047 |
ilies | +0.045 |
per | +0.045 |
pper | +0.045 |
opsy | +0.044 |
Neg logit contributions | |
Paula | -0.048 |
irens | -0.045 |
Paul | -0.044 |
spe | -0.042 |
flashing | -0.041 |
Token |
|
Token ID | 198 |
Feature activation | +3.37e-01 |
Pos logit contributions | |
Paula | +0.024 |
irens | +0.024 |
Paul | +0.023 |
Sarah | +0.022 |
John | +0.022 |
Neg logit contributions | |
ppy | -0.025 |
pper | -0.024 |
ilies | -0.024 |
per | -0.024 |
opsy | -0.024 |
Token | They |
Token ID | 2990 |
Feature activation | +5.20e-02 |
Pos logit contributions | |
irens | +0.024 |
Paula | +0.023 |
Paul | +0.022 |
Sarah | +0.021 |
iren | +0.021 |
Neg logit contributions | |
ilies | -0.025 |
ppy | -0.025 |
per | -0.024 |
opsy | -0.024 |
pper | -0.024 |
Token | heard |
Token ID | 2982 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
Paula | +0.003 |
Paul | +0.003 |
Sarah | +0.003 |
John | +0.003 |
James | +0.003 |
Neg logit contributions | |
ectar | -0.005 |
ilies | -0.004 |
opsy | -0.004 |
izards | -0.004 |
per | -0.004 |
Token | s |
Token ID | 264 |
Feature activation | +9.03e-01 |
Pos logit contributions | |
Paula | +0.131 |
irens | +0.124 |
spe | +0.123 |
Paul | +0.115 |
iren | +0.112 |
Neg logit contributions | |
ily | -0.107 |
ilies | -0.106 |
ppy | -0.102 |
worm | -0.101 |
like | -0.100 |
Token | irens |
Token ID | 42917 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
Paula | +0.030 |
Paul | +0.026 |
spe | +0.023 |
John | +0.021 |
James | +0.020 |
Neg logit contributions | |
ppy | -0.105 |
per | -0.104 |
pper | -0.103 |
ilies | -0.102 |
ily | -0.102 |
Token | and |
Token ID | 290 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
Paula | +1.378 |
Paul | +1.295 |
irens | +1.269 |
spe | +1.268 |
Sarah | +1.221 |
Neg logit contributions | |
per | -0.580 |
ppy | -0.560 |
like | -0.542 |
pper | -0.516 |
ilies | -0.481 |
Token | saw |
Token ID | 2497 |
Feature activation | -6.26e-01 |
Pos logit contributions | |
Paula | +0.010 |
irens | +0.010 |
Paul | +0.009 |
parent | +0.009 |
iren | +0.009 |
Neg logit contributions | |
careful | -0.007 |
wait | -0.007 |
like | -0.007 |
ppy | -0.007 |
dove | -0.007 |
Token | lights |
Token ID | 7588 |
Feature activation | +5.24e+00 |
Pos logit contributions | |
ily | +0.034 |
like | +0.033 |
ilies | +0.033 |
worm | +0.031 |
per | +0.030 |
Neg logit contributions | |
Paula | -0.059 |
spe | -0.052 |
irens | -0.052 |
Paul | -0.051 |
Ellen | -0.050 |
Token | . |
Token ID | 13 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
Paula | +0.553 |
Paul | +0.515 |
irens | +0.512 |
Ellen | +0.504 |
spe | +0.503 |
Neg logit contributions | |
like | -0.229 |
per | -0.210 |
ppy | -0.182 |
pper | -0.178 |
ilies | -0.177 |
Token | They |
Token ID | 1119 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
Paula | +0.009 |
irens | +0.009 |
Paul | +0.008 |
spe | +0.008 |
flashing | +0.008 |
Neg logit contributions | |
ilies | -0.008 |
ily | -0.008 |
ppy | -0.008 |
per | -0.007 |
ectar | -0.007 |
Token | when |
Token ID | 618 |
Feature activation | -2.53e+00 |
Pos logit contributions | |
ppy | +0.161 |
opsy | +0.159 |
ilies | +0.159 |
pper | +0.158 |
ectar | +0.154 |
Neg logit contributions | |
Paula | -0.204 |
Paul | -0.197 |
irens | -0.193 |
John | -0.186 |
Sarah | -0.184 |
Token | he |
Token ID | 339 |
Feature activation | -2.60e-01 |
Pos logit contributions | |
ilies | +0.146 |
ily | +0.145 |
opsy | +0.141 |
like | +0.141 |
ppy | +0.140 |
Neg logit contributions | |
irens | -0.205 |
Paula | -0.201 |
iren | -0.188 |
Paul | -0.183 |
spe | -0.182 |
Token | heard |
Token ID | 2982 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
ppy | +0.018 |
pper | +0.018 |
ilies | +0.018 |
per | +0.018 |
opsy | +0.018 |
Neg logit contributions | |
Paula | -0.020 |
irens | -0.019 |
Paul | -0.019 |
spe | -0.017 |
John | -0.017 |
Token | loud |
Token ID | 7812 |
Feature activation | +3.54e+00 |
Pos logit contributions | |
Paula | +0.165 |
spe | +0.162 |
irens | +0.162 |
Smith | +0.154 |
iren | +0.153 |
Neg logit contributions | |
ily | -0.061 |
like | -0.061 |
ilies | -0.053 |
per | -0.052 |
dove | -0.047 |
Token | s |
Token ID | 264 |
Feature activation | +9.12e-01 |
Pos logit contributions | |
Paula | +0.225 |
spe | +0.222 |
irens | +0.205 |
Paul | +0.201 |
Sarah | +0.199 |
Neg logit contributions | |
like | -0.268 |
per | -0.265 |
ilies | -0.250 |
ppy | -0.247 |
ily | -0.245 |
Token | irens |
Token ID | 42917 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
Paula | +0.021 |
Paul | +0.019 |
spe | +0.017 |
John | +0.016 |
James | +0.015 |
Neg logit contributions | |
per | -0.111 |
ppy | -0.109 |
ilies | -0.107 |
pper | -0.107 |
ily | -0.105 |
Token | . |
Token ID | 13 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
Paula | +1.274 |
irens | +1.214 |
Paul | +1.193 |
spe | +1.193 |
Sarah | +1.172 |
Neg logit contributions | |
per | -0.635 |
like | -0.607 |
ppy | -0.591 |
pper | -0.577 |
ilies | -0.522 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | -7.63e-01 |
Pos logit contributions | |
irens | +0.010 |
bul | +0.010 |
spe | +0.010 |
iren | +0.009 |
Paula | +0.009 |
Neg logit contributions | |
ily | -0.006 |
like | -0.005 |
per | -0.005 |
both | -0.005 |
middle | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | -3.86e-01 |
Pos logit contributions | |
ppy | +0.030 |
per | +0.029 |
ily | +0.029 |
pper | +0.028 |
like | +0.028 |
Neg logit contributions | |
irens | -0.074 |
Paula | -0.074 |
spe | -0.071 |
iren | -0.069 |
Paul | -0.068 |
Token | an |
Token ID | 281 |
Feature activation | -3.25e-01 |
Pos logit contributions | |
ily | +0.022 |
like | +0.021 |
wait | +0.021 |
ppy | +0.020 |
careful | +0.020 |
Neg logit contributions | |
irens | -0.031 |
iren | -0.029 |
Paula | -0.029 |
spe | -0.029 |
parent | -0.028 |
Token | ambulance |
Token ID | 22536 |
Feature activation | +2.44e+00 |
Pos logit contributions | |
per | +0.024 |
ily | +0.024 |
ppy | +0.023 |
ilies | +0.023 |
pper | +0.023 |
Neg logit contributions | |
Paula | -0.022 |
spe | -0.020 |
irens | -0.020 |
Paul | -0.020 |
Smith | -0.019 |
Token | car |
Token ID | 1097 |
Feature activation | +2.72e+00 |
Pos logit contributions | |
Paula | +0.427 |
irens | +0.396 |
Paul | +0.378 |
Sarah | +0.370 |
spe | +0.369 |
Neg logit contributions | |
per | -0.383 |
pper | -0.360 |
like | -0.357 |
ppy | -0.352 |
ilies | -0.346 |
Token | . |
Token ID | 13 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
Paula | +0.285 |
Paul | +0.260 |
irens | +0.256 |
Sarah | +0.254 |
Ellen | +0.252 |
Neg logit contributions | |
per | -0.126 |
pper | -0.114 |
like | -0.113 |
ppy | -0.101 |
ily | -0.098 |
Token | They |
Token ID | 1119 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
Paula | +0.008 |
irens | +0.008 |
Paul | +0.008 |
spe | +0.007 |
John | +0.007 |
Neg logit contributions | |
ppy | -0.008 |
ilies | -0.007 |
per | -0.007 |
pper | -0.007 |
opsy | -0.007 |
Token | had |
Token ID | 550 |
Feature activation | -7.41e-03 |
Pos logit contributions | |
Paula | +0.116 |
irens | +0.115 |
Paul | +0.107 |
iren | +0.103 |
Ellen | +0.103 |
Neg logit contributions | |
ppy | -0.083 |
per | -0.080 |
pper | -0.079 |
ilies | -0.076 |
ily | -0.074 |
Token | s |
Token ID | 264 |
Feature activation | +9.03e-01 |
Pos logit contributions | |
careful | +0.000 |
per | +0.000 |
like | +0.000 |
ppy | +0.000 |
pper | +0.000 |
Neg logit contributions | |
Paula | -0.001 |
irens | -0.001 |
Paul | -0.001 |
spe | -0.001 |
iren | -0.001 |
Token | irens |
Token ID | 42917 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
Paula | +0.020 |
Paul | +0.015 |
spe | +0.014 |
James | +0.012 |
John | +0.011 |
Neg logit contributions | |
ppy | -0.114 |
per | -0.113 |
pper | -0.112 |
ilies | -0.111 |
ily | -0.109 |
Token | and |
Token ID | 290 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
Paula | +1.431 |
irens | +1.330 |
Paul | +1.319 |
Sarah | +1.301 |
Sarah | +1.290 |
Neg logit contributions | |
per | -0.535 |
like | -0.498 |
pper | -0.470 |
ppy | -0.464 |
ilies | -0.407 |
Token | lights |
Token ID | 7588 |
Feature activation | +5.24e+00 |
Pos logit contributions | |
Paula | +0.010 |
irens | +0.010 |
Paul | +0.009 |
spe | +0.009 |
Ellen | +0.009 |
Neg logit contributions | |
per | -0.008 |
ppy | -0.008 |
careful | -0.008 |
like | -0.008 |
ilies | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
Paula | +0.551 |
irens | +0.525 |
Paul | +0.507 |
Ellen | +0.504 |
Sarah | +0.496 |
Neg logit contributions | |
like | -0.217 |
per | -0.206 |
pper | -0.189 |
ilies | -0.185 |
ppy | -0.183 |
Token | They |
Token ID | 1119 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
Paula | +0.008 |
irens | +0.008 |
Paul | +0.008 |
spe | +0.007 |
John | +0.007 |
Neg logit contributions | |
ppy | -0.009 |
pper | -0.009 |
ilies | -0.009 |
per | -0.009 |
opsy | -0.008 |
Token | were |
Token ID | 547 |
Feature activation | +4.86e-02 |
Pos logit contributions | |
Paula | +0.117 |
irens | +0.116 |
Paul | +0.107 |
iren | +0.103 |
Sarah | +0.102 |
Neg logit contributions | |
ppy | -0.085 |
per | -0.082 |
pper | -0.081 |
ilies | -0.078 |
like | -0.076 |
Token | a |
Token ID | 257 |
Feature activation | -7.72e-01 |
Pos logit contributions | |
Paula | +0.155 |
irens | +0.150 |
spe | +0.146 |
Paul | +0.144 |
iren | +0.137 |
Neg logit contributions | |
like | -0.068 |
ily | -0.065 |
ppy | -0.065 |
per | -0.065 |
ilies | -0.062 |
Token | fire |
Token ID | 2046 |
Feature activation | +2.73e-01 |
Pos logit contributions | |
worm | +0.047 |
opsy | +0.046 |
ppy | +0.045 |
pper | +0.043 |
dove | +0.043 |
Neg logit contributions | |
Paula | -0.066 |
irens | -0.065 |
Paul | -0.061 |
spe | -0.057 |
John | -0.057 |
Token | truck |
Token ID | 7779 |
Feature activation | +3.36e+00 |
Pos logit contributions | |
Paula | +0.021 |
irens | +0.019 |
Paul | +0.018 |
Ellen | +0.017 |
spe | +0.017 |
Neg logit contributions | |
pper | -0.020 |
per | -0.019 |
ppy | -0.019 |
opsy | -0.018 |
like | -0.018 |
Token | with |
Token ID | 351 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
Paula | +0.327 |
irens | +0.311 |
telev | +0.297 |
spe | +0.288 |
Paul | +0.286 |
Neg logit contributions | |
pper | -0.168 |
like | -0.167 |
per | -0.163 |
ppy | -0.153 |
wait | -0.139 |
Token | s |
Token ID | 264 |
Feature activation | +9.03e-01 |
Pos logit contributions | |
Paula | +0.022 |
irens | +0.021 |
Paul | +0.020 |
spe | +0.020 |
Ellen | +0.020 |
Neg logit contributions | |
ppy | -0.010 |
per | -0.010 |
ilies | -0.010 |
ily | -0.010 |
worm | -0.010 |
Token | irens |
Token ID | 42917 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
Paula | +0.027 |
Paul | +0.023 |
spe | +0.020 |
James | +0.019 |
John | +0.019 |
Neg logit contributions | |
ppy | -0.107 |
per | -0.105 |
pper | -0.105 |
ilies | -0.103 |
opsy | -0.102 |
Token | that |
Token ID | 326 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
Paula | +1.336 |
spe | +1.252 |
Paul | +1.235 |
irens | +1.234 |
Sarah | +1.203 |
Neg logit contributions | |
per | -0.617 |
pper | -0.575 |
ppy | -0.573 |
like | -0.559 |
ilies | -0.513 |
Token | went |
Token ID | 1816 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
Paula | +0.103 |
irens | +0.098 |
Paul | +0.093 |
spe | +0.091 |
Ellen | +0.089 |
Neg logit contributions | |
ppy | -0.069 |
pper | -0.067 |
per | -0.067 |
ilies | -0.066 |
ily | -0.063 |
Token | " |
Token ID | 366 |
Feature activation | -6.57e-01 |
Pos logit contributions | |
Paula | +0.205 |
irens | +0.189 |
parent | +0.185 |
Paul | +0.184 |
James | +0.183 |
Neg logit contributions | |
like | -0.110 |
per | -0.106 |
ppy | -0.095 |
pper | -0.091 |
careful | -0.087 |
Token | we |
Token ID | 732 |
Feature activation | +3.79e+00 |
Pos logit contributions | |
ppy | +0.044 |
pper | +0.043 |
per | +0.041 |
ilies | +0.040 |
ily | +0.040 |
Neg logit contributions | |
Paula | -0.052 |
Paul | -0.049 |
irens | -0.049 |
James | -0.046 |
Sarah | -0.046 |
Token | e |
Token ID | 68 |
Feature activation | -4.14e+00 |
Pos logit contributions | |
Paula | +0.345 |
irens | +0.337 |
Paul | +0.332 |
John | +0.316 |
spe | +0.313 |
Neg logit contributions | |
opsy | -0.201 |
per | -0.195 |
ppy | -0.194 |
ilies | -0.194 |
pper | -0.193 |
Token | flashing |
Token ID | 25293 |
Feature activation | +8.65e+00 |
Pos logit contributions | |
Paula | +0.020 |
irens | +0.020 |
spe | +0.019 |
Smith | +0.019 |
Paul | +0.019 |
Neg logit contributions | |
ily | -0.008 |
like | -0.008 |
careful | -0.008 |
per | -0.007 |
ilies | -0.007 |
Token | lights |
Token ID | 7588 |
Feature activation | +5.22e+00 |
Pos logit contributions | |
Paula | +0.634 |
spe | +0.581 |
James | +0.579 |
Paul | +0.565 |
Smith | +0.559 |
Neg logit contributions | |
like | -0.625 |
ily | -0.583 |
per | -0.562 |
aw | -0.561 |
feathers | -0.543 |
Token | and |
Token ID | 290 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
Paula | +0.546 |
irens | +0.519 |
Paul | +0.516 |
spe | +0.510 |
James | +0.508 |
Neg logit contributions | |
like | -0.211 |
per | -0.199 |
pper | -0.177 |
ppy | -0.173 |
ilies | -0.167 |
Token | loud |
Token ID | 7812 |
Feature activation | +3.52e+00 |
Pos logit contributions | |
Paula | +0.009 |
irens | +0.008 |
spe | +0.008 |
Paul | +0.008 |
Smith | +0.008 |
Neg logit contributions | |
like | -0.006 |
ily | -0.006 |
ilies | -0.006 |
careful | -0.006 |
worm | -0.006 |
Token | s |
Token ID | 264 |
Feature activation | +8.79e-01 |
Pos logit contributions | |
Paula | +0.235 |
spe | +0.216 |
Paul | +0.204 |
irens | +0.202 |
Sarah | +0.202 |
Neg logit contributions | |
like | -0.264 |
per | -0.254 |
ily | -0.238 |
feathers | -0.235 |
ilies | -0.235 |
Token | irens |
Token ID | 42917 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
Paula | +0.023 |
Paul | +0.020 |
spe | +0.019 |
James | +0.018 |
John | +0.017 |
Neg logit contributions | |
per | -0.105 |
ppy | -0.104 |
pper | -0.103 |
ilies | -0.103 |
ily | -0.101 |
Token | . |
Token ID | 13 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
Paula | +1.315 |
Paul | +1.230 |
irens | +1.221 |
Sarah | +1.206 |
spe | +1.202 |
Neg logit contributions | |
per | -0.621 |
like | -0.564 |
ppy | -0.562 |
pper | -0.555 |
ilies | -0.524 |
Token | She |
Token ID | 1375 |
Feature activation | -8.06e-01 |
Pos logit contributions | |
Paula | +0.009 |
irens | +0.009 |
Paul | +0.008 |
spe | +0.008 |
flashing | +0.008 |
Neg logit contributions | |
ppy | -0.007 |
per | -0.007 |
ilies | -0.006 |
opsy | -0.006 |
pper | -0.006 |
Token | asked |
Token ID | 1965 |
Feature activation | +6.20e-01 |
Pos logit contributions | |
ppy | +0.051 |
per | +0.048 |
pper | +0.048 |
opsy | +0.047 |
ilies | +0.046 |
Neg logit contributions | |
irens | -0.066 |
Paula | -0.066 |
Paul | -0.062 |
spe | -0.059 |
iren | -0.059 |
Token | her |
Token ID | 607 |
Feature activation | -4.67e-01 |
Pos logit contributions | |
Paula | +0.052 |
irens | +0.051 |
Paul | +0.048 |
spe | +0.048 |
iren | +0.046 |
Neg logit contributions | |
ppy | -0.038 |
per | -0.037 |
ilies | -0.036 |
pper | -0.036 |
opsy | -0.036 |
Token | mom |
Token ID | 1995 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
ppy | +0.025 |
opsy | +0.024 |
ily | +0.023 |
ilies | +0.023 |
ectar | +0.023 |
Neg logit contributions | |
irens | -0.043 |
Paula | -0.042 |
Paul | -0.039 |
iren | -0.039 |
spe | -0.039 |
Token | did |
Token ID | 750 |
Feature activation | -3.77e+00 |
Pos logit contributions | |
Paula | +0.098 |
Paul | +0.093 |
irens | +0.091 |
spe | +0.087 |
John | +0.087 |
Neg logit contributions | |
ectar | -0.093 |
ppy | -0.093 |
opsy | -0.092 |
ilies | -0.091 |
pper | -0.091 |
Token | not |
Token ID | 407 |
Feature activation | -5.31e-01 |
Pos logit contributions | |
ppy | +0.165 |
ilies | +0.159 |
pper | +0.158 |
per | +0.158 |
opsy | +0.154 |
Neg logit contributions | |
Paula | -0.384 |
irens | -0.376 |
Paul | -0.368 |
spe | -0.352 |
John | -0.347 |
Token | hear |
Token ID | 3285 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
opsy | +0.031 |
ilies | +0.031 |
ectar | +0.030 |
pper | +0.030 |
ppy | +0.030 |
Neg logit contributions | |
Paula | -0.046 |
irens | -0.043 |
Paul | -0.043 |
Sarah | -0.041 |
John | -0.041 |
Token | the |
Token ID | 262 |
Feature activation | +7.81e-01 |
Pos logit contributions | |
Paula | +0.140 |
irens | +0.137 |
spe | +0.135 |
Paul | +0.132 |
iren | +0.126 |
Neg logit contributions | |
per | -0.106 |
ilies | -0.105 |
ily | -0.103 |
ppy | -0.103 |
like | -0.102 |
Token | s |
Token ID | 264 |
Feature activation | +8.57e-01 |
Pos logit contributions | |
Paula | +0.052 |
irens | +0.050 |
Paul | +0.049 |
spe | +0.046 |
John | +0.045 |
Neg logit contributions | |
ppy | -0.062 |
pper | -0.060 |
ilies | -0.060 |
per | -0.060 |
opsy | -0.059 |
Token | irens |
Token ID | 42917 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
Paula | +0.027 |
Paul | +0.026 |
John | +0.023 |
spe | +0.022 |
James | +0.022 |
Neg logit contributions | |
per | -0.100 |
ppy | -0.097 |
ily | -0.096 |
pper | -0.096 |
ilies | -0.096 |
Token | of |
Token ID | 286 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
Paula | +1.333 |
Paul | +1.263 |
irens | +1.252 |
spe | +1.227 |
Sarah | +1.203 |
Neg logit contributions | |
per | -0.643 |
ppy | -0.612 |
pper | -0.589 |
like | -0.572 |
ilies | -0.521 |
Token | the |
Token ID | 262 |
Feature activation | +7.76e-01 |
Pos logit contributions | |
Paula | +0.104 |
irens | +0.101 |
spe | +0.097 |
Paul | +0.095 |
iren | +0.091 |
Neg logit contributions | |
per | -0.083 |
ilies | -0.082 |
ppy | -0.080 |
ily | -0.079 |
pper | -0.078 |
Token | ambulance |
Token ID | 22536 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
Paula | +0.055 |
irens | +0.053 |
Paul | +0.051 |
spe | +0.048 |
John | +0.047 |
Neg logit contributions | |
ppy | -0.059 |
pper | -0.057 |
ilies | -0.057 |
per | -0.057 |
opsy | -0.057 |
Token | that |
Token ID | 326 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
Paula | +0.230 |
irens | +0.218 |
Paul | +0.217 |
Ellen | +0.207 |
Sarah | +0.207 |
Neg logit contributions | |
per | -0.119 |
pper | -0.113 |
ppy | -0.110 |
like | -0.106 |
wait | -0.102 |
Token | came |
Token ID | 1625 |
Feature activation | +3.88e+00 |
Pos logit contributions | |
Paula | +0.096 |
irens | +0.093 |
Paul | +0.087 |
spe | +0.085 |
Ellen | +0.083 |
Neg logit contributions | |
ppy | -0.071 |
ilies | -0.070 |
ily | -0.069 |
per | -0.068 |
pper | -0.068 |
Token | , |
Token ID | 11 |
Feature activation | -4.56e-01 |
Pos logit contributions | |
Paula | +0.065 |
irens | +0.063 |
spe | +0.061 |
Paul | +0.061 |
flashing | +0.059 |
Neg logit contributions | |
ppy | -0.028 |
ilies | -0.027 |
ily | -0.027 |
per | -0.026 |
pper | -0.025 |
Token | they |
Token ID | 484 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
ppy | +0.033 |
ily | +0.032 |
ilies | +0.032 |
opsy | +0.031 |
per | +0.030 |
Neg logit contributions | |
irens | -0.032 |
spe | -0.030 |
Paula | -0.030 |
iren | -0.030 |
flashing | -0.026 |
Token | both |
Token ID | 1111 |
Feature activation | -4.83e+00 |
Pos logit contributions | |
Paula | +0.158 |
Paul | +0.153 |
Sarah | +0.147 |
John | +0.144 |
spe | +0.142 |
Neg logit contributions | |
ectar | -0.173 |
opsy | -0.169 |
ilies | -0.168 |
pper | -0.167 |
per | -0.162 |
Token | heard |
Token ID | 2982 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
pper | +0.338 |
ppy | +0.337 |
opsy | +0.335 |
ilies | +0.335 |
per | +0.332 |
Neg logit contributions | |
Paula | -0.359 |
Paul | -0.341 |
irens | -0.335 |
spe | -0.326 |
John | -0.322 |
Token | s |
Token ID | 264 |
Feature activation | +8.45e-01 |
Pos logit contributions | |
Paula | +0.175 |
irens | +0.173 |
spe | +0.171 |
iren | +0.166 |
parent | +0.163 |
Neg logit contributions | |
ily | -0.048 |
like | -0.044 |
ilies | -0.038 |
dove | -0.034 |
per | -0.032 |
Token | irens |
Token ID | 42917 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
Paula | +0.031 |
Paul | +0.029 |
spe | +0.027 |
John | +0.025 |
James | +0.024 |
Neg logit contributions | |
ppy | -0.092 |
per | -0.092 |
pper | -0.090 |
ilies | -0.089 |
opsy | -0.088 |
Token | . |
Token ID | 13 |
Feature activation | +5.51e-02 |
Pos logit contributions | |
Paula | +1.160 |
spe | +1.088 |
irens | +1.075 |
Paul | +1.068 |
parent | +1.058 |
Neg logit contributions | |
like | -0.758 |
per | -0.715 |
ppy | -0.648 |
pper | -0.638 |
wait | -0.626 |
Token | The |
Token ID | 383 |
Feature activation | -3.40e-01 |
Pos logit contributions | |
spe | +0.004 |
irens | +0.004 |
Paula | +0.004 |
iren | +0.004 |
bul | +0.004 |
Neg logit contributions | |
ppy | -0.003 |
per | -0.003 |
ilies | -0.003 |
ily | -0.003 |
opsy | -0.003 |
Token | police |
Token ID | 1644 |
Feature activation | +5.50e+00 |
Pos logit contributions | |
careful | +0.020 |
ily | +0.019 |
like | +0.019 |
ars | +0.019 |
wait | +0.019 |
Neg logit contributions | |
Paula | -0.029 |
James | -0.026 |
Paul | -0.025 |
John | -0.025 |
Ellen | -0.025 |
Token | had |
Token ID | 550 |
Feature activation | -6.63e-02 |
Pos logit contributions | |
igsaw | +0.460 |
irens | +0.443 |
Paula | +0.430 |
parent | +0.417 |
telev | +0.409 |
Neg logit contributions | |
ppy | -0.296 |
like | -0.295 |
per | -0.289 |
pper | -0.277 |
still | -0.265 |
Token | come |
Token ID | 1282 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
careful | +0.003 |
like | +0.003 |
per | +0.003 |
ily | +0.003 |
still | +0.003 |
Neg logit contributions | |
Paula | -0.006 |
irens | -0.006 |
iren | -0.006 |
igsaw | -0.006 |
parent | -0.006 |
Token | Tim |
Token ID | 5045 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
Paula | +0.004 |
irens | +0.004 |
Paul | +0.004 |
spe | +0.004 |
flashing | +0.004 |
Neg logit contributions | |
ppy | -0.004 |
ilies | -0.004 |
per | -0.004 |
ily | -0.004 |
ectar | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | +6.39e-02 |
Pos logit contributions | |
Paula | +0.149 |
irens | +0.145 |
Paul | +0.142 |
John | +0.135 |
Sarah | +0.134 |
Neg logit contributions | |
ilies | -0.112 |
ppy | -0.111 |
opsy | -0.111 |
pper | -0.109 |
ectar | -0.108 |
Token | Sam |
Token ID | 3409 |
Feature activation | +5.62e-01 |
Pos logit contributions | |
Paula | +0.005 |
irens | +0.005 |
Paul | +0.005 |
spe | +0.005 |
flashing | +0.004 |
Neg logit contributions | |
ppy | -0.004 |
ilies | -0.004 |
per | -0.004 |
ily | -0.004 |
pper | -0.004 |
Token | heard |
Token ID | 2982 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
Paula | +0.033 |
flashing | +0.031 |
Paul | +0.031 |
John | +0.030 |
irens | +0.030 |
Neg logit contributions | |
ectar | -0.049 |
opsy | -0.047 |
ilies | -0.046 |
pper | -0.044 |
oll | -0.044 |
Token | s |
Token ID | 264 |
Feature activation | +8.55e-01 |
Pos logit contributions | |
Paula | +0.145 |
irens | +0.140 |
spe | +0.136 |
Paul | +0.130 |
iren | +0.129 |
Neg logit contributions | |
ily | -0.084 |
ilies | -0.082 |
like | -0.079 |
per | -0.078 |
worm | -0.078 |
Token | irens |
Token ID | 42917 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
Paula | +0.016 |
Paul | +0.013 |
spe | +0.010 |
John | +0.009 |
James | +0.008 |
Neg logit contributions | |
ppy | -0.112 |
per | -0.111 |
pper | -0.109 |
ilies | -0.109 |
ily | -0.108 |
Token | and |
Token ID | 290 |
Feature activation | +6.17e-02 |
Pos logit contributions | |
Paula | +1.338 |
Paul | +1.268 |
irens | +1.243 |
spe | +1.230 |
Sarah | +1.216 |
Neg logit contributions | |
per | -0.624 |
ppy | -0.587 |
like | -0.558 |
pper | -0.558 |
ilies | -0.521 |
Token | saw |
Token ID | 2497 |
Feature activation | -6.79e-01 |
Pos logit contributions | |
Paula | +0.005 |
irens | +0.005 |
Paul | +0.005 |
spe | +0.004 |
Ellen | +0.004 |
Neg logit contributions | |
ppy | -0.004 |
per | -0.004 |
ilies | -0.004 |
worm | -0.004 |
ily | -0.004 |
Token | fire |
Token ID | 2046 |
Feature activation | +2.38e-01 |
Pos logit contributions | |
ily | +0.039 |
ilies | +0.038 |
like | +0.038 |
per | +0.037 |
worm | +0.036 |
Neg logit contributions | |
Paula | -0.060 |
irens | -0.054 |
Paul | -0.054 |
Ellen | -0.054 |
spe | -0.053 |
Token | trucks |
Token ID | 13960 |
Feature activation | +6.09e+00 |
Pos logit contributions | |
Paula | +0.022 |
Paul | +0.020 |
Ellen | +0.020 |
Sarah | +0.020 |
Sarah | +0.020 |
Neg logit contributions | |
per | -0.013 |
pper | -0.013 |
ppy | -0.012 |
like | -0.012 |
ilies | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | +7.01e-02 |
Pos logit contributions | |
Paula | +0.598 |
Paul | +0.560 |
irens | +0.552 |
Sarah | +0.534 |
Ellen | +0.531 |
Neg logit contributions | |
per | -0.309 |
pper | -0.293 |
ppy | -0.292 |
like | -0.280 |
ilies | -0.277 |
Token | scared |
Token ID | 12008 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
ppy | +0.171 |
per | +0.167 |
pper | +0.166 |
ilies | +0.164 |
opsy | +0.161 |
Neg logit contributions | |
Paula | -0.230 |
irens | -0.223 |
Paul | -0.217 |
spe | -0.206 |
Sarah | -0.202 |
Token | . |
Token ID | 13 |
Feature activation | +5.63e-02 |
Pos logit contributions | |
Paula | +0.204 |
irens | +0.196 |
Paul | +0.189 |
spe | +0.187 |
Ellen | +0.179 |
Neg logit contributions | |
ppy | -0.101 |
per | -0.098 |
ily | -0.094 |
ilies | -0.092 |
pper | -0.091 |
Token | They |
Token ID | 1119 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
Paula | +0.004 |
irens | +0.004 |
Paul | +0.004 |
spe | +0.004 |
Ellen | +0.004 |
Neg logit contributions | |
ilies | -0.004 |
per | -0.004 |
ppy | -0.004 |
ily | -0.004 |
ectar | -0.004 |
Token | hear |
Token ID | 3285 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
Paula | +0.095 |
Paul | +0.092 |
flashing | +0.088 |
irens | +0.087 |
John | +0.087 |
Neg logit contributions | |
ectar | -0.098 |
opsy | -0.092 |
ilies | -0.091 |
ppy | -0.090 |
pper | -0.089 |
Token | s |
Token ID | 264 |
Feature activation | +8.40e-01 |
Pos logit contributions | |
Paula | +0.136 |
irens | +0.130 |
Paul | +0.125 |
spe | +0.123 |
Ellen | +0.117 |
Neg logit contributions | |
ppy | -0.112 |
per | -0.111 |
ilies | -0.110 |
ily | -0.107 |
pper | -0.107 |
Token | irens |
Token ID | 42917 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
Paula | +0.015 |
Paul | +0.011 |
spe | +0.010 |
John | +0.007 |
Sarah | +0.007 |
Neg logit contributions | |
per | -0.110 |
ppy | -0.109 |
pper | -0.107 |
ilies | -0.106 |
ily | -0.105 |
Token | . |
Token ID | 13 |
Feature activation | +6.53e-02 |
Pos logit contributions | |
Paula | +1.333 |
Paul | +1.258 |
irens | +1.249 |
spe | +1.239 |
Sarah | +1.213 |
Neg logit contributions | |
per | -0.627 |
ppy | -0.591 |
like | -0.560 |
pper | -0.556 |
ilies | -0.503 |
Token | The |
Token ID | 383 |
Feature activation | -3.41e-01 |
Pos logit contributions | |
Paula | +0.005 |
irens | +0.005 |
Paul | +0.005 |
spe | +0.004 |
James | +0.004 |
Neg logit contributions | |
ilies | -0.004 |
ppy | -0.004 |
per | -0.004 |
opsy | -0.004 |
ily | -0.004 |
Token | police |
Token ID | 1644 |
Feature activation | +5.50e+00 |
Pos logit contributions | |
ily | +0.021 |
per | +0.021 |
wait | +0.021 |
ars | +0.021 |
careful | +0.021 |
Neg logit contributions | |
Paula | -0.027 |
Paul | -0.024 |
irens | -0.023 |
spe | -0.023 |
James | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | +6.39e-02 |
Pos logit contributions | |
Paula | +0.454 |
irens | +0.431 |
spe | +0.430 |
Paul | +0.422 |
Sarah | +0.406 |
Neg logit contributions | |
per | -0.348 |
ppy | -0.335 |
pper | -0.327 |
ilies | -0.313 |
like | -0.306 |
Token | the |
Token ID | 262 |
Feature activation | +7.54e-01 |
Pos logit contributions | |
Paula | +0.005 |
irens | +0.005 |
Paul | +0.005 |
spe | +0.004 |
James | +0.004 |
Neg logit contributions | |
ppy | -0.004 |
ilies | -0.004 |
per | -0.004 |
pper | -0.004 |
ily | -0.004 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | -8.45e-01 |
Pos logit contributions | |
Paula | +0.004 |
irens | +0.004 |
Paul | +0.004 |
spe | +0.004 |
flashing | +0.004 |
Neg logit contributions | |
ppy | -0.003 |
opsy | -0.003 |
ilies | -0.003 |
per | -0.003 |
ectar | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | -4.56e-01 |
Pos logit contributions | |
ppy | +0.054 |
pper | +0.053 |
per | +0.052 |
ilies | +0.052 |
opsy | +0.051 |
Neg logit contributions | |
Paula | -0.071 |
irens | -0.068 |
Paul | -0.068 |
Sarah | -0.064 |
John | -0.064 |
Token | they |
Token ID | 484 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
ppy | +0.033 |
pper | +0.033 |
per | +0.032 |
ilies | +0.032 |
opsy | +0.031 |
Neg logit contributions | |
Paula | -0.035 |
irens | -0.033 |
Paul | -0.033 |
Sarah | -0.031 |
John | -0.031 |
Token | heard |
Token ID | 2982 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
Paula | +0.154 |
irens | +0.149 |
Paul | +0.145 |
Sarah | +0.135 |
John | +0.134 |
Neg logit contributions | |
ppy | -0.179 |
pper | -0.177 |
ilies | -0.177 |
per | -0.176 |
opsy | -0.176 |
Token | s |
Token ID | 264 |
Feature activation | +8.40e-01 |
Pos logit contributions | |
spe | +0.171 |
Paula | +0.170 |
irens | +0.167 |
Paul | +0.159 |
iren | +0.159 |
Neg logit contributions | |
ily | -0.055 |
ilies | -0.045 |
like | -0.044 |
ppy | -0.044 |
per | -0.040 |
Token | irens |
Token ID | 42917 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
Paula | +0.019 |
Paul | +0.016 |
spe | +0.013 |
irens | +0.013 |
John | +0.012 |
Neg logit contributions | |
ppy | -0.104 |
per | -0.103 |
pper | -0.102 |
ilies | -0.102 |
opsy | -0.101 |
Token | and |
Token ID | 290 |
Feature activation | +7.27e-02 |
Pos logit contributions | |
Paula | +1.210 |
spe | +1.149 |
Paul | +1.122 |
irens | +1.122 |
Sarah | +1.095 |
Neg logit contributions | |
per | -0.690 |
like | -0.688 |
ppy | -0.641 |
pper | -0.630 |
wait | -0.573 |
Token | a |
Token ID | 257 |
Feature activation | -8.19e-01 |
Pos logit contributions | |
irens | +0.006 |
Paula | +0.006 |
igsaw | +0.006 |
parent | +0.006 |
channel | +0.005 |
Neg logit contributions | |
like | -0.004 |
dove | -0.004 |
wait | -0.004 |
ppy | -0.004 |
careful | -0.004 |
Token | cop |
Token ID | 2243 |
Feature activation | +6.46e+00 |
Pos logit contributions | |
worm | +0.045 |
dove | +0.044 |
careful | +0.041 |
opsy | +0.041 |
ppy | +0.040 |
Neg logit contributions | |
Paula | -0.068 |
irens | -0.067 |
Paul | -0.063 |
John | -0.062 |
James | -0.062 |
Token | car |
Token ID | 1097 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
irens | +0.539 |
Paula | +0.524 |
lights | +0.502 |
igsaw | +0.473 |
telev | +0.471 |
Neg logit contributions | |
per | -0.362 |
like | -0.357 |
wait | -0.347 |
ppy | -0.342 |
pper | -0.341 |
Token | pulled |
Token ID | 5954 |
Feature activation | +2.74e+00 |
Pos logit contributions | |
Paula | +0.251 |
igsaw | +0.238 |
James | +0.228 |
telev | +0.225 |
parent | +0.224 |
Neg logit contributions | |
like | -0.137 |
per | -0.127 |
wait | -0.121 |
pper | -0.119 |
at | -0.112 |
Token | purse |
Token ID | 27517 |
Feature activation | +1.94e+00 |
Pos logit contributions | |
Paula | +0.057 |
irens | +0.055 |
Paul | +0.054 |
John | +0.050 |
James | +0.050 |
Neg logit contributions | |
ppy | -0.049 |
ilies | -0.049 |
per | -0.048 |
pper | -0.048 |
ily | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | +5.61e-02 |
Pos logit contributions | |
Paula | +0.172 |
irens | +0.167 |
Paul | +0.163 |
spe | +0.154 |
John | +0.153 |
Neg logit contributions | |
ppy | -0.112 |
pper | -0.108 |
per | -0.107 |
ilies | -0.107 |
opsy | -0.105 |
Token | She |
Token ID | 1375 |
Feature activation | -8.63e-01 |
Pos logit contributions | |
Paula | +0.004 |
irens | +0.004 |
Paul | +0.004 |
spe | +0.004 |
John | +0.004 |
Neg logit contributions | |
ppy | -0.004 |
pper | -0.004 |
ilies | -0.004 |
per | -0.004 |
opsy | -0.004 |
Token | heard |
Token ID | 2982 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
ppy | +0.062 |
pper | +0.060 |
ilies | +0.060 |
opsy | +0.060 |
per | +0.060 |
Neg logit contributions | |
Paula | -0.064 |
irens | -0.062 |
Paul | -0.060 |
John | -0.056 |
Sarah | -0.056 |
Token | s |
Token ID | 264 |
Feature activation | +8.32e-01 |
Pos logit contributions | |
Paula | +0.139 |
spe | +0.137 |
irens | +0.137 |
Paul | +0.130 |
iren | +0.125 |
Neg logit contributions | |
ily | -0.084 |
ilies | -0.081 |
ppy | -0.080 |
per | -0.079 |
ectar | -0.074 |
Token | irens |
Token ID | 42917 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
Paula | +0.018 |
Paul | +0.016 |
spe | +0.014 |
John | +0.012 |
James | +0.011 |
Neg logit contributions | |
ppy | -0.104 |
per | -0.104 |
pper | -0.102 |
ily | -0.102 |
ilies | -0.101 |
Token | outside |
Token ID | 2354 |
Feature activation | +5.33e-01 |
Pos logit contributions | |
Paula | +1.329 |
Paul | +1.272 |
spe | +1.248 |
irens | +1.227 |
Sarah | +1.210 |
Neg logit contributions | |
per | -0.627 |
ppy | -0.583 |
like | -0.553 |
pper | -0.543 |
ilies | -0.488 |
Token | . |
Token ID | 13 |
Feature activation | +4.40e-02 |
Pos logit contributions | |
Paula | +0.057 |
irens | +0.054 |
Paul | +0.054 |
Ellen | +0.052 |
spe | +0.052 |
Neg logit contributions | |
ily | -0.020 |
per | -0.019 |
like | -0.019 |
ppy | -0.017 |
ectar | -0.016 |
Token | She |
Token ID | 1375 |
Feature activation | -8.68e-01 |
Pos logit contributions | |
Paula | +0.004 |
irens | +0.004 |
Paul | +0.003 |
spe | +0.003 |
James | +0.003 |
Neg logit contributions | |
ppy | -0.003 |
per | -0.003 |
ilies | -0.003 |
opsy | -0.003 |
ily | -0.003 |
Token | ran |
Token ID | 4966 |
Feature activation | +2.86e+00 |
Pos logit contributions | |
ppy | +0.061 |
per | +0.059 |
pper | +0.058 |
ilies | +0.057 |
opsy | +0.057 |
Neg logit contributions | |
Paula | -0.067 |
irens | -0.065 |
Paul | -0.063 |
spe | -0.059 |
John | -0.059 |
Token | to |
Token ID | 284 |
Feature activation | +5.85e-01 |
Pos logit contributions | |
Paula | +0.236 |
irens | +0.226 |
Paul | +0.223 |
spe | +0.219 |
James | +0.206 |
Neg logit contributions | |
per | -0.176 |
ppy | -0.175 |
ily | -0.171 |
like | -0.168 |
ilies | -0.165 |
Token | They |
Token ID | 1119 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
Paula | +0.003 |
irens | +0.003 |
Paul | +0.002 |
spe | +0.002 |
John | +0.002 |
Neg logit contributions | |
ppy | -0.003 |
pper | -0.003 |
ilies | -0.003 |
per | -0.003 |
opsy | -0.003 |
Token | hear |
Token ID | 3285 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
Paula | +0.115 |
Paul | +0.110 |
irens | +0.109 |
John | +0.104 |
James | +0.104 |
Neg logit contributions | |
opsy | -0.074 |
ppy | -0.074 |
ectar | -0.073 |
pper | -0.072 |
ilies | -0.072 |
Token | horns |
Token ID | 26771 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
Paula | +0.144 |
irens | +0.138 |
spe | +0.135 |
Paul | +0.133 |
iren | +0.128 |
Neg logit contributions | |
per | -0.095 |
ily | -0.094 |
ilies | -0.093 |
like | -0.090 |
ppy | -0.090 |
Token | and |
Token ID | 290 |
Feature activation | +4.96e-02 |
Pos logit contributions | |
Paula | +0.022 |
Paul | +0.021 |
irens | +0.020 |
spe | +0.020 |
James | +0.020 |
Neg logit contributions | |
per | -0.010 |
ppy | -0.009 |
pper | -0.009 |
like | -0.009 |
ilies | -0.009 |
Token | s |
Token ID | 264 |
Feature activation | +8.17e-01 |
Pos logit contributions | |
Paula | +0.004 |
irens | +0.004 |
Paul | +0.004 |
spe | +0.004 |
John | +0.003 |
Neg logit contributions | |
ppy | -0.003 |
per | -0.003 |
pper | -0.003 |
ilies | -0.003 |
opsy | -0.003 |
Token | irens |
Token ID | 42917 |
Feature activation | +1.32e+01 |
Pos logit contributions | |
Paula | +0.014 |
spe | +0.012 |
Paul | +0.011 |
John | +0.007 |
James | +0.007 |
Neg logit contributions | |
per | -0.106 |
ppy | -0.106 |
pper | -0.103 |
ilies | -0.103 |
ily | -0.100 |
Token | . |
Token ID | 13 |
Feature activation | +3.82e-02 |
Pos logit contributions | |
Paula | +1.302 |
Paul | +1.240 |
irens | +1.222 |
spe | +1.201 |
Sarah | +1.192 |
Neg logit contributions | |
per | -0.650 |
ppy | -0.615 |
pper | -0.583 |
ilies | -0.568 |
like | -0.566 |
Token | They |
Token ID | 1119 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
Paula | +0.003 |
irens | +0.003 |
Paul | +0.003 |
spe | +0.002 |
John | +0.002 |
Neg logit contributions | |
ppy | -0.003 |
per | -0.003 |
ilies | -0.003 |
opsy | -0.003 |
pper | -0.003 |
Token | are |
Token ID | 389 |
Feature activation | -2.75e+00 |
Pos logit contributions | |
Paula | +0.116 |
irens | +0.111 |
Paul | +0.111 |
John | +0.105 |
spe | +0.104 |
Neg logit contributions | |
ppy | -0.075 |
opsy | -0.073 |
pper | -0.073 |
ilies | -0.073 |
ectar | -0.072 |
Token | scared |
Token ID | 12008 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
ppy | +0.176 |
pper | +0.170 |
ilies | +0.168 |
per | +0.168 |
opsy | +0.168 |
Neg logit contributions | |
Paula | -0.228 |
irens | -0.221 |
Paul | -0.216 |
spe | -0.203 |
John | -0.203 |
Token | . |
Token ID | 13 |
Feature activation | +4.30e-02 |
Pos logit contributions | |
Paula | +0.220 |
spe | +0.205 |
Paul | +0.204 |
irens | +0.203 |
Ellen | +0.195 |
Neg logit contributions | |
ily | -0.084 |
ppy | -0.084 |
per | -0.082 |
like | -0.081 |
ilies | -0.075 |
Token | coming |
Token ID | 2406 |
Feature activation | +8.05e+00 |
Pos logit contributions | |
Paula | +0.565 |
irens | +0.544 |
Paul | +0.536 |
spe | +0.508 |
Sarah | +0.504 |
Neg logit contributions | |
ppy | -0.325 |
pper | -0.315 |
per | -0.315 |
ilies | -0.311 |
opsy | -0.302 |
Token | . |
Token ID | 13 |
Feature activation | +5.68e-02 |
Pos logit contributions | |
Paula | +0.711 |
irens | +0.689 |
Paul | +0.663 |
spe | +0.628 |
James | +0.626 |
Neg logit contributions | |
ppy | -0.462 |
per | -0.456 |
pper | -0.443 |
ilies | -0.437 |
ily | -0.427 |
Token | They |
Token ID | 1119 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
Paula | +0.004 |
irens | +0.004 |
spe | +0.004 |
Paul | +0.004 |
flashing | +0.004 |
Neg logit contributions | |
ppy | -0.004 |
per | -0.004 |
ilies | -0.004 |
ily | -0.004 |
pper | -0.004 |
Token | had |
Token ID | 550 |
Feature activation | -7.18e-02 |
Pos logit contributions | |
Paula | +0.099 |
irens | +0.096 |
Paul | +0.093 |
spe | +0.088 |
John | +0.087 |
Neg logit contributions | |
ppy | -0.090 |
pper | -0.087 |
ilies | -0.086 |
per | -0.086 |
opsy | -0.086 |
Token | s |
Token ID | 264 |
Feature activation | +8.43e-01 |
Pos logit contributions | |
per | +0.004 |
ppy | +0.003 |
ilies | +0.003 |
pper | +0.003 |
careful | +0.003 |
Neg logit contributions | |
Paula | -0.007 |
irens | -0.007 |
Paul | -0.006 |
spe | -0.006 |
iren | -0.006 |
Token | irens |
Token ID | 42917 |
Feature activation | +1.32e+01 |
Pos logit contributions | |
Paula | +0.019 |
spe | +0.014 |
Paul | +0.014 |
Ellen | +0.009 |
James | +0.009 |
Neg logit contributions | |
ppy | -0.108 |
per | -0.106 |
pper | -0.105 |
ilies | -0.104 |
ily | -0.102 |
Token | and |
Token ID | 290 |
Feature activation | +4.77e-02 |
Pos logit contributions | |
Paula | +1.405 |
Paul | +1.310 |
irens | +1.289 |
spe | +1.278 |
Sarah | +1.260 |
Neg logit contributions | |
per | -0.564 |
like | -0.521 |
ppy | -0.504 |
pper | -0.496 |
ilies | -0.442 |
Token | lights |
Token ID | 7588 |
Feature activation | +5.17e+00 |
Pos logit contributions | |
Paula | +0.003 |
irens | +0.003 |
Paul | +0.003 |
spe | +0.003 |
Ellen | +0.003 |
Neg logit contributions | |
per | -0.004 |
ppy | -0.003 |
like | -0.003 |
ilies | -0.003 |
careful | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +4.80e-02 |
Pos logit contributions | |
Paula | +0.538 |
irens | +0.506 |
Paul | +0.498 |
Ellen | +0.488 |
spe | +0.485 |
Neg logit contributions | |
per | -0.218 |
like | -0.218 |
pper | -0.201 |
ilies | -0.199 |
ppy | -0.197 |
Token | They |
Token ID | 1119 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
Paula | +0.004 |
irens | +0.003 |
Paul | +0.003 |
spe | +0.003 |
flashing | +0.003 |
Neg logit contributions | |
ppy | -0.003 |
per | -0.003 |
ilies | -0.003 |
ily | -0.003 |
ectar | -0.003 |
Token | were |
Token ID | 547 |
Feature activation | -3.43e-02 |
Pos logit contributions | |
Paula | +0.106 |
irens | +0.103 |
Paul | +0.099 |
spe | +0.092 |
John | +0.092 |
Neg logit contributions | |
ppy | -0.085 |
per | -0.082 |
pper | -0.081 |
ilies | -0.081 |
opsy | -0.078 |
Token | tower |
Token ID | 10580 |
Feature activation | +4.40e-01 |
Pos logit contributions | |
Paula | +0.053 |
irens | +0.050 |
Paul | +0.049 |
spe | +0.046 |
Sarah | +0.045 |
Neg logit contributions | |
ppy | -0.054 |
pper | -0.053 |
per | -0.053 |
ilies | -0.053 |
opsy | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | +4.39e-02 |
Pos logit contributions | |
Paula | +0.040 |
irens | +0.039 |
Paul | +0.038 |
John | +0.036 |
Sarah | +0.036 |
Neg logit contributions | |
ppy | -0.024 |
per | -0.024 |
pper | -0.023 |
ilies | -0.023 |
opsy | -0.022 |
Token | They |
Token ID | 1119 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
Paula | +0.003 |
irens | +0.003 |
Paul | +0.003 |
spe | +0.003 |
John | +0.003 |
Neg logit contributions | |
ppy | -0.003 |
ilies | -0.003 |
per | -0.003 |
opsy | -0.003 |
ectar | -0.003 |
Token | heard |
Token ID | 2982 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
Paula | +0.104 |
irens | +0.101 |
Paul | +0.098 |
spe | +0.092 |
John | +0.091 |
Neg logit contributions | |
ppy | -0.086 |
per | -0.083 |
pper | -0.083 |
ilies | -0.082 |
opsy | -0.080 |
Token | s |
Token ID | 264 |
Feature activation | +8.25e-01 |
Pos logit contributions | |
Paula | +0.133 |
irens | +0.130 |
spe | +0.129 |
Paul | +0.124 |
iren | +0.119 |
Neg logit contributions | |
ily | -0.092 |
ilies | -0.090 |
per | -0.086 |
ppy | -0.084 |
like | -0.083 |
Token | irens |
Token ID | 42917 |
Feature activation | +1.32e+01 |
Pos logit contributions | |
Paula | +0.014 |
Paul | +0.011 |
spe | +0.009 |
John | +0.007 |
James | +0.006 |
Neg logit contributions | |
ppy | -0.109 |
per | -0.109 |
pper | -0.107 |
ilies | -0.106 |
ily | -0.105 |
Token | and |
Token ID | 290 |
Feature activation | +4.58e-02 |
Pos logit contributions | |
Paula | +1.397 |
Paul | +1.336 |
spe | +1.311 |
Sarah | +1.293 |
irens | +1.275 |
Neg logit contributions | |
per | -0.573 |
ppy | -0.508 |
like | -0.496 |
pper | -0.490 |
ilies | -0.426 |
Token | saw |
Token ID | 2497 |
Feature activation | -7.16e-01 |
Pos logit contributions | |
Paula | +0.003 |
irens | +0.003 |
Paul | +0.003 |
Smith | +0.003 |
parent | +0.003 |
Neg logit contributions | |
like | -0.003 |
ppy | -0.003 |
per | -0.003 |
wait | -0.003 |
careful | -0.003 |
Token | fire |
Token ID | 2046 |
Feature activation | +2.01e-01 |
Pos logit contributions | |
ily | +0.043 |
ilies | +0.042 |
per | +0.041 |
like | +0.040 |
ectar | +0.038 |
Neg logit contributions | |
Paula | -0.061 |
Paul | -0.056 |
spe | -0.055 |
irens | -0.055 |
Ellen | -0.054 |
Token | trucks |
Token ID | 13960 |
Feature activation | +6.06e+00 |
Pos logit contributions | |
Paula | +0.018 |
Ellen | +0.017 |
Sarah | +0.017 |
Sarah | +0.016 |
Paul | +0.016 |
Neg logit contributions | |
per | -0.012 |
pper | -0.011 |
ppy | -0.011 |
like | -0.011 |
bles | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | +4.96e-02 |
Pos logit contributions | |
Paula | +0.602 |
Paul | +0.566 |
irens | +0.556 |
Sarah | +0.541 |
Ellen | +0.535 |
Neg logit contributions | |
per | -0.300 |
pper | -0.283 |
ppy | -0.282 |
like | -0.265 |
ilies | -0.265 |
Token | trucks |
Token ID | 13960 |
Feature activation | +6.06e+00 |
Pos logit contributions | |
Paula | +0.020 |
Paul | +0.018 |
irens | +0.018 |
Sarah | +0.018 |
Ellen | +0.018 |
Neg logit contributions | |
per | -0.011 |
ppy | -0.011 |
pper | -0.010 |
ilies | -0.010 |
like | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
Paula | +0.562 |
irens | +0.538 |
Paul | +0.533 |
spe | +0.504 |
Sarah | +0.501 |
Neg logit contributions | |
ppy | -0.326 |
per | -0.320 |
pper | -0.314 |
ilies | -0.310 |
opsy | -0.294 |
Token | They |
Token ID | 1119 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
Paula | +0.002 |
irens | +0.002 |
Paul | +0.002 |
John | +0.002 |
spe | +0.002 |
Neg logit contributions | |
ppy | -0.002 |
pper | -0.002 |
opsy | -0.002 |
per | -0.002 |
ilies | -0.002 |
Token | hear |
Token ID | 3285 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
Paula | +0.116 |
irens | +0.113 |
Paul | +0.111 |
John | +0.105 |
Sarah | +0.104 |
Neg logit contributions | |
ppy | -0.073 |
opsy | -0.070 |
pper | -0.070 |
ilies | -0.069 |
ectar | -0.068 |
Token | s |
Token ID | 264 |
Feature activation | +8.24e-01 |
Pos logit contributions | |
Paula | +0.132 |
irens | +0.128 |
Paul | +0.125 |
spe | +0.122 |
Sarah | +0.116 |
Neg logit contributions | |
ppy | -0.113 |
ilies | -0.113 |
per | -0.111 |
ily | -0.108 |
pper | -0.107 |
Token | irens |
Token ID | 42917 |
Feature activation | +1.32e+01 |
Pos logit contributions | |
Paula | +0.016 |
Paul | +0.014 |
spe | +0.013 |
John | +0.010 |
Sarah | +0.008 |
Neg logit contributions | |
per | -0.106 |
ppy | -0.105 |
ilies | -0.103 |
pper | -0.103 |
ily | -0.102 |
Token | and |
Token ID | 290 |
Feature activation | +3.93e-02 |
Pos logit contributions | |
Paula | +1.389 |
Paul | +1.330 |
spe | +1.298 |
Sarah | +1.277 |
irens | +1.275 |
Neg logit contributions | |
per | -0.573 |
ppy | -0.518 |
like | -0.502 |
pper | -0.481 |
ilies | -0.452 |
Token | h |
Token ID | 289 |
Feature activation | -4.37e+00 |
Pos logit contributions | |
Paula | +0.003 |
irens | +0.003 |
Paul | +0.003 |
spe | +0.003 |
Ellen | +0.003 |
Neg logit contributions | |
ppy | -0.003 |
per | -0.003 |
ilies | -0.003 |
like | -0.003 |
wait | -0.003 |
Token | oses |
Token ID | 4629 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
ppy | +0.303 |
pper | +0.291 |
opsy | +0.289 |
ilies | +0.287 |
per | +0.285 |
Neg logit contributions | |
Paula | -0.338 |
irens | -0.336 |
Paul | -0.320 |
John | -0.298 |
spe | -0.297 |
Token | . |
Token ID | 13 |
Feature activation | +4.07e-02 |
Pos logit contributions | |
ppy | +0.114 |
per | +0.112 |
pper | +0.110 |
ilies | +0.108 |
opsy | +0.103 |
Neg logit contributions | |
Paula | -0.191 |
irens | -0.184 |
Paul | -0.181 |
spe | -0.172 |
Sarah | -0.170 |
Token | They |
Token ID | 1119 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
Paula | +0.003 |
irens | +0.003 |
Paul | +0.003 |
spe | +0.002 |
John | +0.002 |
Neg logit contributions | |
ppy | -0.003 |
pper | -0.003 |
ilies | -0.003 |
per | -0.003 |
opsy | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | +3.93e-02 |
Pos logit contributions | |
ppy | +0.083 |
per | +0.081 |
pper | +0.080 |
ilies | +0.079 |
opsy | +0.078 |
Neg logit contributions | |
Paula | -0.145 |
irens | -0.142 |
Paul | -0.137 |
spe | -0.131 |
John | -0.128 |
Token | saw |
Token ID | 2497 |
Feature activation | -7.06e-01 |
Pos logit contributions | |
Paula | +0.002 |
Paul | +0.002 |
John | +0.002 |
Sarah | +0.002 |
James | +0.002 |
Neg logit contributions | |
pper | -0.004 |
ilies | -0.004 |
ppy | -0.004 |
ectar | -0.004 |
oll | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | -8.37e-01 |
Pos logit contributions | |
ppy | +0.045 |
pper | +0.044 |
per | +0.043 |
opsy | +0.042 |
ilies | +0.042 |
Neg logit contributions | |
Paula | -0.059 |
irens | -0.057 |
Paul | -0.057 |
John | -0.054 |
Sarah | -0.054 |
Token | brown |
Token ID | 7586 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
ppy | +0.061 |
pper | +0.059 |
ilies | +0.059 |
per | +0.058 |
ily | +0.056 |
Neg logit contributions | |
Paula | -0.062 |
irens | -0.059 |
Paul | -0.058 |
spe | -0.055 |
flashing | -0.055 |
Token | honey |
Token ID | 12498 |
Feature activation | -6.03e+00 |
Pos logit contributions | |
Smith | +0.017 |
Paula | +0.017 |
Sarah | +0.017 |
Paul | +0.016 |
irens | +0.016 |
Neg logit contributions | |
ppy | -0.016 |
ilies | -0.016 |
pper | -0.015 |
per | -0.015 |
ars | -0.015 |
Token | comb |
Token ID | 24011 |
Feature activation | -1.13e+01 |
Pos logit contributions | |
ppy | +0.336 |
ilies | +0.336 |
careful | +0.305 |
ily | +0.305 |
imes | +0.304 |
Neg logit contributions | |
flashing | -0.455 |
Paula | -0.445 |
Smith | -0.445 |
irens | -0.434 |
lf | -0.427 |
Token | . |
Token ID | 13 |
Feature activation | +4.31e-02 |
Pos logit contributions | |
ppy | +0.826 |
ilies | +0.776 |
opsy | +0.751 |
pper | +0.750 |
per | +0.737 |
Neg logit contributions | |
Paula | -0.775 |
irens | -0.764 |
flashing | -0.760 |
Paul | -0.738 |
John | -0.735 |
Token | He |
Token ID | 679 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
Paula | +0.003 |
irens | +0.003 |
Paul | +0.003 |
spe | +0.003 |
flashing | +0.003 |
Neg logit contributions | |
ppy | -0.003 |
per | -0.003 |
ilies | -0.003 |
opsy | -0.003 |
ily | -0.003 |
Token | was |
Token ID | 373 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
Paula | +0.009 |
Paul | +0.008 |
John | +0.008 |
Sarah | +0.008 |
spe | +0.008 |
Neg logit contributions | |
ilies | -0.015 |
ectar | -0.015 |
per | -0.015 |
pper | -0.014 |
seeker | -0.014 |
Token | very |
Token ID | 845 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
Paula | +0.096 |
Paul | +0.091 |
John | +0.091 |
Sarah | +0.089 |
flashing | +0.087 |
Neg logit contributions | |
ppy | -0.121 |
pper | -0.119 |
opsy | -0.112 |
ilies | -0.111 |
per | -0.109 |
Token | happy |
Token ID | 3772 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
pper | +0.020 |
ppy | +0.020 |
per | +0.018 |
ilies | +0.018 |
ive | +0.018 |
Neg logit contributions | |
John | -0.013 |
Sarah | -0.012 |
Paula | -0.012 |
Paul | -0.012 |
spe | -0.012 |
Token | crawled |
Token ID | 45668 |
Feature activation | -2.23e+00 |
Pos logit contributions | |
ectar | +0.076 |
ilies | +0.075 |
pper | +0.074 |
per | +0.071 |
imes | +0.071 |
Neg logit contributions | |
Paula | -0.047 |
Paul | -0.044 |
John | -0.042 |
Sarah | -0.042 |
bell | -0.041 |
Token | into |
Token ID | 656 |
Feature activation | -2.88e-01 |
Pos logit contributions | |
pper | +0.184 |
ppy | +0.176 |
opsy | +0.175 |
ilies | +0.170 |
per | +0.168 |
Neg logit contributions | |
Paula | -0.140 |
Sarah | -0.135 |
irens | -0.132 |
Paul | -0.130 |
John | -0.126 |
Token | her |
Token ID | 607 |
Feature activation | -5.20e-01 |
Pos logit contributions | |
ppy | +0.020 |
ilies | +0.019 |
opsy | +0.019 |
per | +0.019 |
pper | +0.019 |
Neg logit contributions | |
Paula | -0.022 |
irens | -0.021 |
Paul | -0.021 |
spe | -0.020 |
John | -0.019 |
Token | little |
Token ID | 1310 |
Feature activation | +2.27e+00 |
Pos logit contributions | |
ppy | +0.035 |
pper | +0.035 |
ilies | +0.033 |
per | +0.033 |
ily | +0.032 |
Neg logit contributions | |
Paula | -0.039 |
irens | -0.036 |
Sarah | -0.036 |
Paul | -0.036 |
Smith | -0.035 |
Token | honey |
Token ID | 12498 |
Feature activation | -6.02e+00 |
Pos logit contributions | |
Paula | +0.173 |
irens | +0.166 |
Paul | +0.161 |
Sarah | +0.158 |
John | +0.155 |
Neg logit contributions | |
pper | -0.160 |
ppy | -0.152 |
ilies | -0.148 |
ily | -0.143 |
per | -0.143 |
Token | comb |
Token ID | 24011 |
Feature activation | -1.13e+01 |
Pos logit contributions | |
ilies | +0.419 |
ily | +0.398 |
ppy | +0.393 |
imes | +0.387 |
pper | +0.374 |
Neg logit contributions | |
Paula | -0.427 |
irens | -0.409 |
Paul | -0.398 |
flashing | -0.396 |
Smith | -0.391 |
Token | bed |
Token ID | 3996 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
ppy | +0.809 |
ilies | +0.794 |
pper | +0.779 |
opsy | +0.756 |
ily | +0.752 |
Neg logit contributions | |
Paula | -0.783 |
irens | -0.758 |
Sarah | -0.739 |
Paul | -0.739 |
flashing | -0.737 |
Token | and |
Token ID | 290 |
Feature activation | +5.77e-02 |
Pos logit contributions | |
ppy | +0.102 |
pper | +0.099 |
ilies | +0.099 |
opsy | +0.098 |
per | +0.097 |
Neg logit contributions | |
Paula | -0.148 |
irens | -0.144 |
Paul | -0.141 |
Sarah | -0.133 |
John | -0.133 |
Token | fell |
Token ID | 3214 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
Paula | +0.003 |
Sarah | +0.003 |
John | +0.003 |
Paul | +0.003 |
James | +0.003 |
Neg logit contributions | |
pper | -0.005 |
ilies | -0.005 |
ectar | -0.004 |
ppy | -0.004 |
per | -0.004 |
Token | fast |
Token ID | 3049 |
Feature activation | +3.21e+00 |
Pos logit contributions | |
Paula | +0.008 |
irens | +0.008 |
Paul | +0.007 |
Sarah | +0.007 |
flashing | +0.006 |
Neg logit contributions | |
ppy | -0.015 |
pper | -0.015 |
ilies | -0.014 |
opsy | -0.014 |
per | -0.014 |
Token | asleep |
Token ID | 16039 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
Paula | +0.191 |
irens | +0.184 |
Paul | +0.177 |
spe | +0.164 |
John | +0.159 |
Neg logit contributions | |
ppy | -0.280 |
per | -0.277 |
ilies | -0.269 |
pper | -0.267 |
ectar | -0.266 |
Token | pay |
Token ID | 1414 |
Feature activation | +3.30e+00 |
Pos logit contributions | |
ectar | +0.219 |
ppy | +0.218 |
ilies | +0.217 |
opsy | +0.208 |
per | +0.207 |
Neg logit contributions | |
Paul | -0.199 |
Paula | -0.195 |
John | -0.184 |
Sarah | -0.177 |
irens | -0.175 |
Token | me |
Token ID | 502 |
Feature activation | -4.63e+00 |
Pos logit contributions | |
Paula | +0.262 |
Paul | +0.260 |
irens | +0.254 |
John | +0.246 |
Sarah | +0.241 |
Neg logit contributions | |
ppy | -0.216 |
opsy | -0.214 |
ectar | -0.211 |
ilies | -0.209 |
pper | -0.206 |
Token | with |
Token ID | 351 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
ppy | +0.305 |
ilies | +0.297 |
ectar | +0.294 |
opsy | +0.294 |
pper | +0.284 |
Neg logit contributions | |
Paula | -0.364 |
Paul | -0.358 |
irens | -0.351 |
John | -0.339 |
Sarah | -0.332 |
Token | some |
Token ID | 617 |
Feature activation | -3.58e+00 |
Pos logit contributions | |
Paula | +0.015 |
irens | +0.015 |
Paul | +0.014 |
John | +0.014 |
spe | +0.014 |
Neg logit contributions | |
ppy | -0.012 |
pper | -0.011 |
ilies | -0.011 |
per | -0.011 |
opsy | -0.011 |
Token | honey |
Token ID | 12498 |
Feature activation | -6.00e+00 |
Pos logit contributions | |
ppy | +0.263 |
pper | +0.255 |
opsy | +0.250 |
per | +0.248 |
ilies | +0.244 |
Neg logit contributions | |
Paula | -0.260 |
irens | -0.256 |
Paul | -0.250 |
John | -0.234 |
spe | -0.232 |
Token | comb |
Token ID | 24011 |
Feature activation | -1.13e+01 |
Pos logit contributions | |
opsy | +0.421 |
ilies | +0.417 |
ppy | +0.408 |
ectar | +0.398 |
pper | +0.395 |
Neg logit contributions | |
Paula | -0.426 |
Paul | -0.414 |
irens | -0.398 |
John | -0.397 |
Sarah | -0.392 |
Token | . |
Token ID | 13 |
Feature activation | +7.98e-02 |
Pos logit contributions | |
ppy | +0.758 |
opsy | +0.750 |
ilies | +0.738 |
pper | +0.736 |
per | +0.714 |
Neg logit contributions | |
Paula | -0.850 |
Paul | -0.819 |
irens | -0.802 |
John | -0.797 |
Sarah | -0.787 |
Token | â |
Token ID | 22940 |
Feature activation | -3.19e+00 |
Pos logit contributions | |
Paula | +0.006 |
Paul | +0.006 |
irens | +0.006 |
John | +0.005 |
Sarah | +0.005 |
Neg logit contributions | |
ppy | -0.006 |
pper | -0.006 |
ilies | -0.005 |
opsy | -0.005 |
per | -0.005 |
Token | € |
Token ID | 26391 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
ppy | +0.223 |
ilies | +0.219 |
opsy | +0.217 |
ectar | +0.214 |
per | +0.214 |
Neg logit contributions | |
Paula | -0.242 |
irens | -0.236 |
Paul | -0.230 |
John | -0.216 |
Sarah | -0.215 |
Token | John |
Token ID | 1757 |
Feature activation | +8.29e+00 |
Pos logit contributions | |
ppy | +0.110 |
pper | +0.107 |
per | +0.106 |
opsy | +0.104 |
ilies | +0.102 |
Neg logit contributions | |
Paula | -0.130 |
irens | -0.125 |
Paul | -0.119 |
spe | -0.117 |
flashing | -0.114 |
Token | agreed |
Token ID | 4987 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
Paula | +0.529 |
Paul | +0.515 |
John | +0.486 |
irens | +0.481 |
spe | +0.476 |
Neg logit contributions | |
ilies | -0.654 |
ectar | -0.647 |
ppy | -0.634 |
opsy | -0.631 |
per | -0.623 |
Token | they |
Token ID | 484 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
ppy | +0.032 |
ilies | +0.032 |
pper | +0.032 |
opsy | +0.031 |
per | +0.031 |
Neg logit contributions | |
Paula | -0.034 |
irens | -0.033 |
Paul | -0.031 |
iren | -0.030 |
spe | -0.029 |
Token | found |
Token ID | 1043 |
Feature activation | -3.09e+00 |
Pos logit contributions | |
Paula | +0.142 |
Paul | +0.141 |
Sarah | +0.136 |
John | +0.136 |
James | +0.124 |
Neg logit contributions | |
ectar | -0.190 |
oll | -0.182 |
ider | -0.182 |
ilies | -0.180 |
opsy | -0.177 |
Token | lots |
Token ID | 6041 |
Feature activation | +6.19e-01 |
Pos logit contributions | |
ppy | +0.207 |
per | +0.198 |
pper | +0.197 |
opsy | +0.189 |
ilies | +0.185 |
Neg logit contributions | |
Paula | -0.245 |
Paul | -0.240 |
John | -0.234 |
irens | -0.232 |
Sarah | -0.229 |
Token | of |
Token ID | 286 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
Paula | +0.041 |
irens | +0.041 |
Paul | +0.039 |
Sarah | +0.037 |
John | +0.037 |
Neg logit contributions | |
ppy | -0.049 |
pper | -0.048 |
per | -0.047 |
opsy | -0.047 |
ilies | -0.045 |
Token | honey |
Token ID | 12498 |
Feature activation | -6.00e+00 |
Pos logit contributions | |
Paula | +0.103 |
irens | +0.100 |
Paul | +0.098 |
Sarah | +0.096 |
flashing | +0.095 |
Neg logit contributions | |
pper | -0.086 |
ppy | -0.084 |
per | -0.083 |
opsy | -0.076 |
ilies | -0.074 |
Token | comb |
Token ID | 24011 |
Feature activation | -1.13e+01 |
Pos logit contributions | |
ppy | +0.386 |
opsy | +0.382 |
pper | +0.360 |
ilies | +0.360 |
ily | +0.357 |
Neg logit contributions | |
Paula | -0.435 |
irens | -0.419 |
Sarah | -0.410 |
Paul | -0.407 |
John | -0.399 |
Token | and |
Token ID | 290 |
Feature activation | +7.89e-02 |
Pos logit contributions | |
ppy | +0.721 |
pper | +0.681 |
opsy | +0.681 |
per | +0.668 |
ilies | +0.668 |
Neg logit contributions | |
Paula | -0.905 |
irens | -0.870 |
Paul | -0.864 |
John | -0.842 |
Sarah | -0.839 |
Token | honey |
Token ID | 12498 |
Feature activation | -5.99e+00 |
Pos logit contributions | |
Paula | +0.006 |
Paul | +0.006 |
irens | +0.006 |
John | +0.006 |
Sarah | +0.005 |
Neg logit contributions | |
ppy | -0.005 |
per | -0.005 |
pper | -0.005 |
opsy | -0.005 |
ilies | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +7.02e-02 |
Pos logit contributions | |
ppy | +0.390 |
opsy | +0.388 |
pper | +0.377 |
per | +0.376 |
ilies | +0.373 |
Neg logit contributions | |
Paula | -0.433 |
Paul | -0.407 |
Sarah | -0.403 |
irens | -0.400 |
John | -0.397 |
Token | They |
Token ID | 1119 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
Paula | +0.006 |
irens | +0.005 |
spe | +0.005 |
Paul | +0.005 |
bul | +0.005 |
Neg logit contributions | |
ilies | -0.004 |
ily | -0.004 |
per | -0.004 |
ppy | -0.004 |
opsy | -0.004 |
Token | shared |
Token ID | 4888 |
Feature activation | +3.47e-01 |
Pos logit contributions | |
John | +0.074 |
Paula | +0.074 |
Sarah | +0.072 |
Paul | +0.072 |
James | +0.069 |
Neg logit contributions | |
ectar | -0.110 |
ider | -0.109 |
per | -0.109 |
ute | -0.107 |
ilies | -0.105 |
Token | It |
Token ID | 632 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
Paula | +0.004 |
irens | +0.004 |
spe | +0.004 |
Paul | +0.004 |
flashing | +0.003 |
Neg logit contributions | |
per | -0.004 |
ilies | -0.004 |
ppy | -0.003 |
pper | -0.003 |
opsy | -0.003 |
Token | was |
Token ID | 373 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
ppy | +0.189 |
pper | +0.185 |
ilies | +0.184 |
per | +0.184 |
opsy | +0.183 |
Neg logit contributions | |
Paula | -0.130 |
irens | -0.124 |
Paul | -0.121 |
spe | -0.111 |
John | -0.111 |
Token | full |
Token ID | 1336 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
Paula | +0.130 |
irens | +0.128 |
Paul | +0.123 |
spe | +0.118 |
James | +0.115 |
Neg logit contributions | |
ppy | -0.089 |
pper | -0.086 |
ilies | -0.086 |
opsy | -0.086 |
per | -0.085 |
Token | of |
Token ID | 286 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
ppy | +0.121 |
pper | +0.118 |
opsy | +0.117 |
ilies | +0.117 |
per | +0.117 |
Neg logit contributions | |
Paula | -0.122 |
irens | -0.118 |
Paul | -0.116 |
John | -0.109 |
Sarah | -0.109 |
Token | honey |
Token ID | 12498 |
Feature activation | -6.00e+00 |
Pos logit contributions | |
Paula | +0.109 |
irens | +0.105 |
Paul | +0.103 |
John | +0.098 |
Sarah | +0.097 |
Neg logit contributions | |
ppy | -0.080 |
pper | -0.078 |
per | -0.076 |
opsy | -0.075 |
ilies | -0.075 |
Token | comb |
Token ID | 24011 |
Feature activation | -1.13e+01 |
Pos logit contributions | |
ppy | +0.409 |
ilies | +0.400 |
opsy | +0.394 |
pper | +0.393 |
per | +0.390 |
Neg logit contributions | |
Paula | -0.455 |
Paul | -0.437 |
irens | -0.436 |
John | -0.413 |
Sarah | -0.409 |
Token | and |
Token ID | 290 |
Feature activation | +7.89e-02 |
Pos logit contributions | |
ppy | +0.722 |
pper | +0.693 |
ilies | +0.692 |
opsy | +0.685 |
per | +0.684 |
Neg logit contributions | |
Paula | -0.923 |
irens | -0.891 |
Paul | -0.881 |
John | -0.836 |
spe | -0.829 |
Token | it |
Token ID | 340 |
Feature activation | -2.57e-01 |
Pos logit contributions | |
Paula | +0.006 |
irens | +0.006 |
Paul | +0.006 |
spe | +0.006 |
iren | +0.005 |
Neg logit contributions | |
ppy | -0.005 |
ilies | -0.005 |
pper | -0.005 |
opsy | -0.005 |
per | -0.005 |
Token | smelled |
Token ID | 37609 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
ppy | +0.021 |
pper | +0.020 |
ilies | +0.020 |
per | +0.020 |
opsy | +0.020 |
Neg logit contributions | |
Paula | -0.017 |
irens | -0.016 |
Paul | -0.016 |
spe | -0.014 |
John | -0.014 |
Token | so |
Token ID | 523 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
Paula | +0.142 |
irens | +0.137 |
Paul | +0.135 |
John | +0.128 |
Sarah | +0.127 |
Neg logit contributions | |
ppy | -0.108 |
pper | -0.105 |
per | -0.103 |
opsy | -0.103 |
ilies | -0.103 |
Token | sweet |
Token ID | 6029 |
Feature activation | -8.61e-01 |
Pos logit contributions | |
Paula | +0.082 |
Paul | +0.079 |
irens | +0.078 |
John | +0.077 |
Sarah | +0.076 |
Neg logit contributions | |
ppy | -0.065 |
pper | -0.065 |
per | -0.063 |
ilies | -0.061 |
opsy | -0.060 |
Token | and |
Token ID | 290 |
Feature activation | +8.57e-02 |
Pos logit contributions | |
ppy | +0.159 |
ilies | +0.156 |
opsy | +0.154 |
pper | +0.153 |
per | +0.152 |
Neg logit contributions | |
Paula | -0.225 |
irens | -0.217 |
Paul | -0.215 |
John | -0.203 |
Sarah | -0.202 |
Token | he |
Token ID | 339 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
Paula | +0.006 |
Paul | +0.006 |
irens | +0.006 |
Sarah | +0.006 |
John | +0.006 |
Neg logit contributions | |
ppy | -0.006 |
per | -0.006 |
pper | -0.006 |
ilies | -0.006 |
opsy | -0.006 |
Token | split |
Token ID | 6626 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
ilies | +0.021 |
ectar | +0.021 |
per | +0.021 |
ppy | +0.021 |
opsy | +0.020 |
Neg logit contributions | |
Paula | -0.018 |
Paul | -0.018 |
Sarah | -0.017 |
spe | -0.017 |
John | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | +7.86e-01 |
Pos logit contributions | |
ppy | +0.128 |
pper | +0.125 |
ilies | +0.124 |
opsy | +0.124 |
per | +0.123 |
Neg logit contributions | |
Paula | -0.099 |
Paul | -0.096 |
irens | -0.094 |
John | -0.092 |
Sarah | -0.089 |
Token | honey |
Token ID | 12498 |
Feature activation | -5.98e+00 |
Pos logit contributions | |
chaos | +0.052 |
million | +0.052 |
stairs | +0.051 |
Paula | +0.050 |
doorway | +0.050 |
Neg logit contributions | |
pper | -0.055 |
per | -0.053 |
ilies | -0.052 |
ppy | -0.051 |
amb | -0.049 |
Token | comb |
Token ID | 24011 |
Feature activation | -1.13e+01 |
Pos logit contributions | |
opsy | +0.419 |
ppy | +0.408 |
ilies | +0.404 |
pper | +0.396 |
ani | +0.383 |
Neg logit contributions | |
spe | -0.366 |
Paula | -0.366 |
Sarah | -0.364 |
Smith | -0.363 |
John | -0.354 |
Token | in |
Token ID | 287 |
Feature activation | -2.79e+00 |
Pos logit contributions | |
ppy | +0.793 |
opsy | +0.776 |
pper | +0.762 |
ilies | +0.751 |
ily | +0.701 |
Neg logit contributions | |
Paula | -0.773 |
spe | -0.756 |
John | -0.752 |
Sarah | -0.749 |
Paul | -0.744 |
Token | two |
Token ID | 734 |
Feature activation | -8.71e-01 |
Pos logit contributions | |
ppy | +0.180 |
pper | +0.175 |
per | +0.174 |
opsy | +0.171 |
ilies | +0.171 |
Neg logit contributions | |
Paula | -0.231 |
irens | -0.226 |
Paul | -0.216 |
spe | -0.206 |
iren | -0.200 |
Token | with |
Token ID | 351 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
pper | +0.063 |
ppy | +0.062 |
opsy | +0.061 |
ilies | +0.059 |
per | +0.058 |
Neg logit contributions | |
Paula | -0.060 |
irens | -0.059 |
Paul | -0.057 |
John | -0.055 |
spe | -0.055 |
Token | his |
Token ID | 465 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
Paula | +0.016 |
irens | +0.016 |
Paul | +0.016 |
John | +0.015 |
Sarah | +0.015 |
Neg logit contributions | |
ppy | -0.014 |
per | -0.013 |
pper | -0.013 |
opsy | -0.013 |
ilies | -0.013 |
Token | tiny |
Token ID | 7009 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
Paula | +0.096 |
Sarah | +0.095 |
John | +0.094 |
Paul | +0.093 |
Smith | +0.091 |
Neg logit contributions | |
per | -0.119 |
pper | -0.115 |
ppy | -0.115 |
ilies | -0.113 |
ily | -0.107 |
Token | one |
Token ID | 530 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
ppy | +0.381 |
pper | +0.379 |
per | +0.376 |
opsy | +0.373 |
ilies | +0.372 |
Neg logit contributions | |
Paula | -0.444 |
irens | -0.435 |
Paul | -0.429 |
Sarah | -0.425 |
John | -0.418 |
Token | half |
Token ID | 2063 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
ppy | +0.097 |
per | +0.097 |
ilies | +0.094 |
pper | +0.092 |
opsy | +0.089 |
Neg logit contributions | |
John | -0.083 |
Paula | -0.082 |
Sarah | -0.082 |
Paul | -0.081 |
Smith | -0.080 |
Token | of |
Token ID | 286 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
John | +0.083 |
Sarah | +0.078 |
Paul | +0.077 |
spe | +0.077 |
Smith | +0.077 |
Neg logit contributions | |
opsy | -0.135 |
ppy | -0.134 |
per | -0.128 |
ilies | -0.128 |
pper | -0.128 |
Token | the |
Token ID | 262 |
Feature activation | +8.22e-01 |
Pos logit contributions | |
Paula | +0.071 |
John | +0.068 |
Paul | +0.067 |
Sarah | +0.067 |
irens | +0.064 |
Neg logit contributions | |
ppy | -0.121 |
per | -0.118 |
pper | -0.116 |
ilies | -0.114 |
oll | -0.112 |
Token | honey |
Token ID | 12498 |
Feature activation | -5.96e+00 |
Pos logit contributions | |
million | +0.053 |
chaos | +0.053 |
Paula | +0.053 |
Sarah | +0.053 |
stairs | +0.051 |
Neg logit contributions | |
per | -0.057 |
pper | -0.056 |
ppy | -0.053 |
ilies | -0.052 |
izards | -0.049 |
Token | comb |
Token ID | 24011 |
Feature activation | -1.13e+01 |
Pos logit contributions | |
ppy | +0.419 |
ilies | +0.409 |
opsy | +0.408 |
per | +0.393 |
pper | +0.392 |
Neg logit contributions | |
Paula | -0.398 |
John | -0.393 |
Paul | -0.390 |
Sarah | -0.387 |
Smith | -0.381 |
Token | and |
Token ID | 290 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
ppy | +0.799 |
opsy | +0.759 |
ilies | +0.757 |
pper | +0.752 |
per | +0.734 |
Neg logit contributions | |
Paula | -0.785 |
Paul | -0.765 |
John | -0.765 |
Sarah | -0.750 |
spe | -0.746 |
Token | kept |
Token ID | 4030 |
Feature activation | -3.54e+00 |
Pos logit contributions | |
Paula | +0.010 |
Paul | +0.010 |
John | +0.010 |
Sarah | +0.010 |
irens | +0.009 |
Neg logit contributions | |
ppy | -0.010 |
ilies | -0.009 |
per | -0.009 |
opsy | -0.009 |
pper | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | +8.34e-01 |
Pos logit contributions | |
ppy | +0.257 |
pper | +0.249 |
ilies | +0.248 |
per | +0.247 |
opsy | +0.246 |
Neg logit contributions | |
Paula | -0.262 |
irens | -0.253 |
Paul | -0.247 |
spe | -0.230 |
John | -0.229 |
Token | other |
Token ID | 584 |
Feature activation | -3.69e+00 |
Pos logit contributions | |
chaos | +0.053 |
million | +0.052 |
ticking | +0.051 |
doorway | +0.050 |
stairs | +0.050 |
Neg logit contributions | |
per | -0.060 |
ppy | -0.059 |
pper | -0.059 |
ilies | -0.057 |
izards | -0.055 |
Token | one |
Token ID | 530 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
pper | +0.224 |
per | +0.221 |
ppy | +0.218 |
ute | +0.204 |
ilies | +0.203 |
Neg logit contributions | |
Paula | -0.266 |
Sarah | -0.257 |
million | -0.252 |
John | -0.252 |
Smith | -0.252 |
Token | big |
Token ID | 1263 |
Feature activation | +2.40e+00 |
Pos logit contributions | |
ppy | +0.057 |
pper | +0.055 |
ilies | +0.054 |
per | +0.054 |
opsy | +0.054 |
Neg logit contributions | |
Paula | -0.066 |
irens | -0.064 |
Paul | -0.062 |
spe | -0.058 |
John | -0.058 |
Token | , |
Token ID | 11 |
Feature activation | -4.70e-01 |
Pos logit contributions | |
Paula | +0.180 |
Paul | +0.177 |
irens | +0.176 |
Sarah | +0.172 |
John | +0.169 |
Neg logit contributions | |
ppy | -0.166 |
pper | -0.162 |
ilies | -0.157 |
per | -0.156 |
opsy | -0.150 |
Token | beautiful |
Token ID | 4950 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
ppy | +0.034 |
pper | +0.033 |
ilies | +0.032 |
per | +0.032 |
opsy | +0.032 |
Neg logit contributions | |
Paula | -0.035 |
irens | -0.034 |
Paul | -0.033 |
spe | -0.031 |
John | -0.031 |
Token | l |
Token ID | 300 |
Feature activation | -4.21e+00 |
Pos logit contributions | |
Paula | +0.009 |
Paul | +0.008 |
irens | +0.008 |
Sarah | +0.008 |
John | +0.008 |
Neg logit contributions | |
ppy | -0.008 |
pper | -0.008 |
per | -0.007 |
ilies | -0.007 |
opsy | -0.007 |
Token | ily |
Token ID | 813 |
Feature activation | -7.44e+00 |
Pos logit contributions | |
wait | +0.191 |
pper | +0.163 |
ppy | +0.157 |
thirst | +0.153 |
opsy | +0.149 |
Neg logit contributions | |
iren | -0.364 |
irens | -0.361 |
doorway | -0.311 |
smart | -0.296 |
stairs | -0.296 |
Token | pad |
Token ID | 14841 |
Feature activation | -1.12e+01 |
Pos logit contributions | |
ilies | +0.506 |
per | +0.503 |
ppy | +0.503 |
opsy | +0.503 |
pper | +0.500 |
Neg logit contributions | |
Paula | -0.528 |
Paul | -0.513 |
Sarah | -0.498 |
irens | -0.490 |
John | -0.484 |
Token | that |
Token ID | 326 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
opsy | +0.750 |
ppy | +0.731 |
ilies | +0.722 |
per | +0.688 |
pper | +0.669 |
Neg logit contributions | |
irens | -0.792 |
Paula | -0.784 |
Paul | -0.782 |
spe | -0.775 |
Smith | -0.759 |
Token | he |
Token ID | 339 |
Feature activation | -3.33e-01 |
Pos logit contributions | |
Paula | +0.084 |
irens | +0.081 |
Paul | +0.079 |
spe | +0.074 |
John | +0.073 |
Neg logit contributions | |
ppy | -0.077 |
pper | -0.074 |
ilies | -0.074 |
per | -0.074 |
opsy | -0.074 |
Token | found |
Token ID | 1043 |
Feature activation | -3.13e+00 |
Pos logit contributions | |
per | +0.024 |
ilies | +0.024 |
ectar | +0.024 |
pper | +0.024 |
ppy | +0.024 |
Neg logit contributions | |
Paula | -0.022 |
Paul | -0.022 |
Sarah | -0.021 |
James | -0.020 |
John | -0.020 |
Token | in |
Token ID | 287 |
Feature activation | -2.85e+00 |
Pos logit contributions | |
ppy | +0.211 |
pper | +0.201 |
opsy | +0.200 |
per | +0.196 |
ilies | +0.187 |
Neg logit contributions | |
Paul | -0.228 |
irens | -0.226 |
Paula | -0.224 |
John | -0.222 |
Sarah | -0.221 |
Token | the |
Token ID | 262 |
Feature activation | +7.29e-01 |
Pos logit contributions | |
per | +0.118 |
worm | +0.116 |
like | +0.114 |
opsy | +0.114 |
ily | +0.113 |
Neg logit contributions | |
Paula | -0.289 |
irens | -0.282 |
spe | -0.276 |
Paul | -0.268 |
iren | -0.263 |
Token | the |
Token ID | 262 |
Feature activation | +7.39e-01 |
Pos logit contributions | |
ppy | +0.024 |
ilies | +0.024 |
opsy | +0.023 |
per | +0.023 |
pper | +0.023 |
Neg logit contributions | |
Paula | -0.048 |
irens | -0.047 |
Paul | -0.045 |
spe | -0.045 |
James | -0.043 |
Token | door |
Token ID | 3420 |
Feature activation | +3.26e+00 |
Pos logit contributions | |
Paula | +0.055 |
irens | +0.053 |
Paul | +0.052 |
spe | +0.049 |
John | +0.049 |
Neg logit contributions | |
ppy | -0.054 |
pper | -0.052 |
ilies | -0.052 |
per | -0.051 |
opsy | -0.051 |
Token | had |
Token ID | 550 |
Feature activation | -9.35e-02 |
Pos logit contributions | |
Paula | +0.264 |
irens | +0.260 |
Paul | +0.250 |
telev | +0.241 |
James | +0.239 |
Neg logit contributions | |
like | -0.196 |
per | -0.193 |
ily | -0.192 |
ppy | -0.185 |
comb | -0.184 |
Token | a |
Token ID | 257 |
Feature activation | -8.36e-01 |
Pos logit contributions | |
ppy | +0.006 |
per | +0.006 |
ily | +0.006 |
ilies | +0.006 |
pper | +0.006 |
Neg logit contributions | |
Paula | -0.007 |
irens | -0.007 |
Paul | -0.007 |
spe | -0.007 |
James | -0.007 |
Token | big |
Token ID | 1263 |
Feature activation | +2.40e+00 |
Pos logit contributions | |
opsy | +0.049 |
ppy | +0.049 |
worm | +0.049 |
per | +0.048 |
pper | +0.048 |
Neg logit contributions | |
irens | -0.069 |
Paula | -0.069 |
Paul | -0.065 |
James | -0.062 |
spe | -0.062 |
Token | pad |
Token ID | 14841 |
Feature activation | -1.12e+01 |
Pos logit contributions | |
Paula | +0.206 |
irens | +0.202 |
Paul | +0.193 |
spe | +0.188 |
James | +0.181 |
Neg logit contributions | |
per | -0.137 |
ppy | -0.137 |
pper | -0.134 |
opsy | -0.134 |
ilies | -0.132 |
Token | lock |
Token ID | 5354 |
Feature activation | -3.46e+00 |
Pos logit contributions | |
ppy | +0.596 |
opsy | +0.588 |
ilies | +0.588 |
ectar | +0.546 |
pper | +0.531 |
Neg logit contributions | |
Paula | -0.984 |
irens | -0.975 |
Paul | -0.964 |
flashing | -0.939 |
spe | -0.932 |
Token | on |
Token ID | 319 |
Feature activation | +3.78e-01 |
Pos logit contributions | |
ppy | +0.221 |
opsy | +0.217 |
ilies | +0.216 |
ectar | +0.210 |
pper | +0.207 |
Neg logit contributions | |
Paula | -0.283 |
irens | -0.274 |
Paul | -0.269 |
flashing | -0.257 |
spe | -0.254 |
Token | it |
Token ID | 340 |
Feature activation | -2.96e-01 |
Pos logit contributions | |
irens | +0.034 |
Paula | +0.032 |
spe | +0.032 |
iren | +0.031 |
Paul | +0.030 |
Neg logit contributions | |
per | -0.020 |
like | -0.020 |
ily | -0.020 |
comb | -0.019 |
ilies | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +3.68e-02 |
Pos logit contributions | |
ppy | +0.016 |
per | +0.016 |
pper | +0.016 |
ilies | +0.016 |
opsy | +0.015 |
Neg logit contributions | |
Paula | -0.027 |
irens | -0.026 |
Paul | -0.025 |
spe | -0.024 |
John | -0.024 |
Token | Sarah |
Token ID | 10490 |
Feature activation | +9.19e+00 |
Pos logit contributions | |
spe | +0.004 |
irens | +0.003 |
channel | +0.003 |
lights | +0.003 |
Paula | +0.003 |
Neg logit contributions | |
ily | -0.002 |
like | -0.001 |
ilies | -0.001 |
ppy | -0.001 |
ectar | -0.001 |
Token | here |
Token ID | 994 |
Feature activation | -5.80e-01 |
Pos logit contributions | |
opsy | +0.223 |
ppy | +0.214 |
pper | +0.209 |
ilies | +0.204 |
per | +0.198 |
Neg logit contributions | |
Paula | -0.324 |
Paul | -0.313 |
John | -0.306 |
Sarah | -0.304 |
irens | -0.303 |
Token | on |
Token ID | 319 |
Feature activation | +3.78e-01 |
Pos logit contributions | |
opsy | +0.043 |
ppy | +0.042 |
pper | +0.042 |
per | +0.040 |
ilies | +0.039 |
Neg logit contributions | |
Paula | -0.040 |
Paul | -0.039 |
John | -0.038 |
Sarah | -0.038 |
James | -0.037 |
Token | my |
Token ID | 616 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
Paula | +0.031 |
irens | +0.030 |
Paul | +0.030 |
John | +0.028 |
Sarah | +0.028 |
Neg logit contributions | |
ppy | -0.024 |
pper | -0.023 |
ilies | -0.023 |
per | -0.023 |
opsy | -0.023 |
Token | l |
Token ID | 300 |
Feature activation | -4.21e+00 |
Pos logit contributions | |
ppy | +0.011 |
pper | +0.011 |
ilies | +0.011 |
per | +0.010 |
ectar | +0.010 |
Neg logit contributions | |
Paula | -0.012 |
Paul | -0.012 |
irens | -0.011 |
John | -0.011 |
James | -0.011 |
Token | ily |
Token ID | 813 |
Feature activation | -7.43e+00 |
Pos logit contributions | |
wait | +0.169 |
ppy | +0.160 |
careful | +0.150 |
pper | +0.148 |
thirst | +0.147 |
Neg logit contributions | |
iren | -0.358 |
irens | -0.352 |
doorway | -0.342 |
Sarah | -0.328 |
igsaw | -0.325 |
Token | pad |
Token ID | 14841 |
Feature activation | -1.12e+01 |
Pos logit contributions | |
opsy | +0.419 |
ilies | +0.404 |
ectar | +0.388 |
per | +0.387 |
ppy | +0.383 |
Neg logit contributions | |
Paula | -0.628 |
Paul | -0.605 |
Sarah | -0.594 |
Smith | -0.586 |
John | -0.580 |
Token | ." |
Token ID | 526 |
Feature activation | -7.06e-01 |
Pos logit contributions | |
opsy | +0.887 |
ppy | +0.850 |
ilies | +0.839 |
thirst | +0.801 |
pollen | +0.795 |
Neg logit contributions | |
Smith | -0.647 |
Paul | -0.645 |
Paula | -0.641 |
spe | -0.627 |
John | -0.616 |
Token | Lily |
Token ID | 20037 |
Feature activation | +7.96e-01 |
Pos logit contributions | |
pper | +0.059 |
ppy | +0.059 |
opsy | +0.057 |
ilies | +0.057 |
per | +0.056 |
Neg logit contributions | |
Paula | -0.044 |
Paul | -0.042 |
irens | -0.042 |
Sarah | -0.040 |
John | -0.039 |
Token | looked |
Token ID | 3114 |
Feature activation | +4.52e+00 |
Pos logit contributions | |
Paula | +0.049 |
Paul | +0.047 |
irens | +0.045 |
John | +0.043 |
Sarah | +0.043 |
Neg logit contributions | |
ilies | -0.066 |
ppy | -0.065 |
ectar | -0.065 |
pper | -0.064 |
opsy | -0.064 |
Token | at |
Token ID | 379 |
Feature activation | -4.51e+00 |
Pos logit contributions | |
Paula | +0.406 |
Paul | +0.391 |
Sarah | +0.379 |
irens | +0.377 |
John | +0.376 |
Neg logit contributions | |
pper | -0.259 |
ppy | -0.255 |
ilies | -0.242 |
per | -0.241 |
opsy | -0.241 |
Token | the |
Token ID | 262 |
Feature activation | +7.38e-01 |
Pos logit contributions | |
ppy | +0.286 |
opsy | +0.278 |
pper | +0.276 |
ilies | +0.273 |
per | +0.272 |
Neg logit contributions | |
irens | -0.372 |
Paula | -0.362 |
spe | -0.354 |
Paul | -0.337 |
iren | -0.334 |
Token | paints |
Token ID | 32281 |
Feature activation | +6.81e-01 |
Pos logit contributions | |
Paula | +0.097 |
irens | +0.095 |
Paul | +0.092 |
spe | +0.088 |
John | +0.086 |
Neg logit contributions | |
ppy | -0.084 |
ilies | -0.083 |
per | -0.081 |
pper | -0.080 |
opsy | -0.080 |
Token | , |
Token ID | 11 |
Feature activation | -4.76e-01 |
Pos logit contributions | |
Paula | +0.060 |
irens | +0.058 |
Paul | +0.057 |
spe | +0.054 |
Sarah | +0.054 |
Neg logit contributions | |
ppy | -0.039 |
pper | -0.038 |
opsy | -0.038 |
ilies | -0.037 |
per | -0.037 |
Token | and |
Token ID | 290 |
Feature activation | +4.23e-02 |
Pos logit contributions | |
ppy | +0.027 |
ilies | +0.026 |
per | +0.026 |
pper | +0.026 |
opsy | +0.025 |
Neg logit contributions | |
Paula | -0.042 |
irens | -0.041 |
Paul | -0.040 |
spe | -0.038 |
John | -0.038 |
Token | a |
Token ID | 257 |
Feature activation | -8.42e-01 |
Pos logit contributions | |
Paula | +0.004 |
irens | +0.004 |
Paul | +0.003 |
spe | +0.003 |
iren | +0.003 |
Neg logit contributions | |
ppy | -0.002 |
ilies | -0.002 |
per | -0.002 |
ily | -0.002 |
pper | -0.002 |
Token | huge |
Token ID | 3236 |
Feature activation | +2.78e+00 |
Pos logit contributions | |
ppy | +0.062 |
ilies | +0.060 |
opsy | +0.059 |
pper | +0.059 |
per | +0.059 |
Neg logit contributions | |
Paula | -0.060 |
irens | -0.059 |
Paul | -0.057 |
John | -0.053 |
spe | -0.053 |
Token | pad |
Token ID | 14841 |
Feature activation | -1.12e+01 |
Pos logit contributions | |
Paula | +0.214 |
irens | +0.211 |
Paul | +0.200 |
spe | +0.191 |
John | +0.186 |
Neg logit contributions | |
ppy | -0.187 |
per | -0.182 |
pper | -0.181 |
ilies | -0.180 |
opsy | -0.180 |
Token | of |
Token ID | 286 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
ppy | +0.696 |
opsy | +0.693 |
ilies | +0.669 |
pper | +0.655 |
per | +0.650 |
Neg logit contributions | |
Paula | -0.924 |
irens | -0.890 |
Paul | -0.886 |
spe | -0.849 |
Sarah | -0.848 |
Token | paper |
Token ID | 3348 |
Feature activation | +3.30e-01 |
Pos logit contributions | |
Paula | +0.100 |
irens | +0.099 |
Paul | +0.094 |
spe | +0.091 |
iren | +0.088 |
Neg logit contributions | |
ilies | -0.080 |
ppy | -0.080 |
per | -0.077 |
ily | -0.077 |
pper | -0.076 |
Token | . |
Token ID | 13 |
Feature activation | +4.08e-02 |
Pos logit contributions | |
Paula | +0.027 |
irens | +0.026 |
Paul | +0.026 |
Sarah | +0.024 |
spe | +0.024 |
Neg logit contributions | |
ppy | -0.021 |
opsy | -0.021 |
pper | -0.020 |
ilies | -0.020 |
per | -0.020 |
Token | She |
Token ID | 1375 |
Feature activation | -8.63e-01 |
Pos logit contributions | |
irens | +0.003 |
Paula | +0.003 |
spe | +0.003 |
Paul | +0.003 |
flashing | +0.003 |
Neg logit contributions | |
ppy | -0.002 |
ilies | -0.002 |
opsy | -0.002 |
ily | -0.002 |
ectar | -0.002 |
Token | couldn |
Token ID | 3521 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
ppy | +0.067 |
pper | +0.066 |
ilies | +0.066 |
opsy | +0.065 |
per | +0.065 |
Neg logit contributions | |
Paula | -0.059 |
irens | -0.056 |
Paul | -0.055 |
spe | -0.052 |
Sarah | -0.051 |
Token | back |
Token ID | 736 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
opsy | +0.245 |
pper | +0.239 |
ppy | +0.237 |
ilies | +0.236 |
ectar | +0.230 |
Neg logit contributions | |
Paula | -0.143 |
Sarah | -0.142 |
Paul | -0.142 |
John | -0.139 |
irens | -0.135 |
Token | onto |
Token ID | 4291 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
ppy | +0.093 |
per | +0.090 |
pper | +0.090 |
ilies | +0.089 |
opsy | +0.089 |
Neg logit contributions | |
Paula | -0.098 |
irens | -0.095 |
Paul | -0.092 |
spe | -0.088 |
John | -0.086 |
Token | his |
Token ID | 465 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
Paula | +0.071 |
irens | +0.068 |
Paul | +0.067 |
John | +0.061 |
Sarah | +0.061 |
Neg logit contributions | |
ppy | -0.087 |
pper | -0.085 |
ilies | -0.084 |
per | -0.084 |
opsy | -0.084 |
Token | l |
Token ID | 300 |
Feature activation | -4.22e+00 |
Pos logit contributions | |
Paula | +0.100 |
flashing | +0.100 |
Paul | +0.099 |
James | +0.097 |
irens | +0.097 |
Neg logit contributions | |
ilies | -0.113 |
ppy | -0.107 |
ectar | -0.103 |
imes | -0.103 |
izards | -0.102 |
Token | ily |
Token ID | 813 |
Feature activation | -7.45e+00 |
Pos logit contributions | |
wait | +0.164 |
prett | +0.154 |
ppy | +0.150 |
anuts | +0.144 |
opsy | +0.142 |
Neg logit contributions | |
iren | -0.392 |
irens | -0.384 |
doorway | -0.355 |
Paula | -0.324 |
Sarah | -0.322 |
Token | pad |
Token ID | 14841 |
Feature activation | -1.12e+01 |
Pos logit contributions | |
ilies | +0.502 |
ectar | +0.467 |
opsy | +0.466 |
ily | +0.464 |
izards | +0.451 |
Neg logit contributions | |
Paula | -0.570 |
Paul | -0.549 |
Sarah | -0.535 |
John | -0.528 |
James | -0.524 |
Token | and |
Token ID | 290 |
Feature activation | +5.03e-02 |
Pos logit contributions | |
ilies | +0.896 |
opsy | +0.869 |
ppy | +0.852 |
izards | +0.817 |
ily | +0.809 |
Neg logit contributions | |
Paula | -0.652 |
Paul | -0.646 |
Smith | -0.635 |
irens | -0.629 |
John | -0.621 |
Token | stood |
Token ID | 6204 |
Feature activation | -2.68e+00 |
Pos logit contributions | |
Paula | +0.004 |
Paul | +0.003 |
Sarah | +0.003 |
John | +0.003 |
irens | +0.003 |
Neg logit contributions | |
ilies | -0.004 |
ppy | -0.004 |
ectar | -0.004 |
pper | -0.004 |
per | -0.004 |
Token | there |
Token ID | 612 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
ppy | +0.238 |
ilies | +0.231 |
pper | +0.222 |
imes | +0.215 |
izards | +0.212 |
Neg logit contributions | |
Paula | -0.161 |
Paul | -0.156 |
Sarah | -0.153 |
John | -0.151 |
irens | -0.149 |
Token | , |
Token ID | 11 |
Feature activation | -4.81e-01 |
Pos logit contributions | |
Paula | +0.100 |
Paul | +0.097 |
John | +0.094 |
irens | +0.092 |
Sarah | +0.092 |
Neg logit contributions | |
ppy | -0.072 |
ilies | -0.070 |
pper | -0.069 |
opsy | -0.069 |
per | -0.065 |
Token | feeling |
Token ID | 4203 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
ppy | +0.037 |
ilies | +0.036 |
pper | +0.035 |
opsy | +0.035 |
per | +0.035 |
Neg logit contributions | |
Paula | -0.034 |
irens | -0.032 |
Paul | -0.032 |
John | -0.030 |
Sarah | -0.030 |
Token | hopped |
Token ID | 45230 |
Feature activation | -2.68e+00 |
Pos logit contributions | |
Paula | +0.013 |
irens | +0.013 |
Paul | +0.013 |
spe | +0.012 |
John | +0.012 |
Neg logit contributions | |
ppy | -0.014 |
pper | -0.014 |
ilies | -0.014 |
per | -0.014 |
opsy | -0.014 |
Token | from |
Token ID | 422 |
Feature activation | -3.89e-01 |
Pos logit contributions | |
ppy | +0.232 |
pper | +0.229 |
opsy | +0.226 |
oll | +0.214 |
ilies | +0.212 |
Neg logit contributions | |
Paula | -0.161 |
Paul | -0.158 |
irens | -0.151 |
Sarah | -0.150 |
John | -0.150 |
Token | one |
Token ID | 530 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
ppy | +0.028 |
pper | +0.028 |
ilies | +0.027 |
per | +0.027 |
opsy | +0.027 |
Neg logit contributions | |
Paula | -0.029 |
Paul | -0.028 |
irens | -0.028 |
John | -0.026 |
Sarah | -0.026 |
Token | l |
Token ID | 300 |
Feature activation | -4.19e+00 |
Pos logit contributions | |
ppy | +0.097 |
pper | +0.094 |
ilies | +0.092 |
per | +0.091 |
opsy | +0.091 |
Neg logit contributions | |
Paula | -0.101 |
irens | -0.099 |
Paul | -0.097 |
spe | -0.093 |
Sarah | -0.093 |
Token | ily |
Token ID | 813 |
Feature activation | -7.42e+00 |
Pos logit contributions | |
wait | +0.182 |
opsy | +0.170 |
pper | +0.163 |
prett | +0.161 |
ppy | +0.152 |
Neg logit contributions | |
irens | -0.389 |
iren | -0.376 |
doorway | -0.347 |
Paula | -0.336 |
Paul | -0.336 |
Token | pad |
Token ID | 14841 |
Feature activation | -1.12e+01 |
Pos logit contributions | |
ilies | +0.334 |
opsy | +0.328 |
ppy | +0.322 |
ectar | +0.322 |
ily | +0.311 |
Neg logit contributions | |
Paula | -0.730 |
Paul | -0.711 |
irens | -0.681 |
John | -0.679 |
Sarah | -0.679 |
Token | to |
Token ID | 284 |
Feature activation | +6.01e-01 |
Pos logit contributions | |
opsy | +0.854 |
ppy | +0.846 |
ilies | +0.835 |
ily | +0.781 |
pper | +0.781 |
Neg logit contributions | |
Paul | -0.730 |
Paula | -0.729 |
irens | -0.697 |
John | -0.694 |
spe | -0.683 |
Token | the |
Token ID | 262 |
Feature activation | +7.54e-01 |
Pos logit contributions | |
Paula | +0.044 |
Paul | +0.043 |
Sarah | +0.042 |
John | +0.042 |
irens | +0.040 |
Neg logit contributions | |
ppy | -0.044 |
pper | -0.042 |
opsy | -0.042 |
ilies | -0.041 |
ectar | -0.041 |
Token | other |
Token ID | 584 |
Feature activation | -3.79e+00 |
Pos logit contributions | |
chaos | +0.050 |
Paula | +0.049 |
Smith | +0.048 |
Paul | +0.048 |
flashing | +0.048 |
Neg logit contributions | |
ppy | -0.059 |
oll | -0.053 |
pper | -0.052 |
ectar | -0.051 |
int | -0.050 |
Token | . |
Token ID | 13 |
Feature activation | +4.03e-02 |
Pos logit contributions | |
ppy | +0.291 |
pper | +0.284 |
opsy | +0.271 |
oll | +0.269 |
ilies | +0.266 |
Neg logit contributions | |
Paul | -0.240 |
Paula | -0.238 |
John | -0.237 |
Sarah | -0.234 |
Smith | -0.229 |
Token | It |
Token ID | 632 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
Paula | +0.002 |
Paul | +0.002 |
irens | +0.002 |
Sarah | +0.002 |
John | +0.002 |
Neg logit contributions | |
pper | -0.004 |
ppy | -0.004 |
opsy | -0.003 |
per | -0.003 |
ilies | -0.003 |
Token | bug |
Token ID | 5434 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
ppy | +0.013 |
pper | +0.012 |
ilies | +0.012 |
per | +0.011 |
oll | +0.011 |
Neg logit contributions | |
irens | -0.015 |
Paul | -0.015 |
Paula | -0.015 |
Sarah | -0.015 |
Smith | -0.014 |
Token | on |
Token ID | 319 |
Feature activation | +3.84e-01 |
Pos logit contributions | |
ppy | +0.188 |
ilies | +0.183 |
opsy | +0.183 |
pper | +0.180 |
per | +0.173 |
Neg logit contributions | |
irens | -0.223 |
Paula | -0.214 |
flashing | -0.210 |
Paul | -0.207 |
Sarah | -0.202 |
Token | a |
Token ID | 257 |
Feature activation | -8.38e-01 |
Pos logit contributions | |
Paula | +0.026 |
irens | +0.026 |
Paul | +0.025 |
John | +0.024 |
Sarah | +0.023 |
Neg logit contributions | |
ppy | -0.030 |
ilies | -0.029 |
pper | -0.029 |
opsy | -0.028 |
per | -0.028 |
Token | l |
Token ID | 300 |
Feature activation | -4.21e+00 |
Pos logit contributions | |
ilies | +0.057 |
ppy | +0.056 |
oll | +0.054 |
ectar | +0.053 |
ily | +0.052 |
Neg logit contributions | |
Paula | -0.056 |
flashing | -0.056 |
Smith | -0.054 |
Paul | -0.054 |
irens | -0.054 |
Token | ily |
Token ID | 813 |
Feature activation | -7.43e+00 |
Pos logit contributions | |
anuts | +0.151 |
ppy | +0.147 |
careful | +0.139 |
pper | +0.138 |
wait | +0.138 |
Neg logit contributions | |
iren | -0.374 |
irens | -0.373 |
doorway | -0.352 |
flashing | -0.324 |
chaos | -0.314 |
Token | pad |
Token ID | 14841 |
Feature activation | -1.12e+01 |
Pos logit contributions | |
ilies | +0.312 |
opsy | +0.306 |
ectar | +0.304 |
per | +0.283 |
ppy | +0.280 |
Neg logit contributions | |
Paula | -0.693 |
flashing | -0.673 |
Sarah | -0.672 |
Paul | -0.671 |
irens | -0.661 |
Token | . |
Token ID | 13 |
Feature activation | +3.73e-02 |
Pos logit contributions | |
opsy | +0.789 |
ppy | +0.776 |
ilies | +0.773 |
pollen | +0.720 |
per | +0.719 |
Neg logit contributions | |
irens | -0.729 |
Paul | -0.706 |
Paula | -0.706 |
flashing | -0.699 |
Smith | -0.697 |
Token | Ho |
Token ID | 9544 |
Feature activation | -7.01e+00 |
Pos logit contributions | |
Paula | +0.002 |
Paul | +0.002 |
irens | +0.002 |
Sarah | +0.002 |
John | +0.002 |
Neg logit contributions | |
pper | -0.003 |
ppy | -0.003 |
ilies | -0.003 |
opsy | -0.003 |
ily | -0.003 |
Token | ppy |
Token ID | 14097 |
Feature activation | -8.90e+00 |
Pos logit contributions | |
wait | +0.205 |
ilies | +0.171 |
middle | +0.157 |
careful | +0.147 |
ily | +0.135 |
Neg logit contributions | |
Paula | -0.729 |
Smith | -0.713 |
irens | -0.702 |
Paul | -0.698 |
jo | -0.693 |
Token | tried |
Token ID | 3088 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
ilies | +0.711 |
ppy | +0.701 |
pper | +0.693 |
ectar | +0.689 |
per | +0.683 |
Neg logit contributions | |
Paula | -0.566 |
Paul | -0.542 |
irens | -0.537 |
spe | -0.510 |
Sarah | -0.502 |
Token | to |
Token ID | 284 |
Feature activation | +5.91e-01 |
Pos logit contributions | |
Paula | +0.104 |
irens | +0.101 |
Paul | +0.098 |
spe | +0.091 |
John | +0.090 |
Neg logit contributions | |
ppy | -0.116 |
pper | -0.113 |
ilies | -0.112 |
per | -0.112 |
opsy | -0.112 |
Token | ist |
Token ID | 396 |
Feature activation | -2.76e-01 |
Pos logit contributions | |
opsy | +0.140 |
ppy | +0.139 |
ilies | +0.139 |
imes | +0.133 |
ectar | +0.131 |
Neg logit contributions | |
bell | -0.194 |
Paula | -0.191 |
iren | -0.185 |
spe | -0.185 |
irens | -0.183 |
Token | ening |
Token ID | 3101 |
Feature activation | +6.08e+00 |
Pos logit contributions | |
ppy | +0.023 |
pper | +0.023 |
per | +0.023 |
ilies | +0.023 |
opsy | +0.022 |
Neg logit contributions | |
Paula | -0.017 |
irens | -0.016 |
Paul | -0.016 |
spe | -0.014 |
Sarah | -0.014 |
Token | golden |
Token ID | 10861 |
Feature activation | +8.50e-01 |
Pos logit contributions | |
Paula | +0.466 |
Paul | +0.463 |
spe | +0.445 |
irens | +0.443 |
flashing | +0.443 |
Neg logit contributions | |
ppy | -0.401 |
pper | -0.382 |
ilies | -0.372 |
opsy | -0.353 |
ily | -0.352 |
Token | l |
Token ID | 300 |
Feature activation | -4.21e+00 |
Pos logit contributions | |
Paula | +0.063 |
Paul | +0.062 |
irens | +0.062 |
flashing | +0.062 |
spe | +0.060 |
Neg logit contributions | |
ppy | -0.058 |
pper | -0.053 |
ilies | -0.052 |
opsy | -0.051 |
ily | -0.050 |
Token | ily |
Token ID | 813 |
Feature activation | -7.43e+00 |
Pos logit contributions | |
wait | +0.199 |
opsy | +0.174 |
decide | +0.174 |
thirst | +0.166 |
remember | +0.166 |
Neg logit contributions | |
iren | -0.323 |
irens | -0.316 |
doorway | -0.313 |
stairs | -0.289 |
ons | -0.278 |
Token | pad |
Token ID | 14841 |
Feature activation | -1.12e+01 |
Pos logit contributions | |
ilies | +0.317 |
opsy | +0.315 |
ectar | +0.311 |
ppy | +0.301 |
per | +0.296 |
Neg logit contributions | |
Paula | -0.698 |
Paul | -0.678 |
Sarah | -0.655 |
John | -0.654 |
Smith | -0.651 |
Token | . |
Token ID | 13 |
Feature activation | +4.31e-02 |
Pos logit contributions | |
opsy | +0.769 |
ppy | +0.752 |
ilies | +0.734 |
thirst | +0.682 |
per | +0.682 |
Neg logit contributions | |
Paul | -0.756 |
Paula | -0.753 |
irens | -0.746 |
Smith | -0.745 |
spe | -0.744 |
Token | The |
Token ID | 383 |
Feature activation | -3.58e-01 |
Pos logit contributions | |
Paula | +0.003 |
irens | +0.003 |
Paul | +0.003 |
spe | +0.003 |
flashing | +0.003 |
Neg logit contributions | |
ppy | -0.003 |
per | -0.003 |
ilies | -0.003 |
opsy | -0.003 |
pper | -0.003 |
Token | frog |
Token ID | 21264 |
Feature activation | -5.99e+00 |
Pos logit contributions | |
ppy | +0.028 |
pper | +0.028 |
per | +0.025 |
ilies | +0.024 |
oll | +0.023 |
Neg logit contributions | |
stairs | -0.023 |
cartoon | -0.022 |
hallway | -0.022 |
chaos | -0.021 |
flash | -0.021 |
Token | hopped |
Token ID | 45230 |
Feature activation | -2.70e+00 |
Pos logit contributions | |
ilies | +0.525 |
pper | +0.522 |
ppy | +0.522 |
ectar | +0.517 |
per | +0.512 |
Neg logit contributions | |
Paula | -0.323 |
Paul | -0.313 |
spe | -0.298 |
irens | -0.296 |
John | -0.288 |
Token | up |
Token ID | 510 |
Feature activation | +2.62e+00 |
Pos logit contributions | |
opsy | +0.236 |
ppy | +0.234 |
pper | +0.232 |
ilies | +0.228 |
ectar | +0.223 |
Neg logit contributions | |
Paula | -0.154 |
Paul | -0.148 |
Sarah | -0.143 |
irens | -0.143 |
John | -0.140 |
Token | top |
Token ID | 1353 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
Paula | +0.033 |
irens | +0.032 |
Paul | +0.031 |
spe | +0.030 |
Sarah | +0.029 |
Neg logit contributions | |
ppy | -0.025 |
per | -0.024 |
pper | -0.024 |
ilies | -0.024 |
opsy | -0.024 |
Token | of |
Token ID | 286 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
ppy | +0.117 |
pper | +0.113 |
per | +0.113 |
ilies | +0.113 |
opsy | +0.112 |
Neg logit contributions | |
Paula | -0.137 |
irens | -0.132 |
Paul | -0.129 |
spe | -0.120 |
John | -0.120 |
Token | a |
Token ID | 257 |
Feature activation | -8.36e-01 |
Pos logit contributions | |
Paula | +0.086 |
irens | +0.082 |
Paul | +0.081 |
John | +0.076 |
Sarah | +0.075 |
Neg logit contributions | |
ppy | -0.101 |
pper | -0.098 |
per | -0.097 |
ilies | -0.097 |
opsy | -0.096 |
Token | l |
Token ID | 300 |
Feature activation | -4.20e+00 |
Pos logit contributions | |
ppy | +0.067 |
ilies | +0.067 |
pper | +0.065 |
per | +0.064 |
ily | +0.064 |
Neg logit contributions | |
Paula | -0.052 |
Paul | -0.050 |
irens | -0.049 |
Smith | -0.048 |
spe | -0.048 |
Token | illy |
Token ID | 6548 |
Feature activation | -2.81e+00 |
Pos logit contributions | |
wait | +0.165 |
ppy | +0.160 |
anuts | +0.153 |
pper | +0.152 |
okay | +0.151 |
Neg logit contributions | |
iren | -0.377 |
irens | -0.368 |
doorway | -0.353 |
stairs | -0.323 |
chairs | -0.315 |
Token | pad |
Token ID | 14841 |
Feature activation | -1.12e+01 |
Pos logit contributions | |
ppy | +0.204 |
ilies | +0.201 |
ectar | +0.197 |
opsy | +0.194 |
ily | +0.194 |
Neg logit contributions | |
Paul | -0.174 |
Paula | -0.174 |
John | -0.174 |
Sarah | -0.170 |
Smith | -0.167 |
Token | . |
Token ID | 13 |
Feature activation | +4.03e-02 |
Pos logit contributions | |
opsy | +0.823 |
ppy | +0.811 |
ilies | +0.788 |
pper | +0.747 |
per | +0.743 |
Neg logit contributions | |
Paula | -0.707 |
Paul | -0.703 |
irens | -0.700 |
Smith | -0.683 |
spe | -0.682 |
Token | One |
Token ID | 1881 |
Feature activation | -9.08e-01 |
Pos logit contributions | |
Paula | +0.003 |
irens | +0.003 |
Paul | +0.003 |
spe | +0.003 |
John | +0.003 |
Neg logit contributions | |
ppy | -0.003 |
ilies | -0.003 |
pper | -0.003 |
per | -0.003 |
opsy | -0.003 |
Token | day |
Token ID | 1110 |
Feature activation | +6.44e-01 |
Pos logit contributions | |
per | +0.088 |
pper | +0.085 |
ppy | +0.084 |
ilies | +0.084 |
bles | +0.079 |
Neg logit contributions | |
Paula | -0.042 |
Sarah | -0.041 |
Paul | -0.041 |
John | -0.039 |
irens | -0.038 |
Token | however |
Token ID | 2158 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
Paula | +0.047 |
Paul | +0.047 |
irens | +0.045 |
John | +0.045 |
Sarah | +0.044 |
Neg logit contributions | |
ppy | -0.048 |
pper | -0.047 |
per | -0.047 |
ilies | -0.046 |
opsy | -0.045 |
Token | a |
Token ID | 257 |
Feature activation | -8.30e-01 |
Pos logit contributions | |
Paula | +0.016 |
Paul | +0.016 |
irens | +0.016 |
John | +0.015 |
Sarah | +0.015 |
Neg logit contributions | |
ppy | -0.016 |
pper | -0.016 |
per | -0.016 |
opsy | -0.016 |
ilies | -0.016 |
Token | stayed |
Token ID | 9658 |
Feature activation | -4.17e+00 |
Pos logit contributions | |
ectar | +0.025 |
ilies | +0.025 |
ppy | +0.024 |
per | +0.024 |
ily | +0.024 |
Neg logit contributions | |
Paula | -0.021 |
Paul | -0.021 |
John | -0.020 |
spe | -0.020 |
Sarah | -0.020 |
Token | on |
Token ID | 319 |
Feature activation | +3.78e-01 |
Pos logit contributions | |
ppy | +0.286 |
pper | +0.277 |
ilies | +0.273 |
ily | +0.260 |
imes | +0.255 |
Neg logit contributions | |
Paula | -0.322 |
Paul | -0.306 |
irens | -0.304 |
Sarah | -0.303 |
John | -0.295 |
Token | the |
Token ID | 262 |
Feature activation | +7.42e-01 |
Pos logit contributions | |
Paula | +0.030 |
irens | +0.029 |
Paul | +0.028 |
spe | +0.027 |
iren | +0.026 |
Neg logit contributions | |
per | -0.024 |
ppy | -0.024 |
opsy | -0.024 |
ilies | -0.024 |
pper | -0.023 |
Token | l |
Token ID | 300 |
Feature activation | -4.21e+00 |
Pos logit contributions | |
stairs | +0.046 |
hallway | +0.044 |
flashing | +0.044 |
Paula | +0.044 |
doorway | +0.042 |
Neg logit contributions | |
ppy | -0.063 |
oll | -0.056 |
ilies | -0.055 |
ectar | -0.055 |
pper | -0.055 |
Token | ily |
Token ID | 813 |
Feature activation | -7.43e+00 |
Pos logit contributions | |
wait | +0.175 |
ppy | +0.173 |
anuts | +0.165 |
okay | +0.163 |
prett | +0.162 |
Neg logit contributions | |
iren | -0.362 |
irens | -0.361 |
doorway | -0.351 |
stairs | -0.316 |
anne | -0.315 |
Token | pad |
Token ID | 14841 |
Feature activation | -1.12e+01 |
Pos logit contributions | |
ilies | +0.389 |
opsy | +0.380 |
ily | +0.377 |
ectar | +0.371 |
ppy | +0.365 |
Neg logit contributions | |
Paula | -0.665 |
Paul | -0.639 |
Sarah | -0.621 |
John | -0.619 |
James | -0.610 |
Token | and |
Token ID | 290 |
Feature activation | +4.82e-02 |
Pos logit contributions | |
opsy | +0.779 |
ppy | +0.770 |
ilies | +0.767 |
ily | +0.733 |
pper | +0.714 |
Neg logit contributions | |
Paula | -0.773 |
Paul | -0.764 |
John | -0.746 |
Sarah | -0.736 |
Smith | -0.735 |
Token | watched |
Token ID | 7342 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
Paula | +0.004 |
Paul | +0.003 |
John | +0.003 |
Sarah | +0.003 |
James | +0.003 |
Neg logit contributions | |
ppy | -0.003 |
ilies | -0.003 |
pper | -0.003 |
ectar | -0.003 |
per | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | +7.36e-01 |
Pos logit contributions | |
ppy | +0.075 |
pper | +0.074 |
opsy | +0.071 |
ilies | +0.070 |
per | +0.069 |
Neg logit contributions | |
Paula | -0.075 |
Paul | -0.072 |
John | -0.070 |
Sarah | -0.070 |
irens | -0.069 |
Token | butterfly |
Token ID | 35113 |
Feature activation | -9.94e-01 |
Pos logit contributions | |
Paula | +0.053 |
irens | +0.052 |
flashing | +0.051 |
Paul | +0.050 |
spe | +0.050 |
Neg logit contributions | |
ppy | -0.053 |
pper | -0.051 |
per | -0.050 |
oll | -0.047 |
ilies | -0.046 |
Token | . |
Token ID | 13 |
Feature activation | +3.84e-02 |
Pos logit contributions | |
pper | +0.060 |
ppy | +0.060 |
opsy | +0.058 |
ily | +0.057 |
oll | +0.056 |
Neg logit contributions | |
Paula | -0.078 |
Sarah | -0.074 |
Paul | -0.073 |
spe | -0.071 |
flashing | -0.070 |
Token | him |
Token ID | 683 |
Feature activation | -3.65e-01 |
Pos logit contributions | |
ppy | +0.000 |
ilies | +0.000 |
per | +0.000 |
pper | +0.000 |
opsy | +0.000 |
Neg logit contributions | |
Paula | -0.001 |
irens | -0.001 |
Paul | -0.001 |
spe | -0.001 |
John | -0.001 |
Token | on |
Token ID | 319 |
Feature activation | +3.89e-01 |
Pos logit contributions | |
ppy | +0.023 |
pper | +0.022 |
opsy | +0.022 |
ilies | +0.022 |
per | +0.022 |
Neg logit contributions | |
Paula | -0.030 |
Paul | -0.029 |
irens | -0.029 |
John | -0.027 |
spe | -0.027 |
Token | the |
Token ID | 262 |
Feature activation | +7.38e-01 |
Pos logit contributions | |
Paula | +0.034 |
irens | +0.034 |
spe | +0.032 |
iren | +0.031 |
Paul | +0.031 |
Neg logit contributions | |
per | -0.021 |
ppy | -0.020 |
opsy | -0.020 |
ilies | -0.020 |
like | -0.020 |
Token | l |
Token ID | 300 |
Feature activation | -4.20e+00 |
Pos logit contributions | |
irens | +0.048 |
Paula | +0.047 |
stairs | +0.047 |
Paul | +0.047 |
flashing | +0.046 |
Neg logit contributions | |
ppy | -0.059 |
ilies | -0.053 |
ectar | -0.053 |
pper | -0.053 |
per | -0.051 |
Token | ily |
Token ID | 813 |
Feature activation | -7.43e+00 |
Pos logit contributions | |
wait | +0.168 |
prett | +0.168 |
ppy | +0.167 |
okay | +0.162 |
careful | +0.159 |
Neg logit contributions | |
irens | -0.377 |
iren | -0.374 |
doorway | -0.335 |
chairs | -0.321 |
igsaw | -0.319 |
Token | pad |
Token ID | 14841 |
Feature activation | -1.12e+01 |
Pos logit contributions | |
opsy | +0.424 |
ilies | +0.420 |
ppy | +0.417 |
pper | +0.412 |
per | +0.412 |
Neg logit contributions | |
Paula | -0.638 |
Paul | -0.615 |
irens | -0.596 |
Sarah | -0.584 |
John | -0.583 |
Token | . |
Token ID | 13 |
Feature activation | +4.40e-02 |
Pos logit contributions | |
opsy | +0.797 |
ppy | +0.786 |
ilies | +0.772 |
pper | +0.738 |
per | +0.734 |
Neg logit contributions | |
Paula | -0.777 |
Paul | -0.768 |
irens | -0.756 |
spe | -0.743 |
John | -0.729 |
Token | They |
Token ID | 1119 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
Paula | +0.003 |
irens | +0.003 |
Paul | +0.003 |
Sarah | +0.003 |
John | +0.003 |
Neg logit contributions | |
ppy | -0.004 |
pper | -0.004 |
opsy | -0.003 |
per | -0.003 |
ilies | -0.003 |
Token | all |
Token ID | 477 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
Paula | +0.102 |
Paul | +0.098 |
irens | +0.096 |
John | +0.092 |
Sarah | +0.092 |
Neg logit contributions | |
ppy | -0.085 |
per | -0.084 |
pper | -0.083 |
opsy | -0.083 |
ilies | -0.082 |
Token | rested |
Token ID | 29263 |
Feature activation | -5.69e+00 |
Pos logit contributions | |
ppy | +0.138 |
pper | +0.132 |
ilies | +0.132 |
per | +0.131 |
opsy | +0.131 |
Neg logit contributions | |
Paula | -0.151 |
irens | -0.147 |
Paul | -0.142 |
spe | -0.133 |
John | -0.132 |
Token | together |
Token ID | 1978 |
Feature activation | +6.21e-01 |
Pos logit contributions | |
ppy | +0.358 |
ilies | +0.354 |
opsy | +0.350 |
pper | +0.347 |
per | +0.331 |
Neg logit contributions | |
Paula | -0.456 |
Paul | -0.452 |
John | -0.439 |
irens | -0.439 |
Sarah | -0.437 |
Token | opped |
Token ID | 38333 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
ppy | +0.311 |
per | +0.295 |
ilies | +0.293 |
wait | +0.288 |
pper | +0.284 |
Neg logit contributions | |
irens | -0.430 |
Paula | -0.413 |
Paul | -0.408 |
spe | -0.388 |
iren | -0.382 |
Token | onto |
Token ID | 4291 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
ppy | +0.133 |
ectar | +0.130 |
ilies | +0.130 |
pper | +0.130 |
opsy | +0.129 |
Neg logit contributions | |
John | -0.064 |
Paul | -0.064 |
irens | -0.063 |
Sarah | -0.063 |
Paula | -0.061 |
Token | the |
Token ID | 262 |
Feature activation | +7.41e-01 |
Pos logit contributions | |
Paula | +0.067 |
Paul | +0.063 |
irens | +0.062 |
John | +0.059 |
Sarah | +0.059 |
Neg logit contributions | |
ppy | -0.093 |
pper | -0.090 |
ilies | -0.088 |
opsy | -0.088 |
per | -0.087 |
Token | l |
Token ID | 300 |
Feature activation | -4.21e+00 |
Pos logit contributions | |
stairs | +0.044 |
flashing | +0.043 |
hallway | +0.041 |
doorway | +0.041 |
irens | +0.040 |
Neg logit contributions | |
ppy | -0.062 |
ectar | -0.056 |
pper | -0.055 |
ilies | -0.055 |
oll | -0.055 |
Token | ily |
Token ID | 813 |
Feature activation | -7.43e+00 |
Pos logit contributions | |
wait | +0.171 |
ppy | +0.170 |
okay | +0.168 |
anuts | +0.168 |
esty | +0.164 |
Neg logit contributions | |
iren | -0.374 |
irens | -0.365 |
doorway | -0.359 |
stairs | -0.325 |
ons | -0.302 |
Token | pad |
Token ID | 14841 |
Feature activation | -1.12e+01 |
Pos logit contributions | |
opsy | +0.433 |
ilies | +0.424 |
ily | +0.413 |
ectar | +0.407 |
seeker | +0.399 |
Neg logit contributions | |
Paula | -0.565 |
Paul | -0.541 |
Sarah | -0.533 |
Smith | -0.532 |
John | -0.528 |
Token | . |
Token ID | 13 |
Feature activation | +4.18e-02 |
Pos logit contributions | |
opsy | +0.863 |
ilies | +0.835 |
ppy | +0.825 |
thirst | +0.782 |
izards | +0.778 |
Neg logit contributions | |
Paula | -0.619 |
irens | -0.616 |
Paul | -0.614 |
Smith | -0.612 |
spe | -0.608 |
Token |
|
Token ID | 198 |
Feature activation | +2.78e-01 |
Pos logit contributions | |
Paula | +0.003 |
irens | +0.003 |
Paul | +0.003 |
spe | +0.003 |
flashing | +0.003 |
Neg logit contributions | |
ppy | -0.003 |
per | -0.003 |
ilies | -0.003 |
pper | -0.003 |
opsy | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +2.72e-01 |
Pos logit contributions | |
Paula | +0.022 |
irens | +0.021 |
Paul | +0.021 |
spe | +0.020 |
flashing | +0.020 |
Neg logit contributions | |
ppy | -0.018 |
per | -0.018 |
opsy | -0.018 |
ilies | -0.018 |
pper | -0.017 |
Token | Suddenly |
Token ID | 38582 |
Feature activation | -3.12e+00 |
Pos logit contributions | |
Paula | +0.022 |
irens | +0.022 |
Paul | +0.021 |
spe | +0.021 |
Sarah | +0.020 |
Neg logit contributions | |
ilies | -0.018 |
opsy | -0.017 |
ppy | -0.017 |
pper | -0.017 |
ectar | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | -4.64e-01 |
Pos logit contributions | |
ppy | +0.238 |
pper | +0.237 |
ilies | +0.232 |
per | +0.230 |
opsy | +0.228 |
Neg logit contributions | |
Paul | -0.220 |
Paula | -0.217 |
John | -0.212 |
Sarah | -0.211 |
irens | -0.207 |
Token | and |
Token ID | 290 |
Feature activation | +4.96e-02 |
Pos logit contributions | |
Paula | +0.143 |
irens | +0.139 |
Paul | +0.134 |
spe | +0.127 |
John | +0.125 |
Neg logit contributions | |
ppy | -0.081 |
per | -0.076 |
ilies | -0.074 |
pper | -0.074 |
ily | -0.073 |
Token | wanted |
Token ID | 2227 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
Paula | +0.004 |
irens | +0.004 |
Paul | +0.004 |
spe | +0.004 |
John | +0.004 |
Neg logit contributions | |
ppy | -0.003 |
pper | -0.003 |
ilies | -0.003 |
per | -0.003 |
opsy | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | +7.41e-01 |
Pos logit contributions | |
Paula | +0.001 |
irens | +0.001 |
spe | +0.001 |
Paul | +0.001 |
iren | +0.001 |
Neg logit contributions | |
ppy | -0.001 |
ily | -0.001 |
ilies | -0.001 |
per | -0.001 |
pper | -0.001 |
Token | l |
Token ID | 300 |
Feature activation | -4.21e+00 |
Pos logit contributions | |
Paula | +0.057 |
Paul | +0.055 |
irens | +0.054 |
Sarah | +0.053 |
chaos | +0.050 |
Neg logit contributions | |
ppy | -0.050 |
pper | -0.050 |
per | -0.047 |
opsy | -0.045 |
ilies | -0.043 |
Token | ily |
Token ID | 813 |
Feature activation | -7.43e+00 |
Pos logit contributions | |
wait | +0.173 |
thirst | +0.157 |
opsy | +0.144 |
pper | +0.141 |
per | +0.137 |
Neg logit contributions | |
iren | -0.343 |
irens | -0.338 |
Sarah | -0.334 |
smart | -0.319 |
chairs | -0.308 |
Token | pad |
Token ID | 14841 |
Feature activation | -1.12e+01 |
Pos logit contributions | |
ppy | +0.529 |
pper | +0.514 |
per | +0.511 |
ilies | +0.510 |
opsy | +0.509 |
Neg logit contributions | |
Paula | -0.552 |
irens | -0.533 |
Paul | -0.521 |
spe | -0.488 |
John | -0.486 |
Token | for |
Token ID | 329 |
Feature activation | -5.89e+00 |
Pos logit contributions | |
opsy | +0.948 |
ppy | +0.948 |
per | +0.920 |
ilies | +0.917 |
pper | +0.909 |
Neg logit contributions | |
Paula | -0.635 |
Paul | -0.621 |
irens | -0.610 |
spe | -0.605 |
Sarah | -0.591 |
Token | himself |
Token ID | 2241 |
Feature activation | -2.95e+00 |
Pos logit contributions | |
ppy | +0.360 |
pper | +0.345 |
ilies | +0.343 |
opsy | +0.341 |
per | +0.341 |
Neg logit contributions | |
Paula | -0.504 |
irens | -0.490 |
Paul | -0.484 |
John | -0.456 |
Sarah | -0.454 |
Token | . |
Token ID | 13 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
ppy | +0.181 |
pper | +0.175 |
opsy | +0.173 |
ilies | +0.173 |
per | +0.173 |
Neg logit contributions | |
Paula | -0.248 |
irens | -0.240 |
Paul | -0.236 |
spe | -0.224 |
John | -0.222 |
Token |
|
Token ID | 198 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
Paula | +0.003 |
irens | +0.003 |
Paul | +0.003 |
spe | +0.003 |
flashing | +0.003 |
Neg logit contributions | |
ppy | -0.003 |
opsy | -0.003 |
ilies | -0.003 |
per | -0.002 |
pper | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | +2.84e-01 |
Pos logit contributions | |
Paula | +0.024 |
irens | +0.023 |
Paul | +0.022 |
spe | +0.022 |
flashing | +0.022 |
Neg logit contributions | |
ppy | -0.018 |
opsy | -0.017 |
ilies | -0.017 |
ily | -0.017 |
ectar | -0.017 |
Token | said |
Token ID | 531 |
Feature activation | -2.35e+00 |
Pos logit contributions | |
Paula | +0.175 |
Paul | +0.167 |
irens | +0.158 |
John | +0.156 |
Sarah | +0.155 |
Neg logit contributions | |
ectar | -0.165 |
opsy | -0.164 |
ilies | -0.161 |
ppy | -0.160 |
pper | -0.157 |
Token | happily |
Token ID | 18177 |
Feature activation | -2.46e+00 |
Pos logit contributions | |
ppy | +0.126 |
per | +0.123 |
pper | +0.121 |
ilies | +0.120 |
opsy | +0.117 |
Neg logit contributions | |
Paula | -0.212 |
irens | -0.211 |
Paul | -0.200 |
spe | -0.190 |
iren | -0.190 |
Token | . |
Token ID | 13 |
Feature activation | +3.68e-02 |
Pos logit contributions | |
ppy | +0.157 |
opsy | +0.155 |
ilies | +0.152 |
pper | +0.152 |
per | +0.149 |
Neg logit contributions | |
Paula | -0.204 |
irens | -0.195 |
Paul | -0.195 |
spe | -0.185 |
John | -0.183 |
Token | Now |
Token ID | 2735 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
Paula | +0.003 |
irens | +0.003 |
spe | +0.003 |
flashing | +0.003 |
Paul | +0.003 |
Neg logit contributions | |
per | -0.002 |
ppy | -0.002 |
pper | -0.002 |
ilies | -0.002 |
opsy | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -4.69e-01 |
Pos logit contributions | |
irens | +0.117 |
Paula | +0.115 |
iren | +0.108 |
spe | +0.105 |
Paul | +0.105 |
Neg logit contributions | |
ppy | -0.071 |
per | -0.071 |
pper | -0.068 |
ilies | -0.067 |
opsy | -0.064 |
Token | Billy |
Token ID | 15890 |
Feature activation | +7.40e+00 |
Pos logit contributions | |
ppy | +0.029 |
pper | +0.029 |
per | +0.029 |
ilies | +0.028 |
opsy | +0.028 |
Neg logit contributions | |
irens | -0.037 |
Paula | -0.037 |
Paul | -0.033 |
iren | -0.033 |
spe | -0.033 |
Token | and |
Token ID | 290 |
Feature activation | +5.05e-02 |
Pos logit contributions | |
Paula | +0.641 |
irens | +0.632 |
Paul | +0.604 |
spe | +0.578 |
flashing | +0.570 |
Neg logit contributions | |
ppy | -0.443 |
pper | -0.426 |
per | -0.423 |
ilies | -0.410 |
opsy | -0.409 |
Token | his |
Token ID | 465 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
irens | +0.004 |
Paula | +0.004 |
iren | +0.004 |
flashing | +0.004 |
Paul | +0.003 |
Neg logit contributions | |
ppy | -0.003 |
ily | -0.003 |
per | -0.003 |
worm | -0.003 |
ilies | -0.003 |
Token | friends |
Token ID | 2460 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
Paula | +0.154 |
irens | +0.151 |
Paul | +0.149 |
spe | +0.141 |
John | +0.141 |
Neg logit contributions | |
ppy | -0.071 |
pper | -0.068 |
per | -0.067 |
ilies | -0.067 |
opsy | -0.065 |
Token | had |
Token ID | 550 |
Feature activation | -8.46e-02 |
Pos logit contributions | |
ppy | +0.183 |
opsy | +0.179 |
ilies | +0.179 |
pper | +0.178 |
per | +0.177 |
Neg logit contributions | |
Paula | -0.192 |
Paul | -0.182 |
irens | -0.180 |
spe | -0.170 |
John | -0.170 |
Token | a |
Token ID | 257 |
Feature activation | -8.39e-01 |
Pos logit contributions | |
ppy | +0.004 |
ilies | +0.004 |
per | +0.004 |
pper | +0.004 |
ily | +0.004 |
Neg logit contributions | |
Paula | -0.008 |
irens | -0.008 |
Paul | -0.007 |
iren | -0.007 |
spe | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
irens | +0.086 |
Paula | +0.086 |
Paul | +0.082 |
spe | +0.078 |
iren | +0.077 |
Neg logit contributions | |
ppy | -0.057 |
ilies | -0.056 |
per | -0.056 |
pper | -0.055 |
ectar | -0.053 |
Token | It |
Token ID | 632 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
Paula | +0.012 |
Paul | +0.011 |
irens | +0.011 |
Sarah | +0.011 |
John | +0.011 |
Neg logit contributions | |
pper | -0.015 |
ppy | -0.015 |
per | -0.014 |
opsy | -0.014 |
ilies | -0.013 |
Token | seemed |
Token ID | 3947 |
Feature activation | +3.50e+00 |
Pos logit contributions | |
ppy | +0.160 |
per | +0.153 |
ilies | +0.151 |
pper | +0.150 |
opsy | +0.149 |
Neg logit contributions | |
irens | -0.144 |
Paula | -0.138 |
Paul | -0.128 |
iren | -0.123 |
John | -0.121 |
Token | to |
Token ID | 284 |
Feature activation | +7.40e-01 |
Pos logit contributions | |
Paula | +0.364 |
irens | +0.360 |
Paul | +0.348 |
spe | +0.333 |
John | +0.330 |
Neg logit contributions | |
ppy | -0.146 |
ilies | -0.137 |
per | -0.136 |
pper | -0.135 |
opsy | -0.133 |
Token | be |
Token ID | 307 |
Feature activation | -4.05e+00 |
Pos logit contributions | |
Paula | +0.067 |
Paul | +0.064 |
irens | +0.062 |
John | +0.060 |
Sarah | +0.060 |
Neg logit contributions | |
ppy | -0.043 |
ilies | -0.041 |
pper | -0.041 |
opsy | -0.041 |
per | -0.040 |
Token | coming |
Token ID | 2406 |
Feature activation | +8.18e+00 |
Pos logit contributions | |
ppy | +0.261 |
per | +0.251 |
ilies | +0.251 |
pper | +0.248 |
opsy | +0.242 |
Neg logit contributions | |
Paula | -0.325 |
irens | -0.324 |
Paul | -0.309 |
spe | -0.299 |
iren | -0.291 |
Token | from |
Token ID | 422 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
Paula | +0.678 |
irens | +0.647 |
Paul | +0.644 |
spe | +0.604 |
Sarah | +0.601 |
Neg logit contributions | |
ppy | -0.522 |
pper | -0.508 |
ilies | -0.506 |
opsy | -0.501 |
per | -0.498 |
Token | her |
Token ID | 607 |
Feature activation | -3.74e-01 |
Pos logit contributions | |
ppy | +0.016 |
pper | +0.015 |
ilies | +0.015 |
per | +0.015 |
opsy | +0.015 |
Neg logit contributions | |
Paula | -0.020 |
irens | -0.019 |
Paul | -0.019 |
spe | -0.018 |
John | -0.017 |
Token | ears |
Token ID | 11368 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
ppy | +0.024 |
pper | +0.024 |
ilies | +0.023 |
per | +0.023 |
opsy | +0.023 |
Neg logit contributions | |
Paula | -0.030 |
irens | -0.029 |
Paul | -0.029 |
spe | -0.027 |
John | -0.027 |
Token | ! |
Token ID | 0 |
Feature activation | +2.80e+00 |
Pos logit contributions | |
ppy | +0.134 |
pper | +0.130 |
ilies | +0.129 |
opsy | +0.128 |
per | +0.128 |
Neg logit contributions | |
Paula | -0.186 |
irens | -0.180 |
Paul | -0.176 |
spe | -0.166 |
John | -0.165 |
Token | Sandy |
Token ID | 17581 |
Feature activation | -4.22e+00 |
Pos logit contributions | |
Paula | +0.188 |
Paul | +0.182 |
irens | +0.178 |
Sarah | +0.177 |
John | +0.170 |
Neg logit contributions | |
pper | -0.237 |
ppy | -0.229 |
per | -0.211 |
opsy | -0.205 |
comb | -0.203 |
Token | backyard |
Token ID | 24296 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
Paula | +0.124 |
Paul | +0.118 |
irens | +0.116 |
Sarah | +0.111 |
John | +0.111 |
Neg logit contributions | |
ppy | -0.109 |
pper | -0.107 |
ilies | -0.105 |
per | -0.103 |
ectar | -0.100 |
Token | . |
Token ID | 13 |
Feature activation | +8.35e-02 |
Pos logit contributions | |
ppy | +0.113 |
pper | +0.109 |
opsy | +0.108 |
ilies | +0.107 |
per | +0.104 |
Neg logit contributions | |
Paula | -0.142 |
irens | -0.137 |
Paul | -0.136 |
spe | -0.130 |
John | -0.129 |
Token | One |
Token ID | 1881 |
Feature activation | -8.83e-01 |
Pos logit contributions | |
irens | +0.008 |
Paula | +0.008 |
iren | +0.007 |
flashing | +0.007 |
spe | +0.007 |
Neg logit contributions | |
like | -0.004 |
ily | -0.003 |
ilies | -0.003 |
per | -0.003 |
wait | -0.003 |
Token | day |
Token ID | 1110 |
Feature activation | +6.75e-01 |
Pos logit contributions | |
ppy | +0.072 |
per | +0.071 |
pper | +0.071 |
ilies | +0.070 |
opsy | +0.068 |
Neg logit contributions | |
Paula | -0.057 |
Paul | -0.054 |
irens | -0.054 |
spe | -0.051 |
John | -0.050 |
Token | , |
Token ID | 11 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
irens | +0.069 |
Paula | +0.068 |
iren | +0.065 |
spe | +0.063 |
channel | +0.063 |
Neg logit contributions | |
like | -0.026 |
ily | -0.025 |
wait | -0.024 |
careful | -0.022 |
ilies | -0.022 |
Token | John |
Token ID | 1757 |
Feature activation | +8.30e+00 |
Pos logit contributions | |
like | +0.031 |
wait | +0.030 |
careful | +0.028 |
then | +0.027 |
ily | +0.027 |
Neg logit contributions | |
irens | -0.030 |
iren | -0.029 |
Paula | -0.027 |
channel | -0.026 |
jo | -0.025 |
Token | noticed |
Token ID | 6810 |
Feature activation | -4.47e-01 |
Pos logit contributions | |
Paula | +0.667 |
irens | +0.652 |
Paul | +0.628 |
spe | +0.591 |
John | +0.585 |
Neg logit contributions | |
ppy | -0.553 |
pper | -0.530 |
per | -0.528 |
opsy | -0.521 |
ilies | -0.521 |
Token | a |
Token ID | 257 |
Feature activation | -7.81e-01 |
Pos logit contributions | |
ily | +0.018 |
ilies | +0.018 |
worm | +0.018 |
per | +0.018 |
ppy | +0.018 |
Neg logit contributions | |
Paula | -0.045 |
irens | -0.043 |
spe | -0.042 |
Paul | -0.041 |
iren | -0.040 |
Token | helpless |
Token ID | 21144 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
opsy | +0.048 |
ppy | +0.048 |
worm | +0.047 |
pper | +0.047 |
per | +0.046 |
Neg logit contributions | |
Paula | -0.064 |
irens | -0.064 |
Paul | -0.060 |
John | -0.057 |
James | -0.057 |
Token | little |
Token ID | 1310 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
Paula | +0.137 |
irens | +0.134 |
Paul | +0.129 |
spe | +0.122 |
James | +0.121 |
Neg logit contributions | |
ppy | -0.104 |
per | -0.101 |
opsy | -0.100 |
pper | -0.100 |
ilies | -0.099 |
Token | insect |
Token ID | 12833 |
Feature activation | -4.23e+00 |
Pos logit contributions | |
Paula | +0.189 |
irens | +0.184 |
Paul | +0.179 |
spe | +0.168 |
John | +0.167 |
Neg logit contributions | |
ppy | -0.149 |
pper | -0.143 |
per | -0.143 |
ilies | -0.142 |
opsy | -0.142 |
Token | own |
Token ID | 898 |
Feature activation | -3.56e+00 |
Pos logit contributions | |
ppy | +0.025 |
pper | +0.023 |
per | +0.023 |
ilies | +0.023 |
opsy | +0.022 |
Neg logit contributions | |
Paula | -0.037 |
Paul | -0.035 |
irens | -0.035 |
spe | -0.033 |
John | -0.033 |
Token | dove |
Token ID | 38862 |
Feature activation | -7.43e+00 |
Pos logit contributions | |
ppy | +0.236 |
pper | +0.227 |
per | +0.224 |
ilies | +0.224 |
opsy | +0.221 |
Neg logit contributions | |
Paula | -0.286 |
irens | -0.274 |
Paul | -0.271 |
spe | -0.255 |
John | -0.253 |
Token | . |
Token ID | 13 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
ppy | +0.449 |
pper | +0.432 |
ilies | +0.431 |
per | +0.429 |
opsy | +0.428 |
Neg logit contributions | |
Paula | -0.631 |
irens | -0.611 |
Paul | -0.600 |
spe | -0.567 |
Sarah | -0.563 |
Token | She |
Token ID | 1375 |
Feature activation | -7.76e-01 |
Pos logit contributions | |
Paula | +0.010 |
irens | +0.010 |
Paul | +0.010 |
spe | +0.009 |
flashing | +0.009 |
Neg logit contributions | |
ppy | -0.009 |
ilies | -0.009 |
opsy | -0.008 |
per | -0.008 |
pper | -0.008 |
Token | had |
Token ID | 550 |
Feature activation | +7.58e-04 |
Pos logit contributions | |
ppy | +0.061 |
ilies | +0.060 |
ectar | +0.059 |
pper | +0.059 |
per | +0.059 |
Neg logit contributions | |
Paula | -0.052 |
Paul | -0.049 |
irens | -0.047 |
spe | -0.047 |
John | -0.045 |
Token | taken |
Token ID | 2077 |
Feature activation | +7.85e+00 |
Pos logit contributions | |
Paula | +0.000 |
irens | +0.000 |
Paul | +0.000 |
spe | +0.000 |
John | +0.000 |
Neg logit contributions | |
ppy | -0.000 |
ilies | -0.000 |
pper | -0.000 |
per | -0.000 |
opsy | -0.000 |
Token | the |
Token ID | 262 |
Feature activation | +8.06e-01 |
Pos logit contributions | |
Paula | +0.623 |
irens | +0.610 |
Paul | +0.588 |
spe | +0.560 |
John | +0.549 |
Neg logit contributions | |
ppy | -0.519 |
ilies | -0.501 |
per | -0.499 |
pper | -0.499 |
opsy | -0.493 |
Token | first |
Token ID | 717 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
Paula | +0.069 |
Paul | +0.065 |
irens | +0.063 |
flashing | +0.062 |
spe | +0.062 |
Neg logit contributions | |
ppy | -0.047 |
per | -0.043 |
ectar | -0.043 |
ilies | -0.043 |
pper | -0.041 |
Token | step |
Token ID | 2239 |
Feature activation | +2.20e+00 |
Pos logit contributions | |
ppy | +0.080 |
ilies | +0.076 |
per | +0.076 |
pper | +0.074 |
ectar | +0.073 |
Neg logit contributions | |
Paula | -0.079 |
Paul | -0.074 |
irens | -0.073 |
spe | -0.071 |
flashing | -0.070 |
Token | towards |
Token ID | 3371 |
Feature activation | +1.89e+00 |
Pos logit contributions | |
Paula | +0.183 |
irens | +0.176 |
Paul | +0.174 |
spe | +0.164 |
John | +0.164 |
Neg logit contributions | |
ppy | -0.137 |
pper | -0.132 |
ilies | -0.132 |
opsy | -0.132 |
per | -0.131 |
Token | her |
Token ID | 607 |
Feature activation | -4.72e-01 |
Pos logit contributions | |
irens | +0.153 |
Paula | +0.153 |
Paul | +0.143 |
spe | +0.142 |
iren | +0.137 |
Neg logit contributions | |
ppy | -0.120 |
ilies | -0.117 |
per | -0.116 |
pper | -0.115 |
opsy | -0.114 |
Token | reminded |
Token ID | 14516 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
ppy | +0.217 |
ilies | +0.215 |
pper | +0.214 |
opsy | +0.212 |
ectar | +0.211 |
Neg logit contributions | |
Paula | -0.232 |
Paul | -0.220 |
irens | -0.218 |
John | -0.207 |
Sarah | -0.204 |
Token | each |
Token ID | 1123 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
irens | +0.086 |
Paula | +0.086 |
Paul | +0.081 |
spe | +0.080 |
flashing | +0.078 |
Neg logit contributions | |
ppy | -0.063 |
ilies | -0.061 |
opsy | -0.061 |
per | -0.061 |
pper | -0.060 |
Token | other |
Token ID | 584 |
Feature activation | -3.74e+00 |
Pos logit contributions | |
ppy | +0.106 |
ilies | +0.100 |
opsy | +0.098 |
pper | +0.098 |
per | +0.097 |
Neg logit contributions | |
irens | -0.257 |
Paula | -0.257 |
Paul | -0.248 |
spe | -0.240 |
flashing | -0.234 |
Token | that |
Token ID | 326 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
ppy | +0.233 |
ilies | +0.223 |
per | +0.221 |
opsy | +0.218 |
pper | +0.218 |
Neg logit contributions | |
Paula | -0.308 |
irens | -0.307 |
Paul | -0.291 |
spe | -0.276 |
iren | -0.274 |
Token | no |
Token ID | 645 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
irens | +0.095 |
Paula | +0.093 |
Paul | +0.088 |
iren | +0.086 |
spe | +0.083 |
Neg logit contributions | |
ppy | -0.071 |
ilies | -0.070 |
ectar | -0.070 |
opsy | -0.069 |
per | -0.068 |
Token | matter |
Token ID | 2300 |
Feature activation | +8.70e+00 |
Pos logit contributions | |
like | +0.108 |
ppy | +0.107 |
opsy | +0.107 |
worm | +0.106 |
wait | +0.106 |
Neg logit contributions | |
irens | -0.111 |
Paula | -0.105 |
iren | -0.098 |
Paul | -0.096 |
flashing | -0.090 |
Token | what |
Token ID | 644 |
Feature activation | +4.40e+00 |
Pos logit contributions | |
irens | +0.719 |
Paula | +0.701 |
iren | +0.652 |
Paul | +0.646 |
spe | +0.642 |
Neg logit contributions | |
ppy | -0.551 |
ilies | -0.524 |
per | -0.523 |
ectar | -0.514 |
pper | -0.511 |
Token | they |
Token ID | 484 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
irens | +0.368 |
Paula | +0.364 |
Paul | +0.340 |
iren | +0.338 |
spe | +0.330 |
Neg logit contributions | |
ppy | -0.254 |
ilies | -0.248 |
opsy | -0.248 |
per | -0.244 |
ectar | -0.235 |
Token | faced |
Token ID | 7452 |
Feature activation | -2.83e-01 |
Pos logit contributions | |
Paula | +0.194 |
Paul | +0.186 |
irens | +0.186 |
John | +0.176 |
James | +0.173 |
Neg logit contributions | |
ppy | -0.147 |
opsy | -0.144 |
ilies | -0.144 |
pper | -0.143 |
ectar | -0.143 |
Token | , |
Token ID | 11 |
Feature activation | -3.98e-01 |
Pos logit contributions | |
per | +0.013 |
ppy | +0.013 |
ilies | +0.013 |
like | +0.012 |
pper | +0.012 |
Neg logit contributions | |
irens | -0.027 |
Paula | -0.027 |
Paul | -0.025 |
iren | -0.025 |
flashing | -0.024 |
Token | love |
Token ID | 1842 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
ppy | +0.025 |
ilies | +0.024 |
per | +0.024 |
opsy | +0.024 |
ectar | +0.024 |
Neg logit contributions | |
irens | -0.032 |
Paula | -0.032 |
Paul | -0.030 |
iren | -0.029 |
flashing | -0.028 |
Token | the |
Token ID | 262 |
Feature activation | +7.36e-01 |
Pos logit contributions | |
ppy | +0.025 |
per | +0.025 |
ilies | +0.024 |
pper | +0.024 |
opsy | +0.024 |
Neg logit contributions | |
Paula | -0.032 |
irens | -0.031 |
Paul | -0.030 |
spe | -0.029 |
John | -0.028 |
Token | darkness |
Token ID | 11854 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
Paula | +0.053 |
irens | +0.050 |
Paul | +0.050 |
spe | +0.046 |
Sarah | +0.046 |
Neg logit contributions | |
ppy | -0.055 |
pper | -0.054 |
ilies | -0.054 |
per | -0.053 |
opsy | -0.053 |
Token | . |
Token ID | 13 |
Feature activation | +3.55e-02 |
Pos logit contributions | |
Paula | +0.161 |
irens | +0.157 |
Paul | +0.153 |
spe | +0.144 |
John | +0.144 |
Neg logit contributions | |
ppy | -0.111 |
pper | -0.107 |
per | -0.106 |
ilies | -0.106 |
opsy | -0.105 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.10e+00 |
Pos logit contributions | |
irens | +0.003 |
Paula | +0.003 |
Paul | +0.003 |
spe | +0.002 |
flashing | +0.002 |
Neg logit contributions | |
ppy | -0.002 |
ilies | -0.002 |
opsy | -0.002 |
per | -0.002 |
ectar | -0.002 |
Token | Bob |
Token ID | 18861 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
ily | +0.242 |
ppy | +0.241 |
ilies | +0.238 |
per | +0.235 |
opsy | +0.232 |
Neg logit contributions | |
Paula | -0.343 |
irens | -0.331 |
spe | -0.322 |
Paul | -0.318 |
flashing | -0.312 |
Token | was |
Token ID | 373 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
ppy | +0.074 |
pper | +0.071 |
per | +0.071 |
ilies | +0.071 |
opsy | +0.071 |
Neg logit contributions | |
Paula | -0.081 |
irens | -0.078 |
Paul | -0.076 |
spe | -0.071 |
John | -0.071 |
Token | a |
Token ID | 257 |
Feature activation | -8.31e-01 |
Pos logit contributions | |
Paula | +0.114 |
Paul | +0.108 |
irens | +0.107 |
John | +0.103 |
Sarah | +0.103 |
Neg logit contributions | |
ppy | -0.107 |
pper | -0.105 |
ilies | -0.101 |
per | -0.100 |
opsy | -0.099 |
Token | young |
Token ID | 1862 |
Feature activation | +4.92e-01 |
Pos logit contributions | |
ilies | +0.059 |
ppy | +0.058 |
pper | +0.058 |
per | +0.056 |
ectar | +0.056 |
Neg logit contributions | |
Paula | -0.063 |
Paul | -0.060 |
irens | -0.057 |
John | -0.056 |
spe | -0.056 |
Token | boy |
Token ID | 2933 |
Feature activation | +2.85e+00 |
Pos logit contributions | |
Paula | +0.032 |
irens | +0.030 |
Paul | +0.030 |
spe | +0.027 |
John | +0.027 |
Neg logit contributions | |
ppy | -0.040 |
pper | -0.039 |
ilies | -0.039 |
per | -0.039 |
opsy | -0.039 |
Token | who |
Token ID | 508 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
Paula | +0.226 |
Paul | +0.220 |
irens | +0.215 |
John | +0.208 |
Sarah | +0.206 |
Neg logit contributions | |
ppy | -0.185 |
ilies | -0.183 |
pper | -0.181 |
opsy | -0.180 |
ectar | -0.175 |
Token | loved |
Token ID | 6151 |
Feature activation | -9.47e-01 |
Pos logit contributions | |
Paula | +0.154 |
irens | +0.149 |
Paul | +0.147 |
spe | +0.139 |
John | +0.138 |
Neg logit contributions | |
ppy | -0.110 |
pper | -0.107 |
ilies | -0.106 |
per | -0.105 |
opsy | -0.104 |
Token | Before |
Token ID | 8421 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
Paula | +0.024 |
irens | +0.023 |
Paul | +0.022 |
Sarah | +0.021 |
John | +0.021 |
Neg logit contributions | |
ppy | -0.019 |
pper | -0.018 |
per | -0.018 |
opsy | -0.018 |
ilies | -0.018 |
Token | he |
Token ID | 339 |
Feature activation | -3.21e-01 |
Pos logit contributions | |
wait | +0.092 |
like | +0.089 |
ily | +0.083 |
opsy | +0.083 |
thirst | +0.083 |
Neg logit contributions | |
spe | -0.149 |
Paula | -0.138 |
irens | -0.137 |
channel | -0.134 |
telev | -0.133 |
Token | left |
Token ID | 1364 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
ppy | +0.020 |
pper | +0.019 |
per | +0.019 |
ily | +0.019 |
opsy | +0.018 |
Neg logit contributions | |
irens | -0.028 |
Paula | -0.027 |
Paul | -0.025 |
spe | -0.025 |
flashing | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | -4.56e-01 |
Pos logit contributions | |
per | +0.060 |
ily | +0.054 |
like | +0.054 |
ppy | +0.053 |
ilies | +0.053 |
Neg logit contributions | |
irens | -0.132 |
Paula | -0.130 |
spe | -0.127 |
flashing | -0.119 |
Paul | -0.119 |
Token | he |
Token ID | 339 |
Feature activation | -3.19e-01 |
Pos logit contributions | |
like | +0.028 |
wait | +0.027 |
careful | +0.027 |
still | +0.026 |
then | +0.024 |
Neg logit contributions | |
irens | -0.036 |
iren | -0.034 |
Paula | -0.034 |
channel | -0.032 |
spe | -0.031 |
Token | heard |
Token ID | 2982 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
ppy | +0.023 |
pper | +0.022 |
per | +0.022 |
ilies | +0.022 |
opsy | +0.022 |
Neg logit contributions | |
Paula | -0.024 |
irens | -0.023 |
Paul | -0.023 |
spe | -0.021 |
John | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | +7.65e-01 |
Pos logit contributions | |
irens | +0.147 |
Paula | +0.147 |
spe | +0.139 |
iren | +0.135 |
Paul | +0.135 |
Neg logit contributions | |
ily | -0.076 |
ilies | -0.074 |
per | -0.073 |
ppy | -0.072 |
like | -0.071 |
Token | voice |
Token ID | 3809 |
Feature activation | +3.05e+00 |
Pos logit contributions | |
Paula | +0.051 |
irens | +0.049 |
Paul | +0.047 |
spe | +0.044 |
John | +0.044 |
Neg logit contributions | |
ily | -0.057 |
per | -0.056 |
ppy | -0.056 |
ilies | -0.056 |
opsy | -0.055 |
Token | again |
Token ID | 757 |
Feature activation | +9.94e-01 |
Pos logit contributions | |
irens | +0.282 |
Paula | +0.276 |
Paul | +0.261 |
spe | +0.254 |
telev | +0.250 |
Neg logit contributions | |
like | -0.149 |
ppy | -0.148 |
per | -0.147 |
ilies | -0.144 |
pper | -0.142 |
Token | : |
Token ID | 25 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
Paula | +0.084 |
irens | +0.081 |
Paul | +0.080 |
John | +0.075 |
Sarah | +0.075 |
Neg logit contributions | |
ppy | -0.062 |
pper | -0.059 |
ilies | -0.059 |
opsy | -0.059 |
per | -0.058 |
Token | " |
Token ID | 366 |
Feature activation | -7.01e-01 |
Pos logit contributions | |
ppy | +0.015 |
ilies | +0.015 |
per | +0.015 |
pper | +0.015 |
opsy | +0.015 |
Neg logit contributions | |
irens | -0.018 |
Paula | -0.018 |
Paul | -0.016 |
spe | -0.016 |
iren | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | +7.35e-01 |
Pos logit contributions | |
like | +0.144 |
per | +0.140 |
ily | +0.137 |
ectar | +0.136 |
worm | +0.135 |
Neg logit contributions | |
irens | -0.261 |
Paula | -0.257 |
iren | -0.245 |
spe | -0.243 |
Paul | -0.240 |
Token | sky |
Token ID | 6766 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
Paula | +0.057 |
irens | +0.055 |
Paul | +0.054 |
spe | +0.050 |
John | +0.050 |
Neg logit contributions | |
ppy | -0.050 |
per | -0.048 |
pper | -0.048 |
opsy | -0.048 |
ilies | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | +3.65e-02 |
Pos logit contributions | |
ppy | +0.101 |
opsy | +0.099 |
pper | +0.099 |
ilies | +0.098 |
per | +0.096 |
Neg logit contributions | |
Paula | -0.119 |
Paul | -0.113 |
irens | -0.112 |
John | -0.107 |
Sarah | -0.107 |
Token | When |
Token ID | 1649 |
Feature activation | +5.80e-01 |
Pos logit contributions | |
Paula | +0.003 |
irens | +0.003 |
Paul | +0.002 |
spe | +0.002 |
John | +0.002 |
Neg logit contributions | |
ppy | -0.003 |
pper | -0.003 |
ilies | -0.003 |
opsy | -0.003 |
per | -0.003 |
Token | she |
Token ID | 673 |
Feature activation | -5.54e-01 |
Pos logit contributions | |
irens | +0.050 |
Paula | +0.050 |
Paul | +0.047 |
spe | +0.047 |
James | +0.046 |
Neg logit contributions | |
opsy | -0.032 |
ilies | -0.031 |
ppy | -0.031 |
pper | -0.031 |
ily | -0.030 |
Token | woke |
Token ID | 19092 |
Feature activation | +2.81e+00 |
Pos logit contributions | |
ilies | +0.052 |
ppy | +0.052 |
ectar | +0.051 |
per | +0.051 |
pper | +0.051 |
Neg logit contributions | |
Paula | -0.027 |
Paul | -0.025 |
irens | -0.024 |
John | -0.023 |
Sarah | -0.022 |
Token | up |
Token ID | 510 |
Feature activation | +2.62e+00 |
Pos logit contributions | |
Paula | +0.178 |
Paul | +0.166 |
irens | +0.166 |
John | +0.153 |
Sarah | +0.153 |
Neg logit contributions | |
ppy | -0.243 |
pper | -0.234 |
ilies | -0.233 |
per | -0.230 |
ectar | -0.227 |
Token | , |
Token ID | 11 |
Feature activation | -4.72e-01 |
Pos logit contributions | |
Paula | +0.250 |
irens | +0.250 |
Paul | +0.239 |
spe | +0.234 |
James | +0.228 |
Neg logit contributions | |
ppy | -0.122 |
per | -0.120 |
ilies | -0.120 |
opsy | -0.120 |
ectar | -0.115 |
Token | she |
Token ID | 673 |
Feature activation | -5.42e-01 |
Pos logit contributions | |
opsy | +0.027 |
ppy | +0.027 |
ilies | +0.027 |
pper | +0.026 |
per | +0.026 |
Neg logit contributions | |
irens | -0.040 |
Paula | -0.040 |
Paul | -0.038 |
spe | -0.037 |
iren | -0.036 |
Token | saw |
Token ID | 2497 |
Feature activation | -7.10e-01 |
Pos logit contributions | |
ilies | +0.045 |
ectar | +0.044 |
ppy | +0.044 |
pper | +0.043 |
imes | +0.043 |
Neg logit contributions | |
Paula | -0.034 |
Paul | -0.032 |
spe | -0.029 |
John | -0.029 |
Sarah | -0.029 |
Token | the |
Token ID | 262 |
Feature activation | +7.41e-01 |
Pos logit contributions | |
ppy | +0.042 |
pper | +0.041 |
ilies | +0.041 |
per | +0.040 |
opsy | +0.040 |
Neg logit contributions | |
Paula | -0.062 |
irens | -0.060 |
Paul | -0.058 |
spe | -0.055 |
John | -0.055 |
Token | hear |
Token ID | 3285 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
Paula | +0.183 |
Paul | +0.173 |
irens | +0.170 |
John | +0.167 |
Sarah | +0.166 |
Neg logit contributions | |
ectar | -0.149 |
opsy | -0.147 |
pper | -0.142 |
ppy | -0.141 |
ilies | -0.141 |
Token | a |
Token ID | 257 |
Feature activation | -8.19e-01 |
Pos logit contributions | |
Paula | +0.171 |
irens | +0.165 |
spe | +0.161 |
Paul | +0.159 |
iren | +0.152 |
Neg logit contributions | |
ilies | -0.077 |
ppy | -0.076 |
ily | -0.076 |
per | -0.075 |
opsy | -0.072 |
Token | loud |
Token ID | 7812 |
Feature activation | +3.48e+00 |
Pos logit contributions | |
ppy | +0.058 |
opsy | +0.057 |
per | +0.056 |
pper | +0.056 |
worm | +0.056 |
Neg logit contributions | |
Paula | -0.059 |
irens | -0.058 |
Paul | -0.055 |
John | -0.052 |
spe | -0.051 |
Token | crack |
Token ID | 8469 |
Feature activation | -2.43e+00 |
Pos logit contributions | |
Paula | +0.232 |
irens | +0.222 |
Paul | +0.214 |
spe | +0.209 |
John | +0.196 |
Neg logit contributions | |
ppy | -0.272 |
per | -0.269 |
pper | -0.264 |
ilies | -0.257 |
opsy | -0.256 |
Token | . |
Token ID | 13 |
Feature activation | +5.67e-02 |
Pos logit contributions | |
ppy | +0.146 |
per | +0.141 |
pper | +0.141 |
ilies | +0.140 |
opsy | +0.138 |
Neg logit contributions | |
Paula | -0.211 |
irens | -0.203 |
Paul | -0.201 |
John | -0.188 |
spe | -0.188 |
Token | They |
Token ID | 1119 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
Paula | +0.005 |
irens | +0.005 |
Paul | +0.004 |
spe | +0.004 |
James | +0.004 |
Neg logit contributions | |
ilies | -0.003 |
ppy | -0.003 |
ily | -0.003 |
per | -0.003 |
opsy | -0.003 |
Token | run |
Token ID | 1057 |
Feature activation | +7.72e-01 |
Pos logit contributions | |
Paula | +0.109 |
Paul | +0.104 |
irens | +0.102 |
John | +0.099 |
flashing | +0.097 |
Neg logit contributions | |
ectar | -0.083 |
opsy | -0.081 |
ppy | -0.081 |
ilies | -0.081 |
pper | -0.080 |
Token | to |
Token ID | 284 |
Feature activation | +6.14e-01 |
Pos logit contributions | |
Paula | +0.060 |
irens | +0.058 |
Paul | +0.057 |
spe | +0.054 |
John | +0.053 |
Neg logit contributions | |
ppy | -0.052 |
per | -0.051 |
pper | -0.050 |
ilies | -0.050 |
ily | -0.049 |
Token | the |
Token ID | 262 |
Feature activation | +7.52e-01 |
Pos logit contributions | |
Paula | +0.049 |
irens | +0.048 |
Paul | +0.045 |
spe | +0.045 |
iren | +0.043 |
Neg logit contributions | |
ppy | -0.040 |
per | -0.039 |
opsy | -0.039 |
ilies | -0.039 |
pper | -0.039 |
Token | tree |
Token ID | 5509 |
Feature activation | -2.39e+00 |
Pos logit contributions | |
Paula | +0.051 |
flashing | +0.048 |
irens | +0.048 |
Paul | +0.047 |
Smith | +0.047 |
Neg logit contributions | |
ppy | -0.062 |
ilies | -0.059 |
pper | -0.057 |
ectar | -0.056 |
izards | -0.055 |
Token | house |
Token ID | 2156 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
ppy | +0.139 |
pper | +0.134 |
per | +0.134 |
ilies | +0.133 |
opsy | +0.132 |
Neg logit contributions | |
Paula | -0.210 |
irens | -0.205 |
Paul | -0.200 |
spe | -0.189 |
John | -0.188 |
Token | animals |
Token ID | 4695 |
Feature activation | -4.91e-01 |
Pos logit contributions | |
ppy | +0.208 |
pper | +0.194 |
per | +0.190 |
comb | +0.186 |
ilies | +0.186 |
Neg logit contributions | |
Paula | -0.339 |
irens | -0.333 |
Paul | -0.333 |
spe | -0.318 |
Sarah | -0.310 |
Token | in |
Token ID | 287 |
Feature activation | -2.85e+00 |
Pos logit contributions | |
ppy | +0.033 |
pper | +0.032 |
opsy | +0.032 |
ilies | +0.032 |
per | +0.032 |
Neg logit contributions | |
Paula | -0.038 |
irens | -0.037 |
Paul | -0.036 |
spe | -0.035 |
John | -0.034 |
Token | the |
Token ID | 262 |
Feature activation | +7.42e-01 |
Pos logit contributions | |
per | +0.162 |
ppy | +0.161 |
ilies | +0.158 |
opsy | +0.154 |
ily | +0.152 |
Neg logit contributions | |
irens | -0.244 |
Paula | -0.232 |
iren | -0.225 |
Paul | -0.221 |
spe | -0.221 |
Token | pond |
Token ID | 16723 |
Feature activation | -6.86e+00 |
Pos logit contributions | |
Paula | +0.060 |
irens | +0.059 |
Paul | +0.057 |
spe | +0.055 |
Sarah | +0.054 |
Neg logit contributions | |
ppy | -0.049 |
pper | -0.047 |
ilies | -0.047 |
per | -0.045 |
ectar | -0.045 |
Token | were |
Token ID | 547 |
Feature activation | -3.32e-02 |
Pos logit contributions | |
ppy | +0.490 |
pper | +0.475 |
ilies | +0.472 |
per | +0.471 |
opsy | +0.469 |
Neg logit contributions | |
Paula | -0.510 |
irens | -0.493 |
Paul | -0.481 |
spe | -0.449 |
John | -0.446 |
Token | sad |
Token ID | 6507 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
ppy | +0.002 |
ilies | +0.002 |
per | +0.002 |
pper | +0.002 |
opsy | +0.002 |
Neg logit contributions | |
Paula | -0.003 |
irens | -0.003 |
Paul | -0.003 |
spe | -0.003 |
iren | -0.003 |
Token | that |
Token ID | 326 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
Paula | +0.118 |
irens | +0.116 |
Paul | +0.112 |
spe | +0.107 |
John | +0.105 |
Neg logit contributions | |
ppy | -0.068 |
per | -0.065 |
ilies | -0.065 |
opsy | -0.064 |
pper | -0.064 |
Token | the |
Token ID | 262 |
Feature activation | +7.34e-01 |
Pos logit contributions | |
irens | +0.085 |
Paula | +0.082 |
iren | +0.077 |
Paul | +0.077 |
flashing | +0.074 |
Neg logit contributions | |
ilies | -0.074 |
ppy | -0.074 |
per | -0.071 |
ily | -0.071 |
opsy | -0.070 |
Token | sw |
Token ID | 1509 |
Feature activation | -8.10e+00 |
Pos logit contributions | |
Paula | +0.064 |
irens | +0.062 |
Paul | +0.061 |
spe | +0.058 |
John | +0.057 |
Neg logit contributions | |
ppy | -0.044 |
pper | -0.042 |
ilies | -0.041 |
per | -0.041 |
opsy | -0.040 |
Token | an |
Token ID | 272 |
Feature activation | -3.78e+00 |
Pos logit contributions | |
ppy | +0.368 |
ilies | +0.328 |
pper | +0.326 |
comb | +0.325 |
ectar | +0.321 |
Neg logit contributions | |
irens | -0.795 |
Paula | -0.777 |
Paul | -0.770 |
spe | -0.758 |
iren | -0.732 |
Token | was |
Token ID | 373 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
ppy | +0.262 |
pper | +0.253 |
per | +0.249 |
ilies | +0.246 |
opsy | +0.245 |
Neg logit contributions | |
Paula | -0.293 |
irens | -0.290 |
Paul | -0.271 |
iren | -0.254 |
spe | -0.252 |
Token | my |
Token ID | 1820 |
Feature activation | +4.87e+00 |
Pos logit contributions | |
ppy | +0.019 |
per | +0.018 |
pper | +0.018 |
opsy | +0.018 |
ilies | +0.018 |
Neg logit contributions | |
Paula | -0.019 |
irens | -0.018 |
Paul | -0.017 |
spe | -0.017 |
flashing | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +8.33e-02 |
Pos logit contributions | |
Paula | +0.390 |
irens | +0.373 |
Paul | +0.373 |
John | +0.353 |
Sarah | +0.349 |
Neg logit contributions | |
ilies | -0.314 |
ppy | -0.312 |
pper | -0.305 |
opsy | -0.304 |
per | -0.298 |
Token | It |
Token ID | 632 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
irens | +0.007 |
Paula | +0.006 |
Paul | +0.006 |
spe | +0.006 |
iren | +0.006 |
Neg logit contributions | |
ppy | -0.005 |
ilies | -0.005 |
ectar | -0.005 |
opsy | -0.005 |
per | -0.005 |
Token | was |
Token ID | 373 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
ppy | +0.152 |
opsy | +0.149 |
ily | +0.147 |
worm | +0.146 |
pper | +0.142 |
Neg logit contributions | |
irens | -0.165 |
Paula | -0.165 |
Paul | -0.153 |
spe | -0.142 |
iren | -0.141 |
Token | a |
Token ID | 257 |
Feature activation | -7.85e-01 |
Pos logit contributions | |
Paula | +0.143 |
irens | +0.140 |
Paul | +0.137 |
spe | +0.132 |
James | +0.128 |
Neg logit contributions | |
ppy | -0.077 |
ilies | -0.075 |
opsy | -0.074 |
per | -0.074 |
pper | -0.074 |
Token | rainy |
Token ID | 37259 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
ppy | +0.050 |
pper | +0.048 |
opsy | +0.048 |
per | +0.048 |
ilies | +0.047 |
Neg logit contributions | |
Paula | -0.065 |
irens | -0.063 |
Paul | -0.061 |
spe | -0.058 |
John | -0.057 |
Token | day |
Token ID | 1110 |
Feature activation | +6.75e-01 |
Pos logit contributions | |
ppy | +0.100 |
opsy | +0.097 |
pper | +0.097 |
per | +0.097 |
ilies | +0.096 |
Neg logit contributions | |
Paula | -0.123 |
irens | -0.118 |
Paul | -0.114 |
spe | -0.110 |
James | -0.106 |
Token | and |
Token ID | 290 |
Feature activation | +8.43e-02 |
Pos logit contributions | |
Paula | +0.058 |
irens | +0.056 |
Paul | +0.056 |
Sarah | +0.053 |
John | +0.053 |
Neg logit contributions | |
ppy | -0.041 |
pper | -0.040 |
opsy | -0.039 |
ilies | -0.039 |
per | -0.039 |
Token | there |
Token ID | 612 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
Paula | +0.006 |
irens | +0.006 |
Paul | +0.006 |
spe | +0.006 |
iren | +0.006 |
Neg logit contributions | |
ppy | -0.006 |
opsy | -0.006 |
ilies | -0.006 |
pper | -0.006 |
per | -0.006 |
Token | were |
Token ID | 547 |
Feature activation | -8.47e-03 |
Pos logit contributions | |
Paula | +0.076 |
irens | +0.076 |
Paul | +0.070 |
spe | +0.066 |
iren | +0.065 |
Neg logit contributions | |
ppy | -0.099 |
opsy | -0.098 |
ilies | -0.098 |
per | -0.096 |
pper | -0.096 |
Token | many |
Token ID | 867 |
Feature activation | -6.94e-01 |
Pos logit contributions | |
ppy | +0.000 |
ilies | +0.000 |
opsy | +0.000 |
ectar | +0.000 |
ily | +0.000 |
Neg logit contributions | |
Paula | -0.001 |
irens | -0.001 |
Paul | -0.001 |
spe | -0.001 |
James | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | -3.50e-01 |
Pos logit contributions | |
ily | +0.219 |
ilies | +0.187 |
ppy | +0.185 |
like | +0.180 |
oll | +0.179 |
Neg logit contributions | |
Paula | -0.563 |
irens | -0.541 |
spe | -0.528 |
Paul | -0.525 |
flashing | -0.509 |
Token | Tim |
Token ID | 5045 |
Feature activation | +1.94e+00 |
Pos logit contributions | |
opsy | +0.026 |
worm | +0.025 |
ily | +0.025 |
pper | +0.025 |
like | +0.025 |
Neg logit contributions | |
Paula | -0.024 |
irens | -0.023 |
spe | -0.021 |
iren | -0.021 |
channel | -0.020 |
Token | my |
Token ID | 1820 |
Feature activation | +4.94e+00 |
Pos logit contributions | |
Paula | +0.121 |
irens | +0.120 |
Paul | +0.109 |
spe | +0.106 |
flashing | +0.106 |
Neg logit contributions | |
ppy | -0.176 |
pper | -0.165 |
opsy | -0.160 |
per | -0.160 |
ily | -0.156 |
Token | went |
Token ID | 1816 |
Feature activation | +2.20e+00 |
Pos logit contributions | |
Paula | +0.350 |
irens | +0.338 |
Paul | +0.338 |
John | +0.325 |
Sarah | +0.324 |
Neg logit contributions | |
ilies | -0.353 |
ppy | -0.350 |
ectar | -0.348 |
pper | -0.345 |
opsy | -0.340 |
Token | back |
Token ID | 736 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
Paula | +0.188 |
Paul | +0.179 |
irens | +0.179 |
spe | +0.174 |
John | +0.165 |
Neg logit contributions | |
ppy | -0.135 |
per | -0.131 |
ilies | -0.130 |
pper | -0.127 |
opsy | -0.126 |
Token | outside |
Token ID | 2354 |
Feature activation | +6.65e-01 |
Pos logit contributions | |
ppy | +0.064 |
per | +0.063 |
opsy | +0.062 |
ectar | +0.061 |
ilies | +0.061 |
Neg logit contributions | |
Paula | -0.108 |
irens | -0.105 |
spe | -0.104 |
Paul | -0.100 |
flashing | -0.096 |
Token | to |
Token ID | 284 |
Feature activation | +7.29e-01 |
Pos logit contributions | |
Paula | +0.059 |
irens | +0.058 |
Paul | +0.055 |
spe | +0.053 |
Ellen | +0.052 |
Neg logit contributions | |
per | -0.037 |
ppy | -0.036 |
ectar | -0.036 |
like | -0.036 |
ilies | -0.035 |
Token | play |
Token ID | 711 |
Feature activation | -2.83e-01 |
Pos logit contributions | |
Paula | +0.064 |
irens | +0.062 |
Paul | +0.060 |
spe | +0.058 |
flashing | +0.056 |
Neg logit contributions | |
ppy | -0.043 |
ilies | -0.041 |
per | -0.041 |
opsy | -0.041 |
pper | -0.041 |
Token | with |
Token ID | 351 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
pper | +0.027 |
opsy | +0.026 |
ppy | +0.026 |
ilies | +0.025 |
ani | +0.024 |
Neg logit contributions | |
Sarah | -0.014 |
irens | -0.013 |
James | -0.012 |
cartoon | -0.012 |
John | -0.011 |
Token | his |
Token ID | 465 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
Paula | +0.017 |
irens | +0.016 |
Paul | +0.015 |
spe | +0.014 |
John | +0.014 |
Neg logit contributions | |
ppy | -0.026 |
pper | -0.026 |
ilies | -0.026 |
per | -0.026 |
opsy | -0.026 |
Token | vehicles |
Token ID | 5672 |
Feature activation | +6.40e+00 |
Pos logit contributions | |
Paula | +0.151 |
irens | +0.145 |
Paul | +0.136 |
spe | +0.134 |
iren | +0.130 |
Neg logit contributions | |
opsy | -0.098 |
worm | -0.095 |
ilies | -0.094 |
ppy | -0.094 |
pper | -0.093 |
Token |
|
Token ID | 198 |
Feature activation | +3.02e-01 |
Pos logit contributions | |
Paula | +0.024 |
irens | +0.024 |
Paul | +0.023 |
spe | +0.023 |
flashing | +0.022 |
Neg logit contributions | |
ppy | -0.018 |
opsy | -0.018 |
ilies | -0.018 |
per | -0.018 |
ily | -0.018 |
Token | " |
Token ID | 1 |
Feature activation | -3.51e+00 |
Pos logit contributions | |
irens | +0.023 |
Paula | +0.022 |
Paul | +0.021 |
Sarah | +0.021 |
John | +0.020 |
Neg logit contributions | |
ilies | -0.023 |
ppy | -0.023 |
ectar | -0.022 |
opsy | -0.021 |
oll | -0.021 |
Token | Good |
Token ID | 10248 |
Feature activation | +3.30e+00 |
Pos logit contributions | |
ppy | +0.268 |
pper | +0.258 |
ilies | +0.258 |
per | +0.256 |
opsy | +0.256 |
Neg logit contributions | |
Paula | -0.250 |
irens | -0.240 |
Paul | -0.233 |
spe | -0.221 |
John | -0.216 |
Token | idea |
Token ID | 2126 |
Feature activation | +7.16e-01 |
Pos logit contributions | |
Paul | +0.213 |
Paula | +0.211 |
John | +0.199 |
Sarah | +0.198 |
irens | +0.195 |
Neg logit contributions | |
ilies | -0.267 |
ppy | -0.264 |
pper | -0.258 |
per | -0.255 |
opsy | -0.255 |
Token | !" |
Token ID | 2474 |
Feature activation | +6.96e-02 |
Pos logit contributions | |
Paula | +0.064 |
irens | +0.062 |
Paul | +0.060 |
spe | +0.058 |
Sarah | +0.056 |
Neg logit contributions | |
ppy | -0.040 |
pper | -0.039 |
ilies | -0.039 |
per | -0.038 |
opsy | -0.038 |
Token | Tom |
Token ID | 4186 |
Feature activation | +6.83e-01 |
Pos logit contributions | |
Paula | +0.005 |
irens | +0.005 |
Paul | +0.005 |
spe | +0.005 |
flashing | +0.004 |
Neg logit contributions | |
ppy | -0.005 |
ilies | -0.005 |
per | -0.005 |
pper | -0.005 |
opsy | -0.005 |
Token | said |
Token ID | 531 |
Feature activation | -2.31e+00 |
Pos logit contributions | |
Paula | +0.057 |
irens | +0.056 |
Paul | +0.054 |
spe | +0.051 |
John | +0.050 |
Neg logit contributions | |
ppy | -0.043 |
pper | -0.042 |
per | -0.041 |
ilies | -0.041 |
opsy | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | +7.28e-02 |
Pos logit contributions | |
ppy | +0.105 |
per | +0.101 |
ily | +0.098 |
ilies | +0.096 |
pper | +0.090 |
Neg logit contributions | |
irens | -0.231 |
Paula | -0.230 |
Paul | -0.216 |
flashing | -0.210 |
iren | -0.207 |
Token | " |
Token ID | 366 |
Feature activation | -6.94e-01 |
Pos logit contributions | |
Paula | +0.007 |
irens | +0.006 |
Paul | +0.006 |
flashing | +0.006 |
spe | +0.006 |
Neg logit contributions | |
ily | -0.004 |
per | -0.004 |
ppy | -0.004 |
ilies | -0.003 |
ectar | -0.003 |
Token | I |
Token ID | 40 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
opsy | +0.053 |
ppy | +0.053 |
ilies | +0.053 |
pper | +0.052 |
oll | +0.049 |
Neg logit contributions | |
Paula | -0.047 |
irens | -0.046 |
spe | -0.043 |
Paul | -0.043 |
James | -0.041 |
Token | found |
Token ID | 1043 |
Feature activation | -3.08e+00 |
Pos logit contributions | |
Paula | +0.098 |
Paul | +0.096 |
spe | +0.092 |
Sarah | +0.090 |
John | +0.090 |
Neg logit contributions | |
ilies | -0.057 |
pper | -0.056 |
opsy | -0.056 |
ectar | -0.055 |
ppy | -0.055 |
Token | His |
Token ID | 6653 |
Feature activation | +9.85e-02 |
Pos logit contributions | |
irens | +0.039 |
Paula | +0.039 |
Paul | +0.037 |
Sarah | +0.036 |
spe | +0.036 |
Neg logit contributions | |
ilies | -0.029 |
ppy | -0.028 |
ectar | -0.028 |
opsy | -0.027 |
pper | -0.027 |
Token | dad |
Token ID | 9955 |
Feature activation | +8.95e-01 |
Pos logit contributions | |
Paula | +0.008 |
irens | +0.008 |
Paul | +0.008 |
John | +0.007 |
spe | +0.007 |
Neg logit contributions | |
ppy | -0.006 |
pper | -0.006 |
ilies | -0.006 |
per | -0.006 |
ectar | -0.006 |
Token | sighed |
Token ID | 21893 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
Paula | +0.071 |
irens | +0.068 |
Paul | +0.065 |
spe | +0.062 |
flashing | +0.062 |
Neg logit contributions | |
ppy | -0.062 |
per | -0.060 |
pper | -0.060 |
ily | -0.057 |
ilies | -0.054 |
Token | and |
Token ID | 290 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
Paula | +0.103 |
irens | +0.101 |
Paul | +0.098 |
spe | +0.093 |
flashing | +0.091 |
Neg logit contributions | |
ppy | -0.057 |
per | -0.056 |
pper | -0.053 |
opsy | -0.052 |
ilies | -0.052 |
Token | looked |
Token ID | 3114 |
Feature activation | +4.68e+00 |
Pos logit contributions | |
Paula | +0.018 |
irens | +0.018 |
Paul | +0.017 |
spe | +0.016 |
Smith | +0.016 |
Neg logit contributions | |
ppy | -0.013 |
per | -0.013 |
pper | -0.013 |
ily | -0.013 |
ilies | -0.012 |
Token | at |
Token ID | 379 |
Feature activation | -4.35e+00 |
Pos logit contributions | |
Paula | +0.497 |
irens | +0.485 |
Paul | +0.476 |
spe | +0.455 |
John | +0.452 |
Neg logit contributions | |
ppy | -0.189 |
pper | -0.178 |
ilies | -0.177 |
per | -0.177 |
opsy | -0.175 |
Token | his |
Token ID | 465 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
ily | +0.267 |
ppy | +0.255 |
per | +0.251 |
pper | +0.235 |
like | +0.231 |
Neg logit contributions | |
spe | -0.361 |
irens | -0.359 |
Paula | -0.357 |
flashing | -0.329 |
iren | -0.327 |
Token | laptop |
Token ID | 13224 |
Feature activation | +3.98e+00 |
Pos logit contributions | |
Paula | +0.148 |
irens | +0.145 |
Paul | +0.139 |
Sarah | +0.130 |
spe | +0.130 |
Neg logit contributions | |
per | -0.097 |
ppy | -0.097 |
opsy | -0.095 |
pper | -0.094 |
ilies | -0.092 |
Token | . |
Token ID | 13 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
Paula | +0.375 |
irens | +0.362 |
Paul | +0.355 |
spe | +0.335 |
John | +0.330 |
Neg logit contributions | |
ppy | -0.205 |
per | -0.205 |
pper | -0.195 |
ilies | -0.192 |
ily | -0.191 |
Token | He |
Token ID | 679 |
Feature activation | +3.05e-01 |
Pos logit contributions | |
Paula | +0.014 |
irens | +0.013 |
spe | +0.013 |
Paul | +0.012 |
flashing | +0.012 |
Neg logit contributions | |
per | -0.012 |
ily | -0.012 |
ppy | -0.012 |
ilies | -0.011 |
ectar | -0.011 |
Token | had |
Token ID | 550 |
Feature activation | +4.51e-02 |
Pos logit contributions | |
Paula | +0.024 |
irens | +0.023 |
Paul | +0.022 |
spe | +0.021 |
John | +0.021 |
Neg logit contributions | |
ppy | -0.021 |
pper | -0.020 |
per | -0.020 |
opsy | -0.020 |
ilies | -0.020 |
Token | what |
Token ID | 644 |
Feature activation | +4.45e+00 |
Pos logit contributions | |
irens | +0.738 |
Paula | +0.726 |
iren | +0.672 |
spe | +0.666 |
flashing | +0.648 |
Neg logit contributions | |
ppy | -0.538 |
per | -0.508 |
pper | -0.500 |
ily | -0.498 |
ilies | -0.492 |
Token | , |
Token ID | 11 |
Feature activation | -3.67e-01 |
Pos logit contributions | |
Paula | +0.393 |
irens | +0.387 |
iren | +0.355 |
spe | +0.342 |
Paul | +0.341 |
Neg logit contributions | |
like | -0.233 |
ppy | -0.227 |
careful | -0.224 |
per | -0.222 |
ily | -0.219 |
Token | he |
Token ID | 339 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
ppy | +0.023 |
per | +0.023 |
ilies | +0.023 |
pper | +0.022 |
opsy | +0.022 |
Neg logit contributions | |
Paula | -0.029 |
irens | -0.029 |
iren | -0.026 |
Paul | -0.026 |
spe | -0.025 |
Token | could |
Token ID | 714 |
Feature activation | -4.01e+00 |
Pos logit contributions | |
ppy | +0.015 |
pper | +0.015 |
per | +0.015 |
ilies | +0.014 |
opsy | +0.014 |
Neg logit contributions | |
Paula | -0.018 |
irens | -0.018 |
Paul | -0.017 |
spe | -0.016 |
Sarah | -0.016 |
Token | always |
Token ID | 1464 |
Feature activation | -3.19e+00 |
Pos logit contributions | |
ppy | +0.213 |
pper | +0.204 |
ilies | +0.202 |
per | +0.202 |
opsy | +0.200 |
Neg logit contributions | |
Paula | -0.373 |
irens | -0.364 |
Paul | -0.356 |
spe | -0.338 |
John | -0.336 |
Token | face |
Token ID | 1986 |
Feature activation | -3.53e+00 |
Pos logit contributions | |
ppy | +0.154 |
ilies | +0.146 |
pper | +0.146 |
opsy | +0.144 |
per | +0.144 |
Neg logit contributions | |
Paula | -0.312 |
irens | -0.303 |
Paul | -0.299 |
John | -0.283 |
spe | -0.283 |
Token | his |
Token ID | 465 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
ppy | +0.224 |
per | +0.220 |
pper | +0.215 |
ilies | +0.210 |
opsy | +0.205 |
Neg logit contributions | |
Paula | -0.282 |
irens | -0.278 |
Paul | -0.261 |
spe | -0.252 |
iren | -0.252 |
Token | fears |
Token ID | 10251 |
Feature activation | +3.26e+00 |
Pos logit contributions | |
Paula | +0.144 |
irens | +0.138 |
Paul | +0.138 |
spe | +0.130 |
John | +0.129 |
Neg logit contributions | |
ppy | -0.102 |
ilies | -0.099 |
pper | -0.098 |
per | -0.096 |
opsy | -0.096 |
Token | just |
Token ID | 655 |
Feature activation | -5.39e+00 |
Pos logit contributions | |
Paula | +0.282 |
irens | +0.275 |
Paul | +0.268 |
spe | +0.253 |
John | +0.252 |
Neg logit contributions | |
ppy | -0.193 |
pper | -0.186 |
per | -0.184 |
ilies | -0.184 |
opsy | -0.182 |
Token | like |
Token ID | 588 |
Feature activation | -6.16e+00 |
Pos logit contributions | |
ppy | +0.276 |
pper | +0.264 |
ilies | +0.262 |
opsy | +0.262 |
per | +0.260 |
Neg logit contributions | |
Paula | -0.511 |
irens | -0.496 |
Paul | -0.490 |
John | -0.464 |
spe | -0.464 |
Token | the |
Token ID | 262 |
Feature activation | +8.75e-01 |
Pos logit contributions | |
per | +0.350 |
ppy | +0.350 |
pper | +0.338 |
ilies | +0.338 |
ily | +0.334 |
Neg logit contributions | |
Paula | -0.525 |
irens | -0.518 |
Paul | -0.475 |
spe | -0.472 |
iren | -0.469 |
Token | hill |
Token ID | 12788 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
Paula | +0.053 |
irens | +0.051 |
Paul | +0.048 |
Sarah | +0.045 |
spe | +0.045 |
Neg logit contributions | |
ppy | -0.057 |
per | -0.056 |
opsy | -0.055 |
pper | -0.055 |
ilies | -0.055 |
Token | and |
Token ID | 290 |
Feature activation | +7.27e-02 |
Pos logit contributions | |
Paula | +0.099 |
irens | +0.096 |
Paul | +0.094 |
spe | +0.089 |
John | +0.089 |
Neg logit contributions | |
ppy | -0.058 |
pper | -0.056 |
per | -0.056 |
ilies | -0.056 |
opsy | -0.055 |
Token | was |
Token ID | 373 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
Paula | +0.006 |
irens | +0.006 |
Paul | +0.005 |
spe | +0.005 |
iren | +0.005 |
Neg logit contributions | |
ppy | -0.005 |
pper | -0.005 |
per | -0.005 |
ilies | -0.005 |
opsy | -0.005 |
Token | very |
Token ID | 845 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
Paula | +0.127 |
Paul | +0.121 |
irens | +0.118 |
John | +0.116 |
Sarah | +0.115 |
Neg logit contributions | |
ppy | -0.095 |
pper | -0.091 |
ilies | -0.090 |
opsy | -0.090 |
per | -0.089 |
Token | grateful |
Token ID | 14066 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
ppy | +0.014 |
pper | +0.013 |
ilies | +0.013 |
per | +0.012 |
opsy | +0.012 |
Neg logit contributions | |
Paula | -0.018 |
Paul | -0.018 |
irens | -0.018 |
spe | -0.017 |
John | -0.017 |
Token | for |
Token ID | 329 |
Feature activation | -5.87e+00 |
Pos logit contributions | |
Paula | +0.086 |
Paul | +0.085 |
John | +0.081 |
Sarah | +0.080 |
irens | +0.080 |
Neg logit contributions | |
ppy | -0.068 |
pper | -0.066 |
ilies | -0.066 |
opsy | -0.066 |
per | -0.063 |
Token | the |
Token ID | 262 |
Feature activation | +7.62e-01 |
Pos logit contributions | |
ppy | +0.451 |
pper | +0.438 |
ilies | +0.436 |
per | +0.435 |
opsy | +0.433 |
Neg logit contributions | |
Paula | -0.410 |
irens | -0.395 |
Paul | -0.385 |
spe | -0.357 |
John | -0.356 |
Token | wonderful |
Token ID | 7932 |
Feature activation | +1.86e+00 |
Pos logit contributions | |
Paula | +0.063 |
Paul | +0.060 |
irens | +0.060 |
spe | +0.057 |
John | +0.056 |
Neg logit contributions | |
ppy | -0.049 |
ilies | -0.047 |
pper | -0.047 |
per | -0.046 |
opsy | -0.045 |
Token | experience |
Token ID | 1998 |
Feature activation | +3.68e+00 |
Pos logit contributions | |
Paula | +0.151 |
irens | +0.150 |
Paul | +0.140 |
spe | +0.134 |
James | +0.133 |
Neg logit contributions | |
opsy | -0.115 |
per | -0.113 |
ectar | -0.111 |
worm | -0.111 |
ilies | -0.110 |
Token | he |
Token ID | 339 |
Feature activation | -3.22e-01 |
Pos logit contributions | |
Paula | +0.325 |
irens | +0.322 |
Paul | +0.304 |
spe | +0.290 |
flashing | +0.288 |
Neg logit contributions | |
ppy | -0.208 |
per | -0.202 |
pper | -0.202 |
ilies | -0.200 |
opsy | -0.193 |
Token | had |
Token ID | 550 |
Feature activation | -5.09e-02 |
Pos logit contributions | |
ppy | +0.023 |
pper | +0.023 |
ilies | +0.023 |
per | +0.023 |
opsy | +0.022 |
Neg logit contributions | |
Paula | -0.024 |
irens | -0.023 |
Paul | -0.022 |
spe | -0.021 |
John | -0.021 |
Token | his |
Token ID | 465 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
ppy | +0.111 |
ilies | +0.108 |
pper | +0.106 |
opsy | +0.106 |
per | +0.105 |
Neg logit contributions | |
Paula | -0.221 |
Paul | -0.214 |
irens | -0.211 |
John | -0.204 |
Sarah | -0.203 |
Token | belt |
Token ID | 10999 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
Paula | +0.134 |
irens | +0.130 |
Paul | +0.127 |
spe | +0.119 |
John | +0.119 |
Neg logit contributions | |
ppy | -0.103 |
pper | -0.100 |
ilies | -0.100 |
per | -0.099 |
opsy | -0.098 |
Token | was |
Token ID | 373 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
ppy | +0.022 |
pper | +0.022 |
per | +0.022 |
ily | +0.021 |
ilies | +0.021 |
Neg logit contributions | |
Paula | -0.021 |
irens | -0.021 |
Paul | -0.020 |
John | -0.018 |
James | -0.018 |
Token | too |
Token ID | 1165 |
Feature activation | -2.88e+00 |
Pos logit contributions | |
Paula | +0.160 |
irens | +0.158 |
Paul | +0.151 |
spe | +0.144 |
John | +0.142 |
Neg logit contributions | |
ppy | -0.070 |
pper | -0.067 |
per | -0.065 |
like | -0.062 |
careful | -0.062 |
Token | big |
Token ID | 1263 |
Feature activation | +2.50e+00 |
Pos logit contributions | |
ppy | +0.165 |
opsy | +0.158 |
pper | +0.157 |
per | +0.155 |
careful | +0.148 |
Neg logit contributions | |
Paula | -0.262 |
irens | -0.250 |
Paul | -0.246 |
James | -0.235 |
spe | -0.231 |
Token | for |
Token ID | 329 |
Feature activation | -5.79e+00 |
Pos logit contributions | |
Paula | +0.264 |
irens | +0.260 |
Paul | +0.246 |
spe | +0.245 |
iren | +0.235 |
Neg logit contributions | |
ppy | -0.105 |
per | -0.100 |
pper | -0.095 |
like | -0.093 |
ily | -0.091 |
Token | him |
Token ID | 683 |
Feature activation | -2.79e-01 |
Pos logit contributions | |
ppy | +0.291 |
per | +0.284 |
ily | +0.271 |
like | +0.254 |
ilies | +0.252 |
Neg logit contributions | |
spe | -0.560 |
Paula | -0.503 |
irens | -0.485 |
iren | -0.455 |
flashing | -0.452 |
Token | and |
Token ID | 290 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
ppy | +0.018 |
pper | +0.018 |
per | +0.017 |
ilies | +0.017 |
opsy | +0.017 |
Neg logit contributions | |
Paula | -0.023 |
irens | -0.022 |
Paul | -0.021 |
spe | -0.020 |
James | -0.020 |
Token | he |
Token ID | 339 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
Paula | +0.013 |
irens | +0.013 |
iren | +0.012 |
spe | +0.012 |
Paul | +0.011 |
Neg logit contributions | |
ppy | -0.009 |
pper | -0.009 |
worm | -0.009 |
careful | -0.009 |
per | -0.009 |
Token | kept |
Token ID | 4030 |
Feature activation | -3.53e+00 |
Pos logit contributions | |
ppy | +0.017 |
ilies | +0.017 |
per | +0.016 |
ectar | +0.016 |
pper | +0.016 |
Neg logit contributions | |
Paula | -0.018 |
Paul | -0.017 |
irens | -0.017 |
spe | -0.016 |
John | -0.016 |
Token | tri |
Token ID | 1333 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
per | +0.239 |
pper | +0.229 |
ppy | +0.224 |
careful | +0.223 |
like | +0.220 |
Neg logit contributions | |
Paula | -0.262 |
spe | -0.257 |
irens | -0.257 |
Paul | -0.237 |
iren | -0.230 |
Token | zoo |
Token ID | 26626 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
Paula | +0.057 |
irens | +0.055 |
Paul | +0.054 |
spe | +0.050 |
Sarah | +0.050 |
Neg logit contributions | |
ppy | -0.070 |
pper | -0.068 |
ilies | -0.067 |
per | -0.067 |
opsy | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
Paula | +0.118 |
irens | +0.114 |
Paul | +0.112 |
spe | +0.105 |
John | +0.105 |
Neg logit contributions | |
ppy | -0.081 |
pper | -0.078 |
ilies | -0.077 |
per | -0.077 |
opsy | -0.076 |
Token | He |
Token ID | 679 |
Feature activation | +2.95e-01 |
Pos logit contributions | |
Paula | +0.011 |
irens | +0.010 |
Paul | +0.010 |
John | +0.009 |
Sarah | +0.009 |
Neg logit contributions | |
ppy | -0.012 |
pper | -0.012 |
ilies | -0.011 |
per | -0.011 |
opsy | -0.011 |
Token | can |
Token ID | 460 |
Feature activation | -2.73e+00 |
Pos logit contributions | |
Paula | +0.019 |
irens | +0.018 |
Paul | +0.018 |
James | +0.017 |
spe | +0.017 |
Neg logit contributions | |
opsy | -0.023 |
ectar | -0.023 |
ppy | -0.023 |
pper | -0.023 |
ilies | -0.023 |
Token | 't |
Token ID | 470 |
Feature activation | +3.77e+00 |
Pos logit contributions | |
ppy | +0.187 |
ectar | +0.187 |
ilies | +0.185 |
opsy | +0.182 |
pper | +0.176 |
Neg logit contributions | |
Paula | -0.217 |
Paul | -0.208 |
irens | -0.199 |
James | -0.197 |
John | -0.197 |
Token | wait |
Token ID | 4043 |
Feature activation | -5.78e+00 |
Pos logit contributions | |
Paul | +0.346 |
Paula | +0.346 |
John | +0.343 |
Sarah | +0.341 |
James | +0.339 |
Neg logit contributions | |
ectar | -0.197 |
opsy | -0.184 |
ilies | -0.183 |
imes | -0.175 |
oll | -0.168 |
Token | to |
Token ID | 284 |
Feature activation | +6.81e-01 |
Pos logit contributions | |
ppy | +0.406 |
pper | +0.400 |
opsy | +0.395 |
ilies | +0.388 |
per | +0.387 |
Neg logit contributions | |
Paula | -0.440 |
irens | -0.421 |
Paul | -0.418 |
John | -0.395 |
Sarah | -0.394 |
Token | see |
Token ID | 766 |
Feature activation | +5.51e-01 |
Pos logit contributions | |
Paula | +0.057 |
irens | +0.055 |
Paul | +0.055 |
John | +0.052 |
Sarah | +0.052 |
Neg logit contributions | |
ppy | -0.042 |
pper | -0.041 |
ilies | -0.041 |
opsy | -0.040 |
per | -0.040 |
Token | the |
Token ID | 262 |
Feature activation | +8.21e-01 |
Pos logit contributions | |
Paula | +0.042 |
irens | +0.041 |
Paul | +0.038 |
spe | +0.038 |
iren | +0.036 |
Neg logit contributions | |
ppy | -0.038 |
ilies | -0.037 |
per | -0.037 |
pper | -0.037 |
ily | -0.036 |
Token | lion |
Token ID | 18744 |
Feature activation | +2.20e-01 |
Pos logit contributions | |
Paula | +0.058 |
irens | +0.056 |
Paul | +0.055 |
spe | +0.051 |
Sarah | +0.051 |
Neg logit contributions | |
ppy | -0.063 |
pper | -0.060 |
per | -0.060 |
opsy | -0.060 |
ilies | -0.059 |
Token | . |
Token ID | 13 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
Paula | +0.017 |
irens | +0.016 |
Paul | +0.016 |
James | +0.015 |
spe | +0.015 |
Neg logit contributions | |
opsy | -0.015 |
ppy | -0.015 |
ilies | -0.014 |
pper | -0.014 |
per | -0.014 |
Token | said |
Token ID | 531 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
Paula | +0.019 |
Paul | +0.018 |
Sarah | +0.017 |
John | +0.017 |
James | +0.017 |
Neg logit contributions | |
ilies | -0.027 |
ectar | -0.026 |
pper | -0.026 |
ppy | -0.026 |
per | -0.026 |
Token | , |
Token ID | 11 |
Feature activation | -3.17e-01 |
Pos logit contributions | |
pper | +0.169 |
ppy | +0.168 |
opsy | +0.164 |
ilies | +0.161 |
per | +0.159 |
Neg logit contributions | |
Paula | -0.154 |
Paul | -0.153 |
John | -0.149 |
Sarah | -0.146 |
irens | -0.144 |
Token | " |
Token ID | 366 |
Feature activation | -5.78e-01 |
Pos logit contributions | |
ppy | +0.027 |
ilies | +0.027 |
pper | +0.027 |
opsy | +0.027 |
per | +0.026 |
Neg logit contributions | |
Paula | -0.019 |
Paul | -0.018 |
irens | -0.018 |
John | -0.017 |
Sarah | -0.017 |
Token | It |
Token ID | 1026 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
ilies | +0.042 |
ppy | +0.041 |
per | +0.041 |
pper | +0.039 |
opsy | +0.039 |
Neg logit contributions | |
Paula | -0.043 |
irens | -0.041 |
Paul | -0.040 |
spe | -0.038 |
Sarah | -0.038 |
Token | smells |
Token ID | 25760 |
Feature activation | +5.09e+00 |
Pos logit contributions | |
ppy | +0.081 |
pper | +0.078 |
ilies | +0.077 |
per | +0.077 |
opsy | +0.077 |
Neg logit contributions | |
Paula | -0.111 |
irens | -0.108 |
Paul | -0.106 |
spe | -0.100 |
John | -0.099 |
Token | like |
Token ID | 588 |
Feature activation | -6.12e+00 |
Pos logit contributions | |
Paula | +0.402 |
Paul | +0.386 |
Sarah | +0.385 |
irens | +0.382 |
John | +0.380 |
Neg logit contributions | |
pper | -0.327 |
per | -0.321 |
ppy | -0.318 |
opsy | -0.315 |
ilies | -0.299 |
Token | rain |
Token ID | 6290 |
Feature activation | -2.30e+00 |
Pos logit contributions | |
ppy | +0.362 |
pper | +0.349 |
per | +0.346 |
ilies | +0.344 |
opsy | +0.343 |
Neg logit contributions | |
Paula | -0.533 |
irens | -0.518 |
Paul | -0.511 |
John | -0.483 |
Sarah | -0.480 |
Token | ." |
Token ID | 526 |
Feature activation | -5.29e-01 |
Pos logit contributions | |
ppy | +0.185 |
pper | +0.176 |
ilies | +0.173 |
per | +0.173 |
ectar | +0.171 |
Neg logit contributions | |
Paula | -0.157 |
irens | -0.149 |
Paul | -0.145 |
spe | -0.137 |
James | -0.130 |
Token |
|
Token ID | 198 |
Feature activation | +4.59e-01 |
Pos logit contributions | |
ily | +0.027 |
per | +0.025 |
like | +0.024 |
oll | +0.024 |
ilies | +0.024 |
Neg logit contributions | |
Paula | -0.049 |
spe | -0.045 |
irens | -0.044 |
flashing | -0.044 |
bul | -0.043 |
Token |
|
Token ID | 198 |
Feature activation | +4.50e-01 |
Pos logit contributions | |
Paula | +0.040 |
irens | +0.038 |
Paul | +0.037 |
spe | +0.036 |
flashing | +0.035 |
Neg logit contributions | |
ppy | -0.027 |
opsy | -0.026 |
ily | -0.026 |
pper | -0.026 |
ilies | -0.026 |
Token | Tom |
Token ID | 13787 |
Feature activation | +4.69e-01 |
Pos logit contributions | |
Paula | +0.038 |
irens | +0.037 |
Paul | +0.036 |
spe | +0.034 |
John | +0.034 |
Neg logit contributions | |
ppy | -0.028 |
pper | -0.027 |
per | -0.027 |
opsy | -0.026 |
ilies | -0.026 |
Token | He |
Token ID | 679 |
Feature activation | +1.97e-01 |
Pos logit contributions | |
Paula | +0.006 |
irens | +0.006 |
spe | +0.005 |
flashing | +0.005 |
Paul | +0.005 |
Neg logit contributions | |
ily | -0.004 |
per | -0.004 |
ppy | -0.004 |
opsy | -0.004 |
worm | -0.004 |
Token | had |
Token ID | 550 |
Feature activation | -6.87e-02 |
Pos logit contributions | |
Paula | +0.014 |
Paul | +0.013 |
irens | +0.013 |
John | +0.012 |
Sarah | +0.012 |
Neg logit contributions | |
ppy | -0.015 |
ilies | -0.015 |
pper | -0.014 |
ectar | -0.014 |
per | -0.014 |
Token | never |
Token ID | 1239 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
ppy | +0.004 |
pper | +0.004 |
per | +0.004 |
ilies | +0.004 |
opsy | +0.004 |
Neg logit contributions | |
Paula | -0.006 |
irens | -0.006 |
Paul | -0.006 |
spe | -0.006 |
John | -0.006 |
Token | seen |
Token ID | 1775 |
Feature activation | +6.81e-01 |
Pos logit contributions | |
Paula | +0.088 |
irens | +0.085 |
Paul | +0.083 |
spe | +0.078 |
Ellen | +0.077 |
Neg logit contributions | |
ppy | -0.061 |
pper | -0.058 |
ilies | -0.058 |
per | -0.057 |
ily | -0.057 |
Token | anything |
Token ID | 1997 |
Feature activation | +2.59e+00 |
Pos logit contributions | |
Paula | +0.059 |
irens | +0.057 |
Paul | +0.056 |
spe | +0.054 |
John | +0.052 |
Neg logit contributions | |
ppy | -0.040 |
pper | -0.039 |
ilies | -0.039 |
per | -0.039 |
opsy | -0.038 |
Token | like |
Token ID | 588 |
Feature activation | -6.26e+00 |
Pos logit contributions | |
Paula | +0.277 |
Paul | +0.272 |
John | +0.267 |
Sarah | +0.267 |
James | +0.266 |
Neg logit contributions | |
opsy | -0.090 |
ppy | -0.087 |
pper | -0.083 |
ilies | -0.080 |
per | -0.079 |
Token | it |
Token ID | 340 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
ppy | +0.419 |
pper | +0.409 |
ilies | +0.407 |
per | +0.405 |
opsy | +0.399 |
Neg logit contributions | |
Paula | -0.496 |
irens | -0.480 |
Paul | -0.463 |
spe | -0.440 |
Sarah | -0.427 |
Token | before |
Token ID | 878 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
opsy | +0.018 |
ppy | +0.018 |
ilies | +0.017 |
per | +0.017 |
ectar | +0.017 |
Neg logit contributions | |
Paula | -0.020 |
Paul | -0.020 |
Sarah | -0.019 |
irens | -0.019 |
John | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | +7.14e-02 |
Pos logit contributions | |
per | +0.093 |
ppy | +0.089 |
opsy | +0.089 |
pper | +0.088 |
ilies | +0.088 |
Neg logit contributions | |
Paula | -0.204 |
irens | -0.195 |
spe | -0.194 |
Paul | -0.187 |
Ellen | -0.181 |
Token | He |
Token ID | 679 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
Paula | +0.006 |
irens | +0.006 |
spe | +0.006 |
flashing | +0.006 |
Paul | +0.005 |
Neg logit contributions | |
ily | -0.004 |
opsy | -0.004 |
ppy | -0.004 |
per | -0.004 |
worm | -0.004 |
Token | decided |
Token ID | 3066 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
Paula | +0.015 |
Paul | +0.014 |
irens | +0.014 |
John | +0.013 |
Sarah | +0.013 |
Neg logit contributions | |
ilies | -0.017 |
ppy | -0.017 |
ectar | -0.016 |
pper | -0.016 |
per | -0.016 |