Token | said |
Token ID | 531 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
handed | +0.081 |
amusement | +0.066 |
vendor | +0.065 |
sandbox | +0.065 |
view | +0.064 |
Neg logit contributions | |
ches | -0.048 |
Tiny | -0.046 |
Ho | -0.040 |
Pig | -0.040 |
fl | -0.039 |
Token | the |
Token ID | 262 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
ches | +0.172 |
Tiny | +0.163 |
Ho | +0.155 |
fl | +0.154 |
Pig | +0.150 |
Neg logit contributions | |
handed | -0.250 |
sandbox | -0.195 |
vendor | -0.191 |
amusement | -0.190 |
view | -0.190 |
Token | voice |
Token ID | 3809 |
Feature activation | +2.55e-04 |
Pos logit contributions | |
ches | +0.022 |
Tiny | +0.020 |
py | +0.019 |
fl | +0.019 |
Ho | +0.019 |
Neg logit contributions | |
handed | -0.028 |
vendor | -0.021 |
sandbox | -0.021 |
Mike | -0.021 |
amusement | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
ches | +0.000 |
Tiny | +0.000 |
fl | +0.000 |
py | +0.000 |
wand | +0.000 |
Neg logit contributions | |
handed | -0.001 |
sandbox | -0.000 |
vendor | -0.000 |
Mike | -0.000 |
amusement | -0.000 |
Token | Pat |
Token ID | 3208 |
Feature activation | -2.27e-02 |
Pos logit contributions | |
ches | +0.026 |
py | +0.024 |
fl | +0.024 |
Tiny | +0.023 |
wand | +0.023 |
Neg logit contributions | |
handed | -0.031 |
Mike | -0.024 |
vendor | -0.024 |
amusement | -0.023 |
sandbox | -0.023 |
Token | ches |
Token ID | 2052 |
Feature activation | +4.80e-01 |
Pos logit contributions | |
handed | +0.047 |
Mike | +0.037 |
vendor | +0.037 |
sandbox | +0.036 |
amusement | +0.036 |
Neg logit contributions | |
ches | -0.036 |
Tiny | -0.033 |
py | -0.032 |
fl | -0.031 |
Ho | -0.031 |
Token | saw |
Token ID | 2497 |
Feature activation | -9.28e-02 |
Pos logit contributions | |
ches | +0.767 |
fl | +0.699 |
Tiny | +0.683 |
py | +0.670 |
Ho | +0.637 |
Neg logit contributions | |
handed | -1.011 |
vendor | -0.791 |
sandbox | -0.790 |
amusement | -0.774 |
Mike | -0.766 |
Token | that |
Token ID | 326 |
Feature activation | -8.97e-02 |
Pos logit contributions | |
handed | +0.202 |
sandbox | +0.164 |
vendor | +0.162 |
apartment | +0.161 |
unatt | +0.160 |
Neg logit contributions | |
Tiny | -0.132 |
ches | -0.132 |
fl | -0.127 |
Ho | -0.125 |
Pig | -0.119 |
Token | it |
Token ID | 340 |
Feature activation | +3.46e-02 |
Pos logit contributions | |
handed | +0.210 |
sandbox | +0.163 |
view | +0.162 |
Mike | +0.160 |
vendor | +0.159 |
Neg logit contributions | |
Tiny | -0.122 |
ches | -0.119 |
fl | -0.116 |
Ho | -0.113 |
Sparkle | -0.111 |
Token | was |
Token ID | 373 |
Feature activation | -6.29e-02 |
Pos logit contributions | |
ches | +0.062 |
Tiny | +0.057 |
fl | +0.056 |
py | +0.055 |
Ho | +0.054 |
Neg logit contributions | |
handed | -0.067 |
vendor | -0.051 |
Mike | -0.050 |
sandbox | -0.050 |
amusement | -0.050 |
Token | a |
Token ID | 257 |
Feature activation | +8.91e-03 |
Pos logit contributions | |
handed | +0.129 |
amusement | +0.101 |
vendor | +0.101 |
Mike | +0.101 |
sandbox | +0.099 |
Neg logit contributions | |
ches | -0.101 |
Tiny | -0.091 |
py | -0.089 |
Ho | -0.086 |
fl | -0.086 |
Token | is |
Token ID | 318 |
Feature activation | -2.52e-02 |
Pos logit contributions | |
ches | +0.202 |
Tiny | +0.190 |
py | +0.183 |
Ho | +0.176 |
Cotton | +0.176 |
Neg logit contributions | |
handed | -0.281 |
sandbox | -0.219 |
vendor | -0.219 |
Mike | -0.217 |
amusement | -0.214 |
Token | angry |
Token ID | 7954 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
handed | +0.042 |
vendor | +0.032 |
amusement | +0.031 |
unatt | +0.031 |
sandbox | +0.030 |
Neg logit contributions | |
ches | -0.047 |
Tiny | -0.044 |
Chick | -0.043 |
py | -0.043 |
Ho | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | +2.56e-02 |
Pos logit contributions | |
handed | +0.266 |
Mike | +0.203 |
vendor | +0.201 |
sandbox | +0.198 |
amusement | +0.197 |
Neg logit contributions | |
ches | -0.246 |
Tiny | -0.228 |
py | -0.221 |
fl | -0.219 |
Cotton | -0.218 |
Token | It |
Token ID | 632 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
ches | +0.039 |
Tiny | +0.038 |
py | +0.035 |
Ho | +0.035 |
fl | +0.035 |
Neg logit contributions | |
handed | -0.055 |
sandbox | -0.043 |
vendor | -0.043 |
Mike | -0.042 |
amusement | -0.042 |
Token | scree |
Token ID | 43465 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
Tiny | +0.024 |
Pig | +0.024 |
Cotton | +0.024 |
Chick | +0.023 |
iny | +0.022 |
Neg logit contributions | |
handed | -0.024 |
sandbox | -0.020 |
aw | -0.020 |
vendor | -0.019 |
lobby | -0.018 |
Token | ches |
Token ID | 2052 |
Feature activation | +4.78e-01 |
Pos logit contributions | |
Cotton | +0.115 |
Pig | +0.112 |
basil | +0.111 |
cabbage | +0.110 |
Sparkle | +0.108 |
Neg logit contributions | |
handed | -0.255 |
aw | -0.229 |
he | -0.212 |
anymore | -0.211 |
Mike | -0.209 |
Token | . |
Token ID | 13 |
Feature activation | +2.58e-02 |
Pos logit contributions | |
ches | +0.726 |
Tiny | +0.648 |
py | +0.623 |
fl | +0.620 |
Ho | +0.595 |
Neg logit contributions | |
handed | -1.067 |
sandbox | -0.867 |
vendor | -0.860 |
Mike | -0.850 |
amusement | -0.845 |
Token | It |
Token ID | 632 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
ches | +0.041 |
Tiny | +0.039 |
py | +0.037 |
Ho | +0.037 |
fl | +0.037 |
Neg logit contributions | |
handed | -0.054 |
vendor | -0.042 |
sandbox | -0.042 |
Mike | -0.041 |
amusement | -0.041 |
Token | fl |
Token ID | 781 |
Feature activation | +4.10e-01 |
Pos logit contributions | |
Tiny | +0.020 |
Pig | +0.020 |
Cotton | +0.019 |
ches | +0.019 |
py | +0.019 |
Neg logit contributions | |
handed | -0.021 |
sandbox | -0.018 |
aw | -0.017 |
vendor | -0.016 |
lobby | -0.016 |
Token | aps |
Token ID | 1686 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
ches | +0.767 |
Tiny | +0.712 |
py | +0.697 |
fl | +0.689 |
Ho | +0.678 |
Neg logit contributions | |
handed | -0.751 |
vendor | -0.556 |
Mike | -0.555 |
sandbox | -0.552 |
amusement | -0.546 |
Token | its |
Token ID | 663 |
Feature activation | +8.63e-02 |
Pos logit contributions | |
handed | +0.289 |
vendor | +0.228 |
Mike | +0.228 |
sandbox | +0.227 |
amusement | +0.225 |
Neg logit contributions | |
ches | -0.185 |
Tiny | -0.168 |
py | -0.163 |
fl | -0.161 |
Ho | -0.157 |
Token | ost |
Token ID | 455 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
ches | +0.543 |
Tiny | +0.520 |
py | +0.505 |
fl | +0.497 |
Ho | +0.493 |
Neg logit contributions | |
handed | -0.669 |
Mike | -0.513 |
vendor | -0.500 |
amusement | -0.488 |
sandbox | -0.487 |
Token | ing |
Token ID | 278 |
Feature activation | +4.47e-02 |
Pos logit contributions | |
ches | +0.157 |
Tiny | +0.137 |
py | +0.135 |
fl | +0.134 |
Ho | +0.126 |
Neg logit contributions | |
handed | -0.253 |
sandbox | -0.206 |
vendor | -0.205 |
amusement | -0.203 |
Mike | -0.202 |
Token | on |
Token ID | 319 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
ches | +0.065 |
Tiny | +0.058 |
fl | +0.058 |
py | +0.057 |
Ho | +0.054 |
Neg logit contributions | |
handed | -0.101 |
vendor | -0.081 |
Mike | -0.080 |
sandbox | -0.080 |
amusement | -0.079 |
Token | their |
Token ID | 511 |
Feature activation | -6.85e-02 |
Pos logit contributions | |
ches | +0.259 |
Tiny | +0.243 |
fl | +0.235 |
py | +0.229 |
Ho | +0.225 |
Neg logit contributions | |
handed | -0.383 |
vendor | -0.302 |
sandbox | -0.301 |
Mike | -0.300 |
amusement | -0.292 |
Token | per |
Token ID | 583 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
handed | +0.153 |
Mike | +0.121 |
vendor | +0.119 |
sandbox | +0.118 |
amusement | +0.117 |
Neg logit contributions | |
ches | -0.101 |
Tiny | -0.092 |
fl | -0.090 |
py | -0.089 |
Ho | -0.086 |
Token | ches |
Token ID | 2052 |
Feature activation | +4.74e-01 |
Pos logit contributions | |
Cotton | +0.106 |
py | +0.104 |
Sparkle | +0.103 |
cabbage | +0.102 |
basil | +0.101 |
Neg logit contributions | |
handed | -0.301 |
aw | -0.280 |
anymore | -0.247 |
out | -0.243 |
unatt | -0.243 |
Token | . |
Token ID | 13 |
Feature activation | +8.59e-03 |
Pos logit contributions | |
ches | +0.732 |
Tiny | +0.644 |
fl | +0.633 |
py | +0.620 |
Ho | +0.594 |
Neg logit contributions | |
handed | -1.041 |
sandbox | -0.836 |
vendor | -0.835 |
Mike | -0.831 |
amusement | -0.820 |
Token | They |
Token ID | 1119 |
Feature activation | +5.19e-02 |
Pos logit contributions | |
ches | +0.014 |
py | +0.013 |
fl | +0.013 |
Tiny | +0.013 |
wand | +0.012 |
Neg logit contributions | |
handed | -0.017 |
Mike | -0.014 |
vendor | -0.014 |
sandbox | -0.013 |
amusement | -0.013 |
Token | were |
Token ID | 547 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
ches | +0.081 |
Tiny | +0.075 |
py | +0.073 |
Ho | +0.071 |
Cotton | +0.071 |
Neg logit contributions | |
handed | -0.107 |
vendor | -0.084 |
Mike | -0.084 |
sandbox | -0.084 |
amusement | -0.083 |
Token | all |
Token ID | 477 |
Feature activation | +9.13e-02 |
Pos logit contributions | |
handed | +0.372 |
Mike | +0.292 |
vendor | +0.291 |
sandbox | +0.288 |
amusement | +0.285 |
Neg logit contributions | |
ches | -0.380 |
Tiny | -0.346 |
py | -0.338 |
Cotton | -0.337 |
Ho | -0.335 |
Token | very |
Token ID | 845 |
Feature activation | -2.78e-02 |
Pos logit contributions | |
ches | +0.149 |
Tiny | +0.137 |
py | +0.134 |
fl | +0.132 |
Ho | +0.130 |
Neg logit contributions | |
handed | -0.188 |
vendor | -0.144 |
Mike | -0.144 |
sandbox | -0.143 |
amusement | -0.142 |
Token | new |
Token ID | 649 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
fl | +0.085 |
ches | +0.084 |
Tiny | +0.083 |
Ho | +0.078 |
wool | +0.075 |
Neg logit contributions | |
handed | -0.153 |
Mike | -0.123 |
apartment | -0.119 |
he | -0.119 |
lobby | -0.118 |
Token | dresses |
Token ID | 27309 |
Feature activation | +7.70e-02 |
Pos logit contributions | |
fl | +0.035 |
wand | +0.032 |
wool | +0.031 |
ches | +0.031 |
Tiny | +0.030 |
Neg logit contributions | |
handed | -0.067 |
Mike | -0.053 |
he | -0.051 |
Jane | -0.050 |
view | -0.049 |
Token | and |
Token ID | 290 |
Feature activation | +3.65e-02 |
Pos logit contributions | |
ches | +0.095 |
fl | +0.083 |
wand | +0.082 |
Tiny | +0.079 |
Sparkle | +0.078 |
Neg logit contributions | |
handed | -0.196 |
Mike | -0.161 |
sandbox | -0.156 |
vendor | -0.155 |
amusement | -0.154 |
Token | no |
Token ID | 645 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
ches | +0.059 |
Tiny | +0.054 |
py | +0.052 |
fl | +0.052 |
Ho | +0.051 |
Neg logit contributions | |
handed | -0.076 |
Mike | -0.059 |
vendor | -0.059 |
sandbox | -0.058 |
amusement | -0.058 |
Token | it |
Token ID | 340 |
Feature activation | +4.34e-02 |
Pos logit contributions | |
handed | +0.261 |
sandbox | +0.208 |
vendor | +0.206 |
amusement | +0.205 |
Mike | +0.203 |
Neg logit contributions | |
ches | -0.198 |
Tiny | -0.179 |
py | -0.175 |
fl | -0.167 |
Ho | -0.167 |
Token | ches |
Token ID | 2052 |
Feature activation | +4.73e-01 |
Pos logit contributions | |
ches | +0.067 |
Tiny | +0.059 |
fl | +0.058 |
py | +0.058 |
Ho | +0.056 |
Neg logit contributions | |
handed | -0.095 |
Mike | -0.076 |
vendor | -0.075 |
amusement | -0.075 |
sandbox | -0.074 |
Token | . |
Token ID | 13 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
ches | +0.754 |
fl | +0.645 |
py | +0.645 |
Tiny | +0.634 |
Sparkle | +0.609 |
Neg logit contributions | |
handed | -1.020 |
Mike | -0.833 |
sandbox | -0.829 |
amaz | -0.827 |
vendor | -0.824 |
Token | They |
Token ID | 1119 |
Feature activation | +6.16e-02 |
Pos logit contributions | |
ches | +0.031 |
py | +0.028 |
fl | +0.028 |
Tiny | +0.028 |
wand | +0.026 |
Neg logit contributions | |
handed | -0.039 |
Mike | -0.030 |
vendor | -0.030 |
sandbox | -0.030 |
amusement | -0.030 |
Token | play |
Token ID | 711 |
Feature activation | -3.26e-02 |
Pos logit contributions | |
ches | +0.108 |
fl | +0.100 |
Tiny | +0.098 |
py | +0.095 |
Ho | +0.095 |
Neg logit contributions | |
handed | -0.123 |
Mike | -0.095 |
vendor | -0.091 |
Jane | -0.089 |
amusement | -0.089 |
Token | together |
Token ID | 1978 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
handed | +0.072 |
vendor | +0.056 |
Mike | +0.056 |
aren | +0.054 |
view | +0.054 |
Neg logit contributions | |
ches | -0.051 |
fl | -0.048 |
Tiny | -0.045 |
wand | -0.045 |
py | -0.044 |
Token | until |
Token ID | 1566 |
Feature activation | +4.97e-02 |
Pos logit contributions | |
handed | +0.029 |
Mike | +0.023 |
vendor | +0.022 |
sandbox | +0.022 |
amusement | +0.022 |
Neg logit contributions | |
ches | -0.020 |
Tiny | -0.018 |
py | -0.017 |
fl | -0.017 |
Ho | -0.017 |
Token | was |
Token ID | 373 |
Feature activation | -6.56e-02 |
Pos logit contributions | |
handed | +0.039 |
vendor | +0.031 |
Mike | +0.031 |
sandbox | +0.030 |
amusement | +0.030 |
Neg logit contributions | |
ches | -0.031 |
Tiny | -0.029 |
py | -0.028 |
Ho | -0.027 |
fl | -0.027 |
Token | cl |
Token ID | 537 |
Feature activation | +8.90e-02 |
Pos logit contributions | |
handed | +0.133 |
sandbox | +0.102 |
Mike | +0.102 |
vendor | +0.102 |
amusement | +0.100 |
Neg logit contributions | |
ches | -0.108 |
Tiny | -0.100 |
py | -0.098 |
fl | -0.098 |
Ho | -0.095 |
Token | apping |
Token ID | 5912 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
ches | +0.147 |
Tiny | +0.139 |
py | +0.137 |
Ho | +0.134 |
fl | +0.133 |
Neg logit contributions | |
handed | -0.170 |
sandbox | -0.135 |
vendor | -0.134 |
amusement | -0.128 |
lobby | -0.127 |
Token | for |
Token ID | 329 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
handed | +0.245 |
sandbox | +0.192 |
vendor | +0.192 |
Mike | +0.191 |
amusement | +0.188 |
Neg logit contributions | |
ches | -0.152 |
Tiny | -0.138 |
fl | -0.135 |
py | -0.133 |
Ho | -0.127 |
Token | Pat |
Token ID | 3208 |
Feature activation | -3.18e-02 |
Pos logit contributions | |
ches | +0.261 |
Tiny | +0.257 |
fl | +0.246 |
Ho | +0.238 |
Sparkle | +0.233 |
Neg logit contributions | |
handed | -0.438 |
sandbox | -0.351 |
vendor | -0.350 |
view | -0.337 |
lobby | -0.335 |
Token | ches |
Token ID | 2052 |
Feature activation | +4.73e-01 |
Pos logit contributions | |
handed | +0.074 |
Mike | +0.059 |
vendor | +0.059 |
sandbox | +0.059 |
amusement | +0.058 |
Neg logit contributions | |
ches | -0.041 |
Tiny | -0.038 |
py | -0.037 |
fl | -0.036 |
Ho | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | +9.61e-03 |
Pos logit contributions | |
ches | +0.705 |
Tiny | +0.625 |
fl | +0.622 |
py | +0.606 |
Sparkle | +0.581 |
Neg logit contributions | |
handed | -1.059 |
sandbox | -0.849 |
vendor | -0.848 |
Mike | -0.837 |
amusement | -0.817 |
Token | Pat |
Token ID | 3208 |
Feature activation | -3.23e-02 |
Pos logit contributions | |
ches | +0.018 |
py | +0.016 |
wand | +0.016 |
fl | +0.016 |
Cotton | +0.015 |
Neg logit contributions | |
handed | -0.017 |
Mike | -0.014 |
vendor | -0.013 |
Mike | -0.013 |
sandbox | -0.013 |
Token | ches |
Token ID | 2052 |
Feature activation | +4.73e-01 |
Pos logit contributions | |
handed | +0.078 |
Mike | +0.066 |
Jane | +0.064 |
vendor | +0.063 |
aren | +0.062 |
Neg logit contributions | |
Tiny | -0.032 |
ches | -0.032 |
Cotton | -0.031 |
Ho | -0.030 |
cartridge | -0.030 |
Token | smiled |
Token ID | 13541 |
Feature activation | -5.99e-02 |
Pos logit contributions | |
ches | +0.790 |
fl | +0.727 |
Tiny | +0.704 |
py | +0.697 |
Ho | +0.663 |
Neg logit contributions | |
handed | -0.965 |
vendor | -0.748 |
sandbox | -0.743 |
amusement | -0.727 |
Mike | -0.720 |
Token | and |
Token ID | 290 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
handed | +0.140 |
Mike | +0.110 |
sandbox | +0.110 |
vendor | +0.109 |
lobby | +0.108 |
Neg logit contributions | |
ches | -0.082 |
Tiny | -0.078 |
fl | -0.077 |
py | -0.074 |
Ho | -0.073 |
Token | were |
Token ID | 547 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
ches | +0.070 |
Tiny | +0.066 |
fl | +0.064 |
py | +0.063 |
Ho | +0.062 |
Neg logit contributions | |
handed | -0.093 |
Mike | -0.072 |
vendor | -0.070 |
amusement | -0.070 |
sandbox | -0.069 |
Token | playing |
Token ID | 2712 |
Feature activation | +5.50e-02 |
Pos logit contributions | |
handed | +0.383 |
sandbox | +0.312 |
vendor | +0.303 |
lobby | +0.302 |
Mike | +0.294 |
Neg logit contributions | |
ches | -0.373 |
py | -0.332 |
Tiny | -0.329 |
iny | -0.322 |
Cotton | -0.318 |
Token | together |
Token ID | 1978 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
fl | +0.071 |
ches | +0.067 |
Tiny | +0.064 |
wand | +0.062 |
hopping | +0.061 |
Neg logit contributions | |
handed | -0.136 |
aren | -0.112 |
view | -0.111 |
vendor | -0.110 |
Mike | -0.107 |
Token | and |
Token ID | 290 |
Feature activation | +2.64e-02 |
Pos logit contributions | |
handed | +0.041 |
Mike | +0.033 |
vendor | +0.033 |
amusement | +0.032 |
sandbox | +0.032 |
Neg logit contributions | |
ches | -0.023 |
fl | -0.021 |
Tiny | -0.020 |
py | -0.019 |
wand | -0.019 |
Token | Pat |
Token ID | 3208 |
Feature activation | -2.66e-02 |
Pos logit contributions | |
ches | +0.036 |
Tiny | +0.034 |
fl | +0.033 |
Ho | +0.032 |
py | +0.031 |
Neg logit contributions | |
handed | -0.062 |
Mike | -0.048 |
view | -0.048 |
vendor | -0.048 |
amusement | -0.048 |
Token | ches |
Token ID | 2052 |
Feature activation | +4.73e-01 |
Pos logit contributions | |
handed | +0.058 |
vendor | +0.046 |
sandbox | +0.046 |
amusement | +0.045 |
lobby | +0.045 |
Neg logit contributions | |
ches | -0.042 |
Tiny | -0.037 |
fl | -0.037 |
Ho | -0.035 |
py | -0.034 |
Token | saw |
Token ID | 2497 |
Feature activation | -9.53e-02 |
Pos logit contributions | |
ches | +0.713 |
fl | +0.680 |
py | +0.618 |
Tiny | +0.611 |
hopping | +0.583 |
Neg logit contributions | |
handed | -1.014 |
vendor | -0.826 |
amaz | -0.813 |
sandbox | -0.797 |
apartment | -0.787 |
Token | a |
Token ID | 257 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
handed | +0.232 |
apartment | +0.191 |
aren | +0.188 |
vendor | +0.187 |
unatt | +0.187 |
Neg logit contributions | |
Tiny | -0.119 |
fl | -0.116 |
ches | -0.110 |
Ho | -0.103 |
wool | -0.102 |
Token | y |
Token ID | 331 |
Feature activation | +3.74e-02 |
Pos logit contributions | |
ches | +0.014 |
fl | +0.013 |
Tiny | +0.013 |
Ho | +0.012 |
wand | +0.012 |
Neg logit contributions | |
handed | -0.025 |
Mike | -0.019 |
view | -0.019 |
amusement | -0.018 |
Jane | -0.018 |
Token | ummy |
Token ID | 13513 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
ches | +0.058 |
Tiny | +0.052 |
fl | +0.051 |
py | +0.051 |
Ho | +0.050 |
Neg logit contributions | |
handed | -0.078 |
Mike | -0.061 |
sandbox | -0.061 |
lobby | -0.060 |
amusement | -0.059 |
Token | piece |
Token ID | 3704 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
handed | +0.038 |
Mike | +0.028 |
vendor | +0.028 |
sandbox | +0.028 |
amusement | +0.028 |
Neg logit contributions | |
ches | -0.039 |
Tiny | -0.036 |
py | -0.036 |
fl | -0.035 |
Ho | -0.035 |
Token | for |
Token ID | 329 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
handed | +0.245 |
sandbox | +0.192 |
vendor | +0.192 |
Mike | +0.191 |
amusement | +0.188 |
Neg logit contributions | |
ches | -0.152 |
Tiny | -0.138 |
fl | -0.135 |
py | -0.133 |
Ho | -0.127 |
Token | Pat |
Token ID | 3208 |
Feature activation | -3.18e-02 |
Pos logit contributions | |
ches | +0.261 |
Tiny | +0.257 |
fl | +0.246 |
Ho | +0.238 |
Sparkle | +0.233 |
Neg logit contributions | |
handed | -0.438 |
sandbox | -0.351 |
vendor | -0.350 |
view | -0.337 |
lobby | -0.335 |
Token | ches |
Token ID | 2052 |
Feature activation | +4.73e-01 |
Pos logit contributions | |
handed | +0.074 |
Mike | +0.059 |
vendor | +0.059 |
sandbox | +0.059 |
amusement | +0.058 |
Neg logit contributions | |
ches | -0.041 |
Tiny | -0.038 |
py | -0.037 |
fl | -0.036 |
Ho | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | +9.61e-03 |
Pos logit contributions | |
ches | +0.705 |
Tiny | +0.625 |
fl | +0.622 |
py | +0.606 |
Sparkle | +0.581 |
Neg logit contributions | |
handed | -1.059 |
sandbox | -0.849 |
vendor | -0.848 |
Mike | -0.837 |
amusement | -0.817 |
Token | Pat |
Token ID | 3208 |
Feature activation | -3.23e-02 |
Pos logit contributions | |
ches | +0.018 |
py | +0.016 |
wand | +0.016 |
fl | +0.016 |
Cotton | +0.015 |
Neg logit contributions | |
handed | -0.017 |
Mike | -0.014 |
vendor | -0.013 |
Mike | -0.013 |
sandbox | -0.013 |
Token | ches |
Token ID | 2052 |
Feature activation | +4.73e-01 |
Pos logit contributions | |
handed | +0.078 |
Mike | +0.066 |
Jane | +0.064 |
vendor | +0.063 |
aren | +0.062 |
Neg logit contributions | |
Tiny | -0.032 |
ches | -0.032 |
Cotton | -0.031 |
Ho | -0.030 |
cartridge | -0.030 |
Token | smiled |
Token ID | 13541 |
Feature activation | -5.99e-02 |
Pos logit contributions | |
ches | +0.790 |
fl | +0.727 |
Tiny | +0.704 |
py | +0.697 |
Ho | +0.663 |
Neg logit contributions | |
handed | -0.965 |
vendor | -0.748 |
sandbox | -0.743 |
amusement | -0.727 |
Mike | -0.720 |
Token | and |
Token ID | 290 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
handed | +0.140 |
Mike | +0.110 |
sandbox | +0.110 |
vendor | +0.109 |
lobby | +0.108 |
Neg logit contributions | |
ches | -0.082 |
Tiny | -0.078 |
fl | -0.077 |
py | -0.074 |
Ho | -0.073 |
Token | thanked |
Token ID | 26280 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
ches | +0.036 |
py | +0.033 |
Tiny | +0.031 |
iny | +0.031 |
agons | +0.030 |
Neg logit contributions | |
handed | -0.045 |
vendor | -0.037 |
Mike | -0.037 |
sandbox | -0.036 |
amusement | -0.035 |
Token | her |
Token ID | 607 |
Feature activation | +6.29e-02 |
Pos logit contributions | |
handed | +0.029 |
Mike | +0.024 |
amusement | +0.023 |
vendor | +0.023 |
sandbox | +0.022 |
Neg logit contributions | |
ches | -0.025 |
py | -0.022 |
Tiny | -0.022 |
fl | -0.022 |
wand | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | +6.87e-03 |
Pos logit contributions | |
ches | +0.109 |
py | +0.100 |
Tiny | +0.098 |
Cotton | +0.096 |
fl | +0.094 |
Neg logit contributions | |
handed | -0.117 |
vendor | -0.090 |
amusement | -0.089 |
Mike | -0.087 |
amaz | -0.086 |
Token | had |
Token ID | 550 |
Feature activation | +3.21e-02 |
Pos logit contributions | |
ches | +0.031 |
Tiny | +0.029 |
py | +0.028 |
Ho | +0.027 |
fl | +0.027 |
Neg logit contributions | |
handed | -0.037 |
vendor | -0.028 |
amusement | -0.028 |
Mike | -0.028 |
sandbox | -0.028 |
Token | him |
Token ID | 683 |
Feature activation | -7.93e-02 |
Pos logit contributions | |
ches | +0.046 |
fl | +0.044 |
Tiny | +0.044 |
py | +0.041 |
Ho | +0.040 |
Neg logit contributions | |
handed | -0.070 |
Mike | -0.055 |
vendor | -0.055 |
lobby | -0.054 |
apartment | -0.054 |
Token | in |
Token ID | 287 |
Feature activation | +5.80e-02 |
Pos logit contributions | |
handed | +0.180 |
Mike | +0.146 |
sandbox | +0.144 |
lobby | +0.144 |
vendor | +0.142 |
Neg logit contributions | |
ches | -0.113 |
Tiny | -0.100 |
fl | -0.100 |
py | -0.096 |
hopping | -0.092 |
Token | its |
Token ID | 663 |
Feature activation | +8.15e-02 |
Pos logit contributions | |
fl | +0.060 |
Tiny | +0.059 |
ches | +0.055 |
Sparkle | +0.053 |
hopping | +0.053 |
Neg logit contributions | |
handed | -0.145 |
sandbox | -0.120 |
vendor | -0.119 |
aw | -0.117 |
Mike | -0.116 |
Token | clut |
Token ID | 32170 |
Feature activation | +6.32e-02 |
Pos logit contributions | |
ches | +0.127 |
fl | +0.125 |
Tiny | +0.124 |
wand | +0.118 |
py | +0.117 |
Neg logit contributions | |
handed | -0.168 |
Mike | -0.143 |
Jane | -0.129 |
vendor | -0.126 |
he | -0.125 |
Token | ches |
Token ID | 2052 |
Feature activation | +4.70e-01 |
Pos logit contributions | |
ches | +0.073 |
Tiny | +0.072 |
py | +0.069 |
Chick | +0.067 |
Ho | +0.067 |
Neg logit contributions | |
handed | -0.152 |
Mike | -0.121 |
amusement | -0.117 |
view | -0.116 |
he | -0.115 |
Token | . |
Token ID | 13 |
Feature activation | +9.75e-03 |
Pos logit contributions | |
ches | +0.722 |
Tiny | +0.632 |
fl | +0.623 |
py | +0.611 |
wand | +0.581 |
Neg logit contributions | |
handed | -1.031 |
vendor | -0.846 |
sandbox | -0.840 |
Mike | -0.834 |
amusement | -0.817 |
Token | He |
Token ID | 679 |
Feature activation | +9.30e-04 |
Pos logit contributions | |
ches | +0.017 |
fl | +0.015 |
py | +0.015 |
wand | +0.015 |
Tiny | +0.014 |
Neg logit contributions | |
handed | -0.019 |
Mike | -0.015 |
vendor | -0.015 |
sandbox | -0.014 |
amusement | -0.014 |
Token | was |
Token ID | 373 |
Feature activation | -5.83e-02 |
Pos logit contributions | |
ches | +0.001 |
Tiny | +0.001 |
py | +0.001 |
Cotton | +0.001 |
Ho | +0.001 |
Neg logit contributions | |
handed | -0.002 |
vendor | -0.001 |
Mike | -0.001 |
sandbox | -0.001 |
amusement | -0.001 |
Token | never |
Token ID | 1239 |
Feature activation | -3.54e-02 |
Pos logit contributions | |
handed | +0.114 |
Mike | +0.087 |
vendor | +0.086 |
sandbox | +0.085 |
amusement | +0.085 |
Neg logit contributions | |
ches | -0.101 |
Tiny | -0.093 |
py | -0.091 |
fl | -0.089 |
Ho | -0.088 |
Token | seen |
Token ID | 1775 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
handed | +0.061 |
Mike | +0.048 |
vendor | +0.046 |
lobby | +0.046 |
sandbox | +0.044 |
Neg logit contributions | |
ches | -0.068 |
Tiny | -0.065 |
py | -0.063 |
fl | -0.063 |
hopping | -0.061 |
Token | , |
Token ID | 11 |
Feature activation | -9.18e-02 |
Pos logit contributions | |
handed | +0.075 |
vendor | +0.059 |
amusement | +0.059 |
Mike | +0.058 |
sandbox | +0.058 |
Neg logit contributions | |
ches | -0.056 |
Tiny | -0.050 |
py | -0.049 |
fl | -0.047 |
Ho | -0.047 |
Token | like |
Token ID | 588 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
handed | +0.188 |
vendor | +0.149 |
sandbox | +0.148 |
Mike | +0.147 |
amusement | +0.145 |
Neg logit contributions | |
ches | -0.148 |
Tiny | -0.132 |
py | -0.131 |
fl | -0.127 |
Ho | -0.126 |
Token | making |
Token ID | 1642 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
handed | +0.268 |
vendor | +0.209 |
sandbox | +0.208 |
Mike | +0.208 |
amusement | +0.205 |
Neg logit contributions | |
ches | -0.201 |
Tiny | -0.182 |
py | -0.178 |
fl | -0.174 |
Ho | -0.172 |
Token | warm |
Token ID | 5814 |
Feature activation | +3.24e-02 |
Pos logit contributions | |
ches | +0.288 |
Tiny | +0.267 |
fl | +0.259 |
py | +0.257 |
Ho | +0.250 |
Neg logit contributions | |
handed | -0.425 |
vendor | -0.336 |
Mike | -0.333 |
sandbox | -0.330 |
amusement | -0.328 |
Token | lun |
Token ID | 14678 |
Feature activation | +4.25e-02 |
Pos logit contributions | |
ches | +0.045 |
Tiny | +0.041 |
py | +0.041 |
fl | +0.041 |
Ho | +0.039 |
Neg logit contributions | |
handed | -0.074 |
Mike | -0.059 |
vendor | -0.059 |
amusement | -0.058 |
sandbox | -0.058 |
Token | ches |
Token ID | 2052 |
Feature activation | +4.68e-01 |
Pos logit contributions | |
cartridge | +0.036 |
Tiny | +0.035 |
py | +0.033 |
iny | +0.033 |
wand | +0.032 |
Neg logit contributions | |
handed | -0.101 |
anymore | -0.091 |
he | -0.087 |
aw | -0.086 |
Mike | -0.085 |
Token | and |
Token ID | 290 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
ches | +0.732 |
Tiny | +0.665 |
py | +0.646 |
fl | +0.639 |
Ho | +0.622 |
Neg logit contributions | |
handed | -1.002 |
vendor | -0.785 |
Mike | -0.783 |
sandbox | -0.783 |
amusement | -0.773 |
Token | playing |
Token ID | 2712 |
Feature activation | +4.82e-02 |
Pos logit contributions | |
ches | +0.027 |
py | +0.025 |
iny | +0.025 |
Tiny | +0.024 |
agons | +0.023 |
Neg logit contributions | |
handed | -0.033 |
sandbox | -0.028 |
vendor | -0.027 |
amaz | -0.027 |
amusement | -0.027 |
Token | games |
Token ID | 1830 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
ches | +0.072 |
Tiny | +0.065 |
fl | +0.065 |
py | +0.063 |
Ho | +0.062 |
Neg logit contributions | |
handed | -0.107 |
vendor | -0.085 |
Mike | -0.084 |
amusement | -0.084 |
sandbox | -0.083 |
Token | . |
Token ID | 13 |
Feature activation | +4.82e-03 |
Pos logit contributions | |
handed | +0.232 |
Mike | +0.185 |
vendor | +0.183 |
amusement | +0.182 |
sandbox | +0.181 |
Neg logit contributions | |
ches | -0.154 |
Tiny | -0.137 |
fl | -0.136 |
py | -0.132 |
Ho | -0.127 |
Token |
|
Token ID | 198 |
Feature activation | +3.17e-02 |
Pos logit contributions | |
ches | +0.008 |
fl | +0.008 |
wand | +0.008 |
py | +0.008 |
iny | +0.007 |
Neg logit contributions | |
handed | -0.009 |
Mike | -0.007 |
vendor | -0.007 |
sandbox | -0.007 |
Mike | -0.007 |
Token | paper |
Token ID | 3348 |
Feature activation | +8.94e-02 |
Pos logit contributions | |
ches | +0.008 |
Tiny | +0.007 |
py | +0.007 |
fl | +0.007 |
Ho | +0.007 |
Neg logit contributions | |
handed | -0.013 |
Mike | -0.011 |
vendor | -0.011 |
sandbox | -0.010 |
amusement | -0.010 |
Token | flew |
Token ID | 13112 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
ches | +0.131 |
Tiny | +0.118 |
fl | +0.115 |
py | +0.115 |
Ho | +0.110 |
Neg logit contributions | |
handed | -0.200 |
Mike | -0.157 |
vendor | -0.157 |
sandbox | -0.155 |
amusement | -0.155 |
Token | away |
Token ID | 1497 |
Feature activation | -9.04e-02 |
Pos logit contributions | |
ches | +0.408 |
Tiny | +0.374 |
py | +0.363 |
fl | +0.362 |
Ho | +0.353 |
Neg logit contributions | |
handed | -0.524 |
vendor | -0.403 |
Mike | -0.402 |
sandbox | -0.400 |
amusement | -0.398 |
Token | . |
Token ID | 13 |
Feature activation | -2.39e-03 |
Pos logit contributions | |
handed | +0.222 |
amusement | +0.180 |
Mike | +0.179 |
sandbox | +0.179 |
vendor | +0.177 |
Neg logit contributions | |
ches | -0.112 |
fl | -0.112 |
Tiny | -0.099 |
hopping | -0.097 |
wand | -0.097 |
Token | Pat |
Token ID | 3208 |
Feature activation | -4.09e-02 |
Pos logit contributions | |
handed | +0.004 |
Mike | +0.003 |
vendor | +0.003 |
Jane | +0.003 |
sandbox | +0.003 |
Neg logit contributions | |
ches | -0.004 |
py | -0.004 |
fl | -0.004 |
wand | -0.004 |
Cotton | -0.004 |
Token | ches |
Token ID | 2052 |
Feature activation | +4.67e-01 |
Pos logit contributions | |
handed | +0.109 |
Mike | +0.093 |
Jane | +0.090 |
vendor | +0.089 |
sandbox | +0.089 |
Neg logit contributions | |
Tiny | -0.032 |
ches | -0.032 |
Cotton | -0.030 |
Ho | -0.029 |
py | -0.029 |
Token | was |
Token ID | 373 |
Feature activation | -7.51e-02 |
Pos logit contributions | |
ches | +0.706 |
fl | +0.647 |
Tiny | +0.608 |
py | +0.607 |
Ho | +0.569 |
Neg logit contributions | |
handed | -1.028 |
sandbox | -0.834 |
vendor | -0.834 |
lobby | -0.816 |
amusement | -0.814 |
Token | disappointed |
Token ID | 11679 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
handed | +0.171 |
Mike | +0.138 |
sandbox | +0.137 |
vendor | +0.137 |
view | +0.134 |
Neg logit contributions | |
ches | -0.106 |
fl | -0.101 |
Tiny | -0.098 |
hopping | -0.096 |
py | -0.094 |
Token | - |
Token ID | 532 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
handed | +0.494 |
amusement | +0.392 |
sandbox | +0.391 |
lobby | +0.386 |
Mike | +0.385 |
Neg logit contributions | |
ches | -0.224 |
Tiny | -0.215 |
fl | -0.206 |
Ho | -0.201 |
wand | -0.190 |
Token | she |
Token ID | 673 |
Feature activation | +4.48e-03 |
Pos logit contributions | |
handed | +0.331 |
view | +0.254 |
Mike | +0.253 |
sandbox | +0.250 |
lobby | +0.250 |
Neg logit contributions | |
Tiny | -0.186 |
ches | -0.179 |
fl | -0.176 |
Ho | -0.171 |
Sparkle | -0.169 |
Token | wanted |
Token ID | 2227 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
ches | +0.007 |
Tiny | +0.006 |
py | +0.006 |
Ho | +0.006 |
Cotton | +0.006 |
Neg logit contributions | |
handed | -0.009 |
vendor | -0.007 |
sandbox | -0.007 |
Mike | -0.007 |
amusement | -0.007 |
Token | â |
Token ID | 22940 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
handed | +0.126 |
Mike | +0.104 |
Mike | +0.099 |
vendor | +0.096 |
Jane | +0.095 |
Neg logit contributions | |
ches | -0.137 |
fl | -0.124 |
wand | -0.121 |
py | -0.120 |
arella | -0.111 |
Token | € |
Token ID | 26391 |
Feature activation | -5.45e-02 |
Pos logit contributions | |
ches | +0.212 |
Tiny | +0.200 |
py | +0.196 |
fl | +0.194 |
Ho | +0.192 |
Neg logit contributions | |
handed | -0.203 |
Mike | -0.149 |
vendor | -0.146 |
amusement | -0.146 |
sandbox | -0.144 |
Token |
|
Token ID | 198 |
Feature activation | +2.51e-02 |
Pos logit contributions | |
handed | +0.119 |
vendor | +0.094 |
amusement | +0.093 |
sandbox | +0.092 |
Mike | +0.091 |
Neg logit contributions | |
ches | -0.082 |
Tiny | -0.077 |
fl | -0.071 |
py | -0.071 |
Ho | -0.070 |
Token |
|
Token ID | 198 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
ches | +0.043 |
Tiny | +0.039 |
py | +0.039 |
fl | +0.039 |
wand | +0.038 |
Neg logit contributions | |
handed | -0.049 |
Mike | -0.040 |
Jane | -0.037 |
vendor | -0.037 |
amusement | -0.037 |
Token | Pat |
Token ID | 12130 |
Feature activation | -5.39e-03 |
Pos logit contributions | |
ches | +0.047 |
Tiny | +0.043 |
py | +0.043 |
wand | +0.043 |
fl | +0.042 |
Neg logit contributions | |
handed | -0.043 |
Mike | -0.035 |
Jane | -0.032 |
amusement | -0.031 |
vendor | -0.031 |
Token | ches |
Token ID | 2052 |
Feature activation | +4.66e-01 |
Pos logit contributions | |
handed | +0.014 |
Mike | +0.012 |
Jane | +0.011 |
vendor | +0.011 |
amusement | +0.011 |
Neg logit contributions | |
Tiny | -0.005 |
ches | -0.005 |
Cotton | -0.004 |
cartridge | -0.004 |
py | -0.004 |
Token | gave |
Token ID | 2921 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
ches | +0.743 |
fl | +0.666 |
Tiny | +0.666 |
py | +0.655 |
Ho | +0.624 |
Neg logit contributions | |
handed | -0.979 |
vendor | -0.765 |
sandbox | -0.761 |
amusement | -0.751 |
Mike | -0.748 |
Token | the |
Token ID | 262 |
Feature activation | -6.61e-03 |
Pos logit contributions | |
ches | +0.267 |
py | +0.242 |
Tiny | +0.242 |
fl | +0.235 |
Cotton | +0.229 |
Neg logit contributions | |
handed | -0.357 |
Mike | -0.279 |
vendor | -0.275 |
amusement | -0.272 |
sandbox | -0.271 |
Token | little |
Token ID | 1310 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
handed | +0.015 |
vendor | +0.012 |
sandbox | +0.012 |
amusement | +0.012 |
lobby | +0.012 |
Neg logit contributions | |
ches | -0.009 |
Tiny | -0.008 |
py | -0.008 |
Ho | -0.008 |
Cotton | -0.008 |
Token | girl |
Token ID | 2576 |
Feature activation | -4.54e-02 |
Pos logit contributions | |
ches | +0.199 |
Tiny | +0.183 |
py | +0.182 |
fl | +0.174 |
Cotton | +0.172 |
Neg logit contributions | |
handed | -0.207 |
vendor | -0.158 |
sandbox | -0.157 |
amusement | -0.155 |
Mike | -0.155 |
Token | a |
Token ID | 257 |
Feature activation | -8.32e-03 |
Pos logit contributions | |
handed | +0.106 |
Mike | +0.087 |
sandbox | +0.087 |
lobby | +0.085 |
vendor | +0.085 |
Neg logit contributions | |
ches | -0.060 |
Tiny | -0.055 |
fl | -0.053 |
py | -0.051 |
Ho | -0.050 |
Token | mouth |
Token ID | 5422 |
Feature activation | +1.73e-03 |
Pos logit contributions | |
ches | +0.030 |
Tiny | +0.028 |
fl | +0.027 |
py | +0.027 |
Ho | +0.027 |
Neg logit contributions | |
handed | -0.032 |
Mike | -0.024 |
vendor | -0.023 |
amusement | -0.023 |
sandbox | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
ches | +0.003 |
Tiny | +0.002 |
fl | +0.002 |
py | +0.002 |
Ho | +0.002 |
Neg logit contributions | |
handed | -0.004 |
Mike | -0.003 |
sandbox | -0.003 |
vendor | -0.003 |
amusement | -0.003 |
Token | He |
Token ID | 679 |
Feature activation | -6.83e-03 |
Pos logit contributions | |
ches | +0.017 |
Tiny | +0.016 |
py | +0.015 |
Ho | +0.015 |
fl | +0.015 |
Neg logit contributions | |
handed | -0.021 |
vendor | -0.016 |
sandbox | -0.016 |
Mike | -0.016 |
amusement | -0.016 |
Token | cr |
Token ID | 1067 |
Feature activation | +8.77e-02 |
Pos logit contributions | |
handed | +0.014 |
sandbox | +0.011 |
vendor | +0.011 |
amusement | +0.011 |
lobby | +0.011 |
Neg logit contributions | |
ches | -0.010 |
Tiny | -0.010 |
py | -0.010 |
Pig | -0.009 |
iny | -0.009 |
Token | un |
Token ID | 403 |
Feature activation | +3.03e-01 |
Pos logit contributions | |
ches | +0.155 |
fl | +0.145 |
Tiny | +0.145 |
Ho | +0.140 |
Cotton | +0.139 |
Neg logit contributions | |
handed | -0.163 |
Mike | -0.125 |
view | -0.120 |
sandbox | -0.119 |
vendor | -0.119 |
Token | ches |
Token ID | 2052 |
Feature activation | +4.65e-01 |
Pos logit contributions | |
ches | +0.473 |
Tiny | +0.429 |
py | +0.419 |
fl | +0.414 |
Ho | +0.405 |
Neg logit contributions | |
handed | -0.649 |
vendor | -0.509 |
Mike | -0.506 |
sandbox | -0.505 |
amusement | -0.502 |
Token | them |
Token ID | 606 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
ches | +0.652 |
fl | +0.551 |
Tiny | +0.549 |
py | +0.532 |
Ho | +0.510 |
Neg logit contributions | |
handed | -1.089 |
sandbox | -0.934 |
vendor | -0.913 |
amusement | -0.908 |
Mike | -0.897 |
Token | and |
Token ID | 290 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
ches | +0.198 |
Tiny | +0.181 |
py | +0.176 |
fl | +0.171 |
Ho | +0.169 |
Neg logit contributions | |
handed | -0.315 |
vendor | -0.249 |
Mike | -0.248 |
sandbox | -0.248 |
amusement | -0.245 |
Token | swall |
Token ID | 20301 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
ches | +0.028 |
py | +0.028 |
iny | +0.027 |
agons | +0.025 |
Pig | +0.025 |
Neg logit contributions | |
handed | -0.033 |
sandbox | -0.031 |
amusement | -0.029 |
vendor | -0.029 |
amaz | -0.028 |
Token | ows |
Token ID | 1666 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
ches | +0.297 |
Tiny | +0.280 |
py | +0.276 |
fl | +0.273 |
Ho | +0.267 |
Neg logit contributions | |
handed | -0.218 |
Mike | -0.151 |
vendor | -0.151 |
sandbox | -0.150 |
amusement | -0.147 |
Token | them |
Token ID | 606 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
handed | +0.248 |
sandbox | +0.195 |
vendor | +0.194 |
Mike | +0.194 |
amusement | +0.192 |
Neg logit contributions | |
ches | -0.170 |
Tiny | -0.154 |
py | -0.149 |
fl | -0.148 |
Ho | -0.145 |
Token | ball |
Token ID | 2613 |
Feature activation | -3.17e-02 |
Pos logit contributions | |
handed | +0.000 |
sandbox | +0.000 |
vendor | +0.000 |
amusement | +0.000 |
alley | +0.000 |
Neg logit contributions | |
ches | -0.000 |
py | -0.000 |
Tiny | -0.000 |
Chick | -0.000 |
Cotton | -0.000 |
Token | and |
Token ID | 290 |
Feature activation | +1.98e-02 |
Pos logit contributions | |
handed | +0.067 |
Mike | +0.052 |
vendor | +0.052 |
amusement | +0.051 |
sandbox | +0.051 |
Neg logit contributions | |
ches | -0.050 |
Tiny | -0.046 |
fl | -0.045 |
py | -0.045 |
Ho | -0.044 |
Token | the |
Token ID | 262 |
Feature activation | -9.22e-04 |
Pos logit contributions | |
ches | +0.032 |
py | +0.029 |
Tiny | +0.028 |
iny | +0.028 |
Pig | +0.027 |
Neg logit contributions | |
handed | -0.039 |
sandbox | -0.033 |
vendor | -0.032 |
amusement | -0.032 |
lobby | -0.030 |
Token | dog |
Token ID | 3290 |
Feature activation | -7.26e-02 |
Pos logit contributions | |
handed | +0.002 |
Mike | +0.001 |
vendor | +0.001 |
sandbox | +0.001 |
amusement | +0.001 |
Neg logit contributions | |
ches | -0.002 |
Tiny | -0.001 |
py | -0.001 |
fl | -0.001 |
Ho | -0.001 |
Token | fet |
Token ID | 11351 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
handed | +0.150 |
sandbox | +0.117 |
vendor | +0.116 |
Mike | +0.114 |
amusement | +0.114 |
Neg logit contributions | |
ches | -0.114 |
Tiny | -0.106 |
py | -0.104 |
Cotton | -0.099 |
fl | -0.099 |
Token | ches |
Token ID | 2052 |
Feature activation | +4.64e-01 |
Pos logit contributions | |
py | +0.129 |
hopping | +0.118 |
asking | +0.112 |
Pig | +0.110 |
carniv | +0.109 |
Neg logit contributions | |
handed | -0.327 |
he | -0.287 |
aw | -0.282 |
Mike | -0.278 |
anymore | -0.277 |
Token | it |
Token ID | 340 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
ches | +0.716 |
Tiny | +0.648 |
py | +0.624 |
fl | +0.624 |
Ho | +0.607 |
Neg logit contributions | |
handed | -1.008 |
vendor | -0.794 |
sandbox | -0.791 |
Mike | -0.787 |
amusement | -0.781 |
Token | . |
Token ID | 13 |
Feature activation | +7.09e-03 |
Pos logit contributions | |
ches | +0.041 |
Tiny | +0.038 |
py | +0.038 |
Cotton | +0.036 |
Pig | +0.036 |
Neg logit contributions | |
handed | -0.045 |
vendor | -0.034 |
sandbox | -0.033 |
Mike | -0.033 |
amusement | -0.033 |
Token | They |
Token ID | 1119 |
Feature activation | +4.15e-02 |
Pos logit contributions | |
ches | +0.012 |
py | +0.011 |
Tiny | +0.010 |
fl | +0.010 |
wand | +0.010 |
Neg logit contributions | |
handed | -0.014 |
vendor | -0.011 |
Mike | -0.011 |
sandbox | -0.011 |
amusement | -0.011 |
Token | push |
Token ID | 4574 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
ches | +0.072 |
Tiny | +0.066 |
fl | +0.065 |
py | +0.065 |
Ho | +0.064 |
Neg logit contributions | |
handed | -0.082 |
Mike | -0.062 |
vendor | -0.061 |
sandbox | -0.060 |
amusement | -0.060 |
Token | the |
Token ID | 262 |
Feature activation | -3.05e-03 |
Pos logit contributions | |
ches | +0.025 |
Tiny | +0.023 |
fl | +0.022 |
Ho | +0.022 |
py | +0.020 |
Neg logit contributions | |
handed | -0.042 |
sandbox | -0.033 |
lobby | -0.033 |
vendor | -0.033 |
view | -0.033 |
Token | smelled |
Token ID | 37609 |
Feature activation | -8.42e-02 |
Pos logit contributions | |
ches | +0.003 |
Tiny | +0.003 |
py | +0.003 |
fl | +0.003 |
Ho | +0.003 |
Neg logit contributions | |
handed | -0.003 |
sandbox | -0.002 |
Mike | -0.002 |
vendor | -0.002 |
amusement | -0.002 |
Token | something |
Token ID | 1223 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
handed | +0.189 |
sandbox | +0.152 |
Mike | +0.152 |
apartment | +0.151 |
aren | +0.151 |
Neg logit contributions | |
Tiny | -0.117 |
ches | -0.117 |
fl | -0.114 |
Ho | -0.112 |
Cotton | -0.110 |
Token | amazing |
Token ID | 4998 |
Feature activation | -7.47e-02 |
Pos logit contributions | |
handed | +0.438 |
Mike | +0.338 |
vendor | +0.338 |
sandbox | +0.336 |
amusement | +0.333 |
Neg logit contributions | |
ches | -0.339 |
Tiny | -0.310 |
py | -0.302 |
fl | -0.300 |
Ho | -0.292 |
Token | . |
Token ID | 13 |
Feature activation | -5.29e-03 |
Pos logit contributions | |
handed | +0.154 |
sandbox | +0.119 |
vendor | +0.118 |
Mike | +0.116 |
amusement | +0.115 |
Neg logit contributions | |
ches | -0.125 |
fl | -0.114 |
Tiny | -0.113 |
py | -0.111 |
wand | -0.107 |
Token | Pat |
Token ID | 3208 |
Feature activation | -4.56e-02 |
Pos logit contributions | |
handed | +0.009 |
vendor | +0.007 |
Mike | +0.007 |
sandbox | +0.007 |
amusement | +0.007 |
Neg logit contributions | |
ches | -0.010 |
py | -0.009 |
fl | -0.009 |
Tiny | -0.008 |
wand | -0.008 |
Token | ches |
Token ID | 2052 |
Feature activation | +4.63e-01 |
Pos logit contributions | |
handed | +0.131 |
Mike | +0.115 |
Jane | +0.114 |
aw | +0.113 |
ill | +0.113 |
Neg logit contributions | |
carniv | -0.018 |
cartridge | -0.018 |
Cotton | -0.017 |
Tiny | -0.017 |
Ho | -0.016 |
Token | knew |
Token ID | 2993 |
Feature activation | +1.00e-02 |
Pos logit contributions | |
ches | +0.699 |
fl | +0.624 |
Tiny | +0.615 |
py | +0.611 |
Ho | +0.569 |
Neg logit contributions | |
handed | -1.025 |
vendor | -0.814 |
sandbox | -0.812 |
Mike | -0.800 |
amusement | -0.800 |
Token | that |
Token ID | 326 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
ches | +0.017 |
Tiny | +0.016 |
py | +0.016 |
fl | +0.016 |
Ho | +0.015 |
Neg logit contributions | |
handed | -0.020 |
vendor | -0.015 |
Mike | -0.015 |
sandbox | -0.015 |
amusement | -0.015 |
Token | y |
Token ID | 331 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
handed | +0.246 |
Mike | +0.193 |
vendor | +0.189 |
sandbox | +0.188 |
view | +0.187 |
Neg logit contributions | |
ches | -0.155 |
Tiny | -0.152 |
fl | -0.146 |
Ho | -0.140 |
py | -0.140 |
Token | ummy |
Token ID | 13513 |
Feature activation | -3.60e-02 |
Pos logit contributions | |
ches | +0.033 |
Tiny | +0.031 |
py | +0.030 |
fl | +0.029 |
Ho | +0.029 |
Neg logit contributions | |
handed | -0.042 |
vendor | -0.032 |
sandbox | -0.032 |
Mike | -0.032 |
amusement | -0.032 |
Token | smell |
Token ID | 8508 |
Feature activation | -3.23e-02 |
Pos logit contributions | |
handed | +0.079 |
Mike | +0.061 |
sandbox | +0.060 |
view | +0.060 |
lobby | +0.059 |
Neg logit contributions | |
ches | -0.057 |
Tiny | -0.052 |
fl | -0.051 |
wand | -0.049 |
py | -0.049 |
Token | friends |
Token ID | 2460 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
ches | +0.046 |
Tiny | +0.042 |
py | +0.041 |
fl | +0.041 |
Ho | +0.040 |
Neg logit contributions | |
handed | -0.077 |
Mike | -0.061 |
vendor | -0.061 |
sandbox | -0.061 |
amusement | -0.060 |
Token | and |
Token ID | 290 |
Feature activation | +3.64e-03 |
Pos logit contributions | |
ches | +0.080 |
fl | +0.075 |
hopping | +0.071 |
Sparkle | +0.069 |
Tiny | +0.069 |
Neg logit contributions | |
handed | -0.291 |
lobby | -0.248 |
sandbox | -0.247 |
Mike | -0.245 |
view | -0.242 |
Token | shared |
Token ID | 4888 |
Feature activation | +2.20e-02 |
Pos logit contributions | |
ches | +0.005 |
Tiny | +0.005 |
py | +0.005 |
fl | +0.005 |
Ho | +0.005 |
Neg logit contributions | |
handed | -0.008 |
Mike | -0.006 |
vendor | -0.006 |
sandbox | -0.006 |
amusement | -0.006 |
Token | their |
Token ID | 511 |
Feature activation | -8.73e-02 |
Pos logit contributions | |
fl | +0.022 |
ches | +0.020 |
carrots | +0.019 |
wool | +0.019 |
Tiny | +0.019 |
Neg logit contributions | |
handed | -0.057 |
Mike | -0.048 |
alley | -0.047 |
apartment | -0.047 |
umbling | -0.047 |
Token | lun |
Token ID | 14678 |
Feature activation | +3.44e-02 |
Pos logit contributions | |
handed | +0.188 |
Mike | +0.156 |
aren | +0.145 |
lobby | +0.144 |
apartment | +0.142 |
Neg logit contributions | |
fl | -0.121 |
ches | -0.121 |
Tiny | -0.113 |
wool | -0.113 |
Ho | -0.112 |
Token | ches |
Token ID | 2052 |
Feature activation | +4.63e-01 |
Pos logit contributions | |
Tiny | +0.034 |
py | +0.033 |
carniv | +0.032 |
Sparkle | +0.032 |
cartridge | +0.031 |
Neg logit contributions | |
handed | -0.084 |
anymore | -0.068 |
Mike | -0.066 |
aw | -0.065 |
out | -0.064 |
Token | every |
Token ID | 790 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
ches | +0.676 |
Tiny | +0.578 |
fl | +0.571 |
py | +0.558 |
Sparkle | +0.539 |
Neg logit contributions | |
handed | -1.051 |
sandbox | -0.878 |
Mike | -0.878 |
vendor | -0.853 |
lobby | -0.846 |
Token | day |
Token ID | 1110 |
Feature activation | +7.12e-02 |
Pos logit contributions | |
ches | +0.158 |
Tiny | +0.153 |
py | +0.147 |
Ho | +0.146 |
Chick | +0.141 |
Neg logit contributions | |
handed | -0.211 |
vendor | -0.172 |
amusement | -0.163 |
sandbox | -0.162 |
lobby | -0.159 |
Token | on |
Token ID | 319 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
ches | +0.113 |
Tiny | +0.103 |
py | +0.100 |
fl | +0.100 |
Ho | +0.098 |
Neg logit contributions | |
handed | -0.149 |
Mike | -0.116 |
vendor | -0.116 |
sandbox | -0.115 |
amusement | -0.114 |
Token | the |
Token ID | 262 |
Feature activation | -9.93e-03 |
Pos logit contributions | |
ches | +0.225 |
Tiny | +0.213 |
Ho | +0.198 |
fl | +0.197 |
Cotton | +0.194 |
Neg logit contributions | |
handed | -0.351 |
sandbox | -0.281 |
Mike | -0.275 |
amusement | -0.273 |
vendor | -0.273 |
Token | bus |
Token ID | 1323 |
Feature activation | -6.49e-02 |
Pos logit contributions | |
handed | +0.020 |
Mike | +0.017 |
aren | +0.016 |
Jane | +0.016 |
apartment | +0.015 |
Neg logit contributions | |
ches | -0.016 |
fl | -0.015 |
Tiny | -0.014 |
Ho | -0.014 |
wand | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | -4.52e-03 |
Pos logit contributions | |
handed | +0.033 |
Mike | +0.026 |
vendor | +0.026 |
sandbox | +0.026 |
amusement | +0.026 |
Neg logit contributions | |
ches | -0.026 |
Tiny | -0.024 |
py | -0.023 |
fl | -0.023 |
Ho | -0.023 |
Token | story |
Token ID | 1621 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
handed | +0.008 |
amusement | +0.007 |
vendor | +0.007 |
temple | +0.007 |
lobby | +0.007 |
Neg logit contributions | |
ches | -0.008 |
py | -0.007 |
Cotton | -0.007 |
Tiny | -0.007 |
Ho | -0.007 |
Token | is |
Token ID | 318 |
Feature activation | -4.52e-02 |
Pos logit contributions | |
ches | +0.180 |
Tiny | +0.153 |
fl | +0.153 |
wand | +0.151 |
py | +0.150 |
Neg logit contributions | |
handed | -0.336 |
sandbox | -0.265 |
Mike | -0.265 |
amusement | -0.262 |
amaz | -0.254 |
Token | : |
Token ID | 25 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
handed | +0.075 |
vendor | +0.058 |
amusement | +0.055 |
Mike | +0.055 |
apartment | +0.052 |
Neg logit contributions | |
ches | -0.089 |
py | -0.081 |
Tiny | -0.080 |
fl | -0.079 |
Cotton | -0.079 |
Token | Mat |
Token ID | 6550 |
Feature activation | +5.05e-02 |
Pos logit contributions | |
ches | +0.046 |
py | +0.040 |
agons | +0.038 |
fl | +0.038 |
Cotton | +0.037 |
Neg logit contributions | |
handed | -0.038 |
vendor | -0.031 |
amusement | -0.031 |
Mike | -0.030 |
apartment | -0.030 |
Token | ches |
Token ID | 2052 |
Feature activation | +4.63e-01 |
Pos logit contributions | |
ches | +0.073 |
Tiny | +0.068 |
py | +0.067 |
fl | +0.066 |
Ho | +0.065 |
Neg logit contributions | |
handed | -0.111 |
Mike | -0.086 |
vendor | -0.086 |
sandbox | -0.085 |
amusement | -0.084 |
Token | are |
Token ID | 389 |
Feature activation | -1.99e-01 |
Pos logit contributions | |
ches | +0.732 |
Tiny | +0.658 |
fl | +0.645 |
py | +0.639 |
wand | +0.601 |
Neg logit contributions | |
handed | -0.979 |
vendor | -0.799 |
Mike | -0.791 |
sandbox | -0.785 |
amusement | -0.778 |
Token | not |
Token ID | 407 |
Feature activation | +2.96e-02 |
Pos logit contributions | |
handed | +0.242 |
vendor | +0.218 |
else | +0.190 |
temple | +0.184 |
amusement | +0.181 |
Neg logit contributions | |
ches | -0.428 |
arella | -0.386 |
bee | -0.380 |
tle | -0.377 |
py | -0.376 |
Token | toys |
Token ID | 14958 |
Feature activation | +5.33e-04 |
Pos logit contributions | |
ches | +0.059 |
py | +0.050 |
arella | +0.050 |
tle | +0.050 |
agons | +0.049 |
Neg logit contributions | |
handed | -0.045 |
vendor | -0.037 |
yet | -0.035 |
sandbox | -0.034 |
anymore | -0.034 |
Token | and |
Token ID | 290 |
Feature activation | +1.37e-02 |
Pos logit contributions | |
ches | +0.001 |
wand | +0.001 |
fl | +0.001 |
Tiny | +0.001 |
Ho | +0.000 |
Neg logit contributions | |
handed | -0.001 |
Mike | -0.001 |
Jane | -0.001 |
lobby | -0.001 |
view | -0.001 |
Token | can |
Token ID | 460 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
ches | +0.021 |
Tiny | +0.020 |
py | +0.019 |
fl | +0.019 |
Ho | +0.019 |
Neg logit contributions | |
handed | -0.030 |
Mike | -0.023 |
vendor | -0.023 |
amusement | -0.022 |
sandbox | -0.022 |
Token | a |
Token ID | 257 |
Feature activation | -4.82e-03 |
Pos logit contributions | |
ches | +0.118 |
Ho | +0.115 |
fl | +0.115 |
Tiny | +0.109 |
wool | +0.102 |
Neg logit contributions | |
handed | -0.205 |
view | -0.160 |
unatt | -0.156 |
lobby | -0.155 |
apartment | -0.154 |
Token | ball |
Token ID | 2613 |
Feature activation | -3.58e-02 |
Pos logit contributions | |
handed | +0.011 |
Mike | +0.008 |
amusement | +0.008 |
view | +0.008 |
lobby | +0.008 |
Neg logit contributions | |
ches | -0.007 |
fl | -0.007 |
Tiny | -0.007 |
wand | -0.006 |
Ho | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
handed | +0.077 |
vendor | +0.060 |
Mike | +0.060 |
sandbox | +0.059 |
amusement | +0.059 |
Neg logit contributions | |
ches | -0.055 |
Tiny | -0.051 |
py | -0.049 |
fl | -0.049 |
Ho | -0.048 |
Token | he |
Token ID | 339 |
Feature activation | -1.50e-02 |
Pos logit contributions | |
ches | +0.024 |
Tiny | +0.022 |
py | +0.022 |
fl | +0.021 |
Ho | +0.020 |
Neg logit contributions | |
handed | -0.035 |
sandbox | -0.028 |
vendor | -0.028 |
Mike | -0.028 |
amusement | -0.028 |
Token | fet |
Token ID | 11351 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
handed | +0.028 |
sandbox | +0.024 |
vendor | +0.023 |
amusement | +0.022 |
lobby | +0.022 |
Neg logit contributions | |
py | -0.023 |
ches | -0.023 |
Tiny | -0.023 |
Pig | -0.022 |
iny | -0.022 |
Token | ches |
Token ID | 2052 |
Feature activation | +4.62e-01 |
Pos logit contributions | |
py | +0.114 |
hopping | +0.098 |
Pig | +0.095 |
carniv | +0.093 |
asking | +0.093 |
Neg logit contributions | |
handed | -0.344 |
he | -0.301 |
Mike | -0.300 |
aw | -0.297 |
anymore | -0.291 |
Token | it |
Token ID | 340 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
ches | +0.704 |
Tiny | +0.622 |
fl | +0.606 |
py | +0.590 |
Ho | +0.582 |
Neg logit contributions | |
handed | -1.035 |
vendor | -0.829 |
sandbox | -0.827 |
amusement | -0.812 |
Mike | -0.807 |
Token | . |
Token ID | 13 |
Feature activation | +4.39e-03 |
Pos logit contributions | |
ches | +0.035 |
Tiny | +0.033 |
py | +0.032 |
fl | +0.031 |
Cotton | +0.031 |
Neg logit contributions | |
handed | -0.042 |
vendor | -0.032 |
sandbox | -0.032 |
Mike | -0.031 |
amusement | -0.031 |
Token | They |
Token ID | 1119 |
Feature activation | +3.74e-02 |
Pos logit contributions | |
ches | +0.007 |
Tiny | +0.006 |
py | +0.006 |
fl | +0.006 |
Ho | +0.006 |
Neg logit contributions | |
handed | -0.010 |
vendor | -0.007 |
Mike | -0.007 |
sandbox | -0.007 |
amusement | -0.007 |
Token | run |
Token ID | 1057 |
Feature activation | +4.99e-02 |
Pos logit contributions | |
ches | +0.066 |
Tiny | +0.061 |
py | +0.060 |
fl | +0.059 |
Ho | +0.058 |
Neg logit contributions | |
handed | -0.072 |
vendor | -0.054 |
sandbox | -0.054 |
Mike | -0.054 |
amusement | -0.053 |
Token | and |
Token ID | 290 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
ches | +0.069 |
fl | +0.059 |
Tiny | +0.058 |
Ho | +0.056 |
Sparkle | +0.054 |
Neg logit contributions | |
handed | -0.124 |
view | -0.101 |
amaz | -0.099 |
amusement | -0.097 |
lobby | -0.097 |
Token | , |
Token ID | 11 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
ches | +0.028 |
Tiny | +0.025 |
fl | +0.025 |
py | +0.025 |
Ho | +0.024 |
Neg logit contributions | |
handed | -0.037 |
Mike | -0.029 |
vendor | -0.029 |
sandbox | -0.029 |
amusement | -0.029 |
Token | â |
Token ID | 22940 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
handed | +0.034 |
Mike | +0.026 |
vendor | +0.026 |
sandbox | +0.026 |
amusement | +0.026 |
Neg logit contributions | |
ches | -0.029 |
Tiny | -0.026 |
py | -0.026 |
fl | -0.025 |
Ho | -0.025 |
Token | € |
Token ID | 26391 |
Feature activation | -5.86e-02 |
Pos logit contributions | |
ches | +0.215 |
Tiny | +0.205 |
py | +0.200 |
Ho | +0.198 |
fl | +0.197 |
Neg logit contributions | |
handed | -0.185 |
Mike | -0.132 |
amusement | -0.130 |
vendor | -0.129 |
view | -0.126 |
Token | said |
Token ID | 531 |
Feature activation | +8.94e-02 |
Pos logit contributions | |
handed | +0.123 |
vendor | +0.096 |
sandbox | +0.095 |
amusement | +0.095 |
Mike | +0.094 |
Neg logit contributions | |
ches | -0.093 |
Tiny | -0.087 |
fl | -0.081 |
py | -0.081 |
Ho | -0.080 |
Token | Pat |
Token ID | 3208 |
Feature activation | -4.42e-02 |
Pos logit contributions | |
ches | +0.147 |
Tiny | +0.134 |
py | +0.132 |
fl | +0.130 |
Ho | +0.126 |
Neg logit contributions | |
handed | -0.181 |
Mike | -0.141 |
vendor | -0.140 |
sandbox | -0.138 |
amusement | -0.138 |
Token | ches |
Token ID | 2052 |
Feature activation | +4.62e-01 |
Pos logit contributions | |
handed | +0.104 |
Mike | +0.087 |
vendor | +0.085 |
Jane | +0.083 |
amusement | +0.083 |
Neg logit contributions | |
Tiny | -0.046 |
ches | -0.046 |
fl | -0.044 |
Cotton | -0.044 |
py | -0.044 |
Token | . |
Token ID | 13 |
Feature activation | -3.45e-03 |
Pos logit contributions | |
ches | +0.708 |
Tiny | +0.621 |
fl | +0.614 |
py | +0.610 |
Sparkle | +0.575 |
Neg logit contributions | |
handed | -1.008 |
vendor | -0.802 |
sandbox | -0.800 |
amusement | -0.786 |
Mike | -0.777 |
Token | She |
Token ID | 1375 |
Feature activation | -3.52e-02 |
Pos logit contributions | |
handed | +0.006 |
Mike | +0.005 |
vendor | +0.005 |
Mike | +0.005 |
Jane | +0.005 |
Neg logit contributions | |
ches | -0.006 |
fl | -0.006 |
py | -0.006 |
wand | -0.006 |
Cotton | -0.005 |
Token | went |
Token ID | 1816 |
Feature activation | +6.42e-02 |
Pos logit contributions | |
handed | +0.067 |
vendor | +0.055 |
Mike | +0.053 |
sandbox | +0.053 |
amusement | +0.053 |
Neg logit contributions | |
ches | -0.054 |
py | -0.053 |
Tiny | -0.052 |
Cotton | -0.049 |
Ho | -0.049 |
Token | back |
Token ID | 736 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
ches | +0.109 |
Tiny | +0.103 |
py | +0.103 |
Cotton | +0.098 |
Ho | +0.097 |
Neg logit contributions | |
handed | -0.125 |
vendor | -0.096 |
Mike | -0.096 |
lobby | -0.094 |
sandbox | -0.094 |
Token | on |
Token ID | 319 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
handed | +0.302 |
Mike | +0.232 |
vendor | +0.232 |
sandbox | +0.231 |
amusement | +0.229 |
Neg logit contributions | |
ches | -0.242 |
Tiny | -0.223 |
py | -0.217 |
fl | -0.215 |
Ho | -0.211 |
Token | babies |
Token ID | 11903 |
Feature activation | +8.59e-02 |
Pos logit contributions | |
ches | +0.080 |
Tiny | +0.073 |
py | +0.072 |
Cotton | +0.070 |
Ho | +0.069 |
Neg logit contributions | |
handed | -0.078 |
sandbox | -0.060 |
amusement | -0.059 |
vendor | -0.059 |
Mike | -0.059 |
Token | in |
Token ID | 287 |
Feature activation | +5.46e-02 |
Pos logit contributions | |
ches | +0.129 |
Tiny | +0.120 |
py | +0.116 |
fl | +0.113 |
Ho | +0.113 |
Neg logit contributions | |
handed | -0.184 |
vendor | -0.144 |
sandbox | -0.143 |
Mike | -0.143 |
amusement | -0.141 |
Token | their |
Token ID | 511 |
Feature activation | -8.26e-02 |
Pos logit contributions | |
ches | +0.084 |
Tiny | +0.077 |
py | +0.075 |
fl | +0.074 |
Ho | +0.072 |
Neg logit contributions | |
handed | -0.118 |
vendor | -0.092 |
Mike | -0.092 |
sandbox | -0.092 |
amusement | -0.090 |
Token | p |
Token ID | 279 |
Feature activation | -3.64e-02 |
Pos logit contributions | |
handed | +0.173 |
Mike | +0.136 |
Jane | +0.129 |
vendor | +0.129 |
view | +0.127 |
Neg logit contributions | |
ches | -0.134 |
Tiny | -0.126 |
fl | -0.124 |
Ho | -0.119 |
py | -0.119 |
Token | ou |
Token ID | 280 |
Feature activation | +1.52e-02 |
Pos logit contributions | |
handed | +0.074 |
Mike | +0.057 |
vendor | +0.057 |
amusement | +0.057 |
sandbox | +0.056 |
Neg logit contributions | |
ches | -0.061 |
Tiny | -0.054 |
py | -0.054 |
fl | -0.053 |
Ho | -0.051 |
Token | ches |
Token ID | 2052 |
Feature activation | +4.59e-01 |
Pos logit contributions | |
Ho | +0.023 |
Cotton | +0.022 |
cartridge | +0.021 |
Chick | +0.021 |
Tiny | +0.021 |
Neg logit contributions | |
handed | -0.027 |
anymore | -0.022 |
aw | -0.021 |
view | -0.021 |
he | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | -1.26e-03 |
Pos logit contributions | |
ches | +0.680 |
Tiny | +0.620 |
py | +0.603 |
fl | +0.594 |
Ho | +0.582 |
Neg logit contributions | |
handed | -1.008 |
vendor | -0.791 |
Mike | -0.789 |
sandbox | -0.785 |
amusement | -0.780 |
Token | There |
Token ID | 1318 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
handed | +0.002 |
amusement | +0.002 |
vendor | +0.002 |
Mike | +0.002 |
sandbox | +0.002 |
Neg logit contributions | |
ches | -0.002 |
fl | -0.002 |
wand | -0.002 |
py | -0.002 |
tle | -0.002 |
Token | are |
Token ID | 389 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
ches | +0.023 |
Tiny | +0.022 |
fl | +0.021 |
py | +0.021 |
Ho | +0.020 |
Neg logit contributions | |
handed | -0.020 |
Mike | -0.015 |
vendor | -0.014 |
amusement | -0.014 |
view | -0.014 |
Token | ko |
Token ID | 41727 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
handed | +0.428 |
sandbox | +0.342 |
vendor | +0.339 |
Mike | +0.336 |
amusement | +0.335 |
Neg logit contributions | |
ches | -0.309 |
Tiny | -0.278 |
py | -0.271 |
fl | -0.265 |
Cotton | -0.261 |
Token | al |
Token ID | 282 |
Feature activation | +6.39e-02 |
Pos logit contributions | |
ches | +0.253 |
Tiny | +0.233 |
Cotton | +0.227 |
fl | +0.225 |
Ho | +0.218 |
Neg logit contributions | |
handed | -0.269 |
Mike | -0.209 |
vendor | -0.197 |
lobby | -0.195 |
Mike | -0.191 |
Token | learned |
Token ID | 4499 |
Feature activation | -7.38e-02 |
Pos logit contributions | |
handed | +0.134 |
vendor | +0.102 |
Mike | +0.102 |
sandbox | +0.101 |
amusement | +0.100 |
Neg logit contributions | |
ches | -0.112 |
Tiny | -0.104 |
py | -0.101 |
fl | -0.099 |
Ho | -0.098 |
Token | that |
Token ID | 326 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
handed | +0.146 |
sandbox | +0.114 |
Mike | +0.113 |
vendor | +0.112 |
lobby | +0.111 |
Neg logit contributions | |
ches | -0.127 |
Tiny | -0.119 |
fl | -0.116 |
py | -0.113 |
Ho | -0.112 |
Token | if |
Token ID | 611 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
handed | +0.218 |
vendor | +0.169 |
amusement | +0.167 |
sandbox | +0.167 |
Mike | +0.166 |
Neg logit contributions | |
ches | -0.184 |
py | -0.161 |
Tiny | -0.161 |
fl | -0.151 |
Cotton | -0.151 |
Token | something |
Token ID | 1223 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
handed | +0.026 |
vendor | +0.020 |
Mike | +0.020 |
sandbox | +0.020 |
amusement | +0.020 |
Neg logit contributions | |
ches | -0.025 |
py | -0.022 |
Tiny | -0.022 |
fl | -0.021 |
iny | -0.021 |
Token | it |
Token ID | 340 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
handed | +0.413 |
vendor | +0.319 |
sandbox | +0.314 |
Mike | +0.314 |
amusement | +0.309 |
Neg logit contributions | |
ches | -0.351 |
Tiny | -0.326 |
py | -0.323 |
fl | -0.309 |
Ho | -0.307 |
Token | ches |
Token ID | 2052 |
Feature activation | +4.59e-01 |
Pos logit contributions | |
Tiny | +0.025 |
ches | +0.025 |
py | +0.025 |
Cotton | +0.024 |
Ho | +0.023 |
Neg logit contributions | |
handed | -0.030 |
vendor | -0.022 |
sandbox | -0.021 |
lobby | -0.021 |
anymore | -0.021 |
Token | , |
Token ID | 11 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
ches | +0.829 |
Tiny | +0.742 |
fl | +0.738 |
py | +0.733 |
Ho | +0.703 |
Neg logit contributions | |
handed | -0.878 |
Mike | -0.690 |
sandbox | -0.684 |
vendor | -0.680 |
amusement | -0.678 |
Token | it |
Token ID | 340 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
handed | +0.038 |
vendor | +0.029 |
sandbox | +0.029 |
Mike | +0.029 |
amusement | +0.029 |
Neg logit contributions | |
ches | -0.033 |
py | -0.030 |
Tiny | -0.030 |
fl | -0.028 |
Ho | -0.028 |
Token | 's |
Token ID | 338 |
Feature activation | +3.06e-02 |
Pos logit contributions | |
Tiny | +0.025 |
ches | +0.025 |
py | +0.024 |
Cotton | +0.024 |
agons | +0.024 |
Neg logit contributions | |
handed | -0.026 |
sandbox | -0.018 |
anymore | -0.018 |
vendor | -0.018 |
amaz | -0.018 |
Token | best |
Token ID | 1266 |
Feature activation | +4.66e-02 |
Pos logit contributions | |
ches | +0.053 |
Tiny | +0.049 |
py | +0.047 |
arella | +0.047 |
Sparkle | +0.047 |
Neg logit contributions | |
handed | -0.052 |
anymore | -0.041 |
yet | -0.041 |
badly | -0.040 |
unatt | -0.039 |
Token | to |
Token ID | 284 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
ches | +0.078 |
Tiny | +0.072 |
py | +0.070 |
fl | +0.068 |
Ho | +0.068 |
Neg logit contributions | |
handed | -0.093 |
vendor | -0.072 |
sandbox | -0.071 |
Mike | -0.070 |
amusement | -0.070 |
Token | back |
Token ID | 736 |
Feature activation | -1.61e-01 |
Pos logit contributions | |
ches | +0.060 |
Tiny | +0.055 |
fl | +0.054 |
Ho | +0.052 |
py | +0.050 |
Neg logit contributions | |
handed | -0.087 |
view | -0.071 |
sandbox | -0.071 |
amaz | -0.070 |
vendor | -0.069 |
Token | home |
Token ID | 1363 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
handed | +0.357 |
Mike | +0.293 |
amusement | +0.292 |
view | +0.286 |
sandbox | +0.285 |
Neg logit contributions | |
ches | -0.238 |
Tiny | -0.219 |
fl | -0.217 |
py | -0.203 |
Ho | -0.203 |
Token | , |
Token ID | 11 |
Feature activation | -3.02e-02 |
Pos logit contributions | |
handed | +0.342 |
amusement | +0.295 |
Mike | +0.289 |
sandbox | +0.288 |
vendor | +0.286 |
Neg logit contributions | |
ches | -0.193 |
fl | -0.180 |
Tiny | -0.177 |
Ho | -0.165 |
py | -0.161 |
Token | empty |
Token ID | 6565 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
handed | +0.064 |
vendor | +0.050 |
sandbox | +0.050 |
Mike | +0.050 |
amusement | +0.049 |
Neg logit contributions | |
ches | -0.047 |
Tiny | -0.043 |
py | -0.042 |
fl | -0.041 |
Ho | -0.040 |
Token | - |
Token ID | 12 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
handed | +0.429 |
Mike | +0.356 |
amusement | +0.340 |
aren | +0.337 |
Jane | +0.337 |
Neg logit contributions | |
fl | -0.242 |
ches | -0.242 |
Tiny | -0.233 |
py | -0.226 |
wand | -0.222 |
Token | handed |
Token ID | 13638 |
Feature activation | -5.64e-01 |
Pos logit contributions | |
amaz | +0.012 |
amusement | +0.012 |
handed | +0.011 |
sandbox | +0.011 |
lobby | +0.010 |
Neg logit contributions | |
ches | -0.029 |
py | -0.026 |
Tiny | -0.026 |
ch | -0.025 |
iny | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
handed | +1.251 |
amusement | +1.022 |
Mike | +1.008 |
sandbox | +1.003 |
view | +0.997 |
Neg logit contributions | |
ches | -0.827 |
Tiny | -0.752 |
fl | -0.718 |
Ho | -0.714 |
py | -0.711 |
Token |
|
Token ID | 198 |
Feature activation | +4.81e-03 |
Pos logit contributions | |
handed | +0.045 |
sandbox | +0.037 |
amusement | +0.037 |
Mike | +0.036 |
vendor | +0.036 |
Neg logit contributions | |
Tiny | -0.027 |
ches | -0.027 |
Ho | -0.025 |
Cotton | -0.024 |
Pig | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | +4.88e-03 |
Pos logit contributions | |
ches | +0.007 |
Tiny | +0.007 |
py | +0.006 |
fl | +0.006 |
Ho | +0.006 |
Neg logit contributions | |
handed | -0.011 |
vendor | -0.008 |
Mike | -0.008 |
sandbox | -0.008 |
amusement | -0.008 |
Token | Nature |
Token ID | 46934 |
Feature activation | +6.59e-02 |
Pos logit contributions | |
ches | +0.010 |
Tiny | +0.009 |
py | +0.009 |
fl | +0.009 |
Ho | +0.008 |
Neg logit contributions | |
handed | -0.009 |
Mike | -0.006 |
vendor | -0.006 |
sandbox | -0.006 |
amusement | -0.006 |
Token | was |
Token ID | 373 |
Feature activation | -9.56e-02 |
Pos logit contributions | |
ches | +0.108 |
Tiny | +0.097 |
fl | +0.097 |
py | +0.095 |
Ho | +0.091 |
Neg logit contributions | |
handed | -0.138 |
Mike | -0.107 |
vendor | -0.107 |
sandbox | -0.106 |
amusement | -0.105 |
Token | he |
Token ID | 339 |
Feature activation | -3.37e-02 |
Pos logit contributions | |
handed | +0.002 |
view | +0.002 |
apartment | +0.002 |
sandbox | +0.002 |
vendor | +0.002 |
Neg logit contributions | |
Tiny | -0.001 |
ches | -0.001 |
fl | -0.001 |
Ho | -0.001 |
py | -0.001 |
Token | was |
Token ID | 373 |
Feature activation | -9.24e-02 |
Pos logit contributions | |
handed | +0.088 |
sandbox | +0.073 |
amusement | +0.071 |
Mike | +0.071 |
lobby | +0.071 |
Neg logit contributions | |
ches | -0.041 |
Tiny | -0.033 |
fl | -0.032 |
py | -0.032 |
Ho | -0.030 |
Token | caught |
Token ID | 4978 |
Feature activation | -6.49e-02 |
Pos logit contributions | |
handed | +0.181 |
vendor | +0.139 |
Mike | +0.136 |
amusement | +0.133 |
unatt | +0.130 |
Neg logit contributions | |
ches | -0.165 |
Tiny | -0.145 |
py | -0.140 |
Chick | -0.140 |
iny | -0.138 |
Token | red |
Token ID | 2266 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
handed | +0.159 |
sandbox | +0.144 |
apartment | +0.140 |
lobby | +0.139 |
vendor | +0.136 |
Neg logit contributions | |
ches | -0.070 |
fl | -0.066 |
Tiny | -0.061 |
Ho | -0.060 |
hopping | -0.059 |
Token | - |
Token ID | 12 |
Feature activation | -8.12e-03 |
Pos logit contributions | |
ches | +0.187 |
Tiny | +0.168 |
fl | +0.166 |
Ho | +0.165 |
py | +0.165 |
Neg logit contributions | |
handed | -0.287 |
Mike | -0.227 |
vendor | -0.225 |
sandbox | -0.223 |
amusement | -0.220 |
Token | handed |
Token ID | 13638 |
Feature activation | -5.62e-01 |
Pos logit contributions | |
handed | +0.016 |
sandbox | +0.013 |
amusement | +0.012 |
vendor | +0.012 |
Mike | +0.012 |
Neg logit contributions | |
ches | -0.014 |
Tiny | -0.013 |
py | -0.012 |
fl | -0.012 |
Ho | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
handed | +1.462 |
Mike | +1.246 |
aren | +1.225 |
lobby | +1.215 |
view | +1.215 |
Neg logit contributions | |
ches | -0.564 |
fl | -0.501 |
wand | -0.488 |
Tiny | -0.482 |
Ho | -0.473 |
Token | But |
Token ID | 887 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
handed | +0.038 |
sandbox | +0.030 |
amusement | +0.030 |
Mike | +0.030 |
vendor | +0.030 |
Neg logit contributions | |
ches | -0.021 |
Tiny | -0.020 |
Ho | -0.019 |
py | -0.018 |
Cotton | -0.018 |
Token | he |
Token ID | 339 |
Feature activation | -3.45e-02 |
Pos logit contributions | |
Tiny | +0.028 |
fl | +0.027 |
ches | +0.027 |
Ho | +0.025 |
hopping | +0.025 |
Neg logit contributions | |
handed | -0.062 |
view | -0.050 |
sandbox | -0.049 |
lobby | -0.048 |
apartment | -0.048 |
Token | didn |
Token ID | 1422 |
Feature activation | +4.67e-02 |
Pos logit contributions | |
handed | +0.085 |
sandbox | +0.067 |
lobby | +0.067 |
view | +0.066 |
amusement | +0.065 |
Neg logit contributions | |
ches | -0.048 |
fl | -0.043 |
Tiny | -0.041 |
Ho | -0.039 |
py | -0.039 |
Token | 't |
Token ID | 470 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
ches | +0.056 |
py | +0.053 |
Tiny | +0.052 |
wand | +0.050 |
fl | +0.050 |
Neg logit contributions | |
handed | -0.112 |
Mike | -0.093 |
he | -0.090 |
amusement | -0.089 |
vendor | -0.088 |
Token | want |
Token ID | 765 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
ches | +0.065 |
Tiny | +0.060 |
py | +0.059 |
fl | +0.058 |
Ho | +0.057 |
Neg logit contributions | |
handed | -0.073 |
vendor | -0.056 |
Mike | -0.055 |
sandbox | -0.055 |
amusement | -0.055 |
Token | to |
Token ID | 284 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
handed | +0.069 |
sandbox | +0.056 |
Mike | +0.055 |
lobby | +0.055 |
amusement | +0.054 |
Neg logit contributions | |
ches | -0.047 |
Tiny | -0.044 |
py | -0.041 |
Ho | -0.041 |
Sparkle | -0.041 |
Token | go |
Token ID | 467 |
Feature activation | +4.21e-02 |
Pos logit contributions | |
handed | +0.028 |
sandbox | +0.023 |
Mike | +0.023 |
lobby | +0.023 |
vendor | +0.022 |
Neg logit contributions | |
ches | -0.018 |
Tiny | -0.017 |
py | -0.016 |
fl | -0.016 |
Ho | -0.016 |
Token | empty |
Token ID | 6565 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
ches | +0.063 |
Tiny | +0.057 |
py | +0.055 |
fl | +0.054 |
Ho | +0.053 |
Neg logit contributions | |
handed | -0.092 |
Mike | -0.075 |
sandbox | -0.074 |
vendor | -0.072 |
amusement | -0.072 |
Token | - |
Token ID | 12 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
handed | +0.357 |
Mike | +0.302 |
sandbox | +0.283 |
aren | +0.283 |
view | +0.282 |
Neg logit contributions | |
ches | -0.223 |
Tiny | -0.210 |
fl | -0.207 |
py | -0.205 |
hopping | -0.203 |
Token | handed |
Token ID | 13638 |
Feature activation | -5.41e-01 |
Pos logit contributions | |
ches | +0.031 |
Tiny | +0.029 |
Cotton | +0.028 |
py | +0.028 |
iny | +0.028 |
Neg logit contributions | |
handed | -0.003 |
sandbox | -0.003 |
amaz | -0.003 |
amusement | -0.002 |
lobby | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
handed | +1.261 |
Mike | +1.087 |
amusement | +1.064 |
sandbox | +1.061 |
view | +1.056 |
Neg logit contributions | |
ches | -0.694 |
Tiny | -0.620 |
fl | -0.589 |
Sparkle | -0.586 |
py | -0.580 |
Token | so |
Token ID | 523 |
Feature activation | -9.58e-02 |
Pos logit contributions | |
handed | +0.025 |
Mike | +0.019 |
lobby | +0.019 |
view | +0.019 |
sandbox | +0.018 |
Neg logit contributions | |
ches | -0.016 |
Tiny | -0.016 |
fl | -0.015 |
Ho | -0.015 |
Cotton | -0.014 |
Token | she |
Token ID | 673 |
Feature activation | +7.78e-03 |
Pos logit contributions | |
handed | +0.210 |
sandbox | +0.165 |
amusement | +0.164 |
view | +0.162 |
Mike | +0.162 |
Neg logit contributions | |
ches | -0.141 |
Tiny | -0.137 |
fl | -0.129 |
Ho | -0.128 |
Cotton | -0.125 |
Token | sque |
Token ID | 35776 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
ches | +0.011 |
py | +0.011 |
Tiny | +0.011 |
iny | +0.010 |
Ho | +0.010 |
Neg logit contributions | |
handed | -0.016 |
vendor | -0.013 |
Mike | -0.013 |
sandbox | -0.012 |
amusement | -0.012 |
Token | aked |
Token ID | 4335 |
Feature activation | +2.51e-02 |
Pos logit contributions | |
ches | +0.254 |
py | +0.228 |
Tiny | +0.225 |
fl | +0.220 |
Ho | +0.212 |
Neg logit contributions | |
handed | -0.361 |
vendor | -0.288 |
sandbox | -0.283 |
amusement | -0.282 |
Mike | -0.282 |
Token | came |
Token ID | 1625 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
handed | +0.095 |
vendor | +0.074 |
Mike | +0.074 |
sandbox | +0.074 |
amusement | +0.073 |
Neg logit contributions | |
ches | -0.066 |
Tiny | -0.060 |
py | -0.058 |
fl | -0.057 |
Ho | -0.056 |
Token | back |
Token ID | 736 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
ches | +0.189 |
Tiny | +0.172 |
py | +0.168 |
fl | +0.164 |
Ho | +0.163 |
Neg logit contributions | |
handed | -0.248 |
vendor | -0.194 |
Mike | -0.192 |
sandbox | -0.191 |
amusement | -0.189 |
Token | home |
Token ID | 1363 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
handed | +0.331 |
sandbox | +0.273 |
Mike | +0.266 |
amusement | +0.262 |
lobby | +0.257 |
Neg logit contributions | |
ches | -0.197 |
Tiny | -0.190 |
fl | -0.182 |
Ho | -0.175 |
py | -0.173 |
Token | empty |
Token ID | 6565 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
handed | +0.284 |
Mike | +0.238 |
sandbox | +0.238 |
amusement | +0.237 |
vendor | +0.230 |
Neg logit contributions | |
ches | -0.162 |
fl | -0.150 |
Tiny | -0.147 |
py | -0.140 |
Ho | -0.138 |
Token | - |
Token ID | 12 |
Feature activation | +1.72e-02 |
Pos logit contributions | |
handed | +0.371 |
Mike | +0.315 |
amusement | +0.300 |
sandbox | +0.295 |
vendor | +0.294 |
Neg logit contributions | |
ches | -0.181 |
fl | -0.177 |
Tiny | -0.167 |
wand | -0.163 |
py | -0.161 |
Token | handed |
Token ID | 13638 |
Feature activation | -5.36e-01 |
Pos logit contributions | |
ches | +0.051 |
iny | +0.047 |
py | +0.046 |
tle | +0.045 |
fl | +0.045 |
Neg logit contributions | |
amaz | -0.010 |
amusement | -0.006 |
sandbox | -0.006 |
handed | -0.003 |
Mike | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | +3.44e-02 |
Pos logit contributions | |
handed | +1.265 |
Mike | +1.129 |
amusement | +1.109 |
sandbox | +1.080 |
amaz | +1.078 |
Neg logit contributions | |
ches | -0.654 |
Tiny | -0.552 |
fl | -0.551 |
py | -0.529 |
Cotton | -0.514 |
Token | told |
Token ID | 1297 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
fl | +0.044 |
ches | +0.044 |
Tiny | +0.042 |
Ho | +0.040 |
Sparkle | +0.039 |
Neg logit contributions | |
handed | -0.078 |
Mike | -0.065 |
amusement | -0.061 |
apartment | -0.061 |
aw | -0.061 |
Token | Lily |
Token ID | 20037 |
Feature activation | +4.23e-02 |
Pos logit contributions | |
ches | +0.281 |
Tiny | +0.273 |
Ho | +0.254 |
py | +0.250 |
fl | +0.249 |
Neg logit contributions | |
handed | -0.532 |
sandbox | -0.439 |
Mike | -0.427 |
vendor | -0.424 |
lobby | -0.417 |
Token | the |
Token ID | 262 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
ches | +0.064 |
Tiny | +0.057 |
py | +0.053 |
Sparkle | +0.053 |
fl | +0.052 |
Neg logit contributions | |
handed | -0.095 |
Mike | -0.080 |
sandbox | -0.078 |
amusement | -0.077 |
vendor | -0.077 |
Token | news |
Token ID | 1705 |
Feature activation | -7.27e-02 |
Pos logit contributions | |
ches | +0.016 |
fl | +0.015 |
Tiny | +0.015 |
py | +0.015 |
wand | +0.015 |
Neg logit contributions | |
handed | -0.021 |
Mike | -0.017 |
sandbox | -0.016 |
amusement | -0.016 |
Jane | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
handed | +0.060 |
Mike | +0.047 |
vendor | +0.045 |
amusement | +0.045 |
view | +0.044 |
Neg logit contributions | |
ches | -0.043 |
Tiny | -0.040 |
fl | -0.039 |
py | -0.038 |
Ho | -0.038 |
Token | swing |
Token ID | 9628 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
handed | +0.030 |
Mike | +0.022 |
ill | +0.022 |
Jane | +0.021 |
view | +0.020 |
Neg logit contributions | |
fl | -0.027 |
wand | -0.026 |
Ho | -0.025 |
wool | -0.025 |
Tiny | -0.025 |
Token | , |
Token ID | 11 |
Feature activation | -2.79e-02 |
Pos logit contributions | |
handed | +0.362 |
Mike | +0.280 |
vendor | +0.279 |
amusement | +0.277 |
sandbox | +0.275 |
Neg logit contributions | |
ches | -0.264 |
Tiny | -0.241 |
py | -0.235 |
fl | -0.234 |
Ho | -0.228 |
Token | and |
Token ID | 290 |
Feature activation | -2.60e-04 |
Pos logit contributions | |
handed | +0.056 |
sandbox | +0.045 |
vendor | +0.045 |
Mike | +0.043 |
amusement | +0.043 |
Neg logit contributions | |
ches | -0.046 |
Tiny | -0.042 |
py | -0.041 |
fl | -0.039 |
Ho | -0.039 |
Token | the |
Token ID | 262 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
handed | +0.001 |
Mike | +0.000 |
vendor | +0.000 |
sandbox | +0.000 |
amusement | +0.000 |
Neg logit contributions | |
ches | -0.000 |
Tiny | -0.000 |
fl | -0.000 |
py | -0.000 |
Ho | -0.000 |
Token | sandbox |
Token ID | 35204 |
Feature activation | -4.56e-01 |
Pos logit contributions | |
handed | +0.034 |
Mike | +0.025 |
Jane | +0.024 |
ill | +0.024 |
view | +0.023 |
Neg logit contributions | |
fl | -0.031 |
wand | -0.029 |
Ho | -0.029 |
Tiny | -0.028 |
ches | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
handed | +1.118 |
Mike | +0.901 |
amusement | +0.881 |
vendor | +0.872 |
view | +0.857 |
Neg logit contributions | |
ches | -0.583 |
fl | -0.553 |
py | -0.527 |
wand | -0.519 |
Tiny | -0.507 |
Token | They |
Token ID | 1119 |
Feature activation | +2.42e-02 |
Pos logit contributions | |
handed | +0.031 |
vendor | +0.025 |
Mike | +0.025 |
sandbox | +0.024 |
amusement | +0.024 |
Neg logit contributions | |
ches | -0.020 |
Tiny | -0.018 |
py | -0.017 |
Ho | -0.017 |
fl | -0.017 |
Token | are |
Token ID | 389 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
ches | +0.038 |
fl | +0.035 |
Tiny | +0.034 |
Ho | +0.034 |
py | +0.033 |
Neg logit contributions | |
handed | -0.053 |
Mike | -0.041 |
vendor | -0.039 |
Jane | -0.039 |
view | -0.039 |
Token | calm |
Token ID | 9480 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
handed | +0.410 |
sandbox | +0.330 |
vendor | +0.322 |
Mike | +0.319 |
amusement | +0.313 |
Neg logit contributions | |
ches | -0.383 |
Tiny | -0.335 |
py | -0.328 |
Ho | -0.317 |
Cotton | -0.317 |
Token | and |
Token ID | 290 |
Feature activation | -1.47e-03 |
Pos logit contributions | |
handed | +0.391 |
Mike | +0.328 |
amusement | +0.317 |
lobby | +0.314 |
Jane | +0.313 |
Neg logit contributions | |
fl | -0.098 |
Tiny | -0.094 |
Ho | -0.085 |
wool | -0.083 |
hopping | -0.082 |
Token | a |
Token ID | 257 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
handed | +0.070 |
Mike | +0.053 |
apartment | +0.052 |
vendor | +0.052 |
view | +0.051 |
Neg logit contributions | |
fl | -0.038 |
ches | -0.038 |
Tiny | -0.036 |
Ho | -0.034 |
Sparkle | -0.034 |
Token | swing |
Token ID | 9628 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
handed | +0.047 |
view | +0.032 |
Mike | +0.032 |
ill | +0.031 |
unatt | +0.030 |
Neg logit contributions | |
fl | -0.042 |
wand | -0.039 |
ches | -0.038 |
Tiny | -0.038 |
Ho | -0.037 |
Token | , |
Token ID | 11 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
handed | +0.357 |
Mike | +0.275 |
vendor | +0.272 |
amusement | +0.271 |
sandbox | +0.270 |
Neg logit contributions | |
ches | -0.261 |
Tiny | -0.237 |
fl | -0.232 |
py | -0.231 |
Ho | -0.224 |
Token | and |
Token ID | 290 |
Feature activation | -6.03e-03 |
Pos logit contributions | |
handed | +0.068 |
Mike | +0.054 |
vendor | +0.053 |
amusement | +0.052 |
sandbox | +0.051 |
Neg logit contributions | |
ches | -0.041 |
fl | -0.038 |
Tiny | -0.037 |
py | -0.036 |
Ho | -0.035 |
Token | a |
Token ID | 257 |
Feature activation | -2.38e-02 |
Pos logit contributions | |
handed | +0.016 |
view | +0.012 |
unatt | +0.012 |
Mike | +0.012 |
apartment | +0.012 |
Neg logit contributions | |
fl | -0.007 |
Tiny | -0.006 |
ches | -0.006 |
Ho | -0.006 |
Sparkle | -0.006 |
Token | sandbox |
Token ID | 35204 |
Feature activation | -4.55e-01 |
Pos logit contributions | |
handed | +0.043 |
view | +0.028 |
Mike | +0.028 |
unatt | +0.027 |
Jane | +0.026 |
Neg logit contributions | |
fl | -0.048 |
wand | -0.045 |
Tiny | -0.044 |
ches | -0.044 |
Ho | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
handed | +1.039 |
Mike | +0.821 |
amusement | +0.800 |
vendor | +0.792 |
sandbox | +0.775 |
Neg logit contributions | |
ches | -0.631 |
fl | -0.586 |
py | -0.570 |
Tiny | -0.563 |
wand | -0.552 |
Token |
|
Token ID | 198 |
Feature activation | +7.59e-03 |
Pos logit contributions | |
handed | +0.034 |
vendor | +0.027 |
sandbox | +0.026 |
Mike | +0.026 |
amusement | +0.026 |
Neg logit contributions | |
ches | -0.029 |
py | -0.027 |
fl | -0.026 |
Tiny | -0.026 |
wand | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | +7.56e-03 |
Pos logit contributions | |
ches | +0.012 |
py | +0.012 |
Tiny | +0.012 |
fl | +0.011 |
Ho | +0.011 |
Neg logit contributions | |
handed | -0.015 |
Mike | -0.012 |
vendor | -0.012 |
sandbox | -0.012 |
amusement | -0.012 |
Token | L |
Token ID | 43 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
py | +0.014 |
ches | +0.013 |
Tiny | +0.013 |
wand | +0.012 |
Sparkle | +0.012 |
Neg logit contributions | |
handed | -0.014 |
Mike | -0.012 |
Jane | -0.011 |
sandbox | -0.011 |
vendor | -0.011 |
Token | ily |
Token ID | 813 |
Feature activation | -3.08e-02 |
Pos logit contributions | |
handed | +0.038 |
vendor | +0.029 |
Mike | +0.029 |
sandbox | +0.028 |
amusement | +0.028 |
Neg logit contributions | |
ches | -0.033 |
Tiny | -0.030 |
py | -0.030 |
fl | -0.029 |
Ho | -0.029 |
Token | noise |
Token ID | 7838 |
Feature activation | -4.82e-02 |
Pos logit contributions | |
handed | +0.049 |
Mike | +0.038 |
vendor | +0.036 |
sandbox | +0.036 |
amusement | +0.036 |
Neg logit contributions | |
ches | -0.039 |
Tiny | -0.036 |
fl | -0.036 |
py | -0.035 |
Ho | -0.034 |
Token | and |
Token ID | 290 |
Feature activation | -5.69e-04 |
Pos logit contributions | |
handed | +0.132 |
sandbox | +0.108 |
vendor | +0.105 |
Mike | +0.104 |
aren | +0.103 |
Neg logit contributions | |
ches | -0.052 |
fl | -0.046 |
Tiny | -0.042 |
py | -0.040 |
tle | -0.037 |
Token | came |
Token ID | 1625 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
handed | +0.001 |
amusement | +0.001 |
view | +0.001 |
vendor | +0.001 |
sandbox | +0.001 |
Neg logit contributions | |
ches | -0.001 |
Tiny | -0.001 |
fl | -0.001 |
py | -0.001 |
Ho | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | -3.85e-02 |
Pos logit contributions | |
ches | +0.169 |
Tiny | +0.154 |
py | +0.151 |
fl | +0.147 |
Cotton | +0.147 |
Neg logit contributions | |
handed | -0.207 |
Mike | -0.162 |
vendor | -0.159 |
sandbox | -0.158 |
amusement | -0.157 |
Token | the |
Token ID | 262 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
handed | +0.085 |
aw | +0.072 |
sandbox | +0.070 |
view | +0.069 |
lobby | +0.068 |
Neg logit contributions | |
Tiny | -0.052 |
fl | -0.048 |
Pig | -0.047 |
Ho | -0.046 |
asking | -0.045 |
Token | sandbox |
Token ID | 35204 |
Feature activation | -4.55e-01 |
Pos logit contributions | |
handed | +0.049 |
Mike | +0.045 |
Jane | +0.043 |
ill | +0.043 |
aren | +0.042 |
Neg logit contributions | |
fl | -0.029 |
ches | -0.029 |
wool | -0.028 |
Tiny | -0.028 |
Ho | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
handed | +1.237 |
Mike | +1.086 |
aren | +1.084 |
amusement | +1.052 |
Whe | +1.046 |
Neg logit contributions | |
fl | -0.308 |
ches | -0.306 |
asking | -0.289 |
hopping | -0.259 |
wand | -0.241 |
Token | She |
Token ID | 1375 |
Feature activation | -4.61e-02 |
Pos logit contributions | |
handed | +0.032 |
view | +0.027 |
amusement | +0.027 |
unatt | +0.027 |
Mike | +0.027 |
Neg logit contributions | |
Tiny | -0.008 |
Ho | -0.007 |
Po | -0.007 |
Sparkle | -0.006 |
Cotton | -0.006 |
Token | saw |
Token ID | 2497 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
handed | +0.092 |
amusement | +0.067 |
view | +0.066 |
sandbox | +0.066 |
vendor | +0.065 |
Neg logit contributions | |
ches | -0.082 |
Tiny | -0.076 |
fl | -0.075 |
py | -0.071 |
Ho | -0.071 |
Token | Tom |
Token ID | 4186 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
handed | +0.314 |
view | +0.251 |
sandbox | +0.251 |
amusement | +0.247 |
apartment | +0.246 |
Neg logit contributions | |
ches | -0.164 |
Tiny | -0.154 |
Ho | -0.149 |
fl | -0.144 |
Cotton | -0.140 |
Token | and |
Token ID | 290 |
Feature activation | -1.16e-03 |
Pos logit contributions | |
handed | +0.525 |
vendor | +0.417 |
Mike | +0.416 |
sandbox | +0.413 |
amusement | +0.411 |
Neg logit contributions | |
ches | -0.342 |
Tiny | -0.307 |
fl | -0.301 |
py | -0.296 |
Ho | -0.288 |
Token | benches |
Token ID | 43183 |
Feature activation | -3.66e-01 |
Pos logit contributions | |
handed | +0.082 |
view | +0.066 |
Mike | +0.066 |
vendor | +0.065 |
lobby | +0.063 |
Neg logit contributions | |
Tiny | -0.040 |
ches | -0.036 |
fl | -0.035 |
Ho | -0.034 |
Sparkle | -0.033 |
Token | , |
Token ID | 11 |
Feature activation | -2.89e-02 |
Pos logit contributions | |
handed | +0.921 |
view | +0.737 |
vendor | +0.731 |
Mike | +0.723 |
amusement | +0.721 |
Neg logit contributions | |
ches | -0.461 |
fl | -0.412 |
Tiny | -0.411 |
hopping | -0.380 |
Ho | -0.369 |
Token | and |
Token ID | 290 |
Feature activation | -4.58e-03 |
Pos logit contributions | |
handed | +0.062 |
Mike | +0.048 |
vendor | +0.048 |
sandbox | +0.047 |
amusement | +0.047 |
Neg logit contributions | |
ches | -0.045 |
Tiny | -0.041 |
py | -0.040 |
fl | -0.040 |
Ho | -0.039 |
Token | in |
Token ID | 287 |
Feature activation | +3.74e-02 |
Pos logit contributions | |
handed | +0.011 |
Mike | +0.009 |
aren | +0.008 |
vendor | +0.008 |
apartment | +0.008 |
Neg logit contributions | |
ches | -0.006 |
Tiny | -0.006 |
fl | -0.006 |
Ho | -0.006 |
Sparkle | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
Tiny | +0.042 |
ches | +0.039 |
fl | +0.037 |
Sparkle | +0.036 |
Ho | +0.035 |
Neg logit contributions | |
handed | -0.103 |
view | -0.083 |
Mike | -0.082 |
vendor | -0.080 |
Jane | -0.077 |
Token | sandbox |
Token ID | 35204 |
Feature activation | -4.55e-01 |
Pos logit contributions | |
handed | +0.040 |
Mike | +0.034 |
ill | +0.031 |
Jane | +0.031 |
aren | +0.031 |
Neg logit contributions | |
fl | -0.016 |
wool | -0.016 |
Tiny | -0.015 |
Ho | -0.015 |
ches | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
handed | +1.179 |
vendor | +0.983 |
amusement | +0.969 |
view | +0.953 |
Mike | +0.946 |
Neg logit contributions | |
fl | -0.478 |
ches | -0.473 |
Tiny | -0.430 |
wand | -0.424 |
py | -0.401 |
Token | They |
Token ID | 1119 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
handed | +0.035 |
vendor | +0.028 |
Mike | +0.028 |
sandbox | +0.028 |
amusement | +0.027 |
Neg logit contributions | |
ches | -0.023 |
Tiny | -0.021 |
Ho | -0.020 |
py | -0.020 |
fl | -0.019 |
Token | asked |
Token ID | 1965 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
ches | +0.031 |
fl | +0.030 |
Tiny | +0.029 |
Ho | +0.028 |
Pig | +0.027 |
Neg logit contributions | |
handed | -0.045 |
view | -0.034 |
vendor | -0.034 |
Mike | -0.033 |
amusement | -0.033 |
Token | other |
Token ID | 584 |
Feature activation | +7.27e-02 |
Pos logit contributions | |
ches | +0.281 |
Tiny | +0.273 |
Ho | +0.247 |
Sparkle | +0.237 |
py | +0.236 |
Neg logit contributions | |
handed | -0.554 |
Mike | -0.443 |
sandbox | -0.441 |
view | -0.440 |
vendor | -0.438 |
Token | people |
Token ID | 661 |
Feature activation | -3.26e-02 |
Pos logit contributions | |
ches | +0.086 |
Tiny | +0.080 |
py | +0.074 |
Ho | +0.073 |
fl | +0.072 |
Neg logit contributions | |
handed | -0.186 |
Mike | -0.151 |
sandbox | -0.148 |
vendor | -0.148 |
view | -0.147 |
Token |
|
Token ID | 198 |
Feature activation | +9.91e-03 |
Pos logit contributions | |
ches | +0.015 |
Tiny | +0.014 |
py | +0.013 |
fl | +0.013 |
Ho | +0.013 |
Neg logit contributions | |
handed | -0.022 |
vendor | -0.018 |
sandbox | -0.018 |
Mike | -0.018 |
amusement | -0.017 |
Token | He |
Token ID | 1544 |
Feature activation | -7.01e-02 |
Pos logit contributions | |
ches | +0.017 |
Tiny | +0.015 |
fl | +0.015 |
py | +0.014 |
iny | +0.014 |
Neg logit contributions | |
handed | -0.020 |
amaz | -0.015 |
amusement | -0.015 |
lobby | -0.015 |
vendor | -0.015 |
Token | ran |
Token ID | 4966 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
handed | +0.142 |
sandbox | +0.119 |
vendor | +0.117 |
Mike | +0.114 |
amusement | +0.113 |
Neg logit contributions | |
ches | -0.103 |
Tiny | -0.100 |
py | -0.100 |
Cotton | -0.093 |
Pig | -0.091 |
Token | to |
Token ID | 284 |
Feature activation | -3.80e-02 |
Pos logit contributions | |
ches | +0.050 |
Tiny | +0.045 |
fl | +0.044 |
Ho | +0.043 |
Sparkle | +0.041 |
Neg logit contributions | |
handed | -0.100 |
view | -0.084 |
amaz | -0.083 |
sandbox | -0.081 |
amusement | -0.081 |
Token | the |
Token ID | 262 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
handed | +0.079 |
sandbox | +0.061 |
Mike | +0.061 |
vendor | +0.061 |
amusement | +0.060 |
Neg logit contributions | |
ches | -0.061 |
Tiny | -0.057 |
py | -0.055 |
fl | -0.054 |
Ho | -0.054 |
Token | sandbox |
Token ID | 35204 |
Feature activation | -4.54e-01 |
Pos logit contributions | |
handed | +0.033 |
Mike | +0.025 |
aren | +0.024 |
amusement | +0.023 |
view | +0.023 |
Neg logit contributions | |
fl | -0.036 |
ches | -0.036 |
wand | -0.035 |
Tiny | -0.035 |
Ho | -0.034 |
Token | and |
Token ID | 290 |
Feature activation | -5.84e-03 |
Pos logit contributions | |
handed | +1.233 |
Mike | +1.060 |
aren | +1.039 |
amusement | +1.007 |
vendor | +0.985 |
Neg logit contributions | |
fl | -0.430 |
ches | -0.364 |
wand | -0.357 |
hopping | -0.347 |
asking | -0.327 |
Token | dug |
Token ID | 18735 |
Feature activation | +4.15e-02 |
Pos logit contributions | |
handed | +0.013 |
sandbox | +0.010 |
vendor | +0.010 |
Mike | +0.010 |
amusement | +0.010 |
Neg logit contributions | |
ches | -0.008 |
py | -0.008 |
Tiny | -0.008 |
Cotton | -0.007 |
iny | -0.007 |
Token | with |
Token ID | 351 |
Feature activation | -6.73e-02 |
Pos logit contributions | |
ches | +0.053 |
fl | +0.049 |
Ho | +0.047 |
Tiny | +0.046 |
wand | +0.046 |
Neg logit contributions | |
handed | -0.105 |
vendor | -0.085 |
aren | -0.083 |
amusement | -0.083 |
view | -0.082 |
Token | his |
Token ID | 465 |
Feature activation | -3.64e-03 |
Pos logit contributions | |
handed | +0.164 |
vendor | +0.132 |
view | +0.132 |
Mike | +0.129 |
amusement | +0.129 |
Neg logit contributions | |
ches | -0.087 |
Tiny | -0.086 |
Ho | -0.080 |
fl | -0.079 |
Pig | -0.076 |
Token | hands |
Token ID | 2832 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
handed | +0.007 |
sandbox | +0.006 |
vendor | +0.005 |
lobby | +0.005 |
amusement | +0.005 |
Neg logit contributions | |
ches | -0.006 |
py | -0.006 |
Tiny | -0.006 |
Cotton | -0.006 |
Sparkle | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
handed | +0.452 |
Mike | +0.355 |
amusement | +0.349 |
vendor | +0.347 |
sandbox | +0.343 |
Neg logit contributions | |
ches | -0.279 |
Tiny | -0.251 |
fl | -0.250 |
py | -0.241 |
Ho | -0.236 |
Token | the |
Token ID | 262 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
handed | +0.053 |
Mike | +0.039 |
amusement | +0.037 |
amaz | +0.037 |
vendor | +0.036 |
Neg logit contributions | |
ches | -0.050 |
fl | -0.046 |
Tiny | -0.045 |
Ho | -0.044 |
py | -0.044 |
Token | slide |
Token ID | 10649 |
Feature activation | -2.95e-01 |
Pos logit contributions | |
handed | +0.032 |
amusement | +0.022 |
amaz | +0.022 |
Mike | +0.021 |
lobby | +0.020 |
Neg logit contributions | |
ches | -0.039 |
fl | -0.038 |
py | -0.036 |
Tiny | -0.036 |
Ho | -0.035 |
Token | and |
Token ID | 290 |
Feature activation | -3.17e-03 |
Pos logit contributions | |
handed | +0.643 |
Mike | +0.498 |
amusement | +0.494 |
vendor | +0.494 |
sandbox | +0.491 |
Neg logit contributions | |
ches | -0.456 |
Tiny | -0.413 |
fl | -0.403 |
py | -0.401 |
Ho | -0.389 |
Token | the |
Token ID | 262 |
Feature activation | -1.97e-02 |
Pos logit contributions | |
handed | +0.007 |
Mike | +0.006 |
view | +0.005 |
apartment | +0.005 |
unatt | +0.005 |
Neg logit contributions | |
fl | -0.005 |
ches | -0.005 |
Tiny | -0.004 |
Ho | -0.004 |
Sparkle | -0.004 |
Token | sandbox |
Token ID | 35204 |
Feature activation | -4.54e-01 |
Pos logit contributions | |
handed | +0.036 |
amusement | +0.025 |
amaz | +0.024 |
Mike | +0.024 |
apartment | +0.023 |
Neg logit contributions | |
fl | -0.039 |
ches | -0.036 |
wand | -0.034 |
Ho | -0.034 |
py | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
handed | +1.083 |
Mike | +0.869 |
amusement | +0.854 |
vendor | +0.841 |
aren | +0.822 |
Neg logit contributions | |
ches | -0.587 |
fl | -0.568 |
py | -0.524 |
wand | -0.516 |
Tiny | -0.510 |
Token | They |
Token ID | 1119 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
handed | +0.022 |
sandbox | +0.019 |
vendor | +0.018 |
lobby | +0.018 |
amusement | +0.018 |
Neg logit contributions | |
ches | -0.026 |
py | -0.026 |
fl | -0.023 |
Tiny | -0.023 |
tle | -0.023 |
Token | saw |
Token ID | 2497 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
ches | +0.036 |
Tiny | +0.033 |
py | +0.032 |
fl | +0.032 |
Ho | +0.032 |
Neg logit contributions | |
handed | -0.042 |
Mike | -0.032 |
vendor | -0.032 |
sandbox | -0.032 |
amusement | -0.032 |
Token | some |
Token ID | 617 |
Feature activation | -2.32e-03 |
Pos logit contributions | |
handed | +0.392 |
har | +0.334 |
aren | +0.317 |
apartment | +0.314 |
unatt | +0.310 |
Neg logit contributions | |
fl | -0.101 |
wool | -0.083 |
Ho | -0.082 |
Tiny | -0.082 |
carrots | -0.075 |
Token | flowers |
Token ID | 12734 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
handed | +0.006 |
unatt | +0.005 |
amaz | +0.005 |
Mike | +0.004 |
aren | +0.004 |
Neg logit contributions | |
fl | -0.003 |
wool | -0.002 |
ches | -0.002 |
py | -0.002 |
hopping | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -3.09e-02 |
Pos logit contributions | |
handed | +0.467 |
Mike | +0.367 |
amusement | +0.360 |
vendor | +0.355 |
view | +0.353 |
Neg logit contributions | |
ches | -0.268 |
fl | -0.240 |
Tiny | -0.239 |
py | -0.229 |
Ho | -0.225 |
Token | the |
Token ID | 262 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
handed | +0.060 |
Mike | +0.044 |
amaz | +0.043 |
amusement | +0.043 |
view | +0.042 |
Neg logit contributions | |
ches | -0.056 |
fl | -0.052 |
Tiny | -0.051 |
Ho | -0.049 |
py | -0.049 |
Token | slide |
Token ID | 10649 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
handed | +0.028 |
amusement | +0.019 |
Mike | +0.018 |
amaz | +0.018 |
view | +0.017 |
Neg logit contributions | |
ches | -0.040 |
fl | -0.039 |
py | -0.037 |
Tiny | -0.037 |
wand | -0.036 |
Token | and |
Token ID | 290 |
Feature activation | -6.00e-03 |
Pos logit contributions | |
handed | +0.635 |
Mike | +0.491 |
vendor | +0.488 |
amusement | +0.487 |
sandbox | +0.485 |
Neg logit contributions | |
ches | -0.458 |
Tiny | -0.416 |
py | -0.405 |
fl | -0.404 |
Ho | -0.393 |
Token | the |
Token ID | 262 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
handed | +0.015 |
Mike | +0.011 |
view | +0.011 |
apartment | +0.011 |
unatt | +0.010 |
Neg logit contributions | |
fl | -0.008 |
ches | -0.008 |
Tiny | -0.008 |
Ho | -0.007 |
Sparkle | -0.007 |
Token | sandbox |
Token ID | 35204 |
Feature activation | -4.54e-01 |
Pos logit contributions | |
handed | +0.034 |
amusement | +0.023 |
amaz | +0.022 |
Mike | +0.022 |
view | +0.021 |
Neg logit contributions | |
fl | -0.039 |
ches | -0.036 |
wand | -0.035 |
py | -0.034 |
Ho | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -1.95e-02 |
Pos logit contributions | |
handed | +1.031 |
Mike | +0.816 |
amusement | +0.802 |
vendor | +0.796 |
view | +0.770 |
Neg logit contributions | |
ches | -0.644 |
fl | -0.599 |
py | -0.575 |
Tiny | -0.568 |
wand | -0.560 |
Token | They |
Token ID | 1119 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
handed | +0.038 |
vendor | +0.030 |
sandbox | +0.030 |
Mike | +0.030 |
amusement | +0.029 |
Neg logit contributions | |
ches | -0.033 |
Tiny | -0.030 |
py | -0.030 |
fl | -0.030 |
Ho | -0.029 |
Token | have |
Token ID | 423 |
Feature activation | +3.06e-02 |
Pos logit contributions | |
ches | +0.033 |
Tiny | +0.030 |
fl | +0.030 |
py | +0.030 |
Ho | +0.029 |
Neg logit contributions | |
handed | -0.040 |
Mike | -0.030 |
vendor | -0.030 |
amusement | -0.030 |
sandbox | -0.029 |
Token | fun |
Token ID | 1257 |
Feature activation | -9.76e-02 |
Pos logit contributions | |
ches | +0.039 |
fl | +0.038 |
Tiny | +0.036 |
hopping | +0.034 |
wool | +0.033 |
Neg logit contributions | |
handed | -0.076 |
view | -0.057 |
vendor | -0.057 |
amaz | -0.057 |
apartment | -0.057 |
Token | with |
Token ID | 351 |
Feature activation | -6.74e-02 |
Pos logit contributions | |
handed | +0.188 |
Mike | +0.153 |
sandbox | +0.152 |
vendor | +0.147 |
amusement | +0.144 |
Neg logit contributions | |
ches | -0.170 |
Tiny | -0.158 |
Cotton | -0.152 |
py | -0.151 |
Ho | -0.149 |
Token | , |
Token ID | 11 |
Feature activation | -2.84e-02 |
Pos logit contributions | |
handed | +0.470 |
Mike | +0.376 |
amaz | +0.366 |
amusement | +0.364 |
vendor | +0.361 |
Neg logit contributions | |
ches | -0.259 |
fl | -0.236 |
Tiny | -0.228 |
py | -0.218 |
Ho | -0.213 |
Token | the |
Token ID | 262 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
handed | +0.068 |
Mike | +0.053 |
apartment | +0.051 |
amaz | +0.051 |
amusement | +0.050 |
Neg logit contributions | |
ches | -0.039 |
fl | -0.037 |
Tiny | -0.036 |
py | -0.034 |
Ho | -0.034 |
Token | slide |
Token ID | 10649 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
handed | +0.027 |
Mike | +0.019 |
amusement | +0.019 |
amaz | +0.019 |
aren | +0.018 |
Neg logit contributions | |
fl | -0.029 |
ches | -0.028 |
py | -0.027 |
Ho | -0.027 |
Tiny | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | -3.85e-03 |
Pos logit contributions | |
handed | +0.699 |
Mike | +0.541 |
amusement | +0.541 |
amaz | +0.537 |
view | +0.531 |
Neg logit contributions | |
ches | -0.415 |
fl | -0.369 |
Tiny | -0.358 |
py | -0.343 |
Ho | -0.337 |
Token | the |
Token ID | 262 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
handed | +0.010 |
apartment | +0.007 |
view | +0.007 |
Mike | +0.007 |
amaz | +0.007 |
Neg logit contributions | |
fl | -0.005 |
ches | -0.005 |
Tiny | -0.005 |
Sparkle | -0.004 |
Ho | -0.004 |
Token | sandbox |
Token ID | 35204 |
Feature activation | -4.54e-01 |
Pos logit contributions | |
handed | +0.030 |
Mike | +0.021 |
aren | +0.021 |
amusement | +0.021 |
Jane | +0.021 |
Neg logit contributions | |
fl | -0.028 |
Ho | -0.026 |
Tiny | -0.025 |
wand | -0.025 |
py | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -1.49e-02 |
Pos logit contributions | |
handed | +1.110 |
Mike | +0.896 |
amusement | +0.876 |
vendor | +0.869 |
aren | +0.857 |
Neg logit contributions | |
ches | -0.558 |
fl | -0.542 |
py | -0.495 |
wand | -0.489 |
Tiny | -0.475 |
Token | She |
Token ID | 1375 |
Feature activation | -4.89e-02 |
Pos logit contributions | |
handed | +0.031 |
Mike | +0.024 |
vendor | +0.024 |
sandbox | +0.024 |
amusement | +0.024 |
Neg logit contributions | |
ches | -0.024 |
Tiny | -0.022 |
py | -0.021 |
fl | -0.021 |
Ho | -0.021 |
Token | wanted |
Token ID | 2227 |
Feature activation | -4.61e-03 |
Pos logit contributions | |
handed | +0.098 |
vendor | +0.078 |
Mike | +0.077 |
sandbox | +0.077 |
amusement | +0.075 |
Neg logit contributions | |
ches | -0.080 |
Tiny | -0.074 |
py | -0.073 |
fl | -0.069 |
Cotton | -0.069 |
Token | to |
Token ID | 284 |
Feature activation | -3.76e-02 |
Pos logit contributions | |
handed | +0.012 |
amaz | +0.010 |
view | +0.009 |
apartment | +0.009 |
lobby | +0.009 |
Neg logit contributions | |
ches | -0.005 |
Tiny | -0.005 |
Ho | -0.005 |
fl | -0.005 |
Sparkle | -0.005 |
Token | join |
Token ID | 4654 |
Feature activation | -5.82e-02 |
Pos logit contributions | |
handed | +0.083 |
Mike | +0.064 |
lobby | +0.063 |
sandbox | +0.062 |
vendor | +0.062 |
Neg logit contributions | |
ches | -0.056 |
Tiny | -0.052 |
fl | -0.051 |
py | -0.050 |
Ho | -0.048 |
Token | and |
Token ID | 290 |
Feature activation | +3.50e-03 |
Pos logit contributions | |
handed | +0.720 |
amaz | +0.573 |
Mike | +0.572 |
Jane | +0.555 |
amusement | +0.555 |
Neg logit contributions | |
ches | -0.389 |
fl | -0.339 |
Tiny | -0.324 |
Ho | -0.309 |
py | -0.299 |
Token | a |
Token ID | 257 |
Feature activation | -2.19e-02 |
Pos logit contributions | |
fl | +0.004 |
Tiny | +0.004 |
Ho | +0.004 |
ches | +0.004 |
Sparkle | +0.004 |
Neg logit contributions | |
handed | -0.009 |
Mike | -0.007 |
aren | -0.007 |
view | -0.007 |
he | -0.007 |
Token | swing |
Token ID | 9628 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
handed | +0.045 |
Mike | +0.033 |
view | +0.032 |
amusement | +0.031 |
Jane | +0.031 |
Neg logit contributions | |
fl | -0.036 |
ches | -0.035 |
Tiny | -0.035 |
wand | -0.034 |
Ho | -0.034 |
Token | and |
Token ID | 290 |
Feature activation | +3.01e-03 |
Pos logit contributions | |
handed | +0.368 |
Mike | +0.290 |
vendor | +0.287 |
amusement | +0.285 |
sandbox | +0.285 |
Neg logit contributions | |
ches | -0.242 |
Tiny | -0.218 |
fl | -0.213 |
py | -0.213 |
Ho | -0.206 |
Token | a |
Token ID | 257 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
fl | +0.004 |
Tiny | +0.004 |
ches | +0.004 |
Ho | +0.004 |
Sparkle | +0.003 |
Neg logit contributions | |
handed | -0.007 |
Mike | -0.006 |
vendor | -0.005 |
aren | -0.005 |
view | -0.005 |
Token | sandbox |
Token ID | 35204 |
Feature activation | -4.53e-01 |
Pos logit contributions | |
handed | +0.040 |
Mike | +0.028 |
view | +0.027 |
amusement | +0.026 |
Jane | +0.026 |
Neg logit contributions | |
fl | -0.042 |
ches | -0.041 |
Tiny | -0.040 |
wand | -0.040 |
Ho | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
handed | +1.005 |
Mike | +0.799 |
vendor | +0.775 |
amusement | +0.774 |
sandbox | +0.766 |
Neg logit contributions | |
ches | -0.657 |
fl | -0.597 |
Tiny | -0.593 |
py | -0.590 |
Ho | -0.569 |
Token | You |
Token ID | 921 |
Feature activation | +8.87e-02 |
Pos logit contributions | |
handed | +0.020 |
sandbox | +0.015 |
amusement | +0.015 |
Mike | +0.015 |
vendor | +0.015 |
Neg logit contributions | |
ches | -0.021 |
fl | -0.019 |
py | -0.018 |
wand | -0.017 |
Tiny | -0.017 |
Token | can |
Token ID | 460 |
Feature activation | +8.93e-02 |
Pos logit contributions | |
ches | +0.150 |
py | +0.140 |
Tiny | +0.139 |
fl | +0.133 |
Ho | +0.133 |
Neg logit contributions | |
handed | -0.170 |
vendor | -0.129 |
sandbox | -0.129 |
amusement | -0.128 |
Mike | -0.127 |
Token | observe |
Token ID | 12414 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
ches | +0.157 |
Tiny | +0.145 |
py | +0.142 |
fl | +0.140 |
Ho | +0.137 |
Neg logit contributions | |
handed | -0.174 |
vendor | -0.131 |
Mike | -0.131 |
sandbox | -0.130 |
amusement | -0.129 |
Token | more |
Token ID | 517 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
handed | +0.320 |
Mike | +0.249 |
vendor | +0.246 |
sandbox | +0.244 |
amusement | +0.241 |
Neg logit contributions | |
ches | -0.211 |
Tiny | -0.194 |
fl | -0.191 |
py | -0.185 |
Ho | -0.182 |
Token | and |
Token ID | 290 |
Feature activation | -1.18e-03 |
Pos logit contributions | |
handed | +0.399 |
Mike | +0.313 |
amusement | +0.305 |
vendor | +0.303 |
view | +0.301 |
Neg logit contributions | |
ches | -0.226 |
fl | -0.207 |
Tiny | -0.201 |
py | -0.194 |
Ho | -0.189 |
Token | the |
Token ID | 262 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
handed | +0.002 |
sandbox | +0.002 |
vendor | +0.002 |
amusement | +0.002 |
Mike | +0.002 |
Neg logit contributions | |
ches | -0.002 |
py | -0.002 |
Tiny | -0.002 |
Ho | -0.002 |
iny | -0.002 |
Token | slide |
Token ID | 10649 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
handed | +0.028 |
Mike | +0.020 |
amusement | +0.019 |
view | +0.019 |
vendor | +0.018 |
Neg logit contributions | |
ches | -0.031 |
fl | -0.031 |
Tiny | -0.030 |
wand | -0.029 |
py | -0.029 |
Token | and |
Token ID | 290 |
Feature activation | -1.37e-03 |
Pos logit contributions | |
handed | +0.725 |
Mike | +0.563 |
amusement | +0.561 |
view | +0.545 |
vendor | +0.545 |
Neg logit contributions | |
ches | -0.398 |
fl | -0.365 |
Tiny | -0.332 |
py | -0.323 |
Ho | -0.322 |
Token | the |
Token ID | 262 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
handed | +0.003 |
Mike | +0.002 |
view | +0.002 |
vendor | +0.002 |
Jane | +0.002 |
Neg logit contributions | |
ches | -0.002 |
fl | -0.002 |
Tiny | -0.002 |
Ho | -0.002 |
Sparkle | -0.002 |
Token | sandbox |
Token ID | 35204 |
Feature activation | -4.53e-01 |
Pos logit contributions | |
handed | +0.029 |
Mike | +0.020 |
amusement | +0.019 |
view | +0.019 |
Jane | +0.018 |
Neg logit contributions | |
fl | -0.032 |
ches | -0.031 |
wand | -0.030 |
Tiny | -0.030 |
wool | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
handed | +1.150 |
Mike | +0.927 |
amusement | +0.902 |
vendor | +0.893 |
aren | +0.881 |
Neg logit contributions | |
ches | -0.523 |
fl | -0.522 |
wand | -0.491 |
py | -0.459 |
Tiny | -0.437 |
Token | They |
Token ID | 1119 |
Feature activation | +2.22e-02 |
Pos logit contributions | |
handed | +0.035 |
Mike | +0.028 |
unatt | +0.027 |
sandbox | +0.027 |
vendor | +0.027 |
Neg logit contributions | |
Tiny | -0.015 |
ches | -0.014 |
Ho | -0.014 |
Sparkle | -0.013 |
Cotton | -0.013 |
Token | have |
Token ID | 423 |
Feature activation | +2.88e-02 |
Pos logit contributions | |
ches | +0.036 |
fl | +0.033 |
Tiny | +0.032 |
py | +0.031 |
Ho | +0.031 |
Neg logit contributions | |
handed | -0.048 |
Mike | -0.036 |
vendor | -0.035 |
amusement | -0.035 |
view | -0.035 |
Token | fun |
Token ID | 1257 |
Feature activation | -9.96e-02 |
Pos logit contributions | |
fl | +0.039 |
ches | +0.037 |
Tiny | +0.037 |
wool | +0.034 |
Ho | +0.034 |
Neg logit contributions | |
handed | -0.072 |
he | -0.055 |
view | -0.053 |
vendor | -0.053 |
Mike | -0.052 |
Token | and |
Token ID | 290 |
Feature activation | -1.63e-03 |
Pos logit contributions | |
handed | +0.231 |
vendor | +0.180 |
Mike | +0.177 |
amusement | +0.176 |
sandbox | +0.173 |
Neg logit contributions | |
ches | -0.139 |
fl | -0.132 |
Tiny | -0.127 |
py | -0.126 |
wand | -0.120 |
Token | can |
Token ID | 460 |
Feature activation | +8.87e-02 |
Pos logit contributions | |
ches | +0.031 |
Tiny | +0.028 |
fl | +0.028 |
py | +0.027 |
Ho | +0.027 |
Neg logit contributions | |
handed | -0.039 |
vendor | -0.030 |
sandbox | -0.030 |
Mike | -0.030 |
amusement | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
ches | +0.127 |
Tiny | +0.117 |
py | +0.113 |
Cotton | +0.110 |
Ho | +0.110 |
Neg logit contributions | |
handed | -0.195 |
vendor | -0.156 |
Mike | -0.156 |
sandbox | -0.154 |
amusement | -0.153 |
Token | This |
Token ID | 770 |
Feature activation | -2.64e-02 |
Pos logit contributions | |
handed | +0.028 |
vendor | +0.022 |
Mike | +0.022 |
sandbox | +0.022 |
amusement | +0.022 |
Neg logit contributions | |
ches | -0.023 |
fl | -0.021 |
py | -0.020 |
Tiny | -0.020 |
wand | -0.019 |
Token | is |
Token ID | 318 |
Feature activation | -6.13e-02 |
Pos logit contributions | |
handed | +0.060 |
Mike | +0.048 |
amusement | +0.046 |
vendor | +0.045 |
aren | +0.045 |
Neg logit contributions | |
ches | -0.039 |
fl | -0.037 |
py | -0.035 |
wand | -0.034 |
Tiny | -0.034 |
Token | my |
Token ID | 616 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
handed | +0.117 |
vendor | +0.095 |
sandbox | +0.094 |
amusement | +0.093 |
Mike | +0.091 |
Neg logit contributions | |
ches | -0.101 |
py | -0.091 |
Tiny | -0.090 |
tle | -0.088 |
fl | -0.087 |
Token | sandbox |
Token ID | 35204 |
Feature activation | -4.53e-01 |
Pos logit contributions | |
fl | +0.165 |
wand | +0.164 |
Ho | +0.147 |
Tiny | +0.147 |
wool | +0.146 |
Neg logit contributions | |
handed | -0.230 |
Mike | -0.189 |
Jane | -0.177 |
aren | -0.174 |
he | -0.170 |
Token | . |
Token ID | 13 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
handed | +1.091 |
Mike | +0.899 |
amusement | +0.862 |
aren | +0.856 |
vendor | +0.846 |
Neg logit contributions | |
ches | -0.586 |
fl | -0.523 |
Tiny | -0.517 |
py | -0.513 |
wand | -0.497 |
Token | I |
Token ID | 314 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
handed | +0.026 |
vendor | +0.020 |
Mike | +0.020 |
amusement | +0.020 |
sandbox | +0.020 |
Neg logit contributions | |
ches | -0.025 |
fl | -0.023 |
py | -0.022 |
Tiny | -0.021 |
wand | -0.021 |
Token | don |
Token ID | 836 |
Feature activation | -6.87e-03 |
Pos logit contributions | |
ches | +0.031 |
Tiny | +0.028 |
fl | +0.028 |
py | +0.027 |
Ho | +0.027 |
Neg logit contributions | |
handed | -0.035 |
vendor | -0.027 |
Mike | -0.027 |
sandbox | -0.027 |
amusement | -0.027 |
Token | 't |
Token ID | 470 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
handed | +0.011 |
he | +0.011 |
Mike | +0.010 |
vendor | +0.009 |
anymore | +0.009 |
Neg logit contributions | |
wand | -0.012 |
Ho | -0.012 |
Pig | -0.011 |
ches | -0.011 |
py | -0.011 |
Token | care |
Token ID | 1337 |
Feature activation | -7.33e-02 |
Pos logit contributions | |
ches | +0.293 |
Tiny | +0.269 |
py | +0.265 |
fl | +0.260 |
Ho | +0.253 |
Neg logit contributions | |
handed | -0.374 |
Mike | -0.290 |
sandbox | -0.288 |
vendor | -0.287 |
lobby | -0.284 |
Token | and |
Token ID | 290 |
Feature activation | -3.47e-03 |
Pos logit contributions | |
handed | +0.626 |
view | +0.470 |
Mike | +0.463 |
amaz | +0.454 |
Jane | +0.454 |
Neg logit contributions | |
fl | -0.168 |
ches | -0.150 |
Tiny | -0.127 |
hopping | -0.118 |
Ho | -0.118 |
Token | slides |
Token ID | 19392 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
handed | +0.007 |
apartment | +0.005 |
unatt | +0.005 |
Mike | +0.005 |
view | +0.005 |
Neg logit contributions | |
fl | -0.006 |
ches | -0.006 |
Tiny | -0.006 |
py | -0.006 |
Ho | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | -3.36e-03 |
Pos logit contributions | |
handed | +0.507 |
Mike | +0.390 |
amusement | +0.390 |
amaz | +0.390 |
vendor | +0.382 |
Neg logit contributions | |
ches | -0.279 |
fl | -0.261 |
Tiny | -0.239 |
py | -0.227 |
Ho | -0.226 |
Token | a |
Token ID | 257 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
handed | +0.007 |
apartment | +0.006 |
amaz | +0.005 |
aren | +0.005 |
unatt | +0.005 |
Neg logit contributions | |
fl | -0.005 |
ches | -0.005 |
Tiny | -0.005 |
hopping | -0.004 |
Ho | -0.004 |
Token | big |
Token ID | 1263 |
Feature activation | +2.28e-02 |
Pos logit contributions | |
handed | +0.050 |
view | +0.038 |
ill | +0.037 |
Mike | +0.037 |
aren | +0.036 |
Neg logit contributions | |
fl | -0.033 |
ches | -0.030 |
wand | -0.030 |
Tiny | -0.029 |
Ho | -0.028 |
Token | sandbox |
Token ID | 35204 |
Feature activation | -4.53e-01 |
Pos logit contributions | |
fl | +0.040 |
ches | +0.037 |
Tiny | +0.036 |
wand | +0.036 |
wool | +0.035 |
Neg logit contributions | |
handed | -0.048 |
Mike | -0.033 |
view | -0.033 |
Jane | -0.032 |
ill | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
handed | +1.095 |
Mike | +0.882 |
amusement | +0.843 |
vendor | +0.840 |
Jane | +0.838 |
Neg logit contributions | |
ches | -0.549 |
fl | -0.544 |
wand | -0.495 |
py | -0.476 |
Tiny | -0.472 |
Token | They |
Token ID | 1119 |
Feature activation | +2.10e-02 |
Pos logit contributions | |
handed | +0.036 |
unatt | +0.029 |
view | +0.027 |
Mike | +0.027 |
temple | +0.026 |
Neg logit contributions | |
Tiny | -0.015 |
ches | -0.014 |
Ho | -0.014 |
Cotton | -0.013 |
Chick | -0.013 |
Token | ran |
Token ID | 4966 |
Feature activation | +3.84e-02 |
Pos logit contributions | |
ches | +0.030 |
fl | +0.029 |
Tiny | +0.028 |
Ho | +0.027 |
Pig | +0.026 |
Neg logit contributions | |
handed | -0.049 |
view | -0.036 |
amusement | -0.036 |
Mike | -0.035 |
amaz | -0.035 |
Token | out |
Token ID | 503 |
Feature activation | +3.78e-02 |
Pos logit contributions | |
ches | +0.043 |
fl | +0.040 |
Tiny | +0.034 |
ch | +0.033 |
hopping | +0.032 |
Neg logit contributions | |
handed | -0.104 |
amaz | -0.086 |
view | -0.085 |
amusement | -0.080 |
unatt | -0.080 |
Token | of |
Token ID | 286 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
fl | +0.040 |
ches | +0.039 |
Tiny | +0.035 |
hopping | +0.033 |
Tw | +0.032 |
Neg logit contributions | |
handed | -0.106 |
vendor | -0.081 |
view | -0.080 |
sandbox | -0.080 |
unatt | -0.079 |
Token | for |
Token ID | 329 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
handed | +0.215 |
sandbox | +0.169 |
amaz | +0.169 |
amusement | +0.169 |
Mike | +0.167 |
Neg logit contributions | |
ches | -0.122 |
fl | -0.108 |
Tiny | -0.107 |
Ho | -0.102 |
py | -0.099 |
Token | the |
Token ID | 262 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
ches | +0.240 |
Tiny | +0.220 |
py | +0.213 |
fl | +0.211 |
Ho | +0.207 |
Neg logit contributions | |
handed | -0.346 |
vendor | -0.270 |
Mike | -0.270 |
sandbox | -0.270 |
amusement | -0.266 |
Token | car |
Token ID | 1097 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
handed | +0.039 |
Mike | +0.030 |
Jane | +0.030 |
ill | +0.029 |
view | +0.028 |
Neg logit contributions | |
fl | -0.029 |
ches | -0.028 |
wand | -0.028 |
Ho | -0.027 |
Tiny | -0.027 |
Token | ," |
Token ID | 553 |
Feature activation | -3.17e-04 |
Pos logit contributions | |
handed | +0.267 |
Mike | +0.218 |
sandbox | +0.215 |
amusement | +0.214 |
amaz | +0.213 |
Neg logit contributions | |
ches | -0.171 |
Tiny | -0.149 |
fl | -0.147 |
py | -0.142 |
Ho | -0.138 |
Token | the |
Token ID | 262 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
handed | +0.001 |
view | +0.001 |
Mike | +0.001 |
sandbox | +0.001 |
lobby | +0.001 |
Neg logit contributions | |
Tiny | -0.000 |
Ho | -0.000 |
ches | -0.000 |
Cotton | -0.000 |
fl | -0.000 |
Token | vendor |
Token ID | 18371 |
Feature activation | -4.53e-01 |
Pos logit contributions | |
handed | +0.021 |
ill | +0.016 |
unatt | +0.013 |
view | +0.012 |
anymore | +0.012 |
Neg logit contributions | |
fl | -0.046 |
asking | -0.043 |
cowboy | -0.043 |
Cotton | -0.042 |
hopping | -0.042 |
Token | said |
Token ID | 531 |
Feature activation | +7.62e-02 |
Pos logit contributions | |
handed | +1.095 |
sandbox | +0.880 |
Mike | +0.874 |
vendor | +0.873 |
amusement | +0.870 |
Neg logit contributions | |
ches | -0.581 |
Tiny | -0.516 |
fl | -0.498 |
py | -0.495 |
Ho | -0.481 |
Token | . |
Token ID | 13 |
Feature activation | -9.71e-03 |
Pos logit contributions | |
ches | +0.130 |
Tiny | +0.121 |
py | +0.119 |
fl | +0.116 |
Ho | +0.114 |
Neg logit contributions | |
handed | -0.151 |
Mike | -0.114 |
vendor | -0.114 |
sandbox | -0.112 |
amusement | -0.111 |
Token |
|
Token ID | 198 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
handed | +0.019 |
vendor | +0.015 |
Mike | +0.015 |
amusement | +0.014 |
sandbox | +0.014 |
Neg logit contributions | |
ches | -0.017 |
py | -0.015 |
fl | -0.015 |
agons | -0.014 |
Tiny | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
ches | +0.019 |
py | +0.018 |
Tiny | +0.018 |
fl | +0.017 |
Cotton | +0.017 |
Neg logit contributions | |
handed | -0.024 |
Mike | -0.018 |
vendor | -0.018 |
sandbox | -0.018 |
amusement | -0.018 |
Token | Ben |
Token ID | 11696 |
Feature activation | -6.13e-02 |
Pos logit contributions | |
ches | +0.017 |
Tiny | +0.016 |
Ho | +0.015 |
fl | +0.015 |
py | +0.015 |
Neg logit contributions | |
handed | -0.026 |
vendor | -0.020 |
amusement | -0.020 |
Mike | -0.020 |
sandbox | -0.020 |
Token | time |
Token ID | 640 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
handed | +0.057 |
Mike | +0.047 |
vendor | +0.047 |
sandbox | +0.047 |
amusement | +0.047 |
Neg logit contributions | |
ches | -0.021 |
Tiny | -0.017 |
fl | -0.017 |
py | -0.016 |
iny | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | -2.83e-02 |
Pos logit contributions | |
ches | +0.153 |
Tiny | +0.148 |
Ho | +0.140 |
fl | +0.140 |
py | +0.138 |
Neg logit contributions | |
handed | -0.293 |
amusement | -0.229 |
view | -0.227 |
sandbox | -0.227 |
vendor | -0.227 |
Token | there |
Token ID | 612 |
Feature activation | -7.39e-02 |
Pos logit contributions | |
handed | +0.054 |
vendor | +0.041 |
Mike | +0.041 |
sandbox | +0.040 |
amusement | +0.040 |
Neg logit contributions | |
ches | -0.050 |
Tiny | -0.047 |
py | -0.046 |
fl | -0.045 |
Ho | -0.044 |
Token | was |
Token ID | 373 |
Feature activation | -9.26e-02 |
Pos logit contributions | |
handed | +0.206 |
unatt | +0.164 |
vendor | +0.162 |
amusement | +0.162 |
view | +0.162 |
Neg logit contributions | |
fl | -0.070 |
ches | -0.067 |
Tiny | -0.064 |
Ho | -0.064 |
hopping | -0.058 |
Token | a |
Token ID | 257 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
handed | +0.194 |
vendor | +0.150 |
Mike | +0.149 |
sandbox | +0.149 |
amusement | +0.147 |
Neg logit contributions | |
ches | -0.148 |
Tiny | -0.136 |
py | -0.132 |
fl | -0.131 |
Ho | -0.128 |
Token | vendor |
Token ID | 18371 |
Feature activation | -4.53e-01 |
Pos logit contributions | |
handed | +0.042 |
vendor | +0.037 |
sandbox | +0.036 |
amusement | +0.034 |
Mike | +0.034 |
Neg logit contributions | |
ches | -0.032 |
py | -0.029 |
Tiny | -0.027 |
Sparkle | -0.025 |
Cotton | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
handed | +0.943 |
vendor | +0.731 |
Mike | +0.727 |
sandbox | +0.723 |
amusement | +0.716 |
Neg logit contributions | |
ches | -0.728 |
Tiny | -0.667 |
py | -0.651 |
fl | -0.641 |
Ho | -0.630 |
Token | He |
Token ID | 679 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
handed | +0.040 |
Mike | +0.032 |
sandbox | +0.032 |
unatt | +0.031 |
view | +0.030 |
Neg logit contributions | |
Tiny | -0.023 |
ches | -0.021 |
Cotton | -0.021 |
Ho | -0.020 |
Pig | -0.020 |
Token | sold |
Token ID | 2702 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
handed | +0.076 |
Mike | +0.059 |
sandbox | +0.059 |
amusement | +0.059 |
vendor | +0.059 |
Neg logit contributions | |
ches | -0.046 |
Tiny | -0.041 |
fl | -0.040 |
py | -0.040 |
Ho | -0.039 |
Token | cold |
Token ID | 4692 |
Feature activation | -4.05e-02 |
Pos logit contributions | |
handed | +0.393 |
apartment | +0.319 |
ill | +0.311 |
unatt | +0.310 |
amaz | +0.306 |
Neg logit contributions | |
fl | -0.170 |
carrots | -0.169 |
Ho | -0.161 |
asking | -0.160 |
potatoes | -0.158 |
Token | treats |
Token ID | 18432 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
handed | +0.091 |
amaz | +0.075 |
Mike | +0.074 |
ill | +0.073 |
amusement | +0.072 |
Neg logit contributions | |
wool | -0.048 |
fl | -0.046 |
ches | -0.044 |
potatoes | -0.044 |
pigs | -0.044 |
Token | want |
Token ID | 765 |
Feature activation | -5.07e-02 |
Pos logit contributions | |
ches | +0.052 |
fl | +0.047 |
Tiny | +0.046 |
Ho | +0.045 |
py | +0.044 |
Neg logit contributions | |
handed | -0.061 |
Mike | -0.046 |
vendor | -0.046 |
sandbox | -0.045 |
Jane | -0.045 |
Token | to |
Token ID | 284 |
Feature activation | -3.42e-02 |
Pos logit contributions | |
handed | +0.141 |
amaz | +0.120 |
he | +0.118 |
Mike | +0.117 |
sandbox | +0.116 |
Neg logit contributions | |
Tiny | -0.042 |
Ho | -0.040 |
wool | -0.038 |
Pig | -0.038 |
ches | -0.036 |
Token | play |
Token ID | 711 |
Feature activation | -6.66e-02 |
Pos logit contributions | |
handed | +0.084 |
Mike | +0.067 |
vendor | +0.066 |
Jane | +0.065 |
unatt | +0.065 |
Neg logit contributions | |
ches | -0.041 |
fl | -0.040 |
Tiny | -0.040 |
Ho | -0.037 |
Pig | -0.036 |
Token | in |
Token ID | 287 |
Feature activation | +3.95e-02 |
Pos logit contributions | |
handed | +0.169 |
aren | +0.148 |
aney | +0.147 |
spe | +0.143 |
amaz | +0.139 |
Neg logit contributions | |
fl | -0.077 |
ches | -0.062 |
wand | -0.061 |
Ho | -0.060 |
wool | -0.060 |
Token | the |
Token ID | 262 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
fl | +0.011 |
Tiny | +0.004 |
wool | +0.003 |
asking | +0.002 |
ro | +0.002 |
Neg logit contributions | |
handed | -0.126 |
aney | -0.116 |
ome | -0.113 |
ival | -0.112 |
attered | -0.111 |
Token | sandbox |
Token ID | 35204 |
Feature activation | -4.52e-01 |
Pos logit contributions | |
handed | +0.028 |
ill | +0.023 |
aney | +0.022 |
Mike | +0.021 |
Megan | +0.020 |
Neg logit contributions | |
fl | -0.025 |
ch | -0.022 |
wool | -0.022 |
ro | -0.021 |
hopping | -0.021 |
Token | , |
Token ID | 11 |
Feature activation | -2.52e-02 |
Pos logit contributions | |
handed | +1.306 |
aren | +1.131 |
Mike | +1.116 |
view | +1.087 |
amusement | +1.077 |
Neg logit contributions | |
fl | -0.327 |
ches | -0.260 |
hopping | -0.235 |
for | -0.230 |
wand | -0.227 |
Token | but |
Token ID | 475 |
Feature activation | -4.58e-02 |
Pos logit contributions | |
handed | +0.068 |
Mike | +0.053 |
view | +0.051 |
aren | +0.050 |
amusement | +0.049 |
Neg logit contributions | |
fl | -0.029 |
Tiny | -0.029 |
ches | -0.028 |
Ho | -0.027 |
hopping | -0.025 |
Token | it |
Token ID | 340 |
Feature activation | +9.90e-03 |
Pos logit contributions | |
handed | +0.124 |
view | +0.094 |
alley | +0.088 |
Mike | +0.088 |
aren | +0.087 |
Neg logit contributions | |
Tiny | -0.053 |
fl | -0.052 |
Ho | -0.049 |
ches | -0.046 |
Sparkle | -0.045 |
Token | is |
Token ID | 318 |
Feature activation | -5.59e-02 |
Pos logit contributions | |
ches | +0.017 |
Tiny | +0.016 |
py | +0.015 |
fl | +0.015 |
Ho | +0.015 |
Neg logit contributions | |
handed | -0.020 |
Mike | -0.015 |
vendor | -0.015 |
amusement | -0.015 |
sandbox | -0.015 |
Token | too |
Token ID | 1165 |
Feature activation | +6.11e-03 |
Pos logit contributions | |
handed | +0.088 |
vendor | +0.073 |
sandbox | +0.073 |
anymore | +0.072 |
amusement | +0.072 |
Neg logit contributions | |
ches | -0.104 |
Tiny | -0.097 |
py | -0.096 |
tle | -0.093 |
Cotton | -0.092 |
Token | playing |
Token ID | 2712 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
handed | +0.260 |
Mike | +0.208 |
amusement | +0.200 |
vendor | +0.199 |
view | +0.198 |
Neg logit contributions | |
ches | -0.137 |
fl | -0.133 |
hopping | -0.120 |
Tiny | -0.117 |
wand | -0.117 |
Token | . |
Token ID | 13 |
Feature activation | -4.41e-03 |
Pos logit contributions | |
fl | +0.021 |
ches | +0.020 |
wand | +0.019 |
py | +0.018 |
Tiny | +0.018 |
Neg logit contributions | |
handed | -0.040 |
Mike | -0.031 |
aren | -0.031 |
vendor | -0.031 |
view | -0.031 |
Token | They |
Token ID | 1119 |
Feature activation | +3.17e-02 |
Pos logit contributions | |
handed | +0.012 |
unatt | +0.009 |
view | +0.009 |
Mike | +0.009 |
amusement | +0.009 |
Neg logit contributions | |
Tiny | -0.005 |
ches | -0.004 |
Ho | -0.004 |
Sparkle | -0.004 |
Cotton | -0.004 |
Token | see |
Token ID | 766 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
fl | +0.050 |
ches | +0.050 |
wand | +0.046 |
Tiny | +0.046 |
py | +0.045 |
Neg logit contributions | |
handed | -0.068 |
Mike | -0.051 |
Jane | -0.048 |
lobby | -0.048 |
view | -0.048 |
Token | the |
Token ID | 262 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
handed | +0.453 |
Mike | +0.359 |
aren | +0.344 |
alley | +0.340 |
view | +0.338 |
Neg logit contributions | |
fl | -0.207 |
ches | -0.204 |
Tiny | -0.200 |
Ho | -0.190 |
Sparkle | -0.185 |
Token | sandbox |
Token ID | 35204 |
Feature activation | -4.52e-01 |
Pos logit contributions | |
handed | +0.033 |
Mike | +0.026 |
Jane | +0.025 |
aren | +0.025 |
view | +0.024 |
Neg logit contributions | |
fl | -0.015 |
wand | -0.013 |
wool | -0.012 |
py | -0.012 |
ches | -0.012 |
Token | , |
Token ID | 11 |
Feature activation | -2.26e-02 |
Pos logit contributions | |
handed | +1.184 |
Mike | +0.952 |
aren | +0.907 |
vendor | +0.895 |
amusement | +0.889 |
Neg logit contributions | |
fl | -0.512 |
ches | -0.493 |
wand | -0.464 |
py | -0.443 |
hopping | -0.431 |
Token | the |
Token ID | 262 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
handed | +0.060 |
Mike | +0.047 |
aren | +0.044 |
view | +0.044 |
vendor | +0.044 |
Neg logit contributions | |
fl | -0.026 |
Tiny | -0.025 |
ches | -0.024 |
py | -0.024 |
Ho | -0.023 |
Token | sees |
Token ID | 7224 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
handed | +0.029 |
Mike | +0.022 |
Jane | +0.021 |
aren | +0.021 |
view | +0.021 |
Neg logit contributions | |
fl | -0.017 |
wand | -0.016 |
Ho | -0.015 |
Tiny | -0.015 |
py | -0.015 |
Token | aw |
Token ID | 707 |
Feature activation | -3.62e-01 |
Pos logit contributions | |
handed | +0.439 |
apartment | +0.336 |
Mike | +0.335 |
amusement | +0.330 |
vendor | +0.328 |
Neg logit contributions | |
fl | -0.212 |
ches | -0.211 |
py | -0.186 |
wand | -0.183 |
hopping | -0.177 |
Token | , |
Token ID | 11 |
Feature activation | -2.37e-02 |
Pos logit contributions | |
handed | +0.714 |
vendor | +0.551 |
Mike | +0.546 |
sandbox | +0.546 |
amusement | +0.538 |
Neg logit contributions | |
ches | -0.606 |
Tiny | -0.564 |
py | -0.553 |
fl | -0.537 |
Ho | -0.532 |
Token | higher |
Token ID | 2440 |
Feature activation | -1.00e-01 |
Pos logit contributions | |
ches | +0.382 |
Tiny | +0.358 |
py | +0.355 |
Cotton | +0.337 |
fl | +0.336 |
Neg logit contributions | |
handed | -0.488 |
vendor | -0.372 |
amusement | -0.371 |
Mike | -0.371 |
sandbox | -0.369 |
Token | and |
Token ID | 290 |
Feature activation | -1.85e-03 |
Pos logit contributions | |
handed | +0.199 |
Mike | +0.151 |
vendor | +0.150 |
amusement | +0.148 |
sandbox | +0.148 |
Neg logit contributions | |
ches | -0.173 |
Tiny | -0.160 |
fl | -0.157 |
py | -0.156 |
Ho | -0.151 |
Token | higher |
Token ID | 2440 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
handed | +0.003 |
Mike | +0.003 |
vendor | +0.003 |
amusement | +0.002 |
sandbox | +0.002 |
Neg logit contributions | |
ches | -0.003 |
Tiny | -0.003 |
fl | -0.003 |
py | -0.003 |
Ho | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
handed | +0.186 |
Mike | +0.137 |
vendor | +0.137 |
amusement | +0.135 |
sandbox | +0.135 |
Neg logit contributions | |
ches | -0.188 |
Tiny | -0.175 |
fl | -0.172 |
py | -0.172 |
Ho | -0.167 |
Token | He |
Token ID | 679 |
Feature activation | -2.95e-02 |
Pos logit contributions | |
handed | +0.048 |
vendor | +0.039 |
sandbox | +0.039 |
Mike | +0.039 |
amusement | +0.038 |
Neg logit contributions | |
ches | -0.026 |
py | -0.022 |
Tiny | -0.022 |
fl | -0.022 |
wand | -0.021 |
Token | hopped |
Token ID | 45230 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
handed | +0.052 |
sandbox | +0.045 |
Mike | +0.043 |
vendor | +0.043 |
Jane | +0.040 |
Neg logit contributions | |
Tiny | -0.046 |
ches | -0.045 |
py | -0.045 |
cartridge | -0.045 |
Cotton | -0.044 |
Token | over |
Token ID | 625 |
Feature activation | +5.38e-02 |
Pos logit contributions | |
ches | +0.377 |
py | +0.355 |
Tiny | +0.354 |
Chick | +0.344 |
Pig | +0.338 |
Neg logit contributions | |
handed | -0.408 |
vendor | -0.326 |
Mike | -0.326 |
sandbox | -0.320 |
amusement | -0.306 |
Token | a |
Token ID | 257 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
ches | +0.071 |
fl | +0.070 |
Tiny | +0.070 |
Ho | +0.066 |
hopping | +0.065 |
Neg logit contributions | |
handed | -0.131 |
view | -0.101 |
Mike | -0.099 |
vendor | -0.097 |
amusement | -0.097 |
Token | large |
Token ID | 1588 |
Feature activation | -2.28e-02 |
Pos logit contributions | |
handed | +0.042 |
Mike | +0.032 |
view | +0.032 |
amusement | +0.031 |
Jane | +0.030 |
Neg logit contributions | |
ches | -0.031 |
fl | -0.029 |
Tiny | -0.028 |
Ho | -0.027 |
wand | -0.027 |
Token | log |
Token ID | 2604 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
handed | +0.047 |
Mike | +0.036 |
view | +0.035 |
Jane | +0.034 |
vendor | +0.033 |
Neg logit contributions | |
ches | -0.038 |
fl | -0.036 |
Tiny | -0.036 |
Ho | -0.034 |
wand | -0.034 |
Token | and |
Token ID | 290 |
Feature activation | -3.18e-03 |
Pos logit contributions | |
ches | +0.195 |
Tiny | +0.180 |
py | +0.175 |
fl | +0.171 |
Ho | +0.170 |
Neg logit contributions | |
handed | -0.257 |
vendor | -0.199 |
sandbox | -0.198 |
Mike | -0.198 |
amusement | -0.196 |
Token | They |
Token ID | 1119 |
Feature activation | +4.92e-02 |
Pos logit contributions | |
ches | +0.005 |
Tiny | +0.005 |
py | +0.005 |
Ho | +0.005 |
fl | +0.004 |
Neg logit contributions | |
handed | -0.008 |
Mike | -0.006 |
amusement | -0.006 |
sandbox | -0.006 |
vendor | -0.006 |
Token | walked |
Token ID | 6807 |
Feature activation | +5.70e-02 |
Pos logit contributions | |
ches | +0.074 |
Tiny | +0.067 |
fl | +0.066 |
py | +0.065 |
Ho | +0.064 |
Neg logit contributions | |
handed | -0.109 |
Mike | -0.084 |
sandbox | -0.084 |
vendor | -0.083 |
amusement | -0.083 |
Token | down |
Token ID | 866 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
ches | +0.074 |
Tiny | +0.068 |
fl | +0.066 |
Sparkle | +0.062 |
Ho | +0.061 |
Neg logit contributions | |
handed | -0.141 |
amaz | -0.109 |
view | -0.109 |
amusement | -0.108 |
Mike | -0.108 |
Token | the |
Token ID | 262 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
handed | +0.041 |
view | +0.034 |
sandbox | +0.034 |
amaz | +0.034 |
unatt | +0.033 |
Neg logit contributions | |
Tiny | -0.011 |
fl | -0.010 |
Ho | -0.010 |
ches | -0.009 |
Pig | -0.009 |
Token | a |
Token ID | 257 |
Feature activation | +7.74e-03 |
Pos logit contributions | |
ches | +0.027 |
Tiny | +0.025 |
fl | +0.024 |
py | +0.024 |
Ho | +0.024 |
Neg logit contributions | |
handed | -0.018 |
Mike | -0.012 |
amusement | -0.012 |
vendor | -0.012 |
sandbox | -0.012 |
Token | is |
Token ID | 271 |
Feature activation | +3.02e-01 |
Pos logit contributions | |
ches | +0.011 |
fl | +0.010 |
Tiny | +0.010 |
Ho | +0.009 |
Cotton | +0.009 |
Neg logit contributions | |
handed | -0.017 |
Mike | -0.014 |
amusement | -0.014 |
view | -0.013 |
amaz | -0.013 |
Token | les |
Token ID | 829 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
ches | +0.448 |
Tiny | +0.429 |
Ho | +0.408 |
Sparkle | +0.403 |
py | +0.403 |
Neg logit contributions | |
handed | -0.664 |
vendor | -0.523 |
sandbox | -0.521 |
lobby | -0.507 |
Mike | -0.504 |
Token | and |
Token ID | 290 |
Feature activation | +2.14e-02 |
Pos logit contributions | |
ches | +0.374 |
Tiny | +0.341 |
fl | +0.333 |
py | +0.327 |
Ho | +0.318 |
Neg logit contributions | |
handed | -0.554 |
Mike | -0.439 |
vendor | -0.435 |
sandbox | -0.431 |
amusement | -0.428 |
Token | saw |
Token ID | 2497 |
Feature activation | -9.93e-02 |
Pos logit contributions | |
ches | +0.032 |
Tiny | +0.030 |
fl | +0.029 |
Ho | +0.028 |
Cotton | +0.027 |
Neg logit contributions | |
handed | -0.048 |
Mike | -0.037 |
amusement | -0.037 |
sandbox | -0.036 |
view | -0.036 |
Token | many |
Token ID | 867 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
handed | +0.245 |
unatt | +0.206 |
apartment | +0.205 |
har | +0.203 |
ill | +0.202 |
Neg logit contributions | |
Tiny | -0.106 |
fl | -0.103 |
Ho | -0.100 |
Pig | -0.094 |
wool | -0.093 |
Token | things |
Token ID | 1243 |
Feature activation | +7.91e-03 |
Pos logit contributions | |
handed | +0.092 |
view | +0.072 |
ill | +0.071 |
Mike | +0.070 |
unatt | +0.070 |
Neg logit contributions | |
fl | -0.044 |
ches | -0.044 |
wool | -0.041 |
potatoes | -0.038 |
wand | -0.038 |
Token | One |
Token ID | 1881 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
handed | +0.003 |
vendor | +0.002 |
sandbox | +0.002 |
Mike | +0.002 |
amusement | +0.002 |
Neg logit contributions | |
ches | -0.002 |
fl | -0.002 |
py | -0.002 |
wand | -0.002 |
Tiny | -0.002 |
Token | day |
Token ID | 1110 |
Feature activation | +7.54e-02 |
Pos logit contributions | |
ches | +0.248 |
Tiny | +0.224 |
py | +0.216 |
fl | +0.214 |
Ho | +0.206 |
Neg logit contributions | |
handed | -0.457 |
Mike | -0.366 |
vendor | -0.363 |
sandbox | -0.362 |
amusement | -0.360 |
Token | , |
Token ID | 11 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
ches | +0.135 |
py | +0.122 |
Tiny | +0.122 |
fl | +0.118 |
Cotton | +0.115 |
Neg logit contributions | |
handed | -0.142 |
vendor | -0.110 |
Mike | -0.110 |
sandbox | -0.109 |
amusement | -0.106 |
Token | a |
Token ID | 257 |
Feature activation | -4.49e-03 |
Pos logit contributions | |
handed | +0.025 |
Mike | +0.019 |
vendor | +0.019 |
sandbox | +0.019 |
amusement | +0.019 |
Neg logit contributions | |
ches | -0.019 |
Tiny | -0.017 |
py | -0.017 |
fl | -0.016 |
Ho | -0.016 |
Token | young |
Token ID | 1862 |
Feature activation | +4.89e-02 |
Pos logit contributions | |
handed | +0.009 |
vendor | +0.008 |
sandbox | +0.008 |
amusement | +0.008 |
Mike | +0.007 |
Neg logit contributions | |
ches | -0.007 |
py | -0.006 |
Tiny | -0.006 |
tle | -0.005 |
fl | -0.005 |
Token | rabbit |
Token ID | 22746 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
ches | +0.071 |
py | +0.061 |
agons | +0.057 |
Tiny | +0.056 |
fl | +0.055 |
Neg logit contributions | |
handed | -0.107 |
vendor | -0.093 |
sandbox | -0.089 |
Mike | -0.085 |
amusement | -0.085 |
Token | came |
Token ID | 1625 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
ches | +0.460 |
Tiny | +0.421 |
py | +0.420 |
agons | +0.411 |
Cotton | +0.406 |
Neg logit contributions | |
handed | -0.556 |
sandbox | -0.442 |
vendor | -0.431 |
lobby | -0.408 |
alley | -0.408 |
Token | over |
Token ID | 625 |
Feature activation | +6.78e-02 |
Pos logit contributions | |
ches | +0.256 |
cartridge | +0.256 |
agons | +0.249 |
Chick | +0.246 |
cowboy | +0.246 |
Neg logit contributions | |
handed | -0.152 |
and | -0.096 |
ill | -0.089 |
vendor | -0.088 |
alley | -0.087 |
Token | to |
Token ID | 284 |
Feature activation | -1.99e-02 |
Pos logit contributions | |
ches | +0.113 |
Tiny | +0.103 |
py | +0.102 |
fl | +0.100 |
Ho | +0.098 |
Neg logit contributions | |
handed | -0.138 |
vendor | -0.106 |
Mike | -0.106 |
sandbox | -0.105 |
amusement | -0.104 |
Token | the |
Token ID | 262 |
Feature activation | -4.86e-03 |
Pos logit contributions | |
handed | +0.033 |
vendor | +0.024 |
amusement | +0.024 |
sandbox | +0.023 |
lobby | +0.022 |
Neg logit contributions | |
ches | -0.041 |
py | -0.036 |
wand | -0.036 |
agons | -0.036 |
iny | -0.035 |
Token | fox |
Token ID | 21831 |
Feature activation | +2.88e-01 |
Pos logit contributions | |
handed | +0.010 |
vendor | +0.008 |
sandbox | +0.008 |
amusement | +0.008 |
lobby | +0.007 |
Neg logit contributions | |
ches | -0.008 |
py | -0.007 |
Tiny | -0.007 |
fl | -0.007 |
Cotton | -0.007 |
Token | drew |
Token ID | 9859 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
ches | +0.008 |
Tiny | +0.008 |
py | +0.007 |
Cotton | +0.007 |
Ho | +0.007 |
Neg logit contributions | |
handed | -0.012 |
sandbox | -0.010 |
vendor | -0.010 |
Mike | -0.010 |
amusement | -0.010 |
Token | a |
Token ID | 257 |
Feature activation | -5.26e-03 |
Pos logit contributions | |
ches | +0.153 |
Tiny | +0.144 |
fl | +0.140 |
py | +0.136 |
Ho | +0.133 |
Neg logit contributions | |
handed | -0.241 |
vendor | -0.189 |
Mike | -0.186 |
sandbox | -0.183 |
amusement | -0.182 |
Token | happy |
Token ID | 3772 |
Feature activation | -8.52e-02 |
Pos logit contributions | |
handed | +0.010 |
sandbox | +0.009 |
vendor | +0.009 |
amusement | +0.008 |
amaz | +0.008 |
Neg logit contributions | |
ches | -0.009 |
py | -0.008 |
Tiny | -0.008 |
Cotton | -0.008 |
Pig | -0.008 |
Token | little |
Token ID | 1310 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
handed | +0.177 |
vendor | +0.138 |
sandbox | +0.138 |
Mike | +0.137 |
amusement | +0.136 |
Neg logit contributions | |
ches | -0.138 |
Tiny | -0.124 |
py | -0.122 |
fl | -0.119 |
Ho | -0.118 |
Token | bird |
Token ID | 6512 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
ches | +0.158 |
fl | +0.151 |
Tiny | +0.149 |
wand | +0.144 |
Ho | +0.141 |
Neg logit contributions | |
handed | -0.267 |
Mike | -0.212 |
vendor | -0.202 |
view | -0.199 |
Jane | -0.199 |
Token | flying |
Token ID | 7348 |
Feature activation | +2.63e-01 |
Pos logit contributions | |
ches | +0.059 |
Tiny | +0.056 |
Cotton | +0.055 |
Ho | +0.054 |
Pig | +0.054 |
Neg logit contributions | |
handed | -0.061 |
sandbox | -0.047 |
amusement | -0.045 |
view | -0.045 |
vendor | -0.045 |
Token | in |
Token ID | 287 |
Feature activation | +5.68e-02 |
Pos logit contributions | |
ches | +0.380 |
Tiny | +0.345 |
fl | +0.342 |
py | +0.335 |
Ho | +0.319 |
Neg logit contributions | |
handed | -0.605 |
Mike | -0.474 |
vendor | -0.472 |
sandbox | -0.468 |
amusement | -0.467 |
Token | the |
Token ID | 262 |
Feature activation | -2.94e-03 |
Pos logit contributions | |
fl | +0.064 |
Tiny | +0.064 |
ches | +0.061 |
Sparkle | +0.058 |
py | +0.057 |
Neg logit contributions | |
handed | -0.143 |
Mike | -0.115 |
vendor | -0.115 |
sandbox | -0.112 |
view | -0.111 |
Token | blue |
Token ID | 4171 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
handed | +0.007 |
Mike | +0.006 |
Jane | +0.006 |
aren | +0.006 |
amusement | +0.006 |
Neg logit contributions | |
fl | -0.003 |
ches | -0.003 |
Tiny | -0.003 |
wand | -0.003 |
Ho | -0.003 |
Token | sky |
Token ID | 6766 |
Feature activation | +3.23e-02 |
Pos logit contributions | |
ches | +0.262 |
Tiny | +0.248 |
fl | +0.244 |
Ho | +0.237 |
py | +0.235 |
Neg logit contributions | |
handed | -0.357 |
Mike | -0.281 |
vendor | -0.275 |
amusement | -0.268 |
Jane | -0.265 |
Token | . |
Token ID | 13 |
Feature activation | -9.59e-03 |
Pos logit contributions | |
ches | +0.051 |
Tiny | +0.047 |
py | +0.046 |
fl | +0.046 |
Ho | +0.044 |
Neg logit contributions | |
handed | -0.068 |
Mike | -0.053 |
vendor | -0.053 |
sandbox | -0.052 |
amusement | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | -3.17e-02 |
Pos logit contributions | |
ches | +0.054 |
Tiny | +0.050 |
py | +0.049 |
fl | +0.048 |
Ho | +0.047 |
Neg logit contributions | |
handed | -0.069 |
Mike | -0.054 |
vendor | -0.054 |
sandbox | -0.053 |
amusement | -0.053 |
Token | Put |
Token ID | 5930 |
Feature activation | -3.70e-02 |
Pos logit contributions | |
handed | +0.066 |
vendor | +0.051 |
Mike | +0.051 |
sandbox | +0.051 |
amusement | +0.050 |
Neg logit contributions | |
ches | -0.051 |
Tiny | -0.046 |
py | -0.045 |
fl | -0.045 |
Ho | -0.043 |
Token | them |
Token ID | 606 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
handed | +0.085 |
Mike | +0.069 |
vendor | +0.068 |
amusement | +0.068 |
sandbox | +0.067 |
Neg logit contributions | |
ches | -0.049 |
py | -0.046 |
Tiny | -0.045 |
fl | -0.042 |
Cotton | -0.042 |
Token | in |
Token ID | 287 |
Feature activation | +6.23e-02 |
Pos logit contributions | |
ches | +0.204 |
Tiny | +0.182 |
fl | +0.177 |
Ho | +0.171 |
py | +0.170 |
Neg logit contributions | |
handed | -0.371 |
Mike | -0.299 |
vendor | -0.295 |
sandbox | -0.293 |
amusement | -0.293 |
Token | your |
Token ID | 534 |
Feature activation | +8.66e-02 |
Pos logit contributions | |
ches | +0.091 |
Tiny | +0.085 |
fl | +0.083 |
Ho | +0.080 |
py | +0.079 |
Neg logit contributions | |
handed | -0.139 |
sandbox | -0.108 |
vendor | -0.108 |
Mike | -0.108 |
view | -0.104 |
Token | pig |
Token ID | 12967 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
ches | +0.140 |
Tiny | +0.137 |
wand | +0.134 |
Ho | +0.133 |
fl | +0.133 |
Neg logit contributions | |
handed | -0.181 |
Mike | -0.154 |
Jane | -0.144 |
view | -0.138 |
ill | -0.138 |
Token | gy |
Token ID | 1360 |
Feature activation | +8.08e-02 |
Pos logit contributions | |
ches | +0.333 |
Tiny | +0.302 |
py | +0.290 |
fl | +0.288 |
Ho | +0.282 |
Neg logit contributions | |
handed | -0.506 |
Mike | -0.404 |
vendor | -0.401 |
sandbox | -0.397 |
amusement | -0.397 |
Token | bank |
Token ID | 3331 |
Feature activation | -5.78e-02 |
Pos logit contributions | |
ches | +0.136 |
Tiny | +0.123 |
fl | +0.122 |
py | +0.118 |
Ho | +0.117 |
Neg logit contributions | |
handed | -0.163 |
Mike | -0.133 |
vendor | -0.125 |
amusement | -0.124 |
Jane | -0.124 |
Token | and |
Token ID | 290 |
Feature activation | +3.35e-02 |
Pos logit contributions | |
handed | +0.130 |
Mike | +0.104 |
amusement | +0.104 |
vendor | +0.103 |
sandbox | +0.103 |
Neg logit contributions | |
ches | -0.083 |
Tiny | -0.073 |
fl | -0.072 |
py | -0.070 |
Ho | -0.070 |
Token | it |
Token ID | 340 |
Feature activation | +4.13e-02 |
Pos logit contributions | |
ches | +0.051 |
Tiny | +0.046 |
py | +0.046 |
fl | +0.044 |
Ho | +0.043 |
Neg logit contributions | |
handed | -0.072 |
vendor | -0.057 |
sandbox | -0.057 |
Mike | -0.056 |
amusement | -0.056 |
Token | will |
Token ID | 481 |
Feature activation | +4.96e-03 |
Pos logit contributions | |
ches | +0.060 |
Tiny | +0.058 |
py | +0.058 |
Cotton | +0.055 |
fl | +0.053 |
Neg logit contributions | |
handed | -0.087 |
sandbox | -0.068 |
vendor | -0.066 |
lobby | -0.066 |
Mike | -0.064 |
Token | . |
Token ID | 13 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
ches | +0.208 |
Tiny | +0.192 |
py | +0.188 |
fl | +0.182 |
Ho | +0.181 |
Neg logit contributions | |
handed | -0.257 |
Mike | -0.200 |
vendor | -0.200 |
sandbox | -0.198 |
amusement | -0.197 |
Token | He |
Token ID | 679 |
Feature activation | -4.24e-03 |
Pos logit contributions | |
ches | +0.025 |
Tiny | +0.022 |
py | +0.022 |
fl | +0.022 |
Ho | +0.021 |
Neg logit contributions | |
handed | -0.029 |
vendor | -0.023 |
Mike | -0.023 |
sandbox | -0.023 |
amusement | -0.022 |
Token | puts |
Token ID | 7584 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
handed | +0.008 |
sandbox | +0.007 |
vendor | +0.007 |
amusement | +0.007 |
lobby | +0.006 |
Neg logit contributions | |
ches | -0.007 |
py | -0.007 |
Tiny | -0.007 |
Pig | -0.006 |
iny | -0.006 |
Token | it |
Token ID | 340 |
Feature activation | +2.81e-02 |
Pos logit contributions | |
ches | +0.193 |
py | +0.180 |
Tiny | +0.176 |
iny | +0.171 |
fl | +0.165 |
Neg logit contributions | |
handed | -0.218 |
Mike | -0.173 |
vendor | -0.171 |
sandbox | -0.170 |
amusement | -0.167 |
Token | on |
Token ID | 319 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
ches | +0.041 |
Tiny | +0.039 |
py | +0.036 |
Cotton | +0.036 |
Chick | +0.036 |
Neg logit contributions | |
handed | -0.056 |
sandbox | -0.048 |
vendor | -0.046 |
Mike | -0.045 |
unatt | -0.045 |
Token | his |
Token ID | 465 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
ches | +0.204 |
Tiny | +0.195 |
Ho | +0.186 |
fl | +0.183 |
Cotton | +0.179 |
Neg logit contributions | |
handed | -0.377 |
sandbox | -0.299 |
Mike | -0.298 |
vendor | -0.294 |
view | -0.293 |
Token | head |
Token ID | 1182 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
ches | +0.025 |
Tiny | +0.023 |
py | +0.022 |
fl | +0.022 |
Ho | +0.022 |
Neg logit contributions | |
handed | -0.039 |
Mike | -0.031 |
vendor | -0.031 |
sandbox | -0.031 |
amusement | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
ches | +0.178 |
Tiny | +0.161 |
fl | +0.158 |
py | +0.154 |
Ho | +0.152 |
Neg logit contributions | |
handed | -0.267 |
Mike | -0.209 |
vendor | -0.209 |
sandbox | -0.208 |
amusement | -0.206 |
Token | He |
Token ID | 679 |
Feature activation | -7.05e-03 |
Pos logit contributions | |
ches | +0.020 |
py | +0.018 |
Tiny | +0.018 |
fl | +0.017 |
Ho | +0.017 |
Neg logit contributions | |
handed | -0.023 |
vendor | -0.018 |
Mike | -0.018 |
sandbox | -0.017 |
amusement | -0.017 |
Token | looks |
Token ID | 3073 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
handed | +0.014 |
sandbox | +0.012 |
vendor | +0.012 |
amusement | +0.011 |
Mike | +0.011 |
Neg logit contributions | |
ches | -0.011 |
Tiny | -0.010 |
py | -0.010 |
Pig | -0.010 |
Cotton | -0.010 |
Token | at |
Token ID | 379 |
Feature activation | -8.95e-03 |
Pos logit contributions | |
handed | +0.191 |
vendor | +0.150 |
sandbox | +0.149 |
amusement | +0.146 |
Mike | +0.145 |
Neg logit contributions | |
ches | -0.178 |
py | -0.167 |
Tiny | -0.167 |
Cotton | -0.156 |
Pig | -0.156 |
Token | would |
Token ID | 561 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
ches | +0.029 |
Tiny | +0.027 |
py | +0.026 |
Ho | +0.025 |
Cotton | +0.025 |
Neg logit contributions | |
handed | -0.041 |
vendor | -0.033 |
sandbox | -0.033 |
amusement | -0.032 |
Mike | -0.032 |
Token | laugh |
Token ID | 6487 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
ches | +0.226 |
Tiny | +0.207 |
py | +0.202 |
fl | +0.197 |
Ho | +0.195 |
Neg logit contributions | |
handed | -0.294 |
vendor | -0.229 |
sandbox | -0.227 |
Mike | -0.227 |
amusement | -0.226 |
Token | together |
Token ID | 1978 |
Feature activation | -3.73e-02 |
Pos logit contributions | |
handed | +0.317 |
vendor | +0.244 |
Mike | +0.244 |
sandbox | +0.242 |
amusement | +0.240 |
Neg logit contributions | |
ches | -0.249 |
Tiny | -0.229 |
py | -0.222 |
fl | -0.221 |
Ho | -0.216 |
Token | . |
Token ID | 13 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
handed | +0.085 |
Mike | +0.067 |
sandbox | +0.066 |
vendor | +0.066 |
lobby | +0.065 |
Neg logit contributions | |
ches | -0.055 |
fl | -0.049 |
Tiny | -0.049 |
py | -0.048 |
Sparkle | -0.046 |
Token | The |
Token ID | 383 |
Feature activation | -6.15e-02 |
Pos logit contributions | |
handed | +0.025 |
vendor | +0.019 |
Mike | +0.019 |
sandbox | +0.019 |
amusement | +0.019 |
Neg logit contributions | |
ches | -0.019 |
Tiny | -0.018 |
py | -0.017 |
fl | -0.017 |
Ho | -0.017 |
Token | ancient |
Token ID | 6156 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
handed | +0.138 |
sandbox | +0.110 |
vendor | +0.110 |
amusement | +0.109 |
Mike | +0.107 |
Neg logit contributions | |
ches | -0.090 |
Tiny | -0.081 |
py | -0.079 |
Ho | -0.076 |
fl | -0.075 |
Token | rose |
Token ID | 8278 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
ches | +0.037 |
Tiny | +0.033 |
py | +0.032 |
Ho | +0.031 |
Cotton | +0.030 |
Neg logit contributions | |
handed | -0.050 |
sandbox | -0.041 |
vendor | -0.041 |
amusement | -0.040 |
Mike | -0.039 |
Token | was |
Token ID | 373 |
Feature activation | -8.52e-02 |
Pos logit contributions | |
ches | +0.236 |
Tiny | +0.213 |
fl | +0.211 |
py | +0.208 |
Ho | +0.199 |
Neg logit contributions | |
handed | -0.322 |
Mike | -0.252 |
vendor | -0.248 |
sandbox | -0.247 |
amusement | -0.246 |
Token | very |
Token ID | 845 |
Feature activation | -5.02e-02 |
Pos logit contributions | |
handed | +0.165 |
vendor | +0.125 |
Mike | +0.124 |
sandbox | +0.124 |
amusement | +0.123 |
Neg logit contributions | |
ches | -0.149 |
Tiny | -0.138 |
py | -0.135 |
fl | -0.133 |
Ho | -0.131 |
Token | happy |
Token ID | 3772 |
Feature activation | -9.23e-02 |
Pos logit contributions | |
handed | +0.107 |
Mike | +0.083 |
sandbox | +0.083 |
amusement | +0.080 |
lobby | +0.080 |
Neg logit contributions | |
ches | -0.079 |
fl | -0.074 |
Tiny | -0.073 |
py | -0.070 |
Ho | -0.068 |
Token | because |
Token ID | 780 |
Feature activation | -5.41e-02 |
Pos logit contributions | |
handed | +0.234 |
lobby | +0.184 |
Mike | +0.183 |
sandbox | +0.182 |
amusement | +0.182 |
Neg logit contributions | |
ches | -0.110 |
fl | -0.108 |
Tiny | -0.107 |
Sparkle | -0.101 |
py | -0.100 |
Token | friends |
Token ID | 2460 |
Feature activation | +9.14e-02 |
Pos logit contributions | |
handed | +0.464 |
sandbox | +0.374 |
vendor | +0.371 |
lobby | +0.363 |
amusement | +0.356 |
Neg logit contributions | |
ches | -0.405 |
Tiny | -0.359 |
py | -0.354 |
fl | -0.338 |
Cotton | -0.337 |
Token | who |
Token ID | 508 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
ches | +0.120 |
fl | +0.108 |
Tiny | +0.107 |
py | +0.102 |
Ho | +0.101 |
Neg logit contributions | |
handed | -0.217 |
sandbox | -0.171 |
Mike | -0.170 |
lobby | -0.168 |
vendor | -0.167 |
Token | liked |
Token ID | 8288 |
Feature activation | -5.30e-02 |
Pos logit contributions | |
handed | +0.270 |
amusement | +0.209 |
Mike | +0.208 |
unatt | +0.207 |
amaz | +0.207 |
Neg logit contributions | |
ches | -0.155 |
fl | -0.144 |
Tiny | -0.143 |
Ho | -0.137 |
py | -0.137 |
Token | to |
Token ID | 284 |
Feature activation | -3.44e-02 |
Pos logit contributions | |
handed | +0.107 |
vendor | +0.083 |
Mike | +0.082 |
sandbox | +0.082 |
amusement | +0.081 |
Neg logit contributions | |
ches | -0.089 |
Tiny | -0.081 |
py | -0.079 |
fl | -0.078 |
Ho | -0.077 |
Token | play |
Token ID | 711 |
Feature activation | -6.68e-02 |
Pos logit contributions | |
handed | +0.069 |
Mike | +0.053 |
vendor | +0.052 |
sandbox | +0.052 |
amusement | +0.051 |
Neg logit contributions | |
ches | -0.058 |
Tiny | -0.054 |
py | -0.052 |
fl | -0.052 |
Ho | -0.050 |
Token | in |
Token ID | 287 |
Feature activation | +4.00e-02 |
Pos logit contributions | |
handed | +0.128 |
sandbox | +0.101 |
Mike | +0.097 |
vendor | +0.097 |
amusement | +0.097 |
Neg logit contributions | |
ches | -0.116 |
py | -0.105 |
Tiny | -0.105 |
Cotton | -0.100 |
Ho | -0.099 |
Token | the |
Token ID | 262 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
Tiny | +0.046 |
fl | +0.044 |
ches | +0.040 |
Sparkle | +0.040 |
Ho | +0.040 |
Neg logit contributions | |
handed | -0.103 |
vendor | -0.082 |
Mike | -0.081 |
unatt | -0.081 |
aw | -0.079 |
Token | garden |
Token ID | 11376 |
Feature activation | -5.52e-02 |
Pos logit contributions | |
handed | +0.025 |
Mike | +0.020 |
Jane | +0.018 |
aren | +0.018 |
vendor | +0.018 |
Neg logit contributions | |
Tiny | -0.026 |
fl | -0.026 |
ches | -0.026 |
Ho | -0.025 |
Pig | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
handed | +0.118 |
Mike | +0.092 |
vendor | +0.091 |
sandbox | +0.090 |
amusement | +0.090 |
Neg logit contributions | |
ches | -0.085 |
Tiny | -0.078 |
fl | -0.076 |
py | -0.076 |
Ho | -0.074 |
Token | One |
Token ID | 1881 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
handed | +0.032 |
Mike | +0.027 |
vendor | +0.026 |
sandbox | +0.026 |
amusement | +0.026 |
Neg logit contributions | |
ches | -0.010 |
Tiny | -0.009 |
py | -0.008 |
fl | -0.008 |
Ho | -0.008 |
Token | day |
Token ID | 1110 |
Feature activation | +5.81e-02 |
Pos logit contributions | |
ches | +0.198 |
Tiny | +0.178 |
fl | +0.175 |
py | +0.169 |
iny | +0.159 |
Neg logit contributions | |
handed | -0.469 |
Mike | -0.383 |
sandbox | -0.370 |
amusement | -0.369 |
vendor | -0.367 |
Token | big |
Token ID | 1263 |
Feature activation | +2.10e-02 |
Pos logit contributions | |
handed | +0.042 |
vendor | +0.035 |
sandbox | +0.034 |
amusement | +0.033 |
Mike | +0.033 |
Neg logit contributions | |
ches | -0.032 |
py | -0.029 |
Tiny | -0.028 |
Cotton | -0.026 |
Ho | -0.026 |
Token | whale |
Token ID | 22206 |
Feature activation | -4.78e-02 |
Pos logit contributions | |
ches | +0.031 |
Tiny | +0.029 |
fl | +0.028 |
py | +0.028 |
Ho | +0.027 |
Neg logit contributions | |
handed | -0.046 |
Mike | -0.036 |
vendor | -0.035 |
sandbox | -0.035 |
amusement | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
handed | +0.091 |
vendor | +0.070 |
sandbox | +0.069 |
Mike | +0.069 |
amusement | +0.068 |
Neg logit contributions | |
ches | -0.083 |
Tiny | -0.077 |
py | -0.076 |
Cotton | -0.074 |
Ho | -0.073 |
Token | |
Token ID | 220 |
Feature activation | +5.38e-02 |
Pos logit contributions | |
handed | +0.027 |
vendor | +0.022 |
sandbox | +0.020 |
Mike | +0.019 |
lobby | +0.019 |
Neg logit contributions | |
ches | -0.037 |
py | -0.034 |
wand | -0.033 |
fl | -0.032 |
iny | -0.031 |
Token |
|
Token ID | 198 |
Feature activation | +7.16e-03 |
Pos logit contributions | |
ches | +0.074 |
Tiny | +0.068 |
py | +0.064 |
fl | +0.062 |
Ho | +0.062 |
Neg logit contributions | |
handed | -0.124 |
sandbox | -0.100 |
amusement | -0.100 |
vendor | -0.099 |
Mike | -0.097 |
Token |
|
Token ID | 198 |
Feature activation | +7.23e-03 |
Pos logit contributions | |
ches | +0.012 |
py | +0.011 |
Tiny | +0.011 |
fl | +0.011 |
Ho | +0.011 |
Neg logit contributions | |
handed | -0.014 |
Mike | -0.011 |
vendor | -0.011 |
sandbox | -0.011 |
amusement | -0.011 |
Token | The |
Token ID | 464 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
ches | +0.013 |
py | +0.012 |
Tiny | +0.012 |
Ho | +0.012 |
wand | +0.012 |
Neg logit contributions | |
handed | -0.014 |
vendor | -0.011 |
Mike | -0.010 |
amusement | -0.010 |
sandbox | -0.010 |
Token | child |
Token ID | 1200 |
Feature activation | -2.39e-01 |
Pos logit contributions | |
handed | +0.256 |
Mike | +0.193 |
sandbox | +0.187 |
vendor | +0.183 |
amusement | +0.183 |
Neg logit contributions | |
ches | -0.227 |
fl | -0.217 |
Tiny | -0.216 |
py | -0.210 |
Ho | -0.203 |
Token | was |
Token ID | 373 |
Feature activation | -9.26e-02 |
Pos logit contributions | |
handed | +0.406 |
vendor | +0.326 |
sandbox | +0.323 |
Mike | +0.322 |
lobby | +0.301 |
Neg logit contributions | |
ches | -0.433 |
Tiny | -0.409 |
Cotton | -0.403 |
py | -0.402 |
Ho | -0.390 |
Token | excited |
Token ID | 6568 |
Feature activation | -4.12e-02 |
Pos logit contributions | |
handed | +0.146 |
vendor | +0.113 |
sandbox | +0.108 |
Mike | +0.108 |
amusement | +0.106 |
Neg logit contributions | |
ches | -0.197 |
Tiny | -0.169 |
py | -0.167 |
Cotton | -0.166 |
wand | -0.163 |
Token | , |
Token ID | 11 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
handed | +0.097 |
Mike | +0.077 |
sandbox | +0.077 |
lobby | +0.076 |
amusement | +0.076 |
Neg logit contributions | |
ches | -0.057 |
Tiny | -0.052 |
fl | -0.051 |
py | -0.049 |
Ho | -0.047 |
Token | very |
Token ID | 845 |
Feature activation | -5.02e-02 |
Pos logit contributions | |
handed | +0.184 |
sandbox | +0.140 |
vendor | +0.139 |
Mike | +0.139 |
amusement | +0.136 |
Neg logit contributions | |
ches | -0.146 |
Tiny | -0.134 |
fl | -0.134 |
py | -0.131 |
Ho | -0.126 |
Token | big |
Token ID | 1263 |
Feature activation | +2.73e-02 |
Pos logit contributions | |
handed | +0.108 |
sandbox | +0.082 |
Mike | +0.082 |
amusement | +0.081 |
view | +0.080 |
Neg logit contributions | |
ches | -0.078 |
fl | -0.072 |
Tiny | -0.071 |
Ho | -0.068 |
py | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
ches | +0.035 |
fl | +0.031 |
Tiny | +0.031 |
wand | +0.029 |
Ho | +0.029 |
Neg logit contributions | |
handed | -0.067 |
Mike | -0.056 |
Jane | -0.054 |
sandbox | -0.053 |
vendor | -0.053 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | +1.67e-02 |
Pos logit contributions | |
handed | +0.045 |
sandbox | +0.037 |
amusement | +0.037 |
vendor | +0.037 |
Mike | +0.036 |
Neg logit contributions | |
Tiny | -0.023 |
ches | -0.022 |
Cotton | -0.020 |
Ho | -0.020 |
Pig | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | -2.80e-02 |
Pos logit contributions | |
Tiny | +0.022 |
ches | +0.022 |
Ho | +0.020 |
Pig | +0.020 |
fl | +0.020 |
Neg logit contributions | |
handed | -0.040 |
view | -0.032 |
vendor | -0.032 |
sandbox | -0.031 |
amusement | -0.031 |
Token | he |
Token ID | 339 |
Feature activation | -3.04e-02 |
Pos logit contributions | |
handed | +0.068 |
view | +0.057 |
apartment | +0.056 |
aren | +0.056 |
sandbox | +0.055 |
Neg logit contributions | |
Tiny | -0.035 |
fl | -0.030 |
Pig | -0.030 |
Cotton | -0.030 |
Ho | -0.030 |
Token | spotted |
Token ID | 13489 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
handed | +0.061 |
vendor | +0.046 |
sandbox | +0.045 |
Mike | +0.044 |
amusement | +0.044 |
Neg logit contributions | |
ches | -0.055 |
Tiny | -0.050 |
fl | -0.049 |
py | -0.048 |
Ho | -0.047 |
Token | something |
Token ID | 1223 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
handed | +0.285 |
apartment | +0.239 |
sandbox | +0.235 |
vendor | +0.232 |
Mike | +0.227 |
Neg logit contributions | |
Tiny | -0.189 |
ches | -0.183 |
fl | -0.180 |
Ho | -0.176 |
Pig | -0.175 |
Token | shiny |
Token ID | 22441 |
Feature activation | +6.00e-02 |
Pos logit contributions | |
handed | +0.515 |
apartment | +0.404 |
sandbox | +0.396 |
lobby | +0.390 |
Mike | +0.390 |
Neg logit contributions | |
ches | -0.301 |
fl | -0.295 |
wand | -0.266 |
Tiny | -0.261 |
hopping | -0.260 |
Token | in |
Token ID | 287 |
Feature activation | +4.54e-02 |
Pos logit contributions | |
ches | +0.071 |
fl | +0.068 |
wand | +0.063 |
Tiny | +0.057 |
Ho | +0.057 |
Neg logit contributions | |
handed | -0.154 |
Mike | -0.123 |
apartment | -0.122 |
Jane | -0.121 |
sandbox | -0.121 |
Token | the |
Token ID | 262 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
fl | +0.053 |
Tiny | +0.053 |
ches | +0.052 |
Ho | +0.049 |
Cotton | +0.049 |
Neg logit contributions | |
handed | -0.109 |
sandbox | -0.087 |
Mike | -0.085 |
lobby | -0.084 |
vendor | -0.084 |
Token | upon |
Token ID | 2402 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
ches | +0.061 |
Tiny | +0.056 |
py | +0.055 |
fl | +0.054 |
Ho | +0.053 |
Neg logit contributions | |
handed | -0.084 |
vendor | -0.066 |
Mike | -0.065 |
sandbox | -0.065 |
amusement | -0.064 |
Token | a |
Token ID | 257 |
Feature activation | +5.32e-03 |
Pos logit contributions | |
ches | +0.043 |
fl | +0.042 |
Tiny | +0.042 |
py | +0.038 |
Ho | +0.038 |
Neg logit contributions | |
handed | -0.092 |
vendor | -0.076 |
lobby | -0.074 |
apartment | -0.074 |
aren | -0.074 |
Token | time |
Token ID | 640 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
ches | +0.006 |
Tiny | +0.005 |
fl | +0.005 |
py | +0.005 |
Ho | +0.004 |
Neg logit contributions | |
handed | -0.014 |
Mike | -0.011 |
vendor | -0.011 |
sandbox | -0.011 |
amusement | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | -5.26e-04 |
Pos logit contributions | |
ches | +0.185 |
Tiny | +0.179 |
Ho | +0.170 |
fl | +0.169 |
py | +0.168 |
Neg logit contributions | |
handed | -0.359 |
amusement | -0.282 |
view | -0.279 |
vendor | -0.277 |
sandbox | -0.276 |
Token | there |
Token ID | 612 |
Feature activation | -4.68e-02 |
Pos logit contributions | |
handed | +0.001 |
Mike | +0.001 |
vendor | +0.001 |
sandbox | +0.001 |
amusement | +0.001 |
Neg logit contributions | |
ches | -0.001 |
Tiny | -0.001 |
py | -0.001 |
fl | -0.001 |
Ho | -0.001 |
Token | was |
Token ID | 373 |
Feature activation | -7.05e-02 |
Pos logit contributions | |
handed | +0.134 |
unatt | +0.108 |
amusement | +0.106 |
vendor | +0.106 |
view | +0.105 |
Neg logit contributions | |
fl | -0.041 |
ches | -0.039 |
Tiny | -0.038 |
Ho | -0.037 |
hopping | -0.035 |
Token | a |
Token ID | 257 |
Feature activation | +4.77e-03 |
Pos logit contributions | |
handed | +0.147 |
vendor | +0.114 |
Mike | +0.114 |
sandbox | +0.113 |
amusement | +0.112 |
Neg logit contributions | |
ches | -0.113 |
Tiny | -0.104 |
py | -0.101 |
fl | -0.100 |
Ho | -0.098 |
Token | good |
Token ID | 922 |
Feature activation | +5.69e-02 |
Pos logit contributions | |
ches | +0.007 |
py | +0.006 |
Tiny | +0.006 |
fl | +0.006 |
Sparkle | +0.006 |
Neg logit contributions | |
handed | -0.010 |
vendor | -0.009 |
sandbox | -0.009 |
Mike | -0.008 |
amusement | -0.008 |
Token | cat |
Token ID | 3797 |
Feature activation | +8.65e-02 |
Pos logit contributions | |
ches | +0.100 |
Tiny | +0.090 |
py | +0.089 |
fl | +0.086 |
Cotton | +0.085 |
Neg logit contributions | |
handed | -0.110 |
vendor | -0.088 |
sandbox | -0.086 |
Mike | -0.084 |
amusement | -0.083 |
Token | named |
Token ID | 3706 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
ches | +0.120 |
Tiny | +0.106 |
py | +0.104 |
fl | +0.101 |
Cotton | +0.101 |
Neg logit contributions | |
handed | -0.197 |
vendor | -0.158 |
sandbox | -0.154 |
Mike | -0.154 |
amusement | -0.153 |
Token | Kitty |
Token ID | 21168 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
ches | +0.260 |
py | +0.256 |
wand | +0.248 |
fl | +0.243 |
tle | +0.237 |
Neg logit contributions | |
handed | -0.250 |
vendor | -0.194 |
amusement | -0.191 |
sandbox | -0.188 |
Jane | -0.185 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.06e-02 |
Pos logit contributions | |
ches | +0.073 |
Tiny | +0.066 |
Ho | +0.061 |
Sparkle | +0.060 |
py | +0.060 |
Neg logit contributions | |
handed | -0.143 |
sandbox | -0.120 |
vendor | -0.118 |
amusement | -0.117 |
amaz | -0.114 |
Token | upon |
Token ID | 2402 |
Feature activation | +4.05e-02 |
Pos logit contributions | |
ches | +0.065 |
Tiny | +0.059 |
py | +0.058 |
fl | +0.057 |
Ho | +0.056 |
Neg logit contributions | |
handed | -0.084 |
vendor | -0.066 |
Mike | -0.065 |
sandbox | -0.065 |
amusement | -0.064 |
Token | a |
Token ID | 257 |
Feature activation | +5.04e-03 |
Pos logit contributions | |
ches | +0.048 |
fl | +0.046 |
Tiny | +0.046 |
py | +0.043 |
Ho | +0.042 |
Neg logit contributions | |
handed | -0.097 |
vendor | -0.080 |
lobby | -0.079 |
aren | -0.079 |
Mike | -0.078 |
Token | time |
Token ID | 640 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
ches | +0.005 |
Tiny | +0.004 |
fl | +0.004 |
py | +0.004 |
iny | +0.004 |
Neg logit contributions | |
handed | -0.014 |
Mike | -0.011 |
vendor | -0.011 |
amusement | -0.011 |
sandbox | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | -3.07e-03 |
Pos logit contributions | |
ches | +0.170 |
Tiny | +0.168 |
Ho | +0.159 |
fl | +0.156 |
py | +0.153 |
Neg logit contributions | |
handed | -0.372 |
amusement | -0.295 |
view | -0.293 |
vendor | -0.288 |
sandbox | -0.288 |
Token | there |
Token ID | 612 |
Feature activation | -4.73e-02 |
Pos logit contributions | |
handed | +0.005 |
Mike | +0.004 |
sandbox | +0.004 |
vendor | +0.004 |
amusement | +0.004 |
Neg logit contributions | |
ches | -0.006 |
Tiny | -0.005 |
py | -0.005 |
fl | -0.005 |
Ho | -0.005 |
Token | was |
Token ID | 373 |
Feature activation | -7.00e-02 |
Pos logit contributions | |
handed | +0.135 |
unatt | +0.109 |
har | +0.109 |
amusement | +0.107 |
view | +0.106 |
Neg logit contributions | |
fl | -0.040 |
ches | -0.039 |
Tiny | -0.038 |
Ho | -0.036 |
hopping | -0.035 |
Token | a |
Token ID | 257 |
Feature activation | +5.62e-03 |
Pos logit contributions | |
handed | +0.148 |
vendor | +0.115 |
Mike | +0.115 |
sandbox | +0.114 |
amusement | +0.113 |
Neg logit contributions | |
ches | -0.110 |
Tiny | -0.101 |
py | -0.098 |
fl | -0.097 |
Ho | -0.095 |
Token | boy |
Token ID | 2933 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
ches | +0.008 |
py | +0.007 |
Tiny | +0.007 |
fl | +0.007 |
Sparkle | +0.007 |
Neg logit contributions | |
handed | -0.012 |
vendor | -0.010 |
sandbox | -0.010 |
Mike | -0.010 |
amusement | -0.010 |
Token | named |
Token ID | 3706 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
handed | +0.449 |
vendor | +0.361 |
Mike | +0.357 |
sandbox | +0.356 |
amusement | +0.351 |
Neg logit contributions | |
ches | -0.273 |
Tiny | -0.239 |
py | -0.234 |
fl | -0.225 |
Cotton | -0.225 |
Token | Tim |
Token ID | 5045 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
ches | +0.333 |
py | +0.310 |
fl | +0.306 |
Tiny | +0.306 |
wand | +0.301 |
Neg logit contributions | |
handed | -0.211 |
vendor | -0.147 |
Mike | -0.144 |
sandbox | -0.143 |
amusement | -0.138 |
Token |
|
Token ID | 198 |
Feature activation | +3.32e-02 |
Pos logit contributions | |
ches | +0.062 |
py | +0.058 |
fl | +0.057 |
wand | +0.057 |
Tiny | +0.055 |
Neg logit contributions | |
handed | -0.063 |
Mike | -0.050 |
Jane | -0.048 |
vendor | -0.046 |
he | -0.044 |
Token | M |
Token ID | 44 |
Feature activation | +6.49e-02 |
Pos logit contributions | |
ches | +0.063 |
py | +0.058 |
Tiny | +0.057 |
fl | +0.056 |
Ho | +0.056 |
Neg logit contributions | |
handed | -0.061 |
Mike | -0.047 |
Jane | -0.045 |
vendor | -0.045 |
sandbox | -0.044 |
Token | ama |
Token ID | 1689 |
Feature activation | +7.46e-02 |
Pos logit contributions | |
ches | +0.114 |
Tiny | +0.105 |
py | +0.103 |
fl | +0.101 |
Ho | +0.099 |
Neg logit contributions | |
handed | -0.127 |
sandbox | -0.096 |
Mike | -0.095 |
vendor | -0.095 |
amusement | -0.094 |
Token | and |
Token ID | 290 |
Feature activation | +2.20e-02 |
Pos logit contributions | |
ches | +0.119 |
Tiny | +0.110 |
py | +0.107 |
fl | +0.105 |
Ho | +0.104 |
Neg logit contributions | |
handed | -0.152 |
Mike | -0.120 |
vendor | -0.120 |
sandbox | -0.118 |
amusement | -0.116 |
Token | Little |
Token ID | 7703 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
ches | +0.036 |
Tiny | +0.033 |
py | +0.032 |
fl | +0.032 |
Ho | +0.031 |
Neg logit contributions | |
handed | -0.045 |
vendor | -0.035 |
Mike | -0.035 |
sandbox | -0.035 |
amusement | -0.034 |
Token | Jack |
Token ID | 3619 |
Feature activation | -4.67e-02 |
Pos logit contributions | |
ches | +0.023 |
Tiny | +0.022 |
Pig | +0.020 |
py | +0.020 |
Ho | +0.020 |
Neg logit contributions | |
handed | -0.027 |
view | -0.020 |
lobby | -0.020 |
sandbox | -0.020 |
amusement | -0.020 |
Token | waited |
Token ID | 13488 |
Feature activation | -7.74e-03 |
Pos logit contributions | |
handed | +0.098 |
vendor | +0.075 |
Mike | +0.075 |
amusement | +0.074 |
sandbox | +0.074 |
Neg logit contributions | |
ches | -0.076 |
fl | -0.069 |
Tiny | -0.068 |
py | -0.067 |
Ho | -0.064 |
Token | and |
Token ID | 290 |
Feature activation | +2.08e-02 |
Pos logit contributions | |
handed | +0.017 |
Mike | +0.014 |
vendor | +0.014 |
sandbox | +0.014 |
amusement | +0.014 |
Neg logit contributions | |
ches | -0.011 |
Tiny | -0.010 |
py | -0.010 |
Ho | -0.009 |
Cotton | -0.009 |
Token | waited |
Token ID | 13488 |
Feature activation | -8.14e-03 |
Pos logit contributions | |
ches | +0.031 |
fl | +0.029 |
Tiny | +0.029 |
Ho | +0.027 |
py | +0.027 |
Neg logit contributions | |
handed | -0.046 |
vendor | -0.035 |
sandbox | -0.035 |
amusement | -0.035 |
Mike | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | +7.90e-03 |
Pos logit contributions | |
handed | +0.017 |
Mike | +0.014 |
vendor | +0.013 |
sandbox | +0.013 |
amusement | +0.013 |
Neg logit contributions | |
ches | -0.012 |
Tiny | -0.011 |
py | -0.011 |
Ho | -0.011 |
Cotton | -0.011 |
Token | The |
Token ID | 383 |
Feature activation | -5.18e-02 |
Pos logit contributions | |
ches | +0.012 |
py | +0.011 |
Tiny | +0.011 |
fl | +0.011 |
Ho | +0.011 |
Neg logit contributions | |
handed | -0.017 |
Mike | -0.013 |
vendor | -0.013 |
sandbox | -0.013 |
amusement | -0.013 |
Token | 't |
Token ID | 470 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
ches | +0.061 |
py | +0.058 |
Tiny | +0.057 |
wand | +0.056 |
fl | +0.055 |
Neg logit contributions | |
handed | -0.121 |
Mike | -0.099 |
he | -0.097 |
amusement | -0.095 |
vendor | -0.095 |
Token | listen |
Token ID | 6004 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
ches | +0.376 |
Tiny | +0.352 |
fl | +0.339 |
py | +0.338 |
Ho | +0.336 |
Neg logit contributions | |
handed | -0.322 |
vendor | -0.236 |
amusement | -0.230 |
Mike | -0.230 |
sandbox | -0.228 |
Token | and |
Token ID | 290 |
Feature activation | -4.93e-03 |
Pos logit contributions | |
ches | +0.018 |
Tiny | +0.016 |
fl | +0.016 |
py | +0.015 |
Ho | +0.015 |
Neg logit contributions | |
handed | -0.026 |
sandbox | -0.021 |
Mike | -0.020 |
vendor | -0.020 |
amusement | -0.020 |
Token | kept |
Token ID | 4030 |
Feature activation | +1.74e-02 |
Pos logit contributions | |
handed | +0.012 |
Mike | +0.010 |
view | +0.010 |
apartment | +0.009 |
amusement | +0.009 |
Neg logit contributions | |
ches | -0.006 |
Tiny | -0.006 |
fl | -0.006 |
Ho | -0.006 |
Cotton | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | -1.75e-02 |
Pos logit contributions | |
ches | +0.032 |
py | +0.028 |
iny | +0.027 |
wand | +0.026 |
bee | +0.026 |
Neg logit contributions | |
handed | -0.034 |
vendor | -0.026 |
Mike | -0.025 |
amusement | -0.025 |
Mike | -0.025 |
Token | towel |
Token ID | 24808 |
Feature activation | -4.58e-02 |
Pos logit contributions | |
handed | +0.036 |
Mike | +0.028 |
vendor | +0.028 |
amusement | +0.027 |
sandbox | +0.027 |
Neg logit contributions | |
ches | -0.028 |
fl | -0.026 |
Tiny | -0.026 |
py | -0.025 |
wand | -0.025 |
Token | for |
Token ID | 329 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
handed | +0.097 |
Mike | +0.084 |
amusement | +0.082 |
vendor | +0.081 |
sandbox | +0.079 |
Neg logit contributions | |
ches | -0.071 |
Tiny | -0.063 |
fl | -0.063 |
Ho | -0.060 |
wand | -0.059 |
Token | himself |
Token ID | 2241 |
Feature activation | -8.49e-02 |
Pos logit contributions | |
ches | +0.295 |
Tiny | +0.264 |
py | +0.262 |
fl | +0.259 |
Ho | +0.252 |
Neg logit contributions | |
handed | -0.320 |
Mike | -0.242 |
vendor | -0.240 |
amusement | -0.238 |
sandbox | -0.237 |
Token | . |
Token ID | 13 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
handed | +0.192 |
Mike | +0.160 |
sandbox | +0.159 |
lobby | +0.157 |
amusement | +0.157 |
Neg logit contributions | |
ches | -0.120 |
Tiny | -0.106 |
fl | -0.105 |
py | -0.102 |
Ho | -0.101 |
Token | But |
Token ID | 887 |
Feature activation | +2.62e-02 |
Pos logit contributions | |
handed | +0.047 |
Mike | +0.037 |
vendor | +0.037 |
sandbox | +0.036 |
amusement | +0.036 |
Neg logit contributions | |
ches | -0.034 |
Tiny | -0.031 |
py | -0.030 |
fl | -0.030 |
Ho | -0.029 |
Token | then |
Token ID | 788 |
Feature activation | -3.60e-03 |
Pos logit contributions | |
ches | +0.043 |
Tiny | +0.039 |
py | +0.038 |
fl | +0.037 |
Ho | +0.037 |
Neg logit contributions | |
handed | -0.054 |
vendor | -0.042 |
Mike | -0.042 |
sandbox | -0.042 |
amusement | -0.041 |
Token | hands |
Token ID | 2832 |
Feature activation | -1.36e-03 |
Pos logit contributions | |
ches | +0.077 |
py | +0.072 |
Tiny | +0.072 |
Ho | +0.068 |
Cotton | +0.068 |
Neg logit contributions | |
handed | -0.085 |
vendor | -0.067 |
sandbox | -0.066 |
amaz | -0.064 |
Mike | -0.063 |
Token | and |
Token ID | 290 |
Feature activation | +1.26e-03 |
Pos logit contributions | |
handed | +0.003 |
sandbox | +0.002 |
Mike | +0.002 |
vendor | +0.002 |
amusement | +0.002 |
Neg logit contributions | |
ches | -0.002 |
Tiny | -0.002 |
fl | -0.002 |
py | -0.002 |
Ho | -0.002 |
Token | said |
Token ID | 531 |
Feature activation | +8.08e-02 |
Pos logit contributions | |
ches | +0.002 |
Tiny | +0.002 |
fl | +0.002 |
Ho | +0.002 |
Sparkle | +0.002 |
Neg logit contributions | |
handed | -0.003 |
vendor | -0.002 |
amusement | -0.002 |
Mike | -0.002 |
apartment | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
ches | +0.116 |
Tiny | +0.106 |
py | +0.103 |
fl | +0.102 |
Ho | +0.100 |
Neg logit contributions | |
handed | -0.183 |
sandbox | -0.142 |
Mike | -0.142 |
vendor | -0.142 |
amusement | -0.142 |
Token | " |
Token ID | 366 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
handed | +0.042 |
vendor | +0.041 |
Mike | +0.039 |
Jane | +0.038 |
Mike | +0.037 |
Neg logit contributions | |
ches | -0.048 |
py | -0.045 |
agons | -0.042 |
iny | -0.042 |
Tiny | -0.040 |
Token | I |
Token ID | 40 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
ches | +0.347 |
Tiny | +0.326 |
py | +0.320 |
fl | +0.312 |
Ho | +0.312 |
Neg logit contributions | |
handed | -0.253 |
Mike | -0.184 |
vendor | -0.180 |
sandbox | -0.177 |
amusement | -0.174 |
Token | love |
Token ID | 1842 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
handed | +0.250 |
vendor | +0.186 |
Mike | +0.186 |
sandbox | +0.185 |
amusement | +0.183 |
Neg logit contributions | |
ches | -0.251 |
Tiny | -0.233 |
py | -0.228 |
fl | -0.224 |
Ho | -0.222 |
Token | my |
Token ID | 616 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
handed | +0.183 |
vendor | +0.154 |
Mike | +0.148 |
sandbox | +0.145 |
amusement | +0.143 |
Neg logit contributions | |
ches | -0.190 |
py | -0.166 |
Tiny | -0.165 |
Cotton | -0.162 |
Ho | -0.160 |
Token | tape |
Token ID | 9154 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
ches | +0.169 |
Tiny | +0.155 |
py | +0.150 |
fl | +0.149 |
Ho | +0.145 |
Neg logit contributions | |
handed | -0.245 |
Mike | -0.192 |
vendor | -0.190 |
sandbox | -0.189 |
amusement | -0.188 |
Token | measure |
Token ID | 3953 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
ches | +0.017 |
Tiny | +0.015 |
fl | +0.015 |
py | +0.015 |
wand | +0.014 |
Neg logit contributions | |
handed | -0.030 |
Mike | -0.024 |
amusement | -0.023 |
sandbox | -0.023 |
view | -0.023 |
Token | !" |
Token ID | 2474 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
handed | +0.265 |
amusement | +0.208 |
Mike | +0.207 |
sandbox | +0.206 |
alley | +0.202 |
Neg logit contributions | |
ches | -0.145 |
Tiny | -0.126 |
py | -0.125 |
fl | -0.125 |
wand | -0.118 |
Token | idea |
Token ID | 2126 |
Feature activation | +7.89e-02 |
Pos logit contributions | |
handed | +0.114 |
Mike | +0.092 |
sandbox | +0.092 |
vendor | +0.089 |
lobby | +0.089 |
Neg logit contributions | |
ches | -0.059 |
py | -0.053 |
fl | -0.052 |
Tiny | -0.052 |
wand | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | -5.00e-04 |
Pos logit contributions | |
ches | +0.114 |
Tiny | +0.103 |
fl | +0.102 |
py | +0.099 |
wand | +0.094 |
Neg logit contributions | |
handed | -0.178 |
sandbox | -0.143 |
vendor | -0.140 |
Mike | -0.138 |
lobby | -0.137 |
Token | She |
Token ID | 1375 |
Feature activation | -2.62e-02 |
Pos logit contributions | |
handed | +0.001 |
sandbox | +0.001 |
Mike | +0.001 |
amusement | +0.001 |
vendor | +0.001 |
Neg logit contributions | |
ches | -0.001 |
Tiny | -0.001 |
fl | -0.001 |
Ho | -0.001 |
py | -0.001 |
Token | asked |
Token ID | 1965 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
handed | +0.052 |
vendor | +0.043 |
Mike | +0.042 |
amusement | +0.041 |
sandbox | +0.041 |
Neg logit contributions | |
ches | -0.040 |
py | -0.039 |
Tiny | -0.038 |
Ho | -0.037 |
Cotton | -0.037 |
Token | the |
Token ID | 262 |
Feature activation | +4.00e-03 |
Pos logit contributions | |
ches | +0.337 |
Tiny | +0.317 |
fl | +0.301 |
py | +0.299 |
Ho | +0.291 |
Neg logit contributions | |
handed | -0.558 |
sandbox | -0.449 |
Mike | -0.435 |
vendor | -0.434 |
amusement | -0.431 |
Token | vendor |
Token ID | 18371 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
asking | +0.008 |
fl | +0.008 |
moral | +0.008 |
ro | +0.008 |
carniv | +0.008 |
Neg logit contributions | |
handed | -0.005 |
ill | -0.004 |
unatt | -0.003 |
Mike | -0.003 |
view | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
handed | +0.914 |
Mike | +0.715 |
sandbox | +0.714 |
view | +0.704 |
amusement | +0.697 |
Neg logit contributions | |
ches | -0.645 |
Tiny | -0.597 |
py | -0.576 |
Ho | -0.568 |
fl | -0.565 |
Token | " |
Token ID | 366 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
handed | +0.023 |
vendor | +0.019 |
Mike | +0.019 |
amusement | +0.019 |
Jane | +0.019 |
Neg logit contributions | |
ches | -0.014 |
py | -0.013 |
Tiny | -0.012 |
fl | -0.012 |
agons | -0.011 |
Token | Exc |
Token ID | 40127 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
ches | +0.333 |
Tiny | +0.309 |
py | +0.303 |
fl | +0.301 |
Ho | +0.295 |
Neg logit contributions | |
handed | -0.317 |
sandbox | -0.233 |
vendor | -0.233 |
Mike | -0.232 |
amusement | -0.229 |
Token | use |
Token ID | 1904 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
handed | +0.025 |
sandbox | +0.021 |
vendor | +0.020 |
Mike | +0.020 |
lobby | +0.020 |
Neg logit contributions | |
ches | -0.013 |
Tiny | -0.011 |
fl | -0.011 |
py | -0.010 |
Ho | -0.010 |
Token | me |
Token ID | 502 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
ches | +0.093 |
Tiny | +0.079 |
py | +0.071 |
fl | +0.071 |
Ho | +0.068 |
Neg logit contributions | |
handed | -0.329 |
sandbox | -0.276 |
Mike | -0.273 |
lobby | -0.271 |
vendor | -0.270 |
Token | . |
Token ID | 13 |
Feature activation | +1.41e-03 |
Pos logit contributions | |
handed | +0.019 |
Mike | +0.017 |
vendor | +0.016 |
amusement | +0.016 |
sandbox | +0.016 |
Neg logit contributions | |
ches | -0.010 |
fl | -0.009 |
Tiny | -0.008 |
wand | -0.008 |
py | -0.007 |
Token | It |
Token ID | 632 |
Feature activation | -2.34e-03 |
Pos logit contributions | |
ches | +0.002 |
Tiny | +0.002 |
Ho | +0.002 |
py | +0.002 |
fl | +0.002 |
Neg logit contributions | |
handed | -0.003 |
sandbox | -0.003 |
amusement | -0.003 |
vendor | -0.003 |
Mike | -0.003 |
Token | wasn |
Token ID | 2492 |
Feature activation | +6.24e-02 |
Pos logit contributions | |
handed | +0.005 |
vendor | +0.004 |
Mike | +0.004 |
amusement | +0.004 |
sandbox | +0.004 |
Neg logit contributions | |
ches | -0.004 |
fl | -0.004 |
Tiny | -0.004 |
py | -0.003 |
Ho | -0.003 |
Token | 't |
Token ID | 470 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
ches | +0.073 |
py | +0.067 |
Tiny | +0.067 |
fl | +0.065 |
wand | +0.065 |
Neg logit contributions | |
handed | -0.149 |
Mike | -0.123 |
vendor | -0.120 |
amusement | -0.119 |
he | -0.118 |
Token | cold |
Token ID | 4692 |
Feature activation | -1.66e-02 |
Pos logit contributions | |
ches | +0.353 |
Tiny | +0.322 |
py | +0.315 |
fl | +0.310 |
Ho | +0.307 |
Neg logit contributions | |
handed | -0.393 |
vendor | -0.298 |
Mike | -0.297 |
sandbox | -0.295 |
amusement | -0.291 |
Token | anymore |
Token ID | 7471 |
Feature activation | -3.67e-01 |
Pos logit contributions | |
handed | +0.028 |
Mike | +0.021 |
amusement | +0.021 |
sandbox | +0.020 |
vendor | +0.020 |
Neg logit contributions | |
ches | -0.033 |
Tiny | -0.031 |
py | -0.030 |
fl | -0.030 |
Ho | -0.030 |
Token | - |
Token ID | 532 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
handed | +0.857 |
sandbox | +0.684 |
vendor | +0.683 |
amusement | +0.672 |
Mike | +0.671 |
Neg logit contributions | |
ches | -0.518 |
Tiny | -0.476 |
fl | -0.471 |
py | -0.447 |
Sparkle | -0.435 |
Token | it |
Token ID | 340 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
handed | +0.299 |
Mike | +0.231 |
sandbox | +0.228 |
vendor | +0.227 |
lobby | +0.227 |
Neg logit contributions | |
ches | -0.173 |
Tiny | -0.170 |
fl | -0.166 |
Ho | -0.160 |
Sparkle | -0.156 |
Token | was |
Token ID | 373 |
Feature activation | -7.26e-02 |
Pos logit contributions | |
ches | +0.045 |
Tiny | +0.041 |
fl | +0.041 |
py | +0.039 |
Ho | +0.038 |
Neg logit contributions | |
handed | -0.057 |
vendor | -0.045 |
Mike | -0.045 |
amusement | -0.045 |
sandbox | -0.044 |
Token | warm |
Token ID | 5814 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
handed | +0.142 |
amusement | +0.109 |
vendor | +0.109 |
Mike | +0.109 |
sandbox | +0.108 |
Neg logit contributions | |
ches | -0.122 |
Tiny | -0.110 |
py | -0.108 |
Ho | -0.104 |
fl | -0.103 |
Token | and |
Token ID | 290 |
Feature activation | +1.52e-02 |
Pos logit contributions | |
ches | +0.032 |
fl | +0.030 |
Tiny | +0.029 |
wool | +0.028 |
Pig | +0.028 |
Neg logit contributions | |
handed | -0.074 |
Mike | -0.062 |
amusement | -0.061 |
vendor | -0.060 |
apartment | -0.060 |
Token | so |
Token ID | 523 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
ches | +0.337 |
Tiny | +0.306 |
py | +0.302 |
fl | +0.296 |
Ho | +0.291 |
Neg logit contributions | |
handed | -0.354 |
vendor | -0.269 |
Mike | -0.266 |
sandbox | -0.265 |
amusement | -0.261 |
Token | fearful |
Token ID | 26844 |
Feature activation | -4.59e-02 |
Pos logit contributions | |
handed | +0.278 |
sandbox | +0.232 |
amusement | +0.230 |
Mike | +0.228 |
view | +0.228 |
Neg logit contributions | |
ches | -0.138 |
Tiny | -0.135 |
fl | -0.128 |
Ho | -0.125 |
py | -0.124 |
Token | of |
Token ID | 286 |
Feature activation | -2.93e-02 |
Pos logit contributions | |
handed | +0.081 |
sandbox | +0.060 |
Mike | +0.059 |
amusement | +0.059 |
lobby | +0.059 |
Neg logit contributions | |
ches | -0.091 |
Tiny | -0.083 |
py | -0.083 |
fl | -0.081 |
Ho | -0.079 |
Token | the |
Token ID | 262 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
handed | +0.070 |
vendor | +0.054 |
Mike | +0.054 |
view | +0.054 |
lobby | +0.053 |
Neg logit contributions | |
ches | -0.038 |
fl | -0.038 |
Tiny | -0.037 |
py | -0.035 |
Ho | -0.034 |
Token | slide |
Token ID | 10649 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
handed | +0.037 |
ill | +0.028 |
aney | +0.027 |
keeping | +0.026 |
view | +0.026 |
Neg logit contributions | |
fl | -0.029 |
tw | -0.024 |
ro | -0.024 |
hopping | -0.023 |
wand | -0.023 |
Token | anymore |
Token ID | 7471 |
Feature activation | -3.90e-01 |
Pos logit contributions | |
handed | +0.549 |
amaz | +0.440 |
Mike | +0.431 |
view | +0.403 |
sandbox | +0.403 |
Neg logit contributions | |
ches | -0.558 |
fl | -0.495 |
Tiny | -0.476 |
ch | -0.453 |
Sparkle | -0.449 |
Token | . |
Token ID | 13 |
Feature activation | -2.39e-02 |
Pos logit contributions | |
handed | +0.822 |
vendor | +0.637 |
Mike | +0.636 |
sandbox | +0.634 |
amusement | +0.626 |
Neg logit contributions | |
ches | -0.625 |
Tiny | -0.575 |
py | -0.559 |
fl | -0.554 |
Ho | -0.541 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +3.35e-02 |
Pos logit contributions | |
handed | +0.050 |
vendor | +0.039 |
Mike | +0.039 |
sandbox | +0.039 |
amusement | +0.038 |
Neg logit contributions | |
ches | -0.038 |
Tiny | -0.035 |
py | -0.034 |
fl | -0.034 |
Ho | -0.033 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
ches | +0.046 |
Tiny | +0.042 |
py | +0.040 |
fl | +0.039 |
Ho | +0.039 |
Neg logit contributions | |
handed | -0.077 |
sandbox | -0.062 |
vendor | -0.062 |
amusement | -0.061 |
Mike | -0.060 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
ches | +0.027 |
Tiny | +0.025 |
py | +0.025 |
fl | +0.024 |
Ho | +0.024 |
Neg logit contributions | |
handed | -0.034 |
vendor | -0.027 |
Mike | -0.026 |
sandbox | -0.026 |
amusement | -0.026 |
Token | a |
Token ID | 257 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
ches | +0.021 |
Tiny | +0.020 |
fl | +0.020 |
py | +0.019 |
Ho | +0.019 |
Neg logit contributions | |
handed | -0.038 |
vendor | -0.031 |
Mike | -0.030 |
sandbox | -0.030 |
amusement | -0.030 |
Token | It |
Token ID | 632 |
Feature activation | -4.59e-03 |
Pos logit contributions | |
ches | +0.001 |
Tiny | +0.001 |
Ho | +0.001 |
Cotton | +0.001 |
Sparkle | +0.001 |
Neg logit contributions | |
handed | -0.002 |
sandbox | -0.002 |
amusement | -0.002 |
vendor | -0.002 |
Mike | -0.002 |
Token | still |
Token ID | 991 |
Feature activation | -4.79e-02 |
Pos logit contributions | |
handed | +0.009 |
vendor | +0.007 |
Mike | +0.007 |
sandbox | +0.007 |
amusement | +0.007 |
Neg logit contributions | |
ches | -0.008 |
Tiny | -0.007 |
fl | -0.007 |
py | -0.007 |
Ho | -0.007 |
Token | wasn |
Token ID | 2492 |
Feature activation | +5.79e-02 |
Pos logit contributions | |
handed | +0.119 |
Mike | +0.094 |
vendor | +0.093 |
amusement | +0.091 |
sandbox | +0.090 |
Neg logit contributions | |
ches | -0.063 |
fl | -0.060 |
Tiny | -0.057 |
py | -0.055 |
wand | -0.054 |
Token | 't |
Token ID | 470 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
ches | +0.068 |
py | +0.063 |
Tiny | +0.063 |
Cotton | +0.059 |
fl | +0.059 |
Neg logit contributions | |
handed | -0.138 |
Mike | -0.114 |
vendor | -0.110 |
amusement | -0.110 |
Jane | -0.108 |
Token | done |
Token ID | 1760 |
Feature activation | -6.38e-02 |
Pos logit contributions | |
ches | +0.315 |
Tiny | +0.293 |
fl | +0.287 |
py | +0.285 |
Ho | +0.272 |
Neg logit contributions | |
handed | -0.435 |
Mike | -0.337 |
vendor | -0.335 |
sandbox | -0.333 |
amusement | -0.333 |
Token | yet |
Token ID | 1865 |
Feature activation | -3.51e-01 |
Pos logit contributions | |
handed | +0.141 |
sandbox | +0.120 |
vendor | +0.115 |
Mike | +0.111 |
lobby | +0.111 |
Neg logit contributions | |
ches | -0.094 |
fl | -0.084 |
Tiny | -0.082 |
wand | -0.080 |
py | -0.079 |
Token | . |
Token ID | 13 |
Feature activation | -2.61e-04 |
Pos logit contributions | |
handed | +0.838 |
Mike | +0.693 |
vendor | +0.686 |
sandbox | +0.685 |
lobby | +0.673 |
Neg logit contributions | |
ches | -0.458 |
fl | -0.430 |
Tiny | -0.399 |
hopping | -0.397 |
wand | -0.385 |
Token |
|
Token ID | 198 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
handed | +0.001 |
sandbox | +0.001 |
amusement | +0.001 |
vendor | +0.000 |
view | +0.000 |
Neg logit contributions | |
ches | -0.000 |
Tiny | -0.000 |
Ho | -0.000 |
Sparkle | -0.000 |
Cotton | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
ches | +0.031 |
Tiny | +0.028 |
py | +0.026 |
fl | +0.026 |
Ho | +0.026 |
Neg logit contributions | |
handed | -0.049 |
sandbox | -0.040 |
vendor | -0.040 |
amusement | -0.039 |
lobby | -0.039 |
Token | The |
Token ID | 464 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
ches | +0.035 |
Tiny | +0.032 |
fl | +0.031 |
py | +0.030 |
Cotton | +0.030 |
Neg logit contributions | |
handed | -0.042 |
lobby | -0.034 |
sandbox | -0.033 |
vendor | -0.033 |
amusement | -0.033 |
Token | little |
Token ID | 1310 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
handed | +0.296 |
Mike | +0.245 |
view | +0.236 |
amusement | +0.235 |
sandbox | +0.234 |
Neg logit contributions | |
ches | -0.121 |
fl | -0.110 |
Tiny | -0.109 |
py | -0.104 |
wand | -0.104 |
Token | the |
Token ID | 262 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
handed | +0.065 |
Mike | +0.051 |
vendor | +0.051 |
sandbox | +0.050 |
view | +0.050 |
Neg logit contributions | |
ches | -0.042 |
Tiny | -0.041 |
fl | -0.038 |
Ho | -0.038 |
Cotton | -0.037 |
Token | park |
Token ID | 3952 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
handed | +0.023 |
Mike | +0.018 |
ill | +0.017 |
Jane | +0.017 |
aren | +0.016 |
Neg logit contributions | |
Tiny | -0.026 |
ches | -0.026 |
fl | -0.025 |
Ho | -0.025 |
Cotton | -0.024 |
Token | and |
Token ID | 290 |
Feature activation | -1.11e-03 |
Pos logit contributions | |
handed | +0.602 |
amaz | +0.513 |
Mike | +0.511 |
unatt | +0.490 |
amusement | +0.485 |
Neg logit contributions | |
fl | -0.198 |
hopping | -0.180 |
ches | -0.178 |
Tiny | -0.172 |
asking | -0.156 |
Token | saw |
Token ID | 2497 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
handed | +0.002 |
view | +0.002 |
Mike | +0.002 |
amaz | +0.002 |
amusement | +0.002 |
Neg logit contributions | |
ches | -0.002 |
Tiny | -0.002 |
Ho | -0.002 |
fl | -0.002 |
Sparkle | -0.002 |
Token | a |
Token ID | 257 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
handed | +0.321 |
har | +0.267 |
apartment | +0.266 |
Mike | +0.265 |
unatt | +0.262 |
Neg logit contributions | |
Tiny | -0.128 |
Ho | -0.114 |
fl | -0.113 |
wool | -0.107 |
ches | -0.102 |
Token | vendor |
Token ID | 18371 |
Feature activation | -4.44e-01 |
Pos logit contributions | |
handed | +0.039 |
view | +0.030 |
Mike | +0.030 |
ill | +0.029 |
amusement | +0.029 |
Neg logit contributions | |
ches | -0.021 |
fl | -0.021 |
Tiny | -0.020 |
Ho | -0.019 |
wand | -0.018 |
Token | selling |
Token ID | 6301 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
handed | +0.936 |
Mike | +0.749 |
sandbox | +0.735 |
amusement | +0.725 |
amaz | +0.724 |
Neg logit contributions | |
ches | -0.644 |
fl | -0.607 |
Tiny | -0.574 |
hopping | -0.563 |
Ho | -0.561 |
Token | ice |
Token ID | 4771 |
Feature activation | -2.47e-01 |
Pos logit contributions | |
handed | +0.504 |
unatt | +0.405 |
ill | +0.404 |
har | +0.391 |
Jane | +0.386 |
Neg logit contributions | |
fl | -0.307 |
wool | -0.297 |
carrots | -0.288 |
potatoes | -0.281 |
Ho | -0.277 |
Token | cream |
Token ID | 8566 |
Feature activation | -2.46e-01 |
Pos logit contributions | |
handed | +0.394 |
amaz | +0.294 |
amusement | +0.286 |
sandbox | +0.266 |
Mike | +0.261 |
Neg logit contributions | |
ches | -0.514 |
fl | -0.481 |
Ho | -0.460 |
py | -0.456 |
Tiny | -0.448 |
Token | . |
Token ID | 13 |
Feature activation | -2.05e-02 |
Pos logit contributions | |
handed | +0.575 |
sandbox | +0.455 |
Mike | +0.453 |
amusement | +0.452 |
amaz | +0.451 |
Neg logit contributions | |
ches | -0.327 |
fl | -0.292 |
Tiny | -0.281 |
Ho | -0.280 |
Cotton | -0.266 |
Token | The |
Token ID | 383 |
Feature activation | -6.46e-02 |
Pos logit contributions | |
handed | +0.041 |
vendor | +0.033 |
sandbox | +0.032 |
amusement | +0.032 |
Mike | +0.032 |
Neg logit contributions | |
ches | -0.033 |
py | -0.032 |
fl | -0.030 |
wand | -0.030 |
Ho | -0.029 |