Token | by |
Token ID | 416 |
Feature activation | +1.48e+01 |
Pos logit contributions | |
Un | +0.023 |
David | +0.023 |
Fro | +0.023 |
concession | +0.022 |
Jack | +0.022 |
Neg logit contributions | |
hole | -0.040 |
pieces | -0.040 |
on | -0.039 |
atop | -0.037 |
edges | -0.037 |
Token | one |
Token ID | 530 |
Feature activation | +1.14e+01 |
Pos logit contributions | |
Un | +0.166 |
Jack | +0.164 |
Fro | +0.164 |
David | +0.161 |
Jim | +0.158 |
Neg logit contributions | |
hole | -0.172 |
pieces | -0.167 |
atop | -0.158 |
lid | -0.153 |
edges | -0.153 |
Token | � |
Token ID | 6184 |
Feature activation | +1.22e+01 |
Pos logit contributions | |
Jack | +0.117 |
Un | +0.116 |
concession | +0.116 |
Fro | +0.116 |
Jim | +0.114 |
Neg logit contributions | |
ps | -0.124 |
hole | -0.124 |
atop | -0.117 |
pieces | -0.117 |
tangled | -0.116 |
Token | � |
Token ID | 95 |
Feature activation | +1.31e+01 |
Pos logit contributions | |
Un | +0.123 |
Jack | +0.122 |
David | +0.122 |
Fro | +0.121 |
concession | +0.116 |
Neg logit contributions | |
pieces | -0.161 |
hole | -0.158 |
on | -0.147 |
atop | -0.147 |
edges | -0.146 |
Token | € |
Token ID | 26391 |
Feature activation | +1.85e+01 |
Pos logit contributions | |
Jack | +0.119 |
Un | +0.119 |
Fro | +0.118 |
David | +0.118 |
Jim | +0.112 |
Neg logit contributions | |
hole | -0.182 |
pieces | -0.179 |
atop | -0.170 |
on | -0.169 |
edges | -0.168 |
Token | � |
Token ID | 447 |
Feature activation | +2.22e+01 |
Pos logit contributions | |
concession | +0.169 |
Un | +0.167 |
Da | +0.167 |
Fro | +0.165 |
Jack | +0.165 |
Neg logit contributions | |
hole | -0.251 |
on | -0.237 |
pieces | -0.232 |
atop | -0.226 |
falling | -0.221 |
Token | � |
Token ID | 250 |
Feature activation | +1.71e+01 |
Pos logit contributions | |
Un | +0.208 |
David | +0.206 |
Fro | +0.205 |
Jack | +0.205 |
Kids | +0.197 |
Neg logit contributions | |
hole | -0.294 |
pieces | -0.289 |
atop | -0.278 |
on | -0.276 |
edges | -0.268 |
Token | 1 |
Token ID | 352 |
Feature activation | +5.25e+00 |
Pos logit contributions | |
Fro | +0.114 |
Un | +0.113 |
Jack | +0.112 |
David | +0.112 |
concession | +0.108 |
Neg logit contributions | |
hole | -0.274 |
pieces | -0.273 |
on | -0.259 |
atop | -0.258 |
edges | -0.257 |
Token | , |
Token ID | 11 |
Feature activation | +1.15e+01 |
Pos logit contributions | |
Un | +0.063 |
Fro | +0.061 |
Jim | +0.061 |
David | +0.061 |
Jack | +0.061 |
Neg logit contributions | |
hole | -0.054 |
edges | -0.053 |
pieces | -0.053 |
falling | -0.051 |
tangled | -0.051 |
Token | 2 |
Token ID | 362 |
Feature activation | +5.85e+00 |
Pos logit contributions | |
Un | +0.113 |
David | +0.110 |
Jack | +0.110 |
Fro | +0.107 |
concession | +0.104 |
Neg logit contributions | |
hole | -0.153 |
pieces | -0.150 |
atop | -0.146 |
ps | -0.143 |
edges | -0.141 |
Token | , |
Token ID | 11 |
Feature activation | +1.10e+01 |
Pos logit contributions | |
Un | +0.067 |
David | +0.066 |
Jack | +0.066 |
Jim | +0.066 |
Fro | +0.064 |
Neg logit contributions | |
hole | -0.066 |
atop | -0.062 |
tangled | -0.062 |
ps | -0.062 |
edges | -0.061 |
Token | said |
Token ID | 531 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
hole | +0.098 |
pieces | +0.095 |
edges | +0.091 |
ps | +0.091 |
lid | +0.091 |
Neg logit contributions | |
Fro | -0.113 |
Jack | -0.104 |
Un | -0.104 |
enny | -0.101 |
His | -0.101 |
Token | � |
Token ID | 6184 |
Feature activation | +1.49e+01 |
Pos logit contributions | |
Un | +0.193 |
David | +0.189 |
Jack | +0.189 |
Kids | +0.185 |
concession | +0.185 |
Neg logit contributions | |
on | -0.112 |
pieces | -0.111 |
hole | -0.110 |
atop | -0.102 |
ps | -0.101 |
Token | � |
Token ID | 95 |
Feature activation | +1.32e+01 |
Pos logit contributions | |
Un | +0.153 |
Jack | +0.152 |
David | +0.151 |
Fro | +0.149 |
Jim | +0.144 |
Neg logit contributions | |
hole | -0.190 |
pieces | -0.184 |
atop | -0.175 |
edges | -0.174 |
on | -0.174 |
Token | € |
Token ID | 26391 |
Feature activation | +1.88e+01 |
Pos logit contributions | |
Un | +0.122 |
Jack | +0.122 |
David | +0.121 |
Fro | +0.120 |
Jim | +0.114 |
Neg logit contributions | |
hole | -0.182 |
pieces | -0.181 |
atop | -0.170 |
on | -0.169 |
edges | -0.169 |
Token | � |
Token ID | 129 |
Feature activation | +1.44e+01 |
Pos logit contributions | |
concession | +0.187 |
Da | +0.182 |
Un | +0.181 |
Jack | +0.179 |
Kids | +0.175 |
Neg logit contributions | |
hole | -0.241 |
on | -0.234 |
pieces | -0.219 |
atop | -0.214 |
trunk | -0.211 |
Token | � |
Token ID | 241 |
Feature activation | +2.13e+01 |
Pos logit contributions | |
Jack | +0.136 |
Un | +0.136 |
David | +0.134 |
Fro | +0.133 |
concession | +0.128 |
Neg logit contributions | |
hole | -0.193 |
pieces | -0.190 |
atop | -0.180 |
edges | -0.179 |
on | -0.179 |
Token | I |
Token ID | 40 |
Feature activation | +1.38e+01 |
Pos logit contributions | |
Fro | +0.179 |
concession | +0.175 |
oph | +0.173 |
Un | +0.172 |
Da | +0.168 |
Neg logit contributions | |
on | -0.301 |
hole | -0.300 |
pieces | -0.294 |
edges | -0.279 |
atop | -0.277 |
Token | kept |
Token ID | 4030 |
Feature activation | +6.17e+00 |
Pos logit contributions | |
Un | +0.142 |
David | +0.140 |
Jack | +0.139 |
Fro | +0.136 |
concession | +0.134 |
Neg logit contributions | |
hole | -0.174 |
pieces | -0.171 |
on | -0.166 |
atop | -0.162 |
edges | -0.157 |
Token | my |
Token ID | 616 |
Feature activation | +9.87e+00 |
Pos logit contributions | |
David | +0.075 |
Un | +0.074 |
Jack | +0.074 |
Fro | +0.074 |
Johnny | +0.073 |
Neg logit contributions | |
hole | -0.069 |
pieces | -0.065 |
lid | -0.061 |
edges | -0.061 |
atop | -0.061 |
Token | promise |
Token ID | 6991 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
Fro | +0.119 |
concession | +0.114 |
Un | +0.113 |
Jack | +0.111 |
David | +0.111 |
Neg logit contributions | |
hole | -0.111 |
atop | -0.103 |
pieces | -0.102 |
ps | -0.096 |
edges | -0.095 |
Token | ! |
Token ID | 0 |
Feature activation | +1.61e+01 |
Pos logit contributions | |
Un | +0.027 |
Jack | +0.027 |
David | +0.027 |
Fro | +0.027 |
concession | +0.026 |
Neg logit contributions | |
pieces | -0.021 |
hole | -0.021 |
atop | -0.019 |
edges | -0.019 |
ps | -0.019 |
Token | said |
Token ID | 531 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
hole | +0.193 |
pieces | +0.191 |
edges | +0.187 |
els | +0.187 |
atop | +0.184 |
Neg logit contributions | |
Un | -0.114 |
Jack | -0.113 |
David | -0.111 |
concession | -0.107 |
enny | -0.107 |
Token | � |
Token ID | 6184 |
Feature activation | +1.49e+01 |
Pos logit contributions | |
Un | +0.193 |
David | +0.189 |
Jack | +0.189 |
Kids | +0.185 |
concession | +0.185 |
Neg logit contributions | |
on | -0.112 |
pieces | -0.111 |
hole | -0.110 |
atop | -0.102 |
ps | -0.101 |
Token | � |
Token ID | 95 |
Feature activation | +1.32e+01 |
Pos logit contributions | |
Un | +0.153 |
Jack | +0.152 |
David | +0.151 |
Fro | +0.149 |
Jim | +0.144 |
Neg logit contributions | |
hole | -0.190 |
pieces | -0.184 |
atop | -0.175 |
edges | -0.174 |
on | -0.174 |
Token | € |
Token ID | 26391 |
Feature activation | +1.88e+01 |
Pos logit contributions | |
Un | +0.122 |
Jack | +0.122 |
David | +0.121 |
Fro | +0.120 |
Jim | +0.114 |
Neg logit contributions | |
hole | -0.182 |
pieces | -0.181 |
atop | -0.170 |
on | -0.169 |
edges | -0.169 |
Token | � |
Token ID | 129 |
Feature activation | +1.44e+01 |
Pos logit contributions | |
concession | +0.187 |
Da | +0.182 |
Un | +0.181 |
Jack | +0.179 |
Kids | +0.175 |
Neg logit contributions | |
hole | -0.241 |
on | -0.234 |
pieces | -0.219 |
atop | -0.214 |
trunk | -0.211 |
Token | � |
Token ID | 241 |
Feature activation | +2.13e+01 |
Pos logit contributions | |
Jack | +0.136 |
Un | +0.136 |
David | +0.134 |
Fro | +0.133 |
concession | +0.128 |
Neg logit contributions | |
hole | -0.193 |
pieces | -0.190 |
atop | -0.180 |
edges | -0.179 |
on | -0.179 |
Token | don |
Token ID | 9099 |
Feature activation | +1.34e+01 |
Pos logit contributions | |
Fro | +0.179 |
concession | +0.175 |
oph | +0.173 |
Un | +0.172 |
Da | +0.168 |
Neg logit contributions | |
on | -0.301 |
hole | -0.300 |
pieces | -0.294 |
edges | -0.279 |
atop | -0.277 |
Token | â |
Token ID | 22940 |
Feature activation | +1.58e+01 |
Pos logit contributions | |
Fro | +0.180 |
concession | +0.178 |
Jack | +0.176 |
Da | +0.175 |
Un | +0.175 |
Neg logit contributions | |
on | -0.124 |
hole | -0.118 |
pieces | -0.114 |
button | -0.108 |
hardest | -0.107 |
Token | € |
Token ID | 26391 |
Feature activation | +1.81e+01 |
Pos logit contributions | |
Jack | +0.094 |
Un | +0.093 |
David | +0.093 |
Fro | +0.091 |
Kids | +0.088 |
Neg logit contributions | |
pieces | -0.273 |
hole | -0.268 |
atop | -0.255 |
edges | -0.255 |
on | -0.254 |
Token | ™ |
Token ID | 8151 |
Feature activation | +1.36e+01 |
Pos logit contributions | |
concession | +0.170 |
Jack | +0.170 |
Un | +0.170 |
Da | +0.169 |
David | +0.164 |
Neg logit contributions | |
hole | -0.234 |
on | -0.227 |
pieces | -0.219 |
atop | -0.211 |
ladder | -0.208 |
Token | t |
Token ID | 83 |
Feature activation | +1.05e+01 |
Pos logit contributions | |
Un | +0.151 |
Jack | +0.150 |
David | +0.150 |
Fro | +0.148 |
Jim | +0.143 |
Neg logit contributions | |
hole | -0.161 |
pieces | -0.160 |
atop | -0.149 |
on | -0.148 |
edges | -0.147 |
Token | shouted |
Token ID | 17293 |
Feature activation | +1.07e+01 |
Pos logit contributions | |
bones | +0.180 |
cks | +0.179 |
splits | +0.175 |
hole | +0.175 |
ps | +0.171 |
Neg logit contributions | |
Jack | -0.140 |
Pete | -0.139 |
Benny | -0.139 |
Fro | -0.139 |
him | -0.138 |
Token | � |
Token ID | 6184 |
Feature activation | +1.38e+01 |
Pos logit contributions | |
Un | +0.164 |
Jack | +0.159 |
David | +0.159 |
Kids | +0.157 |
concession | +0.156 |
Neg logit contributions | |
on | -0.085 |
pieces | -0.081 |
hole | -0.081 |
ps | -0.079 |
atop | -0.076 |
Token | � |
Token ID | 95 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
Un | +0.141 |
Da | +0.139 |
Jack | +0.138 |
David | +0.135 |
Jim | +0.131 |
Neg logit contributions | |
hole | -0.178 |
ps | -0.167 |
pieces | -0.163 |
on | -0.163 |
edges | -0.161 |
Token | € |
Token ID | 26391 |
Feature activation | +1.87e+01 |
Pos logit contributions | |
Fro | +0.129 |
Un | +0.128 |
Jack | +0.127 |
Da | +0.124 |
David | +0.123 |
Neg logit contributions | |
hole | -0.177 |
pieces | -0.166 |
ps | -0.164 |
edges | -0.162 |
on | -0.161 |
Token | � |
Token ID | 129 |
Feature activation | +1.45e+01 |
Pos logit contributions | |
concession | +0.182 |
Da | +0.180 |
Un | +0.177 |
Jack | +0.175 |
Kids | +0.173 |
Neg logit contributions | |
hole | -0.245 |
on | -0.235 |
pieces | -0.221 |
atop | -0.219 |
trunk | -0.214 |
Token | � |
Token ID | 241 |
Feature activation | +2.11e+01 |
Pos logit contributions | |
Jack | +0.140 |
Un | +0.135 |
Fro | +0.135 |
Da | +0.134 |
concession | +0.134 |
Neg logit contributions | |
hole | -0.189 |
ps | -0.181 |
pieces | -0.177 |
edges | -0.176 |
atop | -0.174 |
Token | Run |
Token ID | 10987 |
Feature activation | +5.56e+00 |
Pos logit contributions | |
Fro | +0.182 |
oph | +0.180 |
concession | +0.179 |
Un | +0.175 |
Da | +0.173 |
Neg logit contributions | |
on | -0.293 |
hole | -0.289 |
pieces | -0.282 |
edges | -0.269 |
atop | -0.266 |
Token | Tony |
Token ID | 8832 |
Feature activation | +8.00e+00 |
Pos logit contributions | |
Un | +0.078 |
Jack | +0.078 |
David | +0.078 |
Fro | +0.077 |
Jim | +0.075 |
Neg logit contributions | |
hole | -0.049 |
pieces | -0.049 |
on | -0.045 |
atop | -0.044 |
edges | -0.044 |
Token | , |
Token ID | 11 |
Feature activation | +1.05e+01 |
Pos logit contributions | |
Un | +0.105 |
Fro | +0.104 |
Jack | +0.104 |
David | +0.104 |
Kids | +0.100 |
Neg logit contributions | |
pieces | -0.078 |
hole | -0.078 |
atop | -0.072 |
edges | -0.071 |
falling | -0.069 |
Token | Run |
Token ID | 5660 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
Un | +0.101 |
oph | +0.099 |
Jack | +0.099 |
David | +0.097 |
Da | +0.096 |
Neg logit contributions | |
on | -0.137 |
hole | -0.134 |
pieces | -0.133 |
falling | -0.123 |
lid | -0.122 |
Token | ! |
Token ID | 0 |
Feature activation | +1.64e+01 |
Pos logit contributions | |
Jack | +0.029 |
David | +0.029 |
Un | +0.029 |
Fro | +0.028 |
Jim | +0.028 |
Neg logit contributions | |
hole | -0.018 |
pieces | -0.017 |
on | -0.017 |
edges | -0.016 |
ps | -0.016 |
Token | himself |
Token ID | 2241 |
Feature activation | +1.51e+01 |
Pos logit contributions | |
hole | +0.026 |
pieces | +0.026 |
atop | +0.024 |
on | +0.024 |
edges | +0.024 |
Neg logit contributions | |
Un | -0.020 |
Jack | -0.019 |
David | -0.019 |
Fro | -0.019 |
concession | -0.018 |
Token | � |
Token ID | 6184 |
Feature activation | +1.27e+01 |
Pos logit contributions | |
Fro | +0.179 |
Jack | +0.179 |
David | +0.174 |
Un | +0.169 |
Jim | +0.169 |
Neg logit contributions | |
hole | -0.164 |
pieces | -0.160 |
atop | -0.147 |
edges | -0.147 |
ps | -0.146 |
Token | � |
Token ID | 95 |
Feature activation | +1.32e+01 |
Pos logit contributions | |
Un | +0.130 |
Jack | +0.129 |
David | +0.129 |
Fro | +0.127 |
Da | +0.123 |
Neg logit contributions | |
hole | -0.161 |
pieces | -0.153 |
atop | -0.150 |
on | -0.147 |
edges | -0.145 |
Token | € |
Token ID | 26391 |
Feature activation | +1.85e+01 |
Pos logit contributions | |
Fro | +0.121 |
Un | +0.120 |
Jack | +0.120 |
David | +0.119 |
Da | +0.114 |
Neg logit contributions | |
hole | -0.181 |
pieces | -0.175 |
atop | -0.168 |
on | -0.167 |
edges | -0.167 |
Token | � |
Token ID | 129 |
Feature activation | +1.39e+01 |
Pos logit contributions | |
concession | +0.180 |
Da | +0.176 |
Un | +0.176 |
Kids | +0.174 |
Jack | +0.171 |
Neg logit contributions | |
hole | -0.242 |
on | -0.229 |
pieces | -0.217 |
atop | -0.216 |
trunk | -0.210 |
Token | � |
Token ID | 241 |
Feature activation | +2.09e+01 |
Pos logit contributions | |
Jack | +0.127 |
Un | +0.127 |
David | +0.125 |
Fro | +0.124 |
Jim | +0.118 |
Neg logit contributions | |
hole | -0.191 |
pieces | -0.190 |
atop | -0.179 |
on | -0.178 |
edges | -0.177 |
Token | Sometimes |
Token ID | 15468 |
Feature activation | +8.89e+00 |
Pos logit contributions | |
oph | +0.178 |
Fro | +0.175 |
concession | +0.172 |
Da | +0.169 |
David | +0.167 |
Neg logit contributions | |
on | -0.293 |
hole | -0.289 |
pieces | -0.286 |
edges | -0.275 |
atop | -0.268 |
Token | it |
Token ID | 340 |
Feature activation | +5.50e+00 |
Pos logit contributions | |
Un | +0.110 |
Jack | +0.109 |
Fro | +0.109 |
David | +0.108 |
Kids | +0.104 |
Neg logit contributions | |
hole | -0.095 |
pieces | -0.092 |
atop | -0.087 |
ps | -0.086 |
edges | -0.084 |
Token | â |
Token ID | 22940 |
Feature activation | +1.34e+01 |
Pos logit contributions | |
Un | +0.062 |
Fro | +0.061 |
Jack | +0.061 |
concession | +0.059 |
Kids | +0.059 |
Neg logit contributions | |
hole | -0.064 |
pieces | -0.061 |
atop | -0.060 |
edges | -0.057 |
ps | -0.057 |
Token | € |
Token ID | 26391 |
Feature activation | +1.55e+01 |
Pos logit contributions | |
Jack | +0.073 |
Kids | +0.072 |
Un | +0.072 |
David | +0.070 |
Fro | +0.066 |
Neg logit contributions | |
pieces | -0.243 |
hole | -0.237 |
lid | -0.227 |
atop | -0.224 |
edges | -0.224 |
Token | ™ |
Token ID | 8151 |
Feature activation | +1.11e+01 |
Pos logit contributions | |
Un | +0.142 |
Jack | +0.142 |
Fro | +0.139 |
Da | +0.139 |
concession | +0.139 |
Neg logit contributions | |
hole | -0.207 |
pieces | -0.200 |
on | -0.197 |
atop | -0.189 |
falling | -0.188 |
Token |
|
Token ID | 198 |
Feature activation | +2.18e+01 |
Pos logit contributions | |
hole | +0.048 |
pieces | +0.048 |
atop | +0.045 |
edges | +0.044 |
ps | +0.043 |
Neg logit contributions | |
Un | -0.045 |
Jack | -0.045 |
David | -0.045 |
Fro | -0.045 |
concession | -0.043 |
Token |
|
Token ID | 198 |
Feature activation | +1.92e+01 |
Pos logit contributions | |
concession | +0.222 |
oph | +0.220 |
latable | +0.210 |
Fro | +0.207 |
ory | +0.195 |
Neg logit contributions | |
on | -0.254 |
pieces | -0.252 |
hole | -0.248 |
atop | -0.236 |
slot | -0.227 |
Token | â |
Token ID | 22940 |
Feature activation | +1.85e+01 |
Pos logit contributions | |
Fro | +0.182 |
concession | +0.177 |
oph | +0.171 |
latable | +0.170 |
elt | +0.168 |
Neg logit contributions | |
pieces | -0.226 |
hole | -0.225 |
on | -0.219 |
atop | -0.210 |
slot | -0.208 |
Token | € |
Token ID | 26391 |
Feature activation | +2.02e+01 |
Pos logit contributions | |
Jack | +0.130 |
David | +0.127 |
Un | +0.123 |
Jim | +0.118 |
Kids | +0.118 |
Neg logit contributions | |
pieces | -0.312 |
hole | -0.295 |
on | -0.287 |
lid | -0.284 |
edges | -0.283 |
Token | � |
Token ID | 129 |
Feature activation | +1.51e+01 |
Pos logit contributions | |
concession | +0.195 |
Un | +0.193 |
Kids | +0.189 |
Fro | +0.187 |
Jack | +0.186 |
Neg logit contributions | |
hole | -0.270 |
pieces | -0.251 |
on | -0.248 |
atop | -0.243 |
falling | -0.241 |
Token | � |
Token ID | 241 |
Feature activation | +2.08e+01 |
Pos logit contributions | |
Jack | +0.155 |
Un | +0.155 |
David | +0.153 |
Fro | +0.153 |
concession | +0.147 |
Neg logit contributions | |
hole | -0.189 |
pieces | -0.185 |
atop | -0.176 |
edges | -0.175 |
ps | -0.175 |
Token | Come |
Token ID | 16773 |
Feature activation | +6.79e+00 |
Pos logit contributions | |
Fro | +0.168 |
oph | +0.161 |
concession | +0.161 |
Un | +0.157 |
Da | +0.151 |
Neg logit contributions | |
hole | -0.308 |
on | -0.301 |
pieces | -0.298 |
atop | -0.287 |
edges | -0.286 |
Token | on |
Token ID | 319 |
Feature activation | +7.47e+00 |
Pos logit contributions | |
Jack | +0.130 |
Un | +0.130 |
David | +0.129 |
oph | +0.128 |
Da | +0.128 |
Neg logit contributions | |
on | -0.028 |
hole | -0.025 |
pieces | -0.022 |
atop | -0.021 |
falling | -0.019 |
Token | now |
Token ID | 783 |
Feature activation | +5.85e+00 |
Pos logit contributions | |
Un | +0.092 |
Jack | +0.091 |
David | +0.091 |
Fro | +0.089 |
Jim | +0.088 |
Neg logit contributions | |
pieces | -0.079 |
hole | -0.079 |
on | -0.075 |
atop | -0.073 |
falling | -0.072 |
Token | , |
Token ID | 11 |
Feature activation | +9.88e+00 |
Pos logit contributions | |
Un | +0.081 |
Jack | +0.081 |
David | +0.081 |
Fro | +0.080 |
Jim | +0.078 |
Neg logit contributions | |
hole | -0.053 |
pieces | -0.052 |
atop | -0.048 |
on | -0.047 |
edges | -0.047 |
Token | Katie |
Token ID | 24721 |
Feature activation | +8.25e+00 |
Pos logit contributions | |
Un | +0.100 |
Jack | +0.099 |
David | +0.098 |
Da | +0.096 |
Jim | +0.095 |
Neg logit contributions | |
on | -0.124 |
pieces | -0.122 |
hole | -0.122 |
edges | -0.112 |
falling | -0.112 |
Token |
|
Token ID | 198 |
Feature activation | +2.18e+01 |
Pos logit contributions | |
hole | +0.056 |
pieces | +0.055 |
atop | +0.051 |
ps | +0.050 |
edges | +0.050 |
Neg logit contributions | |
Un | -0.060 |
Jack | -0.059 |
David | -0.059 |
Fro | -0.057 |
concession | -0.057 |
Token |
|
Token ID | 198 |
Feature activation | +1.92e+01 |
Pos logit contributions | |
concession | +0.222 |
oph | +0.220 |
latable | +0.210 |
Fro | +0.207 |
ory | +0.195 |
Neg logit contributions | |
on | -0.254 |
pieces | -0.252 |
hole | -0.248 |
atop | -0.236 |
slot | -0.227 |
Token | â |
Token ID | 22940 |
Feature activation | +1.85e+01 |
Pos logit contributions | |
Fro | +0.182 |
concession | +0.177 |
oph | +0.171 |
latable | +0.170 |
elt | +0.168 |
Neg logit contributions | |
pieces | -0.226 |
hole | -0.225 |
on | -0.219 |
atop | -0.210 |
slot | -0.208 |
Token | € |
Token ID | 26391 |
Feature activation | +2.02e+01 |
Pos logit contributions | |
Jack | +0.130 |
David | +0.127 |
Un | +0.123 |
Jim | +0.118 |
Kids | +0.118 |
Neg logit contributions | |
pieces | -0.312 |
hole | -0.295 |
on | -0.287 |
lid | -0.284 |
edges | -0.283 |
Token | � |
Token ID | 129 |
Feature activation | +1.51e+01 |
Pos logit contributions | |
concession | +0.195 |
Un | +0.193 |
Kids | +0.189 |
Fro | +0.187 |
Jack | +0.186 |
Neg logit contributions | |
hole | -0.270 |
pieces | -0.251 |
on | -0.248 |
atop | -0.243 |
falling | -0.241 |
Token | � |
Token ID | 241 |
Feature activation | +2.08e+01 |
Pos logit contributions | |
Jack | +0.155 |
Un | +0.155 |
David | +0.153 |
Fro | +0.153 |
concession | +0.147 |
Neg logit contributions | |
hole | -0.189 |
pieces | -0.185 |
atop | -0.176 |
edges | -0.175 |
ps | -0.175 |
Token | Hello |
Token ID | 15496 |
Feature activation | +5.71e+00 |
Pos logit contributions | |
Fro | +0.168 |
oph | +0.161 |
concession | +0.161 |
Un | +0.157 |
Da | +0.151 |
Neg logit contributions | |
hole | -0.308 |
on | -0.301 |
pieces | -0.298 |
atop | -0.287 |
edges | -0.286 |
Token | ! |
Token ID | 0 |
Feature activation | +1.69e+01 |
Pos logit contributions | |
Fro | +0.068 |
Jack | +0.068 |
Un | +0.067 |
David | +0.066 |
Johnny | +0.065 |
Neg logit contributions | |
pieces | -0.064 |
hole | -0.064 |
ps | -0.061 |
edges | -0.060 |
atop | -0.059 |
Token | Where |
Token ID | 6350 |
Feature activation | +1.05e+01 |
Pos logit contributions | |
Un | +0.147 |
David | +0.144 |
concession | +0.144 |
Fro | +0.144 |
Jack | +0.144 |
Neg logit contributions | |
hole | -0.233 |
pieces | -0.222 |
on | -0.221 |
atop | -0.215 |
edges | -0.207 |
Token | are |
Token ID | 389 |
Feature activation | +7.22e+00 |
Pos logit contributions | |
Un | +0.128 |
Jack | +0.127 |
David | +0.126 |
Fro | +0.126 |
concession | +0.121 |
Neg logit contributions | |
hole | -0.113 |
pieces | -0.112 |
on | -0.104 |
atop | -0.103 |
edges | -0.102 |
Token | you |
Token ID | 345 |
Feature activation | +4.65e+00 |
Pos logit contributions | |
Un | +0.086 |
David | +0.085 |
concession | +0.084 |
Fro | +0.083 |
enny | +0.082 |
Neg logit contributions | |
pieces | -0.082 |
on | -0.076 |
changing | -0.070 |
falling | -0.069 |
hardest | -0.069 |
Token | perfect |
Token ID | 2818 |
Feature activation | +1.00e+01 |
Pos logit contributions | |
Un | +0.220 |
Jack | +0.220 |
David | +0.218 |
Fro | +0.216 |
Jim | +0.209 |
Neg logit contributions | |
hole | -0.185 |
pieces | -0.185 |
atop | -0.170 |
on | -0.169 |
edges | -0.168 |
Token | bag |
Token ID | 6131 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
Fro | +0.128 |
Jack | +0.127 |
Un | +0.125 |
David | +0.124 |
concession | +0.124 |
Neg logit contributions | |
hole | -0.097 |
pieces | -0.090 |
atop | -0.089 |
ps | -0.085 |
els | -0.084 |
Token | � |
Token ID | 6184 |
Feature activation | +9.11e+00 |
Pos logit contributions | |
Jack | +0.020 |
concession | +0.020 |
enny | +0.020 |
Un | +0.020 |
David | +0.019 |
Neg logit contributions | |
hole | -0.014 |
pieces | -0.014 |
ps | -0.014 |
els | -0.014 |
atop | -0.014 |
Token | � |
Token ID | 95 |
Feature activation | +1.14e+01 |
Pos logit contributions | |
Un | +0.089 |
Jack | +0.089 |
David | +0.088 |
Fro | +0.086 |
concession | +0.085 |
Neg logit contributions | |
pieces | -0.120 |
hole | -0.119 |
ps | -0.112 |
atop | -0.112 |
falling | -0.111 |
Token | € |
Token ID | 26391 |
Feature activation | +1.67e+01 |
Pos logit contributions | |
Un | +0.110 |
Jack | +0.110 |
David | +0.109 |
Fro | +0.109 |
Jim | +0.104 |
Neg logit contributions | |
hole | -0.152 |
pieces | -0.150 |
atop | -0.141 |
on | -0.140 |
edges | -0.140 |
Token | � |
Token ID | 447 |
Feature activation | +2.05e+01 |
Pos logit contributions | |
Un | +0.167 |
Fro | +0.165 |
Jack | +0.164 |
concession | +0.164 |
Da | +0.164 |
Neg logit contributions | |
hole | -0.214 |
on | -0.205 |
pieces | -0.199 |
atop | -0.191 |
falling | -0.187 |
Token | � |
Token ID | 250 |
Feature activation | +1.62e+01 |
Pos logit contributions | |
Un | +0.209 |
David | +0.207 |
Fro | +0.206 |
Jack | +0.204 |
Kids | +0.198 |
Neg logit contributions | |
hole | -0.255 |
pieces | -0.252 |
on | -0.242 |
atop | -0.241 |
edges | -0.231 |
Token | now |
Token ID | 783 |
Feature activation | +9.01e+00 |
Pos logit contributions | |
Un | +0.104 |
Jack | +0.103 |
David | +0.102 |
Fro | +0.095 |
Jim | +0.093 |
Neg logit contributions | |
hole | -0.271 |
pieces | -0.270 |
atop | -0.259 |
ps | -0.257 |
falling | -0.256 |
Token | I |
Token ID | 314 |
Feature activation | +9.55e+00 |
Pos logit contributions | |
Un | +0.095 |
David | +0.094 |
Jack | +0.094 |
Fro | +0.093 |
concession | +0.090 |
Neg logit contributions | |
pieces | -0.111 |
hole | -0.110 |
on | -0.107 |
atop | -0.103 |
edges | -0.102 |
Token | 'm |
Token ID | 1101 |
Feature activation | +9.80e+00 |
Pos logit contributions | |
Jack | +0.101 |
Un | +0.099 |
Fro | +0.099 |
David | +0.099 |
Kids | +0.095 |
Neg logit contributions | |
hole | -0.118 |
pieces | -0.118 |
atop | -0.111 |
edges | -0.110 |
ps | -0.109 |
Token | very |
Token ID | 845 |
Feature activation | +1.08e+01 |
Pos logit contributions | |
Fro | +0.103 |
Un | +0.101 |
David | +0.101 |
Jack | +0.100 |
concession | +0.097 |
Neg logit contributions | |
hole | -0.125 |
pieces | -0.125 |
edges | -0.116 |
atop | -0.114 |
lid | -0.114 |
Token | , |
Token ID | 11 |
Feature activation | +1.69e+01 |
Pos logit contributions | |
hole | +0.145 |
ps | +0.141 |
cks | +0.139 |
hardest | +0.130 |
increasing | +0.129 |
Neg logit contributions | |
wife | -0.111 |
Her | -0.109 |
His | -0.108 |
Jack | -0.108 |
Her | -0.107 |
Token | � |
Token ID | 6184 |
Feature activation | +1.28e+01 |
Pos logit contributions | |
Un | +0.176 |
Jack | +0.174 |
David | +0.174 |
Fro | +0.171 |
Kids | +0.166 |
Neg logit contributions | |
pieces | -0.211 |
hole | -0.210 |
on | -0.200 |
atop | -0.196 |
edges | -0.194 |
Token | � |
Token ID | 95 |
Feature activation | +1.24e+01 |
Pos logit contributions | |
Un | +0.133 |
Jack | +0.132 |
David | +0.132 |
Fro | +0.131 |
concession | +0.126 |
Neg logit contributions | |
hole | -0.161 |
pieces | -0.158 |
atop | -0.151 |
on | -0.149 |
edges | -0.147 |
Token | € |
Token ID | 26391 |
Feature activation | +1.83e+01 |
Pos logit contributions | |
Fro | +0.124 |
Un | +0.122 |
Jack | +0.122 |
David | +0.120 |
Jim | +0.116 |
Neg logit contributions | |
hole | -0.161 |
pieces | -0.157 |
atop | -0.149 |
edges | -0.148 |
on | -0.148 |
Token | � |
Token ID | 129 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
concession | +0.172 |
Un | +0.170 |
Da | +0.170 |
Fro | +0.168 |
Jack | +0.168 |
Neg logit contributions | |
hole | -0.246 |
on | -0.231 |
pieces | -0.226 |
atop | -0.221 |
falling | -0.217 |
Token | � |
Token ID | 241 |
Feature activation | +2.04e+01 |
Pos logit contributions | |
Jack | +0.130 |
Un | +0.130 |
David | +0.129 |
Fro | +0.128 |
concession | +0.123 |
Neg logit contributions | |
hole | -0.174 |
pieces | -0.172 |
atop | -0.162 |
on | -0.161 |
edges | -0.161 |
Token | Thanks |
Token ID | 9690 |
Feature activation | +7.41e+00 |
Pos logit contributions | |
Fro | +0.173 |
oph | +0.172 |
concession | +0.167 |
Un | +0.162 |
Da | +0.162 |
Neg logit contributions | |
hole | -0.286 |
on | -0.286 |
pieces | -0.280 |
edges | -0.270 |
atop | -0.263 |
Token | , |
Token ID | 11 |
Feature activation | +1.06e+01 |
Pos logit contributions | |
Un | +0.103 |
Jack | +0.103 |
David | +0.102 |
Fro | +0.102 |
Jim | +0.098 |
Neg logit contributions | |
hole | -0.067 |
pieces | -0.067 |
atop | -0.061 |
on | -0.060 |
edges | -0.060 |
Token | Jack |
Token ID | 3619 |
Feature activation | +8.42e+00 |
Pos logit contributions | |
Un | +0.102 |
David | +0.100 |
Jack | +0.100 |
Fro | +0.096 |
Da | +0.095 |
Neg logit contributions | |
pieces | -0.138 |
on | -0.137 |
hole | -0.136 |
falling | -0.127 |
edges | -0.127 |
Token | . |
Token ID | 13 |
Feature activation | +1.64e+01 |
Pos logit contributions | |
Fro | +0.108 |
Jack | +0.107 |
David | +0.105 |
Un | +0.104 |
Jim | +0.101 |
Neg logit contributions | |
hole | -0.085 |
pieces | -0.082 |
atop | -0.077 |
edges | -0.077 |
lid | -0.075 |
Token | You |
Token ID | 921 |
Feature activation | +8.72e+00 |
Pos logit contributions | |
Un | +0.162 |
David | +0.159 |
concession | +0.158 |
Da | +0.157 |
Jack | +0.157 |
Neg logit contributions | |
hole | -0.208 |
pieces | -0.195 |
on | -0.192 |
edges | -0.183 |
trunk | -0.182 |
Token | , |
Token ID | 11 |
Feature activation | +1.69e+01 |
Pos logit contributions | |
pieces | +0.013 |
on | +0.013 |
hole | +0.012 |
atop | +0.012 |
falling | +0.012 |
Neg logit contributions | |
David | -0.013 |
Jack | -0.013 |
Un | -0.013 |
Jim | -0.012 |
Fro | -0.012 |
Token | � |
Token ID | 6184 |
Feature activation | +1.28e+01 |
Pos logit contributions | |
Un | +0.176 |
Jack | +0.174 |
David | +0.174 |
Fro | +0.171 |
Kids | +0.166 |
Neg logit contributions | |
pieces | -0.211 |
hole | -0.210 |
on | -0.200 |
atop | -0.196 |
edges | -0.194 |
Token | � |
Token ID | 95 |
Feature activation | +1.24e+01 |
Pos logit contributions | |
Un | +0.133 |
Jack | +0.132 |
David | +0.132 |
Fro | +0.131 |
concession | +0.126 |
Neg logit contributions | |
hole | -0.161 |
pieces | -0.158 |
atop | -0.151 |
on | -0.149 |
edges | -0.147 |
Token | € |
Token ID | 26391 |
Feature activation | +1.83e+01 |
Pos logit contributions | |
Fro | +0.124 |
Un | +0.122 |
Jack | +0.122 |
David | +0.120 |
Jim | +0.116 |
Neg logit contributions | |
hole | -0.161 |
pieces | -0.157 |
atop | -0.149 |
edges | -0.148 |
on | -0.148 |
Token | � |
Token ID | 129 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
concession | +0.172 |
Un | +0.170 |
Da | +0.170 |
Fro | +0.168 |
Jack | +0.168 |
Neg logit contributions | |
hole | -0.246 |
on | -0.231 |
pieces | -0.226 |
atop | -0.221 |
falling | -0.217 |
Token | � |
Token ID | 241 |
Feature activation | +2.04e+01 |
Pos logit contributions | |
Jack | +0.130 |
Un | +0.130 |
David | +0.129 |
Fro | +0.128 |
concession | +0.123 |
Neg logit contributions | |
hole | -0.174 |
pieces | -0.172 |
atop | -0.162 |
on | -0.161 |
edges | -0.161 |
Token | We |
Token ID | 1135 |
Feature activation | +1.11e+01 |
Pos logit contributions | |
Fro | +0.173 |
oph | +0.172 |
concession | +0.167 |
Un | +0.162 |
Da | +0.162 |
Neg logit contributions | |
hole | -0.286 |
on | -0.286 |
pieces | -0.280 |
edges | -0.270 |
atop | -0.263 |
Token | use |
Token ID | 779 |
Feature activation | +5.17e+00 |
Pos logit contributions | |
Jack | +0.129 |
David | +0.127 |
Fro | +0.126 |
Un | +0.125 |
Jim | +0.121 |
Neg logit contributions | |
hole | -0.128 |
pieces | -0.126 |
atop | -0.118 |
edges | -0.117 |
ps | -0.115 |
Token | the |
Token ID | 262 |
Feature activation | +8.08e+00 |
Pos logit contributions | |
Jack | +0.051 |
Un | +0.051 |
Fro | +0.050 |
David | +0.050 |
concession | +0.048 |
Neg logit contributions | |
hole | -0.066 |
atop | -0.063 |
lid | -0.058 |
ps | -0.057 |
globe | -0.057 |
Token | gun |
Token ID | 2485 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
concession | +0.095 |
Fro | +0.094 |
shepherd | +0.093 |
wife | +0.092 |
please | +0.089 |
Neg logit contributions | |
hole | -0.086 |
cks | -0.079 |
atop | -0.076 |
bones | -0.074 |
ps | -0.070 |
Token | . |
Token ID | 13 |
Feature activation | +1.53e+01 |
Pos logit contributions | |
hole | +0.015 |
atop | +0.014 |
pieces | +0.014 |
lid | +0.014 |
edges | +0.014 |
Neg logit contributions | |
Un | -0.013 |
concession | -0.013 |
Jack | -0.013 |
Fro | -0.012 |
David | -0.012 |
Token | , |
Token ID | 11 |
Feature activation | +1.69e+01 |
Pos logit contributions | |
hole | +0.153 |
pieces | +0.149 |
atop | +0.147 |
edges | +0.144 |
ps | +0.144 |
Neg logit contributions | |
Fro | -0.101 |
Un | -0.099 |
Jack | -0.098 |
David | -0.095 |
Jim | -0.093 |
Token | � |
Token ID | 6184 |
Feature activation | +1.28e+01 |
Pos logit contributions | |
Un | +0.176 |
Jack | +0.174 |
David | +0.174 |
Fro | +0.171 |
Kids | +0.166 |
Neg logit contributions | |
pieces | -0.211 |
hole | -0.210 |
on | -0.200 |
atop | -0.196 |
edges | -0.194 |
Token | � |
Token ID | 95 |
Feature activation | +1.24e+01 |
Pos logit contributions | |
Un | +0.133 |
Jack | +0.132 |
David | +0.132 |
Fro | +0.131 |
concession | +0.126 |
Neg logit contributions | |
hole | -0.161 |
pieces | -0.158 |
atop | -0.151 |
on | -0.149 |
edges | -0.147 |
Token | € |
Token ID | 26391 |
Feature activation | +1.83e+01 |
Pos logit contributions | |
Fro | +0.124 |
Un | +0.122 |
Jack | +0.122 |
David | +0.120 |
Jim | +0.116 |
Neg logit contributions | |
hole | -0.161 |
pieces | -0.157 |
atop | -0.149 |
edges | -0.148 |
on | -0.148 |
Token | � |
Token ID | 129 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
concession | +0.172 |
Un | +0.170 |
Da | +0.170 |
Fro | +0.168 |
Jack | +0.168 |
Neg logit contributions | |
hole | -0.246 |
on | -0.231 |
pieces | -0.226 |
atop | -0.221 |
falling | -0.217 |
Token | � |
Token ID | 241 |
Feature activation | +2.04e+01 |
Pos logit contributions | |
Jack | +0.130 |
Un | +0.130 |
David | +0.129 |
Fro | +0.128 |
concession | +0.123 |
Neg logit contributions | |
hole | -0.174 |
pieces | -0.172 |
atop | -0.162 |
on | -0.161 |
edges | -0.161 |
Token | It |
Token ID | 1026 |
Feature activation | +8.28e+00 |
Pos logit contributions | |
Fro | +0.173 |
oph | +0.172 |
concession | +0.167 |
Un | +0.162 |
Da | +0.162 |
Neg logit contributions | |
hole | -0.286 |
on | -0.286 |
pieces | -0.280 |
edges | -0.270 |
atop | -0.263 |
Token | means |
Token ID | 1724 |
Feature activation | +7.62e+00 |
Pos logit contributions | |
Jack | +0.080 |
Un | +0.078 |
Fro | +0.078 |
concession | +0.077 |
Kids | +0.076 |
Neg logit contributions | |
hole | -0.106 |
pieces | -0.103 |
atop | -0.102 |
falling | -0.097 |
edges | -0.097 |
Token | I |
Token ID | 314 |
Feature activation | +8.43e+00 |
Pos logit contributions | |
Fro | +0.071 |
Jack | +0.071 |
Un | +0.071 |
David | +0.071 |
Johnny | +0.068 |
Neg logit contributions | |
hole | -0.105 |
pieces | -0.101 |
atop | -0.098 |
ps | -0.096 |
edges | -0.094 |
Token | 'm |
Token ID | 1101 |
Feature activation | +9.12e+00 |
Pos logit contributions | |
Jack | +0.082 |
Fro | +0.082 |
Un | +0.081 |
David | +0.080 |
Kids | +0.077 |
Neg logit contributions | |
pieces | -0.111 |
hole | -0.111 |
atop | -0.104 |
edges | -0.103 |
ps | -0.102 |
Token | gonna |
Token ID | 8066 |
Feature activation | +9.08e+00 |
Pos logit contributions | |
Fro | +0.096 |
Un | +0.096 |
David | +0.093 |
Jack | +0.091 |
concession | +0.089 |
Neg logit contributions | |
hole | -0.120 |
pieces | -0.119 |
lid | -0.109 |
edges | -0.109 |
atop | -0.108 |
Token | , |
Token ID | 11 |
Feature activation | +1.69e+01 |
Pos logit contributions | |
hole | +0.093 |
pieces | +0.091 |
atop | +0.083 |
edges | +0.081 |
ps | +0.080 |
Neg logit contributions | |
Un | -0.165 |
Jack | -0.164 |
David | -0.163 |
Fro | -0.162 |
Kids | -0.157 |
Token | � |
Token ID | 6184 |
Feature activation | +1.28e+01 |
Pos logit contributions | |
Un | +0.176 |
Jack | +0.174 |
David | +0.174 |
Fro | +0.171 |
Kids | +0.166 |
Neg logit contributions | |
pieces | -0.211 |
hole | -0.210 |
on | -0.200 |
atop | -0.196 |
edges | -0.194 |
Token | � |
Token ID | 95 |
Feature activation | +1.24e+01 |
Pos logit contributions | |
Un | +0.133 |
Jack | +0.132 |
David | +0.132 |
Fro | +0.131 |
concession | +0.126 |
Neg logit contributions | |
hole | -0.161 |
pieces | -0.158 |
atop | -0.151 |
on | -0.149 |
edges | -0.147 |
Token | € |
Token ID | 26391 |
Feature activation | +1.83e+01 |
Pos logit contributions | |
Fro | +0.124 |
Un | +0.122 |
Jack | +0.122 |
David | +0.120 |
Jim | +0.116 |
Neg logit contributions | |
hole | -0.161 |
pieces | -0.157 |
atop | -0.149 |
edges | -0.148 |
on | -0.148 |
Token | � |
Token ID | 129 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
concession | +0.172 |
Un | +0.170 |
Da | +0.170 |
Fro | +0.168 |
Jack | +0.168 |
Neg logit contributions | |
hole | -0.246 |
on | -0.231 |
pieces | -0.226 |
atop | -0.221 |
falling | -0.217 |
Token | � |
Token ID | 241 |
Feature activation | +2.04e+01 |
Pos logit contributions | |
Jack | +0.130 |
Un | +0.130 |
David | +0.129 |
Fro | +0.128 |
concession | +0.123 |
Neg logit contributions | |
hole | -0.174 |
pieces | -0.172 |
atop | -0.162 |
on | -0.161 |
edges | -0.161 |
Token | Hope |
Token ID | 34456 |
Feature activation | +8.26e+00 |
Pos logit contributions | |
Fro | +0.173 |
oph | +0.172 |
concession | +0.167 |
Un | +0.162 |
Da | +0.162 |
Neg logit contributions | |
hole | -0.286 |
on | -0.286 |
pieces | -0.280 |
edges | -0.270 |
atop | -0.263 |
Token | you |
Token ID | 345 |
Feature activation | +8.97e+00 |
Pos logit contributions | |
Fro | +0.096 |
Jack | +0.093 |
Un | +0.091 |
concession | +0.090 |
David | +0.089 |
Neg logit contributions | |
hole | -0.093 |
pieces | -0.086 |
atop | -0.082 |
edges | -0.082 |
ps | -0.081 |
Token | 'll |
Token ID | 1183 |
Feature activation | +8.86e+00 |
Pos logit contributions | |
Fro | +0.078 |
Jack | +0.078 |
David | +0.073 |
please | +0.071 |
concession | +0.070 |
Neg logit contributions | |
hole | -0.123 |
atop | -0.122 |
pieces | -0.121 |
els | -0.121 |
bones | -0.121 |
Token | be |
Token ID | 307 |
Feature activation | +6.28e+00 |
Pos logit contributions | |
Un | +0.090 |
Jack | +0.090 |
David | +0.089 |
Fro | +0.088 |
Jim | +0.085 |
Neg logit contributions | |
hole | -0.113 |
pieces | -0.112 |
atop | -0.105 |
on | -0.105 |
edges | -0.104 |
Token | more |
Token ID | 517 |
Feature activation | +5.23e+00 |
Pos logit contributions | |
Jack | +0.073 |
Un | +0.073 |
David | +0.073 |
Fro | +0.072 |
Jim | +0.070 |
Neg logit contributions | |
hole | -0.071 |
pieces | -0.070 |
atop | -0.065 |
edges | -0.065 |
ps | -0.063 |
Token | , |
Token ID | 11 |
Feature activation | +1.69e+01 |
Pos logit contributions | |
concession | +0.032 |
oph | +0.030 |
David | +0.030 |
Fro | +0.029 |
Un | +0.029 |
Neg logit contributions | |
on | -0.039 |
pieces | -0.038 |
hole | -0.037 |
atop | -0.035 |
trunk | -0.035 |
Token | � |
Token ID | 6184 |
Feature activation | +1.28e+01 |
Pos logit contributions | |
Un | +0.176 |
Jack | +0.174 |
David | +0.174 |
Fro | +0.171 |
Kids | +0.166 |
Neg logit contributions | |
pieces | -0.211 |
hole | -0.210 |
on | -0.200 |
atop | -0.196 |
edges | -0.194 |
Token | � |
Token ID | 95 |
Feature activation | +1.24e+01 |
Pos logit contributions | |
Un | +0.133 |
Jack | +0.132 |
David | +0.132 |
Fro | +0.131 |
concession | +0.126 |
Neg logit contributions | |
hole | -0.161 |
pieces | -0.158 |
atop | -0.151 |
on | -0.149 |
edges | -0.147 |
Token | € |
Token ID | 26391 |
Feature activation | +1.83e+01 |
Pos logit contributions | |
Fro | +0.124 |
Un | +0.122 |
Jack | +0.122 |
David | +0.120 |
Jim | +0.116 |
Neg logit contributions | |
hole | -0.161 |
pieces | -0.157 |
atop | -0.149 |
edges | -0.148 |
on | -0.148 |
Token | � |
Token ID | 129 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
concession | +0.172 |
Un | +0.170 |
Da | +0.170 |
Fro | +0.168 |
Jack | +0.168 |
Neg logit contributions | |
hole | -0.246 |
on | -0.231 |
pieces | -0.226 |
atop | -0.221 |
falling | -0.217 |
Token | � |
Token ID | 241 |
Feature activation | +2.04e+01 |
Pos logit contributions | |
Jack | +0.130 |
Un | +0.130 |
David | +0.129 |
Fro | +0.128 |
concession | +0.123 |
Neg logit contributions | |
hole | -0.174 |
pieces | -0.172 |
atop | -0.162 |
on | -0.161 |
edges | -0.161 |
Token | Will |
Token ID | 8743 |
Feature activation | +9.59e+00 |
Pos logit contributions | |
Fro | +0.173 |
oph | +0.172 |
concession | +0.167 |
Un | +0.162 |
Da | +0.162 |
Neg logit contributions | |
hole | -0.286 |
on | -0.286 |
pieces | -0.280 |
edges | -0.270 |
atop | -0.263 |
Token | you |
Token ID | 345 |
Feature activation | +7.41e+00 |
Pos logit contributions | |
Un | +0.105 |
Jack | +0.105 |
Fro | +0.104 |
David | +0.104 |
Jim | +0.100 |
Neg logit contributions | |
hole | -0.114 |
pieces | -0.114 |
atop | -0.107 |
edges | -0.105 |
on | -0.104 |
Token | come |
Token ID | 1282 |
Feature activation | +4.89e+00 |
Pos logit contributions | |
Fro | +0.070 |
Un | +0.068 |
Jack | +0.068 |
David | +0.067 |
concession | +0.065 |
Neg logit contributions | |
pieces | -0.100 |
hole | -0.098 |
edges | -0.093 |
atop | -0.093 |
els | -0.091 |
Token | listen |
Token ID | 6004 |
Feature activation | +3.04e+00 |
Pos logit contributions | |
Un | +0.070 |
Jack | +0.070 |
David | +0.069 |
Fro | +0.069 |
concession | +0.067 |
Neg logit contributions | |
hole | -0.042 |
pieces | -0.042 |
atop | -0.038 |
edges | -0.038 |
on | -0.037 |
Token | to |
Token ID | 284 |
Feature activation | +7.86e+00 |
Pos logit contributions | |
Un | +0.030 |
Jack | +0.030 |
David | +0.029 |
concession | +0.028 |
Fro | +0.028 |
Neg logit contributions | |
pieces | -0.039 |
hole | -0.039 |
atop | -0.037 |
ps | -0.037 |
lid | -0.037 |
Token | , |
Token ID | 11 |
Feature activation | +1.69e+01 |
Pos logit contributions | |
hole | +0.134 |
lid | +0.132 |
ps | +0.132 |
button | +0.127 |
globe | +0.126 |
Neg logit contributions | |
Jack | -0.117 |
Un | -0.116 |
David | -0.112 |
Da | -0.107 |
He | -0.106 |
Token | � |
Token ID | 6184 |
Feature activation | +1.28e+01 |
Pos logit contributions | |
Un | +0.176 |
Jack | +0.174 |
David | +0.174 |
Fro | +0.171 |
Kids | +0.166 |
Neg logit contributions | |
pieces | -0.211 |
hole | -0.210 |
on | -0.200 |
atop | -0.196 |
edges | -0.194 |
Token | � |
Token ID | 95 |
Feature activation | +1.24e+01 |
Pos logit contributions | |
Un | +0.133 |
Jack | +0.132 |
David | +0.132 |
Fro | +0.131 |
concession | +0.126 |
Neg logit contributions | |
hole | -0.161 |
pieces | -0.158 |
atop | -0.151 |
on | -0.149 |
edges | -0.147 |
Token | € |
Token ID | 26391 |
Feature activation | +1.83e+01 |
Pos logit contributions | |
Fro | +0.124 |
Un | +0.122 |
Jack | +0.122 |
David | +0.120 |
Jim | +0.116 |
Neg logit contributions | |
hole | -0.161 |
pieces | -0.157 |
atop | -0.149 |
edges | -0.148 |
on | -0.148 |
Token | � |
Token ID | 129 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
concession | +0.172 |
Un | +0.170 |
Da | +0.170 |
Fro | +0.168 |
Jack | +0.168 |
Neg logit contributions | |
hole | -0.246 |
on | -0.231 |
pieces | -0.226 |
atop | -0.221 |
falling | -0.217 |
Token | � |
Token ID | 241 |
Feature activation | +2.04e+01 |
Pos logit contributions | |
Jack | +0.130 |
Un | +0.130 |
David | +0.129 |
Fro | +0.128 |
concession | +0.123 |
Neg logit contributions | |
hole | -0.174 |
pieces | -0.172 |
atop | -0.162 |
on | -0.161 |
edges | -0.161 |
Token | This |
Token ID | 1212 |
Feature activation | +1.01e+01 |
Pos logit contributions | |
Fro | +0.173 |
oph | +0.172 |
concession | +0.167 |
Un | +0.162 |
Da | +0.162 |
Neg logit contributions | |
hole | -0.286 |
on | -0.286 |
pieces | -0.280 |
edges | -0.270 |
atop | -0.263 |
Token | box |
Token ID | 3091 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
Fro | +0.125 |
concession | +0.124 |
Jack | +0.123 |
David | +0.122 |
Kids | +0.121 |
Neg logit contributions | |
hole | -0.103 |
atop | -0.098 |
ps | -0.094 |
pieces | -0.089 |
edges | -0.089 |
Token | is |
Token ID | 318 |
Feature activation | +7.09e+00 |
Pos logit contributions | |
concession | +0.024 |
Jack | +0.023 |
enny | +0.023 |
addy | +0.022 |
Fro | +0.022 |
Neg logit contributions | |
atop | -0.020 |
hole | -0.020 |
pieces | -0.019 |
falling | -0.019 |
edges | -0.018 |
Token | probably |
Token ID | 2192 |
Feature activation | +7.39e+00 |
Pos logit contributions | |
Un | +0.065 |
concession | +0.060 |
please | +0.059 |
Johnny | +0.059 |
Fro | +0.058 |
Neg logit contributions | |
hole | -0.094 |
els | -0.092 |
ps | -0.091 |
atop | -0.090 |
pieces | -0.088 |
Token | too |
Token ID | 1165 |
Feature activation | +6.48e+00 |
Pos logit contributions | |
Fro | +0.078 |
concession | +0.077 |
Un | +0.076 |
Jack | +0.073 |
David | +0.071 |
Neg logit contributions | |
hole | -0.088 |
pieces | -0.084 |
els | -0.082 |
atop | -0.080 |
lid | -0.080 |
Token | , |
Token ID | 11 |
Feature activation | +1.69e+01 |
Pos logit contributions | |
hole | +0.056 |
pieces | +0.056 |
atop | +0.052 |
edges | +0.052 |
ps | +0.052 |
Neg logit contributions | |
Jack | -0.040 |
Un | -0.039 |
Fro | -0.039 |
David | -0.039 |
Jim | -0.037 |
Token | � |
Token ID | 6184 |
Feature activation | +1.28e+01 |
Pos logit contributions | |
Un | +0.176 |
Jack | +0.174 |
David | +0.174 |
Fro | +0.171 |
Kids | +0.166 |
Neg logit contributions | |
pieces | -0.211 |
hole | -0.210 |
on | -0.200 |
atop | -0.196 |
edges | -0.194 |
Token | � |
Token ID | 95 |
Feature activation | +1.24e+01 |
Pos logit contributions | |
Un | +0.133 |
Jack | +0.132 |
David | +0.132 |
Fro | +0.131 |
concession | +0.126 |
Neg logit contributions | |
hole | -0.161 |
pieces | -0.158 |
atop | -0.151 |
on | -0.149 |
edges | -0.147 |
Token | € |
Token ID | 26391 |
Feature activation | +1.83e+01 |
Pos logit contributions | |
Fro | +0.124 |
Un | +0.122 |
Jack | +0.122 |
David | +0.120 |
Jim | +0.116 |
Neg logit contributions | |
hole | -0.161 |
pieces | -0.157 |
atop | -0.149 |
edges | -0.148 |
on | -0.148 |
Token | � |
Token ID | 129 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
concession | +0.172 |
Un | +0.170 |
Da | +0.170 |
Fro | +0.168 |
Jack | +0.168 |
Neg logit contributions | |
hole | -0.246 |
on | -0.231 |
pieces | -0.226 |
atop | -0.221 |
falling | -0.217 |
Token | � |
Token ID | 241 |
Feature activation | +2.04e+01 |
Pos logit contributions | |
Jack | +0.130 |
Un | +0.130 |
David | +0.129 |
Fro | +0.128 |
concession | +0.123 |
Neg logit contributions | |
hole | -0.174 |
pieces | -0.172 |
atop | -0.162 |
on | -0.161 |
edges | -0.161 |
Token | Let |
Token ID | 5756 |
Feature activation | +9.26e+00 |
Pos logit contributions | |
Fro | +0.173 |
oph | +0.172 |
concession | +0.167 |
Un | +0.162 |
Da | +0.162 |
Neg logit contributions | |
hole | -0.286 |
on | -0.286 |
pieces | -0.280 |
edges | -0.270 |
atop | -0.263 |
Token | me |
Token ID | 502 |
Feature activation | +8.32e+00 |
Pos logit contributions | |
Fro | +0.139 |
Un | +0.138 |
David | +0.136 |
concession | +0.135 |
Da | +0.134 |
Neg logit contributions | |
on | -0.078 |
pieces | -0.076 |
way | -0.069 |
hole | -0.068 |
button | -0.066 |
Token | take |
Token ID | 1011 |
Feature activation | +2.29e+00 |
Pos logit contributions | |
Un | +0.090 |
Jack | +0.088 |
David | +0.088 |
Fro | +0.086 |
Kids | +0.084 |
Neg logit contributions | |
hole | -0.102 |
pieces | -0.101 |
on | -0.099 |
atop | -0.095 |
ps | -0.093 |
Token | a |
Token ID | 257 |
Feature activation | +8.18e+00 |
Pos logit contributions | |
Jack | +0.025 |
Un | +0.024 |
Fro | +0.024 |
David | +0.024 |
Jim | +0.023 |
Neg logit contributions | |
hole | -0.029 |
pieces | -0.028 |
atop | -0.027 |
edges | -0.026 |
falling | -0.026 |
Token | look |
Token ID | 804 |
Feature activation | +3.02e+00 |
Pos logit contributions | |
Jack | +0.091 |
Fro | +0.089 |
Un | +0.087 |
enny | +0.086 |
David | +0.086 |
Neg logit contributions | |
ps | -0.100 |
hole | -0.100 |
pieces | -0.096 |
els | -0.095 |
atop | -0.094 |
Token | , |
Token ID | 11 |
Feature activation | +1.69e+01 |
Pos logit contributions | |
pieces | +0.192 |
hole | +0.186 |
ps | +0.184 |
edges | +0.179 |
els | +0.178 |
Neg logit contributions | |
Un | -0.183 |
His | -0.175 |
Jack | -0.174 |
Fro | -0.173 |
David | -0.173 |
Token | � |
Token ID | 6184 |
Feature activation | +1.28e+01 |
Pos logit contributions | |
Un | +0.176 |
Jack | +0.174 |
David | +0.174 |
Fro | +0.171 |
Kids | +0.166 |
Neg logit contributions | |
pieces | -0.211 |
hole | -0.210 |
on | -0.200 |
atop | -0.196 |
edges | -0.194 |
Token | � |
Token ID | 95 |
Feature activation | +1.24e+01 |
Pos logit contributions | |
Un | +0.133 |
Jack | +0.132 |
David | +0.132 |
Fro | +0.131 |
concession | +0.126 |
Neg logit contributions | |
hole | -0.161 |
pieces | -0.158 |
atop | -0.151 |
on | -0.149 |
edges | -0.147 |
Token | € |
Token ID | 26391 |
Feature activation | +1.83e+01 |
Pos logit contributions | |
Fro | +0.124 |
Un | +0.122 |
Jack | +0.122 |
David | +0.120 |
Jim | +0.116 |
Neg logit contributions | |
hole | -0.161 |
pieces | -0.157 |
atop | -0.149 |
edges | -0.148 |
on | -0.148 |
Token | � |
Token ID | 129 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
concession | +0.172 |
Un | +0.170 |
Da | +0.170 |
Fro | +0.168 |
Jack | +0.168 |
Neg logit contributions | |
hole | -0.246 |
on | -0.231 |
pieces | -0.226 |
atop | -0.221 |
falling | -0.217 |
Token | � |
Token ID | 241 |
Feature activation | +2.04e+01 |
Pos logit contributions | |
Jack | +0.130 |
Un | +0.130 |
David | +0.129 |
Fro | +0.128 |
concession | +0.123 |
Neg logit contributions | |
hole | -0.174 |
pieces | -0.172 |
atop | -0.162 |
on | -0.161 |
edges | -0.161 |
Token | I |
Token ID | 40 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
Fro | +0.173 |
oph | +0.172 |
concession | +0.167 |
Un | +0.162 |
Da | +0.162 |
Neg logit contributions | |
hole | -0.286 |
on | -0.286 |
pieces | -0.280 |
edges | -0.270 |
atop | -0.263 |
Token | was |
Token ID | 373 |
Feature activation | +9.44e+00 |
Pos logit contributions | |
Un | +0.124 |
David | +0.123 |
Jack | +0.122 |
Fro | +0.120 |
concession | +0.116 |
Neg logit contributions | |
hole | -0.181 |
pieces | -0.179 |
on | -0.172 |
atop | -0.169 |
edges | -0.166 |
Token | dreaming |
Token ID | 28368 |
Feature activation | +2.79e+00 |
Pos logit contributions | |
Fro | +0.103 |
Un | +0.102 |
David | +0.102 |
Jack | +0.101 |
concession | +0.097 |
Neg logit contributions | |
hole | -0.117 |
pieces | -0.115 |
atop | -0.106 |
lid | -0.106 |
edges | -0.106 |
Token | about |
Token ID | 546 |
Feature activation | +7.09e+00 |
Pos logit contributions | |
Jack | +0.030 |
Un | +0.029 |
Fro | +0.029 |
concession | +0.028 |
David | +0.028 |
Neg logit contributions | |
hole | -0.034 |
pieces | -0.033 |
atop | -0.032 |
els | -0.031 |
edges | -0.031 |
Token | this |
Token ID | 428 |
Feature activation | +4.83e+00 |
Pos logit contributions | |
Un | +0.078 |
Fro | +0.078 |
Jack | +0.078 |
David | +0.077 |
concession | +0.074 |
Neg logit contributions | |
hole | -0.087 |
pieces | -0.082 |
atop | -0.079 |
lid | -0.078 |
ps | -0.077 |
Token | , |
Token ID | 11 |
Feature activation | +1.69e+01 |
Pos logit contributions | |
hole | +0.060 |
pieces | +0.059 |
atop | +0.055 |
ps | +0.054 |
edges | +0.054 |
Neg logit contributions | |
Jack | -0.070 |
David | -0.070 |
Un | -0.070 |
Fro | -0.069 |
Jim | -0.067 |
Token | � |
Token ID | 6184 |
Feature activation | +1.28e+01 |
Pos logit contributions | |
Un | +0.176 |
Jack | +0.174 |
David | +0.174 |
Fro | +0.171 |
Kids | +0.166 |
Neg logit contributions | |
pieces | -0.211 |
hole | -0.210 |
on | -0.200 |
atop | -0.196 |
edges | -0.194 |
Token | � |
Token ID | 95 |
Feature activation | +1.24e+01 |
Pos logit contributions | |
Un | +0.133 |
Jack | +0.132 |
David | +0.132 |
Fro | +0.131 |
concession | +0.126 |
Neg logit contributions | |
hole | -0.161 |
pieces | -0.158 |
atop | -0.151 |
on | -0.149 |
edges | -0.147 |
Token | € |
Token ID | 26391 |
Feature activation | +1.83e+01 |
Pos logit contributions | |
Fro | +0.124 |
Un | +0.122 |
Jack | +0.122 |
David | +0.120 |
Jim | +0.116 |
Neg logit contributions | |
hole | -0.161 |
pieces | -0.157 |
atop | -0.149 |
edges | -0.148 |
on | -0.148 |
Token | � |
Token ID | 129 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
concession | +0.172 |
Un | +0.170 |
Da | +0.170 |
Fro | +0.168 |
Jack | +0.168 |
Neg logit contributions | |
hole | -0.246 |
on | -0.231 |
pieces | -0.226 |
atop | -0.221 |
falling | -0.217 |
Token | � |
Token ID | 241 |
Feature activation | +2.04e+01 |
Pos logit contributions | |
Jack | +0.130 |
Un | +0.130 |
David | +0.129 |
Fro | +0.128 |
concession | +0.123 |
Neg logit contributions | |
hole | -0.174 |
pieces | -0.172 |
atop | -0.162 |
on | -0.161 |
edges | -0.161 |
Token | That |
Token ID | 2504 |
Feature activation | +1.20e+01 |
Pos logit contributions | |
Fro | +0.173 |
oph | +0.172 |
concession | +0.167 |
Un | +0.162 |
Da | +0.162 |
Neg logit contributions | |
hole | -0.286 |
on | -0.286 |
pieces | -0.280 |
edges | -0.270 |
atop | -0.263 |
Token | â |
Token ID | 22940 |
Feature activation | +1.43e+01 |
Pos logit contributions | |
Jack | +0.142 |
Un | +0.142 |
Fro | +0.140 |
David | +0.140 |
concession | +0.136 |
Neg logit contributions | |
hole | -0.133 |
pieces | -0.129 |
atop | -0.125 |
ps | -0.123 |
edges | -0.120 |
Token | € |
Token ID | 26391 |
Feature activation | +1.64e+01 |
Pos logit contributions | |
Jack | +0.090 |
Kids | +0.090 |
David | +0.087 |
Un | +0.084 |
Johnny | +0.082 |
Neg logit contributions | |
pieces | -0.250 |
hole | -0.240 |
edges | -0.231 |
lid | -0.230 |
ps | -0.229 |
Token | ™ |
Token ID | 8151 |
Feature activation | +1.27e+01 |
Pos logit contributions | |
Un | +0.159 |
Jack | +0.157 |
Fro | +0.156 |
concession | +0.156 |
David | +0.154 |
Neg logit contributions | |
hole | -0.212 |
pieces | -0.205 |
on | -0.203 |
atop | -0.194 |
falling | -0.191 |
Token | s |
Token ID | 82 |
Feature activation | +1.13e+01 |
Pos logit contributions | |
Jack | +0.134 |
Fro | +0.132 |
David | +0.131 |
Un | +0.131 |
Jim | +0.126 |
Neg logit contributions | |
pieces | -0.157 |
hole | -0.156 |
atop | -0.147 |
edges | -0.147 |
ps | -0.146 |
Token | , |
Token ID | 11 |
Feature activation | +1.69e+01 |
Pos logit contributions | |
hole | +0.058 |
pieces | +0.057 |
els | +0.054 |
ool | +0.054 |
edges | +0.053 |
Neg logit contributions | |
Un | -0.023 |
Jack | -0.022 |
Fro | -0.021 |
His | -0.021 |
David | -0.021 |
Token | � |
Token ID | 6184 |
Feature activation | +1.28e+01 |
Pos logit contributions | |
Un | +0.176 |
Jack | +0.174 |
David | +0.174 |
Fro | +0.171 |
Kids | +0.166 |
Neg logit contributions | |
pieces | -0.211 |
hole | -0.210 |
on | -0.200 |
atop | -0.196 |
edges | -0.194 |
Token | � |
Token ID | 95 |
Feature activation | +1.24e+01 |
Pos logit contributions | |
Un | +0.133 |
Jack | +0.132 |
David | +0.132 |
Fro | +0.131 |
concession | +0.126 |
Neg logit contributions | |
hole | -0.161 |
pieces | -0.158 |
atop | -0.151 |
on | -0.149 |
edges | -0.147 |
Token | € |
Token ID | 26391 |
Feature activation | +1.83e+01 |
Pos logit contributions | |
Fro | +0.124 |
Un | +0.122 |
Jack | +0.122 |
David | +0.120 |
Jim | +0.116 |
Neg logit contributions | |
hole | -0.161 |
pieces | -0.157 |
atop | -0.149 |
edges | -0.148 |
on | -0.148 |
Token | � |
Token ID | 129 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
concession | +0.172 |
Un | +0.170 |
Da | +0.170 |
Fro | +0.168 |
Jack | +0.168 |
Neg logit contributions | |
hole | -0.246 |
on | -0.231 |
pieces | -0.226 |
atop | -0.221 |
falling | -0.217 |
Token | � |
Token ID | 241 |
Feature activation | +2.04e+01 |
Pos logit contributions | |
Jack | +0.130 |
Un | +0.130 |
David | +0.129 |
Fro | +0.128 |
concession | +0.123 |
Neg logit contributions | |
hole | -0.174 |
pieces | -0.172 |
atop | -0.162 |
on | -0.161 |
edges | -0.161 |
Token | I |
Token ID | 40 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
Fro | +0.173 |
oph | +0.172 |
concession | +0.167 |
Un | +0.162 |
Da | +0.162 |
Neg logit contributions | |
hole | -0.286 |
on | -0.286 |
pieces | -0.280 |
edges | -0.270 |
atop | -0.263 |
Token | can |
Token ID | 460 |
Feature activation | +1.05e+01 |
Pos logit contributions | |
Un | +0.124 |
David | +0.123 |
Jack | +0.122 |
Fro | +0.120 |
concession | +0.116 |
Neg logit contributions | |
hole | -0.181 |
pieces | -0.179 |
on | -0.172 |
atop | -0.169 |
edges | -0.166 |
Token | fly |
Token ID | 6129 |
Feature activation | -4.31e-01 |
Pos logit contributions | |
Un | +0.131 |
Fro | +0.128 |
David | +0.127 |
Da | +0.127 |
Jack | +0.125 |
Neg logit contributions | |
hole | -0.114 |
pieces | -0.112 |
on | -0.109 |
falling | -0.103 |
edges | -0.103 |
Token | , |
Token ID | 11 |
Feature activation | +9.50e+00 |
Pos logit contributions | |
pieces | +0.004 |
hole | +0.004 |
on | +0.004 |
atop | +0.004 |
edges | +0.004 |
Neg logit contributions | |
David | -0.006 |
Un | -0.006 |
Jack | -0.006 |
Fro | -0.006 |
Kids | -0.006 |
Token | I |
Token ID | 314 |
Feature activation | +8.92e+00 |
Pos logit contributions | |
Un | +0.104 |
Jack | +0.103 |
David | +0.102 |
Da | +0.100 |
Jim | +0.099 |
Neg logit contributions | |
on | -0.113 |
pieces | -0.112 |
hole | -0.111 |
falling | -0.102 |
edges | -0.101 |
Token | , |
Token ID | 11 |
Feature activation | +1.69e+01 |
Pos logit contributions | |
hole | +0.128 |
pieces | +0.125 |
atop | +0.123 |
els | +0.122 |
ps | +0.120 |
Neg logit contributions | |
David | -0.088 |
Un | -0.087 |
Jack | -0.085 |
Fro | -0.084 |
His | -0.083 |
Token | � |
Token ID | 6184 |
Feature activation | +1.28e+01 |
Pos logit contributions | |
Un | +0.176 |
Jack | +0.174 |
David | +0.174 |
Fro | +0.171 |
Kids | +0.166 |
Neg logit contributions | |
pieces | -0.211 |
hole | -0.210 |
on | -0.200 |
atop | -0.196 |
edges | -0.194 |
Token | � |
Token ID | 95 |
Feature activation | +1.24e+01 |
Pos logit contributions | |
Un | +0.133 |
Jack | +0.132 |
David | +0.132 |
Fro | +0.131 |
concession | +0.126 |
Neg logit contributions | |
hole | -0.161 |
pieces | -0.158 |
atop | -0.151 |
on | -0.149 |
edges | -0.147 |
Token | € |
Token ID | 26391 |
Feature activation | +1.83e+01 |
Pos logit contributions | |
Fro | +0.124 |
Un | +0.122 |
Jack | +0.122 |
David | +0.120 |
Jim | +0.116 |
Neg logit contributions | |
hole | -0.161 |
pieces | -0.157 |
atop | -0.149 |
edges | -0.148 |
on | -0.148 |
Token | � |
Token ID | 129 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
concession | +0.172 |
Un | +0.170 |
Da | +0.170 |
Fro | +0.168 |
Jack | +0.168 |
Neg logit contributions | |
hole | -0.246 |
on | -0.231 |
pieces | -0.226 |
atop | -0.221 |
falling | -0.217 |
Token | � |
Token ID | 241 |
Feature activation | +2.04e+01 |
Pos logit contributions | |
Jack | +0.130 |
Un | +0.130 |
David | +0.129 |
Fro | +0.128 |
concession | +0.123 |
Neg logit contributions | |
hole | -0.174 |
pieces | -0.172 |
atop | -0.162 |
on | -0.161 |
edges | -0.161 |
Token | That |
Token ID | 2504 |
Feature activation | +1.20e+01 |
Pos logit contributions | |
Fro | +0.173 |
oph | +0.172 |
concession | +0.167 |
Un | +0.162 |
Da | +0.162 |
Neg logit contributions | |
hole | -0.286 |
on | -0.286 |
pieces | -0.280 |
edges | -0.270 |
atop | -0.263 |
Token | â |
Token ID | 22940 |
Feature activation | +1.43e+01 |
Pos logit contributions | |
Jack | +0.142 |
Un | +0.142 |
Fro | +0.140 |
David | +0.140 |
concession | +0.136 |
Neg logit contributions | |
hole | -0.133 |
pieces | -0.129 |
atop | -0.125 |
ps | -0.123 |
edges | -0.120 |
Token | € |
Token ID | 26391 |
Feature activation | +1.64e+01 |
Pos logit contributions | |
Jack | +0.090 |
Kids | +0.090 |
David | +0.087 |
Un | +0.084 |
Johnny | +0.082 |
Neg logit contributions | |
pieces | -0.250 |
hole | -0.240 |
edges | -0.231 |
lid | -0.230 |
ps | -0.229 |
Token | ™ |
Token ID | 8151 |
Feature activation | +1.27e+01 |
Pos logit contributions | |
Un | +0.159 |
Jack | +0.157 |
Fro | +0.156 |
concession | +0.156 |
David | +0.154 |
Neg logit contributions | |
hole | -0.212 |
pieces | -0.205 |
on | -0.203 |
atop | -0.194 |
falling | -0.191 |
Token | s |
Token ID | 82 |
Feature activation | +1.13e+01 |
Pos logit contributions | |
Jack | +0.134 |
Fro | +0.132 |
David | +0.131 |
Un | +0.131 |
Jim | +0.126 |
Neg logit contributions | |
pieces | -0.157 |
hole | -0.156 |
atop | -0.147 |
edges | -0.147 |
ps | -0.146 |
Token | , |
Token ID | 11 |
Feature activation | +1.69e+01 |
Pos logit contributions | |
Un | +0.030 |
Fro | +0.029 |
Jack | +0.029 |
David | +0.029 |
Da | +0.029 |
Neg logit contributions | |
pieces | -0.058 |
hole | -0.057 |
on | -0.055 |
edges | -0.053 |
atop | -0.053 |
Token | � |
Token ID | 6184 |
Feature activation | +1.28e+01 |
Pos logit contributions | |
Un | +0.176 |
Jack | +0.174 |
David | +0.174 |
Fro | +0.171 |
Kids | +0.166 |
Neg logit contributions | |
pieces | -0.211 |
hole | -0.210 |
on | -0.200 |
atop | -0.196 |
edges | -0.194 |
Token | � |
Token ID | 95 |
Feature activation | +1.24e+01 |
Pos logit contributions | |
Un | +0.133 |
Jack | +0.132 |
David | +0.132 |
Fro | +0.131 |
concession | +0.126 |
Neg logit contributions | |
hole | -0.161 |
pieces | -0.158 |
atop | -0.151 |
on | -0.149 |
edges | -0.147 |
Token | € |
Token ID | 26391 |
Feature activation | +1.83e+01 |
Pos logit contributions | |
Fro | +0.124 |
Un | +0.122 |
Jack | +0.122 |
David | +0.120 |
Jim | +0.116 |
Neg logit contributions | |
hole | -0.161 |
pieces | -0.157 |
atop | -0.149 |
edges | -0.148 |
on | -0.148 |
Token | � |
Token ID | 129 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
concession | +0.172 |
Un | +0.170 |
Da | +0.170 |
Fro | +0.168 |
Jack | +0.168 |
Neg logit contributions | |
hole | -0.246 |
on | -0.231 |
pieces | -0.226 |
atop | -0.221 |
falling | -0.217 |
Token | � |
Token ID | 241 |
Feature activation | +2.04e+01 |
Pos logit contributions | |
Jack | +0.130 |
Un | +0.130 |
David | +0.129 |
Fro | +0.128 |
concession | +0.123 |
Neg logit contributions | |
hole | -0.174 |
pieces | -0.172 |
atop | -0.162 |
on | -0.161 |
edges | -0.161 |
Token | That |
Token ID | 2504 |
Feature activation | +1.20e+01 |
Pos logit contributions | |
Fro | +0.173 |
oph | +0.172 |
concession | +0.167 |
Un | +0.162 |
Da | +0.162 |
Neg logit contributions | |
hole | -0.286 |
on | -0.286 |
pieces | -0.280 |
edges | -0.270 |
atop | -0.263 |
Token | 's |
Token ID | 338 |
Feature activation | +1.15e+01 |
Pos logit contributions | |
Jack | +0.142 |
Un | +0.142 |
Fro | +0.140 |
David | +0.140 |
concession | +0.136 |
Neg logit contributions | |
hole | -0.133 |
pieces | -0.129 |
atop | -0.125 |
ps | -0.123 |
edges | -0.120 |
Token | the |
Token ID | 262 |
Feature activation | +9.42e+00 |
Pos logit contributions | |
Un | +0.118 |
Jack | +0.114 |
Fro | +0.114 |
David | +0.112 |
Johnny | +0.110 |
Neg logit contributions | |
pieces | -0.152 |
hole | -0.151 |
atop | -0.142 |
ps | -0.137 |
edges | -0.136 |
Token | most |
Token ID | 749 |
Feature activation | +7.37e+00 |
Pos logit contributions | |
Un | +0.119 |
David | +0.116 |
Fro | +0.115 |
concession | +0.115 |
Jack | +0.114 |
Neg logit contributions | |
hole | -0.103 |
atop | -0.091 |
pieces | -0.088 |
ps | -0.088 |
els | -0.088 |
Token | important |
Token ID | 1593 |
Feature activation | +7.11e+00 |
Pos logit contributions | |
Jack | +0.084 |
Johnny | +0.080 |
concession | +0.079 |
Un | +0.079 |
Kids | +0.079 |
Neg logit contributions | |
hole | -0.090 |
ps | -0.086 |
pieces | -0.083 |
atop | -0.082 |
lid | -0.080 |
Token | saw |
Token ID | 2497 |
Feature activation | -6.85e+00 |
Pos logit contributions | |
pieces | +0.075 |
hole | +0.074 |
on | +0.070 |
atop | +0.068 |
falling | +0.067 |
Neg logit contributions | |
Un | -0.075 |
David | -0.074 |
Jack | -0.074 |
Fro | -0.073 |
Jim | -0.071 |
Token | some |
Token ID | 617 |
Feature activation | -1.06e+01 |
Pos logit contributions | |
pieces | +0.088 |
hole | +0.086 |
on | +0.084 |
atop | +0.082 |
ps | +0.082 |
Neg logit contributions | |
David | -0.071 |
Jack | -0.070 |
Un | -0.069 |
Kids | -0.066 |
Jim | -0.066 |
Token | more |
Token ID | 517 |
Feature activation | -1.05e+01 |
Pos logit contributions | |
hole | +0.126 |
atop | +0.111 |
ps | +0.106 |
globe | +0.105 |
edges | +0.104 |
Neg logit contributions | |
Fro | -0.124 |
Jack | -0.124 |
concession | -0.123 |
Un | -0.119 |
David | -0.117 |
Token | che |
Token ID | 1125 |
Feature activation | -1.47e+01 |
Pos logit contributions | |
hole | +0.105 |
atop | +0.100 |
ps | +0.097 |
hardest | +0.093 |
lid | +0.088 |
Neg logit contributions | |
Jack | -0.136 |
Fro | -0.132 |
enny | -0.130 |
concession | -0.127 |
Johnny | -0.127 |
Token | wed |
Token ID | 19103 |
Feature activation | -1.66e+01 |
Pos logit contributions | |
hole | +0.146 |
button | +0.139 |
ladder | +0.139 |
changing | +0.138 |
lid | +0.137 |
Neg logit contributions | |
Fro | -0.177 |
Jack | -0.176 |
ackers | -0.175 |
enny | -0.174 |
Jim | -0.170 |
Token | threads |
Token ID | 14390 |
Feature activation | -2.12e+01 |
Pos logit contributions | |
hole | +0.184 |
code | +0.177 |
mant | +0.161 |
slot | +0.161 |
atop | +0.160 |
Neg logit contributions | |
Fro | -0.201 |
him | -0.187 |
Jack | -0.183 |
He | -0.178 |
He | -0.178 |
Token | in |
Token ID | 287 |
Feature activation | -7.98e+00 |
Pos logit contributions | |
edges | +0.239 |
ps | +0.236 |
hole | +0.235 |
lid | +0.235 |
els | +0.232 |
Neg logit contributions | |
Jack | -0.206 |
Fro | -0.199 |
enny | -0.195 |
He | -0.190 |
They | -0.189 |
Token | the |
Token ID | 262 |
Feature activation | -5.57e+00 |
Pos logit contributions | |
hole | +0.117 |
atop | +0.111 |
ool | +0.105 |
hardest | +0.105 |
edges | +0.103 |
Neg logit contributions | |
Fro | -0.071 |
Un | -0.070 |
Jack | -0.065 |
His | -0.064 |
Da | -0.062 |
Token | house |
Token ID | 2156 |
Feature activation | -1.61e+01 |
Pos logit contributions | |
ps | +0.056 |
hole | +0.056 |
cks | +0.055 |
els | +0.054 |
atop | +0.054 |
Neg logit contributions | |
Un | -0.070 |
Jack | -0.070 |
Da | -0.068 |
Fro | -0.067 |
Brother | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | +3.52e+00 |
Pos logit contributions | |
atop | +0.172 |
hole | +0.169 |
ps | +0.167 |
els | +0.167 |
edges | +0.166 |
Neg logit contributions | |
Un | -0.172 |
Jack | -0.172 |
concession | -0.168 |
Fro | -0.167 |
David | -0.164 |
Token | Fl |
Token ID | 1610 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
David | +0.026 |
Jack | +0.024 |
Un | +0.024 |
Jim | +0.023 |
Johnny | +0.022 |
Neg logit contributions | |
pieces | -0.057 |
hole | -0.056 |
on | -0.055 |
edges | -0.055 |
lid | -0.054 |
Token | she |
Token ID | 673 |
Feature activation | -4.87e+00 |
Pos logit contributions | |
hole | +0.035 |
pieces | +0.034 |
edges | +0.032 |
ps | +0.032 |
lid | +0.032 |
Neg logit contributions | |
Jack | -0.022 |
Un | -0.022 |
Fro | -0.022 |
David | -0.022 |
Johnny | -0.021 |
Token | continued |
Token ID | 3767 |
Feature activation | -7.59e+00 |
Pos logit contributions | |
hole | +0.070 |
pieces | +0.067 |
edges | +0.066 |
ps | +0.065 |
atop | +0.065 |
Neg logit contributions | |
Jack | -0.043 |
Un | -0.042 |
Fro | -0.042 |
Kids | -0.040 |
David | -0.040 |
Token | picking |
Token ID | 10868 |
Feature activation | -1.54e+01 |
Pos logit contributions | |
hole | +0.105 |
pieces | +0.094 |
edges | +0.093 |
lid | +0.093 |
wires | +0.088 |
Neg logit contributions | |
Fro | -0.077 |
Un | -0.075 |
His | -0.071 |
David | -0.070 |
His | -0.069 |
Token | her |
Token ID | 607 |
Feature activation | -8.34e+00 |
Pos logit contributions | |
hole | +0.182 |
code | +0.162 |
falling | +0.147 |
edges | +0.145 |
ool | +0.145 |
Neg logit contributions | |
Fro | -0.200 |
Da | -0.175 |
Pete | -0.166 |
Un | -0.166 |
Jack | -0.165 |
Token | p |
Token ID | 279 |
Feature activation | -1.15e+01 |
Pos logit contributions | |
hole | +0.123 |
slot | +0.108 |
code | +0.103 |
bones | +0.101 |
atop | +0.100 |
Neg logit contributions | |
Fro | -0.086 |
Her | -0.070 |
His | -0.070 |
omo | -0.069 |
concession | -0.069 |
Token | ears |
Token ID | 4127 |
Feature activation | -2.12e+01 |
Pos logit contributions | |
hole | +0.125 |
pieces | +0.124 |
atop | +0.115 |
on | +0.114 |
edges | +0.113 |
Neg logit contributions | |
Un | -0.139 |
Fro | -0.138 |
Jack | -0.138 |
David | -0.138 |
concession | -0.133 |
Token | . |
Token ID | 13 |
Feature activation | +4.48e+00 |
Pos logit contributions | |
hole | +0.202 |
ps | +0.201 |
edges | +0.197 |
pieces | +0.197 |
lid | +0.190 |
Neg logit contributions | |
Fro | -0.261 |
Un | -0.258 |
Jack | -0.252 |
His | -0.244 |
enny | -0.240 |
Token | With |
Token ID | 2080 |
Feature activation | -4.73e+00 |
Pos logit contributions | |
concession | +0.032 |
Jim | +0.032 |
Un | +0.032 |
David | +0.032 |
Jack | +0.031 |
Neg logit contributions | |
hole | -0.068 |
on | -0.067 |
pieces | -0.067 |
edges | -0.064 |
atop | -0.064 |
Token | a |
Token ID | 257 |
Feature activation | -3.79e+00 |
Pos logit contributions | |
pieces | +0.056 |
hole | +0.054 |
on | +0.053 |
atop | +0.051 |
falling | +0.051 |
Neg logit contributions | |
Jack | -0.053 |
David | -0.053 |
Un | -0.052 |
Kids | -0.051 |
concession | -0.051 |
Token | smile |
Token ID | 8212 |
Feature activation | -1.27e+01 |
Pos logit contributions | |
hole | +0.043 |
pieces | +0.042 |
ps | +0.040 |
edges | +0.040 |
atop | +0.039 |
Neg logit contributions | |
Un | -0.045 |
Jack | -0.045 |
David | -0.044 |
Fro | -0.044 |
Jim | -0.042 |
Token | on |
Token ID | 319 |
Feature activation | -7.73e+00 |
Pos logit contributions | |
pieces | +0.098 |
hole | +0.095 |
edges | +0.093 |
ps | +0.084 |
wires | +0.084 |
Neg logit contributions | |
Un | -0.197 |
Fro | -0.193 |
His | -0.190 |
David | -0.189 |
Jack | -0.186 |
Token | , |
Token ID | 11 |
Feature activation | -8.42e-01 |
Pos logit contributions | |
hole | +0.106 |
edges | +0.102 |
pieces | +0.101 |
atop | +0.101 |
ool | +0.100 |
Neg logit contributions | |
please | -0.058 |
Fro | -0.051 |
Max | -0.051 |
Ben | -0.050 |
Un | -0.049 |
Token | do |
Token ID | 466 |
Feature activation | -5.66e+00 |
Pos logit contributions | |
pieces | +0.011 |
hole | +0.011 |
on | +0.010 |
edges | +0.010 |
atop | +0.010 |
Neg logit contributions | |
David | -0.009 |
Jack | -0.009 |
Un | -0.009 |
Fro | -0.008 |
concession | -0.008 |
Token | not |
Token ID | 407 |
Feature activation | -5.09e+00 |
Pos logit contributions | |
hole | +0.082 |
pieces | +0.077 |
atop | +0.075 |
edges | +0.072 |
ps | +0.072 |
Neg logit contributions | |
Un | -0.053 |
Fro | -0.053 |
Jack | -0.050 |
David | -0.049 |
Jim | -0.048 |
Token | touch |
Token ID | 3638 |
Feature activation | -1.36e+01 |
Pos logit contributions | |
hole | +0.069 |
pieces | +0.067 |
atop | +0.065 |
ps | +0.061 |
edges | +0.061 |
Neg logit contributions | |
Fro | -0.052 |
Un | -0.051 |
Jack | -0.049 |
David | -0.046 |
His | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | -7.55e+00 |
Pos logit contributions | |
hole | +0.183 |
atop | +0.176 |
code | +0.172 |
globe | +0.166 |
lid | +0.163 |
Neg logit contributions | |
Fro | -0.141 |
Mitt | -0.129 |
Un | -0.116 |
Johnny | -0.116 |
him | -0.115 |
Token | pins |
Token ID | 20567 |
Feature activation | -2.11e+01 |
Pos logit contributions | |
hole | +0.085 |
atop | +0.084 |
ool | +0.073 |
existed | +0.072 |
underneath | +0.072 |
Neg logit contributions | |
Fro | -0.088 |
Ben | -0.087 |
Kids | -0.082 |
Sam | -0.079 |
L | -0.079 |
Token | . |
Token ID | 13 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
hole | +0.227 |
bones | +0.224 |
lid | +0.221 |
atop | +0.219 |
edges | +0.219 |
Neg logit contributions | |
please | -0.228 |
They | -0.214 |
Fro | -0.205 |
He | -0.204 |
Jack | -0.204 |
Token | They |
Token ID | 1119 |
Feature activation | -5.12e+00 |
Pos logit contributions | |
Jack | +0.013 |
David | +0.013 |
concession | +0.013 |
Un | +0.013 |
oph | +0.013 |
Neg logit contributions | |
hole | -0.021 |
pieces | -0.020 |
on | -0.020 |
edges | -0.019 |
ps | -0.019 |
Token | are |
Token ID | 389 |
Feature activation | -6.34e+00 |
Pos logit contributions | |
hole | +0.074 |
atop | +0.074 |
bones | +0.068 |
pieces | +0.067 |
ool | +0.067 |
Neg logit contributions | |
Fro | -0.039 |
please | -0.037 |
They | -0.034 |
him | -0.034 |
He | -0.033 |
Token | not |
Token ID | 407 |
Feature activation | -7.60e+00 |
Pos logit contributions | |
hole | +0.093 |
atop | +0.084 |
globe | +0.083 |
angle | +0.082 |
lid | +0.081 |
Neg logit contributions | |
Fro | -0.058 |
Mitt | -0.053 |
Un | -0.053 |
wife | -0.050 |
uncle | -0.048 |
Token | toys |
Token ID | 14958 |
Feature activation | -1.70e+01 |
Pos logit contributions | |
hole | +0.096 |
lid | +0.093 |
atop | +0.092 |
globe | +0.090 |
els | +0.085 |
Neg logit contributions | |
Fro | -0.082 |
Un | -0.076 |
wife | -0.073 |
Mitt | -0.071 |
Da | -0.068 |
Token | " |
Token ID | 1 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
Fro | +0.029 |
enny | +0.024 |
ascal | +0.022 |
shepherd | +0.022 |
ory | +0.022 |
Neg logit contributions | |
hole | -0.046 |
on | -0.044 |
pieces | -0.044 |
atop | -0.042 |
edges | -0.042 |
Token | Let |
Token ID | 5756 |
Feature activation | -6.24e+00 |
Pos logit contributions | |
Un | +0.016 |
Jack | +0.016 |
David | +0.016 |
Fro | +0.016 |
Jim | +0.015 |
Neg logit contributions | |
hole | -0.029 |
pieces | -0.028 |
atop | -0.027 |
on | -0.027 |
edges | -0.027 |
Token | 's |
Token ID | 338 |
Feature activation | -4.40e+00 |
Pos logit contributions | |
on | +0.050 |
pieces | +0.049 |
way | +0.044 |
button | +0.042 |
hole | +0.041 |
Neg logit contributions | |
Fro | -0.095 |
enny | -0.094 |
concession | -0.094 |
Un | -0.092 |
coach | -0.092 |
Token | replace |
Token ID | 6330 |
Feature activation | -1.47e+01 |
Pos logit contributions | |
pieces | +0.060 |
hole | +0.060 |
atop | +0.056 |
edges | +0.056 |
ps | +0.055 |
Neg logit contributions | |
Un | -0.041 |
Jack | -0.040 |
David | -0.040 |
Fro | -0.040 |
Jim | -0.038 |
Token | the |
Token ID | 262 |
Feature activation | -7.76e+00 |
Pos logit contributions | |
hole | +0.196 |
atop | +0.182 |
falling | +0.179 |
lid | +0.178 |
ool | +0.177 |
Neg logit contributions | |
Fro | -0.146 |
Jack | -0.132 |
Un | -0.131 |
David | -0.128 |
concession | -0.126 |
Token | rocks |
Token ID | 12586 |
Feature activation | -2.11e+01 |
Pos logit contributions | |
hole | +0.101 |
bones | +0.100 |
ool | +0.100 |
els | +0.100 |
ps | +0.097 |
Neg logit contributions | |
fisherman | -0.074 |
wife | -0.068 |
coach | -0.062 |
dolphin | -0.059 |
I | -0.058 |
Token | at |
Token ID | 379 |
Feature activation | -8.53e+00 |
Pos logit contributions | |
ps | +0.218 |
falling | +0.213 |
edges | +0.207 |
lid | +0.207 |
hole | +0.205 |
Neg logit contributions | |
Jack | -0.222 |
They | -0.218 |
He | -0.212 |
concession | -0.210 |
Fro | -0.209 |
Token | the |
Token ID | 262 |
Feature activation | -6.80e+00 |
Pos logit contributions | |
lid | +0.121 |
ps | +0.117 |
hole | +0.117 |
splits | +0.114 |
tangled | +0.111 |
Neg logit contributions | |
Fro | -0.071 |
Pete | -0.064 |
Sammy | -0.061 |
Dad | -0.061 |
Un | -0.060 |
Token | bottom |
Token ID | 4220 |
Feature activation | -1.58e+01 |
Pos logit contributions | |
hole | +0.086 |
els | +0.086 |
ps | +0.084 |
ool | +0.081 |
cks | +0.076 |
Neg logit contributions | |
fisherman | -0.070 |
Jack | -0.066 |
coach | -0.066 |
I | -0.065 |
concession | -0.065 |
Token | of |
Token ID | 286 |
Feature activation | -7.77e+00 |
Pos logit contributions | |
ps | +0.172 |
bones | +0.152 |
falling | +0.152 |
atop | +0.147 |
tangled | +0.147 |
Neg logit contributions | |
Jack | -0.195 |
Jim | -0.184 |
Un | -0.181 |
David | -0.178 |
Fro | -0.177 |
Token | the |
Token ID | 262 |
Feature activation | -6.76e+00 |
Pos logit contributions | |
hole | +0.116 |
atop | +0.104 |
ool | +0.103 |
splits | +0.101 |
lid | +0.101 |
Neg logit contributions | |
Fro | -0.073 |
Un | -0.066 |
David | -0.062 |
Jack | -0.062 |
Pete | -0.062 |
Token | and |
Token ID | 290 |
Feature activation | -3.34e+00 |
Pos logit contributions | |
ps | +0.235 |
els | +0.231 |
edges | +0.227 |
problems | +0.227 |
bones | +0.224 |
Neg logit contributions | |
His | -0.097 |
Jack | -0.097 |
Fro | -0.097 |
He | -0.096 |
His | -0.095 |
Token | found |
Token ID | 1043 |
Feature activation | -8.51e+00 |
Pos logit contributions | |
hole | +0.040 |
pieces | +0.039 |
ps | +0.037 |
edges | +0.037 |
falling | +0.037 |
Neg logit contributions | |
Jack | -0.038 |
Fro | -0.036 |
Un | -0.036 |
David | -0.036 |
Jim | -0.035 |
Token | lots |
Token ID | 6041 |
Feature activation | -1.23e+01 |
Pos logit contributions | |
hole | +0.122 |
falling | +0.119 |
changing | +0.112 |
edges | +0.112 |
ool | +0.110 |
Neg logit contributions | |
Fro | -0.074 |
Jack | -0.071 |
His | -0.068 |
David | -0.066 |
him | -0.064 |
Token | of |
Token ID | 286 |
Feature activation | -8.86e+00 |
Pos logit contributions | |
hole | +0.184 |
ps | +0.182 |
hardest | +0.182 |
atop | +0.177 |
falling | +0.174 |
Neg logit contributions | |
Jack | -0.092 |
concession | -0.090 |
Fro | -0.086 |
enny | -0.082 |
He | -0.081 |
Token | gold |
Token ID | 3869 |
Feature activation | -2.06e+01 |
Pos logit contributions | |
hole | +0.090 |
ps | +0.084 |
atop | +0.083 |
falling | +0.081 |
hardest | +0.080 |
Neg logit contributions | |
Jack | -0.113 |
concession | -0.109 |
Fro | -0.107 |
David | -0.106 |
He | -0.104 |
Token | coins |
Token ID | 10796 |
Feature activation | -2.09e+01 |
Pos logit contributions | |
hardest | +0.170 |
angle | +0.165 |
falling | +0.162 |
changing | +0.162 |
hole | +0.162 |
Neg logit contributions | |
He | -0.257 |
ory | -0.249 |
him | -0.247 |
Fro | -0.247 |
He | -0.243 |
Token | and |
Token ID | 290 |
Feature activation | -5.05e+00 |
Pos logit contributions | |
splits | +0.210 |
edges | +0.206 |
hardest | +0.206 |
connects | +0.204 |
ps | +0.202 |
Neg logit contributions | |
Fro | -0.221 |
ory | -0.219 |
David | -0.216 |
ete | -0.209 |
He | -0.209 |
Token | shiny |
Token ID | 22441 |
Feature activation | -1.54e+01 |
Pos logit contributions | |
hole | +0.039 |
atop | +0.035 |
ps | +0.033 |
falling | +0.033 |
ool | +0.031 |
Neg logit contributions | |
Jack | -0.076 |
Fro | -0.076 |
Un | -0.075 |
David | -0.074 |
concession | -0.073 |
Token | jewels |
Token ID | 42701 |
Feature activation | -1.84e+01 |
Pos logit contributions | |
hole | +0.134 |
falling | +0.132 |
hardest | +0.131 |
socket | +0.121 |
changing | +0.120 |
Neg logit contributions | |
him | -0.188 |
ory | -0.187 |
He | -0.186 |
He | -0.184 |
His | -0.181 |
Token | . |
Token ID | 13 |
Feature activation | +4.32e+00 |
Pos logit contributions | |
hardest | +0.196 |
edges | +0.191 |
els | +0.188 |
ps | +0.185 |
wires | +0.185 |
Neg logit contributions | |
Fro | -0.182 |
He | -0.179 |
enny | -0.179 |
Jack | -0.178 |
ory | -0.175 |
Token |
|
Token ID | 198 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
Un | +0.034 |
David | +0.031 |
Fro | +0.031 |
concession | +0.031 |
Jim | +0.030 |
Neg logit contributions | |
pieces | -0.067 |
hole | -0.066 |
on | -0.064 |
edges | -0.062 |
atop | -0.061 |
Token | to |
Token ID | 284 |
Feature activation | -5.22e+00 |
Pos logit contributions | |
hole | +0.060 |
pieces | +0.058 |
atop | +0.055 |
ps | +0.054 |
edges | +0.054 |
Neg logit contributions | |
Jack | -0.044 |
Un | -0.043 |
Fro | -0.043 |
David | -0.043 |
Kids | -0.041 |
Token | spray |
Token ID | 11662 |
Feature activation | -1.88e+01 |
Pos logit contributions | |
hole | +0.062 |
pieces | +0.059 |
atop | +0.058 |
hardest | +0.053 |
globe | +0.052 |
Neg logit contributions | |
Fro | -0.060 |
Jack | -0.057 |
Kids | -0.057 |
Un | -0.056 |
Mitt | -0.054 |
Token | some |
Token ID | 617 |
Feature activation | -1.19e+01 |
Pos logit contributions | |
hole | +0.291 |
code | +0.267 |
slot | +0.254 |
plank | +0.248 |
atop | +0.246 |
Neg logit contributions | |
Fro | -0.172 |
Mitt | -0.170 |
Sammy | -0.153 |
They | -0.150 |
Sandy | -0.141 |
Token | nice |
Token ID | 3621 |
Feature activation | -1.27e+01 |
Pos logit contributions | |
hole | +0.167 |
globe | +0.152 |
atop | +0.147 |
bones | +0.144 |
code | +0.144 |
Neg logit contributions | |
Mitt | -0.113 |
Fro | -0.112 |
They | -0.109 |
Kids | -0.106 |
Ben | -0.101 |
Token | smelling |
Token ID | 48090 |
Feature activation | -1.60e+01 |
Pos logit contributions | |
hole | +0.150 |
atop | +0.139 |
code | +0.134 |
slot | +0.130 |
hardest | +0.128 |
Neg logit contributions | |
Fro | -0.149 |
Kids | -0.142 |
Da | -0.137 |
Mitt | -0.136 |
Jack | -0.133 |
Token | liquid |
Token ID | 8122 |
Feature activation | -2.09e+01 |
Pos logit contributions | |
hole | +0.234 |
atop | +0.200 |
code | +0.200 |
hardest | +0.192 |
eking | +0.187 |
Neg logit contributions | |
Fro | -0.160 |
Mitt | -0.157 |
Kids | -0.145 |
They | -0.140 |
Da | -0.140 |
Token | around |
Token ID | 1088 |
Feature activation | -1.18e+01 |
Pos logit contributions | |
hole | +0.274 |
slot | +0.246 |
plank | +0.235 |
code | +0.235 |
angle | +0.234 |
Neg logit contributions | |
They | -0.211 |
Fro | -0.210 |
Pete | -0.190 |
Mitt | -0.188 |
Un | -0.184 |
Token | the |
Token ID | 262 |
Feature activation | -6.64e+00 |
Pos logit contributions | |
hole | +0.182 |
code | +0.161 |
hardest | +0.159 |
atop | +0.159 |
ps | +0.157 |
Neg logit contributions | |
Jack | -0.101 |
Fro | -0.097 |
Da | -0.096 |
Mitt | -0.091 |
Un | -0.090 |
Token | garden |
Token ID | 11376 |
Feature activation | -1.58e+01 |
Pos logit contributions | |
hole | +0.080 |
ps | +0.073 |
atop | +0.073 |
bones | +0.073 |
code | +0.071 |
Neg logit contributions | |
Ben | -0.071 |
Mitt | -0.071 |
Fro | -0.070 |
Kids | -0.070 |
They | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | +3.91e+00 |
Pos logit contributions | |
hole | +0.188 |
atop | +0.181 |
ps | +0.176 |
hardest | +0.171 |
bones | +0.171 |
Neg logit contributions | |
Fro | -0.158 |
They | -0.148 |
Un | -0.143 |
Da | -0.142 |
Mitt | -0.138 |
Token | Sarah |
Token ID | 10490 |
Feature activation | -4.16e+00 |
Pos logit contributions | |
Fro | +0.032 |
Un | +0.032 |
concession | +0.031 |
David | +0.030 |
Jack | +0.030 |
Neg logit contributions | |
hole | -0.057 |
pieces | -0.057 |
on | -0.055 |
edges | -0.053 |
atop | -0.053 |
Token | continued |
Token ID | 3767 |
Feature activation | -8.16e+00 |
Pos logit contributions | |
hole | +0.121 |
pieces | +0.117 |
atop | +0.111 |
edges | +0.111 |
lid | +0.110 |
Neg logit contributions | |
Jack | -0.091 |
Un | -0.090 |
Fro | -0.090 |
David | -0.088 |
Kids | -0.085 |
Token | to |
Token ID | 284 |
Feature activation | -5.24e+00 |
Pos logit contributions | |
hole | +0.128 |
pieces | +0.116 |
lid | +0.113 |
edges | +0.112 |
ps | +0.112 |
Neg logit contributions | |
Fro | -0.068 |
Un | -0.064 |
Jack | -0.063 |
David | -0.060 |
Johnny | -0.059 |
Token | make |
Token ID | 787 |
Feature activation | -1.16e+01 |
Pos logit contributions | |
hole | +0.072 |
pieces | +0.072 |
atop | +0.067 |
edges | +0.067 |
on | +0.067 |
Neg logit contributions | |
Jack | -0.048 |
Un | -0.048 |
David | -0.048 |
Fro | -0.048 |
Kids | -0.045 |
Token | fancy |
Token ID | 14996 |
Feature activation | -1.46e+01 |
Pos logit contributions | |
pieces | +0.109 |
hole | +0.097 |
on | +0.096 |
edges | +0.093 |
falling | +0.090 |
Neg logit contributions | |
Un | -0.168 |
David | -0.164 |
Jack | -0.162 |
Kids | -0.160 |
concession | -0.160 |
Token | mud |
Token ID | 17492 |
Feature activation | -1.92e+01 |
Pos logit contributions | |
hole | +0.127 |
pieces | +0.118 |
atop | +0.113 |
ps | +0.107 |
on | +0.106 |
Neg logit contributions | |
Fro | -0.209 |
Jack | -0.209 |
Un | -0.206 |
David | -0.206 |
concession | -0.201 |
Token | designs |
Token ID | 9824 |
Feature activation | -2.09e+01 |
Pos logit contributions | |
hole | +0.179 |
pieces | +0.170 |
atop | +0.160 |
edges | +0.155 |
ps | +0.150 |
Neg logit contributions | |
Fro | -0.263 |
Jack | -0.260 |
Un | -0.260 |
David | -0.255 |
concession | -0.251 |
Token | . |
Token ID | 13 |
Feature activation | +4.59e+00 |
Pos logit contributions | |
hole | +0.197 |
pieces | +0.174 |
edges | +0.173 |
atop | +0.173 |
ps | +0.171 |
Neg logit contributions | |
Jack | -0.276 |
Un | -0.266 |
Fro | -0.265 |
David | -0.265 |
Johnny | -0.259 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.38e+00 |
Pos logit contributions | |
David | +0.049 |
Jim | +0.046 |
Un | +0.046 |
concession | +0.045 |
Jack | +0.045 |
Neg logit contributions | |
pieces | -0.061 |
on | -0.060 |
hole | -0.058 |
edges | -0.055 |
atop | -0.055 |
Token | One |
Token ID | 3198 |
Feature activation | -2.59e+00 |
Pos logit contributions | |
hole | +0.040 |
pieces | +0.040 |
atop | +0.038 |
edges | +0.038 |
ps | +0.038 |
Neg logit contributions | |
Jack | -0.015 |
Un | -0.014 |
David | -0.014 |
Fro | -0.014 |
Jim | -0.013 |
Token | day |
Token ID | 1110 |
Feature activation | -5.35e+00 |
Pos logit contributions | |
pieces | +0.016 |
hole | +0.015 |
on | +0.015 |
atop | +0.014 |
falling | +0.013 |
Neg logit contributions | |
David | -0.044 |
Un | -0.043 |
Jack | -0.043 |
Fro | -0.043 |
Kids | -0.042 |
Token | , |
Token ID | 11 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
hole | +0.051 |
pieces | +0.050 |
atop | +0.045 |
edges | +0.045 |
ps | +0.045 |
Neg logit contributions | |
Jack | -0.074 |
Un | -0.073 |
David | -0.072 |
Fro | -0.072 |
Jim | -0.070 |
Token | that |
Token ID | 326 |
Feature activation | -4.42e+00 |
Pos logit contributions | |
hardest | +0.184 |
edges | +0.178 |
pieces | +0.177 |
ps | +0.176 |
hole | +0.170 |
Neg logit contributions | |
Un | -0.196 |
David | -0.188 |
Jack | -0.186 |
concession | -0.185 |
Johnny | -0.183 |
Token | allowed |
Token ID | 3142 |
Feature activation | -1.07e+01 |
Pos logit contributions | |
hole | +0.068 |
ps | +0.064 |
atop | +0.063 |
els | +0.061 |
edges | +0.061 |
Neg logit contributions | |
Jack | -0.037 |
Un | -0.037 |
Johnny | -0.035 |
Sarah | -0.034 |
Kids | -0.033 |
Token | people |
Token ID | 661 |
Feature activation | -1.34e+01 |
Pos logit contributions | |
hole | +0.174 |
pieces | +0.169 |
atop | +0.165 |
edges | +0.164 |
hardest | +0.159 |
Neg logit contributions | |
Fro | -0.073 |
Jack | -0.072 |
Un | -0.071 |
David | -0.070 |
Kids | -0.069 |
Token | to |
Token ID | 284 |
Feature activation | -6.44e+00 |
Pos logit contributions | |
lid | +0.206 |
pieces | +0.205 |
edges | +0.203 |
hole | +0.198 |
ps | +0.193 |
Neg logit contributions | |
Un | -0.101 |
David | -0.097 |
Jack | -0.093 |
Fro | -0.092 |
concession | -0.091 |
Token | buy |
Token ID | 2822 |
Feature activation | -1.50e+01 |
Pos logit contributions | |
hole | +0.094 |
pieces | +0.092 |
edges | +0.091 |
hardest | +0.089 |
atop | +0.089 |
Neg logit contributions | |
Un | -0.060 |
David | -0.056 |
Fro | -0.053 |
Jack | -0.052 |
His | -0.051 |
Token | stamps |
Token ID | 25560 |
Feature activation | -2.08e+01 |
Pos logit contributions | |
hole | +0.196 |
atop | +0.181 |
falling | +0.179 |
ps | +0.177 |
lid | +0.175 |
Neg logit contributions | |
him | -0.158 |
Un | -0.148 |
Jack | -0.144 |
uncle | -0.143 |
David | -0.142 |
Token | . |
Token ID | 13 |
Feature activation | +4.35e+00 |
Pos logit contributions | |
hole | +0.213 |
edges | +0.210 |
lid | +0.206 |
ps | +0.205 |
pieces | +0.202 |
Neg logit contributions | |
David | -0.238 |
Fro | -0.236 |
Jack | -0.235 |
Un | -0.234 |
They | -0.228 |
Token | Her |
Token ID | 2332 |
Feature activation | -2.26e+00 |
Pos logit contributions | |
David | +0.032 |
Jim | +0.031 |
Un | +0.031 |
Jack | +0.030 |
Tony | +0.030 |
Neg logit contributions | |
pieces | -0.072 |
edges | -0.070 |
hole | -0.069 |
doors | -0.069 |
lid | -0.068 |
Token | mom |
Token ID | 1995 |
Feature activation | -4.67e+00 |
Pos logit contributions | |
pieces | +0.033 |
hole | +0.032 |
on | +0.031 |
edges | +0.031 |
atop | +0.030 |
Neg logit contributions | |
Jack | -0.020 |
Un | -0.020 |
David | -0.020 |
Fro | -0.019 |
Jim | -0.019 |
Token | my |
Token ID | 1820 |
Feature activation | -5.11e+00 |
Pos logit contributions | |
hole | +0.061 |
pieces | +0.061 |
atop | +0.057 |
on | +0.057 |
edges | +0.057 |
Neg logit contributions | |
Un | -0.046 |
Jack | -0.046 |
David | -0.046 |
Fro | -0.045 |
Jim | -0.043 |
Token | gave |
Token ID | 2921 |
Feature activation | -8.71e+00 |
Pos logit contributions | |
hole | +0.069 |
pieces | +0.067 |
edges | +0.066 |
atop | +0.065 |
ps | +0.064 |
Neg logit contributions | |
Jack | -0.050 |
Un | -0.049 |
David | -0.049 |
Johnny | -0.047 |
Fro | -0.047 |
Token | saw |
Token ID | 2497 |
Feature activation | -7.90e+00 |
Pos logit contributions | |
hole | +0.043 |
pieces | +0.043 |
atop | +0.040 |
on | +0.039 |
edges | +0.039 |
Neg logit contributions | |
Un | -0.041 |
Jack | -0.041 |
David | -0.040 |
Fro | -0.040 |
concession | -0.039 |
Token | a |
Token ID | 257 |
Feature activation | -7.66e+00 |
Pos logit contributions | |
pieces | +0.081 |
on | +0.074 |
edges | +0.067 |
wires | +0.065 |
ps | +0.065 |
Neg logit contributions | |
Un | -0.115 |
David | -0.114 |
oph | -0.110 |
Jack | -0.109 |
Jim | -0.108 |
Token | big |
Token ID | 1263 |
Feature activation | -9.15e+00 |
Pos logit contributions | |
hole | +0.075 |
ps | +0.070 |
pieces | +0.067 |
atop | +0.066 |
els | +0.064 |
Neg logit contributions | |
Jack | -0.106 |
Un | -0.104 |
Fro | -0.103 |
concession | -0.102 |
David | -0.102 |
Token | bottle |
Token ID | 9294 |
Feature activation | -1.88e+01 |
Pos logit contributions | |
hole | +0.091 |
ps | +0.084 |
atop | +0.081 |
hardest | +0.078 |
pieces | +0.076 |
Neg logit contributions | |
Fro | -0.123 |
concession | -0.121 |
Jack | -0.120 |
Un | -0.114 |
Johnny | -0.112 |
Token | of |
Token ID | 286 |
Feature activation | -9.87e+00 |
Pos logit contributions | |
hole | +0.239 |
ps | +0.236 |
hardest | +0.235 |
edges | +0.230 |
problems | +0.230 |
Neg logit contributions | |
Jack | -0.184 |
Un | -0.174 |
Fro | -0.173 |
His | -0.169 |
They | -0.166 |
Token | shampoo |
Token ID | 47259 |
Feature activation | -2.08e+01 |
Pos logit contributions | |
hole | +0.126 |
atop | +0.114 |
hardest | +0.106 |
existed | +0.105 |
ps | +0.104 |
Neg logit contributions | |
Fro | -0.114 |
Jack | -0.105 |
Mitt | -0.104 |
concession | -0.103 |
Kids | -0.102 |
Token | on |
Token ID | 319 |
Feature activation | -1.06e+01 |
Pos logit contributions | |
hole | +0.170 |
atop | +0.153 |
hardest | +0.149 |
pieces | +0.144 |
angles | +0.139 |
Neg logit contributions | |
Fro | -0.286 |
addy | -0.277 |
Kids | -0.275 |
Jack | -0.269 |
concession | -0.267 |
Token | the |
Token ID | 262 |
Feature activation | -5.65e+00 |
Pos logit contributions | |
hole | +0.135 |
atop | +0.126 |
pieces | +0.121 |
hardest | +0.120 |
ps | +0.120 |
Neg logit contributions | |
Jack | -0.110 |
Fro | -0.110 |
Un | -0.109 |
His | -0.105 |
David | -0.102 |
Token | shelf |
Token ID | 18316 |
Feature activation | -1.60e+01 |
Pos logit contributions | |
hole | +0.047 |
atop | +0.046 |
els | +0.041 |
cks | +0.041 |
pieces | +0.040 |
Neg logit contributions | |
concession | -0.080 |
Fro | -0.079 |
Un | -0.078 |
addy | -0.078 |
Kids | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | +3.66e+00 |
Pos logit contributions | |
ps | +0.182 |
hardest | +0.177 |
atop | +0.174 |
matters | +0.172 |
els | +0.170 |
Neg logit contributions | |
Jack | -0.158 |
Un | -0.155 |
They | -0.155 |
His | -0.154 |
He | -0.153 |
Token | It |
Token ID | 632 |
Feature activation | -7.32e+00 |
Pos logit contributions | |
Un | +0.040 |
David | +0.039 |
concession | +0.039 |
Johnny | +0.036 |
Billy | +0.035 |
Neg logit contributions | |
pieces | -0.049 |
on | -0.048 |
lid | -0.046 |
edges | -0.045 |
hole | -0.041 |
Token | good |
Token ID | 922 |
Feature activation | -8.70e+00 |
Pos logit contributions | |
hole | +0.098 |
atop | +0.097 |
els | +0.092 |
lid | +0.089 |
bones | +0.088 |
Neg logit contributions | |
Un | -0.075 |
Jack | -0.071 |
Johnny | -0.070 |
please | -0.069 |
They | -0.068 |
Token | to |
Token ID | 284 |
Feature activation | -5.18e+00 |
Pos logit contributions | |
hole | +0.111 |
atop | +0.109 |
slot | +0.098 |
lid | +0.098 |
globe | +0.097 |
Neg logit contributions | |
Fro | -0.084 |
coach | -0.082 |
Jack | -0.081 |
Un | -0.081 |
please | -0.080 |
Token | have |
Token ID | 423 |
Feature activation | -8.44e+00 |
Pos logit contributions | |
hole | +0.067 |
pieces | +0.067 |
atop | +0.065 |
on | +0.061 |
ps | +0.061 |
Neg logit contributions | |
Jack | -0.053 |
Un | -0.053 |
David | -0.051 |
Fro | -0.050 |
Kids | -0.050 |
Token | so |
Token ID | 523 |
Feature activation | -4.34e+00 |
Pos logit contributions | |
hole | +0.121 |
atop | +0.119 |
globe | +0.111 |
lid | +0.106 |
angle | +0.101 |
Neg logit contributions | |
Jack | -0.076 |
Un | -0.072 |
His | -0.070 |
Mitt | -0.070 |
They | -0.069 |
Token | much |
Token ID | 881 |
Feature activation | -8.99e+00 |
Pos logit contributions | |
hole | +0.059 |
atop | +0.053 |
globe | +0.050 |
lid | +0.049 |
ps | +0.048 |
Neg logit contributions | |
Un | -0.049 |
Jack | -0.046 |
They | -0.045 |
Ben | -0.043 |
enny | -0.042 |
Token | clutter |
Token ID | 45343 |
Feature activation | -2.08e+01 |
Pos logit contributions | |
hole | +0.103 |
atop | +0.103 |
globe | +0.092 |
hardest | +0.090 |
cks | +0.086 |
Neg logit contributions | |
Jack | -0.096 |
concession | -0.096 |
His | -0.096 |
Un | -0.095 |
Kids | -0.095 |
Token | . |
Token ID | 13 |
Feature activation | +3.68e+00 |
Pos logit contributions | |
hole | +0.225 |
lid | +0.218 |
atop | +0.213 |
trunk | +0.210 |
ool | +0.210 |
Neg logit contributions | |
Un | -0.222 |
Jack | -0.211 |
concession | -0.211 |
They | -0.208 |
Johnny | -0.202 |
Token | You |
Token ID | 921 |
Feature activation | -5.11e+00 |
Pos logit contributions | |
David | +0.035 |
Fro | +0.035 |
Jack | +0.035 |
Un | +0.034 |
concession | +0.033 |
Neg logit contributions | |
hole | -0.049 |
pieces | -0.048 |
on | -0.046 |
edges | -0.046 |
ps | -0.045 |
Token | can |
Token ID | 460 |
Feature activation | -4.97e+00 |
Pos logit contributions | |
atop | +0.065 |
hole | +0.064 |
bones | +0.060 |
pieces | +0.060 |
edges | +0.059 |
Neg logit contributions | |
Jack | -0.049 |
Un | -0.046 |
enny | -0.046 |
Fro | -0.046 |
please | -0.045 |
Token | trip |
Token ID | 5296 |
Feature activation | -1.58e+01 |
Pos logit contributions | |
pieces | +0.050 |
hole | +0.050 |
on | +0.050 |
falling | +0.048 |
edges | +0.047 |
Neg logit contributions | |
David | -0.065 |
Un | -0.064 |
Fro | -0.064 |
Da | -0.063 |
Jim | -0.063 |
Token | and |
Token ID | 290 |
Feature activation | -3.75e+00 |
Pos logit contributions | |
hole | +0.184 |
pieces | +0.177 |
atop | +0.173 |
lid | +0.167 |
els | +0.166 |
Neg logit contributions | |
Jack | -0.181 |
Un | -0.177 |
David | -0.172 |
concession | -0.169 |
Johnny | -0.167 |
Token | the |
Token ID | 262 |
Feature activation | -4.63e+00 |
Pos logit contributions | |
hole | +0.067 |
pieces | +0.066 |
atop | +0.062 |
on | +0.061 |
edges | +0.061 |
Neg logit contributions | |
Un | -0.052 |
Jack | -0.052 |
David | -0.051 |
Fro | -0.051 |
Jim | -0.048 |
Token | birds |
Token ID | 10087 |
Feature activation | -1.31e+01 |
Pos logit contributions | |
hole | +0.066 |
ool | +0.066 |
cks | +0.063 |
bones | +0.061 |
atop | +0.060 |
Neg logit contributions | |
wife | -0.039 |
Ben | -0.036 |
David | -0.036 |
Jack | -0.035 |
Un | -0.035 |
Token | could |
Token ID | 714 |
Feature activation | -4.71e+00 |
Pos logit contributions | |
pieces | +0.170 |
hole | +0.168 |
edges | +0.167 |
lid | +0.164 |
els | +0.162 |
Neg logit contributions | |
please | -0.102 |
He | -0.102 |
Jack | -0.101 |
They | -0.100 |
Fro | -0.100 |
Token | make |
Token ID | 787 |
Feature activation | -1.10e+01 |
Pos logit contributions | |
hole | +0.056 |
pieces | +0.056 |
atop | +0.052 |
on | +0.052 |
edges | +0.051 |
Neg logit contributions | |
Un | -0.052 |
Jack | -0.052 |
David | -0.052 |
Fro | -0.051 |
Jim | -0.049 |
Token | their |
Token ID | 511 |
Feature activation | -6.86e+00 |
Pos logit contributions | |
hole | +0.148 |
atop | +0.134 |
pieces | +0.132 |
lid | +0.130 |
edges | +0.129 |
Neg logit contributions | |
Fro | -0.109 |
Jack | -0.106 |
Un | -0.105 |
David | -0.104 |
concession | -0.100 |
Token | nests |
Token ID | 44382 |
Feature activation | -2.07e+01 |
Pos logit contributions | |
hole | +0.079 |
atop | +0.075 |
ps | +0.071 |
ool | +0.065 |
cks | +0.065 |
Neg logit contributions | |
Fro | -0.077 |
concession | -0.069 |
wife | -0.068 |
Mitt | -0.068 |
uncle | -0.068 |
Token | and |
Token ID | 290 |
Feature activation | -4.30e+00 |
Pos logit contributions | |
hole | +0.221 |
pieces | +0.212 |
edges | +0.206 |
els | +0.204 |
atop | +0.203 |
Neg logit contributions | |
Fro | -0.228 |
Jack | -0.223 |
Un | -0.223 |
David | -0.219 |
concession | -0.217 |
Token | feel |
Token ID | 1254 |
Feature activation | -9.45e+00 |
Pos logit contributions | |
hole | +0.049 |
pieces | +0.045 |
atop | +0.044 |
ps | +0.042 |
edges | +0.042 |
Neg logit contributions | |
Un | -0.050 |
Fro | -0.049 |
Jack | -0.047 |
concession | -0.047 |
David | -0.047 |
Token | safe |
Token ID | 3338 |
Feature activation | -1.58e+01 |
Pos logit contributions | |
hole | +0.113 |
pieces | +0.106 |
els | +0.099 |
edges | +0.098 |
atop | +0.098 |
Neg logit contributions | |
Fro | -0.107 |
Un | -0.105 |
Jack | -0.104 |
David | -0.101 |
concession | -0.101 |
Token | . |
Token ID | 13 |
Feature activation | +4.12e+00 |
Pos logit contributions | |
hole | +0.169 |
els | +0.164 |
lid | +0.162 |
ool | +0.162 |
pieces | +0.159 |
Neg logit contributions | |
Un | -0.168 |
concession | -0.161 |
Fro | -0.160 |
They | -0.159 |
He | -0.156 |
Token | The |
Token ID | 383 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
Un | +0.038 |
David | +0.037 |
concession | +0.036 |
Kids | +0.036 |
Jack | +0.035 |
Neg logit contributions | |
hole | -0.058 |
pieces | -0.057 |
on | -0.054 |
edges | -0.053 |
trunk | -0.053 |
Token | Let |
Token ID | 5756 |
Feature activation | -5.94e+00 |
Pos logit contributions | |
Fro | +0.012 |
concession | +0.011 |
enny | +0.010 |
Un | +0.010 |
David | +0.010 |
Neg logit contributions | |
pieces | -0.022 |
hole | -0.021 |
on | -0.020 |
edges | -0.020 |
tangled | -0.020 |
Token | 's |
Token ID | 338 |
Feature activation | -4.16e+00 |
Pos logit contributions | |
way | +0.037 |
on | +0.036 |
pieces | +0.035 |
broken | +0.030 |
button | +0.028 |
Neg logit contributions | |
Fro | -0.105 |
enny | -0.104 |
concession | -0.102 |
David | -0.101 |
Tony | -0.101 |
Token | gather |
Token ID | 6431 |
Feature activation | -1.44e+01 |
Pos logit contributions | |
hole | +0.051 |
pieces | +0.050 |
on | +0.049 |
atop | +0.047 |
falling | +0.047 |
Neg logit contributions | |
Un | -0.045 |
Jack | -0.045 |
David | -0.044 |
Fro | -0.043 |
Da | -0.043 |
Token | up |
Token ID | 510 |
Feature activation | -1.10e+01 |
Pos logit contributions | |
hole | +0.189 |
falling | +0.177 |
code | +0.171 |
atop | +0.169 |
changing | +0.164 |
Neg logit contributions | |
Fro | -0.145 |
Jack | -0.140 |
Pete | -0.129 |
concession | -0.125 |
Mitt | -0.121 |
Token | some |
Token ID | 617 |
Feature activation | -1.20e+01 |
Pos logit contributions | |
hole | +0.146 |
atop | +0.131 |
falling | +0.129 |
lid | +0.128 |
pieces | +0.127 |
Neg logit contributions | |
Jack | -0.114 |
Fro | -0.110 |
Jim | -0.106 |
concession | -0.106 |
Un | -0.106 |
Token | rocks |
Token ID | 12586 |
Feature activation | -2.07e+01 |
Pos logit contributions | |
hole | +0.137 |
ps | +0.134 |
bones | +0.125 |
falling | +0.124 |
globe | +0.124 |
Neg logit contributions | |
concession | -0.121 |
please | -0.119 |
Fro | -0.119 |
wife | -0.114 |
Jack | -0.114 |
Token | and |
Token ID | 290 |
Feature activation | -5.61e+00 |
Pos logit contributions | |
lid | +0.219 |
ps | +0.217 |
edges | +0.217 |
falling | +0.215 |
hole | +0.213 |
Neg logit contributions | |
Fro | -0.219 |
Jack | -0.218 |
concession | -0.210 |
enny | -0.209 |
Un | -0.207 |
Token | see |
Token ID | 766 |
Feature activation | -7.32e+00 |
Pos logit contributions | |
hole | +0.059 |
pieces | +0.059 |
atop | +0.054 |
on | +0.053 |
edges | +0.053 |
Neg logit contributions | |
Un | -0.069 |
Jack | -0.069 |
David | -0.068 |
Fro | -0.068 |
concession | -0.066 |
Token | who |
Token ID | 508 |
Feature activation | -5.91e+00 |
Pos logit contributions | |
lid | +0.109 |
bones | +0.108 |
splits | +0.107 |
falling | +0.106 |
hole | +0.106 |
Neg logit contributions | |
Fro | -0.048 |
Un | -0.047 |
Jack | -0.047 |
They | -0.047 |
Johnny | -0.043 |
Token | can |
Token ID | 460 |
Feature activation | -6.44e+00 |
Pos logit contributions | |
ps | +0.073 |
lid | +0.071 |
hole | +0.071 |
falling | +0.069 |
edges | +0.068 |
Neg logit contributions | |
Jack | -0.056 |
concession | -0.052 |
Un | -0.050 |
Fro | -0.050 |
They | -0.050 |
Token | make |
Token ID | 787 |
Feature activation | -1.17e+01 |
Pos logit contributions | |
hole | +0.077 |
pieces | +0.076 |
atop | +0.071 |
edges | +0.070 |
ps | +0.069 |
Neg logit contributions | |
Jack | -0.070 |
Fro | -0.069 |
Un | -0.068 |
David | -0.067 |
Jim | -0.065 |
Token | you |
Token ID | 345 |
Feature activation | -7.48e+00 |
Pos logit contributions | |
hole | +0.086 |
pieces | +0.084 |
atop | +0.078 |
edges | +0.077 |
falling | +0.077 |
Neg logit contributions | |
Jack | -0.073 |
Un | -0.072 |
David | -0.072 |
Fro | -0.071 |
concession | -0.069 |
Token | will |
Token ID | 481 |
Feature activation | -7.50e+00 |
Pos logit contributions | |
hole | +0.105 |
pieces | +0.103 |
atop | +0.100 |
falling | +0.099 |
ps | +0.099 |
Neg logit contributions | |
Jack | -0.065 |
Un | -0.064 |
David | -0.062 |
Fro | -0.061 |
concession | -0.061 |
Token | never |
Token ID | 1239 |
Feature activation | -7.43e+00 |
Pos logit contributions | |
hole | +0.105 |
pieces | +0.105 |
atop | +0.102 |
edges | +0.096 |
els | +0.095 |
Neg logit contributions | |
Jack | -0.068 |
David | -0.067 |
Un | -0.067 |
Fro | -0.066 |
His | -0.065 |
Token | touch |
Token ID | 3638 |
Feature activation | -1.39e+01 |
Pos logit contributions | |
hole | +0.099 |
pieces | +0.098 |
atop | +0.096 |
edges | +0.092 |
ps | +0.092 |
Neg logit contributions | |
Un | -0.071 |
Jack | -0.071 |
Fro | -0.070 |
David | -0.068 |
His | -0.065 |
Token | my |
Token ID | 616 |
Feature activation | -7.23e+00 |
Pos logit contributions | |
hole | +0.214 |
ool | +0.201 |
code | +0.198 |
lid | +0.197 |
atop | +0.195 |
Neg logit contributions | |
him | -0.113 |
Jack | -0.088 |
Max | -0.088 |
Mitt | -0.086 |
Dad | -0.086 |
Token | axe |
Token ID | 23272 |
Feature activation | -2.07e+01 |
Pos logit contributions | |
bones | +0.075 |
ps | +0.073 |
atop | +0.073 |
hole | +0.069 |
ool | +0.068 |
Neg logit contributions | |
Un | -0.084 |
uncle | -0.084 |
Fro | -0.079 |
Mitt | -0.075 |
him | -0.075 |
Token | again |
Token ID | 757 |
Feature activation | -1.17e+01 |
Pos logit contributions | |
atop | +0.248 |
hole | +0.247 |
ps | +0.246 |
els | +0.241 |
lid | +0.240 |
Neg logit contributions | |
Un | -0.206 |
Fro | -0.200 |
Jack | -0.192 |
concession | -0.189 |
His | -0.184 |
Token | ," |
Token ID | 553 |
Feature activation | +4.22e+00 |
Pos logit contributions | |
hole | +0.109 |
ps | +0.106 |
atop | +0.105 |
lid | +0.104 |
pieces | +0.102 |
Neg logit contributions | |
Un | -0.162 |
Jack | -0.154 |
Fro | -0.150 |
David | -0.147 |
Jim | -0.145 |
Token | his |
Token ID | 465 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
concession | +0.050 |
latable | +0.050 |
enny | +0.047 |
oph | +0.047 |
Da | +0.046 |
Neg logit contributions | |
on | -0.052 |
pieces | -0.048 |
hard | -0.047 |
broken | -0.046 |
tangled | -0.045 |
Token | dad |
Token ID | 9955 |
Feature activation | -5.68e+00 |
Pos logit contributions | |
pieces | +0.040 |
hole | +0.040 |
on | +0.038 |
edges | +0.038 |
atop | +0.038 |
Neg logit contributions | |
Jack | -0.011 |
Un | -0.010 |
Fro | -0.010 |
David | -0.010 |
concession | -0.009 |
Token | said |
Token ID | 531 |
Feature activation | -5.25e+00 |
Pos logit contributions | |
hole | +0.075 |
pieces | +0.074 |
edges | +0.072 |
atop | +0.072 |
falling | +0.071 |
Neg logit contributions | |
Jack | -0.055 |
Un | -0.055 |
David | -0.054 |
Fro | -0.053 |
His | -0.052 |
Token | tree |
Token ID | 5509 |
Feature activation | -1.61e+01 |
Pos logit contributions | |
ps | +0.147 |
hole | +0.137 |
hardest | +0.123 |
angles | +0.119 |
code | +0.119 |
Neg logit contributions | |
Un | -0.257 |
He | -0.252 |
Jack | -0.251 |
She | -0.251 |
concession | -0.247 |
Token | with |
Token ID | 351 |
Feature activation | -8.46e+00 |
Pos logit contributions | |
ps | +0.172 |
pieces | +0.162 |
hardest | +0.160 |
edges | +0.158 |
hole | +0.157 |
Neg logit contributions | |
Un | -0.189 |
Fro | -0.181 |
Jack | -0.181 |
They | -0.176 |
He | -0.176 |
Token | many |
Token ID | 867 |
Feature activation | -1.19e+01 |
Pos logit contributions | |
hole | +0.085 |
pieces | +0.076 |
atop | +0.076 |
ps | +0.073 |
falling | +0.072 |
Neg logit contributions | |
Jack | -0.111 |
Fro | -0.111 |
Un | -0.110 |
David | -0.109 |
Jim | -0.106 |
Token | ripe |
Token ID | 29036 |
Feature activation | -1.72e+01 |
Pos logit contributions | |
hole | +0.125 |
ps | +0.122 |
slot | +0.116 |
globe | +0.115 |
code | +0.114 |
Neg logit contributions | |
He | -0.146 |
Jack | -0.143 |
She | -0.137 |
His | -0.136 |
Da | -0.136 |
Token | p |
Token ID | 279 |
Feature activation | -1.20e+01 |
Pos logit contributions | |
hole | +0.185 |
code | +0.175 |
remote | +0.173 |
ps | +0.171 |
slot | +0.163 |
Neg logit contributions | |
Fro | -0.200 |
oph | -0.193 |
He | -0.190 |
Jack | -0.189 |
He | -0.184 |
Token | ears |
Token ID | 4127 |
Feature activation | -2.06e+01 |
Pos logit contributions | |
hole | +0.146 |
pieces | +0.145 |
atop | +0.137 |
ps | +0.137 |
on | +0.136 |
Neg logit contributions | |
Fro | -0.126 |
Un | -0.124 |
Jack | -0.122 |
David | -0.122 |
oph | -0.121 |
Token | . |
Token ID | 13 |
Feature activation | +4.59e+00 |
Pos logit contributions | |
hole | +0.200 |
code | +0.197 |
ps | +0.196 |
hardest | +0.194 |
remote | +0.189 |
Neg logit contributions | |
Un | -0.230 |
They | -0.228 |
Fro | -0.226 |
Jack | -0.221 |
He | -0.220 |
Token | They |
Token ID | 1119 |
Feature activation | -2.87e+00 |
Pos logit contributions | |
Johnny | +0.058 |
David | +0.056 |
Un | +0.052 |
Billy | +0.051 |
Harry | +0.051 |
Neg logit contributions | |
pieces | -0.054 |
on | -0.053 |
edges | -0.050 |
lid | -0.050 |
ladder | -0.046 |
Token | wanted |
Token ID | 2227 |
Feature activation | -8.75e+00 |
Pos logit contributions | |
pieces | +0.034 |
hole | +0.032 |
on | +0.031 |
edges | +0.030 |
trunk | +0.030 |
Neg logit contributions | |
Un | -0.032 |
concession | -0.032 |
Fro | -0.032 |
David | -0.032 |
Kids | -0.031 |
Token | to |
Token ID | 284 |
Feature activation | -5.10e+00 |
Pos logit contributions | |
hole | +0.155 |
atop | +0.142 |
ps | +0.140 |
lid | +0.135 |
ool | +0.135 |
Neg logit contributions | |
Fro | -0.060 |
Jack | -0.056 |
Da | -0.050 |
Jim | -0.050 |
Un | -0.049 |
Token | pick |
Token ID | 2298 |
Feature activation | -1.45e+01 |
Pos logit contributions | |
hole | +0.058 |
pieces | +0.057 |
atop | +0.053 |
edges | +0.052 |
ps | +0.052 |
Neg logit contributions | |
Jack | -0.060 |
Un | -0.060 |
Fro | -0.059 |
David | -0.059 |
Jim | -0.057 |
Token | chest |
Token ID | 7721 |
Feature activation | -1.55e+01 |
Pos logit contributions | |
falling | +0.053 |
ps | +0.052 |
atop | +0.051 |
els | +0.051 |
hole | +0.051 |
Neg logit contributions | |
concession | -0.070 |
His | -0.069 |
He | -0.069 |
Fro | -0.069 |
Jack | -0.069 |
Token | and |
Token ID | 290 |
Feature activation | -3.34e+00 |
Pos logit contributions | |
ps | +0.235 |
els | +0.231 |
edges | +0.227 |
problems | +0.227 |
bones | +0.224 |
Neg logit contributions | |
His | -0.097 |
Jack | -0.097 |
Fro | -0.097 |
He | -0.096 |
His | -0.095 |
Token | found |
Token ID | 1043 |
Feature activation | -8.51e+00 |
Pos logit contributions | |
hole | +0.040 |
pieces | +0.039 |
ps | +0.037 |
edges | +0.037 |
falling | +0.037 |
Neg logit contributions | |
Jack | -0.038 |
Fro | -0.036 |
Un | -0.036 |
David | -0.036 |
Jim | -0.035 |
Token | lots |
Token ID | 6041 |
Feature activation | -1.23e+01 |
Pos logit contributions | |
hole | +0.122 |
falling | +0.119 |
changing | +0.112 |
edges | +0.112 |
ool | +0.110 |
Neg logit contributions | |
Fro | -0.074 |
Jack | -0.071 |
His | -0.068 |
David | -0.066 |
him | -0.064 |
Token | of |
Token ID | 286 |
Feature activation | -8.86e+00 |
Pos logit contributions | |
hole | +0.184 |
ps | +0.182 |
hardest | +0.182 |
atop | +0.177 |
falling | +0.174 |
Neg logit contributions | |
Jack | -0.092 |
concession | -0.090 |
Fro | -0.086 |
enny | -0.082 |
He | -0.081 |
Token | gold |
Token ID | 3869 |
Feature activation | -2.06e+01 |
Pos logit contributions | |
hole | +0.090 |
ps | +0.084 |
atop | +0.083 |
falling | +0.081 |
hardest | +0.080 |
Neg logit contributions | |
Jack | -0.113 |
concession | -0.109 |
Fro | -0.107 |
David | -0.106 |
He | -0.104 |
Token | coins |
Token ID | 10796 |
Feature activation | -2.09e+01 |
Pos logit contributions | |
hardest | +0.170 |
angle | +0.165 |
falling | +0.162 |
changing | +0.162 |
hole | +0.162 |
Neg logit contributions | |
He | -0.257 |
ory | -0.249 |
him | -0.247 |
Fro | -0.247 |
He | -0.243 |
Token | and |
Token ID | 290 |
Feature activation | -5.05e+00 |
Pos logit contributions | |
splits | +0.210 |
edges | +0.206 |
hardest | +0.206 |
connects | +0.204 |
ps | +0.202 |
Neg logit contributions | |
Fro | -0.221 |
ory | -0.219 |
David | -0.216 |
ete | -0.209 |
He | -0.209 |
Token | shiny |
Token ID | 22441 |
Feature activation | -1.54e+01 |
Pos logit contributions | |
hole | +0.039 |
atop | +0.035 |
ps | +0.033 |
falling | +0.033 |
ool | +0.031 |
Neg logit contributions | |
Jack | -0.076 |
Fro | -0.076 |
Un | -0.075 |
David | -0.074 |
concession | -0.073 |
Token | jewels |
Token ID | 42701 |
Feature activation | -1.84e+01 |
Pos logit contributions | |
hole | +0.134 |
falling | +0.132 |
hardest | +0.131 |
socket | +0.121 |
changing | +0.120 |
Neg logit contributions | |
him | -0.188 |
ory | -0.187 |
He | -0.186 |
He | -0.184 |
His | -0.181 |
Token | . |
Token ID | 13 |
Feature activation | +4.32e+00 |
Pos logit contributions | |
hardest | +0.196 |
edges | +0.191 |
els | +0.188 |
ps | +0.185 |
wires | +0.185 |
Neg logit contributions | |
Fro | -0.182 |
He | -0.179 |
enny | -0.179 |
Jack | -0.178 |
ory | -0.175 |
Token | the |
Token ID | 262 |
Feature activation | -6.04e+00 |
Pos logit contributions | |
hole | +0.187 |
lid | +0.170 |
ool | +0.169 |
code | +0.162 |
pieces | +0.162 |
Neg logit contributions | |
Fro | -0.083 |
Un | -0.069 |
him | -0.068 |
Mitt | -0.066 |
Dod | -0.064 |
Token | area |
Token ID | 1989 |
Feature activation | -1.65e+01 |
Pos logit contributions | |
hole | +0.085 |
els | +0.084 |
bones | +0.079 |
cks | +0.077 |
ps | +0.077 |
Neg logit contributions | |
Fro | -0.062 |
He | -0.054 |
He | -0.052 |
shepherd | -0.052 |
Da | -0.052 |
Token | clean |
Token ID | 3424 |
Feature activation | -1.45e+01 |
Pos logit contributions | |
hole | +0.192 |
ps | +0.180 |
els | +0.171 |
atop | +0.170 |
edges | +0.168 |
Neg logit contributions | |
Jack | -0.170 |
Da | -0.166 |
Fro | -0.166 |
His | -0.161 |
They | -0.159 |
Token | by |
Token ID | 416 |
Feature activation | -6.41e+00 |
Pos logit contributions | |
hole | +0.168 |
atop | +0.159 |
ps | +0.156 |
els | +0.152 |
lid | +0.151 |
Neg logit contributions | |
Fro | -0.151 |
concession | -0.139 |
Un | -0.139 |
Jack | -0.138 |
He | -0.135 |
Token | making |
Token ID | 1642 |
Feature activation | -1.20e+01 |
Pos logit contributions | |
hole | +0.090 |
slot | +0.081 |
atop | +0.080 |
code | +0.080 |
lid | +0.080 |
Neg logit contributions | |
Fro | -0.059 |
They | -0.054 |
Un | -0.053 |
Mitt | -0.052 |
him | -0.049 |
Token | stacks |
Token ID | 24285 |
Feature activation | -2.06e+01 |
Pos logit contributions | |
hole | +0.133 |
atop | +0.119 |
pieces | +0.111 |
falling | +0.109 |
globe | +0.108 |
Neg logit contributions | |
Fro | -0.152 |
Jack | -0.148 |
Un | -0.144 |
David | -0.143 |
concession | -0.139 |
Token | of |
Token ID | 286 |
Feature activation | -8.73e+00 |
Pos logit contributions | |
hole | +0.249 |
atop | +0.235 |
edges | +0.225 |
ps | +0.222 |
lid | +0.217 |
Neg logit contributions | |
They | -0.218 |
Un | -0.210 |
Jack | -0.209 |
Fro | -0.202 |
He | -0.201 |
Token | sticks |
Token ID | 16461 |
Feature activation | -2.01e+01 |
Pos logit contributions | |
hole | +0.087 |
atop | +0.083 |
hardest | +0.075 |
ps | +0.074 |
existed | +0.069 |
Neg logit contributions | |
Fro | -0.120 |
Jack | -0.111 |
concession | -0.108 |
Un | -0.108 |
Johnny | -0.106 |
Token | . |
Token ID | 13 |
Feature activation | +4.64e+00 |
Pos logit contributions | |
hole | +0.251 |
atop | +0.236 |
edges | +0.234 |
slot | +0.233 |
els | +0.231 |
Neg logit contributions | |
They | -0.191 |
Fro | -0.187 |
He | -0.177 |
He | -0.166 |
Un | -0.166 |
Token |
|
Token ID | 198 |
Feature activation | +2.30e+00 |
Pos logit contributions | |
Un | +0.038 |
David | +0.038 |
Jack | +0.037 |
Fro | +0.036 |
concession | +0.036 |
Neg logit contributions | |
hole | -0.069 |
pieces | -0.069 |
on | -0.065 |
atop | -0.065 |
edges | -0.064 |
Token |
|
Token ID | 198 |
Feature activation | +3.34e+00 |
Pos logit contributions | |
Un | +0.018 |
concession | +0.018 |
Fro | +0.018 |
David | +0.018 |
Jack | +0.017 |
Neg logit contributions | |
hole | -0.034 |
pieces | -0.034 |
on | -0.033 |
atop | -0.032 |
edges | -0.031 |
Token | a |
Token ID | 257 |
Feature activation | -5.30e+00 |
Pos logit contributions | |
hole | +0.081 |
plank | +0.078 |
ps | +0.078 |
ool | +0.078 |
splits | +0.077 |
Neg logit contributions | |
Un | -0.047 |
Jack | -0.047 |
David | -0.046 |
uncle | -0.044 |
Her | -0.044 |
Token | grand |
Token ID | 4490 |
Feature activation | -5.66e+00 |
Pos logit contributions | |
ps | +0.068 |
els | +0.066 |
hole | +0.062 |
cks | +0.062 |
atop | +0.059 |
Neg logit contributions | |
Un | -0.060 |
Jack | -0.060 |
concession | -0.059 |
David | -0.059 |
Johnny | -0.056 |
Token | hall |
Token ID | 6899 |
Feature activation | -1.67e+01 |
Pos logit contributions | |
atop | +0.067 |
hole | +0.066 |
pieces | +0.065 |
tangled | +0.065 |
on | +0.064 |
Neg logit contributions | |
Jack | -0.061 |
David | -0.061 |
oph | -0.060 |
concession | -0.059 |
Jim | -0.059 |
Token | filled |
Token ID | 5901 |
Feature activation | -1.29e+01 |
Pos logit contributions | |
ps | +0.183 |
pieces | +0.181 |
els | +0.178 |
edges | +0.175 |
lid | +0.172 |
Neg logit contributions | |
concession | -0.187 |
Un | -0.183 |
David | -0.181 |
Jack | -0.181 |
He | -0.170 |
Token | with |
Token ID | 351 |
Feature activation | -9.88e+00 |
Pos logit contributions | |
hole | +0.144 |
ps | +0.134 |
pieces | +0.131 |
hardest | +0.130 |
edges | +0.129 |
Neg logit contributions | |
Jack | -0.157 |
Un | -0.150 |
David | -0.148 |
concession | -0.146 |
Da | -0.144 |
Token | pots |
Token ID | 32195 |
Feature activation | -2.06e+01 |
Pos logit contributions | |
hole | +0.111 |
atop | +0.102 |
ool | +0.101 |
ps | +0.097 |
cks | +0.096 |
Neg logit contributions | |
concession | -0.115 |
David | -0.108 |
Jack | -0.108 |
Un | -0.106 |
Fro | -0.106 |
Token | of |
Token ID | 286 |
Feature activation | -9.24e+00 |
Pos logit contributions | |
hardest | +0.266 |
hole | +0.262 |
edges | +0.260 |
splits | +0.260 |
atop | +0.258 |
Neg logit contributions | |
David | -0.163 |
Jack | -0.162 |
Un | -0.160 |
concession | -0.157 |
He | -0.156 |
Token | the |
Token ID | 262 |
Feature activation | -6.50e+00 |
Pos logit contributions | |
hole | +0.104 |
hardest | +0.096 |
atop | +0.093 |
ps | +0.089 |
existed | +0.084 |
Neg logit contributions | |
concession | -0.112 |
Fro | -0.102 |
uncle | -0.102 |
Un | -0.101 |
Mitt | -0.099 |
Token | magic |
Token ID | 5536 |
Feature activation | -1.51e+01 |
Pos logit contributions | |
cks | +0.075 |
els | +0.070 |
ool | +0.068 |
bones | +0.067 |
atop | +0.066 |
Neg logit contributions | |
concession | -0.076 |
Jack | -0.069 |
addy | -0.063 |
triple | -0.060 |
wife | -0.060 |
Token | liquid |
Token ID | 8122 |
Feature activation | -1.85e+01 |
Pos logit contributions | |
cks | +0.135 |
atop | +0.121 |
ps | +0.121 |
hardest | +0.119 |
tangled | +0.117 |
Neg logit contributions | |
concession | -0.185 |
Jack | -0.176 |
He | -0.172 |
shepherd | -0.170 |
addy | -0.169 |
Token | , |
Token ID | 11 |
Feature activation | -5.33e+00 |
Pos logit contributions | |
hole | +0.190 |
plank | +0.178 |
hardest | +0.176 |
pieces | +0.175 |
edges | +0.174 |
Neg logit contributions | |
They | -0.191 |
Jack | -0.181 |
Un | -0.179 |
concession | -0.179 |
Fro | -0.175 |
Token | load |
Token ID | 3440 |
Feature activation | -1.70e+01 |
Pos logit contributions | |
hole | +0.096 |
atop | +0.089 |
lid | +0.087 |
pieces | +0.087 |
els | +0.086 |
Neg logit contributions | |
Fro | -0.066 |
Un | -0.062 |
Kids | -0.057 |
Jack | -0.057 |
His | -0.055 |
Token | your |
Token ID | 534 |
Feature activation | -8.09e+00 |
Pos logit contributions | |
hole | +0.192 |
atop | +0.157 |
code | +0.150 |
slot | +0.150 |
falling | +0.143 |
Neg logit contributions | |
Fro | -0.216 |
Jack | -0.207 |
concession | -0.201 |
Un | -0.200 |
Da | -0.194 |
Token | wagon |
Token ID | 31860 |
Feature activation | -1.77e+01 |
Pos logit contributions | |
hole | +0.106 |
atop | +0.093 |
slot | +0.089 |
code | +0.086 |
connects | +0.083 |
Neg logit contributions | |
Fro | -0.092 |
His | -0.080 |
omo | -0.077 |
concession | -0.076 |
His | -0.075 |
Token | with |
Token ID | 351 |
Feature activation | -8.28e+00 |
Pos logit contributions | |
hole | +0.184 |
pieces | +0.172 |
atop | +0.171 |
ps | +0.168 |
edges | +0.167 |
Neg logit contributions | |
Jack | -0.207 |
Fro | -0.206 |
concession | -0.205 |
Un | -0.205 |
His | -0.197 |
Token | p |
Token ID | 279 |
Feature activation | -1.13e+01 |
Pos logit contributions | |
hole | +0.108 |
atop | +0.096 |
lid | +0.087 |
hardest | +0.087 |
pieces | +0.087 |
Neg logit contributions | |
Fro | -0.090 |
Un | -0.085 |
Jack | -0.083 |
Mitt | -0.081 |
concession | -0.081 |
Token | ears |
Token ID | 4127 |
Feature activation | -2.06e+01 |
Pos logit contributions | |
hole | +0.130 |
pieces | +0.130 |
atop | +0.122 |
on | +0.119 |
edges | +0.119 |
Neg logit contributions | |
Fro | -0.128 |
Un | -0.126 |
Jack | -0.125 |
David | -0.125 |
concession | -0.121 |
Token | near |
Token ID | 1474 |
Feature activation | -1.00e+01 |
Pos logit contributions | |
hole | +0.222 |
atop | +0.208 |
ps | +0.208 |
pieces | +0.207 |
edges | +0.205 |
Neg logit contributions | |
Un | -0.236 |
Fro | -0.236 |
Jack | -0.233 |
concession | -0.231 |
enny | -0.228 |
Token | a |
Token ID | 257 |
Feature activation | -4.39e+00 |
Pos logit contributions | |
hole | +0.140 |
atop | +0.131 |
pieces | +0.130 |
edges | +0.125 |
ps | +0.124 |
Neg logit contributions | |
Fro | -0.094 |
Jack | -0.091 |
Un | -0.091 |
David | -0.085 |
Jim | -0.083 |
Token | wild |
Token ID | 4295 |
Feature activation | -1.03e+01 |
Pos logit contributions | |
ps | +0.047 |
els | +0.043 |
hole | +0.041 |
atop | +0.040 |
cks | +0.039 |
Neg logit contributions | |
Un | -0.062 |
Fro | -0.061 |
concession | -0.059 |
Jack | -0.058 |
Da | -0.056 |
Token | bear |
Token ID | 6842 |
Feature activation | -1.51e+01 |
Pos logit contributions | |
hole | +0.104 |
ps | +0.098 |
atop | +0.098 |
pieces | +0.097 |
hardest | +0.092 |
Neg logit contributions | |
Fro | -0.136 |
Un | -0.128 |
Jack | -0.127 |
concession | -0.126 |
Da | -0.122 |
Token | . |
Token ID | 13 |
Feature activation | +4.07e+00 |
Pos logit contributions | |
pieces | +0.135 |
hole | +0.133 |
atop | +0.126 |
edges | +0.125 |
ps | +0.124 |
Neg logit contributions | |
Fro | -0.211 |
Un | -0.209 |
Jack | -0.204 |
concession | -0.199 |
David | -0.198 |
Token | to |
Token ID | 284 |
Feature activation | -5.84e+00 |
Pos logit contributions | |
hole | +0.100 |
atop | +0.093 |
pieces | +0.093 |
ps | +0.090 |
lid | +0.089 |
Neg logit contributions | |
Jack | -0.069 |
Fro | -0.069 |
Un | -0.067 |
David | -0.067 |
Da | -0.064 |
Token | be |
Token ID | 307 |
Feature activation | -8.62e+00 |
Pos logit contributions | |
hole | +0.073 |
pieces | +0.073 |
atop | +0.068 |
on | +0.068 |
edges | +0.067 |
Neg logit contributions | |
Un | -0.061 |
Jack | -0.061 |
David | -0.060 |
Fro | -0.060 |
Jim | -0.057 |
Token | careful |
Token ID | 8161 |
Feature activation | -1.23e+01 |
Pos logit contributions | |
hole | +0.099 |
pieces | +0.095 |
atop | +0.089 |
edges | +0.088 |
lid | +0.087 |
Neg logit contributions | |
Un | -0.102 |
Jack | -0.102 |
David | -0.100 |
Fro | -0.100 |
Da | -0.096 |
Token | when |
Token ID | 618 |
Feature activation | -3.91e+00 |
Pos logit contributions | |
hole | +0.140 |
pieces | +0.136 |
atop | +0.129 |
ps | +0.126 |
lid | +0.126 |
Neg logit contributions | |
Jack | -0.142 |
Un | -0.140 |
David | -0.137 |
Fro | -0.135 |
His | -0.133 |
Token | collecting |
Token ID | 13157 |
Feature activation | -1.66e+01 |
Pos logit contributions | |
hole | +0.063 |
pieces | +0.057 |
ps | +0.056 |
lid | +0.056 |
atop | +0.056 |
Neg logit contributions | |
Fro | -0.031 |
Jack | -0.030 |
Un | -0.030 |
David | -0.028 |
Kids | -0.028 |
Token | rocks |
Token ID | 12586 |
Feature activation | -2.06e+01 |
Pos logit contributions | |
hole | +0.156 |
atop | +0.131 |
slot | +0.123 |
globe | +0.121 |
changing | +0.116 |
Neg logit contributions | |
Fro | -0.228 |
Jack | -0.209 |
His | -0.209 |
He | -0.206 |
His | -0.202 |
Token | and |
Token ID | 290 |
Feature activation | -5.44e+00 |
Pos logit contributions | |
hole | +0.241 |
lid | +0.232 |
edges | +0.232 |
ps | +0.226 |
atop | +0.225 |
Neg logit contributions | |
Fro | -0.215 |
Jack | -0.201 |
Un | -0.201 |
enny | -0.199 |
He | -0.194 |
Token | always |
Token ID | 1464 |
Feature activation | -6.52e+00 |
Pos logit contributions | |
pieces | +0.060 |
hole | +0.059 |
on | +0.055 |
edges | +0.054 |
atop | +0.054 |
Neg logit contributions | |
Un | -0.066 |
Jack | -0.065 |
David | -0.065 |
Fro | -0.064 |
Jim | -0.063 |
Token | make |
Token ID | 787 |
Feature activation | -1.15e+01 |
Pos logit contributions | |
hole | +0.094 |
pieces | +0.093 |
atop | +0.088 |
lid | +0.086 |
edges | +0.086 |
Neg logit contributions | |
Un | -0.055 |
Fro | -0.055 |
Jack | -0.054 |
His | -0.053 |
David | -0.052 |
Token | sure |
Token ID | 1654 |
Feature activation | -1.11e+01 |
Pos logit contributions | |
pieces | +0.143 |
hole | +0.138 |
on | +0.132 |
edges | +0.130 |
falling | +0.129 |
Neg logit contributions | |
Un | -0.125 |
David | -0.124 |
Jack | -0.123 |
Fro | -0.120 |
concession | -0.120 |
Token | she |
Token ID | 673 |
Feature activation | -8.21e+00 |
Pos logit contributions | |
hole | +0.155 |
pieces | +0.150 |
atop | +0.145 |
edges | +0.142 |
ps | +0.141 |
Neg logit contributions | |
Jack | -0.101 |
Un | -0.101 |
Fro | -0.100 |
David | -0.099 |
Kids | -0.095 |
Token | . |
Token ID | 13 |
Feature activation | +3.38e+00 |
Pos logit contributions | |
atop | +0.171 |
els | +0.169 |
ool | +0.164 |
pieces | +0.163 |
hole | +0.161 |
Neg logit contributions | |
Un | -0.159 |
Fro | -0.150 |
His | -0.148 |
uncle | -0.147 |
His | -0.143 |
Token | They |
Token ID | 1119 |
Feature activation | -3.57e+00 |
Pos logit contributions | |
concession | +0.032 |
David | +0.032 |
Jack | +0.030 |
Un | +0.030 |
oph | +0.030 |
Neg logit contributions | |
pieces | -0.047 |
on | -0.045 |
lid | -0.045 |
hole | -0.044 |
edges | -0.044 |
Token | had |
Token ID | 550 |
Feature activation | -5.96e+00 |
Pos logit contributions | |
hole | +0.047 |
pieces | +0.046 |
atop | +0.044 |
edges | +0.043 |
on | +0.043 |
Neg logit contributions | |
Un | -0.035 |
Jack | -0.035 |
Fro | -0.035 |
David | -0.035 |
Kids | -0.033 |
Token | to |
Token ID | 284 |
Feature activation | -5.44e+00 |
Pos logit contributions | |
hole | +0.077 |
pieces | +0.073 |
atop | +0.072 |
edges | +0.069 |
ps | +0.068 |
Neg logit contributions | |
Un | -0.062 |
Fro | -0.062 |
Jack | -0.061 |
David | -0.061 |
Jim | -0.058 |
Token | get |
Token ID | 651 |
Feature activation | -9.46e+00 |
Pos logit contributions | |
pieces | +0.058 |
on | +0.058 |
ps | +0.056 |
hole | +0.056 |
falling | +0.056 |
Neg logit contributions | |
Jack | -0.068 |
enny | -0.068 |
David | -0.067 |
concession | -0.065 |
oph | -0.065 |
Token | stitches |
Token ID | 28096 |
Feature activation | -2.06e+01 |
Pos logit contributions | |
hole | +0.099 |
pieces | +0.095 |
atop | +0.089 |
edges | +0.088 |
ps | +0.085 |
Neg logit contributions | |
Un | -0.121 |
Fro | -0.120 |
Jack | -0.120 |
David | -0.118 |
concession | -0.115 |
Token | and |
Token ID | 290 |
Feature activation | -5.41e+00 |
Pos logit contributions | |
edges | +0.228 |
hole | +0.228 |
atop | +0.226 |
lid | +0.225 |
pieces | +0.224 |
Neg logit contributions | |
Un | -0.213 |
Jack | -0.212 |
enny | -0.208 |
Fro | -0.208 |
David | -0.204 |
Token | shots |
Token ID | 6934 |
Feature activation | -1.77e+01 |
Pos logit contributions | |
pieces | +0.053 |
on | +0.049 |
hole | +0.049 |
falling | +0.047 |
edges | +0.046 |
Neg logit contributions | |
Jack | -0.076 |
David | -0.076 |
oph | -0.074 |
Jim | -0.074 |
Fro | -0.073 |
Token | . |
Token ID | 13 |
Feature activation | +2.84e+00 |
Pos logit contributions | |
hole | +0.203 |
hardest | +0.198 |
atop | +0.193 |
els | +0.192 |
edges | +0.192 |
Neg logit contributions | |
Un | -0.188 |
He | -0.186 |
Jack | -0.185 |
Fro | -0.184 |
His | -0.183 |
Token | They |
Token ID | 1119 |
Feature activation | -3.82e+00 |
Pos logit contributions | |
concession | +0.027 |
David | +0.026 |
oph | +0.025 |
Un | +0.025 |
Jim | +0.024 |
Neg logit contributions | |
pieces | -0.040 |
on | -0.039 |
edges | -0.038 |
hole | -0.038 |
lid | -0.037 |
Token | had |
Token ID | 550 |
Feature activation | -6.21e+00 |
Pos logit contributions | |
hole | +0.050 |
pieces | +0.049 |
atop | +0.047 |
edges | +0.046 |
on | +0.045 |
Neg logit contributions | |
Un | -0.038 |
Jack | -0.038 |
David | -0.038 |
Fro | -0.037 |
Kids | -0.036 |
Token | that |
Token ID | 326 |
Feature activation | +1.64e+01 |
Pos logit contributions | |
hole | +0.069 |
pieces | +0.068 |
atop | +0.065 |
edges | +0.064 |
ps | +0.064 |
Neg logit contributions | |
Jack | -0.046 |
Fro | -0.046 |
Un | -0.045 |
David | -0.045 |
Jim | -0.043 |
Token | were |
Token ID | 547 |
Feature activation | +1.07e+01 |
Pos logit contributions | |
Fro | +0.163 |
Jack | +0.162 |
Un | +0.155 |
David | +0.149 |
concession | +0.149 |
Neg logit contributions | |
hole | -0.225 |
atop | -0.202 |
ps | -0.200 |
pieces | -0.196 |
edges | -0.195 |
Token | all |
Token ID | 477 |
Feature activation | +1.06e+01 |
Pos logit contributions | |
Fro | +0.105 |
Un | +0.103 |
Jack | +0.102 |
please | +0.096 |
David | +0.096 |
Neg logit contributions | |
hole | -0.148 |
lid | -0.124 |
globe | -0.121 |
ool | -0.121 |
atop | -0.120 |
Token | mixed |
Token ID | 7668 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
Jack | +0.098 |
Jack | +0.092 |
He | +0.092 |
His | +0.091 |
concession | +0.090 |
Neg logit contributions | |
hole | -0.137 |
lid | -0.130 |
els | -0.128 |
plank | -0.122 |
atop | -0.121 |
Token | up |
Token ID | 510 |
Feature activation | +4.60e+00 |
Pos logit contributions | |
Jack | +0.015 |
Johnny | +0.014 |
They | +0.014 |
oph | +0.013 |
addy | +0.013 |
Neg logit contributions | |
hole | -0.013 |
code | -0.012 |
ps | -0.012 |
slot | -0.012 |
atop | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | +1.89e+01 |
Pos logit contributions | |
Jack | +0.053 |
Fro | +0.051 |
concession | +0.050 |
Un | +0.049 |
David | +0.049 |
Neg logit contributions | |
hole | -0.055 |
atop | -0.048 |
ps | -0.046 |
lid | -0.045 |
code | -0.045 |
Token | Jane |
Token ID | 12091 |
Feature activation | +1.12e+01 |
Pos logit contributions | |
Un | +0.149 |
David | +0.146 |
concession | +0.144 |
Jack | +0.141 |
Fro | +0.139 |
Neg logit contributions | |
hole | -0.274 |
pieces | -0.273 |
on | -0.264 |
atop | -0.256 |
edges | -0.255 |
Token | didn |
Token ID | 1422 |
Feature activation | +9.56e+00 |
Pos logit contributions | |
Jack | +0.105 |
Un | +0.097 |
David | +0.096 |
Fro | +0.095 |
Jim | +0.095 |
Neg logit contributions | |
hole | -0.155 |
pieces | -0.150 |
atop | -0.146 |
lid | -0.145 |
edges | -0.143 |
Token | 't |
Token ID | 470 |
Feature activation | +8.49e+00 |
Pos logit contributions | |
Un | +0.110 |
David | +0.109 |
Jack | +0.109 |
Fro | +0.109 |
Kids | +0.105 |
Neg logit contributions | |
pieces | -0.108 |
hole | -0.108 |
on | -0.102 |
atop | -0.100 |
falling | -0.099 |
Token | know |
Token ID | 760 |
Feature activation | +4.67e+00 |
Pos logit contributions | |
Jack | +0.068 |
David | +0.066 |
Fro | +0.065 |
Un | +0.064 |
Kids | +0.062 |
Neg logit contributions | |
hole | -0.131 |
pieces | -0.129 |
atop | -0.125 |
edges | -0.121 |
els | -0.119 |
Token | what |
Token ID | 644 |
Feature activation | +8.24e+00 |
Pos logit contributions | |
Jack | +0.033 |
Fro | +0.032 |
David | +0.032 |
Un | +0.031 |
Johnny | +0.031 |
Neg logit contributions | |
hole | -0.073 |
atop | -0.070 |
pieces | -0.070 |
ps | -0.070 |
edges | -0.069 |
Token |
|
Token ID | 198 |
Feature activation | +1.89e+01 |
Pos logit contributions | |
Un | +0.177 |
David | +0.176 |
oph | +0.172 |
Jack | +0.170 |
concession | +0.170 |
Neg logit contributions | |
hole | -0.331 |
on | -0.327 |
pieces | -0.323 |
atop | -0.307 |
trunk | -0.307 |
Token |
|
Token ID | 198 |
Feature activation | +1.84e+01 |
Pos logit contributions | |
oph | +0.195 |
concession | +0.194 |
Fro | +0.182 |
Un | +0.181 |
David | +0.179 |
Neg logit contributions | |
pieces | -0.223 |
hole | -0.217 |
on | -0.216 |
atop | -0.200 |
hard | -0.198 |
Token | â |
Token ID | 22940 |
Feature activation | +1.73e+01 |
Pos logit contributions | |
Fro | +0.173 |
oph | +0.169 |
concession | +0.164 |
latable | +0.152 |
ory | +0.150 |
Neg logit contributions | |
pieces | -0.224 |
hole | -0.222 |
on | -0.215 |
atop | -0.209 |
tangled | -0.202 |
Token | € |
Token ID | 26391 |
Feature activation | +1.91e+01 |
Pos logit contributions | |
Jack | +0.116 |
David | +0.112 |
Un | +0.105 |
Jim | +0.105 |
Kids | +0.105 |
Neg logit contributions | |
pieces | -0.306 |
hole | -0.281 |
lid | -0.280 |
on | -0.278 |
edges | -0.273 |
Token | � |
Token ID | 129 |
Feature activation | +1.39e+01 |
Pos logit contributions | |
concession | +0.181 |
Un | +0.180 |
Fro | +0.177 |
Kids | +0.176 |
Jack | +0.176 |
Neg logit contributions | |
hole | -0.258 |
pieces | -0.243 |
on | -0.237 |
atop | -0.233 |
falling | -0.233 |
Token | � |
Token ID | 241 |
Feature activation | +2.03e+01 |
Pos logit contributions | |
Un | +0.138 |
Jack | +0.137 |
David | +0.137 |
Fro | +0.135 |
Jim | +0.130 |
Neg logit contributions | |
pieces | -0.183 |
hole | -0.182 |
atop | -0.170 |
on | -0.170 |
edges | -0.168 |
Token | What |
Token ID | 2061 |
Feature activation | +1.22e+01 |
Pos logit contributions | |
Fro | +0.164 |
oph | +0.157 |
concession | +0.156 |
Un | +0.151 |
David | +0.149 |
Neg logit contributions | |
hole | -0.300 |
on | -0.293 |
pieces | -0.292 |
edges | -0.281 |
atop | -0.280 |
Token | â |
Token ID | 22940 |
Feature activation | +1.29e+01 |
Pos logit contributions | |
Un | +0.147 |
David | +0.146 |
Fro | +0.145 |
Jack | +0.145 |
concession | +0.140 |
Neg logit contributions | |
pieces | -0.133 |
hole | -0.132 |
on | -0.124 |
atop | -0.121 |
edges | -0.120 |
Token | € |
Token ID | 26391 |
Feature activation | +1.54e+01 |
Pos logit contributions | |
Jack | +0.080 |
David | +0.078 |
Kids | +0.074 |
Un | +0.073 |
Johnny | +0.072 |
Neg logit contributions | |
pieces | -0.236 |
hole | -0.223 |
edges | -0.216 |
lid | -0.215 |
matches | -0.212 |
Token | ™ |
Token ID | 8151 |
Feature activation | +1.08e+01 |
Pos logit contributions | |
concession | +0.134 |
Fro | +0.132 |
Un | +0.132 |
Jack | +0.128 |
Kids | +0.127 |
Neg logit contributions | |
hole | -0.212 |
pieces | -0.204 |
on | -0.201 |
atop | -0.196 |
falling | -0.194 |
Token | s |
Token ID | 82 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
Jack | +0.124 |
David | +0.122 |
Fro | +0.121 |
Un | +0.121 |
Jim | +0.117 |
Neg logit contributions | |
pieces | -0.125 |
hole | -0.123 |
edges | -0.117 |
ps | -0.116 |
atop | -0.116 |
Token |
|
Token ID | 198 |
Feature activation | +1.71e+01 |
Pos logit contributions | |
David | +0.160 |
oph | +0.154 |
Un | +0.153 |
Jack | +0.151 |
concession | +0.150 |
Neg logit contributions | |
on | -0.275 |
hole | -0.270 |
pieces | -0.268 |
trunk | -0.261 |
atop | -0.257 |
Token |
|
Token ID | 198 |
Feature activation | +1.74e+01 |
Pos logit contributions | |
oph | +0.182 |
concession | +0.179 |
David | +0.169 |
Fro | +0.167 |
coach | +0.167 |
Neg logit contributions | |
pieces | -0.198 |
on | -0.192 |
hole | -0.189 |
hard | -0.177 |
trunk | -0.176 |
Token | â |
Token ID | 22940 |
Feature activation | +1.61e+01 |
Pos logit contributions | |
Fro | +0.164 |
oph | +0.161 |
concession | +0.154 |
addy | +0.143 |
ory | +0.143 |
Neg logit contributions | |
pieces | -0.208 |
hole | -0.206 |
on | -0.201 |
atop | -0.197 |
trunk | -0.190 |
Token | € |
Token ID | 26391 |
Feature activation | +1.79e+01 |
Pos logit contributions | |
Jack | +0.101 |
David | +0.097 |
Un | +0.094 |
Jim | +0.093 |
Kids | +0.093 |
Neg logit contributions | |
pieces | -0.293 |
hole | -0.268 |
lid | -0.265 |
on | -0.265 |
matches | -0.260 |
Token | � |
Token ID | 129 |
Feature activation | +1.31e+01 |
Pos logit contributions | |
concession | +0.165 |
Un | +0.165 |
Jack | +0.163 |
Kids | +0.162 |
Fro | +0.161 |
Neg logit contributions | |
hole | -0.244 |
pieces | -0.231 |
on | -0.224 |
falling | -0.221 |
atop | -0.220 |
Token | � |
Token ID | 241 |
Feature activation | +1.97e+01 |
Pos logit contributions | |
Un | +0.123 |
David | +0.122 |
Jack | +0.121 |
Fro | +0.120 |
Jim | +0.115 |
Neg logit contributions | |
pieces | -0.180 |
hole | -0.179 |
atop | -0.167 |
on | -0.167 |
edges | -0.165 |
Token | Why |
Token ID | 5195 |
Feature activation | +1.34e+01 |
Pos logit contributions | |
Fro | +0.160 |
oph | +0.158 |
concession | +0.155 |
Un | +0.149 |
David | +0.148 |
Neg logit contributions | |
hole | -0.286 |
on | -0.281 |
pieces | -0.279 |
edges | -0.269 |
atop | -0.267 |
Token | don |
Token ID | 836 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
Un | +0.150 |
Jack | +0.149 |
David | +0.148 |
Fro | +0.147 |
concession | +0.142 |
Neg logit contributions | |
hole | -0.157 |
pieces | -0.157 |
on | -0.147 |
atop | -0.145 |
edges | -0.144 |
Token | â |
Token ID | 22940 |
Feature activation | +1.29e+01 |
Pos logit contributions | |
Fro | +0.136 |
David | +0.135 |
Un | +0.133 |
Jack | +0.133 |
oph | +0.132 |
Neg logit contributions | |
on | -0.098 |
pieces | -0.097 |
hole | -0.094 |
falling | -0.090 |
atop | -0.089 |
Token | € |
Token ID | 26391 |
Feature activation | +1.50e+01 |
Pos logit contributions | |
Jack | +0.071 |
David | +0.070 |
Un | +0.069 |
Fro | +0.066 |
Jim | +0.065 |
Neg logit contributions | |
pieces | -0.235 |
hole | -0.228 |
edges | -0.218 |
lid | -0.216 |
atop | -0.216 |
Token | ™ |
Token ID | 8151 |
Feature activation | +9.37e+00 |
Pos logit contributions | |
concession | +0.131 |
Jack | +0.128 |
Kids | +0.128 |
Fro | +0.128 |
Un | +0.128 |
Neg logit contributions | |
hole | -0.205 |
pieces | -0.197 |
on | -0.195 |
falling | -0.189 |
atop | -0.187 |
Token | dead |
Token ID | 2636 |
Feature activation | +8.73e+00 |
Pos logit contributions | |
pieces | +0.118 |
hole | +0.117 |
ps | +0.112 |
edges | +0.110 |
els | +0.107 |
Neg logit contributions | |
Fro | -0.167 |
Jack | -0.164 |
Un | -0.162 |
David | -0.160 |
Jim | -0.157 |
Token | squirrel |
Token ID | 33039 |
Feature activation | +5.96e+00 |
Pos logit contributions | |
Jack | +0.118 |
Fro | +0.118 |
Un | +0.117 |
David | +0.115 |
concession | +0.113 |
Neg logit contributions | |
hole | -0.081 |
pieces | -0.075 |
atop | -0.072 |
edges | -0.069 |
ps | -0.069 |
Token | in |
Token ID | 287 |
Feature activation | +1.11e+01 |
Pos logit contributions | |
Fro | +0.072 |
Jack | +0.071 |
Un | +0.069 |
David | +0.068 |
enny | +0.067 |
Neg logit contributions | |
hole | -0.063 |
pieces | -0.062 |
edges | -0.059 |
lid | -0.058 |
atop | -0.058 |
Token | the |
Token ID | 262 |
Feature activation | +1.30e+01 |
Pos logit contributions | |
Jack | +0.125 |
Fro | +0.124 |
Un | +0.124 |
David | +0.121 |
Jim | +0.119 |
Neg logit contributions | |
hole | -0.129 |
pieces | -0.122 |
atop | -0.119 |
edges | -0.115 |
hardest | -0.113 |
Token | grass |
Token ID | 8701 |
Feature activation | +4.95e+00 |
Pos logit contributions | |
Fro | +0.168 |
Un | +0.166 |
Jack | +0.165 |
concession | +0.164 |
David | +0.161 |
Neg logit contributions | |
hole | -0.133 |
pieces | -0.118 |
atop | -0.117 |
ps | -0.117 |
els | -0.116 |
Token | . |
Token ID | 13 |
Feature activation | +1.90e+01 |
Pos logit contributions | |
Fro | +0.065 |
Jack | +0.063 |
Un | +0.063 |
David | +0.062 |
Jim | +0.061 |
Neg logit contributions | |
pieces | -0.045 |
hole | -0.044 |
atop | -0.042 |
edges | -0.041 |
ps | -0.041 |
Token | She |
Token ID | 1375 |
Feature activation | +1.22e+01 |
Pos logit contributions | |
David | +0.151 |
Un | +0.151 |
concession | +0.149 |
Jack | +0.144 |
oph | +0.141 |
Neg logit contributions | |
hole | -0.268 |
pieces | -0.264 |
on | -0.262 |
trunk | -0.252 |
atop | -0.250 |
Token | stared |
Token ID | 18484 |
Feature activation | +3.37e+00 |
Pos logit contributions | |
Un | +0.132 |
Jack | +0.132 |
David | +0.131 |
Fro | +0.130 |
Jim | +0.125 |
Neg logit contributions | |
hole | -0.148 |
pieces | -0.148 |
atop | -0.138 |
on | -0.137 |
edges | -0.136 |
Token | at |
Token ID | 379 |
Feature activation | +8.26e+00 |
Pos logit contributions | |
Un | +0.047 |
Jack | +0.047 |
David | +0.047 |
Fro | +0.046 |
Kids | +0.045 |
Neg logit contributions | |
hole | -0.030 |
pieces | -0.030 |
atop | -0.028 |
on | -0.027 |
edges | -0.027 |
Token | it |
Token ID | 340 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
Un | +0.097 |
Jack | +0.097 |
Fro | +0.096 |
David | +0.096 |
Jim | +0.092 |
Neg logit contributions | |
hole | -0.093 |
pieces | -0.092 |
atop | -0.086 |
edges | -0.084 |
on | -0.084 |
Token | for |
Token ID | 329 |
Feature activation | +7.07e+00 |
Pos logit contributions | |
Fro | +0.019 |
Her | +0.019 |
please | +0.019 |
His | +0.019 |
She | +0.019 |
Neg logit contributions | |
bones | -0.033 |
plank | -0.032 |
slot | -0.032 |
hole | -0.032 |
pits | -0.031 |
Token | could |
Token ID | 714 |
Feature activation | +9.19e+00 |
Pos logit contributions | |
Jack | +0.021 |
Fro | +0.021 |
concession | +0.021 |
David | +0.020 |
Un | +0.020 |
Neg logit contributions | |
hole | -0.018 |
edges | -0.017 |
pieces | -0.017 |
ps | -0.017 |
els | -0.017 |
Token | be |
Token ID | 307 |
Feature activation | +5.32e+00 |
Pos logit contributions | |
Un | +0.090 |
Jack | +0.089 |
Fro | +0.088 |
David | +0.088 |
concession | +0.085 |
Neg logit contributions | |
hole | -0.119 |
pieces | -0.117 |
atop | -0.110 |
edges | -0.109 |
lid | -0.107 |
Token | felt |
Token ID | 2936 |
Feature activation | +7.09e+00 |
Pos logit contributions | |
David | +0.060 |
Jack | +0.059 |
Un | +0.058 |
Fro | +0.058 |
Da | +0.057 |
Neg logit contributions | |
hole | -0.063 |
lid | -0.061 |
button | -0.058 |
pieces | -0.056 |
edges | -0.055 |
Token | all |
Token ID | 477 |
Feature activation | +8.82e+00 |
Pos logit contributions | |
Jack | +0.071 |
Fro | +0.070 |
Un | +0.068 |
David | +0.067 |
concession | +0.066 |
Neg logit contributions | |
hole | -0.094 |
pieces | -0.087 |
els | -0.085 |
atop | -0.083 |
lid | -0.083 |
Token | around |
Token ID | 1088 |
Feature activation | +5.71e+00 |
Pos logit contributions | |
Jack | +0.103 |
Da | +0.097 |
Fro | +0.096 |
concession | +0.092 |
His | +0.092 |
Neg logit contributions | |
hole | -0.101 |
lid | -0.096 |
els | -0.093 |
code | -0.088 |
edges | -0.087 |
Token | . |
Token ID | 13 |
Feature activation | +1.90e+01 |
Pos logit contributions | |
Jack | +0.072 |
Fro | +0.069 |
Un | +0.068 |
David | +0.068 |
Jim | +0.067 |
Neg logit contributions | |
hole | -0.063 |
pieces | -0.057 |
lid | -0.054 |
ps | -0.054 |
edges | -0.053 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
Un | +0.166 |
concession | +0.161 |
David | +0.159 |
Fro | +0.153 |
Jack | +0.153 |
Neg logit contributions | |
hole | -0.258 |
pieces | -0.255 |
on | -0.244 |
edges | -0.237 |
atop | -0.236 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.32e+01 |
Pos logit contributions | |
Jack | +0.080 |
Fro | +0.075 |
David | +0.071 |
Jim | +0.070 |
Un | +0.070 |
Neg logit contributions | |
hole | -0.232 |
pieces | -0.228 |
ps | -0.223 |
edges | -0.222 |
falling | -0.221 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.61e+01 |
Pos logit contributions | |
Un | +0.126 |
David | +0.124 |
Jack | +0.123 |
Fro | +0.121 |
concession | +0.118 |
Neg logit contributions | |
hole | -0.178 |
pieces | -0.177 |
on | -0.167 |
atop | -0.166 |
edges | -0.164 |
Token | a |
Token ID | 257 |
Feature activation | +1.40e+01 |
Pos logit contributions | |
Kids | +0.310 |
ete | +0.302 |
oph | +0.302 |
enny | +0.296 |
concession | +0.296 |
Neg logit contributions | |
on | -0.104 |
pieces | -0.097 |
hard | -0.079 |
broken | -0.068 |
glowing | -0.065 |
Token | time |
Token ID | 640 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
concession | +0.281 |
David | +0.279 |
Kids | +0.277 |
Johnny | +0.274 |
Fro | +0.272 |
Neg logit contributions | |
hard | -0.040 |
pieces | -0.037 |
on | -0.031 |
falling | -0.029 |
edges | -0.027 |
Token | for |
Token ID | 329 |
Feature activation | +8.56e+00 |
Pos logit contributions | |
David | +0.033 |
Fro | +0.031 |
Jack | +0.030 |
enny | +0.030 |
Un | +0.030 |
Neg logit contributions | |
els | -0.102 |
pieces | -0.099 |
matches | -0.097 |
globe | -0.094 |
bulb | -0.094 |
Token | dinner |
Token ID | 8073 |
Feature activation | -3.08e-01 |
Pos logit contributions | |
Fro | +0.082 |
Jack | +0.079 |
Un | +0.079 |
David | +0.078 |
Jim | +0.073 |
Neg logit contributions | |
hole | -0.125 |
atop | -0.114 |
pieces | -0.113 |
edges | -0.109 |
els | -0.109 |
Token | . |
Token ID | 13 |
Feature activation | +1.81e+01 |
Pos logit contributions | |
hole | +0.003 |
pieces | +0.003 |
atop | +0.003 |
ps | +0.002 |
els | +0.002 |
Neg logit contributions | |
Jack | -0.004 |
Un | -0.004 |
David | -0.004 |
Fro | -0.004 |
Jim | -0.004 |
Token | She |
Token ID | 1375 |
Feature activation | +1.14e+01 |
Pos logit contributions | |
Un | +0.151 |
concession | +0.148 |
David | +0.148 |
Fro | +0.138 |
Jack | +0.138 |
Neg logit contributions | |
hole | -0.247 |
pieces | -0.246 |
on | -0.238 |
edges | -0.227 |
trunk | -0.227 |
Token | felt |
Token ID | 2936 |
Feature activation | +7.01e+00 |
Pos logit contributions | |
Un | +0.124 |
Jack | +0.124 |
David | +0.124 |
Fro | +0.122 |
Jim | +0.117 |
Neg logit contributions | |
pieces | -0.137 |
hole | -0.136 |
on | -0.127 |
atop | -0.126 |
edges | -0.125 |
Token | a |
Token ID | 257 |
Feature activation | +8.67e+00 |
Pos logit contributions | |
Un | +0.085 |
Jack | +0.085 |
David | +0.084 |
Fro | +0.084 |
concession | +0.081 |
Neg logit contributions | |
hole | -0.076 |
pieces | -0.075 |
atop | -0.070 |
edges | -0.069 |
on | -0.069 |
Token | wave |
Token ID | 6769 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
Un | +0.100 |
Jack | +0.100 |
David | +0.096 |
Fro | +0.096 |
concession | +0.094 |
Neg logit contributions | |
hole | -0.100 |
pieces | -0.097 |
atop | -0.092 |
ps | -0.091 |
els | -0.090 |
Token | of |
Token ID | 286 |
Feature activation | +6.81e+00 |
Pos logit contributions | |
Un | +0.012 |
Jack | +0.012 |
Fro | +0.011 |
David | +0.011 |
concession | +0.011 |
Neg logit contributions | |
pieces | -0.012 |
hole | -0.011 |
edges | -0.011 |
atop | -0.011 |
els | -0.011 |
Token | relief |
Token ID | 8259 |
Feature activation | +2.76e-01 |
Pos logit contributions | |
Jack | +0.068 |
Fro | +0.068 |
David | +0.066 |
Jim | +0.065 |
concession | +0.065 |
Neg logit contributions | |
hole | -0.092 |
pieces | -0.082 |
atop | -0.080 |
edges | -0.079 |
els | -0.078 |
Token | as |
Token ID | 355 |
Feature activation | +8.36e+00 |
Pos logit contributions | |
Un | +0.003 |
Fro | +0.003 |
Jack | +0.003 |
David | +0.003 |
concession | +0.003 |
Neg logit contributions | |
hole | -0.003 |
pieces | -0.003 |
hardest | -0.003 |
edges | -0.003 |
els | -0.003 |
Token | their |
Token ID | 511 |
Feature activation | +7.65e+00 |
Pos logit contributions | |
Jack | +0.060 |
Un | +0.060 |
David | +0.059 |
Fro | +0.058 |
Jim | +0.055 |
Neg logit contributions | |
hole | -0.133 |
pieces | -0.131 |
atop | -0.124 |
edges | -0.124 |
ps | -0.123 |
Token | her |
Token ID | 607 |
Feature activation | +3.46e+00 |
Pos logit contributions | |
Fro | +0.021 |
Jack | +0.019 |
Da | +0.018 |
Un | +0.018 |
Her | +0.018 |
Neg logit contributions | |
hole | -0.031 |
atop | -0.029 |
pieces | -0.027 |
lid | -0.027 |
edges | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | +1.31e+01 |
Pos logit contributions | |
Fro | +0.035 |
uncle | +0.030 |
wife | +0.029 |
coach | +0.029 |
Pete | +0.026 |
Neg logit contributions | |
hole | -0.047 |
bones | -0.046 |
atop | -0.045 |
ps | -0.045 |
ool | -0.043 |
Token | She |
Token ID | 1375 |
Feature activation | +7.85e+00 |
Pos logit contributions | |
concession | +0.126 |
David | +0.113 |
Jack | +0.109 |
oph | +0.109 |
Jim | +0.105 |
Neg logit contributions | |
hole | -0.183 |
on | -0.180 |
pieces | -0.173 |
trunk | -0.170 |
edges | -0.168 |
Token | ran |
Token ID | 4966 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
David | +0.099 |
Jack | +0.096 |
Un | +0.096 |
concession | +0.093 |
Fro | +0.093 |
Neg logit contributions | |
pieces | -0.085 |
hole | -0.081 |
on | -0.079 |
ps | -0.076 |
tangled | -0.075 |
Token | after |
Token ID | 706 |
Feature activation | +4.52e+00 |
Pos logit contributions | |
David | +0.015 |
Un | +0.015 |
Jack | +0.015 |
Fro | +0.014 |
Jim | +0.014 |
Neg logit contributions | |
pieces | -0.011 |
hole | -0.011 |
on | -0.010 |
edges | -0.010 |
atop | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | +5.72e+00 |
Pos logit contributions | |
Un | +0.053 |
David | +0.053 |
Jack | +0.052 |
Fro | +0.051 |
concession | +0.050 |
Neg logit contributions | |
pieces | -0.051 |
hole | -0.050 |
on | -0.048 |
atop | -0.047 |
falling | -0.046 |
Token | dog |
Token ID | 3290 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
Fro | +0.037 |
Un | +0.035 |
Jack | +0.035 |
concession | +0.034 |
Kids | +0.034 |
Neg logit contributions | |
hole | -0.098 |
atop | -0.094 |
els | -0.092 |
ps | -0.091 |
pieces | -0.089 |
Token | and |
Token ID | 290 |
Feature activation | +6.07e+00 |
Pos logit contributions | |
hole | +0.018 |
atop | +0.018 |
pieces | +0.018 |
edges | +0.018 |
ool | +0.017 |
Neg logit contributions | |
Fro | -0.015 |
Un | -0.012 |
His | -0.012 |
Her | -0.012 |
Jack | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | +5.57e+00 |
Pos logit contributions | |
Fro | +0.068 |
Un | +0.067 |
Jack | +0.067 |
David | +0.066 |
Kids | +0.063 |
Neg logit contributions | |
hole | -0.073 |
pieces | -0.071 |
atop | -0.068 |
edges | -0.067 |
ps | -0.065 |
Token | man |
Token ID | 582 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
Fro | +0.041 |
Un | +0.040 |
Jack | +0.039 |
David | +0.038 |
concession | +0.038 |
Neg logit contributions | |
hole | -0.090 |
atop | -0.085 |
els | -0.083 |
ps | -0.082 |
pieces | -0.081 |
Token | and |
Token ID | 290 |
Feature activation | +6.18e+00 |
Pos logit contributions | |
edges | +0.017 |
hardest | +0.017 |
atop | +0.016 |
hole | +0.016 |
ool | +0.016 |
Neg logit contributions | |
Fro | -0.013 |
His | -0.012 |
His | -0.011 |
Her | -0.011 |
him | -0.011 |
Token | " |
Token ID | 1 |
Feature activation | +1.44e+01 |
Pos logit contributions | |
Fro | +0.161 |
concession | +0.153 |
oph | +0.150 |
enny | +0.136 |
David | +0.135 |
Neg logit contributions | |
pieces | -0.192 |
hole | -0.192 |
on | -0.185 |
atop | -0.178 |
tangled | -0.178 |
Token | Please |
Token ID | 5492 |
Feature activation | +9.45e+00 |
Pos logit contributions | |
Fro | +0.107 |
David | +0.105 |
concession | +0.105 |
enny | +0.102 |
ete | +0.101 |
Neg logit contributions | |
pieces | -0.230 |
tangled | -0.215 |
on | -0.213 |
edges | -0.210 |
ps | -0.209 |
Token | , |
Token ID | 11 |
Feature activation | +1.00e+01 |
Pos logit contributions | |
Fro | +0.119 |
Jack | +0.116 |
Un | +0.113 |
David | +0.111 |
Kids | +0.109 |
Neg logit contributions | |
hole | -0.105 |
pieces | -0.101 |
atop | -0.096 |
edges | -0.092 |
ps | -0.089 |
Token | Lily |
Token ID | 20037 |
Feature activation | +6.18e+00 |
Pos logit contributions | |
Un | +0.084 |
David | +0.084 |
Jack | +0.082 |
Fro | +0.077 |
concession | +0.077 |
Neg logit contributions | |
pieces | -0.146 |
hole | -0.145 |
on | -0.141 |
lid | -0.137 |
falling | -0.137 |
Token | , |
Token ID | 11 |
Feature activation | +9.26e+00 |
Pos logit contributions | |
Un | +0.087 |
Jack | +0.087 |
Fro | +0.086 |
David | +0.086 |
Jim | +0.083 |
Neg logit contributions | |
hole | -0.055 |
pieces | -0.055 |
atop | -0.050 |
edges | -0.049 |
on | -0.048 |
Token | you |
Token ID | 345 |
Feature activation | +5.75e+00 |
Pos logit contributions | |
David | +0.102 |
Un | +0.101 |
Jack | +0.099 |
Fro | +0.095 |
Da | +0.094 |
Neg logit contributions | |
pieces | -0.112 |
hole | -0.110 |
on | -0.110 |
falling | -0.106 |
ps | -0.104 |
Token | have |
Token ID | 423 |
Feature activation | +4.88e+00 |
Pos logit contributions | |
Fro | +0.065 |
Jack | +0.063 |
Jim | +0.058 |
enny | +0.058 |
Un | +0.058 |
Neg logit contributions | |
hole | -0.072 |
atop | -0.068 |
pieces | -0.066 |
bones | -0.065 |
edges | -0.064 |
Token | to |
Token ID | 284 |
Feature activation | +5.44e+00 |
Pos logit contributions | |
Jack | +0.047 |
Un | +0.046 |
Fro | +0.046 |
David | +0.046 |
Jim | +0.044 |
Neg logit contributions | |
hole | -0.067 |
pieces | -0.063 |
atop | -0.062 |
ps | -0.060 |
edges | -0.059 |
Token | take |
Token ID | 1011 |
Feature activation | -6.21e-01 |
Pos logit contributions | |
Un | +0.063 |
David | +0.062 |
Jack | +0.061 |
concession | +0.060 |
Fro | +0.059 |
Neg logit contributions | |
pieces | -0.063 |
hole | -0.061 |
on | -0.059 |
ps | -0.058 |
falling | -0.058 |
Token | the |
Token ID | 262 |
Feature activation | +4.30e+00 |
Pos logit contributions | |
hole | +0.009 |
atop | +0.009 |
globe | +0.008 |
pieces | +0.008 |
changing | +0.008 |
Neg logit contributions | |
Jack | -0.006 |
Fro | -0.006 |
Un | -0.005 |
Da | -0.005 |
Jim | -0.005 |
Token | pill |
Token ID | 9582 |
Feature activation | -3.68e+00 |
Pos logit contributions | |
Fro | +0.051 |
coach | +0.048 |
concession | +0.047 |
wife | +0.046 |
shepherd | +0.045 |
Neg logit contributions | |
hole | -0.053 |
atop | -0.047 |
cks | -0.044 |
els | -0.044 |
bones | -0.043 |
Token | kept |
Token ID | 4030 |
Feature activation | +1.08e+01 |
Pos logit contributions | |
hole | +0.159 |
pieces | +0.158 |
atop | +0.146 |
edges | +0.145 |
ps | +0.144 |
Neg logit contributions | |
Fro | -0.206 |
Un | -0.205 |
Jack | -0.205 |
David | -0.204 |
concession | -0.197 |
Token | asking |
Token ID | 4737 |
Feature activation | +6.54e+00 |
Pos logit contributions | |
Jack | +0.130 |
Fro | +0.130 |
David | +0.129 |
Un | +0.127 |
Johnny | +0.123 |
Neg logit contributions | |
hole | -0.120 |
pieces | -0.113 |
edges | -0.106 |
lid | -0.105 |
atop | -0.105 |
Token | until |
Token ID | 1566 |
Feature activation | +1.20e+01 |
Pos logit contributions | |
Jack | +0.076 |
Fro | +0.074 |
David | +0.074 |
Un | +0.074 |
His | +0.071 |
Neg logit contributions | |
hole | -0.075 |
pieces | -0.073 |
atop | -0.069 |
edges | -0.068 |
falling | -0.067 |
Token | finally |
Token ID | 3443 |
Feature activation | +1.07e+01 |
Pos logit contributions | |
Jack | +0.096 |
David | +0.094 |
Un | +0.094 |
Fro | +0.094 |
Jim | +0.089 |
Neg logit contributions | |
hole | -0.184 |
pieces | -0.179 |
edges | -0.170 |
ps | -0.169 |
atop | -0.168 |
Token | , |
Token ID | 11 |
Feature activation | +1.24e+01 |
Pos logit contributions | |
Jack | +0.112 |
David | +0.111 |
Fro | +0.109 |
Un | +0.109 |
Jim | +0.106 |
Neg logit contributions | |
hole | -0.135 |
pieces | -0.134 |
edges | -0.126 |
atop | -0.123 |
ps | -0.122 |
Token | his |
Token ID | 465 |
Feature activation | +1.24e+01 |
Pos logit contributions | |
Jack | +0.127 |
David | +0.127 |
Un | +0.126 |
Fro | +0.124 |
Jim | +0.120 |
Neg logit contributions | |
hole | -0.158 |
pieces | -0.156 |
edges | -0.147 |
ps | -0.146 |
atop | -0.145 |
Token | parents |
Token ID | 3397 |
Feature activation | +7.72e+00 |
Pos logit contributions | |
Un | +0.102 |
Fro | +0.101 |
Jack | +0.101 |
David | +0.101 |
concession | +0.096 |
Neg logit contributions | |
hole | -0.182 |
pieces | -0.177 |
atop | -0.172 |
edges | -0.166 |
ps | -0.166 |
Token | agreed |
Token ID | 4987 |
Feature activation | +4.72e+00 |
Pos logit contributions | |
David | +0.077 |
Jack | +0.077 |
Un | +0.073 |
Johnny | +0.071 |
Jim | +0.070 |
Neg logit contributions | |
pieces | -0.102 |
edges | -0.101 |
hole | -0.100 |
atop | -0.095 |
ps | -0.094 |
Token | to |
Token ID | 284 |
Feature activation | +8.13e+00 |
Pos logit contributions | |
Jack | +0.055 |
David | +0.054 |
Un | +0.053 |
Fro | +0.052 |
Jim | +0.051 |
Neg logit contributions | |
hole | -0.055 |
pieces | -0.053 |
atop | -0.050 |
edges | -0.050 |
ps | -0.049 |
Token | help |
Token ID | 1037 |
Feature activation | +2.94e+00 |
Pos logit contributions | |
Un | +0.088 |
Jack | +0.087 |
David | +0.087 |
Fro | +0.086 |
concession | +0.083 |
Neg logit contributions | |
hole | -0.098 |
pieces | -0.098 |
atop | -0.091 |
on | -0.091 |
edges | -0.090 |
Token | him |
Token ID | 683 |
Feature activation | +3.83e+00 |
Pos logit contributions | |
Jack | +0.027 |
Un | +0.027 |
David | +0.027 |
Fro | +0.026 |
Jim | +0.025 |
Neg logit contributions | |
hole | -0.041 |
pieces | -0.040 |
atop | -0.038 |
edges | -0.038 |
ps | -0.037 |
Token | , |
Token ID | 11 |
Feature activation | +1.45e+01 |
Pos logit contributions | |
Un | +0.193 |
David | +0.189 |
Jack | +0.189 |
Kids | +0.185 |
concession | +0.185 |
Neg logit contributions | |
on | -0.112 |
pieces | -0.111 |
hole | -0.110 |
atop | -0.102 |
ps | -0.101 |
Token | " |
Token ID | 366 |
Feature activation | +1.82e+01 |
Pos logit contributions | |
Un | +0.136 |
Jack | +0.135 |
David | +0.134 |
Fro | +0.133 |
concession | +0.127 |
Neg logit contributions | |
hole | -0.195 |
pieces | -0.195 |
on | -0.183 |
atop | -0.183 |
edges | -0.181 |
Token | Yes |
Token ID | 5297 |
Feature activation | +1.13e+01 |
Pos logit contributions | |
Fro | +0.167 |
concession | +0.156 |
oph | +0.145 |
addy | +0.143 |
ete | +0.141 |
Neg logit contributions | |
hole | -0.233 |
pieces | -0.231 |
atop | -0.221 |
on | -0.220 |
edges | -0.217 |
Token | , |
Token ID | 11 |
Feature activation | +1.29e+01 |
Pos logit contributions | |
Jack | +0.148 |
Fro | +0.146 |
David | +0.146 |
Un | +0.146 |
Jim | +0.142 |
Neg logit contributions | |
hole | -0.111 |
pieces | -0.110 |
atop | -0.103 |
edges | -0.099 |
ps | -0.098 |
Token | m |
Token ID | 285 |
Feature activation | +1.06e+01 |
Pos logit contributions | |
Un | +0.120 |
David | +0.119 |
Jack | +0.117 |
Fro | +0.114 |
concession | +0.113 |
Neg logit contributions | |
pieces | -0.174 |
hole | -0.172 |
on | -0.170 |
edges | -0.160 |
falling | -0.159 |
Token | ama |
Token ID | 1689 |
Feature activation | +7.27e+00 |
Pos logit contributions | |
Un | +0.091 |
Jack | +0.090 |
David | +0.090 |
Fro | +0.089 |
concession | +0.085 |
Neg logit contributions | |
pieces | -0.154 |
hole | -0.154 |
atop | -0.145 |
edges | -0.144 |
on | -0.144 |
Token | . |
Token ID | 13 |
Feature activation | +1.59e+01 |
Pos logit contributions | |
Fro | +0.095 |
Un | +0.094 |
Jack | +0.089 |
David | +0.089 |
Kids | +0.087 |
Neg logit contributions | |
hole | -0.073 |
pieces | -0.071 |
edges | -0.068 |
atop | -0.067 |
tangled | -0.067 |
Token | I |
Token ID | 314 |
Feature activation | +1.07e+01 |
Pos logit contributions | |
David | +0.154 |
Jack | +0.152 |
Un | +0.147 |
Fro | +0.142 |
Jim | +0.142 |
Neg logit contributions | |
hole | -0.213 |
pieces | -0.205 |
on | -0.200 |
edges | -0.193 |
atop | -0.193 |
Token | can |
Token ID | 460 |
Feature activation | +8.61e+00 |
Pos logit contributions | |
Jack | +0.097 |
Fro | +0.096 |
Un | +0.095 |
David | +0.093 |
Kids | +0.090 |
Neg logit contributions | |
hole | -0.151 |
pieces | -0.150 |
atop | -0.142 |
edges | -0.141 |
falling | -0.137 |
Token | be |
Token ID | 307 |
Feature activation | +4.04e+00 |
Pos logit contributions | |
Un | +0.105 |
David | +0.104 |
Jack | +0.104 |
Fro | +0.103 |
Kids | +0.100 |
Neg logit contributions | |
hole | -0.093 |
pieces | -0.093 |
on | -0.087 |
edges | -0.085 |
atop | -0.085 |
Token | quiet |
Token ID | 5897 |
Feature activation | -8.84e-01 |
Pos logit contributions | |
Fro | +0.046 |
Un | +0.046 |
Jack | +0.046 |
David | +0.046 |
Jim | +0.044 |
Neg logit contributions | |
hole | -0.048 |
pieces | -0.046 |
atop | -0.043 |
edges | -0.043 |
lid | -0.043 |
Token | He |
Token ID | 679 |
Feature activation | +6.18e+00 |
Pos logit contributions | |
David | +0.075 |
Fro | +0.071 |
Un | +0.071 |
Kids | +0.070 |
Jack | +0.070 |
Neg logit contributions | |
pieces | -0.209 |
hole | -0.206 |
on | -0.197 |
edges | -0.196 |
atop | -0.193 |
Token | understood |
Token ID | 7247 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
Jack | +0.064 |
Un | +0.061 |
Fro | +0.060 |
David | +0.059 |
Jim | +0.058 |
Neg logit contributions | |
hole | -0.082 |
atop | -0.075 |
pieces | -0.074 |
falling | -0.074 |
edges | -0.073 |
Token | that |
Token ID | 326 |
Feature activation | +3.76e+00 |
Pos logit contributions | |
Jack | +0.017 |
Un | +0.017 |
David | +0.017 |
Fro | +0.017 |
Jim | +0.016 |
Neg logit contributions | |
hole | -0.023 |
pieces | -0.022 |
edges | -0.021 |
atop | -0.021 |
ps | -0.020 |
Token | it |
Token ID | 340 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
Jack | +0.031 |
Un | +0.030 |
Fro | +0.030 |
David | +0.029 |
Jim | +0.028 |
Neg logit contributions | |
hole | -0.058 |
ps | -0.054 |
atop | -0.053 |
lid | -0.052 |
els | -0.052 |
Token | was |
Token ID | 373 |
Feature activation | +5.85e-01 |
Pos logit contributions | |
hole | +0.003 |
atop | +0.003 |
pieces | +0.003 |
edges | +0.003 |
lid | +0.003 |
Neg logit contributions | |
Jack | -0.002 |
Un | -0.002 |
Fro | -0.002 |
concession | -0.002 |
David | -0.002 |
Token | much |
Token ID | 881 |
Feature activation | -6.76e-01 |
Pos logit contributions | |
Un | +0.005 |
Jack | +0.005 |
His | +0.005 |
Fro | +0.005 |
David | +0.005 |
Neg logit contributions | |
hole | -0.009 |
lid | -0.008 |
globe | -0.008 |
pieces | -0.007 |
els | -0.007 |
Token | too |
Token ID | 1165 |
Feature activation | +8.07e-01 |
Pos logit contributions | |
hole | +0.007 |
ps | +0.006 |
atop | +0.006 |
globe | +0.006 |
edges | +0.006 |
Neg logit contributions | |
Jack | -0.009 |
Un | -0.008 |
Fro | -0.008 |
concession | -0.008 |
David | -0.008 |
Token | cold |
Token ID | 4692 |
Feature activation | -5.80e+00 |
Pos logit contributions | |
Un | +0.010 |
Jack | +0.010 |
David | +0.009 |
He | +0.009 |
Johnny | +0.009 |
Neg logit contributions | |
hole | -0.010 |
bones | -0.008 |
els | -0.008 |
ps | -0.008 |
atop | -0.008 |
Token | outside |
Token ID | 2354 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
hole | +0.067 |
pieces | +0.065 |
atop | +0.062 |
lid | +0.061 |
edges | +0.060 |
Neg logit contributions | |
Jack | -0.065 |
Un | -0.065 |
Fro | -0.063 |
David | -0.063 |
concession | -0.062 |
Token | , |
Token ID | 11 |
Feature activation | +3.75e+00 |
Pos logit contributions | |
hole | +0.022 |
pieces | +0.020 |
atop | +0.020 |
ps | +0.020 |
edges | +0.019 |
Neg logit contributions | |
Jack | -0.022 |
Un | -0.021 |
Fro | -0.021 |
David | -0.021 |
concession | -0.021 |
Token | and |
Token ID | 290 |
Feature activation | +5.35e+00 |
Pos logit contributions | |
Un | +0.041 |
Jack | +0.041 |
David | +0.041 |
Fro | +0.040 |
Kids | +0.039 |
Neg logit contributions | |
pieces | -0.045 |
hole | -0.045 |
on | -0.042 |
atop | -0.041 |
edges | -0.041 |
Token | was |
Token ID | 373 |
Feature activation | -8.40e-01 |
Pos logit contributions | |
David | +0.021 |
Un | +0.020 |
Jack | +0.020 |
oph | +0.020 |
Jim | +0.020 |
Neg logit contributions | |
on | -0.021 |
pieces | -0.021 |
hole | -0.020 |
falling | -0.020 |
ps | -0.020 |
Token | scared |
Token ID | 12008 |
Feature activation | -7.50e+00 |
Pos logit contributions | |
pieces | +0.008 |
hole | +0.008 |
on | +0.008 |
falling | +0.008 |
atop | +0.008 |
Neg logit contributions | |
Jack | -0.011 |
David | -0.011 |
Un | -0.011 |
Fro | -0.010 |
Kids | -0.010 |
Token | of |
Token ID | 286 |
Feature activation | -4.01e+00 |
Pos logit contributions | |
pieces | +0.078 |
hole | +0.077 |
on | +0.073 |
edges | +0.071 |
atop | +0.070 |
Neg logit contributions | |
David | -0.096 |
Un | -0.095 |
Jack | -0.095 |
Fro | -0.093 |
concession | -0.092 |
Token | the |
Token ID | 262 |
Feature activation | -3.25e+00 |
Pos logit contributions | |
hole | +0.055 |
atop | +0.052 |
ool | +0.049 |
lid | +0.049 |
pieces | +0.049 |
Neg logit contributions | |
Fro | -0.043 |
Un | -0.041 |
Jack | -0.039 |
David | -0.038 |
Johnny | -0.037 |
Token | dark |
Token ID | 3223 |
Feature activation | -8.90e+00 |
Pos logit contributions | |
hole | +0.042 |
atop | +0.038 |
els | +0.037 |
cks | +0.036 |
ps | +0.035 |
Neg logit contributions | |
Fro | -0.038 |
concession | -0.036 |
Un | -0.036 |
coach | -0.035 |
Kids | -0.035 |
Token | , |
Token ID | 11 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
hole | +0.086 |
pieces | +0.082 |
atop | +0.082 |
ps | +0.078 |
hardest | +0.075 |
Neg logit contributions | |
Un | -0.116 |
Jack | -0.115 |
Fro | -0.113 |
David | -0.112 |
Kids | -0.111 |
Token | loud |
Token ID | 7812 |
Feature activation | -7.36e+00 |
Pos logit contributions | |
hole | +0.014 |
pieces | +0.013 |
els | +0.013 |
ps | +0.013 |
atop | +0.013 |
Neg logit contributions | |
Fro | -0.012 |
Un | -0.012 |
Jack | -0.012 |
David | -0.011 |
His | -0.011 |
Token | noises |
Token ID | 26782 |
Feature activation | -1.04e+01 |
Pos logit contributions | |
hole | +0.089 |
atop | +0.083 |
pieces | +0.081 |
hardest | +0.077 |
els | +0.077 |
Neg logit contributions | |
Un | -0.083 |
Fro | -0.082 |
Jack | -0.080 |
concession | -0.078 |
Kids | -0.078 |
Token | , |
Token ID | 11 |
Feature activation | -9.04e-01 |
Pos logit contributions | |
hole | +0.136 |
lid | +0.134 |
hardest | +0.130 |
pieces | +0.124 |
ool | +0.123 |
Neg logit contributions | |
Un | -0.108 |
Jack | -0.107 |
Fro | -0.107 |
him | -0.106 |
His | -0.100 |
Token | and |
Token ID | 290 |
Feature activation | +7.32e-01 |
Pos logit contributions | |
hole | +0.012 |
pieces | +0.011 |
atop | +0.011 |
edges | +0.011 |
lid | +0.011 |
Neg logit contributions | |
Fro | -0.009 |
Un | -0.009 |
Jack | -0.009 |
David | -0.009 |
concession | -0.008 |
Token | even |
Token ID | 772 |
Feature activation | -3.86e+00 |
Pos logit contributions | |
David | +0.006 |
Un | +0.006 |
Jack | +0.006 |
Fro | +0.006 |
oph | +0.006 |
Neg logit contributions | |
pieces | -0.011 |
hole | -0.010 |
on | -0.010 |
falling | -0.010 |
edges | -0.010 |
Token | € |
Token ID | 26391 |
Feature activation | +6.76e+00 |
Pos logit contributions | |
Jack | +0.039 |
David | +0.037 |
Kids | +0.035 |
Un | +0.035 |
Johnny | +0.034 |
Neg logit contributions | |
pieces | -0.099 |
hole | -0.092 |
lid | -0.092 |
atop | -0.090 |
on | -0.090 |
Token | Jill |
Token ID | 22772 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
Un | +0.065 |
Kids | +0.065 |
Da | +0.063 |
concession | +0.063 |
Fro | +0.063 |
Neg logit contributions | |
hole | -0.091 |
pieces | -0.090 |
on | -0.085 |
edges | -0.084 |
falling | -0.082 |
Token | laughed |
Token ID | 13818 |
Feature activation | -3.48e+00 |
Pos logit contributions | |
Jack | +0.016 |
Fro | +0.015 |
David | +0.015 |
Un | +0.014 |
Jim | +0.014 |
Neg logit contributions | |
hole | -0.023 |
pieces | -0.023 |
atop | -0.022 |
edges | -0.022 |
ps | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | +4.72e+00 |
Pos logit contributions | |
on | +0.031 |
pieces | +0.029 |
hole | +0.027 |
atop | +0.026 |
edges | +0.025 |
Neg logit contributions | |
Kids | -0.052 |
David | -0.052 |
Un | -0.052 |
concession | -0.051 |
Jack | -0.050 |
Token | said |
Token ID | 531 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
Un | +0.051 |
Jack | +0.051 |
David | +0.051 |
Fro | +0.050 |
concession | +0.049 |
Neg logit contributions | |
hole | -0.057 |
pieces | -0.057 |
on | -0.053 |
atop | -0.053 |
edges | -0.052 |
Token | , |
Token ID | 11 |
Feature activation | +3.43e+00 |
Pos logit contributions | |
Jack | +0.021 |
Un | +0.021 |
Fro | +0.020 |
David | +0.020 |
Jim | +0.020 |
Neg logit contributions | |
hole | -0.014 |
pieces | -0.014 |
atop | -0.012 |
edges | -0.012 |
on | -0.012 |
Token | � |
Token ID | 6184 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
Un | +0.029 |
Jack | +0.029 |
David | +0.029 |
Fro | +0.029 |
Jim | +0.027 |
Neg logit contributions | |
hole | -0.051 |
pieces | -0.050 |
atop | -0.047 |
edges | -0.047 |
ps | -0.047 |
Token | � |
Token ID | 95 |
Feature activation | +4.08e+00 |
Pos logit contributions | |
Un | +0.029 |
David | +0.029 |
Jack | +0.029 |
Fro | +0.028 |
Da | +0.028 |
Neg logit contributions | |
hole | -0.034 |
pieces | -0.033 |
atop | -0.032 |
on | -0.032 |
edges | -0.031 |
Token | € |
Token ID | 26391 |
Feature activation | +5.54e+00 |
Pos logit contributions | |
Un | +0.039 |
Fro | +0.038 |
Jack | +0.038 |
David | +0.038 |
concession | +0.036 |
Neg logit contributions | |
hole | -0.055 |
pieces | -0.054 |
edges | -0.051 |
atop | -0.051 |
on | -0.050 |
Token | � |
Token ID | 129 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
Kids | +0.064 |
Da | +0.063 |
Fro | +0.063 |
concession | +0.062 |
Un | +0.062 |
Neg logit contributions | |
hole | -0.064 |
pieces | -0.063 |
on | -0.060 |
edges | -0.057 |
falling | -0.056 |
Token | � |
Token ID | 241 |
Feature activation | +9.03e+00 |
Pos logit contributions | |
Un | +0.013 |
Jack | +0.013 |
David | +0.013 |
Fro | +0.013 |
Kids | +0.012 |
Neg logit contributions | |
hole | -0.024 |
pieces | -0.024 |
atop | -0.023 |
on | -0.022 |
edges | -0.022 |
Token | It |
Token ID | 632 |
Feature activation | +5.81e+00 |
Pos logit contributions | |
concession | +0.135 |
Un | +0.133 |
David | +0.131 |
Jack | +0.127 |
oph | +0.124 |
Neg logit contributions | |
pieces | -0.201 |
hole | -0.201 |
on | -0.195 |
edges | -0.187 |
lid | -0.186 |
Token | went |
Token ID | 1816 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
Fro | +0.047 |
Jack | +0.043 |
His | +0.043 |
Un | +0.041 |
concession | +0.040 |
Neg logit contributions | |
hole | -0.090 |
els | -0.089 |
ps | -0.082 |
pieces | -0.081 |
ool | -0.081 |
Token | far |
Token ID | 1290 |
Feature activation | -3.77e-01 |
Pos logit contributions | |
Fro | +0.027 |
His | +0.024 |
concession | +0.024 |
Un | +0.023 |
Jack | +0.023 |
Neg logit contributions | |
hole | -0.027 |
els | -0.026 |
pieces | -0.026 |
edges | -0.024 |
ps | -0.024 |
Token | and |
Token ID | 290 |
Feature activation | +7.62e+00 |
Pos logit contributions | |
els | +0.004 |
pieces | +0.004 |
ps | +0.004 |
lid | +0.004 |
hole | +0.004 |
Neg logit contributions | |
Fro | -0.004 |
He | -0.004 |
Jack | -0.004 |
Un | -0.004 |
David | -0.004 |
Token | wide |
Token ID | 3094 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
Fro | +0.078 |
Jack | +0.073 |
Un | +0.068 |
concession | +0.067 |
Johnny | +0.066 |
Neg logit contributions | |
els | -0.115 |
hole | -0.107 |
ool | -0.106 |
ps | -0.099 |
pieces | -0.094 |
Token | on |
Token ID | 319 |
Feature activation | +2.85e+00 |
Pos logit contributions | |
Fro | +0.001 |
him | +0.001 |
Pete | +0.001 |
He | +0.001 |
His | +0.001 |
Neg logit contributions | |
els | -0.002 |
ool | -0.002 |
pieces | -0.002 |
splits | -0.002 |
slot | -0.002 |
Token | the |
Token ID | 262 |
Feature activation | +6.40e+00 |
Pos logit contributions | |
Fro | +0.025 |
Un | +0.024 |
David | +0.023 |
Jack | +0.023 |
Jim | +0.022 |
Neg logit contributions | |
hole | -0.044 |
ool | -0.038 |
hardest | -0.038 |
atop | -0.038 |
els | -0.037 |
Token | big |
Token ID | 1263 |
Feature activation | +5.70e+00 |
Pos logit contributions | |
Un | +0.068 |
David | +0.067 |
concession | +0.067 |
Fro | +0.065 |
Jack | +0.065 |
Neg logit contributions | |
hole | -0.078 |
els | -0.075 |
atop | -0.072 |
pieces | -0.070 |
tangled | -0.069 |
Token | blue |
Token ID | 4171 |
Feature activation | +2.98e+00 |
Pos logit contributions | |
Fro | +0.052 |
David | +0.052 |
Un | +0.051 |
Jack | +0.050 |
concession | +0.050 |
Neg logit contributions | |
hole | -0.078 |
ps | -0.077 |
atop | -0.073 |
tangled | -0.071 |
els | -0.071 |
Token | sea |
Token ID | 5417 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
Jack | +0.035 |
Un | +0.035 |
David | +0.035 |
Fro | +0.034 |
Kids | +0.033 |
Neg logit contributions | |
hole | -0.033 |
pieces | -0.032 |
atop | -0.031 |
ps | -0.030 |
edges | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | +1.40e+01 |
Pos logit contributions | |
els | +0.017 |
ps | +0.016 |
hole | +0.015 |
ool | +0.015 |
tangled | +0.015 |
Neg logit contributions | |
Fro | -0.013 |
Jack | -0.013 |
He | -0.012 |
They | -0.012 |
Pete | -0.012 |
Token | but |
Token ID | 475 |
Feature activation | +7.11e+00 |
Pos logit contributions | |
Un | +0.058 |
Jack | +0.057 |
David | +0.057 |
Fro | +0.055 |
Kids | +0.055 |
Neg logit contributions | |
pieces | -0.053 |
hole | -0.052 |
on | -0.050 |
falling | -0.048 |
edges | -0.048 |
Token | he |
Token ID | 339 |
Feature activation | +4.82e+00 |
Pos logit contributions | |
David | +0.057 |
Un | +0.057 |
Jack | +0.055 |
oph | +0.053 |
Kids | +0.053 |
Neg logit contributions | |
pieces | -0.109 |
on | -0.106 |
hole | -0.105 |
falling | -0.101 |
atop | -0.100 |
Token | did |
Token ID | 750 |
Feature activation | +3.11e+00 |
Pos logit contributions | |
David | +0.057 |
Un | +0.056 |
Jack | +0.055 |
concession | +0.054 |
Fro | +0.054 |
Neg logit contributions | |
pieces | -0.054 |
on | -0.051 |
hole | -0.051 |
atop | -0.048 |
tangled | -0.048 |
Token | not |
Token ID | 407 |
Feature activation | +4.11e+00 |
Pos logit contributions | |
His | +0.023 |
Un | +0.021 |
Jack | +0.020 |
Fro | +0.020 |
David | +0.019 |
Neg logit contributions | |
hole | -0.052 |
pieces | -0.051 |
lid | -0.047 |
edges | -0.047 |
atop | -0.047 |
Token | know |
Token ID | 760 |
Feature activation | -3.91e-01 |
Pos logit contributions | |
Un | +0.025 |
Jack | +0.025 |
Fro | +0.025 |
David | +0.024 |
Jim | +0.023 |
Neg logit contributions | |
hole | -0.070 |
pieces | -0.070 |
atop | -0.069 |
edges | -0.066 |
lid | -0.066 |
Token | how |
Token ID | 703 |
Feature activation | +4.86e+00 |
Pos logit contributions | |
hole | +0.006 |
atop | +0.006 |
lid | +0.006 |
edges | +0.006 |
pieces | +0.006 |
Neg logit contributions | |
Fro | -0.002 |
His | -0.002 |
Un | -0.002 |
Jack | -0.002 |
Johnny | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | +1.12e+01 |
Pos logit contributions | |
Jack | +0.054 |
Un | +0.054 |
Fro | +0.052 |
His | +0.052 |
David | +0.051 |
Neg logit contributions | |
hole | -0.063 |
pieces | -0.056 |
atop | -0.056 |
ps | -0.054 |
lid | -0.054 |
Token | He |
Token ID | 679 |
Feature activation | +5.09e+00 |
Pos logit contributions | |
David | +0.070 |
concession | +0.068 |
oph | +0.066 |
Jack | +0.065 |
Un | +0.065 |
Neg logit contributions | |
pieces | -0.193 |
hole | -0.186 |
edges | -0.184 |
on | -0.183 |
lid | -0.181 |
Token | asked |
Token ID | 1965 |
Feature activation | +3.21e+00 |
Pos logit contributions | |
David | +0.059 |
Jack | +0.058 |
Un | +0.058 |
Fro | +0.057 |
concession | +0.057 |
Neg logit contributions | |
pieces | -0.058 |
hole | -0.057 |
on | -0.054 |
edges | -0.053 |
ps | -0.052 |
Token | , |
Token ID | 11 |
Feature activation | +5.70e+00 |
Pos logit contributions | |
Fro | +0.023 |
Jack | +0.021 |
Un | +0.021 |
His | +0.020 |
Jim | +0.019 |
Neg logit contributions | |
hole | -0.056 |
pieces | -0.052 |
atop | -0.051 |
tangled | -0.050 |
edges | -0.050 |
Token | " |
Token ID | 366 |
Feature activation | +9.07e+00 |
Pos logit contributions | |
Un | +0.035 |
His | +0.033 |
Fro | +0.032 |
His | +0.031 |
Jack | +0.030 |
Neg logit contributions | |
hole | -0.105 |
pieces | -0.102 |
ool | -0.098 |
els | -0.098 |
edges | -0.097 |
Token | drove |
Token ID | 10357 |
Feature activation | -8.59e+00 |
Pos logit contributions | |
lid | +0.178 |
mant | +0.175 |
hardest | +0.174 |
ool | +0.173 |
hole | +0.171 |
Neg logit contributions | |
played | -0.054 |
They | -0.042 |
He | -0.037 |
Jack | -0.036 |
each | -0.035 |
Token | over |
Token ID | 625 |
Feature activation | -4.00e+00 |
Pos logit contributions | |
ool | +0.111 |
els | +0.108 |
hole | +0.108 |
teaspoon | +0.106 |
lid | +0.105 |
Neg logit contributions | |
him | -0.082 |
They | -0.078 |
He | -0.077 |
His | -0.074 |
each | -0.074 |
Token | it |
Token ID | 340 |
Feature activation | -7.62e+00 |
Pos logit contributions | |
ps | +0.056 |
hole | +0.055 |
ool | +0.055 |
atop | +0.053 |
eking | +0.052 |
Neg logit contributions | |
Jack | -0.034 |
Un | -0.034 |
him | -0.032 |
They | -0.032 |
Jim | -0.031 |
Token | , |
Token ID | 11 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
els | +0.132 |
ps | +0.125 |
bones | +0.122 |
cks | +0.118 |
ocol | +0.118 |
Neg logit contributions | |
He | -0.051 |
She | -0.046 |
As | -0.044 |
They | -0.042 |
As | -0.041 |
Token | hon |
Token ID | 3032 |
Feature activation | -9.97e+00 |
Pos logit contributions | |
ool | +0.020 |
els | +0.019 |
hole | +0.018 |
lid | +0.017 |
code | +0.017 |
Neg logit contributions | |
Fro | -0.011 |
concession | -0.010 |
He | -0.010 |
Un | -0.010 |
Jack | -0.010 |
Token | king |
Token ID | 3364 |
Feature activation | -1.06e+01 |
Pos logit contributions | |
pieces | +0.116 |
hole | +0.113 |
atop | +0.107 |
lid | +0.106 |
edges | +0.104 |
Neg logit contributions | |
Un | -0.112 |
Fro | -0.111 |
David | -0.111 |
Jack | -0.109 |
His | -0.104 |
Token | and |
Token ID | 290 |
Feature activation | +2.97e-01 |
Pos logit contributions | |
hole | +0.157 |
ool | +0.144 |
tangled | +0.137 |
atop | +0.135 |
hardest | +0.133 |
Neg logit contributions | |
Fro | -0.089 |
His | -0.088 |
They | -0.087 |
His | -0.086 |
He | -0.081 |
Token | be |
Token ID | 307 |
Feature activation | -4.76e+00 |
Pos logit contributions | |
Fro | +0.003 |
Un | +0.003 |
Max | +0.003 |
him | +0.003 |
Jack | +0.003 |
Neg logit contributions | |
ool | -0.004 |
hole | -0.004 |
lid | -0.004 |
els | -0.004 |
ently | -0.003 |
Token | eping |
Token ID | 7213 |
Feature activation | -1.21e+01 |
Pos logit contributions | |
lid | +0.061 |
ool | +0.060 |
hole | +0.060 |
button | +0.056 |
globe | +0.054 |
Neg logit contributions | |
Un | -0.051 |
They | -0.048 |
David | -0.048 |
Billy | -0.048 |
Johnny | -0.047 |
Token | . |
Token ID | 13 |
Feature activation | +8.22e+00 |
Pos logit contributions | |
hole | +0.161 |
atop | +0.146 |
ool | +0.143 |
tangled | +0.142 |
bulb | +0.141 |
Neg logit contributions | |
Fro | -0.110 |
His | -0.110 |
His | -0.109 |
They | -0.105 |
Un | -0.105 |
Token |
|
Token ID | 198 |
Feature activation | +6.03e+00 |
Pos logit contributions | |
Jack | +0.045 |
Fro | +0.041 |
His | +0.041 |
Un | +0.041 |
David | +0.039 |
Neg logit contributions | |
hole | -0.148 |
pieces | -0.147 |
ps | -0.145 |
falling | -0.143 |
atop | -0.142 |
Token | . |
Token ID | 13 |
Feature activation | +9.93e+00 |
Pos logit contributions | |
Un | +0.018 |
Jack | +0.018 |
Fro | +0.018 |
David | +0.018 |
Jim | +0.018 |
Neg logit contributions | |
hole | -0.009 |
pieces | -0.009 |
atop | -0.008 |
edges | -0.008 |
ps | -0.008 |
Token | They |
Token ID | 1119 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
David | +0.101 |
oph | +0.099 |
concession | +0.099 |
Jack | +0.098 |
Johnny | +0.095 |
Neg logit contributions | |
on | -0.133 |
pieces | -0.127 |
hole | -0.120 |
falling | -0.119 |
ladder | -0.119 |
Token | jumped |
Token ID | 11687 |
Feature activation | -6.48e+00 |
Pos logit contributions | |
Un | +0.027 |
David | +0.027 |
Jack | +0.026 |
concession | +0.026 |
Fro | +0.026 |
Neg logit contributions | |
pieces | -0.027 |
hole | -0.026 |
on | -0.026 |
ps | -0.025 |
atop | -0.025 |
Token | into |
Token ID | 656 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
hole | +0.067 |
pieces | +0.063 |
atop | +0.058 |
edges | +0.058 |
els | +0.057 |
Neg logit contributions | |
Fro | -0.085 |
Jack | -0.083 |
Un | -0.081 |
concession | -0.080 |
Kids | -0.078 |
Token | the |
Token ID | 262 |
Feature activation | -7.45e-01 |
Pos logit contributions | |
pieces | +0.022 |
hole | +0.021 |
on | +0.020 |
edges | +0.019 |
ps | +0.019 |
Neg logit contributions | |
David | -0.031 |
Un | -0.031 |
Jack | -0.031 |
Fro | -0.030 |
enny | -0.030 |
Token | pool |
Token ID | 5933 |
Feature activation | -1.07e+01 |
Pos logit contributions | |
hole | +0.008 |
pieces | +0.008 |
atop | +0.007 |
ps | +0.007 |
falling | +0.007 |
Neg logit contributions | |
Jack | -0.009 |
Fro | -0.009 |
concession | -0.009 |
Un | -0.009 |
Kids | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | +9.78e-01 |
Pos logit contributions | |
hole | +0.159 |
atop | +0.157 |
els | +0.155 |
tangled | +0.151 |
hardest | +0.148 |
Neg logit contributions | |
Fro | -0.080 |
Jack | -0.078 |
concession | -0.076 |
Un | -0.076 |
They | -0.074 |
Token | threw |
Token ID | 9617 |
Feature activation | -7.40e+00 |
Pos logit contributions | |
Fro | +0.010 |
Jack | +0.010 |
Un | +0.010 |
Kids | +0.010 |
concession | +0.010 |
Neg logit contributions | |
hole | -0.013 |
pieces | -0.012 |
atop | -0.011 |
edges | -0.011 |
els | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | -9.14e-01 |
Pos logit contributions | |
hole | +0.098 |
pieces | +0.093 |
atop | +0.090 |
edges | +0.087 |
falling | +0.087 |
Neg logit contributions | |
Fro | -0.075 |
Un | -0.074 |
Jack | -0.074 |
David | -0.071 |
Jim | -0.070 |
Token | ball |
Token ID | 2613 |
Feature activation | -1.12e+01 |
Pos logit contributions | |
hole | +0.010 |
atop | +0.009 |
ps | +0.009 |
els | +0.009 |
pieces | +0.009 |
Neg logit contributions | |
Jack | -0.011 |
Fro | -0.011 |
Un | -0.011 |
His | -0.011 |
concession | -0.010 |
Token | for |
Token ID | 329 |
Feature activation | -2.63e+00 |
Pos logit contributions | |
hole | +0.108 |
pieces | +0.102 |
atop | +0.098 |
edges | +0.096 |
hardest | +0.093 |
Neg logit contributions | |
Fro | -0.150 |
Un | -0.144 |
Jack | -0.143 |
His | -0.141 |
concession | -0.138 |
Token | dog |
Token ID | 3290 |
Feature activation | -6.80e+00 |
Pos logit contributions | |
on | +0.009 |
hard | +0.008 |
pieces | +0.008 |
broken | +0.008 |
matches | +0.007 |
Neg logit contributions | |
enny | -0.010 |
oph | -0.010 |
ete | -0.010 |
Tony | -0.009 |
ree | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +6.40e+00 |
Pos logit contributions | |
hole | +0.080 |
pieces | +0.074 |
atop | +0.070 |
ps | +0.070 |
edges | +0.070 |
Neg logit contributions | |
Fro | -0.085 |
Un | -0.082 |
Jack | -0.081 |
Kids | -0.080 |
His | -0.078 |
Token | Tim |
Token ID | 5045 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
David | +0.047 |
Un | +0.047 |
Jack | +0.045 |
concession | +0.045 |
oph | +0.044 |
Neg logit contributions | |
pieces | -0.098 |
hole | -0.097 |
on | -0.095 |
edges | -0.092 |
falling | -0.091 |
Token | liked |
Token ID | 8288 |
Feature activation | -4.36e+00 |
Pos logit contributions | |
pieces | +0.019 |
hole | +0.019 |
on | +0.018 |
edges | +0.017 |
atop | +0.017 |
Neg logit contributions | |
Un | -0.015 |
David | -0.015 |
Jack | -0.015 |
Fro | -0.015 |
Jim | -0.014 |
Token | to |
Token ID | 284 |
Feature activation | -2.66e+00 |
Pos logit contributions | |
pieces | +0.053 |
hole | +0.050 |
on | +0.050 |
falling | +0.048 |
edges | +0.048 |
Neg logit contributions | |
Jack | -0.048 |
David | -0.048 |
Un | -0.047 |
Fro | -0.046 |
Jim | -0.046 |
Token | play |
Token ID | 711 |
Feature activation | -8.69e+00 |
Pos logit contributions | |
pieces | +0.033 |
on | +0.032 |
trunk | +0.032 |
falling | +0.032 |
ps | +0.031 |
Neg logit contributions | |
David | -0.028 |
Jack | -0.028 |
Un | -0.028 |
enny | -0.027 |
oph | -0.027 |
Token | outside |
Token ID | 2354 |
Feature activation | -9.42e+00 |
Pos logit contributions | |
pieces | +0.074 |
hole | +0.073 |
on | +0.067 |
atop | +0.067 |
edges | +0.065 |
Neg logit contributions | |
Un | -0.126 |
David | -0.126 |
Jack | -0.125 |
Fro | -0.124 |
Jim | -0.121 |
Token | in |
Token ID | 287 |
Feature activation | -5.24e+00 |
Pos logit contributions | |
hole | +0.093 |
pieces | +0.087 |
ps | +0.082 |
atop | +0.081 |
edges | +0.080 |
Neg logit contributions | |
Jack | -0.125 |
Un | -0.124 |
Fro | -0.123 |
concession | -0.119 |
David | -0.119 |
Token | the |
Token ID | 262 |
Feature activation | -4.62e+00 |
Pos logit contributions | |
pieces | +0.045 |
on | +0.041 |
hole | +0.040 |
falling | +0.037 |
edges | +0.037 |
Neg logit contributions | |
David | -0.079 |
oph | -0.079 |
Un | -0.078 |
Fro | -0.078 |
enny | -0.078 |
Token | dirt |
Token ID | 13647 |
Feature activation | -1.50e+01 |
Pos logit contributions | |
pieces | +0.037 |
on | +0.036 |
edges | +0.033 |
hole | +0.033 |
atop | +0.032 |
Neg logit contributions | |
David | -0.071 |
Fro | -0.069 |
enny | -0.069 |
Jack | -0.068 |
Un | -0.067 |
Token | with |
Token ID | 351 |
Feature activation | -6.66e+00 |
Pos logit contributions | |
hole | +0.189 |
atop | +0.180 |
els | +0.175 |
ps | +0.174 |
bones | +0.174 |
Neg logit contributions | |
Fro | -0.135 |
They | -0.132 |
Jack | -0.130 |
He | -0.129 |
He | -0.128 |
Token | looked |
Token ID | 3114 |
Feature activation | -8.03e+00 |
Pos logit contributions | |
pieces | +0.020 |
on | +0.019 |
hole | +0.018 |
edges | +0.018 |
broken | +0.018 |
Neg logit contributions | |
David | -0.024 |
concession | -0.024 |
Fro | -0.023 |
Un | -0.023 |
Kids | -0.023 |
Token | closely |
Token ID | 7173 |
Feature activation | -1.08e+01 |
Pos logit contributions | |
hole | +0.060 |
pieces | +0.060 |
on | +0.054 |
atop | +0.054 |
edges | +0.052 |
Neg logit contributions | |
Un | -0.125 |
David | -0.124 |
Jack | -0.124 |
Fro | -0.122 |
Kids | -0.120 |
Token | and |
Token ID | 290 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
hole | +0.139 |
pieces | +0.136 |
ps | +0.132 |
code | +0.131 |
falling | +0.128 |
Neg logit contributions | |
Jack | -0.107 |
Un | -0.103 |
David | -0.102 |
Fro | -0.101 |
Jim | -0.099 |
Token | saw |
Token ID | 2497 |
Feature activation | -6.06e+00 |
Pos logit contributions | |
hole | +0.024 |
pieces | +0.024 |
on | +0.023 |
atop | +0.023 |
falling | +0.023 |
Neg logit contributions | |
Un | -0.023 |
Fro | -0.022 |
David | -0.022 |
concession | -0.022 |
Kids | -0.022 |
Token | that |
Token ID | 326 |
Feature activation | -4.41e+00 |
Pos logit contributions | |
pieces | +0.071 |
hole | +0.071 |
on | +0.066 |
atop | +0.065 |
edges | +0.065 |
Neg logit contributions | |
Un | -0.068 |
Jack | -0.068 |
David | -0.068 |
Fro | -0.067 |
Kids | -0.065 |
Token | it |
Token ID | 340 |
Feature activation | -9.61e+00 |
Pos logit contributions | |
hole | +0.049 |
pieces | +0.046 |
ps | +0.045 |
atop | +0.044 |
edges | +0.043 |
Neg logit contributions | |
Jack | -0.054 |
Fro | -0.053 |
Un | -0.053 |
David | -0.053 |
Jim | -0.051 |
Token | was |
Token ID | 373 |
Feature activation | -7.60e+00 |
Pos logit contributions | |
hole | +0.133 |
pieces | +0.128 |
atop | +0.123 |
edges | +0.121 |
ps | +0.121 |
Neg logit contributions | |
Jack | -0.091 |
Un | -0.088 |
Fro | -0.088 |
David | -0.084 |
Johnny | -0.083 |
Token | a |
Token ID | 257 |
Feature activation | -6.81e+00 |
Pos logit contributions | |
hole | +0.092 |
pieces | +0.091 |
atop | +0.085 |
edges | +0.084 |
on | +0.083 |
Neg logit contributions | |
Un | -0.083 |
Jack | -0.083 |
David | -0.082 |
Fro | -0.082 |
Jim | -0.078 |
Token | sm |
Token ID | 895 |
Feature activation | -1.01e+01 |
Pos logit contributions | |
hole | +0.078 |
ps | +0.073 |
pieces | +0.070 |
els | +0.069 |
atop | +0.068 |
Neg logit contributions | |
Jack | -0.086 |
Un | -0.085 |
Fro | -0.085 |
concession | -0.082 |
David | -0.081 |
Token | elly |
Token ID | 6148 |
Feature activation | -1.24e+01 |
Pos logit contributions | |
hole | +0.106 |
pieces | +0.102 |
on | +0.097 |
edges | +0.094 |
ool | +0.094 |
Neg logit contributions | |
David | -0.128 |
Un | -0.124 |
Fro | -0.123 |
Harry | -0.122 |
concession | -0.121 |
Token | sausage |
Token ID | 31889 |
Feature activation | -1.72e+01 |
Pos logit contributions | |
hole | +0.152 |
ps | +0.139 |
cks | +0.136 |
atop | +0.136 |
pieces | +0.132 |
Neg logit contributions | |
Fro | -0.149 |
concession | -0.132 |
Jack | -0.132 |
Mitt | -0.129 |
Kids | -0.129 |
Token | He |
Token ID | 679 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
David | +0.051 |
Jack | +0.049 |
Un | +0.048 |
Da | +0.045 |
oph | +0.045 |
Neg logit contributions | |
pieces | -0.138 |
hole | -0.134 |
on | -0.132 |
edges | -0.131 |
trunk | -0.129 |
Token | wants |
Token ID | 3382 |
Feature activation | -4.23e+00 |
Pos logit contributions | |
David | +0.019 |
Jack | +0.019 |
Un | +0.019 |
Fro | +0.018 |
Jim | +0.018 |
Neg logit contributions | |
pieces | -0.012 |
hole | -0.012 |
on | -0.011 |
atop | -0.011 |
edges | -0.011 |
Token | to |
Token ID | 284 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
hole | +0.067 |
atop | +0.062 |
pieces | +0.059 |
lid | +0.058 |
ps | +0.058 |
Neg logit contributions | |
Fro | -0.038 |
Un | -0.036 |
Jack | -0.035 |
David | -0.033 |
Jim | -0.033 |
Token | hug |
Token ID | 16225 |
Feature activation | -7.36e+00 |
Pos logit contributions | |
on | +0.013 |
trunk | +0.013 |
pieces | +0.013 |
falling | +0.012 |
ring | +0.012 |
Neg logit contributions | |
David | -0.017 |
enny | -0.016 |
Jack | -0.016 |
oph | -0.016 |
concession | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
hole | +0.093 |
pieces | +0.089 |
atop | +0.085 |
edges | +0.083 |
ps | +0.082 |
Neg logit contributions | |
Un | -0.081 |
Fro | -0.080 |
Jack | -0.079 |
David | -0.077 |
Kids | -0.075 |
Token | dog |
Token ID | 3290 |
Feature activation | -8.74e+00 |
Pos logit contributions | |
hole | +0.020 |
atop | +0.019 |
pieces | +0.019 |
ps | +0.018 |
els | +0.018 |
Neg logit contributions | |
Un | -0.009 |
Fro | -0.009 |
Jack | -0.009 |
David | -0.009 |
concession | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +7.03e+00 |
Pos logit contributions | |
hole | +0.104 |
pieces | +0.096 |
atop | +0.096 |
edges | +0.095 |
lid | +0.092 |
Neg logit contributions | |
Fro | -0.111 |
His | -0.100 |
His | -0.097 |
Her | -0.094 |
Un | -0.093 |
Token | He |
Token ID | 679 |
Feature activation | +7.19e-01 |
Pos logit contributions | |
David | +0.047 |
Da | +0.045 |
Un | +0.045 |
concession | +0.044 |
Harry | +0.044 |
Neg logit contributions | |
pieces | -0.118 |
hole | -0.113 |
on | -0.112 |
edges | -0.111 |
trunk | -0.109 |
Token | runs |
Token ID | 4539 |
Feature activation | -5.91e+00 |
Pos logit contributions | |
David | +0.010 |
Jack | +0.010 |
Un | +0.010 |
Johnny | +0.010 |
Da | +0.010 |
Neg logit contributions | |
pieces | -0.006 |
hole | -0.006 |
on | -0.006 |
ps | -0.006 |
edges | -0.006 |
Token | to |
Token ID | 284 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
pieces | +0.062 |
hole | +0.061 |
on | +0.059 |
atop | +0.056 |
edges | +0.056 |
Neg logit contributions | |
David | -0.074 |
Un | -0.074 |
Jack | -0.073 |
Fro | -0.072 |
Jim | -0.070 |
Token | the |
Token ID | 262 |
Feature activation | -7.36e-01 |
Pos logit contributions | |
pieces | +0.017 |
hole | +0.017 |
on | +0.016 |
atop | +0.016 |
falling | +0.016 |
Neg logit contributions | |
Un | -0.022 |
Jack | -0.022 |
David | -0.022 |
Fro | -0.021 |
Kids | -0.021 |