Token | having |
Token ID | 1719 |
Feature activation | -3.63e-02 |
Pos logit contributions | |
ished | +0.011 |
lee | +0.010 |
aming | +0.010 |
ady | +0.010 |
amed | +0.010 |
Neg logit contributions | |
hill | -0.007 |
road | -0.007 |
taxi | -0.007 |
ladder | -0.007 |
ride | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | +7.72e-02 |
Pos logit contributions | |
ished | +0.087 |
lee | +0.086 |
ady | +0.082 |
aming | +0.081 |
amed | +0.079 |
Neg logit contributions | |
hill | -0.062 |
road | -0.057 |
ladder | -0.056 |
taxi | -0.055 |
visit | -0.054 |
Token | party |
Token ID | 2151 |
Feature activation | +9.05e-02 |
Pos logit contributions | |
hill | +0.131 |
Zig | +0.126 |
road | +0.120 |
cliff | +0.118 |
ladder | +0.115 |
Neg logit contributions | |
ished | -0.159 |
lee | -0.157 |
ady | -0.155 |
aming | -0.143 |
ealing | -0.143 |
Token | in |
Token ID | 287 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
hill | +0.166 |
ladder | +0.152 |
road | +0.151 |
cliff | +0.149 |
taxi | +0.148 |
Neg logit contributions | |
lee | -0.200 |
ished | -0.200 |
aming | -0.190 |
ady | -0.188 |
amed | -0.183 |
Token | the |
Token ID | 262 |
Feature activation | +4.27e-02 |
Pos logit contributions | |
hill | +0.196 |
ladder | +0.181 |
road | +0.180 |
taxi | +0.177 |
cliff | +0.175 |
Neg logit contributions | |
lee | -0.254 |
ished | -0.249 |
aming | -0.243 |
ady | -0.241 |
amed | -0.230 |
Token | barn |
Token ID | 25203 |
Feature activation | +1.88e-01 |
Pos logit contributions | |
hill | +0.060 |
Zig | +0.055 |
ladder | +0.055 |
road | +0.053 |
cliff | +0.053 |
Neg logit contributions | |
lee | -0.105 |
ished | -0.104 |
aming | -0.102 |
ady | -0.101 |
amed | -0.098 |
Token | ?" |
Token ID | 1701 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
hill | +0.348 |
road | +0.320 |
ladder | +0.319 |
taxi | +0.318 |
cliff | +0.315 |
Neg logit contributions | |
lee | -0.402 |
ished | -0.399 |
aming | -0.378 |
ady | -0.376 |
amed | -0.364 |
Token |
|
Token ID | 198 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
hill | +0.019 |
road | +0.017 |
ladder | +0.017 |
cliff | +0.017 |
taxi | +0.017 |
Neg logit contributions | |
lee | -0.024 |
ished | -0.024 |
ady | -0.023 |
aming | -0.023 |
amed | -0.022 |
Token |
|
Token ID | 198 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
hill | +0.025 |
road | +0.023 |
ladder | +0.023 |
taxi | +0.022 |
cliff | +0.022 |
Neg logit contributions | |
lee | -0.041 |
ished | -0.041 |
aming | -0.039 |
ady | -0.039 |
amed | -0.038 |
Token | L |
Token ID | 43 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
hill | +0.031 |
road | +0.029 |
ladder | +0.029 |
cliff | +0.028 |
mountain | +0.028 |
Neg logit contributions | |
lee | -0.031 |
ished | -0.030 |
ady | -0.030 |
aming | -0.030 |
amed | -0.028 |
Token | ily |
Token ID | 813 |
Feature activation | -9.80e-02 |
Pos logit contributions | |
ished | +0.211 |
lee | +0.209 |
aming | +0.196 |
amed | +0.186 |
ealing | +0.183 |
Neg logit contributions | |
hill | -0.363 |
taxi | -0.354 |
ladder | -0.350 |
road | -0.345 |
slide | -0.343 |
Token | store |
Token ID | 3650 |
Feature activation | +7.23e-02 |
Pos logit contributions | |
hill | +0.196 |
ladder | +0.177 |
taxi | +0.174 |
road | +0.174 |
cliff | +0.173 |
Neg logit contributions | |
lee | -0.306 |
ished | -0.301 |
aming | -0.295 |
ady | -0.293 |
amed | -0.276 |
Token | the |
Token ID | 262 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
hill | +0.134 |
cliff | +0.122 |
road | +0.120 |
ladder | +0.119 |
taxi | +0.117 |
Neg logit contributions | |
ished | -0.160 |
lee | -0.156 |
aming | -0.149 |
ady | -0.146 |
amed | -0.142 |
Token | corn |
Token ID | 11676 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
hill | +0.019 |
Zig | +0.019 |
taxi | +0.017 |
ladder | +0.017 |
road | +0.017 |
Neg logit contributions | |
ished | -0.027 |
aming | -0.026 |
ady | -0.026 |
lee | -0.026 |
ins | -0.025 |
Token | in |
Token ID | 287 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
hill | +0.218 |
taxi | +0.212 |
subway | +0.211 |
road | +0.209 |
Zig | +0.208 |
Neg logit contributions | |
ished | -0.252 |
lee | -0.239 |
aming | -0.232 |
ady | -0.229 |
ins | -0.221 |
Token | the |
Token ID | 262 |
Feature activation | +4.41e-02 |
Pos logit contributions | |
hill | +0.197 |
ladder | +0.181 |
road | +0.180 |
taxi | +0.179 |
cliff | +0.177 |
Neg logit contributions | |
ished | -0.261 |
lee | -0.260 |
ady | -0.251 |
aming | -0.251 |
amed | -0.238 |
Token | barn |
Token ID | 25203 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
Zig | +0.061 |
hill | +0.058 |
taxi | +0.054 |
ladder | +0.053 |
ride | +0.053 |
Neg logit contributions | |
ished | -0.105 |
aming | -0.105 |
lee | -0.103 |
ady | -0.103 |
amed | -0.098 |
Token | , |
Token ID | 11 |
Feature activation | +5.50e-02 |
Pos logit contributions | |
hill | +0.335 |
road | +0.306 |
ladder | +0.305 |
taxi | +0.300 |
cliff | +0.299 |
Neg logit contributions | |
lee | -0.407 |
ished | -0.406 |
aming | -0.383 |
ady | -0.382 |
amed | -0.369 |
Token | so |
Token ID | 523 |
Feature activation | +4.74e-02 |
Pos logit contributions | |
hill | +0.110 |
cliff | +0.101 |
road | +0.101 |
ladder | +0.100 |
taxi | +0.099 |
Neg logit contributions | |
ished | -0.110 |
lee | -0.110 |
aming | -0.103 |
ady | -0.102 |
amed | -0.098 |
Token | it |
Token ID | 340 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
hill | +0.070 |
cliff | +0.062 |
taxi | +0.062 |
ladder | +0.062 |
road | +0.062 |
Neg logit contributions | |
lee | -0.118 |
ished | -0.117 |
aming | -0.111 |
ady | -0.110 |
amed | -0.107 |
Token | stays |
Token ID | 14768 |
Feature activation | -5.53e-02 |
Pos logit contributions | |
hill | +0.026 |
ladder | +0.024 |
road | +0.023 |
cliff | +0.023 |
taxi | +0.023 |
Neg logit contributions | |
lee | -0.040 |
ished | -0.039 |
aming | -0.038 |
ady | -0.038 |
amed | -0.036 |
Token | good |
Token ID | 922 |
Feature activation | -6.91e-02 |
Pos logit contributions | |
lee | +0.134 |
ished | +0.133 |
ady | +0.126 |
aming | +0.124 |
amed | +0.121 |
Neg logit contributions | |
hill | -0.092 |
ladder | -0.083 |
road | -0.083 |
taxi | -0.082 |
cliff | -0.080 |
Token | flashlight |
Token ID | 38371 |
Feature activation | +8.74e-02 |
Pos logit contributions | |
hill | +0.089 |
cliff | +0.081 |
road | +0.081 |
ladder | +0.078 |
taxi | +0.078 |
Neg logit contributions | |
lee | -0.135 |
ished | -0.132 |
ady | -0.131 |
aming | -0.128 |
amed | -0.124 |
Token | , |
Token ID | 11 |
Feature activation | +6.04e-02 |
Pos logit contributions | |
hill | +0.169 |
cliff | +0.156 |
road | +0.155 |
ladder | +0.154 |
taxi | +0.152 |
Neg logit contributions | |
lee | -0.177 |
ished | -0.175 |
ady | -0.164 |
aming | -0.163 |
amed | -0.158 |
Token | and |
Token ID | 290 |
Feature activation | +4.47e-02 |
Pos logit contributions | |
hill | +0.127 |
road | +0.117 |
ladder | +0.117 |
taxi | +0.115 |
cliff | +0.115 |
Neg logit contributions | |
lee | -0.119 |
ished | -0.118 |
aming | -0.111 |
ady | -0.110 |
amed | -0.106 |
Token | a |
Token ID | 257 |
Feature activation | +5.32e-02 |
Pos logit contributions | |
hill | +0.082 |
road | +0.075 |
ladder | +0.075 |
taxi | +0.074 |
cliff | +0.073 |
Neg logit contributions | |
ished | -0.099 |
lee | -0.099 |
aming | -0.094 |
ady | -0.094 |
amed | -0.090 |
Token | magn |
Token ID | 7842 |
Feature activation | +3.25e-02 |
Pos logit contributions | |
hill | +0.081 |
cliff | +0.073 |
road | +0.073 |
ladder | +0.071 |
taxi | +0.071 |
Neg logit contributions | |
lee | -0.125 |
ished | -0.121 |
ady | -0.121 |
aming | -0.117 |
amed | -0.114 |
Token | ifying |
Token ID | 4035 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
hill | +0.046 |
road | +0.040 |
ladder | +0.040 |
taxi | +0.039 |
cliff | +0.039 |
Neg logit contributions | |
lee | -0.087 |
ished | -0.087 |
ady | -0.082 |
aming | -0.082 |
amed | -0.080 |
Token | glass |
Token ID | 5405 |
Feature activation | +5.31e-02 |
Pos logit contributions | |
hill | +0.304 |
subway | +0.284 |
road | +0.281 |
Zig | +0.280 |
taxi | +0.279 |
Neg logit contributions | |
ished | -0.381 |
lee | -0.377 |
aming | -0.372 |
ady | -0.365 |
amed | -0.361 |
Token | . |
Token ID | 13 |
Feature activation | +6.63e-03 |
Pos logit contributions | |
hill | +0.101 |
ladder | +0.092 |
road | +0.092 |
taxi | +0.091 |
cliff | +0.090 |
Neg logit contributions | |
lee | -0.115 |
ished | -0.114 |
aming | -0.108 |
ady | -0.108 |
amed | -0.104 |
Token |
|
Token ID | 198 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
hill | +0.012 |
road | +0.011 |
ladder | +0.011 |
cliff | +0.011 |
taxi | +0.011 |
Neg logit contributions | |
lee | -0.015 |
ished | -0.014 |
ady | -0.014 |
aming | -0.014 |
amed | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | +2.07e-02 |
Pos logit contributions | |
hill | +0.032 |
ladder | +0.029 |
road | +0.029 |
taxi | +0.028 |
cliff | +0.027 |
Neg logit contributions | |
ished | -0.065 |
lee | -0.064 |
aming | -0.061 |
ady | -0.060 |
amed | -0.059 |
Token | As |
Token ID | 1722 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
hill | +0.041 |
road | +0.037 |
ladder | +0.037 |
taxi | +0.037 |
cliff | +0.037 |
Neg logit contributions | |
lee | -0.044 |
ished | -0.044 |
aming | -0.041 |
ady | -0.040 |
amed | -0.039 |
Token | pile |
Token ID | 14540 |
Feature activation | +9.39e-02 |
Pos logit contributions | |
hill | +0.038 |
road | +0.035 |
Zig | +0.035 |
visit | +0.035 |
ladder | +0.035 |
Neg logit contributions | |
lee | -0.062 |
ished | -0.059 |
aming | -0.057 |
ady | -0.057 |
pled | -0.055 |
Token | of |
Token ID | 286 |
Feature activation | +7.38e-02 |
Pos logit contributions | |
hill | +0.161 |
ladder | +0.147 |
road | +0.144 |
taxi | +0.142 |
cliff | +0.141 |
Neg logit contributions | |
lee | -0.226 |
ished | -0.222 |
ady | -0.210 |
aming | -0.205 |
ealing | -0.203 |
Token | hay |
Token ID | 27678 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
hill | +0.134 |
road | +0.122 |
ladder | +0.121 |
taxi | +0.119 |
cliff | +0.119 |
Neg logit contributions | |
lee | -0.166 |
ished | -0.166 |
aming | -0.157 |
ady | -0.156 |
amed | -0.151 |
Token | in |
Token ID | 287 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
hill | +0.261 |
taxi | +0.242 |
road | +0.241 |
ladder | +0.239 |
cliff | +0.236 |
Neg logit contributions | |
ished | -0.325 |
lee | -0.323 |
aming | -0.311 |
ady | -0.305 |
amed | -0.298 |
Token | the |
Token ID | 262 |
Feature activation | +3.82e-02 |
Pos logit contributions | |
hill | +0.183 |
ladder | +0.167 |
road | +0.166 |
taxi | +0.166 |
cliff | +0.163 |
Neg logit contributions | |
lee | -0.245 |
ished | -0.243 |
aming | -0.234 |
ady | -0.233 |
amed | -0.225 |
Token | barn |
Token ID | 25203 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
hill | +0.058 |
cliff | +0.053 |
ladder | +0.053 |
risk | +0.053 |
taxi | +0.052 |
Neg logit contributions | |
ished | -0.089 |
ady | -0.089 |
lee | -0.088 |
aming | -0.088 |
amed | -0.086 |
Token | . |
Token ID | 13 |
Feature activation | +4.09e-04 |
Pos logit contributions | |
hill | +0.347 |
road | +0.319 |
ladder | +0.318 |
taxi | +0.310 |
cliff | +0.310 |
Neg logit contributions | |
ished | -0.383 |
lee | -0.383 |
aming | -0.359 |
ady | -0.359 |
amed | -0.346 |
Token | She |
Token ID | 1375 |
Feature activation | +7.53e-03 |
Pos logit contributions | |
hill | +0.001 |
taxi | +0.001 |
road | +0.001 |
cliff | +0.001 |
ladder | +0.001 |
Neg logit contributions | |
lee | -0.001 |
ished | -0.001 |
ady | -0.001 |
aming | -0.001 |
amed | -0.001 |
Token | was |
Token ID | 373 |
Feature activation | -4.94e-02 |
Pos logit contributions | |
hill | +0.014 |
ladder | +0.013 |
road | +0.012 |
walk | +0.012 |
taxi | +0.012 |
Neg logit contributions | |
ished | -0.017 |
aming | -0.016 |
lee | -0.016 |
ady | -0.016 |
ealing | -0.016 |
Token | very |
Token ID | 845 |
Feature activation | +5.10e-03 |
Pos logit contributions | |
ished | +0.104 |
lee | +0.104 |
ady | +0.095 |
ins | +0.094 |
aming | +0.093 |
Neg logit contributions | |
ladder | -0.095 |
hill | -0.094 |
road | -0.091 |
taxi | -0.090 |
cliff | -0.087 |
Token | hungry |
Token ID | 14720 |
Feature activation | -6.82e-02 |
Pos logit contributions | |
ladder | +0.011 |
hill | +0.011 |
road | +0.010 |
taxi | +0.010 |
cliff | +0.010 |
Neg logit contributions | |
ished | -0.010 |
lee | -0.010 |
ice | -0.010 |
pled | -0.009 |
ady | -0.009 |
Token | sort |
Token ID | 3297 |
Feature activation | -3.06e-02 |
Pos logit contributions | |
hill | +0.184 |
road | +0.167 |
ladder | +0.162 |
taxi | +0.161 |
ride | +0.160 |
Neg logit contributions | |
lee | -0.319 |
ished | -0.318 |
aming | -0.302 |
ady | -0.300 |
amed | -0.295 |
Token | the |
Token ID | 262 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
lee | +0.075 |
ished | +0.073 |
aming | +0.071 |
ady | +0.071 |
amed | +0.069 |
Neg logit contributions | |
hill | -0.050 |
ladder | -0.045 |
road | -0.045 |
taxi | -0.045 |
cliff | -0.043 |
Token | hay |
Token ID | 27678 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
hill | +0.025 |
road | +0.023 |
ladder | +0.023 |
taxi | +0.022 |
mountain | +0.022 |
Neg logit contributions | |
lee | -0.036 |
ished | -0.035 |
aming | -0.034 |
amed | -0.033 |
ady | -0.033 |
Token | in |
Token ID | 287 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
hill | +0.291 |
road | +0.268 |
taxi | +0.267 |
ladder | +0.266 |
cliff | +0.264 |
Neg logit contributions | |
ished | -0.328 |
lee | -0.324 |
aming | -0.304 |
ady | -0.303 |
amed | -0.294 |
Token | the |
Token ID | 262 |
Feature activation | +3.94e-02 |
Pos logit contributions | |
hill | +0.173 |
road | +0.155 |
ladder | +0.154 |
cliff | +0.145 |
mountain | +0.142 |
Neg logit contributions | |
lee | -0.259 |
ished | -0.257 |
aming | -0.240 |
amed | -0.240 |
ady | -0.235 |
Token | barn |
Token ID | 25203 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
hill | +0.055 |
road | +0.049 |
ladder | +0.049 |
taxi | +0.048 |
cliff | +0.048 |
Neg logit contributions | |
lee | -0.105 |
ished | -0.105 |
ady | -0.100 |
aming | -0.100 |
amed | -0.097 |
Token | . |
Token ID | 13 |
Feature activation | +1.22e-02 |
Pos logit contributions | |
hill | +0.325 |
road | +0.296 |
ladder | +0.296 |
cliff | +0.283 |
taxi | +0.282 |
Neg logit contributions | |
lee | -0.407 |
ished | -0.402 |
aming | -0.387 |
ady | -0.385 |
amed | -0.372 |
Token | That |
Token ID | 1320 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
hill | +0.022 |
road | +0.021 |
ladder | +0.020 |
taxi | +0.020 |
visit | +0.020 |
Neg logit contributions | |
lee | -0.027 |
ished | -0.027 |
aming | -0.026 |
ady | -0.025 |
amed | -0.025 |
Token | is |
Token ID | 318 |
Feature activation | -2.89e-02 |
Pos logit contributions | |
ished | +0.021 |
lee | +0.020 |
ady | +0.019 |
aming | +0.019 |
amed | +0.018 |
Neg logit contributions | |
hill | -0.021 |
road | -0.020 |
ladder | -0.020 |
taxi | -0.020 |
cliff | -0.020 |
Token | your |
Token ID | 534 |
Feature activation | -3.12e-02 |
Pos logit contributions | |
ished | +0.058 |
lee | +0.057 |
aming | +0.054 |
ady | +0.053 |
amed | +0.052 |
Neg logit contributions | |
hill | -0.060 |
road | -0.055 |
ladder | -0.054 |
cliff | -0.054 |
taxi | -0.054 |
Token | chore |
Token ID | 30569 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
ished | +0.064 |
lee | +0.063 |
ady | +0.061 |
aming | +0.056 |
ins | +0.056 |
Neg logit contributions | |
hill | -0.056 |
cliff | -0.053 |
taxi | -0.050 |
mountain | -0.050 |
road | -0.050 |
Token | the |
Token ID | 262 |
Feature activation | +6.07e-03 |
Pos logit contributions | |
hill | +0.016 |
road | +0.015 |
ladder | +0.015 |
taxi | +0.015 |
cliff | +0.015 |
Neg logit contributions | |
ished | -0.020 |
lee | -0.019 |
ady | -0.019 |
aming | -0.019 |
amed | -0.018 |
Token | other |
Token ID | 584 |
Feature activation | +6.55e-03 |
Pos logit contributions | |
hill | +0.011 |
ladder | +0.010 |
road | +0.010 |
cliff | +0.010 |
taxi | +0.010 |
Neg logit contributions | |
lee | -0.013 |
ished | -0.013 |
ady | -0.013 |
aming | -0.013 |
amed | -0.012 |
Token | animals |
Token ID | 4695 |
Feature activation | +6.98e-02 |
Pos logit contributions | |
hill | +0.012 |
ladder | +0.011 |
road | +0.011 |
cliff | +0.011 |
taxi | +0.011 |
Neg logit contributions | |
lee | -0.015 |
ished | -0.014 |
ady | -0.014 |
aming | -0.014 |
amed | -0.013 |
Token | in |
Token ID | 287 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
hill | +0.129 |
road | +0.119 |
ladder | +0.118 |
taxi | +0.117 |
cliff | +0.116 |
Neg logit contributions | |
lee | -0.153 |
ished | -0.152 |
aming | -0.143 |
ady | -0.143 |
amed | -0.137 |
Token | the |
Token ID | 262 |
Feature activation | +3.36e-02 |
Pos logit contributions | |
hill | +0.189 |
ladder | +0.175 |
road | +0.173 |
cliff | +0.169 |
taxi | +0.168 |
Neg logit contributions | |
lee | -0.229 |
ished | -0.214 |
ady | -0.213 |
aming | -0.212 |
amed | -0.203 |
Token | barn |
Token ID | 25203 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
hill | +0.041 |
ladder | +0.038 |
taxi | +0.036 |
road | +0.036 |
cliff | +0.035 |
Neg logit contributions | |
lee | -0.092 |
ished | -0.090 |
aming | -0.088 |
ady | -0.088 |
amed | -0.086 |
Token | . |
Token ID | 13 |
Feature activation | +4.05e-03 |
Pos logit contributions | |
hill | +0.336 |
road | +0.308 |
ladder | +0.307 |
taxi | +0.301 |
cliff | +0.298 |
Neg logit contributions | |
ished | -0.385 |
lee | -0.383 |
aming | -0.362 |
ady | -0.361 |
amed | -0.347 |
Token |
|
Token ID | 198 |
Feature activation | +2.42e-02 |
Pos logit contributions | |
hill | +0.008 |
road | +0.007 |
cliff | +0.007 |
ladder | +0.007 |
taxi | +0.007 |
Neg logit contributions | |
lee | -0.009 |
ished | -0.009 |
ady | -0.008 |
aming | -0.008 |
amed | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | +2.21e-02 |
Pos logit contributions | |
hill | +0.039 |
ladder | +0.035 |
road | +0.035 |
taxi | +0.034 |
cliff | +0.034 |
Neg logit contributions | |
ished | -0.060 |
lee | -0.059 |
aming | -0.056 |
ady | -0.056 |
amed | -0.054 |
Token | The |
Token ID | 464 |
Feature activation | -2.50e-02 |
Pos logit contributions | |
hill | +0.038 |
road | +0.035 |
ladder | +0.035 |
taxi | +0.034 |
cliff | +0.034 |
Neg logit contributions | |
lee | -0.051 |
ished | -0.051 |
aming | -0.049 |
ady | -0.048 |
amed | -0.047 |
Token | end |
Token ID | 886 |
Feature activation | +7.91e-03 |
Pos logit contributions | |
ished | +0.056 |
lee | +0.056 |
ady | +0.052 |
aming | +0.051 |
amed | +0.050 |
Neg logit contributions | |
hill | -0.049 |
road | -0.045 |
ladder | -0.044 |
taxi | -0.044 |
cliff | -0.043 |
Token | He |
Token ID | 679 |
Feature activation | -6.40e-03 |
Pos logit contributions | |
lee | +0.028 |
ady | +0.026 |
ished | +0.025 |
amed | +0.025 |
brothers | +0.023 |
Neg logit contributions | |
hill | -0.033 |
cliff | -0.033 |
risk | -0.031 |
taxi | -0.031 |
mountain | -0.030 |
Token | took |
Token ID | 1718 |
Feature activation | -8.84e-02 |
Pos logit contributions | |
lee | +0.014 |
ished | +0.014 |
ady | +0.014 |
aming | +0.014 |
ealing | +0.013 |
Neg logit contributions | |
hill | -0.011 |
ladder | -0.011 |
ride | -0.011 |
road | -0.010 |
mountain | -0.010 |
Token | out |
Token ID | 503 |
Feature activation | +6.77e-02 |
Pos logit contributions | |
lee | +0.203 |
ished | +0.202 |
aming | +0.191 |
ady | +0.191 |
amed | +0.184 |
Neg logit contributions | |
hill | -0.157 |
road | -0.143 |
ladder | -0.143 |
taxi | -0.140 |
cliff | -0.139 |
Token | a |
Token ID | 257 |
Feature activation | +7.17e-02 |
Pos logit contributions | |
hill | +0.132 |
road | +0.120 |
cliff | +0.119 |
ladder | +0.119 |
taxi | +0.116 |
Neg logit contributions | |
lee | -0.146 |
ished | -0.144 |
ady | -0.136 |
aming | -0.135 |
amed | -0.131 |
Token | magn |
Token ID | 7842 |
Feature activation | +3.58e-02 |
Pos logit contributions | |
Zig | +0.115 |
hill | +0.115 |
road | +0.110 |
Raven | +0.108 |
cliff | +0.106 |
Neg logit contributions | |
ady | -0.149 |
lee | -0.147 |
ished | -0.143 |
aming | -0.136 |
amed | -0.136 |
Token | ifying |
Token ID | 4035 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
hill | +0.047 |
road | +0.042 |
ladder | +0.042 |
cliff | +0.040 |
taxi | +0.040 |
Neg logit contributions | |
ished | -0.099 |
lee | -0.099 |
ady | -0.094 |
aming | -0.094 |
amed | -0.092 |
Token | glass |
Token ID | 5405 |
Feature activation | +5.62e-02 |
Pos logit contributions | |
hill | +0.308 |
Zig | +0.292 |
subway | +0.288 |
walk | +0.285 |
taxi | +0.285 |
Neg logit contributions | |
lee | -0.341 |
ished | -0.337 |
aming | -0.327 |
ady | -0.322 |
amed | -0.320 |
Token | and |
Token ID | 290 |
Feature activation | +4.48e-02 |
Pos logit contributions | |
hill | +0.115 |
road | +0.107 |
ladder | +0.106 |
taxi | +0.105 |
cliff | +0.105 |
Neg logit contributions | |
ished | -0.112 |
lee | -0.112 |
ady | -0.104 |
aming | -0.104 |
amed | -0.100 |
Token | a |
Token ID | 257 |
Feature activation | +6.79e-02 |
Pos logit contributions | |
hill | +0.093 |
ladder | +0.087 |
road | +0.086 |
cliff | +0.084 |
mountain | +0.084 |
Neg logit contributions | |
lee | -0.089 |
ished | -0.089 |
ady | -0.084 |
aming | -0.084 |
amed | -0.079 |
Token | notebook |
Token ID | 20922 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
hill | +0.115 |
Zig | +0.108 |
road | +0.107 |
cliff | +0.104 |
taxi | +0.104 |
Neg logit contributions | |
lee | -0.143 |
ady | -0.139 |
ished | -0.137 |
aming | -0.134 |
amed | -0.131 |
Token | and |
Token ID | 290 |
Feature activation | +4.94e-02 |
Pos logit contributions | |
hill | +0.191 |
road | +0.175 |
ladder | +0.174 |
taxi | +0.171 |
cliff | +0.171 |
Neg logit contributions | |
lee | -0.222 |
ished | -0.221 |
aming | -0.209 |
ady | -0.209 |
amed | -0.201 |
Token | is |
Token ID | 318 |
Feature activation | -1.76e-02 |
Pos logit contributions | |
hill | +0.186 |
road | +0.172 |
ladder | +0.171 |
taxi | +0.170 |
cliff | +0.169 |
Neg logit contributions | |
lee | -0.166 |
ished | -0.166 |
ady | -0.155 |
aming | -0.154 |
amed | -0.147 |
Token | some |
Token ID | 617 |
Feature activation | +4.52e-02 |
Pos logit contributions | |
lee | +0.039 |
ished | +0.039 |
ady | +0.036 |
aming | +0.036 |
amed | +0.035 |
Neg logit contributions | |
hill | -0.033 |
road | -0.030 |
ladder | -0.030 |
taxi | -0.029 |
cliff | -0.029 |
Token | wool |
Token ID | 25749 |
Feature activation | +9.91e-02 |
Pos logit contributions | |
hill | +0.085 |
road | +0.078 |
ladder | +0.078 |
taxi | +0.077 |
cliff | +0.076 |
Neg logit contributions | |
lee | -0.098 |
ished | -0.098 |
aming | -0.092 |
ady | -0.092 |
amed | -0.089 |
Token | in |
Token ID | 287 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
hill | +0.176 |
road | +0.162 |
ladder | +0.161 |
taxi | +0.158 |
cliff | +0.158 |
Neg logit contributions | |
ished | -0.222 |
lee | -0.221 |
aming | -0.209 |
ady | -0.208 |
amed | -0.200 |
Token | the |
Token ID | 262 |
Feature activation | +3.51e-02 |
Pos logit contributions | |
hill | +0.189 |
ladder | +0.173 |
road | +0.172 |
taxi | +0.169 |
cliff | +0.169 |
Neg logit contributions | |
lee | -0.248 |
ished | -0.245 |
ady | -0.237 |
aming | -0.235 |
amed | -0.223 |
Token | barn |
Token ID | 25203 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
hill | +0.048 |
ladder | +0.043 |
cliff | +0.041 |
road | +0.041 |
taxi | +0.041 |
Neg logit contributions | |
lee | -0.093 |
ished | -0.092 |
ady | -0.090 |
aming | -0.090 |
amed | -0.086 |
Token | ". |
Token ID | 1911 |
Feature activation | -5.23e-04 |
Pos logit contributions | |
hill | +0.337 |
road | +0.309 |
ladder | +0.308 |
taxi | +0.302 |
cliff | +0.301 |
Neg logit contributions | |
lee | -0.381 |
ished | -0.379 |
aming | -0.357 |
ady | -0.357 |
amed | -0.345 |
Token | |
Token ID | 220 |
Feature activation | -4.45e-02 |
Pos logit contributions | |
lee | +0.001 |
ished | +0.001 |
ady | +0.001 |
aming | +0.001 |
amed | +0.001 |
Neg logit contributions | |
hill | -0.001 |
road | -0.001 |
ladder | -0.001 |
cliff | -0.001 |
taxi | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
ished | +0.123 |
lee | +0.122 |
aming | +0.119 |
ady | +0.115 |
amed | +0.115 |
Neg logit contributions | |
hill | -0.057 |
ladder | -0.052 |
road | -0.050 |
taxi | -0.048 |
visit | -0.047 |
Token |
|
Token ID | 198 |
Feature activation | +1.86e-02 |
Pos logit contributions | |
hill | +0.033 |
road | +0.030 |
ladder | +0.030 |
cliff | +0.030 |
taxi | +0.029 |
Neg logit contributions | |
lee | -0.042 |
ished | -0.041 |
aming | -0.040 |
ady | -0.040 |
amed | -0.039 |
Token | Jess |
Token ID | 34648 |
Feature activation | -5.83e-02 |
Pos logit contributions | |
hill | +0.037 |
road | +0.034 |
ladder | +0.034 |
cliff | +0.033 |
taxi | +0.033 |
Neg logit contributions | |
lee | -0.038 |
ished | -0.038 |
aming | -0.038 |
ady | -0.037 |
amed | -0.036 |
Token | a |
Token ID | 257 |
Feature activation | +6.16e-02 |
Pos logit contributions | |
lee | +0.071 |
ished | +0.070 |
aming | +0.067 |
ady | +0.066 |
amed | +0.064 |
Neg logit contributions | |
hill | -0.057 |
road | -0.052 |
ladder | -0.052 |
taxi | -0.051 |
cliff | -0.051 |
Token | big |
Token ID | 1263 |
Feature activation | +2.66e-02 |
Pos logit contributions | |
Zig | +0.110 |
hill | +0.110 |
risk | +0.107 |
cliff | +0.101 |
road | +0.101 |
Neg logit contributions | |
lee | -0.113 |
ady | -0.109 |
aming | -0.106 |
ins | -0.101 |
ished | -0.100 |
Token | cross |
Token ID | 3272 |
Feature activation | +3.87e-02 |
Pos logit contributions | |
hill | +0.039 |
Zig | +0.036 |
ladder | +0.035 |
road | +0.035 |
risk | +0.035 |
Neg logit contributions | |
lee | -0.063 |
ished | -0.060 |
ady | -0.058 |
aming | -0.058 |
pled | -0.056 |
Token | on |
Token ID | 319 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
hill | +0.068 |
Zig | +0.063 |
taxi | +0.063 |
ladder | +0.062 |
visit | +0.062 |
Neg logit contributions | |
ished | -0.079 |
aming | -0.078 |
lee | -0.077 |
ady | -0.077 |
amed | -0.073 |
Token | the |
Token ID | 262 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
hill | +0.255 |
ladder | +0.240 |
risk | +0.240 |
taxi | +0.236 |
bike | +0.233 |
Neg logit contributions | |
lee | -0.215 |
aming | -0.201 |
ice | -0.196 |
ady | -0.193 |
ished | -0.190 |
Token | barn |
Token ID | 25203 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
hill | +0.045 |
taxi | +0.042 |
ladder | +0.041 |
Zig | +0.040 |
visit | +0.040 |
Neg logit contributions | |
lee | -0.074 |
ished | -0.072 |
aming | -0.071 |
ady | -0.070 |
ice | -0.068 |
Token | door |
Token ID | 3420 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
hill | +0.360 |
taxi | +0.332 |
ladder | +0.332 |
road | +0.330 |
cliff | +0.330 |
Neg logit contributions | |
lee | -0.347 |
ished | -0.343 |
ady | -0.323 |
aming | -0.323 |
amed | -0.310 |
Token | . |
Token ID | 13 |
Feature activation | +3.78e-04 |
Pos logit contributions | |
hill | +0.222 |
road | +0.204 |
ladder | +0.203 |
taxi | +0.201 |
cliff | +0.200 |
Neg logit contributions | |
lee | -0.225 |
ished | -0.224 |
aming | -0.211 |
ady | -0.210 |
amed | -0.202 |
Token | Lucy |
Token ID | 22162 |
Feature activation | -7.65e-02 |
Pos logit contributions | |
hill | +0.001 |
cliff | +0.001 |
risk | +0.001 |
mountain | +0.001 |
road | +0.001 |
Neg logit contributions | |
lee | -0.001 |
ished | -0.001 |
ady | -0.001 |
aming | -0.001 |
amed | -0.001 |
Token | went |
Token ID | 1816 |
Feature activation | -3.99e-02 |
Pos logit contributions | |
ished | +0.163 |
lee | +0.159 |
ady | +0.154 |
aming | +0.153 |
amed | +0.146 |
Neg logit contributions | |
hill | -0.151 |
ladder | -0.137 |
road | -0.135 |
ride | -0.132 |
taxi | -0.132 |
Token | to |
Token ID | 284 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
ished | +0.083 |
lee | +0.081 |
aming | +0.080 |
amed | +0.080 |
ady | +0.078 |
Neg logit contributions | |
hill | -0.071 |
cliff | -0.066 |
mountain | -0.065 |
road | -0.064 |
subway | -0.062 |
Token | She |
Token ID | 1375 |
Feature activation | -4.50e-03 |
Pos logit contributions | |
ished | +0.023 |
lee | +0.023 |
aming | +0.022 |
ady | +0.022 |
amed | +0.021 |
Neg logit contributions | |
hill | -0.016 |
ladder | -0.015 |
road | -0.015 |
taxi | -0.014 |
cliff | -0.014 |
Token | takes |
Token ID | 2753 |
Feature activation | -8.90e-02 |
Pos logit contributions | |
aming | +0.010 |
lee | +0.010 |
ady | +0.009 |
ished | +0.009 |
amed | +0.009 |
Neg logit contributions | |
hill | -0.009 |
ladder | -0.008 |
mountain | -0.008 |
walk | -0.008 |
ride | -0.008 |
Token | out |
Token ID | 503 |
Feature activation | +6.21e-02 |
Pos logit contributions | |
lee | +0.211 |
ished | +0.209 |
aming | +0.200 |
ady | +0.200 |
amed | +0.192 |
Neg logit contributions | |
hill | -0.149 |
ladder | -0.137 |
road | -0.136 |
taxi | -0.135 |
cliff | -0.132 |
Token | a |
Token ID | 257 |
Feature activation | +6.69e-02 |
Pos logit contributions | |
hill | +0.114 |
ladder | +0.105 |
road | +0.104 |
taxi | +0.103 |
cliff | +0.102 |
Neg logit contributions | |
ished | -0.138 |
lee | -0.138 |
aming | -0.131 |
ady | -0.130 |
amed | -0.126 |
Token | magn |
Token ID | 7842 |
Feature activation | +3.46e-02 |
Pos logit contributions | |
Zig | +0.113 |
hill | +0.108 |
road | +0.105 |
Raven | +0.104 |
cliff | +0.100 |
Neg logit contributions | |
ady | -0.133 |
ished | -0.130 |
lee | -0.130 |
pled | -0.123 |
aming | -0.121 |
Token | ifying |
Token ID | 4035 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
hill | +0.046 |
ladder | +0.040 |
road | +0.040 |
taxi | +0.039 |
cliff | +0.039 |
Neg logit contributions | |
lee | -0.095 |
ished | -0.095 |
aming | -0.090 |
ady | -0.090 |
amed | -0.088 |
Token | glass |
Token ID | 5405 |
Feature activation | +5.85e-02 |
Pos logit contributions | |
hill | +0.309 |
subway | +0.292 |
taxi | +0.288 |
road | +0.288 |
Zig | +0.285 |
Neg logit contributions | |
ished | -0.366 |
lee | -0.363 |
aming | -0.349 |
ady | -0.346 |
amed | -0.343 |
Token | . |
Token ID | 13 |
Feature activation | -3.94e-03 |
Pos logit contributions | |
hill | +0.115 |
ladder | +0.106 |
road | +0.104 |
taxi | +0.102 |
cliff | +0.102 |
Neg logit contributions | |
lee | -0.126 |
ished | -0.124 |
aming | -0.118 |
ady | -0.118 |
amed | -0.113 |
Token | She |
Token ID | 1375 |
Feature activation | +8.99e-03 |
Pos logit contributions | |
lee | +0.008 |
ished | +0.008 |
ady | +0.008 |
aming | +0.007 |
amed | +0.007 |
Neg logit contributions | |
hill | -0.008 |
road | -0.007 |
cliff | -0.007 |
taxi | -0.007 |
mountain | -0.007 |
Token | looks |
Token ID | 3073 |
Feature activation | -6.60e-02 |
Pos logit contributions | |
hill | +0.017 |
ladder | +0.017 |
mountain | +0.016 |
visit | +0.016 |
walk | +0.016 |
Neg logit contributions | |
aming | -0.019 |
lee | -0.019 |
ished | -0.018 |
ady | -0.018 |
amed | -0.018 |
Token | at |
Token ID | 379 |
Feature activation | +5.34e-02 |
Pos logit contributions | |
lee | +0.148 |
ady | +0.147 |
ished | +0.145 |
aming | +0.144 |
ins | +0.139 |
Neg logit contributions | |
hill | -0.110 |
ladder | -0.108 |
taxi | -0.104 |
visit | -0.101 |
mountain | -0.100 |
Token | are |
Token ID | 389 |
Feature activation | -1.13e-03 |
Pos logit contributions | |
hill | +0.047 |
road | +0.043 |
ladder | +0.043 |
taxi | +0.043 |
cliff | +0.042 |
Neg logit contributions | |
ished | -0.044 |
lee | -0.044 |
ady | -0.042 |
aming | -0.041 |
amed | -0.040 |
Token | you |
Token ID | 345 |
Feature activation | -3.79e-02 |
Pos logit contributions | |
lee | +0.002 |
ished | +0.002 |
ins | +0.002 |
aming | +0.002 |
brothers | +0.002 |
Neg logit contributions | |
hill | -0.002 |
road | -0.002 |
distance | -0.002 |
ladder | -0.002 |
bench | -0.002 |
Token | doing |
Token ID | 1804 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
ished | +0.080 |
lee | +0.078 |
aming | +0.073 |
ady | +0.072 |
amed | +0.070 |
Neg logit contributions | |
hill | -0.077 |
road | -0.070 |
ladder | -0.070 |
cliff | -0.069 |
taxi | -0.069 |
Token | in |
Token ID | 287 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
lee | +0.036 |
ished | +0.036 |
aming | +0.034 |
ady | +0.034 |
amed | +0.032 |
Neg logit contributions | |
hill | -0.028 |
road | -0.025 |
ladder | -0.025 |
taxi | -0.025 |
cliff | -0.025 |
Token | my |
Token ID | 616 |
Feature activation | +4.71e-03 |
Pos logit contributions | |
hill | +0.189 |
ladder | +0.173 |
road | +0.172 |
taxi | +0.168 |
Zig | +0.166 |
Neg logit contributions | |
lee | -0.210 |
ished | -0.202 |
ady | -0.201 |
aming | -0.201 |
pled | -0.187 |
Token | barn |
Token ID | 25203 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
hill | +0.007 |
Zig | +0.007 |
ladder | +0.006 |
cliff | +0.006 |
road | +0.006 |
Neg logit contributions | |
lee | -0.011 |
ished | -0.011 |
ady | -0.010 |
aming | -0.010 |
ice | -0.010 |
Token | ?". |
Token ID | 43634 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
hill | +0.325 |
road | +0.298 |
ladder | +0.297 |
taxi | +0.295 |
cliff | +0.293 |
Neg logit contributions | |
lee | -0.375 |
ished | -0.372 |
aming | -0.354 |
ady | -0.352 |
amed | -0.340 |
Token |
|
Token ID | 198 |
Feature activation | +2.48e-02 |
Pos logit contributions | |
hill | +0.022 |
cliff | +0.020 |
ladder | +0.020 |
road | +0.020 |
taxi | +0.020 |
Neg logit contributions | |
lee | -0.025 |
ished | -0.023 |
ady | -0.023 |
aming | -0.022 |
amed | -0.022 |
Token |
|
Token ID | 198 |
Feature activation | +2.33e-02 |
Pos logit contributions | |
hill | +0.039 |
road | +0.035 |
ladder | +0.035 |
taxi | +0.034 |
cliff | +0.034 |
Neg logit contributions | |
ished | -0.062 |
lee | -0.062 |
aming | -0.059 |
ady | -0.058 |
amed | -0.057 |
Token | Jake |
Token ID | 43930 |
Feature activation | -9.49e-02 |
Pos logit contributions | |
hill | +0.047 |
road | +0.043 |
ladder | +0.043 |
taxi | +0.042 |
cliff | +0.042 |
Neg logit contributions | |
lee | -0.048 |
ished | -0.048 |
aming | -0.045 |
ady | -0.045 |
amed | -0.043 |
Token | smiled |
Token ID | 13541 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
ished | +0.208 |
lee | +0.196 |
ins | +0.194 |
aming | +0.192 |
ady | +0.189 |
Neg logit contributions | |
hill | -0.170 |
ride | -0.161 |
walk | -0.158 |
visit | -0.157 |
taxi | -0.155 |
Token | a |
Token ID | 257 |
Feature activation | +6.99e-02 |
Pos logit contributions | |
lee | +0.054 |
ished | +0.053 |
ady | +0.051 |
aming | +0.050 |
amed | +0.049 |
Neg logit contributions | |
hill | -0.035 |
ladder | -0.032 |
taxi | -0.032 |
road | -0.031 |
cliff | -0.031 |
Token | small |
Token ID | 1402 |
Feature activation | +1.80e-03 |
Pos logit contributions | |
Zig | +0.125 |
hill | +0.120 |
cliff | +0.112 |
road | +0.112 |
Raven | +0.110 |
Neg logit contributions | |
ady | -0.126 |
lee | -0.123 |
ished | -0.121 |
aming | -0.118 |
pled | -0.113 |
Token | axe |
Token ID | 23272 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
hill | +0.003 |
Zig | +0.003 |
road | +0.003 |
cliff | +0.003 |
taxi | +0.003 |
Neg logit contributions | |
ished | -0.004 |
lee | -0.004 |
ady | -0.004 |
aming | -0.004 |
pled | -0.004 |
Token | in |
Token ID | 287 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
hill | +0.234 |
road | +0.217 |
ladder | +0.217 |
taxi | +0.216 |
cliff | +0.216 |
Neg logit contributions | |
ished | -0.241 |
lee | -0.236 |
aming | -0.225 |
ady | -0.222 |
pled | -0.213 |
Token | the |
Token ID | 262 |
Feature activation | +4.25e-02 |
Pos logit contributions | |
hill | +0.200 |
ladder | +0.187 |
taxi | +0.185 |
cliff | +0.184 |
road | +0.183 |
Neg logit contributions | |
lee | -0.220 |
ady | -0.215 |
ished | -0.214 |
aming | -0.212 |
pled | -0.195 |
Token | barn |
Token ID | 25203 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
Zig | +0.058 |
hill | +0.055 |
risk | +0.053 |
prize | +0.053 |
ladder | +0.052 |
Neg logit contributions | |
ady | -0.098 |
aming | -0.097 |
lee | -0.096 |
ished | -0.096 |
amed | -0.093 |
Token | . |
Token ID | 13 |
Feature activation | +8.76e-03 |
Pos logit contributions | |
hill | +0.348 |
road | +0.320 |
ladder | +0.319 |
taxi | +0.315 |
cliff | +0.314 |
Neg logit contributions | |
lee | -0.367 |
ished | -0.366 |
aming | -0.344 |
ady | -0.344 |
amed | -0.331 |
Token | He |
Token ID | 679 |
Feature activation | +4.71e-03 |
Pos logit contributions | |
hill | +0.016 |
taxi | +0.014 |
cliff | +0.014 |
road | +0.014 |
ladder | +0.014 |
Neg logit contributions | |
lee | -0.020 |
ished | -0.019 |
ady | -0.018 |
aming | -0.018 |
amed | -0.017 |
Token | was |
Token ID | 373 |
Feature activation | -4.53e-02 |
Pos logit contributions | |
hill | +0.009 |
ladder | +0.009 |
road | +0.008 |
ride | +0.008 |
mountain | +0.008 |
Neg logit contributions | |
lee | -0.010 |
ished | -0.010 |
ady | -0.010 |
aming | -0.010 |
ealing | -0.009 |
Token | so |
Token ID | 523 |
Feature activation | +3.12e-02 |
Pos logit contributions | |
ished | +0.109 |
lee | +0.108 |
ady | +0.100 |
ins | +0.098 |
aming | +0.098 |
Neg logit contributions | |
hill | -0.074 |
ladder | -0.071 |
taxi | -0.070 |
road | -0.069 |
ride | -0.067 |
Token | excited |
Token ID | 6568 |
Feature activation | -9.26e-02 |
Pos logit contributions | |
ladder | +0.067 |
hill | +0.066 |
taxi | +0.063 |
road | +0.061 |
cliff | +0.060 |
Neg logit contributions | |
lee | -0.062 |
ished | -0.061 |
ice | -0.057 |
ady | -0.055 |
inda | -0.054 |
Token | a |
Token ID | 257 |
Feature activation | +7.17e-02 |
Pos logit contributions | |
ished | +0.011 |
lee | +0.011 |
aming | +0.011 |
ady | +0.011 |
amed | +0.010 |
Neg logit contributions | |
hill | -0.009 |
cliff | -0.009 |
road | -0.009 |
ladder | -0.008 |
taxi | -0.008 |
Token | candle |
Token ID | 26839 |
Feature activation | +9.78e-02 |
Pos logit contributions | |
Zig | +0.123 |
hill | +0.113 |
Raven | +0.113 |
subway | +0.106 |
cliff | +0.106 |
Neg logit contributions | |
lee | -0.132 |
ady | -0.131 |
ished | -0.125 |
aming | -0.125 |
mistakes | -0.121 |
Token | burning |
Token ID | 9482 |
Feature activation | +7.39e-04 |
Pos logit contributions | |
hill | +0.190 |
road | +0.177 |
ladder | +0.174 |
cliff | +0.174 |
taxi | +0.171 |
Neg logit contributions | |
ished | -0.198 |
lee | -0.194 |
aming | -0.189 |
ady | -0.186 |
amed | -0.178 |
Token | in |
Token ID | 287 |
Feature activation | +9.60e-02 |
Pos logit contributions | |
hill | +0.001 |
road | +0.001 |
ladder | +0.001 |
taxi | +0.001 |
cliff | +0.001 |
Neg logit contributions | |
ished | -0.002 |
lee | -0.002 |
aming | -0.002 |
ady | -0.002 |
amed | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
hill | +0.165 |
ladder | +0.150 |
road | +0.149 |
taxi | +0.148 |
cliff | +0.147 |
Neg logit contributions | |
lee | -0.222 |
ished | -0.220 |
aming | -0.211 |
ady | -0.211 |
amed | -0.202 |
Token | barn |
Token ID | 25203 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
hill | +0.048 |
Zig | +0.044 |
cliff | +0.043 |
ladder | +0.042 |
ride | +0.042 |
Neg logit contributions | |
lee | -0.076 |
ished | -0.076 |
aming | -0.075 |
ady | -0.074 |
amed | -0.071 |
Token | and |
Token ID | 290 |
Feature activation | +4.80e-02 |
Pos logit contributions | |
hill | +0.357 |
road | +0.329 |
ladder | +0.328 |
taxi | +0.322 |
cliff | +0.321 |
Neg logit contributions | |
lee | -0.358 |
ished | -0.358 |
aming | -0.335 |
ady | -0.335 |
amed | -0.323 |
Token | it |
Token ID | 340 |
Feature activation | +2.22e-02 |
Pos logit contributions | |
hill | +0.086 |
road | +0.078 |
ladder | +0.078 |
taxi | +0.077 |
cliff | +0.077 |
Neg logit contributions | |
lee | -0.109 |
ished | -0.109 |
ady | -0.103 |
aming | -0.103 |
amed | -0.099 |
Token | had |
Token ID | 550 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
hill | +0.040 |
ladder | +0.037 |
road | +0.036 |
mountain | +0.035 |
cliff | +0.035 |
Neg logit contributions | |
lee | -0.051 |
ished | -0.050 |
ady | -0.050 |
aming | -0.048 |
amed | -0.046 |
Token | started |
Token ID | 2067 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
hill | +0.021 |
road | +0.019 |
ladder | +0.019 |
taxi | +0.019 |
cliff | +0.019 |
Neg logit contributions | |
lee | -0.024 |
ished | -0.023 |
ady | -0.022 |
aming | -0.022 |
amed | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | +6.12e-04 |
Pos logit contributions | |
lee | +0.043 |
ished | +0.043 |
aming | +0.041 |
ady | +0.041 |
amed | +0.040 |
Neg logit contributions | |
hill | -0.016 |
road | -0.014 |
ladder | -0.014 |
cliff | -0.013 |
taxi | -0.013 |
Token | a |
Token ID | 257 |
Feature activation | +6.53e-02 |
Pos logit contributions | |
hill | +0.017 |
ladder | +0.015 |
road | +0.015 |
taxi | +0.015 |
cliff | +0.015 |
Neg logit contributions | |
lee | -0.017 |
ished | -0.017 |
aming | -0.016 |
ady | -0.016 |
amed | -0.016 |
Token | flashlight |
Token ID | 38371 |
Feature activation | +9.01e-02 |
Pos logit contributions | |
hill | +0.099 |
risk | +0.095 |
Zig | +0.093 |
cliff | +0.092 |
road | +0.091 |
Neg logit contributions | |
ady | -0.138 |
lee | -0.137 |
ished | -0.129 |
aming | -0.125 |
amed | -0.123 |
Token | and |
Token ID | 290 |
Feature activation | +5.46e-02 |
Pos logit contributions | |
hill | +0.180 |
cliff | +0.169 |
road | +0.166 |
ladder | +0.165 |
risk | +0.164 |
Neg logit contributions | |
lee | -0.172 |
ished | -0.171 |
ady | -0.159 |
aming | -0.158 |
amed | -0.153 |
Token | a |
Token ID | 257 |
Feature activation | +6.43e-02 |
Pos logit contributions | |
hill | +0.109 |
ladder | +0.099 |
road | +0.099 |
cliff | +0.099 |
taxi | +0.098 |
Neg logit contributions | |
lee | -0.113 |
ished | -0.111 |
ady | -0.105 |
aming | -0.105 |
amed | -0.101 |
Token | magn |
Token ID | 7842 |
Feature activation | +3.56e-02 |
Pos logit contributions | |
hill | +0.095 |
risk | +0.091 |
Zig | +0.089 |
cliff | +0.088 |
road | +0.087 |
Neg logit contributions | |
lee | -0.139 |
ady | -0.137 |
ished | -0.130 |
aming | -0.127 |
amed | -0.125 |
Token | ifying |
Token ID | 4035 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
hill | +0.052 |
road | +0.047 |
ladder | +0.046 |
cliff | +0.045 |
taxi | +0.045 |
Neg logit contributions | |
ished | -0.094 |
lee | -0.093 |
ady | -0.089 |
aming | -0.088 |
amed | -0.086 |
Token | glass |
Token ID | 5405 |
Feature activation | +5.11e-02 |
Pos logit contributions | |
subway | +0.297 |
hill | +0.295 |
taxi | +0.289 |
ride | +0.289 |
Zig | +0.288 |
Neg logit contributions | |
ished | -0.320 |
lee | -0.320 |
aming | -0.319 |
amed | -0.307 |
ady | -0.305 |
Token | to |
Token ID | 284 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
hill | +0.097 |
road | +0.090 |
ladder | +0.089 |
taxi | +0.089 |
cliff | +0.088 |
Neg logit contributions | |
ished | -0.107 |
lee | -0.107 |
ady | -0.100 |
aming | -0.099 |
amed | -0.096 |
Token | examine |
Token ID | 10716 |
Feature activation | -3.18e-02 |
Pos logit contributions | |
hill | +0.198 |
ladder | +0.175 |
cliff | +0.174 |
taxi | +0.173 |
road | +0.170 |
Neg logit contributions | |
lee | -0.316 |
ished | -0.309 |
ady | -0.298 |
aming | -0.294 |
amed | -0.286 |
Token | their |
Token ID | 511 |
Feature activation | +8.97e-03 |
Pos logit contributions | |
lee | +0.069 |
ished | +0.069 |
aming | +0.065 |
ady | +0.065 |
amed | +0.063 |
Neg logit contributions | |
hill | -0.060 |
road | -0.055 |
ladder | -0.055 |
taxi | -0.054 |
cliff | -0.053 |
Token | finds |
Token ID | 7228 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
hill | +0.017 |
ladder | +0.016 |
taxi | +0.016 |
cliff | +0.016 |
road | +0.016 |
Neg logit contributions | |
lee | -0.018 |
ished | -0.018 |
ady | -0.017 |
aming | -0.016 |
ins | -0.016 |
Token | had |
Token ID | 550 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
hill | +0.020 |
road | +0.018 |
ladder | +0.018 |
cliff | +0.017 |
taxi | +0.017 |
Neg logit contributions | |
ished | -0.024 |
lee | -0.024 |
ady | -0.023 |
aming | -0.023 |
ins | -0.022 |
Token | a |
Token ID | 257 |
Feature activation | +7.34e-02 |
Pos logit contributions | |
hill | +0.022 |
road | +0.020 |
ladder | +0.020 |
taxi | +0.020 |
cliff | +0.020 |
Neg logit contributions | |
ished | -0.035 |
lee | -0.035 |
ady | -0.033 |
aming | -0.033 |
amed | -0.032 |
Token | seat |
Token ID | 5852 |
Feature activation | +5.91e-02 |
Pos logit contributions | |
hill | +0.119 |
Zig | +0.112 |
road | +0.111 |
Raven | +0.109 |
cliff | +0.108 |
Neg logit contributions | |
lee | -0.158 |
ished | -0.152 |
ady | -0.150 |
aming | -0.145 |
amed | -0.140 |
Token | inside |
Token ID | 2641 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
hill | +0.108 |
road | +0.099 |
cliff | +0.099 |
ladder | +0.097 |
taxi | +0.096 |
Neg logit contributions | |
ished | -0.129 |
lee | -0.128 |
aming | -0.121 |
ady | -0.119 |
amed | -0.117 |
Token | the |
Token ID | 262 |
Feature activation | +3.41e-02 |
Pos logit contributions | |
hill | +0.242 |
ladder | +0.223 |
road | +0.223 |
taxi | +0.221 |
cliff | +0.220 |
Neg logit contributions | |
lee | -0.254 |
ished | -0.252 |
aming | -0.239 |
ady | -0.237 |
amed | -0.228 |
Token | fort |
Token ID | 6285 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
hill | +0.044 |
Zig | +0.044 |
Raven | +0.041 |
risk | +0.041 |
taxi | +0.041 |
Neg logit contributions | |
ished | -0.082 |
lee | -0.082 |
aming | -0.079 |
ady | -0.079 |
amed | -0.074 |
Token | . |
Token ID | 13 |
Feature activation | +7.03e-03 |
Pos logit contributions | |
hill | +0.348 |
road | +0.320 |
ladder | +0.319 |
taxi | +0.317 |
cliff | +0.316 |
Neg logit contributions | |
lee | -0.359 |
ished | -0.357 |
aming | -0.336 |
ady | -0.335 |
amed | -0.323 |
Token |
|
Token ID | 198 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
hill | +0.013 |
ladder | +0.011 |
road | +0.011 |
taxi | +0.011 |
cliff | +0.011 |
Neg logit contributions | |
lee | -0.016 |
ished | -0.016 |
ady | -0.015 |
aming | -0.015 |
amed | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | +2.20e-02 |
Pos logit contributions | |
hill | +0.037 |
road | +0.033 |
ladder | +0.033 |
taxi | +0.032 |
cliff | +0.032 |
Neg logit contributions | |
lee | -0.061 |
ished | -0.060 |
aming | -0.057 |
ady | -0.057 |
amed | -0.056 |
Token | Suddenly |
Token ID | 38582 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
hill | +0.041 |
road | +0.037 |
ladder | +0.037 |
taxi | +0.037 |
cliff | +0.036 |
Neg logit contributions | |
ished | -0.049 |
lee | -0.049 |
aming | -0.046 |
ady | -0.046 |
amed | -0.044 |
Token | , |
Token ID | 11 |
Feature activation | +4.92e-02 |
Pos logit contributions | |
hill | +0.041 |
road | +0.037 |
ladder | +0.037 |
cliff | +0.036 |
taxi | +0.036 |
Neg logit contributions | |
lee | -0.047 |
ished | -0.046 |
aming | -0.044 |
ady | -0.043 |
amed | -0.042 |
Token | saw |
Token ID | 2497 |
Feature activation | -2.63e-02 |
Pos logit contributions | |
ished | +0.176 |
lee | +0.176 |
aming | +0.169 |
amed | +0.167 |
bled | +0.166 |
Neg logit contributions | |
hill | -0.134 |
visit | -0.127 |
road | -0.123 |
walk | -0.122 |
ladder | -0.119 |
Token | a |
Token ID | 257 |
Feature activation | +5.91e-02 |
Pos logit contributions | |
ished | +0.062 |
lee | +0.061 |
ady | +0.058 |
aming | +0.058 |
amed | +0.056 |
Neg logit contributions | |
hill | -0.045 |
ladder | -0.042 |
taxi | -0.041 |
road | -0.041 |
cliff | -0.040 |
Token | lock |
Token ID | 5793 |
Feature activation | +6.51e-02 |
Pos logit contributions | |
Zig | +0.111 |
hill | +0.105 |
risk | +0.101 |
road | +0.101 |
cliff | +0.098 |
Neg logit contributions | |
lee | -0.101 |
aming | -0.095 |
ady | -0.095 |
pled | -0.095 |
ished | -0.092 |
Token | on |
Token ID | 319 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
hill | +0.108 |
road | +0.098 |
ladder | +0.096 |
cliff | +0.096 |
taxi | +0.094 |
Neg logit contributions | |
ished | -0.154 |
lee | -0.153 |
aming | -0.148 |
ady | -0.146 |
amed | -0.142 |
Token | the |
Token ID | 262 |
Feature activation | +2.66e-02 |
Pos logit contributions | |
hill | +0.252 |
cliff | +0.236 |
road | +0.229 |
ladder | +0.228 |
mountain | +0.227 |
Neg logit contributions | |
lee | -0.228 |
aming | -0.216 |
ished | -0.211 |
ice | -0.209 |
ady | -0.208 |
Token | barn |
Token ID | 25203 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
Zig | +0.047 |
hill | +0.047 |
cliff | +0.046 |
Sorry | +0.046 |
bel | +0.045 |
Neg logit contributions | |
aming | -0.045 |
ady | -0.044 |
ice | -0.042 |
amed | -0.042 |
lee | -0.042 |
Token | door |
Token ID | 3420 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
hill | +0.358 |
cliff | +0.331 |
taxi | +0.329 |
road | +0.329 |
ladder | +0.327 |
Neg logit contributions | |
lee | -0.339 |
ished | -0.331 |
aming | -0.316 |
ady | -0.314 |
amed | -0.303 |
Token | . |
Token ID | 13 |
Feature activation | +9.32e-03 |
Pos logit contributions | |
hill | +0.227 |
road | +0.208 |
ladder | +0.208 |
cliff | +0.205 |
taxi | +0.205 |
Neg logit contributions | |
lee | -0.223 |
ished | -0.223 |
aming | -0.209 |
ady | -0.209 |
amed | -0.201 |
Token | She |
Token ID | 1375 |
Feature activation | +8.51e-03 |
Pos logit contributions | |
hill | +0.020 |
cliff | +0.018 |
mountain | +0.018 |
risk | +0.018 |
road | +0.018 |
Neg logit contributions | |
lee | -0.018 |
ished | -0.017 |
ady | -0.016 |
aming | -0.016 |
amed | -0.015 |
Token | was |
Token ID | 373 |
Feature activation | -4.61e-02 |
Pos logit contributions | |
hill | +0.017 |
ladder | +0.016 |
road | +0.016 |
taxi | +0.015 |
mountain | +0.015 |
Neg logit contributions | |
ished | -0.018 |
lee | -0.017 |
aming | -0.017 |
ady | -0.017 |
ins | -0.016 |
Token | sad |
Token ID | 6507 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
ished | +0.089 |
lee | +0.088 |
ady | +0.082 |
aming | +0.081 |
ins | +0.078 |
Neg logit contributions | |
hill | -0.099 |
ladder | -0.094 |
road | -0.093 |
taxi | -0.092 |
cliff | -0.090 |
Token | to |
Token ID | 284 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
ished | +0.205 |
lee | +0.204 |
aming | +0.196 |
ady | +0.196 |
amed | +0.191 |
Neg logit contributions | |
hill | -0.088 |
road | -0.076 |
ladder | -0.076 |
cliff | -0.074 |
taxi | -0.073 |
Token | the |
Token ID | 262 |
Feature activation | -8.13e-04 |
Pos logit contributions | |
hill | +0.207 |
ladder | +0.193 |
road | +0.192 |
cliff | +0.185 |
taxi | +0.184 |
Neg logit contributions | |
lee | -0.229 |
ished | -0.217 |
ady | -0.204 |
aming | -0.202 |
ins | -0.194 |
Token | animals |
Token ID | 4695 |
Feature activation | +6.95e-02 |
Pos logit contributions | |
lee | +0.002 |
ished | +0.002 |
ady | +0.002 |
aming | +0.002 |
amed | +0.002 |
Neg logit contributions | |
hill | -0.001 |
ladder | -0.001 |
taxi | -0.001 |
cliff | -0.001 |
road | -0.001 |
Token | in |
Token ID | 287 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
hill | +0.138 |
road | +0.128 |
ladder | +0.127 |
taxi | +0.126 |
cliff | +0.125 |
Neg logit contributions | |
lee | -0.142 |
ished | -0.142 |
ady | -0.133 |
aming | -0.133 |
amed | -0.127 |
Token | the |
Token ID | 262 |
Feature activation | +3.58e-02 |
Pos logit contributions | |
hill | +0.169 |
ladder | +0.154 |
road | +0.153 |
taxi | +0.150 |
cliff | +0.149 |
Neg logit contributions | |
lee | -0.238 |
ished | -0.234 |
ady | -0.225 |
aming | -0.225 |
amed | -0.215 |
Token | barn |
Token ID | 25203 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
hill | +0.046 |
ladder | +0.044 |
taxi | +0.042 |
cliff | +0.041 |
road | +0.040 |
Neg logit contributions | |
lee | -0.093 |
ished | -0.091 |
aming | -0.090 |
ady | -0.089 |
amed | -0.086 |
Token | . |
Token ID | 13 |
Feature activation | +9.66e-03 |
Pos logit contributions | |
hill | +0.334 |
road | +0.304 |
ladder | +0.304 |
taxi | +0.297 |
cliff | +0.296 |
Neg logit contributions | |
ished | -0.380 |
lee | -0.380 |
aming | -0.359 |
ady | -0.358 |
amed | -0.346 |
Token | â |
Token ID | 22940 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
hill | +0.019 |
cliff | +0.018 |
ladder | +0.017 |
road | +0.017 |
taxi | +0.017 |
Neg logit contributions | |
lee | -0.020 |
ished | -0.019 |
ady | -0.018 |
aming | -0.018 |
amed | -0.017 |
Token | € |
Token ID | 26391 |
Feature activation | -9.28e-02 |
Pos logit contributions | |
lee | +0.367 |
ished | +0.362 |
ady | +0.340 |
aming | +0.337 |
ins | +0.332 |
Neg logit contributions | |
hill | -0.321 |
road | -0.294 |
ladder | -0.287 |
ride | -0.285 |
cliff | -0.283 |
Token | Joe |
Token ID | 5689 |
Feature activation | -6.84e-02 |
Pos logit contributions | |
lee | +0.211 |
ished | +0.204 |
ady | +0.195 |
aming | +0.194 |
amed | +0.186 |
Neg logit contributions | |
hill | -0.164 |
taxi | -0.152 |
road | -0.151 |
ladder | -0.151 |
cliff | -0.150 |
Token | was |
Token ID | 373 |
Feature activation | -4.62e-02 |
Pos logit contributions | |
ished | +0.132 |
lee | +0.132 |
aming | +0.128 |
ady | +0.125 |
amed | +0.121 |
Neg logit contributions | |
hill | -0.145 |
ladder | -0.133 |
taxi | -0.132 |
road | -0.132 |
ride | -0.132 |
Token | animals |
Token ID | 4695 |
Feature activation | +7.34e-02 |
Pos logit contributions | |
hill | +0.008 |
road | +0.007 |
ladder | +0.007 |
taxi | +0.007 |
cliff | +0.007 |
Neg logit contributions | |
lee | -0.011 |
ished | -0.011 |
aming | -0.010 |
ady | -0.010 |
amed | -0.010 |
Token | in |
Token ID | 287 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
hill | +0.141 |
road | +0.131 |
ladder | +0.130 |
taxi | +0.128 |
cliff | +0.127 |
Neg logit contributions | |
lee | -0.155 |
ished | -0.155 |
aming | -0.145 |
ady | -0.145 |
amed | -0.140 |
Token | the |
Token ID | 262 |
Feature activation | +3.33e-02 |
Pos logit contributions | |
hill | +0.189 |
ladder | +0.175 |
road | +0.173 |
taxi | +0.171 |
cliff | +0.171 |
Neg logit contributions | |
lee | -0.225 |
ished | -0.215 |
aming | -0.211 |
ady | -0.210 |
amed | -0.201 |
Token | big |
Token ID | 1263 |
Feature activation | +4.32e-02 |
Pos logit contributions | |
hill | +0.044 |
ladder | +0.039 |
road | +0.038 |
cliff | +0.038 |
taxi | +0.037 |
Neg logit contributions | |
lee | -0.089 |
ished | -0.088 |
aming | -0.086 |
ady | -0.085 |
amed | -0.083 |
Token | blue |
Token ID | 4171 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
hill | +0.055 |
road | +0.048 |
ladder | +0.048 |
taxi | +0.047 |
cliff | +0.047 |
Neg logit contributions | |
lee | -0.120 |
ished | -0.119 |
ady | -0.114 |
aming | -0.114 |
amed | -0.110 |
Token | barn |
Token ID | 25203 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
hill | +0.024 |
ladder | +0.022 |
road | +0.022 |
cliff | +0.022 |
taxi | +0.022 |
Neg logit contributions | |
lee | -0.044 |
ished | -0.043 |
ady | -0.042 |
aming | -0.041 |
ins | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | +7.85e-03 |
Pos logit contributions | |
hill | +0.335 |
road | +0.306 |
ladder | +0.306 |
taxi | +0.298 |
cliff | +0.297 |
Neg logit contributions | |
ished | -0.378 |
lee | -0.377 |
aming | -0.356 |
ady | -0.355 |
amed | -0.342 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -9.37e-03 |
Pos logit contributions | |
hill | +0.014 |
road | +0.012 |
ladder | +0.012 |
cliff | +0.012 |
taxi | +0.012 |
Neg logit contributions | |
lee | -0.018 |
ished | -0.018 |
ady | -0.017 |
aming | -0.017 |
amed | -0.017 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.36e-02 |
Pos logit contributions | |
lee | +0.021 |
ished | +0.021 |
ady | +0.020 |
aming | +0.020 |
amed | +0.019 |
Neg logit contributions | |
hill | -0.016 |
ladder | -0.015 |
road | -0.015 |
taxi | -0.015 |
cliff | -0.015 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
hill | +0.068 |
road | +0.061 |
ladder | +0.061 |
taxi | +0.060 |
cliff | +0.059 |
Neg logit contributions | |
lee | -0.109 |
ished | -0.108 |
aming | -0.103 |
ady | -0.103 |
amed | -0.100 |
Token | a |
Token ID | 257 |
Feature activation | +4.76e-02 |
Pos logit contributions | |
lee | +0.026 |
ished | +0.026 |
aming | +0.025 |
ady | +0.025 |
amed | +0.024 |
Neg logit contributions | |
hill | -0.028 |
ladder | -0.026 |
cliff | -0.025 |
road | -0.025 |
taxi | -0.025 |
Token | big |
Token ID | 1263 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
Zig | +0.110 |
hill | +0.105 |
risk | +0.101 |
road | +0.100 |
stack | +0.099 |
Neg logit contributions | |
lee | -0.116 |
aming | -0.105 |
ady | -0.105 |
ealing | -0.102 |
ished | -0.101 |
Token | red |
Token ID | 2266 |
Feature activation | +4.80e-03 |
Pos logit contributions | |
hill | +0.033 |
Zig | +0.032 |
taxi | +0.031 |
road | +0.031 |
ladder | +0.031 |
Neg logit contributions | |
lee | -0.064 |
ished | -0.060 |
ady | -0.058 |
aming | -0.058 |
pled | -0.056 |
Token | ball |
Token ID | 2613 |
Feature activation | +8.04e-02 |
Pos logit contributions | |
Zig | +0.007 |
bel | +0.007 |
Raven | +0.006 |
hill | +0.006 |
risk | +0.006 |
Neg logit contributions | |
lee | -0.010 |
ished | -0.010 |
ady | -0.010 |
aming | -0.009 |
ice | -0.009 |
Token | near |
Token ID | 1474 |
Feature activation | +6.07e-02 |
Pos logit contributions | |
hill | +0.142 |
road | +0.130 |
ladder | +0.130 |
taxi | +0.128 |
cliff | +0.128 |
Neg logit contributions | |
ished | -0.181 |
lee | -0.179 |
aming | -0.171 |
ady | -0.170 |
amed | -0.164 |
Token | the |
Token ID | 262 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
hill | +0.109 |
ladder | +0.099 |
road | +0.098 |
taxi | +0.097 |
cliff | +0.097 |
Neg logit contributions | |
lee | -0.139 |
ished | -0.138 |
aming | -0.131 |
ady | -0.131 |
amed | -0.126 |
Token | barn |
Token ID | 25203 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
hill | +0.033 |
ladder | +0.032 |
Zig | +0.032 |
taxi | +0.032 |
risk | +0.031 |
Neg logit contributions | |
lee | -0.074 |
aming | -0.073 |
ished | -0.072 |
ady | -0.071 |
amed | -0.069 |
Token | . |
Token ID | 13 |
Feature activation | +6.05e-03 |
Pos logit contributions | |
hill | +0.364 |
taxi | +0.339 |
ladder | +0.337 |
road | +0.335 |
cliff | +0.335 |
Neg logit contributions | |
lee | -0.331 |
ished | -0.327 |
aming | -0.307 |
ady | -0.306 |
amed | -0.294 |
Token | He |
Token ID | 679 |
Feature activation | +4.23e-03 |
Pos logit contributions | |
hill | +0.011 |
ladder | +0.010 |
road | +0.010 |
taxi | +0.010 |
cliff | +0.010 |
Neg logit contributions | |
lee | -0.014 |
ished | -0.014 |
aming | -0.013 |
ady | -0.013 |
amed | -0.012 |
Token | wanted |
Token ID | 2227 |
Feature activation | -7.65e-02 |
Pos logit contributions | |
hill | +0.009 |
ladder | +0.008 |
road | +0.008 |
taxi | +0.008 |
ride | +0.008 |
Neg logit contributions | |
ished | -0.009 |
aming | -0.008 |
lee | -0.008 |
ady | -0.008 |
ins | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
lee | +0.206 |
ished | +0.205 |
aming | +0.189 |
pled | +0.187 |
ady | +0.185 |
Neg logit contributions | |
hill | -0.102 |
taxi | -0.089 |
cliff | -0.089 |
road | -0.089 |
ladder | -0.087 |
Token | play |
Token ID | 711 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
hill | +0.146 |
subway | +0.135 |
mountain | +0.133 |
taxi | +0.131 |
cliff | +0.130 |
Neg logit contributions | |
lee | -0.269 |
ished | -0.262 |
ady | -0.245 |
aming | -0.243 |
amed | -0.235 |
Token | Many |
Token ID | 4650 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
hill | +0.017 |
road | +0.015 |
ladder | +0.015 |
cliff | +0.015 |
taxi | +0.015 |
Neg logit contributions | |
lee | -0.022 |
ished | -0.022 |
ady | -0.021 |
aming | -0.020 |
amed | -0.020 |
Token | animals |
Token ID | 4695 |
Feature activation | +5.09e-02 |
Pos logit contributions | |
ished | +0.054 |
lee | +0.054 |
aming | +0.052 |
ady | +0.052 |
amed | +0.050 |
Neg logit contributions | |
hill | -0.037 |
road | -0.033 |
taxi | -0.033 |
ladder | -0.033 |
cliff | -0.032 |
Token | lived |
Token ID | 5615 |
Feature activation | -2.42e-02 |
Pos logit contributions | |
hill | +0.097 |
road | +0.090 |
ladder | +0.089 |
cliff | +0.087 |
taxi | +0.087 |
Neg logit contributions | |
lee | -0.108 |
ished | -0.108 |
aming | -0.101 |
ady | -0.101 |
amed | -0.098 |
Token | inside |
Token ID | 2641 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
ished | +0.059 |
lee | +0.059 |
aming | +0.056 |
ady | +0.056 |
amed | +0.055 |
Neg logit contributions | |
hill | -0.039 |
road | -0.036 |
ladder | -0.035 |
cliff | -0.035 |
taxi | -0.035 |
Token | the |
Token ID | 262 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
hill | +0.201 |
road | +0.183 |
ladder | +0.183 |
taxi | +0.180 |
cliff | +0.179 |
Neg logit contributions | |
lee | -0.264 |
ished | -0.263 |
aming | -0.249 |
ady | -0.249 |
amed | -0.240 |
Token | barn |
Token ID | 25203 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
hill | +0.036 |
cliff | +0.032 |
ladder | +0.032 |
ride | +0.031 |
road | +0.031 |
Neg logit contributions | |
lee | -0.072 |
ished | -0.070 |
ady | -0.069 |
aming | -0.069 |
amed | -0.066 |
Token | . |
Token ID | 13 |
Feature activation | +8.23e-03 |
Pos logit contributions | |
hill | +0.338 |
road | +0.310 |
ladder | +0.309 |
taxi | +0.306 |
cliff | +0.305 |
Neg logit contributions | |
lee | -0.367 |
ished | -0.364 |
aming | -0.343 |
ady | -0.343 |
amed | -0.330 |
Token | They |
Token ID | 1119 |
Feature activation | +7.75e-03 |
Pos logit contributions | |
hill | +0.015 |
ladder | +0.013 |
cliff | +0.013 |
road | +0.013 |
taxi | +0.013 |
Neg logit contributions | |
lee | -0.019 |
ished | -0.018 |
ady | -0.018 |
aming | -0.017 |
amed | -0.017 |
Token | were |
Token ID | 547 |
Feature activation | -2.85e-02 |
Pos logit contributions | |
hill | +0.015 |
road | +0.014 |
ladder | +0.014 |
taxi | +0.014 |
cliff | +0.014 |
Neg logit contributions | |
lee | -0.016 |
ished | -0.016 |
aming | -0.015 |
ady | -0.015 |
amed | -0.014 |
Token | all |
Token ID | 477 |
Feature activation | +5.17e-03 |
Pos logit contributions | |
ished | +0.067 |
lee | +0.064 |
ady | +0.061 |
aming | +0.061 |
oking | +0.059 |
Neg logit contributions | |
hill | -0.050 |
taxi | -0.048 |
road | -0.047 |
ladder | -0.047 |
visit | -0.044 |
Token | friends |
Token ID | 2460 |
Feature activation | +9.17e-03 |
Pos logit contributions | |
hill | +0.010 |
ladder | +0.009 |
road | +0.009 |
taxi | +0.009 |
cliff | +0.009 |
Neg logit contributions | |
lee | -0.011 |
ished | -0.011 |
aming | -0.010 |
ady | -0.010 |
amed | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
hill | +0.051 |
road | +0.047 |
ladder | +0.047 |
taxi | +0.047 |
ride | +0.046 |
Neg logit contributions | |
ished | -0.048 |
lee | -0.047 |
ady | -0.047 |
aming | -0.045 |
amed | -0.043 |
Token | owner |
Token ID | 4870 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
lee | +0.061 |
ished | +0.061 |
aming | +0.059 |
ady | +0.059 |
amed | +0.057 |
Neg logit contributions | |
hill | -0.025 |
ladder | -0.021 |
road | -0.021 |
cliff | -0.021 |
taxi | -0.021 |
Token | smiled |
Token ID | 13541 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
hill | +0.036 |
ride | +0.034 |
taxi | +0.034 |
ladder | +0.034 |
road | +0.033 |
Neg logit contributions | |
lee | -0.031 |
ady | -0.030 |
ished | -0.030 |
aming | -0.029 |
amed | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | -1.93e-03 |
Pos logit contributions | |
ished | +0.425 |
ady | +0.423 |
lee | +0.416 |
ice | +0.415 |
ins | +0.405 |
Neg logit contributions | |
hill | -0.192 |
taxi | -0.189 |
ride | -0.184 |
visit | -0.175 |
walk | -0.168 |
Token | " |
Token ID | 366 |
Feature activation | +1.67e-04 |
Pos logit contributions | |
ished | +0.005 |
aming | +0.005 |
ady | +0.005 |
lee | +0.005 |
amed | +0.005 |
Neg logit contributions | |
hill | -0.003 |
ladder | -0.003 |
road | -0.003 |
taxi | -0.003 |
ride | -0.003 |
Token | Looks |
Token ID | 41102 |
Feature activation | -2.26e-01 |
Pos logit contributions | |
hill | +0.000 |
road | +0.000 |
ladder | +0.000 |
taxi | +0.000 |
cliff | +0.000 |
Neg logit contributions | |
lee | -0.000 |
ished | -0.000 |
ady | -0.000 |
aming | -0.000 |
amed | -0.000 |
Token | like |
Token ID | 588 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
lee | +0.658 |
aming | +0.626 |
ady | +0.616 |
ins | +0.615 |
ished | +0.614 |
Neg logit contributions | |
ride | -0.234 |
road | -0.233 |
risk | -0.229 |
hill | -0.227 |
visit | -0.208 |
Token | you |
Token ID | 345 |
Feature activation | -5.12e-02 |
Pos logit contributions | |
lee | +0.047 |
ished | +0.046 |
aming | +0.043 |
ady | +0.043 |
ins | +0.042 |
Neg logit contributions | |
hill | -0.039 |
ladder | -0.035 |
cliff | -0.035 |
road | -0.034 |
taxi | -0.034 |
Token | two |
Token ID | 734 |
Feature activation | -7.32e-03 |
Pos logit contributions | |
lee | +0.115 |
ished | +0.114 |
aming | +0.108 |
ady | +0.107 |
amed | +0.103 |
Neg logit contributions | |
hill | -0.095 |
road | -0.087 |
ladder | -0.086 |
taxi | -0.085 |
cliff | -0.084 |
Token | are |
Token ID | 389 |
Feature activation | -5.10e-03 |
Pos logit contributions | |
lee | +0.016 |
ished | +0.016 |
aming | +0.015 |
ady | +0.015 |
amed | +0.015 |
Neg logit contributions | |
hill | -0.014 |
road | -0.013 |
taxi | -0.013 |
ladder | -0.013 |
ride | -0.012 |
Token | friends |
Token ID | 2460 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
ished | +0.012 |
lee | +0.012 |
aming | +0.011 |
ady | +0.011 |
amed | +0.011 |
Neg logit contributions | |
hill | -0.009 |
taxi | -0.008 |
road | -0.008 |
ladder | -0.008 |
ride | -0.008 |
Token | smiled |
Token ID | 13541 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
lee | +0.100 |
ished | +0.099 |
ady | +0.098 |
aming | +0.095 |
ealing | +0.092 |
Neg logit contributions | |
hill | -0.116 |
ride | -0.109 |
slide | -0.105 |
taxi | -0.105 |
walk | -0.105 |
Token | . |
Token ID | 13 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
ished | +0.430 |
lee | +0.411 |
ady | +0.409 |
ice | +0.401 |
ins | +0.390 |
Neg logit contributions | |
hill | -0.216 |
taxi | -0.208 |
visit | -0.203 |
ride | -0.200 |
walk | -0.193 |
Token |
|
Token ID | 198 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
ished | +0.036 |
aming | +0.034 |
lee | +0.034 |
ady | +0.034 |
lling | +0.033 |
Neg logit contributions | |
hill | -0.016 |
ladder | -0.015 |
road | -0.014 |
taxi | -0.013 |
visit | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
hill | +0.014 |
road | +0.012 |
ladder | +0.012 |
visit | +0.012 |
taxi | +0.012 |
Neg logit contributions | |
ished | -0.034 |
lee | -0.033 |
aming | -0.032 |
ady | -0.032 |
amed | -0.031 |
Token | " |
Token ID | 1 |
Feature activation | -3.21e-03 |
Pos logit contributions | |
hill | +0.018 |
road | +0.016 |
ladder | +0.016 |
taxi | +0.015 |
cliff | +0.015 |
Neg logit contributions | |
ished | -0.029 |
lee | -0.028 |
ady | -0.025 |
pled | -0.025 |
ins | -0.024 |
Token | Looks |
Token ID | 41102 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
lee | +0.008 |
ished | +0.008 |
ady | +0.008 |
aming | +0.008 |
amed | +0.008 |
Neg logit contributions | |
hill | -0.005 |
road | -0.004 |
taxi | -0.004 |
ladder | -0.004 |
cliff | -0.004 |
Token | like |
Token ID | 588 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
lee | +0.641 |
aming | +0.626 |
bled | +0.606 |
ins | +0.598 |
ady | +0.596 |
Neg logit contributions | |
risk | -0.217 |
ride | -0.215 |
road | -0.207 |
hill | -0.199 |
visit | -0.192 |
Token | the |
Token ID | 262 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
lee | +0.046 |
ished | +0.045 |
aming | +0.042 |
ady | +0.042 |
amed | +0.041 |
Neg logit contributions | |
hill | -0.030 |
ladder | -0.027 |
road | -0.027 |
taxi | -0.027 |
cliff | -0.026 |
Token | frog |
Token ID | 21264 |
Feature activation | +9.41e-03 |
Pos logit contributions | |
ished | +0.027 |
lee | +0.027 |
ady | +0.026 |
aming | +0.026 |
amed | +0.025 |
Neg logit contributions | |
hill | -0.015 |
road | -0.013 |
ladder | -0.013 |
taxi | -0.013 |
cliff | -0.013 |
Token | likes |
Token ID | 7832 |
Feature activation | -5.29e-02 |
Pos logit contributions | |
hill | +0.017 |
ladder | +0.016 |
road | +0.016 |
taxi | +0.016 |
cliff | +0.015 |
Neg logit contributions | |
lee | -0.021 |
ished | -0.021 |
aming | -0.020 |
ady | -0.020 |
amed | -0.019 |
Token | you |
Token ID | 345 |
Feature activation | -4.06e-02 |
Pos logit contributions | |
lee | +0.131 |
ished | +0.129 |
ady | +0.122 |
aming | +0.121 |
ins | +0.117 |
Neg logit contributions | |
hill | -0.083 |
ladder | -0.075 |
road | -0.074 |
taxi | -0.073 |
cliff | -0.073 |
Token | smiled |
Token ID | 13541 |
Feature activation | -1.61e-01 |
Pos logit contributions | |
ady | +0.041 |
ished | +0.038 |
lee | +0.037 |
ealing | +0.036 |
aming | +0.036 |
Neg logit contributions | |
hill | -0.039 |
ride | -0.039 |
slide | -0.037 |
taxi | -0.037 |
walk | -0.037 |
Token | and |
Token ID | 290 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
ished | +0.432 |
ice | +0.411 |
ins | +0.411 |
ady | +0.409 |
wy | +0.406 |
Neg logit contributions | |
visit | -0.197 |
hill | -0.196 |
taxi | -0.195 |
ride | -0.193 |
walk | -0.189 |
Token | said |
Token ID | 531 |
Feature activation | -8.04e-02 |
Pos logit contributions | |
hill | +0.029 |
ride | +0.029 |
ladder | +0.028 |
taxi | +0.028 |
visit | +0.028 |
Neg logit contributions | |
ady | -0.028 |
ished | -0.028 |
aming | -0.026 |
lee | -0.026 |
ealing | -0.025 |
Token | , |
Token ID | 11 |
Feature activation | +3.15e-02 |
Pos logit contributions | |
ished | +0.211 |
ady | +0.210 |
lee | +0.206 |
aming | +0.206 |
amed | +0.196 |
Neg logit contributions | |
hill | -0.109 |
ladder | -0.106 |
taxi | -0.104 |
cliff | -0.096 |
ride | -0.094 |
Token | " |
Token ID | 366 |
Feature activation | +4.86e-04 |
Pos logit contributions | |
hill | +0.056 |
ladder | +0.052 |
road | +0.051 |
taxi | +0.050 |
cliff | +0.050 |
Neg logit contributions | |
ished | -0.073 |
lee | -0.071 |
ady | -0.069 |
aming | -0.068 |
amed | -0.065 |
Token | Looks |
Token ID | 41102 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
hill | +0.001 |
road | +0.001 |
ladder | +0.001 |
taxi | +0.001 |
cliff | +0.001 |
Neg logit contributions | |
lee | -0.001 |
ished | -0.001 |
ady | -0.001 |
aming | -0.001 |
amed | -0.001 |
Token | like |
Token ID | 588 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
lee | +0.651 |
aming | +0.626 |
ady | +0.611 |
ished | +0.604 |
ins | +0.601 |
Neg logit contributions | |
hill | -0.202 |
ride | -0.202 |
road | -0.202 |
risk | -0.201 |
visit | -0.189 |
Token | you |
Token ID | 345 |
Feature activation | -4.81e-02 |
Pos logit contributions | |
lee | +0.045 |
ished | +0.044 |
aming | +0.042 |
ady | +0.041 |
amed | +0.040 |
Neg logit contributions | |
hill | -0.031 |
road | -0.028 |
ladder | -0.028 |
taxi | -0.028 |
cliff | -0.028 |
Token | like |
Token ID | 588 |
Feature activation | -5.43e-03 |
Pos logit contributions | |
ished | +0.115 |
lee | +0.112 |
aming | +0.109 |
ady | +0.107 |
amed | +0.105 |
Neg logit contributions | |
hill | -0.081 |
road | -0.073 |
ladder | -0.072 |
visit | -0.072 |
ride | -0.072 |
Token | the |
Token ID | 262 |
Feature activation | +2.46e-03 |
Pos logit contributions | |
lee | +0.012 |
ished | +0.012 |
aming | +0.011 |
ady | +0.011 |
ins | +0.011 |
Neg logit contributions | |
hill | -0.010 |
road | -0.009 |
ladder | -0.009 |
cliff | -0.009 |
taxi | -0.009 |
Token | wardrobe |
Token ID | 35593 |
Feature activation | +8.39e-02 |
Pos logit contributions | |
hill | +0.005 |
ladder | +0.004 |
road | +0.004 |
cliff | +0.004 |
taxi | +0.004 |
Neg logit contributions | |
lee | -0.005 |
ished | -0.005 |
aming | -0.005 |
ady | -0.005 |
amed | -0.005 |
Token | 's |
Token ID | 338 |
Feature activation | -4.22e-02 |
Pos logit contributions | |
ished | +0.317 |
lee | +0.304 |
aming | +0.301 |
ady | +0.295 |
ins | +0.293 |
Neg logit contributions | |
hill | -0.207 |
ladder | -0.193 |
ride | -0.185 |
taxi | -0.181 |
road | -0.181 |
Token | mom |
Token ID | 1995 |
Feature activation | -3.11e-02 |
Pos logit contributions | |
ished | +0.098 |
lee | +0.095 |
ady | +0.093 |
aming | +0.091 |
amed | +0.090 |
Neg logit contributions | |
hill | -0.076 |
ladder | -0.071 |
taxi | -0.069 |
road | -0.068 |
ride | -0.066 |
Token | my |
Token ID | 1820 |
Feature activation | -6.94e-02 |
Pos logit contributions | |
ished | +0.055 |
lee | +0.054 |
aming | +0.051 |
ady | +0.050 |
amed | +0.050 |
Neg logit contributions | |
hill | -0.068 |
ladder | -0.065 |
taxi | -0.064 |
ride | -0.064 |
road | -0.062 |
Token | . |
Token ID | 13 |
Feature activation | -1.29e-03 |
Pos logit contributions | |
ished | +0.169 |
lee | +0.163 |
aming | +0.161 |
ady | +0.160 |
amed | +0.157 |
Neg logit contributions | |
hill | -0.106 |
ladder | -0.101 |
taxi | -0.101 |
ride | -0.098 |
walk | -0.094 |
Token | " |
Token ID | 366 |
Feature activation | +1.55e-03 |
Pos logit contributions | |
ished | +0.003 |
lee | +0.003 |
aming | +0.003 |
ady | +0.003 |
amed | +0.003 |
Neg logit contributions | |
hill | -0.002 |
ladder | -0.002 |
road | -0.002 |
taxi | -0.002 |
cliff | -0.002 |
Token | Looks |
Token ID | 41102 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
hill | +0.003 |
road | +0.002 |
ladder | +0.002 |
taxi | +0.002 |
cliff | +0.002 |
Neg logit contributions | |
lee | -0.004 |
ished | -0.004 |
ady | -0.003 |
aming | -0.003 |
amed | -0.003 |
Token | like |
Token ID | 588 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
lee | +0.579 |
aming | +0.554 |
ished | +0.543 |
ady | +0.541 |
ins | +0.536 |
Neg logit contributions | |
road | -0.283 |
hill | -0.281 |
risk | -0.279 |
ride | -0.268 |
ladder | -0.266 |
Token | the |
Token ID | 262 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
lee | +0.051 |
ished | +0.050 |
aming | +0.048 |
ady | +0.047 |
amed | +0.046 |
Neg logit contributions | |
hill | -0.034 |
road | -0.031 |
ladder | -0.030 |
taxi | -0.030 |
cliff | -0.029 |
Token | wave |
Token ID | 6769 |
Feature activation | -5.20e-02 |
Pos logit contributions | |
ished | +0.027 |
lee | +0.027 |
ady | +0.025 |
aming | +0.025 |
amed | +0.025 |
Neg logit contributions | |
hill | -0.016 |
road | -0.014 |
ladder | -0.014 |
cliff | -0.014 |
taxi | -0.014 |
Token | destroyed |
Token ID | 6572 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
ished | +0.114 |
lee | +0.113 |
ady | +0.107 |
aming | +0.106 |
amed | +0.103 |
Neg logit contributions | |
hill | -0.095 |
road | -0.087 |
ladder | -0.086 |
cliff | -0.085 |
taxi | -0.085 |
Token | all |
Token ID | 477 |
Feature activation | +2.35e-02 |
Pos logit contributions | |
lee | +0.049 |
ished | +0.049 |
aming | +0.046 |
ady | +0.046 |
amed | +0.044 |
Neg logit contributions | |
hill | -0.040 |
road | -0.037 |
ladder | -0.037 |
taxi | -0.036 |
cliff | -0.036 |
Token | laughed |
Token ID | 13818 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
ished | +0.149 |
ady | +0.144 |
lee | +0.142 |
aming | +0.139 |
ealing | +0.132 |
Neg logit contributions | |
hill | -0.124 |
ride | -0.118 |
visit | -0.114 |
walk | -0.112 |
taxi | -0.111 |
Token | and |
Token ID | 290 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
lee | +0.389 |
ished | +0.388 |
ady | +0.385 |
ins | +0.357 |
aming | +0.354 |
Neg logit contributions | |
hill | -0.227 |
ride | -0.220 |
taxi | -0.219 |
while | -0.212 |
visit | -0.210 |
Token | said |
Token ID | 531 |
Feature activation | -8.31e-02 |
Pos logit contributions | |
hill | +0.041 |
ride | +0.037 |
ladder | +0.037 |
taxi | +0.037 |
road | +0.037 |
Neg logit contributions | |
ished | -0.035 |
ady | -0.034 |
lee | -0.033 |
aming | -0.033 |
amed | -0.031 |
Token | , |
Token ID | 11 |
Feature activation | +3.12e-02 |
Pos logit contributions | |
ished | +0.200 |
lee | +0.199 |
ady | +0.190 |
aming | +0.190 |
amed | +0.184 |
Neg logit contributions | |
hill | -0.138 |
ladder | -0.125 |
road | -0.124 |
taxi | -0.123 |
cliff | -0.122 |
Token | " |
Token ID | 366 |
Feature activation | +3.83e-03 |
Pos logit contributions | |
hill | +0.061 |
road | +0.056 |
ladder | +0.056 |
taxi | +0.055 |
cliff | +0.055 |
Neg logit contributions | |
lee | -0.066 |
ished | -0.065 |
aming | -0.061 |
ady | -0.061 |
amed | -0.059 |
Token | Looks |
Token ID | 41102 |
Feature activation | -2.17e-01 |
Pos logit contributions | |
hill | +0.006 |
road | +0.006 |
ladder | +0.006 |
taxi | +0.006 |
cliff | +0.006 |
Neg logit contributions | |
lee | -0.009 |
ished | -0.009 |
ady | -0.009 |
aming | -0.008 |
amed | -0.008 |
Token | like |
Token ID | 588 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
lee | +0.617 |
aming | +0.592 |
ady | +0.586 |
ished | +0.584 |
ins | +0.579 |
Neg logit contributions | |
risk | -0.221 |
hill | -0.220 |
road | -0.220 |
ride | -0.219 |
ladder | -0.201 |
Token | that |
Token ID | 326 |
Feature activation | +4.58e-02 |
Pos logit contributions | |
lee | +0.031 |
ished | +0.030 |
aming | +0.029 |
ady | +0.028 |
pled | +0.028 |
Neg logit contributions | |
hill | -0.024 |
road | -0.022 |
ladder | -0.022 |
taxi | -0.022 |
cliff | -0.021 |
Token | apple |
Token ID | 17180 |
Feature activation | +8.21e-02 |
Pos logit contributions | |
hill | +0.082 |
road | +0.075 |
ladder | +0.075 |
taxi | +0.073 |
cliff | +0.073 |
Neg logit contributions | |
lee | -0.104 |
ished | -0.104 |
aming | -0.098 |
ady | -0.098 |
amed | -0.095 |
Token | made |
Token ID | 925 |
Feature activation | -4.26e-02 |
Pos logit contributions | |
hill | +0.147 |
road | +0.135 |
taxi | +0.134 |
ladder | +0.134 |
cliff | +0.132 |
Neg logit contributions | |
ished | -0.181 |
lee | -0.179 |
aming | -0.169 |
ady | -0.167 |
amed | -0.162 |
Token | you |
Token ID | 345 |
Feature activation | -4.29e-02 |
Pos logit contributions | |
ished | +0.113 |
lee | +0.113 |
ady | +0.109 |
aming | +0.108 |
amed | +0.105 |
Neg logit contributions | |
hill | -0.060 |
road | -0.053 |
ladder | -0.053 |
taxi | -0.052 |
cliff | -0.051 |
Token | table |
Token ID | 3084 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
hill | +0.054 |
cliff | +0.051 |
taxi | +0.050 |
ladder | +0.049 |
Zig | +0.049 |
Neg logit contributions | |
lee | -0.063 |
ished | -0.063 |
aming | -0.061 |
ady | -0.060 |
amed | -0.058 |
Token | . |
Token ID | 13 |
Feature activation | -1.50e-02 |
Pos logit contributions | |
hill | +0.193 |
road | +0.176 |
ladder | +0.176 |
taxi | +0.169 |
mountain | +0.166 |
Neg logit contributions | |
lee | -0.243 |
ished | -0.241 |
aming | -0.228 |
ady | -0.228 |
amed | -0.222 |
Token |
|
Token ID | 198 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
ished | +0.036 |
lee | +0.036 |
aming | +0.034 |
ady | +0.034 |
amed | +0.033 |
Neg logit contributions | |
hill | -0.025 |
road | -0.023 |
ladder | -0.023 |
taxi | -0.022 |
cliff | -0.022 |
Token |
|
Token ID | 198 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
hill | +0.018 |
road | +0.016 |
ladder | +0.016 |
taxi | +0.016 |
cliff | +0.016 |
Neg logit contributions | |
ished | -0.032 |
lee | -0.032 |
aming | -0.030 |
ady | -0.030 |
amed | -0.030 |
Token | " |
Token ID | 1 |
Feature activation | -1.72e-03 |
Pos logit contributions | |
hill | +0.023 |
road | +0.021 |
ladder | +0.021 |
taxi | +0.021 |
cliff | +0.020 |
Neg logit contributions | |
ished | -0.024 |
lee | -0.024 |
ady | -0.022 |
aming | -0.022 |
amed | -0.021 |
Token | Looks |
Token ID | 41102 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
lee | +0.004 |
ished | +0.004 |
aming | +0.004 |
ady | +0.004 |
amed | +0.004 |
Neg logit contributions | |
hill | -0.003 |
ladder | -0.003 |
road | -0.003 |
cliff | -0.003 |
taxi | -0.003 |
Token | like |
Token ID | 588 |
Feature activation | -1.66e-02 |
Pos logit contributions | |
lee | +0.619 |
aming | +0.593 |
bled | +0.571 |
ished | +0.571 |
ady | +0.569 |
Neg logit contributions | |
road | -0.221 |
risk | -0.216 |
ride | -0.209 |
jail | -0.209 |
ladder | -0.207 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
lee | +0.038 |
ished | +0.038 |
aming | +0.036 |
ady | +0.036 |
amed | +0.034 |
Neg logit contributions | |
hill | -0.029 |
road | -0.026 |
ladder | -0.026 |
cliff | -0.026 |
taxi | -0.026 |
Token | has |
Token ID | 468 |
Feature activation | +3.08e-03 |
Pos logit contributions | |
lee | +0.034 |
ished | +0.033 |
aming | +0.031 |
amed | +0.031 |
ady | +0.030 |
Neg logit contributions | |
hill | -0.025 |
taxi | -0.024 |
ladder | -0.024 |
road | -0.023 |
ride | -0.023 |
Token | some |
Token ID | 617 |
Feature activation | +5.24e-02 |
Pos logit contributions | |
hill | +0.006 |
road | +0.005 |
ladder | +0.005 |
taxi | +0.005 |
cliff | +0.005 |
Neg logit contributions | |
ished | -0.007 |
lee | -0.007 |
aming | -0.006 |
ady | -0.006 |
amed | -0.006 |
Token | mail |
Token ID | 6920 |
Feature activation | +4.50e-02 |
Pos logit contributions | |
hill | +0.090 |
ladder | +0.083 |
road | +0.083 |
taxi | +0.082 |
cliff | +0.080 |
Neg logit contributions | |
lee | -0.122 |
ished | -0.122 |
aming | -0.116 |
ady | -0.115 |
amed | -0.113 |
Token | . |
Token ID | 13 |
Feature activation | -4.13e-03 |
Pos logit contributions | |
hill | +0.156 |
road | +0.145 |
ladder | +0.144 |
taxi | +0.143 |
cliff | +0.142 |
Neg logit contributions | |
lee | -0.154 |
ished | -0.152 |
aming | -0.143 |
ady | -0.142 |
amed | -0.136 |
Token | The |
Token ID | 383 |
Feature activation | -3.82e-02 |
Pos logit contributions | |
ished | +0.010 |
lee | +0.009 |
aming | +0.009 |
ady | +0.009 |
amed | +0.009 |
Neg logit contributions | |
hill | -0.007 |
road | -0.007 |
ladder | -0.007 |
taxi | -0.006 |
cliff | -0.006 |
Token | water |
Token ID | 1660 |
Feature activation | +2.82e-02 |
Pos logit contributions | |
lee | +0.091 |
ished | +0.090 |
aming | +0.086 |
ady | +0.085 |
amed | +0.083 |
Neg logit contributions | |
hill | -0.062 |
road | -0.056 |
ladder | -0.056 |
taxi | -0.055 |
cliff | -0.055 |
Token | was |
Token ID | 373 |
Feature activation | -4.13e-02 |
Pos logit contributions | |
hill | +0.055 |
ladder | +0.050 |
road | +0.049 |
cliff | +0.049 |
taxi | +0.048 |
Neg logit contributions | |
lee | -0.061 |
ished | -0.061 |
ady | -0.060 |
aming | -0.058 |
amed | -0.055 |
Token | so |
Token ID | 523 |
Feature activation | +3.44e-02 |
Pos logit contributions | |
lee | +0.095 |
ished | +0.095 |
ady | +0.088 |
ins | +0.087 |
ealing | +0.085 |
Neg logit contributions | |
hill | -0.073 |
road | -0.069 |
ladder | -0.068 |
taxi | -0.067 |
cliff | -0.064 |
Token | grim |
Token ID | 18288 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
hill | +0.053 |
ladder | +0.050 |
road | +0.049 |
taxi | +0.047 |
cliff | +0.047 |
Neg logit contributions | |
ished | -0.087 |
lee | -0.085 |
ady | -0.082 |
ice | -0.080 |
ins | -0.079 |
Token | y |
Token ID | 88 |
Feature activation | -5.03e-02 |
Pos logit contributions | |
lee | +0.331 |
pled | +0.297 |
ished | +0.286 |
explains | +0.279 |
teaches | +0.274 |
Neg logit contributions | |
road | -0.446 |
hill | -0.420 |
roof | -0.415 |
taxi | -0.412 |
ladder | -0.406 |
Token | and |
Token ID | 290 |
Feature activation | +2.43e-02 |
Pos logit contributions | |
lee | +0.111 |
ished | +0.109 |
aming | +0.104 |
ady | +0.104 |
amed | +0.103 |
Neg logit contributions | |
hill | -0.096 |
ladder | -0.087 |
road | -0.087 |
mountain | -0.084 |
cliff | -0.084 |
Token | gross |
Token ID | 10319 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
hill | +0.046 |
ladder | +0.043 |
road | +0.042 |
taxi | +0.042 |
cliff | +0.041 |
Neg logit contributions | |
ished | -0.053 |
lee | -0.053 |
ady | -0.051 |
aming | -0.049 |
amed | -0.048 |
Token | , |
Token ID | 11 |
Feature activation | +3.33e-02 |
Pos logit contributions | |
lee | +0.253 |
ady | +0.253 |
amed | +0.246 |
aming | +0.246 |
ished | +0.245 |
Neg logit contributions | |
hill | -0.166 |
ladder | -0.148 |
road | -0.144 |
mountain | -0.139 |
cliff | -0.137 |
Token | but |
Token ID | 475 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
hill | +0.056 |
ladder | +0.052 |
road | +0.051 |
taxi | +0.050 |
cliff | +0.049 |
Neg logit contributions | |
ished | -0.080 |
lee | -0.078 |
ady | -0.076 |
aming | -0.075 |
amed | -0.074 |
Token | The |
Token ID | 464 |
Feature activation | -2.39e-02 |
Pos logit contributions | |
hill | +0.050 |
ladder | +0.046 |
road | +0.046 |
taxi | +0.045 |
cliff | +0.045 |
Neg logit contributions | |
lee | -0.054 |
ished | -0.054 |
aming | -0.053 |
ady | -0.051 |
amed | -0.051 |
Token | brave |
Token ID | 14802 |
Feature activation | -7.67e-02 |
Pos logit contributions | |
lee | +0.043 |
ished | +0.042 |
aming | +0.041 |
ady | +0.040 |
amed | +0.039 |
Neg logit contributions | |
hill | -0.053 |
road | -0.049 |
ladder | -0.049 |
taxi | -0.048 |
cliff | -0.048 |
Token | little |
Token ID | 1310 |
Feature activation | -3.41e-02 |
Pos logit contributions | |
lee | +0.176 |
ished | +0.176 |
aming | +0.166 |
ady | +0.166 |
amed | +0.161 |
Neg logit contributions | |
hill | -0.136 |
road | -0.124 |
ladder | -0.124 |
taxi | -0.122 |
cliff | -0.121 |
Token | girl |
Token ID | 2576 |
Feature activation | -2.65e-02 |
Pos logit contributions | |
lee | +0.067 |
ished | +0.066 |
aming | +0.063 |
ady | +0.061 |
amed | +0.060 |
Neg logit contributions | |
hill | -0.069 |
road | -0.065 |
ladder | -0.064 |
taxi | -0.063 |
cliff | -0.063 |
Token | was |
Token ID | 373 |
Feature activation | -4.90e-02 |
Pos logit contributions | |
ished | +0.056 |
ady | +0.055 |
lee | +0.055 |
aming | +0.053 |
ins | +0.051 |
Neg logit contributions | |
hill | -0.052 |
ride | -0.047 |
ladder | -0.047 |
taxi | -0.046 |
mountain | -0.046 |
Token | el |
Token ID | 1288 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
ished | +0.113 |
lee | +0.110 |
ady | +0.103 |
ins | +0.102 |
ice | +0.101 |
Neg logit contributions | |
taxi | -0.084 |
hill | -0.083 |
ladder | -0.081 |
ride | -0.080 |
cliff | -0.078 |
Token | ated |
Token ID | 515 |
Feature activation | -6.83e-02 |
Pos logit contributions | |
lee | +0.265 |
ished | +0.223 |
ady | +0.202 |
ealing | +0.198 |
oking | +0.197 |
Neg logit contributions | |
hill | -0.546 |
cliff | -0.525 |
road | -0.520 |
taxi | -0.511 |
mountain | -0.493 |
Token | . |
Token ID | 13 |
Feature activation | +2.47e-03 |
Pos logit contributions | |
lee | +0.158 |
ished | +0.156 |
ady | +0.151 |
aming | +0.148 |
ins | +0.143 |
Neg logit contributions | |
hill | -0.119 |
ladder | -0.109 |
taxi | -0.108 |
road | -0.108 |
cliff | -0.104 |
Token | She |
Token ID | 1375 |
Feature activation | +9.07e-04 |
Pos logit contributions | |
hill | +0.004 |
road | +0.004 |
ladder | +0.004 |
taxi | +0.004 |
cliff | +0.004 |
Neg logit contributions | |
lee | -0.006 |
ished | -0.006 |
aming | -0.005 |
ady | -0.005 |
amed | -0.005 |
Token | had |
Token ID | 550 |
Feature activation | -3.61e-03 |
Pos logit contributions | |
hill | +0.002 |
ride | +0.002 |
visit | +0.002 |
road | +0.002 |
ladder | +0.002 |
Neg logit contributions | |
ished | -0.002 |
lee | -0.002 |
ady | -0.002 |
aming | -0.002 |
ins | -0.002 |
Token | managed |
Token ID | 5257 |
Feature activation | -3.03e-02 |
Pos logit contributions | |
ished | +0.007 |
lee | +0.007 |
ady | +0.007 |
aming | +0.007 |
ins | +0.007 |
Neg logit contributions | |
hill | -0.007 |
ride | -0.007 |
visit | -0.006 |
ladder | -0.006 |
road | -0.006 |
Token | view |
Token ID | 1570 |
Feature activation | +7.47e-02 |
Pos logit contributions | |
lee | +0.059 |
ished | +0.059 |
ady | +0.056 |
aming | +0.056 |
amed | +0.054 |
Neg logit contributions | |
hill | -0.037 |
road | -0.033 |
ladder | -0.033 |
taxi | -0.032 |
cliff | -0.032 |
Token | ! |
Token ID | 0 |
Feature activation | +1.53e-02 |
Pos logit contributions | |
hill | +0.144 |
road | +0.134 |
taxi | +0.133 |
ladder | +0.132 |
cliff | +0.130 |
Neg logit contributions | |
lee | -0.154 |
ished | -0.154 |
aming | -0.144 |
ady | -0.142 |
amed | -0.137 |
Token | â |
Token ID | 22940 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
hill | +0.027 |
ladder | +0.025 |
road | +0.025 |
taxi | +0.024 |
cliff | +0.024 |
Neg logit contributions | |
ished | -0.035 |
lee | -0.034 |
aming | -0.033 |
ady | -0.032 |
amed | -0.032 |
Token | € |
Token ID | 26391 |
Feature activation | -9.63e-02 |
Pos logit contributions | |
lee | +0.377 |
ished | +0.376 |
ady | +0.351 |
aming | +0.347 |
ins | +0.341 |
Neg logit contributions | |
hill | -0.333 |
road | -0.306 |
ladder | -0.299 |
cliff | -0.298 |
ride | -0.295 |
Token | exclaimed |
Token ID | 33552 |
Feature activation | -7.96e-02 |
Pos logit contributions | |
lee | +0.218 |
ished | +0.212 |
ady | +0.205 |
aming | +0.204 |
amed | +0.196 |
Neg logit contributions | |
hill | -0.173 |
road | -0.159 |
taxi | -0.158 |
ladder | -0.157 |
cliff | -0.154 |
Token | Maj |
Token ID | 12390 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
ished | +0.193 |
lee | +0.190 |
ady | +0.184 |
aming | +0.183 |
amed | +0.177 |
Neg logit contributions | |
hill | -0.129 |
ladder | -0.117 |
taxi | -0.116 |
road | -0.115 |
cliff | -0.114 |
Token | a |
Token ID | 64 |
Feature activation | -8.45e-02 |
Pos logit contributions | |
lee | +0.372 |
aming | +0.366 |
ished | +0.360 |
amed | +0.335 |
ady | +0.326 |
Neg logit contributions | |
hill | -0.434 |
taxi | -0.414 |
mountain | -0.411 |
ladder | -0.410 |
ride | -0.403 |
Token | . |
Token ID | 13 |
Feature activation | +1.96e-04 |
Pos logit contributions | |
ished | +0.206 |
aming | +0.202 |
lee | +0.200 |
ady | +0.197 |
amed | +0.190 |
Neg logit contributions | |
hill | -0.133 |
ladder | -0.121 |
taxi | -0.119 |
mountain | -0.119 |
cliff | -0.116 |
Token | She |
Token ID | 1375 |
Feature activation | -2.64e-03 |
Pos logit contributions | |
hill | +0.000 |
ladder | +0.000 |
road | +0.000 |
taxi | +0.000 |
cliff | +0.000 |
Neg logit contributions | |
ished | -0.000 |
lee | -0.000 |
aming | -0.000 |
ady | -0.000 |
amed | -0.000 |
Token | looked |
Token ID | 3114 |
Feature activation | -8.34e-02 |
Pos logit contributions | |
ished | +0.006 |
lee | +0.006 |
aming | +0.006 |
ady | +0.006 |
ealing | +0.005 |
Neg logit contributions | |
hill | -0.005 |
ride | -0.004 |
mountain | -0.004 |
ladder | -0.004 |
visit | -0.004 |
Token | up |
Token ID | 510 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
lee | +0.167 |
ished | +0.167 |
ady | +0.161 |
aming | +0.159 |
ins | +0.152 |
Neg logit contributions | |
hill | -0.166 |
ladder | -0.154 |
taxi | -0.154 |
road | -0.152 |
cliff | -0.152 |
Token | Mom |
Token ID | 29252 |
Feature activation | -5.02e-02 |
Pos logit contributions | |
hill | +0.010 |
ladder | +0.009 |
road | +0.009 |
taxi | +0.009 |
cliff | +0.009 |
Neg logit contributions | |
lee | -0.014 |
ished | -0.014 |
aming | -0.014 |
ady | -0.014 |
amed | -0.013 |
Token | my |
Token ID | 1820 |
Feature activation | -8.01e-02 |
Pos logit contributions | |
ished | +0.095 |
lee | +0.092 |
aming | +0.090 |
ady | +0.089 |
amed | +0.086 |
Neg logit contributions | |
hill | -0.109 |
ladder | -0.102 |
taxi | -0.098 |
road | -0.098 |
ride | -0.097 |
Token | , |
Token ID | 11 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
ished | +0.204 |
lee | +0.202 |
aming | +0.198 |
ady | +0.196 |
amed | +0.193 |
Neg logit contributions | |
hill | -0.121 |
ladder | -0.111 |
taxi | -0.106 |
ride | -0.104 |
road | -0.103 |
Token | look |
Token ID | 804 |
Feature activation | -6.41e-02 |
Pos logit contributions | |
hill | +0.065 |
road | +0.064 |
cliff | +0.058 |
distance | +0.058 |
ladder | +0.057 |
Neg logit contributions | |
lee | -0.094 |
ished | -0.089 |
pled | -0.084 |
oking | -0.082 |
aming | -0.081 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
ished | +0.133 |
lee | +0.133 |
aming | +0.126 |
ady | +0.125 |
amed | +0.121 |
Neg logit contributions | |
hill | -0.127 |
ladder | -0.116 |
road | -0.116 |
taxi | -0.115 |
cliff | -0.114 |
Token | N |
Token ID | 399 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
hill | +0.034 |
road | +0.031 |
ladder | +0.031 |
cliff | +0.030 |
taxi | +0.030 |
Neg logit contributions | |
lee | -0.035 |
ished | -0.035 |
ady | -0.032 |
aming | -0.032 |
amed | -0.031 |
Token | ana |
Token ID | 2271 |
Feature activation | -8.64e-02 |
Pos logit contributions | |
ished | +0.343 |
lee | +0.320 |
aming | +0.309 |
amed | +0.305 |
lling | +0.300 |
Neg logit contributions | |
hill | -0.467 |
taxi | -0.457 |
road | -0.450 |
ladder | -0.448 |
church | -0.447 |
Token | said |
Token ID | 531 |
Feature activation | -8.33e-02 |
Pos logit contributions | |
ished | +0.169 |
aming | +0.164 |
ady | +0.162 |
lee | +0.159 |
ealing | +0.153 |
Neg logit contributions | |
hill | -0.163 |
ladder | -0.158 |
ride | -0.157 |
taxi | -0.157 |
walk | -0.154 |
Token | with |
Token ID | 351 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
ished | +0.207 |
lee | +0.201 |
ady | +0.200 |
aming | +0.196 |
ins | +0.192 |
Neg logit contributions | |
hill | -0.123 |
ladder | -0.120 |
taxi | -0.117 |
ride | -0.113 |
visit | -0.111 |
Token | excitement |
Token ID | 14067 |
Feature activation | -6.12e-02 |
Pos logit contributions | |
ished | +0.051 |
lee | +0.049 |
ins | +0.049 |
ice | +0.046 |
ady | +0.046 |
Neg logit contributions | |
hill | -0.031 |
visit | -0.029 |
ride | -0.028 |
road | -0.028 |
ladder | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | +3.29e-03 |
Pos logit contributions | |
ished | +0.152 |
lee | +0.146 |
ady | +0.145 |
amed | +0.145 |
aming | +0.143 |
Neg logit contributions | |
hill | -0.093 |
taxi | -0.087 |
ladder | -0.086 |
ride | -0.085 |
road | -0.084 |
Token | " |
Token ID | 1 |
Feature activation | -1.23e-03 |
Pos logit contributions | |
hill | +0.030 |
ladder | +0.028 |
taxi | +0.027 |
road | +0.027 |
visit | +0.026 |
Neg logit contributions | |
lee | -0.037 |
ished | -0.037 |
ady | -0.032 |
pled | -0.032 |
lling | -0.032 |
Token | I |
Token ID | 40 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
aming | +0.003 |
ished | +0.003 |
ady | +0.003 |
amed | +0.003 |
lee | +0.003 |
Neg logit contributions | |
hill | -0.002 |
mountain | -0.002 |
road | -0.002 |
cliff | -0.002 |
ladder | -0.002 |
Token | want |
Token ID | 765 |
Feature activation | -6.07e-02 |
Pos logit contributions | |
lee | +0.050 |
ished | +0.049 |
aming | +0.047 |
ady | +0.046 |
amed | +0.045 |
Neg logit contributions | |
hill | -0.045 |
ladder | -0.041 |
road | -0.041 |
taxi | -0.041 |
visit | -0.040 |
Token | steak |
Token ID | 26320 |
Feature activation | +6.95e-02 |
Pos logit contributions | |
ished | +0.136 |
lee | +0.136 |
ady | +0.126 |
aming | +0.126 |
ice | +0.122 |
Neg logit contributions | |
hill | -0.108 |
road | -0.099 |
taxi | -0.097 |
ladder | -0.096 |
cliff | -0.096 |
Token | . |
Token ID | 13 |
Feature activation | +7.03e-03 |
Pos logit contributions | |
hill | +0.125 |
road | +0.114 |
ladder | +0.113 |
taxi | +0.112 |
cliff | +0.112 |
Neg logit contributions | |
ished | -0.156 |
lee | -0.156 |
aming | -0.147 |
ady | -0.147 |
amed | -0.142 |
Token | Ste |
Token ID | 2441 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
hill | +0.012 |
road | +0.011 |
cliff | +0.011 |
ladder | +0.011 |
taxi | +0.011 |
Neg logit contributions | |
ished | -0.016 |
lee | -0.016 |
ady | -0.016 |
aming | -0.015 |
amed | -0.015 |
Token | ak |
Token ID | 461 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
lee | +0.378 |
ished | +0.341 |
aming | +0.335 |
lling | +0.327 |
bled | +0.312 |
Neg logit contributions | |
hill | -0.400 |
road | -0.385 |
ride | -0.381 |
ladder | -0.379 |
risk | -0.378 |
Token | is |
Token ID | 318 |
Feature activation | -2.92e-02 |
Pos logit contributions | |
hill | +0.033 |
ladder | +0.031 |
road | +0.030 |
taxi | +0.030 |
cliff | +0.029 |
Neg logit contributions | |
lee | -0.042 |
ished | -0.042 |
aming | -0.041 |
ady | -0.040 |
amed | -0.039 |
Token | good |
Token ID | 922 |
Feature activation | -7.36e-02 |
Pos logit contributions | |
lee | +0.066 |
ished | +0.066 |
aming | +0.062 |
ady | +0.062 |
amed | +0.061 |
Neg logit contributions | |
hill | -0.053 |
ladder | -0.049 |
road | -0.049 |
taxi | -0.048 |
mountain | -0.047 |
Token | . |
Token ID | 13 |
Feature activation | +8.71e-03 |
Pos logit contributions | |
lee | +0.179 |
ished | +0.175 |
aming | +0.169 |
ady | +0.167 |
amed | +0.164 |
Neg logit contributions | |
hill | -0.122 |
road | -0.112 |
ladder | -0.112 |
taxi | -0.109 |
ride | -0.106 |
Token | It |
Token ID | 632 |
Feature activation | +4.24e-02 |
Pos logit contributions | |
hill | +0.014 |
road | +0.012 |
ladder | +0.012 |
cliff | +0.012 |
taxi | +0.012 |
Neg logit contributions | |
lee | -0.021 |
ished | -0.021 |
ady | -0.020 |
aming | -0.020 |
amed | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
ished | +0.081 |
lee | +0.080 |
ady | +0.076 |
ins | +0.072 |
aming | +0.071 |
Neg logit contributions | |
hill | -0.066 |
ladder | -0.060 |
taxi | -0.060 |
cliff | -0.060 |
road | -0.059 |
Token | best |
Token ID | 1266 |
Feature activation | -3.35e-02 |
Pos logit contributions | |
ished | +0.019 |
lee | +0.019 |
aming | +0.018 |
ady | +0.017 |
amed | +0.016 |
Neg logit contributions | |
hill | -0.033 |
ladder | -0.031 |
road | -0.031 |
taxi | -0.031 |
cliff | -0.031 |
Token | at |
Token ID | 379 |
Feature activation | +9.38e-02 |
Pos logit contributions | |
lee | +0.068 |
ished | +0.067 |
ady | +0.064 |
aming | +0.063 |
ins | +0.059 |
Neg logit contributions | |
hill | -0.067 |
road | -0.062 |
ladder | -0.061 |
cliff | -0.060 |
taxi | -0.059 |
Token | sliding |
Token ID | 22292 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
hill | +0.180 |
taxi | +0.173 |
road | +0.171 |
ladder | +0.168 |
subway | +0.167 |
Neg logit contributions | |
lee | -0.183 |
ished | -0.178 |
aming | -0.167 |
amed | -0.160 |
ady | -0.157 |
Token | !" |
Token ID | 2474 |
Feature activation | -8.53e-03 |
Pos logit contributions | |
ished | +0.035 |
lee | +0.035 |
aming | +0.033 |
ady | +0.032 |
amed | +0.032 |
Neg logit contributions | |
hill | -0.030 |
ladder | -0.028 |
road | -0.028 |
taxi | -0.027 |
cliff | -0.027 |
Token | Gr |
Token ID | 1902 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
lee | +0.019 |
ished | +0.019 |
ady | +0.018 |
aming | +0.018 |
amed | +0.018 |
Neg logit contributions | |
hill | -0.015 |
ladder | -0.014 |
road | -0.014 |
cliff | -0.014 |
taxi | -0.014 |
Token | anny |
Token ID | 7737 |
Feature activation | -5.57e-02 |
Pos logit contributions | |
ished | +0.402 |
lee | +0.399 |
pled | +0.389 |
ealing | +0.375 |
aming | +0.373 |
Neg logit contributions | |
hill | -0.322 |
road | -0.317 |
| | -0.288 |
visit | -0.271 |
church | -0.269 |
Token | smiled |
Token ID | 13541 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
ished | +0.098 |
ady | +0.097 |
lee | +0.095 |
aming | +0.095 |
ins | +0.088 |
Neg logit contributions | |
hill | -0.117 |
ride | -0.114 |
taxi | -0.113 |
slide | -0.109 |
ladder | -0.109 |
Token | and |
Token ID | 290 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
ished | +0.420 |
ins | +0.400 |
ady | +0.399 |
ice | +0.395 |
lee | +0.393 |
Neg logit contributions | |
taxi | -0.240 |
hill | -0.240 |
ride | -0.231 |
visit | -0.228 |
walk | -0.220 |
Token | said |
Token ID | 531 |
Feature activation | -8.31e-02 |
Pos logit contributions | |
hill | +0.026 |
ride | +0.025 |
taxi | +0.024 |
ladder | +0.024 |
slide | +0.024 |
Neg logit contributions | |
ished | -0.024 |
ady | -0.024 |
aming | -0.023 |
lee | -0.022 |
ins | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | +3.05e-02 |
Pos logit contributions | |
ished | +0.219 |
ady | +0.217 |
lee | +0.215 |
aming | +0.214 |
amed | +0.202 |
Neg logit contributions | |
hill | -0.114 |
taxi | -0.108 |
ladder | -0.108 |
cliff | -0.099 |
ride | -0.098 |
Token | mas |
Token ID | 12422 |
Feature activation | -7.56e-02 |
Pos logit contributions | |
lee | +0.151 |
ished | +0.150 |
aming | +0.142 |
ady | +0.142 |
amed | +0.137 |
Neg logit contributions | |
hill | -0.109 |
road | -0.099 |
ladder | -0.098 |
taxi | -0.098 |
cliff | -0.096 |
Token | ala |
Token ID | 6081 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
ished | +0.131 |
lee | +0.131 |
aming | +0.120 |
ady | +0.119 |
amed | +0.115 |
Neg logit contributions | |
hill | -0.173 |
road | -0.162 |
ladder | -0.162 |
taxi | -0.161 |
cliff | -0.160 |
Token | and |
Token ID | 290 |
Feature activation | +3.67e-02 |
Pos logit contributions | |
ished | +0.205 |
lee | +0.205 |
aming | +0.194 |
ady | +0.192 |
amed | +0.187 |
Neg logit contributions | |
hill | -0.204 |
ladder | -0.188 |
road | -0.187 |
taxi | -0.186 |
cliff | -0.184 |
Token | r |
Token ID | 374 |
Feature activation | -7.25e-02 |
Pos logit contributions | |
hill | +0.068 |
road | +0.062 |
ladder | +0.062 |
taxi | +0.061 |
cliff | +0.061 |
Neg logit contributions | |
ished | -0.080 |
lee | -0.080 |
aming | -0.076 |
ady | -0.075 |
amed | -0.073 |
Token | as |
Token ID | 292 |
Feature activation | +3.40e-03 |
Pos logit contributions | |
lee | +0.091 |
ished | +0.083 |
ady | +0.078 |
pled | +0.077 |
ealing | +0.077 |
Neg logit contributions | |
hill | -0.178 |
road | -0.170 |
taxi | -0.168 |
ladder | -0.168 |
cliff | -0.165 |
Token | el |
Token ID | 1288 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
hill | +0.006 |
road | +0.006 |
ladder | +0.006 |
taxi | +0.006 |
cliff | +0.006 |
Neg logit contributions | |
lee | -0.007 |
ished | -0.007 |
aming | -0.007 |
ady | -0.007 |
amed | -0.007 |
Token | h |
Token ID | 289 |
Feature activation | -6.51e-02 |
Pos logit contributions | |
lee | +0.258 |
ished | +0.235 |
pled | +0.233 |
oking | +0.206 |
aming | +0.204 |
Neg logit contributions | |
road | -0.493 |
hill | -0.492 |
cliff | -0.478 |
taxi | -0.475 |
flight | -0.469 |
Token | an |
Token ID | 272 |
Feature activation | -8.16e-03 |
Pos logit contributions | |
ished | +0.111 |
lee | +0.108 |
aming | +0.107 |
ady | +0.107 |
amed | +0.104 |
Neg logit contributions | |
hill | -0.134 |
ladder | -0.130 |
taxi | -0.130 |
cliff | -0.127 |
road | -0.127 |
Token | out |
Token ID | 448 |
Feature activation | -8.96e-03 |
Pos logit contributions | |
lee | +0.017 |
ished | +0.017 |
aming | +0.016 |
ady | +0.016 |
amed | +0.015 |
Neg logit contributions | |
hill | -0.016 |
road | -0.015 |
ladder | -0.015 |
taxi | -0.014 |
cliff | -0.014 |
Token | and |
Token ID | 290 |
Feature activation | +4.09e-02 |
Pos logit contributions | |
lee | +0.019 |
ished | +0.019 |
aming | +0.018 |
ady | +0.018 |
amed | +0.017 |
Neg logit contributions | |
hill | -0.018 |
road | -0.016 |
ladder | -0.016 |
taxi | -0.016 |
cliff | -0.016 |
Token | b |
Token ID | 275 |
Feature activation | -8.40e-02 |
Pos logit contributions | |
hill | +0.076 |
road | +0.069 |
ladder | +0.069 |
taxi | +0.068 |
cliff | +0.067 |
Neg logit contributions | |
lee | -0.090 |
ished | -0.090 |
aming | -0.085 |
ady | -0.085 |
amed | -0.082 |
Token | when |
Token ID | 618 |
Feature activation | +7.79e-02 |
Pos logit contributions | |
lee | +0.069 |
ished | +0.069 |
aming | +0.065 |
ady | +0.065 |
amed | +0.062 |
Neg logit contributions | |
hill | -0.058 |
road | -0.053 |
ladder | -0.053 |
taxi | -0.052 |
cliff | -0.052 |
Token | I |
Token ID | 314 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
hill | +0.101 |
road | +0.090 |
ladder | +0.089 |
taxi | +0.088 |
cliff | +0.086 |
Neg logit contributions | |
lee | -0.214 |
ished | -0.211 |
aming | -0.202 |
ady | -0.201 |
amed | -0.195 |
Token | 'm |
Token ID | 1101 |
Feature activation | -3.67e-02 |
Pos logit contributions | |
hill | +0.020 |
road | +0.018 |
ladder | +0.018 |
taxi | +0.017 |
cliff | +0.017 |
Neg logit contributions | |
lee | -0.034 |
ished | -0.034 |
aming | -0.032 |
ady | -0.032 |
amed | -0.031 |
Token | elderly |
Token ID | 13830 |
Feature activation | -6.32e-02 |
Pos logit contributions | |
ished | +0.080 |
lee | +0.079 |
ady | +0.074 |
aming | +0.073 |
amed | +0.071 |
Neg logit contributions | |
hill | -0.068 |
ladder | -0.064 |
taxi | -0.064 |
road | -0.063 |
ride | -0.063 |
Token | ." |
Token ID | 526 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
lee | +0.142 |
ished | +0.142 |
ady | +0.135 |
aming | +0.134 |
amed | +0.131 |
Neg logit contributions | |
hill | -0.116 |
road | -0.106 |
ladder | -0.105 |
taxi | -0.104 |
cliff | -0.102 |
Token | Gr |
Token ID | 1902 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
ished | +0.025 |
lee | +0.025 |
aming | +0.024 |
ady | +0.024 |
amed | +0.023 |
Neg logit contributions | |
hill | -0.017 |
road | -0.015 |
ladder | -0.015 |
taxi | -0.015 |
mountain | -0.015 |
Token | anny |
Token ID | 7737 |
Feature activation | -5.46e-02 |
Pos logit contributions | |
lee | +0.397 |
ished | +0.395 |
pled | +0.377 |
aming | +0.369 |
ealing | +0.368 |
Neg logit contributions | |
hill | -0.331 |
road | -0.321 |
| | -0.289 |
je | -0.279 |
mountain | -0.276 |
Token | laughed |
Token ID | 13818 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
ished | +0.095 |
ady | +0.094 |
lee | +0.093 |
aming | +0.093 |
amed | +0.086 |
Neg logit contributions | |
hill | -0.117 |
ride | -0.114 |
taxi | -0.113 |
ladder | -0.110 |
slide | -0.109 |
Token | and |
Token ID | 290 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
ished | +0.418 |
ady | +0.418 |
lee | +0.411 |
ins | +0.402 |
ice | +0.389 |
Neg logit contributions | |
taxi | -0.217 |
ride | -0.216 |
hill | -0.210 |
while | -0.207 |
visit | -0.203 |
Token | said |
Token ID | 531 |
Feature activation | -8.27e-02 |
Pos logit contributions | |
hill | +0.032 |
ride | +0.030 |
taxi | +0.029 |
ladder | +0.029 |
road | +0.029 |
Neg logit contributions | |
ished | -0.026 |
ady | -0.025 |
aming | -0.025 |
lee | -0.024 |
ins | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | +3.08e-02 |
Pos logit contributions | |
ished | +0.220 |
ady | +0.220 |
aming | +0.218 |
lee | +0.217 |
amed | +0.206 |
Neg logit contributions | |
hill | -0.110 |
ladder | -0.105 |
taxi | -0.105 |
cliff | -0.095 |
ride | -0.094 |
Token | will |
Token ID | 481 |
Feature activation | +2.90e-02 |
Pos logit contributions | |
hill | +0.007 |
ladder | +0.006 |
road | +0.006 |
taxi | +0.006 |
ride | +0.006 |
Neg logit contributions | |
lee | -0.009 |
ished | -0.009 |
aming | -0.008 |
ady | -0.008 |
amed | -0.008 |
Token | come |
Token ID | 1282 |
Feature activation | -5.98e-02 |
Pos logit contributions | |
hill | +0.052 |
cliff | +0.048 |
ladder | +0.047 |
mountain | +0.047 |
subway | +0.044 |
Neg logit contributions | |
lee | -0.059 |
ady | -0.058 |
aming | -0.057 |
ished | -0.055 |
amed | -0.053 |
Token | back |
Token ID | 736 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
ished | +0.110 |
lee | +0.109 |
ady | +0.108 |
aming | +0.107 |
amed | +0.103 |
Neg logit contributions | |
hill | -0.129 |
cliff | -0.124 |
mountain | -0.118 |
ladder | -0.116 |
subway | -0.114 |
Token | soon |
Token ID | 2582 |
Feature activation | -5.63e-03 |
Pos logit contributions | |
hill | +0.023 |
cliff | +0.021 |
ladder | +0.021 |
road | +0.021 |
mountain | +0.021 |
Neg logit contributions | |
lee | -0.023 |
ished | -0.023 |
ady | -0.022 |
aming | -0.022 |
amed | -0.021 |
Token | ." |
Token ID | 526 |
Feature activation | +5.55e-03 |
Pos logit contributions | |
lee | +0.013 |
ished | +0.013 |
ady | +0.012 |
aming | +0.012 |
amed | +0.012 |
Neg logit contributions | |
hill | -0.010 |
road | -0.009 |
ladder | -0.009 |
taxi | -0.009 |
cliff | -0.009 |
Token | N |
Token ID | 399 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
hill | +0.012 |
ladder | +0.011 |
road | +0.011 |
cliff | +0.011 |
taxi | +0.011 |
Neg logit contributions | |
lee | -0.011 |
ished | -0.011 |
aming | -0.010 |
ady | -0.010 |
amed | -0.010 |
Token | ana |
Token ID | 2271 |
Feature activation | -8.45e-02 |
Pos logit contributions | |
ished | +0.352 |
lee | +0.324 |
aming | +0.314 |
lling | +0.311 |
amed | +0.309 |
Neg logit contributions | |
hill | -0.448 |
taxi | -0.435 |
road | -0.432 |
church | -0.429 |
visit | -0.428 |
Token | said |
Token ID | 531 |
Feature activation | -8.32e-02 |
Pos logit contributions | |
ished | +0.149 |
lee | +0.148 |
ady | +0.143 |
aming | +0.143 |
amed | +0.134 |
Neg logit contributions | |
hill | -0.192 |
taxi | -0.183 |
ladder | -0.180 |
ride | -0.179 |
road | -0.177 |
Token | . |
Token ID | 13 |
Feature activation | -7.07e-03 |
Pos logit contributions | |
lee | +0.216 |
ished | +0.216 |
ady | +0.210 |
aming | +0.206 |
amed | +0.199 |
Neg logit contributions | |
taxi | -0.112 |
ladder | -0.111 |
hill | -0.109 |
ride | -0.107 |
walk | -0.104 |
Token |
|
Token ID | 198 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
ished | +0.018 |
lee | +0.018 |
aming | +0.017 |
ady | +0.017 |
amed | +0.017 |
Neg logit contributions | |
hill | -0.010 |
ladder | -0.009 |
road | -0.009 |
taxi | -0.009 |
cliff | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | +1.27e-02 |
Pos logit contributions | |
hill | +0.020 |
road | +0.018 |
ladder | +0.018 |
taxi | +0.017 |
cliff | +0.017 |
Neg logit contributions | |
ished | -0.033 |
lee | -0.033 |
aming | -0.032 |
ady | -0.032 |
amed | -0.031 |
Token | " |
Token ID | 1 |
Feature activation | -7.60e-03 |
Pos logit contributions | |
hill | +0.020 |
road | +0.018 |
ladder | +0.018 |
cliff | +0.018 |
taxi | +0.018 |
Neg logit contributions | |
lee | -0.021 |
ished | -0.021 |
ady | -0.019 |
aming | -0.019 |
amed | -0.019 |
Token | Stop |
Token ID | 19485 |
Feature activation | -6.30e-02 |
Pos logit contributions | |
ished | +0.018 |
lee | +0.017 |
aming | +0.017 |
ady | +0.017 |
amed | +0.017 |
Neg logit contributions | |
hill | -0.013 |
ladder | -0.012 |
road | -0.012 |
taxi | -0.012 |
cliff | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | +4.68e-02 |
Pos logit contributions | |
lee | +0.132 |
ished | +0.129 |
aming | +0.120 |
ady | +0.117 |
pled | +0.116 |
Neg logit contributions | |
hill | -0.122 |
road | -0.115 |
ladder | -0.113 |
taxi | -0.112 |
cliff | -0.108 |
Token | stop |
Token ID | 2245 |
Feature activation | -6.78e-02 |
Pos logit contributions | |
hill | +0.083 |
road | +0.077 |
ladder | +0.076 |
mountain | +0.074 |
visit | +0.073 |
Neg logit contributions | |
ished | -0.102 |
lee | -0.100 |
aming | -0.095 |
ady | -0.093 |
ealing | -0.092 |
Token | !" |
Token ID | 2474 |
Feature activation | +8.79e-03 |
Pos logit contributions | |
lee | +0.145 |
ished | +0.143 |
aming | +0.135 |
ady | +0.133 |
amed | +0.128 |
Neg logit contributions | |
hill | -0.130 |
road | -0.120 |
ladder | -0.120 |
taxi | -0.119 |
cliff | -0.115 |
Token | L |
Token ID | 406 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
hill | +0.023 |
taxi | +0.022 |
road | +0.022 |
ladder | +0.022 |
mountain | +0.021 |
Neg logit contributions | |
lee | -0.012 |
ished | -0.011 |
aming | -0.010 |
ady | -0.010 |
ealing | -0.009 |
Token | ila |
Token ID | 10102 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
ished | +0.375 |
lee | +0.369 |
ady | +0.353 |
aming | +0.352 |
amed | +0.340 |
Neg logit contributions | |
hill | -0.449 |
cliff | -0.422 |
ladder | -0.420 |
taxi | -0.414 |
road | -0.407 |
Token | and |
Token ID | 290 |
Feature activation | +8.41e-03 |
Pos logit contributions | |
lee | +0.246 |
ished | +0.240 |
aming | +0.232 |
ady | +0.227 |
amed | +0.220 |
Neg logit contributions | |
hill | -0.336 |
ladder | -0.315 |
road | -0.312 |
cliff | -0.307 |
taxi | -0.305 |
Token | Ben |
Token ID | 3932 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
hill | +0.022 |
road | +0.021 |
ladder | +0.021 |
taxi | +0.021 |
cliff | +0.020 |
Neg logit contributions | |
lee | -0.012 |
ished | -0.012 |
aming | -0.011 |
ady | -0.010 |
amed | -0.010 |
Token | shout |
Token ID | 9191 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
ished | +0.279 |
lee | +0.276 |
ady | +0.263 |
aming | +0.260 |
amed | +0.253 |
Neg logit contributions | |
hill | -0.216 |
ladder | -0.197 |
road | -0.196 |
mountain | -0.195 |
cliff | -0.194 |
Token | . |
Token ID | 13 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
lee | +0.376 |
ished | +0.374 |
aming | +0.364 |
ady | +0.360 |
amed | +0.360 |
Neg logit contributions | |
hill | -0.189 |
taxi | -0.175 |
road | -0.175 |
walk | -0.175 |
ladder | -0.174 |
Token | turned |
Token ID | 2900 |
Feature activation | -7.63e-02 |
Pos logit contributions | |
ished | +0.007 |
aming | +0.007 |
ady | +0.007 |
lee | +0.007 |
ins | +0.007 |
Neg logit contributions | |
hill | -0.007 |
ride | -0.006 |
mountain | -0.006 |
cliff | -0.006 |
ladder | -0.006 |
Token | to |
Token ID | 284 |
Feature activation | +9.05e-02 |
Pos logit contributions | |
ished | +0.194 |
lee | +0.191 |
ady | +0.187 |
aming | +0.185 |
ins | +0.177 |
Neg logit contributions | |
hill | -0.112 |
taxi | -0.103 |
ladder | -0.102 |
cliff | -0.099 |
road | -0.099 |
Token | his |
Token ID | 465 |
Feature activation | -3.83e-02 |
Pos logit contributions | |
hill | +0.154 |
ladder | +0.140 |
road | +0.139 |
taxi | +0.137 |
cliff | +0.136 |
Neg logit contributions | |
lee | -0.215 |
ished | -0.212 |
ady | -0.201 |
aming | -0.201 |
amed | -0.195 |
Token | mother |
Token ID | 2802 |
Feature activation | -4.98e-03 |
Pos logit contributions | |
lee | +0.069 |
ished | +0.066 |
ady | +0.063 |
aming | +0.062 |
amed | +0.060 |
Neg logit contributions | |
hill | -0.086 |
road | -0.081 |
ladder | -0.080 |
cliff | -0.079 |
taxi | -0.079 |
Token | , |
Token ID | 11 |
Feature activation | +3.90e-02 |
Pos logit contributions | |
ished | +0.010 |
lee | +0.010 |
aming | +0.009 |
ady | +0.009 |
amed | +0.009 |
Neg logit contributions | |
hill | -0.010 |
ladder | -0.010 |
road | -0.010 |
taxi | -0.010 |
cliff | -0.009 |
Token | grinning |
Token ID | 45702 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
hill | +0.073 |
ladder | +0.068 |
taxi | +0.067 |
road | +0.066 |
ride | +0.065 |
Neg logit contributions | |
ished | -0.087 |
lee | -0.083 |
aming | -0.079 |
ady | -0.079 |
amed | -0.076 |
Token | , |
Token ID | 11 |
Feature activation | +2.78e-02 |
Pos logit contributions | |
ished | +0.531 |
ins | +0.494 |
lee | +0.494 |
ice | +0.494 |
oking | +0.488 |
Neg logit contributions | |
taxi | -0.253 |
hill | -0.235 |
visit | -0.231 |
road | -0.231 |
ride | -0.214 |
Token | and |
Token ID | 290 |
Feature activation | +2.86e-02 |
Pos logit contributions | |
hill | +0.058 |
ride | +0.055 |
ladder | +0.055 |
taxi | +0.055 |
road | +0.054 |
Neg logit contributions | |
ished | -0.056 |
lee | -0.050 |
ady | -0.049 |
ins | -0.048 |
aming | -0.048 |
Token | said |
Token ID | 531 |
Feature activation | -8.16e-02 |
Pos logit contributions | |
hill | +0.057 |
taxi | +0.054 |
ladder | +0.053 |
ride | +0.053 |
cliff | +0.052 |
Neg logit contributions | |
ished | -0.059 |
ady | -0.055 |
lee | -0.055 |
aming | -0.053 |
ice | -0.052 |
Token | , |
Token ID | 11 |
Feature activation | +3.13e-02 |
Pos logit contributions | |
ished | +0.196 |
lee | +0.192 |
ady | +0.187 |
aming | +0.186 |
amed | +0.177 |
Neg logit contributions | |
hill | -0.135 |
ladder | -0.125 |
taxi | -0.124 |
road | -0.121 |
cliff | -0.120 |
Token | � |
Token ID | 6184 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
hill | +0.063 |
ladder | +0.058 |
road | +0.058 |
taxi | +0.057 |
cliff | +0.057 |
Neg logit contributions | |
ished | -0.064 |
lee | -0.064 |
aming | -0.060 |
ady | -0.060 |
amed | -0.057 |
Token | a |
Token ID | 257 |
Feature activation | +6.75e-02 |
Pos logit contributions | |
lee | +0.184 |
ished | +0.184 |
aming | +0.174 |
ady | +0.173 |
amed | +0.168 |
Neg logit contributions | |
hill | -0.143 |
road | -0.131 |
ladder | -0.130 |
taxi | -0.128 |
cliff | -0.127 |
Token | break |
Token ID | 2270 |
Feature activation | -8.85e-03 |
Pos logit contributions | |
stack | +0.106 |
Zig | +0.099 |
road | +0.099 |
subway | +0.098 |
hill | +0.097 |
Neg logit contributions | |
lee | -0.137 |
ady | -0.133 |
ished | -0.127 |
sure | -0.123 |
inda | -0.121 |
Token | ". |
Token ID | 1911 |
Feature activation | +1.85e-03 |
Pos logit contributions | |
lee | +0.020 |
ished | +0.020 |
aming | +0.019 |
ady | +0.019 |
amed | +0.018 |
Neg logit contributions | |
hill | -0.016 |
ladder | -0.014 |
road | -0.014 |
cliff | -0.014 |
taxi | -0.014 |
Token | So |
Token ID | 1406 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
hill | +0.003 |
road | +0.003 |
taxi | +0.003 |
ladder | +0.003 |
mountain | +0.003 |
Neg logit contributions | |
lee | -0.004 |
ished | -0.004 |
ady | -0.004 |
aming | -0.004 |
amed | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | +4.75e-02 |
Pos logit contributions | |
hill | +0.029 |
road | +0.027 |
taxi | +0.026 |
ladder | +0.026 |
cliff | +0.025 |
Neg logit contributions | |
lee | -0.034 |
ished | -0.032 |
aming | -0.031 |
ady | -0.030 |
amed | -0.030 |
Token | Maj |
Token ID | 12390 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
hill | +0.081 |
road | +0.075 |
ladder | +0.073 |
taxi | +0.072 |
mountain | +0.069 |
Neg logit contributions | |
lee | -0.114 |
ished | -0.108 |
aming | -0.105 |
ady | -0.103 |
amed | -0.103 |
Token | a |
Token ID | 64 |
Feature activation | -7.97e-02 |
Pos logit contributions | |
lee | +0.368 |
ished | +0.364 |
aming | +0.353 |
ady | +0.331 |
arre | +0.330 |
Neg logit contributions | |
hill | -0.407 |
mountain | -0.390 |
taxi | -0.389 |
ride | -0.385 |
visit | -0.379 |
Token | flew |
Token ID | 13112 |
Feature activation | -5.07e-02 |
Pos logit contributions | |
ished | +0.173 |
ady | +0.172 |
lee | +0.170 |
aming | +0.169 |
ealing | +0.163 |
Neg logit contributions | |
hill | -0.135 |
ladder | -0.127 |
ride | -0.127 |
walk | -0.126 |
mountain | -0.125 |
Token | down |
Token ID | 866 |
Feature activation | +5.50e-02 |
Pos logit contributions | |
lee | +0.112 |
ished | +0.112 |
aming | +0.106 |
ady | +0.106 |
amed | +0.102 |
Neg logit contributions | |
hill | -0.094 |
road | -0.086 |
ladder | -0.086 |
taxi | -0.085 |
cliff | -0.084 |
Token | and |
Token ID | 290 |
Feature activation | +3.00e-02 |
Pos logit contributions | |
hill | +0.115 |
road | +0.110 |
taxi | +0.106 |
subway | +0.105 |
cliff | +0.104 |
Neg logit contributions | |
lee | -0.101 |
ished | -0.099 |
aming | -0.095 |
amed | -0.092 |
ady | -0.088 |
Token | landed |
Token ID | 11406 |
Feature activation | -9.25e-03 |
Pos logit contributions | |
hill | +0.063 |
ladder | +0.059 |
road | +0.058 |
taxi | +0.058 |
cliff | +0.058 |
Neg logit contributions | |
lee | -0.059 |
ished | -0.058 |
ady | -0.055 |
aming | -0.055 |
amed | -0.052 |
Token | ! |
Token ID | 0 |
Feature activation | +6.21e-03 |
Pos logit contributions | |
lee | +0.215 |
ished | +0.214 |
aming | +0.201 |
ady | +0.199 |
amed | +0.193 |
Neg logit contributions | |
hill | -0.192 |
ladder | -0.176 |
road | -0.176 |
taxi | -0.173 |
cliff | -0.173 |
Token |
|
Token ID | 198 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
hill | +0.010 |
ladder | +0.009 |
road | +0.009 |
cliff | +0.009 |
taxi | +0.009 |
Neg logit contributions | |
ished | -0.015 |
lee | -0.015 |
aming | -0.014 |
ady | -0.014 |
amed | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | +2.20e-02 |
Pos logit contributions | |
hill | +0.032 |
road | +0.028 |
ladder | +0.028 |
cliff | +0.028 |
taxi | +0.027 |
Neg logit contributions | |
ished | -0.062 |
lee | -0.061 |
aming | -0.058 |
ady | -0.058 |
amed | -0.056 |
Token | Suddenly |
Token ID | 38582 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
hill | +0.038 |
road | +0.035 |
ladder | +0.034 |
cliff | +0.034 |
taxi | +0.034 |
Neg logit contributions | |
ished | -0.051 |
lee | -0.051 |
ady | -0.048 |
aming | -0.048 |
amed | -0.046 |
Token | , |
Token ID | 11 |
Feature activation | +4.97e-02 |
Pos logit contributions | |
hill | +0.040 |
road | +0.037 |
ladder | +0.036 |
taxi | +0.036 |
cliff | +0.035 |
Neg logit contributions | |
lee | -0.043 |
ished | -0.041 |
aming | -0.039 |
ady | -0.039 |
amed | -0.038 |
Token | Maj |
Token ID | 12390 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
hill | +0.097 |
road | +0.092 |
ladder | +0.087 |
taxi | +0.087 |
mountain | +0.082 |
Neg logit contributions | |
lee | -0.104 |
aming | -0.095 |
ished | -0.094 |
amed | -0.093 |
inda | -0.090 |
Token | a |
Token ID | 64 |
Feature activation | -7.74e-02 |
Pos logit contributions | |
ished | +0.356 |
lee | +0.356 |
aming | +0.339 |
ady | +0.318 |
amed | +0.314 |
Neg logit contributions | |
hill | -0.413 |
mountain | -0.396 |
taxi | -0.394 |
ride | -0.390 |
road | -0.387 |
Token | spotted |
Token ID | 13489 |
Feature activation | -3.29e-02 |
Pos logit contributions | |
ished | +0.177 |
lee | +0.168 |
ady | +0.166 |
aming | +0.162 |
ins | +0.160 |
Neg logit contributions | |
hill | -0.134 |
cliff | -0.122 |
mountain | -0.121 |
ladder | -0.120 |
walk | -0.119 |
Token | a |
Token ID | 257 |
Feature activation | +6.34e-02 |
Pos logit contributions | |
ished | +0.080 |
lee | +0.078 |
ady | +0.075 |
aming | +0.074 |
amed | +0.072 |
Neg logit contributions | |
hill | -0.055 |
cliff | -0.050 |
ladder | -0.050 |
road | -0.049 |
taxi | -0.049 |
Token | rad |
Token ID | 2511 |
Feature activation | +5.48e-02 |
Pos logit contributions | |
hill | +0.108 |
Zig | +0.105 |
road | +0.102 |
taxi | +0.100 |
visit | +0.100 |
Neg logit contributions | |
lee | -0.126 |
aming | -0.120 |
ady | -0.120 |
ished | -0.115 |
amed | -0.113 |
Token | ish |
Token ID | 680 |
Feature activation | +5.76e-02 |
Pos logit contributions | |
hill | +0.159 |
road | +0.155 |
taxi | +0.152 |
subway | +0.152 |
ladder | +0.152 |
Neg logit contributions | |
ished | -0.062 |
lee | -0.059 |
ady | -0.052 |
aming | -0.051 |
amed | -0.047 |
Token | Stay |
Token ID | 25681 |
Feature activation | -9.67e-02 |
Pos logit contributions | |
aming | +0.006 |
ished | +0.006 |
amed | +0.005 |
lee | +0.005 |
ady | +0.005 |
Neg logit contributions | |
hill | -0.005 |
mountain | -0.005 |
ladder | -0.005 |
cliff | -0.005 |
road | -0.005 |
Token | away |
Token ID | 1497 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
ished | +0.210 |
lee | +0.207 |
aming | +0.205 |
ady | +0.202 |
amed | +0.194 |
Neg logit contributions | |
hill | -0.178 |
ladder | -0.163 |
road | -0.161 |
cliff | -0.161 |
mountain | -0.159 |
Token | , |
Token ID | 11 |
Feature activation | +4.08e-02 |
Pos logit contributions | |
hill | +0.056 |
road | +0.052 |
ladder | +0.052 |
taxi | +0.051 |
cliff | +0.051 |
Neg logit contributions | |
lee | -0.059 |
ished | -0.058 |
aming | -0.055 |
ady | -0.055 |
amed | -0.053 |
Token | dog |
Token ID | 3290 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
hill | +0.082 |
cliff | +0.074 |
ladder | +0.074 |
mountain | +0.074 |
road | +0.074 |
Neg logit contributions | |
lee | -0.080 |
ished | -0.078 |
aming | -0.072 |
ady | -0.071 |
ealing | -0.071 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
hill | +0.036 |
road | +0.033 |
cliff | +0.033 |
ladder | +0.033 |
subway | +0.033 |
Neg logit contributions | |
ished | -0.033 |
lee | -0.032 |
ady | -0.030 |
aming | -0.030 |
pled | -0.029 |
Token | L |
Token ID | 406 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
hill | +0.028 |
cliff | +0.026 |
mountain | +0.026 |
ladder | +0.026 |
road | +0.025 |
Neg logit contributions | |
lee | -0.024 |
ished | -0.023 |
aming | -0.022 |
ady | -0.021 |
ealing | -0.020 |
Token | ila |
Token ID | 10102 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
ished | +0.369 |
lee | +0.362 |
aming | +0.347 |
ady | +0.343 |
amed | +0.334 |
Neg logit contributions | |
hill | -0.443 |
ladder | -0.415 |
taxi | -0.413 |
cliff | -0.412 |
road | -0.405 |
Token | says |
Token ID | 1139 |
Feature activation | -8.09e-02 |
Pos logit contributions | |
lee | +0.219 |
ished | +0.214 |
aming | +0.213 |
ady | +0.204 |
amed | +0.200 |
Neg logit contributions | |
hill | -0.352 |
ladder | -0.334 |
road | -0.328 |
taxi | -0.324 |
cliff | -0.321 |
Token | . |
Token ID | 13 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
ished | +0.192 |
lee | +0.190 |
ady | +0.184 |
aming | +0.184 |
amed | +0.177 |
Neg logit contributions | |
hill | -0.134 |
ladder | -0.127 |
taxi | -0.123 |
road | -0.122 |
ride | -0.120 |
Token | She |
Token ID | 1375 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
lee | +0.024 |
ished | +0.023 |
ady | +0.022 |
aming | +0.022 |
amed | +0.022 |
Neg logit contributions | |
hill | -0.026 |
mountain | -0.024 |
cliff | -0.024 |
road | -0.023 |
taxi | -0.023 |
Token | uses |
Token ID | 3544 |
Feature activation | -6.91e-02 |
Pos logit contributions | |
lee | +0.022 |
aming | +0.022 |
ished | +0.021 |
ady | +0.021 |
amed | +0.021 |
Neg logit contributions | |
hill | -0.020 |
ladder | -0.019 |
taxi | -0.018 |
road | -0.018 |
cliff | -0.018 |
Token | rainbow |
Token ID | 27223 |
Feature activation | +4.14e-02 |
Pos logit contributions | |
hill | +0.015 |
road | +0.014 |
ladder | +0.014 |
taxi | +0.014 |
cliff | +0.014 |
Neg logit contributions | |
lee | -0.019 |
ished | -0.019 |
aming | -0.018 |
ady | -0.018 |
amed | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +5.83e-03 |
Pos logit contributions | |
hill | +0.081 |
road | +0.075 |
ladder | +0.074 |
taxi | +0.074 |
cliff | +0.073 |
Neg logit contributions | |
lee | -0.086 |
ished | -0.086 |
aming | -0.080 |
ady | -0.080 |
amed | -0.077 |
Token | They |
Token ID | 1119 |
Feature activation | +6.10e-03 |
Pos logit contributions | |
hill | +0.010 |
road | +0.009 |
ladder | +0.009 |
taxi | +0.009 |
cliff | +0.009 |
Neg logit contributions | |
ished | -0.014 |
lee | -0.014 |
aming | -0.013 |
ady | -0.013 |
amed | -0.013 |
Token | reached |
Token ID | 4251 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
hill | +0.011 |
ride | +0.010 |
walk | +0.010 |
ladder | +0.010 |
visit | +0.010 |
Neg logit contributions | |
ished | -0.014 |
ady | -0.013 |
lee | -0.013 |
ealing | -0.013 |
ins | -0.013 |
Token | their |
Token ID | 511 |
Feature activation | -8.46e-03 |
Pos logit contributions | |
hill | +0.021 |
road | +0.020 |
ladder | +0.019 |
taxi | +0.019 |
cliff | +0.019 |
Neg logit contributions | |
lee | -0.027 |
ished | -0.026 |
ady | -0.025 |
aming | -0.025 |
amed | -0.024 |
Token | shelter |
Token ID | 11772 |
Feature activation | +9.72e-02 |
Pos logit contributions | |
lee | +0.019 |
ished | +0.019 |
aming | +0.018 |
ady | +0.018 |
amed | +0.017 |
Neg logit contributions | |
hill | -0.015 |
road | -0.014 |
ladder | -0.013 |
taxi | -0.013 |
cliff | -0.013 |
Token | soon |
Token ID | 2582 |
Feature activation | +9.36e-03 |
Pos logit contributions | |
hill | +0.201 |
ladder | +0.186 |
road | +0.184 |
taxi | +0.182 |
cliff | +0.181 |
Neg logit contributions | |
ished | -0.195 |
lee | -0.195 |
ady | -0.183 |
aming | -0.182 |
amed | -0.175 |
Token | after |
Token ID | 706 |
Feature activation | +8.16e-02 |
Pos logit contributions | |
hill | +0.017 |
ladder | +0.016 |
road | +0.016 |
cliff | +0.015 |
taxi | +0.015 |
Neg logit contributions | |
ished | -0.021 |
lee | -0.021 |
ady | -0.019 |
aming | -0.019 |
amed | -0.019 |
Token | and |
Token ID | 290 |
Feature activation | +6.03e-02 |
Pos logit contributions | |
hill | +0.144 |
road | +0.131 |
ladder | +0.131 |
taxi | +0.129 |
cliff | +0.128 |
Neg logit contributions | |
lee | -0.188 |
ished | -0.186 |
aming | -0.177 |
ady | -0.176 |
amed | -0.170 |
Token | had |
Token ID | 550 |
Feature activation | +9.26e-03 |
Pos logit contributions | |
hill | +0.106 |
ladder | +0.097 |
road | +0.096 |
taxi | +0.095 |
cliff | +0.094 |
Neg logit contributions | |
lee | -0.138 |
ished | -0.138 |
ady | -0.131 |
aming | -0.131 |
amed | -0.126 |
Token | their |
Token ID | 511 |
Feature activation | +3.01e-03 |
Pos logit contributions | |
hill | +0.014 |
ride | +0.013 |
ladder | +0.013 |
road | +0.013 |
visit | +0.013 |
Neg logit contributions | |
ished | -0.023 |
lee | -0.022 |
ady | -0.022 |
ins | -0.022 |
aming | -0.022 |
Token | a |
Token ID | 257 |
Feature activation | +6.77e-02 |
Pos logit contributions | |
lee | +0.049 |
ished | +0.048 |
ady | +0.046 |
aming | +0.045 |
amed | +0.044 |
Neg logit contributions | |
hill | -0.033 |
ladder | -0.030 |
road | -0.029 |
taxi | -0.029 |
ride | -0.028 |
Token | way |
Token ID | 835 |
Feature activation | +7.76e-02 |
Pos logit contributions | |
hill | +0.128 |
cliff | +0.121 |
road | +0.119 |
ladder | +0.118 |
Zig | +0.118 |
Neg logit contributions | |
lee | -0.129 |
ished | -0.127 |
aming | -0.122 |
ady | -0.122 |
amed | -0.117 |
Token | to |
Token ID | 284 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
hill | +0.101 |
road | +0.087 |
cliff | +0.087 |
mountain | +0.086 |
ladder | +0.085 |
Neg logit contributions | |
lee | -0.214 |
ished | -0.210 |
ady | -0.198 |
aming | -0.198 |
amed | -0.194 |
Token | have |
Token ID | 423 |
Feature activation | +1.81e-03 |
Pos logit contributions | |
hill | +0.155 |
ladder | +0.138 |
cliff | +0.134 |
taxi | +0.133 |
road | +0.133 |
Neg logit contributions | |
lee | -0.371 |
ished | -0.367 |
aming | -0.354 |
ady | -0.348 |
amed | -0.341 |
Token | fun |
Token ID | 1257 |
Feature activation | -5.19e-03 |
Pos logit contributions | |
hill | +0.003 |
ride | +0.003 |
road | +0.003 |
visit | +0.003 |
taxi | +0.003 |
Neg logit contributions | |
ished | -0.004 |
lee | -0.004 |
ady | -0.004 |
aming | -0.004 |
amed | -0.004 |
Token | on |
Token ID | 319 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
ished | +0.010 |
lee | +0.010 |
aming | +0.010 |
ady | +0.010 |
ins | +0.009 |
Neg logit contributions | |
hill | -0.010 |
ladder | -0.010 |
road | -0.010 |
cliff | -0.010 |
taxi | -0.010 |
Token | his |
Token ID | 465 |
Feature activation | +6.58e-03 |
Pos logit contributions | |
hill | +0.212 |
ladder | +0.197 |
road | +0.194 |
taxi | +0.193 |
cliff | +0.192 |
Neg logit contributions | |
lee | -0.266 |
ished | -0.253 |
aming | -0.243 |
ady | -0.234 |
pled | -0.227 |
Token | boring |
Token ID | 14262 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
hill | +0.008 |
ladder | +0.007 |
cliff | +0.007 |
taxi | +0.007 |
road | +0.007 |
Neg logit contributions | |
lee | -0.018 |
ished | -0.018 |
aming | -0.017 |
ady | -0.017 |
amed | -0.016 |
Token | vacation |
Token ID | 14600 |
Feature activation | +8.25e-02 |
Pos logit contributions | |
lee | +0.034 |
ished | +0.034 |
aming | +0.033 |
ady | +0.033 |
amed | +0.032 |
Neg logit contributions | |
hill | -0.011 |
ladder | -0.010 |
cliff | -0.010 |
taxi | -0.010 |
road | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +3.46e-03 |
Pos logit contributions | |
hill | +0.174 |
ladder | +0.168 |
cliff | +0.166 |
taxi | +0.165 |
subway | +0.161 |
Neg logit contributions | |
ished | -0.147 |
lee | -0.144 |
aming | -0.134 |
ady | -0.132 |
amed | -0.128 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -7.94e-03 |
Pos logit contributions | |
hill | +0.006 |
ladder | +0.005 |
road | +0.005 |
taxi | +0.005 |
cliff | +0.005 |
Neg logit contributions | |
lee | -0.009 |
ished | -0.008 |
aming | -0.008 |
ady | -0.008 |
amed | -0.008 |
Token | They |
Token ID | 1119 |
Feature activation | +4.18e-03 |
Pos logit contributions | |
hill | +0.008 |
ladder | +0.008 |
road | +0.008 |
taxi | +0.007 |
visit | +0.007 |
Neg logit contributions | |
ished | -0.013 |
lee | -0.013 |
aming | -0.013 |
ady | -0.013 |
amed | -0.012 |
Token | worked |
Token ID | 3111 |
Feature activation | -7.68e-02 |
Pos logit contributions | |
hill | +0.008 |
ride | +0.008 |
ladder | +0.008 |
walk | +0.007 |
trip | +0.007 |
Neg logit contributions | |
ady | -0.008 |
ished | -0.008 |
lee | -0.008 |
ealing | -0.008 |
aming | -0.008 |
Token | together |
Token ID | 1978 |
Feature activation | +3.41e-02 |
Pos logit contributions | |
lee | +0.176 |
ished | +0.171 |
ady | +0.167 |
aming | +0.162 |
amed | +0.154 |
Neg logit contributions | |
hill | -0.135 |
ladder | -0.125 |
road | -0.121 |
taxi | -0.118 |
ride | -0.116 |
Token | in |
Token ID | 287 |
Feature activation | +8.34e-02 |
Pos logit contributions | |
hill | +0.050 |
road | +0.044 |
ladder | +0.044 |
taxi | +0.043 |
cliff | +0.043 |
Neg logit contributions | |
lee | -0.089 |
ished | -0.089 |
aming | -0.085 |
ady | -0.084 |
amed | -0.082 |
Token | the |
Token ID | 262 |
Feature activation | +2.76e-02 |
Pos logit contributions | |
hill | +0.133 |
ladder | +0.119 |
road | +0.118 |
cliff | +0.114 |
taxi | +0.113 |
Neg logit contributions | |
ished | -0.212 |
lee | -0.211 |
ady | -0.199 |
aming | -0.197 |
amed | -0.194 |
Token | kitchen |
Token ID | 9592 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
hill | +0.049 |
ladder | +0.044 |
road | +0.044 |
taxi | +0.043 |
cliff | +0.042 |
Neg logit contributions | |
lee | -0.065 |
ished | -0.065 |
ady | -0.061 |
aming | -0.061 |
amed | -0.060 |
Token | and |
Token ID | 290 |
Feature activation | +5.39e-02 |
Pos logit contributions | |
hill | +0.207 |
ladder | +0.189 |
road | +0.189 |
mountain | +0.182 |
taxi | +0.181 |
Neg logit contributions | |
lee | -0.232 |
ished | -0.231 |
ady | -0.218 |
aming | -0.217 |
amed | -0.211 |
Token | made |
Token ID | 925 |
Feature activation | -3.93e-02 |
Pos logit contributions | |
hill | +0.113 |
road | +0.104 |
ladder | +0.104 |
taxi | +0.102 |
cliff | +0.102 |
Neg logit contributions | |
lee | -0.107 |
ished | -0.107 |
ady | -0.100 |
aming | -0.100 |
amed | -0.096 |
Token | a |
Token ID | 257 |
Feature activation | +5.64e-02 |
Pos logit contributions | |
lee | +0.092 |
ished | +0.091 |
ady | +0.091 |
aming | +0.087 |
amed | +0.085 |
Neg logit contributions | |
hill | -0.062 |
ladder | -0.060 |
taxi | -0.058 |
road | -0.057 |
visit | -0.057 |
Token | delicious |
Token ID | 12625 |
Feature activation | +8.33e-03 |
Pos logit contributions | |
hill | +0.112 |
Zig | +0.108 |
road | +0.105 |
taxi | +0.104 |
subway | +0.103 |
Neg logit contributions | |
lee | -0.105 |
ished | -0.100 |
aming | -0.099 |
ady | -0.096 |
ins | -0.091 |
Token | cake |
Token ID | 12187 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
hill | +0.014 |
taxi | +0.012 |
road | +0.012 |
Zig | +0.012 |
cliff | +0.012 |
Neg logit contributions | |
lee | -0.019 |
ished | -0.019 |
aming | -0.018 |
ady | -0.018 |
ins | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | +2.35e-02 |
Pos logit contributions | |
hill | +0.021 |
ladder | +0.017 |
visit | +0.016 |
road | +0.016 |
cliff | +0.015 |
Neg logit contributions | |
ished | -0.083 |
lee | -0.076 |
pled | -0.075 |
ice | -0.075 |
aming | -0.075 |
Token | When |
Token ID | 2215 |
Feature activation | +6.87e-02 |
Pos logit contributions | |
hill | +0.032 |
To | +0.030 |
Under | +0.029 |
road | +0.028 |
cliff | +0.027 |
Neg logit contributions | |
ished | -0.062 |
lee | -0.058 |
pled | -0.055 |
ice | -0.054 |
ps | -0.054 |
Token | they |
Token ID | 484 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
hill | +0.098 |
road | +0.087 |
ladder | +0.087 |
taxi | +0.085 |
cliff | +0.084 |
Neg logit contributions | |
lee | -0.182 |
ished | -0.181 |
aming | -0.173 |
ady | -0.171 |
amed | -0.167 |
Token | reached |
Token ID | 4251 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
lee | +0.024 |
ished | +0.024 |
aming | +0.023 |
ady | +0.022 |
amed | +0.022 |
Neg logit contributions | |
hill | -0.020 |
road | -0.019 |
ladder | -0.019 |
taxi | -0.018 |
cliff | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | +8.86e-04 |
Pos logit contributions | |
hill | +0.020 |
road | +0.019 |
taxi | +0.019 |
ladder | +0.018 |
ride | +0.018 |
Neg logit contributions | |
lee | -0.023 |
ished | -0.022 |
aming | -0.021 |
ady | -0.021 |
amed | -0.020 |
Token | bridge |
Token ID | 7696 |
Feature activation | +8.88e-02 |
Pos logit contributions | |
hill | +0.001 |
taxi | +0.001 |
Raven | +0.001 |
ladder | +0.001 |
Zig | +0.001 |
Neg logit contributions | |
lee | -0.002 |
ished | -0.002 |
aming | -0.002 |
ady | -0.002 |
amed | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | +5.35e-02 |
Pos logit contributions | |
hill | +0.164 |
road | +0.153 |
taxi | +0.150 |
cliff | +0.149 |
ladder | +0.149 |
Neg logit contributions | |
lee | -0.186 |
ished | -0.184 |
aming | -0.173 |
ady | -0.169 |
amed | -0.169 |
Token | it |
Token ID | 340 |
Feature activation | +1.27e-02 |
Pos logit contributions | |
hill | +0.073 |
ladder | +0.064 |
cliff | +0.063 |
road | +0.062 |
taxi | +0.061 |
Neg logit contributions | |
ished | -0.146 |
lee | -0.144 |
ady | -0.142 |
aming | -0.139 |
amed | -0.135 |
Token | was |
Token ID | 373 |
Feature activation | -3.97e-02 |
Pos logit contributions | |
hill | +0.024 |
ladder | +0.021 |
road | +0.021 |
mountain | +0.021 |
cliff | +0.021 |
Neg logit contributions | |
lee | -0.029 |
ished | -0.028 |
ady | -0.028 |
aming | -0.027 |
ins | -0.026 |
Token | covered |
Token ID | 5017 |
Feature activation | -6.47e-02 |
Pos logit contributions | |
ished | +0.092 |
lee | +0.091 |
ady | +0.085 |
ins | +0.085 |
aming | +0.081 |
Neg logit contributions | |
hill | -0.072 |
ladder | -0.067 |
road | -0.067 |
taxi | -0.066 |
cliff | -0.063 |
Token | in |
Token ID | 287 |
Feature activation | +7.39e-02 |
Pos logit contributions | |
lee | +0.182 |
ished | +0.176 |
ady | +0.171 |
ealing | +0.165 |
oking | +0.163 |
Neg logit contributions | |
hill | -0.086 |
road | -0.073 |
taxi | -0.072 |
ladder | -0.071 |
cliff | -0.070 |
Token | the |
Token ID | 262 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
hill | +0.087 |
road | +0.078 |
ladder | +0.077 |
taxi | +0.076 |
cliff | +0.076 |
Neg logit contributions | |
ished | -0.139 |
lee | -0.138 |
ady | -0.131 |
aming | -0.130 |
amed | -0.126 |
Token | end |
Token ID | 886 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
hill | +0.030 |
ladder | +0.029 |
Zig | +0.029 |
cliff | +0.028 |
taxi | +0.028 |
Neg logit contributions | |
lee | -0.020 |
aming | -0.020 |
ady | -0.019 |
ished | -0.019 |
amed | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | +5.91e-02 |
Pos logit contributions | |
hill | +0.045 |
road | +0.041 |
ladder | +0.041 |
cliff | +0.040 |
taxi | +0.040 |
Neg logit contributions | |
lee | -0.055 |
ished | -0.053 |
ady | -0.051 |
aming | -0.051 |
amed | -0.049 |
Token | Tom |
Token ID | 4186 |
Feature activation | -8.04e-02 |
Pos logit contributions | |
hill | +0.106 |
cliff | +0.097 |
road | +0.096 |
mountain | +0.093 |
ladder | +0.093 |
Neg logit contributions | |
ished | -0.123 |
lee | -0.119 |
aming | -0.116 |
amed | -0.112 |
pled | -0.111 |
Token | 's |
Token ID | 338 |
Feature activation | -2.19e-02 |
Pos logit contributions | |
ady | +0.166 |
ished | +0.166 |
lee | +0.159 |
ins | +0.157 |
aming | +0.156 |
Neg logit contributions | |
hill | -0.151 |
ride | -0.142 |
mountain | -0.138 |
walk | -0.137 |
visit | -0.137 |
Token | room |
Token ID | 2119 |
Feature activation | +8.98e-02 |
Pos logit contributions | |
ished | +0.049 |
lee | +0.048 |
ady | +0.046 |
aming | +0.046 |
amed | +0.045 |
Neg logit contributions | |
hill | -0.042 |
road | -0.038 |
ladder | -0.038 |
taxi | -0.037 |
cliff | -0.037 |
Token | was |
Token ID | 373 |
Feature activation | -3.93e-02 |
Pos logit contributions | |
hill | +0.153 |
ladder | +0.142 |
road | +0.137 |
taxi | +0.133 |
mountain | +0.131 |
Neg logit contributions | |
ady | -0.211 |
lee | -0.211 |
ished | -0.208 |
aming | -0.203 |
ins | -0.197 |
Token | destroyed |
Token ID | 6572 |
Feature activation | -3.47e-02 |
Pos logit contributions | |
lee | +0.082 |
ished | +0.081 |
ins | +0.076 |
ady | +0.076 |
oking | +0.072 |
Neg logit contributions | |
hill | -0.075 |
taxi | -0.074 |
ladder | -0.073 |
road | -0.072 |
ride | -0.071 |
Token | and |
Token ID | 290 |
Feature activation | +5.16e-02 |
Pos logit contributions | |
lee | +0.073 |
ished | +0.072 |
ady | +0.068 |
aming | +0.067 |
amed | +0.066 |
Neg logit contributions | |
hill | -0.070 |
road | -0.064 |
ladder | -0.063 |
taxi | -0.062 |
cliff | -0.061 |
Token | he |
Token ID | 339 |
Feature activation | +5.49e-03 |
Pos logit contributions | |
hill | +0.074 |
road | +0.065 |
ladder | +0.065 |
taxi | +0.064 |
cliff | +0.063 |
Neg logit contributions | |
ished | -0.136 |
lee | -0.136 |
ady | -0.131 |
aming | -0.130 |
amed | -0.126 |
Token | learned |
Token ID | 4499 |
Feature activation | -5.91e-02 |
Pos logit contributions | |
hill | +0.010 |
road | +0.010 |
ladder | +0.009 |
taxi | +0.009 |
ride | +0.009 |
Neg logit contributions | |
ished | -0.012 |
ady | -0.011 |
lee | -0.011 |
aming | -0.011 |
ins | -0.011 |
Token | started |
Token ID | 2067 |
Feature activation | -3.36e-02 |
Pos logit contributions | |
hill | +0.035 |
ladder | +0.034 |
ride | +0.033 |
walk | +0.032 |
slide | +0.032 |
Neg logit contributions | |
ady | -0.042 |
ished | -0.040 |
lee | -0.040 |
aming | -0.039 |
ealing | -0.037 |
Token | washing |
Token ID | 20518 |
Feature activation | -4.63e-02 |
Pos logit contributions | |
lee | +0.095 |
ished | +0.095 |
aming | +0.091 |
ady | +0.091 |
amed | +0.088 |
Neg logit contributions | |
hill | -0.041 |
road | -0.036 |
ladder | -0.036 |
taxi | -0.035 |
cliff | -0.035 |
Token | the |
Token ID | 262 |
Feature activation | +3.57e-03 |
Pos logit contributions | |
lee | +0.123 |
ady | +0.122 |
ished | +0.121 |
aming | +0.117 |
amed | +0.114 |
Neg logit contributions | |
hill | -0.063 |
ladder | -0.058 |
road | -0.053 |
taxi | -0.051 |
visit | -0.050 |
Token | dishes |
Token ID | 16759 |
Feature activation | +6.49e-02 |
Pos logit contributions | |
hill | +0.007 |
road | +0.007 |
ladder | +0.007 |
mountain | +0.006 |
taxi | +0.006 |
Neg logit contributions | |
lee | -0.008 |
ished | -0.007 |
amed | -0.007 |
aming | -0.007 |
ady | -0.007 |
Token | with |
Token ID | 351 |
Feature activation | -2.75e-04 |
Pos logit contributions | |
hill | +0.115 |
ladder | +0.105 |
road | +0.100 |
mountain | +0.096 |
taxi | +0.096 |
Neg logit contributions | |
lee | -0.158 |
ished | -0.151 |
ady | -0.148 |
aming | -0.144 |
amed | -0.141 |
Token | her |
Token ID | 607 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
ished | +0.001 |
lee | +0.001 |
ady | +0.001 |
aming | +0.001 |
amed | +0.001 |
Neg logit contributions | |
hill | -0.000 |
ladder | -0.000 |
taxi | -0.000 |
road | -0.000 |
ride | -0.000 |
Token | mom |
Token ID | 1995 |
Feature activation | +9.09e-03 |
Pos logit contributions | |
lee | +0.040 |
ished | +0.040 |
aming | +0.038 |
ady | +0.038 |
amed | +0.037 |
Neg logit contributions | |
hill | -0.035 |
ladder | -0.032 |
road | -0.032 |
taxi | -0.031 |
cliff | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | +3.95e-03 |
Pos logit contributions | |
hill | +0.016 |
ladder | +0.014 |
ride | +0.014 |
road | +0.014 |
taxi | +0.014 |
Neg logit contributions | |
lee | -0.021 |
ished | -0.021 |
aming | -0.019 |
ady | -0.019 |
amed | -0.019 |
Token | While |
Token ID | 2893 |
Feature activation | +4.30e-02 |
Pos logit contributions | |
hill | +0.007 |
ladder | +0.006 |
road | +0.006 |
taxi | +0.006 |
visit | +0.006 |
Neg logit contributions | |
ished | -0.010 |
lee | -0.009 |
aming | -0.009 |
ady | -0.009 |
amed | -0.009 |
Token | washing |
Token ID | 20518 |
Feature activation | -3.79e-02 |
Pos logit contributions | |
hill | +0.066 |
road | +0.060 |
cliff | +0.058 |
ladder | +0.058 |
taxi | +0.058 |
Neg logit contributions | |
lee | -0.109 |
ished | -0.106 |
aming | -0.103 |
ady | -0.101 |
amed | -0.099 |
Token | , |
Token ID | 11 |
Feature activation | +5.68e-02 |
Pos logit contributions | |
lee | +0.096 |
ady | +0.095 |
ished | +0.094 |
aming | +0.091 |
amed | +0.087 |
Neg logit contributions | |
hill | -0.059 |
ladder | -0.054 |
taxi | -0.050 |
road | -0.050 |
cliff | -0.050 |
Token | brave |
Token ID | 14802 |
Feature activation | -9.43e-02 |
Pos logit contributions | |
lee | +0.234 |
ished | +0.228 |
ady | +0.213 |
aming | +0.204 |
ice | +0.201 |
Neg logit contributions | |
hill | -0.203 |
taxi | -0.197 |
ride | -0.192 |
road | -0.191 |
ladder | -0.190 |
Token | can |
Token ID | 460 |
Feature activation | +6.82e-02 |
Pos logit contributions | |
lee | +0.233 |
aming | +0.228 |
ished | +0.226 |
ady | +0.221 |
amed | +0.214 |
Neg logit contributions | |
hill | -0.150 |
road | -0.138 |
visit | -0.134 |
ride | -0.132 |
ladder | -0.131 |
Token | bring |
Token ID | 2222 |
Feature activation | -5.08e-02 |
Pos logit contributions | |
hill | +0.109 |
ladder | +0.099 |
road | +0.098 |
taxi | +0.098 |
cliff | +0.096 |
Neg logit contributions | |
lee | -0.165 |
ished | -0.163 |
aming | -0.156 |
ady | -0.155 |
amed | -0.151 |
Token | lots |
Token ID | 6041 |
Feature activation | +7.65e-03 |
Pos logit contributions | |
ished | +0.110 |
lee | +0.110 |
aming | +0.104 |
ady | +0.103 |
amed | +0.100 |
Neg logit contributions | |
hill | -0.094 |
road | -0.087 |
ladder | -0.086 |
taxi | -0.085 |
visit | -0.085 |
Token | of |
Token ID | 286 |
Feature activation | +8.15e-02 |
Pos logit contributions | |
hill | +0.015 |
road | +0.013 |
ladder | +0.013 |
taxi | +0.013 |
cliff | +0.013 |
Neg logit contributions | |
lee | -0.016 |
ished | -0.016 |
aming | -0.015 |
ady | -0.015 |
amed | -0.015 |
Token | rewards |
Token ID | 11530 |
Feature activation | +2.14e-02 |
Pos logit contributions | |
hill | +0.159 |
ladder | +0.145 |
road | +0.145 |
cliff | +0.143 |
taxi | +0.142 |
Neg logit contributions | |
lee | -0.172 |
ished | -0.172 |
ady | -0.162 |
aming | -0.161 |
amed | -0.154 |
Token | . |
Token ID | 13 |
Feature activation | +4.16e-03 |
Pos logit contributions | |
hill | +0.040 |
road | +0.036 |
ladder | +0.036 |
taxi | +0.036 |
cliff | +0.035 |
Neg logit contributions | |
ished | -0.047 |
lee | -0.047 |
aming | -0.044 |
ady | -0.044 |
amed | -0.043 |
Token | |
Token ID | 220 |
Feature activation | -3.26e-02 |
Pos logit contributions | |
hill | +0.005 |
visit | +0.005 |
road | +0.004 |
ladder | +0.004 |
cliff | +0.004 |
Neg logit contributions | |
ished | -0.012 |
lee | -0.011 |
aming | -0.011 |
ady | -0.011 |
ins | -0.011 |
Token | Jill |
Token ID | 22772 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
ished | +0.086 |
lee | +0.086 |
aming | +0.082 |
ady | +0.081 |
amed | +0.079 |
Neg logit contributions | |
hill | -0.046 |
ladder | -0.041 |
road | -0.040 |
taxi | -0.039 |
cliff | -0.039 |
Token | felt |
Token ID | 2936 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
ished | +0.246 |
ady | +0.241 |
lee | +0.237 |
aming | +0.227 |
ealing | +0.223 |
Neg logit contributions | |
hill | -0.199 |
ride | -0.181 |
mountain | -0.177 |
cliff | -0.172 |
visit | -0.171 |
Token | so |
Token ID | 523 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
lee | +0.266 |
ished | +0.266 |
ady | +0.252 |
ins | +0.251 |
ice | +0.243 |
Neg logit contributions | |
ride | -0.198 |
taxi | -0.197 |
visit | -0.196 |
hill | -0.195 |
cliff | -0.185 |
Token | car |
Token ID | 1097 |
Feature activation | +6.49e-02 |
Pos logit contributions | |
ished | +0.004 |
lee | +0.004 |
ady | +0.004 |
aming | +0.004 |
pled | +0.004 |
Neg logit contributions | |
hill | -0.003 |
road | -0.003 |
cliff | -0.003 |
Zig | -0.003 |
taxi | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -6.26e-03 |
Pos logit contributions | |
hill | +0.123 |
cliff | +0.116 |
ladder | +0.112 |
mountain | +0.112 |
road | +0.110 |
Neg logit contributions | |
ished | -0.134 |
lee | -0.127 |
aming | -0.124 |
ady | -0.124 |
amed | -0.120 |
Token | He |
Token ID | 679 |
Feature activation | -1.43e-03 |
Pos logit contributions | |
lee | +0.014 |
ished | +0.014 |
ady | +0.013 |
aming | +0.013 |
amed | +0.013 |
Neg logit contributions | |
hill | -0.011 |
road | -0.010 |
ladder | -0.010 |
cliff | -0.010 |
taxi | -0.010 |
Token | ran |
Token ID | 4966 |
Feature activation | -8.85e-02 |
Pos logit contributions | |
lee | +0.003 |
ished | +0.003 |
aming | +0.003 |
ady | +0.003 |
amed | +0.003 |
Neg logit contributions | |
hill | -0.003 |
ladder | -0.003 |
road | -0.003 |
ride | -0.003 |
taxi | -0.003 |
Token | away |
Token ID | 1497 |
Feature activation | +2.70e-02 |
Pos logit contributions | |
ished | +0.207 |
lee | +0.200 |
amed | +0.190 |
aming | +0.189 |
ady | +0.186 |
Neg logit contributions | |
hill | -0.151 |
cliff | -0.139 |
mountain | -0.136 |
ladder | -0.134 |
road | -0.130 |
Token | and |
Token ID | 290 |
Feature activation | +3.36e-02 |
Pos logit contributions | |
hill | +0.055 |
cliff | +0.050 |
road | +0.050 |
taxi | +0.050 |
ladder | +0.050 |
Neg logit contributions | |
lee | -0.053 |
ished | -0.052 |
ady | -0.050 |
aming | -0.049 |
amed | -0.048 |
Token | hid |
Token ID | 24519 |
Feature activation | -7.67e-02 |
Pos logit contributions | |
hill | +0.077 |
road | +0.072 |
ladder | +0.072 |
cliff | +0.071 |
taxi | +0.071 |
Neg logit contributions | |
lee | -0.059 |
ished | -0.059 |
aming | -0.054 |
ady | -0.053 |
amed | -0.052 |
Token | behind |
Token ID | 2157 |
Feature activation | +5.78e-02 |
Pos logit contributions | |
lee | +0.178 |
ished | +0.178 |
aming | +0.168 |
ady | +0.167 |
amed | +0.162 |
Neg logit contributions | |
hill | -0.134 |
road | -0.122 |
ladder | -0.121 |
taxi | -0.119 |
cliff | -0.119 |
Token | a |
Token ID | 257 |
Feature activation | +7.73e-02 |
Pos logit contributions | |
hill | +0.112 |
ladder | +0.103 |
cliff | +0.103 |
road | +0.101 |
taxi | +0.099 |
Neg logit contributions | |
lee | -0.121 |
ished | -0.118 |
aming | -0.108 |
ady | -0.108 |
pled | -0.104 |
Token | tree |
Token ID | 5509 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
Zig | +0.101 |
risk | +0.099 |
visit | +0.097 |
subway | +0.094 |
Raven | +0.094 |
Neg logit contributions | |
lee | -0.171 |
ady | -0.164 |
aming | -0.158 |
ished | -0.158 |
pled | -0.155 |
Token | . |
Token ID | 13 |
Feature activation | -6.55e-03 |
Pos logit contributions | |
hill | +0.314 |
subway | +0.309 |
cliff | +0.304 |
taxi | +0.304 |
ladder | +0.300 |
Neg logit contributions | |
ished | -0.224 |
ady | -0.211 |
lee | -0.207 |
aming | -0.204 |
pled | -0.194 |
Token | moment |
Token ID | 2589 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
hill | +0.100 |
Zig | +0.096 |
subway | +0.096 |
stack | +0.096 |
cliff | +0.096 |
Neg logit contributions | |
ady | -0.105 |
lee | -0.103 |
sure | -0.098 |
pled | -0.098 |
convinced | -0.098 |
Token | . |
Token ID | 13 |
Feature activation | +1.52e-03 |
Pos logit contributions | |
hill | +0.033 |
ladder | +0.031 |
road | +0.031 |
taxi | +0.030 |
cliff | +0.030 |
Neg logit contributions | |
ished | -0.039 |
lee | -0.039 |
aming | -0.036 |
ady | -0.036 |
amed | -0.035 |
Token | When |
Token ID | 1649 |
Feature activation | +5.39e-02 |
Pos logit contributions | |
hill | +0.003 |
road | +0.003 |
ladder | +0.003 |
cliff | +0.003 |
taxi | +0.003 |
Neg logit contributions | |
lee | -0.003 |
ished | -0.003 |
ady | -0.003 |
aming | -0.003 |
amed | -0.003 |
Token | she |
Token ID | 673 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
hill | +0.111 |
road | +0.105 |
hunt | +0.096 |
cliff | +0.095 |
bel | +0.095 |
Neg logit contributions | |
lee | -0.101 |
aming | -0.082 |
ished | -0.079 |
illy | -0.079 |
sisters | -0.079 |
Token | woke |
Token ID | 19092 |
Feature activation | -4.28e-02 |
Pos logit contributions | |
ished | +0.031 |
lee | +0.031 |
aming | +0.030 |
ady | +0.029 |
amed | +0.028 |
Neg logit contributions | |
hill | -0.028 |
road | -0.027 |
ladder | -0.026 |
taxi | -0.026 |
cliff | -0.026 |
Token | up |
Token ID | 510 |
Feature activation | -6.36e-03 |
Pos logit contributions | |
lee | +0.065 |
ished | +0.063 |
aming | +0.057 |
ady | +0.055 |
amed | +0.052 |
Neg logit contributions | |
hill | -0.111 |
road | -0.104 |
ladder | -0.103 |
cliff | -0.101 |
taxi | -0.100 |
Token | , |
Token ID | 11 |
Feature activation | +4.77e-02 |
Pos logit contributions | |
lee | +0.013 |
ished | +0.012 |
aming | +0.011 |
ady | +0.011 |
pled | +0.011 |
Neg logit contributions | |
hill | -0.013 |
road | -0.012 |
cliff | -0.012 |
ladder | -0.012 |
subway | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
hill | +0.088 |
road | +0.083 |
mountain | +0.077 |
hunt | +0.076 |
cliff | +0.076 |
Neg logit contributions | |
lee | -0.101 |
ished | -0.092 |
aming | -0.090 |
amed | -0.087 |
pled | -0.085 |
Token | box |
Token ID | 3091 |
Feature activation | +9.37e-02 |
Pos logit contributions | |
lee | +0.045 |
ished | +0.044 |
aming | +0.044 |
ady | +0.042 |
amed | +0.041 |
Neg logit contributions | |
hill | -0.030 |
Zig | -0.028 |
ladder | -0.027 |
road | -0.027 |
cliff | -0.027 |
Token | was |
Token ID | 373 |
Feature activation | -3.74e-02 |
Pos logit contributions | |
hill | +0.174 |
ladder | +0.161 |
road | +0.157 |
taxi | +0.156 |
cliff | +0.153 |
Neg logit contributions | |
ished | -0.209 |
lee | -0.208 |
ady | -0.200 |
aming | -0.194 |
ins | -0.189 |
Token | gone |
Token ID | 3750 |
Feature activation | -3.04e-02 |
Pos logit contributions | |
ished | +0.081 |
lee | +0.081 |
ady | +0.076 |
ins | +0.074 |
aming | +0.072 |
Neg logit contributions | |
hill | -0.071 |
ladder | -0.067 |
road | -0.066 |
taxi | -0.066 |
cliff | -0.063 |
Token | the |
Token ID | 262 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
ished | +0.298 |
lee | +0.295 |
ady | +0.287 |
aming | +0.271 |
ice | +0.266 |
Neg logit contributions | |
hill | -0.207 |
taxi | -0.192 |
ladder | -0.192 |
road | -0.189 |
cliff | -0.188 |
Token | wind |
Token ID | 2344 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
lee | +0.049 |
ished | +0.049 |
aming | +0.047 |
ady | +0.046 |
amed | +0.045 |
Neg logit contributions | |
hill | -0.044 |
ladder | -0.040 |
road | -0.040 |
taxi | -0.040 |
cliff | -0.040 |
Token | in |
Token ID | 287 |
Feature activation | +7.71e-02 |
Pos logit contributions | |
ished | +0.058 |
lee | +0.057 |
aming | +0.055 |
ady | +0.054 |
amed | +0.053 |
Neg logit contributions | |
hill | -0.041 |
road | -0.037 |
ladder | -0.037 |
cliff | -0.036 |
taxi | -0.036 |
Token | her |
Token ID | 607 |
Feature activation | -8.66e-03 |
Pos logit contributions | |
hill | +0.164 |
road | +0.152 |
ladder | +0.151 |
taxi | +0.149 |
cliff | +0.149 |
Neg logit contributions | |
ished | -0.151 |
lee | -0.151 |
ady | -0.140 |
aming | -0.140 |
amed | -0.135 |
Token | hair |
Token ID | 4190 |
Feature activation | +4.65e-02 |
Pos logit contributions | |
lee | +0.016 |
ished | +0.016 |
ady | +0.015 |
aming | +0.015 |
amed | +0.014 |
Neg logit contributions | |
hill | -0.019 |
road | -0.018 |
ladder | -0.018 |
cliff | -0.018 |
taxi | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | +3.84e-02 |
Pos logit contributions | |
hill | +0.095 |
ladder | +0.087 |
road | +0.087 |
taxi | +0.086 |
cliff | +0.086 |
Neg logit contributions | |
lee | -0.093 |
ished | -0.093 |
aming | -0.088 |
ady | -0.087 |
amed | -0.084 |
Token | the |
Token ID | 262 |
Feature activation | -1.56e-03 |
Pos logit contributions | |
hill | +0.073 |
taxi | +0.067 |
road | +0.067 |
ladder | +0.067 |
cliff | +0.066 |
Neg logit contributions | |
lee | -0.082 |
ished | -0.081 |
aming | -0.076 |
ady | -0.075 |
amed | -0.074 |
Token | sun |
Token ID | 4252 |
Feature activation | +6.12e-02 |
Pos logit contributions | |
lee | +0.003 |
ished | +0.003 |
aming | +0.003 |
ady | +0.003 |
amed | +0.003 |
Neg logit contributions | |
hill | -0.003 |
taxi | -0.002 |
ladder | -0.002 |
cliff | -0.002 |
road | -0.002 |
Token | on |
Token ID | 319 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
hill | +0.119 |
cliff | +0.111 |
road | +0.109 |
ladder | +0.108 |
taxi | +0.107 |
Neg logit contributions | |
lee | -0.126 |
ished | -0.125 |
aming | -0.118 |
amed | -0.114 |
ady | -0.113 |
Token | her |
Token ID | 607 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
hill | +0.254 |
ladder | +0.236 |
road | +0.233 |
cliff | +0.233 |
mountain | +0.230 |
Neg logit contributions | |
lee | -0.178 |
ished | -0.170 |
aming | -0.162 |
ady | -0.159 |
amed | -0.149 |
Token | face |
Token ID | 1986 |
Feature activation | -1.68e-02 |
Pos logit contributions | |
lee | +0.021 |
ished | +0.020 |
aming | +0.018 |
ady | +0.018 |
amed | +0.018 |
Neg logit contributions | |
hill | -0.025 |
road | -0.023 |
ladder | -0.023 |
taxi | -0.023 |
cliff | -0.023 |
Token | ara |
Token ID | 3301 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
lee | +0.179 |
ished | +0.179 |
ady | +0.168 |
aming | +0.167 |
amed | +0.162 |
Neg logit contributions | |
hill | -0.210 |
ladder | -0.195 |
road | -0.194 |
taxi | -0.193 |
cliff | -0.191 |
Token | was |
Token ID | 373 |
Feature activation | -5.83e-02 |
Pos logit contributions | |
lee | +0.257 |
ished | +0.255 |
aming | +0.249 |
ady | +0.242 |
bled | +0.241 |
Neg logit contributions | |
hill | -0.168 |
ride | -0.155 |
slide | -0.153 |
walk | -0.152 |
bench | -0.149 |
Token | very |
Token ID | 845 |
Feature activation | -1.75e-03 |
Pos logit contributions | |
ished | +0.139 |
lee | +0.138 |
ins | +0.130 |
oking | +0.127 |
ice | +0.125 |
Neg logit contributions | |
taxi | -0.095 |
jail | -0.090 |
ladder | -0.088 |
church | -0.087 |
ride | -0.086 |
Token | sad |
Token ID | 6507 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
ished | +0.004 |
lee | +0.004 |
ice | +0.004 |
pled | +0.004 |
ady | +0.004 |
Neg logit contributions | |
ladder | -0.003 |
taxi | -0.003 |
hill | -0.003 |
cliff | -0.003 |
church | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | +3.05e-02 |
Pos logit contributions | |
lee | +0.239 |
ished | +0.237 |
ady | +0.220 |
aming | +0.220 |
amed | +0.207 |
Neg logit contributions | |
hill | -0.267 |
road | -0.247 |
ladder | -0.247 |
cliff | -0.245 |
taxi | -0.244 |
Token | scared |
Token ID | 12008 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
hill | +0.069 |
ladder | +0.065 |
taxi | +0.065 |
road | +0.065 |
cliff | +0.064 |
Neg logit contributions | |
ished | -0.054 |
ady | -0.053 |
lee | -0.052 |
aming | -0.051 |
ins | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | -5.25e-03 |
Pos logit contributions | |
lee | +0.224 |
ished | +0.223 |
aming | +0.209 |
ady | +0.209 |
amed | +0.201 |
Neg logit contributions | |
hill | -0.217 |
road | -0.200 |
ladder | -0.199 |
taxi | -0.196 |
cliff | -0.196 |
Token | She |
Token ID | 1375 |
Feature activation | +3.94e-04 |
Pos logit contributions | |
ished | +0.013 |
lee | +0.012 |
aming | +0.012 |
ady | +0.012 |
amed | +0.012 |
Neg logit contributions | |
hill | -0.009 |
road | -0.008 |
ladder | -0.008 |
taxi | -0.008 |
cliff | -0.008 |
Token | did |
Token ID | 750 |
Feature activation | -1.49e-02 |
Pos logit contributions | |
hill | +0.001 |
road | +0.001 |
taxi | +0.001 |
ladder | +0.001 |
ride | +0.001 |
Neg logit contributions | |
ished | -0.001 |
aming | -0.001 |
lee | -0.001 |
ady | -0.001 |
ins | -0.001 |
Token | not |
Token ID | 407 |
Feature activation | -5.00e-02 |
Pos logit contributions | |
lee | +0.030 |
ished | +0.029 |
aming | +0.028 |
ady | +0.027 |
amed | +0.026 |
Neg logit contributions | |
hill | -0.031 |
road | -0.028 |
ladder | -0.028 |
cliff | -0.028 |
taxi | -0.028 |
Token | like |
Token ID | 588 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
lee | +0.109 |
ished | +0.108 |
aming | +0.102 |
ady | +0.100 |
amed | +0.099 |
Neg logit contributions | |
hill | -0.095 |
road | -0.089 |
ride | -0.087 |
taxi | -0.087 |
ladder | -0.087 |
Token | dolphins |
Token ID | 36798 |
Feature activation | +2.50e-02 |
Pos logit contributions | |
hill | +0.039 |
road | +0.036 |
ladder | +0.036 |
taxi | +0.035 |
cliff | +0.035 |
Neg logit contributions | |
ished | -0.043 |
lee | -0.043 |
aming | -0.041 |
ady | -0.041 |
amed | -0.039 |
Token | and |
Token ID | 290 |
Feature activation | +4.34e-02 |
Pos logit contributions | |
hill | +0.047 |
road | +0.043 |
ladder | +0.043 |
taxi | +0.042 |
cliff | +0.042 |
Neg logit contributions | |
ished | -0.054 |
lee | -0.054 |
ady | -0.051 |
aming | -0.051 |
amed | -0.049 |
Token | how |
Token ID | 703 |
Feature activation | +4.52e-02 |
Pos logit contributions | |
hill | +0.084 |
road | +0.077 |
ladder | +0.077 |
cliff | +0.076 |
taxi | +0.075 |
Neg logit contributions | |
ished | -0.093 |
lee | -0.091 |
aming | -0.088 |
ady | -0.088 |
amed | -0.084 |
Token | they |
Token ID | 484 |
Feature activation | -3.16e-03 |
Pos logit contributions | |
hill | +0.064 |
ladder | +0.057 |
road | +0.057 |
taxi | +0.055 |
cliff | +0.055 |
Neg logit contributions | |
ished | -0.119 |
lee | -0.119 |
ady | -0.114 |
aming | -0.114 |
amed | -0.110 |
Token | can |
Token ID | 460 |
Feature activation | +6.65e-02 |
Pos logit contributions | |
lee | +0.006 |
ished | +0.006 |
ady | +0.006 |
aming | +0.006 |
amed | +0.006 |
Neg logit contributions | |
hill | -0.006 |
road | -0.006 |
ladder | -0.006 |
cliff | -0.006 |
taxi | -0.006 |
Token | be |
Token ID | 307 |
Feature activation | -6.83e-02 |
Pos logit contributions | |
hill | +0.101 |
ladder | +0.093 |
taxi | +0.090 |
road | +0.090 |
subway | +0.089 |
Neg logit contributions | |
lee | -0.165 |
ished | -0.160 |
aming | -0.155 |
ady | -0.155 |
amed | -0.150 |
Token | friends |
Token ID | 2460 |
Feature activation | +5.89e-03 |
Pos logit contributions | |
ished | +0.143 |
lee | +0.139 |
ady | +0.131 |
aming | +0.130 |
amed | +0.125 |
Neg logit contributions | |
hill | -0.130 |
visit | -0.122 |
road | -0.122 |
ride | -0.121 |
taxi | -0.121 |
Token | with |
Token ID | 351 |
Feature activation | -2.31e-03 |
Pos logit contributions | |
hill | +0.012 |
road | +0.011 |
ladder | +0.011 |
cliff | +0.011 |
taxi | +0.011 |
Neg logit contributions | |
lee | -0.012 |
ished | -0.012 |
aming | -0.011 |
ady | -0.011 |
amed | -0.011 |
Token | humans |
Token ID | 5384 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
ished | +0.005 |
lee | +0.005 |
aming | +0.005 |
ady | +0.005 |
amed | +0.005 |
Neg logit contributions | |
hill | -0.004 |
road | -0.004 |
ladder | -0.004 |
taxi | -0.004 |
cliff | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +8.15e-04 |
Pos logit contributions | |
hill | +0.031 |
road | +0.029 |
ladder | +0.028 |
taxi | +0.028 |
cliff | +0.028 |
Neg logit contributions | |
lee | -0.034 |
ished | -0.034 |
aming | -0.032 |
ady | -0.032 |
amed | -0.031 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
hill | +0.001 |
road | +0.001 |
ladder | +0.001 |
visit | +0.001 |
taxi | +0.001 |
Neg logit contributions | |
ished | -0.002 |
lee | -0.002 |
aming | -0.002 |
ady | -0.002 |
amed | -0.002 |
Token | The |
Token ID | 383 |
Feature activation | -3.22e-02 |
Pos logit contributions | |
hill | +0.011 |
ladder | +0.010 |
road | +0.010 |
cliff | +0.010 |
taxi | +0.010 |
Neg logit contributions | |
lee | -0.014 |
ished | -0.013 |
ady | -0.013 |
aming | -0.013 |
amed | -0.012 |
Token | castle |
Token ID | 16669 |
Feature activation | +7.63e-02 |
Pos logit contributions | |
lee | +0.070 |
ished | +0.068 |
aming | +0.067 |
ady | +0.066 |
amed | +0.063 |
Neg logit contributions | |
hill | -0.052 |
ladder | -0.047 |
Zig | -0.047 |
road | -0.047 |
cliff | -0.046 |
Token | was |
Token ID | 373 |
Feature activation | -4.58e-02 |
Pos logit contributions | |
hill | +0.154 |
road | +0.141 |
ladder | +0.141 |
taxi | +0.139 |
cliff | +0.139 |
Neg logit contributions | |
lee | -0.155 |
ished | -0.155 |
aming | -0.145 |
ady | -0.145 |
amed | -0.140 |
Token | empty |
Token ID | 6565 |
Feature activation | -2.63e-02 |
Pos logit contributions | |
lee | +0.093 |
ished | +0.093 |
ady | +0.087 |
aming | +0.086 |
amed | +0.083 |
Neg logit contributions | |
hill | -0.093 |
ladder | -0.086 |
road | -0.086 |
taxi | -0.085 |
cliff | -0.084 |
Token | without |
Token ID | 1231 |
Feature activation | -4.70e-02 |
Pos logit contributions | |
lee | +0.052 |
ished | +0.052 |
aming | +0.049 |
ady | +0.048 |
amed | +0.046 |
Neg logit contributions | |
hill | -0.053 |
ladder | -0.049 |
road | -0.049 |
taxi | -0.049 |
cliff | -0.049 |
Token | her |
Token ID | 607 |
Feature activation | -4.02e-02 |
Pos logit contributions | |
ished | +0.106 |
lee | +0.105 |
ady | +0.100 |
aming | +0.100 |
amed | +0.097 |
Neg logit contributions | |
hill | -0.084 |
road | -0.078 |
ladder | -0.077 |
taxi | -0.076 |
cliff | -0.076 |
Token | and |
Token ID | 290 |
Feature activation | +4.97e-02 |
Pos logit contributions | |
lee | +0.091 |
ished | +0.087 |
ady | +0.083 |
aming | +0.081 |
amed | +0.077 |
Neg logit contributions | |
hill | -0.070 |
road | -0.063 |
cliff | -0.063 |
taxi | -0.062 |
ladder | -0.062 |
Token | everyone |
Token ID | 2506 |
Feature activation | -4.66e-02 |
Pos logit contributions | |
hill | +0.092 |
road | +0.084 |
ladder | +0.083 |
taxi | +0.083 |
cliff | +0.082 |
Neg logit contributions | |
lee | -0.110 |
ished | -0.109 |
aming | -0.102 |
ady | -0.101 |
amed | -0.098 |
Token | was |
Token ID | 373 |
Feature activation | -4.03e-02 |
Pos logit contributions | |
lee | +0.098 |
ished | +0.098 |
aming | +0.091 |
ady | +0.091 |
amed | +0.088 |
Neg logit contributions | |
hill | -0.091 |
road | -0.084 |
ladder | -0.083 |
taxi | -0.082 |
cliff | -0.082 |
Token | very |
Token ID | 845 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
ished | +0.085 |
lee | +0.083 |
ady | +0.076 |
aming | +0.074 |
ice | +0.074 |
Neg logit contributions | |
taxi | -0.076 |
hill | -0.076 |
ladder | -0.074 |
road | -0.072 |
ride | -0.071 |
Token | sad |
Token ID | 6507 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
hill | +0.028 |
ladder | +0.027 |
taxi | +0.027 |
road | +0.027 |
cliff | +0.026 |
Neg logit contributions | |
ished | -0.025 |
lee | -0.024 |
ady | -0.023 |
aming | -0.022 |
amed | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
ished | +0.230 |
lee | +0.229 |
ady | +0.217 |
aming | +0.217 |
amed | +0.210 |
Neg logit contributions | |
hill | -0.206 |
road | -0.190 |
ladder | -0.189 |
taxi | -0.187 |
cliff | -0.184 |
Token |
|
Token ID | 198 |
Feature activation | +9.67e-03 |
Pos logit contributions | |
lee | +0.031 |
ished | +0.030 |
ady | +0.029 |
aming | +0.028 |
amed | +0.027 |
Neg logit contributions | |
hill | -0.026 |
cliff | -0.024 |
road | -0.024 |
taxi | -0.024 |
ladder | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | +8.85e-03 |
Pos logit contributions | |
hill | +0.015 |
road | +0.014 |
ladder | +0.014 |
taxi | +0.013 |
cliff | +0.013 |
Neg logit contributions | |
lee | -0.024 |
ished | -0.024 |
aming | -0.023 |
ady | -0.023 |
amed | -0.022 |
Token | Mom |
Token ID | 29252 |
Feature activation | -4.77e-02 |
Pos logit contributions | |
hill | +0.018 |
road | +0.017 |
ladder | +0.017 |
taxi | +0.017 |
cliff | +0.017 |
Neg logit contributions | |
lee | -0.018 |
ished | -0.018 |
aming | -0.017 |
ady | -0.017 |
amed | -0.016 |
Token | hears |
Token ID | 23180 |
Feature activation | -4.20e-02 |
Pos logit contributions | |
lee | +0.099 |
ished | +0.099 |
ady | +0.095 |
aming | +0.094 |
amed | +0.091 |
Neg logit contributions | |
hill | -0.094 |
ladder | -0.086 |
road | -0.085 |
taxi | -0.084 |
ride | -0.082 |
Token | Lily |
Token ID | 20037 |
Feature activation | -9.42e-02 |
Pos logit contributions | |
ished | +0.095 |
lee | +0.094 |
ady | +0.090 |
aming | +0.090 |
amed | +0.087 |
Neg logit contributions | |
hill | -0.076 |
road | -0.069 |
ladder | -0.069 |
taxi | -0.068 |
cliff | -0.067 |
Token | cry |
Token ID | 3960 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
lee | +0.185 |
ished | +0.183 |
aming | +0.172 |
ady | +0.170 |
amed | +0.167 |
Neg logit contributions | |
hill | -0.198 |
road | -0.185 |
ladder | -0.184 |
taxi | -0.182 |
ride | -0.180 |
Token | . |
Token ID | 13 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
ished | +0.306 |
lee | +0.304 |
ady | +0.289 |
aming | +0.286 |
amed | +0.281 |
Neg logit contributions | |
hill | -0.279 |
road | -0.261 |
taxi | -0.253 |
ladder | -0.253 |
ride | -0.245 |
Token | She |
Token ID | 1375 |
Feature activation | +1.64e-03 |
Pos logit contributions | |
lee | +0.028 |
ished | +0.027 |
aming | +0.026 |
ady | +0.026 |
amed | +0.025 |
Neg logit contributions | |
hill | -0.023 |
road | -0.021 |
ladder | -0.021 |
cliff | -0.020 |
taxi | -0.020 |
Token | runs |
Token ID | 4539 |
Feature activation | -9.07e-02 |
Pos logit contributions | |
hill | +0.003 |
ladder | +0.003 |
taxi | +0.003 |
road | +0.003 |
visit | +0.003 |
Neg logit contributions | |
ished | -0.003 |
lee | -0.003 |
aming | -0.003 |
ady | -0.003 |
amed | -0.003 |
Token | to |
Token ID | 284 |
Feature activation | +9.22e-02 |
Pos logit contributions | |
ished | +0.237 |
lee | +0.225 |
amed | +0.221 |
aming | +0.218 |
ady | +0.212 |
Neg logit contributions | |
hill | -0.134 |
cliff | -0.124 |
mountain | -0.122 |
ladder | -0.111 |
road | -0.108 |
Token | now |
Token ID | 783 |
Feature activation | +2.79e-02 |
Pos logit contributions | |
ished | +0.058 |
lee | +0.056 |
aming | +0.054 |
ady | +0.054 |
amed | +0.050 |
Neg logit contributions | |
hill | -0.057 |
road | -0.053 |
taxi | -0.053 |
ladder | -0.052 |
subway | -0.052 |
Token | !" |
Token ID | 2474 |
Feature activation | +2.81e-03 |
Pos logit contributions | |
hill | +0.056 |
road | +0.051 |
ladder | +0.051 |
taxi | +0.051 |
cliff | +0.050 |
Neg logit contributions | |
ished | -0.056 |
lee | -0.056 |
ady | -0.053 |
aming | -0.053 |
amed | -0.050 |
Token |
|
Token ID | 198 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
hill | +0.005 |
cliff | +0.005 |
mountain | +0.005 |
road | +0.005 |
taxi | +0.005 |
Neg logit contributions | |
lee | -0.006 |
ished | -0.006 |
ady | -0.006 |
aming | -0.006 |
amed | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
hill | +0.027 |
road | +0.024 |
ladder | +0.024 |
taxi | +0.024 |
cliff | +0.024 |
Neg logit contributions | |
ished | -0.045 |
lee | -0.045 |
aming | -0.043 |
ady | -0.043 |
amed | -0.041 |
Token | Then |
Token ID | 6423 |
Feature activation | +3.04e-02 |
Pos logit contributions | |
hill | +0.035 |
road | +0.032 |
ladder | +0.032 |
taxi | +0.031 |
cliff | +0.031 |
Neg logit contributions | |
lee | -0.041 |
ished | -0.040 |
ady | -0.038 |
aming | -0.038 |
amed | -0.037 |
Token | Lily |
Token ID | 20037 |
Feature activation | -9.87e-02 |
Pos logit contributions | |
hill | +0.061 |
road | +0.057 |
cliff | +0.055 |
ladder | +0.054 |
mountain | +0.054 |
Neg logit contributions | |
lee | -0.063 |
ished | -0.060 |
aming | -0.057 |
amed | -0.055 |
ady | -0.055 |
Token | heard |
Token ID | 2982 |
Feature activation | -2.80e-02 |
Pos logit contributions | |
ished | +0.217 |
lee | +0.207 |
ins | +0.207 |
aming | +0.204 |
ady | +0.202 |
Neg logit contributions | |
hill | -0.181 |
ride | -0.167 |
walk | -0.163 |
road | -0.163 |
ladder | -0.162 |
Token | a |
Token ID | 257 |
Feature activation | +6.08e-02 |
Pos logit contributions | |
lee | +0.060 |
ished | +0.059 |
aming | +0.056 |
ady | +0.056 |
amed | +0.054 |
Neg logit contributions | |
hill | -0.055 |
road | -0.050 |
ladder | -0.050 |
taxi | -0.049 |
cliff | -0.049 |
Token | voice |
Token ID | 3809 |
Feature activation | +1.24e-03 |
Pos logit contributions | |
hill | +0.119 |
road | +0.112 |
cliff | +0.110 |
ladder | +0.109 |
taxi | +0.107 |
Neg logit contributions | |
lee | -0.119 |
aming | -0.106 |
ady | -0.103 |
ished | -0.103 |
ealing | -0.100 |
Token | from |
Token ID | 422 |
Feature activation | +3.34e-02 |
Pos logit contributions | |
hill | +0.003 |
road | +0.002 |
ladder | +0.002 |
cliff | +0.002 |
taxi | +0.002 |
Neg logit contributions | |
lee | -0.003 |
ished | -0.002 |
aming | -0.002 |
ady | -0.002 |
amed | -0.002 |
Token | the |
Token ID | 262 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
hill | +0.063 |
ladder | +0.058 |
road | +0.057 |
mountain | +0.056 |
cliff | +0.055 |
Neg logit contributions | |
lee | -0.070 |
aming | -0.069 |
ished | -0.067 |
ady | -0.064 |
amed | -0.064 |
Token | s |
Token ID | 82 |
Feature activation | -3.04e-02 |
Pos logit contributions | |
ished | +0.016 |
lee | +0.015 |
ady | +0.014 |
aming | +0.014 |
amed | +0.014 |
Neg logit contributions | |
hill | -0.019 |
road | -0.017 |
ladder | -0.017 |
cliff | -0.017 |
taxi | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | +4.18e-02 |
Pos logit contributions | |
lee | +0.063 |
ished | +0.063 |
aming | +0.059 |
ady | +0.059 |
amed | +0.057 |
Neg logit contributions | |
hill | -0.060 |
road | -0.055 |
ladder | -0.055 |
taxi | -0.054 |
cliff | -0.054 |
Token | make |
Token ID | 787 |
Feature activation | -4.98e-02 |
Pos logit contributions | |
hill | +0.071 |
cliff | +0.064 |
ladder | +0.062 |
road | +0.062 |
mountain | +0.061 |
Neg logit contributions | |
lee | -0.096 |
ished | -0.089 |
ady | -0.085 |
amed | -0.083 |
sisters | -0.082 |
Token | each |
Token ID | 1123 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
ished | +0.107 |
lee | +0.107 |
aming | +0.104 |
ady | +0.103 |
amed | +0.099 |
Neg logit contributions | |
hill | -0.091 |
taxi | -0.087 |
road | -0.085 |
ladder | -0.085 |
ride | -0.083 |
Token | other |
Token ID | 584 |
Feature activation | -2.62e-02 |
Pos logit contributions | |
lee | +0.090 |
ished | +0.089 |
aming | +0.084 |
ady | +0.082 |
amed | +0.081 |
Neg logit contributions | |
hill | -0.084 |
road | -0.078 |
ladder | -0.077 |
taxi | -0.077 |
cliff | -0.075 |
Token | laugh |
Token ID | 6487 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
ished | +0.057 |
lee | +0.057 |
aming | +0.054 |
ady | +0.054 |
amed | +0.052 |
Neg logit contributions | |
hill | -0.050 |
road | -0.045 |
taxi | -0.045 |
ladder | -0.045 |
ride | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | -5.56e-03 |
Pos logit contributions | |
lee | +0.328 |
ished | +0.321 |
ady | +0.310 |
aming | +0.302 |
amed | +0.288 |
Neg logit contributions | |
hill | -0.207 |
road | -0.189 |
ride | -0.186 |
taxi | -0.186 |
visit | -0.182 |
Token | They |
Token ID | 1119 |
Feature activation | -2.89e-03 |
Pos logit contributions | |
ished | +0.013 |
lee | +0.013 |
aming | +0.013 |
ady | +0.013 |
amed | +0.012 |
Neg logit contributions | |
hill | -0.009 |
road | -0.008 |
ladder | -0.008 |
taxi | -0.008 |
cliff | -0.008 |
Token | had |
Token ID | 550 |
Feature activation | -1.35e-03 |
Pos logit contributions | |
ished | +0.006 |
lee | +0.006 |
aming | +0.006 |
ady | +0.006 |
amed | +0.005 |
Neg logit contributions | |
hill | -0.006 |
road | -0.005 |
taxi | -0.005 |
ladder | -0.005 |
ride | -0.005 |
Token | fun |
Token ID | 1257 |
Feature activation | -3.44e-02 |
Pos logit contributions | |
ished | +0.003 |
lee | +0.003 |
ady | +0.003 |
aming | +0.003 |
ins | +0.003 |
Neg logit contributions | |
hill | -0.002 |
road | -0.002 |
ride | -0.002 |
ladder | -0.002 |
visit | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | +4.92e-02 |
Pos logit contributions | |
lee | +0.069 |
ished | +0.069 |
aming | +0.064 |
ady | +0.064 |
amed | +0.062 |
Neg logit contributions | |
hill | -0.072 |
road | -0.066 |
ladder | -0.066 |
taxi | -0.065 |
cliff | -0.065 |
Token |
|
Token ID | 198 |
Feature activation | +2.51e-02 |
Pos logit contributions | |
hill | +0.011 |
ladder | +0.010 |
road | +0.010 |
taxi | +0.009 |
cliff | +0.009 |
Neg logit contributions | |
lee | -0.014 |
ished | -0.014 |
aming | -0.014 |
ady | -0.014 |
amed | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | +2.31e-02 |
Pos logit contributions | |
hill | +0.037 |
road | +0.033 |
ladder | +0.033 |
taxi | +0.033 |
cliff | +0.032 |
Neg logit contributions | |
lee | -0.065 |
ished | -0.065 |
aming | -0.061 |
ady | -0.061 |
amed | -0.060 |
Token | " |
Token ID | 1 |
Feature activation | +2.05e-03 |
Pos logit contributions | |
hill | +0.043 |
road | +0.039 |
ladder | +0.038 |
taxi | +0.038 |
cliff | +0.037 |
Neg logit contributions | |
lee | -0.054 |
ished | -0.053 |
ady | -0.047 |
aming | -0.046 |
pled | -0.046 |
Token | Come |
Token ID | 16773 |
Feature activation | -9.59e-02 |
Pos logit contributions | |
hill | +0.003 |
ladder | +0.003 |
road | +0.003 |
cliff | +0.003 |
mountain | +0.003 |
Neg logit contributions | |
ished | -0.005 |
lee | -0.005 |
aming | -0.005 |
ady | -0.005 |
amed | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | +3.71e-02 |
Pos logit contributions | |
ished | +0.178 |
lee | +0.177 |
aming | +0.174 |
amed | +0.167 |
ady | +0.165 |
Neg logit contributions | |
hill | -0.191 |
road | -0.186 |
cliff | -0.181 |
mountain | -0.178 |
ladder | -0.175 |
Token | Ben |
Token ID | 3932 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
hill | +0.075 |
road | +0.070 |
ladder | +0.069 |
taxi | +0.068 |
cliff | +0.067 |
Neg logit contributions | |
lee | -0.074 |
ished | -0.074 |
aming | -0.068 |
ady | -0.067 |
amed | -0.065 |
Token | , |
Token ID | 11 |
Feature activation | +4.51e-02 |
Pos logit contributions | |
lee | +0.306 |
ished | +0.306 |
aming | +0.290 |
ady | +0.287 |
amed | +0.284 |
Neg logit contributions | |
hill | -0.216 |
road | -0.194 |
ladder | -0.194 |
taxi | -0.189 |
cliff | -0.187 |
Token | let |
Token ID | 1309 |
Feature activation | -2.61e-02 |
Pos logit contributions | |
hill | +0.091 |
road | +0.087 |
taxi | +0.083 |
ladder | +0.082 |
mountain | +0.082 |
Neg logit contributions | |
ished | -0.085 |
lee | -0.084 |
aming | -0.076 |
pled | -0.076 |
oking | -0.075 |
Token | 's |
Token ID | 338 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
lee | +0.046 |
ished | +0.045 |
ady | +0.043 |
aming | +0.043 |
amed | +0.042 |
Neg logit contributions | |
hill | -0.060 |
taxi | -0.057 |
ride | -0.056 |
visit | -0.056 |
road | -0.055 |
Token | go |
Token ID | 467 |
Feature activation | -1.78e-02 |
Pos logit contributions | |
lee | +0.083 |
ished | +0.078 |
ady | +0.076 |
aming | +0.076 |
amed | +0.072 |
Neg logit contributions | |
hill | -0.052 |
ladder | -0.047 |
road | -0.047 |
subway | -0.046 |
mountain | -0.046 |
Token | closer |
Token ID | 5699 |
Feature activation | -4.23e-03 |
Pos logit contributions | |
ished | +0.033 |
lee | +0.033 |
ady | +0.031 |
aming | +0.030 |
amed | +0.029 |
Neg logit contributions | |
hill | -0.036 |
cliff | -0.034 |
mountain | -0.033 |
road | -0.033 |
subway | -0.033 |
Token | L |
Token ID | 43 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
hill | +0.020 |
road | +0.018 |
ladder | +0.018 |
taxi | +0.018 |
cliff | +0.017 |
Neg logit contributions | |
ished | -0.031 |
lee | -0.030 |
ady | -0.027 |
aming | -0.027 |
amed | -0.026 |
Token | ily |
Token ID | 813 |
Feature activation | -9.58e-02 |
Pos logit contributions | |
lee | +0.244 |
ished | +0.212 |
aming | +0.211 |
ady | +0.207 |
amed | +0.204 |
Neg logit contributions | |
hill | -0.347 |
jail | -0.338 |
road | -0.335 |
ladder | -0.334 |
slide | -0.332 |
Token | and |
Token ID | 290 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
lee | +0.228 |
ished | +0.223 |
ady | +0.218 |
aming | +0.214 |
ins | +0.203 |
Neg logit contributions | |
hill | -0.161 |
road | -0.144 |
ladder | -0.143 |
slide | -0.143 |
bench | -0.141 |
Token | Ben |
Token ID | 3932 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
hill | +0.049 |
road | +0.046 |
ladder | +0.045 |
cliff | +0.045 |
ride | +0.045 |
Neg logit contributions | |
ished | -0.044 |
lee | -0.043 |
ady | -0.043 |
aming | -0.043 |
amed | -0.041 |
Token | felt |
Token ID | 2936 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
lee | +0.240 |
ished | +0.238 |
ady | +0.229 |
aming | +0.223 |
ins | +0.216 |
Neg logit contributions | |
hill | -0.198 |
taxi | -0.179 |
road | -0.178 |
ladder | -0.178 |
ride | -0.176 |
Token | angry |
Token ID | 7954 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
lee | +0.257 |
ished | +0.255 |
ady | +0.244 |
ins | +0.229 |
aming | +0.228 |
Neg logit contributions | |
taxi | -0.237 |
hill | -0.233 |
road | -0.228 |
ladder | -0.226 |
jail | -0.226 |
Token | and |
Token ID | 290 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
lee | +0.331 |
ished | +0.326 |
ady | +0.318 |
amed | +0.312 |
aming | +0.312 |
Neg logit contributions | |
hill | -0.199 |
road | -0.183 |
taxi | -0.172 |
ladder | -0.171 |
ride | -0.169 |
Token | scared |
Token ID | 12008 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
hill | +0.048 |
ladder | +0.045 |
road | +0.045 |
taxi | +0.045 |
cliff | +0.044 |
Neg logit contributions | |
ished | -0.036 |
lee | -0.035 |
ady | -0.034 |
aming | -0.033 |
amed | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
ished | +0.269 |
lee | +0.268 |
ady | +0.259 |
aming | +0.255 |
amed | +0.251 |
Neg logit contributions | |
hill | -0.193 |
road | -0.180 |
taxi | -0.178 |
ladder | -0.174 |
ride | -0.167 |
Token | They |
Token ID | 1119 |
Feature activation | +1.47e-03 |
Pos logit contributions | |
ished | +0.037 |
lee | +0.036 |
aming | +0.036 |
ady | +0.035 |
amed | +0.035 |
Neg logit contributions | |
hill | -0.022 |
road | -0.021 |
ladder | -0.020 |
taxi | -0.019 |
cliff | -0.019 |
Token | did |
Token ID | 750 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
hill | +0.003 |
road | +0.003 |
ladder | +0.003 |
taxi | +0.003 |
cliff | +0.003 |
Neg logit contributions | |
ished | -0.003 |
lee | -0.003 |
ady | -0.003 |
aming | -0.003 |
amed | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +9.15e-03 |
Pos logit contributions | |
lee | +0.007 |
ished | +0.007 |
ady | +0.007 |
aming | +0.007 |
amed | +0.007 |
Neg logit contributions | |
hill | -0.006 |
road | -0.005 |
ladder | -0.005 |
cliff | -0.005 |
taxi | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +8.01e-03 |
Pos logit contributions | |
hill | +0.013 |
road | +0.011 |
ladder | +0.011 |
taxi | +0.011 |
cliff | +0.011 |
Neg logit contributions | |
ished | -0.025 |
lee | -0.025 |
aming | -0.023 |
ady | -0.023 |
amed | -0.022 |
Token | L |
Token ID | 43 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
hill | +0.013 |
road | +0.012 |
ladder | +0.012 |
taxi | +0.012 |
ride | +0.011 |
Neg logit contributions | |
lee | -0.020 |
ished | -0.020 |
aming | -0.018 |
ady | -0.018 |
pled | -0.017 |
Token | ily |
Token ID | 813 |
Feature activation | -9.97e-02 |
Pos logit contributions | |
lee | +0.250 |
ished | +0.218 |
aming | +0.211 |
ady | +0.205 |
ealing | +0.203 |
Neg logit contributions | |
hill | -0.333 |
jail | -0.324 |
ladder | -0.323 |
road | -0.322 |
slide | -0.322 |
Token | feels |
Token ID | 5300 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
aming | +0.224 |
lee | +0.223 |
ady | +0.221 |
ished | +0.214 |
amed | +0.212 |
Neg logit contributions | |
hill | -0.191 |
ladder | -0.172 |
slide | -0.172 |
road | -0.167 |
walk | -0.163 |
Token | angry |
Token ID | 7954 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
ished | +0.255 |
lee | +0.250 |
ady | +0.240 |
ins | +0.232 |
aming | +0.225 |
Neg logit contributions | |
hill | -0.188 |
jail | -0.188 |
taxi | -0.186 |
road | -0.185 |
ladder | -0.181 |
Token | . |
Token ID | 13 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
lee | +0.322 |
ished | +0.321 |
ady | +0.315 |
aming | +0.311 |
amed | +0.309 |
Neg logit contributions | |
hill | -0.217 |
road | -0.202 |
ladder | -0.188 |
taxi | -0.188 |
ride | -0.186 |
Token | She |
Token ID | 1375 |
Feature activation | +4.95e-04 |
Pos logit contributions | |
ished | +0.028 |
lee | +0.027 |
aming | +0.027 |
ady | +0.026 |
amed | +0.026 |
Neg logit contributions | |
hill | -0.018 |
ladder | -0.017 |
road | -0.017 |
taxi | -0.016 |
cliff | -0.016 |
Token | tries |
Token ID | 8404 |
Feature activation | -5.24e-02 |
Pos logit contributions | |
hill | +0.001 |
road | +0.001 |
ladder | +0.001 |
slide | +0.001 |
walk | +0.001 |
Neg logit contributions | |
aming | -0.001 |
ished | -0.001 |
lee | -0.001 |
ady | -0.001 |
amed | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | +8.94e-02 |
Pos logit contributions | |
lee | +0.160 |
ished | +0.160 |
aming | +0.153 |
ady | +0.152 |
amed | +0.149 |
Neg logit contributions | |
hill | -0.054 |
road | -0.045 |
ladder | -0.044 |
cliff | -0.044 |
taxi | -0.043 |
Token | grab |
Token ID | 5552 |
Feature activation | -7.57e-02 |
Pos logit contributions | |
hill | +0.135 |
ladder | +0.121 |
cliff | +0.119 |
mountain | +0.119 |
road | +0.119 |
Neg logit contributions | |
ished | -0.222 |
lee | -0.221 |
aming | -0.213 |
ady | -0.211 |
amed | -0.204 |
Token | . |
Token ID | 13 |
Feature activation | +6.90e-04 |
Pos logit contributions | |
lee | +0.148 |
ished | +0.147 |
aming | +0.139 |
ady | +0.139 |
amed | +0.133 |
Neg logit contributions | |
hill | -0.130 |
road | -0.119 |
ladder | -0.118 |
cliff | -0.117 |
taxi | -0.116 |
Token | Acc |
Token ID | 6366 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
hill | +0.001 |
road | +0.001 |
ladder | +0.001 |
taxi | +0.001 |
cliff | +0.001 |
Neg logit contributions | |
ished | -0.002 |
lee | -0.002 |
aming | -0.001 |
ady | -0.001 |
amed | -0.001 |
Token | idents |
Token ID | 3231 |
Feature activation | -6.76e-02 |
Pos logit contributions | |
ished | +0.354 |
lee | +0.354 |
aming | +0.350 |
ady | +0.350 |
amed | +0.339 |
Neg logit contributions | |
hill | -0.213 |
road | -0.194 |
ladder | -0.193 |
taxi | -0.188 |
ride | -0.180 |
Token | happen |
Token ID | 1645 |
Feature activation | -7.05e-02 |
Pos logit contributions | |
ished | +0.166 |
lee | +0.166 |
aming | +0.158 |
ady | +0.156 |
amed | +0.153 |
Neg logit contributions | |
hill | -0.107 |
ladder | -0.096 |
road | -0.095 |
cliff | -0.095 |
taxi | -0.094 |
Token | . |
Token ID | 13 |
Feature activation | +1.04e-03 |
Pos logit contributions | |
lee | +0.194 |
ished | +0.191 |
ady | +0.190 |
amed | +0.188 |
aming | +0.187 |
Neg logit contributions | |
hill | -0.094 |
ladder | -0.079 |
road | -0.078 |
mountain | -0.076 |
visit | -0.075 |
Token | â |
Token ID | 22940 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
hill | +0.002 |
ladder | +0.002 |
road | +0.002 |
taxi | +0.002 |
visit | +0.002 |
Neg logit contributions | |
ished | -0.002 |
lee | -0.002 |
aming | -0.002 |
ady | -0.002 |
amed | -0.002 |
Token | € |
Token ID | 26391 |
Feature activation | -9.96e-02 |
Pos logit contributions | |
lee | +0.373 |
ished | +0.358 |
ady | +0.339 |
ins | +0.333 |
aming | +0.331 |
Neg logit contributions | |
hill | -0.350 |
road | -0.326 |
ladder | -0.316 |
ride | -0.316 |
taxi | -0.311 |
Token |
|
Token ID | 198 |
Feature activation | +2.66e-02 |
Pos logit contributions | |
lee | +0.230 |
ished | +0.228 |
aming | +0.216 |
ady | +0.216 |
amed | +0.209 |
Neg logit contributions | |
hill | -0.174 |
road | -0.159 |
ladder | -0.157 |
taxi | -0.157 |
cliff | -0.156 |
Token |
|
Token ID | 198 |
Feature activation | +2.53e-02 |
Pos logit contributions | |
hill | +0.036 |
road | +0.032 |
ladder | +0.032 |
taxi | +0.030 |
cliff | +0.030 |
Neg logit contributions | |
ished | -0.074 |
lee | -0.071 |
aming | -0.068 |
ady | -0.068 |
amed | -0.067 |
Token | She |
Token ID | 3347 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
hill | +0.041 |
road | +0.037 |
ladder | +0.037 |
taxi | +0.036 |
cliff | +0.036 |
Neg logit contributions | |
ished | -0.062 |
lee | -0.062 |
ady | -0.058 |
aming | -0.057 |
amed | -0.056 |
Token | took |
Token ID | 1718 |
Feature activation | -9.05e-02 |
Pos logit contributions | |
ride | +0.027 |
hill | +0.026 |
walk | +0.025 |
drive | +0.025 |
taxi | +0.025 |
Neg logit contributions | |
lee | -0.036 |
ady | -0.036 |
aming | -0.036 |
ealing | -0.034 |
ins | -0.034 |