Token | Suddenly |
Token ID | 38582 |
Feature activation | +2.49e+00 |
Pos logit contributions | |
beast | +0.023 |
gut | +0.023 |
mistake | +0.022 |
swollen | +0.021 |
therm | +0.021 |
Neg logit contributions | |
parties | -0.024 |
tea | -0.021 |
celebrate | -0.018 |
recess | -0.018 |
picnic | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
beast | +0.100 |
gut | +0.100 |
mistake | +0.094 |
inct | +0.093 |
therm | +0.093 |
Neg logit contributions | |
parties | -0.066 |
tea | -0.057 |
recess | -0.046 |
celebrate | -0.043 |
party | -0.042 |
Token | the |
Token ID | 262 |
Feature activation | +4.39e+00 |
Pos logit contributions | |
gut | +0.049 |
beast | +0.048 |
mistake | +0.048 |
swollen | +0.046 |
inct | +0.046 |
Neg logit contributions | |
parties | -0.030 |
tea | -0.027 |
recess | -0.021 |
celebrate | -0.020 |
picnic | -0.019 |
Token | rh |
Token ID | 9529 |
Feature activation | +5.52e+00 |
Pos logit contributions | |
gut | +0.157 |
rum | +0.152 |
mistake | +0.136 |
outage | +0.136 |
severe | +0.135 |
Neg logit contributions | |
parties | -0.122 |
tea | -0.119 |
picnic | -0.111 |
party | -0.107 |
celebrate | -0.106 |
Token | in |
Token ID | 259 |
Feature activation | +5.62e+00 |
Pos logit contributions | |
gut | +0.240 |
plea | +0.220 |
severe | +0.215 |
iously | +0.214 |
swollen | +0.214 |
Neg logit contributions | |
parties | -0.106 |
tea | -0.088 |
acky | -0.082 |
together | -0.076 |
P | -0.075 |
Token | oc |
Token ID | 420 |
Feature activation | +9.07e+00 |
Pos logit contributions | |
gut | +0.215 |
swollen | +0.188 |
mistake | +0.185 |
rum | +0.184 |
vain | +0.184 |
Neg logit contributions | |
played | -0.125 |
picnic | -0.118 |
tea | -0.118 |
play | -0.111 |
acky | -0.105 |
Token | eros |
Token ID | 27498 |
Feature activation | +4.81e+00 |
Pos logit contributions | |
severe | +0.304 |
outage | +0.294 |
swollen | +0.283 |
gut | +0.280 |
reckless | +0.274 |
Neg logit contributions | |
parties | -0.255 |
tea | -0.218 |
ils | -0.196 |
sandwiches | -0.193 |
recess | -0.186 |
Token | heard |
Token ID | 2982 |
Feature activation | +3.52e+00 |
Pos logit contributions | |
gut | +0.192 |
rum | +0.178 |
swollen | +0.176 |
therm | +0.173 |
mistake | +0.173 |
Neg logit contributions | |
parties | -0.116 |
tea | -0.109 |
played | -0.098 |
party | -0.092 |
picnic | -0.092 |
Token | his |
Token ID | 465 |
Feature activation | +1.94e+00 |
Pos logit contributions | |
gut | +0.153 |
beast | +0.147 |
swollen | +0.146 |
cut | +0.143 |
mistake | +0.143 |
Neg logit contributions | |
parties | -0.084 |
tea | -0.075 |
celebrate | -0.054 |
recess | -0.052 |
party | -0.051 |
Token | friends |
Token ID | 2460 |
Feature activation | -9.98e-01 |
Pos logit contributions | |
gut | +0.054 |
beast | +0.053 |
cut | +0.052 |
swollen | +0.051 |
inct | +0.050 |
Neg logit contributions | |
parties | -0.071 |
tea | -0.066 |
recess | -0.055 |
party | -0.055 |
celebrate | -0.055 |
Token | laughing |
Token ID | 14376 |
Feature activation | -8.10e-01 |
Pos logit contributions | |
parties | +0.033 |
tea | +0.029 |
recess | +0.025 |
celebrate | +0.024 |
party | +0.024 |
Neg logit contributions | |
beast | -0.033 |
gut | -0.033 |
inct | -0.031 |
swollen | -0.031 |
mistake | -0.031 |
Token |
|
Token ID | 198 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
gut | +0.013 |
beast | +0.013 |
mistake | +0.012 |
therm | +0.012 |
inct | +0.012 |
Neg logit contributions | |
parties | -0.014 |
tea | -0.013 |
recess | -0.011 |
celebrate | -0.011 |
party | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | +6.13e-01 |
Pos logit contributions | |
gut | +0.007 |
beast | +0.007 |
mistake | +0.007 |
inct | +0.007 |
therm | +0.007 |
Neg logit contributions | |
parties | -0.008 |
tea | -0.007 |
recess | -0.006 |
party | -0.006 |
celebrate | -0.006 |
Token | The |
Token ID | 464 |
Feature activation | +4.18e+00 |
Pos logit contributions | |
gut | +0.021 |
beast | +0.021 |
mistake | +0.019 |
inct | +0.019 |
therm | +0.019 |
Neg logit contributions | |
parties | -0.020 |
tea | -0.018 |
recess | -0.015 |
celebrate | -0.015 |
party | -0.014 |
Token | rh |
Token ID | 9529 |
Feature activation | +4.82e+00 |
Pos logit contributions | |
gut | +0.136 |
rum | +0.123 |
beast | +0.119 |
inct | +0.119 |
severe | +0.118 |
Neg logit contributions | |
parties | -0.137 |
tea | -0.129 |
party | -0.112 |
picnic | -0.112 |
celebrate | -0.110 |
Token | in |
Token ID | 259 |
Feature activation | +5.15e+00 |
Pos logit contributions | |
gut | +0.195 |
iously | +0.183 |
inct | +0.182 |
plea | +0.182 |
severe | +0.180 |
Neg logit contributions | |
parties | -0.116 |
tea | -0.101 |
recess | -0.080 |
party | -0.078 |
acky | -0.076 |
Token | oc |
Token ID | 420 |
Feature activation | +9.01e+00 |
Pos logit contributions | |
gut | +0.162 |
severe | +0.149 |
swollen | +0.144 |
reckless | +0.140 |
rum | +0.138 |
Neg logit contributions | |
tea | -0.154 |
parties | -0.146 |
played | -0.144 |
picnic | -0.138 |
party | -0.135 |
Token | eros |
Token ID | 27498 |
Feature activation | +4.57e+00 |
Pos logit contributions | |
severe | +0.303 |
swollen | +0.286 |
outage | +0.276 |
reckless | +0.272 |
gut | +0.268 |
Neg logit contributions | |
parties | -0.252 |
tea | -0.219 |
sandwiches | -0.192 |
recess | -0.191 |
ils | -0.191 |
Token | walked |
Token ID | 6807 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
gut | +0.178 |
severe | +0.166 |
beast | +0.164 |
swollen | +0.163 |
therm | +0.162 |
Neg logit contributions | |
parties | -0.122 |
tea | -0.116 |
played | -0.098 |
party | -0.093 |
picnic | -0.091 |
Token | to |
Token ID | 284 |
Feature activation | +3.58e+00 |
Pos logit contributions | |
parties | +0.053 |
tea | +0.047 |
recess | +0.040 |
celebrate | +0.038 |
party | +0.037 |
Neg logit contributions | |
beast | -0.050 |
gut | -0.050 |
mistake | -0.047 |
swollen | -0.046 |
therm | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | +6.34e-01 |
Pos logit contributions | |
beast | +0.131 |
gut | +0.130 |
mistake | +0.126 |
therm | +0.121 |
passer | +0.121 |
Neg logit contributions | |
parties | -0.106 |
tea | -0.090 |
recess | -0.075 |
ets | -0.072 |
ils | -0.069 |
Token | fence |
Token ID | 13990 |
Feature activation | +4.07e+00 |
Pos logit contributions | |
gut | +0.020 |
rum | +0.020 |
inct | +0.019 |
beast | +0.018 |
swollen | +0.018 |
Neg logit contributions | |
parties | -0.020 |
tea | -0.020 |
picnic | -0.018 |
party | -0.018 |
celebrate | -0.017 |
Token | ," |
Token ID | 553 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
gut | +0.064 |
beast | +0.064 |
inct | +0.061 |
mistake | +0.060 |
therm | +0.060 |
Neg logit contributions | |
parties | -0.041 |
tea | -0.036 |
recess | -0.028 |
celebrate | -0.027 |
party | -0.026 |
Token | said |
Token ID | 531 |
Feature activation | -7.05e-01 |
Pos logit contributions | |
gut | +0.043 |
beast | +0.042 |
mistake | +0.040 |
inct | +0.039 |
severe | +0.038 |
Neg logit contributions | |
parties | -0.033 |
tea | -0.030 |
celebrate | -0.024 |
recess | -0.023 |
party | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | +3.31e+00 |
Pos logit contributions | |
parties | +0.020 |
tea | +0.018 |
recess | +0.014 |
celebrate | +0.014 |
party | +0.013 |
Neg logit contributions | |
gut | -0.027 |
beast | -0.027 |
mistake | -0.026 |
inct | -0.025 |
swollen | -0.025 |
Token | rh |
Token ID | 9529 |
Feature activation | +5.03e+00 |
Pos logit contributions | |
gut | +0.104 |
rum | +0.093 |
beast | +0.093 |
inct | +0.091 |
mistake | +0.089 |
Neg logit contributions | |
parties | -0.113 |
tea | -0.103 |
celebrate | -0.091 |
picnic | -0.090 |
party | -0.090 |
Token | in |
Token ID | 259 |
Feature activation | +5.27e+00 |
Pos logit contributions | |
gut | +0.204 |
inct | +0.189 |
plea | +0.186 |
swollen | +0.186 |
mistake | +0.185 |
Neg logit contributions | |
parties | -0.117 |
tea | -0.097 |
acky | -0.082 |
celebrate | -0.080 |
together | -0.079 |
Token | oc |
Token ID | 420 |
Feature activation | +9.01e+00 |
Pos logit contributions | |
swollen | +0.214 |
gut | +0.210 |
distress | +0.200 |
severe | +0.199 |
beast | +0.198 |
Neg logit contributions | |
tea | -0.090 |
parties | -0.086 |
together | -0.085 |
played | -0.069 |
play | -0.068 |
Token | eros |
Token ID | 27498 |
Feature activation | +4.90e+00 |
Pos logit contributions | |
outage | +0.276 |
severe | +0.273 |
swollen | +0.264 |
gut | +0.263 |
confusion | +0.260 |
Neg logit contributions | |
parties | -0.259 |
tea | -0.220 |
ils | -0.212 |
sandwiches | -0.198 |
recess | -0.188 |
Token | , |
Token ID | 11 |
Feature activation | +3.29e-01 |
Pos logit contributions | |
gut | +0.218 |
beast | +0.216 |
swollen | +0.208 |
inct | +0.208 |
mistake | +0.207 |
Neg logit contributions | |
parties | -0.094 |
tea | -0.087 |
party | -0.057 |
celebrate | -0.056 |
played | -0.055 |
Token | feeling |
Token ID | 4203 |
Feature activation | +2.16e+00 |
Pos logit contributions | |
gut | +0.012 |
beast | +0.011 |
severe | +0.011 |
mistake | +0.011 |
inct | +0.011 |
Neg logit contributions | |
tea | -0.010 |
parties | -0.009 |
played | -0.008 |
Lily | -0.008 |
together | -0.008 |
Token | discouraged |
Token ID | 30170 |
Feature activation | +3.61e+00 |
Pos logit contributions | |
beast | +0.063 |
inct | +0.063 |
mistake | +0.062 |
gut | +0.061 |
adult | +0.059 |
Neg logit contributions | |
parties | -0.073 |
tea | -0.072 |
celebrate | -0.057 |
played | -0.056 |
wear | -0.056 |
Token | . |
Token ID | 13 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
beast | +0.178 |
gut | +0.178 |
inct | +0.177 |
swollen | +0.170 |
passer | +0.169 |
Neg logit contributions | |
parties | -0.056 |
tea | -0.048 |
together | -0.033 |
wear | -0.023 |
party | -0.023 |
Token | € |
Token ID | 26391 |
Feature activation | -6.98e-01 |
Pos logit contributions | |
gut | +0.039 |
beast | +0.038 |
swollen | +0.037 |
mistake | +0.036 |
therm | +0.036 |
Neg logit contributions | |
parties | -0.028 |
tea | -0.023 |
recess | -0.020 |
celebrate | -0.020 |
wear | -0.018 |
Token | replied |
Token ID | 8712 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
parties | +0.022 |
tea | +0.019 |
recess | +0.016 |
celebrate | +0.016 |
picnic | +0.015 |
Neg logit contributions | |
beast | -0.024 |
gut | -0.024 |
mistake | -0.023 |
therm | -0.023 |
swollen | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | +3.41e+00 |
Pos logit contributions | |
parties | +0.040 |
tea | +0.036 |
celebrate | +0.026 |
recess | +0.025 |
wear | +0.024 |
Neg logit contributions | |
beast | -0.073 |
gut | -0.070 |
mistake | -0.069 |
inct | -0.068 |
swollen | -0.067 |
Token | rh |
Token ID | 9529 |
Feature activation | +5.04e+00 |
Pos logit contributions | |
gut | +0.109 |
beast | +0.094 |
inct | +0.093 |
severe | +0.093 |
swollen | +0.093 |
Neg logit contributions | |
parties | -0.116 |
tea | -0.113 |
party | -0.096 |
picnic | -0.096 |
celebrate | -0.094 |
Token | in |
Token ID | 259 |
Feature activation | +5.49e+00 |
Pos logit contributions | |
gut | +0.204 |
inct | +0.194 |
plea | +0.193 |
mistake | +0.190 |
swollen | +0.188 |
Neg logit contributions | |
parties | -0.123 |
tea | -0.106 |
recess | -0.085 |
ils | -0.081 |
acky | -0.080 |
Token | oc |
Token ID | 420 |
Feature activation | +8.99e+00 |
Pos logit contributions | |
gut | +0.171 |
swollen | +0.165 |
severe | +0.155 |
distress | +0.155 |
injured | +0.154 |
Neg logit contributions | |
tea | -0.164 |
parties | -0.151 |
played | -0.134 |
play | -0.129 |
celebrate | -0.129 |
Token | eros |
Token ID | 27498 |
Feature activation | +4.60e+00 |
Pos logit contributions | |
severe | +0.291 |
swollen | +0.272 |
outage | +0.268 |
terrified | +0.267 |
frightening | +0.260 |
Neg logit contributions | |
parties | -0.262 |
tea | -0.230 |
ils | -0.217 |
recess | -0.204 |
sandwiches | -0.193 |
Token | sadly |
Token ID | 21098 |
Feature activation | +5.55e+00 |
Pos logit contributions | |
gut | +0.160 |
beast | +0.156 |
inct | +0.152 |
mistake | +0.150 |
swollen | +0.149 |
Neg logit contributions | |
parties | -0.134 |
tea | -0.128 |
party | -0.097 |
celebrate | -0.097 |
played | -0.096 |
Token | . |
Token ID | 13 |
Feature activation | +5.09e-01 |
Pos logit contributions | |
beast | +0.246 |
gut | +0.238 |
inct | +0.227 |
passer | +0.226 |
swollen | +0.223 |
Neg logit contributions | |
parties | -0.119 |
tea | -0.107 |
wear | -0.080 |
celebrate | -0.079 |
together | -0.078 |
Token | It |
Token ID | 632 |
Feature activation | +4.02e+00 |
Pos logit contributions | |
beast | +0.021 |
gut | +0.020 |
mistake | +0.020 |
swollen | +0.019 |
cut | +0.019 |
Neg logit contributions | |
parties | -0.013 |
tea | -0.011 |
celebrate | -0.009 |
recess | -0.009 |
Lilly | -0.009 |
Token | had |
Token ID | 550 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
gut | +0.157 |
beast | +0.152 |
passer | +0.141 |
adult | +0.141 |
mistake | +0.141 |
Neg logit contributions | |
tea | -0.098 |
parties | -0.092 |
played | -0.084 |
wear | -0.073 |
together | -0.072 |
Token |
|
Token ID | 198 |
Feature activation | +5.72e-01 |
Pos logit contributions | |
gut | +0.037 |
beast | +0.036 |
mistake | +0.036 |
swollen | +0.035 |
severe | +0.034 |
Neg logit contributions | |
parties | -0.021 |
tea | -0.018 |
celebrate | -0.015 |
together | -0.014 |
recess | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
beast | +0.020 |
gut | +0.020 |
mistake | +0.019 |
inct | +0.018 |
therm | +0.018 |
Neg logit contributions | |
parties | -0.018 |
tea | -0.016 |
recess | -0.014 |
celebrate | -0.013 |
party | -0.013 |
Token | The |
Token ID | 464 |
Feature activation | +4.53e+00 |
Pos logit contributions | |
beast | +0.041 |
gut | +0.041 |
swollen | +0.040 |
therm | +0.040 |
mistake | +0.039 |
Neg logit contributions | |
parties | -0.033 |
tea | -0.026 |
celebrate | -0.025 |
games | -0.023 |
recess | -0.023 |
Token | rh |
Token ID | 9529 |
Feature activation | +5.34e+00 |
Pos logit contributions | |
beast | +0.113 |
gut | +0.112 |
mistake | +0.102 |
inct | +0.101 |
therm | +0.100 |
Neg logit contributions | |
parties | -0.190 |
tea | -0.173 |
recess | -0.153 |
celebrate | -0.149 |
party | -0.146 |
Token | in |
Token ID | 259 |
Feature activation | +5.56e+00 |
Pos logit contributions | |
gut | +0.218 |
plea | +0.203 |
inct | +0.201 |
iously | +0.200 |
swollen | +0.197 |
Neg logit contributions | |
parties | -0.127 |
tea | -0.105 |
acky | -0.087 |
together | -0.086 |
celebrate | -0.085 |
Token | oc |
Token ID | 420 |
Feature activation | +8.98e+00 |
Pos logit contributions | |
gut | +0.178 |
swollen | +0.159 |
reckless | +0.156 |
mistake | +0.154 |
severe | +0.152 |
Neg logit contributions | |
parties | -0.157 |
tea | -0.155 |
played | -0.151 |
celebrate | -0.143 |
play | -0.140 |
Token | eros |
Token ID | 27498 |
Feature activation | +4.83e+00 |
Pos logit contributions | |
severe | +0.270 |
swollen | +0.256 |
outage | +0.254 |
gut | +0.253 |
reckless | +0.247 |
Neg logit contributions | |
parties | -0.287 |
tea | -0.238 |
ils | -0.229 |
recess | -0.221 |
party | -0.205 |
Token | looked |
Token ID | 3114 |
Feature activation | -4.51e-01 |
Pos logit contributions | |
gut | +0.193 |
swollen | +0.188 |
beast | +0.184 |
therm | +0.181 |
rum | +0.179 |
Neg logit contributions | |
parties | -0.112 |
tea | -0.105 |
played | -0.102 |
party | -0.089 |
picnic | -0.086 |
Token | up |
Token ID | 510 |
Feature activation | +3.99e+00 |
Pos logit contributions | |
parties | +0.014 |
tea | +0.012 |
recess | +0.011 |
celebrate | +0.010 |
party | +0.009 |
Neg logit contributions | |
gut | -0.017 |
beast | -0.016 |
mistake | -0.015 |
inct | -0.015 |
therm | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | -3.29e-01 |
Pos logit contributions | |
beast | +0.155 |
gut | +0.152 |
mistake | +0.144 |
swollen | +0.143 |
inct | +0.143 |
Neg logit contributions | |
parties | -0.111 |
tea | -0.097 |
recess | -0.079 |
celebrate | -0.075 |
picnic | -0.073 |
Token | hopeful |
Token ID | 17836 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
parties | +0.007 |
tea | +0.006 |
together | +0.006 |
played | +0.005 |
wear | +0.005 |
Neg logit contributions | |
beast | -0.014 |
gut | -0.014 |
rum | -0.014 |
severe | -0.013 |
inct | -0.013 |
Token | here |
Token ID | 994 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
parties | +0.030 |
tea | +0.026 |
celebrate | +0.021 |
recess | +0.020 |
party | +0.020 |
Neg logit contributions | |
gut | -0.042 |
beast | -0.041 |
swollen | -0.041 |
inct | -0.040 |
mistake | -0.040 |
Token | ?" |
Token ID | 1701 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
beast | +0.044 |
gut | +0.043 |
inct | +0.041 |
mistake | +0.041 |
swollen | +0.041 |
Neg logit contributions | |
parties | -0.032 |
tea | -0.027 |
recess | -0.022 |
celebrate | -0.022 |
party | -0.021 |
Token | The |
Token ID | 383 |
Feature activation | +4.37e+00 |
Pos logit contributions | |
swollen | +0.109 |
gut | +0.109 |
mistake | +0.106 |
severe | +0.105 |
therm | +0.105 |
Neg logit contributions | |
parties | -0.061 |
tea | -0.053 |
celebrate | -0.044 |
played | -0.042 |
recess | -0.041 |
Token | rh |
Token ID | 9529 |
Feature activation | +5.51e+00 |
Pos logit contributions | |
gut | +0.146 |
inct | +0.130 |
rum | +0.130 |
beast | +0.125 |
swollen | +0.121 |
Neg logit contributions | |
parties | -0.140 |
tea | -0.127 |
celebrate | -0.117 |
picnic | -0.114 |
party | -0.113 |
Token | in |
Token ID | 259 |
Feature activation | +5.96e+00 |
Pos logit contributions | |
gut | +0.237 |
inct | +0.216 |
iously | +0.215 |
plea | +0.215 |
therm | +0.212 |
Neg logit contributions | |
parties | -0.116 |
tea | -0.091 |
together | -0.083 |
acky | -0.080 |
celebrate | -0.077 |
Token | oc |
Token ID | 420 |
Feature activation | +8.97e+00 |
Pos logit contributions | |
gut | +0.221 |
swollen | +0.191 |
outage | +0.190 |
severe | +0.189 |
rier | +0.187 |
Neg logit contributions | |
played | -0.135 |
named | -0.126 |
together | -0.124 |
play | -0.123 |
tea | -0.122 |
Token | eros |
Token ID | 27498 |
Feature activation | +5.45e+00 |
Pos logit contributions | |
severe | +0.286 |
outage | +0.284 |
gut | +0.275 |
swollen | +0.270 |
therm | +0.260 |
Neg logit contributions | |
parties | -0.267 |
tea | -0.217 |
recess | -0.203 |
ils | -0.202 |
sandwiches | -0.200 |
Token | replied |
Token ID | 8712 |
Feature activation | -4.89e-01 |
Pos logit contributions | |
gut | +0.241 |
severe | +0.223 |
rum | +0.223 |
beast | +0.219 |
outage | +0.218 |
Neg logit contributions | |
parties | -0.101 |
tea | -0.093 |
played | -0.090 |
party | -0.078 |
together | -0.077 |
Token | , |
Token ID | 11 |
Feature activation | +3.50e-01 |
Pos logit contributions | |
parties | +0.010 |
tea | +0.009 |
together | +0.006 |
celebrate | +0.006 |
recess | +0.006 |
Neg logit contributions | |
beast | -0.022 |
gut | -0.021 |
inct | -0.021 |
mistake | -0.021 |
swollen | -0.021 |
Token | " |
Token ID | 366 |
Feature activation | +3.72e+00 |
Pos logit contributions | |
beast | +0.014 |
gut | +0.013 |
severe | +0.013 |
outage | +0.013 |
mistake | +0.013 |
Neg logit contributions | |
parties | -0.008 |
tea | -0.008 |
together | -0.007 |
them | -0.007 |
played | -0.007 |
Token | I |
Token ID | 40 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
gut | +0.157 |
mistake | +0.155 |
beast | +0.153 |
therm | +0.144 |
swollen | +0.143 |
Neg logit contributions | |
parties | -0.086 |
tea | -0.071 |
celebrate | -0.057 |
recess | -0.056 |
wear | -0.056 |
Token | doing |
Token ID | 1804 |
Feature activation | -9.61e-01 |
Pos logit contributions | |
parties | +0.086 |
tea | +0.076 |
party | +0.063 |
celebrate | +0.063 |
recess | +0.061 |
Neg logit contributions | |
gut | -0.097 |
beast | -0.092 |
swollen | -0.090 |
inct | -0.089 |
severe | -0.088 |
Token | ?" |
Token ID | 1701 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
parties | +0.027 |
tea | +0.023 |
recess | +0.019 |
celebrate | +0.018 |
party | +0.017 |
Neg logit contributions | |
beast | -0.037 |
gut | -0.037 |
mistake | -0.035 |
inct | -0.035 |
swollen | -0.035 |
Token | The |
Token ID | 383 |
Feature activation | +4.10e+00 |
Pos logit contributions | |
gut | +0.075 |
beast | +0.074 |
mistake | +0.070 |
swollen | +0.070 |
severe | +0.068 |
Neg logit contributions | |
parties | -0.052 |
tea | -0.045 |
celebrate | -0.037 |
recess | -0.036 |
party | -0.034 |
Token | rh |
Token ID | 9529 |
Feature activation | +5.44e+00 |
Pos logit contributions | |
gut | +0.121 |
beast | +0.112 |
inct | +0.107 |
rum | +0.107 |
severe | +0.102 |
Neg logit contributions | |
parties | -0.152 |
tea | -0.143 |
party | -0.123 |
picnic | -0.122 |
celebrate | -0.122 |
Token | in |
Token ID | 259 |
Feature activation | +5.97e+00 |
Pos logit contributions | |
gut | +0.216 |
inct | +0.195 |
beast | +0.193 |
plea | +0.193 |
therm | +0.193 |
Neg logit contributions | |
parties | -0.143 |
tea | -0.123 |
celebrate | -0.101 |
together | -0.098 |
acky | -0.098 |
Token | oc |
Token ID | 420 |
Feature activation | +8.94e+00 |
Pos logit contributions | |
gut | +0.196 |
reckless | +0.163 |
severe | +0.163 |
beast | +0.162 |
adult | +0.160 |
Neg logit contributions | |
tea | -0.172 |
played | -0.170 |
parties | -0.167 |
play | -0.155 |
together | -0.154 |
Token | eros |
Token ID | 27498 |
Feature activation | +5.47e+00 |
Pos logit contributions | |
severe | +0.276 |
gut | +0.270 |
outage | +0.263 |
swollen | +0.256 |
therm | +0.251 |
Neg logit contributions | |
parties | -0.274 |
tea | -0.235 |
ils | -0.213 |
sandwiches | -0.212 |
recess | -0.203 |
Token | replied |
Token ID | 8712 |
Feature activation | -7.51e-01 |
Pos logit contributions | |
gut | +0.232 |
beast | +0.218 |
severe | +0.212 |
therm | +0.207 |
swollen | +0.207 |
Neg logit contributions | |
parties | -0.132 |
tea | -0.117 |
played | -0.100 |
party | -0.094 |
celebrate | -0.091 |
Token | , |
Token ID | 11 |
Feature activation | +2.89e-01 |
Pos logit contributions | |
parties | +0.019 |
tea | +0.017 |
recess | +0.013 |
celebrate | +0.013 |
party | +0.012 |
Neg logit contributions | |
beast | -0.031 |
gut | -0.031 |
mistake | -0.029 |
inct | -0.029 |
therm | -0.029 |
Token | " |
Token ID | 366 |
Feature activation | +3.58e+00 |
Pos logit contributions | |
gut | +0.010 |
beast | +0.010 |
severe | +0.009 |
inct | +0.009 |
mistake | +0.009 |
Neg logit contributions | |
parties | -0.009 |
tea | -0.009 |
together | -0.007 |
played | -0.007 |
celebrate | -0.007 |
Token | I |
Token ID | 40 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
gut | +0.142 |
beast | +0.139 |
mistake | +0.134 |
inct | +0.128 |
therm | +0.128 |
Neg logit contributions | |
parties | -0.097 |
tea | -0.085 |
celebrate | -0.069 |
recess | -0.068 |
wear | -0.064 |
Token | ?" |
Token ID | 1701 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
parties | +0.023 |
tea | +0.020 |
recess | +0.016 |
celebrate | +0.016 |
party | +0.015 |
Neg logit contributions | |
gut | -0.032 |
beast | -0.031 |
inct | -0.030 |
mistake | -0.029 |
swollen | -0.029 |
Token | |
Token ID | 220 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
swollen | +0.088 |
gut | +0.088 |
mistake | +0.086 |
severe | +0.086 |
beast | +0.086 |
Neg logit contributions | |
parties | -0.049 |
tea | -0.040 |
recess | -0.033 |
celebrate | -0.032 |
wear | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | +6.75e-01 |
Pos logit contributions | |
beast | +0.061 |
gut | +0.061 |
mistake | +0.057 |
therm | +0.057 |
swollen | +0.057 |
Neg logit contributions | |
parties | -0.048 |
tea | -0.042 |
recess | -0.035 |
celebrate | -0.033 |
party | -0.032 |
Token |
|
Token ID | 198 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
gut | +0.029 |
beast | +0.029 |
mistake | +0.028 |
swollen | +0.028 |
therm | +0.028 |
Neg logit contributions | |
parties | -0.015 |
tea | -0.012 |
recess | -0.010 |
celebrate | -0.010 |
wear | -0.009 |
Token | The |
Token ID | 464 |
Feature activation | +4.62e+00 |
Pos logit contributions | |
mistake | +0.044 |
therm | +0.043 |
gut | +0.043 |
swollen | +0.043 |
beast | +0.042 |
Neg logit contributions | |
parties | -0.028 |
P | -0.023 |
tea | -0.022 |
celebrate | -0.022 |
recess | -0.022 |
Token | refrigerator |
Token ID | 30500 |
Feature activation | +8.92e+00 |
Pos logit contributions | |
gut | +0.144 |
rum | +0.137 |
beast | +0.130 |
inct | +0.129 |
shadow | +0.126 |
Neg logit contributions | |
tea | -0.165 |
parties | -0.161 |
party | -0.145 |
birthday | -0.132 |
picnic | -0.129 |
Token | replied |
Token ID | 8712 |
Feature activation | +9.07e-02 |
Pos logit contributions | |
gut | +0.342 |
rum | +0.316 |
beast | +0.313 |
severe | +0.310 |
mistake | +0.300 |
Neg logit contributions | |
parties | -0.242 |
tea | -0.231 |
party | -0.200 |
celebrate | -0.186 |
picnic | -0.184 |
Token | , |
Token ID | 11 |
Feature activation | +9.91e-01 |
Pos logit contributions | |
beast | +0.004 |
gut | +0.004 |
mistake | +0.004 |
swollen | +0.004 |
inct | +0.004 |
Neg logit contributions | |
parties | -0.002 |
tea | -0.002 |
recess | -0.001 |
celebrate | -0.001 |
party | -0.001 |
Token | " |
Token ID | 366 |
Feature activation | +3.89e+00 |
Pos logit contributions | |
beast | +0.036 |
gut | +0.036 |
severe | +0.035 |
mistake | +0.034 |
swollen | +0.033 |
Neg logit contributions | |
parties | -0.027 |
tea | -0.027 |
together | -0.023 |
them | -0.021 |
celebrate | -0.021 |
Token | Yes |
Token ID | 5297 |
Feature activation | -2.75e+00 |
Pos logit contributions | |
gut | +0.132 |
beast | +0.132 |
mistake | +0.126 |
therm | +0.122 |
inct | +0.121 |
Neg logit contributions | |
parties | -0.127 |
tea | -0.112 |
recess | -0.096 |
celebrate | -0.094 |
party | -0.089 |
Token | , |
Token ID | 11 |
Feature activation | -2.14e+00 |
Pos logit contributions | |
parties | +0.085 |
tea | +0.072 |
recess | +0.065 |
celebrate | +0.060 |
picnic | +0.057 |
Neg logit contributions | |
beast | -0.099 |
gut | -0.098 |
mistake | -0.093 |
therm | -0.091 |
inct | -0.090 |
Token | Billy |
Token ID | 15890 |
Feature activation | +8.00e-01 |
Pos logit contributions | |
parties | +0.044 |
tea | +0.039 |
celebrate | +0.031 |
recess | +0.030 |
wear | +0.029 |
Neg logit contributions | |
beast | -0.061 |
gut | -0.059 |
mistake | -0.057 |
inct | -0.057 |
swollen | -0.056 |
Token | . |
Token ID | 13 |
Feature activation | +6.09e-01 |
Pos logit contributions | |
gut | +0.029 |
beast | +0.029 |
inct | +0.028 |
mistake | +0.027 |
swollen | +0.027 |
Neg logit contributions | |
parties | -0.023 |
tea | -0.021 |
celebrate | -0.016 |
wear | -0.016 |
recess | -0.016 |
Token | The |
Token ID | 383 |
Feature activation | +3.42e+00 |
Pos logit contributions | |
beast | +0.022 |
gut | +0.022 |
mistake | +0.021 |
swollen | +0.020 |
inct | +0.020 |
Neg logit contributions | |
parties | -0.018 |
tea | -0.016 |
celebrate | -0.013 |
recess | -0.013 |
wear | -0.012 |
Token | rh |
Token ID | 9529 |
Feature activation | +4.78e+00 |
Pos logit contributions | |
gut | +0.117 |
rum | +0.106 |
inct | +0.103 |
outage | +0.101 |
severe | +0.100 |
Neg logit contributions | |
parties | -0.106 |
tea | -0.103 |
picnic | -0.092 |
party | -0.090 |
celebrate | -0.087 |
Token | in |
Token ID | 259 |
Feature activation | +5.05e+00 |
Pos logit contributions | |
gut | +0.205 |
plea | +0.190 |
inct | +0.188 |
severe | +0.184 |
iously | +0.182 |
Neg logit contributions | |
parties | -0.107 |
tea | -0.086 |
recess | -0.070 |
acky | -0.069 |
together | -0.069 |
Token | oc |
Token ID | 420 |
Feature activation | +8.88e+00 |
Pos logit contributions | |
gut | +0.171 |
reckless | +0.148 |
severe | +0.147 |
vain | +0.145 |
mistake | +0.145 |
Neg logit contributions | |
tea | -0.134 |
played | -0.132 |
parties | -0.126 |
play | -0.121 |
picnic | -0.120 |
Token | eros |
Token ID | 27498 |
Feature activation | +4.43e+00 |
Pos logit contributions | |
severe | +0.302 |
outage | +0.292 |
swollen | +0.275 |
gut | +0.274 |
reckless | +0.266 |
Neg logit contributions | |
parties | -0.253 |
tea | -0.213 |
ils | -0.207 |
recess | -0.188 |
sandwiches | -0.182 |
Token | looked |
Token ID | 3114 |
Feature activation | -9.23e-01 |
Pos logit contributions | |
gut | +0.168 |
beast | +0.160 |
inct | +0.153 |
mistake | +0.150 |
severe | +0.150 |
Neg logit contributions | |
parties | -0.123 |
tea | -0.115 |
played | -0.096 |
party | -0.092 |
picnic | -0.090 |
Token | at |
Token ID | 379 |
Feature activation | +3.13e+00 |
Pos logit contributions | |
parties | +0.029 |
tea | +0.025 |
recess | +0.022 |
celebrate | +0.020 |
party | +0.019 |
Neg logit contributions | |
gut | -0.033 |
beast | -0.032 |
mistake | -0.031 |
therm | -0.030 |
swollen | -0.030 |
Token | Billy |
Token ID | 15890 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
beast | +0.128 |
gut | +0.127 |
mistake | +0.121 |
therm | +0.119 |
inct | +0.119 |
Neg logit contributions | |
parties | -0.081 |
tea | -0.069 |
recess | -0.056 |
celebrate | -0.053 |
party | -0.050 |
Token | with |
Token ID | 351 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
parties | +0.029 |
tea | +0.026 |
celebrate | +0.021 |
picnic | +0.020 |
wear | +0.020 |
Neg logit contributions | |
gut | -0.046 |
beast | -0.045 |
inct | -0.043 |
swollen | -0.042 |
severe | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | -7.72e-01 |
Pos logit contributions | |
gut | +0.067 |
beast | +0.066 |
inct | +0.062 |
cut | +0.062 |
passer | +0.062 |
Neg logit contributions | |
parties | -0.034 |
tea | -0.028 |
celebrate | -0.020 |
recess | -0.020 |
party | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -8.48e-01 |
Pos logit contributions | |
parties | +0.021 |
tea | +0.018 |
recess | +0.015 |
celebrate | +0.015 |
party | +0.013 |
Neg logit contributions | |
gut | -0.030 |
beast | -0.030 |
mistake | -0.029 |
swollen | -0.029 |
therm | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | +4.73e-02 |
Pos logit contributions | |
parties | +0.028 |
tea | +0.024 |
recess | +0.021 |
party | +0.019 |
celebrate | +0.019 |
Neg logit contributions | |
beast | -0.029 |
gut | -0.029 |
mistake | -0.027 |
inct | -0.027 |
therm | -0.027 |
Token | The |
Token ID | 464 |
Feature activation | +3.72e+00 |
Pos logit contributions | |
beast | +0.001 |
gut | +0.001 |
mistake | +0.001 |
therm | +0.001 |
inct | +0.001 |
Neg logit contributions | |
parties | -0.002 |
tea | -0.002 |
recess | -0.001 |
celebrate | -0.001 |
party | -0.001 |
Token | av |
Token ID | 1196 |
Feature activation | +5.59e+00 |
Pos logit contributions | |
gut | +0.117 |
rum | +0.114 |
inct | +0.113 |
beast | +0.112 |
outage | +0.110 |
Neg logit contributions | |
parties | -0.127 |
tea | -0.116 |
party | -0.105 |
picnic | -0.100 |
celebrate | -0.099 |
Token | oc |
Token ID | 420 |
Feature activation | +8.88e+00 |
Pos logit contributions | |
gut | +0.153 |
mistake | +0.137 |
beast | +0.137 |
therm | +0.129 |
reckless | +0.126 |
Neg logit contributions | |
parties | -0.214 |
tea | -0.195 |
celebrate | -0.179 |
ils | -0.176 |
ets | -0.174 |
Token | ados |
Token ID | 22484 |
Feature activation | +3.73e+00 |
Pos logit contributions | |
outage | +0.380 |
gut | +0.358 |
severe | +0.358 |
swollen | +0.353 |
therm | +0.352 |
Neg logit contributions | |
parties | -0.196 |
tea | -0.144 |
ils | -0.143 |
recess | -0.121 |
party | -0.120 |
Token | were |
Token ID | 547 |
Feature activation | +5.69e-01 |
Pos logit contributions | |
gut | +0.166 |
beast | +0.163 |
mistake | +0.156 |
inct | +0.154 |
therm | +0.153 |
Neg logit contributions | |
parties | -0.086 |
tea | -0.070 |
celebrate | -0.060 |
party | -0.056 |
picnic | -0.052 |
Token | shared |
Token ID | 4888 |
Feature activation | -3.24e+00 |
Pos logit contributions | |
beast | +0.022 |
gut | +0.021 |
mistake | +0.020 |
inct | +0.020 |
therm | +0.020 |
Neg logit contributions | |
parties | -0.017 |
tea | -0.015 |
celebrate | -0.012 |
recess | -0.012 |
party | -0.011 |
Token | with |
Token ID | 351 |
Feature activation | -2.74e+00 |
Pos logit contributions | |
parties | +0.096 |
tea | +0.084 |
recess | +0.072 |
celebrate | +0.066 |
party | +0.064 |
Neg logit contributions | |
beast | -0.118 |
gut | -0.118 |
mistake | -0.112 |
inct | -0.109 |
therm | -0.109 |
Token | everybody |
Token ID | 7288 |
Feature activation | -2.78e+00 |
Pos logit contributions | |
parties | +0.071 |
tea | +0.062 |
celebrate | +0.050 |
recess | +0.049 |
party | +0.044 |
Neg logit contributions | |
beast | -0.112 |
gut | -0.111 |
therm | -0.106 |
inct | -0.105 |
mistake | -0.104 |
Token |
|
Token ID | 198 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
beast | +0.011 |
gut | +0.011 |
mistake | +0.010 |
therm | +0.010 |
inct | +0.010 |
Neg logit contributions | |
parties | -0.013 |
tea | -0.012 |
recess | -0.010 |
celebrate | -0.010 |
party | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | +8.35e-01 |
Pos logit contributions | |
beast | +0.009 |
gut | +0.009 |
mistake | +0.009 |
inct | +0.008 |
therm | +0.008 |
Neg logit contributions | |
parties | -0.008 |
tea | -0.007 |
recess | -0.006 |
celebrate | -0.005 |
party | -0.005 |
Token | The |
Token ID | 464 |
Feature activation | +4.09e+00 |
Pos logit contributions | |
beast | +0.027 |
gut | +0.027 |
mistake | +0.026 |
therm | +0.025 |
inct | +0.025 |
Neg logit contributions | |
parties | -0.028 |
tea | -0.025 |
recess | -0.022 |
celebrate | -0.021 |
party | -0.020 |
Token | rh |
Token ID | 9529 |
Feature activation | +4.56e+00 |
Pos logit contributions | |
gut | +0.131 |
rum | +0.117 |
inct | +0.117 |
beast | +0.117 |
severe | +0.110 |
Neg logit contributions | |
parties | -0.140 |
tea | -0.131 |
picnic | -0.112 |
celebrate | -0.112 |
party | -0.112 |
Token | in |
Token ID | 259 |
Feature activation | +4.92e+00 |
Pos logit contributions | |
gut | +0.188 |
iously | +0.181 |
plea | +0.177 |
inct | +0.175 |
severe | +0.174 |
Neg logit contributions | |
parties | -0.104 |
tea | -0.089 |
recess | -0.073 |
together | -0.069 |
acky | -0.068 |
Token | oc |
Token ID | 420 |
Feature activation | +8.87e+00 |
Pos logit contributions | |
gut | +0.165 |
severe | +0.144 |
reckless | +0.141 |
adult | +0.140 |
rum | +0.139 |
Neg logit contributions | |
tea | -0.144 |
parties | -0.133 |
played | -0.132 |
picnic | -0.127 |
play | -0.124 |
Token | eros |
Token ID | 27498 |
Feature activation | +4.28e+00 |
Pos logit contributions | |
severe | +0.299 |
swollen | +0.274 |
outage | +0.273 |
gut | +0.269 |
reckless | +0.269 |
Neg logit contributions | |
parties | -0.244 |
tea | -0.212 |
recess | -0.189 |
ils | -0.186 |
sandwiches | -0.184 |
Token | nodded |
Token ID | 14464 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
gut | +0.165 |
beast | +0.153 |
severe | +0.151 |
swollen | +0.150 |
rum | +0.150 |
Neg logit contributions | |
parties | -0.119 |
tea | -0.112 |
played | -0.091 |
party | -0.091 |
celebrate | -0.088 |
Token | its |
Token ID | 663 |
Feature activation | +2.85e+00 |
Pos logit contributions | |
gut | +0.063 |
beast | +0.063 |
mistake | +0.060 |
therm | +0.059 |
swollen | +0.058 |
Neg logit contributions | |
parties | -0.069 |
tea | -0.058 |
recess | -0.053 |
celebrate | -0.048 |
picnic | -0.048 |
Token | head |
Token ID | 1182 |
Feature activation | +3.32e+00 |
Pos logit contributions | |
beast | +0.089 |
cut | +0.084 |
gut | +0.083 |
passer | +0.083 |
therm | +0.081 |
Neg logit contributions | |
tea | -0.081 |
parties | -0.078 |
birthday | -0.069 |
play | -0.069 |
pretend | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
beast | +0.134 |
gut | +0.133 |
mistake | +0.126 |
inct | +0.125 |
therm | +0.124 |
Neg logit contributions | |
parties | -0.086 |
tea | -0.075 |
recess | -0.059 |
celebrate | -0.057 |
party | -0.054 |
Token |
|
Token ID | 198 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
beast | +0.044 |
gut | +0.043 |
mistake | +0.041 |
swollen | +0.041 |
therm | +0.041 |
Neg logit contributions | |
parties | -0.029 |
tea | -0.024 |
recess | -0.019 |
celebrate | -0.019 |
picnic | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | +6.85e-01 |
Pos logit contributions | |
beast | +0.007 |
gut | +0.007 |
mistake | +0.006 |
inct | +0.006 |
therm | +0.006 |
Neg logit contributions | |
parties | -0.006 |
tea | -0.005 |
recess | -0.004 |
celebrate | -0.004 |
party | -0.004 |
Token | The |
Token ID | 464 |
Feature activation | +4.33e+00 |
Pos logit contributions | |
beast | +0.025 |
gut | +0.024 |
mistake | +0.023 |
therm | +0.023 |
inct | +0.023 |
Neg logit contributions | |
parties | -0.021 |
tea | -0.017 |
celebrate | -0.016 |
recess | -0.015 |
picnic | -0.014 |
Token | rh |
Token ID | 9529 |
Feature activation | +5.02e+00 |
Pos logit contributions | |
gut | +0.137 |
rum | +0.129 |
inct | +0.122 |
outage | +0.121 |
beast | +0.119 |
Neg logit contributions | |
parties | -0.140 |
tea | -0.140 |
picnic | -0.123 |
party | -0.121 |
celebrate | -0.115 |
Token | in |
Token ID | 259 |
Feature activation | +5.29e+00 |
Pos logit contributions | |
gut | +0.213 |
plea | +0.196 |
inct | +0.196 |
therm | +0.194 |
severe | +0.193 |
Neg logit contributions | |
parties | -0.100 |
tea | -0.087 |
acky | -0.078 |
together | -0.075 |
P | -0.074 |
Token | oc |
Token ID | 420 |
Feature activation | +8.86e+00 |
Pos logit contributions | |
gut | +0.181 |
outage | +0.162 |
reckless | +0.162 |
swollen | +0.160 |
vain | +0.158 |
Neg logit contributions | |
played | -0.133 |
tea | -0.133 |
together | -0.123 |
acky | -0.122 |
picnic | -0.121 |
Token | eros |
Token ID | 27498 |
Feature activation | +4.80e+00 |
Pos logit contributions | |
outage | +0.297 |
severe | +0.296 |
swollen | +0.271 |
therm | +0.267 |
gut | +0.266 |
Neg logit contributions | |
parties | -0.230 |
ils | -0.204 |
tea | -0.201 |
sandwiches | -0.180 |
recess | -0.170 |
Token | couldn |
Token ID | 3521 |
Feature activation | +3.50e+00 |
Pos logit contributions | |
gut | +0.180 |
beast | +0.174 |
swollen | +0.171 |
therm | +0.169 |
severe | +0.165 |
Neg logit contributions | |
parties | -0.136 |
tea | -0.122 |
played | -0.105 |
party | -0.101 |
celebrate | -0.100 |
Token | 't |
Token ID | 470 |
Feature activation | +2.72e+00 |
Pos logit contributions | |
gut | +0.103 |
beast | +0.100 |
swollen | +0.093 |
mistake | +0.093 |
inct | +0.091 |
Neg logit contributions | |
parties | -0.130 |
tea | -0.118 |
celebrate | -0.102 |
recess | -0.100 |
wear | -0.096 |
Token | believe |
Token ID | 1975 |
Feature activation | -3.93e-01 |
Pos logit contributions | |
beast | +0.099 |
gut | +0.098 |
mistake | +0.093 |
therm | +0.090 |
cut | +0.089 |
Neg logit contributions | |
parties | -0.082 |
tea | -0.073 |
recess | -0.061 |
picnic | -0.055 |
celebrate | -0.054 |
Token | his |
Token ID | 465 |
Feature activation | +9.97e-01 |
Pos logit contributions | |
parties | +0.013 |
tea | +0.011 |
recess | +0.009 |
celebrate | +0.009 |
party | +0.009 |
Neg logit contributions | |
beast | -0.014 |
gut | -0.013 |
inct | -0.013 |
mistake | -0.013 |
therm | -0.013 |
Token | happy |
Token ID | 3772 |
Feature activation | +3.01e-01 |
Pos logit contributions | |
beast | +0.013 |
gut | +0.013 |
mistake | +0.012 |
inct | +0.012 |
therm | +0.012 |
Neg logit contributions | |
parties | -0.010 |
tea | -0.009 |
recess | -0.007 |
celebrate | -0.007 |
party | -0.007 |
Token | ." |
Token ID | 526 |
Feature activation | +4.97e-01 |
Pos logit contributions | |
beast | +0.012 |
gut | +0.012 |
mistake | +0.011 |
therm | +0.011 |
inct | +0.011 |
Neg logit contributions | |
parties | -0.008 |
tea | -0.007 |
recess | -0.006 |
celebrate | -0.006 |
party | -0.006 |
Token | The |
Token ID | 383 |
Feature activation | +4.04e+00 |
Pos logit contributions | |
gut | +0.018 |
beast | +0.018 |
mistake | +0.017 |
swollen | +0.017 |
therm | +0.017 |
Neg logit contributions | |
parties | -0.015 |
tea | -0.013 |
celebrate | -0.011 |
recess | -0.011 |
picnic | -0.010 |
Token | rh |
Token ID | 9529 |
Feature activation | +5.21e+00 |
Pos logit contributions | |
gut | +0.125 |
rum | +0.114 |
beast | +0.113 |
inct | +0.109 |
severe | +0.104 |
Neg logit contributions | |
parties | -0.142 |
tea | -0.137 |
party | -0.119 |
picnic | -0.119 |
celebrate | -0.117 |
Token | in |
Token ID | 259 |
Feature activation | +5.57e+00 |
Pos logit contributions | |
gut | +0.213 |
inct | +0.196 |
plea | +0.195 |
therm | +0.194 |
beast | +0.192 |
Neg logit contributions | |
parties | -0.119 |
tea | -0.102 |
together | -0.086 |
acky | -0.085 |
celebrate | -0.083 |
Token | oc |
Token ID | 420 |
Feature activation | +8.85e+00 |
Pos logit contributions | |
gut | +0.189 |
vain | +0.172 |
reckless | +0.168 |
mistake | +0.164 |
swollen | +0.164 |
Neg logit contributions | |
played | -0.145 |
tea | -0.143 |
together | -0.130 |
celebrate | -0.128 |
play | -0.127 |
Token | eros |
Token ID | 27498 |
Feature activation | +4.98e+00 |
Pos logit contributions | |
severe | +0.280 |
outage | +0.267 |
gut | +0.258 |
swollen | +0.257 |
therm | +0.255 |
Neg logit contributions | |
parties | -0.268 |
tea | -0.228 |
ils | -0.209 |
sandwiches | -0.202 |
recess | -0.196 |
Token | did |
Token ID | 750 |
Feature activation | +2.55e+00 |
Pos logit contributions | |
gut | +0.189 |
beast | +0.183 |
swollen | +0.182 |
therm | +0.178 |
mistake | +0.174 |
Neg logit contributions | |
parties | -0.129 |
tea | -0.121 |
played | -0.110 |
party | -0.099 |
celebrate | -0.098 |
Token | what |
Token ID | 644 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
beast | +0.086 |
gut | +0.083 |
swollen | +0.081 |
therm | +0.080 |
cut | +0.080 |
Neg logit contributions | |
parties | -0.084 |
tea | -0.075 |
celebrate | -0.063 |
recess | -0.061 |
wear | -0.060 |
Token | the |
Token ID | 262 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
parties | +0.006 |
tea | +0.005 |
celebrate | +0.004 |
recess | +0.004 |
played | +0.004 |
Neg logit contributions | |
gut | -0.009 |
beast | -0.009 |
inct | -0.009 |
swollen | -0.009 |
plea | -0.009 |
Token | bird |
Token ID | 6512 |
Feature activation | +3.55e+00 |
Pos logit contributions | |
gut | +0.046 |
beast | +0.042 |
rum | +0.041 |
inct | +0.039 |
swollen | +0.039 |
Neg logit contributions | |
parties | -0.056 |
tea | -0.054 |
picnic | -0.047 |
party | -0.047 |
celebrate | -0.046 |
Token |
|
Token ID | 198 |
Feature activation | -5.18e-01 |
Pos logit contributions | |
parties | +0.006 |
tea | +0.005 |
recess | +0.004 |
celebrate | +0.004 |
played | +0.004 |
Neg logit contributions | |
beast | -0.007 |
gut | -0.007 |
mistake | -0.007 |
swollen | -0.007 |
severe | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
parties | +0.014 |
tea | +0.012 |
recess | +0.009 |
celebrate | +0.009 |
wear | +0.008 |
Neg logit contributions | |
beast | -0.021 |
gut | -0.020 |
mistake | -0.019 |
therm | -0.019 |
swollen | -0.019 |
Token | The |
Token ID | 464 |
Feature activation | +3.98e+00 |
Pos logit contributions | |
beast | +0.010 |
mistake | +0.010 |
gut | +0.010 |
therm | +0.010 |
swollen | +0.010 |
Neg logit contributions | |
parties | -0.008 |
tea | -0.007 |
celebrate | -0.006 |
P | -0.006 |
recess | -0.006 |
Token | rh |
Token ID | 9529 |
Feature activation | +4.83e+00 |
Pos logit contributions | |
gut | +0.120 |
rum | +0.111 |
inct | +0.108 |
beast | +0.107 |
severe | +0.101 |
Neg logit contributions | |
parties | -0.139 |
tea | -0.132 |
picnic | -0.119 |
party | -0.116 |
celebrate | -0.116 |
Token | in |
Token ID | 259 |
Feature activation | +5.04e+00 |
Pos logit contributions | |
gut | +0.198 |
inct | +0.183 |
plea | +0.182 |
therm | +0.180 |
iously | +0.177 |
Neg logit contributions | |
parties | -0.107 |
tea | -0.090 |
acky | -0.078 |
together | -0.076 |
celebrate | -0.075 |
Token | oc |
Token ID | 420 |
Feature activation | +8.85e+00 |
Pos logit contributions | |
gut | +0.172 |
outage | +0.157 |
reckless | +0.153 |
vain | +0.151 |
rier | +0.149 |
Neg logit contributions | |
played | -0.125 |
tea | -0.121 |
picnic | -0.117 |
play | -0.113 |
parties | -0.113 |
Token | eros |
Token ID | 27498 |
Feature activation | +4.26e+00 |
Pos logit contributions | |
outage | +0.277 |
severe | +0.276 |
swollen | +0.258 |
therm | +0.257 |
gut | +0.257 |
Neg logit contributions | |
parties | -0.260 |
tea | -0.220 |
ils | -0.211 |
sandwiches | -0.199 |
recess | -0.197 |
Token | was |
Token ID | 373 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
gut | +0.163 |
beast | +0.156 |
therm | +0.153 |
swollen | +0.153 |
inct | +0.149 |
Neg logit contributions | |
parties | -0.112 |
tea | -0.102 |
played | -0.091 |
celebrate | -0.086 |
party | -0.085 |
Token | so |
Token ID | 523 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
beast | +0.074 |
gut | +0.073 |
inct | +0.071 |
therm | +0.068 |
mistake | +0.068 |
Neg logit contributions | |
parties | -0.053 |
tea | -0.048 |
celebrate | -0.039 |
party | -0.036 |
picnic | -0.036 |
Token | delighted |
Token ID | 20707 |
Feature activation | +2.57e+00 |
Pos logit contributions | |
beast | +0.059 |
gut | +0.057 |
inct | +0.055 |
mistake | +0.054 |
therm | +0.054 |
Neg logit contributions | |
parties | -0.039 |
tea | -0.034 |
celebrate | -0.029 |
wear | -0.026 |
picnic | -0.026 |
Token | with |
Token ID | 351 |
Feature activation | +6.15e-01 |
Pos logit contributions | |
gut | +0.097 |
beast | +0.096 |
swollen | +0.092 |
mistake | +0.092 |
inct | +0.090 |
Neg logit contributions | |
parties | -0.074 |
tea | -0.064 |
recess | -0.051 |
celebrate | -0.050 |
party | -0.050 |
Token | help |
Token ID | 1037 |
Feature activation | +6.02e-01 |
Pos logit contributions | |
gut | +0.163 |
beast | +0.161 |
inct | +0.155 |
swollen | +0.150 |
mistake | +0.149 |
Neg logit contributions | |
parties | -0.117 |
tea | -0.102 |
celebrate | -0.086 |
recess | -0.081 |
party | -0.080 |
Token | . |
Token ID | 13 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
gut | +0.028 |
beast | +0.028 |
inct | +0.028 |
swollen | +0.027 |
guardian | +0.026 |
Neg logit contributions | |
parties | -0.012 |
tea | -0.009 |
celebrate | -0.007 |
wear | -0.006 |
recess | -0.006 |
Token | The |
Token ID | 383 |
Feature activation | +3.64e+00 |
Pos logit contributions | |
gut | +0.005 |
beast | +0.005 |
mistake | +0.005 |
swollen | +0.005 |
guardian | +0.005 |
Neg logit contributions | |
parties | -0.002 |
tea | -0.002 |
played | -0.001 |
celebrate | -0.001 |
together | -0.001 |
Token | rh |
Token ID | 9529 |
Feature activation | +5.15e+00 |
Pos logit contributions | |
gut | +0.104 |
beast | +0.096 |
inct | +0.095 |
rum | +0.090 |
therm | +0.090 |
Neg logit contributions | |
parties | -0.139 |
tea | -0.128 |
party | -0.111 |
recess | -0.110 |
picnic | -0.109 |
Token | in |
Token ID | 259 |
Feature activation | +5.22e+00 |
Pos logit contributions | |
gut | +0.203 |
therm | +0.186 |
inct | +0.183 |
beast | +0.183 |
plea | +0.182 |
Neg logit contributions | |
parties | -0.137 |
tea | -0.115 |
together | -0.097 |
celebrate | -0.095 |
party | -0.095 |
Token | oc |
Token ID | 420 |
Feature activation | +8.84e+00 |
Pos logit contributions | |
gut | +0.183 |
reckless | +0.157 |
swollen | +0.155 |
outage | +0.152 |
mistake | +0.152 |
Neg logit contributions | |
tea | -0.136 |
parties | -0.133 |
played | -0.130 |
acky | -0.120 |
picnic | -0.119 |
Token | eros |
Token ID | 27498 |
Feature activation | +4.77e+00 |
Pos logit contributions | |
severe | +0.255 |
gut | +0.243 |
swollen | +0.242 |
therm | +0.239 |
outage | +0.238 |
Neg logit contributions | |
parties | -0.309 |
tea | -0.270 |
ils | -0.249 |
recess | -0.247 |
sandwiches | -0.232 |
Token | was |
Token ID | 373 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
gut | +0.205 |
outage | +0.191 |
beast | +0.190 |
rum | +0.187 |
swollen | +0.183 |
Neg logit contributions | |
played | -0.099 |
tea | -0.092 |
parties | -0.089 |
shared | -0.080 |
together | -0.079 |
Token | very |
Token ID | 845 |
Feature activation | +3.48e+00 |
Pos logit contributions | |
beast | +0.088 |
gut | +0.088 |
inct | +0.086 |
mistake | +0.080 |
adult | +0.079 |
Neg logit contributions | |
parties | -0.056 |
tea | -0.055 |
together | -0.044 |
celebrate | -0.043 |
played | -0.043 |
Token | happy |
Token ID | 3772 |
Feature activation | +8.70e-01 |
Pos logit contributions | |
beast | +0.143 |
gut | +0.133 |
inct | +0.132 |
der | +0.124 |
therm | +0.123 |
Neg logit contributions | |
parties | -0.084 |
tea | -0.079 |
wear | -0.067 |
celebrate | -0.066 |
together | -0.066 |
Token | and |
Token ID | 290 |
Feature activation | +7.68e-01 |
Pos logit contributions | |
gut | +0.035 |
beast | +0.035 |
inct | +0.034 |
swollen | +0.034 |
severe | +0.033 |
Neg logit contributions | |
parties | -0.023 |
tea | -0.020 |
party | -0.015 |
together | -0.015 |
recess | -0.015 |
Token | ," |
Token ID | 553 |
Feature activation | -3.38e-02 |
Pos logit contributions | |
gut | +0.115 |
beast | +0.115 |
inct | +0.110 |
therm | +0.108 |
mistake | +0.108 |
Neg logit contributions | |
parties | -0.074 |
tea | -0.063 |
recess | -0.049 |
celebrate | -0.048 |
party | -0.046 |
Token | said |
Token ID | 531 |
Feature activation | -7.64e-01 |
Pos logit contributions | |
parties | +0.001 |
tea | +0.001 |
recess | +0.001 |
celebrate | +0.001 |
party | +0.001 |
Neg logit contributions | |
beast | -0.001 |
gut | -0.001 |
mistake | -0.001 |
inct | -0.001 |
therm | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | +3.56e+00 |
Pos logit contributions | |
parties | +0.021 |
tea | +0.018 |
recess | +0.015 |
celebrate | +0.014 |
party | +0.014 |
Neg logit contributions | |
beast | -0.030 |
gut | -0.030 |
mistake | -0.028 |
therm | -0.028 |
inct | -0.028 |
Token | rh |
Token ID | 9529 |
Feature activation | +4.77e+00 |
Pos logit contributions | |
gut | +0.125 |
rum | +0.116 |
severe | +0.107 |
outage | +0.107 |
inct | +0.107 |
Neg logit contributions | |
parties | -0.100 |
tea | -0.094 |
celebrate | -0.085 |
party | -0.084 |
picnic | -0.082 |
Token | in |
Token ID | 259 |
Feature activation | +5.39e+00 |
Pos logit contributions | |
gut | +0.195 |
inct | +0.183 |
plea | +0.181 |
mistake | +0.178 |
swollen | +0.177 |
Neg logit contributions | |
parties | -0.107 |
tea | -0.091 |
acky | -0.075 |
recess | -0.073 |
together | -0.073 |
Token | oc |
Token ID | 420 |
Feature activation | +8.84e+00 |
Pos logit contributions | |
gut | +0.185 |
swollen | +0.174 |
shattering | +0.161 |
refusing | +0.160 |
severe | +0.159 |
Neg logit contributions | |
together | -0.124 |
tea | -0.124 |
parties | -0.119 |
played | -0.114 |
named | -0.114 |
Token | eros |
Token ID | 27498 |
Feature activation | +4.79e+00 |
Pos logit contributions | |
severe | +0.289 |
outage | +0.278 |
swollen | +0.270 |
gut | +0.267 |
confusion | +0.263 |
Neg logit contributions | |
parties | -0.229 |
tea | -0.193 |
ils | -0.190 |
sandwiches | -0.171 |
recess | -0.170 |
Token | . |
Token ID | 13 |
Feature activation | +3.73e-01 |
Pos logit contributions | |
gut | +0.216 |
beast | +0.214 |
inct | +0.206 |
swollen | +0.205 |
mistake | +0.204 |
Neg logit contributions | |
parties | -0.084 |
tea | -0.079 |
together | -0.053 |
played | -0.051 |
celebrate | -0.050 |
Token | " |
Token ID | 366 |
Feature activation | +3.55e+00 |
Pos logit contributions | |
beast | +0.011 |
gut | +0.011 |
mistake | +0.010 |
swollen | +0.010 |
cut | +0.010 |
Neg logit contributions | |
parties | -0.014 |
tea | -0.013 |
celebrate | -0.011 |
recess | -0.011 |
party | -0.010 |
Token | I |
Token ID | 40 |
Feature activation | -2.26e+00 |
Pos logit contributions | |
beast | +0.129 |
gut | +0.127 |
mistake | +0.123 |
therm | +0.118 |
swollen | +0.116 |
Neg logit contributions | |
parties | -0.104 |
tea | -0.092 |
celebrate | -0.080 |
recess | -0.078 |
wear | -0.073 |
Token | 'm |
Token ID | 1101 |
Feature activation | +4.57e-01 |
Pos logit contributions | |
parties | +0.064 |
tea | +0.060 |
celebrate | +0.051 |
wear | +0.050 |
picnic | +0.047 |
Neg logit contributions | |
beast | -0.081 |
gut | -0.081 |
swollen | -0.077 |
passer | -0.076 |
mistake | -0.075 |
Token | Suddenly |
Token ID | 38582 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
beast | +0.007 |
gut | +0.007 |
therm | +0.007 |
swollen | +0.007 |
mistake | +0.007 |
Neg logit contributions | |
parties | -0.009 |
tea | -0.007 |
recess | -0.007 |
celebrate | -0.007 |
party | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
gut | +0.094 |
beast | +0.093 |
mistake | +0.090 |
swollen | +0.089 |
inct | +0.088 |
Neg logit contributions | |
parties | -0.054 |
tea | -0.047 |
recess | -0.037 |
celebrate | -0.034 |
picnic | -0.033 |
Token | the |
Token ID | 262 |
Feature activation | +4.31e+00 |
Pos logit contributions | |
gut | +0.042 |
mistake | +0.041 |
beast | +0.040 |
swollen | +0.040 |
rum | +0.039 |
Neg logit contributions | |
parties | -0.029 |
tea | -0.026 |
recess | -0.021 |
celebrate | -0.020 |
picnic | -0.019 |
Token | rh |
Token ID | 9529 |
Feature activation | +5.31e+00 |
Pos logit contributions | |
gut | +0.133 |
rum | +0.124 |
mistake | +0.120 |
beast | +0.119 |
inct | +0.116 |
Neg logit contributions | |
parties | -0.152 |
tea | -0.141 |
picnic | -0.126 |
party | -0.124 |
celebrate | -0.124 |
Token | in |
Token ID | 259 |
Feature activation | +5.39e+00 |
Pos logit contributions | |
gut | +0.224 |
plea | +0.208 |
mistake | +0.204 |
iously | +0.203 |
inct | +0.203 |
Neg logit contributions | |
parties | -0.117 |
tea | -0.097 |
acky | -0.083 |
together | -0.080 |
celebrate | -0.078 |
Token | oc |
Token ID | 420 |
Feature activation | +8.84e+00 |
Pos logit contributions | |
gut | +0.190 |
rum | +0.175 |
rier | +0.172 |
mistake | +0.172 |
vain | +0.171 |
Neg logit contributions | |
played | -0.136 |
tea | -0.134 |
picnic | -0.125 |
party | -0.120 |
play | -0.120 |
Token | eros |
Token ID | 27498 |
Feature activation | +4.66e+00 |
Pos logit contributions | |
severe | +0.279 |
swollen | +0.266 |
outage | +0.263 |
gut | +0.253 |
reckless | +0.253 |
Neg logit contributions | |
parties | -0.276 |
tea | -0.231 |
ils | -0.211 |
recess | -0.205 |
celebrate | -0.201 |
Token | saw |
Token ID | 2497 |
Feature activation | +2.71e+00 |
Pos logit contributions | |
gut | +0.180 |
rum | +0.170 |
swollen | +0.166 |
therm | +0.165 |
cut | +0.165 |
Neg logit contributions | |
parties | -0.120 |
tea | -0.112 |
played | -0.097 |
party | -0.095 |
picnic | -0.092 |
Token | something |
Token ID | 1223 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
gut | +0.117 |
inct | +0.112 |
beast | +0.111 |
swollen | +0.110 |
mistake | +0.110 |
Neg logit contributions | |
parties | -0.064 |
tea | -0.058 |
recess | -0.042 |
celebrate | -0.039 |
party | -0.039 |
Token | in |
Token ID | 287 |
Feature activation | +2.49e+00 |
Pos logit contributions | |
beast | +0.075 |
gut | +0.075 |
mistake | +0.070 |
therm | +0.069 |
inct | +0.069 |
Neg logit contributions | |
parties | -0.080 |
tea | -0.071 |
recess | -0.061 |
celebrate | -0.059 |
party | -0.057 |
Token | the |
Token ID | 262 |
Feature activation | -9.66e-01 |
Pos logit contributions | |
gut | +0.114 |
inct | +0.106 |
iously | +0.105 |
ache | +0.105 |
ometer | +0.104 |
Neg logit contributions | |
parties | -0.051 |
tea | -0.046 |
recess | -0.038 |
picnic | -0.032 |
celebrate | -0.032 |
Token | Finally |
Token ID | 11158 |
Feature activation | +2.85e+00 |
Pos logit contributions | |
beast | +0.028 |
gut | +0.027 |
mistake | +0.025 |
therm | +0.025 |
inct | +0.025 |
Neg logit contributions | |
parties | -0.028 |
tea | -0.025 |
recess | -0.021 |
celebrate | -0.021 |
party | -0.020 |
Token | , |
Token ID | 11 |
Feature activation | +5.75e-01 |
Pos logit contributions | |
beast | +0.077 |
gut | +0.076 |
mistake | +0.073 |
therm | +0.071 |
passer | +0.069 |
Neg logit contributions | |
parties | -0.105 |
tea | -0.086 |
ets | -0.085 |
outfits | -0.082 |
recess | -0.082 |
Token | the |
Token ID | 262 |
Feature activation | +3.78e+00 |
Pos logit contributions | |
beast | +0.023 |
gut | +0.023 |
rum | +0.021 |
mistake | +0.021 |
inct | +0.021 |
Neg logit contributions | |
parties | -0.016 |
tea | -0.015 |
recess | -0.011 |
celebrate | -0.011 |
party | -0.011 |
Token | rh |
Token ID | 9529 |
Feature activation | +5.12e+00 |
Pos logit contributions | |
gut | +0.153 |
beast | +0.143 |
rum | +0.143 |
severe | +0.140 |
mistake | +0.139 |
Neg logit contributions | |
parties | -0.099 |
tea | -0.094 |
picnic | -0.080 |
celebrate | -0.078 |
party | -0.077 |
Token | in |
Token ID | 259 |
Feature activation | +5.10e+00 |
Pos logit contributions | |
gut | +0.210 |
plea | +0.192 |
inct | +0.191 |
mistake | +0.190 |
swollen | +0.189 |
Neg logit contributions | |
parties | -0.126 |
tea | -0.107 |
celebrate | -0.086 |
acky | -0.086 |
party | -0.085 |
Token | oc |
Token ID | 420 |
Feature activation | +8.83e+00 |
Pos logit contributions | |
gut | +0.164 |
swollen | +0.148 |
mistake | +0.145 |
vain | +0.143 |
reckless | +0.142 |
Neg logit contributions | |
tea | -0.163 |
parties | -0.150 |
played | -0.150 |
picnic | -0.145 |
celebrate | -0.141 |
Token | eros |
Token ID | 27498 |
Feature activation | +4.36e+00 |
Pos logit contributions | |
severe | +0.276 |
swollen | +0.263 |
reckless | +0.248 |
terrified | +0.246 |
outage | +0.244 |
Neg logit contributions | |
parties | -0.285 |
tea | -0.250 |
ils | -0.220 |
recess | -0.219 |
birthday | -0.216 |
Token | remembered |
Token ID | 12086 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
gut | +0.166 |
beast | +0.160 |
swollen | +0.158 |
therm | +0.156 |
inct | +0.153 |
Neg logit contributions | |
parties | -0.123 |
tea | -0.109 |
party | -0.090 |
celebrate | -0.087 |
picnic | -0.087 |
Token | that |
Token ID | 326 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
beast | +0.097 |
inct | +0.095 |
gut | +0.092 |
cut | +0.092 |
passer | +0.089 |
Neg logit contributions | |
parties | -0.057 |
tea | -0.049 |
party | -0.036 |
celebrate | -0.033 |
recess | -0.033 |
Token | animals |
Token ID | 4695 |
Feature activation | -4.68e-01 |
Pos logit contributions | |
parties | +0.032 |
tea | +0.029 |
celebrate | +0.022 |
party | +0.020 |
together | +0.019 |
Neg logit contributions | |
gut | -0.059 |
beast | -0.058 |
inct | -0.057 |
therm | -0.054 |
severe | -0.054 |
Token | often |
Token ID | 1690 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
parties | +0.012 |
tea | +0.010 |
celebrate | +0.008 |
party | +0.007 |
recess | +0.007 |
Neg logit contributions | |
gut | -0.020 |
beast | -0.020 |
inct | -0.019 |
therm | -0.019 |
swollen | -0.018 |
Token | name |
Token ID | 1438 |
Feature activation | +2.83e-02 |
Pos logit contributions | |
parties | +0.008 |
tea | +0.007 |
picnic | +0.005 |
birthday | +0.005 |
party | +0.005 |
Neg logit contributions | |
gut | -0.010 |
swollen | -0.010 |
severe | -0.010 |
beast | -0.010 |
inct | -0.010 |
Token | ?" |
Token ID | 1701 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
beast | +0.001 |
gut | +0.001 |
mistake | +0.001 |
inct | +0.001 |
therm | +0.001 |
Neg logit contributions | |
parties | -0.001 |
tea | -0.001 |
recess | -0.001 |
celebrate | -0.001 |
party | -0.001 |
Token | The |
Token ID | 383 |
Feature activation | +3.98e+00 |
Pos logit contributions | |
gut | +0.081 |
beast | +0.078 |
mistake | +0.077 |
swollen | +0.076 |
inct | +0.074 |
Neg logit contributions | |
parties | -0.055 |
tea | -0.048 |
recess | -0.039 |
celebrate | -0.039 |
party | -0.036 |
Token | rh |
Token ID | 9529 |
Feature activation | +4.92e+00 |
Pos logit contributions | |
gut | +0.140 |
rum | +0.127 |
inct | +0.126 |
severe | +0.121 |
iously | +0.119 |
Neg logit contributions | |
parties | -0.127 |
tea | -0.120 |
party | -0.104 |
celebrate | -0.102 |
picnic | -0.102 |
Token | in |
Token ID | 259 |
Feature activation | +5.61e+00 |
Pos logit contributions | |
gut | +0.213 |
plea | +0.196 |
inct | +0.195 |
iously | +0.193 |
severe | +0.192 |
Neg logit contributions | |
parties | -0.106 |
tea | -0.088 |
acky | -0.074 |
together | -0.073 |
celebrate | -0.068 |
Token | oc |
Token ID | 420 |
Feature activation | +8.81e+00 |
Pos logit contributions | |
gut | +0.204 |
swollen | +0.177 |
reckless | +0.175 |
severe | +0.175 |
outage | +0.171 |
Neg logit contributions | |
played | -0.140 |
tea | -0.138 |
parties | -0.134 |
play | -0.130 |
together | -0.125 |
Token | eros |
Token ID | 27498 |
Feature activation | +5.14e+00 |
Pos logit contributions | |
severe | +0.294 |
outage | +0.279 |
gut | +0.273 |
swollen | +0.272 |
confusion | +0.265 |
Neg logit contributions | |
parties | -0.253 |
tea | -0.211 |
ils | -0.199 |
sandwiches | -0.190 |
recess | -0.181 |
Token | was |
Token ID | 373 |
Feature activation | +2.50e+00 |
Pos logit contributions | |
gut | +0.221 |
beast | +0.201 |
severe | +0.200 |
swollen | +0.198 |
rum | +0.198 |
Neg logit contributions | |
parties | -0.118 |
tea | -0.105 |
played | -0.097 |
party | -0.088 |
picnic | -0.083 |
Token | surprised |
Token ID | 6655 |
Feature activation | +3.34e+00 |
Pos logit contributions | |
gut | +0.085 |
beast | +0.084 |
inct | +0.080 |
mistake | +0.078 |
therm | +0.077 |
Neg logit contributions | |
parties | -0.081 |
tea | -0.074 |
recess | -0.059 |
celebrate | -0.058 |
party | -0.058 |
Token | that |
Token ID | 326 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
beast | +0.108 |
gut | +0.108 |
mistake | +0.101 |
therm | +0.098 |
inct | +0.098 |
Neg logit contributions | |
parties | -0.115 |
tea | -0.103 |
recess | -0.090 |
celebrate | -0.086 |
picnic | -0.083 |
Token | the |
Token ID | 262 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
parties | +0.003 |
tea | +0.003 |
celebrate | +0.002 |
recess | +0.002 |
party | +0.002 |
Neg logit contributions | |
gut | -0.005 |
swollen | -0.005 |
inct | -0.005 |
beast | -0.005 |
mistake | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | -4.68e-01 |
Pos logit contributions | |
parties | +0.006 |
tea | +0.005 |
recess | +0.004 |
celebrate | +0.004 |
picnic | +0.004 |
Neg logit contributions | |
gut | -0.010 |
beast | -0.010 |
mistake | -0.010 |
swollen | -0.010 |
therm | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | +3.52e-01 |
Pos logit contributions | |
parties | +0.016 |
tea | +0.014 |
recess | +0.012 |
party | +0.011 |
celebrate | +0.011 |
Neg logit contributions | |
beast | -0.016 |
gut | -0.015 |
mistake | -0.014 |
therm | -0.014 |
inct | -0.014 |
Token | The |
Token ID | 464 |
Feature activation | +3.88e+00 |
Pos logit contributions | |
gut | +0.011 |
beast | +0.011 |
mistake | +0.011 |
therm | +0.010 |
inct | +0.010 |
Neg logit contributions | |
parties | -0.012 |
tea | -0.011 |
recess | -0.009 |
celebrate | -0.009 |
picnic | -0.009 |
Token | rh |
Token ID | 9529 |
Feature activation | +4.82e+00 |
Pos logit contributions | |
gut | +0.122 |
rum | +0.116 |
inct | +0.111 |
outage | +0.110 |
beast | +0.107 |
Neg logit contributions | |
parties | -0.125 |
tea | -0.120 |
party | -0.108 |
picnic | -0.106 |
celebrate | -0.106 |
Token | in |
Token ID | 259 |
Feature activation | +5.14e+00 |
Pos logit contributions | |
gut | +0.213 |
plea | +0.199 |
iously | +0.197 |
inct | +0.194 |
swollen | +0.193 |
Neg logit contributions | |
parties | -0.083 |
tea | -0.065 |
together | -0.065 |
acky | -0.062 |
P | -0.061 |
Token | oc |
Token ID | 420 |
Feature activation | +8.80e+00 |
Pos logit contributions | |
gut | +0.184 |
swollen | +0.176 |
outage | +0.174 |
unstable | +0.169 |
vain | +0.168 |
Neg logit contributions | |
played | -0.106 |
celebrate | -0.094 |
play | -0.093 |
together | -0.092 |
parties | -0.091 |
Token | eros |
Token ID | 27498 |
Feature activation | +4.39e+00 |
Pos logit contributions | |
outage | +0.297 |
severe | +0.297 |
swollen | +0.285 |
therm | +0.269 |
gut | +0.265 |
Neg logit contributions | |
parties | -0.228 |
ils | -0.191 |
tea | -0.181 |
recess | -0.168 |
celebrate | -0.161 |
Token | was |
Token ID | 373 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
gut | +0.165 |
swollen | +0.162 |
therm | +0.159 |
beast | +0.158 |
rum | +0.153 |
Neg logit contributions | |
parties | -0.116 |
tea | -0.101 |
played | -0.095 |
party | -0.089 |
celebrate | -0.088 |
Token | so |
Token ID | 523 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
beast | +0.065 |
gut | +0.064 |
inct | +0.062 |
therm | +0.062 |
cut | +0.060 |
Neg logit contributions | |
parties | -0.041 |
tea | -0.037 |
celebrate | -0.032 |
together | -0.029 |
party | -0.028 |
Token | excited |
Token ID | 6568 |
Feature activation | +2.80e+00 |
Pos logit contributions | |
beast | +0.062 |
gut | +0.059 |
therm | +0.058 |
mistake | +0.058 |
inct | +0.058 |
Neg logit contributions | |
parties | -0.029 |
celebrate | -0.024 |
tea | -0.024 |
together | -0.022 |
played | -0.021 |
Token | that |
Token ID | 326 |
Feature activation | +4.55e-02 |
Pos logit contributions | |
beast | +0.102 |
gut | +0.102 |
mistake | +0.096 |
inct | +0.095 |
therm | +0.095 |
Neg logit contributions | |
parties | -0.085 |
tea | -0.074 |
recess | -0.062 |
celebrate | -0.059 |
party | -0.057 |
Token | y |
Token ID | 331 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
parties | +0.087 |
tea | +0.076 |
recess | +0.064 |
celebrate | +0.061 |
party | +0.059 |
Neg logit contributions | |
beast | -0.105 |
gut | -0.104 |
mistake | -0.098 |
therm | -0.097 |
inct | -0.097 |
Token | ummy |
Token ID | 13513 |
Feature activation | -3.16e+00 |
Pos logit contributions | |
parties | +0.011 |
tea | +0.010 |
recess | +0.009 |
celebrate | +0.009 |
costumes | +0.009 |
Neg logit contributions | |
therm | -0.008 |
gut | -0.008 |
beast | -0.008 |
mistake | -0.008 |
severe | -0.007 |
Token | spaghetti |
Token ID | 48241 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
parties | +0.101 |
tea | +0.086 |
recess | +0.082 |
celebrate | +0.073 |
wear | +0.070 |
Neg logit contributions | |
beast | -0.104 |
gut | -0.101 |
therm | -0.099 |
mistake | -0.097 |
passer | -0.095 |
Token | with |
Token ID | 351 |
Feature activation | -3.13e+00 |
Pos logit contributions | |
parties | +0.050 |
tea | +0.043 |
recess | +0.040 |
celebrate | +0.036 |
wear | +0.034 |
Neg logit contributions | |
beast | -0.057 |
gut | -0.056 |
mistake | -0.054 |
therm | -0.053 |
cut | -0.053 |
Token | meat |
Token ID | 6174 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
parties | +0.093 |
tea | +0.083 |
recess | +0.065 |
celebrate | +0.065 |
party | +0.062 |
Neg logit contributions | |
beast | -0.119 |
gut | -0.119 |
inct | -0.112 |
mistake | -0.111 |
swollen | -0.108 |
Token | balls |
Token ID | 21591 |
Feature activation | -9.06e+00 |
Pos logit contributions | |
parties | +0.041 |
tea | +0.036 |
recess | +0.030 |
celebrate | +0.028 |
party | +0.027 |
Neg logit contributions | |
beast | -0.056 |
gut | -0.056 |
mistake | -0.053 |
therm | -0.052 |
inct | -0.052 |
Token | ." |
Token ID | 526 |
Feature activation | +1.97e-01 |
Pos logit contributions | |
parties | +0.273 |
tea | +0.241 |
recess | +0.208 |
celebrate | +0.196 |
picnic | +0.189 |
Neg logit contributions | |
beast | -0.315 |
gut | -0.313 |
therm | -0.296 |
mistake | -0.295 |
cut | -0.290 |
Token | Lily |
Token ID | 20037 |
Feature activation | -8.13e-01 |
Pos logit contributions | |
mistake | +0.004 |
inct | +0.003 |
therm | +0.003 |
Start | +0.003 |
Fear | +0.003 |
Neg logit contributions | |
parties | -0.008 |
recess | -0.007 |
ils | -0.007 |
seas | -0.007 |
camping | -0.007 |
Token | was |
Token ID | 373 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
parties | +0.029 |
tea | +0.026 |
recess | +0.022 |
celebrate | +0.021 |
party | +0.021 |
Neg logit contributions | |
beast | -0.025 |
gut | -0.025 |
mistake | -0.023 |
therm | -0.023 |
inct | -0.023 |
Token | so |
Token ID | 523 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
beast | +0.066 |
gut | +0.064 |
inct | +0.061 |
mistake | +0.060 |
therm | +0.059 |
Neg logit contributions | |
parties | -0.054 |
tea | -0.048 |
celebrate | -0.039 |
recess | -0.038 |
party | -0.037 |
Token | happy |
Token ID | 3772 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
gut | +0.038 |
beast | +0.037 |
mistake | +0.036 |
therm | +0.036 |
swollen | +0.035 |
Neg logit contributions | |
parties | -0.040 |
tea | -0.035 |
recess | -0.033 |
outfits | -0.028 |
celebrate | -0.028 |
Token | Lily |
Token ID | 20037 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
parties | +0.022 |
ils | +0.019 |
recess | +0.019 |
tea | +0.018 |
ets | +0.017 |
Neg logit contributions | |
therm | -0.014 |
mistake | -0.014 |
gut | -0.014 |
passer | -0.013 |
inct | -0.013 |
Token | loved |
Token ID | 6151 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
parties | +0.065 |
tea | +0.056 |
recess | +0.054 |
celebrate | +0.045 |
picnic | +0.043 |
Neg logit contributions | |
gut | -0.067 |
beast | -0.066 |
mistake | -0.064 |
therm | -0.064 |
cut | -0.062 |
Token | spaghetti |
Token ID | 48241 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
parties | +0.009 |
tea | +0.007 |
recess | +0.006 |
celebrate | +0.006 |
party | +0.006 |
Neg logit contributions | |
beast | -0.011 |
gut | -0.011 |
mistake | -0.010 |
therm | -0.010 |
swollen | -0.010 |
Token | with |
Token ID | 351 |
Feature activation | -3.18e+00 |
Pos logit contributions | |
parties | +0.060 |
recess | +0.054 |
tea | +0.051 |
celebrate | +0.048 |
wear | +0.046 |
Neg logit contributions | |
beast | -0.053 |
therm | -0.052 |
mistake | -0.051 |
gut | -0.051 |
passer | -0.048 |
Token | meat |
Token ID | 6174 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
parties | +0.084 |
tea | +0.069 |
recess | +0.060 |
celebrate | +0.056 |
party | +0.053 |
Neg logit contributions | |
beast | -0.125 |
gut | -0.124 |
therm | -0.121 |
swollen | -0.119 |
mistake | -0.118 |
Token | balls |
Token ID | 21591 |
Feature activation | -8.65e+00 |
Pos logit contributions | |
parties | +0.046 |
tea | +0.040 |
recess | +0.034 |
celebrate | +0.033 |
party | +0.031 |
Neg logit contributions | |
beast | -0.056 |
gut | -0.055 |
mistake | -0.052 |
therm | -0.052 |
swollen | -0.051 |
Token | . |
Token ID | 13 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
parties | +0.282 |
tea | +0.248 |
recess | +0.220 |
celebrate | +0.207 |
picnic | +0.200 |
Neg logit contributions | |
beast | -0.282 |
gut | -0.280 |
therm | -0.271 |
mistake | -0.263 |
cut | -0.260 |
Token | She |
Token ID | 1375 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
parties | +0.047 |
tea | +0.042 |
recess | +0.040 |
ils | +0.039 |
celebrate | +0.037 |
Neg logit contributions | |
gut | -0.030 |
therm | -0.030 |
mistake | -0.028 |
inct | -0.028 |
Owl | -0.027 |
Token | watched |
Token ID | 7342 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
parties | +0.006 |
recess | +0.005 |
tea | +0.005 |
celebrate | +0.004 |
dates | +0.004 |
Neg logit contributions | |
gut | -0.006 |
beast | -0.006 |
therm | -0.006 |
mistake | -0.006 |
cut | -0.005 |
Token | as |
Token ID | 355 |
Feature activation | -3.93e-01 |
Pos logit contributions | |
parties | +0.033 |
tea | +0.027 |
recess | +0.025 |
picnic | +0.023 |
party | +0.022 |
Neg logit contributions | |
gut | -0.040 |
beast | -0.040 |
therm | -0.039 |
mistake | -0.038 |
cut | -0.037 |
Token | her |
Token ID | 607 |
Feature activation | +3.36e-02 |
Pos logit contributions | |
parties | +0.009 |
tea | +0.008 |
recess | +0.006 |
celebrate | +0.006 |
party | +0.006 |
Neg logit contributions | |
beast | -0.017 |
gut | -0.017 |
mistake | -0.016 |
therm | -0.016 |
swollen | -0.016 |
Token | food |
Token ID | 2057 |
Feature activation | +7.44e-01 |
Pos logit contributions | |
parties | +0.050 |
tea | +0.043 |
recess | +0.035 |
celebrate | +0.033 |
party | +0.032 |
Neg logit contributions | |
beast | -0.073 |
gut | -0.072 |
mistake | -0.068 |
inct | -0.068 |
therm | -0.068 |
Token | , |
Token ID | 11 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
beast | +0.027 |
gut | +0.027 |
therm | +0.026 |
mistake | +0.026 |
cut | +0.025 |
Neg logit contributions | |
parties | -0.022 |
tea | -0.018 |
recess | -0.017 |
celebrate | -0.015 |
picnic | -0.014 |
Token | spaghetti |
Token ID | 48241 |
Feature activation | -6.29e-01 |
Pos logit contributions | |
parties | +0.069 |
tea | +0.062 |
recess | +0.050 |
celebrate | +0.049 |
party | +0.048 |
Neg logit contributions | |
beast | -0.079 |
gut | -0.078 |
inct | -0.073 |
mistake | -0.073 |
swollen | -0.072 |
Token | with |
Token ID | 351 |
Feature activation | -2.88e+00 |
Pos logit contributions | |
parties | +0.019 |
tea | +0.016 |
recess | +0.015 |
celebrate | +0.014 |
outfits | +0.013 |
Neg logit contributions | |
beast | -0.021 |
therm | -0.021 |
mistake | -0.021 |
gut | -0.020 |
cut | -0.020 |
Token | meat |
Token ID | 6174 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
parties | +0.087 |
tea | +0.075 |
recess | +0.065 |
celebrate | +0.061 |
party | +0.059 |
Neg logit contributions | |
beast | -0.104 |
gut | -0.103 |
therm | -0.098 |
mistake | -0.097 |
swollen | -0.096 |
Token | balls |
Token ID | 21591 |
Feature activation | -8.54e+00 |
Pos logit contributions | |
parties | +0.034 |
tea | +0.029 |
recess | +0.025 |
celebrate | +0.023 |
party | +0.022 |
Neg logit contributions | |
beast | -0.045 |
gut | -0.045 |
mistake | -0.043 |
therm | -0.043 |
inct | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | -7.22e-01 |
Pos logit contributions | |
parties | +0.261 |
tea | +0.227 |
recess | +0.202 |
celebrate | +0.192 |
picnic | +0.180 |
Neg logit contributions | |
beast | -0.285 |
gut | -0.281 |
therm | -0.278 |
cut | -0.269 |
mistake | -0.269 |
Token | It |
Token ID | 632 |
Feature activation | +3.32e+00 |
Pos logit contributions | |
parties | +0.021 |
tea | +0.016 |
recess | +0.016 |
ils | +0.014 |
celebrate | +0.014 |
Neg logit contributions | |
therm | -0.026 |
gut | -0.025 |
mistake | -0.025 |
inct | -0.025 |
swollen | -0.025 |
Token | smelled |
Token ID | 37609 |
Feature activation | +2.40e+00 |
Pos logit contributions | |
gut | +0.109 |
beast | +0.107 |
mistake | +0.099 |
inct | +0.098 |
therm | +0.095 |
Neg logit contributions | |
parties | -0.114 |
tea | -0.106 |
celebrate | -0.086 |
recess | -0.085 |
party | -0.084 |
Token | so |
Token ID | 523 |
Feature activation | +6.26e-01 |
Pos logit contributions | |
therm | +0.064 |
beast | +0.064 |
mistake | +0.062 |
gut | +0.061 |
severe | +0.058 |
Neg logit contributions | |
parties | -0.087 |
recess | -0.080 |
tea | -0.072 |
outfits | -0.068 |
dates | -0.068 |
Token | delicious |
Token ID | 12625 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
gut | +0.019 |
mistake | +0.019 |
therm | +0.018 |
swollen | +0.018 |
beast | +0.018 |
Neg logit contributions | |
parties | -0.021 |
recess | -0.018 |
tea | -0.017 |
outfits | -0.015 |
celebrate | -0.015 |
Token | food |
Token ID | 2057 |
Feature activation | +8.07e-01 |
Pos logit contributions | |
parties | +0.035 |
tea | +0.029 |
recess | +0.025 |
celebrate | +0.024 |
party | +0.022 |
Neg logit contributions | |
beast | -0.045 |
gut | -0.045 |
mistake | -0.043 |
therm | -0.043 |
swollen | -0.042 |
Token | - |
Token ID | 532 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
beast | +0.031 |
gut | +0.031 |
mistake | +0.029 |
therm | +0.029 |
swollen | +0.028 |
Neg logit contributions | |
parties | -0.023 |
tea | -0.019 |
recess | -0.017 |
celebrate | -0.015 |
picnic | -0.014 |
Token | spaghetti |
Token ID | 48241 |
Feature activation | -7.95e-01 |
Pos logit contributions | |
beast | +0.010 |
gut | +0.010 |
inct | +0.010 |
mistake | +0.010 |
swollen | +0.009 |
Neg logit contributions | |
parties | -0.008 |
tea | -0.007 |
recess | -0.005 |
celebrate | -0.005 |
picnic | -0.005 |
Token | with |
Token ID | 351 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
parties | +0.025 |
tea | +0.021 |
recess | +0.020 |
celebrate | +0.018 |
outfits | +0.017 |
Neg logit contributions | |
beast | -0.027 |
therm | -0.025 |
gut | -0.025 |
mistake | -0.025 |
cut | -0.024 |
Token | meat |
Token ID | 6174 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
parties | +0.081 |
tea | +0.068 |
recess | +0.064 |
celebrate | +0.058 |
party | +0.056 |
Neg logit contributions | |
beast | -0.086 |
gut | -0.084 |
therm | -0.083 |
swollen | -0.082 |
mistake | -0.080 |
Token | balls |
Token ID | 21591 |
Feature activation | -8.46e+00 |
Pos logit contributions | |
parties | +0.038 |
tea | +0.033 |
recess | +0.028 |
celebrate | +0.027 |
party | +0.025 |
Neg logit contributions | |
beast | -0.049 |
gut | -0.049 |
mistake | -0.046 |
therm | -0.046 |
inct | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | -6.21e-01 |
Pos logit contributions | |
parties | +0.247 |
tea | +0.213 |
recess | +0.183 |
celebrate | +0.173 |
picnic | +0.163 |
Neg logit contributions | |
beast | -0.308 |
gut | -0.304 |
therm | -0.294 |
mistake | -0.290 |
cut | -0.284 |
Token |
|
Token ID | 198 |
Feature activation | -6.39e-01 |
Pos logit contributions | |
parties | +0.020 |
tea | +0.016 |
recess | +0.016 |
ils | +0.015 |
celebrate | +0.014 |
Neg logit contributions | |
therm | -0.019 |
gut | -0.019 |
inct | -0.019 |
mistake | -0.019 |
swollen | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
parties | +0.027 |
tea | +0.024 |
recess | +0.023 |
ils | +0.022 |
party | +0.021 |
Neg logit contributions | |
mistake | -0.013 |
therm | -0.013 |
gut | -0.013 |
beast | -0.012 |
severe | -0.012 |
Token | L |
Token ID | 43 |
Feature activation | -7.36e-01 |
Pos logit contributions | |
mistake | +0.006 |
gut | +0.006 |
therm | +0.006 |
swollen | +0.006 |
severe | +0.006 |
Neg logit contributions | |
parties | -0.007 |
tea | -0.006 |
recess | -0.006 |
celebrate | -0.005 |
party | -0.005 |
Token | ily |
Token ID | 813 |
Feature activation | -8.11e-01 |
Pos logit contributions | |
parties | +0.019 |
tea | +0.016 |
celebrate | +0.014 |
recess | +0.014 |
picnic | +0.014 |
Neg logit contributions | |
mistake | -0.028 |
adult | -0.027 |
beast | -0.026 |
gut | -0.026 |
rum | -0.025 |
Token | make |
Token ID | 787 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
parties | +0.058 |
tea | +0.053 |
celebrate | +0.041 |
wear | +0.039 |
picnic | +0.039 |
Neg logit contributions | |
beast | -0.086 |
gut | -0.084 |
inct | -0.084 |
passer | -0.081 |
mistake | -0.080 |
Token | snow |
Token ID | 6729 |
Feature activation | -2.48e+00 |
Pos logit contributions | |
parties | +0.065 |
tea | +0.055 |
recess | +0.053 |
celebrate | +0.047 |
party | +0.046 |
Neg logit contributions | |
beast | -0.066 |
gut | -0.066 |
mistake | -0.063 |
therm | -0.061 |
inct | -0.060 |
Token | men |
Token ID | 3653 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
parties | +0.089 |
recess | +0.077 |
tea | +0.076 |
celebrate | +0.067 |
wear | +0.066 |
Neg logit contributions | |
beast | -0.070 |
gut | -0.069 |
cut | -0.066 |
therm | -0.066 |
mistake | -0.066 |
Token | and |
Token ID | 290 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
parties | +0.047 |
tea | +0.041 |
recess | +0.037 |
celebrate | +0.032 |
picnic | +0.031 |
Neg logit contributions | |
gut | -0.055 |
beast | -0.055 |
mistake | -0.052 |
therm | -0.052 |
cut | -0.051 |
Token | snow |
Token ID | 6729 |
Feature activation | -3.31e+00 |
Pos logit contributions | |
parties | +0.076 |
tea | +0.070 |
celebrate | +0.054 |
picnic | +0.051 |
party | +0.051 |
Neg logit contributions | |
beast | -0.104 |
gut | -0.101 |
inct | -0.101 |
passer | -0.097 |
mistake | -0.095 |
Token | balls |
Token ID | 21591 |
Feature activation | -8.32e+00 |
Pos logit contributions | |
parties | +0.119 |
tea | +0.106 |
recess | +0.098 |
celebrate | +0.091 |
wear | +0.088 |
Neg logit contributions | |
beast | -0.096 |
gut | -0.095 |
mistake | -0.089 |
therm | -0.089 |
cut | -0.089 |
Token | ." |
Token ID | 526 |
Feature activation | -6.23e-01 |
Pos logit contributions | |
parties | +0.253 |
tea | +0.223 |
recess | +0.199 |
celebrate | +0.179 |
picnic | +0.175 |
Neg logit contributions | |
beast | -0.282 |
gut | -0.280 |
mistake | -0.269 |
therm | -0.268 |
cut | -0.264 |
Token |
|
Token ID | 198 |
Feature activation | -5.39e-01 |
Pos logit contributions | |
parties | +0.021 |
tea | +0.018 |
recess | +0.016 |
celebrate | +0.015 |
party | +0.015 |
Neg logit contributions | |
gut | -0.020 |
beast | -0.019 |
mistake | -0.019 |
therm | -0.018 |
inct | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
parties | +0.019 |
tea | +0.017 |
recess | +0.015 |
celebrate | +0.014 |
party | +0.014 |
Neg logit contributions | |
gut | -0.017 |
beast | -0.016 |
mistake | -0.016 |
therm | -0.015 |
cut | -0.015 |
Token | L |
Token ID | 43 |
Feature activation | -5.27e-01 |
Pos logit contributions | |
beast | +0.004 |
gut | +0.004 |
mistake | +0.004 |
inct | +0.003 |
therm | +0.003 |
Neg logit contributions | |
parties | -0.003 |
tea | -0.003 |
recess | -0.002 |
celebrate | -0.002 |
party | -0.002 |
Token | ily |
Token ID | 813 |
Feature activation | -4.37e-01 |
Pos logit contributions | |
parties | +0.010 |
tea | +0.008 |
recess | +0.006 |
celebrate | +0.006 |
picnic | +0.005 |
Neg logit contributions | |
beast | -0.025 |
gut | -0.025 |
mistake | -0.024 |
therm | -0.024 |
inct | -0.023 |
Token | of |
Token ID | 286 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
parties | +0.098 |
tea | +0.088 |
recess | +0.079 |
ackers | +0.074 |
picnic | +0.073 |
Neg logit contributions | |
beast | -0.089 |
gut | -0.087 |
cut | -0.085 |
ringing | -0.084 |
therm | -0.084 |
Token | all |
Token ID | 477 |
Feature activation | -3.63e+00 |
Pos logit contributions | |
parties | +0.048 |
tea | +0.042 |
recess | +0.033 |
celebrate | +0.032 |
party | +0.030 |
Neg logit contributions | |
beast | -0.075 |
gut | -0.075 |
inct | -0.071 |
mistake | -0.070 |
therm | -0.070 |
Token | the |
Token ID | 262 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
parties | +0.111 |
tea | +0.096 |
recess | +0.084 |
picnic | +0.081 |
party | +0.079 |
Neg logit contributions | |
beast | -0.124 |
gut | -0.124 |
mistake | -0.117 |
cut | -0.117 |
therm | -0.116 |
Token | fun |
Token ID | 1257 |
Feature activation | -4.85e+00 |
Pos logit contributions | |
parties | +0.076 |
tea | +0.068 |
recess | +0.061 |
celebrate | +0.057 |
picnic | +0.055 |
Neg logit contributions | |
beast | -0.075 |
gut | -0.073 |
mistake | -0.071 |
therm | -0.069 |
inct | -0.068 |
Token | things |
Token ID | 1243 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
parties | +0.142 |
tea | +0.127 |
ackers | +0.117 |
recess | +0.116 |
celebrate | +0.113 |
Neg logit contributions | |
beast | -0.147 |
mistake | -0.142 |
gut | -0.141 |
ringing | -0.135 |
therm | -0.134 |
Token | they |
Token ID | 484 |
Feature activation | -8.31e+00 |
Pos logit contributions | |
parties | +0.044 |
tea | +0.038 |
recess | +0.033 |
picnic | +0.028 |
celebrate | +0.028 |
Neg logit contributions | |
gut | -0.069 |
beast | -0.069 |
therm | -0.067 |
cut | -0.066 |
swollen | -0.064 |
Token | did |
Token ID | 750 |
Feature activation | -2.78e+00 |
Pos logit contributions | |
parties | +0.267 |
tea | +0.253 |
recess | +0.209 |
sandwiches | +0.195 |
picnic | +0.193 |
Neg logit contributions | |
beast | -0.257 |
gut | -0.249 |
mistake | -0.234 |
injured | -0.234 |
reckless | -0.232 |
Token | . |
Token ID | 13 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
parties | +0.086 |
tea | +0.081 |
recess | +0.070 |
picnic | +0.065 |
celebrate | +0.062 |
Neg logit contributions | |
gut | -0.089 |
mistake | -0.089 |
beast | -0.086 |
inct | -0.080 |
plea | -0.079 |
Token | Ch |
Token ID | 609 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
parties | +0.049 |
recess | +0.041 |
tea | +0.040 |
party | +0.037 |
celebrate | +0.036 |
Neg logit contributions | |
gut | -0.046 |
inct | -0.045 |
therm | -0.045 |
beast | -0.043 |
Owl | -0.043 |
Token | oo |
Token ID | 2238 |
Feature activation | +8.36e-01 |
Pos logit contributions | |
parties | +0.068 |
tea | +0.061 |
recess | +0.053 |
celebrate | +0.051 |
party | +0.049 |
Neg logit contributions | |
beast | -0.063 |
gut | -0.063 |
mistake | -0.059 |
therm | -0.058 |
inct | -0.058 |
Token | - |
Token ID | 12 |
Feature activation | +5.36e+00 |
Pos logit contributions | |
beast | +0.028 |
gut | +0.028 |
mistake | +0.026 |
therm | +0.025 |
inct | +0.025 |
Neg logit contributions | |
parties | -0.028 |
tea | -0.025 |
recess | -0.021 |
celebrate | -0.020 |
party | -0.020 |
Token | every |
Token ID | 790 |
Feature activation | -3.90e+00 |
Pos logit contributions | |
parties | +0.121 |
tea | +0.105 |
recess | +0.087 |
celebrate | +0.081 |
picnic | +0.078 |
Neg logit contributions | |
beast | -0.178 |
gut | -0.178 |
mistake | -0.170 |
therm | -0.164 |
inct | -0.164 |
Token | day |
Token ID | 1110 |
Feature activation | -3.22e+00 |
Pos logit contributions | |
parties | +0.090 |
tea | +0.074 |
recess | +0.057 |
celebrate | +0.053 |
party | +0.049 |
Neg logit contributions | |
beast | -0.171 |
gut | -0.169 |
mistake | -0.163 |
therm | -0.160 |
inct | -0.159 |
Token | and |
Token ID | 290 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
parties | +0.103 |
tea | +0.091 |
recess | +0.077 |
celebrate | +0.075 |
picnic | +0.073 |
Neg logit contributions | |
gut | -0.107 |
beast | -0.107 |
mistake | -0.100 |
injured | -0.097 |
therm | -0.096 |
Token | shared |
Token ID | 4888 |
Feature activation | -4.37e+00 |
Pos logit contributions | |
parties | +0.046 |
tea | +0.040 |
recess | +0.033 |
celebrate | +0.031 |
party | +0.030 |
Neg logit contributions | |
beast | -0.067 |
gut | -0.066 |
mistake | -0.063 |
therm | -0.062 |
inct | -0.062 |
Token | everything |
Token ID | 2279 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
parties | +0.113 |
recess | +0.093 |
tea | +0.093 |
acky | +0.082 |
dates | +0.081 |
Neg logit contributions | |
mistake | -0.157 |
gut | -0.150 |
beast | -0.145 |
severe | -0.136 |
swollen | -0.136 |
Token | they |
Token ID | 484 |
Feature activation | -8.30e+00 |
Pos logit contributions | |
parties | +0.054 |
tea | +0.045 |
recess | +0.044 |
picnic | +0.039 |
ashed | +0.038 |
Neg logit contributions | |
gut | -0.060 |
mistake | -0.058 |
beast | -0.056 |
guardian | -0.055 |
therm | -0.054 |
Token | had |
Token ID | 550 |
Feature activation | -6.27e+00 |
Pos logit contributions | |
parties | +0.257 |
tea | +0.255 |
sandwiches | +0.209 |
outfits | +0.209 |
text | +0.203 |
Neg logit contributions | |
reckless | -0.221 |
beast | -0.219 |
disob | -0.215 |
Owl | -0.212 |
injured | -0.212 |
Token | . |
Token ID | 13 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
parties | +0.191 |
tea | +0.188 |
recess | +0.162 |
celebrate | +0.150 |
acky | +0.146 |
Neg logit contributions | |
mistake | -0.184 |
gut | -0.170 |
injured | -0.163 |
beast | -0.160 |
reckless | -0.158 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -8.00e-01 |
Pos logit contributions | |
parties | +0.085 |
tea | +0.071 |
recess | +0.068 |
outfits | +0.067 |
costumes | +0.065 |
Neg logit contributions | |
beast | -0.053 |
mistake | -0.053 |
gut | -0.052 |
Owl | -0.051 |
severe | -0.050 |
Token | Once |
Token ID | 7454 |
Feature activation | -3.14e+00 |
Pos logit contributions | |
parties | +0.032 |
tea | +0.029 |
recess | +0.025 |
outfits | +0.025 |
ets | +0.025 |
Neg logit contributions | |
beast | -0.020 |
gut | -0.018 |
inct | -0.018 |
mistake | -0.017 |
therm | -0.017 |
Token | upon |
Token ID | 2402 |
Feature activation | +2.64e+00 |
Pos logit contributions | |
parties | +0.135 |
tea | +0.124 |
celebrate | +0.111 |
recess | +0.109 |
ils | +0.108 |
Neg logit contributions | |
gut | -0.069 |
mistake | -0.068 |
beast | -0.066 |
therm | -0.061 |
adult | -0.060 |
Token | menu |
Token ID | 6859 |
Feature activation | +2.20e+00 |
Pos logit contributions | |
parties | +0.046 |
tea | +0.041 |
recess | +0.034 |
celebrate | +0.033 |
party | +0.032 |
Neg logit contributions | |
beast | -0.052 |
gut | -0.052 |
mistake | -0.048 |
inct | -0.048 |
therm | -0.048 |
Token | with |
Token ID | 351 |
Feature activation | -2.30e+00 |
Pos logit contributions | |
beast | +0.074 |
mistake | +0.072 |
gut | +0.071 |
therm | +0.071 |
cut | +0.069 |
Neg logit contributions | |
parties | -0.066 |
tea | -0.057 |
recess | -0.053 |
celebrate | -0.047 |
dates | -0.045 |
Token | pictures |
Token ID | 5986 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
parties | +0.070 |
tea | +0.059 |
recess | +0.054 |
celebrate | +0.050 |
party | +0.048 |
Neg logit contributions | |
beast | -0.079 |
gut | -0.077 |
therm | -0.076 |
mistake | -0.073 |
passer | -0.073 |
Token | of |
Token ID | 286 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
parties | +0.083 |
tea | +0.074 |
recess | +0.068 |
celebrate | +0.062 |
picnic | +0.060 |
Neg logit contributions | |
beast | -0.079 |
cut | -0.074 |
gut | -0.074 |
therm | -0.074 |
passer | -0.070 |
Token | food |
Token ID | 2057 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
parties | +0.047 |
tea | +0.041 |
recess | +0.035 |
celebrate | +0.033 |
party | +0.032 |
Neg logit contributions | |
beast | -0.058 |
gut | -0.057 |
mistake | -0.054 |
therm | -0.054 |
swollen | -0.053 |
Token | they |
Token ID | 484 |
Feature activation | -8.30e+00 |
Pos logit contributions | |
parties | +0.062 |
tea | +0.054 |
recess | +0.047 |
celebrate | +0.044 |
picnic | +0.041 |
Neg logit contributions | |
beast | -0.076 |
gut | -0.075 |
mistake | -0.071 |
therm | -0.071 |
cut | -0.069 |
Token | liked |
Token ID | 8288 |
Feature activation | -2.25e+00 |
Pos logit contributions | |
parties | +0.255 |
tea | +0.231 |
recess | +0.195 |
celebrate | +0.176 |
picnic | +0.175 |
Neg logit contributions | |
beast | -0.287 |
gut | -0.278 |
therm | -0.264 |
mistake | -0.263 |
cut | -0.259 |
Token | , |
Token ID | 11 |
Feature activation | -3.64e+00 |
Pos logit contributions | |
parties | +0.067 |
tea | +0.060 |
recess | +0.055 |
celebrate | +0.051 |
wear | +0.049 |
Neg logit contributions | |
beast | -0.075 |
gut | -0.072 |
mistake | -0.070 |
therm | -0.069 |
cut | -0.067 |
Token | such |
Token ID | 884 |
Feature activation | -5.27e-01 |
Pos logit contributions | |
parties | +0.114 |
tea | +0.101 |
recess | +0.081 |
celebrate | +0.080 |
party | +0.078 |
Neg logit contributions | |
beast | -0.131 |
gut | -0.131 |
inct | -0.123 |
mistake | -0.122 |
swollen | -0.121 |
Token | as |
Token ID | 355 |
Feature activation | -6.73e-01 |
Pos logit contributions | |
parties | +0.015 |
tea | +0.013 |
recess | +0.011 |
celebrate | +0.010 |
party | +0.010 |
Neg logit contributions | |
beast | -0.020 |
gut | -0.020 |
mistake | -0.019 |
therm | -0.019 |
inct | -0.019 |
Token | pizza |
Token ID | 14256 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
parties | +0.020 |
tea | +0.017 |
recess | +0.014 |
celebrate | +0.014 |
party | +0.013 |
Neg logit contributions | |
beast | -0.025 |
gut | -0.025 |
mistake | -0.023 |
therm | -0.023 |
swollen | -0.023 |
Token | snow |
Token ID | 6729 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
parties | +0.016 |
tea | +0.013 |
recess | +0.013 |
celebrate | +0.011 |
ets | +0.011 |
Neg logit contributions | |
beast | -0.020 |
gut | -0.019 |
mistake | -0.019 |
therm | -0.019 |
swollen | -0.018 |
Token | men |
Token ID | 3653 |
Feature activation | -7.31e-01 |
Pos logit contributions | |
parties | +0.059 |
tea | +0.053 |
recess | +0.048 |
celebrate | +0.046 |
party | +0.044 |
Neg logit contributions | |
beast | -0.042 |
gut | -0.041 |
inct | -0.038 |
mistake | -0.038 |
therm | -0.038 |
Token | and |
Token ID | 290 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
parties | +0.023 |
tea | +0.020 |
recess | +0.019 |
celebrate | +0.017 |
picnic | +0.016 |
Neg logit contributions | |
beast | -0.024 |
gut | -0.024 |
therm | -0.023 |
mistake | -0.023 |
cut | -0.023 |
Token | throwing |
Token ID | 9644 |
Feature activation | -4.27e+00 |
Pos logit contributions | |
parties | +0.066 |
tea | +0.062 |
recess | +0.047 |
celebrate | +0.046 |
costumes | +0.046 |
Neg logit contributions | |
gut | -0.080 |
beast | -0.079 |
inct | -0.078 |
passer | -0.075 |
mistake | -0.074 |
Token | snow |
Token ID | 6729 |
Feature activation | -3.21e+00 |
Pos logit contributions | |
parties | +0.137 |
tea | +0.113 |
recess | +0.107 |
celebrate | +0.101 |
party | +0.098 |
Neg logit contributions | |
beast | -0.141 |
gut | -0.136 |
therm | -0.133 |
mistake | -0.130 |
inct | -0.129 |
Token | balls |
Token ID | 21591 |
Feature activation | -8.16e+00 |
Pos logit contributions | |
parties | +0.096 |
tea | +0.083 |
recess | +0.073 |
celebrate | +0.070 |
party | +0.066 |
Neg logit contributions | |
beast | -0.115 |
gut | -0.113 |
therm | -0.108 |
inct | -0.107 |
cut | -0.107 |
Token | . |
Token ID | 13 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
parties | +0.251 |
tea | +0.217 |
recess | +0.193 |
celebrate | +0.184 |
picnic | +0.170 |
Neg logit contributions | |
beast | -0.278 |
gut | -0.274 |
therm | -0.264 |
cut | -0.263 |
mistake | -0.262 |
Token |
|
Token ID | 198 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
parties | +0.049 |
tea | +0.042 |
recess | +0.034 |
celebrate | +0.034 |
ils | +0.034 |
Neg logit contributions | |
gut | -0.065 |
beast | -0.064 |
inct | -0.061 |
therm | -0.060 |
Owl | -0.060 |
Token |
|
Token ID | 198 |
Feature activation | -2.92e-01 |
Pos logit contributions | |
parties | +0.036 |
tea | +0.032 |
recess | +0.026 |
celebrate | +0.025 |
party | +0.024 |
Neg logit contributions | |
beast | -0.047 |
gut | -0.047 |
mistake | -0.044 |
therm | -0.044 |
inct | -0.044 |
Token | " |
Token ID | 1 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
parties | +0.010 |
tea | +0.009 |
recess | +0.008 |
celebrate | +0.007 |
picnic | +0.007 |
Neg logit contributions | |
beast | -0.010 |
gut | -0.010 |
mistake | -0.009 |
therm | -0.009 |
inct | -0.009 |
Token | Wow |
Token ID | 22017 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
beast | +0.063 |
gut | +0.059 |
inct | +0.058 |
passer | +0.057 |
therm | +0.057 |
Neg logit contributions | |
parties | -0.078 |
tea | -0.069 |
recess | -0.060 |
celebrate | -0.057 |
party | -0.057 |
Token | people |
Token ID | 661 |
Feature activation | +2.57e-01 |
Pos logit contributions | |
parties | +0.034 |
tea | +0.029 |
recess | +0.022 |
celebrate | +0.022 |
party | +0.021 |
Neg logit contributions | |
gut | -0.058 |
beast | -0.057 |
inct | -0.056 |
swollen | -0.055 |
therm | -0.054 |
Token | use |
Token ID | 779 |
Feature activation | -2.70e+00 |
Pos logit contributions | |
gut | +0.011 |
beast | +0.011 |
inct | +0.011 |
swollen | +0.011 |
therm | +0.011 |
Neg logit contributions | |
parties | -0.006 |
tea | -0.005 |
celebrate | -0.004 |
wear | -0.004 |
party | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | +4.27e-01 |
Pos logit contributions | |
parties | +0.072 |
tea | +0.060 |
celebrate | +0.048 |
recess | +0.047 |
party | +0.045 |
Neg logit contributions | |
beast | -0.111 |
gut | -0.110 |
inct | -0.106 |
swollen | -0.105 |
passer | -0.102 |
Token | buy |
Token ID | 2822 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
beast | +0.017 |
gut | +0.017 |
inct | +0.016 |
swollen | +0.016 |
therm | +0.015 |
Neg logit contributions | |
parties | -0.012 |
tea | -0.010 |
celebrate | -0.009 |
recess | -0.008 |
wear | -0.008 |
Token | things |
Token ID | 1243 |
Feature activation | -2.40e+00 |
Pos logit contributions | |
parties | +0.060 |
tea | +0.052 |
recess | +0.043 |
celebrate | +0.042 |
party | +0.040 |
Neg logit contributions | |
beast | -0.072 |
gut | -0.072 |
inct | -0.067 |
mistake | -0.067 |
swollen | -0.067 |
Token | they |
Token ID | 484 |
Feature activation | -8.15e+00 |
Pos logit contributions | |
parties | +0.062 |
tea | +0.052 |
celebrate | +0.040 |
recess | +0.040 |
party | +0.038 |
Neg logit contributions | |
beast | -0.101 |
gut | -0.099 |
inct | -0.094 |
swollen | -0.094 |
mistake | -0.094 |
Token | want |
Token ID | 765 |
Feature activation | -3.51e+00 |
Pos logit contributions | |
parties | +0.236 |
tea | +0.197 |
celebrate | +0.168 |
recess | +0.159 |
wear | +0.158 |
Neg logit contributions | |
beast | -0.317 |
gut | -0.314 |
swollen | -0.297 |
inct | -0.295 |
therm | -0.292 |
Token | or |
Token ID | 393 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
parties | +0.097 |
tea | +0.085 |
recess | +0.070 |
celebrate | +0.066 |
party | +0.063 |
Neg logit contributions | |
beast | -0.135 |
gut | -0.134 |
mistake | -0.127 |
therm | -0.125 |
inct | -0.125 |
Token | need |
Token ID | 761 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
parties | +0.032 |
tea | +0.028 |
celebrate | +0.024 |
wear | +0.023 |
party | +0.023 |
Neg logit contributions | |
beast | -0.038 |
inct | -0.037 |
gut | -0.037 |
swollen | -0.034 |
therm | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
parties | +0.053 |
tea | +0.045 |
recess | +0.036 |
celebrate | +0.035 |
party | +0.033 |
Neg logit contributions | |
beast | -0.083 |
gut | -0.083 |
inct | -0.079 |
swollen | -0.078 |
mistake | -0.078 |
Token |
|
Token ID | 198 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
parties | +0.039 |
tea | +0.034 |
recess | +0.027 |
celebrate | +0.026 |
party | +0.025 |
Neg logit contributions | |
beast | -0.052 |
gut | -0.052 |
mistake | -0.049 |
swollen | -0.048 |
inct | -0.048 |
Token | and |
Token ID | 290 |
Feature activation | -6.38e-01 |
Pos logit contributions | |
parties | +0.008 |
tea | +0.007 |
recess | +0.006 |
celebrate | +0.005 |
party | +0.005 |
Neg logit contributions | |
gut | -0.009 |
beast | -0.009 |
mistake | -0.009 |
therm | -0.008 |
inct | -0.008 |
Token | enjoyed |
Token ID | 8359 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
parties | +0.017 |
tea | +0.015 |
recess | +0.012 |
celebrate | +0.011 |
party | +0.011 |
Neg logit contributions | |
beast | -0.026 |
gut | -0.026 |
therm | -0.024 |
mistake | -0.024 |
inct | -0.024 |
Token | the |
Token ID | 262 |
Feature activation | -8.51e-01 |
Pos logit contributions | |
parties | +0.054 |
tea | +0.047 |
recess | +0.040 |
celebrate | +0.038 |
party | +0.037 |
Neg logit contributions | |
gut | -0.063 |
beast | -0.063 |
mistake | -0.059 |
therm | -0.058 |
inct | -0.057 |
Token | beautiful |
Token ID | 4950 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
parties | +0.027 |
tea | +0.024 |
recess | +0.020 |
celebrate | +0.019 |
party | +0.018 |
Neg logit contributions | |
beast | -0.030 |
gut | -0.029 |
mistake | -0.028 |
therm | -0.027 |
inct | -0.027 |
Token | sights |
Token ID | 21343 |
Feature activation | -8.79e-01 |
Pos logit contributions | |
parties | +0.048 |
tea | +0.042 |
recess | +0.036 |
celebrate | +0.035 |
picnic | +0.034 |
Neg logit contributions | |
gut | -0.048 |
beast | -0.047 |
swollen | -0.045 |
therm | -0.044 |
passer | -0.044 |
Token | they |
Token ID | 484 |
Feature activation | -8.14e+00 |
Pos logit contributions | |
parties | +0.028 |
tea | +0.025 |
recess | +0.022 |
celebrate | +0.021 |
outfits | +0.020 |
Neg logit contributions | |
beast | -0.029 |
gut | -0.027 |
therm | -0.026 |
cut | -0.026 |
ringing | -0.026 |
Token | found |
Token ID | 1043 |
Feature activation | -3.34e+00 |
Pos logit contributions | |
parties | +0.265 |
tea | +0.252 |
recess | +0.196 |
sandwiches | +0.189 |
picnic | +0.188 |
Neg logit contributions | |
beast | -0.265 |
gut | -0.252 |
therm | -0.234 |
Owl | -0.233 |
injured | -0.233 |
Token | . |
Token ID | 13 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
parties | +0.095 |
tea | +0.086 |
recess | +0.071 |
celebrate | +0.068 |
picnic | +0.064 |
Neg logit contributions | |
beast | -0.123 |
gut | -0.121 |
mistake | -0.115 |
cut | -0.113 |
therm | -0.112 |
Token | At |
Token ID | 1629 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
parties | +0.061 |
tea | +0.054 |
recess | +0.050 |
skirts | +0.046 |
outfits | +0.046 |
Neg logit contributions | |
beast | -0.045 |
gut | -0.044 |
inct | -0.043 |
mistake | -0.042 |
therm | -0.040 |
Token | the |
Token ID | 262 |
Feature activation | +6.18e-01 |
Pos logit contributions | |
beast | +0.064 |
gut | +0.063 |
mistake | +0.060 |
therm | +0.058 |
inct | +0.058 |
Neg logit contributions | |
parties | -0.066 |
tea | -0.059 |
recess | -0.049 |
celebrate | -0.048 |
party | -0.046 |
Token | end |
Token ID | 886 |
Feature activation | +3.18e+00 |
Pos logit contributions | |
beast | +0.020 |
gut | +0.020 |
mistake | +0.019 |
swollen | +0.019 |
therm | +0.018 |
Neg logit contributions | |
parties | -0.021 |
tea | -0.019 |
recess | -0.016 |
celebrate | -0.016 |
party | -0.015 |
Token | of |
Token ID | 286 |
Feature activation | -2.39e-01 |
Pos logit contributions | |
beast | +0.054 |
gut | +0.053 |
inct | +0.050 |
mistake | +0.050 |
swollen | +0.050 |
Neg logit contributions | |
parties | -0.041 |
tea | -0.036 |
recess | -0.029 |
celebrate | -0.028 |
party | -0.027 |
Token | all |
Token ID | 477 |
Feature activation | -2.24e+00 |
Pos logit contributions | |
parties | +0.005 |
tea | +0.004 |
recess | +0.003 |
celebrate | +0.003 |
party | +0.003 |
Neg logit contributions | |
beast | -0.011 |
gut | -0.011 |
inct | -0.011 |
swollen | -0.011 |
therm | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
parties | +0.057 |
tea | +0.049 |
celebrate | +0.037 |
recess | +0.035 |
party | +0.034 |
Neg logit contributions | |
beast | -0.095 |
gut | -0.094 |
inct | -0.090 |
swollen | -0.090 |
therm | -0.089 |
Token | wonderful |
Token ID | 7932 |
Feature activation | -2.50e+00 |
Pos logit contributions | |
parties | +0.038 |
tea | +0.033 |
recess | +0.029 |
celebrate | +0.027 |
party | +0.026 |
Neg logit contributions | |
beast | -0.041 |
gut | -0.041 |
mistake | -0.039 |
therm | -0.038 |
inct | -0.037 |
Token | things |
Token ID | 1243 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
parties | +0.086 |
tea | +0.077 |
recess | +0.069 |
celebrate | +0.065 |
picnic | +0.061 |
Neg logit contributions | |
beast | -0.078 |
gut | -0.076 |
mistake | -0.074 |
therm | -0.070 |
inct | -0.069 |
Token | they |
Token ID | 484 |
Feature activation | -8.13e+00 |
Pos logit contributions | |
parties | +0.030 |
tea | +0.026 |
recess | +0.021 |
celebrate | +0.019 |
picnic | +0.018 |
Neg logit contributions | |
beast | -0.046 |
gut | -0.046 |
mistake | -0.044 |
therm | -0.043 |
inct | -0.043 |
Token | had |
Token ID | 550 |
Feature activation | -6.70e+00 |
Pos logit contributions | |
parties | +0.244 |
tea | +0.226 |
recess | +0.185 |
picnic | +0.169 |
outfits | +0.164 |
Neg logit contributions | |
beast | -0.281 |
gut | -0.277 |
mistake | -0.261 |
cut | -0.255 |
guardian | -0.253 |
Token | done |
Token ID | 1760 |
Feature activation | -2.95e+00 |
Pos logit contributions | |
parties | +0.198 |
tea | +0.189 |
recess | +0.156 |
picnic | +0.144 |
celebrate | +0.142 |
Neg logit contributions | |
gut | -0.214 |
beast | -0.213 |
mistake | -0.211 |
inct | -0.196 |
injured | -0.196 |
Token | . |
Token ID | 13 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
parties | +0.085 |
tea | +0.076 |
recess | +0.064 |
picnic | +0.060 |
celebrate | +0.059 |
Neg logit contributions | |
gut | -0.107 |
beast | -0.105 |
mistake | -0.103 |
inct | -0.098 |
therm | -0.097 |
Token | It |
Token ID | 632 |
Feature activation | +3.05e+00 |
Pos logit contributions | |
parties | +0.040 |
tea | +0.035 |
recess | +0.030 |
celebrate | +0.028 |
party | +0.028 |
Neg logit contributions | |
beast | -0.044 |
gut | -0.044 |
mistake | -0.041 |
inct | -0.041 |
therm | -0.041 |
Token | was |
Token ID | 373 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
gut | +0.114 |
beast | +0.111 |
rum | +0.104 |
therm | +0.103 |
swollen | +0.102 |
Neg logit contributions | |
parties | -0.090 |
tea | -0.082 |
celebrate | -0.069 |
party | -0.065 |
together | -0.063 |
Token | . |
Token ID | 13 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
parties | +0.061 |
tea | +0.054 |
recess | +0.046 |
celebrate | +0.043 |
picnic | +0.041 |
Neg logit contributions | |
beast | -0.078 |
gut | -0.078 |
mistake | -0.075 |
cut | -0.074 |
therm | -0.073 |
Token | They |
Token ID | 1119 |
Feature activation | -3.34e+00 |
Pos logit contributions | |
parties | +0.048 |
tea | +0.041 |
recess | +0.035 |
ils | +0.033 |
party | +0.033 |
Neg logit contributions | |
gut | -0.062 |
beast | -0.061 |
therm | -0.059 |
Owl | -0.058 |
inct | -0.058 |
Token | share |
Token ID | 2648 |
Feature activation | -3.05e+00 |
Pos logit contributions | |
parties | +0.090 |
tea | +0.077 |
recess | +0.064 |
celebrate | +0.058 |
party | +0.057 |
Neg logit contributions | |
beast | -0.132 |
gut | -0.131 |
mistake | -0.124 |
therm | -0.123 |
swollen | -0.122 |
Token | their |
Token ID | 511 |
Feature activation | -2.74e+00 |
Pos logit contributions | |
parties | +0.065 |
tea | +0.051 |
recess | +0.044 |
celebrate | +0.038 |
party | +0.036 |
Neg logit contributions | |
gut | -0.134 |
beast | -0.133 |
mistake | -0.131 |
therm | -0.128 |
swollen | -0.127 |
Token | snow |
Token ID | 6729 |
Feature activation | -3.22e+00 |
Pos logit contributions | |
parties | +0.087 |
tea | +0.075 |
recess | +0.066 |
celebrate | +0.063 |
party | +0.059 |
Neg logit contributions | |
gut | -0.094 |
beast | -0.094 |
mistake | -0.090 |
therm | -0.088 |
swollen | -0.086 |
Token | balls |
Token ID | 21591 |
Feature activation | -8.12e+00 |
Pos logit contributions | |
parties | +0.143 |
tea | +0.129 |
recess | +0.123 |
celebrate | +0.116 |
wear | +0.112 |
Neg logit contributions | |
beast | -0.067 |
gut | -0.067 |
cut | -0.062 |
mistake | -0.061 |
therm | -0.059 |
Token | and |
Token ID | 290 |
Feature activation | -3.00e+00 |
Pos logit contributions | |
parties | +0.270 |
tea | +0.240 |
recess | +0.208 |
celebrate | +0.197 |
picnic | +0.190 |
Neg logit contributions | |
beast | -0.266 |
gut | -0.265 |
mistake | -0.250 |
therm | -0.245 |
cut | -0.243 |
Token | snow |
Token ID | 6729 |
Feature activation | -3.25e+00 |
Pos logit contributions | |
parties | +0.078 |
tea | +0.068 |
recess | +0.054 |
celebrate | +0.051 |
party | +0.049 |
Neg logit contributions | |
beast | -0.122 |
gut | -0.122 |
mistake | -0.115 |
inct | -0.115 |
therm | -0.114 |
Token | men |
Token ID | 3653 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
parties | +0.122 |
tea | +0.109 |
recess | +0.100 |
celebrate | +0.095 |
wear | +0.091 |
Neg logit contributions | |
beast | -0.091 |
gut | -0.090 |
cut | -0.085 |
mistake | -0.084 |
therm | -0.083 |
Token | . |
Token ID | 13 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
parties | +0.042 |
tea | +0.036 |
recess | +0.030 |
celebrate | +0.028 |
party | +0.027 |
Neg logit contributions | |
beast | -0.059 |
gut | -0.059 |
mistake | -0.056 |
inct | -0.055 |
therm | -0.055 |
Token | They |
Token ID | 1119 |
Feature activation | -3.43e+00 |
Pos logit contributions | |
parties | +0.063 |
tea | +0.053 |
ils | +0.049 |
recess | +0.047 |
party | +0.046 |
Neg logit contributions | |
gut | -0.070 |
Owl | -0.067 |
therm | -0.066 |
beast | -0.066 |
inct | -0.065 |
Token | They |
Token ID | 1119 |
Feature activation | -3.16e+00 |
Pos logit contributions | |
parties | +0.032 |
tea | +0.026 |
ils | +0.024 |
recess | +0.024 |
celebrate | +0.022 |
Neg logit contributions | |
gut | -0.038 |
therm | -0.036 |
beast | -0.036 |
mistake | -0.036 |
inct | -0.036 |
Token | had |
Token ID | 550 |
Feature activation | -3.07e+00 |
Pos logit contributions | |
parties | +0.097 |
tea | +0.085 |
recess | +0.074 |
celebrate | +0.067 |
party | +0.065 |
Neg logit contributions | |
beast | -0.111 |
gut | -0.109 |
therm | -0.105 |
cut | -0.105 |
mistake | -0.103 |
Token | fun |
Token ID | 1257 |
Feature activation | -5.86e+00 |
Pos logit contributions | |
parties | +0.098 |
tea | +0.082 |
recess | +0.078 |
celebrate | +0.071 |
party | +0.068 |
Neg logit contributions | |
beast | -0.103 |
gut | -0.099 |
therm | -0.098 |
swollen | -0.096 |
mistake | -0.095 |
Token | throwing |
Token ID | 9644 |
Feature activation | -3.83e+00 |
Pos logit contributions | |
parties | +0.189 |
tea | +0.164 |
recess | +0.143 |
celebrate | +0.137 |
party | +0.132 |
Neg logit contributions | |
beast | -0.201 |
gut | -0.199 |
mistake | -0.190 |
therm | -0.187 |
inct | -0.185 |
Token | snow |
Token ID | 6729 |
Feature activation | -2.94e+00 |
Pos logit contributions | |
parties | +0.133 |
tea | +0.115 |
recess | +0.103 |
celebrate | +0.099 |
party | +0.096 |
Neg logit contributions | |
beast | -0.122 |
gut | -0.119 |
therm | -0.112 |
mistake | -0.112 |
inct | -0.111 |
Token | balls |
Token ID | 21591 |
Feature activation | -8.11e+00 |
Pos logit contributions | |
parties | +0.101 |
tea | +0.088 |
recess | +0.084 |
celebrate | +0.080 |
wear | +0.075 |
Neg logit contributions | |
beast | -0.090 |
cut | -0.086 |
gut | -0.086 |
inct | -0.083 |
therm | -0.082 |
Token | at |
Token ID | 379 |
Feature activation | -7.10e-01 |
Pos logit contributions | |
parties | +0.263 |
tea | +0.228 |
recess | +0.206 |
celebrate | +0.202 |
picnic | +0.185 |
Neg logit contributions | |
beast | -0.255 |
gut | -0.248 |
cut | -0.243 |
mistake | -0.239 |
therm | -0.236 |
Token | each |
Token ID | 1123 |
Feature activation | -3.73e+00 |
Pos logit contributions | |
parties | +0.022 |
tea | +0.020 |
recess | +0.016 |
celebrate | +0.016 |
ets | +0.016 |
Neg logit contributions | |
beast | -0.025 |
gut | -0.024 |
mistake | -0.023 |
therm | -0.023 |
inct | -0.022 |
Token | other |
Token ID | 584 |
Feature activation | -5.48e-02 |
Pos logit contributions | |
parties | +0.135 |
tea | +0.120 |
recess | +0.105 |
celebrate | +0.101 |
party | +0.097 |
Neg logit contributions | |
beast | -0.115 |
gut | -0.112 |
mistake | -0.107 |
therm | -0.105 |
inct | -0.102 |
Token | . |
Token ID | 13 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
parties | +0.001 |
tea | +0.001 |
recess | +0.001 |
celebrate | +0.001 |
party | +0.001 |
Neg logit contributions | |
beast | -0.002 |
gut | -0.002 |
mistake | -0.002 |
therm | -0.002 |
inct | -0.002 |
Token | But |
Token ID | 887 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
parties | +0.051 |
tea | +0.043 |
recess | +0.039 |
ils | +0.038 |
celebrate | +0.038 |
Neg logit contributions | |
gut | -0.039 |
beast | -0.037 |
mistake | -0.037 |
inct | -0.037 |
therm | -0.035 |
Token | good |
Token ID | 922 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
parties | +0.064 |
tea | +0.055 |
recess | +0.049 |
picnic | +0.047 |
ets | +0.047 |
Neg logit contributions | |
gut | -0.056 |
beast | -0.053 |
mistake | -0.052 |
guardian | -0.050 |
reckless | -0.050 |
Token | friends |
Token ID | 2460 |
Feature activation | -4.89e+00 |
Pos logit contributions | |
parties | +0.043 |
tea | +0.039 |
recess | +0.033 |
celebrate | +0.033 |
picnic | +0.032 |
Neg logit contributions | |
gut | -0.046 |
beast | -0.044 |
mistake | -0.043 |
reckless | -0.041 |
adult | -0.040 |
Token | and |
Token ID | 290 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
parties | +0.168 |
tea | +0.153 |
recess | +0.137 |
celebrate | +0.125 |
picnic | +0.123 |
Neg logit contributions | |
gut | -0.152 |
beast | -0.151 |
mistake | -0.141 |
cut | -0.134 |
therm | -0.133 |
Token | share |
Token ID | 2648 |
Feature activation | -3.35e+00 |
Pos logit contributions | |
parties | +0.049 |
tea | +0.042 |
recess | +0.033 |
celebrate | +0.032 |
party | +0.030 |
Neg logit contributions | |
beast | -0.083 |
gut | -0.082 |
inct | -0.078 |
mistake | -0.078 |
therm | -0.078 |
Token | everything |
Token ID | 2279 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
parties | +0.091 |
tea | +0.078 |
recess | +0.069 |
celebrate | +0.061 |
party | +0.060 |
Neg logit contributions | |
mistake | -0.125 |
gut | -0.124 |
beast | -0.123 |
swollen | -0.116 |
therm | -0.115 |
Token | they |
Token ID | 484 |
Feature activation | -8.09e+00 |
Pos logit contributions | |
parties | +0.049 |
tea | +0.041 |
recess | +0.040 |
picnic | +0.035 |
celebrate | +0.033 |
Neg logit contributions | |
gut | -0.054 |
mistake | -0.053 |
cut | -0.051 |
beast | -0.051 |
swollen | -0.050 |
Token | found |
Token ID | 1043 |
Feature activation | -3.39e+00 |
Pos logit contributions | |
parties | +0.260 |
tea | +0.252 |
recess | +0.209 |
text | +0.198 |
outfits | +0.198 |
Neg logit contributions | |
beast | -0.229 |
reckless | -0.224 |
gut | -0.223 |
injured | -0.219 |
mistake | -0.216 |
Token | in |
Token ID | 287 |
Feature activation | +9.92e-02 |
Pos logit contributions | |
parties | +0.097 |
tea | +0.086 |
recess | +0.073 |
celebrate | +0.068 |
picnic | +0.065 |
Neg logit contributions | |
beast | -0.124 |
gut | -0.123 |
mistake | -0.119 |
cut | -0.115 |
therm | -0.114 |
Token | the |
Token ID | 262 |
Feature activation | -2.58e+00 |
Pos logit contributions | |
beast | +0.004 |
gut | +0.004 |
inct | +0.004 |
therm | +0.004 |
cut | +0.004 |
Neg logit contributions | |
parties | -0.003 |
tea | -0.002 |
recess | -0.002 |
celebrate | -0.002 |
party | -0.002 |
Token | park |
Token ID | 3952 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
parties | +0.078 |
tea | +0.072 |
recess | +0.059 |
party | +0.058 |
picnic | +0.057 |
Neg logit contributions | |
gut | -0.094 |
beast | -0.093 |
rum | -0.090 |
swollen | -0.089 |
inct | -0.088 |
Token | from |
Token ID | 422 |
Feature activation | +6.44e-01 |
Pos logit contributions | |
parties | +0.053 |
tea | +0.046 |
recess | +0.038 |
celebrate | +0.036 |
picnic | +0.034 |
Neg logit contributions | |
beast | -0.076 |
gut | -0.076 |
mistake | -0.072 |
therm | -0.070 |
inct | -0.070 |
Token | snow |
Token ID | 6729 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
parties | +0.068 |
recess | +0.057 |
tea | +0.056 |
dates | +0.050 |
party | +0.050 |
Neg logit contributions | |
gut | -0.069 |
beast | -0.068 |
mistake | -0.066 |
therm | -0.065 |
injured | -0.062 |
Token | men |
Token ID | 3653 |
Feature activation | -5.42e-01 |
Pos logit contributions | |
parties | +0.053 |
tea | +0.047 |
recess | +0.045 |
celebrate | +0.040 |
wear | +0.039 |
Neg logit contributions | |
beast | -0.042 |
gut | -0.042 |
mistake | -0.040 |
cut | -0.039 |
inct | -0.039 |
Token | , |
Token ID | 11 |
Feature activation | -3.04e+00 |
Pos logit contributions | |
parties | +0.018 |
tea | +0.016 |
recess | +0.015 |
celebrate | +0.013 |
picnic | +0.013 |
Neg logit contributions | |
gut | -0.017 |
beast | -0.017 |
mistake | -0.016 |
therm | -0.016 |
cut | -0.016 |
Token | threw |
Token ID | 9617 |
Feature activation | -3.90e+00 |
Pos logit contributions | |
parties | +0.095 |
tea | +0.088 |
celebrate | +0.068 |
recess | +0.067 |
party | +0.066 |
Neg logit contributions | |
beast | -0.111 |
gut | -0.110 |
inct | -0.107 |
mistake | -0.103 |
passer | -0.101 |
Token | snow |
Token ID | 6729 |
Feature activation | -2.76e+00 |
Pos logit contributions | |
parties | +0.127 |
tea | +0.108 |
recess | +0.097 |
celebrate | +0.092 |
party | +0.090 |
Neg logit contributions | |
beast | -0.131 |
gut | -0.128 |
therm | -0.123 |
mistake | -0.121 |
swollen | -0.119 |
Token | balls |
Token ID | 21591 |
Feature activation | -8.09e+00 |
Pos logit contributions | |
parties | +0.086 |
tea | +0.075 |
recess | +0.067 |
celebrate | +0.063 |
party | +0.059 |
Neg logit contributions | |
beast | -0.095 |
gut | -0.095 |
therm | -0.089 |
cut | -0.088 |
mistake | -0.088 |
Token | , |
Token ID | 11 |
Feature activation | -3.42e+00 |
Pos logit contributions | |
parties | +0.279 |
tea | +0.246 |
recess | +0.220 |
celebrate | +0.214 |
picnic | +0.199 |
Neg logit contributions | |
beast | -0.244 |
gut | -0.244 |
mistake | -0.230 |
therm | -0.228 |
cut | -0.227 |
Token | and |
Token ID | 290 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
parties | +0.105 |
tea | +0.095 |
recess | +0.076 |
celebrate | +0.074 |
party | +0.072 |
Neg logit contributions | |
beast | -0.125 |
gut | -0.124 |
inct | -0.117 |
mistake | -0.116 |
therm | -0.113 |
Token | slid |
Token ID | 27803 |
Feature activation | -3.14e+00 |
Pos logit contributions | |
parties | +0.051 |
tea | +0.045 |
recess | +0.038 |
celebrate | +0.036 |
party | +0.035 |
Neg logit contributions | |
beast | -0.059 |
gut | -0.059 |
mistake | -0.055 |
inct | -0.055 |
therm | -0.055 |
Token | down |
Token ID | 866 |
Feature activation | +5.82e-01 |
Pos logit contributions | |
parties | +0.099 |
tea | +0.085 |
recess | +0.074 |
celebrate | +0.072 |
party | +0.069 |
Neg logit contributions | |
beast | -0.109 |
gut | -0.108 |
therm | -0.102 |
inct | -0.101 |
swollen | -0.100 |
Token | hills |
Token ID | 18639 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
gut | +0.024 |
beast | +0.024 |
inct | +0.023 |
mistake | +0.023 |
swollen | +0.022 |
Neg logit contributions | |
parties | -0.016 |
tea | -0.014 |
recess | -0.011 |
together | -0.010 |
celebrate | -0.010 |
Token | on |
Token ID | 319 |
Feature activation | +3.15e-01 |
Pos logit contributions | |
parties | +0.057 |
tea | +0.049 |
recess | +0.037 |
celebrate | +0.037 |
picnic | +0.037 |
Neg logit contributions | |
beast | -0.083 |
gut | -0.082 |
inct | -0.078 |
therm | -0.077 |
severe | -0.076 |
Token | , |
Token ID | 11 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
beast | +0.012 |
gut | +0.012 |
mistake | +0.011 |
therm | +0.011 |
inct | +0.011 |
Neg logit contributions | |
parties | -0.009 |
tea | -0.008 |
recess | -0.007 |
celebrate | -0.006 |
party | -0.006 |
Token | they |
Token ID | 484 |
Feature activation | -4.01e+00 |
Pos logit contributions | |
parties | +0.056 |
tea | +0.049 |
recess | +0.039 |
celebrate | +0.038 |
party | +0.036 |
Neg logit contributions | |
beast | -0.083 |
gut | -0.083 |
mistake | -0.079 |
inct | -0.078 |
therm | -0.078 |
Token | shared |
Token ID | 4888 |
Feature activation | -4.08e+00 |
Pos logit contributions | |
parties | +0.111 |
tea | +0.097 |
recess | +0.079 |
celebrate | +0.074 |
party | +0.071 |
Neg logit contributions | |
beast | -0.156 |
gut | -0.155 |
mistake | -0.146 |
therm | -0.145 |
inct | -0.145 |
Token | everything |
Token ID | 2279 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
parties | +0.112 |
tea | +0.094 |
recess | +0.085 |
celebrate | +0.074 |
party | +0.073 |
Neg logit contributions | |
beast | -0.153 |
gut | -0.153 |
mistake | -0.149 |
swollen | -0.142 |
therm | -0.142 |
Token | they |
Token ID | 484 |
Feature activation | -8.07e+00 |
Pos logit contributions | |
parties | +0.029 |
tea | +0.024 |
recess | +0.022 |
celebrate | +0.018 |
picnic | +0.018 |
Neg logit contributions | |
gut | -0.047 |
beast | -0.046 |
mistake | -0.044 |
cut | -0.044 |
therm | -0.043 |
Token | found |
Token ID | 1043 |
Feature activation | -3.40e+00 |
Pos logit contributions | |
parties | +0.247 |
tea | +0.232 |
recess | +0.184 |
outfits | +0.177 |
birthday | +0.171 |
Neg logit contributions | |
beast | -0.267 |
gut | -0.258 |
reckless | -0.253 |
injured | -0.252 |
Owl | -0.251 |
Token | in |
Token ID | 287 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
parties | +0.098 |
tea | +0.086 |
recess | +0.071 |
celebrate | +0.068 |
party | +0.065 |
Neg logit contributions | |
beast | -0.128 |
gut | -0.127 |
mistake | -0.120 |
therm | -0.119 |
inct | -0.119 |
Token | the |
Token ID | 262 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
beast | +0.004 |
gut | +0.004 |
inct | +0.004 |
therm | +0.004 |
cut | +0.004 |
Neg logit contributions | |
parties | -0.003 |
tea | -0.003 |
recess | -0.002 |
celebrate | -0.002 |
party | -0.002 |
Token | forest |
Token ID | 8222 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
parties | +0.066 |
tea | +0.064 |
picnic | +0.053 |
party | +0.053 |
recess | +0.050 |
Neg logit contributions | |
gut | -0.072 |
rum | -0.071 |
beast | -0.068 |
cut | -0.067 |
iously | -0.067 |
Token | and |
Token ID | 290 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
parties | +0.004 |
tea | +0.003 |
celebrate | +0.002 |
party | +0.002 |
together | +0.002 |
Neg logit contributions | |
beast | -0.007 |
gut | -0.007 |
inct | -0.006 |
therm | -0.006 |
swollen | -0.006 |
Token | ty |
Token ID | 774 |
Feature activation | -5.88e-01 |
Pos logit contributions | |
gut | +0.090 |
beast | +0.088 |
inct | +0.085 |
therm | +0.084 |
mistake | +0.084 |
Neg logit contributions | |
parties | -0.058 |
tea | -0.051 |
celebrate | -0.037 |
party | -0.037 |
played | -0.037 |
Token | and |
Token ID | 290 |
Feature activation | +4.29e-01 |
Pos logit contributions | |
parties | +0.017 |
tea | +0.015 |
played | +0.012 |
celebrate | +0.012 |
party | +0.011 |
Neg logit contributions | |
beast | -0.023 |
gut | -0.022 |
inct | -0.021 |
therm | -0.021 |
swollen | -0.021 |
Token | Sammy |
Token ID | 44067 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
beast | +0.015 |
gut | +0.015 |
mistake | +0.014 |
therm | +0.014 |
inct | +0.014 |
Neg logit contributions | |
parties | -0.014 |
tea | -0.012 |
recess | -0.010 |
celebrate | -0.010 |
party | -0.010 |
Token | shared |
Token ID | 4888 |
Feature activation | -3.88e+00 |
Pos logit contributions | |
parties | +0.054 |
tea | +0.048 |
recess | +0.041 |
picnic | +0.035 |
birthday | +0.034 |
Neg logit contributions | |
beast | -0.078 |
gut | -0.077 |
cut | -0.075 |
mistake | -0.074 |
Owl | -0.073 |
Token | everything |
Token ID | 2279 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
parties | +0.103 |
tea | +0.088 |
recess | +0.087 |
ets | +0.079 |
dates | +0.077 |
Neg logit contributions | |
mistake | -0.132 |
gut | -0.129 |
beast | -0.123 |
therm | -0.122 |
severe | -0.122 |
Token | they |
Token ID | 484 |
Feature activation | -8.05e+00 |
Pos logit contributions | |
parties | +0.030 |
tea | +0.025 |
recess | +0.024 |
picnic | +0.019 |
celebrate | +0.018 |
Neg logit contributions | |
gut | -0.049 |
beast | -0.047 |
cut | -0.047 |
mistake | -0.046 |
guardian | -0.046 |
Token | found |
Token ID | 1043 |
Feature activation | -3.17e+00 |
Pos logit contributions | |
tea | +0.232 |
parties | +0.232 |
sandwiches | +0.181 |
recess | +0.176 |
outfits | +0.176 |
Neg logit contributions | |
Owl | -0.241 |
beast | -0.240 |
reckless | -0.231 |
Eagle | -0.231 |
gut | -0.228 |
Token | and |
Token ID | 290 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
parties | +0.097 |
tea | +0.086 |
recess | +0.075 |
celebrate | +0.069 |
picnic | +0.067 |
Neg logit contributions | |
beast | -0.111 |
gut | -0.111 |
mistake | -0.105 |
cut | -0.102 |
therm | -0.102 |
Token | were |
Token ID | 547 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
parties | +0.048 |
tea | +0.041 |
recess | +0.033 |
celebrate | +0.030 |
party | +0.029 |
Neg logit contributions | |
gut | -0.080 |
beast | -0.079 |
mistake | -0.075 |
swollen | -0.074 |
cut | -0.074 |
Token | the |
Token ID | 262 |
Feature activation | -4.48e-02 |
Pos logit contributions | |
parties | +0.048 |
tea | +0.042 |
recess | +0.037 |
picnic | +0.034 |
celebrate | +0.034 |
Neg logit contributions | |
gut | -0.049 |
beast | -0.048 |
mistake | -0.046 |
swollen | -0.045 |
therm | -0.045 |
Token | happiest |
Token ID | 49414 |
Feature activation | -9.10e-01 |
Pos logit contributions | |
parties | +0.001 |
tea | +0.001 |
recess | +0.001 |
celebrate | +0.001 |
party | +0.001 |
Neg logit contributions | |
beast | -0.002 |
gut | -0.002 |
mistake | -0.002 |
therm | -0.002 |
inct | -0.002 |
Token | liked |
Token ID | 8288 |
Feature activation | -2.62e-01 |
Pos logit contributions | |
parties | +0.126 |
tea | +0.117 |
recess | +0.103 |
dates | +0.101 |
sandwiches | +0.101 |
Neg logit contributions | |
claw | -0.061 |
beast | -0.057 |
gut | -0.056 |
panicked | -0.055 |
shaking | -0.055 |
Token | to |
Token ID | 284 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
parties | +0.010 |
tea | +0.009 |
recess | +0.008 |
celebrate | +0.007 |
party | +0.007 |
Neg logit contributions | |
beast | -0.007 |
gut | -0.007 |
mistake | -0.007 |
therm | -0.007 |
swollen | -0.007 |
Token | examine |
Token ID | 10716 |
Feature activation | -7.39e-01 |
Pos logit contributions | |
claw | +0.043 |
gut | +0.043 |
beast | +0.042 |
mistake | +0.041 |
spike | +0.039 |
Neg logit contributions | |
parties | -0.053 |
tea | -0.047 |
text | -0.044 |
ils | -0.043 |
ouses | -0.043 |
Token | different |
Token ID | 1180 |
Feature activation | -5.59e-02 |
Pos logit contributions | |
parties | +0.025 |
tea | +0.021 |
recess | +0.020 |
celebrate | +0.019 |
party | +0.018 |
Neg logit contributions | |
beast | -0.024 |
gut | -0.023 |
therm | -0.022 |
mistake | -0.022 |
swollen | -0.021 |
Token | things |
Token ID | 1243 |
Feature activation | -2.48e+00 |
Pos logit contributions | |
parties | +0.002 |
tea | +0.002 |
recess | +0.001 |
celebrate | +0.001 |
party | +0.001 |
Neg logit contributions | |
beast | -0.002 |
gut | -0.002 |
mistake | -0.002 |
swollen | -0.002 |
therm | -0.002 |
Token | they |
Token ID | 484 |
Feature activation | -8.02e+00 |
Pos logit contributions | |
parties | +0.075 |
tea | +0.066 |
recess | +0.057 |
celebrate | +0.053 |
party | +0.051 |
Neg logit contributions | |
beast | -0.089 |
gut | -0.089 |
mistake | -0.083 |
therm | -0.083 |
cut | -0.082 |
Token | found |
Token ID | 1043 |
Feature activation | -3.51e+00 |
Pos logit contributions | |
parties | +0.258 |
tea | +0.239 |
recess | +0.197 |
celebrate | +0.184 |
picnic | +0.183 |
Neg logit contributions | |
beast | -0.260 |
gut | -0.258 |
mistake | -0.239 |
therm | -0.239 |
swollen | -0.237 |
Token | , |
Token ID | 11 |
Feature activation | -3.02e+00 |
Pos logit contributions | |
parties | +0.101 |
tea | +0.089 |
recess | +0.078 |
celebrate | +0.073 |
picnic | +0.069 |
Neg logit contributions | |
beast | -0.127 |
gut | -0.126 |
mistake | -0.120 |
cut | -0.117 |
therm | -0.116 |
Token | like |
Token ID | 588 |
Feature activation | -2.58e+00 |
Pos logit contributions | |
parties | +0.080 |
tea | +0.069 |
recess | +0.056 |
celebrate | +0.052 |
party | +0.050 |
Neg logit contributions | |
beast | -0.121 |
gut | -0.121 |
mistake | -0.114 |
therm | -0.113 |
inct | -0.113 |
Token | leaves |
Token ID | 5667 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
parties | +0.080 |
tea | +0.070 |
recess | +0.062 |
celebrate | +0.058 |
party | +0.056 |
Neg logit contributions | |
beast | -0.089 |
gut | -0.087 |
mistake | -0.083 |
therm | -0.082 |
swollen | -0.081 |
Token | , |
Token ID | 11 |
Feature activation | -3.38e+00 |
Pos logit contributions | |
parties | +0.052 |
tea | +0.045 |
recess | +0.034 |
celebrate | +0.033 |
party | +0.033 |
Neg logit contributions | |
beast | -0.078 |
gut | -0.078 |
inct | -0.074 |
mistake | -0.073 |
swollen | -0.072 |
Token | He |
Token ID | 679 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
beast | +0.001 |
gut | +0.001 |
mistake | +0.001 |
swollen | +0.001 |
therm | +0.001 |
Neg logit contributions | |
parties | -0.001 |
tea | -0.001 |
recess | -0.000 |
celebrate | -0.000 |
party | -0.000 |
Token | cooked |
Token ID | 15847 |
Feature activation | -9.51e-01 |
Pos logit contributions | |
gut | +0.037 |
beast | +0.036 |
inct | +0.034 |
mistake | +0.033 |
swollen | +0.033 |
Neg logit contributions | |
parties | -0.032 |
tea | -0.029 |
celebrate | -0.024 |
party | -0.023 |
recess | -0.023 |
Token | spaghetti |
Token ID | 48241 |
Feature activation | -5.36e-01 |
Pos logit contributions | |
parties | +0.028 |
tea | +0.024 |
recess | +0.021 |
celebrate | +0.019 |
party | +0.018 |
Neg logit contributions | |
beast | -0.035 |
gut | -0.035 |
mistake | -0.033 |
therm | -0.033 |
inct | -0.032 |
Token | with |
Token ID | 351 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
parties | +0.016 |
tea | +0.013 |
recess | +0.012 |
celebrate | +0.011 |
picnic | +0.010 |
Neg logit contributions | |
beast | -0.019 |
gut | -0.019 |
mistake | -0.018 |
therm | -0.018 |
cut | -0.018 |
Token | meat |
Token ID | 6174 |
Feature activation | -9.14e-01 |
Pos logit contributions | |
parties | +0.060 |
tea | +0.051 |
recess | +0.044 |
celebrate | +0.041 |
party | +0.040 |
Neg logit contributions | |
beast | -0.078 |
gut | -0.077 |
therm | -0.073 |
mistake | -0.073 |
swollen | -0.072 |
Token | balls |
Token ID | 21591 |
Feature activation | -7.99e+00 |
Pos logit contributions | |
parties | +0.026 |
tea | +0.023 |
recess | +0.018 |
celebrate | +0.018 |
party | +0.017 |
Neg logit contributions | |
beast | -0.035 |
gut | -0.035 |
mistake | -0.033 |
inct | -0.033 |
swollen | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | -3.30e-01 |
Pos logit contributions | |
parties | +0.246 |
tea | +0.216 |
recess | +0.184 |
celebrate | +0.174 |
party | +0.167 |
Neg logit contributions | |
beast | -0.283 |
gut | -0.281 |
mistake | -0.265 |
therm | -0.263 |
inct | -0.260 |
Token | His |
Token ID | 2399 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
parties | +0.011 |
tea | +0.009 |
recess | +0.008 |
celebrate | +0.008 |
party | +0.007 |
Neg logit contributions | |
beast | -0.012 |
gut | -0.011 |
swollen | -0.011 |
mistake | -0.011 |
severe | -0.011 |
Token | friends |
Token ID | 2460 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
beast | +0.071 |
gut | +0.070 |
inct | +0.067 |
swollen | +0.066 |
cut | +0.066 |
Neg logit contributions | |
parties | -0.054 |
tea | -0.049 |
recess | -0.038 |
picnic | -0.037 |
celebrate | -0.037 |
Token | all |
Token ID | 477 |
Feature activation | -8.74e-01 |
Pos logit contributions | |
parties | +0.043 |
tea | +0.037 |
celebrate | +0.029 |
recess | +0.029 |
party | +0.028 |
Neg logit contributions | |
beast | -0.065 |
gut | -0.065 |
mistake | -0.060 |
inct | -0.060 |
therm | -0.060 |
Token | said |
Token ID | 531 |
Feature activation | -3.01e+00 |
Pos logit contributions | |
parties | +0.026 |
tea | +0.022 |
celebrate | +0.018 |
recess | +0.018 |
party | +0.017 |
Neg logit contributions | |
gut | -0.033 |
beast | -0.033 |
inct | -0.031 |
mistake | -0.031 |
therm | -0.030 |
Token | under |
Token ID | 739 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
parties | +0.046 |
tea | +0.040 |
recess | +0.033 |
celebrate | +0.031 |
party | +0.029 |
Neg logit contributions | |
beast | -0.068 |
gut | -0.067 |
therm | -0.064 |
mistake | -0.064 |
inct | -0.064 |
Token | the |
Token ID | 262 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
gut | +0.008 |
beast | +0.008 |
cut | +0.008 |
inct | +0.008 |
mistake | +0.007 |
Neg logit contributions | |
parties | -0.004 |
tea | -0.004 |
recess | -0.003 |
celebrate | -0.003 |
picnic | -0.003 |
Token | couch |
Token ID | 18507 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
parties | +0.043 |
tea | +0.040 |
party | +0.033 |
recess | +0.033 |
picnic | +0.033 |
Neg logit contributions | |
gut | -0.037 |
beast | -0.037 |
inct | -0.036 |
mistake | -0.033 |
cut | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
beast | +0.064 |
gut | +0.064 |
inct | +0.061 |
swollen | +0.061 |
mistake | +0.060 |
Neg logit contributions | |
parties | -0.032 |
tea | -0.027 |
party | -0.019 |
recess | -0.019 |
celebrate | -0.018 |
Token | The |
Token ID | 383 |
Feature activation | +3.24e+00 |
Pos logit contributions | |
beast | +0.007 |
swollen | +0.006 |
gut | +0.006 |
mistake | +0.006 |
ache | +0.006 |
Neg logit contributions | |
parties | -0.003 |
tea | -0.003 |
celebrate | -0.002 |
recess | -0.002 |
played | -0.002 |
Token | nos |
Token ID | 43630 |
Feature activation | +6.40e+00 |
Pos logit contributions | |
rum | +0.101 |
inct | +0.095 |
beast | +0.093 |
gut | +0.091 |
iously | +0.091 |
Neg logit contributions | |
parties | -0.112 |
tea | -0.110 |
party | -0.096 |
picnic | -0.093 |
recess | -0.092 |
Token | y |
Token ID | 88 |
Feature activation | +3.00e+00 |
Pos logit contributions | |
gut | +0.267 |
beast | +0.256 |
therm | +0.253 |
swollen | +0.253 |
inct | +0.250 |
Neg logit contributions | |
parties | -0.152 |
tea | -0.124 |
recess | -0.102 |
celebrate | -0.094 |
ashed | -0.091 |
Token | person |
Token ID | 1048 |
Feature activation | +4.85e+00 |
Pos logit contributions | |
swollen | +0.087 |
gut | +0.086 |
rum | +0.085 |
inct | +0.084 |
beast | +0.080 |
Neg logit contributions | |
tea | -0.097 |
parties | -0.097 |
party | -0.089 |
picnic | -0.085 |
play | -0.078 |
Token | was |
Token ID | 373 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
gut | +0.180 |
beast | +0.177 |
inct | +0.172 |
therm | +0.168 |
swollen | +0.168 |
Neg logit contributions | |
parties | -0.136 |
tea | -0.122 |
played | -0.104 |
party | -0.101 |
celebrate | -0.100 |
Token | determined |
Token ID | 5295 |
Feature activation | +4.00e+00 |
Pos logit contributions | |
beast | +0.055 |
gut | +0.053 |
inct | +0.053 |
therm | +0.050 |
cut | +0.050 |
Neg logit contributions | |
parties | -0.044 |
tea | -0.041 |
celebrate | -0.034 |
party | -0.032 |
together | -0.032 |
Token | to |
Token ID | 284 |
Feature activation | +3.20e+00 |
Pos logit contributions | |
beast | +0.165 |
inct | +0.163 |
swollen | +0.161 |
gut | +0.160 |
therm | +0.159 |
Neg logit contributions | |
parties | -0.102 |
tea | -0.082 |
recess | -0.066 |
together | -0.066 |
party | -0.065 |
Token | snap |
Token ID | 45380 |
Feature activation | +8.89e-03 |
Pos logit contributions | |
gut | +0.066 |
beast | +0.062 |
mistake | +0.060 |
inct | +0.056 |
swollen | +0.055 |
Neg logit contributions | |
parties | -0.057 |
tea | -0.051 |
celebrate | -0.045 |
recess | -0.045 |
wear | -0.043 |
Token | '. |
Token ID | 4458 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
gut | +0.000 |
beast | +0.000 |
mistake | +0.000 |
swollen | +0.000 |
inct | +0.000 |
Neg logit contributions | |
parties | -0.000 |
tea | -0.000 |
celebrate | -0.000 |
recess | -0.000 |
party | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
gut | +0.061 |
beast | +0.060 |
mistake | +0.058 |
swollen | +0.057 |
therm | +0.056 |
Neg logit contributions | |
parties | -0.041 |
tea | -0.037 |
celebrate | -0.030 |
recess | -0.030 |
together | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | +6.00e-01 |
Pos logit contributions | |
parties | +0.004 |
tea | +0.003 |
recess | +0.003 |
celebrate | +0.003 |
wear | +0.002 |
Neg logit contributions | |
gut | -0.006 |
beast | -0.006 |
mistake | -0.006 |
therm | -0.006 |
swollen | -0.006 |
Token | The |
Token ID | 464 |
Feature activation | +4.01e+00 |
Pos logit contributions | |
gut | +0.021 |
beast | +0.021 |
mistake | +0.020 |
therm | +0.020 |
swollen | +0.020 |
Neg logit contributions | |
parties | -0.017 |
tea | -0.015 |
celebrate | -0.013 |
recess | -0.013 |
P | -0.013 |
Token | egg |
Token ID | 5935 |
Feature activation | +6.29e+00 |
Pos logit contributions | |
rum | +0.132 |
gut | +0.124 |
shadow | +0.112 |
inct | +0.111 |
aned | +0.109 |
Neg logit contributions | |
tea | -0.134 |
party | -0.115 |
parties | -0.114 |
picnic | -0.111 |
wedding | -0.108 |
Token | looked |
Token ID | 3114 |
Feature activation | +4.14e-01 |
Pos logit contributions | |
gut | +0.253 |
beast | +0.236 |
passer | +0.233 |
rum | +0.232 |
severe | +0.228 |
Neg logit contributions | |
tea | -0.145 |
parties | -0.138 |
played | -0.125 |
party | -0.117 |
together | -0.112 |
Token | around |
Token ID | 1088 |
Feature activation | +3.14e+00 |
Pos logit contributions | |
beast | +0.016 |
gut | +0.016 |
mistake | +0.015 |
inct | +0.015 |
therm | +0.015 |
Neg logit contributions | |
parties | -0.012 |
tea | -0.010 |
recess | -0.009 |
celebrate | -0.008 |
party | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | +8.70e-01 |
Pos logit contributions | |
beast | +0.115 |
gut | +0.115 |
mistake | +0.108 |
inct | +0.107 |
therm | +0.107 |
Neg logit contributions | |
parties | -0.094 |
tea | -0.083 |
recess | -0.069 |
celebrate | -0.066 |
party | -0.064 |
Token | saw |
Token ID | 2497 |
Feature activation | +2.60e+00 |
Pos logit contributions | |
gut | +0.034 |
beast | +0.034 |
therm | +0.032 |
mistake | +0.032 |
inct | +0.032 |
Neg logit contributions | |
parties | -0.023 |
tea | -0.021 |
recess | -0.017 |
celebrate | -0.016 |
party | -0.016 |
Token | a |
Token ID | 257 |
Feature activation | -2.72e-01 |
Pos logit contributions | |
gut | +0.115 |
beast | +0.110 |
inct | +0.108 |
mistake | +0.108 |
passer | +0.107 |
Neg logit contributions | |
parties | -0.054 |
tea | -0.052 |
celebrate | -0.035 |
together | -0.035 |
recess | -0.033 |
Token | water |
Token ID | 1660 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
gut | +0.039 |
beast | +0.039 |
inct | +0.037 |
swollen | +0.036 |
severe | +0.036 |
Neg logit contributions | |
parties | -0.039 |
tea | -0.035 |
recess | -0.030 |
party | -0.028 |
celebrate | -0.028 |
Token | was |
Token ID | 373 |
Feature activation | +2.70e+00 |
Pos logit contributions | |
gut | +0.052 |
beast | +0.051 |
swollen | +0.048 |
rum | +0.048 |
inct | +0.047 |
Neg logit contributions | |
parties | -0.045 |
tea | -0.042 |
party | -0.034 |
celebrate | -0.033 |
picnic | -0.033 |
Token | gone |
Token ID | 3750 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
beast | +0.110 |
gut | +0.108 |
inct | +0.099 |
passer | +0.099 |
mistake | +0.097 |
Neg logit contributions | |
parties | -0.076 |
tea | -0.068 |
celebrate | -0.054 |
party | -0.053 |
recess | -0.052 |
Token | , |
Token ID | 11 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
beast | +0.051 |
gut | +0.051 |
mistake | +0.048 |
therm | +0.048 |
inct | +0.048 |
Neg logit contributions | |
parties | -0.037 |
tea | -0.032 |
recess | -0.027 |
celebrate | -0.025 |
party | -0.024 |
Token | the |
Token ID | 262 |
Feature activation | +3.05e+00 |
Pos logit contributions | |
beast | +0.009 |
gut | +0.009 |
mistake | +0.009 |
inct | +0.009 |
therm | +0.009 |
Neg logit contributions | |
parties | -0.006 |
tea | -0.005 |
recess | -0.004 |
celebrate | -0.004 |
party | -0.003 |
Token | town |
Token ID | 3240 |
Feature activation | +6.45e+00 |
Pos logit contributions | |
beast | +0.097 |
gut | +0.097 |
rum | +0.096 |
inct | +0.094 |
swollen | +0.091 |
Neg logit contributions | |
parties | -0.103 |
tea | -0.099 |
recess | -0.082 |
party | -0.081 |
picnic | -0.079 |
Token | was |
Token ID | 373 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
gut | +0.234 |
beast | +0.233 |
swollen | +0.228 |
rum | +0.224 |
therm | +0.224 |
Neg logit contributions | |
parties | -0.185 |
tea | -0.164 |
party | -0.145 |
celebrate | -0.143 |
played | -0.141 |
Token | very |
Token ID | 845 |
Feature activation | +3.68e+00 |
Pos logit contributions | |
beast | +0.090 |
gut | +0.087 |
inct | +0.083 |
passer | +0.081 |
Eagle | +0.079 |
Neg logit contributions | |
parties | -0.066 |
tea | -0.062 |
celebrate | -0.050 |
party | -0.049 |
picnic | -0.047 |
Token | happy |
Token ID | 3772 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
beast | +0.147 |
gut | +0.131 |
Eagle | +0.131 |
Owl | +0.131 |
inct | +0.130 |
Neg logit contributions | |
parties | -0.103 |
tea | -0.101 |
celebrate | -0.081 |
party | -0.078 |
together | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
beast | +0.038 |
gut | +0.038 |
mistake | +0.036 |
therm | +0.035 |
inct | +0.035 |
Neg logit contributions | |
parties | -0.030 |
tea | -0.026 |
recess | -0.022 |
celebrate | -0.021 |
party | -0.020 |
Token | They |
Token ID | 1119 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
beast | +0.004 |
gut | +0.004 |
mistake | +0.004 |
swollen | +0.004 |
therm | +0.004 |
Neg logit contributions | |
parties | -0.003 |
tea | -0.002 |
recess | -0.002 |
celebrate | -0.002 |
party | -0.002 |
Token | ?" |
Token ID | 1701 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
parties | +0.058 |
tea | +0.051 |
recess | +0.043 |
celebrate | +0.040 |
picnic | +0.039 |
Neg logit contributions | |
beast | -0.072 |
gut | -0.071 |
mistake | -0.067 |
therm | -0.066 |
inct | -0.066 |
Token | |
Token ID | 220 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
beast | +0.069 |
gut | +0.068 |
swollen | +0.065 |
mistake | +0.065 |
inct | +0.064 |
Neg logit contributions | |
parties | -0.052 |
tea | -0.045 |
celebrate | -0.037 |
recess | -0.037 |
party | -0.035 |
Token |
|
Token ID | 198 |
Feature activation | +2.99e-01 |
Pos logit contributions | |
gut | +0.035 |
beast | +0.035 |
inct | +0.033 |
mistake | +0.033 |
therm | +0.033 |
Neg logit contributions | |
parties | -0.038 |
tea | -0.034 |
recess | -0.030 |
party | -0.028 |
celebrate | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | +8.86e-01 |
Pos logit contributions | |
gut | +0.009 |
beast | +0.009 |
therm | +0.009 |
mistake | +0.009 |
inct | +0.009 |
Neg logit contributions | |
parties | -0.010 |
tea | -0.009 |
recess | -0.008 |
picnic | -0.007 |
party | -0.007 |
Token | The |
Token ID | 464 |
Feature activation | +4.37e+00 |
Pos logit contributions | |
beast | +0.032 |
gut | +0.031 |
mistake | +0.029 |
swollen | +0.028 |
therm | +0.028 |
Neg logit contributions | |
parties | -0.028 |
tea | -0.025 |
celebrate | -0.022 |
recess | -0.021 |
party | -0.020 |
Token | man |
Token ID | 582 |
Feature activation | +6.20e+00 |
Pos logit contributions | |
gut | +0.102 |
beast | +0.100 |
swollen | +0.097 |
inct | +0.093 |
therm | +0.092 |
Neg logit contributions | |
parties | -0.192 |
tea | -0.173 |
recess | -0.158 |
celebrate | -0.152 |
party | -0.151 |
Token | smiled |
Token ID | 13541 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
gut | +0.245 |
beast | +0.236 |
severe | +0.226 |
swollen | +0.224 |
therm | +0.221 |
Neg logit contributions | |
parties | -0.166 |
tea | -0.148 |
played | -0.127 |
celebrate | -0.124 |
party | -0.123 |
Token | and |
Token ID | 290 |
Feature activation | +9.15e-02 |
Pos logit contributions | |
gut | +0.043 |
beast | +0.043 |
mistake | +0.040 |
therm | +0.040 |
inct | +0.040 |
Neg logit contributions | |
parties | -0.044 |
tea | -0.039 |
recess | -0.034 |
celebrate | -0.032 |
party | -0.031 |
Token | answered |
Token ID | 9373 |
Feature activation | +5.15e-01 |
Pos logit contributions | |
gut | +0.003 |
beast | +0.003 |
mistake | +0.003 |
inct | +0.003 |
therm | +0.003 |
Neg logit contributions | |
parties | -0.003 |
tea | -0.003 |
recess | -0.002 |
celebrate | -0.002 |
party | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -4.53e-01 |
Pos logit contributions | |
gut | +0.019 |
beast | +0.019 |
mistake | +0.018 |
therm | +0.018 |
swollen | +0.018 |
Neg logit contributions | |
parties | -0.015 |
tea | -0.013 |
recess | -0.011 |
celebrate | -0.010 |
picnic | -0.010 |
Token | " |
Token ID | 366 |
Feature activation | +3.50e+00 |
Pos logit contributions | |
parties | +0.012 |
tea | +0.012 |
together | +0.010 |
Lily | +0.010 |
celebrate | +0.010 |
Neg logit contributions | |
beast | -0.017 |
gut | -0.017 |
severe | -0.016 |
inct | -0.016 |
rum | -0.015 |
Token | hello |
Token ID | 23748 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
parties | +0.106 |
tea | +0.094 |
celebrate | +0.076 |
recess | +0.072 |
party | +0.072 |
Neg logit contributions | |
beast | -0.129 |
gut | -0.129 |
swollen | -0.124 |
mistake | -0.124 |
inct | -0.124 |
Token | . |
Token ID | 13 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
beast | +0.050 |
gut | +0.049 |
swollen | +0.049 |
mistake | +0.048 |
therm | +0.047 |
Neg logit contributions | |
parties | -0.035 |
tea | -0.030 |
celebrate | -0.023 |
party | -0.023 |
recess | -0.022 |
Token |
|
Token ID | 198 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
mistake | +0.007 |
gut | +0.007 |
beast | +0.007 |
swollen | +0.007 |
severe | +0.007 |
Neg logit contributions | |
parties | -0.004 |
tea | -0.003 |
together | -0.002 |
celebrate | -0.002 |
recess | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | +5.41e-01 |
Pos logit contributions | |
parties | +0.007 |
tea | +0.006 |
recess | +0.005 |
celebrate | +0.005 |
picnic | +0.004 |
Neg logit contributions | |
beast | -0.009 |
gut | -0.009 |
mistake | -0.009 |
therm | -0.009 |
inct | -0.009 |
Token | The |
Token ID | 464 |
Feature activation | +4.38e+00 |
Pos logit contributions | |
beast | +0.019 |
gut | +0.019 |
mistake | +0.018 |
therm | +0.018 |
inct | +0.017 |
Neg logit contributions | |
parties | -0.016 |
tea | -0.014 |
celebrate | -0.013 |
recess | -0.012 |
picnic | -0.012 |
Token | stranger |
Token ID | 16195 |
Feature activation | +5.86e+00 |
Pos logit contributions | |
gut | +0.129 |
rum | +0.126 |
inct | +0.122 |
mistake | +0.118 |
beast | +0.117 |
Neg logit contributions | |
parties | -0.160 |
tea | -0.149 |
party | -0.134 |
picnic | -0.134 |
recess | -0.126 |
Token | smiled |
Token ID | 13541 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
gut | +0.223 |
beast | +0.215 |
swollen | +0.208 |
mistake | +0.207 |
severe | +0.207 |
Neg logit contributions | |
parties | -0.165 |
tea | -0.148 |
played | -0.124 |
celebrate | -0.123 |
party | -0.122 |
Token | at |
Token ID | 379 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
beast | +0.065 |
gut | +0.064 |
mistake | +0.062 |
swollen | +0.061 |
therm | +0.061 |
Neg logit contributions | |
parties | -0.046 |
tea | -0.041 |
recess | -0.031 |
celebrate | -0.031 |
party | -0.030 |
Token | the |
Token ID | 262 |
Feature activation | +7.18e-01 |
Pos logit contributions | |
beast | +0.069 |
gut | +0.068 |
mistake | +0.064 |
therm | +0.063 |
inct | +0.063 |
Neg logit contributions | |
parties | -0.060 |
tea | -0.053 |
recess | -0.044 |
celebrate | -0.042 |
party | -0.041 |
Token | bird |
Token ID | 6512 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
gut | +0.019 |
inct | +0.018 |
rum | +0.018 |
beast | +0.017 |
swollen | +0.017 |
Neg logit contributions | |
parties | -0.029 |
tea | -0.026 |
party | -0.024 |
picnic | -0.023 |
celebrate | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | -9.62e-01 |
Pos logit contributions | |
gut | +0.098 |
beast | +0.098 |
inct | +0.093 |
swollen | +0.092 |
mistake | +0.091 |
Neg logit contributions | |
parties | -0.062 |
tea | -0.056 |
celebrate | -0.043 |
party | -0.041 |
wear | -0.040 |
Token | is |
Token ID | 318 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
beast | +0.003 |
gut | +0.003 |
mistake | +0.003 |
therm | +0.003 |
inct | +0.003 |
Neg logit contributions | |
parties | -0.003 |
tea | -0.003 |
recess | -0.002 |
celebrate | -0.002 |
party | -0.002 |
Token | that |
Token ID | 326 |
Feature activation | -2.95e+00 |
Pos logit contributions | |
gut | +0.046 |
beast | +0.046 |
mistake | +0.043 |
inct | +0.043 |
swollen | +0.042 |
Neg logit contributions | |
parties | -0.039 |
tea | -0.035 |
recess | -0.029 |
celebrate | -0.028 |
party | -0.027 |
Token | ?" |
Token ID | 1701 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
tea | +0.064 |
parties | +0.063 |
picnic | +0.053 |
party | +0.053 |
celebrate | +0.051 |
Neg logit contributions | |
gut | -0.125 |
swollen | -0.120 |
rum | -0.117 |
passer | -0.116 |
vain | -0.114 |
Token |
|
Token ID | 198 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
beast | +0.058 |
gut | +0.057 |
mistake | +0.053 |
inct | +0.053 |
swollen | +0.053 |
Neg logit contributions | |
parties | -0.037 |
tea | -0.032 |
recess | -0.025 |
celebrate | -0.025 |
party | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | +5.89e-01 |
Pos logit contributions | |
parties | +0.000 |
tea | +0.000 |
recess | +0.000 |
celebrate | +0.000 |
party | +0.000 |
Neg logit contributions | |
beast | -0.001 |
gut | -0.001 |
mistake | -0.000 |
inct | -0.000 |
therm | -0.000 |
Token | " |
Token ID | 1 |
Feature activation | +2.28e+00 |
Pos logit contributions | |
beast | +0.023 |
gut | +0.022 |
mistake | +0.021 |
swollen | +0.021 |
inct | +0.021 |
Neg logit contributions | |
parties | -0.017 |
tea | -0.015 |
celebrate | -0.012 |
recess | -0.012 |
party | -0.012 |
Token | That |
Token ID | 2504 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
gut | +0.089 |
beast | +0.085 |
mistake | +0.085 |
rum | +0.080 |
iously | +0.078 |
Neg logit contributions | |
parties | -0.062 |
tea | -0.057 |
celebrate | -0.048 |
recess | -0.045 |
picnic | -0.044 |
Token | is |
Token ID | 318 |
Feature activation | +2.71e+00 |
Pos logit contributions | |
gut | +0.095 |
rum | +0.092 |
swollen | +0.090 |
passer | +0.089 |
reckless | +0.088 |
Neg logit contributions | |
tea | -0.037 |
parties | -0.034 |
picnic | -0.029 |
party | -0.024 |
celebrate | -0.023 |
Token | a |
Token ID | 257 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
gut | +0.128 |
rum | +0.123 |
beast | +0.122 |
passer | +0.118 |
cut | +0.118 |
Neg logit contributions | |
tea | -0.057 |
parties | -0.054 |
picnic | -0.035 |
celebrate | -0.034 |
party | -0.033 |
Token | bird |
Token ID | 6512 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
tea | +0.042 |
parties | +0.041 |
picnic | +0.036 |
party | +0.036 |
birthday | +0.034 |
Neg logit contributions | |
gut | -0.041 |
rum | -0.041 |
iously | -0.037 |
inct | -0.037 |
beast | -0.037 |
Token | 's |
Token ID | 338 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
gut | +0.061 |
beast | +0.059 |
inct | +0.057 |
mistake | +0.056 |
passer | +0.056 |
Neg logit contributions | |
parties | -0.030 |
tea | -0.027 |
celebrate | -0.019 |
party | -0.018 |
wear | -0.018 |
Token | The |
Token ID | 464 |
Feature activation | +3.73e+00 |
Pos logit contributions | |
parties | +0.006 |
tea | +0.005 |
recess | +0.004 |
celebrate | +0.004 |
party | +0.004 |
Neg logit contributions | |
beast | -0.009 |
gut | -0.009 |
mistake | -0.009 |
inct | -0.009 |
therm | -0.009 |
Token | referee |
Token ID | 22957 |
Feature activation | +4.01e+00 |
Pos logit contributions | |
gut | +0.126 |
beast | +0.124 |
inct | +0.119 |
swollen | +0.116 |
therm | +0.116 |
Neg logit contributions | |
parties | -0.124 |
tea | -0.112 |
recess | -0.097 |
party | -0.091 |
celebrate | -0.090 |
Token | is |
Token ID | 318 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
gut | +0.131 |
beast | +0.130 |
therm | +0.120 |
swollen | +0.120 |
mistake | +0.120 |
Neg logit contributions | |
parties | -0.137 |
tea | -0.122 |
recess | -0.104 |
celebrate | -0.101 |
party | -0.099 |
Token | calm |
Token ID | 9480 |
Feature activation | +5.41e-01 |
Pos logit contributions | |
beast | +0.066 |
gut | +0.065 |
inct | +0.064 |
passer | +0.060 |
therm | +0.059 |
Neg logit contributions | |
parties | -0.066 |
tea | -0.062 |
celebrate | -0.052 |
recess | -0.051 |
together | -0.050 |
Token | and |
Token ID | 290 |
Feature activation | +7.68e-01 |
Pos logit contributions | |
beast | +0.020 |
gut | +0.020 |
inct | +0.019 |
swollen | +0.019 |
therm | +0.019 |
Neg logit contributions | |
parties | -0.016 |
tea | -0.014 |
recess | -0.011 |
party | -0.011 |
celebrate | -0.011 |
Token | kind |
Token ID | 1611 |
Feature activation | +2.71e+00 |
Pos logit contributions | |
beast | +0.027 |
inct | +0.026 |
gut | +0.026 |
rum | +0.026 |
Owl | +0.025 |
Neg logit contributions | |
parties | -0.023 |
tea | -0.021 |
play | -0.019 |
played | -0.018 |
wear | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
gut | +0.117 |
beast | +0.117 |
inct | +0.111 |
swollen | +0.111 |
therm | +0.110 |
Neg logit contributions | |
parties | -0.064 |
tea | -0.054 |
recess | -0.042 |
party | -0.038 |
celebrate | -0.038 |
Token | He |
Token ID | 679 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
beast | +0.000 |
cut | +0.000 |
gut | +0.000 |
swollen | +0.000 |
therm | +0.000 |
Neg logit contributions | |
parties | -0.000 |
tea | -0.000 |
recess | -0.000 |
celebrate | -0.000 |
party | -0.000 |
Token | smiles |
Token ID | 21845 |
Feature activation | +3.32e-01 |
Pos logit contributions | |
gut | +0.042 |
beast | +0.042 |
inct | +0.039 |
swollen | +0.039 |
severe | +0.039 |
Neg logit contributions | |
parties | -0.040 |
tea | -0.036 |
party | -0.028 |
recess | -0.028 |
picnic | -0.028 |
Token | at |
Token ID | 379 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
beast | +0.013 |
gut | +0.013 |
cut | +0.013 |
inct | +0.013 |
mistake | +0.013 |
Neg logit contributions | |
parties | -0.009 |
tea | -0.007 |
recess | -0.006 |
together | -0.006 |
celebrate | -0.006 |
Token | Ben |
Token ID | 3932 |
Feature activation | -3.17e+00 |
Pos logit contributions | |
beast | +0.073 |
gut | +0.070 |
mistake | +0.069 |
adult | +0.066 |
therm | +0.065 |
Neg logit contributions | |
parties | -0.041 |
tea | -0.034 |
celebrate | -0.027 |
picnic | -0.026 |
party | -0.025 |
Token | When |
Token ID | 2215 |
Feature activation | +3.30e+00 |
Pos logit contributions | |
beast | +0.013 |
swollen | +0.013 |
gut | +0.013 |
thorn | +0.013 |
mistake | +0.013 |
Neg logit contributions | |
parties | -0.007 |
celebrate | -0.006 |
tea | -0.006 |
together | -0.006 |
recess | -0.005 |
Token | they |
Token ID | 484 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
gut | +0.139 |
beast | +0.135 |
swollen | +0.134 |
mistake | +0.130 |
cut | +0.129 |
Neg logit contributions | |
parties | -0.084 |
tea | -0.072 |
recess | -0.060 |
celebrate | -0.059 |
picnic | -0.055 |
Token | were |
Token ID | 547 |
Feature activation | +9.34e-01 |
Pos logit contributions | |
parties | +0.043 |
tea | +0.037 |
celebrate | +0.029 |
recess | +0.028 |
party | +0.026 |
Neg logit contributions | |
gut | -0.083 |
beast | -0.082 |
mistake | -0.078 |
inct | -0.077 |
swollen | -0.077 |
Token | done |
Token ID | 1760 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
beast | +0.035 |
gut | +0.035 |
mistake | +0.033 |
inct | +0.033 |
therm | +0.033 |
Neg logit contributions | |
parties | -0.028 |
tea | -0.024 |
recess | -0.020 |
celebrate | -0.019 |
party | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | +3.23e-02 |
Pos logit contributions | |
beast | +0.050 |
gut | +0.050 |
mistake | +0.047 |
inct | +0.047 |
therm | +0.047 |
Neg logit contributions | |
parties | -0.038 |
tea | -0.033 |
recess | -0.027 |
celebrate | -0.026 |
party | -0.025 |
Token | John |
Token ID | 1757 |
Feature activation | +9.90e-01 |
Pos logit contributions | |
gut | +0.001 |
beast | +0.001 |
swollen | +0.001 |
thorn | +0.001 |
mistake | +0.001 |
Neg logit contributions | |
parties | -0.001 |
tea | -0.001 |
celebrate | -0.001 |
together | -0.001 |
recess | -0.000 |
Token | said |
Token ID | 531 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
gut | +0.040 |
beast | +0.039 |
inct | +0.036 |
swollen | +0.036 |
mistake | +0.036 |
Neg logit contributions | |
parties | -0.025 |
tea | -0.023 |
played | -0.020 |
celebrate | -0.019 |
party | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | +4.46e-01 |
Pos logit contributions | |
parties | +0.045 |
tea | +0.039 |
recess | +0.031 |
celebrate | +0.030 |
party | +0.029 |
Neg logit contributions | |
beast | -0.064 |
gut | -0.064 |
mistake | -0.060 |
therm | -0.060 |
inct | -0.060 |
Token | " |
Token ID | 366 |
Feature activation | +3.62e+00 |
Pos logit contributions | |
beast | +0.017 |
gut | +0.017 |
swollen | +0.016 |
mistake | +0.016 |
distress | +0.016 |
Neg logit contributions | |
parties | -0.011 |
tea | -0.010 |
together | -0.010 |
they | -0.010 |
celebrate | -0.010 |
Token | Look |
Token ID | 8567 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
beast | +0.121 |
gut | +0.120 |
mistake | +0.112 |
therm | +0.111 |
inct | +0.111 |
Neg logit contributions | |
parties | -0.121 |
tea | -0.108 |
recess | -0.092 |
celebrate | -0.088 |
party | -0.085 |
Token | Daddy |
Token ID | 26926 |
Feature activation | -6.49e-01 |
Pos logit contributions | |
parties | +0.065 |
tea | +0.057 |
recess | +0.054 |
ets | +0.047 |
dates | +0.046 |
Neg logit contributions | |
beast | -0.059 |
gut | -0.058 |
mistake | -0.055 |
inct | -0.052 |
therm | -0.051 |
Token | it |
Token ID | 340 |
Feature activation | +6.61e-01 |
Pos logit contributions | |
parties | +0.044 |
tea | +0.038 |
recess | +0.031 |
celebrate | +0.030 |
party | +0.029 |
Neg logit contributions | |
beast | -0.058 |
gut | -0.058 |
mistake | -0.054 |
therm | -0.054 |
inct | -0.054 |
Token | . |
Token ID | 13 |
Feature activation | -6.84e-01 |
Pos logit contributions | |
beast | +0.023 |
gut | +0.023 |
mistake | +0.022 |
inct | +0.022 |
therm | +0.022 |
Neg logit contributions | |
parties | -0.020 |
tea | -0.017 |
recess | -0.015 |
celebrate | -0.014 |
picnic | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | -7.38e-01 |
Pos logit contributions | |
parties | +0.019 |
tea | +0.016 |
recess | +0.013 |
celebrate | +0.012 |
picnic | +0.012 |
Neg logit contributions | |
beast | -0.027 |
gut | -0.027 |
mistake | -0.026 |
swollen | -0.025 |
therm | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
parties | +0.029 |
tea | +0.026 |
recess | +0.023 |
party | +0.022 |
celebrate | +0.022 |
Neg logit contributions | |
beast | -0.020 |
gut | -0.020 |
mistake | -0.019 |
inct | -0.018 |
therm | -0.018 |
Token | Sam |
Token ID | 16305 |
Feature activation | +5.47e-02 |
Pos logit contributions | |
beast | +0.006 |
gut | +0.006 |
mistake | +0.005 |
inct | +0.005 |
therm | +0.005 |
Neg logit contributions | |
parties | -0.006 |
tea | -0.006 |
recess | -0.005 |
celebrate | -0.005 |
party | -0.005 |
Token | was |
Token ID | 373 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
gut | +0.002 |
beast | +0.002 |
mistake | +0.002 |
inct | +0.002 |
therm | +0.002 |
Neg logit contributions | |
parties | -0.002 |
tea | -0.001 |
recess | -0.001 |
celebrate | -0.001 |
party | -0.001 |
Token | curious |
Token ID | 11040 |
Feature activation | +3.45e+00 |
Pos logit contributions | |
beast | +0.061 |
gut | +0.060 |
inct | +0.059 |
cut | +0.057 |
therm | +0.056 |
Neg logit contributions | |
parties | -0.052 |
tea | -0.047 |
celebrate | -0.038 |
played | -0.037 |
party | -0.036 |
Token | but |
Token ID | 475 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
beast | +0.124 |
gut | +0.123 |
mistake | +0.116 |
therm | +0.114 |
inct | +0.112 |
Neg logit contributions | |
parties | -0.106 |
tea | -0.093 |
recess | -0.080 |
celebrate | -0.076 |
picnic | -0.072 |
Token | she |
Token ID | 673 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
inct | +0.086 |
gut | +0.085 |
beast | +0.084 |
therm | +0.081 |
mistake | +0.081 |
Neg logit contributions | |
parties | -0.049 |
tea | -0.041 |
played | -0.036 |
wear | -0.035 |
recess | -0.033 |
Token | was |
Token ID | 373 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
parties | +0.030 |
played | +0.026 |
tea | +0.025 |
drew | +0.022 |
wear | +0.022 |
Neg logit contributions | |
gut | -0.041 |
rum | -0.040 |
beast | -0.039 |
swollen | -0.039 |
therm | -0.039 |
Token | also |
Token ID | 635 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
inct | +0.052 |
beast | +0.050 |
cut | +0.049 |
gut | +0.049 |
therm | +0.047 |
Neg logit contributions | |
parties | -0.046 |
tea | -0.042 |
played | -0.035 |
wear | -0.034 |
together | -0.034 |
Token | gloomy |
Token ID | 46400 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
gut | +0.110 |
beast | +0.109 |
inct | +0.105 |
passer | +0.104 |
mistake | +0.103 |
Neg logit contributions | |
tea | -0.062 |
parties | -0.062 |
played | -0.045 |
recess | -0.043 |
play | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
beast | +0.103 |
gut | +0.103 |
mistake | +0.097 |
inct | +0.096 |
therm | +0.096 |
Neg logit contributions | |
parties | -0.065 |
tea | -0.056 |
recess | -0.044 |
celebrate | -0.042 |
party | -0.040 |
Token | She |
Token ID | 1375 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
beast | +0.017 |
gut | +0.017 |
mistake | +0.016 |
cut | +0.016 |
severe | +0.016 |
Neg logit contributions | |
parties | -0.010 |
tea | -0.009 |
recess | -0.007 |
wear | -0.007 |
celebrate | -0.007 |
Token | was |
Token ID | 373 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
beast | +0.051 |
gut | +0.050 |
inct | +0.048 |
rum | +0.048 |
cut | +0.048 |
Neg logit contributions | |
parties | -0.031 |
tea | -0.029 |
played | -0.028 |
wear | -0.025 |
danced | -0.024 |
Token | getting |
Token ID | 1972 |
Feature activation | +2.69e+00 |
Pos logit contributions | |
beast | +0.058 |
inct | +0.055 |
gut | +0.054 |
passer | +0.053 |
rum | +0.052 |
Neg logit contributions | |
parties | -0.045 |
tea | -0.045 |
played | -0.035 |
together | -0.035 |
outside | -0.035 |
Token | worried |
Token ID | 7960 |
Feature activation | +4.26e+00 |
Pos logit contributions | |
beast | +0.103 |
gut | +0.096 |
rum | +0.095 |
mistake | +0.094 |
inct | +0.090 |
Neg logit contributions | |
parties | -0.079 |
tea | -0.078 |
together | -0.063 |
recess | -0.057 |
them | -0.057 |
Token | that |
Token ID | 326 |
Feature activation | +3.95e-01 |
Pos logit contributions | |
beast | +0.161 |
gut | +0.160 |
inct | +0.151 |
mistake | +0.150 |
therm | +0.149 |
Neg logit contributions | |
parties | -0.124 |
tea | -0.108 |
recess | -0.088 |
celebrate | -0.085 |
party | -0.082 |
Token | the |
Token ID | 262 |
Feature activation | +3.26e+00 |
Pos logit contributions | |
beast | +0.016 |
gut | +0.016 |
inct | +0.015 |
cut | +0.015 |
mistake | +0.015 |
Neg logit contributions | |
parties | -0.011 |
tea | -0.009 |
recess | -0.007 |
celebrate | -0.007 |
party | -0.007 |
Token | party |
Token ID | 2151 |
Feature activation | +5.13e-01 |
Pos logit contributions | |
rum | +0.122 |
gut | +0.115 |
beast | +0.108 |
passer | +0.107 |
severe | +0.107 |
Neg logit contributions | |
tea | -0.099 |
parties | -0.098 |
party | -0.080 |
picnic | -0.079 |
celebrate | -0.073 |
Token | might |
Token ID | 1244 |
Feature activation | +4.95e+00 |
Pos logit contributions | |
beast | +0.018 |
gut | +0.018 |
inct | +0.017 |
swollen | +0.017 |
therm | +0.017 |
Neg logit contributions | |
parties | -0.016 |
tea | -0.014 |
party | -0.012 |
celebrate | -0.012 |
recess | -0.012 |
Token | be |
Token ID | 307 |
Feature activation | +5.33e-01 |
Pos logit contributions | |
beast | +0.199 |
gut | +0.196 |
inct | +0.188 |
passer | +0.188 |
Eagle | +0.183 |
Neg logit contributions | |
parties | -0.134 |
tea | -0.123 |
celebrate | -0.102 |
wear | -0.101 |
play | -0.101 |
Token | some |
Token ID | 617 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
beast | +0.042 |
gut | +0.042 |
mistake | +0.039 |
inct | +0.039 |
rum | +0.038 |
Neg logit contributions | |
parties | -0.031 |
tea | -0.029 |
recess | -0.022 |
celebrate | -0.021 |
together | -0.021 |
Token | big |
Token ID | 1263 |
Feature activation | -5.94e-01 |
Pos logit contributions | |
parties | +0.054 |
tea | +0.051 |
recess | +0.041 |
celebrate | +0.040 |
picnic | +0.039 |
Neg logit contributions | |
beast | -0.051 |
gut | -0.051 |
mistake | -0.047 |
inct | -0.047 |
swollen | -0.046 |
Token | balloons |
Token ID | 35485 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
parties | +0.020 |
tea | +0.019 |
recess | +0.015 |
party | +0.015 |
picnic | +0.015 |
Neg logit contributions | |
gut | -0.020 |
beast | -0.019 |
inct | -0.018 |
mistake | -0.018 |
severe | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
parties | +0.047 |
tea | +0.041 |
recess | +0.035 |
celebrate | +0.033 |
party | +0.032 |
Neg logit contributions | |
beast | -0.054 |
gut | -0.053 |
mistake | -0.050 |
therm | -0.050 |
inct | -0.049 |
Token | some |
Token ID | 617 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
parties | +0.052 |
tea | +0.047 |
recess | +0.036 |
celebrate | +0.036 |
picnic | +0.035 |
Neg logit contributions | |
beast | -0.064 |
gut | -0.064 |
mistake | -0.060 |
inct | -0.060 |
therm | -0.057 |
Token | paint |
Token ID | 7521 |
Feature activation | -6.42e-02 |
Pos logit contributions | |
parties | +0.052 |
tea | +0.046 |
recess | +0.039 |
celebrate | +0.037 |
party | +0.036 |
Neg logit contributions | |
beast | -0.057 |
gut | -0.057 |
mistake | -0.053 |
inct | -0.052 |
therm | -0.052 |
Token | and |
Token ID | 290 |
Feature activation | -8.77e-01 |
Pos logit contributions | |
parties | +0.002 |
tea | +0.002 |
recess | +0.001 |
celebrate | +0.001 |
party | +0.001 |
Neg logit contributions | |
beast | -0.002 |
gut | -0.002 |
therm | -0.002 |
mistake | -0.002 |
inct | -0.002 |
Token | some |
Token ID | 617 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
parties | +0.025 |
tea | +0.022 |
recess | +0.017 |
celebrate | +0.017 |
party | +0.016 |
Neg logit contributions | |
gut | -0.034 |
beast | -0.034 |
mistake | -0.032 |
inct | -0.032 |
therm | -0.031 |
Token | fuzzy |
Token ID | 34669 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
parties | +0.069 |
tea | +0.062 |
recess | +0.051 |
celebrate | +0.050 |
party | +0.048 |
Neg logit contributions | |
beast | -0.077 |
gut | -0.077 |
mistake | -0.072 |
inct | -0.072 |
therm | -0.071 |
Token | things |
Token ID | 1243 |
Feature activation | -7.11e-01 |
Pos logit contributions | |
parties | +0.064 |
tea | +0.057 |
recess | +0.048 |
celebrate | +0.046 |
party | +0.045 |
Neg logit contributions | |
beast | -0.068 |
gut | -0.067 |
mistake | -0.063 |
inct | -0.062 |
therm | -0.062 |
Token | . |
Token ID | 13 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
parties | +0.020 |
tea | +0.018 |
recess | +0.015 |
celebrate | +0.014 |
party | +0.013 |
Neg logit contributions | |
beast | -0.026 |
gut | -0.026 |
therm | -0.025 |
mistake | -0.025 |
cut | -0.024 |
Token | � |
Token ID | 6184 |
Feature activation | +3.67e-01 |
Pos logit contributions | |
parties | +0.038 |
tea | +0.033 |
recess | +0.025 |
celebrate | +0.024 |
party | +0.023 |
Neg logit contributions | |
beast | -0.063 |
gut | -0.062 |
mistake | -0.060 |
inct | -0.059 |
swollen | -0.058 |
Token | � |
Token ID | 95 |
Feature activation | +5.92e-03 |
Pos logit contributions | |
gut | +0.011 |
beast | +0.011 |
inct | +0.010 |
therm | +0.010 |
mistake | +0.010 |
Neg logit contributions | |
parties | -0.013 |
tea | -0.012 |
recess | -0.010 |
celebrate | -0.010 |
picnic | -0.010 |
Token | € |
Token ID | 26391 |
Feature activation | -4.87e-01 |
Pos logit contributions | |
beast | +0.000 |
gut | +0.000 |
mistake | +0.000 |
inct | +0.000 |
therm | +0.000 |
Neg logit contributions | |
parties | -0.000 |
tea | -0.000 |
recess | -0.000 |
celebrate | -0.000 |
party | -0.000 |
Token | � |
Token ID | 129 |
Feature activation | +8.89e-01 |
Pos logit contributions | |
parties | +0.019 |
tea | +0.018 |
recess | +0.015 |
celebrate | +0.015 |
party | +0.014 |
Neg logit contributions | |
beast | -0.013 |
gut | -0.013 |
inct | -0.012 |
mistake | -0.012 |
therm | -0.012 |
Token | � |
Token ID | 241 |
Feature activation | +2.86e+00 |
Pos logit contributions | |
beast | +0.027 |
gut | +0.027 |
swollen | +0.026 |
therm | +0.025 |
mistake | +0.025 |
Neg logit contributions | |
parties | -0.031 |
tea | -0.028 |
recess | -0.024 |
celebrate | -0.024 |
wear | -0.023 |
Token | I |
Token ID | 40 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
beast | +0.091 |
gut | +0.090 |
mistake | +0.084 |
therm | +0.083 |
inct | +0.082 |
Neg logit contributions | |
parties | -0.100 |
tea | -0.089 |
recess | -0.076 |
celebrate | -0.074 |
party | -0.072 |
Token | will |
Token ID | 481 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
parties | +0.034 |
tea | +0.030 |
recess | +0.025 |
celebrate | +0.025 |
party | +0.024 |
Neg logit contributions | |
beast | -0.035 |
gut | -0.034 |
mistake | -0.032 |
swollen | -0.032 |
inct | -0.032 |
Token | help |
Token ID | 1037 |
Feature activation | +9.10e-01 |
Pos logit contributions | |
beast | +0.096 |
gut | +0.093 |
inct | +0.092 |
swollen | +0.091 |
mistake | +0.090 |
Neg logit contributions | |
parties | -0.064 |
tea | -0.053 |
wear | -0.049 |
celebrate | -0.048 |
play | -0.043 |
Token | you |
Token ID | 345 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
beast | +0.040 |
gut | +0.039 |
inct | +0.038 |
swollen | +0.037 |
mistake | +0.037 |
Neg logit contributions | |
parties | -0.021 |
tea | -0.017 |
recess | -0.013 |
celebrate | -0.013 |
wear | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -2.82e-01 |
Pos logit contributions | |
parties | +0.042 |
tea | +0.035 |
celebrate | +0.030 |
wear | +0.029 |
party | +0.027 |
Neg logit contributions | |
beast | -0.066 |
gut | -0.065 |
swollen | -0.064 |
passer | -0.063 |
inct | -0.062 |
Token | â |
Token ID | 22940 |
Feature activation | +6.09e-01 |
Pos logit contributions | |
parties | +0.009 |
tea | +0.008 |
recess | +0.007 |
celebrate | +0.006 |
party | +0.006 |
Neg logit contributions | |
beast | -0.010 |
gut | -0.010 |
mistake | -0.009 |
swollen | -0.009 |
inct | -0.009 |
Token | very |
Token ID | 845 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
beast | +0.014 |
gut | +0.014 |
inct | +0.013 |
mistake | +0.013 |
therm | +0.013 |
Neg logit contributions | |
parties | -0.012 |
tea | -0.011 |
recess | -0.009 |
celebrate | -0.009 |
party | -0.008 |
Token | careful |
Token ID | 8161 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
beast | +0.074 |
gut | +0.072 |
inct | +0.069 |
mistake | +0.068 |
therm | +0.067 |
Neg logit contributions | |
parties | -0.069 |
tea | -0.062 |
recess | -0.052 |
celebrate | -0.050 |
wear | -0.048 |
Token | with |
Token ID | 351 |
Feature activation | -6.57e-01 |
Pos logit contributions | |
beast | +0.084 |
gut | +0.081 |
inct | +0.081 |
swollen | +0.079 |
passer | +0.078 |
Neg logit contributions | |
parties | -0.075 |
tea | -0.065 |
celebrate | -0.052 |
wear | -0.052 |
together | -0.051 |
Token | his |
Token ID | 465 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
parties | +0.017 |
tea | +0.015 |
recess | +0.011 |
costumes | +0.010 |
them | +0.010 |
Neg logit contributions | |
inct | -0.028 |
beast | -0.028 |
gut | -0.027 |
passer | -0.026 |
plea | -0.026 |
Token | cars |
Token ID | 5006 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
parties | +0.035 |
tea | +0.032 |
recess | +0.026 |
party | +0.025 |
costumes | +0.025 |
Neg logit contributions | |
beast | -0.046 |
gut | -0.045 |
inct | -0.044 |
passer | -0.044 |
severe | -0.043 |
Token | and |
Token ID | 290 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
parties | +0.049 |
tea | +0.039 |
recess | +0.030 |
party | +0.029 |
celebrate | +0.029 |
Neg logit contributions | |
beast | -0.087 |
gut | -0.085 |
inct | -0.085 |
swollen | -0.082 |
passer | -0.082 |
Token | never |
Token ID | 1239 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
parties | +0.044 |
tea | +0.037 |
recess | +0.031 |
costumes | +0.030 |
played | +0.030 |
Neg logit contributions | |
inct | -0.054 |
beast | -0.053 |
gut | -0.052 |
passer | -0.050 |
therm | -0.049 |
Token | let |
Token ID | 1309 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
gut | +0.067 |
beast | +0.066 |
inct | +0.063 |
therm | +0.062 |
mistake | +0.061 |
Neg logit contributions | |
parties | -0.050 |
tea | -0.043 |
recess | -0.036 |
celebrate | -0.034 |
party | -0.034 |
Token | anyone |
Token ID | 2687 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
parties | +0.056 |
tea | +0.048 |
recess | +0.037 |
celebrate | +0.036 |
party | +0.034 |
Neg logit contributions | |
gut | -0.090 |
beast | -0.090 |
inct | -0.085 |
swollen | -0.085 |
mistake | -0.085 |
Token | else |
Token ID | 2073 |
Feature activation | -8.29e-01 |
Pos logit contributions | |
beast | +0.008 |
gut | +0.008 |
mistake | +0.008 |
therm | +0.008 |
swollen | +0.008 |
Neg logit contributions | |
parties | -0.007 |
tea | -0.006 |
recess | -0.005 |
celebrate | -0.005 |
party | -0.005 |
Token | touch |
Token ID | 3638 |
Feature activation | -6.10e-01 |
Pos logit contributions | |
parties | +0.023 |
tea | +0.020 |
recess | +0.016 |
celebrate | +0.015 |
party | +0.015 |
Neg logit contributions | |
beast | -0.032 |
gut | -0.032 |
mistake | -0.030 |
inct | -0.030 |
therm | -0.030 |
Token | friends |
Token ID | 2460 |
Feature activation | -3.93e+00 |
Pos logit contributions | |
parties | +0.101 |
tea | +0.086 |
recess | +0.085 |
celebrate | +0.080 |
wear | +0.074 |
Neg logit contributions | |
beast | -0.095 |
gut | -0.095 |
mistake | -0.093 |
therm | -0.090 |
swollen | -0.089 |
Token | , |
Token ID | 11 |
Feature activation | -2.84e+00 |
Pos logit contributions | |
parties | +0.134 |
tea | +0.113 |
recess | +0.106 |
picnic | +0.101 |
ets | +0.099 |
Neg logit contributions | |
gut | -0.125 |
mistake | -0.124 |
beast | -0.123 |
Owl | -0.120 |
adult | -0.120 |
Token | Max |
Token ID | 5436 |
Feature activation | -2.74e+00 |
Pos logit contributions | |
parties | +0.099 |
ils | +0.083 |
ets | +0.080 |
tea | +0.079 |
outfits | +0.075 |
Neg logit contributions | |
Owl | -0.084 |
beast | -0.082 |
Eagle | -0.080 |
Wise | -0.079 |
therm | -0.076 |
Token | and |
Token ID | 290 |
Feature activation | -8.34e-01 |
Pos logit contributions | |
parties | +0.108 |
tea | +0.094 |
recess | +0.090 |
celebrate | +0.082 |
picnic | +0.078 |
Neg logit contributions | |
beast | -0.074 |
gut | -0.073 |
mistake | -0.072 |
therm | -0.069 |
Owl | -0.068 |
Token | Sally |
Token ID | 25737 |
Feature activation | -2.81e+00 |
Pos logit contributions | |
parties | +0.027 |
tea | +0.023 |
ets | +0.023 |
irl | +0.022 |
ils | +0.022 |
Neg logit contributions | |
Owl | -0.025 |
passer | -0.024 |
beast | -0.024 |
Eagle | -0.024 |
Wise | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | -8.40e-01 |
Pos logit contributions | |
parties | +0.098 |
tea | +0.083 |
recess | +0.079 |
celebrate | +0.071 |
picnic | +0.070 |
Neg logit contributions | |
gut | -0.085 |
mistake | -0.084 |
cut | -0.084 |
beast | -0.084 |
therm | -0.083 |
Token | They |
Token ID | 1119 |
Feature activation | -2.81e+00 |
Pos logit contributions | |
parties | +0.025 |
tea | +0.023 |
seas | +0.021 |
ils | +0.020 |
celebrate | +0.019 |
Neg logit contributions | |
Owl | -0.025 |
Wise | -0.024 |
gut | -0.024 |
beast | -0.024 |
mistake | -0.024 |
Token | lived |
Token ID | 5615 |
Feature activation | -7.79e-01 |
Pos logit contributions | |
parties | +0.101 |
recess | +0.091 |
tea | +0.091 |
pic | +0.086 |
seas | +0.082 |
Neg logit contributions | |
cut | -0.063 |
injured | -0.060 |
panicked | -0.060 |
reckless | -0.060 |
claw | -0.060 |
Token | by |
Token ID | 416 |
Feature activation | +4.01e+00 |
Pos logit contributions | |
parties | +0.021 |
tea | +0.018 |
recess | +0.015 |
celebrate | +0.014 |
picnic | +0.014 |
Neg logit contributions | |
gut | -0.030 |
beast | -0.029 |
mistake | -0.028 |
inct | -0.028 |
therm | -0.028 |
Token | the |
Token ID | 262 |
Feature activation | -8.61e-01 |
Pos logit contributions | |
gut | +0.174 |
beast | +0.171 |
inct | +0.162 |
cut | +0.161 |
swollen | +0.160 |
Neg logit contributions | |
parties | -0.103 |
tea | -0.090 |
recess | -0.069 |
celebrate | -0.064 |
picnic | -0.062 |
Token | beach |
Token ID | 10481 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
parties | +0.028 |
tea | +0.025 |
party | +0.021 |
recess | +0.021 |
picnic | +0.020 |
Neg logit contributions | |
gut | -0.031 |
beast | -0.030 |
rum | -0.030 |
inct | -0.029 |
swollen | -0.029 |
Token | her |
Token ID | 607 |
Feature activation | +3.52e-01 |
Pos logit contributions | |
beast | +0.038 |
therm | +0.038 |
gut | +0.037 |
swollen | +0.037 |
cut | +0.036 |
Neg logit contributions | |
parties | -0.044 |
tea | -0.038 |
recess | -0.034 |
picnic | -0.033 |
celebrate | -0.032 |
Token | toys |
Token ID | 14958 |
Feature activation | -4.34e-01 |
Pos logit contributions | |
beast | +0.013 |
gut | +0.013 |
therm | +0.012 |
mistake | +0.012 |
swollen | +0.012 |
Neg logit contributions | |
parties | -0.011 |
tea | -0.009 |
recess | -0.008 |
celebrate | -0.008 |
party | -0.007 |
Token | off |
Token ID | 572 |
Feature activation | +5.86e-01 |
Pos logit contributions | |
parties | +0.012 |
tea | +0.010 |
recess | +0.008 |
celebrate | +0.008 |
party | +0.008 |
Neg logit contributions | |
beast | -0.017 |
gut | -0.017 |
mistake | -0.016 |
inct | -0.016 |
therm | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | -6.12e-01 |
Pos logit contributions | |
beast | +0.021 |
gut | +0.021 |
mistake | +0.020 |
therm | +0.020 |
swollen | +0.019 |
Neg logit contributions | |
parties | -0.017 |
tea | -0.015 |
recess | -0.013 |
celebrate | -0.012 |
picnic | -0.012 |
Token | shelf |
Token ID | 18316 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
parties | +0.026 |
tea | +0.023 |
recess | +0.021 |
celebrate | +0.020 |
party | +0.020 |
Neg logit contributions | |
beast | -0.015 |
gut | -0.015 |
inct | -0.014 |
mistake | -0.014 |
swollen | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
beast | +0.109 |
gut | +0.109 |
inct | +0.102 |
mistake | +0.102 |
swollen | +0.101 |
Neg logit contributions | |
parties | -0.064 |
tea | -0.054 |
recess | -0.043 |
celebrate | -0.040 |
party | -0.039 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
parties | +0.005 |
tea | +0.005 |
recess | +0.004 |
celebrate | +0.004 |
party | +0.003 |
Neg logit contributions | |
beast | -0.008 |
gut | -0.008 |
mistake | -0.008 |
therm | -0.008 |
inct | -0.008 |
Token | was |
Token ID | 373 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
parties | +0.039 |
played | +0.036 |
tea | +0.035 |
drew | +0.028 |
danced | +0.028 |
Neg logit contributions | |
beast | -0.068 |
gut | -0.068 |
guardian | -0.065 |
inct | -0.064 |
cut | -0.062 |
Token | very |
Token ID | 845 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
beast | +0.053 |
inct | +0.051 |
gut | +0.049 |
cut | +0.048 |
rum | +0.045 |
Neg logit contributions | |
parties | -0.043 |
tea | -0.041 |
played | -0.034 |
together | -0.032 |
celebrate | -0.032 |
Token | sad |
Token ID | 6507 |
Feature activation | +3.78e+00 |
Pos logit contributions | |
beast | +0.086 |
inct | +0.078 |
gut | +0.078 |
cut | +0.073 |
Eagle | +0.073 |
Neg logit contributions | |
parties | -0.095 |
tea | -0.088 |
wear | -0.074 |
celebrate | -0.073 |
recess | -0.072 |
Token | and |
Token ID | 290 |
Feature activation | +1.75e+00 |
Pos logit contributions | |
beast | +0.146 |
gut | +0.144 |
inct | +0.138 |
cut | +0.135 |
swollen | +0.135 |
Neg logit contributions | |
parties | -0.110 |
tea | -0.095 |
recess | -0.075 |
party | -0.073 |
celebrate | -0.072 |
Token | . |
Token ID | 13 |
Feature activation | -8.12e-01 |
Pos logit contributions | |
beast | +0.060 |
gut | +0.057 |
cut | +0.057 |
therm | +0.056 |
Owl | +0.055 |
Neg logit contributions | |
parties | -0.048 |
tea | -0.039 |
recess | -0.036 |
celebrate | -0.032 |
picnic | -0.032 |
Token | The |
Token ID | 383 |
Feature activation | +2.50e+00 |
Pos logit contributions | |
parties | +0.025 |
tea | +0.022 |
recess | +0.018 |
celebrate | +0.018 |
picnic | +0.017 |
Neg logit contributions | |
gut | -0.029 |
beast | -0.029 |
mistake | -0.028 |
inct | -0.027 |
swollen | -0.027 |
Token | pigeon |
Token ID | 46387 |
Feature activation | +4.07e+00 |
Pos logit contributions | |
beast | +0.061 |
mistake | +0.057 |
gut | +0.056 |
injured | +0.055 |
crow | +0.054 |
Neg logit contributions | |
parties | -0.102 |
tea | -0.092 |
recess | -0.083 |
ets | -0.080 |
ils | -0.080 |
Token | had |
Token ID | 550 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
beast | +0.133 |
gut | +0.127 |
cut | +0.125 |
mistake | +0.121 |
therm | +0.119 |
Neg logit contributions | |
parties | -0.136 |
tea | -0.118 |
recess | -0.112 |
celebrate | -0.097 |
picnic | -0.095 |
Token | many |
Token ID | 867 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
parties | +0.064 |
tea | +0.056 |
recess | +0.048 |
celebrate | +0.044 |
party | +0.043 |
Neg logit contributions | |
beast | -0.081 |
gut | -0.081 |
mistake | -0.076 |
therm | -0.075 |
swollen | -0.075 |
Token | friends |
Token ID | 2460 |
Feature activation | -4.32e+00 |
Pos logit contributions | |
parties | +0.029 |
tea | +0.026 |
recess | +0.022 |
celebrate | +0.021 |
party | +0.020 |
Neg logit contributions | |
beast | -0.037 |
gut | -0.037 |
mistake | -0.035 |
therm | -0.035 |
swollen | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
parties | +0.153 |
tea | +0.139 |
recess | +0.130 |
picnic | +0.119 |
sandwiches | +0.113 |
Neg logit contributions | |
gut | -0.118 |
beast | -0.118 |
cut | -0.113 |
reckless | -0.111 |
guardian | -0.111 |
Token | They |
Token ID | 1119 |
Feature activation | -2.79e+00 |
Pos logit contributions | |
parties | +0.033 |
tea | +0.030 |
recess | +0.025 |
ils | +0.024 |
party | +0.023 |
Neg logit contributions | |
beast | -0.040 |
gut | -0.038 |
Owl | -0.037 |
inct | -0.037 |
therm | -0.036 |
Token | played |
Token ID | 2826 |
Feature activation | -5.35e+00 |
Pos logit contributions | |
parties | +0.081 |
tea | +0.076 |
recess | +0.068 |
outfits | +0.060 |
dates | +0.060 |
Neg logit contributions | |
cut | -0.087 |
beast | -0.086 |
gut | -0.086 |
Owl | -0.084 |
Eagle | -0.082 |
Token | together |
Token ID | 1978 |
Feature activation | -3.84e+00 |
Pos logit contributions | |
parties | +0.147 |
tea | +0.125 |
ouses | +0.125 |
recess | +0.124 |
ackers | +0.122 |
Neg logit contributions | |
beast | -0.160 |
warning | -0.156 |
gut | -0.155 |
mistake | -0.153 |
inct | -0.147 |
Token | every |
Token ID | 790 |
Feature activation | -3.55e+00 |
Pos logit contributions | |
parties | +0.107 |
tea | +0.093 |
recess | +0.083 |
celebrate | +0.076 |
picnic | +0.073 |
Neg logit contributions | |
beast | -0.137 |
gut | -0.134 |
mistake | -0.132 |
cut | -0.128 |
inct | -0.123 |
Token |
|
Token ID | 198 |
Feature activation | -2.60e-02 |
Pos logit contributions | |
parties | +0.032 |
tea | +0.028 |
recess | +0.024 |
celebrate | +0.023 |
party | +0.023 |
Neg logit contributions | |
beast | -0.031 |
gut | -0.030 |
mistake | -0.029 |
inct | -0.028 |
therm | -0.028 |
Token | Tim |
Token ID | 14967 |
Feature activation | -7.64e-02 |
Pos logit contributions | |
parties | +0.001 |
tea | +0.001 |
celebrate | +0.001 |
recess | +0.001 |
picnic | +0.001 |
Neg logit contributions | |
beast | -0.001 |
gut | -0.001 |
mistake | -0.001 |
therm | -0.001 |
inct | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
parties | +0.002 |
tea | +0.002 |
played | +0.002 |
celebrate | +0.002 |
party | +0.002 |
Neg logit contributions | |
gut | -0.003 |
beast | -0.003 |
inct | -0.003 |
therm | -0.003 |
swollen | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
gut | +0.046 |
inct | +0.046 |
beast | +0.045 |
plea | +0.044 |
umbling | +0.043 |
Neg logit contributions | |
parties | -0.019 |
tea | -0.017 |
played | -0.014 |
play | -0.014 |
recess | -0.013 |
Token | squirrel |
Token ID | 33039 |
Feature activation | +2.27e+00 |
Pos logit contributions | |
rum | +0.075 |
gut | +0.073 |
inct | +0.073 |
iously | +0.069 |
outage | +0.068 |
Neg logit contributions | |
parties | -0.052 |
tea | -0.050 |
picnic | -0.046 |
party | -0.043 |
play | -0.043 |
Token | played |
Token ID | 2826 |
Feature activation | -5.18e+00 |
Pos logit contributions | |
beast | +0.084 |
gut | +0.082 |
cut | +0.082 |
therm | +0.079 |
injured | +0.078 |
Neg logit contributions | |
parties | -0.063 |
tea | -0.053 |
recess | -0.047 |
celebrate | -0.041 |
costumes | -0.040 |
Token | with |
Token ID | 351 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
parties | +0.140 |
tea | +0.118 |
recess | +0.109 |
outfits | +0.097 |
notebooks | +0.096 |
Neg logit contributions | |
beast | -0.186 |
gut | -0.186 |
mistake | -0.182 |
cut | -0.173 |
warning | -0.171 |
Token | the |
Token ID | 262 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
parties | +0.071 |
tea | +0.064 |
recess | +0.050 |
celebrate | +0.049 |
picnic | +0.047 |
Neg logit contributions | |
beast | -0.096 |
gut | -0.096 |
inct | -0.092 |
mistake | -0.091 |
therm | -0.090 |
Token | cone |
Token ID | 27763 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
parties | +0.046 |
tea | +0.043 |
picnic | +0.042 |
sandwiches | +0.038 |
party | +0.036 |
Neg logit contributions | |
gut | -0.052 |
inct | -0.050 |
beast | -0.046 |
guardian | -0.045 |
rum | -0.045 |
Token | , |
Token ID | 11 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
beast | +0.065 |
gut | +0.065 |
mistake | +0.062 |
cut | +0.062 |
therm | +0.061 |
Neg logit contributions | |
parties | -0.034 |
tea | -0.028 |
recess | -0.023 |
celebrate | -0.021 |
picnic | -0.019 |
Token | rolling |
Token ID | 10708 |
Feature activation | -2.68e+00 |
Pos logit contributions | |
parties | +0.068 |
tea | +0.061 |
celebrate | +0.048 |
recess | +0.048 |
party | +0.048 |
Neg logit contributions | |
gut | -0.080 |
beast | -0.080 |
inct | -0.077 |
therm | -0.074 |
severe | -0.073 |
Token | to |
Token ID | 284 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
parties | +0.013 |
tea | +0.012 |
recess | +0.011 |
celebrate | +0.010 |
ets | +0.010 |
Neg logit contributions | |
beast | -0.009 |
gut | -0.009 |
mistake | -0.008 |
therm | -0.008 |
adult | -0.008 |
Token | run |
Token ID | 1057 |
Feature activation | -3.28e+00 |
Pos logit contributions | |
gut | +0.050 |
mistake | +0.050 |
adult | +0.047 |
beast | +0.047 |
warning | +0.046 |
Neg logit contributions | |
parties | -0.041 |
tea | -0.038 |
text | -0.034 |
ils | -0.032 |
ouses | -0.032 |
Token | and |
Token ID | 290 |
Feature activation | -2.69e+00 |
Pos logit contributions | |
parties | +0.111 |
tea | +0.105 |
celebrate | +0.093 |
birthday | +0.089 |
recess | +0.089 |
Neg logit contributions | |
beast | -0.084 |
gut | -0.080 |
mistake | -0.077 |
adult | -0.072 |
law | -0.071 |
Token | jump |
Token ID | 4391 |
Feature activation | -3.81e+00 |
Pos logit contributions | |
parties | +0.056 |
tea | +0.046 |
recess | +0.034 |
celebrate | +0.031 |
party | +0.029 |
Neg logit contributions | |
beast | -0.119 |
gut | -0.119 |
mistake | -0.114 |
therm | -0.113 |
swollen | -0.113 |
Token | and |
Token ID | 290 |
Feature activation | -2.31e+00 |
Pos logit contributions | |
parties | +0.125 |
tea | +0.110 |
recess | +0.106 |
celebrate | +0.103 |
wedding | +0.102 |
Neg logit contributions | |
beast | -0.098 |
cut | -0.098 |
inct | -0.094 |
gut | -0.093 |
therm | -0.092 |
Token | throw |
Token ID | 3714 |
Feature activation | -4.52e+00 |
Pos logit contributions | |
parties | +0.051 |
tea | +0.042 |
recess | +0.033 |
celebrate | +0.029 |
party | +0.028 |
Neg logit contributions | |
beast | -0.101 |
gut | -0.100 |
mistake | -0.096 |
therm | -0.094 |
swollen | -0.094 |
Token | balls |
Token ID | 11333 |
Feature activation | -4.77e+00 |
Pos logit contributions | |
parties | +0.145 |
tea | +0.122 |
celebrate | +0.115 |
recess | +0.114 |
wedding | +0.111 |
Neg logit contributions | |
beast | -0.132 |
mistake | -0.124 |
therm | -0.121 |
gut | -0.119 |
ang | -0.118 |
Token | . |
Token ID | 13 |
Feature activation | -9.47e-01 |
Pos logit contributions | |
parties | +0.158 |
tea | +0.143 |
celebrate | +0.128 |
pic | +0.125 |
picnic | +0.124 |
Neg logit contributions | |
cut | -0.134 |
gut | -0.126 |
beast | -0.126 |
mistake | -0.120 |
claw | -0.120 |
Token | He |
Token ID | 679 |
Feature activation | +6.26e-01 |
Pos logit contributions | |
parties | +0.031 |
tea | +0.027 |
recess | +0.024 |
ils | +0.022 |
celebrate | +0.022 |
Neg logit contributions | |
beast | -0.031 |
gut | -0.031 |
therm | -0.030 |
mistake | -0.029 |
inct | -0.029 |
Token | was |
Token ID | 373 |
Feature activation | +6.62e-01 |
Pos logit contributions | |
beast | +0.020 |
gut | +0.020 |
mistake | +0.019 |
cut | +0.018 |
therm | +0.018 |
Neg logit contributions | |
parties | -0.021 |
tea | -0.018 |
recess | -0.016 |
celebrate | -0.015 |
birthday | -0.015 |
Token | very |
Token ID | 845 |
Feature activation | +2.18e+00 |
Pos logit contributions | |
gut | +0.020 |
beast | +0.020 |
mistake | +0.019 |
therm | +0.019 |
swollen | +0.019 |
Neg logit contributions | |
parties | -0.023 |
tea | -0.020 |
recess | -0.018 |
celebrate | -0.017 |
picnic | -0.017 |
Token | day |
Token ID | 1110 |
Feature activation | +6.13e-01 |
Pos logit contributions | |
beast | +0.020 |
passer | +0.019 |
mistake | +0.019 |
adult | +0.019 |
therm | +0.019 |
Neg logit contributions | |
parties | -0.018 |
tea | -0.015 |
skirts | -0.013 |
outfits | -0.013 |
ils | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
gut | +0.018 |
beast | +0.018 |
therm | +0.017 |
mistake | +0.017 |
passer | +0.017 |
Neg logit contributions | |
parties | -0.021 |
tea | -0.018 |
recess | -0.016 |
celebrate | -0.016 |
ets | -0.015 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
therm | +0.008 |
mistake | +0.007 |
gut | +0.007 |
swollen | +0.007 |
beast | +0.007 |
Neg logit contributions | |
parties | -0.006 |
ets | -0.005 |
ils | -0.005 |
recess | -0.005 |
ag | -0.004 |
Token | wanted |
Token ID | 2227 |
Feature activation | -5.20e-02 |
Pos logit contributions | |
parties | +0.058 |
tea | +0.053 |
recess | +0.049 |
celebrate | +0.044 |
dates | +0.041 |
Neg logit contributions | |
beast | -0.052 |
gut | -0.050 |
mistake | -0.050 |
therm | -0.048 |
injured | -0.047 |
Token | to |
Token ID | 284 |
Feature activation | +3.37e+00 |
Pos logit contributions | |
parties | +0.002 |
tea | +0.002 |
recess | +0.001 |
celebrate | +0.001 |
party | +0.001 |
Neg logit contributions | |
beast | -0.002 |
gut | -0.002 |
inct | -0.002 |
cut | -0.002 |
swollen | -0.002 |
Token | wear |
Token ID | 5806 |
Feature activation | -4.80e+00 |
Pos logit contributions | |
gut | +0.131 |
beast | +0.130 |
mistake | +0.128 |
therm | +0.124 |
adult | +0.123 |
Neg logit contributions | |
parties | -0.080 |
tea | -0.070 |
recess | -0.059 |
ets | -0.057 |
ashed | -0.056 |
Token | her |
Token ID | 607 |
Feature activation | -2.60e+00 |
Pos logit contributions | |
parties | +0.149 |
tea | +0.129 |
recess | +0.114 |
celebrate | +0.109 |
ashed | +0.107 |
Neg logit contributions | |
gut | -0.159 |
beast | -0.155 |
therm | -0.151 |
swollen | -0.148 |
mistake | -0.146 |
Token | favorite |
Token ID | 4004 |
Feature activation | -3.12e+00 |
Pos logit contributions | |
parties | +0.069 |
tea | +0.058 |
recess | +0.048 |
celebrate | +0.045 |
party | +0.042 |
Neg logit contributions | |
beast | -0.104 |
gut | -0.104 |
mistake | -0.098 |
therm | -0.098 |
inct | -0.097 |
Token | dress |
Token ID | 6576 |
Feature activation | -2.24e+00 |
Pos logit contributions | |
parties | +0.104 |
tea | +0.087 |
recess | +0.082 |
ets | +0.077 |
celebrate | +0.075 |
Neg logit contributions | |
beast | -0.099 |
therm | -0.098 |
gut | -0.096 |
mistake | -0.094 |
swollen | -0.091 |
Token | to |
Token ID | 284 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
parties | +0.068 |
tea | +0.061 |
recess | +0.054 |
celebrate | +0.050 |
picnic | +0.048 |
Neg logit contributions | |
gut | -0.077 |
beast | -0.077 |
mistake | -0.073 |
therm | -0.073 |
cut | -0.070 |
Token | school |
Token ID | 1524 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
mistake | +0.058 |
gut | +0.058 |
beast | +0.056 |
therm | +0.054 |
adult | +0.053 |
Neg logit contributions | |
parties | -0.042 |
tea | -0.037 |
ashed | -0.033 |
recess | -0.032 |
ets | -0.030 |
Token | will |
Token ID | 481 |
Feature activation | +8.18e-01 |
Pos logit contributions | |
parties | +0.135 |
tea | +0.116 |
recess | +0.099 |
celebrate | +0.083 |
party | +0.082 |
Neg logit contributions | |
beast | -0.197 |
gut | -0.194 |
mistake | -0.184 |
therm | -0.183 |
cut | -0.183 |
Token | come |
Token ID | 1282 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
gut | +0.028 |
beast | +0.028 |
mistake | +0.026 |
therm | +0.026 |
inct | +0.025 |
Neg logit contributions | |
parties | -0.026 |
tea | -0.023 |
recess | -0.020 |
celebrate | -0.018 |
party | -0.018 |
Token | back |
Token ID | 736 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
parties | +0.051 |
tea | +0.046 |
recess | +0.041 |
ils | +0.039 |
ets | +0.036 |
Neg logit contributions | |
gut | -0.044 |
beast | -0.041 |
guardian | -0.039 |
mistake | -0.039 |
Owl | -0.039 |
Token | tomorrow |
Token ID | 9439 |
Feature activation | -2.93e+00 |
Pos logit contributions | |
beast | +0.071 |
gut | +0.071 |
mistake | +0.067 |
therm | +0.066 |
inct | +0.066 |
Neg logit contributions | |
parties | -0.060 |
tea | -0.053 |
recess | -0.044 |
celebrate | -0.042 |
party | -0.040 |
Token | and |
Token ID | 290 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
parties | +0.086 |
tea | +0.076 |
recess | +0.065 |
celebrate | +0.060 |
party | +0.057 |
Neg logit contributions | |
beast | -0.105 |
gut | -0.105 |
mistake | -0.100 |
therm | -0.098 |
inct | -0.096 |
Token | play |
Token ID | 711 |
Feature activation | -6.01e+00 |
Pos logit contributions | |
parties | +0.028 |
tea | +0.023 |
celebrate | +0.018 |
recess | +0.016 |
picnic | +0.016 |
Neg logit contributions | |
beast | -0.067 |
gut | -0.066 |
inct | -0.064 |
mistake | -0.063 |
therm | -0.063 |
Token | with |
Token ID | 351 |
Feature activation | -3.36e+00 |
Pos logit contributions | |
parties | +0.202 |
tea | +0.172 |
recess | +0.162 |
ashed | +0.160 |
notebooks | +0.153 |
Neg logit contributions | |
beast | -0.179 |
gut | -0.173 |
mistake | -0.167 |
cut | -0.161 |
inct | -0.154 |
Token | you |
Token ID | 345 |
Feature activation | -5.21e+00 |
Pos logit contributions | |
parties | +0.095 |
tea | +0.081 |
recess | +0.068 |
celebrate | +0.063 |
party | +0.061 |
Neg logit contributions | |
beast | -0.130 |
gut | -0.129 |
mistake | -0.122 |
therm | -0.120 |
swollen | -0.119 |
Token | ," |
Token ID | 553 |
Feature activation | -4.64e-01 |
Pos logit contributions | |
parties | +0.162 |
tea | +0.144 |
recess | +0.126 |
celebrate | +0.115 |
picnic | +0.111 |
Neg logit contributions | |
beast | -0.178 |
gut | -0.178 |
mistake | -0.170 |
therm | -0.165 |
swollen | -0.163 |
Token | they |
Token ID | 484 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
parties | +0.012 |
tea | +0.010 |
recess | +0.008 |
celebrate | +0.008 |
party | +0.007 |
Neg logit contributions | |
beast | -0.019 |
gut | -0.019 |
mistake | -0.018 |
inct | -0.018 |
therm | -0.018 |
Token | say |
Token ID | 910 |
Feature activation | -3.23e+00 |
Pos logit contributions | |
parties | +0.045 |
tea | +0.041 |
recess | +0.038 |
dates | +0.036 |
ils | +0.034 |
Neg logit contributions | |
gut | -0.044 |
beast | -0.044 |
mistake | -0.042 |
therm | -0.042 |
plea | -0.041 |