Token | that |
Token ID | 326 |
Feature activation | +9.81e-02 |
Pos logit contributions | |
oppy | +0.429 |
opsy | +0.408 |
chy | +0.402 |
rig | +0.379 |
bur | +0.378 |
Neg logit contributions | |
folk | -0.189 |
sale | -0.186 |
letters | -0.180 |
digits | -0.169 |
Amen | -0.164 |
Token | day |
Token ID | 1110 |
Feature activation | +4.47e-03 |
Pos logit contributions | |
opsy | +0.238 |
oppy | +0.234 |
rig | +0.216 |
Remy | +0.212 |
bur | +0.210 |
Neg logit contributions | |
folk | -0.109 |
digits | -0.104 |
sale | -0.102 |
customer | -0.098 |
Amen | -0.097 |
Token | on |
Token ID | 319 |
Feature activation | -5.02e-02 |
Pos logit contributions | |
opsy | +0.010 |
oppy | +0.010 |
Remy | +0.009 |
bur | +0.009 |
chy | +0.009 |
Neg logit contributions | |
folk | -0.006 |
digits | -0.006 |
sale | -0.005 |
customer | -0.005 |
letters | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
sale | +0.036 |
folk | +0.034 |
customer | +0.028 |
digits | +0.026 |
letters | +0.025 |
Neg logit contributions | |
rig | -0.143 |
opsy | -0.139 |
oppy | -0.132 |
chy | -0.129 |
bur | -0.124 |
Token | Fl |
Token ID | 1610 |
Feature activation | +3.94e-01 |
Pos logit contributions | |
sale | +0.045 |
folk | +0.044 |
customer | +0.044 |
ones | +0.040 |
one | +0.040 |
Neg logit contributions | |
rig | -0.045 |
opsy | -0.042 |
oppy | -0.042 |
web | -0.040 |
chy | -0.040 |
Token | opsy |
Token ID | 44522 |
Feature activation | +5.67e-01 |
Pos logit contributions | |
chy | +0.351 |
rig | +0.325 |
web | +0.311 |
webs | +0.299 |
rubble | +0.271 |
Neg logit contributions | |
ams | -0.772 |
be | -0.743 |
ones | -0.743 |
h | -0.734 |
customer | -0.731 |
Token | and |
Token ID | 290 |
Feature activation | +4.36e-02 |
Pos logit contributions | |
opsy | +1.306 |
oppy | +1.279 |
Remy | +1.178 |
rig | +1.172 |
chy | +1.125 |
Neg logit contributions | |
folk | -0.734 |
digits | -0.695 |
sale | -0.627 |
letters | -0.619 |
cheering | -0.585 |
Token | Mr |
Token ID | 1770 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
rig | +0.082 |
oppy | +0.077 |
opsy | +0.076 |
chy | +0.072 |
web | +0.070 |
Neg logit contributions | |
folk | -0.070 |
sale | -0.070 |
customer | -0.068 |
Amen | -0.067 |
letters | -0.065 |
Token | . |
Token ID | 13 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
rig | +0.402 |
oppy | +0.390 |
opsy | +0.389 |
bur | +0.369 |
web | +0.359 |
Neg logit contributions | |
folk | -0.164 |
boxes | -0.148 |
ones | -0.148 |
sale | -0.147 |
letters | -0.146 |
Token | Snow |
Token ID | 7967 |
Feature activation | +3.41e-01 |
Pos logit contributions | |
oppy | +0.017 |
rig | +0.017 |
chy | +0.017 |
opsy | +0.016 |
ches | +0.015 |
Neg logit contributions | |
sale | -0.023 |
folk | -0.022 |
customer | -0.021 |
digits | -0.021 |
Amen | -0.021 |
Token | man |
Token ID | 805 |
Feature activation | -1.93e-01 |
Pos logit contributions | |
oppy | +0.701 |
opsy | +0.700 |
rig | +0.671 |
bur | +0.644 |
Remy | +0.639 |
Neg logit contributions | |
folk | -0.523 |
boxes | -0.415 |
sale | -0.396 |
winners | -0.391 |
cheering | -0.390 |
Token | orns |
Token ID | 19942 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
opsy | +0.401 |
oppy | +0.388 |
Remy | +0.359 |
bur | +0.349 |
rig | +0.349 |
Neg logit contributions | |
folk | -0.427 |
digits | -0.418 |
sale | -0.402 |
customer | -0.394 |
bench | -0.390 |
Token | and |
Token ID | 290 |
Feature activation | +4.16e-02 |
Pos logit contributions | |
opsy | +0.371 |
oppy | +0.361 |
Remy | +0.333 |
bur | +0.320 |
uffy | +0.318 |
Neg logit contributions | |
folk | -0.211 |
sale | -0.201 |
digits | -0.182 |
customer | -0.179 |
epad | -0.178 |
Token | never |
Token ID | 1239 |
Feature activation | +5.74e-03 |
Pos logit contributions | |
opsy | +0.095 |
oppy | +0.093 |
rig | +0.082 |
Remy | +0.081 |
uffy | +0.081 |
Neg logit contributions | |
folk | -0.058 |
sale | -0.055 |
digits | -0.053 |
letters | -0.052 |
epad | -0.048 |
Token | met |
Token ID | 1138 |
Feature activation | -8.09e-02 |
Pos logit contributions | |
opsy | +0.013 |
oppy | +0.013 |
Remy | +0.012 |
chy | +0.011 |
uffy | +0.011 |
Neg logit contributions | |
folk | -0.008 |
digits | -0.008 |
sale | -0.007 |
letters | -0.007 |
epad | -0.007 |
Token | Fl |
Token ID | 1610 |
Feature activation | +3.93e-01 |
Pos logit contributions | |
folk | +0.108 |
sale | +0.106 |
digits | +0.103 |
letters | +0.099 |
customer | +0.094 |
Neg logit contributions | |
oppy | -0.182 |
opsy | -0.181 |
rig | -0.161 |
chy | -0.159 |
uffy | -0.156 |
Token | opsy |
Token ID | 44522 |
Feature activation | +5.67e-01 |
Pos logit contributions | |
chy | +0.277 |
rig | +0.261 |
Remy | +0.196 |
web | +0.173 |
ches | +0.159 |
Neg logit contributions | |
customer | -0.912 |
digits | -0.909 |
sale | -0.908 |
folk | -0.878 |
ones | -0.877 |
Token | . |
Token ID | 13 |
Feature activation | +8.46e-03 |
Pos logit contributions | |
opsy | +1.376 |
oppy | +1.340 |
Remy | +1.247 |
rig | +1.222 |
bur | +1.207 |
Neg logit contributions | |
folk | -0.649 |
sale | -0.599 |
digits | -0.570 |
customer | -0.555 |
letters | -0.525 |
Token | He |
Token ID | 679 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
opsy | +0.019 |
oppy | +0.019 |
rig | +0.017 |
chy | +0.017 |
web | +0.016 |
Neg logit contributions | |
sale | -0.010 |
Amen | -0.010 |
folk | -0.010 |
| -0.009 |
digits | -0.009 |
Token | was |
Token ID | 373 |
Feature activation | +6.28e-02 |
Pos logit contributions | |
folk | +0.011 |
digits | +0.010 |
sale | +0.010 |
Amen | +0.010 |
asked | +0.010 |
Neg logit contributions | |
opsy | -0.028 |
oppy | -0.027 |
uffy | -0.025 |
rig | -0.024 |
Remy | -0.024 |
Token | glad |
Token ID | 9675 |
Feature activation | -3.22e-03 |
Pos logit contributions | |
opsy | +0.137 |
oppy | +0.136 |
Remy | +0.125 |
chy | +0.121 |
bur | +0.120 |
Neg logit contributions | |
folk | -0.092 |
digits | -0.088 |
sale | -0.084 |
customer | -0.077 |
Amen | -0.077 |
Token | for |
Token ID | 329 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
sale | +0.003 |
one | +0.003 |
them | +0.003 |
folk | +0.003 |
cheering | +0.003 |
Neg logit contributions | |
rig | -0.008 |
oppy | -0.008 |
opsy | -0.008 |
chy | -0.008 |
web | -0.007 |
Token | happy |
Token ID | 3772 |
Feature activation | -6.59e-02 |
Pos logit contributions | |
folk | +0.148 |
digits | +0.143 |
sale | +0.134 |
customer | +0.123 |
prices | +0.122 |
Neg logit contributions | |
opsy | -0.193 |
oppy | -0.191 |
Remy | -0.177 |
chy | -0.174 |
bur | -0.169 |
Token | and |
Token ID | 290 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
folk | +0.084 |
sale | +0.083 |
digits | +0.080 |
cheering | +0.074 |
letters | +0.073 |
Neg logit contributions | |
opsy | -0.152 |
oppy | -0.150 |
chy | -0.136 |
rig | -0.132 |
bur | -0.131 |
Token | grateful |
Token ID | 14066 |
Feature activation | -6.08e-04 |
Pos logit contributions | |
opsy | +0.091 |
oppy | +0.090 |
uffy | +0.081 |
chy | +0.080 |
rig | +0.080 |
Neg logit contributions | |
folk | -0.047 |
digits | -0.044 |
sale | -0.042 |
cheering | -0.042 |
letters | -0.041 |
Token | to |
Token ID | 284 |
Feature activation | +6.54e-03 |
Pos logit contributions | |
sale | +0.001 |
folk | +0.001 |
letters | +0.001 |
digits | +0.001 |
customer | +0.001 |
Neg logit contributions | |
opsy | -0.001 |
oppy | -0.001 |
bur | -0.001 |
rig | -0.001 |
chy | -0.001 |
Token | Fl |
Token ID | 1610 |
Feature activation | +3.92e-01 |
Pos logit contributions | |
opsy | +0.013 |
oppy | +0.012 |
chy | +0.011 |
Remy | +0.011 |
bur | +0.011 |
Neg logit contributions | |
folk | -0.011 |
digits | -0.011 |
sale | -0.010 |
customer | -0.010 |
Amen | -0.010 |
Token | opsy |
Token ID | 44522 |
Feature activation | +5.66e-01 |
Pos logit contributions | |
chy | +0.318 |
web | +0.293 |
rig | +0.272 |
Remy | +0.236 |
cob | +0.235 |
Neg logit contributions | |
sale | -0.823 |
digits | -0.812 |
customer | -0.811 |
be | -0.801 |
folk | -0.798 |
Token | for |
Token ID | 329 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
opsy | +1.343 |
oppy | +1.279 |
Remy | +1.183 |
bur | +1.163 |
rig | +1.159 |
Neg logit contributions | |
folk | -0.675 |
digits | -0.650 |
sale | -0.641 |
letters | -0.636 |
customer | -0.573 |
Token | saving |
Token ID | 8914 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
opsy | +0.255 |
oppy | +0.251 |
Remy | +0.250 |
bur | +0.236 |
chy | +0.227 |
Neg logit contributions | |
folk | -0.201 |
digits | -0.195 |
sale | -0.162 |
prices | -0.160 |
epad | -0.159 |
Token | him |
Token ID | 683 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
sale | +0.025 |
folk | +0.025 |
digits | +0.025 |
customer | +0.023 |
letters | +0.023 |
Neg logit contributions | |
opsy | -0.042 |
oppy | -0.041 |
chy | -0.038 |
Remy | -0.037 |
rig | -0.037 |
Token | from |
Token ID | 422 |
Feature activation | +1.98e-01 |
Pos logit contributions | |
opsy | +0.034 |
oppy | +0.032 |
Remy | +0.030 |
uffy | +0.029 |
bur | +0.028 |
Neg logit contributions | |
sale | -0.015 |
folk | -0.014 |
letters | -0.013 |
digits | -0.013 |
prices | -0.012 |
Token | melting |
Token ID | 24203 |
Feature activation | +9.15e-02 |
Pos logit contributions | |
opsy | +0.417 |
oppy | +0.413 |
Remy | +0.375 |
chy | +0.369 |
bur | +0.363 |
Neg logit contributions | |
folk | -0.306 |
digits | -0.295 |
sale | -0.287 |
customer | -0.263 |
letters | -0.259 |
Token | patient |
Token ID | 5827 |
Feature activation | -6.85e-02 |
Pos logit contributions | |
opsy | +0.029 |
oppy | +0.029 |
Remy | +0.027 |
chy | +0.027 |
bur | +0.026 |
Neg logit contributions | |
digits | -0.021 |
folk | -0.021 |
customer | -0.020 |
sale | -0.020 |
letters | -0.018 |
Token | little |
Token ID | 1310 |
Feature activation | +6.79e-02 |
Pos logit contributions | |
folk | +0.122 |
digits | +0.121 |
sale | +0.109 |
epad | +0.104 |
letters | +0.103 |
Neg logit contributions | |
opsy | -0.129 |
oppy | -0.127 |
Remy | -0.116 |
bur | -0.113 |
chy | -0.112 |
Token | bunny |
Token ID | 44915 |
Feature activation | +2.56e-01 |
Pos logit contributions | |
opsy | +0.146 |
oppy | +0.144 |
Remy | +0.134 |
chy | +0.132 |
bur | +0.130 |
Neg logit contributions | |
folk | -0.099 |
digits | -0.096 |
sale | -0.090 |
customer | -0.087 |
letters | -0.084 |
Token | named |
Token ID | 3706 |
Feature activation | +6.21e-04 |
Pos logit contributions | |
opsy | +0.598 |
oppy | +0.585 |
chy | +0.541 |
Remy | +0.541 |
bur | +0.539 |
Neg logit contributions | |
folk | -0.303 |
digits | -0.299 |
sale | -0.291 |
Amen | -0.270 |
customer | -0.263 |
Token | Fl |
Token ID | 1610 |
Feature activation | +3.91e-01 |
Pos logit contributions | |
opsy | +0.001 |
oppy | +0.001 |
rig | +0.001 |
chy | +0.001 |
bur | +0.001 |
Neg logit contributions | |
sale | -0.001 |
customer | -0.001 |
digits | -0.001 |
folk | -0.001 |
Robert | -0.001 |
Token | oppy |
Token ID | 26696 |
Feature activation | +5.66e-01 |
Pos logit contributions | |
rig | +0.293 |
chy | +0.268 |
refreshing | +0.260 |
Remy | +0.204 |
web | +0.196 |
Neg logit contributions | |
customer | -0.932 |
digits | -0.878 |
sale | -0.873 |
buyer | -0.849 |
Amen | -0.826 |
Token | . |
Token ID | 13 |
Feature activation | +7.35e-03 |
Pos logit contributions | |
opsy | +1.431 |
oppy | +1.376 |
Remy | +1.289 |
bur | +1.276 |
rig | +1.250 |
Neg logit contributions | |
digits | -0.571 |
folk | -0.563 |
sale | -0.551 |
letters | -0.502 |
Amen | -0.487 |
Token | Fl |
Token ID | 1610 |
Feature activation | +3.92e-01 |
Pos logit contributions | |
opsy | +0.012 |
oppy | +0.011 |
rig | +0.010 |
chy | +0.009 |
uffy | +0.009 |
Neg logit contributions | |
sale | -0.014 |
folk | -0.014 |
digits | -0.014 |
customer | -0.013 |
Amen | -0.013 |
Token | oppy |
Token ID | 26696 |
Feature activation | +5.66e-01 |
Pos logit contributions | |
rig | +0.326 |
chy | +0.266 |
refreshing | +0.261 |
Remy | +0.235 |
iqu | +0.201 |
Neg logit contributions | |
customer | -0.829 |
digits | -0.807 |
ones | -0.776 |
three | -0.771 |
Amen | -0.764 |
Token | lived |
Token ID | 5615 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
opsy | +1.364 |
oppy | +1.311 |
Remy | +1.238 |
rig | +1.228 |
bur | +1.190 |
Neg logit contributions | |
digits | -0.602 |
folk | -0.600 |
sale | -0.562 |
letters | -0.552 |
Amen | -0.517 |
Token | in |
Token ID | 287 |
Feature activation | +3.67e-02 |
Pos logit contributions | |
opsy | +0.371 |
oppy | +0.371 |
chy | +0.342 |
Remy | +0.341 |
rig | +0.333 |
Neg logit contributions | |
digits | -0.217 |
folk | -0.214 |
sale | -0.206 |
customer | -0.200 |
letters | -0.189 |
Token | to |
Token ID | 284 |
Feature activation | +5.97e-03 |
Pos logit contributions | |
folk | +0.037 |
digits | +0.036 |
sale | +0.035 |
letters | +0.032 |
customer | +0.032 |
Neg logit contributions | |
opsy | -0.056 |
oppy | -0.055 |
Remy | -0.050 |
chy | -0.049 |
bur | -0.049 |
Token | play |
Token ID | 711 |
Feature activation | +5.70e-02 |
Pos logit contributions | |
opsy | +0.013 |
oppy | +0.013 |
Remy | +0.012 |
chy | +0.012 |
uffy | +0.011 |
Neg logit contributions | |
digits | -0.008 |
folk | -0.008 |
sale | -0.008 |
number | -0.007 |
cheering | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | +3.87e-02 |
Pos logit contributions | |
opsy | +0.129 |
oppy | +0.128 |
Remy | +0.118 |
chy | +0.116 |
bur | +0.115 |
Neg logit contributions | |
folk | -0.080 |
digits | -0.077 |
sale | -0.072 |
customer | -0.066 |
letters | -0.065 |
Token | asked |
Token ID | 1965 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
opsy | +0.083 |
oppy | +0.082 |
Remy | +0.074 |
chy | +0.073 |
bur | +0.072 |
Neg logit contributions | |
folk | -0.058 |
digits | -0.056 |
sale | -0.053 |
letters | -0.049 |
customer | -0.048 |
Token | Fl |
Token ID | 1610 |
Feature activation | +3.93e-01 |
Pos logit contributions | |
folk | +0.353 |
sale | +0.352 |
digits | +0.339 |
letters | +0.318 |
customer | +0.311 |
Neg logit contributions | |
opsy | -0.496 |
oppy | -0.485 |
rig | -0.437 |
chy | -0.437 |
bur | -0.426 |
Token | oppy |
Token ID | 26696 |
Feature activation | +5.66e-01 |
Pos logit contributions | |
chy | +0.237 |
Remy | +0.217 |
rig | +0.186 |
refreshing | +0.161 |
itch | +0.127 |
Neg logit contributions | |
digits | -0.991 |
sale | -0.949 |
gu | -0.936 |
customer | -0.931 |
Amen | -0.924 |
Token | to |
Token ID | 284 |
Feature activation | +6.07e-03 |
Pos logit contributions | |
oppy | +1.130 |
opsy | +1.130 |
Remy | +1.040 |
chy | +1.011 |
bur | +1.001 |
Neg logit contributions | |
folk | -0.934 |
digits | -0.907 |
sale | -0.842 |
customer | -0.787 |
letters | -0.773 |
Token | join |
Token ID | 4654 |
Feature activation | -4.82e-02 |
Pos logit contributions | |
opsy | +0.013 |
oppy | +0.013 |
Remy | +0.012 |
chy | +0.012 |
bur | +0.012 |
Neg logit contributions | |
folk | -0.009 |
digits | -0.009 |
sale | -0.008 |
customer | -0.007 |
letters | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +6.25e-03 |
Pos logit contributions | |
folk | +0.068 |
digits | +0.064 |
sale | +0.062 |
letters | +0.056 |
customer | +0.056 |
Neg logit contributions | |
opsy | -0.108 |
oppy | -0.106 |
chy | -0.099 |
Remy | -0.096 |
bur | -0.095 |
Token | Fl |
Token ID | 1610 |
Feature activation | +3.94e-01 |
Pos logit contributions | |
opsy | +0.011 |
oppy | +0.011 |
chy | +0.009 |
rig | +0.008 |
uffy | +0.008 |
Neg logit contributions | |
sale | -0.012 |
folk | -0.012 |
digits | -0.011 |
letters | -0.010 |
customer | -0.010 |
Token | oppy |
Token ID | 26696 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
chy | +0.291 |
rig | +0.279 |
Remy | +0.259 |
refreshing | +0.255 |
Tig | +0.185 |
Neg logit contributions | |
digits | -0.859 |
gu | -0.812 |
winners | -0.809 |
sale | -0.792 |
ka | -0.788 |
Token | named |
Token ID | 3706 |
Feature activation | +6.21e-04 |
Pos logit contributions | |
opsy | +0.598 |
oppy | +0.585 |
chy | +0.541 |
Remy | +0.541 |
bur | +0.539 |
Neg logit contributions | |
folk | -0.303 |
digits | -0.299 |
sale | -0.291 |
Amen | -0.270 |
customer | -0.263 |
Token | Fl |
Token ID | 1610 |
Feature activation | +3.91e-01 |
Pos logit contributions | |
opsy | +0.001 |
oppy | +0.001 |
rig | +0.001 |
chy | +0.001 |
bur | +0.001 |
Neg logit contributions | |
sale | -0.001 |
customer | -0.001 |
digits | -0.001 |
folk | -0.001 |
Robert | -0.001 |
Token | oppy |
Token ID | 26696 |
Feature activation | +5.66e-01 |
Pos logit contributions | |
rig | +0.293 |
chy | +0.268 |
refreshing | +0.260 |
Remy | +0.204 |
web | +0.196 |
Neg logit contributions | |
customer | -0.932 |
digits | -0.878 |
sale | -0.873 |
buyer | -0.849 |
Amen | -0.826 |
Token | . |
Token ID | 13 |
Feature activation | +7.35e-03 |
Pos logit contributions | |
opsy | +1.431 |
oppy | +1.376 |
Remy | +1.289 |
bur | +1.276 |
rig | +1.250 |
Neg logit contributions | |
digits | -0.571 |
folk | -0.563 |
sale | -0.551 |
letters | -0.502 |
Amen | -0.487 |
Token | Fl |
Token ID | 1610 |
Feature activation | +3.92e-01 |
Pos logit contributions | |
opsy | +0.012 |
oppy | +0.011 |
rig | +0.010 |
chy | +0.009 |
uffy | +0.009 |
Neg logit contributions | |
sale | -0.014 |
folk | -0.014 |
digits | -0.014 |
customer | -0.013 |
Amen | -0.013 |
Token | oppy |
Token ID | 26696 |
Feature activation | +5.66e-01 |
Pos logit contributions | |
rig | +0.326 |
chy | +0.266 |
refreshing | +0.261 |
Remy | +0.235 |
iqu | +0.201 |
Neg logit contributions | |
customer | -0.829 |
digits | -0.807 |
ones | -0.776 |
three | -0.771 |
Amen | -0.764 |
Token | lived |
Token ID | 5615 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
opsy | +1.364 |
oppy | +1.311 |
Remy | +1.238 |
rig | +1.228 |
bur | +1.190 |
Neg logit contributions | |
digits | -0.602 |
folk | -0.600 |
sale | -0.562 |
letters | -0.552 |
Amen | -0.517 |
Token | in |
Token ID | 287 |
Feature activation | +3.67e-02 |
Pos logit contributions | |
opsy | +0.371 |
oppy | +0.371 |
chy | +0.342 |
Remy | +0.341 |
rig | +0.333 |
Neg logit contributions | |
digits | -0.217 |
folk | -0.214 |
sale | -0.206 |
customer | -0.200 |
letters | -0.189 |
Token | a |
Token ID | 257 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
opsy | +0.099 |
oppy | +0.097 |
uffy | +0.092 |
rig | +0.092 |
chy | +0.088 |
Neg logit contributions | |
sale | -0.031 |
digits | -0.030 |
letters | -0.029 |
customer | -0.029 |
Amen | -0.028 |
Token | modest |
Token ID | 12949 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
opsy | +0.032 |
oppy | +0.031 |
rig | +0.031 |
chy | +0.030 |
Remy | +0.029 |
Neg logit contributions | |
digits | -0.021 |
sale | -0.020 |
customer | -0.020 |
shop | -0.019 |
Amen | -0.019 |
Token | little |
Token ID | 1310 |
Feature activation | +6.98e-02 |
Pos logit contributions | |
sale | +0.017 |
folk | +0.017 |
digits | +0.017 |
customer | +0.016 |
letters | +0.015 |
Neg logit contributions | |
opsy | -0.022 |
oppy | -0.022 |
Remy | -0.021 |
chy | -0.021 |
rig | -0.020 |
Token | very |
Token ID | 845 |
Feature activation | -2.98e-02 |
Pos logit contributions | |
opsy | +0.140 |
oppy | +0.139 |
Remy | +0.128 |
chy | +0.121 |
bur | +0.121 |
Neg logit contributions | |
folk | -0.082 |
digits | -0.078 |
sale | -0.077 |
cheering | -0.069 |
Amen | -0.068 |
Token | sad |
Token ID | 6507 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
folk | +0.034 |
digits | +0.032 |
sale | +0.032 |
Amen | +0.030 |
cheering | +0.030 |
Neg logit contributions | |
opsy | -0.071 |
oppy | -0.068 |
Remy | -0.066 |
chy | -0.062 |
uffy | -0.062 |
Token | and |
Token ID | 290 |
Feature activation | +4.08e-02 |
Pos logit contributions | |
opsy | +0.049 |
oppy | +0.047 |
Remy | +0.043 |
rig | +0.042 |
chy | +0.041 |
Neg logit contributions | |
sale | -0.021 |
folk | -0.020 |
digits | -0.019 |
cheering | -0.018 |
customer | -0.017 |
Token | missed |
Token ID | 6825 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
opsy | +0.097 |
oppy | +0.093 |
rig | +0.087 |
uffy | +0.084 |
Remy | +0.080 |
Neg logit contributions | |
folk | -0.052 |
Amen | -0.046 |
begged | -0.045 |
sale | -0.045 |
asked | -0.044 |
Token | Fl |
Token ID | 1610 |
Feature activation | +3.92e-01 |
Pos logit contributions | |
opsy | +0.249 |
oppy | +0.248 |
rig | +0.233 |
chy | +0.218 |
uffy | +0.217 |
Neg logit contributions | |
sale | -0.111 |
folk | -0.103 |
digits | -0.097 |
cheering | -0.097 |
letters | -0.093 |
Token | opsy |
Token ID | 44522 |
Feature activation | +5.66e-01 |
Pos logit contributions | |
rig | +0.319 |
chy | +0.290 |
web | +0.231 |
Remy | +0.207 |
ches | +0.190 |
Neg logit contributions | |
digits | -0.826 |
customer | -0.826 |
ones | -0.813 |
sale | -0.806 |
folk | -0.802 |
Token | . |
Token ID | 13 |
Feature activation | +8.32e-03 |
Pos logit contributions | |
opsy | +1.317 |
oppy | +1.265 |
Remy | +1.176 |
rig | +1.160 |
bur | +1.145 |
Neg logit contributions | |
folk | -0.715 |
digits | -0.645 |
sale | -0.642 |
letters | -0.632 |
customer | -0.587 |
Token | He |
Token ID | 679 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
opsy | +0.021 |
oppy | +0.020 |
rig | +0.018 |
Remy | +0.018 |
chy | +0.018 |
Neg logit contributions | |
folk | -0.009 |
sale | -0.009 |
digits | -0.008 |
prices | -0.007 |
customer | -0.007 |
Token | wished |
Token ID | 16555 |
Feature activation | +1.37e-02 |
Pos logit contributions | |
folk | +0.012 |
asked | +0.012 |
digits | +0.012 |
begged | +0.012 |
sale | +0.011 |
Neg logit contributions | |
opsy | -0.031 |
oppy | -0.030 |
uffy | -0.028 |
Remy | -0.027 |
rig | -0.027 |
Token | he |
Token ID | 339 |
Feature activation | +4.36e-02 |
Pos logit contributions | |
rig | +0.037 |
opsy | +0.036 |
oppy | +0.036 |
web | +0.032 |
uffy | +0.032 |
Neg logit contributions | |
folk | -0.011 |
sale | -0.011 |
one | -0.010 |
them | -0.010 |
ones | -0.009 |
Token | had |
Token ID | 550 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
opsy | +0.112 |
oppy | +0.109 |
rig | +0.103 |
Remy | +0.100 |
bur | +0.100 |
Neg logit contributions | |
folk | -0.048 |
digits | -0.041 |
sale | -0.039 |
letters | -0.037 |
salesman | -0.036 |
Token | oppy |
Token ID | 26696 |
Feature activation | +5.66e-01 |
Pos logit contributions | |
chy | +0.237 |
Remy | +0.217 |
rig | +0.186 |
refreshing | +0.161 |
itch | +0.127 |
Neg logit contributions | |
digits | -0.991 |
sale | -0.949 |
gu | -0.936 |
customer | -0.931 |
Amen | -0.924 |
Token | to |
Token ID | 284 |
Feature activation | +6.07e-03 |
Pos logit contributions | |
oppy | +1.130 |
opsy | +1.130 |
Remy | +1.040 |
chy | +1.011 |
bur | +1.001 |
Neg logit contributions | |
folk | -0.934 |
digits | -0.907 |
sale | -0.842 |
customer | -0.787 |
letters | -0.773 |
Token | join |
Token ID | 4654 |
Feature activation | -4.82e-02 |
Pos logit contributions | |
opsy | +0.013 |
oppy | +0.013 |
Remy | +0.012 |
chy | +0.012 |
bur | +0.012 |
Neg logit contributions | |
folk | -0.009 |
digits | -0.009 |
sale | -0.008 |
customer | -0.007 |
letters | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +6.25e-03 |
Pos logit contributions | |
folk | +0.068 |
digits | +0.064 |
sale | +0.062 |
letters | +0.056 |
customer | +0.056 |
Neg logit contributions | |
opsy | -0.108 |
oppy | -0.106 |
chy | -0.099 |
Remy | -0.096 |
bur | -0.095 |
Token | Fl |
Token ID | 1610 |
Feature activation | +3.94e-01 |
Pos logit contributions | |
opsy | +0.011 |
oppy | +0.011 |
chy | +0.009 |
rig | +0.008 |
uffy | +0.008 |
Neg logit contributions | |
sale | -0.012 |
folk | -0.012 |
digits | -0.011 |
letters | -0.010 |
customer | -0.010 |
Token | oppy |
Token ID | 26696 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
chy | +0.291 |
rig | +0.279 |
Remy | +0.259 |
refreshing | +0.255 |
Tig | +0.185 |
Neg logit contributions | |
digits | -0.859 |
gu | -0.812 |
winners | -0.809 |
sale | -0.792 |
ka | -0.788 |
Token | thought |
Token ID | 1807 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
opsy | +1.278 |
oppy | +1.261 |
Remy | +1.170 |
chy | +1.143 |
bur | +1.126 |
Neg logit contributions | |
folk | -0.763 |
digits | -0.747 |
sale | -0.697 |
customer | -0.645 |
letters | -0.640 |
Token | for |
Token ID | 329 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
sale | +0.135 |
them | +0.123 |
one | +0.110 |
folk | +0.110 |
choosing | +0.104 |
Neg logit contributions | |
opsy | -0.415 |
oppy | -0.385 |
rig | -0.360 |
bur | -0.346 |
uffy | -0.343 |
Token | a |
Token ID | 257 |
Feature activation | +1.72e-02 |
Pos logit contributions | |
opsy | +0.225 |
oppy | +0.224 |
Remy | +0.215 |
bur | +0.205 |
chy | +0.203 |
Neg logit contributions | |
folk | -0.240 |
digits | -0.229 |
sale | -0.204 |
jury | -0.200 |
epad | -0.198 |
Token | moment |
Token ID | 2589 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
opsy | +0.039 |
oppy | +0.039 |
chy | +0.036 |
Remy | +0.036 |
bur | +0.035 |
Neg logit contributions | |
folk | -0.025 |
digits | -0.023 |
sale | -0.021 |
letters | -0.020 |
prices | -0.019 |
Token | and |
Token ID | 290 |
Feature activation | +4.01e-02 |
Pos logit contributions | |
folk | +0.217 |
digits | +0.210 |
sale | +0.200 |
customer | +0.186 |
letters | +0.183 |
Neg logit contributions | |
opsy | -0.297 |
oppy | -0.294 |
Remy | -0.268 |
chy | -0.263 |
bur | -0.263 |
Token | named |
Token ID | 3706 |
Feature activation | +5.46e-02 |
Pos logit contributions | |
opsy | +0.217 |
oppy | +0.215 |
Remy | +0.201 |
bur | +0.193 |
chy | +0.189 |
Neg logit contributions | |
folk | -0.139 |
digits | -0.136 |
sale | -0.129 |
customer | -0.117 |
letters | -0.116 |
Token | Fl |
Token ID | 1610 |
Feature activation | +4.14e-01 |
Pos logit contributions | |
opsy | +0.115 |
oppy | +0.111 |
bur | +0.102 |
chy | +0.100 |
rig | +0.098 |
Neg logit contributions | |
folk | -0.085 |
digits | -0.080 |
sale | -0.080 |
letters | -0.072 |
customer | -0.071 |
Token | opsy |
Token ID | 44522 |
Feature activation | +5.78e-01 |
Pos logit contributions | |
chy | +0.403 |
bur | +0.374 |
opsy | +0.373 |
Remy | +0.371 |
rig | +0.362 |
Neg logit contributions | |
folk | -0.962 |
sale | -0.936 |
digits | -0.935 |
customer | -0.928 |
Amen | -0.895 |
Token | . |
Token ID | 13 |
Feature activation | +4.98e-03 |
Pos logit contributions | |
opsy | +1.374 |
oppy | +1.327 |
rig | +1.275 |
bur | +1.243 |
Remy | +1.227 |
Neg logit contributions | |
folk | -0.650 |
digits | -0.593 |
sale | -0.571 |
Amen | -0.536 |
letters | -0.535 |
Token | Fl |
Token ID | 1610 |
Feature activation | +3.95e-01 |
Pos logit contributions | |
opsy | +0.008 |
oppy | +0.008 |
chy | +0.007 |
Remy | +0.007 |
bur | +0.007 |
Neg logit contributions | |
folk | -0.010 |
digits | -0.010 |
sale | -0.009 |
customer | -0.009 |
Amen | -0.009 |
Token | opsy |
Token ID | 44522 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
chy | +0.228 |
rig | +0.165 |
Remy | +0.137 |
ches | +0.136 |
bur | +0.127 |
Neg logit contributions | |
folk | -1.030 |
digits | -1.024 |
customer | -1.002 |
winners | -0.999 |
sale | -0.998 |
Token | loved |
Token ID | 6151 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
opsy | +1.256 |
oppy | +1.231 |
rig | +1.124 |
Remy | +1.121 |
chy | +1.087 |
Neg logit contributions | |
folk | -0.751 |
digits | -0.697 |
sale | -0.659 |
letters | -0.624 |
Amen | -0.608 |
Token | to |
Token ID | 284 |
Feature activation | -4.62e-03 |
Pos logit contributions | |
oppy | +0.068 |
opsy | +0.068 |
chy | +0.061 |
uffy | +0.059 |
Remy | +0.059 |
Neg logit contributions | |
sale | -0.036 |
digits | -0.036 |
folk | -0.035 |
letters | -0.033 |
customer | -0.032 |
Token | hop |
Token ID | 1725 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
digits | +0.007 |
folk | +0.007 |
sale | +0.007 |
letters | +0.006 |
Amen | +0.006 |
Neg logit contributions | |
opsy | -0.009 |
oppy | -0.009 |
Remy | -0.008 |
uffy | -0.008 |
chy | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | +2.50e-02 |
Pos logit contributions | |
opsy | +0.802 |
oppy | +0.762 |
uffy | +0.696 |
iqu | +0.681 |
Remy | +0.663 |
Neg logit contributions | |
sale | -0.250 |
folk | -0.217 |
one | -0.216 |
letters | -0.199 |
ones | -0.199 |
Token | play |
Token ID | 711 |
Feature activation | +3.95e-02 |
Pos logit contributions | |
opsy | +0.049 |
oppy | +0.048 |
uffy | +0.044 |
rig | +0.044 |
chy | +0.042 |
Neg logit contributions | |
folk | -0.037 |
digits | -0.036 |
Amen | -0.035 |
sale | -0.035 |
number | -0.033 |
Token | , |
Token ID | 11 |
Feature activation | -3.14e-02 |
Pos logit contributions | |
folk | +0.052 |
sale | +0.050 |
digits | +0.046 |
customer | +0.045 |
Amen | +0.039 |
Neg logit contributions | |
opsy | -0.133 |
oppy | -0.128 |
rig | -0.126 |
bur | -0.124 |
chy | -0.121 |
Token | whenever |
Token ID | 8797 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
folk | +0.039 |
customer | +0.035 |
digits | +0.035 |
sale | +0.035 |
cheering | +0.031 |
Neg logit contributions | |
opsy | -0.073 |
oppy | -0.071 |
bur | -0.068 |
rig | -0.067 |
chy | -0.066 |
Token | Anna |
Token ID | 11735 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
opsy | +0.269 |
oppy | +0.261 |
rig | +0.245 |
chy | +0.244 |
bur | +0.244 |
Neg logit contributions | |
folk | -0.148 |
digits | -0.140 |
sale | -0.137 |
customer | -0.131 |
Amen | -0.123 |
Token | took |
Token ID | 1718 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
folk | +0.016 |
digits | +0.015 |
sale | +0.014 |
customer | +0.014 |
number | +0.013 |
Neg logit contributions | |
opsy | -0.031 |
oppy | -0.031 |
Remy | -0.029 |
chy | -0.028 |
rig | -0.027 |
Token | Fl |
Token ID | 1610 |
Feature activation | +3.91e-01 |
Pos logit contributions | |
folk | +0.023 |
digits | +0.023 |
sale | +0.021 |
customer | +0.020 |
letters | +0.020 |
Neg logit contributions | |
opsy | -0.030 |
oppy | -0.029 |
Remy | -0.027 |
chy | -0.026 |
bur | -0.026 |
Token | opsy |
Token ID | 44522 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
chy | +0.286 |
rig | +0.204 |
Remy | +0.183 |
web | +0.152 |
refreshing | +0.126 |
Neg logit contributions | |
customer | -0.966 |
digits | -0.959 |
sale | -0.949 |
ones | -0.919 |
folk | -0.916 |
Token | for |
Token ID | 329 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
opsy | +1.349 |
oppy | +1.270 |
chy | +1.182 |
Remy | +1.179 |
rig | +1.173 |
Neg logit contributions | |
folk | -0.672 |
sale | -0.658 |
letters | -0.642 |
digits | -0.623 |
customer | -0.607 |
Token | a |
Token ID | 257 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
opsy | +0.240 |
oppy | +0.239 |
Remy | +0.225 |
bur | +0.216 |
chy | +0.215 |
Neg logit contributions | |
folk | -0.217 |
digits | -0.209 |
sale | -0.189 |
customer | -0.180 |
epad | -0.179 |
Token | walk |
Token ID | 2513 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
opsy | +0.028 |
oppy | +0.027 |
Remy | +0.025 |
chy | +0.024 |
bur | +0.024 |
Neg logit contributions | |
digits | -0.026 |
folk | -0.026 |
sale | -0.025 |
customer | -0.024 |
epad | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | -3.04e-02 |
Pos logit contributions | |
opsy | +0.437 |
oppy | +0.432 |
Remy | +0.401 |
chy | +0.391 |
bur | +0.389 |
Neg logit contributions | |
folk | -0.302 |
digits | -0.293 |
sale | -0.273 |
customer | -0.252 |
letters | -0.249 |
Token | she |
Token ID | 673 |
Feature activation | +8.58e-02 |
Pos logit contributions | |
folk | +0.045 |
digits | +0.042 |
sale | +0.041 |
customer | +0.040 |
cheering | +0.040 |
Neg logit contributions | |
opsy | -0.064 |
oppy | -0.063 |
rig | -0.057 |
chy | -0.056 |
bur | -0.055 |
Token | hurt |
Token ID | 5938 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
folk | +0.076 |
digits | +0.072 |
sale | +0.069 |
letters | +0.066 |
winners | +0.065 |
Neg logit contributions | |
opsy | -0.110 |
oppy | -0.105 |
Remy | -0.098 |
bur | -0.094 |
rig | -0.092 |
Token | my |
Token ID | 616 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
opsy | +0.250 |
oppy | +0.247 |
Remy | +0.233 |
bur | +0.223 |
chy | +0.221 |
Neg logit contributions | |
folk | -0.214 |
digits | -0.210 |
sale | -0.193 |
customer | -0.183 |
epad | -0.181 |
Token | ear |
Token ID | 1027 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
oppy | +0.033 |
opsy | +0.033 |
Remy | +0.031 |
chy | +0.030 |
rig | +0.029 |
Neg logit contributions | |
digits | -0.022 |
sale | -0.022 |
folk | -0.021 |
letters | -0.021 |
customer | -0.021 |
Token | ." |
Token ID | 526 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
folk | +0.154 |
digits | +0.150 |
sale | +0.139 |
customer | +0.127 |
letters | +0.126 |
Neg logit contributions | |
opsy | -0.264 |
oppy | -0.261 |
Remy | -0.245 |
bur | -0.238 |
chy | -0.236 |
Token | Fl |
Token ID | 1610 |
Feature activation | +3.93e-01 |
Pos logit contributions | |
folk | +0.177 |
| +0.168 |
sale | +0.165 |
digits | +0.163 |
Amen | +0.157 |
Neg logit contributions | |
opsy | -0.235 |
oppy | -0.224 |
rig | -0.221 |
chy | -0.221 |
iqu | -0.195 |
Token | oppy |
Token ID | 26696 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
chy | +0.324 |
rig | +0.289 |
refreshing | +0.222 |
cob | +0.209 |
web | +0.181 |
Neg logit contributions | |
customer | -0.842 |
digits | -0.838 |
folk | -0.829 |
ones | -0.818 |
buyer | -0.816 |
Token | apologized |
Token ID | 22893 |
Feature activation | -5.73e-03 |
Pos logit contributions | |
opsy | +1.227 |
oppy | +1.216 |
Remy | +1.122 |
chy | +1.091 |
bur | +1.083 |
Neg logit contributions | |
folk | -0.811 |
digits | -0.791 |
sale | -0.737 |
letters | -0.688 |
customer | -0.677 |
Token | and |
Token ID | 290 |
Feature activation | +4.18e-02 |
Pos logit contributions | |
folk | +0.007 |
sale | +0.007 |
digits | +0.007 |
letters | +0.007 |
customer | +0.006 |
Neg logit contributions | |
opsy | -0.013 |
oppy | -0.013 |
chy | -0.012 |
Remy | -0.012 |
bur | -0.012 |
Token | they |
Token ID | 484 |
Feature activation | +4.48e-02 |
Pos logit contributions | |
opsy | +0.097 |
oppy | +0.097 |
rig | +0.095 |
chy | +0.094 |
uffy | +0.089 |
Neg logit contributions | |
folk | -0.044 |
begged | -0.041 |
sale | -0.040 |
salesman | -0.040 |
asked | -0.040 |
Token | continued |
Token ID | 3767 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
opsy | +0.103 |
oppy | +0.103 |
rig | +0.101 |
Remy | +0.094 |
chy | +0.093 |
Neg logit contributions | |
folk | -0.049 |
digits | -0.045 |
sale | -0.044 |
numbered | -0.041 |
letters | -0.040 |
Token | to |
Token ID | 284 |
Feature activation | +8.80e-03 |
Pos logit contributions | |
opsy | +0.374 |
oppy | +0.370 |
Remy | +0.339 |
chy | +0.335 |
bur | +0.329 |
Neg logit contributions | |
folk | -0.246 |
digits | -0.243 |
sale | -0.230 |
customer | -0.210 |
letters | -0.210 |
Token | . |
Token ID | 13 |
Feature activation | +6.37e-03 |
Pos logit contributions | |
opsy | +0.227 |
oppy | +0.219 |
bur | +0.200 |
Remy | +0.198 |
rig | +0.192 |
Neg logit contributions | |
sale | -0.098 |
folk | -0.097 |
digits | -0.090 |
letters | -0.083 |
epad | -0.083 |
Token | The |
Token ID | 383 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
opsy | +0.013 |
oppy | +0.012 |
Remy | +0.011 |
chy | +0.011 |
bur | +0.011 |
Neg logit contributions | |
folk | -0.011 |
digits | -0.010 |
sale | -0.010 |
customer | -0.009 |
letters | -0.009 |
Token | car |
Token ID | 1097 |
Feature activation | -7.85e-02 |
Pos logit contributions | |
opsy | +0.206 |
oppy | +0.204 |
Remy | +0.187 |
chy | +0.182 |
bur | +0.180 |
Neg logit contributions | |
folk | -0.178 |
digits | -0.173 |
sale | -0.164 |
customer | -0.153 |
letters | -0.151 |
Token | hit |
Token ID | 2277 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
folk | +0.124 |
digits | +0.117 |
sale | +0.114 |
letters | +0.104 |
customer | +0.101 |
Neg logit contributions | |
opsy | -0.169 |
oppy | -0.166 |
Remy | -0.152 |
bur | -0.143 |
chy | -0.141 |
Token | Fl |
Token ID | 1610 |
Feature activation | +3.91e-01 |
Pos logit contributions | |
oppy | +0.030 |
rig | +0.029 |
opsy | +0.028 |
uffy | +0.024 |
chy | +0.024 |
Neg logit contributions | |
sale | -0.025 |
them | -0.025 |
one | -0.025 |
folk | -0.025 |
letters | -0.024 |
Token | opsy |
Token ID | 44522 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
chy | +0.321 |
rig | +0.294 |
Remy | +0.205 |
refreshing | +0.193 |
othe | +0.174 |
Neg logit contributions | |
ones | -0.834 |
digits | -0.805 |
customer | -0.798 |
gu | -0.784 |
be | -0.779 |
Token | and |
Token ID | 290 |
Feature activation | +3.99e-02 |
Pos logit contributions | |
opsy | +1.375 |
oppy | +1.329 |
Remy | +1.238 |
uffy | +1.140 |
rig | +1.136 |
Neg logit contributions | |
folk | -0.664 |
sale | -0.647 |
digits | -0.643 |
letters | -0.591 |
number | -0.564 |
Token | she |
Token ID | 673 |
Feature activation | +8.56e-02 |
Pos logit contributions | |
oppy | +0.076 |
opsy | +0.076 |
rig | +0.074 |
uffy | +0.063 |
chy | +0.062 |
Neg logit contributions | |
folk | -0.068 |
sale | -0.066 |
letters | -0.060 |
digits | -0.060 |
begged | -0.058 |
Token | was |
Token ID | 373 |
Feature activation | +6.10e-02 |
Pos logit contributions | |
opsy | +0.156 |
oppy | +0.145 |
rig | +0.137 |
Remy | +0.136 |
uffy | +0.116 |
Neg logit contributions | |
folk | -0.156 |
digits | -0.150 |
sale | -0.143 |
letters | -0.135 |
cheering | -0.133 |
Token | gone |
Token ID | 3750 |
Feature activation | +5.71e-02 |
Pos logit contributions | |
opsy | +0.127 |
oppy | +0.124 |
Remy | +0.114 |
bur | +0.105 |
uffy | +0.104 |
Neg logit contributions | |
folk | -0.097 |
digits | -0.093 |
sale | -0.091 |
cheering | -0.084 |
letters | -0.082 |
Token | . |
Token ID | 13 |
Feature activation | +7.49e-03 |
Pos logit contributions | |
opsy | +0.136 |
oppy | +0.131 |
Remy | +0.124 |
rig | +0.114 |
uffy | +0.111 |
Neg logit contributions | |
folk | -0.069 |
sale | -0.069 |
digits | -0.066 |
epad | -0.060 |
cheering | -0.059 |
Token | smooth |
Token ID | 7209 |
Feature activation | +9.63e-02 |
Pos logit contributions | |
digits | +0.145 |
folk | +0.145 |
sale | +0.136 |
letters | +0.131 |
customer | +0.125 |
Neg logit contributions | |
opsy | -0.215 |
oppy | -0.214 |
Remy | -0.192 |
bur | -0.187 |
uffy | -0.186 |
Token | and |
Token ID | 290 |
Feature activation | +3.89e-02 |
Pos logit contributions | |
opsy | +0.196 |
oppy | +0.194 |
Remy | +0.178 |
bur | +0.172 |
chy | +0.171 |
Neg logit contributions | |
folk | -0.155 |
digits | -0.151 |
sale | -0.142 |
letters | -0.132 |
customer | -0.131 |
Token | shiny |
Token ID | 22441 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
opsy | +0.073 |
oppy | +0.072 |
Remy | +0.065 |
bur | +0.063 |
chy | +0.063 |
Neg logit contributions | |
folk | -0.068 |
digits | -0.067 |
sale | -0.063 |
letters | -0.059 |
customer | -0.059 |
Token | . |
Token ID | 13 |
Feature activation | +4.37e-03 |
Pos logit contributions | |
opsy | +0.022 |
oppy | +0.022 |
chy | +0.020 |
Remy | +0.020 |
bur | +0.020 |
Neg logit contributions | |
folk | -0.022 |
digits | -0.021 |
sale | -0.020 |
customer | -0.019 |
jury | -0.019 |
Token | Fl |
Token ID | 1610 |
Feature activation | +3.94e-01 |
Pos logit contributions | |
opsy | +0.009 |
oppy | +0.009 |
chy | +0.008 |
rig | +0.008 |
bur | +0.008 |
Neg logit contributions | |
folk | -0.006 |
digits | -0.006 |
sale | -0.006 |
Amen | -0.006 |
| -0.005 |
Token | oppy |
Token ID | 26696 |
Feature activation | +5.64e-01 |
Pos logit contributions | |
rig | +0.325 |
chy | +0.290 |
Remy | +0.279 |
refreshing | +0.259 |
ches | +0.209 |
Neg logit contributions | |
digits | -0.834 |
customer | -0.792 |
winners | -0.785 |
ka | -0.781 |
ones | -0.775 |
Token | was |
Token ID | 373 |
Feature activation | +6.18e-02 |
Pos logit contributions | |
opsy | +1.244 |
oppy | +1.231 |
Remy | +1.140 |
chy | +1.100 |
bur | +1.082 |
Neg logit contributions | |
folk | -0.779 |
digits | -0.770 |
sale | -0.723 |
letters | -0.668 |
customer | -0.657 |
Token | so |
Token ID | 523 |
Feature activation | -9.14e-02 |
Pos logit contributions | |
opsy | +0.123 |
oppy | +0.115 |
Remy | +0.114 |
chy | +0.112 |
bur | +0.108 |
Neg logit contributions | |
folk | -0.110 |
digits | -0.104 |
sale | -0.096 |
jury | -0.091 |
bench | -0.090 |
Token | happy |
Token ID | 3772 |
Feature activation | -6.12e-02 |
Pos logit contributions | |
digits | +0.097 |
folk | +0.097 |
sale | +0.085 |
cheering | +0.079 |
letters | +0.079 |
Neg logit contributions | |
oppy | -0.229 |
opsy | -0.228 |
Remy | -0.213 |
bur | -0.211 |
chy | -0.209 |
Token | to |
Token ID | 284 |
Feature activation | +7.19e-03 |
Pos logit contributions | |
sale | +0.071 |
digits | +0.070 |
folk | +0.069 |
cheering | +0.063 |
letters | +0.063 |
Neg logit contributions | |
opsy | -0.146 |
oppy | -0.144 |
chy | -0.133 |
rig | -0.128 |
bur | -0.126 |
Token | have |
Token ID | 423 |
Feature activation | -7.49e-02 |
Pos logit contributions | |
opsy | +0.015 |
oppy | +0.015 |
Remy | +0.014 |
chy | +0.014 |
bur | +0.014 |
Neg logit contributions | |
folk | -0.012 |
digits | -0.011 |
sale | -0.010 |
prices | -0.010 |
letters | -0.010 |
Token | too |
Token ID | 1165 |
Feature activation | -8.54e-02 |
Pos logit contributions | |
folk | +0.184 |
digits | +0.182 |
sale | +0.169 |
letters | +0.156 |
customer | +0.155 |
Neg logit contributions | |
opsy | -0.263 |
oppy | -0.260 |
Remy | -0.239 |
bur | -0.232 |
chy | -0.232 |
Token | . |
Token ID | 13 |
Feature activation | +5.59e-03 |
Pos logit contributions | |
digits | +0.100 |
sale | +0.096 |
folk | +0.094 |
letters | +0.083 |
customer | +0.081 |
Neg logit contributions | |
opsy | -0.203 |
oppy | -0.200 |
Remy | -0.184 |
bur | -0.179 |
chy | -0.178 |
Token |
|
Token ID | 198 |
Feature activation | +3.30e-02 |
Pos logit contributions | |
opsy | +0.014 |
oppy | +0.014 |
chy | +0.014 |
rig | +0.013 |
uffy | +0.013 |
Neg logit contributions | |
digits | -0.005 |
folk | -0.005 |
sale | -0.005 |
letters | -0.004 |
Amen | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | +3.27e-02 |
Pos logit contributions | |
opsy | +0.075 |
oppy | +0.073 |
chy | +0.068 |
bur | +0.068 |
Remy | +0.067 |
Neg logit contributions | |
digits | -0.043 |
folk | -0.042 |
sale | -0.040 |
customer | -0.036 |
Amen | -0.036 |
Token | H |
Token ID | 39 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
opsy | +0.069 |
oppy | +0.067 |
chy | +0.062 |
uffy | +0.062 |
bur | +0.061 |
Neg logit contributions | |
digits | -0.045 |
sale | -0.042 |
folk | -0.042 |
customer | -0.039 |
salesman | -0.038 |
Token | oppy |
Token ID | 26696 |
Feature activation | +5.64e-01 |
Pos logit contributions | |
bur | +0.074 |
rig | +0.072 |
Remy | +0.066 |
opsy | +0.063 |
refreshing | +0.054 |
Neg logit contributions | |
digits | -0.325 |
epad | -0.320 |
bench | -0.317 |
sale | -0.314 |
customer | -0.301 |
Token | hopped |
Token ID | 45230 |
Feature activation | +2.20e-01 |
Pos logit contributions | |
opsy | +1.272 |
oppy | +1.257 |
Remy | +1.162 |
chy | +1.135 |
bur | +1.103 |
Neg logit contributions | |
folk | -0.766 |
digits | -0.754 |
sale | -0.694 |
letters | -0.650 |
customer | -0.629 |
Token | and |
Token ID | 290 |
Feature activation | +3.85e-02 |
Pos logit contributions | |
opsy | +0.516 |
oppy | +0.516 |
uffy | +0.458 |
chy | +0.456 |
Remy | +0.444 |
Neg logit contributions | |
digits | -0.271 |
folk | -0.271 |
sale | -0.260 |
customer | -0.244 |
salesman | -0.238 |
Token | hopped |
Token ID | 45230 |
Feature activation | +2.20e-01 |
Pos logit contributions | |
opsy | +0.077 |
oppy | +0.075 |
chy | +0.067 |
uffy | +0.066 |
rig | +0.065 |
Neg logit contributions | |
folk | -0.062 |
digits | -0.062 |
sale | -0.057 |
letters | -0.054 |
cheering | -0.054 |
Token | , |
Token ID | 11 |
Feature activation | -2.83e-02 |
Pos logit contributions | |
opsy | +0.520 |
oppy | +0.514 |
uffy | +0.459 |
chy | +0.448 |
bur | +0.446 |
Neg logit contributions | |
folk | -0.275 |
digits | -0.274 |
sale | -0.265 |
customer | -0.237 |
cheering | -0.235 |
Token | but |
Token ID | 475 |
Feature activation | -5.82e-03 |
Pos logit contributions | |
folk | +0.036 |
digits | +0.035 |
sale | +0.032 |
cheering | +0.031 |
letters | +0.029 |
Neg logit contributions | |
oppy | -0.068 |
opsy | -0.067 |
uffy | -0.059 |
Remy | -0.059 |
chy | -0.057 |
Token | man |
Token ID | 805 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
oppy | +0.489 |
opsy | +0.487 |
bur | +0.449 |
Remy | +0.439 |
rig | +0.433 |
Neg logit contributions | |
folk | -0.661 |
boxes | -0.601 |
sale | -0.592 |
digits | -0.584 |
customer | -0.579 |
Token | back |
Token ID | 736 |
Feature activation | +1.45e-02 |
Pos logit contributions | |
folk | +0.304 |
digits | +0.298 |
sale | +0.279 |
letters | +0.262 |
customer | +0.258 |
Neg logit contributions | |
opsy | -0.417 |
oppy | -0.409 |
Remy | -0.379 |
bur | -0.366 |
chy | -0.363 |
Token | together |
Token ID | 1978 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
opsy | +0.032 |
oppy | +0.031 |
Remy | +0.029 |
chy | +0.027 |
bur | +0.027 |
Neg logit contributions | |
folk | -0.022 |
digits | -0.021 |
sale | -0.020 |
letters | -0.018 |
customer | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +4.38e-03 |
Pos logit contributions | |
folk | +0.183 |
digits | +0.177 |
sale | +0.169 |
customer | +0.154 |
letters | +0.153 |
Neg logit contributions | |
opsy | -0.268 |
oppy | -0.265 |
Remy | -0.243 |
bur | -0.235 |
chy | -0.235 |
Token | Fl |
Token ID | 1610 |
Feature activation | +3.90e-01 |
Pos logit contributions | |
opsy | +0.009 |
oppy | +0.009 |
chy | +0.008 |
rig | +0.007 |
Remy | +0.007 |
Neg logit contributions | |
folk | -0.007 |
sale | -0.007 |
digits | -0.007 |
customer | -0.006 |
Amen | -0.006 |
Token | opsy |
Token ID | 44522 |
Feature activation | +5.64e-01 |
Pos logit contributions | |
chy | +0.354 |
rig | +0.309 |
web | +0.279 |
cob | +0.241 |
rubble | +0.236 |
Neg logit contributions | |
ones | -0.778 |
digits | -0.771 |
ams | -0.768 |
be | -0.763 |
customer | -0.746 |
Token | was |
Token ID | 373 |
Feature activation | +5.87e-02 |
Pos logit contributions | |
opsy | +1.233 |
oppy | +1.215 |
Remy | +1.126 |
chy | +1.073 |
rig | +1.063 |
Neg logit contributions | |
folk | -0.794 |
digits | -0.779 |
sale | -0.702 |
letters | -0.676 |
cheering | -0.640 |
Token | so |
Token ID | 523 |
Feature activation | -9.47e-02 |
Pos logit contributions | |
opsy | +0.119 |
oppy | +0.115 |
chy | +0.110 |
Remy | +0.109 |
bur | +0.106 |
Neg logit contributions | |
folk | -0.097 |
digits | -0.094 |
sale | -0.086 |
jury | -0.082 |
prices | -0.080 |
Token | brave |
Token ID | 14802 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
folk | +0.120 |
digits | +0.117 |
sale | +0.106 |
customer | +0.098 |
letters | +0.096 |
Neg logit contributions | |
opsy | -0.225 |
oppy | -0.223 |
Remy | -0.209 |
bur | -0.204 |
chy | -0.203 |
Token | and |
Token ID | 290 |
Feature activation | +3.59e-02 |
Pos logit contributions | |
opsy | +0.225 |
oppy | +0.220 |
Remy | +0.196 |
chy | +0.194 |
bur | +0.193 |
Neg logit contributions | |
folk | -0.171 |
digits | -0.164 |
sale | -0.156 |
customer | -0.145 |
letters | -0.145 |
Token | worked |
Token ID | 3111 |
Feature activation | -6.18e-02 |
Pos logit contributions | |
oppy | +0.077 |
opsy | +0.077 |
chy | +0.067 |
Remy | +0.067 |
uffy | +0.067 |
Neg logit contributions | |
folk | -0.052 |
digits | -0.051 |
sale | -0.048 |
letters | -0.046 |
customer | -0.046 |
Token | bush |
Token ID | 24484 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
opsy | +0.030 |
oppy | +0.029 |
Remy | +0.026 |
uffy | +0.025 |
chy | +0.025 |
Neg logit contributions | |
digits | -0.026 |
sale | -0.026 |
folk | -0.026 |
customer | -0.024 |
letters | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +5.12e-03 |
Pos logit contributions | |
opsy | +0.436 |
oppy | +0.429 |
Remy | +0.395 |
chy | +0.382 |
bur | +0.378 |
Neg logit contributions | |
folk | -0.292 |
digits | -0.283 |
sale | -0.276 |
customer | -0.249 |
letters | -0.244 |
Token |
|
Token ID | 198 |
Feature activation | +3.13e-02 |
Pos logit contributions | |
opsy | +0.012 |
oppy | +0.012 |
chy | +0.011 |
rig | +0.011 |
bur | +0.011 |
Neg logit contributions | |
digits | -0.006 |
folk | -0.006 |
sale | -0.005 |
customer | -0.005 |
Amen | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +3.09e-02 |
Pos logit contributions | |
opsy | +0.087 |
chy | +0.084 |
oppy | +0.084 |
bur | +0.080 |
rig | +0.077 |
Neg logit contributions | |
digits | -0.020 |
| -0.018 |
sale | -0.017 |
John | -0.016 |
Amen | -0.016 |
Token | P |
Token ID | 47 |
Feature activation | +5.98e-02 |
Pos logit contributions | |
opsy | +0.070 |
chy | +0.068 |
oppy | +0.068 |
rig | +0.066 |
bur | +0.063 |
Neg logit contributions | |
digits | -0.038 |
sale | -0.033 |
John | -0.033 |
customer | -0.033 |
folk | -0.032 |
Token | oppy |
Token ID | 26696 |
Feature activation | +5.64e-01 |
Pos logit contributions | |
chy | +0.062 |
opsy | +0.050 |
refreshing | +0.043 |
bur | +0.042 |
rig | +0.039 |
Neg logit contributions | |
digits | -0.149 |
customer | -0.139 |
sale | -0.138 |
iers | -0.137 |
jury | -0.131 |
Token | cries |
Token ID | 24691 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
opsy | +1.277 |
oppy | +1.219 |
Remy | +1.115 |
chy | +1.079 |
rig | +1.053 |
Neg logit contributions | |
digits | -0.801 |
folk | -0.792 |
sale | -0.750 |
letters | -0.732 |
prices | -0.685 |
Token | . |
Token ID | 13 |
Feature activation | +5.17e-03 |
Pos logit contributions | |
opsy | +0.548 |
oppy | +0.521 |
rig | +0.487 |
uffy | +0.458 |
chy | +0.451 |
Neg logit contributions | |
digits | -0.199 |
letters | -0.194 |
folk | -0.192 |
sale | -0.186 |
cheering | -0.178 |
Token | She |
Token ID | 1375 |
Feature activation | +3.15e-02 |
Pos logit contributions | |
opsy | +0.012 |
oppy | +0.012 |
rig | +0.011 |
chy | +0.011 |
bur | +0.010 |
Neg logit contributions | |
digits | -0.006 |
sale | -0.006 |
folk | -0.006 |
customer | -0.005 |
letters | -0.005 |
Token | has |
Token ID | 468 |
Feature activation | +4.19e-02 |
Pos logit contributions | |
opsy | +0.081 |
oppy | +0.077 |
rig | +0.069 |
Remy | +0.068 |
uffy | +0.066 |
Neg logit contributions | |
digits | -0.033 |
letters | -0.033 |
number | -0.030 |
begged | -0.028 |
prices | -0.026 |
Token | no |
Token ID | 645 |
Feature activation | -5.67e-02 |
Pos logit contributions | |
opsy | +0.090 |
oppy | +0.088 |
Remy | +0.081 |
chy | +0.079 |
bur | +0.078 |
Neg logit contributions | |
folk | -0.063 |
digits | -0.062 |
sale | -0.059 |
letters | -0.054 |
customer | -0.054 |
Token | ! |
Token ID | 0 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
opsy | +0.353 |
oppy | +0.345 |
Remy | +0.326 |
bur | +0.324 |
chy | +0.318 |
Neg logit contributions | |
folk | -0.219 |
digits | -0.214 |
sale | -0.199 |
letters | -0.184 |
cheering | -0.178 |
Token | Be |
Token ID | 1355 |
Feature activation | -5.23e-02 |
Pos logit contributions | |
opsy | +0.221 |
oppy | +0.217 |
Remy | +0.201 |
chy | +0.200 |
bur | +0.199 |
Neg logit contributions | |
folk | -0.158 |
digits | -0.152 |
sale | -0.144 |
customer | -0.133 |
letters | -0.130 |
Token | careful |
Token ID | 8161 |
Feature activation | +4.67e-02 |
Pos logit contributions | |
folk | +0.098 |
digits | +0.094 |
sale | +0.088 |
salesman | +0.082 |
customer | +0.082 |
Neg logit contributions | |
opsy | -0.096 |
oppy | -0.095 |
Remy | -0.087 |
chy | -0.085 |
bur | -0.083 |
Token | , |
Token ID | 11 |
Feature activation | -2.86e-02 |
Pos logit contributions | |
opsy | +0.097 |
oppy | +0.096 |
Remy | +0.089 |
chy | +0.087 |
bur | +0.086 |
Neg logit contributions | |
folk | -0.074 |
digits | -0.072 |
sale | -0.067 |
customer | -0.062 |
letters | -0.061 |
Token | Fl |
Token ID | 1610 |
Feature activation | +3.92e-01 |
Pos logit contributions | |
sale | +0.042 |
folk | +0.041 |
letters | +0.038 |
customer | +0.037 |
digits | +0.037 |
Neg logit contributions | |
opsy | -0.058 |
oppy | -0.057 |
rig | -0.052 |
uffy | -0.049 |
chy | -0.048 |
Token | oppy |
Token ID | 26696 |
Feature activation | +5.64e-01 |
Pos logit contributions | |
chy | +0.291 |
rig | +0.280 |
cob | +0.223 |
web | +0.215 |
refreshing | +0.209 |
Neg logit contributions | |
customer | -0.820 |
ones | -0.815 |
sale | -0.813 |
digits | -0.813 |
winners | -0.779 |
Token | . |
Token ID | 13 |
Feature activation | +7.48e-03 |
Pos logit contributions | |
opsy | +1.323 |
oppy | +1.290 |
Remy | +1.212 |
bur | +1.187 |
chy | +1.155 |
Neg logit contributions | |
folk | -0.710 |
digits | -0.690 |
sale | -0.653 |
letters | -0.593 |
customer | -0.584 |
Token | You |
Token ID | 921 |
Feature activation | -5.12e-02 |
Pos logit contributions | |
opsy | +0.017 |
oppy | +0.016 |
chy | +0.015 |
Remy | +0.015 |
bur | +0.015 |
Neg logit contributions | |
folk | -0.011 |
digits | -0.010 |
sale | -0.009 |
Amen | -0.009 |
customer | -0.009 |
Token | hurt |
Token ID | 5938 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
folk | +0.076 |
digits | +0.072 |
sale | +0.069 |
letters | +0.066 |
winners | +0.065 |
Neg logit contributions | |
opsy | -0.110 |
oppy | -0.105 |
Remy | -0.098 |
bur | -0.094 |
rig | -0.092 |
Token | my |
Token ID | 616 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
opsy | +0.250 |
oppy | +0.247 |
Remy | +0.233 |
bur | +0.223 |
chy | +0.221 |
Neg logit contributions | |
folk | -0.214 |
digits | -0.210 |
sale | -0.193 |
customer | -0.183 |
epad | -0.181 |
Token | ear |
Token ID | 1027 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
oppy | +0.033 |
opsy | +0.033 |
Remy | +0.031 |
chy | +0.030 |
rig | +0.029 |
Neg logit contributions | |
digits | -0.022 |
sale | -0.022 |
folk | -0.021 |
letters | -0.021 |
customer | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | +2.88e-02 |
Pos logit contributions | |
opsy | +0.060 |
oppy | +0.059 |
Remy | +0.054 |
chy | +0.053 |
bur | +0.053 |
Neg logit contributions | |
folk | -0.047 |
digits | -0.046 |
sale | -0.043 |
customer | -0.040 |
letters | -0.039 |
Token | One |
Token ID | 3198 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
opsy | +0.073 |
oppy | +0.073 |
Remy | +0.068 |
chy | +0.066 |
bur | +0.066 |
Neg logit contributions | |
folk | -0.032 |
digits | -0.031 |
sale | -0.028 |
customer | -0.025 |
letters | -0.025 |
Token | day |
Token ID | 1110 |
Feature activation | +4.13e-03 |
Pos logit contributions | |
one | +0.051 |
color | +0.040 |
three | +0.033 |
minute | +0.029 |
3 | +0.016 |
Neg logit contributions | |
oppy | -0.378 |
iqu | -0.372 |
uffy | -0.362 |
opsy | -0.357 |
istered | -0.353 |
Token | , |
Token ID | 11 |
Feature activation | -2.97e-02 |
Pos logit contributions | |
opsy | +0.011 |
oppy | +0.010 |
rig | +0.010 |
chy | +0.010 |
bur | +0.010 |
Neg logit contributions | |
folk | -0.004 |
digits | -0.004 |
sale | -0.004 |
customer | -0.003 |
letters | -0.003 |
Token | Fl |
Token ID | 1610 |
Feature activation | +3.89e-01 |
Pos logit contributions | |
sale | +0.056 |
folk | +0.055 |
customer | +0.054 |
digits | +0.053 |
one | +0.052 |
Neg logit contributions | |
opsy | -0.047 |
oppy | -0.046 |
rig | -0.046 |
bur | -0.040 |
chy | -0.040 |
Token | opsy |
Token ID | 44522 |
Feature activation | +5.63e-01 |
Pos logit contributions | |
chy | +0.414 |
rig | +0.346 |
rubble | +0.331 |
ches | +0.307 |
webs | +0.285 |
Neg logit contributions | |
digits | -0.699 |
ones | -0.686 |
customer | -0.685 |
three | -0.673 |
be | -0.665 |
Token | hopped |
Token ID | 45230 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
opsy | +1.500 |
oppy | +1.460 |
rig | +1.369 |
chy | +1.367 |
Remy | +1.356 |
Neg logit contributions | |
folk | -0.487 |
digits | -0.479 |
sale | -0.424 |
asked | -0.417 |
letters | -0.396 |
Token | up |
Token ID | 510 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
opsy | +0.549 |
oppy | +0.542 |
uffy | +0.485 |
Remy | +0.465 |
bur | +0.452 |
Neg logit contributions | |
folk | -0.245 |
sale | -0.243 |
digits | -0.235 |
letters | -0.218 |
cheering | -0.217 |
Token | to |
Token ID | 284 |
Feature activation | +4.92e-03 |
Pos logit contributions | |
opsy | +0.244 |
oppy | +0.241 |
Remy | +0.212 |
uffy | +0.206 |
chy | +0.206 |
Neg logit contributions | |
sale | -0.129 |
folk | -0.123 |
digits | -0.121 |
customer | -0.111 |
bench | -0.110 |
Token | the |
Token ID | 262 |
Feature activation | +4.07e-02 |
Pos logit contributions | |
opsy | +0.012 |
oppy | +0.012 |
chy | +0.012 |
uffy | +0.012 |
rig | +0.012 |
Neg logit contributions | |
sale | -0.005 |
digits | -0.005 |
customer | -0.004 |
number | -0.004 |
one | -0.004 |
Token | gate |
Token ID | 8946 |
Feature activation | +2.84e-03 |
Pos logit contributions | |
oppy | +0.088 |
opsy | +0.083 |
uffy | +0.075 |
rig | +0.074 |
Remy | +0.073 |
Neg logit contributions | |
sale | -0.060 |
customer | -0.057 |
number | -0.056 |
letters | -0.055 |
digits | -0.055 |
Token | might |
Token ID | 1244 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
folk | +0.072 |
digits | +0.068 |
sale | +0.066 |
number | +0.064 |
winners | +0.062 |
Neg logit contributions | |
opsy | -0.121 |
oppy | -0.110 |
Remy | -0.106 |
rig | -0.103 |
bur | -0.103 |
Token | get |
Token ID | 651 |
Feature activation | +1.35e-02 |
Pos logit contributions | |
opsy | +0.420 |
oppy | +0.407 |
uffy | +0.380 |
rig | +0.368 |
ilo | +0.349 |
Neg logit contributions | |
sale | -0.191 |
number | -0.179 |
one | -0.163 |
cheering | -0.163 |
winners | -0.154 |
Token | hurt |
Token ID | 5938 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
opsy | +0.028 |
oppy | +0.027 |
Remy | +0.024 |
rig | +0.023 |
chy | +0.023 |
Neg logit contributions | |
digits | -0.020 |
folk | -0.020 |
letters | -0.019 |
sale | -0.019 |
cheering | -0.019 |
Token | ." |
Token ID | 526 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
opsy | +0.268 |
oppy | +0.267 |
Remy | +0.251 |
chy | +0.244 |
bur | +0.242 |
Neg logit contributions | |
folk | -0.181 |
digits | -0.178 |
sale | -0.162 |
customer | -0.152 |
letters | -0.148 |
Token | Fl |
Token ID | 1610 |
Feature activation | +3.90e-01 |
Pos logit contributions | |
folk | +0.226 |
sale | +0.225 |
digits | +0.218 |
prices | +0.199 |
Amen | +0.197 |
Neg logit contributions | |
opsy | -0.214 |
oppy | -0.207 |
chy | -0.182 |
rig | -0.173 |
bur | -0.167 |
Token | opsy |
Token ID | 44522 |
Feature activation | +5.63e-01 |
Pos logit contributions | |
chy | +0.375 |
rig | +0.296 |
ches | +0.245 |
webs | +0.217 |
web | +0.213 |
Neg logit contributions | |
digits | -0.835 |
customer | -0.800 |
begged | -0.777 |
" | -0.775 |
ones | -0.769 |
Token | listened |
Token ID | 16399 |
Feature activation | -7.76e-02 |
Pos logit contributions | |
opsy | +1.247 |
oppy | +1.231 |
Remy | +1.123 |
chy | +1.102 |
bur | +1.085 |
Neg logit contributions | |
folk | -0.743 |
digits | -0.725 |
sale | -0.667 |
letters | -0.638 |
begged | -0.626 |
Token | to |
Token ID | 284 |
Feature activation | +4.96e-03 |
Pos logit contributions | |
folk | +0.126 |
digits | +0.124 |
sale | +0.116 |
customer | +0.108 |
letters | +0.107 |
Neg logit contributions | |
opsy | -0.157 |
oppy | -0.155 |
Remy | -0.143 |
bur | -0.138 |
chy | -0.137 |
Token | her |
Token ID | 607 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
oppy | +0.013 |
opsy | +0.013 |
rig | +0.013 |
uffy | +0.012 |
chy | +0.012 |
Neg logit contributions | |
them | -0.004 |
sale | -0.004 |
folk | -0.004 |
one | -0.003 |
cheering | -0.003 |
Token | friends |
Token ID | 2460 |
Feature activation | -5.31e-02 |
Pos logit contributions | |
oppy | +0.061 |
opsy | +0.060 |
chy | +0.057 |
Remy | +0.056 |
rig | +0.054 |
Neg logit contributions | |
digits | -0.021 |
sale | -0.019 |
letters | -0.019 |
customer | -0.018 |
number | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | +3.87e-02 |
Pos logit contributions | |
folk | +0.075 |
digits | +0.070 |
sale | +0.070 |
customer | +0.062 |
cheering | +0.062 |
Neg logit contributions | |
opsy | -0.120 |
oppy | -0.118 |
Remy | -0.107 |
chy | -0.105 |
bur | -0.104 |
Token | a |
Token ID | 257 |
Feature activation | +1.39e-02 |
Pos logit contributions | |
folk | +0.198 |
digits | +0.195 |
sale | +0.191 |
letters | +0.185 |
customer | +0.180 |
Neg logit contributions | |
opsy | -0.278 |
oppy | -0.272 |
chy | -0.251 |
rig | -0.249 |
Remy | -0.238 |
Token | past |
Token ID | 1613 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
opsy | +0.025 |
oppy | +0.025 |
Remy | +0.023 |
chy | +0.023 |
bur | +0.021 |
Neg logit contributions | |
digits | -0.023 |
sale | -0.023 |
customer | -0.022 |
folk | -0.022 |
letters | -0.022 |
Token | el |
Token ID | 417 |
Feature activation | +6.17e-02 |
Pos logit contributions | |
folk | +0.192 |
digits | +0.188 |
sale | +0.179 |
letters | +0.173 |
customer | +0.169 |
Neg logit contributions | |
opsy | -0.259 |
oppy | -0.258 |
chy | -0.237 |
Remy | -0.231 |
rig | -0.230 |
Token | for |
Token ID | 329 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
oppy | +0.122 |
opsy | +0.122 |
Remy | +0.112 |
bur | +0.108 |
chy | +0.107 |
Neg logit contributions | |
folk | -0.097 |
digits | -0.096 |
sale | -0.092 |
letters | -0.088 |
customer | -0.087 |
Token | Fl |
Token ID | 1610 |
Feature activation | +3.92e-01 |
Pos logit contributions | |
opsy | +0.232 |
oppy | +0.229 |
chy | +0.209 |
rig | +0.203 |
Remy | +0.200 |
Neg logit contributions | |
folk | -0.215 |
digits | -0.210 |
sale | -0.208 |
customer | -0.192 |
letters | -0.190 |
Token | oppy |
Token ID | 26696 |
Feature activation | +5.63e-01 |
Pos logit contributions | |
rig | +0.331 |
chy | +0.315 |
Remy | +0.309 |
refreshing | +0.273 |
Roc | +0.217 |
Neg logit contributions | |
digits | -0.831 |
sale | -0.827 |
customer | -0.823 |
gu | -0.782 |
acher | -0.780 |
Token | to |
Token ID | 284 |
Feature activation | +5.28e-03 |
Pos logit contributions | |
opsy | +1.314 |
oppy | +1.264 |
Remy | +1.164 |
uffy | +1.122 |
rig | +1.117 |
Neg logit contributions | |
sale | -0.730 |
folk | -0.709 |
digits | -0.691 |
letters | -0.652 |
customer | -0.651 |
Token | eat |
Token ID | 4483 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
opsy | +0.011 |
oppy | +0.011 |
uffy | +0.010 |
chy | +0.010 |
rig | +0.010 |
Neg logit contributions | |
sale | -0.008 |
digits | -0.008 |
folk | -0.007 |
customer | -0.007 |
letters | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +3.28e-03 |
Pos logit contributions | |
oppy | +0.263 |
opsy | +0.260 |
uffy | +0.235 |
chy | +0.232 |
rig | +0.232 |
Neg logit contributions | |
sale | -0.131 |
digits | -0.116 |
folk | -0.115 |
letters | -0.112 |
customer | -0.111 |
Token | The |
Token ID | 383 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
opsy | +0.008 |
oppy | +0.008 |
chy | +0.007 |
rig | +0.007 |
bur | +0.007 |
Neg logit contributions | |
digits | -0.004 |
folk | -0.003 |
sale | -0.003 |
customer | -0.003 |
Amen | -0.003 |
Token | past |
Token ID | 1613 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
oppy | +0.162 |
opsy | +0.160 |
Remy | +0.141 |
rig | +0.137 |
uffy | +0.133 |
Neg logit contributions | |
folk | -0.209 |
digits | -0.205 |
sale | -0.203 |
customer | -0.199 |
letters | -0.197 |
Token | |
Token ID | 220 |
Feature activation | -2.26e-01 |
Pos logit contributions | |
opsy | +0.187 |
oppy | +0.186 |
Remy | +0.173 |
bur | +0.166 |
chy | +0.165 |
Neg logit contributions | |
folk | -0.146 |
digits | -0.143 |
sale | -0.132 |
customer | -0.124 |
letters | -0.123 |
Token |
|
Token ID | 198 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
folk | +0.308 |
digits | +0.301 |
sale | +0.282 |
customer | +0.252 |
letters | +0.252 |
Neg logit contributions | |
opsy | -0.510 |
oppy | -0.494 |
bur | -0.466 |
chy | -0.462 |
Remy | -0.457 |
Token |
|
Token ID | 198 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
opsy | +0.032 |
oppy | +0.032 |
Remy | +0.029 |
chy | +0.028 |
bur | +0.028 |
Neg logit contributions | |
folk | -0.027 |
digits | -0.026 |
sale | -0.024 |
customer | -0.023 |
letters | -0.023 |
Token | The |
Token ID | 464 |
Feature activation | +9.07e-02 |
Pos logit contributions | |
opsy | +0.026 |
oppy | +0.026 |
Remy | +0.024 |
chy | +0.023 |
bur | +0.023 |
Neg logit contributions | |
folk | -0.027 |
digits | -0.026 |
sale | -0.025 |
letters | -0.023 |
customer | -0.023 |
Token | towns |
Token ID | 11684 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
opsy | +0.189 |
oppy | +0.187 |
chy | +0.171 |
Remy | +0.171 |
bur | +0.168 |
Neg logit contributions | |
folk | -0.141 |
digits | -0.136 |
sale | -0.125 |
Amen | -0.116 |
bench | -0.114 |
Token | folk |
Token ID | 19956 |
Feature activation | -4.58e-01 |
Pos logit contributions | |
digits | +0.175 |
Amen | +0.173 |
epad | +0.156 |
sale | +0.141 |
Gu | +0.140 |
Neg logit contributions | |
oppy | -0.500 |
opsy | -0.497 |
chy | -0.474 |
Remy | -0.472 |
bur | -0.450 |
Token | searched |
Token ID | 16499 |
Feature activation | -5.82e-02 |
Pos logit contributions | |
digits | +0.742 |
folk | +0.717 |
sale | +0.677 |
customer | +0.651 |
epad | +0.638 |
Neg logit contributions | |
opsy | -0.897 |
oppy | -0.883 |
Remy | -0.814 |
bur | -0.812 |
chy | -0.809 |
Token | high |
Token ID | 1029 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
folk | +0.108 |
digits | +0.105 |
sale | +0.099 |
customer | +0.093 |
letters | +0.092 |
Neg logit contributions | |
opsy | -0.105 |
oppy | -0.103 |
Remy | -0.095 |
chy | -0.092 |
bur | -0.092 |
Token | and |
Token ID | 290 |
Feature activation | +2.33e-02 |
Pos logit contributions | |
opsy | +0.241 |
oppy | +0.239 |
Remy | +0.219 |
chy | +0.211 |
bur | +0.210 |
Neg logit contributions | |
folk | -0.195 |
digits | -0.189 |
sale | -0.179 |
letters | -0.166 |
customer | -0.165 |
Token | low |
Token ID | 1877 |
Feature activation | -2.98e-02 |
Pos logit contributions | |
opsy | +0.045 |
oppy | +0.044 |
Remy | +0.042 |
chy | +0.041 |
bur | +0.040 |
Neg logit contributions | |
folk | -0.039 |
digits | -0.039 |
sale | -0.036 |
customer | -0.034 |
letters | -0.033 |
Token | , |
Token ID | 11 |
Feature activation | -4.52e-02 |
Pos logit contributions | |
folk | +0.047 |
digits | +0.047 |
sale | +0.043 |
customer | +0.041 |
Amen | +0.040 |
Neg logit contributions | |
opsy | -0.060 |
oppy | -0.060 |
Remy | -0.056 |
chy | -0.055 |
bur | -0.054 |
Token | . |
Token ID | 13 |
Feature activation | -8.06e-03 |
Pos logit contributions | |
opsy | +0.258 |
oppy | +0.256 |
chy | +0.241 |
Remy | +0.237 |
bur | +0.237 |
Neg logit contributions | |
digits | -0.266 |
folk | -0.261 |
sale | -0.237 |
customer | -0.228 |
Amen | -0.227 |
Token | He |
Token ID | 679 |
Feature activation | -2.72e-02 |
Pos logit contributions | |
folk | +0.014 |
digits | +0.014 |
sale | +0.013 |
epad | +0.012 |
Amen | +0.012 |
Neg logit contributions | |
oppy | -0.015 |
opsy | -0.015 |
Remy | -0.014 |
bur | -0.013 |
chy | -0.013 |
Token | asked |
Token ID | 1965 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
folk | +0.041 |
digits | +0.039 |
sale | +0.036 |
customer | +0.034 |
epad | +0.033 |
Neg logit contributions | |
opsy | -0.059 |
oppy | -0.058 |
Remy | -0.054 |
bur | -0.053 |
chy | -0.053 |
Token | the |
Token ID | 262 |
Feature activation | +2.98e-02 |
Pos logit contributions | |
digits | +0.449 |
folk | +0.447 |
bench | +0.402 |
Amen | +0.402 |
epad | +0.387 |
Neg logit contributions | |
Remy | -0.449 |
oppy | -0.434 |
opsy | -0.428 |
bur | -0.389 |
chy | -0.387 |
Token | towns |
Token ID | 11684 |
Feature activation | -1.99e-01 |
Pos logit contributions | |
opsy | +0.059 |
oppy | +0.057 |
bur | +0.055 |
Remy | +0.054 |
chy | +0.054 |
Neg logit contributions | |
folk | -0.044 |
digits | -0.043 |
sale | -0.039 |
Amen | -0.039 |
bench | -0.039 |
Token | folk |
Token ID | 19956 |
Feature activation | -4.52e-01 |
Pos logit contributions | |
digits | +0.270 |
Amen | +0.253 |
Gu | +0.244 |
bench | +0.238 |
sale | +0.236 |
Neg logit contributions | |
oppy | -0.376 |
Remy | -0.375 |
opsy | -0.370 |
chy | -0.357 |
rig | -0.343 |
Token | if |
Token ID | 611 |
Feature activation | +2.09e-02 |
Pos logit contributions | |
folk | +0.802 |
digits | +0.796 |
sale | +0.727 |
bench | +0.687 |
Amen | +0.686 |
Neg logit contributions | |
opsy | -0.821 |
oppy | -0.818 |
Remy | -0.767 |
chy | -0.733 |
bur | -0.720 |
Token | they |
Token ID | 484 |
Feature activation | +3.16e-02 |
Pos logit contributions | |
Remy | +0.030 |
oppy | +0.029 |
opsy | +0.028 |
bur | +0.025 |
chy | +0.025 |
Neg logit contributions | |
digits | -0.044 |
folk | -0.042 |
channel | -0.040 |
bench | -0.039 |
Amen | -0.039 |
Token | wanted |
Token ID | 2227 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
opsy | +0.060 |
oppy | +0.059 |
Remy | +0.054 |
bur | +0.051 |
chy | +0.051 |
Neg logit contributions | |
folk | -0.057 |
digits | -0.054 |
sale | -0.051 |
letters | -0.048 |
customer | -0.048 |
Token | him |
Token ID | 683 |
Feature activation | +1.14e-03 |
Pos logit contributions | |
folk | +0.219 |
digits | +0.211 |
Amen | +0.193 |
bench | +0.189 |
sale | +0.186 |
Neg logit contributions | |
oppy | -0.183 |
opsy | -0.181 |
Remy | -0.179 |
bur | -0.169 |
chy | -0.168 |
Token | to |
Token ID | 284 |
Feature activation | -6.47e-03 |
Pos logit contributions | |
opsy | +0.003 |
oppy | +0.003 |
Remy | +0.002 |
bur | +0.002 |
chy | +0.002 |
Neg logit contributions | |
folk | -0.002 |
digits | -0.002 |
sale | -0.001 |
letters | -0.001 |
customer | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | +3.23e-02 |
Pos logit contributions | |
folk | +0.118 |
digits | +0.115 |
sale | +0.107 |
customer | +0.101 |
Amen | +0.099 |
Neg logit contributions | |
opsy | -0.138 |
oppy | -0.137 |
Remy | -0.127 |
chy | -0.124 |
bur | -0.123 |
Token | days |
Token ID | 1528 |
Feature activation | -7.11e-02 |
Pos logit contributions | |
oppy | +0.059 |
opsy | +0.059 |
Remy | +0.058 |
bur | +0.057 |
chy | +0.055 |
Neg logit contributions | |
folk | -0.054 |
digits | -0.053 |
Amen | -0.050 |
sale | -0.048 |
customer | -0.047 |
Token | and |
Token ID | 290 |
Feature activation | +3.26e-02 |
Pos logit contributions | |
folk | +0.115 |
digits | +0.114 |
sale | +0.104 |
customer | +0.099 |
Amen | +0.098 |
Neg logit contributions | |
opsy | -0.141 |
oppy | -0.140 |
Remy | -0.131 |
chy | -0.126 |
bur | -0.126 |
Token | the |
Token ID | 262 |
Feature activation | +3.63e-02 |
Pos logit contributions | |
oppy | +0.058 |
Remy | +0.058 |
opsy | +0.058 |
bur | +0.056 |
chy | +0.055 |
Neg logit contributions | |
folk | -0.053 |
digits | -0.052 |
Amen | -0.051 |
sale | -0.048 |
customer | -0.046 |
Token | towns |
Token ID | 11684 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
opsy | +0.066 |
chy | +0.065 |
oppy | +0.065 |
bur | +0.065 |
Remy | +0.064 |
Neg logit contributions | |
folk | -0.052 |
Amen | -0.051 |
digits | -0.051 |
sale | -0.048 |
bench | -0.047 |
Token | folk |
Token ID | 19956 |
Feature activation | -4.50e-01 |
Pos logit contributions | |
Amen | +0.168 |
digits | +0.153 |
Gu | +0.135 |
epad | +0.132 |
customer | +0.126 |
Neg logit contributions | |
chy | -0.456 |
opsy | -0.455 |
oppy | -0.452 |
Remy | -0.448 |
bur | -0.428 |
Token | looked |
Token ID | 3114 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
digits | +0.760 |
folk | +0.730 |
sale | +0.720 |
Amen | +0.687 |
customer | +0.684 |
Neg logit contributions | |
opsy | -0.811 |
oppy | -0.799 |
chy | -0.774 |
bur | -0.734 |
Remy | -0.732 |
Token | on |
Token ID | 319 |
Feature activation | -5.79e-02 |
Pos logit contributions | |
folk | +0.215 |
digits | +0.208 |
sale | +0.197 |
Amen | +0.188 |
customer | +0.187 |
Neg logit contributions | |
oppy | -0.180 |
opsy | -0.178 |
chy | -0.168 |
Remy | -0.165 |
bur | -0.164 |
Token | . |
Token ID | 13 |
Feature activation | -8.57e-04 |
Pos logit contributions | |
folk | +0.105 |
digits | +0.101 |
Amen | +0.091 |
sale | +0.091 |
customer | +0.089 |
Neg logit contributions | |
opsy | -0.102 |
oppy | -0.101 |
Remy | -0.096 |
bur | -0.095 |
chy | -0.094 |
Token |
|
Token ID | 198 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
digits | +0.001 |
folk | +0.001 |
sale | +0.001 |
epad | +0.001 |
customer | +0.001 |
Neg logit contributions | |
oppy | -0.002 |
Remy | -0.002 |
opsy | -0.002 |
bur | -0.002 |
chy | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
opsy | +0.046 |
oppy | +0.046 |
Remy | +0.043 |
bur | +0.040 |
chy | +0.039 |
Neg logit contributions | |
folk | -0.048 |
digits | -0.047 |
sale | -0.044 |
epad | -0.042 |
letters | -0.042 |
Token | thing |
Token ID | 1517 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
opsy | +0.004 |
oppy | +0.004 |
bur | +0.004 |
Remy | +0.004 |
chy | +0.003 |
Neg logit contributions | |
folk | -0.004 |
digits | -0.004 |
Amen | -0.004 |
sale | -0.004 |
bench | -0.003 |
Token | ever |
Token ID | 1683 |
Feature activation | -2.07e-02 |
Pos logit contributions | |
folk | +0.232 |
digits | +0.223 |
sale | +0.210 |
customer | +0.196 |
letters | +0.195 |
Neg logit contributions | |
opsy | -0.293 |
oppy | -0.290 |
Remy | -0.265 |
bur | -0.259 |
chy | -0.256 |
Token | . |
Token ID | 13 |
Feature activation | +2.64e-05 |
Pos logit contributions | |
folk | +0.035 |
digits | +0.035 |
sale | +0.032 |
customer | +0.030 |
letters | +0.030 |
Neg logit contributions | |
opsy | -0.040 |
oppy | -0.040 |
Remy | -0.037 |
chy | -0.036 |
bur | -0.036 |
Token | The |
Token ID | 383 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
oppy | +0.000 |
opsy | +0.000 |
Remy | +0.000 |
bur | +0.000 |
chy | +0.000 |
Neg logit contributions | |
digits | -0.000 |
folk | -0.000 |
sale | -0.000 |
epad | -0.000 |
customer | -0.000 |
Token | towns |
Token ID | 11684 |
Feature activation | -1.94e-01 |
Pos logit contributions | |
opsy | +0.168 |
oppy | +0.165 |
Remy | +0.161 |
chy | +0.158 |
bur | +0.158 |
Neg logit contributions | |
digits | -0.151 |
Amen | -0.150 |
folk | -0.148 |
sale | -0.144 |
bench | -0.143 |
Token | folk |
Token ID | 19956 |
Feature activation | -4.48e-01 |
Pos logit contributions | |
Amen | +0.133 |
digits | +0.119 |
Gu | +0.113 |
epad | +0.111 |
sale | +0.096 |
Neg logit contributions | |
opsy | -0.476 |
oppy | -0.464 |
Remy | -0.462 |
chy | -0.456 |
bur | -0.439 |
Token | cl |
Token ID | 537 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
digits | +0.786 |
sale | +0.738 |
folk | +0.727 |
customer | +0.701 |
bench | +0.698 |
Neg logit contributions | |
opsy | -0.794 |
oppy | -0.762 |
chy | -0.708 |
bur | -0.705 |
Remy | -0.700 |
Token | apped |
Token ID | 6320 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
digits | +0.206 |
folk | +0.200 |
sale | +0.197 |
customer | +0.184 |
bench | +0.176 |
Neg logit contributions | |
oppy | -0.358 |
opsy | -0.355 |
Remy | -0.317 |
bur | -0.315 |
uffy | -0.306 |
Token | and |
Token ID | 290 |
Feature activation | +3.49e-02 |
Pos logit contributions | |
folk | +0.249 |
digits | +0.247 |
sale | +0.228 |
customer | +0.217 |
jury | +0.212 |
Neg logit contributions | |
opsy | -0.248 |
oppy | -0.246 |
Remy | -0.230 |
chy | -0.224 |
bur | -0.223 |
Token | cheered |
Token ID | 34767 |
Feature activation | -2.45e-01 |
Pos logit contributions | |
opsy | +0.078 |
oppy | +0.078 |
Remy | +0.077 |
bur | +0.074 |
chy | +0.073 |
Neg logit contributions | |
digits | -0.044 |
folk | -0.042 |
sale | -0.041 |
customer | -0.039 |
Amen | -0.037 |
Token | because |
Token ID | 780 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
digits | +0.459 |
folk | +0.454 |
sale | +0.413 |
customer | +0.402 |
bench | +0.400 |
Neg logit contributions | |
opsy | -0.410 |
oppy | -0.409 |
Remy | -0.388 |
chy | -0.380 |
bur | -0.361 |
Token | its |
Token ID | 663 |
Feature activation | +8.72e-02 |
Pos logit contributions | |
opsy | +0.007 |
oppy | +0.007 |
Remy | +0.007 |
chy | +0.006 |
bur | +0.006 |
Neg logit contributions | |
folk | -0.005 |
digits | -0.004 |
sale | -0.004 |
customer | -0.004 |
letters | -0.004 |
Token | bouncing |
Token ID | 30713 |
Feature activation | +8.73e-02 |
Pos logit contributions | |
oppy | +0.191 |
opsy | +0.189 |
Remy | +0.175 |
uffy | +0.166 |
chy | +0.164 |
Neg logit contributions | |
sale | -0.112 |
digits | -0.111 |
number | -0.111 |
cheering | -0.106 |
folk | -0.106 |
Token | , |
Token ID | 11 |
Feature activation | -3.32e-02 |
Pos logit contributions | |
opsy | +0.213 |
oppy | +0.209 |
Remy | +0.190 |
chy | +0.189 |
bur | +0.182 |
Neg logit contributions | |
folk | -0.109 |
sale | -0.100 |
digits | -0.097 |
cheering | -0.087 |
letters | -0.087 |
Token | the |
Token ID | 262 |
Feature activation | +3.46e-02 |
Pos logit contributions | |
digits | +0.056 |
folk | +0.055 |
sale | +0.051 |
customer | +0.048 |
bench | +0.048 |
Neg logit contributions | |
opsy | -0.064 |
oppy | -0.063 |
Remy | -0.060 |
bur | -0.057 |
chy | -0.056 |
Token | towns |
Token ID | 11684 |
Feature activation | -1.91e-01 |
Pos logit contributions | |
oppy | +0.066 |
opsy | +0.065 |
Remy | +0.060 |
chy | +0.058 |
bur | +0.057 |
Neg logit contributions | |
folk | -0.062 |
digits | -0.060 |
sale | -0.056 |
customer | -0.054 |
letters | -0.054 |
Token | folk |
Token ID | 19956 |
Feature activation | -4.44e-01 |
Pos logit contributions | |
digits | +0.291 |
Amen | +0.269 |
sale | +0.261 |
epad | +0.253 |
customer | +0.252 |
Neg logit contributions | |
opsy | -0.372 |
oppy | -0.362 |
chy | -0.344 |
Remy | -0.342 |
bur | -0.331 |
Token | thanked |
Token ID | 26280 |
Feature activation | -4.40e-02 |
Pos logit contributions | |
folk | +0.666 |
digits | +0.658 |
sale | +0.610 |
customer | +0.566 |
letters | +0.557 |
Neg logit contributions | |
opsy | -0.948 |
oppy | -0.935 |
Remy | -0.867 |
chy | -0.849 |
bur | -0.849 |
Token | the |
Token ID | 262 |
Feature activation | +3.49e-02 |
Pos logit contributions | |
digits | +0.078 |
folk | +0.072 |
bench | +0.071 |
customer | +0.066 |
salesman | +0.065 |
Neg logit contributions | |
opsy | -0.079 |
Remy | -0.079 |
oppy | -0.076 |
bur | -0.068 |
Dee | -0.068 |
Token | rope |
Token ID | 17182 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
opsy | +0.070 |
oppy | +0.068 |
Remy | +0.064 |
chy | +0.062 |
bur | +0.061 |
Neg logit contributions | |
folk | -0.057 |
digits | -0.056 |
sale | -0.052 |
Amen | -0.048 |
epad | -0.048 |
Token | for |
Token ID | 329 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
opsy | +0.246 |
oppy | +0.243 |
Remy | +0.220 |
chy | +0.213 |
bur | +0.211 |
Neg logit contributions | |
folk | -0.204 |
digits | -0.194 |
sale | -0.190 |
letters | -0.175 |
customer | -0.174 |
Token | its |
Token ID | 663 |
Feature activation | +8.83e-02 |
Pos logit contributions | |
opsy | +0.246 |
Remy | +0.242 |
oppy | +0.238 |
bur | +0.227 |
chy | +0.217 |
Neg logit contributions | |
folk | -0.193 |
digits | -0.191 |
sale | -0.161 |
prices | -0.158 |
customer | -0.157 |
Token | was |
Token ID | 373 |
Feature activation | +4.65e-02 |
Pos logit contributions | |
folk | +0.194 |
digits | +0.184 |
sale | +0.178 |
customer | +0.162 |
letters | +0.159 |
Neg logit contributions | |
opsy | -0.286 |
oppy | -0.286 |
Remy | -0.263 |
chy | -0.258 |
bur | -0.245 |
Token | very |
Token ID | 845 |
Feature activation | -4.39e-02 |
Pos logit contributions | |
opsy | +0.094 |
oppy | +0.091 |
Remy | +0.086 |
chy | +0.085 |
bur | +0.084 |
Neg logit contributions | |
folk | -0.076 |
digits | -0.075 |
sale | -0.068 |
jury | -0.064 |
letters | -0.064 |
Token | happy |
Token ID | 3772 |
Feature activation | -7.84e-02 |
Pos logit contributions | |
folk | +0.069 |
digits | +0.067 |
sale | +0.062 |
letters | +0.057 |
customer | +0.057 |
Neg logit contributions | |
opsy | -0.093 |
oppy | -0.092 |
Remy | -0.085 |
chy | -0.083 |
bur | -0.083 |
Token | with |
Token ID | 351 |
Feature activation | -9.95e-02 |
Pos logit contributions | |
folk | +0.138 |
digits | +0.135 |
sale | +0.124 |
customer | +0.117 |
letters | +0.116 |
Neg logit contributions | |
opsy | -0.149 |
oppy | -0.147 |
Remy | -0.137 |
bur | -0.133 |
chy | -0.131 |
Token | his |
Token ID | 465 |
Feature activation | -5.11e-02 |
Pos logit contributions | |
folk | +0.141 |
digits | +0.137 |
sale | +0.127 |
customer | +0.117 |
letters | +0.117 |
Neg logit contributions | |
opsy | -0.224 |
oppy | -0.222 |
chy | -0.202 |
Remy | -0.201 |
bur | -0.199 |
Token | sale |
Token ID | 5466 |
Feature activation | -4.25e-01 |
Pos logit contributions | |
sale | +0.074 |
digits | +0.072 |
letters | +0.071 |
folk | +0.068 |
prices | +0.066 |
Neg logit contributions | |
opsy | -0.105 |
oppy | -0.100 |
Remy | -0.100 |
chy | -0.094 |
rig | -0.089 |
Token | . |
Token ID | 13 |
Feature activation | -6.30e-03 |
Pos logit contributions | |
folk | +0.727 |
digits | +0.726 |
sale | +0.658 |
Amen | +0.637 |
customer | +0.625 |
Neg logit contributions | |
opsy | -0.796 |
oppy | -0.792 |
Remy | -0.729 |
chy | -0.720 |
bur | -0.707 |
Token | He |
Token ID | 679 |
Feature activation | -2.64e-02 |
Pos logit contributions | |
folk | +0.009 |
digits | +0.009 |
sale | +0.008 |
customer | +0.007 |
letters | +0.007 |
Neg logit contributions | |
opsy | -0.014 |
oppy | -0.014 |
chy | -0.013 |
Remy | -0.013 |
bur | -0.013 |
Token | had |
Token ID | 550 |
Feature activation | -2.61e-02 |
Pos logit contributions | |
folk | +0.037 |
digits | +0.036 |
sale | +0.034 |
customer | +0.031 |
letters | +0.031 |
Neg logit contributions | |
opsy | -0.058 |
oppy | -0.058 |
Remy | -0.053 |
chy | -0.052 |
uffy | -0.050 |
Token | enough |
Token ID | 1576 |
Feature activation | -9.50e-02 |
Pos logit contributions | |
folk | +0.044 |
digits | +0.042 |
Amen | +0.040 |
jury | +0.036 |
sale | +0.036 |
Neg logit contributions | |
opsy | -0.048 |
Remy | -0.047 |
oppy | -0.047 |
bur | -0.047 |
chy | -0.045 |
Token | money |
Token ID | 1637 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
folk | +0.148 |
digits | +0.142 |
sale | +0.126 |
Amen | +0.124 |
epad | +0.120 |
Neg logit contributions | |
opsy | -0.199 |
oppy | -0.198 |
Remy | -0.184 |
bur | -0.183 |
chy | -0.178 |
Token | his |
Token ID | 465 |
Feature activation | -4.75e-02 |
Pos logit contributions | |
opsy | +0.298 |
oppy | +0.297 |
Remy | +0.287 |
chy | +0.275 |
bur | +0.274 |
Neg logit contributions | |
digits | -0.232 |
folk | -0.230 |
sale | -0.202 |
Amen | -0.197 |
bench | -0.196 |
Token | grandma |
Token ID | 49890 |
Feature activation | -5.03e-02 |
Pos logit contributions | |
folk | +0.085 |
digits | +0.083 |
sale | +0.076 |
customer | +0.073 |
epad | +0.072 |
Neg logit contributions | |
opsy | -0.088 |
oppy | -0.087 |
Remy | -0.080 |
bur | -0.079 |
chy | -0.078 |
Token | got |
Token ID | 1392 |
Feature activation | -4.10e-02 |
Pos logit contributions | |
folk | +0.090 |
digits | +0.089 |
sale | +0.083 |
customer | +0.078 |
Amen | +0.077 |
Neg logit contributions | |
opsy | -0.092 |
oppy | -0.092 |
chy | -0.084 |
Remy | -0.083 |
bur | -0.080 |
Token | it |
Token ID | 340 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
folk | +0.075 |
digits | +0.074 |
Amen | +0.066 |
sale | +0.066 |
customer | +0.064 |
Neg logit contributions | |
oppy | -0.073 |
opsy | -0.073 |
chy | -0.068 |
Remy | -0.066 |
bur | -0.065 |
Token | on |
Token ID | 319 |
Feature activation | -6.18e-02 |
Pos logit contributions | |
folk | +0.188 |
digits | +0.179 |
sale | +0.179 |
letters | +0.160 |
epad | +0.160 |
Neg logit contributions | |
opsy | -0.283 |
oppy | -0.277 |
Remy | -0.260 |
bur | -0.246 |
chy | -0.236 |
Token | sale |
Token ID | 5466 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
folk | +0.111 |
digits | +0.109 |
Amen | +0.093 |
sale | +0.092 |
jury | +0.092 |
Neg logit contributions | |
opsy | -0.111 |
oppy | -0.109 |
Remy | -0.105 |
bur | -0.102 |
chy | -0.101 |
Token | ! |
Token ID | 0 |
Feature activation | +9.97e-02 |
Pos logit contributions | |
digits | +0.857 |
folk | +0.855 |
Amen | +0.823 |
jury | +0.754 |
customer | +0.745 |
Neg logit contributions | |
oppy | -0.612 |
chy | -0.599 |
opsy | -0.596 |
bur | -0.543 |
Remy | -0.539 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
opsy | +0.187 |
oppy | +0.187 |
Remy | +0.178 |
bur | +0.166 |
chy | +0.162 |
Neg logit contributions | |
folk | -0.171 |
digits | -0.170 |
sale | -0.156 |
epad | -0.149 |
letters | -0.149 |
Token | Once |
Token ID | 7454 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
opsy | +0.340 |
oppy | +0.337 |
Remy | +0.311 |
chy | +0.300 |
bur | +0.300 |
Neg logit contributions | |
folk | -0.304 |
digits | -0.297 |
sale | -0.278 |
customer | -0.261 |
letters | -0.258 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
folk | +0.197 |
digits | +0.192 |
sale | +0.182 |
customer | +0.174 |
cheering | +0.168 |
Neg logit contributions | |
opsy | -0.178 |
oppy | -0.168 |
Remy | -0.160 |
bur | -0.160 |
chy | -0.156 |
Token | a |
Token ID | 257 |
Feature activation | -2.15e-03 |
Pos logit contributions | |
opsy | +0.347 |
oppy | +0.344 |
Remy | +0.316 |
bur | +0.309 |
chy | +0.308 |
Neg logit contributions | |
folk | -0.240 |
digits | -0.236 |
sale | -0.220 |
customer | -0.205 |
letters | -0.202 |
Token | were |
Token ID | 547 |
Feature activation | -3.96e-02 |
Pos logit contributions | |
folk | +0.030 |
digits | +0.028 |
sale | +0.027 |
letters | +0.025 |
customer | +0.024 |
Neg logit contributions | |
opsy | -0.046 |
oppy | -0.045 |
Remy | -0.042 |
chy | -0.041 |
bur | -0.041 |
Token | many |
Token ID | 867 |
Feature activation | -6.26e-02 |
Pos logit contributions | |
folk | +0.062 |
digits | +0.060 |
sale | +0.054 |
Amen | +0.053 |
bench | +0.051 |
Neg logit contributions | |
opsy | -0.079 |
oppy | -0.078 |
Remy | -0.076 |
chy | -0.074 |
bur | -0.073 |
Token | beautiful |
Token ID | 4950 |
Feature activation | +4.80e-02 |
Pos logit contributions | |
folk | +0.094 |
digits | +0.091 |
sale | +0.085 |
Amen | +0.081 |
customer | +0.079 |
Neg logit contributions | |
opsy | -0.129 |
oppy | -0.128 |
bur | -0.121 |
Remy | -0.121 |
chy | -0.118 |
Token | apples |
Token ID | 22514 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
opsy | +0.101 |
oppy | +0.100 |
bur | +0.096 |
Remy | +0.094 |
chy | +0.091 |
Neg logit contributions | |
folk | -0.070 |
digits | -0.068 |
sale | -0.064 |
customer | -0.061 |
bench | -0.061 |
Token | for |
Token ID | 329 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
folk | +0.025 |
digits | +0.024 |
sale | +0.023 |
customer | +0.021 |
letters | +0.021 |
Neg logit contributions | |
opsy | -0.031 |
oppy | -0.031 |
Remy | -0.029 |
bur | -0.028 |
chy | -0.028 |
Token | sale |
Token ID | 5466 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
opsy | +0.212 |
oppy | +0.211 |
Remy | +0.208 |
bur | +0.192 |
chy | +0.183 |
Neg logit contributions | |
folk | -0.181 |
digits | -0.181 |
bench | -0.160 |
Amen | -0.159 |
epad | -0.157 |
Token | . |
Token ID | 13 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
digits | +0.833 |
folk | +0.824 |
Amen | +0.821 |
Gu | +0.732 |
customer | +0.730 |
Neg logit contributions | |
oppy | -0.637 |
opsy | -0.619 |
chy | -0.606 |
Remy | -0.574 |
bur | -0.564 |
Token | The |
Token ID | 383 |
Feature activation | +8.79e-02 |
Pos logit contributions | |
folk | +0.022 |
digits | +0.022 |
sale | +0.020 |
epad | +0.019 |
letters | +0.019 |
Neg logit contributions | |
oppy | -0.025 |
opsy | -0.025 |
Remy | -0.024 |
bur | -0.022 |
chy | -0.021 |
Token | pirate |
Token ID | 25868 |
Feature activation | -3.53e-02 |
Pos logit contributions | |
opsy | +0.178 |
oppy | +0.175 |
Remy | +0.163 |
bur | +0.159 |
chy | +0.159 |
Neg logit contributions | |
folk | -0.142 |
digits | -0.139 |
sale | -0.127 |
epad | -0.118 |
customer | -0.118 |
Token | smiled |
Token ID | 13541 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
folk | +0.060 |
digits | +0.059 |
sale | +0.054 |
Amen | +0.052 |
epad | +0.052 |
Neg logit contributions | |
opsy | -0.067 |
oppy | -0.067 |
Remy | -0.062 |
chy | -0.060 |
bur | -0.060 |
Token | and |
Token ID | 290 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
folk | +0.230 |
digits | +0.224 |
sale | +0.210 |
customer | +0.197 |
letters | +0.195 |
Neg logit contributions | |
opsy | -0.265 |
oppy | -0.262 |
Remy | -0.241 |
chy | -0.234 |
bur | -0.233 |
Token | that |
Token ID | 326 |
Feature activation | +8.10e-02 |
Pos logit contributions | |
folk | +0.042 |
digits | +0.041 |
sale | +0.038 |
customer | +0.036 |
bench | +0.036 |
Neg logit contributions | |
opsy | -0.047 |
oppy | -0.046 |
Remy | -0.043 |
bur | -0.041 |
chy | -0.041 |
Token | jewel |
Token ID | 16927 |
Feature activation | +7.66e-02 |
Pos logit contributions | |
opsy | +0.146 |
oppy | +0.145 |
Remy | +0.134 |
chy | +0.131 |
bur | +0.129 |
Neg logit contributions | |
folk | -0.148 |
digits | -0.145 |
sale | -0.136 |
customer | -0.127 |
Amen | -0.125 |
Token | is |
Token ID | 318 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
opsy | +0.163 |
oppy | +0.156 |
Remy | +0.146 |
bur | +0.141 |
chy | +0.133 |
Neg logit contributions | |
folk | -0.115 |
digits | -0.109 |
sale | -0.107 |
letters | -0.103 |
number | -0.098 |
Token | not |
Token ID | 407 |
Feature activation | +1.45e-02 |
Pos logit contributions | |
opsy | +0.077 |
oppy | +0.076 |
Remy | +0.070 |
bur | +0.067 |
chy | +0.067 |
Neg logit contributions | |
folk | -0.060 |
digits | -0.059 |
sale | -0.055 |
customer | -0.051 |
letters | -0.051 |
Token | for |
Token ID | 329 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
opsy | +0.028 |
oppy | +0.028 |
Remy | +0.026 |
chy | +0.025 |
bur | +0.025 |
Neg logit contributions | |
folk | -0.025 |
digits | -0.024 |
sale | -0.022 |
customer | -0.021 |
Amen | -0.021 |
Token | sale |
Token ID | 5466 |
Feature activation | -4.22e-01 |
Pos logit contributions | |
oppy | +0.215 |
opsy | +0.214 |
Remy | +0.204 |
bur | +0.193 |
chy | +0.186 |
Neg logit contributions | |
folk | -0.193 |
digits | -0.187 |
Amen | -0.168 |
bench | -0.163 |
sale | -0.162 |
Token | . |
Token ID | 13 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
folk | +0.768 |
digits | +0.760 |
Amen | +0.689 |
sale | +0.688 |
customer | +0.665 |
Neg logit contributions | |
oppy | -0.748 |
opsy | -0.738 |
chy | -0.684 |
Remy | -0.682 |
bur | -0.647 |
Token | It |
Token ID | 632 |
Feature activation | -8.40e-02 |
Pos logit contributions | |
folk | +0.019 |
digits | +0.018 |
sale | +0.017 |
epad | +0.016 |
bench | +0.016 |
Neg logit contributions | |
oppy | -0.021 |
opsy | -0.021 |
Remy | -0.020 |
bur | -0.019 |
chy | -0.018 |
Token | is |
Token ID | 318 |
Feature activation | +3.84e-02 |
Pos logit contributions | |
folk | +0.134 |
digits | +0.129 |
sale | +0.123 |
letters | +0.119 |
prices | +0.114 |
Neg logit contributions | |
opsy | -0.175 |
oppy | -0.170 |
Remy | -0.155 |
bur | -0.145 |
uffy | -0.139 |
Token | too |
Token ID | 1165 |
Feature activation | -9.79e-02 |
Pos logit contributions | |
opsy | +0.078 |
oppy | +0.077 |
Remy | +0.071 |
bur | +0.069 |
chy | +0.069 |
Neg logit contributions | |
folk | -0.062 |
digits | -0.061 |
sale | -0.057 |
customer | -0.053 |
letters | -0.053 |
Token | special |
Token ID | 2041 |
Feature activation | +7.31e-02 |
Pos logit contributions | |
folk | +0.150 |
digits | +0.149 |
sale | +0.144 |
letters | +0.135 |
customer | +0.132 |
Neg logit contributions | |
opsy | -0.204 |
oppy | -0.200 |
Remy | -0.186 |
bur | -0.179 |
chy | -0.175 |
Token | che |
Token ID | 2395 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
digits | +0.205 |
folk | +0.203 |
sale | +0.183 |
Amen | +0.182 |
customer | +0.176 |
Neg logit contributions | |
opsy | -0.272 |
oppy | -0.263 |
Remy | -0.250 |
chy | -0.245 |
bur | -0.237 |
Token | ap |
Token ID | 499 |
Feature activation | -2.79e-02 |
Pos logit contributions | |
folk | +0.061 |
digits | +0.056 |
epad | +0.055 |
sale | +0.055 |
cheering | +0.054 |
Neg logit contributions | |
oppy | -0.052 |
opsy | -0.051 |
Remy | -0.044 |
uffy | -0.042 |
chy | -0.038 |
Token | basketball |
Token ID | 9669 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
folk | +0.048 |
digits | +0.047 |
sale | +0.042 |
customer | +0.040 |
Amen | +0.040 |
Neg logit contributions | |
oppy | -0.054 |
opsy | -0.053 |
Remy | -0.050 |
chy | -0.048 |
bur | -0.046 |
Token | s |
Token ID | 82 |
Feature activation | +2.01e-01 |
Pos logit contributions | |
folk | +0.018 |
digits | +0.018 |
sale | +0.016 |
customer | +0.015 |
Amen | +0.015 |
Neg logit contributions | |
opsy | -0.026 |
oppy | -0.026 |
Remy | -0.024 |
chy | -0.023 |
bur | -0.023 |
Token | for |
Token ID | 329 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
opsy | +0.393 |
oppy | +0.391 |
Remy | +0.366 |
chy | +0.359 |
bur | +0.354 |
Neg logit contributions | |
folk | -0.328 |
digits | -0.321 |
sale | -0.281 |
Amen | -0.279 |
customer | -0.275 |
Token | sale |
Token ID | 5466 |
Feature activation | -4.21e-01 |
Pos logit contributions | |
oppy | +0.211 |
Remy | +0.211 |
opsy | +0.207 |
bur | +0.194 |
Lucky | +0.183 |
Neg logit contributions | |
folk | -0.194 |
digits | -0.183 |
Amen | -0.163 |
bench | -0.162 |
salesman | -0.157 |
Token | ". |
Token ID | 1911 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
folk | +0.659 |
digits | +0.653 |
Amen | +0.589 |
cheering | +0.557 |
jury | +0.533 |
Neg logit contributions | |
oppy | -0.847 |
opsy | -0.825 |
Remy | -0.790 |
chy | -0.770 |
bur | -0.734 |
Token | Tim |
Token ID | 5045 |
Feature activation | +6.67e-02 |
Pos logit contributions | |
folk | +0.214 |
digits | +0.210 |
epad | +0.203 |
winners | +0.190 |
customer | +0.190 |
Neg logit contributions | |
Remy | -0.169 |
opsy | -0.168 |
oppy | -0.166 |
bur | -0.152 |
chy | -0.143 |
Token | my |
Token ID | 1820 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
opsy | +0.165 |
oppy | +0.164 |
Remy | +0.153 |
chy | +0.150 |
bur | +0.149 |
Neg logit contributions | |
folk | -0.081 |
digits | -0.077 |
sale | -0.071 |
customer | -0.063 |
letters | -0.062 |
Token | asked |
Token ID | 1965 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
folk | +0.252 |
digits | +0.243 |
sale | +0.226 |
epad | +0.219 |
customer | +0.217 |
Neg logit contributions | |
opsy | -0.250 |
oppy | -0.246 |
Remy | -0.226 |
chy | -0.221 |
bur | -0.220 |
Token | his |
Token ID | 465 |
Feature activation | -5.28e-02 |
Pos logit contributions | |
digits | +0.561 |
folk | +0.551 |
winners | +0.492 |
epad | +0.491 |
bench | +0.491 |
Neg logit contributions | |
Remy | -0.361 |
oppy | -0.300 |
Lucky | -0.294 |
opsy | -0.289 |
Mitt | -0.284 |
Token | put |
Token ID | 1234 |
Feature activation | -5.19e-02 |
Pos logit contributions | |
folk | +0.335 |
digits | +0.325 |
sale | +0.302 |
customer | +0.279 |
letters | +0.276 |
Neg logit contributions | |
opsy | -0.468 |
oppy | -0.463 |
Remy | -0.429 |
bur | -0.418 |
chy | -0.418 |
Token | the |
Token ID | 262 |
Feature activation | +3.30e-02 |
Pos logit contributions | |
folk | +0.095 |
digits | +0.093 |
Amen | +0.087 |
sale | +0.086 |
epad | +0.085 |
Neg logit contributions | |
opsy | -0.089 |
oppy | -0.087 |
Remy | -0.085 |
Lucky | -0.079 |
bur | -0.078 |
Token | rake |
Token ID | 47885 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
opsy | +0.049 |
oppy | +0.049 |
Remy | +0.043 |
chy | +0.040 |
bur | +0.040 |
Neg logit contributions | |
digits | -0.070 |
folk | -0.070 |
sale | -0.067 |
customer | -0.064 |
letters | -0.064 |
Token | up |
Token ID | 510 |
Feature activation | +9.12e-02 |
Pos logit contributions | |
opsy | +0.386 |
oppy | +0.382 |
Remy | +0.354 |
chy | +0.336 |
bur | +0.335 |
Neg logit contributions | |
folk | -0.284 |
digits | -0.274 |
sale | -0.261 |
letters | -0.242 |
customer | -0.241 |
Token | for |
Token ID | 329 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
opsy | +0.186 |
oppy | +0.184 |
Remy | +0.171 |
chy | +0.166 |
bur | +0.166 |
Neg logit contributions | |
folk | -0.146 |
digits | -0.143 |
sale | -0.132 |
customer | -0.124 |
letters | -0.122 |
Token | sale |
Token ID | 5466 |
Feature activation | -4.20e-01 |
Pos logit contributions | |
opsy | +0.209 |
oppy | +0.205 |
Remy | +0.203 |
bur | +0.187 |
Lucky | +0.186 |
Neg logit contributions | |
folk | -0.197 |
digits | -0.193 |
Amen | -0.177 |
bench | -0.165 |
salesman | -0.161 |
Token | and |
Token ID | 290 |
Feature activation | +2.71e-02 |
Pos logit contributions | |
folk | +0.873 |
digits | +0.865 |
Amen | +0.859 |
jury | +0.776 |
president | +0.775 |
Neg logit contributions | |
oppy | -0.549 |
opsy | -0.543 |
chy | -0.543 |
bur | -0.500 |
Remy | -0.486 |
Token | waited |
Token ID | 13488 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
opsy | +0.053 |
oppy | +0.053 |
Remy | +0.050 |
bur | +0.049 |
chy | +0.048 |
Neg logit contributions | |
folk | -0.044 |
digits | -0.043 |
sale | -0.038 |
Amen | -0.037 |
customer | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | -7.01e-03 |
Pos logit contributions | |
folk | +0.272 |
digits | +0.261 |
sale | +0.235 |
jury | +0.234 |
Amen | +0.230 |
Neg logit contributions | |
opsy | -0.211 |
oppy | -0.208 |
chy | -0.191 |
bur | -0.186 |
Remy | -0.185 |
Token |
|
Token ID | 198 |
Feature activation | +1.80e-02 |
Pos logit contributions | |
folk | +0.013 |
digits | +0.013 |
Amen | +0.011 |
winners | +0.011 |
sale | +0.011 |
Neg logit contributions | |
opsy | -0.012 |
oppy | -0.012 |
Remy | -0.011 |
bur | -0.010 |
uffy | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | +1.72e-02 |
Pos logit contributions | |
opsy | +0.036 |
oppy | +0.036 |
Remy | +0.033 |
chy | +0.032 |
bur | +0.032 |
Neg logit contributions | |
folk | -0.030 |
digits | -0.029 |
sale | -0.027 |
customer | -0.025 |
letters | -0.025 |
Token | found |
Token ID | 1043 |
Feature activation | -5.87e-02 |
Pos logit contributions | |
folk | +0.167 |
Amen | +0.163 |
digits | +0.163 |
channel | +0.150 |
cheering | +0.143 |
Neg logit contributions | |
oppy | -0.185 |
opsy | -0.179 |
Remy | -0.170 |
chy | -0.159 |
bur | -0.158 |
Token | the |
Token ID | 262 |
Feature activation | +2.73e-02 |
Pos logit contributions | |
folk | +0.099 |
digits | +0.096 |
sale | +0.088 |
Amen | +0.085 |
epad | +0.084 |
Neg logit contributions | |
opsy | -0.114 |
oppy | -0.113 |
Remy | -0.107 |
bur | -0.101 |
chy | -0.101 |
Token | monster |
Token ID | 9234 |
Feature activation | +4.57e-02 |
Pos logit contributions | |
opsy | +0.053 |
oppy | +0.052 |
chy | +0.050 |
bur | +0.049 |
Remy | +0.049 |
Neg logit contributions | |
folk | -0.045 |
digits | -0.044 |
sale | -0.038 |
Amen | -0.038 |
bench | -0.037 |
Token | truck |
Token ID | 7779 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
opsy | +0.095 |
oppy | +0.094 |
Remy | +0.086 |
bur | +0.083 |
chy | +0.083 |
Neg logit contributions | |
folk | -0.072 |
digits | -0.070 |
sale | -0.067 |
customer | -0.062 |
letters | -0.061 |
Token | on |
Token ID | 319 |
Feature activation | -6.36e-02 |
Pos logit contributions | |
folk | +0.212 |
digits | +0.211 |
Amen | +0.188 |
sale | +0.186 |
customer | +0.182 |
Neg logit contributions | |
opsy | -0.228 |
oppy | -0.227 |
chy | -0.212 |
Remy | -0.211 |
bur | -0.205 |
Token | sale |
Token ID | 5466 |
Feature activation | -4.20e-01 |
Pos logit contributions | |
folk | +0.114 |
digits | +0.112 |
Amen | +0.098 |
sale | +0.096 |
epad | +0.095 |
Neg logit contributions | |
opsy | -0.114 |
oppy | -0.113 |
Remy | -0.108 |
bur | -0.104 |
chy | -0.102 |
Token | . |
Token ID | 13 |
Feature activation | -4.44e-03 |
Pos logit contributions | |
Amen | +0.896 |
folk | +0.847 |
digits | +0.845 |
Gu | +0.754 |
president | +0.751 |
Neg logit contributions | |
oppy | -0.549 |
opsy | -0.539 |
chy | -0.531 |
bur | -0.497 |
Remy | -0.494 |
Token | Tim |
Token ID | 5045 |
Feature activation | +6.70e-02 |
Pos logit contributions | |
folk | +0.008 |
digits | +0.008 |
sale | +0.007 |
customer | +0.007 |
epad | +0.007 |
Neg logit contributions | |
opsy | -0.008 |
oppy | -0.008 |
Remy | -0.007 |
bur | -0.007 |
chy | -0.007 |
Token | my |
Token ID | 1820 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
opsy | +0.159 |
oppy | +0.158 |
Remy | +0.145 |
chy | +0.144 |
bur | +0.144 |
Neg logit contributions | |
folk | -0.089 |
digits | -0.085 |
sale | -0.079 |
customer | -0.070 |
jury | -0.070 |
Token | was |
Token ID | 373 |
Feature activation | +5.06e-02 |
Pos logit contributions | |
folk | +0.236 |
digits | +0.230 |
sale | +0.216 |
customer | +0.203 |
letters | +0.201 |
Neg logit contributions | |
opsy | -0.252 |
oppy | -0.249 |
Remy | -0.228 |
chy | -0.221 |
bur | -0.221 |
Token | very |
Token ID | 845 |
Feature activation | -4.15e-02 |
Pos logit contributions | |
opsy | +0.084 |
oppy | +0.080 |
chy | +0.079 |
bur | +0.078 |
Remy | +0.077 |
Neg logit contributions | |
folk | -0.095 |
digits | -0.093 |
occasions | -0.084 |
sale | -0.082 |
boxes | -0.082 |
Token | that |
Token ID | 326 |
Feature activation | +8.50e-02 |
Pos logit contributions | |
opsy | +0.249 |
oppy | +0.245 |
Remy | +0.224 |
bur | +0.214 |
chy | +0.212 |
Neg logit contributions | |
folk | -0.292 |
digits | -0.287 |
sale | -0.266 |
epad | -0.253 |
customer | -0.253 |
Token | day |
Token ID | 1110 |
Feature activation | -5.65e-03 |
Pos logit contributions | |
opsy | +0.176 |
oppy | +0.175 |
Remy | +0.161 |
chy | +0.157 |
bur | +0.157 |
Neg logit contributions | |
folk | -0.135 |
digits | -0.132 |
sale | -0.123 |
customer | -0.115 |
letters | -0.113 |
Token | , |
Token ID | 11 |
Feature activation | -3.83e-02 |
Pos logit contributions | |
folk | +0.008 |
digits | +0.008 |
sale | +0.007 |
customer | +0.007 |
letters | +0.007 |
Neg logit contributions | |
opsy | -0.012 |
oppy | -0.012 |
Remy | -0.011 |
chy | -0.011 |
bur | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | +2.54e-02 |
Pos logit contributions | |
folk | +0.056 |
digits | +0.054 |
sale | +0.053 |
customer | +0.048 |
letters | +0.047 |
Neg logit contributions | |
opsy | -0.084 |
oppy | -0.082 |
Remy | -0.076 |
bur | -0.075 |
chy | -0.074 |
Token | garage |
Token ID | 15591 |
Feature activation | -3.02e-02 |
Pos logit contributions | |
opsy | +0.055 |
oppy | +0.053 |
chy | +0.049 |
Remy | +0.049 |
bur | +0.049 |
Neg logit contributions | |
folk | -0.039 |
digits | -0.037 |
sale | -0.033 |
bench | -0.031 |
epad | -0.031 |
Token | sale |
Token ID | 5466 |
Feature activation | -4.20e-01 |
Pos logit contributions | |
digits | +0.063 |
folk | +0.062 |
Amen | +0.059 |
epad | +0.057 |
winners | +0.057 |
Neg logit contributions | |
chy | -0.041 |
oppy | -0.039 |
opsy | -0.039 |
bur | -0.037 |
uffy | -0.036 |
Token | was |
Token ID | 373 |
Feature activation | +5.06e-02 |
Pos logit contributions | |
Amen | +0.789 |
digits | +0.768 |
folk | +0.757 |
customer | +0.674 |
president | +0.663 |
Neg logit contributions | |
oppy | -0.682 |
opsy | -0.654 |
chy | -0.641 |
uffy | -0.606 |
bur | -0.601 |
Token | over |
Token ID | 625 |
Feature activation | -9.58e-02 |
Pos logit contributions | |
chy | +0.075 |
bur | +0.067 |
oppy | +0.064 |
opsy | +0.063 |
poisonous | +0.060 |
Neg logit contributions | |
folk | -0.107 |
digits | -0.099 |
Amen | -0.091 |
combinations | -0.085 |
cles | -0.084 |
Token | and |
Token ID | 290 |
Feature activation | +2.94e-02 |
Pos logit contributions | |
folk | +0.182 |
digits | +0.175 |
Amen | +0.159 |
sale | +0.154 |
salesman | +0.151 |
Neg logit contributions | |
oppy | -0.171 |
opsy | -0.171 |
Remy | -0.154 |
bur | -0.154 |
chy | -0.148 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
oppy | +0.051 |
opsy | +0.050 |
chy | +0.047 |
Remy | +0.047 |
bur | +0.047 |
Neg logit contributions | |
folk | -0.055 |
digits | -0.055 |
Amen | -0.048 |
customer | -0.047 |
sale | -0.047 |
Token | and |
Token ID | 290 |
Feature activation | +3.03e-02 |
Pos logit contributions | |
opsy | +0.027 |
oppy | +0.027 |
Remy | +0.024 |
bur | +0.024 |
chy | +0.024 |
Neg logit contributions | |
folk | -0.025 |
digits | -0.024 |
sale | -0.022 |
customer | -0.021 |
letters | -0.021 |
Token | old |
Token ID | 1468 |
Feature activation | +4.55e-03 |
Pos logit contributions | |
opsy | +0.032 |
oppy | +0.032 |
Remy | +0.029 |
bur | +0.029 |
chy | +0.029 |
Neg logit contributions | |
folk | -0.018 |
digits | -0.017 |
sale | -0.015 |
epad | -0.014 |
Amen | -0.014 |
Token | toys |
Token ID | 14958 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
oppy | +0.009 |
bur | +0.009 |
opsy | +0.009 |
Remy | +0.008 |
chy | +0.008 |
Neg logit contributions | |
folk | -0.007 |
digits | -0.007 |
Amen | -0.006 |
epad | -0.006 |
Adam | -0.006 |
Token | at |
Token ID | 379 |
Feature activation | +6.49e-02 |
Pos logit contributions | |
folk | +0.259 |
digits | +0.257 |
sale | +0.229 |
Amen | +0.222 |
customer | +0.218 |
Neg logit contributions | |
opsy | -0.275 |
oppy | -0.274 |
Remy | -0.251 |
chy | -0.248 |
bur | -0.244 |
Token | a |
Token ID | 257 |
Feature activation | +7.13e-03 |
Pos logit contributions | |
oppy | +0.114 |
opsy | +0.113 |
Remy | +0.105 |
bur | +0.103 |
chy | +0.097 |
Neg logit contributions | |
folk | -0.124 |
digits | -0.119 |
epad | -0.103 |
Amen | -0.102 |
sale | -0.101 |
Token | garage |
Token ID | 15591 |
Feature activation | -2.90e-02 |
Pos logit contributions | |
opsy | +0.014 |
oppy | +0.014 |
Remy | +0.013 |
bur | +0.012 |
chy | +0.012 |
Neg logit contributions | |
folk | -0.012 |
digits | -0.012 |
sale | -0.011 |
letters | -0.010 |
epad | -0.010 |
Token | sale |
Token ID | 5466 |
Feature activation | -4.19e-01 |
Pos logit contributions | |
digits | +0.060 |
folk | +0.059 |
Amen | +0.055 |
cheering | +0.051 |
letters | +0.051 |
Neg logit contributions | |
chy | -0.039 |
bur | -0.038 |
oppy | -0.038 |
opsy | -0.037 |
Remy | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -2.16e-03 |
Pos logit contributions | |
folk | +0.857 |
digits | +0.834 |
Amen | +0.780 |
cheering | +0.731 |
winners | +0.724 |
Neg logit contributions | |
oppy | -0.625 |
opsy | -0.616 |
chy | -0.586 |
bur | -0.565 |
Remy | -0.552 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
folk | +0.004 |
digits | +0.004 |
sale | +0.004 |
epad | +0.004 |
customer | +0.003 |
Neg logit contributions | |
opsy | -0.004 |
oppy | -0.004 |
Remy | -0.004 |
bur | -0.004 |
chy | -0.003 |
Token | was |
Token ID | 373 |
Feature activation | +5.29e-02 |
Pos logit contributions | |
opsy | +0.032 |
oppy | +0.032 |
Remy | +0.029 |
bur | +0.028 |
chy | +0.028 |
Neg logit contributions | |
folk | -0.034 |
digits | -0.033 |
sale | -0.030 |
epad | -0.029 |
customer | -0.029 |
Token | scared |
Token ID | 12008 |
Feature activation | +9.02e-02 |
Pos logit contributions | |
opsy | +0.078 |
chy | +0.074 |
oppy | +0.073 |
Remy | +0.070 |
bur | +0.069 |
Neg logit contributions | |
folk | -0.112 |
digits | -0.109 |
occasions | -0.098 |
boxes | -0.098 |
sale | -0.097 |
Token | because |
Token ID | 780 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
opsy | +0.193 |
oppy | +0.188 |
Remy | +0.172 |
chy | +0.167 |
bur | +0.167 |
Neg logit contributions | |
folk | -0.139 |
digits | -0.135 |
sale | -0.130 |
customer | -0.120 |
letters | -0.119 |
Token | . |
Token ID | 13 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
folk | +0.224 |
digits | +0.218 |
sale | +0.204 |
customer | +0.189 |
letters | +0.187 |
Neg logit contributions | |
opsy | -0.318 |
oppy | -0.313 |
Remy | -0.290 |
bur | -0.282 |
chy | -0.280 |
Token | He |
Token ID | 679 |
Feature activation | -3.40e-02 |
Pos logit contributions | |
folk | +0.022 |
digits | +0.021 |
sale | +0.019 |
epad | +0.018 |
cheering | +0.018 |
Neg logit contributions | |
oppy | -0.028 |
opsy | -0.028 |
Remy | -0.027 |
chy | -0.025 |
bur | -0.025 |
Token | is |
Token ID | 318 |
Feature activation | +3.02e-02 |
Pos logit contributions | |
digits | +0.052 |
folk | +0.051 |
sale | +0.049 |
letters | +0.046 |
customer | +0.045 |
Neg logit contributions | |
opsy | -0.073 |
oppy | -0.069 |
Remy | -0.065 |
bur | -0.063 |
chy | -0.061 |
Token | not |
Token ID | 407 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
oppy | +0.056 |
opsy | +0.055 |
Remy | +0.054 |
chy | +0.053 |
bur | +0.051 |
Neg logit contributions | |
folk | -0.054 |
digits | -0.051 |
sale | -0.046 |
Amen | -0.044 |
bench | -0.043 |
Token | for |
Token ID | 329 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
oppy | +0.022 |
Remy | +0.022 |
opsy | +0.022 |
chy | +0.021 |
bur | +0.019 |
Neg logit contributions | |
folk | -0.025 |
digits | -0.022 |
Amen | -0.022 |
sale | -0.020 |
salesman | -0.019 |
Token | sale |
Token ID | 5466 |
Feature activation | -4.18e-01 |
Pos logit contributions | |
oppy | +0.203 |
Remy | +0.202 |
opsy | +0.195 |
bur | +0.189 |
Ho | +0.175 |
Neg logit contributions | |
folk | -0.197 |
digits | -0.184 |
Amen | -0.176 |
bench | -0.175 |
boxes | -0.165 |
Token | . |
Token ID | 13 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
folk | +0.869 |
digits | +0.857 |
Amen | +0.848 |
cheering | +0.768 |
Gu | +0.756 |
Neg logit contributions | |
oppy | -0.602 |
chy | -0.560 |
Remy | -0.535 |
opsy | -0.532 |
uffy | -0.509 |
Token | Please |
Token ID | 4222 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
folk | +0.023 |
digits | +0.022 |
sale | +0.020 |
cheering | +0.020 |
epad | +0.019 |
Neg logit contributions | |
oppy | -0.027 |
opsy | -0.026 |
Remy | -0.026 |
bur | -0.023 |
chy | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | -4.80e-02 |
Pos logit contributions | |
folk | +0.360 |
digits | +0.340 |
sale | +0.303 |
epad | +0.300 |
bench | +0.298 |
Neg logit contributions | |
oppy | -0.352 |
opsy | -0.345 |
Remy | -0.323 |
chy | -0.305 |
uffy | -0.297 |
Token | put |
Token ID | 1234 |
Feature activation | -5.67e-02 |
Pos logit contributions | |
folk | +0.083 |
digits | +0.080 |
sale | +0.074 |
bench | +0.069 |
letters | +0.069 |
Neg logit contributions | |
oppy | -0.093 |
opsy | -0.092 |
Remy | -0.087 |
chy | -0.082 |
bur | -0.082 |
Token | him |
Token ID | 683 |
Feature activation | -1.72e-03 |
Pos logit contributions | |
folk | +0.101 |
digits | +0.096 |
epad | +0.089 |
sale | +0.086 |
Amen | +0.085 |
Neg logit contributions | |
oppy | -0.101 |
Remy | -0.101 |
opsy | -0.099 |
bur | -0.091 |
Lucky | -0.089 |
Token | T |
Token ID | 51 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
digits | +0.243 |
folk | +0.242 |
invitations | +0.229 |
epad | +0.222 |
sale | +0.221 |
Neg logit contributions | |
opsy | -0.223 |
Remy | -0.203 |
oppy | -0.195 |
chy | -0.185 |
bur | -0.180 |
Token | oys |
Token ID | 19417 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
opsy | +0.306 |
oppy | +0.303 |
Remy | +0.282 |
bur | +0.276 |
chy | +0.276 |
Neg logit contributions | |
folk | -0.192 |
digits | -0.187 |
sale | -0.174 |
customer | -0.159 |
letters | -0.157 |
Token | and |
Token ID | 290 |
Feature activation | +2.53e-02 |
Pos logit contributions | |
folk | +0.199 |
digits | +0.193 |
sale | +0.177 |
customer | +0.165 |
letters | +0.162 |
Neg logit contributions | |
opsy | -0.273 |
oppy | -0.273 |
Remy | -0.253 |
chy | -0.246 |
bur | -0.244 |
Token | books |
Token ID | 3835 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
oppy | +0.048 |
opsy | +0.047 |
Remy | +0.045 |
bur | +0.043 |
chy | +0.042 |
Neg logit contributions | |
folk | -0.045 |
digits | -0.043 |
sale | -0.040 |
epad | -0.038 |
customer | -0.037 |
Token | for |
Token ID | 329 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
folk | +0.211 |
digits | +0.208 |
sale | +0.187 |
customer | +0.180 |
Amen | +0.178 |
Neg logit contributions | |
oppy | -0.216 |
opsy | -0.215 |
chy | -0.199 |
Remy | -0.199 |
bur | -0.190 |
Token | sale |
Token ID | 5466 |
Feature activation | -4.18e-01 |
Pos logit contributions | |
Remy | +0.199 |
oppy | +0.195 |
opsy | +0.195 |
bur | +0.193 |
Ho | +0.175 |
Neg logit contributions | |
folk | -0.214 |
digits | -0.193 |
bench | -0.179 |
epad | -0.172 |
salesman | -0.171 |
Token | ! |
Token ID | 0 |
Feature activation | +9.41e-02 |
Pos logit contributions | |
folk | +0.827 |
digits | +0.801 |
Amen | +0.771 |
jury | +0.700 |
cheering | +0.696 |
Neg logit contributions | |
oppy | -0.661 |
opsy | -0.623 |
Remy | -0.604 |
chy | -0.594 |
uffy | -0.565 |
Token | One |
Token ID | 1881 |
Feature activation | -9.39e-02 |
Pos logit contributions | |
oppy | +0.178 |
opsy | +0.177 |
Remy | +0.173 |
bur | +0.155 |
Lucky | +0.154 |
Neg logit contributions | |
folk | -0.159 |
digits | -0.154 |
sale | -0.137 |
bench | -0.135 |
epad | -0.134 |
Token | dollar |
Token ID | 8872 |
Feature activation | -2.60e-01 |
Pos logit contributions | |
folk | +0.166 |
digits | +0.161 |
epad | +0.144 |
sale | +0.143 |
jury | +0.139 |
Neg logit contributions | |
opsy | -0.177 |
oppy | -0.175 |
Remy | -0.162 |
chy | -0.159 |
bur | -0.155 |
Token | each |
Token ID | 1123 |
Feature activation | +1.39e-02 |
Pos logit contributions | |
folk | +0.473 |
digits | +0.456 |
Amen | +0.431 |
epad | +0.414 |
cheering | +0.406 |
Neg logit contributions | |
oppy | -0.457 |
opsy | -0.456 |
Remy | -0.435 |
bur | -0.400 |
chy | -0.398 |
Token | !" |
Token ID | 2474 |
Feature activation | -8.69e-02 |
Pos logit contributions | |
opsy | +0.029 |
oppy | +0.028 |
Remy | +0.027 |
chy | +0.026 |
bur | +0.025 |
Neg logit contributions | |
folk | -0.021 |
digits | -0.021 |
Amen | -0.018 |
epad | -0.018 |
sale | -0.018 |
Token | these |
Token ID | 777 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
folk | +0.035 |
digits | +0.035 |
sale | +0.032 |
customer | +0.030 |
letters | +0.030 |
Neg logit contributions | |
opsy | -0.042 |
oppy | -0.041 |
Remy | -0.038 |
chy | -0.037 |
bur | -0.037 |
Token | carrots |
Token ID | 34397 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
folk | +0.199 |
digits | +0.185 |
letters | +0.184 |
sale | +0.180 |
prices | +0.168 |
Neg logit contributions | |
opsy | -0.308 |
oppy | -0.299 |
chy | -0.271 |
Remy | -0.268 |
bur | -0.266 |
Token | are |
Token ID | 389 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
opsy | +0.396 |
oppy | +0.384 |
Remy | +0.359 |
bur | +0.345 |
rig | +0.334 |
Neg logit contributions | |
folk | -0.242 |
sale | -0.225 |
digits | -0.224 |
letters | -0.204 |
winners | -0.196 |
Token | not |
Token ID | 407 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
opsy | +0.035 |
oppy | +0.035 |
Remy | +0.030 |
uffy | +0.030 |
bur | +0.030 |
Neg logit contributions | |
sale | -0.024 |
folk | -0.024 |
digits | -0.024 |
letters | -0.022 |
customer | -0.022 |
Token | for |
Token ID | 329 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
opsy | +0.034 |
oppy | +0.033 |
Remy | +0.031 |
chy | +0.030 |
bur | +0.030 |
Neg logit contributions | |
folk | -0.028 |
digits | -0.028 |
sale | -0.026 |
customer | -0.024 |
letters | -0.024 |
Token | sale |
Token ID | 5466 |
Feature activation | -4.18e-01 |
Pos logit contributions | |
opsy | +0.221 |
oppy | +0.220 |
Remy | +0.209 |
bur | +0.200 |
chy | +0.194 |
Neg logit contributions | |
folk | -0.193 |
digits | -0.190 |
sale | -0.166 |
bench | -0.162 |
customer | -0.162 |
Token | . |
Token ID | 13 |
Feature activation | -7.04e-03 |
Pos logit contributions | |
folk | +0.733 |
digits | +0.727 |
Amen | +0.651 |
sale | +0.645 |
customer | +0.628 |
Neg logit contributions | |
oppy | -0.775 |
opsy | -0.760 |
Remy | -0.709 |
chy | -0.707 |
bur | -0.677 |
Token | They |
Token ID | 1119 |
Feature activation | -3.36e-02 |
Pos logit contributions | |
folk | +0.010 |
digits | +0.010 |
sale | +0.009 |
customer | +0.008 |
letters | +0.008 |
Neg logit contributions | |
opsy | -0.016 |
oppy | -0.015 |
Remy | -0.014 |
chy | -0.014 |
bur | -0.014 |
Token | are |
Token ID | 389 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
folk | +0.054 |
digits | +0.052 |
sale | +0.049 |
letters | +0.046 |
customer | +0.045 |
Neg logit contributions | |
opsy | -0.069 |
oppy | -0.068 |
Remy | -0.062 |
bur | -0.060 |
chy | -0.060 |
Token | for |
Token ID | 329 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
opsy | +0.035 |
oppy | +0.035 |
Remy | +0.031 |
bur | +0.030 |
chy | +0.030 |
Neg logit contributions | |
folk | -0.029 |
digits | -0.028 |
sale | -0.028 |
letters | -0.026 |
customer | -0.026 |
Token | our |
Token ID | 674 |
Feature activation | +6.26e-02 |
Pos logit contributions | |
opsy | +0.244 |
oppy | +0.242 |
Remy | +0.232 |
bur | +0.223 |
chy | +0.216 |
Neg logit contributions | |
folk | -0.175 |
digits | -0.170 |
sale | -0.146 |
bench | -0.143 |
epad | -0.142 |
Token | . |
Token ID | 13 |
Feature activation | -9.16e-03 |
Pos logit contributions | |
opsy | +0.237 |
oppy | +0.230 |
Remy | +0.214 |
bur | +0.210 |
chy | +0.203 |
Neg logit contributions | |
folk | -0.141 |
sale | -0.136 |
digits | -0.135 |
customer | -0.124 |
letters | -0.120 |
Token | He |
Token ID | 679 |
Feature activation | -2.83e-02 |
Pos logit contributions | |
folk | +0.010 |
digits | +0.010 |
Amen | +0.009 |
sale | +0.009 |
customer | +0.009 |
Neg logit contributions | |
opsy | -0.023 |
oppy | -0.022 |
bur | -0.021 |
chy | -0.021 |
rig | -0.020 |
Token | is |
Token ID | 318 |
Feature activation | +3.58e-02 |
Pos logit contributions | |
digits | +0.039 |
folk | +0.039 |
sale | +0.038 |
letters | +0.037 |
customer | +0.035 |
Neg logit contributions | |
opsy | -0.063 |
oppy | -0.060 |
Remy | -0.056 |
bur | -0.054 |
rig | -0.052 |
Token | not |
Token ID | 407 |
Feature activation | +1.72e-02 |
Pos logit contributions | |
opsy | +0.069 |
oppy | +0.068 |
Remy | +0.065 |
chy | +0.063 |
bur | +0.062 |
Neg logit contributions | |
folk | -0.063 |
digits | -0.060 |
sale | -0.054 |
customer | -0.050 |
Amen | -0.050 |
Token | for |
Token ID | 329 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
opsy | +0.030 |
oppy | +0.030 |
Remy | +0.029 |
chy | +0.029 |
bur | +0.027 |
Neg logit contributions | |
folk | -0.033 |
digits | -0.031 |
sale | -0.027 |
Amen | -0.027 |
salesman | -0.026 |
Token | sale |
Token ID | 5466 |
Feature activation | -4.18e-01 |
Pos logit contributions | |
oppy | +0.208 |
Remy | +0.207 |
opsy | +0.204 |
bur | +0.195 |
Ho | +0.181 |
Neg logit contributions | |
folk | -0.206 |
digits | -0.195 |
bench | -0.179 |
Amen | -0.176 |
salesman | -0.173 |
Token | . |
Token ID | 13 |
Feature activation | -8.68e-03 |
Pos logit contributions | |
folk | +0.875 |
digits | +0.867 |
Amen | +0.856 |
cheering | +0.775 |
banners | +0.756 |
Neg logit contributions | |
oppy | -0.594 |
chy | -0.560 |
Remy | -0.530 |
opsy | -0.516 |
uffy | -0.503 |
Token | He |
Token ID | 679 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
folk | +0.013 |
digits | +0.012 |
sale | +0.011 |
customer | +0.011 |
letters | +0.010 |
Neg logit contributions | |
opsy | -0.019 |
oppy | -0.019 |
Remy | -0.017 |
chy | -0.017 |
bur | -0.017 |
Token | is |
Token ID | 318 |
Feature activation | +3.71e-02 |
Pos logit contributions | |
digits | +0.040 |
folk | +0.039 |
sale | +0.038 |
letters | +0.037 |
customer | +0.035 |
Neg logit contributions | |
opsy | -0.060 |
oppy | -0.057 |
Remy | -0.053 |
bur | -0.052 |
chy | -0.049 |
Token | my |
Token ID | 616 |
Feature activation | +2.34e-03 |
Pos logit contributions | |
opsy | +0.069 |
oppy | +0.069 |
Remy | +0.066 |
chy | +0.065 |
bur | +0.063 |
Neg logit contributions | |
folk | -0.066 |
digits | -0.063 |
sale | -0.057 |
Amen | -0.053 |
salesman | -0.053 |
Token | friend |
Token ID | 1545 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
opsy | +0.005 |
oppy | +0.005 |
Remy | +0.004 |
chy | +0.004 |
bur | +0.004 |
Neg logit contributions | |
folk | -0.004 |
digits | -0.004 |
sale | -0.004 |
customer | -0.003 |
letters | -0.003 |
Token | 's |
Token ID | 338 |
Feature activation | +9.19e-02 |
Pos logit contributions | |
folk | +0.198 |
digits | +0.187 |
sale | +0.182 |
letters | +0.174 |
customer | +0.164 |
Neg logit contributions | |
opsy | -0.274 |
oppy | -0.264 |
Remy | -0.247 |
bur | -0.242 |
rig | -0.229 |
Token | yard |
Token ID | 12699 |
Feature activation | +9.43e-02 |
Pos logit contributions | |
opsy | +0.193 |
oppy | +0.191 |
Remy | +0.177 |
chy | +0.171 |
bur | +0.170 |
Neg logit contributions | |
folk | -0.141 |
digits | -0.139 |
sale | -0.132 |
letters | -0.122 |
customer | -0.122 |
Token | was |
Token ID | 373 |
Feature activation | +5.41e-02 |
Pos logit contributions | |
opsy | +0.187 |
oppy | +0.183 |
Remy | +0.169 |
bur | +0.162 |
chy | +0.160 |
Neg logit contributions | |
folk | -0.160 |
digits | -0.155 |
sale | -0.148 |
letters | -0.138 |
customer | -0.136 |
Token | not |
Token ID | 407 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
opsy | +0.093 |
oppy | +0.092 |
chy | +0.089 |
Remy | +0.087 |
bur | +0.086 |
Neg logit contributions | |
folk | -0.103 |
digits | -0.100 |
sale | -0.088 |
Amen | -0.085 |
customer | -0.085 |
Token | for |
Token ID | 329 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
opsy | +0.040 |
oppy | +0.040 |
chy | +0.040 |
Remy | +0.039 |
bur | +0.037 |
Neg logit contributions | |
folk | -0.047 |
digits | -0.046 |
Amen | -0.041 |
sale | -0.040 |
jury | -0.039 |
Token | sale |
Token ID | 5466 |
Feature activation | -4.18e-01 |
Pos logit contributions | |
oppy | +0.216 |
Remy | +0.215 |
opsy | +0.210 |
bur | +0.209 |
chy | +0.190 |
Neg logit contributions | |
folk | -0.194 |
digits | -0.189 |
Amen | -0.186 |
bench | -0.170 |
salesman | -0.164 |
Token | . |
Token ID | 13 |
Feature activation | -1.34e-03 |
Pos logit contributions | |
digits | +0.755 |
folk | +0.755 |
Amen | +0.722 |
sale | +0.654 |
cheering | +0.649 |
Neg logit contributions | |
oppy | -0.732 |
opsy | -0.701 |
chy | -0.683 |
Remy | -0.658 |
bur | -0.632 |
Token | Tim |
Token ID | 5045 |
Feature activation | +7.20e-02 |
Pos logit contributions | |
folk | +0.002 |
digits | +0.002 |
sale | +0.002 |
customer | +0.002 |
epad | +0.002 |
Neg logit contributions | |
opsy | -0.002 |
oppy | -0.002 |
Remy | -0.002 |
bur | -0.002 |
chy | -0.002 |
Token | my |
Token ID | 1820 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
opsy | +0.179 |
oppy | +0.177 |
Remy | +0.165 |
chy | +0.163 |
bur | +0.162 |
Neg logit contributions | |
folk | -0.086 |
digits | -0.083 |
sale | -0.076 |
customer | -0.068 |
letters | -0.066 |
Token | was |
Token ID | 373 |
Feature activation | +5.53e-02 |
Pos logit contributions | |
folk | +0.217 |
digits | +0.211 |
sale | +0.199 |
customer | +0.186 |
letters | +0.185 |
Neg logit contributions | |
opsy | -0.249 |
oppy | -0.246 |
Remy | -0.227 |
chy | -0.219 |
bur | -0.218 |
Token | very |
Token ID | 845 |
Feature activation | -3.52e-02 |
Pos logit contributions | |
opsy | +0.094 |
oppy | +0.091 |
Remy | +0.086 |
chy | +0.086 |
bur | +0.083 |
Neg logit contributions | |
folk | -0.108 |
digits | -0.107 |
sale | -0.096 |
epad | -0.092 |
boxes | -0.092 |
Token | old |
Token ID | 1468 |
Feature activation | +4.50e-03 |
Pos logit contributions | |
folk | +0.075 |
digits | +0.069 |
epad | +0.064 |
Amen | +0.063 |
bench | +0.062 |
Neg logit contributions | |
opsy | -0.102 |
oppy | -0.100 |
Remy | -0.093 |
bur | -0.092 |
uffy | -0.088 |
Token | toys |
Token ID | 14958 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
opsy | +0.008 |
bur | +0.008 |
oppy | +0.008 |
Remy | +0.008 |
uffy | +0.007 |
Neg logit contributions | |
folk | -0.007 |
digits | -0.007 |
Amen | -0.007 |
epad | -0.007 |
Adam | -0.006 |
Token | at |
Token ID | 379 |
Feature activation | +6.58e-02 |
Pos logit contributions | |
folk | +0.263 |
digits | +0.262 |
sale | +0.231 |
Amen | +0.228 |
epad | +0.222 |
Neg logit contributions | |
oppy | -0.272 |
opsy | -0.271 |
Remy | -0.248 |
chy | -0.247 |
bur | -0.240 |
Token | a |
Token ID | 257 |
Feature activation | +7.54e-03 |
Pos logit contributions | |
oppy | +0.117 |
opsy | +0.116 |
Remy | +0.110 |
bur | +0.106 |
chy | +0.100 |
Neg logit contributions | |
folk | -0.123 |
digits | -0.120 |
epad | -0.105 |
Amen | -0.103 |
sale | -0.103 |
Token | garage |
Token ID | 15591 |
Feature activation | -2.65e-02 |
Pos logit contributions | |
opsy | +0.015 |
oppy | +0.015 |
Remy | +0.014 |
bur | +0.014 |
chy | +0.013 |
Neg logit contributions | |
folk | -0.013 |
digits | -0.012 |
sale | -0.011 |
letters | -0.010 |
epad | -0.010 |
Token | sale |
Token ID | 5466 |
Feature activation | -4.17e-01 |
Pos logit contributions | |
digits | +0.055 |
folk | +0.054 |
Amen | +0.050 |
cheering | +0.046 |
letters | +0.046 |
Neg logit contributions | |
chy | -0.037 |
oppy | -0.037 |
bur | -0.036 |
opsy | -0.035 |
Remy | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -1.92e-03 |
Pos logit contributions | |
folk | +0.851 |
digits | +0.829 |
Amen | +0.774 |
cheering | +0.734 |
winners | +0.722 |
Neg logit contributions | |
oppy | -0.637 |
opsy | -0.618 |
chy | -0.594 |
Remy | -0.566 |
bur | -0.560 |
Token | Tim |
Token ID | 5045 |
Feature activation | +7.11e-02 |
Pos logit contributions | |
folk | +0.004 |
digits | +0.004 |
sale | +0.003 |
epad | +0.003 |
customer | +0.003 |
Neg logit contributions | |
opsy | -0.003 |
oppy | -0.003 |
Remy | -0.003 |
bur | -0.003 |
chy | -0.003 |
Token | my |
Token ID | 1820 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
opsy | +0.172 |
oppy | +0.171 |
Remy | +0.158 |
chy | +0.156 |
bur | +0.156 |
Neg logit contributions | |
folk | -0.091 |
digits | -0.086 |
sale | -0.080 |
customer | -0.071 |
jury | -0.070 |
Token | thought |
Token ID | 1807 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
folk | +0.234 |
digits | +0.225 |
sale | +0.207 |
epad | +0.200 |
customer | +0.195 |
Neg logit contributions | |
opsy | -0.252 |
oppy | -0.249 |
Remy | -0.227 |
bur | -0.222 |
chy | -0.221 |
Token | it |
Token ID | 340 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
folk | +0.239 |
digits | +0.236 |
sale | +0.230 |
customer | +0.207 |
letters | +0.207 |
Neg logit contributions | |
opsy | -0.350 |
oppy | -0.340 |
Remy | -0.311 |
bur | -0.309 |
chy | -0.304 |
Token | ens |
Token ID | 641 |
Feature activation | +3.57e-01 |
Pos logit contributions | |
opsy | +0.620 |
oppy | +0.607 |
bur | +0.589 |
rig | +0.580 |
web | +0.524 |
Neg logit contributions | |
folk | -0.725 |
letters | -0.716 |
ones | -0.710 |
customer | -0.698 |
digits | -0.698 |
Token | by |
Token ID | 416 |
Feature activation | +4.97e-02 |
Pos logit contributions | |
opsy | +0.968 |
oppy | +0.914 |
rig | +0.901 |
Remy | +0.823 |
uffy | +0.799 |
Neg logit contributions | |
sale | -0.347 |
folk | -0.291 |
letters | -0.283 |
cheering | -0.270 |
digits | -0.267 |
Token | mistake |
Token ID | 7457 |
Feature activation | +4.69e-02 |
Pos logit contributions | |
opsy | +0.103 |
oppy | +0.100 |
Remy | +0.097 |
bur | +0.093 |
chy | +0.092 |
Neg logit contributions | |
folk | -0.081 |
digits | -0.078 |
sale | -0.069 |
customer | -0.066 |
Amen | -0.066 |
Token | . |
Token ID | 13 |
Feature activation | -5.35e-03 |
Pos logit contributions | |
opsy | +0.118 |
oppy | +0.114 |
rig | +0.106 |
bur | +0.104 |
chy | +0.103 |
Neg logit contributions | |
folk | -0.053 |
sale | -0.050 |
digits | -0.049 |
letters | -0.048 |
cheering | -0.043 |
Token | Mitt |
Token ID | 16627 |
Feature activation | +4.15e-01 |
Pos logit contributions | |
sale | +0.007 |
folk | +0.006 |
digits | +0.006 |
customer | +0.006 |
letters | +0.005 |
Neg logit contributions | |
opsy | -0.012 |
oppy | -0.012 |
rig | -0.011 |
chy | -0.011 |
bur | -0.010 |
Token | ens |
Token ID | 641 |
Feature activation | +3.57e-01 |
Pos logit contributions | |
opsy | +0.646 |
oppy | +0.631 |
bur | +0.611 |
rig | +0.586 |
web | +0.534 |
Neg logit contributions | |
ones | -0.676 |
folk | -0.676 |
digits | -0.667 |
letters | -0.661 |
customer | -0.654 |
Token | was |
Token ID | 373 |
Feature activation | +5.25e-02 |
Pos logit contributions | |
opsy | +0.823 |
oppy | +0.807 |
rig | +0.776 |
Remy | +0.717 |
chy | +0.692 |
Neg logit contributions | |
folk | -0.440 |
sale | -0.425 |
digits | -0.423 |
letters | -0.388 |
begged | -0.379 |
Token | upset |
Token ID | 9247 |
Feature activation | -2.50e-02 |
Pos logit contributions | |
opsy | +0.109 |
oppy | +0.109 |
Remy | +0.099 |
bur | +0.096 |
chy | +0.094 |
Neg logit contributions | |
folk | -0.080 |
digits | -0.079 |
sale | -0.077 |
customer | -0.071 |
cheering | -0.070 |
Token | and |
Token ID | 290 |
Feature activation | +3.01e-02 |
Pos logit contributions | |
sale | +0.023 |
folk | +0.022 |
digits | +0.022 |
cheering | +0.021 |
letters | +0.020 |
Neg logit contributions | |
opsy | -0.067 |
oppy | -0.064 |
rig | -0.058 |
bur | -0.057 |
Remy | -0.056 |
Token | said |
Token ID | 531 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
opsy | +0.071 |
oppy | +0.069 |
rig | +0.065 |
uffy | +0.061 |
bur | +0.057 |
Neg logit contributions | |
folk | -0.034 |
begged | -0.034 |
sale | -0.033 |
asked | -0.032 |
letters | -0.032 |
Token | , |
Token ID | 11 |
Feature activation | -3.94e-02 |
Pos logit contributions | |
folk | +0.022 |
digits | +0.021 |
sale | +0.020 |
customer | +0.018 |
letters | +0.018 |
Neg logit contributions | |
opsy | -0.034 |
oppy | -0.034 |
Remy | -0.031 |
chy | -0.031 |
bur | -0.030 |
Token | Spot |
Token ID | 15899 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
folk | +0.013 |
sale | +0.013 |
digits | +0.012 |
customer | +0.012 |
letters | +0.012 |
Neg logit contributions | |
opsy | -0.015 |
oppy | -0.014 |
chy | -0.013 |
bur | -0.013 |
rig | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | +3.40e-02 |
Pos logit contributions | |
opsy | +0.533 |
oppy | +0.525 |
bur | +0.478 |
rig | +0.470 |
chy | +0.469 |
Neg logit contributions | |
folk | -0.283 |
sale | -0.275 |
digits | -0.261 |
customer | -0.252 |
letters | -0.239 |
Token | a |
Token ID | 257 |
Feature activation | +9.00e-03 |
Pos logit contributions | |
opsy | +0.068 |
oppy | +0.068 |
chy | +0.060 |
rig | +0.059 |
bur | +0.058 |
Neg logit contributions | |
folk | -0.054 |
digits | -0.053 |
sale | -0.052 |
customer | -0.052 |
letters | -0.050 |
Token | cat |
Token ID | 3797 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
opsy | +0.017 |
oppy | +0.017 |
Remy | +0.015 |
chy | +0.015 |
bur | +0.015 |
Neg logit contributions | |
folk | -0.015 |
digits | -0.015 |
sale | -0.014 |
customer | -0.014 |
letters | -0.014 |
Token | named |
Token ID | 3706 |
Feature activation | -6.74e-03 |
Pos logit contributions | |
opsy | +0.261 |
oppy | +0.255 |
Remy | +0.231 |
chy | +0.227 |
bur | +0.224 |
Neg logit contributions | |
folk | -0.138 |
sale | -0.133 |
digits | -0.131 |
customer | -0.116 |
letters | -0.116 |
Token | Fl |
Token ID | 1610 |
Feature activation | +3.84e-01 |
Pos logit contributions | |
folk | +0.012 |
sale | +0.012 |
customer | +0.011 |
digits | +0.010 |
letters | +0.010 |
Neg logit contributions | |
opsy | -0.012 |
oppy | -0.012 |
rig | -0.011 |
bur | -0.011 |
chy | -0.011 |
Token | uffy |
Token ID | 15352 |
Feature activation | +4.61e-01 |
Pos logit contributions | |
chy | +0.350 |
rig | +0.286 |
Remy | +0.286 |
refreshing | +0.264 |
web | +0.254 |
Neg logit contributions | |
folk | -0.909 |
sale | -0.903 |
customer | -0.902 |
digits | -0.897 |
number | -0.842 |
Token | . |
Token ID | 13 |
Feature activation | +2.45e-03 |
Pos logit contributions | |
opsy | +1.120 |
oppy | +1.101 |
bur | +0.995 |
Remy | +0.990 |
chy | +0.969 |
Neg logit contributions | |
folk | -0.555 |
sale | -0.536 |
digits | -0.504 |
customer | -0.470 |
letters | -0.460 |
Token |
|
Token ID | 198 |
Feature activation | +2.49e-02 |
Pos logit contributions | |
oppy | +0.005 |
opsy | +0.005 |
chy | +0.005 |
bur | +0.005 |
Remy | +0.005 |
Neg logit contributions | |
folk | -0.004 |
digits | -0.003 |
sale | -0.003 |
customer | -0.003 |
Amen | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
opsy | +0.048 |
oppy | +0.048 |
Remy | +0.045 |
bur | +0.042 |
chy | +0.042 |
Neg logit contributions | |
folk | -0.043 |
digits | -0.042 |
sale | -0.039 |
letters | -0.037 |
epad | -0.037 |
Token | One |
Token ID | 3198 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
opsy | +0.059 |
oppy | +0.057 |
Remy | +0.056 |
bur | +0.053 |
chy | +0.052 |
Neg logit contributions | |
folk | -0.034 |
digits | -0.032 |
sale | -0.029 |
epad | -0.028 |
letters | -0.027 |
Token | cat |
Token ID | 3797 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
opsy | +0.010 |
oppy | +0.010 |
Remy | +0.009 |
chy | +0.009 |
bur | +0.009 |
Neg logit contributions | |
folk | -0.009 |
digits | -0.009 |
sale | -0.009 |
customer | -0.008 |
letters | -0.008 |
Token | named |
Token ID | 3706 |
Feature activation | -9.07e-04 |
Pos logit contributions | |
opsy | +0.262 |
oppy | +0.259 |
Remy | +0.236 |
bur | +0.233 |
chy | +0.232 |
Neg logit contributions | |
folk | -0.152 |
digits | -0.147 |
sale | -0.143 |
customer | -0.128 |
Amen | -0.127 |
Token | Mitt |
Token ID | 16627 |
Feature activation | +4.19e-01 |
Pos logit contributions | |
sale | +0.001 |
digits | +0.001 |
customer | +0.001 |
folk | +0.001 |
number | +0.001 |
Neg logit contributions | |
opsy | -0.002 |
oppy | -0.002 |
rig | -0.002 |
bur | -0.002 |
chy | -0.002 |
Token | ens |
Token ID | 641 |
Feature activation | +3.58e-01 |
Pos logit contributions | |
oppy | +0.641 |
opsy | +0.631 |
rig | +0.628 |
bur | +0.594 |
web | +0.583 |
Neg logit contributions | |
ones | -0.675 |
folk | -0.665 |
customer | -0.657 |
sale | -0.644 |
bench | -0.628 |
Token | . |
Token ID | 13 |
Feature activation | +4.57e-03 |
Pos logit contributions | |
opsy | +0.945 |
oppy | +0.930 |
rig | +0.857 |
bur | +0.845 |
uffy | +0.828 |
Neg logit contributions | |
sale | -0.342 |
folk | -0.328 |
digits | -0.296 |
customer | -0.275 |
letters | -0.274 |
Token | Mitt |
Token ID | 16627 |
Feature activation | +4.20e-01 |
Pos logit contributions | |
opsy | +0.006 |
oppy | +0.006 |
rig | +0.006 |
chy | +0.005 |
bur | +0.005 |
Neg logit contributions | |
digits | -0.009 |
sale | -0.009 |
folk | -0.009 |
letters | -0.009 |
customer | -0.009 |
Token | ens |
Token ID | 641 |
Feature activation | +3.59e-01 |
Pos logit contributions | |
oppy | +0.642 |
opsy | +0.621 |
bur | +0.589 |
rig | +0.575 |
web | +0.543 |
Neg logit contributions | |
ones | -0.681 |
digits | -0.662 |
bench | -0.657 |
folk | -0.656 |
letters | -0.651 |
Token | loved |
Token ID | 6151 |
Feature activation | +3.92e-02 |
Pos logit contributions | |
opsy | +0.864 |
oppy | +0.858 |
rig | +0.798 |
bur | +0.766 |
Remy | +0.757 |
Neg logit contributions | |
folk | -0.384 |
sale | -0.378 |
digits | -0.367 |
letters | -0.347 |
number | -0.344 |
Token | to |
Token ID | 284 |
Feature activation | +4.61e-03 |
Pos logit contributions | |
oppy | +0.094 |
opsy | +0.094 |
chy | +0.083 |
uffy | +0.081 |
rig | +0.081 |
Neg logit contributions | |
sale | -0.048 |
digits | -0.048 |
folk | -0.046 |
letters | -0.044 |
customer | -0.043 |
Token | play |
Token ID | 711 |
Feature activation | +5.72e-02 |
Pos logit contributions | |
opsy | +0.010 |
oppy | +0.010 |
uffy | +0.009 |
rig | +0.009 |
Remy | +0.009 |
Neg logit contributions | |
sale | -0.006 |
digits | -0.006 |
folk | -0.006 |
number | -0.006 |
letters | -0.006 |
Token | with |
Token ID | 351 |
Feature activation | -8.52e-02 |
Pos logit contributions | |
opsy | +0.137 |
oppy | +0.135 |
chy | +0.122 |
Remy | +0.121 |
bur | +0.119 |
Neg logit contributions | |
folk | -0.071 |
digits | -0.070 |
sale | -0.069 |
customer | -0.061 |
letters | -0.061 |
Token | was |
Token ID | 373 |
Feature activation | +4.71e-02 |
Pos logit contributions | |
digits | +0.144 |
folk | +0.143 |
sale | +0.141 |
number | +0.123 |
cheering | +0.123 |
Neg logit contributions | |
opsy | -0.305 |
oppy | -0.303 |
Remy | -0.279 |
bur | -0.279 |
rig | -0.276 |
Token | a |
Token ID | 257 |
Feature activation | +3.51e-03 |
Pos logit contributions | |
oppy | +0.119 |
opsy | +0.116 |
bur | +0.107 |
chy | +0.106 |
Remy | +0.106 |
Neg logit contributions | |
sale | -0.049 |
digits | -0.048 |
folk | -0.047 |
customer | -0.044 |
letters | -0.042 |
Token | little |
Token ID | 1310 |
Feature activation | +5.96e-02 |
Pos logit contributions | |
oppy | +0.007 |
opsy | +0.007 |
Remy | +0.007 |
chy | +0.007 |
bur | +0.007 |
Neg logit contributions | |
digits | -0.005 |
customer | -0.005 |
folk | -0.005 |
sale | -0.005 |
Amen | -0.004 |
Token | rabbit |
Token ID | 22746 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
opsy | +0.145 |
oppy | +0.144 |
chy | +0.136 |
Remy | +0.135 |
rig | +0.131 |
Neg logit contributions | |
folk | -0.066 |
digits | -0.063 |
customer | -0.058 |
sale | -0.058 |
letters | -0.054 |
Token | named |
Token ID | 3706 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
opsy | +0.491 |
oppy | +0.487 |
Remy | +0.447 |
chy | +0.447 |
rig | +0.445 |
Neg logit contributions | |
folk | -0.198 |
sale | -0.196 |
digits | -0.192 |
Amen | -0.184 |
customer | -0.169 |
Token | Remy |
Token ID | 32912 |
Feature activation | +5.27e-01 |
Pos logit contributions | |
sale | +0.014 |
digits | +0.013 |
customer | +0.012 |
one | +0.012 |
number | +0.012 |
Neg logit contributions | |
opsy | -0.022 |
rig | -0.021 |
oppy | -0.020 |
chy | -0.019 |
uffy | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -7.68e-03 |
Pos logit contributions | |
opsy | +1.454 |
oppy | +1.419 |
rig | +1.354 |
uffy | +1.344 |
bur | +1.339 |
Neg logit contributions | |
sale | -0.295 |
digits | -0.271 |
one | -0.255 |
asked | -0.238 |
letters | -0.236 |
Token | Remy |
Token ID | 32912 |
Feature activation | +5.28e-01 |
Pos logit contributions | |
digits | +0.017 |
ones | +0.017 |
number | +0.016 |
sale | +0.016 |
letters | +0.016 |
Neg logit contributions | |
rig | -0.008 |
opsy | -0.008 |
oppy | -0.008 |
chy | -0.006 |
iqu | -0.006 |
Token | loved |
Token ID | 6151 |
Feature activation | +2.90e-02 |
Pos logit contributions | |
opsy | +1.299 |
oppy | +1.261 |
rig | +1.197 |
bur | +1.159 |
Remy | +1.141 |
Neg logit contributions | |
digits | -0.469 |
sale | -0.454 |
asked | -0.453 |
letters | -0.453 |
number | -0.451 |
Token | to |
Token ID | 284 |
Feature activation | -5.23e-03 |
Pos logit contributions | |
oppy | +0.071 |
opsy | +0.070 |
chy | +0.064 |
uffy | +0.063 |
rig | +0.063 |
Neg logit contributions | |
sale | -0.033 |
digits | -0.033 |
letters | -0.032 |
folk | -0.030 |
customer | -0.029 |
Token | spring |
Token ID | 6076 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
digits | +0.008 |
sale | +0.008 |
number | +0.007 |
folk | +0.007 |
letters | +0.007 |
Neg logit contributions | |
opsy | -0.011 |
oppy | -0.010 |
uffy | -0.010 |
rig | -0.009 |
chy | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -1.05e-05 |
Pos logit contributions | |
opsy | +0.024 |
oppy | +0.024 |
Remy | +0.022 |
bur | +0.021 |
chy | +0.020 |
Neg logit contributions | |
folk | -0.013 |
digits | -0.012 |
sale | -0.012 |
letters | -0.011 |
cheering | -0.011 |
Token | She |
Token ID | 1375 |
Feature activation | +2.86e-02 |
Pos logit contributions | |
folk | +0.000 |
digits | +0.000 |
sale | +0.000 |
letters | +0.000 |
Amen | +0.000 |
Neg logit contributions | |
oppy | -0.000 |
opsy | -0.000 |
chy | -0.000 |
bur | -0.000 |
rig | -0.000 |
Token | wanted |
Token ID | 2227 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
opsy | +0.069 |
oppy | +0.069 |
Remy | +0.062 |
chy | +0.061 |
bur | +0.060 |
Neg logit contributions | |
folk | -0.033 |
digits | -0.032 |
sale | -0.030 |
letters | -0.028 |
Amen | -0.027 |
Token | Kitty |
Token ID | 21168 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
folk | +0.160 |
digits | +0.157 |
sale | +0.150 |
customer | +0.138 |
letters | +0.136 |
Neg logit contributions | |
opsy | -0.227 |
oppy | -0.225 |
Remy | -0.204 |
chy | -0.201 |
bur | -0.200 |
Token | and |
Token ID | 290 |
Feature activation | +3.29e-02 |
Pos logit contributions | |
opsy | +0.553 |
oppy | +0.537 |
bur | +0.489 |
Remy | +0.481 |
rig | +0.476 |
Neg logit contributions | |
sale | -0.219 |
folk | -0.205 |
letters | -0.193 |
digits | -0.187 |
customer | -0.180 |
Token | Bunny |
Token ID | 28683 |
Feature activation | +4.08e-01 |
Pos logit contributions | |
opsy | +0.070 |
oppy | +0.068 |
chy | +0.061 |
rig | +0.061 |
bur | +0.059 |
Neg logit contributions | |
folk | -0.050 |
digits | -0.047 |
sale | -0.046 |
letters | -0.043 |
customer | -0.042 |
Token | to |
Token ID | 284 |
Feature activation | -1.63e-03 |
Pos logit contributions | |
opsy | +0.882 |
oppy | +0.866 |
Remy | +0.797 |
bur | +0.775 |
chy | +0.771 |
Neg logit contributions | |
folk | -0.614 |
digits | -0.595 |
sale | -0.564 |
letters | -0.519 |
customer | -0.516 |
Token | be |
Token ID | 307 |
Feature activation | -4.27e-02 |
Pos logit contributions | |
folk | +0.003 |
digits | +0.003 |
sale | +0.002 |
customer | +0.002 |
letters | +0.002 |
Neg logit contributions | |
opsy | -0.003 |
oppy | -0.003 |
Remy | -0.003 |
chy | -0.003 |
bur | -0.003 |
Token | friends |
Token ID | 2460 |
Feature activation | -6.16e-02 |
Pos logit contributions | |
folk | +0.097 |
digits | +0.091 |
sale | +0.082 |
person | +0.081 |
alphabet | +0.079 |
Neg logit contributions | |
opsy | -0.060 |
oppy | -0.060 |
chy | -0.057 |
bur | -0.056 |
Remy | -0.055 |
Token | again |
Token ID | 757 |
Feature activation | -2.98e-02 |
Pos logit contributions | |
folk | +0.120 |
digits | +0.119 |
sale | +0.105 |
letters | +0.101 |
jury | +0.100 |
Neg logit contributions | |
opsy | -0.107 |
oppy | -0.106 |
Remy | -0.097 |
chy | -0.097 |
bur | -0.096 |
Token | . |
Token ID | 13 |
Feature activation | +9.03e-04 |
Pos logit contributions | |
folk | +0.052 |
digits | +0.050 |
sale | +0.046 |
customer | +0.043 |
letters | +0.043 |
Neg logit contributions | |
opsy | -0.058 |
oppy | -0.058 |
Remy | -0.054 |
bur | -0.052 |
chy | -0.052 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | -5.74e-02 |
Pos logit contributions | |
folk | +0.013 |
digits | +0.012 |
sale | +0.011 |
customer | +0.011 |
letters | +0.010 |
Neg logit contributions | |
opsy | -0.017 |
oppy | -0.017 |
Remy | -0.016 |
bur | -0.015 |
chy | -0.015 |
Token | he |
Token ID | 339 |
Feature activation | +3.08e-02 |
Pos logit contributions | |
folk | +0.082 |
digits | +0.080 |
sale | +0.074 |
customer | +0.068 |
letters | +0.068 |
Neg logit contributions | |
opsy | -0.128 |
oppy | -0.126 |
Remy | -0.117 |
chy | -0.114 |
bur | -0.114 |
Token | saw |
Token ID | 2497 |
Feature activation | -4.85e-02 |
Pos logit contributions | |
opsy | +0.077 |
oppy | +0.075 |
Remy | +0.072 |
rig | +0.069 |
chy | +0.067 |
Neg logit contributions | |
folk | -0.034 |
digits | -0.033 |
sale | -0.029 |
letters | -0.027 |
cheering | -0.026 |
Token | the |
Token ID | 262 |
Feature activation | +2.39e-02 |
Pos logit contributions | |
folk | +0.081 |
digits | +0.080 |
sale | +0.073 |
customer | +0.069 |
Amen | +0.069 |
Neg logit contributions | |
opsy | -0.094 |
oppy | -0.093 |
Remy | -0.089 |
bur | -0.085 |
chy | -0.084 |
Token | book |
Token ID | 1492 |
Feature activation | -8.36e-02 |
Pos logit contributions | |
opsy | +0.052 |
oppy | +0.052 |
Remy | +0.048 |
bur | +0.046 |
chy | +0.045 |
Neg logit contributions | |
folk | -0.035 |
digits | -0.035 |
sale | -0.033 |
letters | -0.031 |
customer | -0.031 |
Token | tucked |
Token ID | 29779 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
folk | +0.135 |
digits | +0.131 |
sale | +0.126 |
letters | +0.115 |
customer | +0.114 |
Neg logit contributions | |
opsy | -0.174 |
oppy | -0.171 |
Remy | -0.158 |
bur | -0.150 |
chy | -0.146 |
Token | away |
Token ID | 1497 |
Feature activation | -2.50e-02 |
Pos logit contributions | |
opsy | +0.296 |
oppy | +0.293 |
Remy | +0.269 |
bur | +0.261 |
chy | +0.261 |
Neg logit contributions | |
folk | -0.226 |
digits | -0.217 |
sale | -0.205 |
letters | -0.191 |
customer | -0.190 |
Token | in |
Token ID | 287 |
Feature activation | +2.89e-02 |
Pos logit contributions | |
folk | +0.038 |
digits | +0.036 |
sale | +0.036 |
letters | +0.032 |
customer | +0.032 |
Neg logit contributions | |
opsy | -0.054 |
oppy | -0.053 |
Remy | -0.049 |
bur | -0.046 |
chy | -0.046 |
Token | a |
Token ID | 257 |
Feature activation | +5.47e-03 |
Pos logit contributions | |
bur | +0.050 |
opsy | +0.049 |
Remy | +0.048 |
oppy | +0.048 |
chy | +0.045 |
Neg logit contributions | |
folk | -0.055 |
digits | -0.052 |
Amen | -0.045 |
jury | -0.044 |
salesman | -0.044 |
Token | corner |
Token ID | 5228 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
opsy | +0.010 |
oppy | +0.010 |
Remy | +0.009 |
chy | +0.009 |
bur | +0.009 |
Neg logit contributions | |
folk | -0.010 |
digits | -0.009 |
sale | -0.009 |
customer | -0.008 |
letters | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -6.55e-03 |
Pos logit contributions | |
folk | +0.280 |
digits | +0.275 |
sale | +0.251 |
Amen | +0.241 |
customer | +0.240 |
Neg logit contributions | |
opsy | -0.302 |
oppy | -0.300 |
Remy | -0.279 |
chy | -0.276 |
bur | -0.275 |
Token | tape |
Token ID | 9154 |
Feature activation | -1.00e-01 |
Pos logit contributions | |
opsy | +0.038 |
oppy | +0.037 |
Remy | +0.036 |
rig | +0.033 |
chy | +0.033 |
Neg logit contributions | |
letters | -0.020 |
sale | -0.019 |
digits | -0.019 |
presents | -0.019 |
folk | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | +2.93e-02 |
Pos logit contributions | |
folk | +0.157 |
digits | +0.153 |
sale | +0.147 |
letters | +0.138 |
customer | +0.136 |
Neg logit contributions | |
opsy | -0.209 |
oppy | -0.205 |
Remy | -0.190 |
chy | -0.182 |
bur | -0.181 |
Token | go |
Token ID | 467 |
Feature activation | +7.29e-02 |
Pos logit contributions | |
opsy | +0.064 |
oppy | +0.063 |
Remy | +0.058 |
chy | +0.056 |
bur | +0.056 |
Neg logit contributions | |
folk | -0.043 |
digits | -0.043 |
sale | -0.039 |
letters | -0.038 |
customer | -0.037 |
Token | for |
Token ID | 329 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
oppy | +0.129 |
opsy | +0.127 |
Remy | +0.119 |
bur | +0.117 |
chy | +0.115 |
Neg logit contributions | |
folk | -0.141 |
digits | -0.132 |
sale | -0.124 |
jury | -0.117 |
customer | -0.117 |
Token | a |
Token ID | 257 |
Feature activation | +8.97e-03 |
Pos logit contributions | |
opsy | +0.216 |
oppy | +0.215 |
Remy | +0.205 |
bur | +0.196 |
chy | +0.193 |
Neg logit contributions | |
folk | -0.219 |
digits | -0.205 |
sale | -0.189 |
customer | -0.182 |
jury | -0.179 |
Token | walk |
Token ID | 2513 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
opsy | +0.015 |
oppy | +0.015 |
Remy | +0.014 |
chy | +0.013 |
bur | +0.013 |
Neg logit contributions | |
digits | -0.017 |
folk | -0.017 |
sale | -0.016 |
customer | -0.016 |
letters | -0.015 |
Token | in |
Token ID | 287 |
Feature activation | +3.02e-02 |
Pos logit contributions | |
opsy | +0.356 |
oppy | +0.353 |
Remy | +0.326 |
chy | +0.326 |
bur | +0.316 |
Neg logit contributions | |
folk | -0.353 |
digits | -0.351 |
sale | -0.315 |
customer | -0.301 |
jury | -0.300 |
Token | the |
Token ID | 262 |
Feature activation | +3.42e-02 |
Pos logit contributions | |
opsy | +0.056 |
oppy | +0.055 |
Remy | +0.052 |
bur | +0.050 |
chy | +0.049 |
Neg logit contributions | |
folk | -0.055 |
digits | -0.053 |
sale | -0.049 |
customer | -0.046 |
epad | -0.046 |
Token | park |
Token ID | 3952 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
oppy | +0.085 |
opsy | +0.083 |
Remy | +0.076 |
chy | +0.075 |
uffy | +0.072 |
Neg logit contributions | |
digits | -0.040 |
sale | -0.037 |
folk | -0.037 |
letters | -0.035 |
customer | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -1.61e-03 |
Pos logit contributions | |
folk | +0.021 |
digits | +0.021 |
sale | +0.018 |
Amen | +0.018 |
jury | +0.018 |
Neg logit contributions | |
opsy | -0.018 |
oppy | -0.018 |
chy | -0.017 |
Remy | -0.016 |
bur | -0.016 |
Token | One |
Token ID | 1881 |
Feature activation | -8.40e-02 |
Pos logit contributions | |
folk | +0.003 |
digits | +0.003 |
sale | +0.002 |
customer | +0.002 |
letters | +0.002 |
Neg logit contributions | |
opsy | -0.003 |
oppy | -0.003 |
Remy | -0.003 |
chy | -0.003 |
bur | -0.003 |
Token | also |
Token ID | 635 |
Feature activation | +3.96e-02 |
Pos logit contributions | |
opsy | +0.176 |
oppy | +0.170 |
Remy | +0.157 |
bur | +0.153 |
chy | +0.152 |
Neg logit contributions | |
folk | -0.111 |
digits | -0.106 |
sale | -0.102 |
letters | -0.094 |
Amen | -0.092 |
Token | made |
Token ID | 925 |
Feature activation | -7.77e-02 |
Pos logit contributions | |
opsy | +0.088 |
oppy | +0.087 |
Remy | +0.081 |
bur | +0.078 |
chy | +0.077 |
Neg logit contributions | |
folk | -0.055 |
digits | -0.054 |
sale | -0.051 |
letters | -0.047 |
customer | -0.047 |
Token | Angel |
Token ID | 3905 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
digits | +0.099 |
sale | +0.096 |
letters | +0.092 |
folk | +0.092 |
customer | +0.085 |
Neg logit contributions | |
opsy | -0.181 |
oppy | -0.178 |
chy | -0.161 |
Remy | -0.161 |
bur | -0.158 |
Token | wet |
Token ID | 9583 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
opsy | +0.276 |
oppy | +0.273 |
Remy | +0.251 |
chy | +0.246 |
bur | +0.243 |
Neg logit contributions | |
folk | -0.224 |
digits | -0.219 |
sale | -0.204 |
customer | -0.190 |
letters | -0.187 |
Token | and |
Token ID | 290 |
Feature activation | +3.61e-02 |
Pos logit contributions | |
opsy | +0.331 |
oppy | +0.327 |
Remy | +0.303 |
chy | +0.294 |
bur | +0.293 |
Neg logit contributions | |
folk | -0.259 |
digits | -0.252 |
sale | -0.235 |
customer | -0.219 |
letters | -0.216 |
Token | dirty |
Token ID | 11841 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
opsy | +0.067 |
oppy | +0.065 |
Remy | +0.059 |
rig | +0.055 |
bur | +0.055 |
Neg logit contributions | |
folk | -0.062 |
digits | -0.061 |
sale | -0.058 |
letters | -0.058 |
cheering | -0.056 |
Token | . |
Token ID | 13 |
Feature activation | +1.82e-03 |
Pos logit contributions | |
opsy | +0.203 |
oppy | +0.200 |
Remy | +0.185 |
chy | +0.184 |
bur | +0.182 |
Neg logit contributions | |
folk | -0.163 |
digits | -0.162 |
sale | -0.147 |
customer | -0.138 |
epad | -0.136 |
Token | She |
Token ID | 1375 |
Feature activation | +3.10e-02 |
Pos logit contributions | |
opsy | +0.004 |
oppy | +0.004 |
Remy | +0.003 |
chy | +0.003 |
bur | +0.003 |
Neg logit contributions | |
folk | -0.003 |
digits | -0.003 |
sale | -0.003 |
letters | -0.002 |
Amen | -0.002 |
Token | could |
Token ID | 714 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
opsy | +0.072 |
oppy | +0.070 |
Remy | +0.066 |
rig | +0.062 |
bur | +0.061 |
Neg logit contributions | |
folk | -0.037 |
digits | -0.036 |
letters | -0.036 |
Amen | -0.033 |
sale | -0.033 |
Token | have |
Token ID | 423 |
Feature activation | -8.01e-02 |
Pos logit contributions | |
opsy | +0.323 |
oppy | +0.322 |
Remy | +0.293 |
rig | +0.289 |
bur | +0.284 |
Neg logit contributions | |
sale | -0.191 |
folk | -0.191 |
digits | -0.188 |
Amen | -0.175 |
customer | -0.173 |
Token | been |
Token ID | 587 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
folk | +0.146 |
digits | +0.143 |
sale | +0.135 |
customer | +0.127 |
letters | +0.126 |
Neg logit contributions | |
opsy | -0.146 |
oppy | -0.145 |
Remy | -0.132 |
chy | -0.127 |
bur | -0.127 |
Token | : |
Token ID | 25 |
Feature activation | -6.98e-03 |
Pos logit contributions | |
folk | +0.020 |
digits | +0.019 |
sale | +0.017 |
bench | +0.017 |
customer | +0.017 |
Neg logit contributions | |
oppy | -0.020 |
opsy | -0.020 |
Remy | -0.018 |
chy | -0.018 |
bur | -0.017 |
Token | " |
Token ID | 366 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
folk | +0.011 |
digits | +0.010 |
sale | +0.009 |
bench | +0.009 |
letters | +0.009 |
Neg logit contributions | |
opsy | -0.015 |
oppy | -0.015 |
Remy | -0.014 |
bur | -0.013 |
chy | -0.013 |
Token | Hello |
Token ID | 15496 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
folk | +0.202 |
digits | +0.199 |
sale | +0.194 |
customer | +0.176 |
letters | +0.173 |
Neg logit contributions | |
opsy | -0.258 |
oppy | -0.245 |
Remy | -0.232 |
chy | -0.226 |
bur | -0.222 |
Token | Ab |
Token ID | 2275 |
Feature activation | +6.88e-02 |
Pos logit contributions | |
folk | +0.261 |
digits | +0.252 |
sale | +0.236 |
bench | +0.223 |
customer | +0.220 |
Neg logit contributions | |
opsy | -0.317 |
oppy | -0.314 |
Remy | -0.287 |
bur | -0.278 |
chy | -0.274 |
Token | ig |
Token ID | 328 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
opsy | +0.079 |
Remy | +0.078 |
oppy | +0.076 |
chy | +0.073 |
bur | +0.068 |
Neg logit contributions | |
digits | -0.162 |
folk | -0.158 |
sale | -0.148 |
customer | -0.144 |
letters | -0.143 |
Token | ail |
Token ID | 603 |
Feature activation | +2.59e-01 |
Pos logit contributions | |
opsy | +0.466 |
oppy | +0.457 |
Remy | +0.424 |
bur | +0.411 |
chy | +0.407 |
Neg logit contributions | |
folk | -0.494 |
digits | -0.482 |
sale | -0.455 |
customer | -0.430 |
letters | -0.424 |
Token | ! |
Token ID | 0 |
Feature activation | +9.99e-02 |
Pos logit contributions | |
opsy | +0.484 |
oppy | +0.483 |
Remy | +0.437 |
chy | +0.430 |
bur | +0.419 |
Neg logit contributions | |
folk | -0.454 |
digits | -0.449 |
sale | -0.413 |
customer | -0.388 |
epad | -0.386 |
Token | I |
Token ID | 314 |
Feature activation | +7.35e-03 |
Pos logit contributions | |
oppy | +0.209 |
opsy | +0.208 |
Remy | +0.194 |
bur | +0.183 |
chy | +0.183 |
Neg logit contributions | |
folk | -0.154 |
digits | -0.152 |
sale | -0.143 |
letters | -0.132 |
bench | -0.131 |
Token | 'm |
Token ID | 1101 |
Feature activation | +6.07e-02 |
Pos logit contributions | |
opsy | +0.014 |
oppy | +0.014 |
Remy | +0.013 |
bur | +0.013 |
chy | +0.013 |
Neg logit contributions | |
folk | -0.012 |
digits | -0.012 |
sale | -0.011 |
customer | -0.010 |
cheering | -0.010 |
Token | glad |
Token ID | 9675 |
Feature activation | -6.19e-03 |
Pos logit contributions | |
opsy | +0.136 |
oppy | +0.135 |
Remy | +0.126 |
chy | +0.123 |
bur | +0.122 |
Neg logit contributions | |
folk | -0.086 |
digits | -0.083 |
sale | -0.077 |
customer | -0.071 |
letters | -0.070 |
Token | you |
Token ID | 345 |
Feature activation | +8.05e-02 |
Pos logit contributions | |
folk | +0.010 |
digits | +0.010 |
sale | +0.009 |
customer | +0.009 |
letters | +0.009 |
Neg logit contributions | |
opsy | -0.012 |
oppy | -0.012 |
Remy | -0.011 |
chy | -0.011 |
bur | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | -3.46e-02 |
Pos logit contributions | |
folk | +0.002 |
digits | +0.002 |
sale | +0.002 |
customer | +0.002 |
letters | +0.002 |
Neg logit contributions | |
opsy | -0.004 |
oppy | -0.004 |
Remy | -0.003 |
chy | -0.003 |
bur | -0.003 |
Token | Annie |
Token ID | 25860 |
Feature activation | -6.59e-02 |
Pos logit contributions | |
folk | +0.048 |
digits | +0.046 |
sale | +0.044 |
customer | +0.040 |
letters | +0.039 |
Neg logit contributions | |
opsy | -0.077 |
oppy | -0.076 |
bur | -0.070 |
Remy | -0.070 |
chy | -0.070 |
Token | was |
Token ID | 373 |
Feature activation | +5.70e-02 |
Pos logit contributions | |
folk | +0.081 |
digits | +0.079 |
sale | +0.073 |
letters | +0.068 |
customer | +0.066 |
Neg logit contributions | |
opsy | -0.152 |
oppy | -0.149 |
Remy | -0.144 |
chy | -0.137 |
rig | -0.135 |
Token | feeling |
Token ID | 4203 |
Feature activation | -5.79e-02 |
Pos logit contributions | |
opsy | +0.131 |
oppy | +0.130 |
Remy | +0.123 |
chy | +0.118 |
bur | +0.116 |
Neg logit contributions | |
folk | -0.074 |
digits | -0.073 |
sale | -0.069 |
customer | -0.064 |
cheering | -0.062 |
Token | very |
Token ID | 845 |
Feature activation | -3.44e-02 |
Pos logit contributions | |
folk | +0.072 |
digits | +0.067 |
sale | +0.067 |
customer | +0.065 |
cheering | +0.063 |
Neg logit contributions | |
opsy | -0.131 |
oppy | -0.128 |
Remy | -0.126 |
uffy | -0.120 |
rig | -0.117 |
Token | brave |
Token ID | 14802 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
digits | +0.045 |
folk | +0.043 |
sale | +0.042 |
cheering | +0.040 |
customer | +0.040 |
Neg logit contributions | |
Remy | -0.070 |
opsy | -0.070 |
oppy | -0.067 |
ches | -0.064 |
uffy | -0.064 |
Token | . |
Token ID | 13 |
Feature activation | +9.44e-04 |
Pos logit contributions | |
opsy | +0.198 |
oppy | +0.196 |
Remy | +0.182 |
chy | +0.178 |
bur | +0.178 |
Neg logit contributions | |
folk | -0.187 |
digits | -0.185 |
sale | -0.171 |
customer | -0.160 |
jury | -0.158 |
Token | She |
Token ID | 1375 |
Feature activation | +2.81e-02 |
Pos logit contributions | |
opsy | +0.002 |
oppy | +0.002 |
Remy | +0.002 |
chy | +0.002 |
bur | +0.002 |
Neg logit contributions | |
folk | -0.002 |
digits | -0.002 |
sale | -0.002 |
jury | -0.001 |
customer | -0.001 |
Token | decided |
Token ID | 3066 |
Feature activation | -8.81e-02 |
Pos logit contributions | |
opsy | +0.066 |
oppy | +0.064 |
Remy | +0.061 |
bur | +0.057 |
chy | +0.057 |
Neg logit contributions | |
digits | -0.036 |
folk | -0.036 |
sale | -0.033 |
letters | -0.030 |
customer | -0.029 |
Token | to |
Token ID | 284 |
Feature activation | +1.93e-03 |
Pos logit contributions | |
folk | +0.144 |
digits | +0.140 |
sale | +0.130 |
customer | +0.122 |
letters | +0.120 |
Neg logit contributions | |
opsy | -0.178 |
oppy | -0.176 |
Remy | -0.163 |
chy | -0.159 |
bur | -0.158 |
Token | escape |
Token ID | 6654 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
opsy | +0.004 |
oppy | +0.003 |
Remy | +0.003 |
bur | +0.003 |
chy | +0.003 |
Neg logit contributions | |
folk | -0.004 |
digits | -0.003 |
sale | -0.003 |
customer | -0.003 |
letters | -0.003 |
Token | One |
Token ID | 1881 |
Feature activation | -8.60e-02 |
Pos logit contributions | |
folk | +0.002 |
digits | +0.002 |
sale | +0.002 |
letters | +0.002 |
epad | +0.002 |
Neg logit contributions | |
opsy | -0.002 |
oppy | -0.002 |
Remy | -0.002 |
bur | -0.002 |
chy | -0.002 |
Token | day |
Token ID | 1110 |
Feature activation | -1.18e-03 |
Pos logit contributions | |
sale | +0.069 |
customer | +0.069 |
folk | +0.068 |
digits | +0.067 |
letters | +0.064 |
Neg logit contributions | |
opsy | -0.229 |
oppy | -0.221 |
Remy | -0.219 |
bur | -0.209 |
rig | -0.203 |
Token | , |
Token ID | 11 |
Feature activation | -3.52e-02 |
Pos logit contributions | |
folk | +0.002 |
digits | +0.002 |
sale | +0.002 |
customer | +0.002 |
letters | +0.002 |
Neg logit contributions | |
oppy | -0.003 |
opsy | -0.002 |
Remy | -0.002 |
chy | -0.002 |
bur | -0.002 |
Token | she |
Token ID | 673 |
Feature activation | +7.93e-02 |
Pos logit contributions | |
folk | +0.063 |
digits | +0.062 |
sale | +0.058 |
customer | +0.054 |
letters | +0.054 |
Neg logit contributions | |
opsy | -0.066 |
oppy | -0.065 |
Remy | -0.059 |
chy | -0.058 |
bur | -0.057 |
Token | found |
Token ID | 1043 |
Feature activation | -5.37e-02 |
Pos logit contributions | |
opsy | +0.183 |
oppy | +0.179 |
Remy | +0.170 |
chy | +0.163 |
bur | +0.160 |
Neg logit contributions | |
folk | -0.105 |
digits | -0.103 |
sale | -0.095 |
customer | -0.087 |
letters | -0.086 |
Token | a |
Token ID | 257 |
Feature activation | +9.43e-03 |
Pos logit contributions | |
folk | +0.064 |
digits | +0.063 |
sale | +0.061 |
letters | +0.059 |
customer | +0.055 |
Neg logit contributions | |
opsy | -0.129 |
oppy | -0.128 |
rig | -0.118 |
Remy | -0.116 |
chy | -0.115 |
Token | veil |
Token ID | 30615 |
Feature activation | +8.10e-02 |
Pos logit contributions | |
opsy | +0.018 |
oppy | +0.018 |
Remy | +0.017 |
chy | +0.016 |
bur | +0.016 |
Neg logit contributions | |
folk | -0.016 |
digits | -0.016 |
sale | -0.015 |
customer | -0.014 |
letters | -0.014 |
Token | in |
Token ID | 287 |
Feature activation | +3.20e-02 |
Pos logit contributions | |
opsy | +0.175 |
oppy | +0.174 |
Remy | +0.160 |
bur | +0.152 |
chy | +0.146 |
Neg logit contributions | |
folk | -0.123 |
sale | -0.117 |
digits | -0.116 |
letters | -0.106 |
customer | -0.106 |
Token | her |
Token ID | 607 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
opsy | +0.053 |
oppy | +0.051 |
Remy | +0.050 |
bur | +0.049 |
chy | +0.046 |
Neg logit contributions | |
folk | -0.064 |
digits | -0.060 |
Amen | -0.053 |
jury | -0.053 |
winners | -0.053 |
Token | sack |
Token ID | 23704 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
opsy | +0.035 |
oppy | +0.034 |
Remy | +0.034 |
chy | +0.033 |
rig | +0.033 |
Neg logit contributions | |
digits | -0.021 |
folk | -0.019 |
customer | -0.019 |
letters | -0.019 |
sale | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -5.72e-05 |
Pos logit contributions | |
opsy | +0.270 |
oppy | +0.267 |
Remy | +0.247 |
chy | +0.239 |
bur | +0.238 |
Neg logit contributions | |
folk | -0.207 |
digits | -0.201 |
sale | -0.189 |
customer | -0.175 |
letters | -0.174 |
Token | , |
Token ID | 11 |
Feature activation | -3.06e-02 |
Pos logit contributions | |
folk | +0.034 |
digits | +0.033 |
sale | +0.031 |
customer | +0.027 |
cheering | +0.027 |
Neg logit contributions | |
opsy | -0.068 |
oppy | -0.066 |
chy | -0.062 |
Remy | -0.062 |
bur | -0.062 |
Token | there |
Token ID | 612 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
folk | +0.025 |
digits | +0.024 |
sale | +0.022 |
customer | +0.019 |
Amen | +0.018 |
Neg logit contributions | |
opsy | -0.083 |
oppy | -0.082 |
bur | -0.078 |
Remy | -0.078 |
chy | -0.076 |
Token | was |
Token ID | 373 |
Feature activation | +6.02e-02 |
Pos logit contributions | |
folk | +0.141 |
digits | +0.140 |
sale | +0.137 |
cheering | +0.118 |
number | +0.118 |
Neg logit contributions | |
opsy | -0.270 |
oppy | -0.267 |
Remy | -0.246 |
rig | -0.243 |
bur | -0.243 |
Token | a |
Token ID | 257 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
oppy | +0.151 |
opsy | +0.149 |
chy | +0.136 |
bur | +0.135 |
uffy | +0.134 |
Neg logit contributions | |
digits | -0.063 |
sale | -0.063 |
folk | -0.060 |
customer | -0.057 |
letters | -0.056 |
Token | little |
Token ID | 1310 |
Feature activation | +7.27e-02 |
Pos logit contributions | |
opsy | +0.036 |
oppy | +0.036 |
Remy | +0.033 |
chy | +0.033 |
bur | +0.032 |
Neg logit contributions | |
digits | -0.026 |
folk | -0.025 |
customer | -0.024 |
sale | -0.024 |
Amen | -0.022 |
Token | boy |
Token ID | 2933 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
opsy | +0.177 |
oppy | +0.175 |
chy | +0.164 |
Remy | +0.164 |
bur | +0.158 |
Neg logit contributions | |
folk | -0.083 |
digits | -0.079 |
sale | -0.073 |
customer | -0.072 |
letters | -0.067 |
Token | named |
Token ID | 3706 |
Feature activation | -6.44e-04 |
Pos logit contributions | |
folk | +0.179 |
digits | +0.175 |
sale | +0.164 |
customer | +0.152 |
letters | +0.150 |
Neg logit contributions | |
opsy | -0.244 |
oppy | -0.241 |
Remy | -0.223 |
chy | -0.218 |
bur | -0.217 |
Token | John |
Token ID | 1757 |
Feature activation | -1.88e-01 |
Pos logit contributions | |
digits | +0.001 |
folk | +0.001 |
sale | +0.001 |
customer | +0.001 |
letters | +0.001 |
Neg logit contributions | |
opsy | -0.001 |
oppy | -0.001 |
bur | -0.001 |
chy | -0.001 |
Remy | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +5.38e-03 |
Pos logit contributions | |
folk | +0.326 |
digits | +0.313 |
sale | +0.294 |
customer | +0.272 |
letters | +0.269 |
Neg logit contributions | |
opsy | -0.367 |
oppy | -0.364 |
Remy | -0.334 |
chy | -0.330 |
bur | -0.323 |
Token | He |
Token ID | 679 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
Remy | +0.009 |
chy | +0.008 |
opsy | +0.007 |
bur | +0.007 |
oppy | +0.007 |
Neg logit contributions | |
folk | -0.010 |
jury | -0.009 |
winners | -0.009 |
judges | -0.009 |
iers | -0.009 |
Token | loved |
Token ID | 6151 |
Feature activation | +4.06e-02 |
Pos logit contributions | |
digits | +0.011 |
sale | +0.010 |
folk | +0.010 |
Amen | +0.010 |
number | +0.010 |
Neg logit contributions | |
opsy | -0.031 |
oppy | -0.030 |
Remy | -0.028 |
rig | -0.028 |
bur | -0.028 |
Token | a |
Token ID | 257 |
Feature activation | -5.18e-03 |
Pos logit contributions | |
opsy | +0.355 |
oppy | +0.353 |
Remy | +0.321 |
bur | +0.318 |
chy | +0.314 |
Neg logit contributions | |
folk | -0.208 |
digits | -0.207 |
sale | -0.194 |
customer | -0.179 |
letters | -0.178 |
Token | time |
Token ID | 640 |
Feature activation | -4.68e-02 |
Pos logit contributions | |
digits | +0.004 |
customer | +0.004 |
sale | +0.004 |
folk | +0.004 |
John | +0.003 |
Neg logit contributions | |
opsy | -0.013 |
oppy | -0.013 |
Remy | -0.013 |
bur | -0.013 |
chy | -0.012 |
Token | , |
Token ID | 11 |
Feature activation | -5.49e-02 |
Pos logit contributions | |
folk | +0.057 |
digits | +0.055 |
sale | +0.053 |
customer | +0.046 |
cheering | +0.045 |
Neg logit contributions | |
opsy | -0.112 |
oppy | -0.109 |
Remy | -0.103 |
bur | -0.102 |
chy | -0.102 |
Token | there |
Token ID | 612 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
folk | +0.065 |
digits | +0.063 |
sale | +0.057 |
customer | +0.053 |
letters | +0.051 |
Neg logit contributions | |
opsy | -0.133 |
oppy | -0.132 |
Remy | -0.124 |
bur | -0.122 |
chy | -0.121 |
Token | was |
Token ID | 373 |
Feature activation | +3.60e-02 |
Pos logit contributions | |
folk | +0.144 |
digits | +0.144 |
sale | +0.142 |
cheering | +0.123 |
number | +0.122 |
Neg logit contributions | |
opsy | -0.327 |
oppy | -0.326 |
rig | -0.301 |
bur | -0.300 |
Remy | -0.300 |
Token | a |
Token ID | 257 |
Feature activation | -6.55e-03 |
Pos logit contributions | |
oppy | +0.089 |
opsy | +0.088 |
chy | +0.081 |
bur | +0.080 |
rig | +0.079 |
Neg logit contributions | |
sale | -0.038 |
digits | -0.038 |
folk | -0.037 |
customer | -0.035 |
letters | -0.033 |
Token | girl |
Token ID | 2576 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
digits | +0.009 |
folk | +0.009 |
customer | +0.009 |
sale | +0.009 |
Amen | +0.008 |
Neg logit contributions | |
oppy | -0.013 |
opsy | -0.013 |
Remy | -0.012 |
chy | -0.012 |
bur | -0.012 |
Token | called |
Token ID | 1444 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
folk | +0.211 |
digits | +0.206 |
sale | +0.192 |
customer | +0.179 |
letters | +0.177 |
Neg logit contributions | |
opsy | -0.282 |
oppy | -0.279 |
Remy | -0.259 |
chy | -0.252 |
bur | -0.251 |
Token | Jill |
Token ID | 22772 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
opsy | +0.048 |
oppy | +0.047 |
Remy | +0.044 |
chy | +0.043 |
bur | +0.043 |
Neg logit contributions | |
folk | -0.039 |
digits | -0.038 |
sale | -0.037 |
customer | -0.034 |
letters | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
folk | +0.184 |
digits | +0.180 |
sale | +0.172 |
customer | +0.158 |
letters | +0.153 |
Neg logit contributions | |
opsy | -0.418 |
oppy | -0.416 |
Remy | -0.399 |
bur | -0.389 |
rig | -0.365 |
Token | She |
Token ID | 1375 |
Feature activation | +8.63e-03 |
Pos logit contributions | |
folk | +0.033 |
digits | +0.029 |
Amen | +0.029 |
sale | +0.028 |
customer | +0.028 |
Neg logit contributions | |
oppy | -0.027 |
opsy | -0.026 |
chy | -0.025 |
Remy | -0.025 |
bur | -0.024 |
Token | feels |
Token ID | 5300 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
digits | +0.031 |
folk | +0.027 |
letters | +0.026 |
number | +0.026 |
prices | +0.025 |
Neg logit contributions | |
opsy | -0.053 |
oppy | -0.052 |
Remy | -0.046 |
uffy | -0.044 |
chy | -0.044 |
Token | the |
Token ID | 262 |
Feature activation | +3.51e-02 |
Pos logit contributions | |
opsy | +0.283 |
oppy | +0.279 |
bur | +0.250 |
Remy | +0.250 |
chy | +0.246 |
Neg logit contributions | |
folk | -0.178 |
digits | -0.174 |
sale | -0.156 |
cheering | -0.153 |
letters | -0.153 |
Token | wind |
Token ID | 2344 |
Feature activation | -6.57e-02 |
Pos logit contributions | |
opsy | +0.073 |
oppy | +0.072 |
Remy | +0.067 |
chy | +0.065 |
bur | +0.065 |
Neg logit contributions | |
folk | -0.055 |
digits | -0.054 |
sale | -0.050 |
customer | -0.047 |
letters | -0.046 |
Token | in |
Token ID | 287 |
Feature activation | +3.93e-02 |
Pos logit contributions | |
folk | +0.102 |
digits | +0.098 |
sale | +0.092 |
customer | +0.085 |
letters | +0.085 |
Neg logit contributions | |
opsy | -0.139 |
oppy | -0.138 |
Remy | -0.128 |
bur | -0.124 |
chy | -0.121 |
Token | his |
Token ID | 465 |
Feature activation | -4.15e-02 |
Pos logit contributions | |
opsy | +0.078 |
oppy | +0.076 |
Remy | +0.075 |
bur | +0.071 |
chy | +0.069 |
Neg logit contributions | |
folk | -0.066 |
digits | -0.063 |
sale | -0.057 |
epad | -0.055 |
Amen | -0.054 |
Token | hair |
Token ID | 4190 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
folk | +0.050 |
digits | +0.049 |
letters | +0.044 |
sale | +0.044 |
customer | +0.041 |
Neg logit contributions | |
opsy | -0.102 |
oppy | -0.100 |
Remy | -0.094 |
chy | -0.092 |
bur | -0.090 |
Token | . |
Token ID | 13 |
Feature activation | -2.49e-03 |
Pos logit contributions | |
folk | +0.228 |
digits | +0.222 |
sale | +0.209 |
letters | +0.192 |
customer | +0.191 |
Neg logit contributions | |
opsy | -0.326 |
oppy | -0.323 |
Remy | -0.299 |
bur | -0.288 |
chy | -0.282 |
Token | He |
Token ID | 679 |
Feature activation | -2.11e-02 |
Pos logit contributions | |
folk | +0.003 |
digits | +0.003 |
sale | +0.003 |
customer | +0.003 |
letters | +0.003 |
Neg logit contributions | |
opsy | -0.006 |
oppy | -0.006 |
chy | -0.005 |
bur | -0.005 |
Remy | -0.005 |
Token | sees |
Token ID | 7224 |
Feature activation | +3.55e-02 |
Pos logit contributions | |
digits | +0.029 |
folk | +0.026 |
letters | +0.025 |
number | +0.025 |
prices | +0.024 |
Neg logit contributions | |
opsy | -0.051 |
oppy | -0.049 |
Remy | -0.044 |
uffy | -0.042 |
bur | -0.041 |
Token | the |
Token ID | 262 |
Feature activation | +3.67e-02 |
Pos logit contributions | |
opsy | +0.071 |
oppy | +0.070 |
Remy | +0.065 |
chy | +0.063 |
bur | +0.063 |
Neg logit contributions | |
folk | -0.059 |
digits | -0.057 |
sale | -0.053 |
customer | -0.050 |
letters | -0.049 |
Token | trees |
Token ID | 7150 |
Feature activation | +3.83e-02 |
Pos logit contributions | |
opsy | +0.061 |
oppy | +0.058 |
bur | +0.057 |
chy | +0.057 |
Remy | +0.056 |
Neg logit contributions | |
folk | -0.070 |
digits | -0.069 |
sale | -0.063 |
Amen | -0.061 |
epad | -0.061 |
Token | could |
Token ID | 714 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
folk | +0.202 |
digits | +0.195 |
sale | +0.186 |
letters | +0.174 |
customer | +0.171 |
Neg logit contributions | |
opsy | -0.226 |
oppy | -0.221 |
Remy | -0.204 |
bur | -0.197 |
chy | -0.193 |
Token | help |
Token ID | 1037 |
Feature activation | -8.66e-03 |
Pos logit contributions | |
opsy | +0.276 |
oppy | +0.272 |
Remy | +0.251 |
bur | +0.242 |
chy | +0.240 |
Neg logit contributions | |
folk | -0.216 |
digits | -0.212 |
sale | -0.202 |
customer | -0.185 |
letters | -0.185 |
Token | him |
Token ID | 683 |
Feature activation | -1.96e-03 |
Pos logit contributions | |
folk | +0.014 |
digits | +0.013 |
sale | +0.012 |
customer | +0.012 |
letters | +0.011 |
Neg logit contributions | |
opsy | -0.018 |
oppy | -0.018 |
Remy | -0.017 |
chy | -0.016 |
bur | -0.016 |
Token | be |
Token ID | 307 |
Feature activation | -5.08e-02 |
Pos logit contributions | |
folk | +0.003 |
digits | +0.003 |
sale | +0.003 |
customer | +0.003 |
letters | +0.003 |
Neg logit contributions | |
opsy | -0.004 |
oppy | -0.004 |
Remy | -0.004 |
chy | -0.003 |
bur | -0.003 |
Token | free |
Token ID | 1479 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
folk | +0.110 |
digits | +0.104 |
sale | +0.099 |
jury | +0.094 |
customer | +0.094 |
Neg logit contributions | |
oppy | -0.074 |
opsy | -0.074 |
Remy | -0.068 |
bur | -0.068 |
chy | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
opsy | +0.241 |
oppy | +0.240 |
Remy | +0.222 |
chy | +0.219 |
bur | +0.217 |
Neg logit contributions | |
folk | -0.213 |
digits | -0.209 |
sale | -0.191 |
customer | -0.181 |
jury | -0.177 |
Token | He |
Token ID | 679 |
Feature activation | -3.17e-02 |
Pos logit contributions | |
folk | +0.030 |
digits | +0.029 |
sale | +0.027 |
customer | +0.025 |
letters | +0.025 |
Neg logit contributions | |
opsy | -0.039 |
oppy | -0.038 |
Remy | -0.036 |
bur | -0.035 |
chy | -0.034 |
Token | was |
Token ID | 373 |
Feature activation | +4.26e-02 |
Pos logit contributions | |
folk | +0.045 |
digits | +0.044 |
sale | +0.041 |
customer | +0.038 |
letters | +0.038 |
Neg logit contributions | |
opsy | -0.071 |
oppy | -0.070 |
Remy | -0.065 |
chy | -0.063 |
bur | -0.062 |
Token | so |
Token ID | 523 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
opsy | +0.068 |
Remy | +0.064 |
chy | +0.064 |
oppy | +0.063 |
bur | +0.063 |
Neg logit contributions | |
folk | -0.085 |
digits | -0.081 |
sale | -0.075 |
jury | -0.074 |
boxes | -0.072 |
Token | happy |
Token ID | 3772 |
Feature activation | -8.08e-02 |
Pos logit contributions | |
folk | +0.180 |
digits | +0.173 |
sale | +0.164 |
customer | +0.150 |
prices | +0.147 |
Neg logit contributions | |
opsy | -0.225 |
oppy | -0.223 |
Remy | -0.206 |
chy | -0.203 |
bur | -0.201 |
Token | . |
Token ID | 13 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
folk | +0.133 |
digits | +0.130 |
sale | +0.124 |
customer | +0.114 |
letters | +0.114 |
Neg logit contributions | |
opsy | -0.163 |
oppy | -0.161 |
Remy | -0.146 |
chy | -0.144 |
bur | -0.142 |
Token | again |
Token ID | 757 |
Feature activation | -3.42e-02 |
Pos logit contributions | |
opsy | +0.236 |
oppy | +0.233 |
Remy | +0.210 |
uffy | +0.203 |
bur | +0.201 |
Neg logit contributions | |
folk | -0.137 |
digits | -0.128 |
letters | -0.123 |
sale | -0.118 |
cheering | -0.111 |
Token | . |
Token ID | 13 |
Feature activation | -5.56e-03 |
Pos logit contributions | |
folk | +0.054 |
digits | +0.052 |
sale | +0.049 |
customer | +0.045 |
letters | +0.045 |
Neg logit contributions | |
opsy | -0.071 |
oppy | -0.070 |
Remy | -0.065 |
chy | -0.063 |
bur | -0.063 |
Token | |
Token ID | 220 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
folk | +0.009 |
digits | +0.009 |
sale | +0.008 |
customer | +0.008 |
winners | +0.008 |
Neg logit contributions | |
oppy | -0.011 |
opsy | -0.011 |
Remy | -0.010 |
bur | -0.010 |
chy | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
folk | +0.295 |
digits | +0.290 |
sale | +0.268 |
customer | +0.239 |
letters | +0.238 |
Neg logit contributions | |
opsy | -0.505 |
oppy | -0.487 |
chy | -0.463 |
bur | -0.460 |
Remy | -0.452 |
Token |
|
Token ID | 198 |
Feature activation | +1.80e-02 |
Pos logit contributions | |
oppy | +0.031 |
opsy | +0.031 |
Remy | +0.030 |
bur | +0.027 |
chy | +0.026 |
Neg logit contributions | |
folk | -0.038 |
digits | -0.036 |
sale | -0.034 |
prices | -0.033 |
letters | -0.033 |
Token | Sarah |
Token ID | 29284 |
Feature activation | -1.93e-01 |
Pos logit contributions | |
opsy | +0.032 |
oppy | +0.032 |
Remy | +0.030 |
bur | +0.028 |
chy | +0.028 |
Neg logit contributions | |
folk | -0.033 |
digits | -0.032 |
sale | -0.030 |
customer | -0.028 |
letters | -0.028 |
Token | thanked |
Token ID | 26280 |
Feature activation | -5.23e-02 |
Pos logit contributions | |
folk | +0.274 |
digits | +0.262 |
sale | +0.239 |
letters | +0.227 |
cheering | +0.220 |
Neg logit contributions | |
oppy | -0.429 |
opsy | -0.429 |
Remy | -0.398 |
chy | -0.372 |
bur | -0.369 |
Token | the |
Token ID | 262 |
Feature activation | +2.90e-02 |
Pos logit contributions | |
folk | +0.087 |
digits | +0.084 |
sale | +0.082 |
letters | +0.075 |
customer | +0.075 |
Neg logit contributions | |
opsy | -0.104 |
oppy | -0.103 |
chy | -0.094 |
Remy | -0.092 |
bur | -0.091 |
Token | bear |
Token ID | 6842 |
Feature activation | +8.85e-02 |
Pos logit contributions | |
opsy | +0.050 |
oppy | +0.049 |
Remy | +0.045 |
chy | +0.043 |
bur | +0.043 |
Neg logit contributions | |
folk | -0.057 |
digits | -0.056 |
sale | -0.053 |
customer | -0.050 |
letters | -0.049 |
Token | and |
Token ID | 290 |
Feature activation | +2.74e-02 |
Pos logit contributions | |
opsy | +0.180 |
oppy | +0.176 |
Remy | +0.157 |
chy | +0.157 |
bur | +0.154 |
Neg logit contributions | |
folk | -0.141 |
digits | -0.134 |
sale | -0.134 |
customer | -0.127 |
letters | -0.125 |
Token | gave |
Token ID | 2921 |
Feature activation | -6.49e-02 |
Pos logit contributions | |
opsy | +0.062 |
oppy | +0.062 |
chy | +0.057 |
uffy | +0.055 |
Remy | +0.055 |
Neg logit contributions | |
folk | -0.037 |
digits | -0.035 |
sale | -0.033 |
letters | -0.032 |
customer | -0.031 |
Token | He |
Token ID | 679 |
Feature activation | -2.12e-02 |
Pos logit contributions | |
folk | +0.007 |
digits | +0.007 |
sale | +0.006 |
customer | +0.006 |
letters | +0.006 |
Neg logit contributions | |
opsy | -0.010 |
oppy | -0.010 |
Remy | -0.010 |
bur | -0.009 |
chy | -0.009 |
Token | laughed |
Token ID | 13818 |
Feature activation | -4.08e-02 |
Pos logit contributions | |
folk | +0.027 |
digits | +0.026 |
sale | +0.024 |
number | +0.023 |
cheering | +0.022 |
Neg logit contributions | |
oppy | -0.049 |
opsy | -0.049 |
Remy | -0.044 |
uffy | -0.043 |
chy | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | +3.04e-02 |
Pos logit contributions | |
folk | +0.060 |
digits | +0.057 |
sale | +0.055 |
letters | +0.051 |
cheering | +0.050 |
Neg logit contributions | |
opsy | -0.091 |
oppy | -0.090 |
Remy | -0.080 |
bur | -0.078 |
chy | -0.078 |
Token | laughed |
Token ID | 13818 |
Feature activation | -4.08e-02 |
Pos logit contributions | |
opsy | +0.066 |
oppy | +0.066 |
chy | +0.059 |
Remy | +0.058 |
bur | +0.057 |
Neg logit contributions | |
folk | -0.045 |
digits | -0.044 |
sale | -0.040 |
cheering | -0.038 |
letters | -0.038 |
Token | . |
Token ID | 13 |
Feature activation | -4.11e-03 |
Pos logit contributions | |
folk | +0.064 |
digits | +0.062 |
sale | +0.058 |
customer | +0.053 |
letters | +0.053 |
Neg logit contributions | |
opsy | -0.086 |
oppy | -0.085 |
Remy | -0.078 |
bur | -0.076 |
chy | -0.076 |
Token | |
Token ID | 220 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
folk | +0.005 |
digits | +0.005 |
sale | +0.005 |
customer | +0.005 |
epad | +0.005 |
Neg logit contributions | |
opsy | -0.009 |
oppy | -0.009 |
Remy | -0.009 |
bur | -0.009 |
chy | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
folk | +0.250 |
digits | +0.247 |
sale | +0.223 |
cheering | +0.197 |
John | +0.195 |
Neg logit contributions | |
opsy | -0.545 |
oppy | -0.524 |
bur | -0.502 |
chy | -0.501 |
Remy | -0.476 |
Token |
|
Token ID | 198 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
opsy | +0.042 |
oppy | +0.042 |
chy | +0.038 |
Remy | +0.038 |
bur | +0.038 |
Neg logit contributions | |
folk | -0.029 |
digits | -0.029 |
sale | -0.027 |
customer | -0.024 |
cheering | -0.024 |
Token | The |
Token ID | 464 |
Feature activation | +9.69e-02 |
Pos logit contributions | |
opsy | +0.037 |
oppy | +0.036 |
Remy | +0.033 |
chy | +0.032 |
bur | +0.032 |
Neg logit contributions | |
folk | -0.034 |
digits | -0.033 |
sale | -0.031 |
customer | -0.029 |
letters | -0.028 |
Token | man |
Token ID | 582 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
opsy | +0.208 |
oppy | +0.207 |
Remy | +0.191 |
chy | +0.186 |
bur | +0.185 |
Neg logit contributions | |
folk | -0.149 |
digits | -0.145 |
sale | -0.134 |
customer | -0.125 |
letters | -0.123 |
Token | was |
Token ID | 373 |
Feature activation | +5.13e-02 |
Pos logit contributions | |
folk | +0.217 |
digits | +0.207 |
sale | +0.194 |
letters | +0.180 |
customer | +0.178 |
Neg logit contributions | |
opsy | -0.297 |
oppy | -0.296 |
Remy | -0.270 |
chy | -0.264 |
bur | -0.258 |
Token | tiny |
Token ID | 7009 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
folk | +0.005 |
digits | +0.005 |
sale | +0.005 |
epad | +0.005 |
bench | +0.005 |
Neg logit contributions | |
opsy | -0.005 |
oppy | -0.005 |
bur | -0.004 |
Remy | -0.004 |
chy | -0.004 |
Token | mouse |
Token ID | 10211 |
Feature activation | +6.81e-02 |
Pos logit contributions | |
opsy | +0.197 |
oppy | +0.196 |
bur | +0.183 |
Remy | +0.178 |
chy | +0.170 |
Neg logit contributions | |
folk | -0.224 |
digits | -0.217 |
sale | -0.199 |
epad | -0.191 |
bench | -0.189 |
Token | . |
Token ID | 13 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
opsy | +0.140 |
oppy | +0.138 |
Remy | +0.127 |
chy | +0.123 |
bur | +0.122 |
Neg logit contributions | |
folk | -0.108 |
digits | -0.106 |
sale | -0.100 |
customer | -0.092 |
letters | -0.092 |
Token | The |
Token ID | 383 |
Feature activation | +8.72e-02 |
Pos logit contributions | |
folk | +0.025 |
digits | +0.025 |
sale | +0.023 |
customer | +0.022 |
letters | +0.021 |
Neg logit contributions | |
opsy | -0.031 |
oppy | -0.030 |
chy | -0.027 |
Remy | -0.027 |
bur | -0.027 |
Token | mouse |
Token ID | 10211 |
Feature activation | +6.87e-02 |
Pos logit contributions | |
opsy | +0.155 |
oppy | +0.153 |
Remy | +0.139 |
chy | +0.135 |
bur | +0.134 |
Neg logit contributions | |
folk | -0.165 |
digits | -0.162 |
sale | -0.153 |
customer | -0.144 |
letters | -0.143 |
Token | asked |
Token ID | 1965 |
Feature activation | -2.57e-01 |
Pos logit contributions | |
opsy | +0.145 |
oppy | +0.143 |
Remy | +0.132 |
chy | +0.128 |
bur | +0.127 |
Neg logit contributions | |
folk | -0.105 |
digits | -0.102 |
sale | -0.095 |
letters | -0.088 |
customer | -0.087 |
Token | the |
Token ID | 262 |
Feature activation | +2.35e-02 |
Pos logit contributions | |
folk | +0.440 |
digits | +0.432 |
sale | +0.394 |
customer | +0.371 |
bench | +0.368 |
Neg logit contributions | |
oppy | -0.499 |
opsy | -0.499 |
Remy | -0.472 |
chy | -0.442 |
bur | -0.441 |
Token | bald |
Token ID | 28273 |
Feature activation | -8.05e-02 |
Pos logit contributions | |
opsy | +0.041 |
oppy | +0.041 |
Remy | +0.037 |
chy | +0.036 |
bur | +0.036 |
Neg logit contributions | |
folk | -0.045 |
digits | -0.044 |
sale | -0.042 |
customer | -0.039 |
letters | -0.039 |
Token | man |
Token ID | 582 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
folk | +0.135 |
digits | +0.132 |
sale | +0.122 |
customer | +0.113 |
letters | +0.112 |
Neg logit contributions | |
opsy | -0.160 |
oppy | -0.158 |
Remy | -0.146 |
chy | -0.142 |
bur | -0.141 |
Token | if |
Token ID | 611 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
folk | +0.269 |
digits | +0.260 |
sale | +0.239 |
bench | +0.227 |
customer | +0.223 |
Neg logit contributions | |
oppy | -0.291 |
opsy | -0.289 |
Remy | -0.272 |
chy | -0.257 |
bur | -0.256 |
Token | he |
Token ID | 339 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
opsy | +0.030 |
oppy | +0.029 |
chy | +0.026 |
Remy | +0.026 |
bur | +0.026 |
Neg logit contributions | |
folk | -0.022 |
digits | -0.021 |
sale | -0.020 |
customer | -0.019 |
letters | -0.019 |
Token | ™ |
Token ID | 8151 |
Feature activation | +3.28e-02 |
Pos logit contributions | |
opsy | +0.076 |
oppy | +0.074 |
Remy | +0.068 |
chy | +0.068 |
bur | +0.067 |
Neg logit contributions | |
folk | -0.066 |
digits | -0.064 |
sale | -0.060 |
customer | -0.056 |
letters | -0.056 |
Token | s |
Token ID | 82 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
opsy | +0.055 |
oppy | +0.054 |
Remy | +0.049 |
bur | +0.048 |
chy | +0.048 |
Neg logit contributions | |
folk | -0.064 |
digits | -0.062 |
sale | -0.060 |
customer | -0.056 |
letters | -0.055 |
Token | go |
Token ID | 467 |
Feature activation | +5.97e-02 |
Pos logit contributions | |
opsy | +0.387 |
oppy | +0.382 |
Remy | +0.353 |
chy | +0.344 |
bur | +0.341 |
Neg logit contributions | |
folk | -0.311 |
digits | -0.303 |
sale | -0.285 |
customer | -0.265 |
letters | -0.262 |
Token | to |
Token ID | 284 |
Feature activation | -8.71e-03 |
Pos logit contributions | |
opsy | +0.118 |
oppy | +0.118 |
Remy | +0.109 |
bur | +0.105 |
chy | +0.104 |
Neg logit contributions | |
folk | -0.100 |
digits | -0.097 |
sale | -0.090 |
letters | -0.084 |
customer | -0.084 |
Token | the |
Token ID | 262 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
sale | +0.012 |
folk | +0.011 |
digits | +0.011 |
customer | +0.011 |
cheering | +0.010 |
Neg logit contributions | |
opsy | -0.020 |
oppy | -0.019 |
chy | -0.018 |
uffy | -0.017 |
rig | -0.017 |
Token | store |
Token ID | 3650 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
oppy | +0.050 |
opsy | +0.050 |
Remy | +0.045 |
chy | +0.044 |
bur | +0.043 |
Neg logit contributions | |
folk | -0.035 |
digits | -0.035 |
sale | -0.034 |
customer | -0.032 |
letters | -0.031 |
Token | and |
Token ID | 290 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
folk | +0.373 |
digits | +0.371 |
Amen | +0.324 |
sale | +0.323 |
epad | +0.321 |
Neg logit contributions | |
opsy | -0.357 |
oppy | -0.355 |
Remy | -0.337 |
bur | -0.320 |
chy | -0.314 |
Token | get |
Token ID | 651 |
Feature activation | +1.35e-03 |
Pos logit contributions | |
opsy | +0.042 |
oppy | +0.041 |
chy | +0.038 |
Remy | +0.037 |
bur | +0.036 |
Neg logit contributions | |
folk | -0.027 |
digits | -0.026 |
sale | -0.025 |
customer | -0.023 |
letters | -0.022 |
Token | you |
Token ID | 345 |
Feature activation | +6.59e-02 |
Pos logit contributions | |
opsy | +0.003 |
oppy | +0.003 |
Remy | +0.003 |
bur | +0.003 |
chy | +0.002 |
Neg logit contributions | |
folk | -0.002 |
digits | -0.002 |
sale | -0.002 |
customer | -0.002 |
epad | -0.002 |
Token | a |
Token ID | 257 |
Feature activation | +4.79e-05 |
Pos logit contributions | |
opsy | +0.142 |
oppy | +0.139 |
Remy | +0.130 |
chy | +0.126 |
bur | +0.125 |
Neg logit contributions | |
folk | -0.097 |
digits | -0.094 |
sale | -0.090 |
customer | -0.083 |
letters | -0.082 |
Token | battery |
Token ID | 6555 |
Feature activation | -9.05e-02 |
Pos logit contributions | |
opsy | +0.000 |
oppy | +0.000 |
Remy | +0.000 |
chy | +0.000 |
bur | +0.000 |
Neg logit contributions | |
folk | -0.000 |
digits | -0.000 |
sale | -0.000 |
customer | -0.000 |
letters | -0.000 |
Token | bons |
Token ID | 23461 |
Feature activation | -1.73e-01 |
Pos logit contributions | |
opsy | +0.008 |
oppy | +0.008 |
Remy | +0.008 |
bur | +0.008 |
chy | +0.007 |
Neg logit contributions | |
folk | -0.004 |
digits | -0.004 |
sale | -0.004 |
customer | -0.003 |
jury | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | +3.48e-02 |
Pos logit contributions | |
folk | +0.288 |
digits | +0.280 |
sale | +0.262 |
customer | +0.245 |
letters | +0.243 |
Neg logit contributions | |
opsy | -0.343 |
oppy | -0.339 |
Remy | -0.313 |
chy | -0.302 |
bur | -0.302 |
Token | shiny |
Token ID | 22441 |
Feature activation | +7.88e-03 |
Pos logit contributions | |
opsy | +0.066 |
oppy | +0.064 |
Remy | +0.062 |
bur | +0.061 |
chy | +0.061 |
Neg logit contributions | |
folk | -0.059 |
digits | -0.056 |
sale | -0.054 |
customer | -0.051 |
Amen | -0.050 |
Token | stars |
Token ID | 5788 |
Feature activation | -3.67e-02 |
Pos logit contributions | |
bur | +0.012 |
opsy | +0.012 |
oppy | +0.012 |
chy | +0.011 |
Remy | +0.011 |
Neg logit contributions | |
folk | -0.014 |
Amen | -0.014 |
digits | -0.013 |
customer | -0.013 |
epad | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | +5.18e-04 |
Pos logit contributions | |
folk | +0.062 |
digits | +0.061 |
sale | +0.056 |
customer | +0.053 |
Amen | +0.052 |
Neg logit contributions | |
opsy | -0.071 |
oppy | -0.070 |
Remy | -0.065 |
chy | -0.064 |
bur | -0.063 |
Token | |
Token ID | 220 |
Feature activation | -2.18e-01 |
Pos logit contributions | |
opsy | +0.001 |
oppy | +0.001 |
Remy | +0.001 |
chy | +0.001 |
bur | +0.001 |
Neg logit contributions | |
folk | -0.001 |
digits | -0.001 |
sale | -0.001 |
customer | -0.001 |
letters | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
folk | +0.281 |
digits | +0.277 |
sale | +0.254 |
letters | +0.228 |
cheering | +0.224 |
Neg logit contributions | |
opsy | -0.498 |
oppy | -0.486 |
chy | -0.465 |
bur | -0.462 |
Remy | -0.456 |
Token |
|
Token ID | 198 |
Feature activation | +2.56e-02 |
Pos logit contributions | |
oppy | +0.059 |
opsy | +0.058 |
chy | +0.056 |
bur | +0.054 |
Remy | +0.053 |
Neg logit contributions | |
digits | -0.034 |
folk | -0.034 |
sale | -0.032 |
cheering | -0.029 |
customer | -0.028 |
Token | When |
Token ID | 2215 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
oppy | +0.053 |
opsy | +0.053 |
chy | +0.049 |
Remy | +0.049 |
bur | +0.048 |
Neg logit contributions | |
folk | -0.039 |
digits | -0.039 |
sale | -0.037 |
customer | -0.034 |
cheering | -0.033 |
Token | it |
Token ID | 340 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
opsy | +0.056 |
oppy | +0.056 |
chy | +0.051 |
Remy | +0.051 |
bur | +0.050 |
Neg logit contributions | |
folk | -0.041 |
digits | -0.040 |
sale | -0.039 |
letters | -0.035 |
customer | -0.035 |
Token | was |
Token ID | 373 |
Feature activation | +5.76e-02 |
Pos logit contributions | |
folk | +0.167 |
digits | +0.163 |
sale | +0.156 |
cheering | +0.137 |
number | +0.137 |
Neg logit contributions | |
opsy | -0.279 |
oppy | -0.274 |
Remy | -0.262 |
bur | -0.239 |
chy | -0.235 |