Token | had |
Token ID | 550 |
Feature activation | +8.13e-02 |
Pos logit contributions | |
cardboard | +0.068 |
lf | +0.064 |
camps | +0.062 |
ts | +0.062 |
fossils | +0.055 |
Neg logit contributions | |
eal | -0.135 |
arre | -0.129 |
umpy | -0.124 |
irk | -0.121 |
back | -0.121 |
Token | ripped |
Token ID | 19551 |
Feature activation | +2.39e-02 |
Pos logit contributions | |
lf | +0.110 |
cardboard | +0.107 |
ts | +0.100 |
camps | +0.096 |
trek | +0.089 |
Neg logit contributions | |
eal | -0.162 |
arre | -0.156 |
umpy | -0.147 |
irk | -0.145 |
vious | -0.142 |
Token | right |
Token ID | 826 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
lf | +0.040 |
cardboard | +0.038 |
ts | +0.037 |
camps | +0.036 |
trek | +0.036 |
Neg logit contributions | |
eal | -0.038 |
arre | -0.037 |
back | -0.036 |
umpy | -0.035 |
irk | -0.034 |
Token | off |
Token ID | 572 |
Feature activation | -4.58e-02 |
Pos logit contributions | |
eal | +0.229 |
arre | +0.218 |
irk | +0.206 |
umpy | +0.206 |
ags | +0.200 |
Neg logit contributions | |
cardboard | -0.183 |
lf | -0.181 |
ts | -0.165 |
camps | -0.160 |
trek | -0.156 |
Token | its |
Token ID | 663 |
Feature activation | -6.32e-02 |
Pos logit contributions | |
eal | +0.076 |
arre | +0.074 |
back | +0.072 |
umpy | +0.072 |
irk | +0.069 |
Neg logit contributions | |
cardboard | -0.069 |
lf | -0.068 |
ts | -0.065 |
camps | -0.065 |
trek | -0.062 |
Token | hinges |
Token ID | 41442 |
Feature activation | +1.91e-01 |
Pos logit contributions | |
eal | +0.105 |
arre | +0.102 |
irk | +0.094 |
umpy | +0.093 |
vious | +0.088 |
Neg logit contributions | |
cardboard | -0.111 |
lf | -0.107 |
ts | -0.101 |
camps | -0.097 |
blasted | -0.095 |
Token | ! |
Token ID | 0 |
Feature activation | -2.84e-02 |
Pos logit contributions | |
lf | +0.304 |
ts | +0.302 |
cardboard | +0.295 |
camps | +0.275 |
trek | +0.270 |
Neg logit contributions | |
arre | -0.314 |
eal | -0.311 |
umpy | -0.299 |
back | -0.291 |
irk | -0.279 |
Token | Lily |
Token ID | 20037 |
Feature activation | +6.18e-03 |
Pos logit contributions | |
eal | +0.055 |
arre | +0.052 |
irk | +0.049 |
umpy | +0.049 |
awk | +0.049 |
Neg logit contributions | |
cardboard | -0.042 |
lf | -0.040 |
ts | -0.035 |
camps | -0.034 |
pretend | -0.033 |
Token | felt |
Token ID | 2936 |
Feature activation | +2.21e-02 |
Pos logit contributions | |
cardboard | +0.009 |
camps | +0.008 |
lf | +0.008 |
ts | +0.008 |
fossils | +0.008 |
Neg logit contributions | |
eal | -0.012 |
arre | -0.012 |
umpy | -0.011 |
back | -0.011 |
disappointed | -0.010 |
Token | bad |
Token ID | 2089 |
Feature activation | -8.83e-02 |
Pos logit contributions | |
cardboard | +0.036 |
ts | +0.033 |
lf | +0.033 |
pretend | +0.032 |
shovel | +0.030 |
Neg logit contributions | |
eal | -0.039 |
arre | -0.039 |
irk | -0.035 |
umpy | -0.035 |
awk | -0.034 |
Token | for |
Token ID | 329 |
Feature activation | -6.82e-02 |
Pos logit contributions | |
eal | +0.169 |
arre | +0.162 |
umpy | +0.151 |
irk | +0.149 |
ags | +0.147 |
Neg logit contributions | |
cardboard | -0.132 |
lf | -0.125 |
ts | -0.116 |
camps | -0.112 |
pretend | -0.110 |
Token | door |
Token ID | 3420 |
Feature activation | +2.43e-02 |
Pos logit contributions | |
cardboard | +0.097 |
camps | +0.093 |
ts | +0.089 |
trek | +0.086 |
lf | +0.085 |
Neg logit contributions | |
eal | -0.142 |
arre | -0.137 |
back | -0.135 |
umpy | -0.135 |
irk | -0.129 |
Token | had |
Token ID | 550 |
Feature activation | +8.03e-02 |
Pos logit contributions | |
camps | +0.036 |
cardboard | +0.034 |
seas | +0.034 |
ts | +0.033 |
fossils | +0.033 |
Neg logit contributions | |
back | -0.041 |
eal | -0.039 |
disappointed | -0.039 |
umpy | -0.037 |
mad | -0.037 |
Token | fallen |
Token ID | 9292 |
Feature activation | +4.21e-02 |
Pos logit contributions | |
lf | +0.116 |
cardboard | +0.104 |
camps | +0.104 |
ts | +0.104 |
trek | +0.095 |
Neg logit contributions | |
eal | -0.142 |
back | -0.140 |
arre | -0.134 |
umpy | -0.130 |
ags | -0.128 |
Token | off |
Token ID | 572 |
Feature activation | -4.54e-02 |
Pos logit contributions | |
lf | +0.084 |
ts | +0.078 |
cardboard | +0.076 |
trek | +0.072 |
fossils | +0.072 |
Neg logit contributions | |
back | -0.056 |
eal | -0.053 |
arre | -0.053 |
umpy | -0.051 |
ags | -0.049 |
Token | its |
Token ID | 663 |
Feature activation | -6.81e-02 |
Pos logit contributions | |
back | +0.068 |
eal | +0.066 |
arre | +0.065 |
umpy | +0.065 |
irk | +0.063 |
Neg logit contributions | |
lf | -0.072 |
ts | -0.070 |
cardboard | -0.070 |
trek | -0.067 |
blasted | -0.066 |
Token | hinges |
Token ID | 41442 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
eal | +0.126 |
arre | +0.121 |
umpy | +0.116 |
irk | +0.114 |
ags | +0.111 |
Neg logit contributions | |
lf | -0.099 |
cardboard | -0.097 |
ts | -0.093 |
camps | -0.086 |
blasted | -0.084 |
Token | and |
Token ID | 290 |
Feature activation | -2.68e-02 |
Pos logit contributions | |
lf | +0.301 |
ts | +0.292 |
cardboard | +0.288 |
camps | +0.268 |
trek | +0.268 |
Neg logit contributions | |
arre | -0.305 |
eal | -0.302 |
umpy | -0.294 |
back | -0.284 |
irk | -0.277 |
Token | Jill |
Token ID | 22772 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
eal | +0.046 |
back | +0.045 |
arre | +0.044 |
disappointed | +0.042 |
mad | +0.042 |
Neg logit contributions | |
lf | -0.038 |
cardboard | -0.037 |
camps | -0.036 |
ts | -0.034 |
trek | -0.034 |
Token | was |
Token ID | 373 |
Feature activation | +2.74e-02 |
Pos logit contributions | |
cardboard | +0.019 |
lf | +0.018 |
camps | +0.017 |
ts | +0.017 |
trek | +0.016 |
Neg logit contributions | |
eal | -0.021 |
arre | -0.020 |
umpy | -0.019 |
irk | -0.019 |
back | -0.018 |
Token | lying |
Token ID | 9105 |
Feature activation | -3.05e-03 |
Pos logit contributions | |
cardboard | +0.044 |
lf | +0.043 |
ts | +0.040 |
camps | +0.039 |
trek | +0.038 |
Neg logit contributions | |
eal | -0.048 |
arre | -0.047 |
umpy | -0.044 |
irk | -0.043 |
ags | -0.041 |
Token | on |
Token ID | 319 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
eal | +0.006 |
arre | +0.006 |
umpy | +0.005 |
irk | +0.005 |
vious | +0.005 |
Neg logit contributions | |
cardboard | -0.004 |
lf | -0.004 |
ts | -0.004 |
camps | -0.004 |
trek | -0.004 |
Token | calling |
Token ID | 4585 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
cardboard | +0.053 |
lf | +0.052 |
ts | +0.049 |
camps | +0.046 |
blasted | +0.045 |
Neg logit contributions | |
eal | -0.075 |
arre | -0.073 |
umpy | -0.069 |
irk | -0.068 |
ags | -0.065 |
Token | out |
Token ID | 503 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
eal | +0.027 |
arre | +0.027 |
umpy | +0.025 |
irk | +0.025 |
vious | +0.024 |
Neg logit contributions | |
cardboard | -0.016 |
lf | -0.016 |
called | -0.013 |
pretend | -0.013 |
ts | -0.013 |
Token | " |
Token ID | 366 |
Feature activation | +1.41e-03 |
Pos logit contributions | |
eal | +0.033 |
arre | +0.032 |
irk | +0.030 |
umpy | +0.030 |
agree | +0.029 |
Neg logit contributions | |
cardboard | -0.020 |
lf | -0.019 |
ts | -0.017 |
camps | -0.017 |
pretend | -0.016 |
Token | Hello |
Token ID | 15496 |
Feature activation | +4.18e-02 |
Pos logit contributions | |
lf | +0.002 |
cardboard | +0.002 |
ts | +0.002 |
blasted | +0.002 |
trek | +0.002 |
Neg logit contributions | |
eal | -0.002 |
arre | -0.002 |
irk | -0.002 |
fault | -0.002 |
ags | -0.002 |
Token | ?". |
Token ID | 43634 |
Feature activation | -2.98e-02 |
Pos logit contributions | |
cardboard | +0.056 |
lf | +0.055 |
ts | +0.050 |
camps | +0.048 |
blasted | +0.046 |
Neg logit contributions | |
eal | -0.085 |
arre | -0.082 |
umpy | -0.077 |
irk | -0.077 |
ags | -0.074 |
Token | A |
Token ID | 317 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
eal | +0.057 |
arre | +0.056 |
umpy | +0.051 |
irk | +0.050 |
ags | +0.050 |
Neg logit contributions | |
cardboard | -0.045 |
lf | -0.044 |
called | -0.038 |
ts | -0.038 |
pretend | -0.038 |
Token | little |
Token ID | 1310 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
cardboard | +0.281 |
lf | +0.276 |
ts | +0.258 |
camps | +0.246 |
blasted | +0.238 |
Neg logit contributions | |
eal | -0.341 |
arre | -0.330 |
umpy | -0.308 |
irk | -0.308 |
ags | -0.290 |
Token | girl |
Token ID | 2576 |
Feature activation | +3.85e-02 |
Pos logit contributions | |
cardboard | +0.207 |
lf | +0.198 |
ts | +0.183 |
camps | +0.179 |
shovel | +0.178 |
Neg logit contributions | |
eal | -0.187 |
arre | -0.185 |
umpy | -0.167 |
ags | -0.165 |
irk | -0.163 |
Token | came |
Token ID | 1625 |
Feature activation | +2.58e-02 |
Pos logit contributions | |
cardboard | +0.048 |
lf | +0.045 |
ts | +0.044 |
camps | +0.042 |
seas | +0.038 |
Neg logit contributions | |
eal | -0.082 |
arre | -0.077 |
irk | -0.074 |
umpy | -0.073 |
disappointed | -0.071 |
Token | running |
Token ID | 2491 |
Feature activation | -8.16e-03 |
Pos logit contributions | |
cardboard | +0.045 |
lf | +0.043 |
trek | +0.041 |
camps | +0.041 |
called | +0.040 |
Neg logit contributions | |
eal | -0.042 |
arre | -0.040 |
irk | -0.037 |
umpy | -0.036 |
vious | -0.035 |
Token | into |
Token ID | 656 |
Feature activation | +4.63e-02 |
Pos logit contributions | |
arre | +0.016 |
eal | +0.016 |
ags | +0.015 |
umpy | +0.015 |
ealing | +0.015 |
Neg logit contributions | |
cardboard | -0.011 |
lf | -0.010 |
pretend | -0.010 |
called | -0.010 |
pretended | -0.010 |
Token | was |
Token ID | 373 |
Feature activation | +3.68e-02 |
Pos logit contributions | |
cardboard | +0.101 |
ts | +0.095 |
lf | +0.094 |
swords | +0.088 |
camps | +0.088 |
Neg logit contributions | |
eal | -0.109 |
arre | -0.108 |
umpy | -0.105 |
back | -0.103 |
irk | -0.101 |
Token | a |
Token ID | 257 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
lf | +0.057 |
ts | +0.053 |
blasted | +0.049 |
cardboard | +0.047 |
fossils | +0.046 |
Neg logit contributions | |
eal | -0.064 |
arre | -0.064 |
ags | -0.063 |
umpy | -0.060 |
back | -0.060 |
Token | little |
Token ID | 1310 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
cardboard | +0.232 |
lf | +0.230 |
ts | +0.229 |
trek | +0.211 |
blasted | +0.210 |
Neg logit contributions | |
eal | -0.296 |
umpy | -0.280 |
arre | -0.280 |
irk | -0.270 |
back | -0.260 |
Token | girl |
Token ID | 2576 |
Feature activation | +3.40e-02 |
Pos logit contributions | |
cardboard | +0.208 |
lf | +0.208 |
ts | +0.197 |
camps | +0.188 |
trek | +0.187 |
Neg logit contributions | |
eal | -0.155 |
arre | -0.147 |
umpy | -0.137 |
irk | -0.137 |
awk | -0.125 |
Token | named |
Token ID | 3706 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
ts | +0.040 |
loot | +0.040 |
camps | +0.039 |
fossils | +0.039 |
cardboard | +0.038 |
Neg logit contributions | |
back | -0.064 |
umpy | -0.064 |
eal | -0.063 |
irk | -0.062 |
disappointed | -0.060 |
Token | A |
Token ID | 317 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
lf | +0.174 |
cardboard | +0.173 |
ts | +0.166 |
camps | +0.157 |
blasted | +0.157 |
Neg logit contributions | |
eal | -0.171 |
arre | -0.163 |
umpy | -0.153 |
irk | -0.149 |
back | -0.145 |
Token | ire |
Token ID | 557 |
Feature activation | +2.67e-02 |
Pos logit contributions | |
cardboard | +0.232 |
lf | +0.227 |
ts | +0.210 |
camps | +0.196 |
blasted | +0.191 |
Neg logit contributions | |
eal | -0.380 |
arre | -0.368 |
umpy | -0.348 |
irk | -0.347 |
vious | -0.328 |
Token | l |
Token ID | 75 |
Feature activation | -3.69e-02 |
Pos logit contributions | |
cardboard | +0.041 |
camps | +0.037 |
lf | +0.036 |
trek | +0.036 |
ts | +0.036 |
Neg logit contributions | |
umpy | -0.046 |
eal | -0.046 |
arre | -0.045 |
irk | -0.043 |
itter | -0.041 |
Token | le |
Token ID | 293 |
Feature activation | -4.58e-02 |
Pos logit contributions | |
eal | +0.060 |
arre | +0.059 |
umpy | +0.059 |
irk | +0.056 |
ags | +0.055 |
Neg logit contributions | |
cardboard | -0.061 |
lf | -0.057 |
camps | -0.054 |
ts | -0.054 |
shovel | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | -9.39e-03 |
Pos logit contributions | |
umpy | +0.072 |
eal | +0.069 |
arre | +0.065 |
itter | +0.065 |
back | +0.065 |
Neg logit contributions | |
cardboard | -0.074 |
dice | -0.069 |
fossils | -0.066 |
swords | -0.066 |
ts | -0.065 |
Token | She |
Token ID | 1375 |
Feature activation | +6.63e-02 |
Pos logit contributions | |
eal | +0.010 |
arre | +0.010 |
irk | +0.009 |
umpy | +0.009 |
ags | +0.008 |
Neg logit contributions | |
cardboard | -0.021 |
lf | -0.021 |
ts | -0.020 |
camps | -0.020 |
blasted | -0.019 |
Token | was |
Token ID | 373 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
cardboard | +0.095 |
lf | +0.094 |
ts | +0.092 |
blasted | +0.081 |
camps | +0.080 |
Neg logit contributions | |
eal | -0.142 |
arre | -0.139 |
back | -0.131 |
umpy | -0.131 |
irk | -0.130 |
Token | a |
Token ID | 257 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
lf | +0.061 |
ts | +0.056 |
blasted | +0.051 |
cardboard | +0.050 |
fossils | +0.047 |
Neg logit contributions | |
eal | -0.075 |
arre | -0.073 |
ags | -0.070 |
awk | -0.068 |
vious | -0.068 |
Token | small |
Token ID | 1402 |
Feature activation | +8.08e-02 |
Pos logit contributions | |
ts | +0.255 |
cardboard | +0.251 |
lf | +0.242 |
blasted | +0.238 |
trek | +0.230 |
Neg logit contributions | |
eal | -0.281 |
umpy | -0.280 |
arre | -0.268 |
irk | -0.263 |
back | -0.249 |
Token | fair |
Token ID | 3148 |
Feature activation | -4.43e-02 |
Pos logit contributions | |
lf | +0.146 |
cardboard | +0.146 |
ts | +0.139 |
blasted | +0.131 |
camps | +0.131 |
Neg logit contributions | |
eal | -0.123 |
arre | -0.117 |
umpy | -0.110 |
irk | -0.110 |
awk | -0.101 |
Token | . |
Token ID | 13 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
eal | +0.069 |
arre | +0.064 |
umpy | +0.063 |
irk | +0.062 |
awk | +0.057 |
Neg logit contributions | |
cardboard | -0.078 |
lf | -0.076 |
ts | -0.076 |
blasted | -0.070 |
trek | -0.070 |
Token | A |
Token ID | 317 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
eal | +0.017 |
umpy | +0.017 |
irk | +0.017 |
arre | +0.016 |
agree | +0.015 |
Neg logit contributions | |
ts | -0.030 |
lf | -0.028 |
cardboard | -0.027 |
blasted | -0.026 |
camps | -0.025 |
Token | little |
Token ID | 1310 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
cardboard | +0.239 |
lf | +0.234 |
ts | +0.226 |
camps | +0.206 |
trek | +0.204 |
Neg logit contributions | |
eal | -0.364 |
arre | -0.348 |
umpy | -0.336 |
irk | -0.333 |
back | -0.311 |
Token | boy |
Token ID | 2933 |
Feature activation | +5.18e-02 |
Pos logit contributions | |
cardboard | +0.202 |
lf | +0.191 |
ts | +0.176 |
camps | +0.174 |
shovel | +0.174 |
Neg logit contributions | |
eal | -0.185 |
arre | -0.184 |
umpy | -0.163 |
irk | -0.162 |
ags | -0.159 |
Token | went |
Token ID | 1816 |
Feature activation | +4.44e-02 |
Pos logit contributions | |
camps | +0.064 |
ts | +0.063 |
cardboard | +0.061 |
miles | +0.060 |
lf | +0.058 |
Neg logit contributions | |
umpy | -0.098 |
eal | -0.097 |
irk | -0.092 |
back | -0.090 |
disappointed | -0.089 |
Token | to |
Token ID | 284 |
Feature activation | +3.33e-04 |
Pos logit contributions | |
cardboard | +0.075 |
lf | +0.070 |
camps | +0.066 |
ts | +0.065 |
trek | +0.064 |
Neg logit contributions | |
eal | -0.078 |
arre | -0.076 |
umpy | -0.070 |
irk | -0.070 |
ealing | -0.067 |
Token | the |
Token ID | 262 |
Feature activation | -3.16e-02 |
Pos logit contributions | |
lf | +0.001 |
cardboard | +0.000 |
ts | +0.000 |
camps | +0.000 |
blasted | +0.000 |
Neg logit contributions | |
eal | -0.001 |
arre | -0.001 |
back | -0.001 |
irk | -0.000 |
ags | -0.000 |
Token | a |
Token ID | 257 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
lf | +0.060 |
ts | +0.058 |
blasted | +0.053 |
cardboard | +0.050 |
hunt | +0.050 |
Neg logit contributions | |
arre | -0.062 |
eal | -0.061 |
ags | -0.059 |
back | -0.058 |
umpy | -0.057 |
Token | dead |
Token ID | 2636 |
Feature activation | -5.70e-03 |
Pos logit contributions | |
ts | +0.235 |
cardboard | +0.233 |
lf | +0.232 |
trek | +0.213 |
blasted | +0.212 |
Neg logit contributions | |
eal | -0.300 |
arre | -0.284 |
umpy | -0.281 |
irk | -0.274 |
back | -0.265 |
Token | wheat |
Token ID | 17135 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
eal | +0.011 |
arre | +0.010 |
umpy | +0.010 |
irk | +0.010 |
ags | +0.009 |
Neg logit contributions | |
cardboard | -0.009 |
lf | -0.008 |
ts | -0.008 |
camps | -0.008 |
blasted | -0.007 |
Token | field |
Token ID | 2214 |
Feature activation | +5.01e-02 |
Pos logit contributions | |
cardboard | +0.017 |
lf | +0.016 |
ts | +0.015 |
camps | +0.015 |
blasted | +0.014 |
Neg logit contributions | |
eal | -0.025 |
arre | -0.024 |
umpy | -0.023 |
irk | -0.022 |
ags | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
cardboard | +0.067 |
lf | +0.066 |
ts | +0.061 |
camps | +0.058 |
blasted | +0.056 |
Neg logit contributions | |
eal | -0.101 |
arre | -0.098 |
umpy | -0.092 |
irk | -0.092 |
ags | -0.088 |
Token | A |
Token ID | 317 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
eal | +0.017 |
arre | +0.016 |
umpy | +0.015 |
irk | +0.015 |
ags | +0.014 |
Neg logit contributions | |
cardboard | -0.026 |
lf | -0.026 |
ts | -0.024 |
camps | -0.024 |
blasted | -0.023 |
Token | young |
Token ID | 1862 |
Feature activation | +6.03e-02 |
Pos logit contributions | |
cardboard | +0.213 |
lf | +0.208 |
ts | +0.189 |
camps | +0.177 |
blasted | +0.170 |
Neg logit contributions | |
eal | -0.396 |
arre | -0.385 |
umpy | -0.364 |
irk | -0.364 |
ags | -0.347 |
Token | boy |
Token ID | 2933 |
Feature activation | +4.84e-02 |
Pos logit contributions | |
cardboard | +0.105 |
lf | +0.100 |
ts | +0.091 |
shovel | +0.089 |
camps | +0.089 |
Neg logit contributions | |
eal | -0.103 |
arre | -0.102 |
ags | -0.093 |
irk | -0.091 |
umpy | -0.089 |
Token | went |
Token ID | 1816 |
Feature activation | +4.37e-02 |
Pos logit contributions | |
camps | +0.064 |
ts | +0.062 |
lf | +0.062 |
cardboard | +0.061 |
dice | +0.058 |
Neg logit contributions | |
eal | -0.084 |
umpy | -0.084 |
crow | -0.080 |
back | -0.080 |
vious | -0.080 |
Token | out |
Token ID | 503 |
Feature activation | -4.72e-03 |
Pos logit contributions | |
cardboard | +0.067 |
called | +0.058 |
lf | +0.058 |
hunt | +0.057 |
camps | +0.056 |
Neg logit contributions | |
eal | -0.087 |
arre | -0.083 |
irk | -0.080 |
umpy | -0.078 |
ealing | -0.077 |
Token | to |
Token ID | 284 |
Feature activation | +2.64e-03 |
Pos logit contributions | |
eal | +0.008 |
arre | +0.007 |
umpy | +0.007 |
irk | +0.007 |
ags | +0.006 |
Neg logit contributions | |
cardboard | -0.008 |
lf | -0.008 |
ts | -0.008 |
camps | -0.007 |
trek | -0.007 |
Token | said |
Token ID | 531 |
Feature activation | -3.29e-02 |
Pos logit contributions | |
cardboard | +0.005 |
lf | +0.005 |
ts | +0.004 |
pretend | +0.004 |
shovel | +0.004 |
Neg logit contributions | |
eal | -0.009 |
arre | -0.009 |
irk | -0.008 |
umpy | -0.008 |
ags | -0.008 |
Token | " |
Token ID | 366 |
Feature activation | -3.03e-03 |
Pos logit contributions | |
eal | +0.080 |
arre | +0.078 |
umpy | +0.076 |
irk | +0.075 |
ags | +0.072 |
Neg logit contributions | |
lf | -0.034 |
cardboard | -0.034 |
ts | -0.029 |
pretend | -0.027 |
called | -0.026 |
Token | Free |
Token ID | 11146 |
Feature activation | -9.51e-03 |
Pos logit contributions | |
eal | +0.006 |
ags | +0.005 |
arre | +0.005 |
ealing | +0.005 |
fault | +0.005 |
Neg logit contributions | |
lf | -0.004 |
cardboard | -0.004 |
pretended | -0.004 |
pretend | -0.003 |
blasted | -0.003 |
Token | Candy |
Token ID | 24680 |
Feature activation | +5.97e-03 |
Pos logit contributions | |
eal | +0.015 |
irk | +0.014 |
arre | +0.014 |
umpy | +0.013 |
vious | +0.013 |
Neg logit contributions | |
lf | -0.017 |
cardboard | -0.017 |
ts | -0.016 |
pretend | -0.015 |
camps | -0.015 |
Token | ". |
Token ID | 1911 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
cardboard | +0.008 |
lf | +0.008 |
ts | +0.007 |
camps | +0.007 |
pretend | +0.007 |
Neg logit contributions | |
eal | -0.012 |
arre | -0.012 |
irk | -0.011 |
umpy | -0.011 |
ags | -0.011 |
Token | A |
Token ID | 317 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
eal | +0.062 |
arre | +0.060 |
umpy | +0.056 |
vious | +0.055 |
ags | +0.055 |
Neg logit contributions | |
cardboard | -0.040 |
lf | -0.040 |
called | -0.038 |
pretend | -0.035 |
named | -0.035 |
Token | man |
Token ID | 582 |
Feature activation | -8.77e-04 |
Pos logit contributions | |
cardboard | +0.265 |
lf | +0.260 |
ts | +0.240 |
camps | +0.229 |
blasted | +0.222 |
Neg logit contributions | |
eal | -0.342 |
arre | -0.331 |
umpy | -0.309 |
irk | -0.309 |
ags | -0.295 |
Token | with |
Token ID | 351 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
eal | +0.002 |
arre | +0.002 |
irk | +0.002 |
disappointed | +0.001 |
umpy | +0.001 |
Neg logit contributions | |
cardboard | -0.001 |
camps | -0.001 |
lf | -0.001 |
ts | -0.001 |
seas | -0.001 |
Token | a |
Token ID | 257 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
eal | +0.037 |
arre | +0.036 |
irk | +0.035 |
umpy | +0.034 |
vious | +0.034 |
Neg logit contributions | |
lf | -0.020 |
cardboard | -0.018 |
ts | -0.018 |
camps | -0.017 |
trek | -0.016 |
Token | beard |
Token ID | 21213 |
Feature activation | +6.61e-02 |
Pos logit contributions | |
cardboard | +0.219 |
lf | +0.200 |
shovel | +0.179 |
pretend | +0.173 |
camps | +0.170 |
Neg logit contributions | |
eal | -0.341 |
arre | -0.330 |
ags | -0.317 |
umpy | -0.315 |
vious | -0.309 |
Token | and |
Token ID | 290 |
Feature activation | -1.14e-02 |
Pos logit contributions | |
cardboard | +0.091 |
lf | +0.086 |
ts | +0.082 |
camps | +0.080 |
trek | +0.075 |
Neg logit contributions | |
eal | -0.130 |
arre | -0.125 |
umpy | -0.118 |
irk | -0.118 |
vious | -0.113 |
Token | day |
Token ID | 1110 |
Feature activation | -3.55e-03 |
Pos logit contributions | |
eal | +0.069 |
arre | +0.068 |
irk | +0.063 |
umpy | +0.063 |
ags | +0.060 |
Neg logit contributions | |
cardboard | -0.046 |
lf | -0.045 |
ts | -0.041 |
camps | -0.039 |
trek | -0.038 |
Token | something |
Token ID | 1223 |
Feature activation | +3.34e-02 |
Pos logit contributions | |
eal | +0.008 |
arre | +0.008 |
irk | +0.007 |
umpy | +0.007 |
ags | +0.007 |
Neg logit contributions | |
cardboard | -0.004 |
lf | -0.004 |
ts | -0.003 |
camps | -0.003 |
trek | -0.003 |
Token | magical |
Token ID | 10883 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
cardboard | +0.046 |
lf | +0.044 |
ts | +0.041 |
camps | +0.038 |
pretend | +0.038 |
Neg logit contributions | |
eal | -0.069 |
arre | -0.067 |
irk | -0.063 |
umpy | -0.062 |
ags | -0.061 |
Token | happened |
Token ID | 3022 |
Feature activation | +4.06e-03 |
Pos logit contributions | |
cardboard | +0.027 |
lf | +0.026 |
ts | +0.024 |
camps | +0.023 |
blasted | +0.022 |
Neg logit contributions | |
eal | -0.040 |
arre | -0.039 |
umpy | -0.036 |
irk | -0.036 |
vious | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -1.49e-02 |
Pos logit contributions | |
cardboard | +0.005 |
lf | +0.005 |
ts | +0.004 |
camps | +0.004 |
pretend | +0.004 |
Neg logit contributions | |
eal | -0.009 |
arre | -0.009 |
irk | -0.008 |
umpy | -0.008 |
ags | -0.008 |
Token | A |
Token ID | 317 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
eal | +0.027 |
arre | +0.025 |
irk | +0.024 |
umpy | +0.023 |
ags | +0.023 |
Neg logit contributions | |
cardboard | -0.024 |
lf | -0.023 |
camps | -0.021 |
trek | -0.021 |
ts | -0.020 |
Token | gentle |
Token ID | 10296 |
Feature activation | -5.80e-02 |
Pos logit contributions | |
cardboard | +0.264 |
lf | +0.260 |
ts | +0.252 |
camps | +0.235 |
blasted | +0.227 |
Neg logit contributions | |
eal | -0.328 |
arre | -0.319 |
umpy | -0.301 |
irk | -0.300 |
vious | -0.279 |
Token | rain |
Token ID | 6290 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
arre | +0.106 |
eal | +0.104 |
ags | +0.098 |
umpy | +0.095 |
Fighting | +0.092 |
Neg logit contributions | |
cardboard | -0.094 |
lf | -0.093 |
trek | -0.079 |
pretend | -0.078 |
ts | -0.077 |
Token | started |
Token ID | 2067 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
eal | +0.027 |
arre | +0.026 |
irk | +0.025 |
ags | +0.025 |
umpy | +0.025 |
Neg logit contributions | |
lf | -0.017 |
cardboard | -0.017 |
ts | -0.015 |
blasted | -0.015 |
pretended | -0.014 |
Token | pouring |
Token ID | 23147 |
Feature activation | -2.83e-02 |
Pos logit contributions | |
cardboard | +0.047 |
lf | +0.044 |
ts | +0.041 |
blasted | +0.040 |
trek | +0.040 |
Neg logit contributions | |
eal | -0.056 |
arre | -0.052 |
irk | -0.050 |
umpy | -0.049 |
ags | -0.048 |
Token | from |
Token ID | 422 |
Feature activation | -4.46e-03 |
Pos logit contributions | |
eal | +0.055 |
arre | +0.052 |
irk | +0.051 |
umpy | +0.049 |
agree | +0.046 |
Neg logit contributions | |
cardboard | -0.043 |
lf | -0.038 |
ts | -0.038 |
camps | -0.036 |
shovel | -0.035 |
Token | there |
Token ID | 612 |
Feature activation | +6.40e-02 |
Pos logit contributions | |
eal | +0.033 |
arre | +0.032 |
irk | +0.030 |
umpy | +0.030 |
vious | +0.029 |
Neg logit contributions | |
cardboard | -0.019 |
lf | -0.018 |
ts | -0.017 |
camps | -0.016 |
blasted | -0.015 |
Token | in |
Token ID | 287 |
Feature activation | +3.74e-02 |
Pos logit contributions | |
cardboard | +0.087 |
lf | +0.083 |
ts | +0.077 |
camps | +0.073 |
blasted | +0.072 |
Neg logit contributions | |
eal | -0.125 |
arre | -0.123 |
umpy | -0.114 |
irk | -0.113 |
back | -0.109 |
Token | the |
Token ID | 262 |
Feature activation | -2.96e-02 |
Pos logit contributions | |
lf | +0.055 |
ts | +0.052 |
blasted | +0.052 |
cardboard | +0.051 |
camps | +0.050 |
Neg logit contributions | |
eal | -0.058 |
back | -0.057 |
arre | -0.056 |
irk | -0.056 |
ags | -0.055 |
Token | garden |
Token ID | 11376 |
Feature activation | +6.46e-02 |
Pos logit contributions | |
eal | +0.070 |
arre | +0.066 |
irk | +0.064 |
umpy | +0.061 |
ealing | +0.060 |
Neg logit contributions | |
cardboard | -0.036 |
lf | -0.031 |
camps | -0.030 |
pretend | -0.027 |
seas | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
cardboard | +0.101 |
lf | +0.099 |
ts | +0.094 |
camps | +0.090 |
blasted | +0.088 |
Neg logit contributions | |
eal | -0.108 |
arre | -0.106 |
umpy | -0.102 |
irk | -0.096 |
back | -0.096 |
Token | A |
Token ID | 317 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
eal | +0.017 |
arre | +0.016 |
umpy | +0.015 |
irk | +0.014 |
agree | +0.014 |
Neg logit contributions | |
lf | -0.030 |
cardboard | -0.030 |
ts | -0.029 |
camps | -0.028 |
blasted | -0.027 |
Token | three |
Token ID | 1115 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
cardboard | +0.226 |
lf | +0.221 |
ts | +0.219 |
camps | +0.197 |
trek | +0.197 |
Neg logit contributions | |
eal | -0.365 |
arre | -0.351 |
umpy | -0.340 |
irk | -0.334 |
back | -0.318 |
Token | year |
Token ID | 614 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
cardboard | +0.182 |
camps | +0.169 |
blasted | +0.168 |
ts | +0.165 |
trek | +0.164 |
Neg logit contributions | |
eal | -0.198 |
umpy | -0.184 |
arre | -0.180 |
irk | -0.173 |
vious | -0.164 |
Token | old |
Token ID | 1468 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
cardboard | +0.024 |
lf | +0.024 |
ts | +0.021 |
camps | +0.020 |
called | +0.020 |
Neg logit contributions | |
eal | -0.023 |
arre | -0.022 |
irk | -0.020 |
ags | -0.019 |
Fighting | -0.019 |
Token | girl |
Token ID | 2576 |
Feature activation | +3.21e-02 |
Pos logit contributions | |
cardboard | +0.029 |
ts | +0.029 |
lf | +0.029 |
camps | +0.028 |
blasted | +0.027 |
Neg logit contributions | |
eal | -0.023 |
arre | -0.021 |
irk | -0.021 |
umpy | -0.020 |
back | -0.019 |
Token | was |
Token ID | 373 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
ts | +0.038 |
camps | +0.037 |
lf | +0.035 |
cardboard | +0.035 |
seas | +0.032 |
Neg logit contributions | |
eal | -0.064 |
back | -0.060 |
irk | -0.059 |
umpy | -0.059 |
disappointed | -0.058 |
Token | there |
Token ID | 612 |
Feature activation | +6.83e-02 |
Pos logit contributions | |
cardboard | +0.086 |
lf | +0.086 |
ts | +0.079 |
trek | +0.078 |
camps | +0.076 |
Neg logit contributions | |
arre | -0.128 |
eal | -0.127 |
umpy | -0.124 |
back | -0.122 |
ags | -0.117 |
Token | was |
Token ID | 373 |
Feature activation | +3.71e-02 |
Pos logit contributions | |
cardboard | +0.105 |
ts | +0.099 |
lf | +0.098 |
swords | +0.092 |
fossils | +0.092 |
Neg logit contributions | |
arre | -0.116 |
eal | -0.114 |
umpy | -0.109 |
back | -0.107 |
irk | -0.104 |
Token | a |
Token ID | 257 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
lf | +0.058 |
ts | +0.057 |
blasted | +0.052 |
hunt | +0.050 |
cardboard | +0.050 |
Neg logit contributions | |
arre | -0.063 |
eal | -0.061 |
ags | -0.059 |
back | -0.059 |
umpy | -0.057 |
Token | theater |
Token ID | 13766 |
Feature activation | +5.71e-02 |
Pos logit contributions | |
ts | +0.246 |
cardboard | +0.241 |
lf | +0.240 |
trek | +0.231 |
blasted | +0.225 |
Neg logit contributions | |
eal | -0.279 |
umpy | -0.266 |
arre | -0.264 |
irk | -0.255 |
back | -0.250 |
Token | . |
Token ID | 13 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
cardboard | +0.083 |
ts | +0.077 |
camps | +0.073 |
lf | +0.073 |
trek | +0.072 |
Neg logit contributions | |
eal | -0.106 |
arre | -0.100 |
umpy | -0.100 |
irk | -0.095 |
back | -0.093 |
Token | A |
Token ID | 317 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
eal | +0.018 |
arre | +0.017 |
umpy | +0.016 |
irk | +0.016 |
agree | +0.015 |
Neg logit contributions | |
ts | -0.027 |
cardboard | -0.027 |
lf | -0.026 |
camps | -0.025 |
trek | -0.024 |
Token | little |
Token ID | 1310 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
cardboard | +0.291 |
ts | +0.290 |
lf | +0.272 |
trek | +0.267 |
camps | +0.260 |
Neg logit contributions | |
eal | -0.292 |
umpy | -0.274 |
arre | -0.267 |
irk | -0.256 |
back | -0.254 |
Token | girl |
Token ID | 2576 |
Feature activation | +3.45e-02 |
Pos logit contributions | |
cardboard | +0.210 |
lf | +0.206 |
ts | +0.194 |
camps | +0.188 |
blasted | +0.184 |
Neg logit contributions | |
eal | -0.166 |
arre | -0.161 |
umpy | -0.146 |
irk | -0.146 |
ags | -0.137 |
Token | named |
Token ID | 3706 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
ts | +0.045 |
camps | +0.042 |
swords | +0.040 |
cardboard | +0.040 |
loot | +0.038 |
Neg logit contributions | |
eal | -0.067 |
umpy | -0.065 |
back | -0.063 |
irk | -0.062 |
itter | -0.061 |
Token | Lucy |
Token ID | 22162 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
cardboard | +0.182 |
lf | +0.179 |
ts | +0.174 |
camps | +0.167 |
trek | +0.161 |
Neg logit contributions | |
eal | -0.169 |
arre | -0.158 |
umpy | -0.152 |
irk | -0.147 |
back | -0.142 |
Token | was |
Token ID | 373 |
Feature activation | +4.33e-02 |
Pos logit contributions | |
cardboard | +0.028 |
lf | +0.027 |
camps | +0.026 |
seas | +0.026 |
dice | +0.025 |
Neg logit contributions | |
eal | -0.030 |
umpy | -0.028 |
back | -0.027 |
arre | -0.026 |
irk | -0.026 |
Token | in |
Token ID | 287 |
Feature activation | +3.78e-02 |
Pos logit contributions | |
cardboard | +0.072 |
ts | +0.069 |
camps | +0.066 |
loot | +0.065 |
trek | +0.065 |
Neg logit contributions | |
umpy | -0.066 |
eal | -0.063 |
arre | -0.062 |
vious | -0.059 |
back | -0.059 |
Token | a |
Token ID | 257 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
lf | +0.060 |
camps | +0.060 |
ts | +0.059 |
cardboard | +0.059 |
loot | +0.059 |
Neg logit contributions | |
eal | -0.048 |
back | -0.048 |
agree | -0.048 |
ags | -0.047 |
arre | -0.047 |
Token | big |
Token ID | 1263 |
Feature activation | +7.40e-02 |
Pos logit contributions | |
cardboard | +0.200 |
lf | +0.198 |
ts | +0.185 |
camps | +0.171 |
trek | +0.166 |
Neg logit contributions | |
eal | -0.341 |
arre | -0.333 |
umpy | -0.315 |
irk | -0.311 |
vious | -0.296 |
Token | park |
Token ID | 3952 |
Feature activation | +4.62e-02 |
Pos logit contributions | |
cardboard | +0.094 |
lf | +0.091 |
ts | +0.081 |
camps | +0.078 |
blasted | +0.075 |
Neg logit contributions | |
eal | -0.159 |
arre | -0.154 |
irk | -0.145 |
umpy | -0.145 |
ags | -0.139 |
Token | . |
Token ID | 13 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
cardboard | +0.077 |
ts | +0.073 |
lf | +0.072 |
camps | +0.071 |
trek | +0.067 |
Neg logit contributions | |
eal | -0.074 |
umpy | -0.070 |
arre | -0.067 |
irk | -0.065 |
back | -0.065 |
Token | A |
Token ID | 317 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
eal | +0.017 |
arre | +0.015 |
umpy | +0.015 |
irk | +0.014 |
vious | +0.014 |
Neg logit contributions | |
lf | -0.031 |
ts | -0.031 |
cardboard | -0.031 |
camps | -0.030 |
blasted | -0.028 |
Token | little |
Token ID | 1310 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
cardboard | +0.237 |
lf | +0.232 |
ts | +0.217 |
camps | +0.204 |
trek | +0.197 |
Neg logit contributions | |
eal | -0.364 |
arre | -0.352 |
umpy | -0.334 |
irk | -0.331 |
vious | -0.313 |
Token | boy |
Token ID | 2933 |
Feature activation | +4.88e-02 |
Pos logit contributions | |
cardboard | +0.179 |
lf | +0.168 |
ts | +0.155 |
shovel | +0.153 |
camps | +0.151 |
Neg logit contributions | |
eal | -0.185 |
arre | -0.184 |
irk | -0.164 |
umpy | -0.163 |
ags | -0.161 |
Token | named |
Token ID | 3706 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
camps | +0.067 |
ts | +0.063 |
cardboard | +0.060 |
miles | +0.058 |
trek | +0.058 |
Neg logit contributions | |
umpy | -0.086 |
eal | -0.085 |
back | -0.080 |
crow | -0.079 |
irk | -0.079 |
Token | Tim |
Token ID | 5045 |
Feature activation | +3.88e-02 |
Pos logit contributions | |
lf | +0.194 |
ts | +0.183 |
cardboard | +0.178 |
hours | +0.171 |
blasted | +0.168 |
Neg logit contributions | |
back | -0.131 |
eal | -0.125 |
umpy | -0.107 |
ags | -0.106 |
agree | -0.104 |
Token | was |
Token ID | 373 |
Feature activation | +4.11e-02 |
Pos logit contributions | |
cardboard | +0.071 |
lf | +0.071 |
ts | +0.066 |
camps | +0.064 |
blasted | +0.063 |
Neg logit contributions | |
eal | -0.057 |
umpy | -0.055 |
arre | -0.054 |
irk | -0.051 |
back | -0.050 |
Token | day |
Token ID | 1110 |
Feature activation | -7.20e-03 |
Pos logit contributions | |
cardboard | +0.149 |
lf | +0.146 |
ts | +0.132 |
camps | +0.129 |
trek | +0.127 |
Neg logit contributions | |
eal | -0.169 |
arre | -0.165 |
irk | -0.151 |
umpy | -0.148 |
ags | -0.145 |
Token | something |
Token ID | 1223 |
Feature activation | +2.93e-02 |
Pos logit contributions | |
eal | +0.012 |
arre | +0.011 |
back | +0.011 |
ealing | +0.011 |
mad | +0.010 |
Neg logit contributions | |
lf | -0.011 |
cardboard | -0.010 |
ts | -0.010 |
camps | -0.010 |
blasted | -0.009 |
Token | special |
Token ID | 2041 |
Feature activation | -5.93e-02 |
Pos logit contributions | |
cardboard | +0.044 |
lf | +0.043 |
ts | +0.039 |
camps | +0.037 |
blasted | +0.036 |
Neg logit contributions | |
eal | -0.056 |
arre | -0.055 |
irk | -0.051 |
umpy | -0.051 |
ags | -0.049 |
Token | happened |
Token ID | 3022 |
Feature activation | +1.60e-03 |
Pos logit contributions | |
eal | +0.100 |
arre | +0.095 |
irk | +0.089 |
umpy | +0.089 |
vious | +0.087 |
Neg logit contributions | |
cardboard | -0.095 |
lf | -0.092 |
ts | -0.090 |
camps | -0.088 |
trek | -0.083 |
Token | . |
Token ID | 13 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
cardboard | +0.002 |
lf | +0.002 |
ts | +0.002 |
camps | +0.002 |
blasted | +0.002 |
Neg logit contributions | |
eal | -0.003 |
arre | -0.003 |
irk | -0.003 |
umpy | -0.003 |
ags | -0.003 |
Token | A |
Token ID | 317 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
eal | +0.018 |
arre | +0.017 |
umpy | +0.016 |
irk | +0.016 |
vious | +0.015 |
Neg logit contributions | |
lf | -0.023 |
cardboard | -0.023 |
ts | -0.022 |
camps | -0.021 |
blasted | -0.020 |
Token | big |
Token ID | 1263 |
Feature activation | +7.49e-02 |
Pos logit contributions | |
cardboard | +0.236 |
lf | +0.232 |
ts | +0.220 |
camps | +0.205 |
trek | +0.197 |
Neg logit contributions | |
eal | -0.360 |
arre | -0.348 |
umpy | -0.330 |
irk | -0.329 |
vious | -0.310 |
Token | bird |
Token ID | 6512 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
cardboard | +0.100 |
lf | +0.098 |
ts | +0.090 |
camps | +0.085 |
blasted | +0.082 |
Neg logit contributions | |
eal | -0.153 |
arre | -0.148 |
umpy | -0.139 |
irk | -0.139 |
ags | -0.132 |
Token | flew |
Token ID | 13112 |
Feature activation | +6.40e-02 |
Pos logit contributions | |
cardboard | +0.025 |
lf | +0.024 |
ts | +0.024 |
camps | +0.024 |
fossils | +0.023 |
Neg logit contributions | |
eal | -0.037 |
arre | -0.036 |
vious | -0.036 |
umpy | -0.035 |
awk | -0.035 |
Token | by |
Token ID | 416 |
Feature activation | -6.00e-02 |
Pos logit contributions | |
cardboard | +0.107 |
called | +0.096 |
blasted | +0.093 |
trek | +0.092 |
lf | +0.092 |
Neg logit contributions | |
eal | -0.111 |
arre | -0.111 |
irk | -0.101 |
umpy | -0.098 |
fault | -0.096 |
Token | and |
Token ID | 290 |
Feature activation | -8.44e-03 |
Pos logit contributions | |
eal | +0.098 |
arre | +0.095 |
umpy | +0.092 |
irk | +0.090 |
back | +0.090 |
Neg logit contributions | |
lf | -0.096 |
cardboard | -0.095 |
ts | -0.088 |
camps | -0.088 |
blasted | -0.084 |
Token | was |
Token ID | 373 |
Feature activation | +3.68e-02 |
Pos logit contributions | |
cardboard | +0.099 |
ts | +0.094 |
lf | +0.090 |
swords | +0.088 |
camps | +0.088 |
Neg logit contributions | |
eal | -0.106 |
arre | -0.103 |
umpy | -0.098 |
back | -0.098 |
irk | -0.097 |
Token | a |
Token ID | 257 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
lf | +0.064 |
ts | +0.060 |
blasted | +0.057 |
hunt | +0.053 |
cardboard | +0.051 |
Neg logit contributions | |
arre | -0.055 |
eal | -0.053 |
ags | -0.053 |
back | -0.053 |
vious | -0.052 |
Token | little |
Token ID | 1310 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
ts | +0.239 |
cardboard | +0.237 |
lf | +0.236 |
trek | +0.217 |
blasted | +0.216 |
Neg logit contributions | |
eal | -0.282 |
arre | -0.267 |
umpy | -0.266 |
irk | -0.260 |
back | -0.249 |
Token | girl |
Token ID | 2576 |
Feature activation | +3.18e-02 |
Pos logit contributions | |
lf | +0.201 |
cardboard | +0.201 |
ts | +0.191 |
camps | +0.182 |
trek | +0.181 |
Neg logit contributions | |
eal | -0.146 |
arre | -0.138 |
irk | -0.128 |
umpy | -0.128 |
awk | -0.116 |
Token | named |
Token ID | 3706 |
Feature activation | +9.81e-02 |
Pos logit contributions | |
loot | +0.044 |
ts | +0.043 |
fossils | +0.041 |
cardboard | +0.041 |
dice | +0.041 |
Neg logit contributions | |
back | -0.056 |
umpy | -0.054 |
irk | -0.053 |
eal | -0.053 |
olding | -0.052 |
Token | A |
Token ID | 317 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
cardboard | +0.160 |
lf | +0.158 |
ts | +0.150 |
camps | +0.143 |
blasted | +0.140 |
Neg logit contributions | |
eal | -0.172 |
arre | -0.165 |
umpy | -0.155 |
irk | -0.153 |
ags | -0.145 |
Token | yl |
Token ID | 2645 |
Feature activation | -2.37e-02 |
Pos logit contributions | |
cardboard | +0.241 |
lf | +0.236 |
ts | +0.217 |
camps | +0.206 |
blasted | +0.199 |
Neg logit contributions | |
eal | -0.361 |
arre | -0.351 |
umpy | -0.329 |
irk | -0.329 |
ags | -0.313 |
Token | a |
Token ID | 64 |
Feature activation | +2.08e-02 |
Pos logit contributions | |
eal | +0.033 |
arre | +0.032 |
umpy | +0.031 |
irk | +0.029 |
vious | +0.028 |
Neg logit contributions | |
cardboard | -0.046 |
lf | -0.042 |
camps | -0.042 |
ts | -0.042 |
shovel | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | -9.71e-03 |
Pos logit contributions | |
cardboard | +0.029 |
ts | +0.026 |
camps | +0.026 |
trek | +0.025 |
blasted | +0.025 |
Neg logit contributions | |
eal | -0.038 |
umpy | -0.038 |
irk | -0.035 |
arre | -0.035 |
back | -0.035 |
Token | She |
Token ID | 1375 |
Feature activation | +6.59e-02 |
Pos logit contributions | |
eal | +0.014 |
arre | +0.014 |
umpy | +0.012 |
irk | +0.012 |
ealing | +0.012 |
Neg logit contributions | |
cardboard | -0.019 |
lf | -0.018 |
camps | -0.017 |
ts | -0.017 |
pretend | -0.017 |
Token | was |
Token ID | 373 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
cardboard | +0.101 |
camps | +0.093 |
lf | +0.091 |
shovel | +0.089 |
ts | +0.087 |
Neg logit contributions | |
eal | -0.117 |
umpy | -0.113 |
arre | -0.109 |
irk | -0.103 |
back | -0.101 |
Token | , |
Token ID | 11 |
Feature activation | +7.54e-02 |
Pos logit contributions | |
cardboard | +0.002 |
lf | +0.002 |
ts | +0.002 |
camps | +0.002 |
blasted | +0.002 |
Neg logit contributions | |
eal | -0.004 |
arre | -0.004 |
umpy | -0.004 |
irk | -0.004 |
back | -0.004 |
Token | something |
Token ID | 1223 |
Feature activation | +3.48e-02 |
Pos logit contributions | |
cardboard | +0.120 |
lf | +0.117 |
ts | +0.114 |
camps | +0.106 |
blasted | +0.105 |
Neg logit contributions | |
eal | -0.124 |
arre | -0.119 |
back | -0.110 |
irk | -0.108 |
umpy | -0.108 |
Token | magical |
Token ID | 10883 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
cardboard | +0.048 |
lf | +0.046 |
ts | +0.043 |
pretend | +0.041 |
called | +0.040 |
Neg logit contributions | |
arre | -0.071 |
eal | -0.071 |
irk | -0.066 |
umpy | -0.065 |
ags | -0.065 |
Token | happened |
Token ID | 3022 |
Feature activation | +6.50e-03 |
Pos logit contributions | |
cardboard | +0.027 |
lf | +0.026 |
ts | +0.025 |
camps | +0.025 |
trek | +0.023 |
Neg logit contributions | |
eal | -0.029 |
arre | -0.027 |
irk | -0.026 |
umpy | -0.026 |
vious | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
cardboard | +0.008 |
lf | +0.008 |
ts | +0.007 |
camps | +0.007 |
trek | +0.007 |
Neg logit contributions | |
eal | -0.014 |
arre | -0.014 |
irk | -0.013 |
umpy | -0.013 |
ags | -0.012 |
Token | A |
Token ID | 317 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
eal | +0.020 |
arre | +0.020 |
umpy | +0.018 |
irk | +0.018 |
ags | +0.017 |
Neg logit contributions | |
cardboard | -0.019 |
lf | -0.018 |
ts | -0.017 |
camps | -0.016 |
blasted | -0.016 |
Token | big |
Token ID | 1263 |
Feature activation | +7.86e-02 |
Pos logit contributions | |
cardboard | +0.251 |
lf | +0.250 |
ts | +0.247 |
camps | +0.227 |
trek | +0.220 |
Neg logit contributions | |
eal | -0.328 |
arre | -0.319 |
umpy | -0.304 |
irk | -0.303 |
back | -0.283 |
Token | rainbow |
Token ID | 27223 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
cardboard | +0.111 |
lf | +0.110 |
ts | +0.104 |
camps | +0.099 |
trek | +0.095 |
Neg logit contributions | |
eal | -0.150 |
arre | -0.145 |
irk | -0.137 |
umpy | -0.136 |
vious | -0.129 |
Token | appeared |
Token ID | 4120 |
Feature activation | +6.40e-02 |
Pos logit contributions | |
cardboard | +0.050 |
ts | +0.048 |
lf | +0.047 |
camps | +0.047 |
trek | +0.044 |
Neg logit contributions | |
eal | -0.079 |
arre | -0.076 |
umpy | -0.072 |
vious | -0.071 |
irk | -0.071 |
Token | in |
Token ID | 287 |
Feature activation | +3.91e-02 |
Pos logit contributions | |
lf | +0.098 |
cardboard | +0.096 |
ts | +0.092 |
camps | +0.082 |
blasted | +0.080 |
Neg logit contributions | |
eal | -0.114 |
arre | -0.112 |
umpy | -0.106 |
back | -0.105 |
irk | -0.104 |
Token | the |
Token ID | 262 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
lf | +0.067 |
ts | +0.067 |
cardboard | +0.062 |
blasted | +0.061 |
trek | +0.061 |
Neg logit contributions | |
back | -0.053 |
arre | -0.050 |
eal | -0.050 |
aback | -0.049 |
ags | -0.048 |
Token | something |
Token ID | 1223 |
Feature activation | +3.20e-02 |
Pos logit contributions | |
camps | +0.123 |
ts | +0.122 |
trek | +0.122 |
hunt | +0.118 |
lf | +0.114 |
Neg logit contributions | |
back | -0.074 |
mad | -0.065 |
disappointed | -0.062 |
Fighting | -0.061 |
eal | -0.061 |
Token | very |
Token ID | 845 |
Feature activation | -3.62e-02 |
Pos logit contributions | |
cardboard | +0.049 |
lf | +0.048 |
ts | +0.044 |
pretend | +0.041 |
blasted | +0.041 |
Neg logit contributions | |
eal | -0.060 |
arre | -0.060 |
irk | -0.055 |
umpy | -0.055 |
ags | -0.053 |
Token | special |
Token ID | 2041 |
Feature activation | -5.74e-02 |
Pos logit contributions | |
arre | +0.067 |
eal | +0.067 |
irk | +0.061 |
ags | +0.059 |
umpy | +0.059 |
Neg logit contributions | |
cardboard | -0.056 |
lf | -0.055 |
ts | -0.050 |
pretend | -0.047 |
camps | -0.047 |
Token | happened |
Token ID | 3022 |
Feature activation | +1.47e-03 |
Pos logit contributions | |
eal | +0.092 |
arre | +0.086 |
irk | +0.083 |
umpy | +0.082 |
vious | +0.080 |
Neg logit contributions | |
cardboard | -0.093 |
ts | -0.091 |
camps | -0.090 |
lf | -0.088 |
trek | -0.084 |
Token | . |
Token ID | 13 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
cardboard | +0.002 |
lf | +0.002 |
ts | +0.002 |
camps | +0.002 |
blasted | +0.002 |
Neg logit contributions | |
eal | -0.003 |
arre | -0.003 |
irk | -0.003 |
umpy | -0.003 |
ags | -0.003 |
Token | A |
Token ID | 317 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
eal | +0.021 |
arre | +0.020 |
umpy | +0.019 |
irk | +0.019 |
agree | +0.018 |
Neg logit contributions | |
lf | -0.027 |
cardboard | -0.027 |
ts | -0.026 |
camps | -0.025 |
blasted | -0.024 |
Token | tiny |
Token ID | 7009 |
Feature activation | +9.51e-02 |
Pos logit contributions | |
cardboard | +0.239 |
ts | +0.239 |
lf | +0.235 |
camps | +0.219 |
trek | +0.214 |
Neg logit contributions | |
eal | -0.343 |
arre | -0.340 |
umpy | -0.320 |
irk | -0.315 |
back | -0.296 |
Token | fairy |
Token ID | 25607 |
Feature activation | +4.24e-02 |
Pos logit contributions | |
lf | +0.152 |
cardboard | +0.152 |
ts | +0.149 |
camps | +0.141 |
trek | +0.138 |
Neg logit contributions | |
eal | -0.162 |
arre | -0.156 |
irk | -0.147 |
umpy | -0.147 |
vious | -0.137 |
Token | appeared |
Token ID | 4120 |
Feature activation | +6.28e-02 |
Pos logit contributions | |
camps | +0.058 |
ts | +0.056 |
cardboard | +0.054 |
fossils | +0.053 |
dice | +0.053 |
Neg logit contributions | |
eal | -0.073 |
back | -0.070 |
vious | -0.068 |
arre | -0.067 |
umpy | -0.066 |
Token | out |
Token ID | 503 |
Feature activation | -5.16e-03 |
Pos logit contributions | |
cardboard | +0.088 |
lf | +0.088 |
ts | +0.081 |
camps | +0.076 |
blasted | +0.074 |
Neg logit contributions | |
eal | -0.121 |
arre | -0.118 |
umpy | -0.112 |
irk | -0.111 |
ags | -0.106 |
Token | of |
Token ID | 286 |
Feature activation | +2.65e-02 |
Pos logit contributions | |
eal | +0.007 |
umpy | +0.007 |
arre | +0.007 |
irk | +0.007 |
vious | +0.007 |
Neg logit contributions | |
lf | -0.009 |
cardboard | -0.009 |
ts | -0.009 |
camps | -0.008 |
shovel | -0.008 |
Token | big |
Token ID | 1263 |
Feature activation | +7.76e-02 |
Pos logit contributions | |
cardboard | +0.222 |
lf | +0.217 |
ts | +0.199 |
camps | +0.190 |
blasted | +0.183 |
Neg logit contributions | |
eal | -0.320 |
arre | -0.310 |
umpy | -0.291 |
irk | -0.291 |
ags | -0.277 |
Token | surprise |
Token ID | 5975 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
cardboard | +0.119 |
lf | +0.115 |
ts | +0.104 |
camps | +0.101 |
blasted | +0.097 |
Neg logit contributions | |
eal | -0.146 |
arre | -0.142 |
umpy | -0.133 |
irk | -0.131 |
ags | -0.125 |
Token | was |
Token ID | 373 |
Feature activation | +3.76e-02 |
Pos logit contributions | |
cardboard | +0.017 |
lf | +0.016 |
ts | +0.016 |
camps | +0.015 |
blasted | +0.014 |
Neg logit contributions | |
eal | -0.020 |
arre | -0.020 |
irk | -0.018 |
umpy | -0.018 |
ags | -0.017 |
Token | there |
Token ID | 612 |
Feature activation | +7.03e-02 |
Pos logit contributions | |
cardboard | +0.048 |
lf | +0.047 |
ts | +0.047 |
camps | +0.043 |
blasted | +0.042 |
Neg logit contributions | |
eal | -0.071 |
arre | -0.069 |
back | -0.066 |
irk | -0.065 |
umpy | -0.065 |
Token | . |
Token ID | 13 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
cardboard | +0.095 |
lf | +0.093 |
ts | +0.086 |
camps | +0.081 |
blasted | +0.079 |
Neg logit contributions | |
eal | -0.142 |
arre | -0.137 |
umpy | -0.129 |
irk | -0.129 |
ags | -0.123 |
Token | A |
Token ID | 317 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
eal | +0.031 |
arre | +0.030 |
umpy | +0.028 |
irk | +0.028 |
ags | +0.027 |
Neg logit contributions | |
cardboard | -0.030 |
lf | -0.030 |
ts | -0.028 |
camps | -0.027 |
trek | -0.026 |
Token | big |
Token ID | 1263 |
Feature activation | +7.68e-02 |
Pos logit contributions | |
ts | +0.273 |
lf | +0.258 |
cardboard | +0.257 |
camps | +0.239 |
trek | +0.238 |
Neg logit contributions | |
eal | -0.305 |
arre | -0.293 |
irk | -0.280 |
umpy | -0.278 |
back | -0.273 |
Token | , |
Token ID | 11 |
Feature activation | +6.19e-02 |
Pos logit contributions | |
cardboard | +0.109 |
lf | +0.108 |
ts | +0.104 |
camps | +0.098 |
trek | +0.095 |
Neg logit contributions | |
eal | -0.146 |
arre | -0.140 |
irk | -0.132 |
umpy | -0.131 |
vious | -0.124 |
Token | fluffy |
Token ID | 39145 |
Feature activation | -5.89e-04 |
Pos logit contributions | |
cardboard | +0.086 |
lf | +0.085 |
ts | +0.079 |
camps | +0.075 |
blasted | +0.073 |
Neg logit contributions | |
eal | -0.121 |
arre | -0.117 |
umpy | -0.110 |
irk | -0.110 |
ags | -0.104 |
Token | bunny |
Token ID | 44915 |
Feature activation | +4.60e-02 |
Pos logit contributions | |
eal | +0.001 |
arre | +0.001 |
irk | +0.001 |
umpy | +0.001 |
ags | +0.001 |
Neg logit contributions | |
cardboard | -0.001 |
lf | -0.001 |
ts | -0.001 |
camps | -0.001 |
blasted | -0.001 |
Token | hopped |
Token ID | 45230 |
Feature activation | +6.28e-02 |
Pos logit contributions | |
cardboard | +0.058 |
lf | +0.058 |
ts | +0.058 |
camps | +0.053 |
fossils | +0.048 |
Neg logit contributions | |
eal | -0.090 |
arre | -0.084 |
umpy | -0.083 |
irk | -0.082 |
back | -0.081 |
Token | , |
Token ID | 11 |
Feature activation | +7.47e-02 |
Pos logit contributions | |
eal | +0.005 |
arre | +0.005 |
umpy | +0.004 |
irk | +0.004 |
awk | +0.004 |
Neg logit contributions | |
cardboard | -0.002 |
lf | -0.002 |
ts | -0.002 |
camps | -0.002 |
blasted | -0.002 |
Token | something |
Token ID | 1223 |
Feature activation | +3.41e-02 |
Pos logit contributions | |
cardboard | +0.111 |
lf | +0.109 |
ts | +0.108 |
camps | +0.103 |
blasted | +0.102 |
Neg logit contributions | |
eal | -0.118 |
arre | -0.115 |
back | -0.112 |
mad | -0.104 |
Fighting | -0.103 |
Token | magical |
Token ID | 10883 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
cardboard | +0.047 |
lf | +0.044 |
ts | +0.042 |
pretend | +0.040 |
called | +0.038 |
Neg logit contributions | |
eal | -0.072 |
arre | -0.072 |
irk | -0.066 |
ags | -0.065 |
umpy | -0.065 |
Token | happened |
Token ID | 3022 |
Feature activation | +7.13e-03 |
Pos logit contributions | |
cardboard | +0.025 |
lf | +0.025 |
ts | +0.024 |
camps | +0.023 |
trek | +0.022 |
Neg logit contributions | |
eal | -0.032 |
arre | -0.030 |
umpy | -0.029 |
irk | -0.029 |
vious | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
cardboard | +0.009 |
lf | +0.009 |
ts | +0.008 |
camps | +0.007 |
trek | +0.007 |
Neg logit contributions | |
eal | -0.016 |
arre | -0.015 |
irk | -0.014 |
umpy | -0.014 |
ags | -0.014 |
Token | A |
Token ID | 317 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
eal | +0.021 |
arre | +0.020 |
irk | +0.019 |
umpy | +0.019 |
ags | +0.018 |
Neg logit contributions | |
cardboard | -0.021 |
lf | -0.020 |
ts | -0.019 |
camps | -0.018 |
trek | -0.018 |
Token | little |
Token ID | 1310 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
cardboard | +0.252 |
lf | +0.251 |
ts | +0.250 |
camps | +0.229 |
trek | +0.222 |
Neg logit contributions | |
eal | -0.323 |
arre | -0.314 |
umpy | -0.300 |
irk | -0.298 |
back | -0.280 |
Token | girl |
Token ID | 2576 |
Feature activation | +3.54e-02 |
Pos logit contributions | |
cardboard | +0.192 |
lf | +0.191 |
ts | +0.179 |
camps | +0.171 |
blasted | +0.167 |
Neg logit contributions | |
eal | -0.187 |
arre | -0.180 |
irk | -0.167 |
umpy | -0.167 |
vious | -0.156 |
Token | has |
Token ID | 468 |
Feature activation | +5.52e-02 |
Pos logit contributions | |
ts | +0.048 |
cardboard | +0.047 |
lf | +0.046 |
camps | +0.045 |
fossils | +0.041 |
Neg logit contributions | |
eal | -0.068 |
irk | -0.061 |
arre | -0.061 |
back | -0.061 |
disappointed | -0.061 |
Token | seen |
Token ID | 1775 |
Feature activation | +7.99e-02 |
Pos logit contributions | |
lf | +0.067 |
cardboard | +0.067 |
ts | +0.061 |
camps | +0.059 |
blasted | +0.054 |
Neg logit contributions | |
eal | -0.116 |
arre | -0.113 |
umpy | -0.108 |
irk | -0.106 |
vious | -0.104 |
Token | the |
Token ID | 262 |
Feature activation | -3.22e-02 |
Pos logit contributions | |
cardboard | +0.107 |
lf | +0.106 |
ts | +0.099 |
camps | +0.096 |
blasted | +0.093 |
Neg logit contributions | |
eal | -0.152 |
arre | -0.149 |
umpy | -0.140 |
irk | -0.139 |
back | -0.135 |
Token | , |
Token ID | 11 |
Feature activation | +7.02e-02 |
Pos logit contributions | |
arre | +0.180 |
eal | +0.176 |
umpy | +0.172 |
ags | +0.171 |
irk | +0.171 |
Neg logit contributions | |
lf | -0.073 |
cardboard | -0.073 |
ts | -0.068 |
pretended | -0.068 |
called | -0.062 |
Token | something |
Token ID | 1223 |
Feature activation | +3.07e-02 |
Pos logit contributions | |
cardboard | +0.093 |
lf | +0.091 |
ts | +0.082 |
camps | +0.079 |
trek | +0.075 |
Neg logit contributions | |
eal | -0.146 |
arre | -0.140 |
irk | -0.133 |
umpy | -0.132 |
ags | -0.127 |
Token | special |
Token ID | 2041 |
Feature activation | -4.40e-02 |
Pos logit contributions | |
cardboard | +0.048 |
lf | +0.046 |
ts | +0.043 |
camps | +0.040 |
pretend | +0.039 |
Neg logit contributions | |
eal | -0.058 |
arre | -0.056 |
irk | -0.052 |
umpy | -0.052 |
ags | -0.050 |
Token | happened |
Token ID | 3022 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
eal | +0.084 |
arre | +0.081 |
umpy | +0.076 |
irk | +0.076 |
vious | +0.073 |
Neg logit contributions | |
cardboard | -0.062 |
lf | -0.062 |
ts | -0.058 |
camps | -0.055 |
trek | -0.053 |
Token | . |
Token ID | 13 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
cardboard | +0.012 |
lf | +0.012 |
ts | +0.010 |
pretend | +0.010 |
camps | +0.009 |
Neg logit contributions | |
eal | -0.024 |
arre | -0.023 |
irk | -0.022 |
umpy | -0.022 |
ags | -0.021 |
Token | A |
Token ID | 317 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
eal | +0.028 |
arre | +0.027 |
irk | +0.025 |
umpy | +0.025 |
ags | +0.024 |
Neg logit contributions | |
cardboard | -0.026 |
lf | -0.026 |
camps | -0.023 |
ts | -0.023 |
trek | -0.023 |
Token | nice |
Token ID | 3621 |
Feature activation | -5.03e-02 |
Pos logit contributions | |
cardboard | +0.258 |
lf | +0.254 |
ts | +0.240 |
camps | +0.226 |
blasted | +0.219 |
Neg logit contributions | |
eal | -0.337 |
arre | -0.326 |
umpy | -0.306 |
irk | -0.306 |
vious | -0.287 |
Token | man |
Token ID | 582 |
Feature activation | +1.51e-03 |
Pos logit contributions | |
eal | +0.090 |
arre | +0.089 |
umpy | +0.082 |
ags | +0.081 |
irk | +0.080 |
Neg logit contributions | |
cardboard | -0.081 |
lf | -0.079 |
ts | -0.069 |
camps | -0.068 |
shovel | -0.068 |
Token | came |
Token ID | 1625 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
cardboard | +0.002 |
lf | +0.002 |
ts | +0.002 |
camps | +0.002 |
seas | +0.002 |
Neg logit contributions | |
eal | -0.003 |
arre | -0.003 |
irk | -0.003 |
back | -0.003 |
umpy | -0.003 |
Token | to |
Token ID | 284 |
Feature activation | +2.36e-03 |
Pos logit contributions | |
cardboard | +0.035 |
lf | +0.032 |
trek | +0.030 |
camps | +0.030 |
ts | +0.030 |
Neg logit contributions | |
eal | -0.046 |
arre | -0.044 |
irk | -0.041 |
umpy | -0.041 |
ags | -0.039 |
Token | the |
Token ID | 262 |
Feature activation | -3.37e-02 |
Pos logit contributions | |
cardboard | +0.003 |
lf | +0.003 |
ts | +0.003 |
camps | +0.003 |
blasted | +0.003 |
Neg logit contributions | |
eal | -0.004 |
arre | -0.004 |
irk | -0.004 |
umpy | -0.004 |
ags | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
ts | +0.073 |
lf | +0.068 |
blasted | +0.066 |
loot | +0.065 |
hunt | +0.064 |
Neg logit contributions | |
back | -0.051 |
arre | -0.046 |
vious | -0.046 |
eal | -0.046 |
umpy | -0.045 |
Token | big |
Token ID | 1263 |
Feature activation | +7.29e-02 |
Pos logit contributions | |
ts | +0.240 |
cardboard | +0.236 |
lf | +0.235 |
trek | +0.218 |
blasted | +0.215 |
Neg logit contributions | |
eal | -0.283 |
arre | -0.274 |
umpy | -0.267 |
irk | -0.258 |
back | -0.250 |
Token | food |
Token ID | 2057 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
cardboard | +0.100 |
lf | +0.099 |
ts | +0.091 |
camps | +0.086 |
blasted | +0.084 |
Neg logit contributions | |
eal | -0.145 |
arre | -0.140 |
umpy | -0.131 |
irk | -0.131 |
ags | -0.125 |
Token | contest |
Token ID | 8414 |
Feature activation | +2.75e-02 |
Pos logit contributions | |
ts | +0.019 |
cardboard | +0.018 |
lf | +0.018 |
trek | +0.017 |
blasted | +0.017 |
Neg logit contributions | |
eal | -0.021 |
arre | -0.021 |
umpy | -0.021 |
irk | -0.021 |
back | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
cardboard | +0.045 |
ts | +0.043 |
lf | +0.042 |
camps | +0.041 |
trek | +0.039 |
Neg logit contributions | |
eal | -0.045 |
arre | -0.043 |
umpy | -0.043 |
irk | -0.040 |
back | -0.039 |
Token | A |
Token ID | 317 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
eal | +0.023 |
arre | +0.023 |
umpy | +0.021 |
irk | +0.021 |
vious | +0.020 |
Neg logit contributions | |
lf | -0.028 |
cardboard | -0.027 |
ts | -0.027 |
camps | -0.025 |
trek | -0.025 |
Token | kind |
Token ID | 1611 |
Feature activation | -5.00e-02 |
Pos logit contributions | |
cardboard | +0.264 |
lf | +0.258 |
ts | +0.248 |
camps | +0.233 |
trek | +0.229 |
Neg logit contributions | |
eal | -0.328 |
arre | -0.319 |
umpy | -0.300 |
irk | -0.295 |
vious | -0.276 |
Token | judge |
Token ID | 5052 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
arre | +0.081 |
eal | +0.081 |
ags | +0.071 |
Fighting | +0.071 |
agree | +0.070 |
Neg logit contributions | |
cardboard | -0.089 |
lf | -0.088 |
ts | -0.075 |
camps | -0.075 |
shovel | -0.075 |
Token | was |
Token ID | 373 |
Feature activation | +3.78e-02 |
Pos logit contributions | |
camps | +0.028 |
cardboard | +0.027 |
ts | +0.026 |
lf | +0.025 |
trek | +0.025 |
Neg logit contributions | |
eal | -0.046 |
umpy | -0.044 |
arre | -0.044 |
vious | -0.043 |
irk | -0.042 |
Token | there |
Token ID | 612 |
Feature activation | +6.48e-02 |
Pos logit contributions | |
cardboard | +0.051 |
lf | +0.050 |
ts | +0.046 |
camps | +0.045 |
trek | +0.043 |
Neg logit contributions | |
eal | -0.075 |
arre | -0.073 |
umpy | -0.069 |
irk | -0.068 |
ags | -0.065 |
Token | to |
Token ID | 284 |
Feature activation | +2.78e-03 |
Pos logit contributions | |
cardboard | +0.101 |
lf | +0.100 |
ts | +0.093 |
camps | +0.090 |
blasted | +0.087 |
Neg logit contributions | |
eal | -0.112 |
arre | -0.110 |
umpy | -0.104 |
irk | -0.100 |
back | -0.097 |
Token | key |
Token ID | 1994 |
Feature activation | +3.97e-02 |
Pos logit contributions | |
ts | +0.048 |
lf | +0.046 |
camps | +0.045 |
cardboard | +0.045 |
pretended | +0.043 |
Neg logit contributions | |
back | -0.064 |
eal | -0.062 |
cheek | -0.058 |
umpy | -0.057 |
disappointed | -0.056 |
Token | could |
Token ID | 714 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
camps | +0.057 |
ts | +0.054 |
cardboard | +0.054 |
trek | +0.050 |
lf | +0.050 |
Neg logit contributions | |
eal | -0.068 |
back | -0.067 |
umpy | -0.066 |
disappointed | -0.064 |
olding | -0.061 |
Token | unlock |
Token ID | 12116 |
Feature activation | +7.63e-02 |
Pos logit contributions | |
camps | +0.126 |
cardboard | +0.126 |
lf | +0.121 |
ts | +0.119 |
hours | +0.111 |
Neg logit contributions | |
eal | -0.189 |
back | -0.179 |
umpy | -0.176 |
agree | -0.176 |
disappointed | -0.171 |
Token | anything |
Token ID | 1997 |
Feature activation | -4.79e-03 |
Pos logit contributions | |
lf | +0.120 |
camps | +0.118 |
cardboard | +0.117 |
trek | +0.112 |
blasted | +0.111 |
Neg logit contributions | |
back | -0.115 |
eal | -0.108 |
ags | -0.102 |
disappointed | -0.100 |
arre | -0.100 |
Token | ! |
Token ID | 0 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
back | +0.008 |
disappointed | +0.007 |
umpy | +0.007 |
eal | +0.007 |
forward | +0.007 |
Neg logit contributions | |
lf | -0.007 |
camps | -0.007 |
cardboard | -0.007 |
trek | -0.007 |
ts | -0.007 |
Token | A |
Token ID | 317 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
eal | +0.020 |
arre | +0.019 |
umpy | +0.018 |
irk | +0.018 |
ags | +0.017 |
Neg logit contributions | |
lf | -0.026 |
cardboard | -0.026 |
ts | -0.025 |
camps | -0.024 |
blasted | -0.023 |
Token | little |
Token ID | 1310 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
ts | +0.315 |
camps | +0.276 |
lf | +0.275 |
trek | +0.275 |
cardboard | +0.266 |
Neg logit contributions | |
back | -0.256 |
eal | -0.250 |
umpy | -0.245 |
cheek | -0.242 |
irk | -0.235 |
Token | girl |
Token ID | 2576 |
Feature activation | +3.22e-02 |
Pos logit contributions | |
lf | +0.222 |
ts | +0.213 |
cardboard | +0.208 |
camps | +0.205 |
trek | +0.201 |
Neg logit contributions | |
eal | -0.121 |
arre | -0.107 |
irk | -0.107 |
umpy | -0.104 |
back | -0.101 |
Token | named |
Token ID | 3706 |
Feature activation | +9.72e-02 |
Pos logit contributions | |
camps | +0.049 |
ts | +0.048 |
swords | +0.043 |
seas | +0.043 |
cardboard | +0.043 |
Neg logit contributions | |
eal | -0.054 |
back | -0.053 |
umpy | -0.050 |
disappointed | -0.049 |
irk | -0.049 |
Token | Quick |
Token ID | 12029 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
lf | +0.180 |
cardboard | +0.180 |
camps | +0.177 |
ts | +0.174 |
trek | +0.166 |
Neg logit contributions | |
eal | -0.138 |
arre | -0.122 |
umpy | -0.122 |
back | -0.122 |
itter | -0.118 |
Token | was |
Token ID | 373 |
Feature activation | +3.78e-02 |
Pos logit contributions | |
cardboard | +0.150 |
lf | +0.148 |
ts | +0.135 |
camps | +0.133 |
blasted | +0.127 |
Neg logit contributions | |
eal | -0.195 |
arre | -0.185 |
irk | -0.175 |
umpy | -0.175 |
vious | -0.168 |
Token | you |
Token ID | 345 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
eal | +0.051 |
arre | +0.050 |
umpy | +0.047 |
irk | +0.046 |
vious | +0.044 |
Neg logit contributions | |
lf | -0.037 |
cardboard | -0.036 |
ts | -0.034 |
camps | -0.030 |
pretend | -0.030 |
Token | for |
Token ID | 329 |
Feature activation | -6.94e-02 |
Pos logit contributions | |
eal | +0.231 |
arre | +0.225 |
irk | +0.213 |
umpy | +0.212 |
vious | +0.204 |
Neg logit contributions | |
cardboard | -0.126 |
lf | -0.122 |
camps | -0.105 |
ts | -0.104 |
blasted | -0.099 |
Token | supplying |
Token ID | 28099 |
Feature activation | -8.35e-03 |
Pos logit contributions | |
eal | +0.125 |
arre | +0.123 |
irk | +0.111 |
ags | +0.111 |
umpy | +0.111 |
Neg logit contributions | |
cardboard | -0.103 |
lf | -0.100 |
camps | -0.096 |
ts | -0.093 |
trek | -0.091 |
Token | us |
Token ID | 514 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
eal | +0.014 |
arre | +0.014 |
umpy | +0.013 |
irk | +0.013 |
vious | +0.012 |
Neg logit contributions | |
cardboard | -0.014 |
lf | -0.014 |
ts | -0.013 |
camps | -0.013 |
pretend | -0.012 |
Token | with |
Token ID | 351 |
Feature activation | -2.82e-02 |
Pos logit contributions | |
eal | +0.282 |
arre | +0.270 |
irk | +0.269 |
umpy | +0.257 |
vious | +0.248 |
Neg logit contributions | |
cardboard | -0.158 |
lf | -0.153 |
ts | -0.145 |
camps | -0.133 |
pretend | -0.129 |
Token | panc |
Token ID | 22154 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
eal | +0.051 |
arre | +0.050 |
umpy | +0.046 |
irk | +0.046 |
ags | +0.045 |
Neg logit contributions | |
lf | -0.042 |
cardboard | -0.042 |
ts | -0.039 |
camps | -0.037 |
blasted | -0.037 |
Token | ake |
Token ID | 539 |
Feature activation | -6.51e-02 |
Pos logit contributions | |
eal | +0.272 |
arre | +0.249 |
irk | +0.237 |
umpy | +0.235 |
vious | +0.218 |
Neg logit contributions | |
lf | -0.455 |
cardboard | -0.454 |
ts | -0.423 |
camps | -0.409 |
trek | -0.401 |
Token | mix |
Token ID | 5022 |
Feature activation | -3.84e-02 |
Pos logit contributions | |
eal | +0.137 |
irk | +0.132 |
arre | +0.132 |
umpy | +0.127 |
vious | +0.122 |
Neg logit contributions | |
lf | -0.087 |
cardboard | -0.084 |
ts | -0.077 |
camps | -0.072 |
pretend | -0.070 |
Token | , |
Token ID | 11 |
Feature activation | +5.50e-02 |
Pos logit contributions | |
eal | +0.080 |
arre | +0.077 |
irk | +0.074 |
umpy | +0.073 |
agree | +0.069 |
Neg logit contributions | |
cardboard | -0.051 |
lf | -0.049 |
ts | -0.046 |
camps | -0.043 |
pretend | -0.040 |
Token | mom |
Token ID | 1995 |
Feature activation | -3.98e-02 |
Pos logit contributions | |
cardboard | +0.070 |
lf | +0.068 |
camps | +0.059 |
pretend | +0.059 |
pretended | +0.057 |
Neg logit contributions | |
eal | -0.117 |
arre | -0.114 |
umpy | -0.111 |
irk | -0.110 |
ags | -0.107 |
Token | and |
Token ID | 290 |
Feature activation | -1.66e-02 |
Pos logit contributions | |
eal | +0.074 |
arre | +0.070 |
irk | +0.067 |
ags | +0.065 |
agree | +0.065 |
Neg logit contributions | |
cardboard | -0.060 |
lf | -0.059 |
camps | -0.053 |
pretended | -0.052 |
ts | -0.052 |
Token | mom |
Token ID | 1995 |
Feature activation | -3.88e-02 |
Pos logit contributions | |
eal | +0.019 |
arre | +0.018 |
irk | +0.017 |
back | +0.017 |
mad | +0.016 |
Neg logit contributions | |
cardboard | -0.019 |
lf | -0.018 |
ts | -0.018 |
camps | -0.017 |
trek | -0.016 |
Token | came |
Token ID | 1625 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
eal | +0.068 |
arre | +0.066 |
irk | +0.059 |
back | +0.058 |
umpy | +0.058 |
Neg logit contributions | |
cardboard | -0.065 |
lf | -0.064 |
ts | -0.061 |
camps | -0.057 |
fossils | -0.056 |
Token | back |
Token ID | 736 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
cardboard | +0.028 |
lf | +0.026 |
ts | +0.025 |
camps | +0.025 |
pretend | +0.024 |
Neg logit contributions | |
eal | -0.026 |
arre | -0.025 |
umpy | -0.024 |
irk | -0.023 |
vious | -0.023 |
Token | with |
Token ID | 351 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
eal | +0.303 |
arre | +0.293 |
umpy | +0.273 |
irk | +0.272 |
vious | +0.258 |
Neg logit contributions | |
cardboard | -0.258 |
lf | -0.252 |
ts | -0.234 |
camps | -0.226 |
trek | -0.218 |
Token | the |
Token ID | 262 |
Feature activation | -3.97e-02 |
Pos logit contributions | |
eal | +0.054 |
arre | +0.052 |
irk | +0.050 |
back | +0.049 |
ags | +0.049 |
Neg logit contributions | |
lf | -0.037 |
cardboard | -0.034 |
blasted | -0.032 |
camps | -0.032 |
ts | -0.031 |
Token | panc |
Token ID | 22154 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
eal | +0.074 |
arre | +0.071 |
irk | +0.067 |
umpy | +0.067 |
ags | +0.063 |
Neg logit contributions | |
lf | -0.059 |
cardboard | -0.059 |
ts | -0.055 |
camps | -0.052 |
blasted | -0.051 |
Token | ake |
Token ID | 539 |
Feature activation | -6.50e-02 |
Pos logit contributions | |
eal | +0.291 |
arre | +0.264 |
irk | +0.250 |
umpy | +0.247 |
vious | +0.239 |
Neg logit contributions | |
lf | -0.448 |
cardboard | -0.439 |
ts | -0.400 |
camps | -0.390 |
trek | -0.384 |
Token | mix |
Token ID | 5022 |
Feature activation | -3.93e-02 |
Pos logit contributions | |
eal | +0.144 |
arre | +0.141 |
irk | +0.134 |
umpy | +0.133 |
ags | +0.128 |
Neg logit contributions | |
cardboard | -0.080 |
lf | -0.079 |
ts | -0.067 |
camps | -0.061 |
pretend | -0.060 |
Token | and |
Token ID | 290 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
eal | +0.088 |
arre | +0.086 |
umpy | +0.082 |
irk | +0.082 |
agree | +0.079 |
Neg logit contributions | |
cardboard | -0.045 |
ts | -0.041 |
lf | -0.041 |
shovel | -0.035 |
called | -0.035 |
Token | they |
Token ID | 484 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
eal | +0.026 |
arre | +0.025 |
irk | +0.023 |
umpy | +0.023 |
back | +0.022 |
Neg logit contributions | |
cardboard | -0.016 |
lf | -0.016 |
camps | -0.015 |
ts | -0.015 |
trek | -0.014 |
Token | made |
Token ID | 925 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
eal | +0.036 |
arre | +0.035 |
back | +0.032 |
disappointed | +0.031 |
irk | +0.031 |
Neg logit contributions | |
cardboard | -0.026 |
lf | -0.026 |
camps | -0.023 |
ts | -0.023 |
fossils | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | -3.42e-04 |
Pos logit contributions | |
eal | +0.050 |
arre | +0.048 |
umpy | +0.043 |
irk | +0.043 |
ags | +0.041 |
Neg logit contributions | |
cardboard | -0.063 |
lf | -0.062 |
ts | -0.059 |
camps | -0.057 |
blasted | -0.056 |
Token | come |
Token ID | 1282 |
Feature activation | -7.68e-02 |
Pos logit contributions | |
eal | +0.001 |
arre | +0.001 |
umpy | +0.001 |
irk | +0.001 |
ags | +0.001 |
Neg logit contributions | |
cardboard | -0.001 |
lf | -0.001 |
ts | -0.000 |
camps | -0.000 |
blasted | -0.000 |
Token | back |
Token ID | 736 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
eal | +0.108 |
arre | +0.103 |
irk | +0.094 |
umpy | +0.094 |
vious | +0.087 |
Neg logit contributions | |
cardboard | -0.154 |
lf | -0.150 |
ts | -0.142 |
camps | -0.139 |
blasted | -0.135 |
Token | with |
Token ID | 351 |
Feature activation | -2.52e-02 |
Pos logit contributions | |
eal | +0.301 |
arre | +0.292 |
umpy | +0.273 |
irk | +0.271 |
back | +0.262 |
Neg logit contributions | |
cardboard | -0.257 |
lf | -0.256 |
ts | -0.243 |
camps | -0.222 |
blasted | -0.222 |
Token | the |
Token ID | 262 |
Feature activation | -3.96e-02 |
Pos logit contributions | |
eal | +0.045 |
arre | +0.043 |
irk | +0.042 |
ags | +0.041 |
back | +0.040 |
Neg logit contributions | |
lf | -0.038 |
cardboard | -0.037 |
blasted | -0.035 |
ts | -0.034 |
camps | -0.033 |
Token | panc |
Token ID | 22154 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
eal | +0.077 |
arre | +0.075 |
irk | +0.072 |
umpy | +0.070 |
agree | +0.068 |
Neg logit contributions | |
cardboard | -0.058 |
lf | -0.055 |
camps | -0.052 |
ts | -0.051 |
pretend | -0.049 |
Token | ake |
Token ID | 539 |
Feature activation | -6.24e-02 |
Pos logit contributions | |
eal | +0.295 |
arre | +0.267 |
irk | +0.259 |
umpy | +0.253 |
vious | +0.245 |
Neg logit contributions | |
lf | -0.433 |
cardboard | -0.418 |
camps | -0.383 |
ts | -0.376 |
trek | -0.368 |
Token | mix |
Token ID | 5022 |
Feature activation | -3.60e-02 |
Pos logit contributions | |
eal | +0.132 |
arre | +0.128 |
irk | +0.122 |
umpy | +0.121 |
ags | +0.116 |
Neg logit contributions | |
cardboard | -0.081 |
lf | -0.080 |
ts | -0.071 |
camps | -0.067 |
blasted | -0.064 |
Token | , |
Token ID | 11 |
Feature activation | +5.72e-02 |
Pos logit contributions | |
eal | +0.079 |
arre | +0.078 |
irk | +0.075 |
umpy | +0.074 |
agree | +0.072 |
Neg logit contributions | |
cardboard | -0.041 |
lf | -0.041 |
ts | -0.038 |
camps | -0.033 |
pretend | -0.033 |
Token | but |
Token ID | 475 |
Feature activation | -1.35e-03 |
Pos logit contributions | |
cardboard | +0.064 |
lf | +0.061 |
ts | +0.055 |
pretended | +0.054 |
pretend | +0.053 |
Neg logit contributions | |
eal | -0.130 |
arre | -0.127 |
umpy | -0.123 |
irk | -0.123 |
awk | -0.117 |
Token | she |
Token ID | 673 |
Feature activation | +1.29e-04 |
Pos logit contributions | |
eal | +0.003 |
arre | +0.003 |
umpy | +0.003 |
irk | +0.003 |
ags | +0.003 |
Neg logit contributions | |
lf | -0.002 |
cardboard | -0.002 |
ts | -0.001 |
pretended | -0.001 |
pretend | -0.001 |
Token | he |
Token ID | 339 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
eal | +0.027 |
arre | +0.026 |
back | +0.025 |
irk | +0.024 |
disappointed | +0.024 |
Neg logit contributions | |
cardboard | -0.020 |
lf | -0.020 |
camps | -0.019 |
ts | -0.019 |
trek | -0.018 |
Token | became |
Token ID | 2627 |
Feature activation | +2.39e-02 |
Pos logit contributions | |
cardboard | +0.027 |
lf | +0.026 |
camps | +0.024 |
ts | +0.023 |
fossils | +0.023 |
Neg logit contributions | |
eal | -0.037 |
arre | -0.037 |
umpy | -0.033 |
back | -0.033 |
disappointed | -0.032 |
Token | a |
Token ID | 257 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
cardboard | +0.038 |
lf | +0.035 |
ts | +0.035 |
camps | +0.032 |
pretend | +0.032 |
Neg logit contributions | |
eal | -0.044 |
arre | -0.042 |
irk | -0.040 |
umpy | -0.040 |
awk | -0.038 |
Token | blue |
Token ID | 4171 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
lf | +0.237 |
cardboard | +0.229 |
ts | +0.222 |
trek | +0.200 |
blasted | +0.200 |
Neg logit contributions | |
eal | -0.276 |
arre | -0.272 |
umpy | -0.253 |
irk | -0.248 |
back | -0.238 |
Token | berry |
Token ID | 8396 |
Feature activation | +3.72e-02 |
Pos logit contributions | |
lf | +0.034 |
ts | +0.033 |
cardboard | +0.033 |
blasted | +0.030 |
trek | +0.030 |
Neg logit contributions | |
eal | -0.030 |
arre | -0.028 |
irk | -0.027 |
umpy | -0.027 |
vious | -0.025 |
Token | panc |
Token ID | 22154 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
cardboard | +0.052 |
lf | +0.051 |
ts | +0.047 |
camps | +0.045 |
blasted | +0.044 |
Neg logit contributions | |
eal | -0.073 |
arre | -0.071 |
umpy | -0.066 |
irk | -0.066 |
ags | -0.063 |
Token | ake |
Token ID | 539 |
Feature activation | -6.10e-02 |
Pos logit contributions | |
eal | +0.241 |
arre | +0.232 |
umpy | +0.203 |
irk | +0.201 |
ags | +0.183 |
Neg logit contributions | |
cardboard | -0.481 |
lf | -0.471 |
ts | -0.456 |
camps | -0.439 |
blasted | -0.435 |
Token | . |
Token ID | 13 |
Feature activation | -2.52e-02 |
Pos logit contributions | |
eal | +0.125 |
arre | +0.121 |
irk | +0.115 |
umpy | +0.113 |
ags | +0.109 |
Neg logit contributions | |
lf | -0.083 |
cardboard | -0.083 |
ts | -0.075 |
camps | -0.072 |
pretend | -0.067 |
Token | It |
Token ID | 632 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
eal | +0.056 |
arre | +0.054 |
irk | +0.052 |
umpy | +0.051 |
awk | +0.049 |
Neg logit contributions | |
cardboard | -0.030 |
lf | -0.029 |
ts | -0.025 |
camps | -0.025 |
blasted | -0.023 |
Token | was |
Token ID | 373 |
Feature activation | +2.97e-02 |
Pos logit contributions | |
eal | +0.028 |
arre | +0.027 |
irk | +0.026 |
umpy | +0.026 |
ags | +0.025 |
Neg logit contributions | |
cardboard | -0.012 |
lf | -0.012 |
ts | -0.010 |
camps | -0.010 |
blasted | -0.009 |
Token | not |
Token ID | 407 |
Feature activation | -3.43e-02 |
Pos logit contributions | |
cardboard | +0.040 |
lf | +0.039 |
ts | +0.035 |
camps | +0.034 |
blasted | +0.032 |
Neg logit contributions | |
eal | -0.061 |
arre | -0.059 |
irk | -0.056 |
umpy | -0.056 |
ags | -0.053 |
Token | into |
Token ID | 656 |
Feature activation | +4.14e-02 |
Pos logit contributions | |
cardboard | +0.029 |
lf | +0.028 |
ts | +0.026 |
camps | +0.025 |
trek | +0.025 |
Neg logit contributions | |
eal | -0.035 |
arre | -0.035 |
umpy | -0.032 |
irk | -0.032 |
ags | -0.030 |
Token | many |
Token ID | 867 |
Feature activation | +8.83e-02 |
Pos logit contributions | |
trek | +0.056 |
lf | +0.055 |
camps | +0.055 |
blasted | +0.053 |
cardboard | +0.053 |
Neg logit contributions | |
arre | -0.077 |
eal | -0.070 |
umpy | -0.069 |
irk | -0.067 |
mad | -0.066 |
Token | pieces |
Token ID | 5207 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
cardboard | +0.125 |
lf | +0.120 |
trek | +0.116 |
camps | +0.116 |
blasted | +0.115 |
Neg logit contributions | |
arre | -0.162 |
eal | -0.160 |
umpy | -0.151 |
irk | -0.144 |
vious | -0.141 |
Token | and |
Token ID | 290 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
cardboard | +0.033 |
lf | +0.033 |
ts | +0.031 |
camps | +0.030 |
trek | +0.029 |
Neg logit contributions | |
eal | -0.044 |
arre | -0.044 |
umpy | -0.041 |
irk | -0.040 |
vious | -0.039 |
Token | the |
Token ID | 262 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
arre | +0.025 |
eal | +0.025 |
back | +0.023 |
disappointed | +0.023 |
mad | +0.023 |
Neg logit contributions | |
camps | -0.024 |
trek | -0.023 |
lf | -0.023 |
cardboard | -0.023 |
ts | -0.021 |
Token | panc |
Token ID | 22154 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
eal | +0.081 |
arre | +0.079 |
irk | +0.075 |
umpy | +0.075 |
agree | +0.071 |
Neg logit contributions | |
cardboard | -0.052 |
lf | -0.050 |
ts | -0.047 |
camps | -0.043 |
blasted | -0.041 |
Token | ake |
Token ID | 539 |
Feature activation | -6.10e-02 |
Pos logit contributions | |
eal | +0.257 |
arre | +0.237 |
irk | +0.218 |
umpy | +0.216 |
vious | +0.199 |
Neg logit contributions | |
cardboard | -0.461 |
lf | -0.459 |
ts | -0.431 |
camps | -0.416 |
blasted | -0.410 |
Token | batter |
Token ID | 4984 |
Feature activation | -4.15e-02 |
Pos logit contributions | |
eal | +0.138 |
arre | +0.134 |
irk | +0.128 |
umpy | +0.127 |
ags | +0.123 |
Neg logit contributions | |
cardboard | -0.068 |
lf | -0.068 |
ts | -0.061 |
camps | -0.055 |
blasted | -0.054 |
Token | spilled |
Token ID | 31896 |
Feature activation | +3.69e-02 |
Pos logit contributions | |
eal | +0.083 |
arre | +0.081 |
umpy | +0.074 |
back | +0.073 |
irk | +0.072 |
Neg logit contributions | |
cardboard | -0.053 |
lf | -0.053 |
camps | -0.050 |
ts | -0.048 |
trek | -0.047 |
Token | all |
Token ID | 477 |
Feature activation | -7.75e-02 |
Pos logit contributions | |
cardboard | +0.051 |
lf | +0.050 |
ts | +0.046 |
camps | +0.044 |
blasted | +0.042 |
Neg logit contributions | |
eal | -0.074 |
arre | -0.071 |
umpy | -0.067 |
irk | -0.067 |
ags | -0.064 |
Token | over |
Token ID | 625 |
Feature activation | -9.91e-02 |
Pos logit contributions | |
eal | +0.123 |
arre | +0.116 |
irk | +0.115 |
ealing | +0.110 |
umpy | +0.106 |
Neg logit contributions | |
cardboard | -0.147 |
lf | -0.132 |
ts | -0.126 |
blasted | -0.122 |
called | -0.122 |
Token | don |
Token ID | 836 |
Feature activation | -8.51e-02 |
Pos logit contributions | |
eal | +0.018 |
arre | +0.018 |
umpy | +0.016 |
irk | +0.016 |
ealing | +0.015 |
Neg logit contributions | |
cardboard | -0.014 |
lf | -0.013 |
ts | -0.012 |
camps | -0.012 |
trek | -0.012 |
Token | 't |
Token ID | 470 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
arre | +0.131 |
eal | +0.127 |
umpy | +0.119 |
agree | +0.113 |
Fighting | +0.111 |
Neg logit contributions | |
cardboard | -0.150 |
lf | -0.141 |
ts | -0.141 |
camps | -0.139 |
trek | -0.136 |
Token | you |
Token ID | 345 |
Feature activation | -9.90e-02 |
Pos logit contributions | |
eal | +0.020 |
arre | +0.019 |
umpy | +0.017 |
irk | +0.017 |
vious | +0.017 |
Neg logit contributions | |
cardboard | -0.021 |
lf | -0.020 |
ts | -0.020 |
camps | -0.019 |
trek | -0.019 |
Token | enter |
Token ID | 3802 |
Feature activation | +5.87e-02 |
Pos logit contributions | |
eal | +0.182 |
arre | +0.177 |
umpy | +0.160 |
awk | +0.156 |
vious | +0.155 |
Neg logit contributions | |
lf | -0.152 |
cardboard | -0.148 |
camps | -0.137 |
ts | -0.137 |
trek | -0.132 |
Token | a |
Token ID | 257 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
lf | +0.095 |
cardboard | +0.093 |
ts | +0.088 |
blasted | +0.087 |
trek | +0.086 |
Neg logit contributions | |
eal | -0.094 |
arre | -0.092 |
ags | -0.086 |
back | -0.085 |
umpy | -0.083 |
Token | panc |
Token ID | 22154 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
cardboard | +0.220 |
lf | +0.217 |
ts | +0.205 |
camps | +0.189 |
blasted | +0.187 |
Neg logit contributions | |
eal | -0.300 |
arre | -0.292 |
umpy | -0.272 |
irk | -0.270 |
vious | -0.256 |
Token | ake |
Token ID | 539 |
Feature activation | -6.03e-02 |
Pos logit contributions | |
eal | +0.252 |
arre | +0.236 |
irk | +0.214 |
umpy | +0.214 |
ags | +0.194 |
Neg logit contributions | |
cardboard | -0.464 |
lf | -0.461 |
ts | -0.433 |
camps | -0.423 |
blasted | -0.414 |
Token | cooking |
Token ID | 10801 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
eal | +0.127 |
arre | +0.123 |
irk | +0.117 |
umpy | +0.116 |
ags | +0.111 |
Neg logit contributions | |
cardboard | -0.077 |
lf | -0.077 |
ts | -0.068 |
camps | -0.066 |
blasted | -0.063 |
Token | contest |
Token ID | 8414 |
Feature activation | +2.68e-02 |
Pos logit contributions | |
eal | +0.031 |
arre | +0.030 |
irk | +0.029 |
umpy | +0.029 |
Fighting | +0.027 |
Neg logit contributions | |
cardboard | -0.019 |
lf | -0.019 |
ts | -0.017 |
camps | -0.016 |
pretend | -0.016 |
Token | ?" |
Token ID | 1701 |
Feature activation | -2.54e-02 |
Pos logit contributions | |
cardboard | +0.038 |
lf | +0.037 |
ts | +0.034 |
camps | +0.032 |
trek | +0.032 |
Neg logit contributions | |
eal | -0.053 |
arre | -0.051 |
umpy | -0.048 |
irk | -0.047 |
vious | -0.045 |
Token | Tim |
Token ID | 5045 |
Feature activation | +3.88e-02 |
Pos logit contributions | |
eal | +0.052 |
arre | +0.050 |
umpy | +0.047 |
irk | +0.047 |
ags | +0.045 |
Neg logit contributions | |
cardboard | -0.034 |
lf | -0.034 |
ts | -0.031 |
camps | -0.029 |
blasted | -0.028 |
Token | not |
Token ID | 407 |
Feature activation | -3.35e-02 |
Pos logit contributions | |
eal | +0.030 |
arre | +0.029 |
umpy | +0.027 |
irk | +0.027 |
ags | +0.026 |
Neg logit contributions | |
cardboard | -0.021 |
lf | -0.020 |
ts | -0.019 |
camps | -0.017 |
blasted | -0.017 |
Token | to |
Token ID | 284 |
Feature activation | -1.82e-04 |
Pos logit contributions | |
eal | +0.068 |
arre | +0.065 |
irk | +0.063 |
umpy | +0.061 |
ags | +0.061 |
Neg logit contributions | |
cardboard | -0.046 |
lf | -0.044 |
ts | -0.041 |
pretend | -0.040 |
shovel | -0.039 |
Token | stop |
Token ID | 2245 |
Feature activation | -9.15e-02 |
Pos logit contributions | |
eal | +0.000 |
arre | +0.000 |
irk | +0.000 |
umpy | +0.000 |
ags | +0.000 |
Neg logit contributions | |
cardboard | -0.000 |
lf | -0.000 |
ts | -0.000 |
camps | -0.000 |
pretend | -0.000 |
Token | stirring |
Token ID | 26547 |
Feature activation | -3.80e-02 |
Pos logit contributions | |
eal | +0.206 |
arre | +0.195 |
umpy | +0.190 |
irk | +0.189 |
agree | +0.179 |
Neg logit contributions | |
cardboard | -0.112 |
lf | -0.108 |
ts | -0.102 |
pretend | -0.093 |
pretended | -0.089 |
Token | the |
Token ID | 262 |
Feature activation | -3.84e-02 |
Pos logit contributions | |
eal | +0.089 |
arre | +0.086 |
irk | +0.083 |
umpy | +0.083 |
vious | +0.078 |
Neg logit contributions | |
cardboard | -0.045 |
lf | -0.043 |
ts | -0.041 |
pretend | -0.037 |
camps | -0.035 |
Token | panc |
Token ID | 22154 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
eal | +0.079 |
arre | +0.078 |
irk | +0.075 |
umpy | +0.074 |
agree | +0.070 |
Neg logit contributions | |
cardboard | -0.052 |
lf | -0.051 |
ts | -0.047 |
pretend | -0.043 |
camps | -0.043 |
Token | ake |
Token ID | 539 |
Feature activation | -5.95e-02 |
Pos logit contributions | |
eal | +0.294 |
irk | +0.267 |
arre | +0.258 |
umpy | +0.253 |
vious | +0.239 |
Neg logit contributions | |
lf | -0.425 |
cardboard | -0.412 |
ts | -0.384 |
camps | -0.372 |
blasted | -0.365 |
Token | batter |
Token ID | 4984 |
Feature activation | -4.04e-02 |
Pos logit contributions | |
eal | +0.131 |
arre | +0.127 |
irk | +0.124 |
umpy | +0.120 |
ags | +0.115 |
Neg logit contributions | |
lf | -0.074 |
cardboard | -0.072 |
ts | -0.065 |
camps | -0.059 |
pretend | -0.059 |
Token | . |
Token ID | 13 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
eal | +0.083 |
arre | +0.080 |
irk | +0.076 |
umpy | +0.075 |
ags | +0.071 |
Neg logit contributions | |
cardboard | -0.055 |
lf | -0.054 |
ts | -0.050 |
camps | -0.047 |
pretend | -0.045 |
Token |
|
Token ID | 198 |
Feature activation | -8.44e-03 |
Pos logit contributions | |
eal | +0.054 |
umpy | +0.052 |
irk | +0.051 |
arre | +0.051 |
vious | +0.051 |
Neg logit contributions | |
cardboard | -0.027 |
lf | -0.026 |
called | -0.023 |
pretend | -0.022 |
pretended | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | -8.00e-03 |
Pos logit contributions | |
eal | +0.015 |
arre | +0.014 |
umpy | +0.013 |
irk | +0.013 |
ags | +0.013 |
Neg logit contributions | |
cardboard | -0.014 |
lf | -0.013 |
ts | -0.012 |
camps | -0.012 |
blasted | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
cardboard | +0.105 |
lf | +0.103 |
ts | +0.096 |
camps | +0.091 |
trek | +0.090 |
Neg logit contributions | |
eal | -0.118 |
arre | -0.114 |
umpy | -0.107 |
irk | -0.105 |
vious | -0.100 |
Token | supply |
Token ID | 5127 |
Feature activation | -8.67e-03 |
Pos logit contributions | |
eal | +0.025 |
arre | +0.024 |
umpy | +0.022 |
irk | +0.022 |
ags | +0.021 |
Neg logit contributions | |
cardboard | -0.017 |
lf | -0.016 |
ts | -0.015 |
camps | -0.014 |
blasted | -0.014 |
Token | us |
Token ID | 514 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
eal | +0.017 |
arre | +0.017 |
umpy | +0.016 |
irk | +0.016 |
agree | +0.015 |
Neg logit contributions | |
cardboard | -0.012 |
lf | -0.011 |
ts | -0.011 |
camps | -0.010 |
pretend | -0.010 |
Token | with |
Token ID | 351 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
eal | +0.264 |
arre | +0.258 |
irk | +0.255 |
umpy | +0.246 |
agree | +0.235 |
Neg logit contributions | |
cardboard | -0.140 |
lf | -0.133 |
ts | -0.130 |
pretend | -0.116 |
camps | -0.115 |
Token | more |
Token ID | 517 |
Feature activation | -6.22e-02 |
Pos logit contributions | |
eal | +0.045 |
arre | +0.043 |
irk | +0.041 |
umpy | +0.041 |
ags | +0.038 |
Neg logit contributions | |
cardboard | -0.031 |
lf | -0.030 |
ts | -0.028 |
camps | -0.026 |
blasted | -0.025 |
Token | panc |
Token ID | 22154 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
eal | +0.118 |
arre | +0.115 |
irk | +0.108 |
umpy | +0.108 |
agree | +0.102 |
Neg logit contributions | |
cardboard | -0.092 |
lf | -0.090 |
ts | -0.084 |
camps | -0.080 |
pretend | -0.076 |
Token | ake |
Token ID | 539 |
Feature activation | -6.00e-02 |
Pos logit contributions | |
eal | +0.256 |
arre | +0.238 |
irk | +0.221 |
umpy | +0.218 |
vious | +0.200 |
Neg logit contributions | |
cardboard | -0.452 |
lf | -0.451 |
ts | -0.424 |
camps | -0.411 |
trek | -0.402 |
Token | mix |
Token ID | 5022 |
Feature activation | -3.31e-02 |
Pos logit contributions | |
eal | +0.122 |
arre | +0.119 |
irk | +0.118 |
umpy | +0.112 |
Fighting | +0.109 |
Neg logit contributions | |
lf | -0.081 |
cardboard | -0.080 |
ts | -0.074 |
camps | -0.069 |
pretend | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
eal | +0.067 |
arre | +0.066 |
irk | +0.063 |
umpy | +0.062 |
agree | +0.059 |
Neg logit contributions | |
cardboard | -0.044 |
lf | -0.043 |
ts | -0.041 |
camps | -0.038 |
pretend | -0.036 |
Token | But |
Token ID | 887 |
Feature activation | +1.34e-03 |
Pos logit contributions | |
eal | +0.036 |
arre | +0.034 |
irk | +0.033 |
umpy | +0.032 |
ags | +0.031 |
Neg logit contributions | |
cardboard | -0.036 |
lf | -0.034 |
camps | -0.032 |
ts | -0.032 |
pretend | -0.031 |
Token | it |
Token ID | 340 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
cardboard | +0.002 |
lf | +0.002 |
ts | +0.002 |
camps | +0.002 |
trek | +0.002 |
Neg logit contributions | |
eal | -0.002 |
arre | -0.002 |
umpy | -0.002 |
irk | -0.002 |
back | -0.002 |
Token | where |
Token ID | 810 |
Feature activation | +4.58e-02 |
Pos logit contributions | |
eal | +0.063 |
arre | +0.060 |
umpy | +0.057 |
irk | +0.057 |
ags | +0.054 |
Neg logit contributions | |
cardboard | -0.044 |
lf | -0.043 |
ts | -0.039 |
camps | -0.038 |
trek | -0.036 |
Token | they |
Token ID | 484 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
cardboard | +0.043 |
lf | +0.040 |
pretend | +0.035 |
ts | +0.034 |
called | +0.032 |
Neg logit contributions | |
eal | -0.118 |
arre | -0.115 |
ags | -0.111 |
umpy | -0.110 |
irk | -0.110 |
Token | could |
Token ID | 714 |
Feature activation | +9.77e-02 |
Pos logit contributions | |
eal | +0.028 |
arre | +0.028 |
umpy | +0.026 |
back | +0.026 |
vious | +0.026 |
Neg logit contributions | |
lf | -0.016 |
cardboard | -0.016 |
camps | -0.014 |
ts | -0.013 |
trek | -0.013 |
Token | get |
Token ID | 651 |
Feature activation | -5.06e-02 |
Pos logit contributions | |
cardboard | +0.143 |
lf | +0.141 |
ts | +0.134 |
camps | +0.127 |
trek | +0.124 |
Neg logit contributions | |
eal | -0.182 |
arre | -0.178 |
umpy | -0.166 |
irk | -0.160 |
ags | -0.158 |
Token | more |
Token ID | 517 |
Feature activation | -6.18e-02 |
Pos logit contributions | |
eal | +0.109 |
arre | +0.107 |
irk | +0.105 |
umpy | +0.104 |
agree | +0.097 |
Neg logit contributions | |
cardboard | -0.065 |
lf | -0.058 |
ts | -0.057 |
pretend | -0.054 |
camps | -0.053 |
Token | panc |
Token ID | 22154 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
arre | +0.110 |
eal | +0.108 |
irk | +0.105 |
umpy | +0.103 |
agree | +0.101 |
Neg logit contributions | |
cardboard | -0.097 |
lf | -0.093 |
ts | -0.088 |
pretend | -0.086 |
camps | -0.085 |
Token | ake |
Token ID | 539 |
Feature activation | -5.94e-02 |
Pos logit contributions | |
eal | +0.264 |
arre | +0.249 |
irk | +0.230 |
umpy | +0.229 |
vious | +0.208 |
Neg logit contributions | |
cardboard | -0.443 |
lf | -0.441 |
ts | -0.416 |
camps | -0.402 |
blasted | -0.393 |
Token | mix |
Token ID | 5022 |
Feature activation | -3.32e-02 |
Pos logit contributions | |
eal | +0.137 |
irk | +0.137 |
arre | +0.133 |
umpy | +0.131 |
Fighting | +0.128 |
Neg logit contributions | |
lf | -0.064 |
cardboard | -0.061 |
ts | -0.053 |
pretend | -0.051 |
camps | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
eal | +0.076 |
arre | +0.075 |
irk | +0.073 |
umpy | +0.072 |
Fighting | +0.069 |
Neg logit contributions | |
cardboard | -0.036 |
lf | -0.033 |
ts | -0.032 |
pretend | -0.029 |
camps | -0.028 |
Token | |
Token ID | 220 |
Feature activation | -9.38e-02 |
Pos logit contributions | |
umpy | +0.040 |
eal | +0.040 |
vious | +0.039 |
arre | +0.038 |
irk | +0.038 |
Neg logit contributions | |
cardboard | -0.024 |
lf | -0.023 |
pretend | -0.021 |
called | -0.020 |
pretended | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -3.57e-03 |
Pos logit contributions | |
eal | +0.192 |
arre | +0.182 |
irk | +0.171 |
vious | +0.170 |
umpy | +0.169 |
Neg logit contributions | |
cardboard | -0.123 |
lf | -0.121 |
ts | -0.108 |
camps | -0.104 |
trek | -0.099 |
Token | pointed |
Token ID | 6235 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
cardboard | +0.116 |
lf | +0.109 |
ts | +0.105 |
camps | +0.097 |
blasted | +0.091 |
Neg logit contributions | |
eal | -0.186 |
arre | -0.175 |
irk | -0.161 |
umpy | -0.159 |
back | -0.154 |
Token | to |
Token ID | 284 |
Feature activation | -3.02e-03 |
Pos logit contributions | |
cardboard | +0.022 |
pretend | +0.020 |
shovel | +0.020 |
swords | +0.019 |
pretended | +0.019 |
Neg logit contributions | |
arre | -0.039 |
umpy | -0.036 |
eal | -0.034 |
agree | -0.034 |
ivering | -0.034 |
Token | the |
Token ID | 262 |
Feature activation | -3.51e-02 |
Pos logit contributions | |
eal | +0.006 |
arre | +0.005 |
irk | +0.005 |
back | +0.005 |
umpy | +0.005 |
Neg logit contributions | |
lf | -0.004 |
cardboard | -0.004 |
ts | -0.004 |
camps | -0.004 |
blasted | -0.004 |
Token | blue |
Token ID | 4171 |
Feature activation | +2.49e-02 |
Pos logit contributions | |
eal | +0.071 |
arre | +0.068 |
umpy | +0.065 |
irk | +0.064 |
vious | +0.061 |
Neg logit contributions | |
cardboard | -0.048 |
lf | -0.047 |
ts | -0.043 |
camps | -0.041 |
blasted | -0.040 |
Token | berry |
Token ID | 8396 |
Feature activation | +4.51e-02 |
Pos logit contributions | |
cardboard | +0.039 |
lf | +0.038 |
ts | +0.036 |
camps | +0.034 |
blasted | +0.034 |
Neg logit contributions | |
eal | -0.045 |
arre | -0.043 |
irk | -0.040 |
umpy | -0.040 |
vious | -0.038 |
Token | panc |
Token ID | 22154 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
cardboard | +0.058 |
lf | +0.057 |
ts | +0.052 |
camps | +0.049 |
blasted | +0.048 |
Neg logit contributions | |
eal | -0.094 |
arre | -0.091 |
umpy | -0.086 |
irk | -0.086 |
ags | -0.082 |
Token | ake |
Token ID | 539 |
Feature activation | -6.03e-02 |
Pos logit contributions | |
eal | +0.264 |
arre | +0.251 |
umpy | +0.230 |
irk | +0.229 |
vious | +0.211 |
Neg logit contributions | |
lf | -0.441 |
cardboard | -0.440 |
ts | -0.409 |
camps | -0.397 |
blasted | -0.390 |
Token | . |
Token ID | 13 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
eal | +0.129 |
arre | +0.127 |
irk | +0.122 |
umpy | +0.122 |
vious | +0.115 |
Neg logit contributions | |
lf | -0.079 |
cardboard | -0.074 |
ts | -0.068 |
camps | -0.061 |
pretend | -0.061 |
Token |
|
Token ID | 198 |
Feature activation | -7.72e-03 |
Pos logit contributions | |
eal | +0.033 |
arre | +0.032 |
irk | +0.030 |
umpy | +0.030 |
vious | +0.029 |
Neg logit contributions | |
lf | -0.032 |
cardboard | -0.032 |
camps | -0.028 |
ts | -0.027 |
blasted | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | -7.39e-03 |
Pos logit contributions | |
eal | +0.016 |
arre | +0.015 |
irk | +0.014 |
vious | +0.014 |
awk | +0.014 |
Neg logit contributions | |
lf | -0.010 |
cardboard | -0.010 |
camps | -0.009 |
trek | -0.008 |
blasted | -0.008 |
Token | Mom |
Token ID | 29252 |
Feature activation | +2.33e-02 |
Pos logit contributions | |
eal | +0.014 |
arre | +0.013 |
irk | +0.013 |
umpy | +0.012 |
ags | +0.012 |
Neg logit contributions | |
lf | -0.012 |
cardboard | -0.011 |
ts | -0.010 |
camps | -0.010 |
blasted | -0.010 |
Token | that |
Token ID | 326 |
Feature activation | -3.52e-03 |
Pos logit contributions | |
eal | +0.124 |
arre | +0.120 |
irk | +0.113 |
umpy | +0.113 |
ags | +0.109 |
Neg logit contributions | |
cardboard | -0.084 |
lf | -0.083 |
ts | -0.075 |
camps | -0.070 |
blasted | -0.068 |
Token | they |
Token ID | 484 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
eal | +0.008 |
arre | +0.008 |
irk | +0.007 |
umpy | +0.007 |
ags | +0.007 |
Neg logit contributions | |
cardboard | -0.004 |
lf | -0.004 |
ts | -0.004 |
camps | -0.004 |
pretend | -0.003 |
Token | ran |
Token ID | 4966 |
Feature activation | +4.42e-02 |
Pos logit contributions | |
eal | +0.030 |
arre | +0.029 |
umpy | +0.028 |
irk | +0.027 |
vious | +0.026 |
Neg logit contributions | |
cardboard | -0.011 |
lf | -0.011 |
ts | -0.010 |
camps | -0.009 |
trek | -0.008 |
Token | out |
Token ID | 503 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
cardboard | +0.069 |
lf | +0.062 |
camps | +0.061 |
pretend | +0.061 |
ts | +0.059 |
Neg logit contributions | |
eal | -0.082 |
arre | -0.077 |
irk | -0.077 |
umpy | -0.076 |
ealing | -0.073 |
Token | of |
Token ID | 286 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
eal | +0.022 |
arre | +0.021 |
irk | +0.020 |
umpy | +0.020 |
ealing | +0.019 |
Neg logit contributions | |
cardboard | -0.017 |
lf | -0.016 |
ts | -0.015 |
camps | -0.014 |
trek | -0.014 |
Token | panc |
Token ID | 22154 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
cardboard | +0.036 |
lf | +0.035 |
ts | +0.033 |
camps | +0.031 |
blasted | +0.031 |
Neg logit contributions | |
eal | -0.046 |
arre | -0.045 |
irk | -0.042 |
umpy | -0.041 |
ags | -0.040 |
Token | ake |
Token ID | 539 |
Feature activation | -5.86e-02 |
Pos logit contributions | |
eal | +0.250 |
arre | +0.234 |
umpy | +0.216 |
irk | +0.215 |
vious | +0.194 |
Neg logit contributions | |
cardboard | -0.457 |
lf | -0.452 |
ts | -0.425 |
camps | -0.416 |
trek | -0.406 |
Token | mix |
Token ID | 5022 |
Feature activation | -3.27e-02 |
Pos logit contributions | |
eal | +0.109 |
arre | +0.106 |
umpy | +0.098 |
irk | +0.097 |
ags | +0.093 |
Neg logit contributions | |
cardboard | -0.088 |
lf | -0.086 |
ts | -0.080 |
camps | -0.077 |
blasted | -0.075 |
Token | . |
Token ID | 13 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
eal | +0.070 |
arre | +0.068 |
irk | +0.065 |
umpy | +0.065 |
agree | +0.061 |
Neg logit contributions | |
cardboard | -0.041 |
lf | -0.039 |
ts | -0.038 |
camps | -0.034 |
pretend | -0.034 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
eal | +0.035 |
arre | +0.034 |
irk | +0.033 |
umpy | +0.033 |
awk | +0.032 |
Neg logit contributions | |
cardboard | -0.023 |
lf | -0.021 |
pretend | -0.019 |
pretended | -0.018 |
camps | -0.018 |
Token | was |
Token ID | 373 |
Feature activation | +3.65e-02 |
Pos logit contributions | |
lf | +0.012 |
cardboard | +0.011 |
ts | +0.010 |
pretended | +0.010 |
pretend | +0.009 |
Neg logit contributions | |
eal | -0.028 |
arre | -0.027 |
irk | -0.027 |
ags | -0.026 |
umpy | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
eal | +0.044 |
arre | +0.043 |
irk | +0.042 |
umpy | +0.041 |
ags | +0.040 |
Neg logit contributions | |
cardboard | -0.023 |
lf | -0.022 |
pretended | -0.020 |
pretend | -0.019 |
ts | -0.019 |
Token | continued |
Token ID | 3767 |
Feature activation | +5.41e-02 |
Pos logit contributions | |
eal | +0.028 |
arre | +0.027 |
irk | +0.025 |
umpy | +0.024 |
back | +0.024 |
Neg logit contributions | |
cardboard | -0.019 |
lf | -0.018 |
ts | -0.017 |
camps | -0.017 |
trek | -0.015 |
Token | to |
Token ID | 284 |
Feature activation | +2.39e-03 |
Pos logit contributions | |
cardboard | +0.086 |
lf | +0.084 |
ts | +0.076 |
pretend | +0.075 |
camps | +0.074 |
Neg logit contributions | |
eal | -0.103 |
arre | -0.098 |
umpy | -0.098 |
irk | -0.096 |
vious | -0.094 |
Token | stir |
Token ID | 11240 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
lf | +0.003 |
cardboard | +0.003 |
ts | +0.003 |
camps | +0.003 |
trek | +0.003 |
Neg logit contributions | |
eal | -0.005 |
arre | -0.005 |
irk | -0.004 |
umpy | -0.004 |
ags | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | -3.66e-02 |
Pos logit contributions | |
eal | +0.066 |
arre | +0.064 |
umpy | +0.062 |
irk | +0.062 |
agree | +0.058 |
Neg logit contributions | |
cardboard | -0.037 |
lf | -0.036 |
ts | -0.035 |
pretend | -0.031 |
camps | -0.030 |
Token | panc |
Token ID | 22154 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
eal | +0.073 |
arre | +0.070 |
irk | +0.068 |
umpy | +0.067 |
vious | +0.063 |
Neg logit contributions | |
cardboard | -0.054 |
lf | -0.052 |
ts | -0.046 |
pretend | -0.044 |
camps | -0.043 |
Token | ake |
Token ID | 539 |
Feature activation | -5.79e-02 |
Pos logit contributions | |
eal | +0.262 |
arre | +0.240 |
irk | +0.229 |
umpy | +0.224 |
vious | +0.209 |
Neg logit contributions | |
lf | -0.443 |
cardboard | -0.439 |
ts | -0.409 |
camps | -0.397 |
blasted | -0.389 |
Token | batter |
Token ID | 4984 |
Feature activation | -3.81e-02 |
Pos logit contributions | |
eal | +0.121 |
arre | +0.117 |
irk | +0.111 |
umpy | +0.110 |
ags | +0.105 |
Neg logit contributions | |
cardboard | -0.077 |
lf | -0.076 |
ts | -0.068 |
camps | -0.064 |
blasted | -0.062 |
Token | . |
Token ID | 13 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
eal | +0.073 |
arre | +0.071 |
umpy | +0.066 |
irk | +0.066 |
ags | +0.064 |
Neg logit contributions | |
cardboard | -0.054 |
lf | -0.053 |
ts | -0.049 |
camps | -0.047 |
trek | -0.046 |
Token | |
Token ID | 220 |
Feature activation | -9.33e-02 |
Pos logit contributions | |
eal | +0.039 |
arre | +0.037 |
irk | +0.036 |
umpy | +0.036 |
vious | +0.035 |
Neg logit contributions | |
cardboard | -0.024 |
lf | -0.023 |
pretend | -0.021 |
camps | -0.020 |
pretended | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -3.91e-03 |
Pos logit contributions | |
eal | +0.187 |
arre | +0.179 |
irk | +0.168 |
vious | +0.165 |
umpy | +0.165 |
Neg logit contributions | |
cardboard | -0.126 |
lf | -0.124 |
ts | -0.111 |
camps | -0.107 |
trek | -0.103 |
Token | . |
Token ID | 13 |
Feature activation | -3.97e-02 |
Pos logit contributions | |
eal | +0.274 |
irk | +0.255 |
arre | +0.250 |
ags | +0.239 |
umpy | +0.233 |
Neg logit contributions | |
cardboard | -0.120 |
lf | -0.112 |
pretend | -0.105 |
ts | -0.102 |
called | -0.100 |
Token | I |
Token ID | 314 |
Feature activation | -4.29e-03 |
Pos logit contributions | |
eal | +0.063 |
arre | +0.061 |
irk | +0.055 |
umpy | +0.054 |
awk | +0.051 |
Neg logit contributions | |
cardboard | -0.070 |
lf | -0.066 |
pretend | -0.063 |
ts | -0.061 |
named | -0.061 |
Token | was |
Token ID | 373 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
eal | +0.009 |
arre | +0.008 |
ags | +0.008 |
irk | +0.008 |
umpy | +0.008 |
Neg logit contributions | |
cardboard | -0.005 |
pretend | -0.005 |
lf | -0.005 |
ts | -0.005 |
pretended | -0.005 |
Token | tired |
Token ID | 10032 |
Feature activation | -4.33e-02 |
Pos logit contributions | |
cardboard | +0.033 |
ts | +0.031 |
lf | +0.030 |
pretend | +0.029 |
called | +0.029 |
Neg logit contributions | |
eal | -0.044 |
irk | -0.042 |
arre | -0.041 |
umpy | -0.039 |
agree | -0.038 |
Token | and |
Token ID | 290 |
Feature activation | -2.71e-02 |
Pos logit contributions | |
eal | +0.096 |
arre | +0.092 |
irk | +0.085 |
ags | +0.083 |
ealing | +0.083 |
Neg logit contributions | |
cardboard | -0.052 |
lf | -0.050 |
ts | -0.044 |
camps | -0.044 |
pretend | -0.041 |
Token | crank |
Token ID | 30425 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
eal | +0.046 |
arre | +0.045 |
irk | +0.042 |
ags | +0.041 |
umpy | +0.040 |
Neg logit contributions | |
cardboard | -0.045 |
lf | -0.044 |
ts | -0.043 |
pretend | -0.040 |
pretended | -0.040 |
Token | y |
Token ID | 88 |
Feature activation | +3.62e-02 |
Pos logit contributions | |
eal | +0.360 |
arre | +0.335 |
ags | +0.329 |
irk | +0.320 |
Aud | +0.317 |
Neg logit contributions | |
called | -0.208 |
ts | -0.198 |
pretended | -0.196 |
lf | -0.196 |
pretend | -0.174 |
Token | . |
Token ID | 13 |
Feature activation | -4.04e-02 |
Pos logit contributions | |
cardboard | +0.040 |
lf | +0.037 |
ts | +0.035 |
pretend | +0.034 |
camps | +0.032 |
Neg logit contributions | |
eal | -0.083 |
arre | -0.079 |
irk | -0.077 |
ags | -0.073 |
agree | -0.072 |
Token | I |
Token ID | 314 |
Feature activation | -8.68e-04 |
Pos logit contributions | |
eal | +0.076 |
arre | +0.074 |
irk | +0.067 |
umpy | +0.065 |
ags | +0.064 |
Neg logit contributions | |
cardboard | -0.061 |
lf | -0.058 |
pretend | -0.052 |
camps | -0.052 |
ts | -0.051 |
Token | appreciate |
Token ID | 9144 |
Feature activation | -9.55e-03 |
Pos logit contributions | |
eal | +0.002 |
arre | +0.002 |
ags | +0.002 |
irk | +0.002 |
umpy | +0.002 |
Neg logit contributions | |
cardboard | -0.001 |
pretended | -0.001 |
pretend | -0.001 |
ts | -0.001 |
lf | -0.001 |
Token | your |
Token ID | 534 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
eal | +0.019 |
arre | +0.018 |
umpy | +0.017 |
irk | +0.017 |
vious | +0.015 |
Neg logit contributions | |
cardboard | -0.014 |
lf | -0.014 |
ts | -0.013 |
camps | -0.012 |
pretend | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -3.60e-02 |
Pos logit contributions | |
eal | +0.259 |
arre | +0.251 |
umpy | +0.235 |
irk | +0.235 |
ags | +0.224 |
Neg logit contributions | |
cardboard | -0.176 |
lf | -0.173 |
ts | -0.159 |
camps | -0.151 |
blasted | -0.146 |
Token | They |
Token ID | 1119 |
Feature activation | +5.56e-02 |
Pos logit contributions | |
eal | +0.081 |
arre | +0.073 |
awk | +0.070 |
vious | +0.069 |
irk | +0.068 |
Neg logit contributions | |
cardboard | -0.043 |
camps | -0.036 |
pretend | -0.034 |
shovel | -0.034 |
called | -0.033 |
Token | are |
Token ID | 389 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
cardboard | +0.083 |
lf | +0.080 |
ts | +0.077 |
camps | +0.071 |
pretend | +0.071 |
Neg logit contributions | |
eal | -0.104 |
arre | -0.101 |
umpy | -0.095 |
irk | -0.095 |
ags | -0.090 |
Token | tired |
Token ID | 10032 |
Feature activation | -4.16e-02 |
Pos logit contributions | |
eal | +0.035 |
arre | +0.033 |
irk | +0.032 |
umpy | +0.030 |
aback | +0.029 |
Neg logit contributions | |
cardboard | -0.024 |
ts | -0.022 |
lf | -0.022 |
called | -0.022 |
pretend | -0.021 |
Token | and |
Token ID | 290 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
eal | +0.080 |
arre | +0.078 |
umpy | +0.074 |
irk | +0.074 |
ags | +0.070 |
Neg logit contributions | |
cardboard | -0.059 |
lf | -0.058 |
ts | -0.054 |
camps | -0.051 |
blasted | -0.050 |
Token | crank |
Token ID | 30425 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
arre | +0.053 |
eal | +0.053 |
irk | +0.048 |
ags | +0.045 |
itter | +0.045 |
Neg logit contributions | |
ts | -0.053 |
cardboard | -0.051 |
pretend | -0.050 |
pretended | -0.049 |
called | -0.048 |
Token | y |
Token ID | 88 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
eal | +0.385 |
joy | +0.334 |
arre | +0.332 |
disapproval | +0.332 |
aback | +0.332 |
Neg logit contributions | |
called | -0.194 |
ts | -0.192 |
lf | -0.182 |
pretended | -0.177 |
pretend | -0.172 |
Token | . |
Token ID | 13 |
Feature activation | -3.58e-02 |
Pos logit contributions | |
cardboard | +0.037 |
lf | +0.034 |
ts | +0.033 |
pretend | +0.031 |
pretended | +0.029 |
Neg logit contributions | |
eal | -0.073 |
arre | -0.069 |
irk | -0.066 |
umpy | -0.064 |
ags | -0.063 |
Token | They |
Token ID | 1119 |
Feature activation | +5.56e-02 |
Pos logit contributions | |
eal | +0.087 |
arre | +0.079 |
awk | +0.076 |
vious | +0.076 |
irk | +0.075 |
Neg logit contributions | |
cardboard | -0.036 |
pretend | -0.030 |
camps | -0.029 |
shovel | -0.028 |
lf | -0.026 |
Token | say |
Token ID | 910 |
Feature activation | -5.60e-02 |
Pos logit contributions | |
cardboard | +0.079 |
lf | +0.074 |
ts | +0.073 |
pretend | +0.070 |
trek | +0.067 |
Neg logit contributions | |
eal | -0.108 |
arre | -0.107 |
umpy | -0.100 |
irk | -0.099 |
ags | -0.094 |
Token | , |
Token ID | 11 |
Feature activation | +4.26e-02 |
Pos logit contributions | |
eal | +0.161 |
arre | +0.157 |
irk | +0.154 |
umpy | +0.154 |
vious | +0.146 |
Neg logit contributions | |
cardboard | -0.038 |
lf | -0.032 |
ts | -0.026 |
pretend | -0.026 |
shovel | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -3.60e-02 |
Pos logit contributions | |
eal | +0.153 |
arre | +0.146 |
irk | +0.142 |
agree | +0.140 |
umpy | +0.135 |
Neg logit contributions | |
cardboard | -0.060 |
lf | -0.058 |
ts | -0.054 |
pretend | -0.051 |
camps | -0.050 |
Token | You |
Token ID | 921 |
Feature activation | -5.48e-02 |
Pos logit contributions | |
eal | +0.061 |
arre | +0.061 |
irk | +0.052 |
ocked | +0.051 |
mad | +0.050 |
Neg logit contributions | |
cardboard | -0.059 |
pretend | -0.052 |
named | -0.050 |
lf | -0.049 |
shovel | -0.049 |
Token | are |
Token ID | 389 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
eal | +0.106 |
arre | +0.104 |
umpy | +0.097 |
irk | +0.095 |
vious | +0.093 |
Neg logit contributions | |
lf | -0.079 |
cardboard | -0.079 |
ts | -0.070 |
camps | -0.070 |
trek | -0.066 |
Token | tired |
Token ID | 10032 |
Feature activation | -4.14e-02 |
Pos logit contributions | |
eal | +0.035 |
arre | +0.034 |
irk | +0.032 |
umpy | +0.031 |
awk | +0.030 |
Neg logit contributions | |
cardboard | -0.025 |
lf | -0.024 |
ts | -0.024 |
pretend | -0.021 |
pretended | -0.021 |
Token | and |
Token ID | 290 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
eal | +0.088 |
arre | +0.084 |
irk | +0.079 |
umpy | +0.077 |
ags | +0.076 |
Neg logit contributions | |
cardboard | -0.053 |
lf | -0.051 |
ts | -0.047 |
camps | -0.045 |
pretend | -0.042 |
Token | crank |
Token ID | 30425 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
arre | +0.054 |
eal | +0.054 |
irk | +0.050 |
ags | +0.047 |
umpy | +0.045 |
Neg logit contributions | |
ts | -0.047 |
cardboard | -0.046 |
pretend | -0.044 |
lf | -0.042 |
called | -0.042 |
Token | y |
Token ID | 88 |
Feature activation | +3.32e-02 |
Pos logit contributions | |
eal | +0.379 |
arre | +0.330 |
irk | +0.330 |
joy | +0.328 |
ags | +0.323 |
Neg logit contributions | |
called | -0.199 |
ts | -0.188 |
lf | -0.184 |
pretend | -0.177 |
pretended | -0.176 |
Token | . |
Token ID | 13 |
Feature activation | -3.66e-02 |
Pos logit contributions | |
cardboard | +0.036 |
lf | +0.033 |
ts | +0.032 |
pretend | +0.031 |
camps | +0.028 |
Neg logit contributions | |
eal | -0.078 |
arre | -0.073 |
irk | -0.071 |
ags | -0.067 |
agree | -0.067 |
Token | You |
Token ID | 921 |
Feature activation | -5.46e-02 |
Pos logit contributions | |
eal | +0.058 |
arre | +0.056 |
irk | +0.049 |
umpy | +0.046 |
ocked | +0.046 |
Neg logit contributions | |
cardboard | -0.065 |
lf | -0.058 |
pretend | -0.057 |
ts | -0.056 |
named | -0.056 |
Token | need |
Token ID | 761 |
Feature activation | -1.71e-03 |
Pos logit contributions | |
eal | +0.109 |
arre | +0.106 |
umpy | +0.099 |
irk | +0.099 |
ags | +0.094 |
Neg logit contributions | |
cardboard | -0.076 |
lf | -0.074 |
ts | -0.068 |
camps | -0.065 |
blasted | -0.063 |
Token | to |
Token ID | 284 |
Feature activation | -8.64e-03 |
Pos logit contributions | |
eal | +0.003 |
arre | +0.003 |
umpy | +0.003 |
irk | +0.003 |
ealing | +0.002 |
Neg logit contributions | |
cardboard | -0.003 |
lf | -0.003 |
ts | -0.003 |
camps | -0.003 |
trek | -0.002 |
Token | the |
Token ID | 262 |
Feature activation | -4.69e-02 |
Pos logit contributions | |
eal | +0.038 |
arre | +0.037 |
umpy | +0.035 |
irk | +0.034 |
awk | +0.033 |
Neg logit contributions | |
cardboard | -0.018 |
lf | -0.016 |
ts | -0.015 |
pretend | -0.013 |
camps | -0.013 |
Token | bag |
Token ID | 6131 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
eal | +0.062 |
arre | +0.061 |
umpy | +0.058 |
irk | +0.057 |
back | +0.052 |
Neg logit contributions | |
lf | -0.090 |
ts | -0.089 |
cardboard | -0.088 |
blasted | -0.084 |
trek | -0.083 |
Token | will |
Token ID | 481 |
Feature activation | -4.59e-02 |
Pos logit contributions | |
cardboard | +0.014 |
ts | +0.013 |
lf | +0.013 |
pretend | +0.012 |
camps | +0.012 |
Neg logit contributions | |
eal | -0.022 |
arre | -0.022 |
irk | -0.021 |
umpy | -0.020 |
Fighting | -0.019 |
Token | not |
Token ID | 407 |
Feature activation | -4.40e-02 |
Pos logit contributions | |
eal | +0.088 |
arre | +0.086 |
irk | +0.079 |
umpy | +0.079 |
vious | +0.075 |
Neg logit contributions | |
cardboard | -0.069 |
lf | -0.066 |
ts | -0.062 |
camps | -0.059 |
pretend | -0.057 |
Token | be |
Token ID | 307 |
Feature activation | -7.21e-02 |
Pos logit contributions | |
eal | +0.077 |
arre | +0.074 |
irk | +0.069 |
umpy | +0.069 |
ags | +0.065 |
Neg logit contributions | |
cardboard | -0.072 |
lf | -0.070 |
ts | -0.066 |
camps | -0.063 |
blasted | -0.061 |
Token | mad |
Token ID | 8805 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
eal | +0.119 |
arre | +0.115 |
irk | +0.107 |
umpy | +0.105 |
agree | +0.099 |
Neg logit contributions | |
cardboard | -0.127 |
lf | -0.121 |
ts | -0.120 |
camps | -0.110 |
pretended | -0.109 |
Token | at |
Token ID | 379 |
Feature activation | -4.76e-02 |
Pos logit contributions | |
eal | +0.347 |
arre | +0.320 |
irk | +0.310 |
agree | +0.297 |
ags | +0.294 |
Neg logit contributions | |
cardboard | -0.251 |
lf | -0.243 |
ts | -0.231 |
camps | -0.229 |
pretend | -0.220 |
Token | them |
Token ID | 606 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
eal | +0.081 |
arre | +0.079 |
irk | +0.073 |
umpy | +0.073 |
ags | +0.071 |
Neg logit contributions | |
cardboard | -0.077 |
lf | -0.076 |
ts | -0.070 |
camps | -0.069 |
blasted | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | -3.85e-02 |
Pos logit contributions | |
eal | +0.264 |
arre | +0.245 |
irk | +0.241 |
umpy | +0.225 |
ags | +0.224 |
Neg logit contributions | |
cardboard | -0.131 |
lf | -0.126 |
ts | -0.115 |
camps | -0.112 |
pretend | -0.111 |
Token | They |
Token ID | 1119 |
Feature activation | +5.73e-02 |
Pos logit contributions | |
eal | +0.084 |
arre | +0.080 |
irk | +0.075 |
awk | +0.074 |
umpy | +0.073 |
Neg logit contributions | |
cardboard | -0.047 |
lf | -0.040 |
pretend | -0.039 |
ts | -0.038 |
camps | -0.038 |
Token | hope |
Token ID | 2911 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
cardboard | +0.085 |
lf | +0.079 |
ts | +0.078 |
pretend | +0.074 |
camps | +0.073 |
Neg logit contributions | |
eal | -0.108 |
arre | -0.107 |
umpy | -0.100 |
irk | -0.099 |
ags | -0.093 |
Token | please |
Token ID | 3387 |
Feature activation | -6.86e-02 |
Pos logit contributions | |
cardboard | +0.061 |
lf | +0.059 |
pretend | +0.055 |
ts | +0.055 |
camps | +0.054 |
Neg logit contributions | |
eal | -0.068 |
arre | -0.067 |
irk | -0.063 |
umpy | -0.063 |
ags | -0.059 |
Token | . |
Token ID | 13 |
Feature activation | -4.43e-02 |
Pos logit contributions | |
eal | +0.125 |
arre | +0.120 |
umpy | +0.112 |
irk | +0.111 |
ags | +0.108 |
Neg logit contributions | |
cardboard | -0.105 |
lf | -0.103 |
ts | -0.096 |
camps | -0.093 |
trek | -0.092 |
Token | Do |
Token ID | 2141 |
Feature activation | +2.63e-02 |
Pos logit contributions | |
eal | +0.060 |
arre | +0.058 |
umpy | +0.052 |
irk | +0.052 |
ags | +0.048 |
Neg logit contributions | |
cardboard | -0.090 |
lf | -0.088 |
ts | -0.083 |
camps | -0.081 |
blasted | -0.079 |
Token | not |
Token ID | 407 |
Feature activation | -4.66e-02 |
Pos logit contributions | |
cardboard | +0.047 |
lf | +0.045 |
camps | +0.044 |
blasted | +0.043 |
trek | +0.042 |
Neg logit contributions | |
eal | -0.039 |
arre | -0.038 |
agree | -0.034 |
umpy | -0.034 |
irk | -0.034 |
Token | be |
Token ID | 307 |
Feature activation | -7.39e-02 |
Pos logit contributions | |
eal | +0.087 |
arre | +0.085 |
umpy | +0.079 |
irk | +0.078 |
vious | +0.075 |
Neg logit contributions | |
cardboard | -0.069 |
lf | -0.068 |
ts | -0.064 |
camps | -0.061 |
trek | -0.059 |
Token | mad |
Token ID | 8805 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
eal | +0.114 |
irk | +0.107 |
arre | +0.106 |
agree | +0.100 |
umpy | +0.099 |
Neg logit contributions | |
cardboard | -0.132 |
called | -0.128 |
ts | -0.126 |
pretend | -0.120 |
named | -0.119 |
Token | . |
Token ID | 13 |
Feature activation | -4.45e-02 |
Pos logit contributions | |
eal | +0.394 |
arre | +0.363 |
irk | +0.358 |
agree | +0.345 |
ags | +0.340 |
Neg logit contributions | |
lf | -0.194 |
cardboard | -0.190 |
ts | -0.173 |
camps | -0.172 |
pretend | -0.171 |
Token | Do |
Token ID | 2141 |
Feature activation | +2.62e-02 |
Pos logit contributions | |
eal | +0.073 |
arre | +0.071 |
irk | +0.065 |
umpy | +0.064 |
ags | +0.060 |
Neg logit contributions | |
cardboard | -0.078 |
lf | -0.075 |
ts | -0.070 |
pretend | -0.068 |
camps | -0.067 |
Token | not |
Token ID | 407 |
Feature activation | -4.69e-02 |
Pos logit contributions | |
cardboard | +0.045 |
lf | +0.044 |
camps | +0.042 |
ts | +0.040 |
blasted | +0.040 |
Neg logit contributions | |
eal | -0.041 |
arre | -0.041 |
umpy | -0.036 |
irk | -0.036 |
agree | -0.036 |
Token | take |
Token ID | 1011 |
Feature activation | -4.52e-02 |
Pos logit contributions | |
eal | +0.089 |
arre | +0.086 |
umpy | +0.081 |
irk | +0.081 |
ags | +0.076 |
Neg logit contributions | |
cardboard | -0.069 |
lf | -0.068 |
ts | -0.063 |
camps | -0.060 |
blasted | -0.058 |
Token | our |
Token ID | 674 |
Feature activation | -8.87e-02 |
Pos logit contributions | |
eal | +0.096 |
arre | +0.094 |
umpy | +0.090 |
irk | +0.089 |
agree | +0.084 |
Neg logit contributions | |
cardboard | -0.059 |
ts | -0.055 |
lf | -0.053 |
pretend | -0.048 |
camps | -0.046 |
Token | . |
Token ID | 13 |
Feature activation | -4.01e-02 |
Pos logit contributions | |
cardboard | +0.009 |
lf | +0.009 |
ts | +0.008 |
camps | +0.008 |
pretend | +0.007 |
Neg logit contributions | |
eal | -0.016 |
arre | -0.015 |
irk | -0.014 |
umpy | -0.014 |
agree | -0.013 |
Token | You |
Token ID | 921 |
Feature activation | -5.52e-02 |
Pos logit contributions | |
eal | +0.057 |
arre | +0.054 |
irk | +0.049 |
umpy | +0.049 |
awk | +0.045 |
Neg logit contributions | |
cardboard | -0.079 |
lf | -0.076 |
ts | -0.072 |
camps | -0.070 |
blasted | -0.068 |
Token | 're |
Token ID | 821 |
Feature activation | -2.89e-02 |
Pos logit contributions | |
eal | +0.103 |
arre | +0.103 |
umpy | +0.095 |
irk | +0.092 |
vious | +0.091 |
Neg logit contributions | |
lf | -0.082 |
cardboard | -0.081 |
ts | -0.073 |
camps | -0.073 |
trek | -0.069 |
Token | tired |
Token ID | 10032 |
Feature activation | -3.88e-02 |
Pos logit contributions | |
eal | +0.056 |
irk | +0.051 |
arre | +0.050 |
umpy | +0.048 |
awk | +0.047 |
Neg logit contributions | |
cardboard | -0.047 |
lf | -0.043 |
ts | -0.043 |
called | -0.039 |
pretend | -0.038 |
Token | and |
Token ID | 290 |
Feature activation | -2.86e-02 |
Pos logit contributions | |
eal | +0.084 |
arre | +0.080 |
irk | +0.076 |
umpy | +0.073 |
ags | +0.073 |
Neg logit contributions | |
cardboard | -0.048 |
lf | -0.046 |
ts | -0.041 |
camps | -0.040 |
pretend | -0.038 |
Token | crank |
Token ID | 30425 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
eal | +0.046 |
arre | +0.044 |
irk | +0.041 |
ags | +0.039 |
umpy | +0.038 |
Neg logit contributions | |
cardboard | -0.051 |
ts | -0.049 |
lf | -0.048 |
pretend | -0.046 |
pretended | -0.045 |
Token | y |
Token ID | 88 |
Feature activation | +3.49e-02 |
Pos logit contributions | |
eal | +0.339 |
ags | +0.299 |
Aud | +0.296 |
irk | +0.296 |
arre | +0.286 |
Neg logit contributions | |
called | -0.221 |
lf | -0.212 |
ts | -0.208 |
pretended | -0.206 |
cardboard | -0.197 |
Token | ." |
Token ID | 526 |
Feature activation | -4.40e-02 |
Pos logit contributions | |
cardboard | +0.041 |
lf | +0.037 |
ts | +0.035 |
pretend | +0.034 |
camps | +0.033 |
Neg logit contributions | |
eal | -0.079 |
irk | -0.073 |
arre | -0.072 |
ags | -0.068 |
agree | -0.066 |
Token |
|
Token ID | 198 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
eal | +0.090 |
arre | +0.085 |
umpy | +0.078 |
vious | +0.077 |
irk | +0.077 |
Neg logit contributions | |
cardboard | -0.062 |
lf | -0.060 |
ts | -0.053 |
shovel | -0.051 |
pretend | -0.050 |
Token |
|
Token ID | 198 |
Feature activation | -1.76e-02 |
Pos logit contributions | |
eal | +0.028 |
arre | +0.027 |
irk | +0.026 |
umpy | +0.026 |
Fighting | +0.024 |
Neg logit contributions | |
cardboard | -0.032 |
lf | -0.031 |
ts | -0.030 |
camps | -0.029 |
blasted | -0.029 |
Token | S |
Token ID | 50 |
Feature activation | -5.76e-02 |
Pos logit contributions | |
eal | +0.032 |
arre | +0.031 |
umpy | +0.029 |
irk | +0.029 |
ags | +0.028 |
Neg logit contributions | |
cardboard | -0.027 |
lf | -0.027 |
ts | -0.025 |
camps | -0.024 |
trek | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | -4.74e-02 |
Pos logit contributions | |
lf | +0.026 |
cardboard | +0.026 |
ts | +0.024 |
camps | +0.024 |
trek | +0.023 |
Neg logit contributions | |
eal | -0.032 |
arre | -0.031 |
irk | -0.029 |
umpy | -0.029 |
back | -0.028 |
Token | car |
Token ID | 1097 |
Feature activation | +4.05e-02 |
Pos logit contributions | |
eal | +0.089 |
arre | +0.086 |
umpy | +0.080 |
irk | +0.080 |
ags | +0.076 |
Neg logit contributions | |
cardboard | -0.072 |
lf | -0.070 |
ts | -0.065 |
camps | -0.062 |
blasted | -0.060 |
Token | and |
Token ID | 290 |
Feature activation | -3.07e-02 |
Pos logit contributions | |
cardboard | +0.062 |
lf | +0.060 |
ts | +0.056 |
camps | +0.053 |
blasted | +0.052 |
Neg logit contributions | |
eal | -0.076 |
arre | -0.073 |
irk | -0.069 |
umpy | -0.068 |
ags | -0.065 |
Token | not |
Token ID | 407 |
Feature activation | -4.30e-02 |
Pos logit contributions | |
eal | +0.067 |
arre | +0.066 |
umpy | +0.063 |
irk | +0.061 |
ealing | +0.059 |
Neg logit contributions | |
cardboard | -0.040 |
lf | -0.038 |
ts | -0.035 |
pretend | -0.033 |
pretended | -0.032 |
Token | be |
Token ID | 307 |
Feature activation | -7.31e-02 |
Pos logit contributions | |
eal | +0.076 |
arre | +0.074 |
umpy | +0.069 |
irk | +0.068 |
vious | +0.065 |
Neg logit contributions | |
cardboard | -0.068 |
lf | -0.067 |
ts | -0.062 |
camps | -0.061 |
trek | -0.059 |
Token | mad |
Token ID | 8805 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
eal | +0.121 |
irk | +0.112 |
arre | +0.111 |
umpy | +0.102 |
agree | +0.102 |
Neg logit contributions | |
cardboard | -0.132 |
ts | -0.127 |
lf | -0.126 |
pretended | -0.116 |
pretend | -0.112 |
Token | at |
Token ID | 379 |
Feature activation | -4.71e-02 |
Pos logit contributions | |
eal | +0.345 |
arre | +0.321 |
irk | +0.308 |
agree | +0.301 |
ags | +0.296 |
Neg logit contributions | |
cardboard | -0.252 |
lf | -0.243 |
camps | -0.229 |
ts | -0.228 |
pretend | -0.217 |
Token | them |
Token ID | 606 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
arre | +0.073 |
eal | +0.072 |
irk | +0.067 |
ags | +0.067 |
back | +0.067 |
Neg logit contributions | |
cardboard | -0.078 |
lf | -0.078 |
camps | -0.073 |
blasted | -0.072 |
trek | -0.071 |
Token | . |
Token ID | 13 |
Feature activation | -3.69e-02 |
Pos logit contributions | |
eal | +0.258 |
arre | +0.243 |
irk | +0.236 |
umpy | +0.226 |
ags | +0.224 |
Neg logit contributions | |
cardboard | -0.133 |
lf | -0.130 |
ts | -0.116 |
camps | -0.113 |
pretend | -0.113 |
Token |
|
Token ID | 198 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
eal | +0.075 |
arre | +0.072 |
umpy | +0.067 |
irk | +0.066 |
vious | +0.065 |
Neg logit contributions | |
cardboard | -0.050 |
lf | -0.047 |
ts | -0.043 |
camps | -0.043 |
pretend | -0.042 |
Token |
|
Token ID | 198 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
eal | +0.024 |
arre | +0.023 |
umpy | +0.021 |
irk | +0.020 |
vious | +0.019 |
Neg logit contributions | |
cardboard | -0.030 |
lf | -0.029 |
ts | -0.028 |
camps | -0.027 |
trek | -0.026 |
Token | You |
Token ID | 921 |
Feature activation | -5.05e-02 |
Pos logit contributions | |
arre | +0.064 |
eal | +0.062 |
irk | +0.055 |
umpy | +0.053 |
ags | +0.053 |
Neg logit contributions | |
cardboard | -0.055 |
lf | -0.050 |
pretend | -0.050 |
blasted | -0.048 |
ts | -0.047 |
Token | are |
Token ID | 389 |
Feature activation | -1.59e-02 |
Pos logit contributions | |
eal | +0.090 |
arre | +0.089 |
umpy | +0.084 |
vious | +0.080 |
awk | +0.080 |
Neg logit contributions | |
lf | -0.082 |
cardboard | -0.078 |
camps | -0.071 |
ts | -0.069 |
trek | -0.066 |
Token | making |
Token ID | 1642 |
Feature activation | -4.64e-02 |
Pos logit contributions | |
eal | +0.031 |
arre | +0.030 |
irk | +0.028 |
umpy | +0.027 |
vious | +0.026 |
Neg logit contributions | |
cardboard | -0.024 |
lf | -0.023 |
ts | -0.022 |
camps | -0.020 |
pretend | -0.020 |
Token | my |
Token ID | 616 |
Feature activation | -7.68e-02 |
Pos logit contributions | |
eal | +0.109 |
irk | +0.103 |
arre | +0.102 |
umpy | +0.100 |
olding | +0.095 |
Neg logit contributions | |
cardboard | -0.052 |
ts | -0.047 |
lf | -0.046 |
pretend | -0.043 |
camps | -0.038 |
Token | dog |
Token ID | 3290 |
Feature activation | +3.80e-03 |
Pos logit contributions | |
arre | +0.147 |
eal | +0.146 |
umpy | +0.138 |
irk | +0.136 |
aback | +0.134 |
Neg logit contributions | |
cardboard | -0.109 |
pretend | -0.107 |
lf | -0.099 |
ts | -0.097 |
hunt | -0.096 |
Token | mad |
Token ID | 8805 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
cardboard | +0.006 |
pretend | +0.006 |
ts | +0.006 |
lf | +0.006 |
hunt | +0.005 |
Neg logit contributions | |
eal | -0.007 |
irk | -0.007 |
arre | -0.006 |
ags | -0.006 |
umpy | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -3.80e-02 |
Pos logit contributions | |
eal | +0.386 |
arre | +0.365 |
irk | +0.351 |
agree | +0.333 |
Fighting | +0.330 |
Neg logit contributions | |
cardboard | -0.209 |
lf | -0.203 |
ts | -0.188 |
camps | -0.178 |
pretend | -0.175 |
Token | He |
Token ID | 679 |
Feature activation | +7.90e-02 |
Pos logit contributions | |
eal | +0.063 |
arre | +0.062 |
irk | +0.054 |
fault | +0.051 |
awk | +0.050 |
Neg logit contributions | |
cardboard | -0.063 |
lf | -0.061 |
called | -0.057 |
pretend | -0.057 |
named | -0.056 |
Token | does |
Token ID | 857 |
Feature activation | -1.97e-02 |
Pos logit contributions | |
cardboard | +0.128 |
lf | +0.127 |
ts | +0.116 |
camps | +0.113 |
trek | +0.109 |
Neg logit contributions | |
eal | -0.139 |
arre | -0.135 |
umpy | -0.125 |
irk | -0.123 |
vious | -0.119 |
Token | not |
Token ID | 407 |
Feature activation | -4.37e-02 |
Pos logit contributions | |
eal | +0.046 |
arre | +0.045 |
irk | +0.043 |
umpy | +0.042 |
vious | +0.040 |
Neg logit contributions | |
cardboard | -0.022 |
lf | -0.021 |
ts | -0.021 |
pretend | -0.020 |
pretended | -0.017 |
Token | like |
Token ID | 588 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
eal | +0.086 |
arre | +0.085 |
irk | +0.079 |
umpy | +0.077 |
ags | +0.074 |
Neg logit contributions | |
cardboard | -0.060 |
ts | -0.057 |
lf | -0.056 |
pretend | -0.054 |
pretended | -0.052 |
Token | They |
Token ID | 1119 |
Feature activation | +5.67e-02 |
Pos logit contributions | |
eal | +0.067 |
arre | +0.065 |
umpy | +0.060 |
irk | +0.059 |
vious | +0.058 |
Neg logit contributions | |
cardboard | -0.052 |
lf | -0.049 |
ts | -0.046 |
camps | -0.044 |
pretend | -0.044 |
Token | told |
Token ID | 1297 |
Feature activation | -3.96e-02 |
Pos logit contributions | |
cardboard | +0.096 |
lf | +0.095 |
camps | +0.091 |
fossils | +0.085 |
dice | +0.084 |
Neg logit contributions | |
eal | -0.087 |
back | -0.086 |
disappointed | -0.085 |
mad | -0.083 |
ashamed | -0.080 |
Token | them |
Token ID | 606 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
eal | +0.075 |
arre | +0.074 |
umpy | +0.070 |
irk | +0.068 |
vious | +0.064 |
Neg logit contributions | |
cardboard | -0.061 |
lf | -0.058 |
ts | -0.056 |
pretend | -0.051 |
camps | -0.051 |
Token | what |
Token ID | 644 |
Feature activation | +1.45e-03 |
Pos logit contributions | |
eal | +0.244 |
arre | +0.233 |
irk | +0.224 |
umpy | +0.222 |
ags | +0.215 |
Neg logit contributions | |
cardboard | -0.142 |
lf | -0.135 |
ts | -0.129 |
camps | -0.115 |
pretend | -0.115 |
Token | they |
Token ID | 484 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
cardboard | +0.002 |
lf | +0.002 |
ts | +0.002 |
camps | +0.002 |
blasted | +0.002 |
Neg logit contributions | |
eal | -0.003 |
arre | -0.003 |
umpy | -0.003 |
irk | -0.003 |
ags | -0.003 |
Token | saw |
Token ID | 2497 |
Feature activation | +9.11e-02 |
Pos logit contributions | |
eal | +0.042 |
arre | +0.041 |
umpy | +0.039 |
irk | +0.039 |
vious | +0.038 |
Neg logit contributions | |
cardboard | -0.025 |
lf | -0.025 |
camps | -0.022 |
ts | -0.021 |
trek | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -3.71e-02 |
Pos logit contributions | |
cardboard | +0.142 |
lf | +0.139 |
camps | +0.126 |
ts | +0.123 |
blasted | +0.122 |
Neg logit contributions | |
eal | -0.163 |
arre | -0.156 |
umpy | -0.150 |
irk | -0.148 |
vious | -0.144 |
Token |
|
Token ID | 198 |
Feature activation | -1.68e-02 |
Pos logit contributions | |
eal | +0.082 |
arre | +0.078 |
vious | +0.072 |
umpy | +0.072 |
awk | +0.071 |
Neg logit contributions | |
cardboard | -0.047 |
lf | -0.040 |
pretend | -0.038 |
called | -0.037 |
shovel | -0.037 |
Token |
|
Token ID | 198 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
eal | +0.025 |
arre | +0.024 |
umpy | +0.023 |
irk | +0.023 |
ags | +0.021 |
Neg logit contributions | |
cardboard | -0.031 |
lf | -0.030 |
ts | -0.029 |
camps | -0.028 |
blasted | -0.027 |
Token | " |
Token ID | 1 |
Feature activation | -2.89e-02 |
Pos logit contributions | |
eal | +0.026 |
arre | +0.026 |
umpy | +0.024 |
irk | +0.023 |
Fighting | +0.022 |
Neg logit contributions | |
cardboard | -0.029 |
lf | -0.028 |
ts | -0.026 |
camps | -0.026 |
trek | -0.025 |
Token | Mom |
Token ID | 29252 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
eal | +0.053 |
arre | +0.050 |
irk | +0.047 |
umpy | +0.047 |
ags | +0.045 |
Neg logit contributions | |
cardboard | -0.046 |
lf | -0.046 |
ts | -0.043 |
camps | -0.040 |
blasted | -0.039 |
Token | day |
Token ID | 1110 |
Feature activation | -8.31e-03 |
Pos logit contributions | |
cardboard | +0.142 |
lf | +0.139 |
ts | +0.129 |
camps | +0.125 |
blasted | +0.121 |
Neg logit contributions | |
eal | -0.155 |
arre | -0.151 |
irk | -0.139 |
umpy | -0.136 |
ags | -0.132 |
Token | , |
Token ID | 11 |
Feature activation | +6.14e-02 |
Pos logit contributions | |
eal | +0.012 |
back | +0.012 |
arre | +0.012 |
ealing | +0.011 |
awk | +0.011 |
Neg logit contributions | |
lf | -0.015 |
cardboard | -0.014 |
ts | -0.013 |
camps | -0.013 |
trek | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | -3.45e-02 |
Pos logit contributions | |
lf | +0.117 |
cardboard | +0.108 |
camps | +0.105 |
trek | +0.100 |
ts | +0.098 |
Neg logit contributions | |
eal | -0.076 |
back | -0.074 |
disappointed | -0.072 |
mad | -0.070 |
ashamed | -0.065 |
Token | hipp |
Token ID | 18568 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
eal | +0.061 |
arre | +0.058 |
umpy | +0.057 |
back | +0.053 |
ags | +0.053 |
Neg logit contributions | |
lf | -0.054 |
cardboard | -0.053 |
ts | -0.049 |
trek | -0.047 |
blasted | -0.046 |
Token | o |
Token ID | 78 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
cardboard | +0.035 |
lf | +0.034 |
ts | +0.032 |
camps | +0.031 |
trek | +0.030 |
Neg logit contributions | |
eal | -0.023 |
arre | -0.023 |
umpy | -0.021 |
irk | -0.020 |
vious | -0.019 |
Token | saw |
Token ID | 2497 |
Feature activation | +9.26e-02 |
Pos logit contributions | |
cardboard | +0.043 |
lf | +0.042 |
ts | +0.038 |
camps | +0.035 |
trek | +0.033 |
Neg logit contributions | |
arre | -0.065 |
eal | -0.065 |
umpy | -0.062 |
vious | -0.058 |
irk | -0.057 |
Token | a |
Token ID | 257 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
lf | +0.116 |
cardboard | +0.111 |
ts | +0.099 |
camps | +0.098 |
trek | +0.095 |
Neg logit contributions | |
eal | -0.191 |
arre | -0.186 |
umpy | -0.175 |
vious | -0.175 |
back | -0.174 |
Token | little |
Token ID | 1310 |
Feature activation | +9.85e-02 |
Pos logit contributions | |
lf | +0.220 |
cardboard | +0.215 |
ts | +0.211 |
camps | +0.191 |
trek | +0.190 |
Neg logit contributions | |
eal | -0.293 |
arre | -0.287 |
umpy | -0.272 |
irk | -0.270 |
back | -0.255 |
Token | bird |
Token ID | 6512 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
cardboard | +0.153 |
lf | +0.144 |
ts | +0.135 |
camps | +0.132 |
shovel | +0.127 |
Neg logit contributions | |
eal | -0.183 |
arre | -0.178 |
umpy | -0.164 |
irk | -0.163 |
ags | -0.160 |
Token | . |
Token ID | 13 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
lf | +0.023 |
cardboard | +0.022 |
ts | +0.021 |
camps | +0.020 |
fossils | +0.019 |
Neg logit contributions | |
eal | -0.028 |
arre | -0.028 |
umpy | -0.027 |
irk | -0.026 |
vious | -0.026 |
Token | The |
Token ID | 383 |
Feature activation | +3.09e-02 |
Pos logit contributions | |
arre | +0.024 |
eal | +0.024 |
irk | +0.021 |
umpy | +0.021 |
ags | +0.020 |
Neg logit contributions | |
cardboard | -0.027 |
lf | -0.026 |
ts | -0.025 |
blasted | -0.024 |
camps | -0.023 |
Token | boy |
Token ID | 2933 |
Feature activation | +4.31e-02 |
Pos logit contributions | |
cardboard | +0.213 |
lf | +0.211 |
ts | +0.199 |
camps | +0.186 |
trek | +0.181 |
Neg logit contributions | |
eal | -0.307 |
arre | -0.300 |
irk | -0.280 |
umpy | -0.280 |
vious | -0.262 |
Token | who |
Token ID | 508 |
Feature activation | +6.67e-02 |
Pos logit contributions | |
cardboard | +0.058 |
lf | +0.057 |
camps | +0.054 |
ts | +0.054 |
blasted | +0.051 |
Neg logit contributions | |
eal | -0.084 |
arre | -0.083 |
irk | -0.079 |
umpy | -0.077 |
mad | -0.077 |
Token | was |
Token ID | 373 |
Feature activation | +3.45e-02 |
Pos logit contributions | |
cardboard | +0.076 |
lf | +0.072 |
ts | +0.066 |
camps | +0.065 |
blasted | +0.060 |
Neg logit contributions | |
eal | -0.147 |
arre | -0.144 |
umpy | -0.137 |
irk | -0.137 |
vious | -0.131 |
Token | playing |
Token ID | 2712 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
cardboard | +0.062 |
lf | +0.061 |
ts | +0.057 |
camps | +0.055 |
blasted | +0.053 |
Neg logit contributions | |
eal | -0.056 |
arre | -0.053 |
irk | -0.049 |
umpy | -0.049 |
ags | -0.046 |
Token | with |
Token ID | 351 |
Feature activation | -2.53e-02 |
Pos logit contributions | |
cardboard | +0.019 |
lf | +0.018 |
ts | +0.016 |
camps | +0.015 |
blasted | +0.015 |
Neg logit contributions | |
eal | -0.034 |
arre | -0.033 |
irk | -0.031 |
umpy | -0.031 |
ags | -0.030 |
Token | a |
Token ID | 257 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
eal | +0.040 |
arre | +0.040 |
back | +0.039 |
ags | +0.038 |
irk | +0.037 |
Neg logit contributions | |
lf | -0.040 |
blasted | -0.038 |
camps | -0.036 |
cardboard | -0.036 |
trek | -0.035 |
Token | toy |
Token ID | 13373 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
cardboard | +0.216 |
lf | +0.214 |
ts | +0.200 |
camps | +0.188 |
blasted | +0.183 |
Neg logit contributions | |
eal | -0.308 |
arre | -0.300 |
umpy | -0.280 |
irk | -0.280 |
ags | -0.265 |
Token | that |
Token ID | 326 |
Feature activation | -7.46e-03 |
Pos logit contributions | |
lf | +0.035 |
ts | +0.035 |
trek | +0.034 |
cardboard | +0.031 |
blasted | +0.031 |
Neg logit contributions | |
back | -0.031 |
arre | -0.028 |
eal | -0.027 |
umpy | -0.027 |
vious | -0.027 |
Token | she |
Token ID | 673 |
Feature activation | -3.22e-03 |
Pos logit contributions | |
eal | +0.016 |
arre | +0.016 |
irk | +0.015 |
umpy | +0.015 |
vious | +0.014 |
Neg logit contributions | |
cardboard | -0.008 |
lf | -0.008 |
ts | -0.008 |
camps | -0.007 |
trek | -0.007 |
Token | wanted |
Token ID | 2227 |
Feature activation | +5.00e-02 |
Pos logit contributions | |
eal | +0.006 |
arre | +0.006 |
umpy | +0.006 |
disappointed | +0.005 |
vious | +0.005 |
Neg logit contributions | |
cardboard | -0.005 |
camps | -0.004 |
lf | -0.004 |
shovel | -0.004 |
seas | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | -3.60e-03 |
Pos logit contributions | |
cardboard | +0.095 |
lf | +0.094 |
ts | +0.089 |
camps | +0.086 |
blasted | +0.084 |
Neg logit contributions | |
eal | -0.072 |
arre | -0.069 |
umpy | -0.064 |
irk | -0.063 |
ags | -0.059 |
Token | his |
Token ID | 465 |
Feature activation | -4.03e-02 |
Pos logit contributions | |
lf | +0.057 |
ts | +0.056 |
blasted | +0.056 |
cardboard | +0.056 |
pretended | +0.054 |
Neg logit contributions | |
eal | -0.036 |
back | -0.035 |
irk | -0.033 |
mad | -0.033 |
ags | -0.033 |
Token | garden |
Token ID | 11376 |
Feature activation | +5.96e-02 |
Pos logit contributions | |
eal | +0.080 |
arre | +0.078 |
umpy | +0.073 |
irk | +0.073 |
ags | +0.069 |
Neg logit contributions | |
cardboard | -0.056 |
lf | -0.055 |
ts | -0.051 |
camps | -0.048 |
blasted | -0.047 |
Token | . |
Token ID | 13 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
cardboard | +0.093 |
lf | +0.092 |
ts | +0.088 |
camps | +0.083 |
blasted | +0.083 |
Neg logit contributions | |
eal | -0.101 |
arre | -0.100 |
umpy | -0.094 |
irk | -0.092 |
ags | -0.088 |
Token |
|
Token ID | 198 |
Feature activation | -5.79e-03 |
Pos logit contributions | |
eal | +0.034 |
arre | +0.033 |
umpy | +0.030 |
irk | +0.030 |
ags | +0.029 |
Neg logit contributions | |
cardboard | -0.030 |
lf | -0.029 |
ts | -0.027 |
camps | -0.026 |
trek | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | -5.12e-03 |
Pos logit contributions | |
eal | +0.009 |
arre | +0.009 |
irk | +0.008 |
umpy | +0.008 |
ags | +0.007 |
Neg logit contributions | |
cardboard | -0.010 |
lf | -0.010 |
ts | -0.010 |
camps | -0.009 |
blasted | -0.009 |
Token | He |
Token ID | 1544 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
eal | +0.009 |
arre | +0.009 |
umpy | +0.008 |
irk | +0.008 |
vious | +0.008 |
Neg logit contributions | |
cardboard | -0.008 |
lf | -0.008 |
ts | -0.008 |
camps | -0.007 |
blasted | -0.007 |
Token | picked |
Token ID | 6497 |
Feature activation | +1.80e-02 |
Pos logit contributions | |
cardboard | +0.160 |
camps | +0.155 |
fossils | +0.140 |
loot | +0.140 |
miles | +0.139 |
Neg logit contributions | |
eal | -0.163 |
arre | -0.143 |
back | -0.138 |
disappointed | -0.138 |
mad | -0.134 |
Token | up |
Token ID | 510 |
Feature activation | -7.23e-02 |
Pos logit contributions | |
lf | +0.031 |
cardboard | +0.031 |
ts | +0.029 |
camps | +0.028 |
blasted | +0.027 |
Neg logit contributions | |
eal | -0.029 |
arre | -0.028 |
irk | -0.026 |
umpy | -0.026 |
vious | -0.024 |
Token | the |
Token ID | 262 |
Feature activation | -3.79e-02 |
Pos logit contributions | |
eal | +0.121 |
arre | +0.113 |
back | +0.112 |
vious | +0.106 |
irk | +0.104 |
Neg logit contributions | |
lf | -0.118 |
cardboard | -0.110 |
camps | -0.106 |
blasted | -0.104 |
ts | -0.103 |
Token | mel |
Token ID | 7758 |
Feature activation | +7.20e-03 |
Pos logit contributions | |
eal | +0.074 |
arre | +0.072 |
umpy | +0.068 |
irk | +0.068 |
vious | +0.065 |
Neg logit contributions | |
lf | -0.052 |
cardboard | -0.051 |
ts | -0.050 |
camps | -0.046 |
blasted | -0.045 |
Token | on |
Token ID | 261 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
cardboard | +0.010 |
ts | +0.009 |
lf | +0.009 |
camps | +0.009 |
trek | +0.008 |
Neg logit contributions | |
eal | -0.014 |
arre | -0.014 |
umpy | -0.013 |
irk | -0.013 |
vious | -0.012 |
Token | you |
Token ID | 345 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
cardboard | +0.075 |
lf | +0.073 |
ts | +0.065 |
camps | +0.063 |
trek | +0.061 |
Neg logit contributions | |
eal | -0.117 |
arre | -0.117 |
irk | -0.108 |
umpy | -0.107 |
ags | -0.106 |
Token | can |
Token ID | 460 |
Feature activation | +3.58e-02 |
Pos logit contributions | |
eal | +0.222 |
arre | +0.214 |
umpy | +0.202 |
irk | +0.197 |
vious | +0.195 |
Neg logit contributions | |
lf | -0.142 |
cardboard | -0.137 |
ts | -0.127 |
camps | -0.125 |
trek | -0.114 |
Token | spin |
Token ID | 7906 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
cardboard | +0.046 |
lf | +0.044 |
ts | +0.040 |
pretend | +0.039 |
blasted | +0.038 |
Neg logit contributions | |
eal | -0.074 |
arre | -0.073 |
irk | -0.070 |
umpy | -0.068 |
ags | -0.065 |
Token | too |
Token ID | 1165 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
eal | +0.032 |
arre | +0.031 |
irk | +0.029 |
umpy | +0.029 |
olding | +0.027 |
Neg logit contributions | |
cardboard | -0.022 |
lf | -0.021 |
ts | -0.020 |
pretend | -0.019 |
camps | -0.018 |
Token | !" |
Token ID | 2474 |
Feature activation | -3.03e-02 |
Pos logit contributions | |
eal | +0.237 |
arre | +0.234 |
umpy | +0.230 |
irk | +0.217 |
vious | +0.210 |
Neg logit contributions | |
cardboard | -0.188 |
lf | -0.187 |
ts | -0.178 |
camps | -0.163 |
blasted | -0.162 |
Token | So |
Token ID | 1406 |
Feature activation | +9.51e-02 |
Pos logit contributions | |
arre | +0.058 |
eal | +0.056 |
umpy | +0.055 |
fault | +0.052 |
irk | +0.052 |
Neg logit contributions | |
cardboard | -0.043 |
lf | -0.040 |
called | -0.038 |
ts | -0.037 |
trek | -0.037 |
Token | the |
Token ID | 262 |
Feature activation | -3.92e-02 |
Pos logit contributions | |
cardboard | +0.113 |
lf | +0.110 |
ts | +0.099 |
camps | +0.093 |
blasted | +0.091 |
Neg logit contributions | |
eal | -0.210 |
arre | -0.206 |
umpy | -0.195 |
irk | -0.194 |
ags | -0.186 |
Token | little |
Token ID | 1310 |
Feature activation | +9.97e-02 |
Pos logit contributions | |
arre | +0.066 |
eal | +0.061 |
irk | +0.057 |
umpy | +0.056 |
ags | +0.055 |
Neg logit contributions | |
cardboard | -0.067 |
lf | -0.066 |
ts | -0.063 |
camps | -0.061 |
pretend | -0.060 |
Token | girl |
Token ID | 2576 |
Feature activation | +2.78e-02 |
Pos logit contributions | |
cardboard | +0.156 |
lf | +0.144 |
camps | +0.143 |
shovel | +0.143 |
pretend | +0.141 |
Neg logit contributions | |
arre | -0.184 |
eal | -0.159 |
ags | -0.154 |
umpy | -0.154 |
agree | -0.151 |
Token | and |
Token ID | 290 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
cardboard | +0.034 |
lf | +0.033 |
ts | +0.030 |
camps | +0.029 |
blasted | +0.028 |
Neg logit contributions | |
eal | -0.059 |
arre | -0.058 |
umpy | -0.054 |
irk | -0.054 |
ags | -0.052 |
Token | the |
Token ID | 262 |
Feature activation | -3.88e-02 |
Pos logit contributions | |
arre | +0.026 |
umpy | +0.026 |
eal | +0.026 |
irk | +0.025 |
ags | +0.024 |
Neg logit contributions | |
lf | -0.012 |
cardboard | -0.012 |
ts | -0.011 |
pretended | -0.010 |
trek | -0.010 |
Token | underneath |
Token ID | 14638 |
Feature activation | +3.15e-02 |
Pos logit contributions | |
eal | +0.017 |
back | +0.016 |
ags | +0.016 |
irk | +0.016 |
mad | +0.015 |
Neg logit contributions | |
lf | -0.015 |
cardboard | -0.015 |
ts | -0.014 |
blasted | -0.014 |
trek | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | -3.22e-02 |
Pos logit contributions | |
lf | +0.046 |
cardboard | +0.045 |
ts | +0.043 |
camps | +0.041 |
blasted | +0.040 |
Neg logit contributions | |
eal | -0.058 |
arre | -0.056 |
irk | -0.053 |
umpy | -0.052 |
ags | -0.050 |
Token | carpet |
Token ID | 20710 |
Feature activation | +8.88e-02 |
Pos logit contributions | |
eal | +0.066 |
arre | +0.064 |
umpy | +0.062 |
irk | +0.061 |
back | +0.059 |
Neg logit contributions | |
lf | -0.039 |
cardboard | -0.038 |
ts | -0.038 |
trek | -0.034 |
camps | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
cardboard | +0.126 |
lf | +0.123 |
ts | +0.116 |
camps | +0.110 |
blasted | +0.105 |
Neg logit contributions | |
eal | -0.171 |
arre | -0.166 |
umpy | -0.156 |
irk | -0.155 |
vious | -0.147 |
Token | The |
Token ID | 383 |
Feature activation | +3.15e-02 |
Pos logit contributions | |
eal | +0.034 |
arre | +0.034 |
umpy | +0.031 |
ags | +0.031 |
irk | +0.031 |
Neg logit contributions | |
cardboard | -0.029 |
lf | -0.027 |
called | -0.025 |
shovel | -0.024 |
pretend | -0.024 |
Token | puppet |
Token ID | 30095 |
Feature activation | +4.49e-03 |
Pos logit contributions | |
lf | +0.052 |
cardboard | +0.052 |
ts | +0.052 |
trek | +0.049 |
camps | +0.047 |
Neg logit contributions | |
eal | -0.052 |
arre | -0.047 |
irk | -0.047 |
umpy | -0.046 |
back | -0.044 |
Token | had |
Token ID | 550 |
Feature activation | +8.56e-02 |
Pos logit contributions | |
cardboard | +0.005 |
lf | +0.005 |
ts | +0.005 |
camps | +0.004 |
blasted | +0.004 |
Neg logit contributions | |
eal | -0.010 |
arre | -0.010 |
umpy | -0.009 |
irk | -0.009 |
ags | -0.009 |
Token | a |
Token ID | 257 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
lf | +0.098 |
cardboard | +0.098 |
ts | +0.088 |
camps | +0.084 |
blasted | +0.079 |
Neg logit contributions | |
eal | -0.188 |
arre | -0.182 |
umpy | -0.173 |
irk | -0.172 |
vious | -0.165 |
Token | friendly |
Token ID | 8030 |
Feature activation | -3.11e-02 |
Pos logit contributions | |
cardboard | +0.209 |
lf | +0.207 |
ts | +0.193 |
camps | +0.181 |
blasted | +0.174 |
Neg logit contributions | |
eal | -0.321 |
arre | -0.312 |
umpy | -0.292 |
irk | -0.292 |
ags | -0.276 |
Token | face |
Token ID | 1986 |
Feature activation | -5.90e-02 |
Pos logit contributions | |
arre | +0.053 |
eal | +0.052 |
umpy | +0.049 |
Fighting | +0.047 |
vious | +0.047 |
Neg logit contributions | |
cardboard | -0.054 |
lf | -0.051 |
pretend | -0.046 |
shovel | -0.045 |
ts | -0.045 |
Token | and |
Token ID | 290 |
Feature activation | -8.96e-03 |
Pos logit contributions | |
eal | +0.114 |
arre | +0.111 |
umpy | +0.105 |
irk | +0.104 |
vious | +0.099 |
Neg logit contributions | |
cardboard | -0.082 |
lf | -0.080 |
ts | -0.076 |
camps | -0.072 |
trek | -0.069 |
Token | and |
Token ID | 290 |
Feature activation | -1.50e-02 |
Pos logit contributions | |
cardboard | +0.062 |
lf | +0.060 |
ts | +0.055 |
camps | +0.052 |
blasted | +0.050 |
Neg logit contributions | |
eal | -0.101 |
arre | -0.098 |
irk | -0.093 |
umpy | -0.092 |
ags | -0.089 |
Token | put |
Token ID | 1234 |
Feature activation | +2.83e-03 |
Pos logit contributions | |
eal | +0.030 |
arre | +0.028 |
irk | +0.027 |
back | +0.026 |
umpy | +0.026 |
Neg logit contributions | |
lf | -0.020 |
cardboard | -0.020 |
ts | -0.018 |
camps | -0.018 |
trek | -0.017 |
Token | it |
Token ID | 340 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
cardboard | +0.004 |
lf | +0.003 |
ts | +0.003 |
camps | +0.003 |
shovel | +0.003 |
Neg logit contributions | |
eal | -0.006 |
arre | -0.006 |
umpy | -0.006 |
irk | -0.006 |
vious | -0.005 |
Token | on |
Token ID | 319 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
eal | +0.258 |
arre | +0.251 |
irk | +0.237 |
umpy | +0.235 |
vious | +0.222 |
Neg logit contributions | |
cardboard | -0.180 |
lf | -0.172 |
ts | -0.162 |
camps | -0.153 |
trek | -0.147 |
Token | a |
Token ID | 257 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
eal | +0.030 |
arre | +0.029 |
umpy | +0.027 |
irk | +0.027 |
ags | +0.026 |
Neg logit contributions | |
cardboard | -0.022 |
lf | -0.021 |
ts | -0.020 |
camps | -0.019 |
blasted | -0.018 |
Token | leaf |
Token ID | 12835 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
cardboard | +0.209 |
lf | +0.207 |
ts | +0.194 |
camps | +0.181 |
blasted | +0.178 |
Neg logit contributions | |
eal | -0.310 |
arre | -0.299 |
umpy | -0.284 |
irk | -0.281 |
ags | -0.267 |
Token | . |
Token ID | 13 |
Feature activation | -1.97e-02 |
Pos logit contributions | |
cardboard | +0.021 |
lf | +0.021 |
ts | +0.019 |
camps | +0.018 |
blasted | +0.017 |
Neg logit contributions | |
eal | -0.032 |
arre | -0.031 |
irk | -0.030 |
umpy | -0.029 |
ags | -0.028 |
Token | She |
Token ID | 1375 |
Feature activation | +5.23e-02 |
Pos logit contributions | |
eal | +0.033 |
arre | +0.032 |
umpy | +0.030 |
irk | +0.030 |
ags | +0.028 |
Neg logit contributions | |
cardboard | -0.033 |
lf | -0.033 |
ts | -0.030 |
camps | -0.029 |
trek | -0.028 |
Token | wanted |
Token ID | 2227 |
Feature activation | +4.85e-02 |
Pos logit contributions | |
lf | +0.074 |
cardboard | +0.073 |
ts | +0.065 |
camps | +0.063 |
blasted | +0.060 |
Neg logit contributions | |
eal | -0.099 |
arre | -0.094 |
vious | -0.085 |
irk | -0.085 |
disappointed | -0.084 |
Token | to |
Token ID | 284 |
Feature activation | -4.26e-03 |
Pos logit contributions | |
cardboard | +0.086 |
lf | +0.084 |
ts | +0.079 |
camps | +0.076 |
trek | +0.074 |
Neg logit contributions | |
eal | -0.080 |
arre | -0.077 |
umpy | -0.071 |
irk | -0.071 |
ags | -0.066 |
Token | weigh |
Token ID | 10164 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
eal | +0.008 |
arre | +0.007 |
umpy | +0.007 |
awk | +0.006 |
back | +0.006 |
Neg logit contributions | |
lf | -0.007 |
cardboard | -0.006 |
camps | -0.006 |
ts | -0.006 |
trek | -0.006 |
Token | From |
Token ID | 3574 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
eal | +0.067 |
arre | +0.063 |
umpy | +0.059 |
irk | +0.059 |
vious | +0.058 |
Neg logit contributions | |
cardboard | -0.045 |
lf | -0.043 |
camps | -0.038 |
pretend | -0.037 |
called | -0.037 |
Token | that |
Token ID | 326 |
Feature activation | -8.01e-03 |
Pos logit contributions | |
cardboard | +0.017 |
lf | +0.016 |
ts | +0.015 |
blasted | +0.015 |
camps | +0.015 |
Neg logit contributions | |
eal | -0.026 |
arre | -0.025 |
irk | -0.024 |
umpy | -0.024 |
ags | -0.023 |
Token | day |
Token ID | 1110 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
eal | +0.016 |
arre | +0.015 |
irk | +0.014 |
umpy | +0.014 |
ags | +0.014 |
Neg logit contributions | |
cardboard | -0.012 |
lf | -0.012 |
camps | -0.011 |
ts | -0.010 |
trek | -0.010 |
Token | on |
Token ID | 319 |
Feature activation | -1.68e-02 |
Pos logit contributions | |
eal | +0.026 |
arre | +0.026 |
umpy | +0.024 |
irk | +0.024 |
ags | +0.023 |
Neg logit contributions | |
cardboard | -0.022 |
lf | -0.021 |
ts | -0.020 |
camps | -0.019 |
trek | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | +5.28e-02 |
Pos logit contributions | |
eal | +0.037 |
arre | +0.036 |
umpy | +0.034 |
irk | +0.034 |
ags | +0.033 |
Neg logit contributions | |
cardboard | -0.018 |
lf | -0.018 |
ts | -0.017 |
camps | -0.016 |
blasted | -0.015 |
Token | Lily |
Token ID | 20037 |
Feature activation | +3.87e-03 |
Pos logit contributions | |
cardboard | +0.077 |
lf | +0.076 |
ts | +0.070 |
camps | +0.067 |
trek | +0.064 |
Neg logit contributions | |
eal | -0.102 |
arre | -0.098 |
irk | -0.092 |
umpy | -0.092 |
ags | -0.088 |
Token | learned |
Token ID | 4499 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
lf | +0.004 |
cardboard | +0.004 |
pretended | +0.004 |
ts | +0.004 |
blasted | +0.004 |
Neg logit contributions | |
eal | -0.009 |
irk | -0.008 |
arre | -0.008 |
ags | -0.008 |
awk | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | -6.21e-03 |
Pos logit contributions | |
eal | +0.044 |
arre | +0.040 |
irk | +0.039 |
umpy | +0.039 |
ealing | +0.039 |
Neg logit contributions | |
cardboard | -0.017 |
called | -0.014 |
lf | -0.013 |
treasure | -0.012 |
pretend | -0.012 |
Token | be |
Token ID | 307 |
Feature activation | -7.14e-02 |
Pos logit contributions | |
eal | +0.012 |
arre | +0.012 |
umpy | +0.011 |
irk | +0.011 |
vious | +0.011 |
Neg logit contributions | |
cardboard | -0.009 |
lf | -0.008 |
ts | -0.008 |
pretend | -0.008 |
trek | -0.007 |
Token | serious |
Token ID | 2726 |
Feature activation | -7.82e-02 |
Pos logit contributions | |
arre | +0.138 |
eal | +0.138 |
irk | +0.131 |
umpy | +0.124 |
awk | +0.123 |
Neg logit contributions | |
cardboard | -0.100 |
ts | -0.094 |
called | -0.092 |
hours | -0.090 |
pretend | -0.089 |
Token | when |
Token ID | 618 |
Feature activation | +5.29e-02 |
Pos logit contributions | |
eal | +0.154 |
arre | +0.150 |
umpy | +0.141 |
irk | +0.140 |
ags | +0.133 |
Neg logit contributions | |
cardboard | -0.109 |
lf | -0.107 |
ts | -0.099 |
camps | -0.094 |
blasted | -0.091 |
Token | crash |
Token ID | 7014 |
Feature activation | +3.37e-02 |
Pos logit contributions | |
eal | +0.033 |
arre | +0.033 |
irk | +0.030 |
umpy | +0.030 |
back | +0.029 |
Neg logit contributions | |
lf | -0.024 |
cardboard | -0.024 |
camps | -0.022 |
ts | -0.022 |
trek | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
cardboard | +0.050 |
lf | +0.050 |
ts | +0.046 |
camps | +0.044 |
trek | +0.043 |
Neg logit contributions | |
eal | -0.062 |
arre | -0.060 |
umpy | -0.056 |
irk | -0.055 |
ags | -0.053 |
Token |
|
Token ID | 198 |
Feature activation | -6.87e-03 |
Pos logit contributions | |
eal | +0.047 |
arre | +0.045 |
umpy | +0.043 |
irk | +0.042 |
awk | +0.041 |
Neg logit contributions | |
cardboard | -0.027 |
camps | -0.025 |
lf | -0.024 |
pretend | -0.023 |
trek | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | -6.34e-03 |
Pos logit contributions | |
eal | +0.012 |
arre | +0.011 |
umpy | +0.011 |
irk | +0.010 |
ags | +0.010 |
Neg logit contributions | |
cardboard | -0.011 |
lf | -0.011 |
ts | -0.011 |
camps | -0.010 |
blasted | -0.010 |
Token | One |
Token ID | 3198 |
Feature activation | +8.57e-02 |
Pos logit contributions | |
eal | +0.015 |
arre | +0.014 |
irk | +0.014 |
umpy | +0.014 |
ags | +0.013 |
Neg logit contributions | |
cardboard | -0.007 |
lf | -0.006 |
ts | -0.006 |
camps | -0.005 |
trek | -0.005 |
Token | day |
Token ID | 1110 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
cardboard | +0.131 |
lf | +0.130 |
camps | +0.115 |
ts | +0.114 |
pretend | +0.111 |
Neg logit contributions | |
eal | -0.162 |
arre | -0.159 |
irk | -0.147 |
ags | -0.142 |
umpy | -0.141 |
Token | , |
Token ID | 11 |
Feature activation | +5.89e-02 |
Pos logit contributions | |
eal | +0.017 |
arre | +0.016 |
back | +0.015 |
ealing | +0.015 |
forward | +0.014 |
Neg logit contributions | |
lf | -0.016 |
cardboard | -0.016 |
ts | -0.015 |
blasted | -0.014 |
camps | -0.014 |
Token | he |
Token ID | 339 |
Feature activation | +1.88e-02 |
Pos logit contributions | |
cardboard | +0.107 |
lf | +0.106 |
ts | +0.099 |
blasted | +0.094 |
trek | +0.093 |
Neg logit contributions | |
eal | -0.087 |
arre | -0.081 |
back | -0.075 |
disappointed | -0.074 |
irk | -0.074 |
Token | saw |
Token ID | 2497 |
Feature activation | +9.20e-02 |
Pos logit contributions | |
cardboard | +0.029 |
ts | +0.028 |
lf | +0.028 |
fossils | +0.025 |
loot | +0.024 |
Neg logit contributions | |
arre | -0.034 |
eal | -0.033 |
umpy | -0.032 |
back | -0.030 |
vious | -0.029 |
Token | a |
Token ID | 257 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
lf | +0.113 |
cardboard | +0.109 |
ts | +0.101 |
camps | +0.098 |
blasted | +0.097 |
Neg logit contributions | |
eal | -0.186 |
arre | -0.182 |
irk | -0.171 |
back | -0.170 |
vious | -0.169 |
Token | new |
Token ID | 649 |
Feature activation | -5.80e-02 |
Pos logit contributions | |
ts | +0.223 |
lf | +0.223 |
cardboard | +0.220 |
trek | +0.201 |
blasted | +0.198 |
Neg logit contributions | |
eal | -0.273 |
arre | -0.268 |
irk | -0.254 |
umpy | -0.252 |
back | -0.245 |
Token | , |
Token ID | 11 |
Feature activation | +6.25e-02 |
Pos logit contributions | |
ts | +0.015 |
cardboard | +0.014 |
lf | +0.014 |
camps | +0.013 |
seas | +0.013 |
Neg logit contributions | |
arre | -0.014 |
eal | -0.014 |
umpy | -0.013 |
back | -0.013 |
vious | -0.013 |
Token | a |
Token ID | 257 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
lf | +0.088 |
cardboard | +0.086 |
ts | +0.081 |
camps | +0.076 |
blasted | +0.072 |
Neg logit contributions | |
eal | -0.119 |
arre | -0.116 |
irk | -0.111 |
back | -0.107 |
umpy | -0.106 |
Token | man |
Token ID | 582 |
Feature activation | -4.78e-05 |
Pos logit contributions | |
cardboard | +0.202 |
lf | +0.196 |
ts | +0.176 |
camps | +0.170 |
blasted | +0.162 |
Neg logit contributions | |
eal | -0.334 |
arre | -0.324 |
irk | -0.304 |
umpy | -0.304 |
ags | -0.295 |
Token | named |
Token ID | 3706 |
Feature activation | +9.97e-02 |
Pos logit contributions | |
eal | +0.000 |
arre | +0.000 |
back | +0.000 |
irk | +0.000 |
aring | +0.000 |
Neg logit contributions | |
cardboard | -0.000 |
lf | -0.000 |
ts | -0.000 |
camps | -0.000 |
seas | -0.000 |
Token | John |
Token ID | 1757 |
Feature activation | +2.76e-02 |
Pos logit contributions | |
lf | +0.179 |
cardboard | +0.173 |
ts | +0.163 |
hours | +0.157 |
blasted | +0.157 |
Neg logit contributions | |
eal | -0.140 |
arre | -0.138 |
back | -0.128 |
umpy | -0.127 |
aback | -0.124 |
Token | , |
Token ID | 11 |
Feature activation | +6.21e-02 |
Pos logit contributions | |
lf | +0.042 |
cardboard | +0.041 |
ts | +0.038 |
seas | +0.037 |
dice | +0.036 |
Neg logit contributions | |
back | -0.048 |
eal | -0.047 |
arre | -0.046 |
umpy | -0.043 |
irk | -0.043 |
Token | came |
Token ID | 1625 |
Feature activation | +2.19e-02 |
Pos logit contributions | |
lf | +0.093 |
cardboard | +0.091 |
ts | +0.086 |
camps | +0.078 |
seas | +0.077 |
Neg logit contributions | |
arre | -0.113 |
eal | -0.113 |
back | -0.105 |
irk | -0.104 |
umpy | -0.104 |
Token | to |
Token ID | 284 |
Feature activation | +1.24e-03 |
Pos logit contributions | |
cardboard | +0.034 |
camps | +0.031 |
lf | +0.030 |
trek | +0.030 |
called | +0.030 |
Neg logit contributions | |
eal | -0.040 |
arre | -0.038 |
irk | -0.037 |
ealing | -0.035 |
agree | -0.035 |
Token | take |
Token ID | 1011 |
Feature activation | -3.72e-02 |
Pos logit contributions | |
lf | +0.002 |
cardboard | +0.002 |
ts | +0.002 |
camps | +0.002 |
blasted | +0.002 |
Neg logit contributions | |
eal | -0.002 |
arre | -0.002 |
irk | -0.002 |
back | -0.002 |
ags | -0.002 |
Token | him |
Token ID | 683 |
Feature activation | -5.48e-02 |
Pos logit contributions | |
eal | +0.074 |
arre | +0.072 |
umpy | +0.066 |
irk | +0.066 |
ags | +0.065 |
Neg logit contributions | |
lf | -0.052 |
cardboard | -0.050 |
ts | -0.046 |
camps | -0.043 |
blasted | -0.043 |
Token | for |
Token ID | 329 |
Feature activation | -6.65e-02 |
Pos logit contributions | |
eal | +0.090 |
arre | +0.084 |
irk | +0.081 |
umpy | +0.079 |
Fighting | +0.075 |
Neg logit contributions | |
cardboard | -0.100 |
lf | -0.091 |
ts | -0.088 |
camps | -0.088 |
trek | -0.085 |
Token | . |
Token ID | 13 |
Feature activation | -3.56e-02 |
Pos logit contributions | |
cardboard | +0.094 |
lf | +0.092 |
ts | +0.085 |
camps | +0.080 |
blasted | +0.077 |
Neg logit contributions | |
eal | -0.139 |
arre | -0.135 |
irk | -0.126 |
umpy | -0.126 |
ags | -0.120 |
Token | He |
Token ID | 679 |
Feature activation | +8.01e-02 |
Pos logit contributions | |
eal | +0.058 |
arre | +0.056 |
umpy | +0.051 |
irk | +0.051 |
ags | +0.048 |
Neg logit contributions | |
cardboard | -0.063 |
lf | -0.062 |
ts | -0.058 |
camps | -0.056 |
blasted | -0.055 |
Token | says |
Token ID | 1139 |
Feature activation | -3.43e-02 |
Pos logit contributions | |
lf | +0.140 |
cardboard | +0.140 |
ts | +0.128 |
camps | +0.124 |
blasted | +0.122 |
Neg logit contributions | |
eal | -0.129 |
arre | -0.125 |
umpy | -0.114 |
irk | -0.113 |
vious | -0.109 |
Token | , |
Token ID | 11 |
Feature activation | +4.39e-02 |
Pos logit contributions | |
eal | +0.101 |
arre | +0.097 |
irk | +0.094 |
umpy | +0.093 |
ags | +0.092 |
Neg logit contributions | |
cardboard | -0.021 |
lf | -0.017 |
ts | -0.014 |
shovel | -0.012 |
swords | -0.012 |
Token | " |
Token ID | 366 |
Feature activation | -8.13e-03 |
Pos logit contributions | |
cardboard | +0.056 |
lf | +0.054 |
ts | +0.048 |
shovel | +0.045 |
pretend | +0.045 |
Neg logit contributions | |
eal | -0.097 |
arre | -0.092 |
irk | -0.088 |
umpy | -0.088 |
awk | -0.087 |
Token | No |
Token ID | 2949 |
Feature activation | -6.46e-02 |
Pos logit contributions | |
eal | +0.014 |
arre | +0.013 |
irk | +0.012 |
ags | +0.012 |
fault | +0.012 |
Neg logit contributions | |
lf | -0.013 |
cardboard | -0.013 |
ts | -0.013 |
pretend | -0.012 |
pretended | -0.012 |
Token | , |
Token ID | 11 |
Feature activation | +4.68e-02 |
Pos logit contributions | |
eal | +0.148 |
arre | +0.145 |
irk | +0.137 |
umpy | +0.136 |
ags | +0.131 |
Neg logit contributions | |
cardboard | -0.069 |
lf | -0.067 |
ts | -0.062 |
camps | -0.057 |
blasted | -0.055 |
Token | you |
Token ID | 345 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
cardboard | +0.075 |
lf | +0.074 |
ts | +0.070 |
camps | +0.068 |
trek | +0.066 |
Neg logit contributions | |
eal | -0.083 |
arre | -0.079 |
irk | -0.073 |
umpy | -0.073 |
vious | -0.070 |
Token | go |
Token ID | 467 |
Feature activation | -4.01e-02 |
Pos logit contributions | |
arre | +0.186 |
eal | +0.173 |
umpy | +0.172 |
vious | +0.161 |
ealing | +0.159 |
Neg logit contributions | |
lf | -0.179 |
cardboard | -0.175 |
camps | -0.162 |
ts | -0.161 |
trek | -0.159 |
Token | away |
Token ID | 1497 |
Feature activation | -9.95e-02 |
Pos logit contributions | |
eal | +0.055 |
arre | +0.053 |
umpy | +0.050 |
back | +0.047 |
irk | +0.047 |
Neg logit contributions | |
cardboard | -0.080 |
lf | -0.080 |
ts | -0.075 |
camps | -0.072 |
trek | -0.072 |
Token | . |
Token ID | 13 |
Feature activation | -3.86e-02 |
Pos logit contributions | |
umpy | +0.160 |
arre | +0.158 |
eal | +0.156 |
back | +0.146 |
irk | +0.146 |
Neg logit contributions | |
cardboard | -0.166 |
lf | -0.161 |
ts | -0.152 |
trek | -0.151 |
blasted | -0.146 |
Token | poke |
Token ID | 22620 |
Feature activation | -1.68e-03 |
Pos logit contributions | |
eal | +0.039 |
arre | +0.037 |
irk | +0.034 |
umpy | +0.034 |
back | +0.034 |
Neg logit contributions | |
cardboard | -0.025 |
lf | -0.025 |
camps | -0.023 |
ts | -0.022 |
trek | -0.022 |
Token | her |
Token ID | 607 |
Feature activation | -7.32e-02 |
Pos logit contributions | |
eal | +0.003 |
arre | +0.003 |
umpy | +0.003 |
irk | +0.003 |
Fighting | +0.003 |
Neg logit contributions | |
cardboard | -0.003 |
lf | -0.002 |
ts | -0.002 |
pretend | -0.002 |
camps | -0.002 |
Token | teeth |
Token ID | 9941 |
Feature activation | +2.42e-02 |
Pos logit contributions | |
eal | +0.136 |
arre | +0.134 |
irk | +0.125 |
umpy | +0.124 |
Fighting | +0.117 |
Neg logit contributions | |
cardboard | -0.114 |
lf | -0.110 |
ts | -0.103 |
camps | -0.098 |
pretend | -0.097 |
Token | . |
Token ID | 13 |
Feature activation | -2.64e-02 |
Pos logit contributions | |
cardboard | +0.032 |
lf | +0.031 |
ts | +0.028 |
camps | +0.026 |
blasted | +0.026 |
Neg logit contributions | |
eal | -0.051 |
arre | -0.049 |
irk | -0.046 |
umpy | -0.046 |
ags | -0.044 |
Token | It |
Token ID | 632 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
eal | +0.054 |
arre | +0.051 |
irk | +0.049 |
awk | +0.048 |
umpy | +0.048 |
Neg logit contributions | |
cardboard | -0.037 |
lf | -0.035 |
ts | -0.031 |
camps | -0.030 |
pretend | -0.029 |
Token | hurt |
Token ID | 5938 |
Feature activation | -6.65e-02 |
Pos logit contributions | |
eal | +0.025 |
arre | +0.024 |
umpy | +0.023 |
irk | +0.022 |
vious | +0.022 |
Neg logit contributions | |
cardboard | -0.015 |
lf | -0.014 |
ts | -0.013 |
camps | -0.013 |
trek | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | -2.08e-02 |
Pos logit contributions | |
eal | +0.134 |
arre | +0.130 |
umpy | +0.122 |
irk | +0.122 |
ags | +0.116 |
Neg logit contributions | |
cardboard | -0.091 |
lf | -0.089 |
ts | -0.082 |
camps | -0.078 |
blasted | -0.075 |
Token | tasted |
Token ID | 29187 |
Feature activation | -6.41e-03 |
Pos logit contributions | |
eal | +0.040 |
arre | +0.039 |
irk | +0.036 |
umpy | +0.036 |
ags | +0.035 |
Neg logit contributions | |
cardboard | -0.031 |
lf | -0.030 |
ts | -0.028 |
camps | -0.026 |
blasted | -0.026 |
Token | bad |
Token ID | 2089 |
Feature activation | -9.07e-02 |
Pos logit contributions | |
eal | +0.013 |
arre | +0.012 |
irk | +0.012 |
umpy | +0.011 |
agree | +0.011 |
Neg logit contributions | |
cardboard | -0.009 |
ts | -0.009 |
lf | -0.009 |
pretend | -0.008 |
camps | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -2.75e-02 |
Pos logit contributions | |
eal | +0.184 |
arre | +0.179 |
umpy | +0.168 |
irk | +0.168 |
ags | +0.160 |
Neg logit contributions | |
cardboard | -0.122 |
lf | -0.119 |
ts | -0.110 |
camps | -0.104 |
blasted | -0.100 |
Token |
|
Token ID | 198 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
eal | +0.057 |
arre | +0.053 |
irk | +0.050 |
awk | +0.049 |
umpy | +0.049 |
Neg logit contributions | |
cardboard | -0.038 |
lf | -0.037 |
ts | -0.032 |
camps | -0.030 |
pretend | -0.030 |
Token | again |
Token ID | 757 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
eal | +0.023 |
arre | +0.023 |
umpy | +0.022 |
irk | +0.021 |
ags | +0.021 |
Neg logit contributions | |
cardboard | -0.012 |
lf | -0.012 |
ts | -0.011 |
camps | -0.010 |
blasted | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
eal | +0.230 |
arre | +0.221 |
irk | +0.208 |
umpy | +0.208 |
back | +0.201 |
Neg logit contributions | |
cardboard | -0.150 |
lf | -0.147 |
ts | -0.133 |
camps | -0.129 |
trek | -0.122 |
Token | He |
Token ID | 679 |
Feature activation | +8.37e-02 |
Pos logit contributions | |
eal | +0.037 |
arre | +0.036 |
irk | +0.033 |
umpy | +0.033 |
vious | +0.032 |
Neg logit contributions | |
cardboard | -0.030 |
lf | -0.030 |
ts | -0.027 |
camps | -0.027 |
trek | -0.027 |
Token | kept |
Token ID | 4030 |
Feature activation | -2.07e-02 |
Pos logit contributions | |
cardboard | +0.102 |
lf | +0.099 |
ts | +0.090 |
camps | +0.085 |
blasted | +0.080 |
Neg logit contributions | |
eal | -0.181 |
arre | -0.175 |
umpy | -0.164 |
irk | -0.164 |
vious | -0.157 |
Token | jumping |
Token ID | 14284 |
Feature activation | -3.82e-02 |
Pos logit contributions | |
eal | +0.049 |
umpy | +0.049 |
arre | +0.048 |
irk | +0.046 |
vious | +0.045 |
Neg logit contributions | |
cardboard | -0.022 |
lf | -0.020 |
ts | -0.019 |
pretend | -0.018 |
camps | -0.017 |
Token | up |
Token ID | 510 |
Feature activation | -7.29e-02 |
Pos logit contributions | |
eal | +0.079 |
arre | +0.077 |
irk | +0.073 |
umpy | +0.073 |
ags | +0.069 |
Neg logit contributions | |
cardboard | -0.050 |
lf | -0.048 |
ts | -0.045 |
camps | -0.043 |
trek | -0.041 |
Token | and |
Token ID | 290 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
eal | +0.139 |
arre | +0.135 |
umpy | +0.126 |
irk | +0.126 |
ags | +0.120 |
Neg logit contributions | |
lf | -0.104 |
cardboard | -0.103 |
ts | -0.095 |
camps | -0.089 |
blasted | -0.088 |
Token | down |
Token ID | 866 |
Feature activation | -6.36e-02 |
Pos logit contributions | |
eal | +0.020 |
arre | +0.019 |
umpy | +0.019 |
irk | +0.019 |
ags | +0.018 |
Neg logit contributions | |
cardboard | -0.021 |
lf | -0.021 |
pretended | -0.021 |
ts | -0.020 |
pretend | -0.020 |
Token | until |
Token ID | 1566 |
Feature activation | +6.91e-02 |
Pos logit contributions | |
eal | +0.141 |
arre | +0.137 |
irk | +0.129 |
umpy | +0.129 |
ags | +0.123 |
Neg logit contributions | |
cardboard | -0.076 |
lf | -0.073 |
ts | -0.067 |
camps | -0.063 |
trek | -0.061 |
Token | he |
Token ID | 339 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
cardboard | +0.078 |
lf | +0.076 |
ts | +0.069 |
camps | +0.065 |
blasted | +0.062 |
Neg logit contributions | |
eal | -0.153 |
arre | -0.148 |
irk | -0.139 |
umpy | -0.139 |
vious | -0.134 |
Token | was |
Token ID | 373 |
Feature activation | +2.96e-02 |
Pos logit contributions | |
cardboard | +0.021 |
lf | +0.020 |
ts | +0.018 |
camps | +0.017 |
shovel | +0.016 |
Neg logit contributions | |
eal | -0.035 |
arre | -0.034 |
umpy | -0.032 |
irk | -0.031 |
vious | -0.030 |
Token | |
Token ID | 220 |
Feature activation | -9.28e-02 |
Pos logit contributions | |
eal | +0.056 |
arre | +0.053 |
umpy | +0.049 |
irk | +0.049 |
ags | +0.046 |
Neg logit contributions | |
cardboard | -0.064 |
lf | -0.063 |
ts | -0.059 |
camps | -0.057 |
blasted | -0.055 |
Token |
|
Token ID | 198 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
eal | +0.217 |
arre | +0.196 |
irk | +0.192 |
vious | +0.192 |
umpy | +0.183 |
Neg logit contributions | |
cardboard | -0.095 |
lf | -0.091 |
ts | -0.079 |
camps | -0.076 |
called | -0.074 |
Token |
|
Token ID | 198 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
eal | +0.028 |
arre | +0.027 |
irk | +0.025 |
vious | +0.025 |
umpy | +0.025 |
Neg logit contributions | |
lf | -0.018 |
cardboard | -0.018 |
ts | -0.016 |
camps | -0.016 |
trek | -0.015 |
Token | J |
Token ID | 41 |
Feature activation | +1.91e-03 |
Pos logit contributions | |
eal | +0.028 |
arre | +0.026 |
irk | +0.025 |
umpy | +0.024 |
ags | +0.024 |
Neg logit contributions | |
lf | -0.019 |
cardboard | -0.018 |
ts | -0.016 |
camps | -0.015 |
trek | -0.015 |
Token | enny |
Token ID | 11870 |
Feature activation | +4.00e-02 |
Pos logit contributions | |
lf | +0.003 |
cardboard | +0.003 |
ts | +0.003 |
blasted | +0.003 |
trek | +0.003 |
Neg logit contributions | |
eal | -0.003 |
arre | -0.003 |
agree | -0.003 |
ags | -0.003 |
Fighting | -0.003 |
Token | 's |
Token ID | 338 |
Feature activation | -7.02e-02 |
Pos logit contributions | |
lf | +0.046 |
cardboard | +0.044 |
ts | +0.040 |
pretended | +0.038 |
trek | +0.037 |
Neg logit contributions | |
eal | -0.087 |
arre | -0.086 |
irk | -0.083 |
umpy | -0.082 |
ags | -0.082 |
Token | friends |
Token ID | 2460 |
Feature activation | +2.48e-02 |
Pos logit contributions | |
eal | +0.117 |
arre | +0.113 |
umpy | +0.105 |
irk | +0.104 |
ags | +0.098 |
Neg logit contributions | |
cardboard | -0.120 |
lf | -0.118 |
ts | -0.110 |
camps | -0.106 |
blasted | -0.103 |
Token | were |
Token ID | 547 |
Feature activation | +5.22e-02 |
Pos logit contributions | |
cardboard | +0.024 |
lf | +0.023 |
ts | +0.021 |
pretended | +0.019 |
blasted | +0.019 |
Neg logit contributions | |
eal | -0.058 |
arre | -0.056 |
umpy | -0.055 |
irk | -0.055 |
ags | -0.054 |
Token | very |
Token ID | 845 |
Feature activation | -3.29e-02 |
Pos logit contributions | |
called | +0.070 |
cardboard | +0.070 |
lf | +0.068 |
named | +0.064 |
ts | +0.064 |
Neg logit contributions | |
eal | -0.106 |
irk | -0.101 |
awk | -0.096 |
arre | -0.095 |
agree | -0.093 |
Token | excited |
Token ID | 6568 |
Feature activation | +6.00e-04 |
Pos logit contributions | |
eal | +0.052 |
arre | +0.050 |
ags | +0.050 |
irk | +0.047 |
agree | +0.047 |
Neg logit contributions | |
cardboard | -0.054 |
lf | -0.054 |
ts | -0.052 |
called | -0.052 |
pretend | -0.050 |
Token | when |
Token ID | 618 |
Feature activation | +6.47e-02 |
Pos logit contributions | |
lf | +0.001 |
cardboard | +0.001 |
ts | +0.001 |
pretend | +0.001 |
camps | +0.001 |
Neg logit contributions | |
eal | -0.001 |
arre | -0.001 |
umpy | -0.001 |
irk | -0.001 |
ags | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
eal | +0.001 |
arre | +0.001 |
umpy | +0.001 |
irk | +0.001 |
ags | +0.001 |
Neg logit contributions | |
cardboard | -0.001 |
lf | -0.001 |
ts | -0.001 |
camps | -0.001 |
blasted | -0.001 |
Token | She |
Token ID | 1375 |
Feature activation | +5.79e-02 |
Pos logit contributions | |
eal | +0.044 |
arre | +0.044 |
awk | +0.041 |
ags | +0.040 |
umpy | +0.040 |
Neg logit contributions | |
cardboard | -0.036 |
lf | -0.034 |
shovel | -0.032 |
camps | -0.031 |
called | -0.031 |
Token | thought |
Token ID | 1807 |
Feature activation | -8.34e-03 |
Pos logit contributions | |
cardboard | +0.078 |
lf | +0.075 |
ts | +0.074 |
camps | +0.069 |
fossils | +0.064 |
Neg logit contributions | |
eal | -0.120 |
arre | -0.113 |
irk | -0.105 |
umpy | -0.104 |
back | -0.102 |
Token | for |
Token ID | 329 |
Feature activation | -6.37e-02 |
Pos logit contributions | |
eal | +0.013 |
arre | +0.012 |
irk | +0.011 |
umpy | +0.011 |
back | +0.011 |
Neg logit contributions | |
lf | -0.014 |
cardboard | -0.014 |
camps | -0.013 |
ts | -0.013 |
blasted | -0.013 |
Token | a |
Token ID | 257 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
back | +0.102 |
aback | +0.097 |
firm | +0.096 |
eal | +0.095 |
agree | +0.093 |
Neg logit contributions | |
camps | -0.097 |
lf | -0.092 |
blasted | -0.088 |
cardboard | -0.084 |
trek | -0.082 |
Token | moment |
Token ID | 2589 |
Feature activation | -6.65e-02 |
Pos logit contributions | |
lf | +0.329 |
cardboard | +0.327 |
ts | +0.318 |
camps | +0.303 |
blasted | +0.296 |
Neg logit contributions | |
eal | -0.193 |
arre | -0.184 |
irk | -0.166 |
umpy | -0.164 |
agree | -0.149 |
Token | and |
Token ID | 290 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
eal | +0.142 |
arre | +0.138 |
umpy | +0.129 |
irk | +0.129 |
ags | +0.124 |
Neg logit contributions | |
cardboard | -0.084 |
lf | -0.083 |
ts | -0.074 |
camps | -0.070 |
trek | -0.068 |
Token | smiled |
Token ID | 13541 |
Feature activation | -2.26e-02 |
Pos logit contributions | |
eal | +0.029 |
arre | +0.028 |
back | +0.026 |
irk | +0.026 |
disappointed | +0.026 |
Neg logit contributions | |
cardboard | -0.018 |
lf | -0.017 |
ts | -0.017 |
camps | -0.017 |
trek | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -2.35e-02 |
Pos logit contributions | |
arre | +0.056 |
eal | +0.055 |
umpy | +0.052 |
awk | +0.051 |
Fighting | +0.051 |
Neg logit contributions | |
cardboard | -0.020 |
pretend | -0.017 |
pretended | -0.017 |
lf | -0.016 |
camps | -0.016 |
Token | |
Token ID | 220 |
Feature activation | -9.69e-02 |
Pos logit contributions | |
eal | +0.043 |
arre | +0.042 |
umpy | +0.039 |
irk | +0.039 |
awk | +0.038 |
Neg logit contributions | |
cardboard | -0.036 |
lf | -0.034 |
camps | -0.031 |
shovel | -0.031 |
trek | -0.030 |
Token |
|
Token ID | 198 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
eal | +0.207 |
arre | +0.190 |
irk | +0.186 |
vious | +0.182 |
umpy | +0.181 |
Neg logit contributions | |
cardboard | -0.121 |
lf | -0.113 |
ts | -0.105 |
camps | -0.100 |
trek | -0.099 |
Token | , |
Token ID | 11 |
Feature activation | +5.05e-02 |
Pos logit contributions | |
eal | +0.219 |
arre | +0.209 |
umpy | +0.197 |
irk | +0.194 |
ealing | +0.192 |
Neg logit contributions | |
cardboard | -0.100 |
lf | -0.100 |
pretend | -0.085 |
ts | -0.080 |
pretended | -0.079 |
Token | but |
Token ID | 475 |
Feature activation | -6.60e-03 |
Pos logit contributions | |
cardboard | +0.043 |
lf | +0.037 |
pretend | +0.037 |
pretended | +0.036 |
called | +0.036 |
Neg logit contributions | |
eal | -0.131 |
arre | -0.129 |
umpy | -0.122 |
ealing | -0.118 |
arling | -0.117 |
Token | she |
Token ID | 673 |
Feature activation | -5.49e-03 |
Pos logit contributions | |
eal | +0.014 |
arre | +0.014 |
umpy | +0.013 |
irk | +0.013 |
awk | +0.013 |
Neg logit contributions | |
lf | -0.009 |
cardboard | -0.008 |
ts | -0.008 |
pretend | -0.007 |
pretended | -0.007 |
Token | made |
Token ID | 925 |
Feature activation | -2.19e-02 |
Pos logit contributions | |
eal | +0.012 |
arre | +0.011 |
umpy | +0.011 |
irk | +0.011 |
vious | +0.010 |
Neg logit contributions | |
cardboard | -0.007 |
lf | -0.007 |
ts | -0.006 |
camps | -0.006 |
blasted | -0.006 |
Token | her |
Token ID | 607 |
Feature activation | -7.25e-02 |
Pos logit contributions | |
eal | +0.053 |
arre | +0.052 |
umpy | +0.051 |
irk | +0.050 |
awk | +0.047 |
Neg logit contributions | |
cardboard | -0.022 |
lf | -0.020 |
ts | -0.019 |
pretend | -0.018 |
camps | -0.016 |
Token | unhappy |
Token ID | 19283 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
eal | +0.154 |
arre | +0.151 |
umpy | +0.142 |
arling | +0.138 |
irk | +0.137 |
Neg logit contributions | |
cardboard | -0.098 |
lf | -0.091 |
pretend | -0.089 |
pretended | -0.079 |
camps | -0.078 |
Token | . |
Token ID | 13 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
eal | +0.337 |
arre | +0.322 |
umpy | +0.301 |
ealing | +0.299 |
awk | +0.299 |
Neg logit contributions | |
lf | -0.122 |
cardboard | -0.107 |
pretend | -0.097 |
called | -0.087 |
ts | -0.086 |
Token | She |
Token ID | 1375 |
Feature activation | +5.11e-02 |
Pos logit contributions | |
eal | +0.054 |
arre | +0.052 |
umpy | +0.046 |
irk | +0.046 |
awk | +0.045 |
Neg logit contributions | |
cardboard | -0.057 |
lf | -0.055 |
pretend | -0.051 |
camps | -0.050 |
ts | -0.049 |
Token | said |
Token ID | 531 |
Feature activation | -2.82e-02 |
Pos logit contributions | |
cardboard | +0.083 |
lf | +0.079 |
camps | +0.074 |
ts | +0.072 |
fossils | +0.069 |
Neg logit contributions | |
eal | -0.093 |
arre | -0.085 |
irk | -0.080 |
disappointed | -0.079 |
vious | -0.079 |
Token | she |
Token ID | 673 |
Feature activation | -5.53e-03 |
Pos logit contributions | |
eal | +0.068 |
arre | +0.066 |
umpy | +0.064 |
irk | +0.062 |
ags | +0.061 |
Neg logit contributions | |
cardboard | -0.032 |
lf | -0.032 |
ts | -0.027 |
pretend | -0.025 |
jewels | -0.023 |
Token | was |
Token ID | 373 |
Feature activation | +2.56e-02 |
Pos logit contributions | |
eal | +0.012 |
arre | +0.012 |
umpy | +0.011 |
irk | +0.011 |
vious | +0.011 |
Neg logit contributions | |
cardboard | -0.007 |
lf | -0.007 |
ts | -0.006 |
camps | -0.006 |
trek | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
eal | +0.024 |
back | +0.023 |
arre | +0.022 |
irk | +0.022 |
umpy | +0.022 |
Neg logit contributions | |
lf | -0.022 |
cardboard | -0.020 |
ts | -0.019 |
camps | -0.019 |
blasted | -0.018 |
Token | carrot |
Token ID | 40562 |
Feature activation | +4.95e-02 |
Pos logit contributions | |
lf | +0.227 |
ts | +0.218 |
camps | +0.201 |
cardboard | +0.200 |
trek | +0.194 |
Neg logit contributions | |
eal | -0.265 |
back | -0.255 |
irk | -0.254 |
umpy | -0.247 |
arre | -0.246 |
Token | . |
Token ID | 13 |
Feature activation | -2.25e-02 |
Pos logit contributions | |
lf | +0.078 |
cardboard | +0.076 |
ts | +0.074 |
trek | +0.069 |
camps | +0.068 |
Neg logit contributions | |
eal | -0.084 |
arre | -0.081 |
back | -0.077 |
umpy | -0.077 |
irk | -0.076 |
Token | He |
Token ID | 679 |
Feature activation | +8.05e-02 |
Pos logit contributions | |
eal | +0.033 |
arre | +0.031 |
irk | +0.029 |
umpy | +0.029 |
ags | +0.027 |
Neg logit contributions | |
lf | -0.042 |
cardboard | -0.042 |
ts | -0.040 |
camps | -0.038 |
blasted | -0.038 |
Token | gives |
Token ID | 3607 |
Feature activation | -4.19e-02 |
Pos logit contributions | |
lf | +0.159 |
cardboard | +0.147 |
ts | +0.138 |
hours | +0.136 |
trek | +0.133 |
Neg logit contributions | |
eal | -0.110 |
arre | -0.102 |
vious | -0.094 |
disappointed | -0.093 |
irk | -0.090 |
Token | it |
Token ID | 340 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
eal | +0.072 |
arre | +0.066 |
irk | +0.063 |
back | +0.062 |
ags | +0.061 |
Neg logit contributions | |
lf | -0.069 |
cardboard | -0.067 |
ts | -0.061 |
camps | -0.061 |
blasted | -0.060 |
Token | to |
Token ID | 284 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
eal | +0.222 |
arre | +0.216 |
umpy | +0.200 |
irk | +0.200 |
ags | +0.188 |
Neg logit contributions | |
cardboard | -0.208 |
lf | -0.201 |
ts | -0.189 |
camps | -0.182 |
blasted | -0.177 |
Token | L |
Token ID | 406 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
eal | +0.022 |
arre | +0.021 |
irk | +0.020 |
umpy | +0.019 |
ags | +0.019 |
Neg logit contributions | |
cardboard | -0.017 |
lf | -0.017 |
ts | -0.016 |
camps | -0.015 |
blasted | -0.015 |
Token | ila |
Token ID | 10102 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
eal | +0.029 |
arre | +0.028 |
irk | +0.026 |
umpy | +0.026 |
agree | +0.025 |
Neg logit contributions | |
lf | -0.032 |
cardboard | -0.032 |
ts | -0.030 |
camps | -0.028 |
trek | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -2.62e-02 |
Pos logit contributions | |
eal | +0.026 |
arre | +0.026 |
irk | +0.024 |
umpy | +0.024 |
agree | +0.023 |
Neg logit contributions | |
cardboard | -0.014 |
lf | -0.013 |
ts | -0.012 |
camps | -0.011 |
pretend | -0.011 |
Token | He |
Token ID | 679 |
Feature activation | +8.00e-02 |
Pos logit contributions | |
eal | +0.044 |
arre | +0.043 |
umpy | +0.039 |
irk | +0.039 |
ags | +0.037 |
Neg logit contributions | |
cardboard | -0.045 |
lf | -0.044 |
ts | -0.041 |
camps | -0.039 |
blasted | -0.038 |
Token | hurt |
Token ID | 5938 |
Feature activation | -6.87e-02 |
Pos logit contributions | |
eal | +0.055 |
arre | +0.053 |
irk | +0.050 |
umpy | +0.050 |
ags | +0.047 |
Neg logit contributions | |
cardboard | -0.041 |
lf | -0.039 |
ts | -0.037 |
pretend | -0.034 |
camps | -0.034 |
Token | your |
Token ID | 534 |
Feature activation | -9.84e-02 |
Pos logit contributions | |
eal | +0.130 |
arre | +0.125 |
irk | +0.115 |
umpy | +0.113 |
ags | +0.108 |
Neg logit contributions | |
cardboard | -0.106 |
lf | -0.103 |
ts | -0.098 |
pretend | -0.088 |
camps | -0.088 |
Token | lip |
Token ID | 10645 |
Feature activation | -5.69e-02 |
Pos logit contributions | |
eal | +0.149 |
arre | +0.146 |
umpy | +0.132 |
irk | +0.131 |
vious | +0.123 |
Neg logit contributions | |
cardboard | -0.187 |
lf | -0.182 |
ts | -0.172 |
camps | -0.164 |
pretend | -0.163 |
Token | . |
Token ID | 13 |
Feature activation | -3.68e-02 |
Pos logit contributions | |
eal | +0.124 |
arre | +0.117 |
irk | +0.114 |
umpy | +0.108 |
ags | +0.107 |
Neg logit contributions | |
cardboard | -0.072 |
lf | -0.071 |
ts | -0.065 |
pretend | -0.059 |
camps | -0.058 |
Token | Or |
Token ID | 1471 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
eal | +0.053 |
arre | +0.051 |
irk | +0.046 |
umpy | +0.046 |
ags | +0.042 |
Neg logit contributions | |
cardboard | -0.071 |
lf | -0.069 |
ts | -0.065 |
camps | -0.063 |
blasted | -0.062 |
Token | it |
Token ID | 340 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
lf | +0.025 |
cardboard | +0.025 |
camps | +0.023 |
ts | +0.022 |
trek | +0.022 |
Neg logit contributions | |
eal | -0.033 |
arre | -0.032 |
irk | -0.030 |
umpy | -0.030 |
vious | -0.029 |
Token | might |
Token ID | 1244 |
Feature activation | -3.61e-03 |
Pos logit contributions | |
eal | +0.252 |
irk | +0.239 |
arre | +0.233 |
umpy | +0.219 |
ags | +0.219 |
Neg logit contributions | |
cardboard | -0.178 |
ts | -0.160 |
lf | -0.159 |
pretend | -0.154 |
blasted | -0.148 |
Token | stick |
Token ID | 4859 |
Feature activation | +3.04e-02 |
Pos logit contributions | |
eal | +0.007 |
arre | +0.007 |
irk | +0.006 |
umpy | +0.006 |
ags | +0.006 |
Neg logit contributions | |
cardboard | -0.006 |
lf | -0.005 |
ts | -0.005 |
camps | -0.005 |
blasted | -0.005 |
Token | to |
Token ID | 284 |
Feature activation | -7.17e-03 |
Pos logit contributions | |
cardboard | +0.046 |
lf | +0.041 |
ts | +0.040 |
pretend | +0.038 |
camps | +0.036 |
Neg logit contributions | |
eal | -0.061 |
arre | -0.057 |
irk | -0.056 |
umpy | -0.053 |
Fighting | -0.050 |
Token | something |
Token ID | 1223 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
eal | +0.014 |
arre | +0.013 |
umpy | +0.013 |
irk | +0.013 |
vious | +0.012 |
Neg logit contributions | |
cardboard | -0.011 |
lf | -0.010 |
ts | -0.010 |
camps | -0.009 |
pretend | -0.009 |
Token | else |
Token ID | 2073 |
Feature activation | -3.85e-03 |
Pos logit contributions | |
cardboard | +0.020 |
lf | +0.019 |
ts | +0.018 |
camps | +0.017 |
blasted | +0.016 |
Neg logit contributions | |
eal | -0.036 |
arre | -0.035 |
irk | -0.033 |
umpy | -0.033 |
ags | -0.031 |
Token | bitter |
Token ID | 12922 |
Feature activation | -9.13e-02 |
Pos logit contributions | |
eal | +0.150 |
arre | +0.140 |
irk | +0.138 |
umpy | +0.132 |
vious | +0.124 |
Neg logit contributions | |
cardboard | -0.124 |
ts | -0.115 |
lf | -0.115 |
camps | -0.107 |
pretend | -0.106 |
Token | . |
Token ID | 13 |
Feature activation | -4.10e-02 |
Pos logit contributions | |
eal | +0.202 |
arre | +0.193 |
irk | +0.185 |
umpy | +0.179 |
Fighting | +0.177 |
Neg logit contributions | |
cardboard | -0.106 |
lf | -0.099 |
ts | -0.093 |
camps | -0.088 |
pretend | -0.087 |
Token | It |
Token ID | 632 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
eal | +0.064 |
arre | +0.061 |
irk | +0.056 |
umpy | +0.056 |
ags | +0.052 |
Neg logit contributions | |
cardboard | -0.074 |
lf | -0.072 |
ts | -0.067 |
camps | -0.066 |
blasted | -0.064 |
Token | 's |
Token ID | 338 |
Feature activation | -7.82e-02 |
Pos logit contributions | |
eal | +0.030 |
irk | +0.029 |
ags | +0.027 |
arre | +0.027 |
umpy | +0.026 |
Neg logit contributions | |
cardboard | -0.017 |
pretend | -0.016 |
matches | -0.015 |
ts | -0.015 |
pretended | -0.015 |
Token | worth |
Token ID | 2861 |
Feature activation | -5.44e-02 |
Pos logit contributions | |
eal | +0.150 |
arre | +0.140 |
irk | +0.137 |
umpy | +0.132 |
vious | +0.124 |
Neg logit contributions | |
cardboard | -0.124 |
lf | -0.114 |
ts | -0.113 |
camps | -0.106 |
pretend | -0.105 |
Token | it |
Token ID | 340 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
eal | +0.105 |
arre | +0.102 |
umpy | +0.097 |
irk | +0.096 |
ags | +0.093 |
Neg logit contributions | |
cardboard | -0.075 |
lf | -0.075 |
camps | -0.067 |
ts | -0.065 |
blasted | -0.064 |
Token | ," |
Token ID | 553 |
Feature activation | +2.84e-02 |
Pos logit contributions | |
eal | +0.267 |
arre | +0.257 |
irk | +0.248 |
umpy | +0.234 |
Fighting | +0.229 |
Neg logit contributions | |
cardboard | -0.159 |
lf | -0.152 |
ts | -0.143 |
camps | -0.131 |
pretend | -0.130 |
Token | the |
Token ID | 262 |
Feature activation | -4.74e-02 |
Pos logit contributions | |
cardboard | +0.031 |
lf | +0.026 |
ts | +0.025 |
pretend | +0.023 |
called | +0.022 |
Neg logit contributions | |
eal | -0.071 |
arre | -0.070 |
umpy | -0.068 |
irk | -0.062 |
ags | -0.062 |
Token | doctor |
Token ID | 6253 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
eal | +0.073 |
arre | +0.072 |
irk | +0.064 |
umpy | +0.063 |
agree | +0.061 |
Neg logit contributions | |
cardboard | -0.089 |
lf | -0.088 |
ts | -0.082 |
camps | -0.079 |
pretend | -0.078 |
Token | said |
Token ID | 531 |
Feature activation | -3.04e-02 |
Pos logit contributions | |
eal | +0.040 |
arre | +0.040 |
umpy | +0.038 |
irk | +0.037 |
ags | +0.037 |
Neg logit contributions | |
lf | -0.023 |
cardboard | -0.023 |
ts | -0.021 |
pretended | -0.020 |
blasted | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -3.72e-02 |
Pos logit contributions | |
arre | +0.076 |
eal | +0.074 |
umpy | +0.069 |
Fighting | +0.068 |
agree | +0.067 |
Neg logit contributions | |
cardboard | -0.027 |
lf | -0.026 |
ts | -0.025 |
pretend | -0.020 |
called | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | -1.75e-02 |
Pos logit contributions | |
ts | +0.077 |
lf | +0.077 |
cardboard | +0.077 |
trek | +0.074 |
blasted | +0.072 |
Neg logit contributions | |
eal | -0.092 |
arre | -0.090 |
back | -0.089 |
irk | -0.087 |
umpy | -0.086 |
Token | she |
Token ID | 673 |
Feature activation | +6.07e-04 |
Pos logit contributions | |
eal | +0.027 |
disappointed | +0.025 |
mad | +0.024 |
arre | +0.024 |
vious | +0.023 |
Neg logit contributions | |
camps | -0.029 |
cardboard | -0.029 |
lf | -0.028 |
ts | -0.027 |
blasted | -0.027 |
Token | wanted |
Token ID | 2227 |
Feature activation | +4.37e-02 |
Pos logit contributions | |
cardboard | +0.001 |
lf | +0.001 |
camps | +0.001 |
ts | +0.001 |
blasted | +0.001 |
Neg logit contributions | |
eal | -0.001 |
arre | -0.001 |
umpy | -0.001 |
irk | -0.001 |
disappointed | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | -8.11e-04 |
Pos logit contributions | |
cardboard | +0.071 |
lf | +0.069 |
ts | +0.065 |
camps | +0.062 |
blasted | +0.060 |
Neg logit contributions | |
eal | -0.078 |
arre | -0.075 |
umpy | -0.070 |
irk | -0.070 |
ags | -0.066 |
Token | try |
Token ID | 1949 |
Feature activation | -2.90e-02 |
Pos logit contributions | |
eal | +0.002 |
arre | +0.001 |
umpy | +0.001 |
irk | +0.001 |
vious | +0.001 |
Neg logit contributions | |
cardboard | -0.001 |
lf | -0.001 |
ts | -0.001 |
camps | -0.001 |
blasted | -0.001 |
Token | it |
Token ID | 340 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
eal | +0.070 |
arre | +0.067 |
umpy | +0.065 |
irk | +0.065 |
vious | +0.062 |
Neg logit contributions | |
cardboard | -0.030 |
lf | -0.028 |
ts | -0.026 |
pretend | -0.023 |
camps | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
eal | +0.255 |
arre | +0.248 |
irk | +0.234 |
umpy | +0.232 |
vious | +0.221 |
Neg logit contributions | |
cardboard | -0.159 |
lf | -0.152 |
ts | -0.140 |
camps | -0.132 |
pretend | -0.128 |
Token |
|
Token ID | 198 |
Feature activation | -7.91e-03 |
Pos logit contributions | |
eal | +0.035 |
arre | +0.034 |
irk | +0.032 |
umpy | +0.032 |
ags | +0.031 |
Neg logit contributions | |
cardboard | -0.038 |
lf | -0.036 |
pretend | -0.034 |
trek | -0.033 |
ts | -0.033 |
Token |
|
Token ID | 198 |
Feature activation | -7.95e-03 |
Pos logit contributions | |
eal | +0.013 |
arre | +0.012 |
umpy | +0.012 |
irk | +0.011 |
ags | +0.011 |
Neg logit contributions | |
cardboard | -0.014 |
lf | -0.014 |
ts | -0.013 |
camps | -0.012 |
blasted | -0.012 |
Token | She |
Token ID | 3347 |
Feature activation | +7.99e-02 |
Pos logit contributions | |
eal | +0.013 |
arre | +0.012 |
umpy | +0.011 |
irk | +0.011 |
vious | +0.010 |
Neg logit contributions | |
cardboard | -0.014 |
lf | -0.014 |
ts | -0.013 |
camps | -0.013 |
blasted | -0.012 |
Token | reached |
Token ID | 4251 |
Feature activation | +5.91e-02 |
Pos logit contributions | |
cardboard | +0.114 |
lf | +0.104 |
ts | +0.101 |
camps | +0.100 |
fossils | +0.092 |
Neg logit contributions | |
eal | -0.160 |
arre | -0.152 |
umpy | -0.135 |
irk | -0.135 |
disappointed | -0.134 |