Token | ades |
Token ID | 2367 |
Feature activation | +7.49e-02 |
Pos logit contributions | |
than | +0.189 |
- | +0.159 |
ies | +0.157 |
ever | +0.154 |
knowing | +0.152 |
Neg logit contributions | |
First | -0.120 |
V | -0.119 |
Here | -0.117 |
Higher | -0.114 |
Grab | -0.111 |
Token | won |
Token ID | 1839 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
than | +0.103 |
ies | +0.098 |
- | +0.090 |
ier | +0.089 |
inches | +0.088 |
Neg logit contributions | |
V | -0.180 |
Here | -0.177 |
Higher | -0.175 |
First | -0.172 |
Grab | -0.167 |
Token | 't |
Token ID | 470 |
Feature activation | +4.35e-02 |
Pos logit contributions | |
than | +0.064 |
ies | +0.059 |
ier | +0.057 |
- | +0.057 |
owing | +0.054 |
Neg logit contributions | |
V | -0.045 |
Higher | -0.042 |
Here | -0.042 |
First | -0.040 |
Grab | -0.039 |
Token | make |
Token ID | 787 |
Feature activation | +4.69e-02 |
Pos logit contributions | |
than | +0.100 |
- | +0.094 |
ies | +0.092 |
owing | +0.089 |
ishing | +0.089 |
Neg logit contributions | |
Many | -0.059 |
V | -0.058 |
First | -0.057 |
The | -0.057 |
Everyone | -0.056 |
Token | a |
Token ID | 257 |
Feature activation | +6.03e-02 |
Pos logit contributions | |
than | +0.102 |
ier | +0.091 |
ies | +0.091 |
- | +0.086 |
owing | +0.086 |
Neg logit contributions | |
V | -0.083 |
Here | -0.079 |
Higher | -0.076 |
First | -0.075 |
Everyone | -0.074 |
Token | candle |
Token ID | 26839 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
than | +0.117 |
owing | +0.113 |
ier | +0.109 |
ge | +0.098 |
ies | +0.096 |
Neg logit contributions | |
V | -0.111 |
Ro | -0.110 |
Here | -0.104 |
Look | -0.101 |
Bang | -0.099 |
Token | ! |
Token ID | 0 |
Feature activation | -8.92e-02 |
Pos logit contributions | |
than | +0.214 |
owing | +0.198 |
ishing | +0.187 |
ies | +0.187 |
ge | +0.184 |
Neg logit contributions | |
The | -0.238 |
It | -0.224 |
Everyone | -0.214 |
Ro | -0.213 |
Two | -0.212 |
Token | â |
Token ID | 22940 |
Feature activation | +9.65e-03 |
Pos logit contributions | |
V | +0.122 |
Here | +0.120 |
Ro | +0.116 |
Higher | +0.112 |
Grab | +0.112 |
Neg logit contributions | |
than | -0.221 |
- | -0.200 |
ier | -0.190 |
ies | -0.185 |
! | -0.179 |
Token | € |
Token ID | 26391 |
Feature activation | -2.54e-02 |
Pos logit contributions | |
than | +0.025 |
- | +0.023 |
ier | +0.023 |
ies | +0.022 |
owing | +0.021 |
Neg logit contributions | |
V | -0.014 |
Here | -0.014 |
Higher | -0.013 |
Ro | -0.013 |
Grab | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -8.36e-02 |
Pos logit contributions | |
Here | +0.044 |
V | +0.043 |
Higher | +0.041 |
Grab | +0.041 |
Ro | +0.039 |
Neg logit contributions | |
than | -0.059 |
- | -0.054 |
ies | -0.053 |
ier | -0.051 |
inches | -0.048 |
Token |
|
Token ID | 198 |
Feature activation | -7.58e-02 |
Pos logit contributions | |
Grab | +0.141 |
Higher | +0.140 |
Here | +0.140 |
Ro | +0.136 |
V | +0.136 |
Neg logit contributions | |
than | -0.163 |
- | -0.160 |
ines | -0.143 |
! | -0.143 |
ever | -0.136 |
Token | , |
Token ID | 11 |
Feature activation | +3.37e-02 |
Pos logit contributions | |
than | +0.149 |
inches | +0.135 |
ge | +0.131 |
supposed | +0.123 |
owing | +0.119 |
Neg logit contributions | |
Everyone | -0.254 |
Here | -0.254 |
The | -0.251 |
Let | -0.250 |
Higher | -0.247 |
Token | it |
Token ID | 340 |
Feature activation | +5.18e-02 |
Pos logit contributions | |
ier | +0.055 |
ies | +0.054 |
than | +0.053 |
ge | +0.052 |
- | +0.050 |
Neg logit contributions | |
Higher | -0.071 |
Here | -0.071 |
V | -0.069 |
The | -0.068 |
Look | -0.066 |
Token | even |
Token ID | 772 |
Feature activation | +5.77e-02 |
Pos logit contributions | |
than | +0.112 |
owing | +0.108 |
ies | +0.108 |
ier | +0.108 |
ge | +0.105 |
Neg logit contributions | |
V | -0.072 |
Here | -0.071 |
The | -0.069 |
Hi | -0.067 |
Higher | -0.067 |
Token | had |
Token ID | 550 |
Feature activation | +5.91e-02 |
Pos logit contributions | |
than | +0.099 |
ies | +0.093 |
owing | +0.093 |
ier | +0.093 |
ge | +0.092 |
Neg logit contributions | |
V | -0.115 |
Here | -0.114 |
Higher | -0.109 |
Grab | -0.105 |
Hi | -0.104 |
Token | real |
Token ID | 1103 |
Feature activation | +9.72e-02 |
Pos logit contributions | |
than | +0.099 |
ies | +0.099 |
owing | +0.094 |
ier | +0.093 |
ge | +0.092 |
Neg logit contributions | |
Here | -0.114 |
V | -0.114 |
Look | -0.111 |
The | -0.111 |
Higher | -0.109 |
Token | fireworks |
Token ID | 26056 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
ge | +0.175 |
ies | +0.171 |
than | +0.168 |
ier | +0.167 |
owing | +0.163 |
Neg logit contributions | |
Everyone | -0.178 |
Many | -0.172 |
Here | -0.169 |
V | -0.169 |
Higher | -0.166 |
Token | ! |
Token ID | 0 |
Feature activation | -8.58e-02 |
Pos logit contributions | |
ies | +0.176 |
ge | +0.175 |
owing | +0.171 |
ishing | +0.158 |
inches | +0.154 |
Neg logit contributions | |
Everyone | -0.259 |
The | -0.259 |
Max | -0.248 |
Here | -0.248 |
It | -0.247 |
Token |
|
Token ID | 198 |
Feature activation | -7.96e-02 |
Pos logit contributions | |
V | +0.124 |
Grab | +0.121 |
Here | +0.121 |
Ro | +0.118 |
Higher | +0.107 |
Neg logit contributions | |
than | -0.212 |
- | -0.194 |
ies | -0.183 |
ever | -0.183 |
! | -0.180 |
Token |
|
Token ID | 198 |
Feature activation | -7.40e-02 |
Pos logit contributions | |
Grab | +0.141 |
Ro | +0.137 |
V | +0.133 |
Here | +0.128 |
First | +0.127 |
Neg logit contributions | |
- | -0.164 |
than | -0.154 |
! | -0.143 |
ines | -0.141 |
ies | -0.139 |
Token | Sarah |
Token ID | 29284 |
Feature activation | +3.22e-02 |
Pos logit contributions | |
V | +0.109 |
Higher | +0.108 |
Grab | +0.108 |
Here | +0.106 |
First | +0.106 |
Neg logit contributions | |
than | -0.168 |
- | -0.163 |
ies | -0.148 |
ier | -0.148 |
ever | -0.147 |
Token | and |
Token ID | 290 |
Feature activation | -4.17e-02 |
Pos logit contributions | |
than | +0.051 |
ier | +0.049 |
ies | +0.046 |
ge | +0.044 |
inches | +0.043 |
Neg logit contributions | |
V | -0.071 |
Here | -0.068 |
The | -0.068 |
Higher | -0.067 |
Grab | -0.066 |
Token | were |
Token ID | 547 |
Feature activation | +3.61e-02 |
Pos logit contributions | |
than | +0.099 |
ies | +0.098 |
ier | +0.091 |
ge | +0.089 |
- | +0.089 |
Neg logit contributions | |
The | -0.147 |
Higher | -0.145 |
Here | -0.143 |
It | -0.135 |
V | -0.133 |
Token | tiny |
Token ID | 7009 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
than | +0.063 |
ies | +0.061 |
ier | +0.059 |
- | +0.059 |
ge | +0.059 |
Neg logit contributions | |
Here | -0.069 |
The | -0.066 |
Higher | -0.064 |
V | -0.064 |
Look | -0.063 |
Token | bits |
Token ID | 10340 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
supposed | +0.176 |
ier | +0.173 |
than | +0.169 |
owing | +0.167 |
ge | +0.164 |
Neg logit contributions | |
The | -0.186 |
Higher | -0.172 |
V | -0.171 |
Here | -0.170 |
Let | -0.167 |
Token | of |
Token ID | 286 |
Feature activation | +6.53e-02 |
Pos logit contributions | |
ies | +0.165 |
than | +0.164 |
- | +0.163 |
ge | +0.158 |
ishing | +0.156 |
Neg logit contributions | |
The | -0.183 |
Higher | -0.178 |
It | -0.175 |
There | -0.172 |
Here | -0.171 |
Token | sparkling |
Token ID | 39072 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
than | +0.139 |
ier | +0.138 |
ies | +0.133 |
- | +0.133 |
owing | +0.132 |
Neg logit contributions | |
Here | -0.112 |
Higher | -0.102 |
V | -0.102 |
Grab | -0.099 |
The | -0.098 |
Token | dust |
Token ID | 8977 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
supposed | +0.236 |
owing | +0.224 |
ge | +0.216 |
than | +0.214 |
ier | +0.212 |
Neg logit contributions | |
Here | -0.190 |
The | -0.189 |
Higher | -0.175 |
Grab | -0.174 |
Spr | -0.173 |
Token | drifting |
Token ID | 38193 |
Feature activation | +9.26e-02 |
Pos logit contributions | |
than | +0.177 |
ge | +0.168 |
ies | +0.167 |
inches | +0.162 |
supposed | +0.160 |
Neg logit contributions | |
The | -0.268 |
It | -0.258 |
Here | -0.256 |
Higher | -0.252 |
Everyone | -0.243 |
Token | in |
Token ID | 287 |
Feature activation | +4.72e-02 |
Pos logit contributions | |
than | +0.125 |
ies | +0.113 |
inches | +0.106 |
ier | +0.106 |
ge | +0.102 |
Neg logit contributions | |
Higher | -0.207 |
The | -0.206 |
First | -0.203 |
Here | -0.202 |
V | -0.202 |
Token | the |
Token ID | 262 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
ier | +0.118 |
ies | +0.118 |
than | +0.116 |
owing | +0.115 |
- | +0.114 |
Neg logit contributions | |
Here | -0.061 |
V | -0.059 |
Look | -0.059 |
The | -0.058 |
Higher | -0.057 |
Token | air |
Token ID | 1633 |
Feature activation | +9.58e-02 |
Pos logit contributions | |
than | +0.048 |
ies | +0.044 |
- | +0.044 |
ier | +0.044 |
ge | +0.040 |
Neg logit contributions | |
V | -0.037 |
Here | -0.036 |
Higher | -0.034 |
Grab | -0.034 |
First | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | -8.83e-02 |
Pos logit contributions | |
than | +0.126 |
ies | +0.124 |
ier | +0.121 |
- | +0.120 |
ge | +0.117 |
Neg logit contributions | |
The | -0.201 |
Here | -0.186 |
Everyone | -0.184 |
Higher | -0.184 |
It | -0.182 |
Token | a |
Token ID | 257 |
Feature activation | +6.20e-02 |
Pos logit contributions | |
than | +0.056 |
ier | +0.053 |
ies | +0.053 |
owing | +0.052 |
ge | +0.052 |
Neg logit contributions | |
Higher | -0.045 |
Here | -0.044 |
V | -0.043 |
The | -0.042 |
First | -0.042 |
Token | rhythm |
Token ID | 18662 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
than | +0.121 |
ier | +0.121 |
ies | +0.118 |
owing | +0.114 |
ge | +0.110 |
Neg logit contributions | |
V | -0.116 |
Here | -0.110 |
Ro | -0.107 |
Hi | -0.106 |
Look | -0.105 |
Token | can |
Token ID | 460 |
Feature activation | +5.67e-02 |
Pos logit contributions | |
than | +0.154 |
ies | +0.149 |
ge | +0.140 |
inches | +0.135 |
owing | +0.132 |
Neg logit contributions | |
Here | -0.216 |
Higher | -0.214 |
V | -0.213 |
Everyone | -0.211 |
The | -0.209 |
Token | fill |
Token ID | 6070 |
Feature activation | +9.30e-02 |
Pos logit contributions | |
than | +0.114 |
ies | +0.111 |
ier | +0.105 |
ishing | +0.104 |
- | +0.103 |
Neg logit contributions | |
V | -0.091 |
The | -0.090 |
Many | -0.087 |
Everyone | -0.087 |
Higher | -0.086 |
Token | a |
Token ID | 257 |
Feature activation | +6.63e-02 |
Pos logit contributions | |
ier | +0.175 |
ge | +0.166 |
owing | +0.165 |
than | +0.164 |
ies | +0.162 |
Neg logit contributions | |
Here | -0.151 |
The | -0.149 |
V | -0.143 |
Look | -0.142 |
Higher | -0.141 |
Token | space |
Token ID | 2272 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
than | +0.134 |
ier | +0.131 |
ies | +0.127 |
owing | +0.126 |
ge | +0.122 |
Neg logit contributions | |
V | -0.127 |
Here | -0.117 |
Look | -0.115 |
Ro | -0.114 |
Bang | -0.110 |
Token | with |
Token ID | 351 |
Feature activation | +7.64e-02 |
Pos logit contributions | |
ies | +0.196 |
owing | +0.188 |
ier | +0.187 |
supposed | +0.185 |
ge | +0.178 |
Neg logit contributions | |
The | -0.262 |
Higher | -0.255 |
V | -0.250 |
Everyone | -0.247 |
Here | -0.243 |
Token | joy |
Token ID | 8716 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
ier | +0.137 |
ies | +0.136 |
than | +0.135 |
ge | +0.131 |
owing | +0.129 |
Neg logit contributions | |
Here | -0.139 |
Higher | -0.135 |
The | -0.132 |
V | -0.130 |
Look | -0.127 |
Token | ! |
Token ID | 0 |
Feature activation | -8.93e-02 |
Pos logit contributions | |
inches | +0.161 |
ge | +0.152 |
than | +0.146 |
owing | +0.141 |
ies | +0.139 |
Neg logit contributions | |
The | -0.230 |
It | -0.226 |
Everyone | -0.221 |
All | -0.205 |
Hi | -0.205 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
V | +0.122 |
Here | +0.116 |
Grab | +0.115 |
Ro | +0.108 |
Higher | +0.105 |
Neg logit contributions | |
than | -0.231 |
- | -0.214 |
ies | -0.204 |
ever | -0.203 |
ier | -0.202 |
Token | Once |
Token ID | 7454 |
Feature activation | -2.53e-02 |
Pos logit contributions | |
V | +0.043 |
Here | +0.042 |
Higher | +0.042 |
Grab | +0.041 |
First | +0.040 |
Neg logit contributions | |
than | -0.069 |
- | -0.066 |
ier | -0.063 |
ies | -0.062 |
ever | -0.060 |
Token | you |
Token ID | 345 |
Feature activation | +2.50e-02 |
Pos logit contributions | |
V | +0.015 |
Here | +0.015 |
Higher | +0.014 |
Grab | +0.014 |
First | +0.013 |
Neg logit contributions | |
than | -0.018 |
- | -0.017 |
ies | -0.017 |
ier | -0.016 |
inches | -0.015 |
Token | help |
Token ID | 1037 |
Feature activation | +3.25e-02 |
Pos logit contributions | |
than | +0.049 |
- | +0.045 |
ies | +0.045 |
ier | +0.045 |
ge | +0.043 |
Neg logit contributions | |
Here | -0.045 |
V | -0.045 |
Higher | -0.043 |
First | -0.042 |
Many | -0.042 |
Token | me |
Token ID | 502 |
Feature activation | +5.94e-02 |
Pos logit contributions | |
than | +0.074 |
ier | +0.068 |
ies | +0.066 |
- | +0.065 |
inches | +0.063 |
Neg logit contributions | |
V | -0.054 |
Here | -0.053 |
Higher | -0.050 |
First | -0.049 |
Grab | -0.048 |
Token | hang |
Token ID | 8181 |
Feature activation | +9.20e-02 |
Pos logit contributions | |
than | +0.120 |
ier | +0.108 |
ies | +0.105 |
- | +0.100 |
ge | +0.098 |
Neg logit contributions | |
Here | -0.102 |
First | -0.102 |
V | -0.100 |
Higher | -0.100 |
The | -0.100 |
Token | my |
Token ID | 616 |
Feature activation | +4.71e-02 |
Pos logit contributions | |
than | +0.172 |
owing | +0.165 |
ge | +0.163 |
ies | +0.163 |
ier | +0.159 |
Neg logit contributions | |
Here | -0.165 |
V | -0.163 |
Higher | -0.159 |
The | -0.156 |
First | -0.150 |
Token | art |
Token ID | 1242 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
than | +0.096 |
ier | +0.095 |
ge | +0.089 |
owing | +0.085 |
ies | +0.083 |
Neg logit contributions | |
Here | -0.084 |
V | -0.080 |
Higher | -0.078 |
Look | -0.075 |
Grab | -0.075 |
Token | ?" |
Token ID | 1701 |
Feature activation | -9.52e-02 |
Pos logit contributions | |
owing | +0.255 |
ge | +0.248 |
than | +0.240 |
ies | +0.232 |
ier | +0.221 |
Neg logit contributions | |
First | -0.248 |
V | -0.247 |
Here | -0.244 |
Higher | -0.236 |
Everyone | -0.235 |
Token | The |
Token ID | 383 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
V | +0.147 |
Grab | +0.136 |
Higher | +0.135 |
Here | +0.133 |
Ro | +0.122 |
Neg logit contributions | |
than | -0.232 |
- | -0.219 |
ies | -0.210 |
ever | -0.198 |
ier | -0.191 |
Token | cat |
Token ID | 3797 |
Feature activation | +1.50e-03 |
Pos logit contributions | |
V | +0.042 |
Here | +0.042 |
Higher | +0.040 |
Grab | +0.040 |
First | +0.038 |
Neg logit contributions | |
than | -0.045 |
- | -0.041 |
ies | -0.041 |
ier | -0.040 |
inches | -0.037 |
Token | said |
Token ID | 531 |
Feature activation | -2.70e-02 |
Pos logit contributions | |
than | +0.003 |
- | +0.003 |
ies | +0.003 |
ier | +0.003 |
inches | +0.003 |
Neg logit contributions | |
V | -0.003 |
Here | -0.003 |
Higher | -0.002 |
Grab | -0.002 |
First | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -3.63e-02 |
Pos logit contributions | |
V | +0.066 |
Here | +0.065 |
Higher | +0.064 |
Grab | +0.062 |
First | +0.060 |
Neg logit contributions | |
than | -0.042 |
ies | -0.037 |
ier | -0.035 |
- | -0.035 |
inches | -0.032 |
Token | together |
Token ID | 1978 |
Feature activation | +7.37e-02 |
Pos logit contributions | |
than | +0.169 |
ies | +0.164 |
owing | +0.158 |
- | +0.155 |
ge | +0.154 |
Neg logit contributions | |
Here | -0.240 |
First | -0.235 |
The | -0.227 |
Everyone | -0.218 |
Grab | -0.212 |
Token | . |
Token ID | 13 |
Feature activation | -8.09e-02 |
Pos logit contributions | |
ier | +0.110 |
than | +0.100 |
ies | +0.100 |
owing | +0.099 |
- | +0.099 |
Neg logit contributions | |
Here | -0.152 |
First | -0.150 |
The | -0.147 |
Everyone | -0.142 |
Higher | -0.142 |
Token | They |
Token ID | 1119 |
Feature activation | -1.59e-02 |
Pos logit contributions | |
Grab | +0.108 |
V | +0.107 |
Here | +0.106 |
Higher | +0.093 |
Ro | +0.088 |
Neg logit contributions | |
than | -0.200 |
ever | -0.170 |
- | -0.165 |
too | -0.160 |
forever | -0.158 |
Token | cannot |
Token ID | 2314 |
Feature activation | +6.63e-03 |
Pos logit contributions | |
V | +0.029 |
Here | +0.028 |
Higher | +0.027 |
Grab | +0.027 |
First | +0.026 |
Neg logit contributions | |
than | -0.035 |
- | -0.032 |
ies | -0.031 |
ier | -0.031 |
inches | -0.029 |
Token | afford |
Token ID | 5368 |
Feature activation | +5.31e-02 |
Pos logit contributions | |
than | +0.015 |
ies | +0.014 |
ier | +0.014 |
- | +0.014 |
inches | +0.013 |
Neg logit contributions | |
V | -0.011 |
Here | -0.011 |
Higher | -0.010 |
Grab | -0.010 |
First | -0.010 |
Token | gold |
Token ID | 3869 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
ies | +0.120 |
owing | +0.114 |
than | +0.114 |
ier | +0.113 |
yt | +0.107 |
Neg logit contributions | |
Here | -0.082 |
The | -0.078 |
V | -0.077 |
First | -0.073 |
Higher | -0.072 |
Token | ice |
Token ID | 4771 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
supposed | +0.220 |
owing | +0.219 |
rying | +0.202 |
believing | +0.200 |
ge | +0.188 |
Neg logit contributions | |
Here | -0.232 |
It | -0.227 |
The | -0.218 |
One | -0.208 |
Red | -0.207 |
Token | cream |
Token ID | 8566 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
believing | +0.232 |
knowing | +0.212 |
supposed | +0.208 |
owing | +0.208 |
inches | +0.196 |
Neg logit contributions | |
Here | -0.188 |
First | -0.168 |
The | -0.163 |
� | -0.161 |
Cream | -0.160 |
Token | . |
Token ID | 13 |
Feature activation | -8.94e-02 |
Pos logit contributions | |
ies | +0.173 |
owing | +0.156 |
ge | +0.150 |
inches | +0.149 |
ines | +0.146 |
Neg logit contributions | |
Here | -0.237 |
First | -0.220 |
The | -0.216 |
Hi | -0.208 |
Spot | -0.207 |
Token |
|
Token ID | 198 |
Feature activation | -7.85e-02 |
Pos logit contributions | |
Grab | +0.133 |
Here | +0.126 |
V | +0.123 |
Higher | +0.122 |
Ro | +0.109 |
Neg logit contributions | |
than | -0.217 |
ever | -0.175 |
- | -0.172 |
too | -0.166 |
anymore | -0.166 |
Token |
|
Token ID | 198 |
Feature activation | -8.19e-02 |
Pos logit contributions | |
Grab | +0.146 |
Higher | +0.140 |
Here | +0.134 |
First | +0.126 |
Ro | +0.123 |
Neg logit contributions | |
than | -0.158 |
- | -0.152 |
ines | -0.134 |
ies | -0.131 |
ever | -0.126 |
Token | to |
Token ID | 284 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
ier | +0.090 |
ge | +0.085 |
ies | +0.082 |
- | +0.081 |
inches | +0.080 |
Neg logit contributions | |
The | -0.107 |
Here | -0.104 |
Higher | -0.102 |
Let | -0.097 |
Look | -0.097 |
Token | be |
Token ID | 307 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
than | +0.045 |
ier | +0.042 |
- | +0.041 |
ies | +0.041 |
inches | +0.037 |
Neg logit contributions | |
V | -0.038 |
Here | -0.038 |
Higher | -0.036 |
Grab | -0.036 |
First | -0.034 |
Token | gold |
Token ID | 3869 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
than | +0.069 |
ies | +0.065 |
ier | +0.065 |
- | +0.064 |
ge | +0.061 |
Neg logit contributions | |
Here | -0.079 |
V | -0.079 |
Higher | -0.074 |
Grab | -0.074 |
Look | -0.072 |
Token | is |
Token ID | 318 |
Feature activation | +5.77e-02 |
Pos logit contributions | |
ge | +0.174 |
than | +0.173 |
supposed | +0.171 |
owing | +0.158 |
inches | +0.157 |
Neg logit contributions | |
The | -0.220 |
Here | -0.213 |
It | -0.209 |
There | -0.208 |
First | -0.205 |
Token | not |
Token ID | 407 |
Feature activation | +7.22e-02 |
Pos logit contributions | |
ies | +0.089 |
- | +0.089 |
than | +0.088 |
ier | +0.087 |
ge | +0.085 |
Neg logit contributions | |
Here | -0.115 |
The | -0.111 |
V | -0.108 |
Higher | -0.108 |
First | -0.105 |
Token | gold |
Token ID | 3869 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
owing | +0.128 |
- | +0.125 |
ines | +0.120 |
ig | +0.117 |
ester | +0.114 |
Neg logit contributions | |
The | -0.121 |
It | -0.114 |
Fin | -0.111 |
Their | -0.110 |
One | -0.109 |
Token | , |
Token ID | 11 |
Feature activation | +2.33e-02 |
Pos logit contributions | |
supposed | +0.202 |
owing | +0.193 |
than | +0.191 |
ge | +0.190 |
ies | +0.189 |
Neg logit contributions | |
The | -0.218 |
It | -0.215 |
One | -0.208 |
There | -0.201 |
A | -0.199 |
Token | but |
Token ID | 475 |
Feature activation | +9.44e-03 |
Pos logit contributions | |
ier | +0.036 |
ies | +0.035 |
- | +0.035 |
than | +0.035 |
ge | +0.033 |
Neg logit contributions | |
Here | -0.050 |
Higher | -0.048 |
The | -0.048 |
V | -0.048 |
Grab | -0.046 |
Token | something |
Token ID | 1223 |
Feature activation | +4.93e-02 |
Pos logit contributions | |
ier | +0.021 |
than | +0.020 |
- | +0.020 |
ies | +0.020 |
owing | +0.019 |
Neg logit contributions | |
Here | -0.014 |
Higher | -0.013 |
The | -0.013 |
V | -0.013 |
First | -0.013 |
Token | else |
Token ID | 2073 |
Feature activation | +5.94e-02 |
Pos logit contributions | |
owing | +0.107 |
ge | +0.107 |
ier | +0.106 |
ies | +0.105 |
than | +0.100 |
Neg logit contributions | |
Here | -0.069 |
The | -0.068 |
It | -0.066 |
Their | -0.061 |
Something | -0.061 |
Token | that |
Token ID | 326 |
Feature activation | +1.74e-02 |
Pos logit contributions | |
ier | +0.126 |
than | +0.122 |
- | +0.117 |
ge | +0.117 |
ies | +0.116 |
Neg logit contributions | |
The | -0.084 |
Here | -0.084 |
It | -0.079 |
Higher | -0.078 |
V | -0.075 |
Token | that |
Token ID | 326 |
Feature activation | +3.42e-02 |
Pos logit contributions | |
than | +0.052 |
ines | +0.051 |
ier | +0.050 |
ies | +0.050 |
- | +0.050 |
Neg logit contributions | |
V | -0.044 |
The | -0.043 |
Higher | -0.041 |
First | -0.041 |
Max | -0.040 |
Token | easy |
Token ID | 2562 |
Feature activation | +8.96e-02 |
Pos logit contributions | |
than | +0.067 |
ier | +0.066 |
ge | +0.063 |
ies | +0.063 |
- | +0.062 |
Neg logit contributions | |
Higher | -0.058 |
Here | -0.057 |
V | -0.057 |
The | -0.057 |
Ro | -0.054 |
Token | to |
Token ID | 284 |
Feature activation | +3.18e-02 |
Pos logit contributions | |
ier | +0.130 |
ge | +0.127 |
than | +0.125 |
owing | +0.123 |
ies | +0.120 |
Neg logit contributions | |
The | -0.171 |
First | -0.170 |
It | -0.168 |
Everyone | -0.166 |
Higher | -0.162 |
Token | make |
Token ID | 787 |
Feature activation | +4.69e-02 |
Pos logit contributions | |
than | +0.071 |
ier | +0.067 |
- | +0.065 |
ies | +0.064 |
inches | +0.060 |
Neg logit contributions | |
V | -0.054 |
Here | -0.051 |
Higher | -0.050 |
Ro | -0.049 |
Many | -0.048 |
Token | the |
Token ID | 262 |
Feature activation | +4.38e-02 |
Pos logit contributions | |
than | +0.102 |
ier | +0.095 |
ies | +0.090 |
ge | +0.088 |
owing | +0.088 |
Neg logit contributions | |
V | -0.082 |
Higher | -0.075 |
Everyone | -0.075 |
Here | -0.074 |
The | -0.074 |
Token | rainbow |
Token ID | 27223 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
than | +0.101 |
ier | +0.096 |
ge | +0.094 |
- | +0.093 |
ies | +0.091 |
Neg logit contributions | |
Here | -0.072 |
V | -0.068 |
Look | -0.067 |
Higher | -0.067 |
The | -0.062 |
Token | stay |
Token ID | 2652 |
Feature activation | +7.72e-02 |
Pos logit contributions | |
than | +0.247 |
ge | +0.231 |
ier | +0.220 |
owing | +0.218 |
supposed | +0.213 |
Neg logit contributions | |
The | -0.216 |
It | -0.202 |
Everyone | -0.202 |
Higher | -0.194 |
Many | -0.190 |
Token | . |
Token ID | 13 |
Feature activation | -3.93e-02 |
Pos logit contributions | |
than | +0.134 |
ier | +0.128 |
owing | +0.117 |
ies | +0.117 |
inches | +0.116 |
Neg logit contributions | |
The | -0.130 |
V | -0.129 |
Here | -0.128 |
Look | -0.120 |
First | -0.120 |
Token | It |
Token ID | 632 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
Here | +0.054 |
Grab | +0.050 |
V | +0.050 |
Higher | +0.045 |
Bang | +0.044 |
Neg logit contributions | |
than | -0.098 |
- | -0.088 |
ever | -0.084 |
inches | -0.082 |
too | -0.081 |
Token | will |
Token ID | 481 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
than | +0.049 |
ier | +0.045 |
ies | +0.044 |
ge | +0.043 |
- | +0.043 |
Neg logit contributions | |
V | -0.033 |
Here | -0.032 |
Grab | -0.031 |
Higher | -0.031 |
First | -0.030 |
Token | always |
Token ID | 1464 |
Feature activation | +4.49e-02 |
Pos logit contributions | |
than | +0.046 |
- | +0.044 |
ies | +0.043 |
ier | +0.043 |
owing | +0.040 |
Neg logit contributions | |
V | -0.032 |
Here | -0.032 |
Higher | -0.030 |
Ro | -0.029 |
Grab | -0.029 |
Token | slide |
Token ID | 10649 |
Feature activation | +5.94e-02 |
Pos logit contributions | |
than | +0.027 |
ier | +0.025 |
ies | +0.025 |
- | +0.025 |
ge | +0.023 |
Neg logit contributions | |
Here | -0.025 |
V | -0.024 |
Higher | -0.023 |
Grab | -0.023 |
Look | -0.022 |
Token | is |
Token ID | 318 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
than | +0.135 |
ier | +0.128 |
ies | +0.127 |
ge | +0.126 |
owing | +0.121 |
Neg logit contributions | |
V | -0.087 |
Here | -0.083 |
The | -0.080 |
Grab | -0.080 |
Higher | -0.079 |
Token | not |
Token ID | 407 |
Feature activation | +3.96e-02 |
Pos logit contributions | |
ies | +0.045 |
ier | +0.045 |
than | +0.044 |
owing | +0.044 |
- | +0.043 |
Neg logit contributions | |
V | -0.047 |
Here | -0.046 |
Higher | -0.043 |
The | -0.042 |
Look | -0.042 |
Token | made |
Token ID | 925 |
Feature activation | +4.93e-02 |
Pos logit contributions | |
ines | +0.081 |
owing | +0.074 |
- | +0.072 |
ier | +0.071 |
than | +0.070 |
Neg logit contributions | |
The | -0.058 |
Max | -0.055 |
Red | -0.053 |
It | -0.052 |
Big | -0.052 |
Token | of |
Token ID | 286 |
Feature activation | +5.39e-02 |
Pos logit contributions | |
than | +0.086 |
ies | +0.077 |
ge | +0.073 |
owing | +0.072 |
ier | +0.071 |
Neg logit contributions | |
V | -0.104 |
Higher | -0.098 |
The | -0.097 |
Everyone | -0.095 |
First | -0.095 |
Token | gold |
Token ID | 3869 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
than | +0.106 |
ier | +0.105 |
ies | +0.102 |
owing | +0.101 |
- | +0.100 |
Neg logit contributions | |
Here | -0.095 |
V | -0.090 |
Higher | -0.088 |
The | -0.088 |
Look | -0.087 |
Token | , |
Token ID | 11 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
supposed | +0.242 |
owing | +0.221 |
asured | +0.202 |
ge | +0.201 |
knowing | +0.196 |
Neg logit contributions | |
It | -0.196 |
The | -0.195 |
One | -0.174 |
He | -0.172 |
Ben | -0.172 |
Token | but |
Token ID | 475 |
Feature activation | +2.26e-02 |
Pos logit contributions | |
ier | +0.045 |
ies | +0.043 |
ge | +0.042 |
than | +0.042 |
owing | +0.042 |
Neg logit contributions | |
Here | -0.044 |
Higher | -0.042 |
V | -0.042 |
The | -0.041 |
It | -0.041 |
Token | it |
Token ID | 340 |
Feature activation | +2.54e-02 |
Pos logit contributions | |
than | +0.051 |
ier | +0.050 |
ies | +0.049 |
owing | +0.049 |
- | +0.049 |
Neg logit contributions | |
Here | -0.032 |
V | -0.030 |
Higher | -0.030 |
The | -0.029 |
First | -0.029 |
Token | is |
Token ID | 318 |
Feature activation | +2.62e-02 |
Pos logit contributions | |
than | +0.062 |
ier | +0.058 |
ies | +0.057 |
ge | +0.057 |
- | +0.054 |
Neg logit contributions | |
V | -0.035 |
Here | -0.034 |
Grab | -0.033 |
The | -0.032 |
Higher | -0.031 |
Token | brilliant |
Token ID | 10457 |
Feature activation | +8.72e-02 |
Pos logit contributions | |
owing | +0.052 |
ies | +0.051 |
ier | +0.051 |
- | +0.049 |
ge | +0.048 |
Neg logit contributions | |
The | -0.044 |
Here | -0.043 |
V | -0.041 |
It | -0.040 |
One | -0.040 |
Token | of |
Token ID | 286 |
Feature activation | +4.07e-02 |
Pos logit contributions | |
ge | +0.139 |
ier | +0.134 |
supposed | +0.133 |
ies | +0.131 |
owing | +0.126 |
Neg logit contributions | |
Here | -0.202 |
The | -0.196 |
Higher | -0.186 |
Hi | -0.185 |
Everyone | -0.182 |
Token | gardens |
Token ID | 25476 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
ier | +0.083 |
than | +0.078 |
ies | +0.076 |
owing | +0.075 |
ge | +0.073 |
Neg logit contributions | |
Here | -0.076 |
Higher | -0.068 |
Many | -0.068 |
V | -0.067 |
Look | -0.066 |
Token | can |
Token ID | 460 |
Feature activation | +3.67e-02 |
Pos logit contributions | |
ies | +0.140 |
than | +0.134 |
inches | +0.129 |
ge | +0.129 |
ier | +0.128 |
Neg logit contributions | |
The | -0.205 |
Higher | -0.198 |
Here | -0.197 |
Everyone | -0.184 |
First | -0.181 |
Token | still |
Token ID | 991 |
Feature activation | +7.73e-02 |
Pos logit contributions | |
than | +0.080 |
ies | +0.074 |
- | +0.072 |
ier | +0.071 |
inches | +0.067 |
Neg logit contributions | |
Here | -0.061 |
V | -0.061 |
Higher | -0.058 |
The | -0.056 |
Grab | -0.056 |
Token | bl |
Token ID | 698 |
Feature activation | +9.88e-02 |
Pos logit contributions | |
than | +0.133 |
ier | +0.129 |
ies | +0.127 |
- | +0.124 |
ishing | +0.120 |
Neg logit contributions | |
The | -0.132 |
Here | -0.124 |
Everyone | -0.122 |
Hi | -0.120 |
Higher | -0.119 |
Token | ossom |
Token ID | 25548 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
than | +0.187 |
inches | +0.169 |
times | +0.164 |
ier | +0.156 |
ge | +0.156 |
Neg logit contributions | |
V | -0.164 |
First | -0.162 |
Everyone | -0.160 |
Higher | -0.158 |
Here | -0.155 |
Token | . |
Token ID | 13 |
Feature activation | -8.87e-02 |
Pos logit contributions | |
ge | +0.207 |
ier | +0.184 |
supposed | +0.182 |
owing | +0.171 |
inches | +0.166 |
Neg logit contributions | |
The | -0.254 |
Here | -0.239 |
Everyone | -0.237 |
Sun | -0.236 |
Higher | -0.235 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -3.30e-02 |
Pos logit contributions | |
V | +0.131 |
Here | +0.124 |
Grab | +0.124 |
Ro | +0.120 |
Bang | +0.113 |
Neg logit contributions | |
than | -0.216 |
- | -0.202 |
ever | -0.185 |
ies | -0.185 |
forever | -0.178 |
Token | Sus |
Token ID | 30746 |
Feature activation | +3.82e-02 |
Pos logit contributions | |
V | +0.048 |
Here | +0.047 |
Higher | +0.047 |
Grab | +0.047 |
First | +0.045 |
Neg logit contributions | |
than | -0.079 |
- | -0.076 |
ier | -0.071 |
ever | -0.069 |
ies | -0.069 |
Token | ie |
Token ID | 494 |
Feature activation | -7.45e-03 |
Pos logit contributions | |
than | +0.053 |
ier | +0.038 |
inches | +0.037 |
- | +0.034 |
ever | +0.033 |
Neg logit contributions | |
V | -0.101 |
Here | -0.097 |
Look | -0.095 |
Higher | -0.094 |
First | -0.092 |
Token | and |
Token ID | 290 |
Feature activation | -4.81e-02 |
Pos logit contributions | |
V | +0.017 |
Here | +0.016 |
Higher | +0.016 |
Grab | +0.016 |
First | +0.015 |
Neg logit contributions | |
than | -0.013 |
- | -0.012 |
ies | -0.012 |
ier | -0.011 |
ever | -0.011 |
Token | nest |
Token ID | 16343 |
Feature activation | +5.99e-02 |
Pos logit contributions | |
ier | +0.120 |
than | +0.111 |
ge | +0.108 |
owing | +0.101 |
ets | +0.099 |
Neg logit contributions | |
The | -0.079 |
Many | -0.069 |
Here | -0.067 |
First | -0.066 |
One | -0.066 |
Token | couldn |
Token ID | 3521 |
Feature activation | -8.10e-04 |
Pos logit contributions | |
than | +0.135 |
ge | +0.122 |
ies | +0.112 |
ier | +0.111 |
inches | +0.111 |
Neg logit contributions | |
The | -0.088 |
Here | -0.084 |
Higher | -0.083 |
Grab | -0.080 |
V | -0.079 |
Token | 't |
Token ID | 470 |
Feature activation | +3.30e-02 |
Pos logit contributions | |
V | +0.001 |
Here | +0.001 |
Higher | +0.001 |
Grab | +0.001 |
First | +0.001 |
Neg logit contributions | |
than | -0.002 |
- | -0.002 |
ier | -0.002 |
ies | -0.002 |
inches | -0.002 |
Token | prevent |
Token ID | 2948 |
Feature activation | +3.52e-02 |
Pos logit contributions | |
than | +0.076 |
ies | +0.070 |
ier | +0.067 |
- | +0.067 |
inches | +0.063 |
Neg logit contributions | |
V | -0.054 |
Higher | -0.051 |
Here | -0.051 |
The | -0.049 |
Ro | -0.049 |
Token | the |
Token ID | 262 |
Feature activation | +3.71e-02 |
Pos logit contributions | |
than | +0.077 |
ier | +0.073 |
ies | +0.069 |
- | +0.068 |
ines | +0.068 |
Neg logit contributions | |
The | -0.055 |
V | -0.053 |
Here | -0.050 |
Higher | -0.049 |
Many | -0.048 |
Token | fire |
Token ID | 2046 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
than | +0.088 |
ier | +0.082 |
ies | +0.079 |
ge | +0.078 |
- | +0.077 |
Neg logit contributions | |
Here | -0.058 |
V | -0.056 |
Many | -0.052 |
Higher | -0.051 |
The | -0.051 |
Token | from |
Token ID | 422 |
Feature activation | +3.12e-02 |
Pos logit contributions | |
than | +0.197 |
ishing | +0.196 |
ines | +0.192 |
asured | +0.182 |
ies | +0.181 |
Neg logit contributions | |
The | -0.276 |
It | -0.231 |
Big | -0.230 |
There | -0.229 |
Dog | -0.222 |
Token | hurting |
Token ID | 21530 |
Feature activation | +6.62e-02 |
Pos logit contributions | |
than | +0.051 |
ier | +0.045 |
- | +0.044 |
ies | +0.043 |
owing | +0.041 |
Neg logit contributions | |
V | -0.068 |
Higher | -0.066 |
Here | -0.066 |
Look | -0.064 |
Ro | -0.064 |
Token | her |
Token ID | 607 |
Feature activation | +7.31e-02 |
Pos logit contributions | |
than | +0.117 |
ier | +0.112 |
ies | +0.107 |
ines | +0.104 |
owing | +0.102 |
Neg logit contributions | |
The | -0.123 |
It | -0.111 |
V | -0.110 |
Higher | -0.109 |
Here | -0.107 |
Token | . |
Token ID | 13 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
ier | +0.126 |
than | +0.119 |
ets | +0.116 |
ines | +0.112 |
supposed | +0.111 |
Neg logit contributions | |
The | -0.127 |
Big | -0.100 |
Mom | -0.098 |
Many | -0.098 |
Every | -0.095 |
Token |
|
Token ID | 198 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
V | +0.191 |
Grab | +0.182 |
Here | +0.175 |
Pull | +0.173 |
Bang | +0.164 |
Neg logit contributions | |
- | -0.207 |
than | -0.205 |
for | -0.177 |
ies | -0.177 |
ever | -0.175 |
Token | waiting |
Token ID | 4953 |
Feature activation | +7.13e-02 |
Pos logit contributions | |
V | +0.027 |
Here | +0.026 |
Grab | +0.026 |
Higher | +0.025 |
First | +0.024 |
Neg logit contributions | |
than | -0.035 |
ier | -0.033 |
ies | -0.032 |
- | -0.030 |
ge | -0.029 |
Token | wouldn |
Token ID | 3636 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
than | +0.084 |
ier | +0.081 |
ge | +0.080 |
owing | +0.075 |
- | +0.073 |
Neg logit contributions | |
Here | -0.161 |
The | -0.159 |
First | -0.157 |
Higher | -0.151 |
Everyone | -0.151 |
Token | 't |
Token ID | 470 |
Feature activation | +3.57e-02 |
Pos logit contributions | |
than | +0.080 |
ies | +0.074 |
- | +0.072 |
ier | +0.072 |
inches | +0.070 |
Neg logit contributions | |
V | -0.029 |
Higher | -0.026 |
Here | -0.026 |
Grab | -0.024 |
First | -0.024 |
Token | get |
Token ID | 651 |
Feature activation | +5.22e-02 |
Pos logit contributions | |
than | +0.079 |
- | +0.075 |
ier | +0.072 |
ies | +0.072 |
inches | +0.071 |
Neg logit contributions | |
First | -0.047 |
V | -0.047 |
The | -0.046 |
Here | -0.046 |
Many | -0.046 |
Token | their |
Token ID | 511 |
Feature activation | +4.48e-02 |
Pos logit contributions | |
than | +0.094 |
ies | +0.081 |
ier | +0.080 |
- | +0.079 |
inches | +0.074 |
Neg logit contributions | |
Here | -0.107 |
V | -0.100 |
Grab | -0.099 |
Higher | -0.099 |
The | -0.099 |
Token | gifts |
Token ID | 13201 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
than | +0.081 |
ier | +0.080 |
- | +0.072 |
ies | +0.072 |
owing | +0.069 |
Neg logit contributions | |
Here | -0.087 |
First | -0.084 |
V | -0.081 |
Higher | -0.080 |
Grab | -0.078 |
Token | now |
Token ID | 783 |
Feature activation | +5.97e-02 |
Pos logit contributions | |
than | +0.192 |
inches | +0.189 |
ies | +0.173 |
- | +0.173 |
ishing | +0.171 |
Neg logit contributions | |
First | -0.251 |
The | -0.249 |
Here | -0.248 |
Everyone | -0.234 |
It | -0.232 |
Token | . |
Token ID | 13 |
Feature activation | -9.29e-02 |
Pos logit contributions | |
than | +0.084 |
owing | +0.079 |
ies | +0.078 |
inches | +0.077 |
- | +0.076 |
Neg logit contributions | |
Here | -0.125 |
The | -0.123 |
First | -0.122 |
Everyone | -0.118 |
V | -0.116 |
Token | Just |
Token ID | 2329 |
Feature activation | -1.08e-03 |
Pos logit contributions | |
V | +0.122 |
Grab | +0.121 |
Here | +0.120 |
Higher | +0.112 |
Ro | +0.104 |
Neg logit contributions | |
than | -0.229 |
- | -0.207 |
ever | -0.199 |
forever | -0.193 |
ies | -0.192 |
Token | as |
Token ID | 355 |
Feature activation | -3.61e-02 |
Pos logit contributions | |
Here | +0.002 |
V | +0.002 |
Higher | +0.002 |
First | +0.002 |
The | +0.002 |
Neg logit contributions | |
ier | -0.002 |
than | -0.002 |
- | -0.002 |
ies | -0.002 |
inches | -0.002 |
Token | they |
Token ID | 484 |
Feature activation | +5.08e-03 |
Pos logit contributions | |
V | +0.062 |
Here | +0.061 |
Higher | +0.059 |
Grab | +0.057 |
First | +0.056 |
Neg logit contributions | |
than | -0.080 |
ier | -0.075 |
ies | -0.075 |
- | -0.074 |
owing | -0.069 |
Token | she |
Token ID | 673 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
ier | +0.016 |
than | +0.015 |
- | +0.014 |
ge | +0.014 |
ies | +0.014 |
Neg logit contributions | |
Higher | -0.014 |
Here | -0.013 |
The | -0.013 |
First | -0.013 |
V | -0.012 |
Token | couldn |
Token ID | 3521 |
Feature activation | -1.70e-03 |
Pos logit contributions | |
than | +0.080 |
ier | +0.075 |
ies | +0.074 |
ge | +0.072 |
- | +0.071 |
Neg logit contributions | |
Here | -0.047 |
V | -0.046 |
Grab | -0.044 |
The | -0.042 |
Higher | -0.041 |
Token | 't |
Token ID | 470 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
V | +0.002 |
Here | +0.002 |
Higher | +0.002 |
Grab | +0.002 |
First | +0.002 |
Neg logit contributions | |
than | -0.005 |
ies | -0.004 |
- | -0.004 |
ier | -0.004 |
inches | -0.004 |
Token | make |
Token ID | 787 |
Feature activation | +3.33e-02 |
Pos logit contributions | |
than | +0.043 |
ies | +0.041 |
ier | +0.040 |
- | +0.039 |
inches | +0.038 |
Neg logit contributions | |
V | -0.031 |
Here | -0.030 |
Many | -0.029 |
Higher | -0.029 |
The | -0.029 |
Token | the |
Token ID | 262 |
Feature activation | +3.92e-02 |
Pos logit contributions | |
than | +0.082 |
ier | +0.077 |
ies | +0.075 |
ge | +0.074 |
owing | +0.073 |
Neg logit contributions | |
The | -0.046 |
V | -0.045 |
Everyone | -0.043 |
Higher | -0.043 |
It | -0.041 |
Token | rainbow |
Token ID | 27223 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
than | +0.094 |
ier | +0.089 |
ge | +0.088 |
- | +0.087 |
ies | +0.086 |
Neg logit contributions | |
Here | -0.061 |
Look | -0.058 |
Higher | -0.058 |
V | -0.058 |
The | -0.053 |
Token | stay |
Token ID | 2652 |
Feature activation | +7.31e-02 |
Pos logit contributions | |
than | +0.226 |
ge | +0.210 |
supposed | +0.202 |
ishing | +0.197 |
ier | +0.196 |
Neg logit contributions | |
The | -0.233 |
It | -0.216 |
Everyone | -0.211 |
Higher | -0.208 |
Many | -0.200 |
Token | , |
Token ID | 11 |
Feature activation | +1.72e-02 |
Pos logit contributions | |
than | +0.119 |
ier | +0.113 |
inches | +0.106 |
ies | +0.105 |
ines | +0.102 |
Neg logit contributions | |
The | -0.130 |
Here | -0.128 |
V | -0.123 |
Higher | -0.118 |
First | -0.116 |
Token | but |
Token ID | 475 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
ier | +0.030 |
ies | +0.028 |
- | +0.027 |
than | +0.026 |
ines | +0.026 |
Neg logit contributions | |
Here | -0.035 |
Higher | -0.035 |
The | -0.035 |
It | -0.033 |
Grab | -0.033 |
Token | she |
Token ID | 673 |
Feature activation | +8.69e-03 |
Pos logit contributions | |
Here | +0.027 |
Higher | +0.026 |
The | +0.026 |
First | +0.025 |
It | +0.024 |
Neg logit contributions | |
than | -0.046 |
ier | -0.046 |
ies | -0.044 |
- | -0.043 |
ge | -0.043 |
Token | was |
Token ID | 373 |
Feature activation | +2.70e-03 |
Pos logit contributions | |
than | +0.020 |
ier | +0.019 |
ies | +0.018 |
ge | +0.018 |
- | +0.018 |
Neg logit contributions | |
Here | -0.012 |
V | -0.012 |
Grab | -0.011 |
The | -0.011 |
First | -0.011 |
Token | You |
Token ID | 1639 |
Feature activation | -1.70e-03 |
Pos logit contributions | |
V | +0.025 |
Higher | +0.025 |
Here | +0.024 |
Grab | +0.022 |
First | +0.022 |
Neg logit contributions | |
than | -0.068 |
- | -0.062 |
ies | -0.061 |
ier | -0.061 |
ever | -0.059 |
Token | can |
Token ID | 460 |
Feature activation | +8.92e-04 |
Pos logit contributions | |
V | +0.003 |
Here | +0.003 |
Higher | +0.003 |
Grab | +0.003 |
First | +0.003 |
Neg logit contributions | |
than | -0.004 |
- | -0.003 |
ies | -0.003 |
ier | -0.003 |
inches | -0.003 |
Token | 't |
Token ID | 470 |
Feature activation | +6.12e-03 |
Pos logit contributions | |
than | +0.002 |
ies | +0.002 |
ier | +0.002 |
- | +0.002 |
inches | +0.002 |
Neg logit contributions | |
V | -0.002 |
Here | -0.001 |
Higher | -0.001 |
Grab | -0.001 |
First | -0.001 |
Token | keep |
Token ID | 1394 |
Feature activation | +4.46e-02 |
Pos logit contributions | |
than | +0.014 |
ies | +0.013 |
ier | +0.013 |
- | +0.013 |
inches | +0.012 |
Neg logit contributions | |
V | -0.010 |
Here | -0.009 |
Higher | -0.009 |
Grab | -0.009 |
Ro | -0.009 |
Token | this |
Token ID | 428 |
Feature activation | +8.23e-02 |
Pos logit contributions | |
than | +0.097 |
ier | +0.093 |
ies | +0.088 |
- | +0.082 |
ines | +0.079 |
Neg logit contributions | |
V | -0.074 |
Here | -0.073 |
Higher | -0.071 |
Look | -0.069 |
Grab | -0.069 |
Token | gold |
Token ID | 3869 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
than | +0.142 |
ier | +0.139 |
ies | +0.134 |
owing | +0.130 |
ge | +0.128 |
Neg logit contributions | |
Here | -0.135 |
The | -0.133 |
Higher | -0.132 |
First | -0.129 |
V | -0.128 |
Token | , |
Token ID | 11 |
Feature activation | +5.97e-03 |
Pos logit contributions | |
supposed | +0.213 |
owing | +0.206 |
ge | +0.198 |
ines | +0.195 |
than | +0.190 |
Neg logit contributions | |
It | -0.210 |
The | -0.208 |
Everyone | -0.198 |
Here | -0.197 |
People | -0.189 |
Token | you |
Token ID | 345 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
than | +0.011 |
ier | +0.011 |
ies | +0.011 |
- | +0.010 |
owing | +0.010 |
Neg logit contributions | |
V | -0.011 |
Here | -0.011 |
Higher | -0.011 |
Grab | -0.011 |
The | -0.010 |
Token | have |
Token ID | 423 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
than | +0.039 |
ier | +0.036 |
ies | +0.035 |
- | +0.035 |
inches | +0.033 |
Neg logit contributions | |
V | -0.026 |
Here | -0.025 |
Higher | -0.024 |
First | -0.024 |
Grab | -0.024 |
Token | to |
Token ID | 284 |
Feature activation | +3.61e-03 |
Pos logit contributions | |
than | +0.040 |
ies | +0.039 |
ier | +0.039 |
- | +0.036 |
owing | +0.036 |
Neg logit contributions | |
Here | -0.050 |
V | -0.048 |
Higher | -0.047 |
The | -0.047 |
First | -0.046 |
Token | give |
Token ID | 1577 |
Feature activation | +4.09e-02 |
Pos logit contributions | |
than | +0.008 |
ier | +0.007 |
- | +0.007 |
ies | +0.007 |
inches | +0.006 |
Neg logit contributions | |
V | -0.006 |
Here | -0.006 |
Grab | -0.006 |
Higher | -0.006 |
First | -0.006 |
Token | mixer |
Token ID | 33938 |
Feature activation | +9.10e-02 |
Pos logit contributions | |
than | +0.186 |
ier | +0.180 |
ge | +0.176 |
owing | +0.171 |
ies | +0.170 |
Neg logit contributions | |
Here | -0.153 |
V | -0.144 |
Look | -0.142 |
Hi | -0.138 |
Higher | -0.137 |
Token | that |
Token ID | 326 |
Feature activation | +1.10e-03 |
Pos logit contributions | |
than | +0.121 |
ge | +0.117 |
ies | +0.116 |
ier | +0.109 |
inches | +0.109 |
Neg logit contributions | |
Here | -0.188 |
The | -0.187 |
It | -0.182 |
V | -0.178 |
Everyone | -0.174 |
Token | could |
Token ID | 714 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
than | +0.002 |
ier | +0.002 |
ies | +0.002 |
- | +0.002 |
ge | +0.002 |
Neg logit contributions | |
Here | -0.002 |
V | -0.002 |
Higher | -0.002 |
Grab | -0.002 |
Look | -0.002 |
Token | make |
Token ID | 787 |
Feature activation | +4.82e-02 |
Pos logit contributions | |
than | +0.032 |
ies | +0.029 |
ier | +0.029 |
- | +0.029 |
inches | +0.027 |
Neg logit contributions | |
V | -0.024 |
Here | -0.024 |
Higher | -0.023 |
Grab | -0.022 |
First | -0.021 |
Token | a |
Token ID | 257 |
Feature activation | +4.92e-02 |
Pos logit contributions | |
than | +0.110 |
ier | +0.103 |
ge | +0.101 |
owing | +0.097 |
ies | +0.095 |
Neg logit contributions | |
V | -0.073 |
Everyone | -0.071 |
Here | -0.069 |
The | -0.069 |
Higher | -0.068 |
Token | rainbow |
Token ID | 27223 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
than | +0.106 |
ier | +0.104 |
owing | +0.102 |
ge | +0.096 |
ies | +0.094 |
Neg logit contributions | |
Here | -0.086 |
V | -0.086 |
Look | -0.082 |
Ro | -0.079 |
Higher | -0.076 |
Token | with |
Token ID | 351 |
Feature activation | +5.54e-02 |
Pos logit contributions | |
ge | +0.220 |
owing | +0.210 |
than | +0.207 |
asured | +0.194 |
ier | +0.194 |
Neg logit contributions | |
Everyone | -0.227 |
The | -0.224 |
V | -0.222 |
It | -0.218 |
Ro | -0.212 |
Token | soap |
Token ID | 19533 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
than | +0.125 |
owing | +0.122 |
ier | +0.121 |
ies | +0.120 |
ge | +0.117 |
Neg logit contributions | |
Here | -0.087 |
Higher | -0.080 |
V | -0.079 |
Look | -0.078 |
The | -0.077 |
Token | bubbles |
Token ID | 25037 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
ge | +0.159 |
owing | +0.157 |
than | +0.157 |
ies | +0.155 |
inches | +0.148 |
Neg logit contributions | |
Here | -0.225 |
The | -0.222 |
It | -0.216 |
V | -0.216 |
Ro | -0.213 |
Token | . |
Token ID | 13 |
Feature activation | -9.18e-02 |
Pos logit contributions | |
ge | +0.170 |
ies | +0.167 |
than | +0.163 |
inches | +0.160 |
owing | +0.159 |
Neg logit contributions | |
Here | -0.231 |
V | -0.226 |
The | -0.224 |
Higher | -0.224 |
First | -0.222 |
Token | Daisy |
Token ID | 40355 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
V | +0.181 |
Ro | +0.153 |
Here | +0.152 |
Bang | +0.147 |
Grab | +0.145 |
Neg logit contributions | |
than | -0.188 |
- | -0.172 |
! | -0.155 |
for | -0.155 |
ies | -0.150 |
Token | she |
Token ID | 673 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
Here | +0.009 |
V | +0.009 |
Higher | +0.008 |
Grab | +0.008 |
Ro | +0.008 |
Neg logit contributions | |
than | -0.011 |
ier | -0.011 |
ies | -0.011 |
- | -0.010 |
owing | -0.010 |
Token | needed |
Token ID | 2622 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
than | +0.027 |
ier | +0.024 |
ies | +0.024 |
- | +0.024 |
ge | +0.023 |
Neg logit contributions | |
V | -0.020 |
Here | -0.019 |
Grab | -0.018 |
Higher | -0.018 |
First | -0.017 |
Token | to |
Token ID | 284 |
Feature activation | -1.44e-03 |
Pos logit contributions | |
than | +0.040 |
ies | +0.038 |
ier | +0.037 |
- | +0.034 |
ge | +0.033 |
Neg logit contributions | |
Here | -0.044 |
V | -0.042 |
Higher | -0.042 |
First | -0.041 |
Grab | -0.040 |
Token | prevent |
Token ID | 2948 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
V | +0.003 |
Here | +0.003 |
Higher | +0.003 |
Grab | +0.003 |
First | +0.002 |
Neg logit contributions | |
than | -0.003 |
- | -0.003 |
ies | -0.003 |
ier | -0.003 |
inches | -0.002 |
Token | the |
Token ID | 262 |
Feature activation | +4.39e-02 |
Pos logit contributions | |
than | +0.063 |
ier | +0.060 |
ies | +0.059 |
- | +0.055 |
owing | +0.055 |
Neg logit contributions | |
V | -0.049 |
Here | -0.046 |
The | -0.045 |
First | -0.044 |
Higher | -0.043 |
Token | shampoo |
Token ID | 47259 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
than | +0.086 |
ier | +0.079 |
ies | +0.077 |
owing | +0.075 |
- | +0.075 |
Neg logit contributions | |
Here | -0.085 |
V | -0.081 |
Ro | -0.077 |
Grab | -0.077 |
Look | -0.075 |
Token | from |
Token ID | 422 |
Feature activation | +4.03e-02 |
Pos logit contributions | |
than | +0.204 |
owing | +0.176 |
ies | +0.170 |
inches | +0.170 |
asured | +0.170 |
Neg logit contributions | |
The | -0.237 |
First | -0.233 |
V | -0.225 |
Here | -0.224 |
Grab | -0.220 |
Token | getting |
Token ID | 1972 |
Feature activation | +5.77e-02 |
Pos logit contributions | |
than | +0.072 |
ier | +0.068 |
ies | +0.067 |
- | +0.063 |
owing | +0.061 |
Neg logit contributions | |
V | -0.082 |
Here | -0.080 |
Look | -0.075 |
Higher | -0.074 |
Grab | -0.074 |
Token | in |
Token ID | 287 |
Feature activation | +4.29e-02 |
Pos logit contributions | |
than | +0.106 |
ies | +0.092 |
ier | +0.089 |
owing | +0.086 |
- | +0.086 |
Neg logit contributions | |
V | -0.122 |
Here | -0.121 |
Grab | -0.114 |
Higher | -0.111 |
Look | -0.110 |
Token | her |
Token ID | 607 |
Feature activation | +4.51e-02 |
Pos logit contributions | |
ier | +0.092 |
ies | +0.092 |
owing | +0.089 |
than | +0.089 |
- | +0.085 |
Neg logit contributions | |
Here | -0.069 |
V | -0.069 |
Look | -0.065 |
The | -0.065 |
First | -0.063 |
Token | eyes |
Token ID | 2951 |
Feature activation | +8.36e-02 |
Pos logit contributions | |
ier | +0.096 |
than | +0.092 |
ies | +0.085 |
ge | +0.081 |
ever | +0.078 |
Neg logit contributions | |
Here | -0.078 |
The | -0.075 |
First | -0.074 |
V | -0.074 |
Higher | -0.070 |
Token | the |
Token ID | 262 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
Here | +0.014 |
V | +0.014 |
Higher | +0.013 |
First | +0.013 |
The | +0.013 |
Neg logit contributions | |
ier | -0.018 |
than | -0.018 |
- | -0.017 |
ies | -0.017 |
owing | -0.016 |
Token | best |
Token ID | 1266 |
Feature activation | +6.31e-02 |
Pos logit contributions | |
than | +0.022 |
ier | +0.020 |
- | +0.020 |
ies | +0.020 |
ge | +0.018 |
Neg logit contributions | |
V | -0.021 |
Here | -0.021 |
Higher | -0.020 |
Grab | -0.020 |
Ro | -0.019 |
Token | treasure |
Token ID | 14068 |
Feature activation | +9.75e-02 |
Pos logit contributions | |
ier | +0.122 |
owing | +0.118 |
ge | +0.115 |
than | +0.109 |
inches | +0.101 |
Neg logit contributions | |
First | -0.109 |
Ro | -0.102 |
V | -0.101 |
Here | -0.099 |
Everyone | -0.099 |
Token | is |
Token ID | 318 |
Feature activation | +3.55e-02 |
Pos logit contributions | |
ge | +0.184 |
than | +0.175 |
owing | +0.172 |
ier | +0.169 |
inches | +0.166 |
Neg logit contributions | |
Here | -0.149 |
The | -0.141 |
Everyone | -0.139 |
People | -0.139 |
It | -0.139 |
Token | not |
Token ID | 407 |
Feature activation | +5.25e-02 |
Pos logit contributions | |
than | +0.069 |
ies | +0.066 |
owing | +0.065 |
ier | +0.065 |
- | +0.065 |
Neg logit contributions | |
Here | -0.062 |
V | -0.061 |
The | -0.057 |
Higher | -0.056 |
First | -0.055 |
Token | gold |
Token ID | 3869 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
owing | +0.100 |
ines | +0.098 |
- | +0.097 |
ier | +0.094 |
ies | +0.092 |
Neg logit contributions | |
The | -0.082 |
Max | -0.082 |
Red | -0.075 |
Blue | -0.075 |
It | -0.074 |
Token | or |
Token ID | 393 |
Feature activation | +4.80e-02 |
Pos logit contributions | |
supposed | +0.203 |
owing | +0.197 |
rying | +0.192 |
ge | +0.188 |
believing | +0.182 |
Neg logit contributions | |
It | -0.205 |
The | -0.201 |
Max | -0.192 |
Blue | -0.191 |
One | -0.190 |
Token | jewels |
Token ID | 42701 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
ier | +0.096 |
- | +0.093 |
than | +0.092 |
owing | +0.092 |
ies | +0.089 |
Neg logit contributions | |
Here | -0.082 |
V | -0.081 |
The | -0.079 |
Higher | -0.078 |
Ro | -0.076 |
Token | . |
Token ID | 13 |
Feature activation | -5.69e-02 |
Pos logit contributions | |
ishing | +0.213 |
ies | +0.206 |
rying | +0.206 |
owing | +0.199 |
inches | +0.195 |
Neg logit contributions | |
The | -0.197 |
Here | -0.183 |
It | -0.182 |
One | -0.170 |
Max | -0.168 |
Token | The |
Token ID | 383 |
Feature activation | +2.44e-03 |
Pos logit contributions | |
Grab | +0.082 |
Here | +0.080 |
Higher | +0.079 |
V | +0.078 |
Quick | +0.071 |
Neg logit contributions | |
than | -0.135 |
ever | -0.116 |
- | -0.107 |
inches | -0.105 |
ier | -0.105 |
Token | best |
Token ID | 1266 |
Feature activation | +6.27e-02 |
Pos logit contributions | |
than | +0.005 |
ier | +0.004 |
- | +0.004 |
ies | +0.004 |
owing | +0.004 |
Neg logit contributions | |
Here | -0.005 |
V | -0.005 |
Higher | -0.005 |
Grab | -0.005 |
Ro | -0.004 |
Token | ms |
Token ID | 907 |
Feature activation | +9.37e-02 |
Pos logit contributions | |
ies | +0.174 |
than | +0.154 |
- | +0.147 |
ge | +0.145 |
inches | +0.145 |
Neg logit contributions | |
Here | -0.203 |
V | -0.202 |
Grab | -0.190 |
Higher | -0.180 |
Ro | -0.179 |
Token | or |
Token ID | 393 |
Feature activation | +3.00e-02 |
Pos logit contributions | |
than | +0.130 |
ies | +0.112 |
- | +0.111 |
inches | +0.111 |
ge | +0.104 |
Neg logit contributions | |
Here | -0.202 |
The | -0.202 |
V | -0.202 |
Higher | -0.196 |
Look | -0.191 |
Token | mess |
Token ID | 2085 |
Feature activation | +9.83e-02 |
Pos logit contributions | |
than | +0.062 |
ier | +0.058 |
- | +0.057 |
ies | +0.056 |
owing | +0.052 |
Neg logit contributions | |
V | -0.055 |
Here | -0.055 |
Grab | -0.053 |
Higher | -0.052 |
First | -0.050 |
Token | up |
Token ID | 510 |
Feature activation | +7.42e-02 |
Pos logit contributions | |
than | +0.147 |
owing | +0.126 |
ies | +0.122 |
inches | +0.120 |
- | +0.117 |
Neg logit contributions | |
V | -0.225 |
Here | -0.214 |
Look | -0.201 |
Max | -0.198 |
Ro | -0.198 |
Token | your |
Token ID | 534 |
Feature activation | +5.71e-02 |
Pos logit contributions | |
than | +0.108 |
ies | +0.108 |
ier | +0.106 |
owing | +0.106 |
- | +0.101 |
Neg logit contributions | |
Here | -0.150 |
The | -0.147 |
V | -0.146 |
Higher | -0.139 |
Ro | -0.137 |
Token | art |
Token ID | 1242 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
ier | +0.115 |
than | +0.114 |
ies | +0.103 |
ge | +0.103 |
owing | +0.100 |
Neg logit contributions | |
Here | -0.097 |
V | -0.096 |
The | -0.089 |
Higher | -0.088 |
Grab | -0.088 |
Token | ." |
Token ID | 526 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
than | +0.221 |
owing | +0.215 |
ge | +0.212 |
ies | +0.206 |
inches | +0.192 |
Neg logit contributions | |
Here | -0.261 |
V | -0.259 |
First | -0.242 |
Everyone | -0.240 |
The | -0.237 |
Token |
|
Token ID | 198 |
Feature activation | -9.29e-02 |
Pos logit contributions | |
V | +0.177 |
Grab | +0.166 |
Higher | +0.166 |
Here | +0.161 |
Ro | +0.148 |
Neg logit contributions | |
than | -0.252 |
- | -0.237 |
ies | -0.226 |
ever | -0.216 |
! | -0.208 |
Token |
|
Token ID | 198 |
Feature activation | -9.03e-02 |
Pos logit contributions | |
Grab | +0.183 |
Higher | +0.179 |
Here | +0.169 |
First | +0.168 |
Pull | +0.158 |
Neg logit contributions | |
- | -0.175 |
than | -0.173 |
! | -0.150 |
ies | -0.144 |
ever | -0.138 |
Token | L |
Token ID | 43 |
Feature activation | -2.03e-03 |
Pos logit contributions | |
Higher | +0.148 |
First | +0.131 |
Here | +0.126 |
Grab | +0.124 |
V | +0.121 |
Neg logit contributions | |
than | -0.212 |
- | -0.197 |
ies | -0.175 |
ever | -0.171 |
ge | -0.170 |
Token | ily |
Token ID | 813 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
V | +0.002 |
Here | +0.002 |
Grab | +0.002 |
Higher | +0.002 |
Many | +0.002 |
Neg logit contributions | |
than | -0.006 |
ies | -0.005 |
- | -0.005 |
ier | -0.005 |
inches | -0.005 |
Token | even |
Token ID | 772 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
than | +0.003 |
ies | +0.003 |
ier | +0.003 |
- | +0.003 |
owing | +0.003 |
Neg logit contributions | |
V | -0.003 |
Here | -0.003 |
Higher | -0.003 |
First | -0.003 |
Grab | -0.003 |
Token | more |
Token ID | 517 |
Feature activation | +5.16e-02 |
Pos logit contributions | |
than | +0.045 |
ier | +0.043 |
- | +0.043 |
ies | +0.042 |
owing | +0.042 |
Neg logit contributions | |
Here | -0.050 |
V | -0.050 |
Higher | -0.047 |
Grab | -0.046 |
First | -0.046 |
Token | fun |
Token ID | 1257 |
Feature activation | +6.43e-02 |
Pos logit contributions | |
owing | +0.100 |
ier | +0.097 |
rying | +0.087 |
ies | +0.087 |
- | +0.085 |
Neg logit contributions | |
Here | -0.099 |
Many | -0.086 |
The | -0.084 |
Higher | -0.082 |
One | -0.082 |
Token | sorting |
Token ID | 29407 |
Feature activation | +7.59e-02 |
Pos logit contributions | |
owing | +0.107 |
ier | +0.102 |
ust | +0.091 |
adow | +0.087 |
rying | +0.085 |
Neg logit contributions | |
Let | -0.103 |
Here | -0.101 |
Everyone | -0.099 |
The | -0.097 |
Higher | -0.097 |
Token | the |
Token ID | 262 |
Feature activation | +3.06e-02 |
Pos logit contributions | |
owing | +0.148 |
ier | +0.139 |
than | +0.135 |
ies | +0.135 |
ge | +0.134 |
Neg logit contributions | |
The | -0.130 |
Here | -0.128 |
Higher | -0.122 |
First | -0.120 |
Everyone | -0.118 |
Token | cand |
Token ID | 2658 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
ier | +0.060 |
than | +0.058 |
owing | +0.055 |
- | +0.054 |
ies | +0.054 |
Neg logit contributions | |
Here | -0.061 |
V | -0.054 |
Grab | -0.053 |
Look | -0.053 |
Hi | -0.052 |
Token | ies |
Token ID | 444 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
owing | +0.150 |
than | +0.136 |
supposed | +0.135 |
believing | +0.132 |
knowing | +0.127 |
Neg logit contributions | |
First | -0.248 |
Everyone | -0.243 |
They | -0.243 |
Here | -0.241 |
The | -0.240 |
Token | again |
Token ID | 757 |
Feature activation | +5.46e-02 |
Pos logit contributions | |
owing | +0.149 |
ies | +0.149 |
ier | +0.144 |
- | +0.143 |
than | +0.140 |
Neg logit contributions | |
The | -0.178 |
It | -0.166 |
Higher | -0.164 |
First | -0.158 |
Let | -0.157 |
Token | , |
Token ID | 11 |
Feature activation | +5.01e-03 |
Pos logit contributions | |
ier | +0.087 |
owing | +0.076 |
ies | +0.072 |
ines | +0.070 |
iness | +0.068 |
Neg logit contributions | |
The | -0.112 |
Sam | -0.104 |
Max | -0.103 |
Max | -0.099 |
First | -0.099 |
Token | and |
Token ID | 290 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
ier | +0.008 |
- | +0.008 |
ies | +0.008 |
owing | +0.007 |
than | +0.007 |
Neg logit contributions | |
Higher | -0.011 |
Here | -0.011 |
The | -0.011 |
V | -0.010 |
Look | -0.010 |
Token | they |
Token ID | 484 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
Here | +0.026 |
V | +0.025 |
Higher | +0.024 |
The | +0.023 |
First | +0.023 |
Neg logit contributions | |
than | -0.058 |
ier | -0.055 |
- | -0.054 |
owing | -0.053 |
ies | -0.052 |
Token | it |
Token ID | 340 |
Feature activation | +2.99e-02 |
Pos logit contributions | |
Here | +0.002 |
V | +0.002 |
Higher | +0.002 |
Grab | +0.002 |
Ro | +0.002 |
Neg logit contributions | |
than | -0.003 |
ier | -0.002 |
ies | -0.002 |
- | -0.002 |
owing | -0.002 |
Token | 's |
Token ID | 338 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
than | +0.064 |
ies | +0.057 |
ier | +0.057 |
- | +0.056 |
ge | +0.053 |
Neg logit contributions | |
V | -0.055 |
Here | -0.054 |
Higher | -0.052 |
Grab | -0.052 |
First | -0.049 |
Token | not |
Token ID | 407 |
Feature activation | +2.07e-02 |
Pos logit contributions | |
than | +0.030 |
ies | +0.029 |
ier | +0.028 |
- | +0.028 |
owing | +0.027 |
Neg logit contributions | |
V | -0.032 |
Here | -0.031 |
Grab | -0.029 |
First | -0.029 |
Higher | -0.029 |
Token | about |
Token ID | 546 |
Feature activation | +3.54e-02 |
Pos logit contributions | |
ines | +0.040 |
ier | +0.038 |
than | +0.038 |
ies | +0.037 |
- | +0.037 |
Neg logit contributions | |
The | -0.033 |
V | -0.033 |
Here | -0.033 |
Grab | -0.032 |
First | -0.032 |
Token | the |
Token ID | 262 |
Feature activation | +4.47e-02 |
Pos logit contributions | |
ier | +0.072 |
than | +0.071 |
ies | +0.069 |
- | +0.067 |
owing | +0.065 |
Neg logit contributions | |
Here | -0.061 |
Higher | -0.060 |
The | -0.059 |
V | -0.058 |
Look | -0.056 |
Token | gold |
Token ID | 3869 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
than | +0.088 |
ies | +0.086 |
ier | +0.083 |
- | +0.082 |
owing | +0.079 |
Neg logit contributions | |
Here | -0.085 |
V | -0.079 |
Higher | -0.076 |
Look | -0.075 |
Hi | -0.074 |
Token | coin |
Token ID | 10752 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
supposed | +0.221 |
ge | +0.203 |
owing | +0.203 |
than | +0.195 |
rying | +0.190 |
Neg logit contributions | |
Here | -0.223 |
It | -0.218 |
The | -0.216 |
Hi | -0.208 |
A | -0.199 |
Token | , |
Token ID | 11 |
Feature activation | -8.49e-03 |
Pos logit contributions | |
than | +0.164 |
ies | +0.163 |
ier | +0.150 |
supposed | +0.149 |
ge | +0.148 |
Neg logit contributions | |
The | -0.208 |
Here | -0.196 |
Max | -0.195 |
It | -0.191 |
Spot | -0.184 |
Token | but |
Token ID | 475 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
Here | +0.019 |
Higher | +0.018 |
V | +0.018 |
Grab | +0.018 |
The | +0.017 |
Neg logit contributions | |
ier | -0.014 |
than | -0.014 |
ies | -0.013 |
- | -0.013 |
ge | -0.012 |
Token | how |
Token ID | 703 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
Here | +0.023 |
Higher | +0.021 |
V | +0.021 |
First | +0.021 |
Grab | +0.021 |
Neg logit contributions | |
than | -0.040 |
ier | -0.039 |
- | -0.038 |
ies | -0.037 |
owing | -0.036 |
Token | he |
Token ID | 339 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
V | +0.040 |
Here | +0.040 |
Higher | +0.039 |
Grab | +0.037 |
First | +0.036 |
Neg logit contributions | |
than | -0.039 |
ies | -0.035 |
ier | -0.034 |
- | -0.034 |
inches | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | -3.87e-02 |
Pos logit contributions | |
than | +0.145 |
- | +0.140 |
ge | +0.140 |
yt | +0.136 |
ies | +0.134 |
Neg logit contributions | |
The | -0.174 |
Here | -0.165 |
Hi | -0.162 |
Ro | -0.161 |
V | -0.158 |
Token | They |
Token ID | 1119 |
Feature activation | +5.67e-03 |
Pos logit contributions | |
Here | +0.063 |
V | +0.061 |
Grab | +0.061 |
Higher | +0.059 |
Quick | +0.055 |
Neg logit contributions | |
than | -0.087 |
too | -0.070 |
ever | -0.069 |
because | -0.068 |
indeed | -0.066 |
Token | are |
Token ID | 389 |
Feature activation | +3.24e-02 |
Pos logit contributions | |
than | +0.014 |
ier | +0.013 |
- | +0.013 |
ge | +0.013 |
ies | +0.013 |
Neg logit contributions | |
Here | -0.008 |
V | -0.008 |
Grab | -0.008 |
Higher | -0.007 |
The | -0.007 |
Token | bad |
Token ID | 2089 |
Feature activation | +5.34e-02 |
Pos logit contributions | |
ier | +0.064 |
ge | +0.063 |
than | +0.063 |
- | +0.062 |
owing | +0.062 |
Neg logit contributions | |
V | -0.056 |
Here | -0.055 |
The | -0.053 |
Ro | -0.050 |
Look | -0.050 |
Token | ducks |
Token ID | 39694 |
Feature activation | +7.76e-02 |
Pos logit contributions | |
ge | +0.085 |
than | +0.078 |
ier | +0.073 |
owing | +0.073 |
- | +0.072 |
Neg logit contributions | |
V | -0.108 |
Here | -0.105 |
Many | -0.104 |
Ro | -0.104 |
First | -0.104 |
Token | ." |
Token ID | 526 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
ge | +0.141 |
ier | +0.118 |
ishing | +0.117 |
yt | +0.117 |
inches | +0.117 |
Neg logit contributions | |
The | -0.136 |
Spot | -0.122 |
It | -0.120 |
Here | -0.119 |
They | -0.118 |
Token | Mom |
Token ID | 11254 |
Feature activation | +2.95e-05 |
Pos logit contributions | |
V | +0.235 |
Grab | +0.230 |
Ro | +0.225 |
Bang | +0.215 |
Here | +0.213 |
Neg logit contributions | |
than | -0.281 |
ies | -0.254 |
- | -0.251 |
for | -0.236 |
and | -0.233 |
Token | my |
Token ID | 1820 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
than | +0.000 |
ier | +0.000 |
- | +0.000 |
ies | +0.000 |
inches | +0.000 |
Neg logit contributions | |
V | -0.000 |
Here | -0.000 |
Higher | -0.000 |
Grab | -0.000 |
First | -0.000 |
Token | says |
Token ID | 1139 |
Feature activation | -3.62e-02 |
Pos logit contributions | |
Here | +0.027 |
V | +0.027 |
Higher | +0.027 |
Grab | +0.025 |
First | +0.025 |
Neg logit contributions | |
than | -0.036 |
- | -0.034 |
ies | -0.034 |
ier | -0.032 |
inches | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | -5.06e-02 |
Pos logit contributions | |
V | +0.088 |
Here | +0.086 |
Higher | +0.086 |
Grab | +0.083 |
First | +0.081 |
Neg logit contributions | |
than | -0.056 |
ies | -0.047 |
ier | -0.046 |
- | -0.045 |
ge | -0.043 |
Token | " |
Token ID | 366 |
Feature activation | -3.92e-02 |
Pos logit contributions | |
V | +0.074 |
Here | +0.071 |
First | +0.068 |
Higher | +0.068 |
Grab | +0.066 |
Neg logit contributions | |
than | -0.127 |
- | -0.118 |
ier | -0.113 |
ies | -0.111 |
ever | -0.108 |
Token | . |
Token ID | 13 |
Feature activation | -4.88e-02 |
Pos logit contributions | |
than | +0.168 |
- | +0.157 |
ies | +0.157 |
ishing | +0.153 |
owing | +0.150 |
Neg logit contributions | |
The | -0.173 |
V | -0.171 |
First | -0.158 |
One | -0.156 |
Here | -0.156 |
Token | The |
Token ID | 383 |
Feature activation | +1.27e-02 |
Pos logit contributions | |
Ol | +0.075 |
V | +0.075 |
Here | +0.074 |
Grab | +0.074 |
Bang | +0.071 |
Neg logit contributions | |
than | -0.112 |
too | -0.090 |
for | -0.088 |
- | -0.087 |
ever | -0.087 |
Token | dolls |
Token ID | 36062 |
Feature activation | +6.35e-02 |
Pos logit contributions | |
than | +0.027 |
ies | +0.025 |
- | +0.025 |
ier | +0.025 |
inches | +0.022 |
Neg logit contributions | |
V | -0.024 |
Here | -0.024 |
Higher | -0.023 |
Grab | -0.022 |
First | -0.021 |
Token | are |
Token ID | 389 |
Feature activation | +3.52e-02 |
Pos logit contributions | |
than | +0.158 |
ier | +0.156 |
ies | +0.153 |
- | +0.150 |
ge | +0.148 |
Neg logit contributions | |
Here | -0.067 |
V | -0.065 |
The | -0.062 |
Grab | -0.061 |
Many | -0.058 |
Token | cool |
Token ID | 3608 |
Feature activation | +8.86e-02 |
Pos logit contributions | |
ier | +0.069 |
owing | +0.067 |
than | +0.066 |
- | +0.065 |
ge | +0.065 |
Neg logit contributions | |
V | -0.063 |
Here | -0.057 |
The | -0.057 |
Max | -0.053 |
Many | -0.053 |
Token | ." |
Token ID | 526 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
ge | +0.149 |
owing | +0.141 |
than | +0.141 |
inches | +0.139 |
ier | +0.137 |
Neg logit contributions | |
The | -0.141 |
First | -0.134 |
V | -0.134 |
It | -0.126 |
One | -0.126 |
Token |
|
Token ID | 198 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
V | +0.242 |
Grab | +0.231 |
Ro | +0.231 |
Bang | +0.223 |
Here | +0.220 |
Neg logit contributions | |
than | -0.269 |
- | -0.222 |
and | -0.215 |
too | -0.214 |
ies | -0.211 |
Token |
|
Token ID | 198 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
Grab | +0.226 |
Here | +0.221 |
Higher | +0.219 |
First | +0.210 |
Ro | +0.206 |
Neg logit contributions | |
- | -0.183 |
than | -0.180 |
! | -0.159 |
and | -0.154 |
ies | -0.153 |
Token | Mom |
Token ID | 29252 |
Feature activation | -7.05e-03 |
Pos logit contributions | |
Higher | +0.141 |
Here | +0.126 |
V | +0.125 |
Many | +0.120 |
Curious | +0.118 |
Neg logit contributions | |
than | -0.274 |
- | -0.254 |
ies | -0.248 |
ever | -0.238 |
inches | -0.234 |
Token | says |
Token ID | 1139 |
Feature activation | -3.94e-02 |
Pos logit contributions | |
V | +0.011 |
Here | +0.011 |
Higher | +0.011 |
Grab | +0.011 |
First | +0.010 |
Neg logit contributions | |
than | -0.017 |
- | -0.016 |
ies | -0.015 |
ier | -0.015 |
inches | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | -6.02e-02 |
Pos logit contributions | |
V | +0.106 |
Here | +0.105 |
Higher | +0.102 |
Grab | +0.101 |
First | +0.098 |
Neg logit contributions | |
than | -0.052 |
- | -0.044 |
ies | -0.044 |
ier | -0.043 |
inches | -0.037 |
Token | some |
Token ID | 617 |
Feature activation | +4.17e-02 |
Pos logit contributions | |
V | +0.008 |
Here | +0.008 |
Higher | +0.008 |
Grab | +0.008 |
First | +0.008 |
Neg logit contributions | |
than | -0.005 |
ier | -0.004 |
ies | -0.004 |
- | -0.004 |
ge | -0.004 |
Token | water |
Token ID | 1660 |
Feature activation | +8.23e-02 |
Pos logit contributions | |
than | +0.090 |
ge | +0.083 |
ier | +0.083 |
ies | +0.081 |
- | +0.078 |
Neg logit contributions | |
Here | -0.075 |
V | -0.070 |
Grab | -0.070 |
First | -0.069 |
Higher | -0.068 |
Token | and |
Token ID | 290 |
Feature activation | +4.56e-03 |
Pos logit contributions | |
than | +0.121 |
ge | +0.119 |
ier | +0.111 |
inches | +0.103 |
supposed | +0.103 |
Neg logit contributions | |
First | -0.173 |
Here | -0.171 |
V | -0.169 |
Ro | -0.166 |
The | -0.166 |
Token | a |
Token ID | 257 |
Feature activation | +8.10e-03 |
Pos logit contributions | |
than | +0.009 |
ies | +0.008 |
- | +0.008 |
ier | +0.008 |
inches | +0.008 |
Neg logit contributions | |
V | -0.009 |
Here | -0.009 |
Higher | -0.008 |
Grab | -0.008 |
Look | -0.008 |
Token | nap |
Token ID | 25422 |
Feature activation | +6.22e-02 |
Pos logit contributions | |
than | +0.018 |
ier | +0.018 |
ies | +0.016 |
- | +0.016 |
ge | +0.016 |
Neg logit contributions | |
Here | -0.014 |
V | -0.014 |
Grab | -0.013 |
Ro | -0.013 |
Look | -0.013 |
Token | ." |
Token ID | 526 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
than | +0.106 |
ge | +0.104 |
owing | +0.096 |
ies | +0.095 |
ier | +0.095 |
Neg logit contributions | |
V | -0.118 |
First | -0.113 |
Here | -0.111 |
Ro | -0.107 |
The | -0.107 |
Token |
|
Token ID | 198 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
V | +0.255 |
Here | +0.253 |
Ro | +0.251 |
Grab | +0.244 |
Bang | +0.234 |
Neg logit contributions | |
- | -0.244 |
than | -0.242 |
ies | -0.232 |
and | -0.215 |
for | -0.213 |
Token |
|
Token ID | 198 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
Grab | +0.242 |
Here | +0.237 |
Higher | +0.228 |
Ro | +0.217 |
Pull | +0.215 |
Neg logit contributions | |
- | -0.206 |
than | -0.181 |
ies | -0.170 |
! | -0.159 |
. | -0.156 |
Token | " |
Token ID | 1 |
Feature activation | -6.89e-02 |
Pos logit contributions | |
Higher | +0.248 |
Here | +0.240 |
V | +0.225 |
Ro | +0.217 |
Grab | +0.217 |
Neg logit contributions | |
- | -0.212 |
than | -0.195 |
ies | -0.187 |
y | -0.151 |
ever | -0.147 |
Token | Not |
Token ID | 3673 |
Feature activation | +2.73e-02 |
Pos logit contributions | |
Higher | +0.093 |
V | +0.090 |
The | +0.086 |
Many | +0.083 |
First | +0.081 |
Neg logit contributions | |
than | -0.174 |
- | -0.149 |
ever | -0.148 |
ier | -0.142 |
ies | -0.140 |
Token | yet |
Token ID | 1865 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
ier | +0.045 |
than | +0.041 |
- | +0.040 |
owing | +0.039 |
ies | +0.038 |
Neg logit contributions | |
The | -0.055 |
V | -0.054 |
First | -0.054 |
Hi | -0.053 |
Here | -0.052 |
Token | Just |
Token ID | 2329 |
Feature activation | +3.61e-03 |
Pos logit contributions | |
V | +0.056 |
Here | +0.055 |
Higher | +0.053 |
Grab | +0.052 |
Look | +0.051 |
Neg logit contributions | |
than | -0.099 |
- | -0.087 |
ever | -0.083 |
too | -0.080 |
for | -0.079 |
Token | one |
Token ID | 530 |
Feature activation | +4.03e-02 |
Pos logit contributions | |
than | +0.008 |
ier | +0.007 |
ies | +0.007 |
- | +0.007 |
ge | +0.006 |
Neg logit contributions | |
V | -0.007 |
Here | -0.007 |
Higher | -0.006 |
Grab | -0.006 |
First | -0.006 |
Token | more |
Token ID | 517 |
Feature activation | +5.77e-02 |
Pos logit contributions | |
ier | +0.075 |
than | +0.070 |
ge | +0.067 |
owing | +0.066 |
ies | +0.064 |
Neg logit contributions | |
V | -0.077 |
Here | -0.075 |
Ro | -0.073 |
The | -0.071 |
Many | -0.071 |
Token | . |
Token ID | 13 |
Feature activation | -4.22e-02 |
Pos logit contributions | |
ier | +0.113 |
owing | +0.106 |
than | +0.104 |
ge | +0.101 |
- | +0.095 |
Neg logit contributions | |
V | -0.104 |
Here | -0.102 |
Many | -0.097 |
First | -0.096 |
Ro | -0.096 |
Token | Please |
Token ID | 4222 |
Feature activation | -2.46e-02 |
Pos logit contributions | |
Here | +0.057 |
V | +0.057 |
Higher | +0.054 |
Grab | +0.052 |
Look | +0.051 |
Neg logit contributions | |
than | -0.101 |
- | -0.088 |
ever | -0.086 |
inches | -0.084 |
too | -0.084 |
Token | ." |
Token ID | 526 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
Here | +0.064 |
V | +0.064 |
Higher | +0.062 |
Grab | +0.061 |
First | +0.059 |
Neg logit contributions | |
than | -0.035 |
- | -0.031 |
ies | -0.030 |
ier | -0.029 |
inches | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | -1.00e-01 |
Pos logit contributions | |
V | +0.233 |
Here | +0.213 |
Grab | +0.211 |
Ro | +0.198 |
Higher | +0.194 |
Neg logit contributions | |
than | -0.261 |
ever | -0.235 |
too | -0.235 |
- | -0.231 |
! | -0.225 |
Token |
|
Token ID | 198 |
Feature activation | -9.76e-02 |
Pos logit contributions | |
Higher | +0.219 |
Grab | +0.216 |
Here | +0.209 |
First | +0.204 |
V | +0.194 |
Neg logit contributions | |
- | -0.157 |
than | -0.140 |
. | -0.129 |
! | -0.127 |
ever | -0.120 |
Token | M |
Token ID | 44 |
Feature activation | -3.52e-02 |
Pos logit contributions | |
Higher | +0.181 |
Here | +0.163 |
Many | +0.157 |
V | +0.156 |
First | +0.154 |
Neg logit contributions | |
than | -0.209 |
- | -0.198 |
ies | -0.171 |
ever | -0.168 |
inches | -0.164 |
Token | ia |
Token ID | 544 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
V | +0.069 |
Here | +0.067 |
Higher | +0.064 |
Grab | +0.064 |
First | +0.062 |
Neg logit contributions | |
than | -0.073 |
ies | -0.063 |
ier | -0.063 |
- | -0.063 |
ge | -0.058 |
Token | shakes |
Token ID | 29959 |
Feature activation | -9.55e-03 |
Pos logit contributions | |
V | +0.033 |
Here | +0.032 |
Higher | +0.031 |
Grab | +0.030 |
First | +0.029 |
Neg logit contributions | |
than | -0.040 |
ies | -0.036 |
- | -0.036 |
ier | -0.036 |
inches | -0.033 |
Token | are |
Token ID | 389 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
than | +0.030 |
ier | +0.029 |
- | +0.028 |
ies | +0.028 |
ge | +0.026 |
Neg logit contributions | |
Here | -0.021 |
V | -0.020 |
Grab | -0.020 |
Higher | -0.019 |
First | -0.019 |
Token | home |
Token ID | 1363 |
Feature activation | +8.01e-02 |
Pos logit contributions | |
ier | +0.073 |
owing | +0.068 |
- | +0.068 |
ies | +0.067 |
ge | +0.067 |
Neg logit contributions | |
The | -0.065 |
V | -0.063 |
Here | -0.063 |
Max | -0.060 |
It | -0.060 |
Token | to |
Token ID | 284 |
Feature activation | -8.67e-03 |
Pos logit contributions | |
than | +0.115 |
ies | +0.109 |
ier | +0.109 |
ishing | +0.101 |
ge | +0.100 |
Neg logit contributions | |
The | -0.166 |
Here | -0.156 |
V | -0.155 |
First | -0.155 |
Higher | -0.153 |
Token | many |
Token ID | 867 |
Feature activation | +6.49e-02 |
Pos logit contributions | |
V | +0.017 |
Here | +0.017 |
Higher | +0.016 |
Grab | +0.016 |
First | +0.015 |
Neg logit contributions | |
than | -0.018 |
- | -0.016 |
ies | -0.016 |
ier | -0.016 |
inches | -0.015 |
Token | animals |
Token ID | 4695 |
Feature activation | +7.49e-02 |
Pos logit contributions | |
ier | +0.122 |
owing | +0.120 |
than | +0.110 |
ge | +0.108 |
ies | +0.107 |
Neg logit contributions | |
Here | -0.113 |
Many | -0.112 |
V | -0.110 |
Ro | -0.109 |
People | -0.102 |
Token | .'" |
Token ID | 11496 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
ies | +0.122 |
ier | +0.121 |
than | +0.120 |
- | +0.114 |
ge | +0.109 |
Neg logit contributions | |
Here | -0.144 |
V | -0.142 |
The | -0.140 |
First | -0.139 |
Higher | -0.138 |
Token |
|
Token ID | 198 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
V | +0.244 |
Grab | +0.218 |
Here | +0.212 |
Ro | +0.211 |
Higher | +0.204 |
Neg logit contributions | |
than | -0.260 |
- | -0.215 |
ever | -0.207 |
and | -0.199 |
. | -0.196 |
Token |
|
Token ID | 198 |
Feature activation | -9.95e-02 |
Pos logit contributions | |
Grab | +0.213 |
Here | +0.206 |
Higher | +0.192 |
V | +0.190 |
Hi | +0.186 |
Neg logit contributions | |
than | -0.206 |
- | -0.200 |
ies | -0.184 |
ines | -0.171 |
! | -0.171 |
Token | Ben |
Token ID | 11696 |
Feature activation | -5.67e-03 |
Pos logit contributions | |
Higher | +0.158 |
Here | +0.156 |
V | +0.147 |
Grab | +0.142 |
Many | +0.135 |
Neg logit contributions | |
than | -0.243 |
- | -0.214 |
ies | -0.211 |
inches | -0.205 |
ge | -0.191 |
Token | and |
Token ID | 290 |
Feature activation | -4.67e-02 |
Pos logit contributions | |
V | +0.017 |
Here | +0.016 |
Higher | +0.016 |
Grab | +0.016 |
First | +0.015 |
Neg logit contributions | |
than | -0.006 |
ies | -0.005 |
- | -0.005 |
ier | -0.005 |
inches | -0.004 |
Token | Mia |
Token ID | 32189 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
V | +0.069 |
Here | +0.067 |
Higher | +0.064 |
Grab | +0.063 |
First | +0.060 |
Neg logit contributions | |
than | -0.119 |
ier | -0.109 |
- | -0.109 |
ies | -0.109 |
inches | -0.101 |
Token | It |
Token ID | 632 |
Feature activation | -3.05e-03 |
Pos logit contributions | |
Here | +0.131 |
V | +0.127 |
Ol | +0.122 |
Higher | +0.121 |
Bang | +0.119 |
Neg logit contributions | |
than | -0.138 |
too | -0.129 |
- | -0.120 |
for | -0.115 |
because | -0.115 |
Token | is |
Token ID | 318 |
Feature activation | +1.93e-02 |
Pos logit contributions | |
V | +0.005 |
Here | +0.005 |
Grab | +0.005 |
Higher | +0.005 |
Ro | +0.005 |
Neg logit contributions | |
than | -0.007 |
ier | -0.006 |
ies | -0.006 |
- | -0.006 |
ge | -0.006 |
Token | sad |
Token ID | 6507 |
Feature activation | +4.14e-02 |
Pos logit contributions | |
ier | +0.035 |
ies | +0.034 |
than | +0.034 |
ge | +0.033 |
- | +0.032 |
Neg logit contributions | |
V | -0.037 |
Here | -0.037 |
The | -0.036 |
Grab | -0.035 |
Look | -0.033 |
Token | and |
Token ID | 290 |
Feature activation | -4.59e-03 |
Pos logit contributions | |
ge | +0.068 |
ier | +0.067 |
owing | +0.066 |
inches | +0.064 |
than | +0.064 |
Neg logit contributions | |
The | -0.080 |
First | -0.077 |
V | -0.074 |
One | -0.074 |
It | -0.073 |
Token | troubled |
Token ID | 17840 |
Feature activation | +4.58e-02 |
Pos logit contributions | |
V | +0.010 |
Here | +0.010 |
Higher | +0.010 |
Grab | +0.009 |
First | +0.009 |
Neg logit contributions | |
than | -0.008 |
ies | -0.008 |
- | -0.007 |
ier | -0.007 |
inches | -0.007 |
Token | ." |
Token ID | 526 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
than | +0.080 |
ge | +0.079 |
ies | +0.078 |
- | +0.076 |
ier | +0.076 |
Neg logit contributions | |
V | -0.087 |
The | -0.086 |
Higher | -0.084 |
First | -0.083 |
Here | -0.081 |
Token |
|
Token ID | 198 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
Grab | +0.204 |
Ro | +0.203 |
V | +0.199 |
Bang | +0.193 |
Here | +0.190 |
Neg logit contributions | |
- | -0.279 |
than | -0.278 |
ies | -0.250 |
anymore | -0.245 |
ever | -0.242 |
Token |
|
Token ID | 198 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
Grab | +0.235 |
Ro | +0.223 |
Higher | +0.222 |
Here | +0.218 |
Pull | +0.212 |
Neg logit contributions | |
- | -0.224 |
than | -0.206 |
ever | -0.187 |
ies | -0.180 |
and | -0.172 |
Token | Mom |
Token ID | 29252 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
Higher | +0.157 |
V | +0.142 |
Here | +0.141 |
Grab | +0.139 |
Ro | +0.136 |
Neg logit contributions | |
than | -0.279 |
- | -0.279 |
ies | -0.250 |
ever | -0.247 |
inches | -0.227 |
Token | looked |
Token ID | 3114 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
Here | +0.023 |
Higher | +0.023 |
V | +0.023 |
Grab | +0.022 |
Ro | +0.021 |
Neg logit contributions | |
- | -0.023 |
than | -0.022 |
ies | -0.022 |
ier | -0.021 |
ever | -0.019 |
Token | at |
Token ID | 379 |
Feature activation | -3.95e-02 |
Pos logit contributions | |
Here | +0.038 |
V | +0.037 |
Higher | +0.037 |
Grab | +0.036 |
First | +0.035 |
Neg logit contributions | |
than | -0.018 |
ier | -0.015 |
ies | -0.014 |
- | -0.014 |
inches | -0.013 |
Token | are |
Token ID | 389 |
Feature activation | +2.60e-02 |
Pos logit contributions | |
than | +0.017 |
ier | +0.016 |
- | +0.016 |
ies | +0.015 |
ge | +0.015 |
Neg logit contributions | |
Here | -0.012 |
V | -0.011 |
Grab | -0.011 |
First | -0.011 |
Higher | -0.011 |
Token | so |
Token ID | 523 |
Feature activation | -7.57e-03 |
Pos logit contributions | |
ier | +0.050 |
than | +0.048 |
ge | +0.047 |
- | +0.047 |
owing | +0.047 |
Neg logit contributions | |
Here | -0.048 |
V | -0.047 |
The | -0.046 |
First | -0.045 |
Grab | -0.044 |
Token | bright |
Token ID | 6016 |
Feature activation | +7.14e-02 |
Pos logit contributions | |
V | +0.018 |
Here | +0.017 |
Higher | +0.017 |
Grab | +0.017 |
First | +0.016 |
Neg logit contributions | |
than | -0.013 |
ies | -0.012 |
- | -0.012 |
ier | -0.011 |
inches | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
ge | +0.115 |
owing | +0.110 |
ies | +0.099 |
ier | +0.098 |
than | +0.096 |
Neg logit contributions | |
The | -0.140 |
V | -0.137 |
First | -0.135 |
It | -0.135 |
Here | -0.134 |
Token | shiny |
Token ID | 22441 |
Feature activation | +7.04e-02 |
Pos logit contributions | |
than | +0.031 |
ier | +0.029 |
- | +0.029 |
ies | +0.028 |
ge | +0.026 |
Neg logit contributions | |
Here | -0.031 |
V | -0.031 |
Higher | -0.029 |
Grab | -0.029 |
First | -0.028 |
Token | ." |
Token ID | 526 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
ge | +0.135 |
owing | +0.130 |
supposed | +0.120 |
ies | +0.115 |
than | +0.114 |
Neg logit contributions | |
The | -0.117 |
Here | -0.115 |
First | -0.114 |
It | -0.113 |
V | -0.113 |
Token | Ben |
Token ID | 3932 |
Feature activation | -2.88e-02 |
Pos logit contributions | |
V | +0.189 |
Here | +0.180 |
Ro | +0.173 |
Grab | +0.164 |
Higher | +0.162 |
Neg logit contributions | |
than | -0.319 |
ies | -0.303 |
- | -0.299 |
anymore | -0.272 |
! | -0.272 |
Token | said |
Token ID | 531 |
Feature activation | -4.25e-02 |
Pos logit contributions | |
Here | +0.057 |
V | +0.057 |
Higher | +0.056 |
Grab | +0.052 |
Ro | +0.050 |
Neg logit contributions | |
than | -0.060 |
- | -0.056 |
ies | -0.055 |
ier | -0.051 |
ever | -0.048 |
Token | , |
Token ID | 11 |
Feature activation | -5.09e-02 |
Pos logit contributions | |
V | +0.109 |
Here | +0.108 |
Higher | +0.106 |
Grab | +0.104 |
First | +0.101 |
Neg logit contributions | |
than | -0.060 |
ies | -0.051 |
ier | -0.051 |
- | -0.050 |
inches | -0.044 |
Token | " |
Token ID | 366 |
Feature activation | -4.05e-02 |
Pos logit contributions | |
Here | +0.075 |
V | +0.074 |
Grab | +0.068 |
Higher | +0.066 |
Ro | +0.064 |
Neg logit contributions | |
than | -0.131 |
- | -0.123 |
ier | -0.117 |
ies | -0.116 |
ever | -0.111 |
Token | No |
Token ID | 2949 |
Feature activation | -1.84e-03 |
Pos logit contributions | |
V | +0.051 |
Higher | +0.051 |
Here | +0.047 |
Grab | +0.046 |
First | +0.045 |
Neg logit contributions | |
than | -0.112 |
- | -0.099 |
ies | -0.094 |
ier | -0.092 |
ever | -0.092 |
Token | . |
Token ID | 13 |
Feature activation | -4.21e-02 |
Pos logit contributions | |
supposed | +0.173 |
than | +0.171 |
ge | +0.155 |
ier | +0.153 |
knowing | +0.144 |
Neg logit contributions | |
The | -0.184 |
First | -0.173 |
It | -0.170 |
V | -0.170 |
One | -0.170 |
Token | It |
Token ID | 632 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
Here | +0.053 |
V | +0.050 |
Grab | +0.048 |
Ol | +0.046 |
Look | +0.044 |
Neg logit contributions | |
than | -0.110 |
too | -0.091 |
- | -0.091 |
inches | -0.089 |
because | -0.088 |
Token | looks |
Token ID | 3073 |
Feature activation | +3.51e-02 |
Pos logit contributions | |
than | +0.031 |
ier | +0.030 |
ge | +0.028 |
ies | +0.028 |
- | +0.027 |
Neg logit contributions | |
V | -0.021 |
Here | -0.020 |
Ro | -0.019 |
Grab | -0.019 |
Higher | -0.019 |
Token | so |
Token ID | 523 |
Feature activation | +9.75e-03 |
Pos logit contributions | |
than | +0.059 |
ier | +0.059 |
ies | +0.059 |
ge | +0.056 |
- | +0.056 |
Neg logit contributions | |
Here | -0.074 |
V | -0.073 |
Higher | -0.070 |
Grab | -0.068 |
The | -0.068 |
Token | pretty |
Token ID | 2495 |
Feature activation | +7.34e-02 |
Pos logit contributions | |
than | +0.018 |
ies | +0.016 |
- | +0.016 |
ier | +0.016 |
inches | +0.014 |
Neg logit contributions | |
V | -0.021 |
Here | -0.021 |
Higher | -0.020 |
Grab | -0.020 |
First | -0.019 |
Token | ." |
Token ID | 526 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
ge | +0.126 |
owing | +0.122 |
ies | +0.121 |
ier | +0.113 |
than | +0.112 |
Neg logit contributions | |
Here | -0.133 |
The | -0.132 |
First | -0.129 |
It | -0.125 |
Higher | -0.124 |
Token | Lily |
Token ID | 20037 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
V | +0.217 |
Here | +0.213 |
Ro | +0.211 |
Grab | +0.208 |
Higher | +0.202 |
Neg logit contributions | |
than | -0.261 |
ies | -0.252 |
- | -0.249 |
! | -0.230 |
for | -0.229 |
Token | says |
Token ID | 1139 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
Here | +0.039 |
V | +0.039 |
Higher | +0.038 |
Grab | +0.036 |
Ro | +0.035 |
Neg logit contributions | |
than | -0.053 |
- | -0.048 |
ies | -0.048 |
ier | -0.046 |
inches | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
V | +0.071 |
Higher | +0.070 |
The | +0.069 |
Here | +0.068 |
First | +0.067 |
Neg logit contributions | |
than | -0.041 |
ies | -0.034 |
ier | -0.034 |
inches | -0.032 |
ge | -0.030 |
Token |
|
Token ID | 198 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
Grab | +0.214 |
Ro | +0.214 |
Here | +0.213 |
V | +0.211 |
Bang | +0.204 |
Neg logit contributions | |
than | -0.208 |
- | -0.189 |
and | -0.177 |
too | -0.174 |
for | -0.168 |
Token |
|
Token ID | 198 |
Feature activation | -9.97e-02 |
Pos logit contributions | |
Grab | +0.212 |
Here | +0.209 |
Higher | +0.197 |
Ro | +0.196 |
V | +0.188 |
Neg logit contributions | |
than | -0.202 |
- | -0.202 |
ies | -0.179 |
! | -0.174 |
and | -0.166 |
Token | hurt |
Token ID | 5938 |
Feature activation | +7.15e-02 |
Pos logit contributions | |
than | +0.072 |
ier | +0.065 |
ies | +0.063 |
- | +0.061 |
ge | +0.060 |
Neg logit contributions | |
Here | -0.067 |
V | -0.064 |
Grab | -0.061 |
The | -0.060 |
Look | -0.059 |
Token | or |
Token ID | 393 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
ies | +0.102 |
ge | +0.101 |
than | +0.099 |
ier | +0.097 |
owing | +0.096 |
Neg logit contributions | |
The | -0.147 |
V | -0.147 |
Many | -0.141 |
Ro | -0.136 |
Hi | -0.135 |
Token | make |
Token ID | 787 |
Feature activation | +2.56e-02 |
Pos logit contributions | |
than | +0.049 |
ier | +0.046 |
- | +0.045 |
ies | +0.045 |
inches | +0.040 |
Neg logit contributions | |
Here | -0.048 |
V | -0.048 |
Higher | -0.045 |
Grab | -0.045 |
Ro | -0.043 |
Token | a |
Token ID | 257 |
Feature activation | +3.47e-02 |
Pos logit contributions | |
than | +0.058 |
ier | +0.052 |
ies | +0.051 |
- | +0.050 |
ge | +0.049 |
Neg logit contributions | |
V | -0.045 |
Here | -0.044 |
Higher | -0.041 |
Grab | -0.040 |
First | -0.040 |
Token | mess |
Token ID | 2085 |
Feature activation | +9.40e-02 |
Pos logit contributions | |
ier | +0.063 |
than | +0.063 |
owing | +0.058 |
ge | +0.054 |
ies | +0.052 |
Neg logit contributions | |
Here | -0.071 |
V | -0.071 |
Look | -0.070 |
Ro | -0.069 |
Many | -0.062 |
Token | ." |
Token ID | 526 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
owing | +0.149 |
ge | +0.144 |
than | +0.143 |
ies | +0.140 |
inches | +0.138 |
Neg logit contributions | |
V | -0.170 |
The | -0.160 |
It | -0.150 |
Max | -0.148 |
Ro | -0.145 |
Token | Mom |
Token ID | 11254 |
Feature activation | -5.97e-03 |
Pos logit contributions | |
V | +0.224 |
Ro | +0.216 |
Grab | +0.216 |
Here | +0.204 |
Bang | +0.200 |
Neg logit contributions | |
than | -0.277 |
- | -0.250 |
ies | -0.240 |
and | -0.224 |
for | -0.223 |
Token | said |
Token ID | 531 |
Feature activation | -3.22e-02 |
Pos logit contributions | |
V | +0.010 |
Here | +0.010 |
Higher | +0.010 |
Grab | +0.010 |
First | +0.009 |
Neg logit contributions | |
than | -0.013 |
- | -0.012 |
ies | -0.012 |
ier | -0.012 |
inches | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
V | +0.075 |
Higher | +0.075 |
Here | +0.074 |
Grab | +0.071 |
First | +0.070 |
Neg logit contributions | |
than | -0.051 |
ies | -0.044 |
ier | -0.042 |
- | -0.041 |
ge | -0.040 |
Token | " |
Token ID | 366 |
Feature activation | -3.83e-02 |
Pos logit contributions | |
V | +0.061 |
Here | +0.059 |
Higher | +0.055 |
First | +0.054 |
Grab | +0.054 |
Neg logit contributions | |
than | -0.108 |
- | -0.101 |
ier | -0.096 |
ies | -0.094 |
ever | -0.092 |
Token | That |
Token ID | 2504 |
Feature activation | +4.69e-03 |
Pos logit contributions | |
Higher | +0.047 |
V | +0.046 |
Here | +0.043 |
First | +0.043 |
Grab | +0.041 |
Neg logit contributions | |
than | -0.107 |
- | -0.094 |
ies | -0.092 |
ier | -0.091 |
ever | -0.088 |
Token | is |
Token ID | 318 |
Feature activation | +2.68e-02 |
Pos logit contributions | |
than | +0.027 |
ier | +0.025 |
ies | +0.024 |
- | +0.024 |
ge | +0.023 |
Neg logit contributions | |
V | -0.023 |
Here | -0.022 |
Grab | -0.021 |
Higher | -0.021 |
Ro | -0.020 |
Token | so |
Token ID | 523 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
ier | +0.051 |
than | +0.050 |
ies | +0.050 |
owing | +0.048 |
- | +0.048 |
Neg logit contributions | |
V | -0.052 |
Here | -0.050 |
The | -0.046 |
Look | -0.046 |
Higher | -0.046 |
Token | nice |
Token ID | 3621 |
Feature activation | +7.91e-02 |
Pos logit contributions | |
than | +0.024 |
ier | +0.023 |
ies | +0.022 |
- | +0.022 |
ge | +0.020 |
Neg logit contributions | |
V | -0.025 |
Here | -0.025 |
Higher | -0.023 |
Grab | -0.023 |
Ro | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
ge | +0.148 |
owing | +0.135 |
ier | +0.128 |
inches | +0.126 |
ies | +0.125 |
Neg logit contributions | |
First | -0.145 |
The | -0.140 |
It | -0.137 |
V | -0.132 |
Ro | -0.128 |
Token | warm |
Token ID | 5814 |
Feature activation | +8.39e-02 |
Pos logit contributions | |
than | +0.045 |
ier | +0.043 |
ies | +0.042 |
- | +0.041 |
ge | +0.039 |
Neg logit contributions | |
V | -0.045 |
Here | -0.044 |
Higher | -0.042 |
Grab | -0.041 |
First | -0.040 |
Token | ." |
Token ID | 526 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
ge | +0.157 |
inches | +0.140 |
ies | +0.139 |
owing | +0.138 |
than | +0.137 |
Neg logit contributions | |
The | -0.150 |
Here | -0.149 |
V | -0.149 |
First | -0.148 |
Hi | -0.146 |
Token |
|
Token ID | 198 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
Ro | +0.229 |
V | +0.228 |
Grab | +0.227 |
Bang | +0.216 |
Here | +0.213 |
Neg logit contributions | |
than | -0.262 |
- | -0.234 |
ies | -0.226 |
anymore | -0.219 |
and | -0.217 |
Token |
|
Token ID | 198 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
Grab | +0.219 |
Ro | +0.213 |
Higher | +0.209 |
Here | +0.207 |
V | +0.192 |
Neg logit contributions | |
- | -0.204 |
than | -0.200 |
! | -0.173 |
ies | -0.171 |
. | -0.166 |
Token | Tom |
Token ID | 13787 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
Higher | +0.138 |
Here | +0.123 |
V | +0.120 |
Ro | +0.117 |
Many | +0.115 |
Neg logit contributions | |
than | -0.288 |
- | -0.272 |
ies | -0.257 |
ever | -0.239 |
inches | -0.235 |
Token | likes |
Token ID | 7832 |
Feature activation | -3.00e-03 |
Pos logit contributions | |
Here | +0.037 |
V | +0.037 |
Higher | +0.036 |
Grab | +0.035 |
Ro | +0.033 |
Neg logit contributions | |
than | -0.037 |
- | -0.034 |
ies | -0.033 |
ier | -0.032 |
inches | -0.030 |
Token | a |
Token ID | 257 |
Feature activation | +3.75e-03 |
Pos logit contributions | |
V | +0.005 |
Here | +0.005 |
Higher | +0.005 |
Grab | +0.005 |
Ro | +0.004 |
Neg logit contributions | |
than | -0.007 |
ier | -0.006 |
ies | -0.006 |
- | -0.006 |
owing | -0.006 |
Token | It |
Token ID | 632 |
Feature activation | +5.35e-03 |
Pos logit contributions | |
V | +0.076 |
Higher | +0.074 |
Here | +0.068 |
Bang | +0.065 |
First | +0.065 |
Neg logit contributions | |
than | -0.095 |
- | -0.087 |
ever | -0.085 |
because | -0.074 |
too | -0.073 |
Token | was |
Token ID | 373 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
than | +0.012 |
ier | +0.011 |
ies | +0.011 |
- | +0.010 |
ge | +0.010 |
Neg logit contributions | |
V | -0.010 |
Here | -0.010 |
Grab | -0.009 |
Ro | -0.009 |
Higher | -0.009 |
Token | a |
Token ID | 257 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
than | +0.052 |
ies | +0.052 |
ier | +0.051 |
ge | +0.050 |
owing | +0.048 |
Neg logit contributions | |
V | -0.041 |
Here | -0.041 |
Grab | -0.038 |
The | -0.038 |
Look | -0.038 |
Token | bad |
Token ID | 2089 |
Feature activation | +5.05e-02 |
Pos logit contributions | |
owing | +0.055 |
than | +0.054 |
ier | +0.054 |
ge | +0.052 |
ies | +0.052 |
Neg logit contributions | |
Here | -0.042 |
Ro | -0.042 |
Look | -0.041 |
V | -0.041 |
Bang | -0.037 |
Token | insect |
Token ID | 12833 |
Feature activation | +7.88e-02 |
Pos logit contributions | |
ge | +0.108 |
than | +0.104 |
owing | +0.101 |
ier | +0.099 |
inches | +0.099 |
Neg logit contributions | |
Ro | -0.075 |
V | -0.073 |
Hi | -0.071 |
Here | -0.071 |
Look | -0.071 |
Token | ." |
Token ID | 526 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
ge | +0.116 |
than | +0.116 |
ies | +0.113 |
ier | +0.109 |
owing | +0.105 |
Neg logit contributions | |
Ro | -0.165 |
V | -0.161 |
Here | -0.156 |
The | -0.150 |
Give | -0.143 |
Token | Her |
Token ID | 2332 |
Feature activation | -3.76e-02 |
Pos logit contributions | |
V | +0.222 |
Grab | +0.213 |
Ro | +0.206 |
Here | +0.198 |
Higher | +0.193 |
Neg logit contributions | |
than | -0.284 |
- | -0.268 |
ies | -0.245 |
anymore | -0.238 |
ever | -0.237 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
V | +0.057 |
Here | +0.056 |
Higher | +0.053 |
Grab | +0.052 |
First | +0.049 |
Neg logit contributions | |
than | -0.094 |
- | -0.087 |
ies | -0.086 |
ier | -0.086 |
inches | -0.080 |
Token | hugged |
Token ID | 41130 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
V | +0.023 |
Here | +0.023 |
Higher | +0.023 |
Grab | +0.021 |
First | +0.021 |
Neg logit contributions | |
than | -0.029 |
- | -0.028 |
ies | -0.027 |
ier | -0.027 |
inches | -0.024 |
Token | her |
Token ID | 607 |
Feature activation | +7.56e-03 |
Pos logit contributions | |
V | +0.034 |
Here | +0.033 |
Higher | +0.032 |
Grab | +0.032 |
First | +0.031 |
Neg logit contributions | |
than | -0.033 |
- | -0.030 |
ies | -0.030 |
ier | -0.030 |
inches | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | -4.11e-02 |
Pos logit contributions | |
than | +0.011 |
ier | +0.010 |
supposed | +0.010 |
ines | +0.009 |
ies | +0.009 |
Neg logit contributions | |
The | -0.017 |
V | -0.016 |
Higher | -0.016 |
Here | -0.015 |
Grab | -0.015 |
Token | Please |
Token ID | 4222 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
Here | +0.080 |
V | +0.076 |
Ol | +0.075 |
Grab | +0.074 |
Higher | +0.073 |
Neg logit contributions | |
than | -0.112 |
- | -0.087 |
too | -0.086 |
for | -0.083 |
ever | -0.082 |
Token | , |
Token ID | 11 |
Feature activation | -5.26e-02 |
Pos logit contributions | |
Here | +0.050 |
V | +0.049 |
Higher | +0.048 |
Grab | +0.047 |
First | +0.045 |
Neg logit contributions | |
than | -0.028 |
- | -0.025 |
ies | -0.025 |
ier | -0.024 |
inches | -0.021 |
Token | Mom |
Token ID | 11254 |
Feature activation | +3.18e-02 |
Pos logit contributions | |
Here | +0.088 |
V | +0.087 |
Higher | +0.084 |
Grab | +0.080 |
Look | +0.076 |
Neg logit contributions | |
than | -0.119 |
ies | -0.112 |
- | -0.112 |
ever | -0.105 |
ier | -0.102 |
Token | , |
Token ID | 11 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
than | +0.055 |
ier | +0.050 |
ies | +0.048 |
- | +0.048 |
inches | +0.046 |
Neg logit contributions | |
V | -0.069 |
Here | -0.066 |
Higher | -0.064 |
First | -0.064 |
Grab | -0.063 |
Token | please |
Token ID | 3387 |
Feature activation | +4.44e-03 |
Pos logit contributions | |
V | +0.020 |
Here | +0.019 |
Higher | +0.018 |
Grab | +0.018 |
First | +0.017 |
Neg logit contributions | |
than | -0.025 |
ies | -0.023 |
- | -0.023 |
ier | -0.023 |
inches | -0.021 |
Token | !" |
Token ID | 2474 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
than | +0.007 |
ies | +0.006 |
- | +0.006 |
ier | +0.006 |
inches | +0.006 |
Neg logit contributions | |
V | -0.010 |
Here | -0.010 |
Higher | -0.010 |
Grab | -0.010 |
First | -0.010 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
V | +0.240 |
Higher | +0.233 |
Here | +0.229 |
Grab | +0.224 |
Ro | +0.224 |
Neg logit contributions | |
than | -0.233 |
too | -0.225 |
- | -0.224 |
ies | -0.222 |
and | -0.212 |
Token | begs |
Token ID | 38609 |
Feature activation | -4.21e-02 |
Pos logit contributions | |
Here | +0.036 |
V | +0.035 |
Higher | +0.034 |
Grab | +0.033 |
Ro | +0.032 |
Neg logit contributions | |
than | -0.038 |
- | -0.034 |
ies | -0.034 |
ier | -0.032 |
inches | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
V | +0.123 |
Here | +0.121 |
Higher | +0.119 |
Grab | +0.117 |
First | +0.114 |
Neg logit contributions | |
than | -0.046 |
ies | -0.037 |
- | -0.036 |
ier | -0.036 |
inches | -0.030 |
Token |
|
Token ID | 198 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
Grab | +0.227 |
Here | +0.220 |
V | +0.219 |
Ro | +0.216 |
Higher | +0.212 |
Neg logit contributions | |
than | -0.203 |
- | -0.185 |
too | -0.170 |
and | -0.166 |
. | -0.165 |
Token |
|
Token ID | 198 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
Grab | +0.231 |
Here | +0.227 |
Ro | +0.221 |
Higher | +0.212 |
V | +0.211 |
Neg logit contributions | |
- | -0.211 |
than | -0.198 |
! | -0.177 |
ies | -0.175 |
. | -0.171 |
Token | play |
Token ID | 711 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
V | +0.010 |
Here | +0.010 |
Higher | +0.010 |
Grab | +0.009 |
First | +0.009 |
Neg logit contributions | |
than | -0.016 |
ier | -0.015 |
ies | -0.015 |
- | -0.015 |
inches | -0.014 |
Token | outside |
Token ID | 2354 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
than | +0.021 |
ies | +0.018 |
- | +0.018 |
ier | +0.018 |
inches | +0.015 |
Neg logit contributions | |
V | -0.043 |
Here | -0.043 |
Higher | -0.042 |
Grab | -0.041 |
First | -0.040 |
Token | with |
Token ID | 351 |
Feature activation | +1.49e-04 |
Pos logit contributions | |
ies | +0.036 |
ier | +0.036 |
than | +0.034 |
ge | +0.030 |
- | +0.030 |
Neg logit contributions | |
The | -0.055 |
V | -0.054 |
Higher | -0.053 |
Here | -0.052 |
First | -0.051 |
Token | his |
Token ID | 465 |
Feature activation | -4.67e-03 |
Pos logit contributions | |
than | +0.000 |
- | +0.000 |
ies | +0.000 |
ier | +0.000 |
inches | +0.000 |
Neg logit contributions | |
V | -0.000 |
Here | -0.000 |
Higher | -0.000 |
Grab | -0.000 |
First | -0.000 |
Token | friends |
Token ID | 2460 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
Here | +0.011 |
V | +0.011 |
Higher | +0.010 |
Grab | +0.010 |
First | +0.010 |
Neg logit contributions | |
than | -0.008 |
ier | -0.007 |
ies | -0.007 |
- | -0.007 |
ge | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
than | +0.022 |
ies | +0.020 |
ier | +0.019 |
- | +0.019 |
inches | +0.017 |
Neg logit contributions | |
V | -0.044 |
Here | -0.043 |
Higher | -0.042 |
First | -0.042 |
The | -0.041 |
Token | He |
Token ID | 679 |
Feature activation | -3.99e-02 |
Pos logit contributions | |
Ro | +0.214 |
Grab | +0.207 |
V | +0.205 |
Here | +0.191 |
Cat | +0.183 |
Neg logit contributions | |
than | -0.283 |
! | -0.249 |
anymore | -0.248 |
too | -0.240 |
and | -0.238 |
Token | asked |
Token ID | 1965 |
Feature activation | -2.82e-02 |
Pos logit contributions | |
Here | +0.080 |
Higher | +0.079 |
V | +0.076 |
Grab | +0.074 |
First | +0.070 |
Neg logit contributions | |
than | -0.078 |
- | -0.078 |
ies | -0.074 |
ever | -0.068 |
ier | -0.065 |
Token | his |
Token ID | 465 |
Feature activation | -2.45e-02 |
Pos logit contributions | |
Here | +0.062 |
V | +0.062 |
Grab | +0.060 |
Higher | +0.058 |
Ro | +0.056 |
Neg logit contributions | |
than | -0.052 |
- | -0.050 |
ies | -0.048 |
ier | -0.046 |
ever | -0.041 |
Token | mum |
Token ID | 25682 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
V | +0.062 |
Here | +0.061 |
Higher | +0.060 |
Grab | +0.059 |
First | +0.057 |
Neg logit contributions | |
than | -0.037 |
- | -0.032 |
ies | -0.032 |
ier | -0.031 |
inches | -0.027 |
Token | if |
Token ID | 611 |
Feature activation | -4.46e-02 |
Pos logit contributions | |
than | +0.017 |
ier | +0.015 |
- | +0.014 |
ies | +0.014 |
inches | +0.013 |
Neg logit contributions | |
V | -0.024 |
Here | -0.024 |
Higher | -0.024 |
First | -0.023 |
Grab | -0.022 |
Token | is |
Token ID | 318 |
Feature activation | +3.59e-02 |
Pos logit contributions | |
than | +0.058 |
ier | +0.054 |
ies | +0.054 |
ge | +0.052 |
owing | +0.051 |
Neg logit contributions | |
V | -0.038 |
Ro | -0.035 |
Here | -0.034 |
Grab | -0.034 |
Bang | -0.033 |
Token | very |
Token ID | 845 |
Feature activation | +3.05e-02 |
Pos logit contributions | |
ier | +0.069 |
owing | +0.068 |
ies | +0.068 |
than | +0.065 |
- | +0.064 |
Neg logit contributions | |
V | -0.067 |
The | -0.060 |
Here | -0.060 |
Bang | -0.059 |
Ro | -0.057 |
Token | high |
Token ID | 1029 |
Feature activation | +8.54e-02 |
Pos logit contributions | |
than | +0.059 |
ier | +0.057 |
ies | +0.055 |
owing | +0.053 |
- | +0.052 |
Neg logit contributions | |
V | -0.059 |
Here | -0.057 |
Higher | -0.053 |
Grab | -0.053 |
Ro | -0.053 |
Token | and |
Token ID | 290 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
ies | +0.130 |
owing | +0.128 |
ge | +0.127 |
inches | +0.117 |
than | +0.116 |
Neg logit contributions | |
V | -0.163 |
The | -0.160 |
First | -0.153 |
Higher | -0.147 |
Here | -0.146 |
Token | fast |
Token ID | 3049 |
Feature activation | +7.21e-02 |
Pos logit contributions | |
than | +0.053 |
ier | +0.052 |
ies | +0.052 |
- | +0.050 |
owing | +0.047 |
Neg logit contributions | |
V | -0.051 |
Here | -0.050 |
Higher | -0.048 |
Grab | -0.047 |
Ro | -0.045 |
Token | ." |
Token ID | 526 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
ies | +0.123 |
owing | +0.119 |
ier | +0.115 |
ge | +0.114 |
- | +0.111 |
Neg logit contributions | |
V | -0.135 |
The | -0.131 |
First | -0.122 |
It | -0.120 |
Here | -0.119 |
Token |
|
Token ID | 198 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
V | +0.210 |
Here | +0.196 |
Grab | +0.193 |
Ro | +0.184 |
Bang | +0.179 |
Neg logit contributions | |
than | -0.291 |
- | -0.259 |
ies | -0.258 |
! | -0.248 |
anymore | -0.242 |
Token |
|
Token ID | 198 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
Grab | +0.230 |
Here | +0.223 |
Higher | +0.212 |
Hi | +0.205 |
V | +0.202 |
Neg logit contributions | |
than | -0.204 |
- | -0.198 |
ies | -0.177 |
! | -0.175 |
ines | -0.169 |
Token | Ben |
Token ID | 11696 |
Feature activation | -1.14e-02 |
Pos logit contributions | |
Higher | +0.137 |
Here | +0.126 |
V | +0.121 |
Many | +0.117 |
First | +0.115 |
Neg logit contributions | |
than | -0.286 |
- | -0.254 |
ies | -0.253 |
inches | -0.240 |
ever | -0.226 |
Token | and |
Token ID | 290 |
Feature activation | -4.92e-02 |
Pos logit contributions | |
V | +0.030 |
Here | +0.030 |
Higher | +0.029 |
Grab | +0.028 |
First | +0.027 |
Neg logit contributions | |
than | -0.016 |
- | -0.014 |
ies | -0.013 |
ier | -0.013 |
inches | -0.011 |
Token | Mia |
Token ID | 32189 |
Feature activation | -2.07e-02 |
Pos logit contributions | |
V | +0.075 |
Here | +0.073 |
Higher | +0.070 |
Grab | +0.068 |
First | +0.064 |
Neg logit contributions | |
than | -0.123 |
- | -0.113 |
ies | -0.113 |
ier | -0.112 |
inches | -0.104 |
Token | Please |
Token ID | 4222 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
Here | +0.089 |
Higher | +0.078 |
Grab | +0.078 |
V | +0.077 |
Ol | +0.075 |
Neg logit contributions | |
than | -0.117 |
too | -0.097 |
ever | -0.096 |
- | -0.092 |
for | -0.090 |
Token | , |
Token ID | 11 |
Feature activation | -4.50e-02 |
Pos logit contributions | |
Here | +0.043 |
V | +0.042 |
Higher | +0.041 |
Grab | +0.040 |
First | +0.039 |
Neg logit contributions | |
than | -0.035 |
- | -0.031 |
ies | -0.031 |
ier | -0.030 |
inches | -0.027 |
Token | Mom |
Token ID | 11254 |
Feature activation | +3.32e-02 |
Pos logit contributions | |
V | +0.072 |
Here | +0.072 |
Higher | +0.068 |
Grab | +0.067 |
Ro | +0.063 |
Neg logit contributions | |
than | -0.106 |
- | -0.099 |
ies | -0.098 |
ier | -0.093 |
ever | -0.091 |
Token | , |
Token ID | 11 |
Feature activation | -1.19e-03 |
Pos logit contributions | |
than | +0.055 |
ier | +0.049 |
ies | +0.048 |
- | +0.047 |
inches | +0.045 |
Neg logit contributions | |
V | -0.076 |
Here | -0.072 |
Higher | -0.070 |
First | -0.069 |
Grab | -0.068 |
Token | please |
Token ID | 3387 |
Feature activation | +6.77e-03 |
Pos logit contributions | |
V | +0.002 |
Here | +0.002 |
Higher | +0.002 |
Grab | +0.002 |
First | +0.002 |
Neg logit contributions | |
than | -0.003 |
ies | -0.003 |
ier | -0.003 |
- | -0.003 |
ge | -0.002 |
Token | ." |
Token ID | 526 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
than | +0.011 |
- | +0.009 |
ies | +0.009 |
ier | +0.009 |
inches | +0.008 |
Neg logit contributions | |
V | -0.017 |
Here | -0.017 |
Higher | -0.016 |
Grab | -0.016 |
First | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
V | +0.223 |
Ro | +0.217 |
Grab | +0.211 |
Here | +0.205 |
Higher | +0.203 |
Neg logit contributions | |
than | -0.271 |
- | -0.259 |
ies | -0.245 |
too | -0.226 |
! | -0.226 |
Token |
|
Token ID | 198 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
Grab | +0.252 |
Higher | +0.238 |
Ro | +0.237 |
Here | +0.237 |
Pull | +0.231 |
Neg logit contributions | |
- | -0.210 |
than | -0.198 |
! | -0.181 |
ies | -0.174 |
ever | -0.174 |
Token | Mom |
Token ID | 29252 |
Feature activation | -7.06e-03 |
Pos logit contributions | |
Higher | +0.149 |
V | +0.133 |
Here | +0.128 |
Ro | +0.125 |
Many | +0.122 |
Neg logit contributions | |
than | -0.290 |
- | -0.282 |
ies | -0.271 |
ever | -0.249 |
inches | -0.243 |
Token | smiled |
Token ID | 13541 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
Here | +0.013 |
V | +0.013 |
Higher | +0.013 |
Grab | +0.012 |
Ro | +0.012 |
Neg logit contributions | |
than | -0.015 |
- | -0.014 |
ies | -0.014 |
ier | -0.014 |
ever | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -8.89e-02 |
Pos logit contributions | |
ge | +0.027 |
than | +0.026 |
supposed | +0.025 |
inches | +0.025 |
ier | +0.025 |
Neg logit contributions | |
The | -0.041 |
V | -0.040 |
Hi | -0.039 |
Here | -0.038 |
It | -0.038 |
Token | think |
Token ID | 892 |
Feature activation | -1.50e-02 |
Pos logit contributions | |
V | +0.072 |
Here | +0.072 |
Grab | +0.067 |
Higher | +0.065 |
Look | +0.063 |
Neg logit contributions | |
than | -0.088 |
ier | -0.082 |
ies | -0.081 |
- | -0.079 |
inches | -0.076 |
Token | they |
Token ID | 484 |
Feature activation | +3.94e-03 |
Pos logit contributions | |
V | +0.023 |
Here | +0.022 |
Higher | +0.021 |
Grab | +0.021 |
First | +0.019 |
Neg logit contributions | |
than | -0.038 |
- | -0.035 |
ies | -0.035 |
ier | -0.034 |
inches | -0.032 |
Token | are |
Token ID | 389 |
Feature activation | +4.21e-02 |
Pos logit contributions | |
than | +0.009 |
ier | +0.008 |
- | +0.008 |
ies | +0.008 |
ge | +0.008 |
Neg logit contributions | |
Here | -0.006 |
V | -0.006 |
Grab | -0.006 |
Higher | -0.006 |
First | -0.006 |
Token | more |
Token ID | 517 |
Feature activation | +5.04e-02 |
Pos logit contributions | |
than | +0.082 |
ge | +0.081 |
ier | +0.080 |
ies | +0.080 |
- | +0.079 |
Neg logit contributions | |
Here | -0.073 |
V | -0.072 |
The | -0.068 |
Grab | -0.065 |
Higher | -0.064 |
Token | fun |
Token ID | 1257 |
Feature activation | +7.54e-02 |
Pos logit contributions | |
ies | +0.096 |
ge | +0.095 |
owing | +0.093 |
ier | +0.091 |
ie | +0.087 |
Neg logit contributions | |
Here | -0.088 |
V | -0.086 |
First | -0.079 |
Many | -0.076 |
The | -0.073 |
Token | ." |
Token ID | 526 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
ge | +0.130 |
owing | +0.123 |
ier | +0.118 |
ies | +0.118 |
ust | +0.112 |
Neg logit contributions | |
The | -0.123 |
V | -0.122 |
First | -0.119 |
Let | -0.119 |
Here | -0.117 |
Token | Mom |
Token ID | 11254 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
V | +0.219 |
Grab | +0.212 |
Ro | +0.209 |
Here | +0.202 |
Bang | +0.194 |
Neg logit contributions | |
than | -0.282 |
- | -0.253 |
ies | -0.248 |
too | -0.235 |
ever | -0.232 |
Token | says |
Token ID | 1139 |
Feature activation | -3.79e-02 |
Pos logit contributions | |
V | +0.018 |
Here | +0.017 |
Higher | +0.017 |
Grab | +0.016 |
First | +0.016 |
Neg logit contributions | |
than | -0.026 |
- | -0.024 |
ies | -0.024 |
ier | -0.023 |
inches | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | -4.76e-02 |
Pos logit contributions | |
V | +0.088 |
Higher | +0.087 |
Here | +0.086 |
Grab | +0.083 |
The | +0.083 |
Neg logit contributions | |
than | -0.062 |
ies | -0.052 |
ier | -0.051 |
- | -0.049 |
ge | -0.048 |
Token | " |
Token ID | 366 |
Feature activation | -4.61e-02 |
Pos logit contributions | |
V | +0.066 |
Here | +0.063 |
Higher | +0.060 |
First | +0.060 |
Grab | +0.058 |
Neg logit contributions | |
than | -0.125 |
- | -0.115 |
ier | -0.112 |
ies | -0.109 |
ever | -0.106 |
Token | No |
Token ID | 2949 |
Feature activation | -8.44e-03 |
Pos logit contributions | |
Higher | +0.056 |
V | +0.055 |
First | +0.053 |
Here | +0.050 |
Grab | +0.050 |
Neg logit contributions | |
than | -0.130 |
- | -0.116 |
ies | -0.113 |
ier | -0.112 |
ever | -0.106 |
Token | ? |
Token ID | 30 |
Feature activation | -4.74e-02 |
Pos logit contributions | |
than | +0.143 |
ier | +0.129 |
ies | +0.128 |
- | +0.124 |
inches | +0.122 |
Neg logit contributions | |
V | -0.100 |
Here | -0.099 |
Higher | -0.097 |
Grab | -0.095 |
First | -0.093 |
Token | Maybe |
Token ID | 6674 |
Feature activation | -5.19e-03 |
Pos logit contributions | |
Here | +0.071 |
V | +0.069 |
Look | +0.065 |
Higher | +0.064 |
Grab | +0.064 |
Neg logit contributions | |
than | -0.099 |
too | -0.093 |
- | -0.091 |
for | -0.086 |
ever | -0.085 |
Token | it |
Token ID | 340 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
V | +0.010 |
Here | +0.010 |
Higher | +0.010 |
Grab | +0.010 |
First | +0.009 |
Neg logit contributions | |
than | -0.011 |
- | -0.010 |
ies | -0.010 |
ier | -0.010 |
ever | -0.009 |
Token | 's |
Token ID | 338 |
Feature activation | +3.89e-02 |
Pos logit contributions | |
than | +0.045 |
ier | +0.041 |
ies | +0.039 |
- | +0.038 |
ge | +0.038 |
Neg logit contributions | |
V | -0.033 |
Here | -0.032 |
Grab | -0.031 |
Higher | -0.030 |
Ro | -0.029 |
Token | mine |
Token ID | 6164 |
Feature activation | +7.36e-02 |
Pos logit contributions | |
ier | +0.073 |
ies | +0.070 |
than | +0.070 |
owing | +0.068 |
- | +0.068 |
Neg logit contributions | |
V | -0.073 |
Here | -0.069 |
Look | -0.068 |
The | -0.066 |
Bang | -0.066 |
Token | ." |
Token ID | 526 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
owing | +0.115 |
than | +0.114 |
- | +0.113 |
ge | +0.113 |
ies | +0.113 |
Neg logit contributions | |
V | -0.125 |
First | -0.123 |
The | -0.120 |
Higher | -0.119 |
Hi | -0.114 |
Token | The |
Token ID | 383 |
Feature activation | -3.96e-02 |
Pos logit contributions | |
V | +0.178 |
Grab | +0.157 |
Ro | +0.155 |
Here | +0.152 |
Higher | +0.151 |
Neg logit contributions | |
- | -0.328 |
than | -0.312 |
ies | -0.293 |
for | -0.285 |
and | -0.278 |
Token | boy |
Token ID | 2933 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
Higher | +0.091 |
Here | +0.091 |
V | +0.090 |
First | +0.090 |
Grab | +0.088 |
Neg logit contributions | |
- | -0.058 |
than | -0.057 |
ies | -0.054 |
ier | -0.050 |
ever | -0.045 |
Token | said |
Token ID | 531 |
Feature activation | -3.24e-02 |
Pos logit contributions | |
V | +0.023 |
Here | +0.023 |
Higher | +0.022 |
Grab | +0.022 |
First | +0.021 |
Neg logit contributions | |
than | -0.030 |
- | -0.028 |
ies | -0.027 |
ier | -0.027 |
inches | -0.025 |
Token | yes |
Token ID | 3763 |
Feature activation | +3.43e-02 |
Pos logit contributions | |
V | +0.086 |
Here | +0.086 |
Higher | +0.083 |
Grab | +0.082 |
First | +0.080 |
Neg logit contributions | |
than | -0.044 |
- | -0.038 |
ies | -0.037 |
ier | -0.037 |
inches | -0.031 |
Token | and |
Token ID | 290 |
Feature activation | -4.45e-02 |
Pos logit contributions | |
inches | +0.060 |
ier | +0.057 |
than | +0.057 |
ies | +0.055 |
ishing | +0.055 |
Neg logit contributions | |
The | -0.064 |
It | -0.060 |
Max | -0.057 |
One | -0.057 |
His | -0.056 |
Token | outside |
Token ID | 2354 |
Feature activation | +2.77e-02 |
Pos logit contributions | |
than | +0.048 |
ier | +0.042 |
ies | +0.041 |
- | +0.039 |
inches | +0.035 |
Neg logit contributions | |
V | -0.102 |
Here | -0.099 |
Higher | -0.097 |
Grab | -0.096 |
Ro | -0.093 |
Token | . |
Token ID | 13 |
Feature activation | -5.14e-02 |
Pos logit contributions | |
than | +0.050 |
ier | +0.045 |
ies | +0.044 |
ge | +0.043 |
- | +0.042 |
Neg logit contributions | |
V | -0.057 |
Here | -0.056 |
Higher | -0.054 |
The | -0.054 |
Grab | -0.053 |
Token | I |
Token ID | 314 |
Feature activation | -2.06e-02 |
Pos logit contributions | |
Here | +0.092 |
V | +0.086 |
Look | +0.079 |
Bang | +0.079 |
Higher | +0.077 |
Neg logit contributions | |
than | -0.092 |
too | -0.086 |
because | -0.086 |
- | -0.084 |
for | -0.079 |
Token | am |
Token ID | 716 |
Feature activation | +3.81e-03 |
Pos logit contributions | |
V | +0.037 |
Here | +0.036 |
Grab | +0.035 |
Ro | +0.033 |
Higher | +0.033 |
Neg logit contributions | |
than | -0.044 |
ier | -0.043 |
ies | -0.040 |
inches | -0.039 |
- | -0.038 |
Token | scared |
Token ID | 12008 |
Feature activation | +2.74e-02 |
Pos logit contributions | |
than | +0.008 |
ier | +0.007 |
ies | +0.007 |
- | +0.007 |
ge | +0.006 |
Neg logit contributions | |
V | -0.007 |
Here | -0.007 |
The | -0.007 |
Higher | -0.007 |
Grab | -0.007 |
Token | ." |
Token ID | 526 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
than | +0.039 |
ier | +0.036 |
ies | +0.035 |
ge | +0.035 |
- | +0.033 |
Neg logit contributions | |
V | -0.065 |
Higher | -0.062 |
The | -0.061 |
Many | -0.061 |
Here | -0.060 |
Token | Her |
Token ID | 2332 |
Feature activation | -3.50e-02 |
Pos logit contributions | |
Grab | +0.211 |
V | +0.211 |
Here | +0.204 |
Ro | +0.202 |
Bang | +0.193 |
Neg logit contributions | |
- | -0.279 |
than | -0.266 |
ies | -0.259 |
for | -0.242 |
and | -0.241 |
Token | mom |
Token ID | 1995 |
Feature activation | -6.83e-03 |
Pos logit contributions | |
Here | +0.063 |
V | +0.061 |
Higher | +0.060 |
Grab | +0.059 |
First | +0.055 |
Neg logit contributions | |
than | -0.078 |
- | -0.073 |
ies | -0.073 |
ier | -0.068 |
ever | -0.065 |
Token | hugged |
Token ID | 41130 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
Here | +0.012 |
V | +0.012 |
Higher | +0.012 |
Grab | +0.011 |
First | +0.011 |
Neg logit contributions | |
than | -0.015 |
- | -0.014 |
ies | -0.014 |
ier | -0.014 |
inches | -0.012 |
Token | her |
Token ID | 607 |
Feature activation | +6.26e-03 |
Pos logit contributions | |
V | +0.029 |
Here | +0.028 |
Higher | +0.028 |
Grab | +0.027 |
First | +0.026 |
Neg logit contributions | |
than | -0.028 |
- | -0.026 |
ies | -0.025 |
ier | -0.024 |
ever | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | -3.92e-02 |
Pos logit contributions | |
ier | +0.010 |
than | +0.009 |
ge | +0.009 |
ines | +0.009 |
supposed | +0.009 |
Neg logit contributions | |
The | -0.014 |
Higher | -0.013 |
V | -0.013 |
First | -0.012 |
Grab | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -4.46e-02 |
Pos logit contributions | |
ines | +0.055 |
owing | +0.054 |
ier | +0.054 |
ies | +0.053 |
ingly | +0.052 |
Neg logit contributions | |
The | -0.045 |
It | -0.040 |
V | -0.040 |
Many | -0.038 |
One | -0.038 |
Token | It |
Token ID | 632 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
Here | +0.068 |
V | +0.066 |
Grab | +0.063 |
Quick | +0.059 |
Higher | +0.059 |
Neg logit contributions | |
than | -0.098 |
ever | -0.085 |
- | -0.080 |
inches | -0.079 |
too | -0.078 |
Token | is |
Token ID | 318 |
Feature activation | +2.82e-02 |
Pos logit contributions | |
than | +0.037 |
ier | +0.035 |
ies | +0.034 |
- | +0.033 |
ge | +0.032 |
Neg logit contributions | |
V | -0.026 |
Here | -0.025 |
Grab | -0.024 |
Higher | -0.024 |
The | -0.024 |
Token | just |
Token ID | 655 |
Feature activation | +3.05e-02 |
Pos logit contributions | |
ier | +0.057 |
owing | +0.056 |
ies | +0.056 |
- | +0.052 |
ge | +0.052 |
Neg logit contributions | |
V | -0.050 |
The | -0.050 |
Here | -0.046 |
Higher | -0.045 |
Bang | -0.045 |
Token | metal |
Token ID | 6147 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
ier | +0.075 |
than | +0.074 |
ies | +0.074 |
- | +0.071 |
ge | +0.071 |
Neg logit contributions | |
The | -0.040 |
Here | -0.039 |
Higher | -0.039 |
V | -0.038 |
First | -0.036 |
Token | ." |
Token ID | 526 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
owing | +0.179 |
ge | +0.179 |
ies | +0.161 |
supposed | +0.160 |
believing | +0.159 |
Neg logit contributions | |
The | -0.168 |
It | -0.160 |
First | -0.155 |
One | -0.150 |
V | -0.149 |
Token | The |
Token ID | 383 |
Feature activation | -4.16e-02 |
Pos logit contributions | |
Ro | +0.245 |
Grab | +0.241 |
V | +0.236 |
Here | +0.232 |
Bang | +0.225 |
Neg logit contributions | |
than | -0.242 |
- | -0.238 |
ies | -0.235 |
and | -0.216 |
for | -0.206 |
Token | boy |
Token ID | 2933 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
V | +0.091 |
Here | +0.090 |
Higher | +0.090 |
Grab | +0.087 |
First | +0.087 |
Neg logit contributions | |
than | -0.071 |
ies | -0.071 |
- | -0.069 |
ier | -0.064 |
inches | -0.058 |
Token | said |
Token ID | 531 |
Feature activation | -3.08e-02 |
Pos logit contributions | |
V | +0.023 |
Here | +0.022 |
Grab | +0.021 |
Higher | +0.021 |
First | +0.020 |
Neg logit contributions | |
than | -0.033 |
ier | -0.030 |
ies | -0.030 |
- | -0.029 |
ge | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | -4.32e-02 |
Pos logit contributions | |
V | +0.080 |
Here | +0.080 |
Higher | +0.078 |
Grab | +0.077 |
First | +0.074 |
Neg logit contributions | |
than | -0.043 |
ies | -0.036 |
ier | -0.036 |
- | -0.036 |
inches | -0.031 |
Token | " |
Token ID | 366 |
Feature activation | -3.89e-02 |
Pos logit contributions | |
Here | +0.069 |
V | +0.067 |
Grab | +0.063 |
Higher | +0.060 |
First | +0.060 |
Neg logit contributions | |
than | -0.104 |
- | -0.099 |
ies | -0.093 |
ier | -0.091 |
too | -0.089 |
Token | fun |
Token ID | 1257 |
Feature activation | +5.87e-02 |
Pos logit contributions | |
than | +0.036 |
ier | +0.034 |
ies | +0.034 |
- | +0.033 |
owing | +0.032 |
Neg logit contributions | |
V | -0.036 |
Here | -0.036 |
Higher | -0.034 |
Many | -0.032 |
Look | -0.032 |
Token | with |
Token ID | 351 |
Feature activation | +3.34e-02 |
Pos logit contributions | |
ge | +0.099 |
ier | +0.095 |
owing | +0.094 |
decision | +0.089 |
supposed | +0.088 |
Neg logit contributions | |
V | -0.110 |
The | -0.108 |
Many | -0.104 |
Ro | -0.103 |
Higher | -0.103 |
Token | paper |
Token ID | 3348 |
Feature activation | +9.63e-02 |
Pos logit contributions | |
than | +0.074 |
ier | +0.069 |
- | +0.067 |
ies | +0.066 |
ge | +0.063 |
Neg logit contributions | |
V | -0.058 |
Here | -0.058 |
Higher | -0.056 |
Grab | -0.053 |
First | -0.051 |
Token | and |
Token ID | 290 |
Feature activation | +3.05e-02 |
Pos logit contributions | |
ge | +0.153 |
supposed | +0.141 |
owing | +0.140 |
ier | +0.134 |
than | +0.132 |
Neg logit contributions | |
Here | -0.187 |
V | -0.186 |
The | -0.184 |
First | -0.183 |
Ro | -0.177 |
Token | scissors |
Token ID | 41786 |
Feature activation | +8.46e-02 |
Pos logit contributions | |
than | +0.063 |
ier | +0.061 |
- | +0.059 |
ies | +0.057 |
ge | +0.055 |
Neg logit contributions | |
V | -0.057 |
Here | -0.057 |
Higher | -0.054 |
Grab | -0.053 |
Ro | -0.052 |
Token | ." |
Token ID | 526 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
supposed | +0.124 |
ge | +0.124 |
owing | +0.120 |
inches | +0.117 |
than | +0.116 |
Neg logit contributions | |
The | -0.159 |
V | -0.152 |
Here | -0.148 |
It | -0.144 |
First | -0.141 |
Token | They |
Token ID | 1119 |
Feature activation | -4.63e-02 |
Pos logit contributions | |
V | +0.188 |
Ro | +0.168 |
Grab | +0.168 |
Here | +0.167 |
Bang | +0.158 |
Neg logit contributions | |
than | -0.319 |
ies | -0.294 |
- | -0.279 |
anymore | -0.269 |
for | -0.263 |
Token | hug |
Token ID | 16225 |
Feature activation | +9.69e-04 |
Pos logit contributions | |
Here | +0.095 |
V | +0.095 |
Grab | +0.090 |
Higher | +0.090 |
Ro | +0.087 |
Neg logit contributions | |
than | -0.086 |
ies | -0.080 |
ier | -0.078 |
- | -0.077 |
ever | -0.071 |
Token | and |
Token ID | 290 |
Feature activation | -3.08e-02 |
Pos logit contributions | |
than | +0.001 |
ier | +0.001 |
ies | +0.001 |
- | +0.001 |
inches | +0.001 |
Neg logit contributions | |
V | -0.003 |
Here | -0.003 |
Higher | -0.003 |
Grab | -0.002 |
First | -0.002 |
Token | laugh |
Token ID | 6487 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
Here | +0.052 |
V | +0.052 |
Higher | +0.050 |
Grab | +0.050 |
First | +0.047 |
Neg logit contributions | |
than | -0.070 |
ier | -0.066 |
- | -0.065 |
ies | -0.064 |
ge | -0.060 |
Token | and |
Token ID | 290 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
than | +0.027 |
ies | +0.023 |
ier | +0.023 |
- | +0.022 |
ge | +0.022 |
Neg logit contributions | |
V | -0.064 |
Here | -0.062 |
Higher | -0.061 |
The | -0.059 |
First | -0.058 |
Token | has |
Token ID | 468 |
Feature activation | +3.44e-02 |
Pos logit contributions | |
than | +0.036 |
ier | +0.033 |
ies | +0.033 |
- | +0.032 |
ge | +0.031 |
Neg logit contributions | |
V | -0.027 |
Here | -0.027 |
Grab | -0.025 |
Ro | -0.024 |
Higher | -0.024 |
Token | a |
Token ID | 257 |
Feature activation | +3.12e-02 |
Pos logit contributions | |
than | +0.078 |
ies | +0.075 |
ier | +0.074 |
- | +0.072 |
owing | +0.072 |
Neg logit contributions | |
V | -0.055 |
Here | -0.055 |
Higher | -0.051 |
Look | -0.049 |
Many | -0.049 |
Token | big |
Token ID | 1263 |
Feature activation | +5.38e-02 |
Pos logit contributions | |
than | +0.066 |
ier | +0.064 |
ies | +0.061 |
owing | +0.060 |
- | +0.058 |
Neg logit contributions | |
V | -0.056 |
Here | -0.055 |
Look | -0.052 |
Ro | -0.049 |
Grab | -0.049 |
Token | flower |
Token ID | 15061 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
ier | +0.118 |
than | +0.118 |
ies | +0.110 |
owing | +0.109 |
ge | +0.109 |
Neg logit contributions | |
V | -0.082 |
Here | -0.081 |
Look | -0.078 |
Grab | -0.076 |
Higher | -0.076 |
Token | on |
Token ID | 319 |
Feature activation | +3.42e-02 |
Pos logit contributions | |
ier | +0.159 |
ge | +0.151 |
owing | +0.143 |
ies | +0.139 |
than | +0.138 |
Neg logit contributions | |
Here | -0.190 |
The | -0.190 |
It | -0.182 |
V | -0.179 |
Ro | -0.178 |
Token | top |
Token ID | 1353 |
Feature activation | +8.31e-02 |
Pos logit contributions | |
than | +0.070 |
ies | +0.064 |
ier | +0.064 |
- | +0.064 |
inches | +0.058 |
Neg logit contributions | |
V | -0.066 |
Here | -0.065 |
Higher | -0.063 |
Grab | -0.061 |
First | -0.059 |
Token | . |
Token ID | 13 |
Feature activation | -3.91e-02 |
Pos logit contributions | |
ies | +0.119 |
owing | +0.117 |
ier | +0.115 |
than | +0.112 |
supposed | +0.107 |
Neg logit contributions | |
The | -0.168 |
First | -0.163 |
Here | -0.163 |
Many | -0.162 |
V | -0.161 |
Token | It |
Token ID | 632 |
Feature activation | +8.76e-03 |
Pos logit contributions | |
Here | +0.047 |
V | +0.046 |
Grab | +0.045 |
Higher | +0.043 |
Quick | +0.042 |
Neg logit contributions | |
than | -0.105 |
- | -0.089 |
ever | -0.087 |
ier | -0.084 |
for | -0.082 |
Token | makes |
Token ID | 1838 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
than | +0.020 |
ier | +0.019 |
ies | +0.018 |
- | +0.018 |
ge | +0.018 |
Neg logit contributions | |
V | -0.014 |
Here | -0.014 |
Grab | -0.013 |
Ro | -0.013 |
Higher | -0.013 |
Token | you |
Token ID | 345 |
Feature activation | +4.94e-02 |
Pos logit contributions | |
than | +0.034 |
ies | +0.031 |
ier | +0.031 |
- | +0.031 |
inches | +0.028 |
Neg logit contributions | |
V | -0.028 |
Here | -0.028 |
Higher | -0.027 |
Grab | -0.026 |
First | -0.025 |
Token | look |
Token ID | 804 |
Feature activation | +5.14e-02 |
Pos logit contributions | |
ier | +0.105 |
than | +0.098 |
- | +0.094 |
ishing | +0.090 |
ies | +0.088 |
Neg logit contributions | |
The | -0.077 |
V | -0.072 |
Many | -0.071 |
It | -0.070 |
Higher | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | -8.60e-02 |
Pos logit contributions | |
ier | +0.082 |
ies | +0.072 |
owing | +0.070 |
inches | +0.070 |
ge | +0.068 |
Neg logit contributions | |
The | -0.087 |
One | -0.076 |
Their | -0.076 |
V | -0.075 |
Sam | -0.075 |
Token | And |
Token ID | 843 |
Feature activation | -2.38e-02 |
Pos logit contributions | |
Grab | +0.131 |
Here | +0.121 |
Higher | +0.120 |
V | +0.115 |
Ro | +0.107 |
Neg logit contributions | |
than | -0.221 |
ever | -0.176 |
anymore | -0.173 |
- | -0.171 |
too | -0.168 |
Token | they |
Token ID | 484 |
Feature activation | -1.61e-02 |
Pos logit contributions | |
V | +0.032 |
Here | +0.032 |
The | +0.032 |
Higher | +0.032 |
First | +0.030 |
Neg logit contributions | |
ier | -0.058 |
than | -0.057 |
- | -0.056 |
ies | -0.056 |
ge | -0.055 |
Token | did |
Token ID | 750 |
Feature activation | -4.21e-03 |
Pos logit contributions | |
V | +0.031 |
Here | +0.031 |
Higher | +0.030 |
Grab | +0.029 |
First | +0.028 |
Neg logit contributions | |
than | -0.033 |
- | -0.030 |
ies | -0.030 |
ier | -0.030 |
ever | -0.027 |
Token | not |
Token ID | 407 |
Feature activation | +9.48e-04 |
Pos logit contributions | |
V | +0.007 |
Higher | +0.007 |
Here | +0.007 |
The | +0.007 |
First | +0.006 |
Neg logit contributions | |
than | -0.009 |
ier | -0.009 |
ies | -0.009 |
- | -0.008 |
owing | -0.008 |
Token | jump |
Token ID | 4391 |
Feature activation | +8.63e-02 |
Pos logit contributions | |
ier | +0.002 |
than | +0.002 |
- | +0.002 |
ies | +0.002 |
inches | +0.002 |
Neg logit contributions | |
V | -0.001 |
Here | -0.001 |
The | -0.001 |
Many | -0.001 |
Hi | -0.001 |
Token | in |
Token ID | 287 |
Feature activation | +4.14e-02 |
Pos logit contributions | |
than | +0.118 |
ies | +0.100 |
owing | +0.099 |
inches | +0.096 |
ier | +0.092 |
Neg logit contributions | |
V | -0.209 |
Higher | -0.197 |
Here | -0.195 |
The | -0.194 |
Ro | -0.190 |
Token | the |
Token ID | 262 |
Feature activation | +2.27e-03 |
Pos logit contributions | |
ier | +0.102 |
owing | +0.098 |
than | +0.098 |
ies | +0.097 |
- | +0.093 |
Neg logit contributions | |
V | -0.056 |
Here | -0.054 |
The | -0.052 |
Higher | -0.049 |
Look | -0.049 |
Token | mud |
Token ID | 17492 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
than | +0.004 |
- | +0.004 |
ies | +0.004 |
ier | +0.004 |
inches | +0.004 |
Neg logit contributions | |
V | -0.005 |
Here | -0.005 |
Higher | -0.005 |
Grab | -0.004 |
First | -0.004 |
Token | again |
Token ID | 757 |
Feature activation | +5.51e-02 |
Pos logit contributions | |
supposed | +0.149 |
than | +0.143 |
owing | +0.128 |
ies | +0.127 |
ge | +0.123 |
Neg logit contributions | |
Here | -0.210 |
Ro | -0.206 |
V | -0.205 |
Look | -0.203 |
The | -0.196 |
Token | . |
Token ID | 13 |
Feature activation | -7.90e-02 |
Pos logit contributions | |
ier | +0.094 |
owing | +0.085 |
ies | +0.085 |
supposed | +0.085 |
than | +0.081 |
Neg logit contributions | |
The | -0.106 |
One | -0.092 |
V | -0.092 |
Max | -0.091 |
They | -0.088 |
Token | " |
Token ID | 366 |
Feature activation | -4.03e-02 |
Pos logit contributions | |
Here | +0.058 |
V | +0.057 |
Grab | +0.054 |
Higher | +0.051 |
Ro | +0.050 |
Neg logit contributions | |
than | -0.109 |
- | -0.103 |
ies | -0.097 |
ier | -0.097 |
ever | -0.093 |
Token | Uh |
Token ID | 34653 |
Feature activation | +3.07e-02 |
Pos logit contributions | |
Higher | +0.047 |
Many | +0.045 |
V | +0.043 |
First | +0.042 |
Their | +0.039 |
Neg logit contributions | |
than | -0.108 |
- | -0.096 |
ies | -0.091 |
ever | -0.091 |
y | -0.088 |
Token | oh |
Token ID | 11752 |
Feature activation | -1.74e-03 |
Pos logit contributions | |
than | +0.059 |
ier | +0.054 |
ies | +0.050 |
ge | +0.049 |
inches | +0.049 |
Neg logit contributions | |
V | -0.060 |
Here | -0.055 |
Grab | -0.054 |
Higher | -0.053 |
First | -0.053 |
Token | , |
Token ID | 11 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
V | +0.003 |
Higher | +0.003 |
The | +0.003 |
First | +0.003 |
Here | +0.003 |
Neg logit contributions | |
than | -0.003 |
ies | -0.003 |
ier | -0.003 |
ge | -0.003 |
- | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
V | +0.021 |
Here | +0.021 |
Higher | +0.020 |
Grab | +0.020 |
First | +0.019 |
Neg logit contributions | |
than | -0.026 |
- | -0.023 |
ies | -0.023 |
ier | -0.023 |
inches | -0.021 |
Token | roof |
Token ID | 9753 |
Feature activation | +8.89e-02 |
Pos logit contributions | |
than | +0.037 |
ier | +0.034 |
- | +0.034 |
ies | +0.034 |
ge | +0.032 |
Neg logit contributions | |
V | -0.024 |
Here | -0.024 |
Higher | -0.023 |
Grab | -0.022 |
First | -0.021 |
Token | has |
Token ID | 468 |
Feature activation | +4.86e-02 |
Pos logit contributions | |
than | +0.211 |
ge | +0.193 |
knowing | +0.190 |
ier | +0.185 |
ies | +0.185 |
Neg logit contributions | |
The | -0.111 |
It | -0.103 |
V | -0.101 |
Higher | -0.098 |
First | -0.094 |
Token | a |
Token ID | 257 |
Feature activation | +4.24e-02 |
Pos logit contributions | |
than | +0.100 |
ies | +0.097 |
ier | +0.091 |
owing | +0.091 |
- | +0.091 |
Neg logit contributions | |
V | -0.081 |
Here | -0.081 |
Higher | -0.081 |
The | -0.081 |
First | -0.077 |
Token | hole |
Token ID | 7604 |
Feature activation | +8.33e-02 |
Pos logit contributions | |
than | +0.095 |
ier | +0.095 |
ies | +0.092 |
owing | +0.090 |
ge | +0.084 |
Neg logit contributions | |
V | -0.066 |
Here | -0.063 |
Ro | -0.062 |
Look | -0.061 |
Higher | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | -4.51e-02 |
Pos logit contributions | |
ies | +0.113 |
than | +0.111 |
ge | +0.100 |
- | +0.100 |
owing | +0.099 |
Neg logit contributions | |
The | -0.168 |
V | -0.163 |
Here | -0.161 |
Higher | -0.161 |
Look | -0.155 |
Token | What |
Token ID | 1867 |
Feature activation | -2.17e-02 |
Pos logit contributions | |
Here | +0.072 |
V | +0.064 |
Grab | +0.060 |
Look | +0.059 |
Higher | +0.057 |
Neg logit contributions | |
than | -0.098 |
- | -0.091 |
ever | -0.085 |
too | -0.085 |
because | -0.084 |
Token | was |
Token ID | 373 |
Feature activation | -4.07e-04 |
Pos logit contributions | |
V | +0.031 |
Here | +0.030 |
Grab | +0.028 |
Higher | +0.028 |
First | +0.027 |
Neg logit contributions | |
than | -0.040 |
ier | -0.036 |
ies | -0.036 |
- | -0.036 |
inches | -0.033 |
Token | happy |
Token ID | 3772 |
Feature activation | +4.15e-02 |
Pos logit contributions | |
Here | +0.001 |
V | +0.001 |
Higher | +0.001 |
Grab | +0.001 |
The | +0.001 |
Neg logit contributions | |
than | -0.001 |
ies | -0.001 |
ier | -0.001 |
- | -0.001 |
ge | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
ier | +0.073 |
ge | +0.071 |
inches | +0.064 |
owing | +0.064 |
ishing | +0.061 |
Neg logit contributions | |
The | -0.079 |
Everyone | -0.075 |
It | -0.074 |
Max | -0.073 |
First | -0.071 |
Token | see |
Token ID | 766 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
V | +0.019 |
Here | +0.019 |
Higher | +0.018 |
Grab | +0.018 |
First | +0.016 |
Neg logit contributions | |
than | -0.022 |
- | -0.020 |
ies | -0.020 |
ier | -0.020 |
ever | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | +3.17e-02 |
Pos logit contributions | |
than | +0.026 |
ier | +0.024 |
ies | +0.024 |
- | +0.023 |
inches | +0.023 |
Neg logit contributions | |
Here | -0.020 |
V | -0.019 |
Higher | -0.019 |
Grab | -0.018 |
First | -0.018 |
Token | water |
Token ID | 1660 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
than | +0.068 |
ier | +0.063 |
- | +0.061 |
ies | +0.061 |
owing | +0.059 |
Neg logit contributions | |
Here | -0.059 |
V | -0.055 |
Higher | -0.052 |
Look | -0.052 |
Grab | -0.052 |
Token | and |
Token ID | 290 |
Feature activation | +4.40e-03 |
Pos logit contributions | |
than | +0.150 |
- | +0.136 |
ier | +0.136 |
ies | +0.134 |
ge | +0.132 |
Neg logit contributions | |
Here | -0.212 |
The | -0.207 |
Hi | -0.202 |
Higher | -0.200 |
Look | -0.195 |
Token | it |
Token ID | 340 |
Feature activation | +2.93e-02 |
Pos logit contributions | |
than | +0.010 |
ier | +0.009 |
- | +0.009 |
ies | +0.009 |
ge | +0.008 |
Neg logit contributions | |
Here | -0.008 |
V | -0.007 |
Higher | -0.007 |
Grab | -0.007 |
First | -0.007 |
Token | picked |
Token ID | 6497 |
Feature activation | +3.47e-02 |
Pos logit contributions | |
than | +0.065 |
ier | +0.060 |
ies | +0.059 |
- | +0.059 |
owing | +0.057 |
Neg logit contributions | |
V | -0.048 |
Here | -0.048 |
Higher | -0.045 |
Grab | -0.044 |
Ro | -0.043 |
Token | itself |
Token ID | 2346 |
Feature activation | +7.17e-02 |
Pos logit contributions | |
than | +0.066 |
ier | +0.060 |
ies | +0.059 |
ge | +0.057 |
- | +0.056 |
Neg logit contributions | |
Here | -0.071 |
V | -0.070 |
Higher | -0.068 |
Look | -0.065 |
Grab | -0.065 |
Token | down |
Token ID | 866 |
Feature activation | +4.44e-02 |
Pos logit contributions | |
than | +0.137 |
ier | +0.130 |
inches | +0.125 |
supposed | +0.123 |
owing | +0.121 |
Neg logit contributions | |
The | -0.116 |
Higher | -0.110 |
Here | -0.106 |
First | -0.104 |
V | -0.104 |
Token | do |
Token ID | 466 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
V | +0.063 |
Here | +0.062 |
Higher | +0.059 |
Grab | +0.058 |
First | +0.056 |
Neg logit contributions | |
than | -0.069 |
ies | -0.062 |
- | -0.062 |
ier | -0.062 |
inches | -0.056 |
Token | not |
Token ID | 407 |
Feature activation | -7.25e-03 |
Pos logit contributions | |
V | +0.020 |
Higher | +0.018 |
Here | +0.018 |
The | +0.017 |
Look | +0.017 |
Neg logit contributions | |
than | -0.029 |
ies | -0.028 |
ier | -0.028 |
owing | -0.026 |
- | -0.026 |
Token | see |
Token ID | 766 |
Feature activation | -2.22e-03 |
Pos logit contributions | |
V | +0.012 |
Many | +0.010 |
Here | +0.010 |
Hi | +0.010 |
Ro | +0.010 |
Neg logit contributions | |
than | -0.016 |
- | -0.016 |
inches | -0.015 |
ies | -0.015 |
ier | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | +2.55e-02 |
Pos logit contributions | |
V | +0.004 |
Here | +0.003 |
Higher | +0.003 |
The | +0.003 |
Look | +0.003 |
Neg logit contributions | |
than | -0.005 |
ier | -0.005 |
ies | -0.005 |
- | -0.005 |
ge | -0.004 |
Token | oct |
Token ID | 19318 |
Feature activation | +6.45e-02 |
Pos logit contributions | |
than | +0.050 |
owing | +0.049 |
ier | +0.048 |
- | +0.048 |
ge | +0.047 |
Neg logit contributions | |
Here | -0.045 |
Look | -0.044 |
V | -0.042 |
Bang | -0.041 |
Hi | -0.040 |
Token | opus |
Token ID | 25790 |
Feature activation | +8.31e-02 |
Pos logit contributions | |
than | +0.157 |
ies | +0.144 |
ier | +0.141 |
- | +0.138 |
ge | +0.137 |
Neg logit contributions | |
V | -0.094 |
Here | -0.092 |
Grab | -0.085 |
Ro | -0.085 |
Higher | -0.084 |
Token | that |
Token ID | 326 |
Feature activation | -3.17e-03 |
Pos logit contributions | |
than | +0.145 |
ier | +0.134 |
ies | +0.120 |
inches | +0.111 |
- | +0.111 |
Neg logit contributions | |
Give | -0.152 |
V | -0.146 |
The | -0.146 |
Here | -0.146 |
Ro | -0.145 |
Token | is |
Token ID | 318 |
Feature activation | +3.13e-02 |
Pos logit contributions | |
V | +0.006 |
Here | +0.006 |
Higher | +0.006 |
Grab | +0.006 |
First | +0.005 |
Neg logit contributions | |
than | -0.007 |
ies | -0.006 |
- | -0.006 |
ier | -0.006 |
inches | -0.005 |
Token | hiding |
Token ID | 11816 |
Feature activation | +5.99e-02 |
Pos logit contributions | |
than | +0.055 |
ier | +0.053 |
ies | +0.053 |
- | +0.051 |
owing | +0.049 |
Neg logit contributions | |
V | -0.062 |
Here | -0.057 |
Look | -0.056 |
Bang | -0.055 |
Grab | -0.054 |
Token | under |
Token ID | 739 |
Feature activation | +2.37e-03 |
Pos logit contributions | |
ier | +0.076 |
than | +0.073 |
owing | +0.072 |
ies | +0.072 |
ge | +0.071 |
Neg logit contributions | |
The | -0.121 |
V | -0.120 |
Look | -0.115 |
Bang | -0.115 |
Here | -0.114 |
Token | the |
Token ID | 262 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
than | +0.005 |
- | +0.005 |
ies | +0.005 |
ier | +0.004 |
inches | +0.004 |
Neg logit contributions | |
V | -0.005 |
Here | -0.004 |
Higher | -0.004 |
Grab | -0.004 |
First | -0.004 |
Token | arre |
Token ID | 9624 |
Feature activation | +6.88e-02 |
Pos logit contributions | |
than | +0.087 |
ier | +0.077 |
ies | +0.077 |
- | +0.077 |
ever | +0.069 |
Neg logit contributions | |
V | -0.098 |
Here | -0.095 |
Higher | -0.092 |
Grab | -0.092 |
First | -0.088 |
Token | ling |
Token ID | 1359 |
Feature activation | +4.39e-02 |
Pos logit contributions | |
than | +0.115 |
ever | +0.096 |
owing | +0.096 |
inches | +0.093 |
ier | +0.092 |
Neg logit contributions | |
V | -0.137 |
Bang | -0.129 |
Grab | -0.127 |
Higher | -0.126 |
Here | -0.123 |
Token | again |
Token ID | 757 |
Feature activation | +4.55e-02 |
Pos logit contributions | |
ier | +0.063 |
ies | +0.060 |
than | +0.060 |
ge | +0.056 |
supposed | +0.054 |
Neg logit contributions | |
Higher | -0.100 |
The | -0.098 |
Here | -0.096 |
V | -0.094 |
Grab | -0.091 |
Token | and |
Token ID | 290 |
Feature activation | -1.50e-02 |
Pos logit contributions | |
ier | +0.085 |
ies | +0.072 |
iest | +0.066 |
iness | +0.064 |
ines | +0.064 |
Neg logit contributions | |
The | -0.088 |
Max | -0.077 |
Higher | -0.074 |
Suddenly | -0.074 |
| -0.073 |
Token | their |
Token ID | 511 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
V | +0.032 |
Here | +0.032 |
Higher | +0.032 |
Grab | +0.031 |
Look | +0.030 |
Neg logit contributions | |
than | -0.027 |
ier | -0.024 |
- | -0.024 |
ies | -0.023 |
owing | -0.021 |
Token | voices |
Token ID | 10839 |
Feature activation | +4.66e-02 |
Pos logit contributions | |
than | +0.022 |
ier | +0.020 |
ies | +0.018 |
- | +0.017 |
ge | +0.017 |
Neg logit contributions | |
Here | -0.021 |
V | -0.021 |
Higher | -0.020 |
Grab | -0.020 |
Ro | -0.019 |
Token | grew |
Token ID | 6348 |
Feature activation | +5.25e-02 |
Pos logit contributions | |
than | +0.111 |
ies | +0.097 |
inches | +0.095 |
- | +0.095 |
ier | +0.093 |
Neg logit contributions | |
V | -0.069 |
Higher | -0.065 |
Here | -0.065 |
Grab | -0.063 |
The | -0.062 |
Token | louder |
Token ID | 27089 |
Feature activation | +7.05e-02 |
Pos logit contributions | |
than | +0.105 |
ies | +0.087 |
- | +0.086 |
ever | +0.084 |
owing | +0.082 |
Neg logit contributions | |
V | -0.096 |
Higher | -0.092 |
Here | -0.090 |
The | -0.088 |
Grab | -0.088 |
Token | and |
Token ID | 290 |
Feature activation | -7.10e-03 |
Pos logit contributions | |
ier | +0.110 |
supposed | +0.107 |
ishing | +0.102 |
ig | +0.102 |
owing | +0.102 |
Neg logit contributions | |
The | -0.129 |
Everyone | -0.116 |
It | -0.115 |
They | -0.115 |
Many | -0.109 |
Token | louder |
Token ID | 27089 |
Feature activation | +4.96e-02 |
Pos logit contributions | |
V | +0.015 |
Here | +0.014 |
Higher | +0.014 |
Grab | +0.013 |
Look | +0.013 |
Neg logit contributions | |
than | -0.014 |
- | -0.013 |
ies | -0.013 |
ier | -0.012 |
owing | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
ier | +0.068 |
supposed | +0.066 |
ies | +0.065 |
owing | +0.064 |
ig | +0.064 |
Neg logit contributions | |
The | -0.103 |
Everyone | -0.096 |
Many | -0.091 |
It | -0.091 |
V | -0.091 |
Token |
|
Token ID | 198 |
Feature activation | -8.50e-02 |
Pos logit contributions | |
V | +0.150 |
Ro | +0.141 |
Here | +0.137 |
Grab | +0.136 |
Bang | +0.132 |
Neg logit contributions | |
than | -0.231 |
- | -0.214 |
! | -0.198 |
ever | -0.196 |
too | -0.195 |
Token |
|
Token ID | 198 |
Feature activation | -8.45e-02 |
Pos logit contributions | |
Grab | +0.146 |
Here | +0.141 |
Higher | +0.135 |
Ro | +0.133 |
V | +0.132 |
Neg logit contributions | |
than | -0.161 |
- | -0.161 |
ever | -0.144 |
ge | -0.140 |
! | -0.139 |
Token | The |
Token ID | 464 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
Higher | +0.126 |
Here | +0.124 |
V | +0.119 |
Grab | +0.115 |
The | +0.114 |
Neg logit contributions | |
than | -0.180 |
- | -0.170 |
ge | -0.153 |
ever | -0.151 |
ies | -0.148 |
Token | sw |
Token ID | 1509 |
Feature activation | +5.89e-02 |
Pos logit contributions | |
V | +0.023 |
Here | +0.023 |
Grab | +0.021 |
Higher | +0.021 |
Look | +0.020 |
Neg logit contributions | |
than | -0.026 |
ier | -0.023 |
ies | -0.023 |
- | -0.023 |
owing | -0.021 |
Token | an |
Token ID | 272 |
Feature activation | +3.02e-02 |
Pos logit contributions | |
than | +0.106 |
ier | +0.093 |
ies | +0.088 |
ever | +0.085 |
- | +0.084 |
Neg logit contributions | |
V | -0.125 |
Here | -0.121 |
Grab | -0.119 |
Higher | -0.117 |
First | -0.114 |
Token | kept |
Token ID | 4030 |
Feature activation | +2.44e-02 |
Pos logit contributions | |
than | +0.067 |
ier | +0.061 |
ies | +0.060 |
- | +0.058 |
inches | +0.056 |
Neg logit contributions | |
V | -0.052 |
Here | -0.051 |
Higher | -0.049 |
Grab | -0.049 |
Ro | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | +2.88e-02 |
Pos logit contributions | |
than | +0.048 |
ier | +0.046 |
ies | +0.044 |
- | +0.041 |
owing | +0.040 |
Neg logit contributions | |
V | -0.050 |
Here | -0.048 |
Higher | -0.047 |
Grab | -0.047 |
Look | -0.046 |
Token | matching |
Token ID | 12336 |
Feature activation | +9.96e-02 |
Pos logit contributions | |
than | +0.062 |
ier | +0.061 |
owing | +0.057 |
ies | +0.056 |
- | +0.055 |
Neg logit contributions | |
Here | -0.052 |
V | -0.049 |
Look | -0.048 |
Grab | -0.048 |
Bang | -0.045 |
Token | toy |
Token ID | 13373 |
Feature activation | +9.16e-02 |
Pos logit contributions | |
ier | +0.215 |
owing | +0.211 |
supposed | +0.202 |
than | +0.201 |
ies | +0.192 |
Neg logit contributions | |
V | -0.152 |
Here | -0.146 |
Higher | -0.143 |
First | -0.141 |
Grab | -0.137 |
Token | all |
Token ID | 477 |
Feature activation | +3.07e-02 |
Pos logit contributions | |
supposed | +0.133 |
ier | +0.132 |
ies | +0.129 |
than | +0.126 |
owing | +0.115 |
Neg logit contributions | |
Higher | -0.181 |
First | -0.181 |
The | -0.175 |
Here | -0.168 |
Max | -0.167 |
Token | to |
Token ID | 284 |
Feature activation | +4.68e-03 |
Pos logit contributions | |
ier | +0.054 |
ies | +0.053 |
inches | +0.053 |
than | +0.050 |
owing | +0.050 |
Neg logit contributions | |
The | -0.057 |
Higher | -0.055 |
Here | -0.052 |
First | -0.051 |
| -0.049 |
Token | of |
Token ID | 286 |
Feature activation | +6.25e-03 |
Pos logit contributions | |
than | +0.008 |
ies | +0.007 |
ier | +0.007 |
- | +0.007 |
ge | +0.006 |
Neg logit contributions | |
V | -0.020 |
Here | -0.020 |
Higher | -0.020 |
Grab | -0.019 |
First | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | +8.02e-03 |
Pos logit contributions | |
than | +0.014 |
- | +0.012 |
ies | +0.012 |
ier | +0.012 |
inches | +0.011 |
Neg logit contributions | |
V | -0.011 |
Here | -0.011 |
Higher | -0.011 |
Grab | -0.011 |
First | -0.010 |
Token | story |
Token ID | 1621 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
than | +0.015 |
ies | +0.013 |
ier | +0.013 |
- | +0.013 |
ge | +0.012 |
Neg logit contributions | |
V | -0.017 |
Here | -0.017 |
Higher | -0.017 |
Grab | -0.016 |
First | -0.016 |
Token | is |
Token ID | 318 |
Feature activation | -5.87e-03 |
Pos logit contributions | |
than | +0.029 |
ier | +0.028 |
ies | +0.028 |
ge | +0.028 |
- | +0.027 |
Neg logit contributions | |
V | -0.021 |
Higher | -0.020 |
First | -0.020 |
Here | -0.019 |
The | -0.019 |
Token | that |
Token ID | 326 |
Feature activation | +2.43e-03 |
Pos logit contributions | |
V | +0.013 |
Here | +0.013 |
Higher | +0.013 |
Grab | +0.012 |
First | +0.012 |
Neg logit contributions | |
than | -0.010 |
ies | -0.009 |
ier | -0.009 |
- | -0.009 |
ge | -0.008 |
Token | it |
Token ID | 340 |
Feature activation | +2.86e-02 |
Pos logit contributions | |
ier | +0.005 |
than | +0.005 |
ies | +0.004 |
- | +0.004 |
ge | +0.004 |
Neg logit contributions | |
Higher | -0.004 |
V | -0.004 |
Here | -0.004 |
First | -0.004 |
Grab | -0.004 |
Token | is |
Token ID | 318 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
than | +0.062 |
ier | +0.056 |
ies | +0.056 |
- | +0.055 |
ge | +0.054 |
Neg logit contributions | |
V | -0.049 |
Here | -0.048 |
Higher | -0.047 |
Grab | -0.046 |
First | -0.046 |
Token | important |
Token ID | 1593 |
Feature activation | +5.42e-02 |
Pos logit contributions | |
than | +0.029 |
ies | +0.028 |
ier | +0.027 |
- | +0.027 |
ge | +0.026 |
Neg logit contributions | |
V | -0.040 |
Here | -0.040 |
Higher | -0.038 |
First | -0.037 |
Grab | -0.037 |
Token | to |
Token ID | 284 |
Feature activation | -4.31e-04 |
Pos logit contributions | |
ier | +0.103 |
ge | +0.101 |
ies | +0.099 |
owing | +0.092 |
ines | +0.092 |
Neg logit contributions | |
First | -0.093 |
The | -0.091 |
It | -0.088 |
Higher | -0.083 |
Everyone | -0.081 |
Token | always |
Token ID | 1464 |
Feature activation | +3.16e-02 |
Pos logit contributions | |
V | +0.001 |
Here | +0.001 |
Higher | +0.001 |
Grab | +0.001 |
First | +0.001 |
Neg logit contributions | |
than | -0.001 |
- | -0.001 |
ies | -0.001 |
ier | -0.001 |
inches | -0.001 |
Token | listen |
Token ID | 6004 |
Feature activation | +4.73e-02 |
Pos logit contributions | |
ier | +0.074 |
ies | +0.073 |
than | +0.072 |
- | +0.069 |
ge | +0.067 |
Neg logit contributions | |
Here | -0.043 |
V | -0.042 |
Grab | -0.041 |
Higher | -0.040 |
First | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | -7.89e-02 |
Pos logit contributions | |
than | +0.039 |
ies | +0.038 |
ier | +0.038 |
ge | +0.037 |
supposed | +0.036 |
Neg logit contributions | |
The | -0.058 |
Everyone | -0.053 |
Many | -0.053 |
It | -0.053 |
Hi | -0.053 |
Token | He |
Token ID | 679 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
V | +0.100 |
Here | +0.097 |
Grab | +0.091 |
Higher | +0.085 |
Ro | +0.084 |
Neg logit contributions | |
than | -0.219 |
- | -0.197 |
ier | -0.185 |
inches | -0.183 |
ies | -0.183 |
Token | didn |
Token ID | 1422 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
V | +0.020 |
Here | +0.020 |
Higher | +0.019 |
Grab | +0.019 |
First | +0.018 |
Neg logit contributions | |
than | -0.026 |
ies | -0.024 |
ier | -0.024 |
- | -0.024 |
inches | -0.022 |
Token | 't |
Token ID | 470 |
Feature activation | -3.75e-04 |
Pos logit contributions | |
Here | +0.022 |
V | +0.022 |
Higher | +0.021 |
Grab | +0.020 |
Look | +0.019 |
Neg logit contributions | |
than | -0.036 |
- | -0.034 |
ier | -0.033 |
ies | -0.032 |
ever | -0.031 |
Token | want |
Token ID | 765 |
Feature activation | +4.45e-03 |
Pos logit contributions | |
V | +0.001 |
Many | +0.000 |
Hi | +0.000 |
Here | +0.000 |
Ro | +0.000 |
Neg logit contributions | |
than | -0.001 |
inches | -0.001 |
ies | -0.001 |
- | -0.001 |
ier | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
than | +0.008 |
ies | +0.007 |
ier | +0.007 |
- | +0.007 |
ge | +0.007 |
Neg logit contributions | |
Here | -0.010 |
V | -0.009 |
Higher | -0.009 |
The | -0.009 |
First | -0.009 |
Token | give |
Token ID | 1577 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
than | +0.023 |
ier | +0.022 |
ies | +0.021 |
- | +0.021 |
inches | +0.020 |
Neg logit contributions | |
V | -0.017 |
Here | -0.016 |
Grab | -0.015 |
Higher | -0.015 |
Ro | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
than | +0.036 |
ier | +0.033 |
ies | +0.032 |
- | +0.031 |
inches | +0.030 |
Neg logit contributions | |
Here | -0.034 |
V | -0.034 |
Higher | -0.033 |
First | -0.031 |
Grab | -0.031 |
Token | bottle |
Token ID | 9294 |
Feature activation | +8.37e-02 |
Pos logit contributions | |
than | +0.038 |
ier | +0.035 |
ies | +0.034 |
- | +0.034 |
ge | +0.032 |
Neg logit contributions | |
Here | -0.032 |
V | -0.030 |
Grab | -0.029 |
Higher | -0.028 |
Look | -0.028 |
Token | away |
Token ID | 1497 |
Feature activation | +6.52e-02 |
Pos logit contributions | |
than | +0.131 |
ies | +0.117 |
ier | +0.113 |
ge | +0.111 |
inches | +0.110 |
Neg logit contributions | |
The | -0.178 |
Here | -0.174 |
Higher | -0.170 |
First | -0.170 |
Grab | -0.166 |
Token | , |
Token ID | 11 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
than | +0.095 |
ies | +0.093 |
ier | +0.093 |
ishing | +0.091 |
- | +0.089 |
Neg logit contributions | |
The | -0.134 |
First | -0.121 |
It | -0.120 |
His | -0.119 |
Everyone | -0.119 |
Token | axe |
Token ID | 23272 |
Feature activation | +8.43e-02 |
Pos logit contributions | |
than | +0.044 |
ier | +0.041 |
ies | +0.040 |
- | +0.039 |
ge | +0.038 |
Neg logit contributions | |
Here | -0.035 |
V | -0.034 |
Higher | -0.032 |
Grab | -0.032 |
Ro | -0.031 |
Token | carefully |
Token ID | 7773 |
Feature activation | +5.83e-02 |
Pos logit contributions | |
than | +0.114 |
ies | +0.112 |
ier | +0.103 |
ge | +0.097 |
- | +0.093 |
Neg logit contributions | |
The | -0.185 |
Higher | -0.184 |
First | -0.182 |
Here | -0.181 |
V | -0.180 |
Token | , |
Token ID | 11 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
ies | +0.076 |
than | +0.074 |
ier | +0.071 |
- | +0.064 |
owing | +0.062 |
Neg logit contributions | |
Here | -0.135 |
V | -0.134 |
First | -0.134 |
The | -0.134 |
Higher | -0.134 |
Token | and |
Token ID | 290 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
V | +0.029 |
Here | +0.028 |
Higher | +0.028 |
Grab | +0.027 |
First | +0.026 |
Neg logit contributions | |
than | -0.025 |
ies | -0.023 |
ier | -0.023 |
- | -0.023 |
inches | -0.020 |
Token | only |
Token ID | 691 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
V | +0.033 |
Here | +0.033 |
Higher | +0.032 |
Grab | +0.031 |
First | +0.030 |
Neg logit contributions | |
than | -0.038 |
ier | -0.035 |
ies | -0.035 |
- | -0.035 |
owing | -0.032 |
Token | in |
Token ID | 287 |
Feature activation | +2.28e-02 |
Pos logit contributions | |
than | +0.022 |
ier | +0.021 |
ies | +0.021 |
- | +0.020 |
owing | +0.018 |
Neg logit contributions | |
V | -0.019 |
Here | -0.019 |
Higher | -0.019 |
First | -0.018 |
Grab | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | +9.36e-03 |
Pos logit contributions | |
ies | +0.048 |
ier | +0.047 |
than | +0.047 |
- | +0.044 |
owing | +0.042 |
Neg logit contributions | |
V | -0.041 |
Here | -0.040 |
Higher | -0.038 |
First | -0.037 |
Grab | -0.037 |
Token | barn |
Token ID | 25203 |
Feature activation | +5.26e-02 |
Pos logit contributions | |
than | +0.020 |
ies | +0.018 |
ier | +0.018 |
- | +0.018 |
ge | +0.016 |
Neg logit contributions | |
Here | -0.018 |
V | -0.018 |
Higher | -0.017 |
Grab | -0.017 |
First | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -8.49e-02 |
Pos logit contributions | |
than | +0.079 |
ies | +0.068 |
- | +0.067 |
ier | +0.066 |
ge | +0.061 |
Neg logit contributions | |
Here | -0.121 |
V | -0.121 |
Higher | -0.119 |
First | -0.117 |
The | -0.115 |
Token | â |
Token ID | 22940 |
Feature activation | +9.85e-03 |
Pos logit contributions | |
Here | +0.096 |
V | +0.095 |
Grab | +0.089 |
Ro | +0.084 |
Higher | +0.083 |
Neg logit contributions | |
than | -0.226 |
- | -0.203 |
ier | -0.196 |
inches | -0.189 |
ge | -0.188 |
Token | € |
Token ID | 26391 |
Feature activation | -2.34e-02 |
Pos logit contributions | |
than | +0.024 |
- | +0.023 |
ier | +0.022 |
ies | +0.021 |
owing | +0.021 |
Neg logit contributions | |
V | -0.015 |
Here | -0.014 |
Higher | -0.014 |
Ro | -0.013 |
Grab | -0.013 |
Token | children |
Token ID | 1751 |
Feature activation | +6.65e-02 |
Pos logit contributions | |
ier | +0.154 |
than | +0.144 |
ge | +0.142 |
ies | +0.140 |
owing | +0.137 |
Neg logit contributions | |
Here | -0.119 |
V | -0.112 |
Hi | -0.110 |
Higher | -0.110 |
Grab | -0.109 |
Token | 's |
Token ID | 338 |
Feature activation | +3.70e-02 |
Pos logit contributions | |
than | +0.107 |
ier | +0.102 |
ies | +0.096 |
inches | +0.090 |
- | +0.087 |
Neg logit contributions | |
Here | -0.130 |
The | -0.128 |
Higher | -0.124 |
There | -0.123 |
V | -0.123 |
Token | theater |
Token ID | 13766 |
Feature activation | +9.94e-02 |
Pos logit contributions | |
ier | +0.074 |
than | +0.068 |
ies | +0.064 |
ge | +0.063 |
- | +0.060 |
Neg logit contributions | |
Here | -0.069 |
Higher | -0.064 |
V | -0.064 |
The | -0.063 |
Many | -0.062 |
Token | inside |
Token ID | 2641 |
Feature activation | +4.24e-02 |
Pos logit contributions | |
ies | +0.132 |
ier | +0.131 |
ge | +0.126 |
than | +0.125 |
inches | +0.119 |
Neg logit contributions | |
The | -0.202 |
Here | -0.196 |
Everyone | -0.196 |
V | -0.192 |
Hi | -0.190 |
Token | . |
Token ID | 13 |
Feature activation | -8.74e-02 |
Pos logit contributions | |
than | +0.059 |
ier | +0.059 |
ies | +0.056 |
- | +0.052 |
ge | +0.051 |
Neg logit contributions | |
Here | -0.097 |
The | -0.096 |
Higher | -0.091 |
Everyone | -0.089 |
V | -0.089 |
Token | |
Token ID | 220 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
V | +0.162 |
Grab | +0.146 |
Ro | +0.146 |
Here | +0.141 |
Bang | +0.136 |
Neg logit contributions | |
than | -0.187 |
- | -0.161 |
ever | -0.159 |
! | -0.152 |
ies | -0.150 |
Token |
|
Token ID | 198 |
Feature activation | -8.31e-02 |
Pos logit contributions | |
Here | +0.048 |
Grab | +0.047 |
V | +0.046 |
Ro | +0.046 |
Hi | +0.042 |
Neg logit contributions | |
than | -0.068 |
- | -0.064 |
ies | -0.061 |
ier | -0.060 |
! | -0.057 |
Token |
|
Token ID | 198 |
Feature activation | -7.75e-02 |
Pos logit contributions | |
Grab | +0.145 |
V | +0.140 |
Ro | +0.132 |
Here | +0.131 |
First | +0.130 |
Neg logit contributions | |
than | -0.167 |
ever | -0.153 |
- | -0.153 |
! | -0.143 |
ines | -0.134 |
Token | Jane |
Token ID | 41083 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
V | +0.158 |
Higher | +0.150 |
Grab | +0.148 |
Here | +0.145 |
First | +0.145 |
Neg logit contributions | |
than | -0.143 |
- | -0.126 |
ever | -0.122 |
ier | -0.116 |
ies | -0.110 |
Token | was |
Token ID | 373 |
Feature activation | -4.52e-03 |
Pos logit contributions | |
than | +0.037 |
ier | +0.033 |
ies | +0.032 |
- | +0.031 |
inches | +0.030 |
Neg logit contributions | |
V | -0.031 |
Here | -0.029 |
Higher | -0.028 |
Grab | -0.028 |
The | -0.027 |
Token | so |
Token ID | 523 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
V | +0.011 |
Here | +0.010 |
Higher | +0.010 |
The | +0.010 |
Grab | +0.010 |
Neg logit contributions | |
than | -0.007 |
ies | -0.007 |
ier | -0.006 |
- | -0.006 |
ge | -0.006 |
Token | go |
Token ID | 467 |
Feature activation | +5.89e-02 |
Pos logit contributions | |
than | +0.153 |
ies | +0.146 |
ishing | +0.137 |
ier | +0.133 |
ge | +0.130 |
Neg logit contributions | |
V | -0.158 |
The | -0.157 |
Higher | -0.154 |
It | -0.151 |
Here | -0.148 |
Token | away |
Token ID | 1497 |
Feature activation | +6.63e-02 |
Pos logit contributions | |
than | +0.086 |
ier | +0.084 |
ies | +0.080 |
inches | +0.076 |
owing | +0.071 |
Neg logit contributions | |
V | -0.136 |
Here | -0.130 |
Grab | -0.128 |
The | -0.127 |
Ro | -0.126 |
Token | . |
Token ID | 13 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
ishing | +0.134 |
ies | +0.133 |
ig | +0.132 |
ines | +0.121 |
ins | +0.121 |
Neg logit contributions | |
The | -0.099 |
He | -0.083 |
One | -0.078 |
They | -0.075 |
Big | -0.074 |
Token |
|
Token ID | 198 |
Feature activation | -9.72e-02 |
Pos logit contributions | |
V | +0.186 |
Grab | +0.184 |
Ro | +0.181 |
Here | +0.178 |
Pull | +0.163 |
Neg logit contributions | |
than | -0.209 |
- | -0.200 |
too | -0.176 |
! | -0.175 |
because | -0.171 |
Token |
|
Token ID | 198 |
Feature activation | -9.76e-02 |
Pos logit contributions | |
Grab | +0.161 |
Here | +0.159 |
V | +0.147 |
Higher | +0.146 |
Ro | +0.137 |
Neg logit contributions | |
than | -0.222 |
- | -0.217 |
ies | -0.193 |
ever | -0.191 |
ge | -0.186 |
Token | Suddenly |
Token ID | 38582 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
Here | +0.174 |
Higher | +0.168 |
Grab | +0.158 |
First | +0.157 |
V | +0.150 |
Neg logit contributions | |
than | -0.208 |
- | -0.190 |
ge | -0.168 |
ever | -0.158 |
ies | -0.151 |
Token | , |
Token ID | 11 |
Feature activation | -7.04e-02 |
Pos logit contributions | |
V | +0.097 |
Here | +0.096 |
Higher | +0.093 |
Grab | +0.092 |
First | +0.089 |
Neg logit contributions | |
than | -0.044 |
- | -0.038 |
ies | -0.037 |
ier | -0.036 |
inches | -0.030 |
Token | the |
Token ID | 262 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
V | +0.141 |
Here | +0.135 |
Higher | +0.128 |
Grab | +0.128 |
First | +0.124 |
Neg logit contributions | |
than | -0.141 |
- | -0.131 |
ies | -0.127 |
ier | -0.127 |
ever | -0.118 |
Token | elephant |
Token ID | 20950 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
V | +0.020 |
Here | +0.019 |
Higher | +0.019 |
Grab | +0.019 |
First | +0.018 |
Neg logit contributions | |
than | -0.022 |
- | -0.020 |
ies | -0.020 |
ier | -0.019 |
inches | -0.018 |
Token | heard |
Token ID | 2982 |
Feature activation | -2.28e-02 |
Pos logit contributions | |
than | +0.040 |
ier | +0.035 |
ies | +0.035 |
- | +0.035 |
ge | +0.032 |
Neg logit contributions | |
V | -0.031 |
Here | -0.030 |
Grab | -0.029 |
Higher | -0.029 |
Ro | -0.027 |
Token | a |
Token ID | 257 |
Feature activation | -2.36e-03 |
Pos logit contributions | |
V | +0.045 |
Here | +0.044 |
Grab | +0.043 |
Higher | +0.042 |
First | +0.041 |
Neg logit contributions | |
than | -0.046 |
- | -0.044 |
ies | -0.041 |
ier | -0.041 |
inches | -0.038 |
Token | open |
Token ID | 1280 |
Feature activation | +7.41e-02 |
Pos logit contributions | |
than | +0.099 |
ier | +0.091 |
ies | +0.089 |
- | +0.089 |
ge | +0.085 |
Neg logit contributions | |
The | -0.141 |
V | -0.139 |
Here | -0.138 |
Look | -0.134 |
Higher | -0.132 |
Token | . |
Token ID | 13 |
Feature activation | -8.49e-02 |
Pos logit contributions | |
ier | +0.117 |
ies | +0.104 |
owing | +0.103 |
supposed | +0.102 |
than | +0.102 |
Neg logit contributions | |
The | -0.150 |
Here | -0.145 |
First | -0.142 |
V | -0.137 |
Look | -0.136 |
Token | She |
Token ID | 1375 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
Grab | +0.147 |
Here | +0.134 |
V | +0.133 |
Higher | +0.132 |
Quick | +0.127 |
Neg logit contributions | |
than | -0.182 |
- | -0.137 |
too | -0.137 |
and | -0.136 |
. | -0.132 |
Token | says |
Token ID | 1139 |
Feature activation | -3.56e-02 |
Pos logit contributions | |
V | +0.043 |
Here | +0.042 |
Higher | +0.040 |
Grab | +0.039 |
First | +0.037 |
Neg logit contributions | |
than | -0.055 |
ier | -0.050 |
ies | -0.050 |
- | -0.050 |
ge | -0.046 |
Token | , |
Token ID | 11 |
Feature activation | -3.46e-02 |
Pos logit contributions | |
Here | +0.099 |
V | +0.099 |
Grab | +0.097 |
Higher | +0.095 |
First | +0.092 |
Neg logit contributions | |
than | -0.045 |
- | -0.041 |
ies | -0.039 |
ier | -0.038 |
ever | -0.033 |
Token | " |
Token ID | 366 |
Feature activation | -2.91e-02 |
Pos logit contributions | |
Here | +0.049 |
V | +0.047 |
Grab | +0.046 |
Higher | +0.043 |
First | +0.042 |
Neg logit contributions | |
than | -0.091 |
- | -0.084 |
ies | -0.082 |
ier | -0.080 |
ever | -0.076 |
Token | Tom |
Token ID | 13787 |
Feature activation | +3.60e-02 |
Pos logit contributions | |
Higher | +0.034 |
First | +0.030 |
Everywhere | +0.027 |
V | +0.026 |
Many | +0.026 |
Neg logit contributions | |
than | -0.080 |
ever | -0.068 |
- | -0.068 |
ies | -0.064 |
ier | -0.063 |
Token | and |
Token ID | 290 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
than | +0.057 |
ier | +0.052 |
ge | +0.049 |
ies | +0.049 |
inches | +0.047 |
Neg logit contributions | |
V | -0.082 |
Here | -0.077 |
The | -0.076 |
Grab | -0.076 |
Higher | -0.074 |
Token | Anna |
Token ID | 11735 |
Feature activation | +3.47e-02 |
Pos logit contributions | |
V | +0.018 |
Here | +0.017 |
Higher | +0.016 |
Grab | +0.016 |
First | +0.015 |
Neg logit contributions | |
than | -0.037 |
- | -0.035 |
ies | -0.035 |
ier | -0.034 |
inches | -0.032 |
Token | , |
Token ID | 11 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
than | +0.062 |
ier | +0.058 |
ge | +0.055 |
ies | +0.054 |
- | +0.053 |
Neg logit contributions | |
V | -0.069 |
The | -0.066 |
Here | -0.065 |
Grab | -0.063 |
Many | -0.063 |
Token | you |
Token ID | 345 |
Feature activation | -6.89e-03 |
Pos logit contributions | |
than | +0.033 |
ier | +0.031 |
ies | +0.030 |
- | +0.030 |
ge | +0.029 |
Neg logit contributions | |
V | -0.024 |
Here | -0.024 |
Higher | -0.023 |
Grab | -0.022 |
Ro | -0.022 |
Token | anymore |
Token ID | 7471 |
Feature activation | +5.02e-02 |
Pos logit contributions | |
ies | +0.090 |
than | +0.079 |
ier | +0.076 |
- | +0.072 |
yt | +0.068 |
Neg logit contributions | |
Here | -0.153 |
V | -0.153 |
The | -0.152 |
Max | -0.151 |
Higher | -0.151 |
Token | . |
Token ID | 13 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
ier | +0.088 |
ishing | +0.082 |
ies | +0.080 |
owing | +0.079 |
inches | +0.078 |
Neg logit contributions | |
The | -0.092 |
Max | -0.088 |
One | -0.085 |
Everyone | -0.084 |
All | -0.084 |
Token | The |
Token ID | 383 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
Grab | +0.138 |
V | +0.136 |
Ro | +0.133 |
Here | +0.129 |
Higher | +0.122 |
Neg logit contributions | |
than | -0.256 |
- | -0.233 |
anymore | -0.220 |
forever | -0.216 |
ever | -0.216 |
Token | end |
Token ID | 886 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
V | +0.027 |
Here | +0.027 |
Higher | +0.026 |
Grab | +0.025 |
First | +0.024 |
Neg logit contributions | |
than | -0.026 |
- | -0.023 |
ies | -0.023 |
ier | -0.023 |
inches | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | -9.65e-02 |
Pos logit contributions | |
V | +0.025 |
The | +0.024 |
Here | +0.024 |
Max | +0.023 |
Higher | +0.022 |
Neg logit contributions | |
ier | -0.016 |
ies | -0.016 |
than | -0.015 |
inches | -0.015 |
ge | -0.014 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -5.08e-02 |
Pos logit contributions | |
V | +0.133 |
Grab | +0.131 |
Here | +0.127 |
Higher | +0.125 |
Ro | +0.124 |
Neg logit contributions | |
than | -0.245 |
- | -0.224 |
ever | -0.210 |
ies | -0.208 |
ier | -0.207 |
Token | Once |
Token ID | 7454 |
Feature activation | -4.12e-02 |
Pos logit contributions | |
Grab | +0.077 |
Here | +0.075 |
V | +0.074 |
Higher | +0.073 |
First | +0.070 |
Neg logit contributions | |
than | -0.120 |
- | -0.115 |
ever | -0.107 |
ier | -0.105 |
! | -0.103 |
Token | upon |
Token ID | 2402 |
Feature activation | -5.01e-02 |
Pos logit contributions | |
V | +0.098 |
Here | +0.096 |
Higher | +0.094 |
Grab | +0.091 |
First | +0.090 |
Neg logit contributions | |
than | -0.067 |
ier | -0.057 |
- | -0.057 |
ies | -0.056 |
inches | -0.052 |
Token | a |
Token ID | 257 |
Feature activation | -4.88e-02 |
Pos logit contributions | |
V | +0.094 |
Here | +0.093 |
Grab | +0.090 |
Higher | +0.090 |
First | +0.086 |
Neg logit contributions | |
than | -0.104 |
- | -0.096 |
ies | -0.092 |
ier | -0.090 |
ever | -0.087 |
Token | time |
Token ID | 640 |
Feature activation | -4.30e-02 |
Pos logit contributions | |
V | +0.121 |
Here | +0.119 |
Higher | +0.116 |
Grab | +0.114 |
First | +0.111 |
Neg logit contributions | |
than | -0.073 |
ier | -0.067 |
ies | -0.066 |
- | -0.063 |
inches | -0.057 |
Token | , |
Token ID | 11 |
Feature activation | -5.39e-02 |
Pos logit contributions | |
V | +0.114 |
Here | +0.112 |
Higher | +0.110 |
Grab | +0.108 |
First | +0.105 |
Neg logit contributions | |
than | -0.058 |
ies | -0.050 |
- | -0.049 |
ier | -0.049 |
inches | -0.042 |
Token | " |
Token ID | 366 |
Feature activation | -3.57e-02 |
Pos logit contributions | |
V | +0.059 |
Here | +0.058 |
Grab | +0.054 |
Higher | +0.052 |
First | +0.050 |
Neg logit contributions | |
than | -0.119 |
- | -0.111 |
ies | -0.106 |
ier | -0.106 |
ever | -0.101 |
Token | Wow |
Token ID | 22017 |
Feature activation | +5.94e-03 |
Pos logit contributions | |
Higher | +0.042 |
V | +0.040 |
First | +0.038 |
Many | +0.037 |
Grab | +0.036 |
Neg logit contributions | |
than | -0.098 |
- | -0.087 |
ier | -0.086 |
ies | -0.085 |
ever | -0.084 |
Token | , |
Token ID | 11 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
than | +0.009 |
ier | +0.008 |
- | +0.007 |
ies | +0.007 |
ge | +0.007 |
Neg logit contributions | |
V | -0.014 |
Higher | -0.014 |
Here | -0.013 |
Grab | -0.013 |
First | -0.013 |
Token | Lily |
Token ID | 20037 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
V | +0.029 |
Here | +0.028 |
Higher | +0.027 |
Grab | +0.026 |
First | +0.025 |
Neg logit contributions | |
than | -0.039 |
- | -0.036 |
ies | -0.036 |
ier | -0.035 |
inches | -0.033 |
Token | ! |
Token ID | 0 |
Feature activation | -4.51e-02 |
Pos logit contributions | |
than | +0.028 |
ge | +0.025 |
ier | +0.025 |
ies | +0.023 |
inches | +0.022 |
Neg logit contributions | |
V | -0.072 |
The | -0.070 |
Here | -0.070 |
Higher | -0.068 |
Grab | -0.067 |
Token | You |
Token ID | 921 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
V | +0.077 |
Here | +0.076 |
Spider | +0.070 |
Grab | +0.070 |
Ro | +0.069 |
Neg logit contributions | |
than | -0.087 |
ever | -0.080 |
- | -0.075 |
ier | -0.074 |
perfectly | -0.069 |
Token | collected |
Token ID | 7723 |
Feature activation | +5.33e-02 |
Pos logit contributions | |
V | +0.033 |
Here | +0.032 |
Higher | +0.031 |
Grab | +0.031 |
First | +0.029 |
Neg logit contributions | |
than | -0.032 |
- | -0.029 |
ies | -0.029 |
ier | -0.029 |
inches | -0.026 |
Token | so |
Token ID | 523 |
Feature activation | +3.43e-03 |
Pos logit contributions | |
than | +0.118 |
ies | +0.116 |
ier | +0.114 |
owing | +0.113 |
- | +0.110 |
Neg logit contributions | |
Here | -0.089 |
V | -0.084 |
Higher | -0.084 |
The | -0.079 |
First | -0.079 |
Token | many |
Token ID | 867 |
Feature activation | +6.64e-02 |
Pos logit contributions | |
than | +0.007 |
- | +0.006 |
ies | +0.006 |
ier | +0.006 |
inches | +0.005 |
Neg logit contributions | |
V | -0.007 |
Here | -0.007 |
Higher | -0.007 |
Grab | -0.007 |
First | -0.006 |
Token | apples |
Token ID | 22514 |
Feature activation | +8.84e-02 |
Pos logit contributions | |
ier | +0.149 |
than | +0.145 |
owing | +0.143 |
ies | +0.141 |
- | +0.139 |
Neg logit contributions | |
Here | -0.102 |
Higher | -0.094 |
Many | -0.091 |
V | -0.091 |
Grab | -0.090 |
Token | . |
Token ID | 13 |
Feature activation | -5.35e-02 |
Pos logit contributions | |
ge | +0.140 |
owing | +0.136 |
ies | +0.133 |
- | +0.126 |
ier | +0.126 |
Neg logit contributions | |
The | -0.160 |
First | -0.151 |
Here | -0.149 |
It | -0.149 |
Higher | -0.141 |
Token | . |
Token ID | 13 |
Feature activation | -9.91e-02 |
Pos logit contributions | |
ies | +0.092 |
ge | +0.088 |
than | +0.087 |
inches | +0.086 |
ier | +0.082 |
Neg logit contributions | |
Here | -0.155 |
V | -0.153 |
The | -0.150 |
First | -0.148 |
Ro | -0.147 |
Token |
|
Token ID | 198 |
Feature activation | -8.75e-02 |
Pos logit contributions | |
Here | +0.165 |
Grab | +0.156 |
V | +0.156 |
Higher | +0.150 |
Give | +0.139 |
Neg logit contributions | |
than | -0.213 |
- | -0.197 |
too | -0.184 |
ies | -0.178 |
anymore | -0.174 |
Token |
|
Token ID | 198 |
Feature activation | -8.51e-02 |
Pos logit contributions | |
Grab | +0.156 |
Here | +0.153 |
Higher | +0.146 |
V | +0.134 |
Pull | +0.131 |
Neg logit contributions | |
- | -0.182 |
than | -0.176 |
ines | -0.163 |
ies | -0.158 |
ever | -0.152 |
Token | One |
Token ID | 3198 |
Feature activation | -6.52e-03 |
Pos logit contributions | |
Here | +0.141 |
Higher | +0.140 |
V | +0.130 |
Grab | +0.128 |
Many | +0.116 |
Neg logit contributions | |
than | -0.194 |
- | -0.186 |
ies | -0.171 |
ever | -0.156 |
ge | -0.155 |
Token | day |
Token ID | 1110 |
Feature activation | -5.28e-02 |
Pos logit contributions | |
V | +0.015 |
Here | +0.014 |
Higher | +0.014 |
Grab | +0.014 |
First | +0.014 |
Neg logit contributions | |
than | -0.011 |
ier | -0.010 |
ge | -0.009 |
ies | -0.009 |
inches | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | -8.34e-02 |
Pos logit contributions | |
Here | +0.159 |
V | +0.158 |
Higher | +0.150 |
Grab | +0.149 |
Look | +0.144 |
Neg logit contributions | |
than | -0.056 |
- | -0.050 |
ies | -0.048 |
ier | -0.043 |
ever | -0.036 |
Token | Sam |
Token ID | 3409 |
Feature activation | -4.83e-03 |
Pos logit contributions | |
V | +0.131 |
Here | +0.130 |
Higher | +0.122 |
Grab | +0.119 |
Ro | +0.111 |
Neg logit contributions | |
than | -0.206 |
ies | -0.190 |
- | -0.190 |
ier | -0.183 |
inches | -0.172 |
Token | and |
Token ID | 290 |
Feature activation | -4.91e-02 |
Pos logit contributions | |
V | +0.013 |
Here | +0.013 |
Higher | +0.012 |
Grab | +0.012 |
First | +0.012 |
Neg logit contributions | |
than | -0.007 |
ier | -0.006 |
ies | -0.006 |
- | -0.005 |
inches | -0.005 |
Token | Mia |
Token ID | 32189 |
Feature activation | -2.30e-02 |
Pos logit contributions | |
Here | +0.082 |
V | +0.080 |
Higher | +0.077 |
Grab | +0.075 |
Look | +0.070 |
Neg logit contributions | |
than | -0.116 |
- | -0.108 |
ies | -0.107 |
ier | -0.101 |
ever | -0.097 |
Token | saw |
Token ID | 2497 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
V | +0.041 |
Here | +0.040 |
Higher | +0.039 |
Grab | +0.038 |
First | +0.036 |
Neg logit contributions | |
than | -0.051 |
ies | -0.047 |
- | -0.047 |
ier | -0.046 |
inches | -0.042 |
Token | a |
Token ID | 257 |
Feature activation | +8.92e-03 |
Pos logit contributions | |
V | +0.070 |
Higher | +0.065 |
Grab | +0.065 |
Here | +0.064 |
Ro | +0.064 |
Neg logit contributions | |
than | -0.050 |
- | -0.047 |
ies | -0.041 |
anymore | -0.040 |
for | -0.040 |
Token | and |
Token ID | 290 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
ier | +0.053 |
owing | +0.044 |
than | +0.044 |
ge | +0.043 |
ies | +0.042 |
Neg logit contributions | |
V | -0.056 |
The | -0.053 |
Many | -0.052 |
Here | -0.052 |
Higher | -0.052 |
Token | play |
Token ID | 711 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
V | +0.018 |
Here | +0.017 |
Higher | +0.017 |
Grab | +0.016 |
First | +0.016 |
Neg logit contributions | |
than | -0.025 |
ier | -0.024 |
ies | -0.023 |
- | -0.023 |
ge | -0.021 |
Token | with |
Token ID | 351 |
Feature activation | -4.23e-03 |
Pos logit contributions | |
V | +0.032 |
Here | +0.032 |
Higher | +0.031 |
Grab | +0.031 |
First | +0.030 |
Neg logit contributions | |
than | -0.016 |
- | -0.014 |
ies | -0.013 |
ier | -0.013 |
inches | -0.011 |
Token | her |
Token ID | 607 |
Feature activation | -1.26e-03 |
Pos logit contributions | |
V | +0.009 |
Here | +0.009 |
Higher | +0.009 |
Grab | +0.009 |
First | +0.008 |
Neg logit contributions | |
than | -0.008 |
- | -0.007 |
ies | -0.007 |
ier | -0.007 |
inches | -0.006 |
Token | toys |
Token ID | 14958 |
Feature activation | +4.95e-02 |
Pos logit contributions | |
V | +0.003 |
Here | +0.003 |
Higher | +0.002 |
The | +0.002 |
Look | +0.002 |
Neg logit contributions | |
than | -0.002 |
ier | -0.002 |
ge | -0.002 |
owing | -0.002 |
- | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
ies | +0.082 |
than | +0.080 |
ge | +0.078 |
inches | +0.076 |
owing | +0.076 |
Neg logit contributions | |
Here | -0.094 |
V | -0.091 |
The | -0.090 |
One | -0.088 |
First | -0.086 |
Token | One |
Token ID | 1881 |
Feature activation | -1.92e-02 |
Pos logit contributions | |
Grab | +0.160 |
Here | +0.145 |
Bang | +0.142 |
Ro | +0.141 |
Quick | +0.139 |
Neg logit contributions | |
than | -0.278 |
for | -0.258 |
anymore | -0.256 |
- | -0.254 |
ies | -0.246 |
Token | day |
Token ID | 1110 |
Feature activation | -6.27e-02 |
Pos logit contributions | |
V | +0.046 |
Here | +0.045 |
Higher | +0.044 |
Grab | +0.043 |
Many | +0.041 |
Neg logit contributions | |
than | -0.031 |
ier | -0.029 |
ies | -0.027 |
- | -0.026 |
inches | -0.025 |
Token | , |
Token ID | 11 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
V | +0.160 |
Here | +0.157 |
Higher | +0.155 |
Grab | +0.151 |
First | +0.147 |
Neg logit contributions | |
than | -0.090 |
ier | -0.078 |
ies | -0.076 |
- | -0.075 |
inches | -0.068 |
Token | she |
Token ID | 673 |
Feature activation | -2.65e-02 |
Pos logit contributions | |
V | +0.235 |
Here | +0.229 |
Grab | +0.223 |
Higher | +0.222 |
Ro | +0.218 |
Neg logit contributions | |
than | -0.172 |
- | -0.161 |
ies | -0.158 |
ever | -0.137 |
for | -0.133 |
Token | had |
Token ID | 550 |
Feature activation | -2.72e-02 |
Pos logit contributions | |
V | +0.050 |
Here | +0.050 |
Higher | +0.050 |
Grab | +0.048 |
First | +0.046 |
Neg logit contributions | |
than | -0.056 |
- | -0.053 |
ies | -0.052 |
ier | -0.048 |
ever | -0.046 |
Token | Let |
Token ID | 5756 |
Feature activation | -2.07e-02 |
Pos logit contributions | |
Higher | +0.071 |
First | +0.069 |
Everyone | +0.062 |
His | +0.061 |
Here | +0.060 |
Neg logit contributions | |
than | -0.179 |
ever | -0.159 |
- | -0.151 |
y | -0.147 |
inches | -0.140 |
Token | 's |
Token ID | 338 |
Feature activation | -2.58e-02 |
Pos logit contributions | |
Higher | +0.059 |
Here | +0.059 |
V | +0.058 |
Grab | +0.056 |
Ro | +0.053 |
Neg logit contributions | |
- | -0.026 |
than | -0.022 |
ies | -0.021 |
! | -0.018 |
ever | -0.017 |
Token | climb |
Token ID | 12080 |
Feature activation | +3.55e-02 |
Pos logit contributions | |
V | +0.051 |
Here | +0.051 |
Grab | +0.048 |
Higher | +0.048 |
Look | +0.045 |
Neg logit contributions | |
than | -0.052 |
- | -0.047 |
ies | -0.046 |
ier | -0.044 |
ever | -0.043 |
Token | the |
Token ID | 262 |
Feature activation | +3.73e-03 |
Pos logit contributions | |
than | +0.074 |
ier | +0.069 |
ies | +0.068 |
- | +0.065 |
owing | +0.062 |
Neg logit contributions | |
V | -0.067 |
Higher | -0.065 |
Here | -0.065 |
Grab | -0.061 |
First | -0.061 |
Token | tree |
Token ID | 5509 |
Feature activation | +5.53e-02 |
Pos logit contributions | |
than | +0.009 |
ies | +0.008 |
ier | +0.008 |
- | +0.008 |
ge | +0.008 |
Neg logit contributions | |
V | -0.006 |
Higher | -0.006 |
Here | -0.006 |
Grab | -0.005 |
First | -0.005 |
Token | !" |
Token ID | 2474 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
ier | +0.111 |
than | +0.109 |
ies | +0.102 |
- | +0.098 |
owing | +0.096 |
Neg logit contributions | |
First | -0.096 |
The | -0.094 |
V | -0.094 |
Higher | -0.090 |
One | -0.085 |
Token | Sam |
Token ID | 3409 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
V | +0.173 |
Here | +0.168 |
Higher | +0.160 |
Grab | +0.150 |
Ro | +0.148 |
Neg logit contributions | |
than | -0.234 |
- | -0.226 |
ies | -0.223 |
! | -0.198 |
too | -0.195 |
Token | says |
Token ID | 1139 |
Feature activation | -3.01e-02 |
Pos logit contributions | |
Here | +0.040 |
V | +0.040 |
Higher | +0.039 |
Grab | +0.037 |
Ro | +0.035 |
Neg logit contributions | |
than | -0.050 |
- | -0.046 |
ies | -0.045 |
ier | -0.043 |
inches | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | -9.85e-02 |
Pos logit contributions | |
V | +0.075 |
The | +0.073 |
Higher | +0.073 |
First | +0.072 |
Here | +0.072 |
Neg logit contributions | |
than | -0.040 |
ier | -0.034 |
ies | -0.034 |
inches | -0.031 |
- | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | -8.98e-02 |
Pos logit contributions | |
Here | +0.171 |
Higher | +0.158 |
V | +0.150 |
Grab | +0.147 |
Ta | +0.133 |
Neg logit contributions | |
than | -0.229 |
- | -0.225 |
too | -0.204 |
ies | -0.198 |
! | -0.190 |
Token |
|
Token ID | 198 |
Feature activation | -8.37e-02 |
Pos logit contributions | |
Here | +0.164 |
Higher | +0.160 |
Grab | +0.150 |
V | +0.138 |
Many | +0.132 |
Neg logit contributions | |
- | -0.201 |
than | -0.194 |
ies | -0.173 |
! | -0.170 |
ever | -0.162 |
Token | liked |
Token ID | 8288 |
Feature activation | +6.36e-03 |
Pos logit contributions | |
V | +0.034 |
Here | +0.033 |
Higher | +0.032 |
Grab | +0.031 |
First | +0.030 |
Neg logit contributions | |
than | -0.044 |
- | -0.041 |
ies | -0.040 |
ier | -0.040 |
inches | -0.036 |
Token | to |
Token ID | 284 |
Feature activation | -3.54e-02 |
Pos logit contributions | |
ier | +0.010 |
than | +0.010 |
ies | +0.009 |
- | +0.009 |
ge | +0.008 |
Neg logit contributions | |
Here | -0.015 |
V | -0.014 |
Higher | -0.014 |
The | -0.014 |
First | -0.014 |
Token | explore |
Token ID | 7301 |
Feature activation | -2.98e-03 |
Pos logit contributions | |
V | +0.063 |
Here | +0.063 |
Grab | +0.059 |
Higher | +0.059 |
Look | +0.055 |
Neg logit contributions | |
than | -0.079 |
- | -0.072 |
ies | -0.071 |
ier | -0.070 |
ever | -0.065 |
Token | . |
Token ID | 13 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
Here | +0.005 |
Higher | +0.005 |
V | +0.005 |
The | +0.005 |
Grab | +0.005 |
Neg logit contributions | |
ier | -0.006 |
than | -0.006 |
ies | -0.006 |
owing | -0.005 |
inches | -0.005 |
Token | One |
Token ID | 1881 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
Grab | +0.128 |
Here | +0.128 |
V | +0.123 |
Ro | +0.113 |
Bang | +0.112 |
Neg logit contributions | |
than | -0.296 |
- | -0.260 |
anymore | -0.254 |
ies | -0.253 |
for | -0.248 |
Token | day |
Token ID | 1110 |
Feature activation | -7.11e-02 |
Pos logit contributions | |
V | +0.023 |
Here | +0.023 |
Higher | +0.023 |
The | +0.022 |
First | +0.022 |
Neg logit contributions | |
than | -0.019 |
ier | -0.018 |
ge | -0.017 |
ies | -0.016 |
inches | -0.016 |
Token | , |
Token ID | 11 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
V | +0.195 |
Here | +0.192 |
Higher | +0.186 |
Grab | +0.185 |
Ro | +0.179 |
Neg logit contributions | |
than | -0.092 |
- | -0.080 |
ies | -0.079 |
ier | -0.075 |
ever | -0.063 |
Token | she |
Token ID | 673 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
V | +0.206 |
Here | +0.202 |
Higher | +0.194 |
Grab | +0.192 |
Ro | +0.183 |
Neg logit contributions | |
than | -0.202 |
- | -0.182 |
ies | -0.182 |
ier | -0.178 |
inches | -0.161 |
Token | saw |
Token ID | 2497 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
V | +0.030 |
Here | +0.030 |
Higher | +0.029 |
Grab | +0.028 |
First | +0.027 |
Neg logit contributions | |
than | -0.036 |
- | -0.033 |
ies | -0.033 |
ier | -0.032 |
ever | -0.029 |
Token | a |
Token ID | 257 |
Feature activation | +5.04e-03 |
Pos logit contributions | |
V | +0.065 |
Grab | +0.061 |
Here | +0.061 |
Higher | +0.060 |
Ro | +0.059 |
Neg logit contributions | |
than | -0.053 |
- | -0.052 |
ies | -0.049 |
ier | -0.048 |
ever | -0.042 |
Token | big |
Token ID | 1263 |
Feature activation | +4.15e-02 |
Pos logit contributions | |
than | +0.011 |
ier | +0.010 |
ies | +0.010 |
owing | +0.009 |
- | +0.009 |
Neg logit contributions | |
V | -0.009 |
Here | -0.009 |
Higher | -0.008 |
Grab | -0.008 |
Look | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -4.95e-02 |
Pos logit contributions | |
than | +0.072 |
ier | +0.068 |
ies | +0.067 |
- | +0.065 |
owing | +0.063 |
Neg logit contributions | |
V | -0.082 |
The | -0.077 |
Here | -0.076 |
Higher | -0.074 |
Ro | -0.073 |
Token | Can |
Token ID | 1680 |
Feature activation | -5.50e-02 |
Pos logit contributions | |
V | +0.088 |
Here | +0.082 |
Grab | +0.080 |
Bang | +0.079 |
Higher | +0.078 |
Neg logit contributions | |
than | -0.096 |
ever | -0.082 |
- | -0.075 |
indeed | -0.074 |
too | -0.073 |
Token | we |
Token ID | 356 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
V | +0.138 |
Here | +0.135 |
Grab | +0.131 |
Higher | +0.128 |
Look | +0.126 |
Neg logit contributions | |
than | -0.074 |
ever | -0.072 |
- | -0.071 |
ies | -0.067 |
anymore | -0.062 |
Token | play |
Token ID | 711 |
Feature activation | +2.66e-02 |
Pos logit contributions | |
V | +0.022 |
Here | +0.021 |
Higher | +0.020 |
Grab | +0.020 |
First | +0.019 |
Neg logit contributions | |
than | -0.029 |
ier | -0.027 |
ies | -0.027 |
- | -0.026 |
ge | -0.024 |
Token | more |
Token ID | 517 |
Feature activation | +4.89e-02 |
Pos logit contributions | |
than | +0.035 |
- | +0.029 |
ies | +0.029 |
ier | +0.029 |
inches | +0.024 |
Neg logit contributions | |
V | -0.072 |
Here | -0.071 |
Higher | -0.070 |
Grab | -0.069 |
First | -0.067 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
than | +0.097 |
ier | +0.094 |
ies | +0.091 |
ge | +0.088 |
owing | +0.088 |
Neg logit contributions | |
V | -0.094 |
Here | -0.090 |
Higher | -0.085 |
Ro | -0.084 |
First | -0.082 |
Token | Daddy |
Token ID | 26926 |
Feature activation | +3.72e-03 |
Pos logit contributions | |
V | +0.195 |
Grab | +0.188 |
Ro | +0.178 |
Here | +0.178 |
Bang | +0.164 |
Neg logit contributions | |
than | -0.267 |
- | -0.258 |
ies | -0.253 |
knowing | -0.241 |
for | -0.229 |
Token | said |
Token ID | 531 |
Feature activation | -3.54e-02 |
Pos logit contributions | |
than | +0.008 |
- | +0.008 |
ies | +0.008 |
ier | +0.007 |
inches | +0.007 |
Neg logit contributions | |
V | -0.007 |
Here | -0.007 |
Higher | -0.006 |
Grab | -0.006 |
First | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | -3.64e-02 |
Pos logit contributions | |
V | +0.087 |
Here | +0.085 |
Higher | +0.085 |
Grab | +0.082 |
First | +0.080 |
Neg logit contributions | |
than | -0.054 |
ies | -0.046 |
ier | -0.044 |
- | -0.043 |
inches | -0.041 |
Token | " |
Token ID | 366 |
Feature activation | -3.64e-02 |
Pos logit contributions | |
V | +0.050 |
Here | +0.049 |
Higher | +0.045 |
First | +0.045 |
Grab | +0.044 |
Neg logit contributions | |
than | -0.096 |
- | -0.090 |
ier | -0.086 |
ies | -0.085 |
ever | -0.082 |
Token | Of |
Token ID | 5189 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
V | +0.052 |
Higher | +0.051 |
First | +0.048 |
Here | +0.045 |
Grab | +0.045 |
Neg logit contributions | |
than | -0.092 |
- | -0.080 |
ever | -0.077 |
ier | -0.075 |
ies | -0.075 |