Token | She |
Token ID | 1375 |
Feature activation | +6.12e-02 |
Pos logit contributions | |
dd | +0.064 |
night | +0.052 |
ez | +0.046 |
pity | +0.044 |
amb | +0.044 |
Neg logit contributions | |
groups | -0.071 |
bars | -0.070 |
towns | -0.063 |
younger | -0.062 |
fighters | -0.062 |
Token | f |
Token ID | 277 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
dd | +0.106 |
night | +0.087 |
ez | +0.079 |
pity | +0.075 |
amb | +0.074 |
Neg logit contributions | |
groups | -0.115 |
bars | -0.112 |
towns | -0.100 |
fighters | -0.100 |
younger | -0.100 |
Token | iddled |
Token ID | 34897 |
Feature activation | +7.35e-02 |
Pos logit contributions | |
groups | +0.358 |
bars | +0.355 |
fighters | +0.313 |
towns | +0.312 |
eight | +0.310 |
Neg logit contributions | |
dd | -0.307 |
night | -0.255 |
ez | -0.224 |
pity | -0.218 |
uggle | -0.210 |
Token | and |
Token ID | 290 |
Feature activation | +8.20e-02 |
Pos logit contributions | |
dd | +0.124 |
night | +0.103 |
ez | +0.090 |
amb | +0.088 |
pity | +0.087 |
Neg logit contributions | |
groups | -0.142 |
bars | -0.137 |
towns | -0.124 |
eight | -0.124 |
younger | -0.122 |
Token | pro |
Token ID | 386 |
Feature activation | +6.10e-02 |
Pos logit contributions | |
dd | +0.147 |
night | +0.125 |
ez | +0.112 |
pity | +0.108 |
faced | +0.104 |
Neg logit contributions | |
groups | -0.146 |
bars | -0.144 |
fighters | -0.131 |
department | -0.128 |
ents | -0.128 |
Token | dd |
Token ID | 1860 |
Feature activation | +5.17e-01 |
Pos logit contributions | |
dd | +0.086 |
night | +0.075 |
pity | +0.063 |
ez | +0.062 |
uggle | +0.061 |
Neg logit contributions | |
bars | -0.127 |
groups | -0.127 |
eight | -0.114 |
ents | -0.113 |
towns | -0.112 |
Token | ed |
Token ID | 276 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
dd | +0.816 |
night | +0.665 |
ez | +0.577 |
- | +0.546 |
amb | +0.500 |
Neg logit contributions | |
groups | -1.107 |
bars | -1.051 |
younger | -0.970 |
ents | -0.954 |
older | -0.952 |
Token | for |
Token ID | 329 |
Feature activation | +2.78e-02 |
Pos logit contributions | |
dd | +0.379 |
night | +0.316 |
ez | +0.284 |
pity | +0.274 |
amb | +0.259 |
Neg logit contributions | |
groups | -0.425 |
bars | -0.420 |
fighters | -0.374 |
towns | -0.369 |
younger | -0.366 |
Token | a |
Token ID | 257 |
Feature activation | +7.52e-02 |
Pos logit contributions | |
dd | +0.044 |
night | +0.034 |
ez | +0.031 |
pity | +0.031 |
it | +0.029 |
Neg logit contributions | |
bars | -0.057 |
groups | -0.055 |
fighters | -0.051 |
department | -0.050 |
lights | -0.050 |
Token | few |
Token ID | 1178 |
Feature activation | -9.30e-02 |
Pos logit contributions | |
dd | +0.142 |
night | +0.122 |
ez | +0.108 |
pity | +0.106 |
faced | +0.103 |
Neg logit contributions | |
groups | -0.128 |
bars | -0.127 |
towns | -0.112 |
fighters | -0.111 |
ents | -0.109 |
Token | minutes |
Token ID | 2431 |
Feature activation | -9.45e-02 |
Pos logit contributions | |
bars | +0.182 |
groups | +0.167 |
fighters | +0.161 |
ains | +0.160 |
ents | +0.159 |
Neg logit contributions | |
dd | -0.155 |
night | -0.131 |
ez | -0.121 |
- | -0.120 |
pity | -0.118 |
Token | . |
Token ID | 13 |
Feature activation | -3.36e-02 |
Pos logit contributions | |
dd | +0.198 |
night | +0.165 |
ez | +0.150 |
amb | +0.138 |
- | +0.138 |
Neg logit contributions | |
groups | -0.221 |
bars | -0.217 |
towns | -0.193 |
lights | -0.193 |
younger | -0.191 |
Token | She |
Token ID | 1375 |
Feature activation | -4.77e-03 |
Pos logit contributions | |
groups | +0.067 |
bars | +0.063 |
lights | +0.060 |
fighters | +0.060 |
bridges | +0.058 |
Neg logit contributions | |
dd | -0.052 |
night | -0.044 |
amb | -0.039 |
ez | -0.038 |
Now | -0.038 |
Token | poked |
Token ID | 47320 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
groups | +0.010 |
bars | +0.009 |
towns | +0.008 |
fighters | +0.008 |
lights | +0.008 |
Neg logit contributions | |
dd | -0.008 |
night | -0.006 |
ez | -0.006 |
pity | -0.005 |
amb | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | +5.78e-02 |
Pos logit contributions | |
dd | +0.207 |
night | +0.175 |
ez | +0.156 |
uggle | +0.152 |
pity | +0.151 |
Neg logit contributions | |
groups | -0.180 |
bars | -0.179 |
eight | -0.157 |
towns | -0.157 |
younger | -0.154 |
Token | pro |
Token ID | 386 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
dd | +0.097 |
night | +0.078 |
ez | +0.072 |
pity | +0.068 |
faced | +0.065 |
Neg logit contributions | |
groups | -0.114 |
bars | -0.110 |
lights | -0.100 |
fighters | -0.099 |
towns | -0.099 |
Token | dd |
Token ID | 1860 |
Feature activation | +4.84e-01 |
Pos logit contributions | |
dd | +0.015 |
night | +0.014 |
pity | +0.009 |
faced | +0.009 |
uggle | +0.009 |
Neg logit contributions | |
bars | -0.072 |
groups | -0.070 |
ents | -0.066 |
eight | -0.065 |
towns | -0.065 |
Token | ed |
Token ID | 276 |
Feature activation | +1.91e-01 |
Pos logit contributions | |
dd | +0.497 |
night | +0.346 |
ez | +0.274 |
- | +0.234 |
oo | +0.207 |
Neg logit contributions | |
groups | -1.302 |
bars | -1.261 |
younger | -1.179 |
older | -1.156 |
lights | -1.156 |
Token | the |
Token ID | 262 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
dd | +0.330 |
night | +0.275 |
ez | +0.249 |
pity | +0.239 |
amb | +0.229 |
Neg logit contributions | |
groups | -0.360 |
bars | -0.353 |
fighters | -0.315 |
towns | -0.311 |
younger | -0.309 |
Token | thick |
Token ID | 6546 |
Feature activation | -7.76e-02 |
Pos logit contributions | |
groups | +0.273 |
bars | +0.269 |
towns | +0.244 |
eight | +0.242 |
younger | +0.241 |
Neg logit contributions | |
dd | -0.233 |
night | -0.188 |
ez | -0.169 |
amb | -0.160 |
pity | -0.159 |
Token | skeleton |
Token ID | 18328 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
groups | +0.149 |
bars | +0.146 |
towns | +0.132 |
eight | +0.131 |
younger | +0.129 |
Neg logit contributions | |
dd | -0.133 |
night | -0.107 |
ez | -0.096 |
amb | -0.093 |
pity | -0.092 |
Token | with |
Token ID | 351 |
Feature activation | +8.54e-02 |
Pos logit contributions | |
dd | +0.152 |
night | +0.125 |
- | +0.117 |
ez | +0.117 |
ggy | +0.111 |
Neg logit contributions | |
groups | -0.211 |
bars | -0.198 |
towns | -0.184 |
lights | -0.183 |
younger | -0.182 |
Token | . |
Token ID | 13 |
Feature activation | -3.98e-02 |
Pos logit contributions | |
dd | +0.044 |
night | +0.034 |
- | +0.032 |
it | +0.029 |
inside | +0.028 |
Neg logit contributions | |
groups | -0.069 |
bars | -0.064 |
towns | -0.063 |
eight | -0.061 |
younger | -0.060 |
Token | |
Token ID | 220 |
Feature activation | +6.50e-03 |
Pos logit contributions | |
groups | +0.080 |
bars | +0.079 |
fighters | +0.071 |
towns | +0.071 |
eight | +0.070 |
Neg logit contributions | |
dd | -0.063 |
night | -0.051 |
ez | -0.045 |
pity | -0.043 |
amb | -0.043 |
Token |
|
Token ID | 198 |
Feature activation | +4.16e-02 |
Pos logit contributions | |
dd | +0.007 |
night | +0.005 |
ez | +0.004 |
uggle | +0.004 |
oo | +0.004 |
Neg logit contributions | |
groups | -0.016 |
bars | -0.015 |
towns | -0.014 |
eight | -0.014 |
fighters | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | +4.06e-02 |
Pos logit contributions | |
dd | +0.059 |
night | +0.043 |
ez | +0.038 |
amb | +0.037 |
oo | +0.036 |
Neg logit contributions | |
groups | -0.093 |
bars | -0.092 |
fighters | -0.084 |
younger | -0.083 |
department | -0.083 |
Token | E |
Token ID | 36 |
Feature activation | +9.83e-02 |
Pos logit contributions | |
dd | +0.080 |
night | +0.062 |
ez | +0.058 |
amb | +0.055 |
oo | +0.055 |
Neg logit contributions | |
bars | -0.071 |
groups | -0.071 |
younger | -0.062 |
fighters | -0.062 |
department | -0.062 |
Token | dd |
Token ID | 1860 |
Feature activation | +4.83e-01 |
Pos logit contributions | |
dd | +0.037 |
night | +0.015 |
ez | +0.003 |
pity | +0.000 |
faced | -0.005 |
Neg logit contributions | |
groups | -0.304 |
bars | -0.303 |
fighters | -0.285 |
towns | -0.284 |
eight | -0.280 |
Token | ie |
Token ID | 494 |
Feature activation | -3.57e-03 |
Pos logit contributions | |
dd | +0.693 |
night | +0.549 |
ez | +0.452 |
- | +0.430 |
oo | +0.416 |
Neg logit contributions | |
groups | -1.074 |
bars | -1.070 |
fastest | -0.985 |
younger | -0.969 |
thirteen | -0.954 |
Token | was |
Token ID | 373 |
Feature activation | +2.60e-02 |
Pos logit contributions | |
groups | +0.007 |
bars | +0.007 |
lights | +0.006 |
fighters | +0.006 |
towns | +0.006 |
Neg logit contributions | |
dd | -0.006 |
night | -0.005 |
ez | -0.005 |
pity | -0.004 |
faced | -0.004 |
Token | about |
Token ID | 546 |
Feature activation | +8.77e-02 |
Pos logit contributions | |
dd | +0.044 |
night | +0.036 |
ez | +0.034 |
pity | +0.032 |
faced | +0.030 |
Neg logit contributions | |
bars | -0.050 |
groups | -0.049 |
lights | -0.044 |
fighters | -0.044 |
towns | -0.043 |
Token | to |
Token ID | 284 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
dd | +0.160 |
night | +0.134 |
ez | +0.121 |
pity | +0.116 |
amb | +0.114 |
Neg logit contributions | |
groups | -0.157 |
bars | -0.155 |
towns | -0.136 |
fighters | -0.136 |
younger | -0.135 |
Token | give |
Token ID | 1577 |
Feature activation | -5.43e-03 |
Pos logit contributions | |
groups | +0.049 |
bars | +0.049 |
fighters | +0.043 |
towns | +0.043 |
younger | +0.042 |
Neg logit contributions | |
dd | -0.050 |
night | -0.042 |
ez | -0.038 |
pity | -0.036 |
amb | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -3.21e-02 |
Pos logit contributions | |
dd | +0.011 |
night | +0.009 |
ez | +0.008 |
pity | +0.008 |
faced | +0.008 |
Neg logit contributions | |
bars | -0.011 |
groups | -0.011 |
fighters | -0.009 |
towns | -0.009 |
lights | -0.009 |
Token | It |
Token ID | 632 |
Feature activation | +2.76e-01 |
Pos logit contributions | |
bars | +0.065 |
groups | +0.065 |
lights | +0.058 |
fighters | +0.057 |
fastest | +0.056 |
Neg logit contributions | |
dd | -0.050 |
night | -0.042 |
ez | -0.037 |
pity | -0.035 |
Now | -0.035 |
Token | poked |
Token ID | 47320 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
dd | +0.430 |
night | +0.334 |
ez | +0.306 |
pity | +0.295 |
faced | +0.288 |
Neg logit contributions | |
bars | -0.593 |
groups | -0.577 |
lights | -0.516 |
towns | -0.509 |
bridges | -0.504 |
Token | and |
Token ID | 290 |
Feature activation | +5.70e-02 |
Pos logit contributions | |
dd | +0.182 |
night | +0.151 |
ez | +0.135 |
amb | +0.129 |
pity | +0.129 |
Neg logit contributions | |
groups | -0.191 |
bars | -0.183 |
eight | -0.167 |
towns | -0.167 |
fighters | -0.164 |
Token | pro |
Token ID | 386 |
Feature activation | +2.14e-02 |
Pos logit contributions | |
dd | +0.098 |
night | +0.079 |
ez | +0.072 |
pity | +0.069 |
faced | +0.067 |
Neg logit contributions | |
bars | -0.112 |
groups | -0.111 |
lights | -0.097 |
towns | -0.096 |
younger | -0.095 |
Token | dd |
Token ID | 1860 |
Feature activation | +4.81e-01 |
Pos logit contributions | |
night | +0.008 |
dd | +0.006 |
uggle | +0.004 |
coincidence | +0.004 |
pity | +0.004 |
Neg logit contributions | |
groups | -0.062 |
bars | -0.060 |
eight | -0.059 |
ents | -0.059 |
towns | -0.057 |
Token | ed |
Token ID | 276 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
dd | +0.514 |
night | +0.352 |
ez | +0.282 |
- | +0.246 |
oo | +0.217 |
Neg logit contributions | |
groups | -1.307 |
bars | -1.277 |
younger | -1.181 |
older | -1.162 |
lights | -1.149 |
Token | at |
Token ID | 379 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
dd | +0.307 |
night | +0.250 |
ez | +0.224 |
pity | +0.216 |
amb | +0.204 |
Neg logit contributions | |
groups | -0.382 |
bars | -0.381 |
fighters | -0.337 |
younger | -0.334 |
towns | -0.334 |
Token | the |
Token ID | 262 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
dd | +0.196 |
night | +0.160 |
ez | +0.149 |
pity | +0.146 |
oo | +0.140 |
Neg logit contributions | |
bars | -0.200 |
groups | -0.193 |
lights | -0.170 |
fighters | -0.169 |
department | -0.167 |
Token | trap |
Token ID | 12840 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
groups | +0.295 |
bars | +0.284 |
towns | +0.259 |
fighters | +0.259 |
eight | +0.258 |
Neg logit contributions | |
dd | -0.234 |
night | -0.190 |
ez | -0.170 |
amb | -0.160 |
pity | -0.159 |
Token | until |
Token ID | 1566 |
Feature activation | -8.10e-02 |
Pos logit contributions | |
dd | +0.230 |
night | +0.192 |
ez | +0.169 |
it | +0.167 |
- | +0.166 |
Neg logit contributions | |
bars | -0.326 |
groups | -0.319 |
towns | -0.290 |
fighters | -0.285 |
lights | -0.284 |
Token | sister |
Token ID | 6621 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
dd | +0.037 |
night | +0.031 |
ez | +0.028 |
pity | +0.027 |
amb | +0.027 |
Neg logit contributions | |
groups | -0.032 |
bars | -0.031 |
towns | -0.027 |
fighters | -0.027 |
younger | -0.027 |
Token | !" |
Token ID | 2474 |
Feature activation | +5.06e-02 |
Pos logit contributions | |
bars | +0.324 |
groups | +0.310 |
lights | +0.286 |
narrower | +0.283 |
towns | +0.280 |
Neg logit contributions | |
dd | -0.248 |
night | -0.224 |
ez | -0.197 |
- | -0.187 |
amb | -0.181 |
Token |
|
Token ID | 198 |
Feature activation | +3.26e-02 |
Pos logit contributions | |
dd | +0.093 |
night | +0.077 |
ez | +0.072 |
pity | +0.069 |
faced | +0.067 |
Neg logit contributions | |
bars | -0.088 |
groups | -0.088 |
eight | -0.076 |
towns | -0.076 |
younger | -0.076 |
Token |
|
Token ID | 198 |
Feature activation | +3.20e-02 |
Pos logit contributions | |
dd | +0.061 |
night | +0.055 |
ez | +0.049 |
pity | +0.048 |
faced | +0.047 |
Neg logit contributions | |
groups | -0.053 |
bars | -0.051 |
eight | -0.046 |
towns | -0.045 |
younger | -0.045 |
Token | E |
Token ID | 36 |
Feature activation | +9.00e-02 |
Pos logit contributions | |
dd | +0.060 |
night | +0.050 |
ez | +0.045 |
pity | +0.043 |
amb | +0.043 |
Neg logit contributions | |
groups | -0.056 |
bars | -0.055 |
towns | -0.049 |
fighters | -0.049 |
younger | -0.048 |
Token | dd |
Token ID | 1860 |
Feature activation | +4.74e-01 |
Pos logit contributions | |
dd | +0.075 |
night | +0.055 |
ez | +0.046 |
pity | +0.044 |
faced | +0.039 |
Neg logit contributions | |
groups | -0.232 |
bars | -0.231 |
fighters | -0.217 |
towns | -0.214 |
eight | -0.213 |
Token | ie |
Token ID | 494 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
dd | +0.717 |
night | +0.604 |
ez | +0.501 |
oo | +0.479 |
- | +0.478 |
Neg logit contributions | |
groups | -1.015 |
bars | -1.003 |
fastest | -0.935 |
towns | -0.912 |
bridges | -0.909 |
Token | went |
Token ID | 1816 |
Feature activation | +3.60e-02 |
Pos logit contributions | |
groups | +0.030 |
bars | +0.029 |
towns | +0.026 |
fighters | +0.026 |
lights | +0.026 |
Neg logit contributions | |
dd | -0.027 |
night | -0.022 |
ez | -0.020 |
pity | -0.019 |
amb | -0.019 |
Token | immediately |
Token ID | 3393 |
Feature activation | +3.44e-02 |
Pos logit contributions | |
dd | +0.068 |
night | +0.055 |
ez | +0.052 |
amb | +0.048 |
pity | +0.048 |
Neg logit contributions | |
groups | -0.064 |
bars | -0.062 |
older | -0.057 |
younger | -0.056 |
towns | -0.056 |
Token | to |
Token ID | 284 |
Feature activation | -3.54e-02 |
Pos logit contributions | |
dd | +0.049 |
night | +0.041 |
ez | +0.035 |
pity | +0.034 |
- | +0.033 |
Neg logit contributions | |
bars | -0.073 |
groups | -0.072 |
fighters | -0.065 |
lights | -0.065 |
ains | -0.064 |
Token | his |
Token ID | 465 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
groups | +0.059 |
bars | +0.055 |
younger | +0.051 |
older | +0.051 |
towns | +0.050 |
Neg logit contributions | |
dd | -0.073 |
night | -0.060 |
ez | -0.056 |
amb | -0.054 |
faced | -0.052 |
Token | fire |
Token ID | 2046 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
groups | +0.254 |
bars | +0.243 |
towns | +0.219 |
fighters | +0.218 |
younger | +0.216 |
Neg logit contributions | |
dd | -0.296 |
night | -0.248 |
ez | -0.231 |
amb | -0.221 |
pity | -0.219 |
Token | work |
Token ID | 1818 |
Feature activation | -1.58e-01 |
Pos logit contributions | |
groups | +0.035 |
bars | +0.034 |
ents | +0.030 |
eight | +0.030 |
narrower | +0.029 |
Neg logit contributions | |
dd | -0.033 |
night | -0.031 |
- | -0.028 |
ez | -0.026 |
ake | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -5.08e-02 |
Pos logit contributions | |
bars | +0.308 |
groups | +0.303 |
towns | +0.270 |
eight | +0.269 |
fighters | +0.269 |
Neg logit contributions | |
dd | -0.246 |
night | -0.224 |
ez | -0.187 |
- | -0.182 |
uggle | -0.180 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
groups | +0.087 |
bars | +0.087 |
eight | +0.078 |
towns | +0.078 |
younger | +0.076 |
Neg logit contributions | |
dd | -0.096 |
night | -0.076 |
ez | -0.072 |
pity | -0.070 |
amb | -0.067 |
Token | E |
Token ID | 36 |
Feature activation | +8.55e-02 |
Pos logit contributions | |
dd | +0.356 |
night | +0.300 |
pity | +0.266 |
ez | +0.266 |
amb | +0.263 |
Neg logit contributions | |
groups | -0.299 |
bars | -0.278 |
towns | -0.261 |
eight | -0.252 |
ill | -0.248 |
Token | dd |
Token ID | 1860 |
Feature activation | +4.68e-01 |
Pos logit contributions | |
dd | +0.107 |
night | +0.088 |
pity | +0.083 |
ez | +0.082 |
faced | +0.074 |
Neg logit contributions | |
bars | -0.178 |
groups | -0.178 |
fighters | -0.166 |
eight | -0.166 |
towns | -0.161 |
Token | ie |
Token ID | 494 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
dd | +0.655 |
night | +0.559 |
ez | +0.444 |
oo | +0.435 |
- | +0.431 |
Neg logit contributions | |
bars | -1.026 |
groups | -1.013 |
fastest | -0.948 |
younger | -0.927 |
thirteen | -0.910 |
Token | was |
Token ID | 373 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
groups | +0.028 |
bars | +0.027 |
eight | +0.025 |
towns | +0.025 |
older | +0.024 |
Neg logit contributions | |
dd | -0.030 |
night | -0.024 |
pity | -0.022 |
ez | -0.022 |
faced | -0.022 |
Token | a |
Token ID | 257 |
Feature activation | +3.49e-02 |
Pos logit contributions | |
dd | +0.029 |
night | +0.025 |
ez | +0.023 |
amb | +0.022 |
uggle | +0.022 |
Neg logit contributions | |
groups | -0.027 |
older | -0.025 |
younger | -0.025 |
bars | -0.024 |
eight | -0.024 |
Token | three |
Token ID | 1115 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
dd | +0.073 |
night | +0.062 |
ez | +0.060 |
amb | +0.056 |
pity | +0.055 |
Neg logit contributions | |
groups | -0.053 |
bars | -0.050 |
younger | -0.046 |
older | -0.046 |
eight | -0.046 |
Token | year |
Token ID | 614 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
groups | +0.427 |
bars | +0.422 |
towns | +0.378 |
fighters | +0.378 |
younger | +0.376 |
Neg logit contributions | |
dd | -0.306 |
night | -0.246 |
ez | -0.217 |
pity | -0.204 |
amb | -0.200 |
Token | um |
Token ID | 388 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
groups | +0.312 |
bars | +0.283 |
towns | +0.267 |
younger | +0.261 |
older | +0.260 |
Neg logit contributions | |
dd | -0.343 |
night | -0.287 |
amb | -0.268 |
ez | -0.259 |
oo | -0.251 |
Token | ma |
Token ID | 2611 |
Feature activation | -7.52e-02 |
Pos logit contributions | |
dd | +0.202 |
night | +0.167 |
ez | +0.142 |
- | +0.132 |
amb | +0.129 |
Neg logit contributions | |
groups | -0.313 |
bars | -0.305 |
towns | -0.280 |
fastest | -0.278 |
fighters | -0.276 |
Token | kissed |
Token ID | 27946 |
Feature activation | +8.85e-02 |
Pos logit contributions | |
groups | +0.153 |
bars | +0.148 |
towns | +0.135 |
lights | +0.134 |
fastest | +0.133 |
Neg logit contributions | |
dd | -0.117 |
night | -0.099 |
ez | -0.089 |
- | -0.081 |
pity | -0.080 |
Token | him |
Token ID | 683 |
Feature activation | +2.39e-01 |
Pos logit contributions | |
dd | +0.133 |
night | +0.109 |
ez | +0.094 |
pity | +0.089 |
amb | +0.086 |
Neg logit contributions | |
groups | -0.187 |
bars | -0.184 |
towns | -0.166 |
fighters | -0.165 |
younger | -0.164 |
Token | good |
Token ID | 922 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
dd | +0.399 |
night | +0.334 |
ez | +0.298 |
pity | +0.282 |
amb | +0.275 |
Neg logit contributions | |
groups | -0.461 |
bars | -0.453 |
towns | -0.405 |
fighters | -0.404 |
eight | -0.399 |
Token | night |
Token ID | 3847 |
Feature activation | +4.28e-01 |
Pos logit contributions | |
dd | +0.240 |
night | +0.187 |
ez | +0.177 |
amb | +0.168 |
pity | +0.167 |
Neg logit contributions | |
groups | -0.256 |
bars | -0.253 |
fighters | -0.227 |
younger | -0.225 |
towns | -0.225 |
Token | and |
Token ID | 290 |
Feature activation | +9.22e-02 |
Pos logit contributions | |
dd | +0.722 |
night | +0.613 |
ez | +0.538 |
pity | +0.505 |
amb | +0.502 |
Neg logit contributions | |
groups | -0.824 |
bars | -0.812 |
towns | -0.723 |
fighters | -0.717 |
younger | -0.709 |
Token | urged |
Token ID | 11643 |
Feature activation | -2.50e-03 |
Pos logit contributions | |
dd | +0.162 |
night | +0.133 |
ez | +0.121 |
pity | +0.115 |
amb | +0.115 |
Neg logit contributions | |
groups | -0.170 |
bars | -0.168 |
towns | -0.149 |
fighters | -0.149 |
younger | -0.148 |
Token | him |
Token ID | 683 |
Feature activation | +2.26e-01 |
Pos logit contributions | |
groups | +0.005 |
bars | +0.005 |
fighters | +0.004 |
towns | +0.004 |
younger | +0.004 |
Neg logit contributions | |
dd | -0.004 |
night | -0.003 |
ez | -0.003 |
pity | -0.003 |
amb | -0.003 |
Token | to |
Token ID | 284 |
Feature activation | +5.30e-03 |
Pos logit contributions | |
dd | +0.359 |
night | +0.317 |
ez | +0.259 |
ggy | +0.252 |
pity | +0.251 |
Neg logit contributions | |
groups | -0.461 |
bars | -0.443 |
towns | -0.398 |
lights | -0.396 |
ains | -0.392 |
Token | go |
Token ID | 467 |
Feature activation | +3.22e-02 |
Pos logit contributions | |
dd | +0.010 |
night | +0.008 |
ez | +0.007 |
pity | +0.007 |
amb | +0.007 |
Neg logit contributions | |
groups | -0.010 |
bars | -0.009 |
towns | -0.008 |
fighters | -0.008 |
younger | -0.008 |
Token | bunny |
Token ID | 44915 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
dd | +0.035 |
night | +0.030 |
pity | +0.025 |
- | +0.024 |
oo | +0.024 |
Neg logit contributions | |
bars | -0.063 |
groups | -0.061 |
lights | -0.055 |
ents | -0.054 |
towns | -0.054 |
Token | said |
Token ID | 531 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
dd | +0.378 |
night | +0.311 |
ez | +0.280 |
pity | +0.269 |
amb | +0.263 |
Neg logit contributions | |
groups | -0.410 |
bars | -0.401 |
towns | -0.358 |
fighters | -0.357 |
eight | -0.355 |
Token | , |
Token ID | 11 |
Feature activation | +5.43e-02 |
Pos logit contributions | |
dd | +0.251 |
night | +0.208 |
ez | +0.192 |
pity | +0.187 |
uggle | +0.186 |
Neg logit contributions | |
groups | -0.216 |
bars | -0.209 |
eight | -0.191 |
fighters | -0.188 |
younger | -0.188 |
Token | " |
Token ID | 366 |
Feature activation | -5.81e-02 |
Pos logit contributions | |
dd | +0.109 |
night | +0.094 |
ez | +0.086 |
pity | +0.083 |
amb | +0.082 |
Neg logit contributions | |
groups | -0.086 |
bars | -0.085 |
towns | -0.073 |
fighters | -0.072 |
lights | -0.072 |
Token | Good |
Token ID | 10248 |
Feature activation | +2.77e-02 |
Pos logit contributions | |
groups | +0.102 |
bars | +0.102 |
towns | +0.090 |
eight | +0.090 |
fighters | +0.090 |
Neg logit contributions | |
dd | -0.104 |
night | -0.089 |
ez | -0.079 |
pity | -0.077 |
uggle | -0.075 |
Token | night |
Token ID | 3847 |
Feature activation | +4.27e-01 |
Pos logit contributions | |
dd | +0.040 |
ez | +0.031 |
to | +0.027 |
faced | +0.026 |
uggle | +0.025 |
Neg logit contributions | |
fastest | -0.052 |
groups | -0.052 |
bars | -0.051 |
fighters | -0.051 |
older | -0.051 |
Token | sheep |
Token ID | 15900 |
Feature activation | +4.43e-02 |
Pos logit contributions | |
dd | +0.761 |
night | +0.622 |
ez | +0.573 |
pity | +0.550 |
amb | +0.537 |
Neg logit contributions | |
groups | -0.779 |
bars | -0.754 |
fighters | -0.680 |
towns | -0.678 |
eight | -0.678 |
Token | ! |
Token ID | 0 |
Feature activation | +5.77e-02 |
Pos logit contributions | |
dd | +0.073 |
night | +0.062 |
ez | +0.054 |
pity | +0.051 |
amb | +0.051 |
Neg logit contributions | |
groups | -0.088 |
bars | -0.087 |
lights | -0.077 |
towns | -0.077 |
ents | -0.076 |
Token | I |
Token ID | 314 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
dd | +0.107 |
night | +0.089 |
ez | +0.081 |
pity | +0.078 |
amb | +0.076 |
Neg logit contributions | |
groups | -0.101 |
bars | -0.100 |
towns | -0.088 |
fighters | -0.088 |
eight | -0.087 |
Token | 'm |
Token ID | 1101 |
Feature activation | +6.94e-02 |
Pos logit contributions | |
groups | +0.169 |
bars | +0.158 |
department | +0.148 |
eight | +0.147 |
fighters | +0.146 |
Neg logit contributions | |
dd | -0.204 |
night | -0.165 |
faced | -0.151 |
ez | -0.151 |
amb | -0.150 |
Token | gonna |
Token ID | 8066 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
dd | +0.122 |
night | +0.104 |
ez | +0.093 |
amb | +0.091 |
pity | +0.089 |
Neg logit contributions | |
groups | -0.125 |
bars | -0.119 |
older | -0.113 |
younger | -0.113 |
eight | -0.110 |
Token | the |
Token ID | 262 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
bars | +0.331 |
groups | +0.316 |
fighters | +0.287 |
department | +0.282 |
fastest | +0.279 |
Neg logit contributions | |
dd | -0.365 |
night | -0.305 |
ez | -0.287 |
pity | -0.283 |
it | -0.261 |
Token | crib |
Token ID | 48083 |
Feature activation | +6.72e-02 |
Pos logit contributions | |
groups | +0.202 |
bars | +0.198 |
towns | +0.173 |
fighters | +0.172 |
younger | +0.171 |
Neg logit contributions | |
dd | -0.236 |
night | -0.200 |
ez | -0.182 |
pity | -0.175 |
amb | -0.173 |
Token | and |
Token ID | 290 |
Feature activation | +7.52e-02 |
Pos logit contributions | |
dd | +0.067 |
night | +0.066 |
- | +0.061 |
it | +0.051 |
amb | +0.049 |
Neg logit contributions | |
bars | -0.152 |
groups | -0.148 |
ains | -0.143 |
narrower | -0.141 |
thirteen | -0.138 |
Token | said |
Token ID | 531 |
Feature activation | +9.52e-02 |
Pos logit contributions | |
dd | +0.141 |
night | +0.117 |
ez | +0.108 |
pity | +0.102 |
amb | +0.102 |
Neg logit contributions | |
groups | -0.131 |
bars | -0.129 |
towns | -0.114 |
fighters | -0.113 |
younger | -0.113 |
Token | good |
Token ID | 922 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
dd | +0.169 |
night | +0.139 |
ez | +0.129 |
pity | +0.122 |
amb | +0.120 |
Neg logit contributions | |
groups | -0.174 |
bars | -0.170 |
eight | -0.154 |
towns | -0.152 |
younger | -0.152 |
Token | night |
Token ID | 3847 |
Feature activation | +4.21e-01 |
Pos logit contributions | |
dd | +0.146 |
ez | +0.099 |
faced | +0.088 |
uggle | +0.080 |
ggy | +0.076 |
Neg logit contributions | |
bars | -0.250 |
older | -0.243 |
department | -0.243 |
fighters | -0.242 |
younger | -0.238 |
Token | to |
Token ID | 284 |
Feature activation | -7.36e-03 |
Pos logit contributions | |
dd | +0.768 |
night | +0.652 |
ez | +0.580 |
pity | +0.555 |
amb | +0.549 |
Neg logit contributions | |
groups | -0.758 |
bars | -0.744 |
towns | -0.656 |
fighters | -0.650 |
younger | -0.644 |
Token | her |
Token ID | 607 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
groups | +0.013 |
bars | +0.013 |
towns | +0.011 |
fighters | +0.011 |
younger | +0.011 |
Neg logit contributions | |
dd | -0.014 |
night | -0.012 |
ez | -0.010 |
pity | -0.010 |
amb | -0.010 |
Token | mom |
Token ID | 1995 |
Feature activation | -6.90e-03 |
Pos logit contributions | |
dd | +0.214 |
night | +0.186 |
ez | +0.163 |
pity | +0.159 |
amb | +0.155 |
Neg logit contributions | |
groups | -0.221 |
bars | -0.220 |
towns | -0.193 |
fighters | -0.192 |
eight | -0.190 |
Token | . |
Token ID | 13 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
groups | +0.015 |
bars | +0.014 |
towns | +0.013 |
narrower | +0.013 |
lights | +0.013 |
Neg logit contributions | |
dd | -0.010 |
night | -0.009 |
ez | -0.007 |
- | -0.007 |
amb | -0.007 |
Token | She |
Token ID | 1375 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
groups | +0.035 |
bars | +0.035 |
towns | +0.031 |
fighters | +0.031 |
younger | +0.030 |
Neg logit contributions | |
dd | -0.030 |
night | -0.025 |
ez | -0.022 |
amb | -0.021 |
pity | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | +7.00e-02 |
Pos logit contributions | |
dd | +0.010 |
night | +0.008 |
ez | +0.008 |
pity | +0.007 |
amb | +0.007 |
Neg logit contributions | |
groups | -0.012 |
bars | -0.012 |
towns | -0.011 |
fighters | -0.011 |
younger | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | +6.46e-02 |
Pos logit contributions | |
dd | +0.108 |
night | +0.095 |
pity | +0.077 |
amb | +0.076 |
ez | +0.075 |
Neg logit contributions | |
groups | -0.143 |
bars | -0.129 |
eight | -0.128 |
towns | -0.126 |
ents | -0.124 |
Token | He |
Token ID | 1544 |
Feature activation | -9.38e-02 |
Pos logit contributions | |
dd | +0.149 |
night | +0.133 |
ez | +0.122 |
pity | +0.120 |
amb | +0.116 |
Neg logit contributions | |
groups | -0.084 |
bars | -0.079 |
towns | -0.069 |
eight | -0.068 |
fighters | -0.066 |
Token | said |
Token ID | 531 |
Feature activation | +9.33e-02 |
Pos logit contributions | |
groups | +0.176 |
bars | +0.173 |
towns | +0.153 |
fighters | +0.153 |
younger | +0.152 |
Neg logit contributions | |
dd | -0.163 |
night | -0.136 |
ez | -0.122 |
pity | -0.117 |
amb | -0.114 |
Token | good |
Token ID | 922 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
dd | +0.161 |
night | +0.131 |
ez | +0.122 |
pity | +0.115 |
amb | +0.113 |
Neg logit contributions | |
groups | -0.175 |
bars | -0.171 |
eight | -0.156 |
towns | -0.154 |
fighters | -0.154 |
Token | night |
Token ID | 3847 |
Feature activation | +4.21e-01 |
Pos logit contributions | |
dd | +0.191 |
ez | +0.138 |
faced | +0.126 |
ggy | +0.122 |
uggle | +0.120 |
Neg logit contributions | |
bars | -0.218 |
fighters | -0.216 |
department | -0.214 |
older | -0.214 |
towns | -0.212 |
Token | to |
Token ID | 284 |
Feature activation | -5.65e-03 |
Pos logit contributions | |
dd | +0.750 |
night | +0.613 |
ez | +0.564 |
pity | +0.545 |
amb | +0.527 |
Neg logit contributions | |
groups | -0.765 |
bars | -0.750 |
fighters | -0.669 |
towns | -0.667 |
eight | -0.666 |
Token | the |
Token ID | 262 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
groups | +0.010 |
bars | +0.010 |
fighters | +0.009 |
towns | +0.009 |
department | +0.009 |
Neg logit contributions | |
dd | -0.010 |
night | -0.009 |
ez | -0.008 |
pity | -0.008 |
amb | -0.007 |
Token | moon |
Token ID | 8824 |
Feature activation | -9.95e-02 |
Pos logit contributions | |
groups | +0.221 |
bars | +0.215 |
towns | +0.193 |
younger | +0.192 |
fighters | +0.191 |
Neg logit contributions | |
dd | -0.222 |
night | -0.184 |
ez | -0.168 |
amb | -0.159 |
pity | -0.159 |
Token | before |
Token ID | 878 |
Feature activation | -2.77e-01 |
Pos logit contributions | |
groups | +0.209 |
bars | +0.201 |
towns | +0.183 |
eight | +0.182 |
younger | +0.181 |
Neg logit contributions | |
dd | -0.143 |
night | -0.130 |
ez | -0.109 |
- | -0.104 |
amb | -0.101 |
Token | he |
Token ID | 339 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
groups | +0.519 |
bars | +0.515 |
towns | +0.450 |
fighters | +0.448 |
younger | +0.448 |
Neg logit contributions | |
dd | -0.491 |
night | -0.407 |
ez | -0.361 |
pity | -0.348 |
oo | -0.337 |
Token | toys |
Token ID | 14958 |
Feature activation | -6.63e-03 |
Pos logit contributions | |
groups | +0.046 |
bars | +0.045 |
towns | +0.040 |
younger | +0.040 |
eight | +0.039 |
Neg logit contributions | |
dd | -0.048 |
night | -0.040 |
ez | -0.036 |
amb | -0.035 |
pity | -0.034 |
Token | and |
Token ID | 290 |
Feature activation | +8.10e-02 |
Pos logit contributions | |
groups | +0.015 |
bars | +0.015 |
eight | +0.014 |
towns | +0.014 |
younger | +0.014 |
Neg logit contributions | |
dd | -0.009 |
night | -0.007 |
- | -0.006 |
ez | -0.005 |
uggle | -0.005 |
Token | hugged |
Token ID | 41130 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
dd | +0.145 |
night | +0.121 |
ez | +0.109 |
pity | +0.104 |
amb | +0.103 |
Neg logit contributions | |
groups | -0.148 |
bars | -0.145 |
towns | -0.129 |
fighters | -0.128 |
younger | -0.127 |
Token | Mom |
Token ID | 11254 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
dd | +0.240 |
night | +0.199 |
ez | +0.189 |
faced | +0.183 |
amb | +0.181 |
Neg logit contributions | |
groups | -0.158 |
bars | -0.148 |
towns | -0.135 |
fighters | -0.129 |
lights | -0.129 |
Token | good |
Token ID | 922 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
groups | +0.027 |
bars | +0.026 |
eight | +0.024 |
towns | +0.024 |
fighters | +0.024 |
Neg logit contributions | |
dd | -0.017 |
night | -0.016 |
ez | -0.014 |
- | -0.013 |
amb | -0.012 |
Token | night |
Token ID | 3847 |
Feature activation | +4.20e-01 |
Pos logit contributions | |
dd | +0.119 |
ggy | +0.073 |
faced | +0.067 |
ez | +0.067 |
amb | +0.063 |
Neg logit contributions | |
bars | -0.294 |
groups | -0.289 |
directions | -0.288 |
department | -0.286 |
fighters | -0.284 |
Token | . |
Token ID | 13 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
dd | +0.802 |
night | +0.673 |
ez | +0.617 |
pity | +0.594 |
amb | +0.583 |
Neg logit contributions | |
groups | -0.716 |
bars | -0.700 |
towns | -0.616 |
fighters | -0.615 |
younger | -0.610 |
Token | They |
Token ID | 1119 |
Feature activation | -3.66e-02 |
Pos logit contributions | |
groups | +0.023 |
bars | +0.022 |
towns | +0.020 |
fighters | +0.020 |
younger | +0.019 |
Neg logit contributions | |
dd | -0.025 |
night | -0.021 |
ez | -0.019 |
pity | -0.018 |
amb | -0.018 |
Token | went |
Token ID | 1816 |
Feature activation | +6.31e-02 |
Pos logit contributions | |
groups | +0.069 |
bars | +0.068 |
towns | +0.060 |
fighters | +0.060 |
younger | +0.060 |
Neg logit contributions | |
dd | -0.063 |
night | -0.053 |
ez | -0.047 |
pity | -0.045 |
amb | -0.044 |
Token | to |
Token ID | 284 |
Feature activation | -1.25e-02 |
Pos logit contributions | |
dd | +0.124 |
night | +0.098 |
ez | +0.096 |
pity | +0.090 |
oof | +0.088 |
Neg logit contributions | |
groups | -0.107 |
bars | -0.103 |
older | -0.095 |
towns | -0.093 |
younger | -0.092 |
Token | their |
Token ID | 511 |
Feature activation | -4.29e-02 |
Pos logit contributions | |
groups | +0.022 |
bars | +0.021 |
younger | +0.020 |
older | +0.019 |
eight | +0.019 |
Neg logit contributions | |
dd | -0.024 |
night | -0.019 |
ez | -0.018 |
amb | -0.017 |
faced | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | +7.80e-02 |
Pos logit contributions | |
dd | +0.174 |
night | +0.142 |
ez | +0.125 |
pity | +0.118 |
amb | +0.116 |
Neg logit contributions | |
groups | -0.230 |
bars | -0.225 |
towns | -0.203 |
fighters | -0.202 |
younger | -0.201 |
Token | said |
Token ID | 531 |
Feature activation | +9.24e-02 |
Pos logit contributions | |
dd | +0.133 |
night | +0.110 |
ez | +0.099 |
pity | +0.094 |
amb | +0.094 |
Neg logit contributions | |
groups | -0.149 |
bars | -0.146 |
towns | -0.130 |
fighters | -0.130 |
younger | -0.129 |
Token | , |
Token ID | 11 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
dd | +0.166 |
night | +0.139 |
ez | +0.128 |
pity | +0.122 |
amb | +0.122 |
Neg logit contributions | |
groups | -0.164 |
bars | -0.162 |
eight | -0.144 |
fighters | -0.143 |
towns | -0.143 |
Token | " |
Token ID | 366 |
Feature activation | -8.27e-02 |
Pos logit contributions | |
dd | +0.042 |
night | +0.035 |
ez | +0.031 |
amb | +0.030 |
it | +0.030 |
Neg logit contributions | |
groups | -0.043 |
bars | -0.042 |
towns | -0.037 |
fastest | -0.037 |
lights | -0.037 |
Token | Good |
Token ID | 10248 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
bars | +0.157 |
groups | +0.156 |
towns | +0.138 |
fighters | +0.138 |
eight | +0.136 |
Neg logit contributions | |
dd | -0.139 |
night | -0.117 |
ez | -0.103 |
pity | -0.100 |
uggle | -0.097 |
Token | night |
Token ID | 3847 |
Feature activation | +4.20e-01 |
Pos logit contributions | |
dd | +0.020 |
ez | +0.014 |
ggy | +0.012 |
faced | +0.012 |
pity | +0.011 |
Neg logit contributions | |
bars | -0.033 |
groups | -0.032 |
older | -0.032 |
younger | -0.031 |
fighters | -0.031 |
Token | , |
Token ID | 11 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
dd | +0.702 |
night | +0.598 |
ez | +0.516 |
pity | +0.493 |
amb | +0.487 |
Neg logit contributions | |
groups | -0.808 |
bars | -0.805 |
towns | -0.705 |
fighters | -0.696 |
ents | -0.693 |
Token | Joe |
Token ID | 5689 |
Feature activation | -5.36e-03 |
Pos logit contributions | |
dd | +0.032 |
night | +0.027 |
ez | +0.023 |
pity | +0.023 |
amb | +0.022 |
Neg logit contributions | |
bars | -0.039 |
groups | -0.039 |
towns | -0.034 |
fighters | -0.034 |
lights | -0.034 |
Token | ." |
Token ID | 526 |
Feature activation | -6.77e-02 |
Pos logit contributions | |
groups | +0.011 |
bars | +0.011 |
lights | +0.010 |
towns | +0.009 |
ents | +0.009 |
Neg logit contributions | |
dd | -0.009 |
night | -0.008 |
ez | -0.007 |
- | -0.006 |
oo | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +5.77e-02 |
Pos logit contributions | |
groups | +0.134 |
bars | +0.132 |
towns | +0.118 |
fighters | +0.117 |
younger | +0.117 |
Neg logit contributions | |
dd | -0.111 |
night | -0.092 |
ez | -0.082 |
pity | -0.077 |
amb | -0.077 |
Token |
|
Token ID | 198 |
Feature activation | +5.45e-02 |
Pos logit contributions | |
dd | +0.098 |
night | +0.089 |
pity | +0.076 |
ez | +0.075 |
faced | +0.072 |
Neg logit contributions | |
groups | -0.105 |
bars | -0.099 |
eight | -0.093 |
towns | -0.092 |
ents | -0.090 |
Token | her |
Token ID | 607 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
groups | +0.030 |
bars | +0.029 |
older | +0.028 |
younger | +0.027 |
eight | +0.027 |
Neg logit contributions | |
dd | -0.031 |
night | -0.025 |
ez | -0.023 |
amb | -0.023 |
ager | -0.022 |
Token | new |
Token ID | 649 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
dd | +0.212 |
night | +0.192 |
ez | +0.165 |
pity | +0.164 |
faced | +0.152 |
Neg logit contributions | |
groups | -0.256 |
bars | -0.255 |
fighters | -0.230 |
towns | -0.227 |
fastest | -0.226 |
Token | doll |
Token ID | 3654 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
groups | +0.041 |
bars | +0.040 |
younger | +0.036 |
older | +0.036 |
towns | +0.036 |
Neg logit contributions | |
dd | -0.038 |
night | -0.030 |
ez | -0.027 |
amb | -0.026 |
pity | -0.026 |
Token | a |
Token ID | 257 |
Feature activation | +7.46e-02 |
Pos logit contributions | |
dd | +0.272 |
night | +0.233 |
ez | +0.201 |
pity | +0.192 |
faced | +0.188 |
Neg logit contributions | |
groups | -0.349 |
bars | -0.346 |
towns | -0.307 |
fighters | -0.306 |
younger | -0.303 |
Token | good |
Token ID | 922 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
dd | +0.152 |
night | +0.114 |
ez | +0.114 |
amb | +0.110 |
oo | +0.103 |
Neg logit contributions | |
groups | -0.126 |
bars | -0.123 |
younger | -0.117 |
older | -0.115 |
eight | -0.115 |
Token | night |
Token ID | 3847 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
dd | +0.262 |
ez | +0.181 |
amb | +0.181 |
cha | +0.170 |
to | +0.164 |
Neg logit contributions | |
older | -0.178 |
younger | -0.175 |
groups | -0.174 |
department | -0.169 |
size | -0.169 |
Token | hug |
Token ID | 16225 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
dd | +0.776 |
night | +0.600 |
ez | +0.575 |
pity | +0.561 |
faced | +0.551 |
Neg logit contributions | |
groups | -0.727 |
bars | -0.709 |
eight | -0.642 |
fighters | -0.638 |
younger | -0.636 |
Token | and |
Token ID | 290 |
Feature activation | +8.14e-02 |
Pos logit contributions | |
dd | +0.196 |
night | +0.160 |
ez | +0.145 |
pity | +0.139 |
amb | +0.136 |
Neg logit contributions | |
groups | -0.212 |
bars | -0.209 |
towns | -0.186 |
fighters | -0.186 |
younger | -0.184 |
Token | put |
Token ID | 1234 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
dd | +0.151 |
night | +0.118 |
ez | +0.117 |
amb | +0.112 |
faced | +0.110 |
Neg logit contributions | |
groups | -0.138 |
bars | -0.137 |
younger | -0.123 |
fighters | -0.122 |
towns | -0.121 |
Token | it |
Token ID | 340 |
Feature activation | +3.18e-01 |
Pos logit contributions | |
dd | +0.044 |
amb | +0.034 |
faced | +0.034 |
night | +0.033 |
ggy | +0.033 |
Neg logit contributions | |
bars | -0.038 |
groups | -0.037 |
older | -0.034 |
forks | -0.034 |
directions | -0.034 |
Token | away |
Token ID | 1497 |
Feature activation | +9.54e-02 |
Pos logit contributions | |
dd | +0.474 |
night | +0.398 |
ez | +0.336 |
pity | +0.315 |
uggle | +0.311 |
Neg logit contributions | |
groups | -0.674 |
bars | -0.660 |
fighters | -0.599 |
towns | -0.598 |
younger | -0.589 |
Token | bed |
Token ID | 3996 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
dd | +0.217 |
night | +0.188 |
ez | +0.163 |
pity | +0.158 |
uggle | +0.154 |
Neg logit contributions | |
groups | -0.237 |
bars | -0.234 |
fighters | -0.210 |
towns | -0.208 |
eight | -0.205 |
Token | and |
Token ID | 290 |
Feature activation | +7.66e-02 |
Pos logit contributions | |
dd | +0.021 |
night | +0.018 |
- | +0.015 |
ez | +0.013 |
uggle | +0.013 |
Neg logit contributions | |
groups | -0.034 |
bars | -0.033 |
towns | -0.031 |
bridges | -0.031 |
fighters | -0.031 |
Token | kissed |
Token ID | 27946 |
Feature activation | +5.27e-02 |
Pos logit contributions | |
dd | +0.139 |
night | +0.114 |
ez | +0.106 |
amb | +0.102 |
pity | +0.100 |
Neg logit contributions | |
groups | -0.137 |
bars | -0.135 |
younger | -0.119 |
fighters | -0.119 |
towns | -0.118 |
Token | her |
Token ID | 607 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
dd | +0.081 |
night | +0.063 |
ez | +0.058 |
amb | +0.054 |
pity | +0.054 |
Neg logit contributions | |
groups | -0.109 |
bars | -0.107 |
towns | -0.096 |
younger | -0.096 |
fighters | -0.096 |
Token | good |
Token ID | 922 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
dd | +0.234 |
night | +0.177 |
ez | +0.173 |
amb | +0.165 |
pity | +0.162 |
Neg logit contributions | |
groups | -0.212 |
bars | -0.209 |
younger | -0.189 |
older | -0.187 |
eight | -0.185 |
Token | night |
Token ID | 3847 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
dd | +0.126 |
ez | +0.071 |
amb | +0.061 |
faced | +0.061 |
oof | +0.057 |
Neg logit contributions | |
bars | -0.287 |
groups | -0.282 |
older | -0.271 |
younger | -0.270 |
department | -0.265 |
Token | . |
Token ID | 13 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
dd | +0.786 |
night | +0.658 |
ez | +0.602 |
pity | +0.578 |
amb | +0.569 |
Neg logit contributions | |
groups | -0.721 |
bars | -0.706 |
towns | -0.622 |
fighters | -0.620 |
younger | -0.616 |
Token |
|
Token ID | 198 |
Feature activation | +6.19e-02 |
Pos logit contributions | |
groups | +0.025 |
bars | +0.025 |
towns | +0.022 |
fighters | +0.022 |
younger | +0.022 |
Neg logit contributions | |
dd | -0.023 |
night | -0.019 |
ez | -0.017 |
pity | -0.016 |
amb | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | +5.84e-02 |
Pos logit contributions | |
dd | +0.100 |
night | +0.088 |
pity | +0.076 |
amb | +0.074 |
ez | +0.073 |
Neg logit contributions | |
groups | -0.116 |
bars | -0.108 |
eight | -0.105 |
ents | -0.101 |
directions | -0.100 |
Token | The |
Token ID | 464 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
dd | +0.124 |
night | +0.109 |
ez | +0.100 |
pity | +0.097 |
amb | +0.095 |
Neg logit contributions | |
groups | -0.086 |
bars | -0.083 |
towns | -0.072 |
eight | -0.071 |
fighters | -0.071 |
Token | little |
Token ID | 1310 |
Feature activation | -6.49e-02 |
Pos logit contributions | |
groups | +0.311 |
bars | +0.303 |
older | +0.262 |
younger | +0.262 |
towns | +0.260 |
Neg logit contributions | |
dd | -0.523 |
night | -0.443 |
ez | -0.425 |
amb | -0.410 |
pity | -0.401 |
Token | them |
Token ID | 606 |
Feature activation | +4.06e-02 |
Pos logit contributions | |
dd | +0.045 |
night | +0.038 |
ez | +0.033 |
pity | +0.032 |
amb | +0.030 |
Neg logit contributions | |
groups | -0.056 |
bars | -0.055 |
fighters | -0.050 |
towns | -0.049 |
eight | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
dd | +0.045 |
night | +0.042 |
- | +0.039 |
it | +0.037 |
ggy | +0.035 |
Neg logit contributions | |
bars | -0.091 |
groups | -0.087 |
eight | -0.083 |
lights | -0.082 |
thirteen | -0.082 |
Token | She |
Token ID | 1375 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
groups | +0.024 |
bars | +0.024 |
fighters | +0.022 |
towns | +0.022 |
fastest | +0.021 |
Neg logit contributions | |
dd | -0.019 |
night | -0.016 |
ez | -0.014 |
amb | -0.013 |
pity | -0.013 |
Token | said |
Token ID | 531 |
Feature activation | +9.17e-02 |
Pos logit contributions | |
dd | +0.027 |
night | +0.023 |
ez | +0.020 |
pity | +0.019 |
amb | +0.019 |
Neg logit contributions | |
groups | -0.030 |
bars | -0.030 |
towns | -0.026 |
fighters | -0.026 |
younger | -0.026 |
Token | good |
Token ID | 922 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
dd | +0.155 |
night | +0.128 |
ez | +0.116 |
pity | +0.110 |
amb | +0.109 |
Neg logit contributions | |
groups | -0.176 |
bars | -0.172 |
towns | -0.154 |
fighters | -0.154 |
eight | -0.153 |
Token | night |
Token ID | 3847 |
Feature activation | +4.17e-01 |
Pos logit contributions | |
dd | +0.186 |
ez | +0.136 |
faced | +0.130 |
ggy | +0.126 |
uggle | +0.123 |
Neg logit contributions | |
bars | -0.243 |
groups | -0.234 |
older | -0.232 |
fighters | -0.230 |
department | -0.230 |
Token | and |
Token ID | 290 |
Feature activation | +7.22e-02 |
Pos logit contributions | |
dd | +0.719 |
night | +0.588 |
ez | +0.537 |
pity | +0.516 |
amb | +0.502 |
Neg logit contributions | |
groups | -0.780 |
bars | -0.765 |
fighters | -0.685 |
towns | -0.683 |
eight | -0.681 |
Token | went |
Token ID | 1816 |
Feature activation | +6.46e-02 |
Pos logit contributions | |
dd | +0.131 |
night | +0.109 |
ez | +0.100 |
pity | +0.095 |
amb | +0.094 |
Neg logit contributions | |
groups | -0.129 |
bars | -0.127 |
towns | -0.112 |
fighters | -0.112 |
younger | -0.111 |
Token | off |
Token ID | 572 |
Feature activation | -3.49e-03 |
Pos logit contributions | |
dd | +0.129 |
night | +0.104 |
ez | +0.100 |
oof | +0.095 |
pity | +0.095 |
Neg logit contributions | |
groups | -0.104 |
bars | -0.101 |
older | -0.093 |
younger | -0.091 |
towns | -0.091 |
Token | to |
Token ID | 284 |
Feature activation | -3.74e-03 |
Pos logit contributions | |
groups | +0.006 |
bars | +0.006 |
fighters | +0.006 |
towns | +0.006 |
younger | +0.005 |
Neg logit contributions | |
dd | -0.006 |
night | -0.005 |
ez | -0.005 |
pity | -0.004 |
amb | -0.004 |
Token | bed |
Token ID | 3996 |
Feature activation | +9.07e-03 |
Pos logit contributions | |
groups | +0.007 |
bars | +0.007 |
younger | +0.007 |
older | +0.007 |
towns | +0.007 |
Neg logit contributions | |
dd | -0.006 |
night | -0.005 |
ez | -0.004 |
amb | -0.004 |
faced | -0.004 |
Token | are |
Token ID | 389 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
groups | +0.332 |
bars | +0.310 |
department | +0.297 |
younger | +0.295 |
towns | +0.293 |
Neg logit contributions | |
dd | -0.317 |
night | -0.263 |
it | -0.248 |
ez | -0.236 |
wish | -0.234 |
Token | the |
Token ID | 262 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
bars | +0.286 |
groups | +0.254 |
towns | +0.236 |
ains | +0.233 |
oes | +0.230 |
Neg logit contributions | |
dd | -0.197 |
it | -0.157 |
pity | -0.153 |
night | -0.145 |
happened | -0.140 |
Token | sun |
Token ID | 4252 |
Feature activation | +6.16e-02 |
Pos logit contributions | |
bars | +0.202 |
groups | +0.200 |
ents | +0.177 |
towns | +0.171 |
lights | +0.170 |
Neg logit contributions | |
dd | -0.209 |
night | -0.186 |
pity | -0.164 |
ez | -0.159 |
wish | -0.154 |
Token | 's |
Token ID | 338 |
Feature activation | -8.40e-02 |
Pos logit contributions | |
dd | +0.082 |
night | +0.073 |
- | +0.065 |
it | +0.061 |
ake | +0.059 |
Neg logit contributions | |
groups | -0.135 |
bars | -0.126 |
towns | -0.119 |
eight | -0.117 |
narrower | -0.117 |
Token | good |
Token ID | 922 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
bars | +0.153 |
groups | +0.152 |
ents | +0.132 |
towns | +0.130 |
fighters | +0.130 |
Neg logit contributions | |
dd | -0.147 |
night | -0.130 |
pity | -0.111 |
ez | -0.109 |
- | -0.106 |
Token | night |
Token ID | 3847 |
Feature activation | +4.15e-01 |
Pos logit contributions | |
dd | +0.190 |
night | +0.165 |
happened | +0.136 |
- | +0.136 |
it | +0.131 |
Neg logit contributions | |
groups | -0.247 |
bars | -0.246 |
towns | -0.223 |
ents | -0.220 |
eight | -0.219 |
Token | kiss |
Token ID | 9245 |
Feature activation | +8.08e-02 |
Pos logit contributions | |
dd | +0.654 |
night | +0.576 |
ez | +0.469 |
- | +0.459 |
pity | +0.440 |
Neg logit contributions | |
groups | -0.839 |
bars | -0.829 |
towns | -0.738 |
younger | -0.727 |
ents | -0.723 |
Token | to |
Token ID | 284 |
Feature activation | -7.84e-03 |
Pos logit contributions | |
dd | +0.133 |
night | +0.113 |
ez | +0.098 |
pity | +0.093 |
oo | +0.090 |
Neg logit contributions | |
groups | -0.160 |
bars | -0.156 |
towns | -0.140 |
younger | -0.138 |
fighters | -0.137 |
Token | the |
Token ID | 262 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
groups | +0.014 |
bars | +0.014 |
fighters | +0.012 |
towns | +0.012 |
department | +0.012 |
Neg logit contributions | |
dd | -0.014 |
night | -0.012 |
ez | -0.011 |
pity | -0.011 |
oo | -0.010 |
Token | earth |
Token ID | 4534 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
groups | +0.199 |
bars | +0.196 |
fighters | +0.169 |
towns | +0.168 |
eight | +0.166 |
Neg logit contributions | |
dd | -0.237 |
night | -0.204 |
ez | -0.183 |
pity | -0.178 |
amb | -0.173 |
Token | ." |
Token ID | 526 |
Feature activation | -6.28e-02 |
Pos logit contributions | |
groups | +0.023 |
bars | +0.023 |
narrower | +0.021 |
eight | +0.021 |
towns | +0.021 |
Neg logit contributions | |
dd | -0.016 |
night | -0.014 |
- | -0.012 |
ez | -0.011 |
it | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
dd | +0.224 |
night | +0.188 |
amb | +0.172 |
ez | +0.171 |
oof | +0.169 |
Neg logit contributions | |
groups | -0.179 |
bars | -0.154 |
towns | -0.149 |
bridges | -0.147 |
lights | -0.144 |
Token | Then |
Token ID | 3244 |
Feature activation | +3.54e-02 |
Pos logit contributions | |
groups | +0.019 |
bars | +0.019 |
towns | +0.017 |
eight | +0.017 |
younger | +0.017 |
Neg logit contributions | |
dd | -0.020 |
night | -0.017 |
ez | -0.015 |
pity | -0.015 |
faced | -0.014 |
Token | she |
Token ID | 673 |
Feature activation | +4.09e-02 |
Pos logit contributions | |
dd | +0.063 |
night | +0.052 |
ez | +0.048 |
pity | +0.045 |
amb | +0.045 |
Neg logit contributions | |
groups | -0.065 |
bars | -0.063 |
towns | -0.057 |
eight | -0.056 |
fighters | -0.056 |
Token | said |
Token ID | 531 |
Feature activation | +8.98e-02 |
Pos logit contributions | |
dd | +0.069 |
night | +0.057 |
ez | +0.050 |
pity | +0.048 |
amb | +0.046 |
Neg logit contributions | |
groups | -0.079 |
bars | -0.079 |
towns | -0.070 |
fighters | -0.069 |
lights | -0.069 |
Token | good |
Token ID | 922 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
dd | +0.155 |
night | +0.124 |
ez | +0.120 |
oof | +0.112 |
pity | +0.111 |
Neg logit contributions | |
groups | -0.165 |
bars | -0.163 |
eight | -0.151 |
older | -0.149 |
younger | -0.149 |
Token | night |
Token ID | 3847 |
Feature activation | +4.15e-01 |
Pos logit contributions | |
dd | +0.150 |
faced | +0.099 |
ez | +0.096 |
ggy | +0.091 |
Here | +0.086 |
Neg logit contributions | |
bars | -0.248 |
older | -0.244 |
oes | -0.242 |
department | -0.241 |
fighters | -0.238 |
Token | to |
Token ID | 284 |
Feature activation | -9.57e-03 |
Pos logit contributions | |
dd | +0.778 |
night | +0.622 |
ez | +0.598 |
pity | +0.585 |
uggle | +0.567 |
Neg logit contributions | |
groups | -0.698 |
bars | -0.687 |
eight | -0.618 |
fighters | -0.616 |
younger | -0.615 |
Token | the |
Token ID | 262 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
groups | +0.016 |
bars | +0.015 |
towns | +0.014 |
younger | +0.014 |
eight | +0.014 |
Neg logit contributions | |
dd | -0.019 |
night | -0.016 |
ez | -0.015 |
amb | -0.014 |
faced | -0.014 |
Token | nos |
Token ID | 43630 |
Feature activation | -2.52e-03 |
Pos logit contributions | |
groups | +0.226 |
bars | +0.223 |
towns | +0.196 |
fighters | +0.195 |
younger | +0.193 |
Neg logit contributions | |
dd | -0.232 |
night | -0.195 |
ez | -0.175 |
pity | -0.169 |
amb | -0.165 |
Token | y |
Token ID | 88 |
Feature activation | -8.34e-02 |
Pos logit contributions | |
groups | +0.004 |
bars | +0.004 |
towns | +0.003 |
fighters | +0.003 |
younger | +0.003 |
Neg logit contributions | |
dd | -0.005 |
night | -0.004 |
ez | -0.004 |
amb | -0.004 |
pity | -0.004 |
Token | little |
Token ID | 1310 |
Feature activation | -6.65e-02 |
Pos logit contributions | |
groups | +0.164 |
bars | +0.161 |
towns | +0.144 |
fighters | +0.144 |
younger | +0.143 |
Neg logit contributions | |
dd | -0.138 |
night | -0.113 |
ez | -0.101 |
pity | -0.096 |
amb | -0.094 |
Token | , |
Token ID | 11 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
dd | +0.369 |
night | +0.306 |
ez | +0.272 |
pity | +0.261 |
amb | +0.255 |
Neg logit contributions | |
groups | -0.413 |
bars | -0.407 |
towns | -0.361 |
fighters | -0.361 |
younger | -0.358 |
Token | Jack |
Token ID | 3619 |
Feature activation | +8.71e-02 |
Pos logit contributions | |
dd | +0.028 |
night | +0.023 |
ez | +0.019 |
pity | +0.019 |
amb | +0.018 |
Neg logit contributions | |
groups | -0.040 |
bars | -0.040 |
fighters | -0.036 |
towns | -0.036 |
eight | -0.035 |
Token | y |
Token ID | 88 |
Feature activation | -8.38e-02 |
Pos logit contributions | |
dd | +0.137 |
night | +0.115 |
ez | +0.101 |
- | +0.093 |
amb | +0.093 |
Neg logit contributions | |
groups | -0.179 |
bars | -0.176 |
lights | -0.160 |
towns | -0.158 |
bridges | -0.158 |
Token | said |
Token ID | 531 |
Feature activation | +8.39e-02 |
Pos logit contributions | |
groups | +0.162 |
bars | +0.160 |
lights | +0.142 |
fighters | +0.141 |
towns | +0.141 |
Neg logit contributions | |
dd | -0.138 |
night | -0.119 |
ez | -0.102 |
pity | -0.097 |
amb | -0.095 |
Token | good |
Token ID | 922 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
dd | +0.127 |
night | +0.097 |
ez | +0.093 |
oof | +0.086 |
amb | +0.085 |
Neg logit contributions | |
groups | -0.173 |
bars | -0.171 |
eight | -0.159 |
younger | -0.156 |
older | -0.156 |
Token | night |
Token ID | 3847 |
Feature activation | +4.14e-01 |
Pos logit contributions | |
dd | +0.134 |
ez | +0.093 |
faced | +0.083 |
ggy | +0.078 |
Here | +0.073 |
Neg logit contributions | |
bars | -0.241 |
older | -0.237 |
department | -0.236 |
fighters | -0.233 |
groups | -0.232 |
Token | to |
Token ID | 284 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
dd | +0.757 |
night | +0.604 |
ez | +0.576 |
pity | +0.558 |
uggle | +0.543 |
Neg logit contributions | |
groups | -0.725 |
bars | -0.711 |
fighters | -0.641 |
eight | -0.638 |
younger | -0.637 |
Token | the |
Token ID | 262 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
groups | +0.024 |
bars | +0.024 |
towns | +0.021 |
younger | +0.021 |
fighters | +0.021 |
Neg logit contributions | |
dd | -0.024 |
night | -0.020 |
ez | -0.018 |
pity | -0.017 |
amb | -0.017 |
Token | skull |
Token ID | 14511 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
groups | +0.228 |
bars | +0.212 |
older | +0.210 |
eight | +0.209 |
younger | +0.206 |
Neg logit contributions | |
dd | -0.252 |
ez | -0.204 |
night | -0.198 |
amb | -0.196 |
oo | -0.190 |
Token | and |
Token ID | 290 |
Feature activation | +6.81e-02 |
Pos logit contributions | |
dd | +0.213 |
night | +0.200 |
ez | +0.159 |
- | +0.157 |
faced | +0.145 |
Neg logit contributions | |
groups | -0.311 |
bars | -0.301 |
towns | -0.280 |
bridges | -0.272 |
ains | -0.272 |
Token | went |
Token ID | 1816 |
Feature activation | +5.78e-02 |
Pos logit contributions | |
dd | +0.131 |
ez | +0.104 |
night | +0.104 |
amb | +0.096 |
faced | +0.095 |
Neg logit contributions | |
groups | -0.111 |
bars | -0.110 |
younger | -0.098 |
fighters | -0.098 |
towns | -0.098 |
Token | anymore |
Token ID | 7471 |
Feature activation | -5.17e-02 |
Pos logit contributions | |
dd | +0.054 |
night | +0.046 |
- | +0.041 |
it | +0.039 |
ez | +0.039 |
Neg logit contributions | |
groups | -0.075 |
bars | -0.073 |
fighters | -0.070 |
ains | -0.069 |
bowls | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
bars | +0.097 |
groups | +0.097 |
towns | +0.085 |
fighters | +0.084 |
younger | +0.084 |
Neg logit contributions | |
dd | -0.090 |
night | -0.075 |
ez | -0.068 |
pity | -0.063 |
ggy | -0.062 |
Token | She |
Token ID | 1375 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
groups | +0.050 |
bars | +0.048 |
fighters | +0.044 |
towns | +0.043 |
lights | +0.043 |
Neg logit contributions | |
dd | -0.033 |
night | -0.027 |
ez | -0.024 |
amb | -0.024 |
pity | -0.023 |
Token | said |
Token ID | 531 |
Feature activation | +8.63e-02 |
Pos logit contributions | |
dd | +0.021 |
night | +0.017 |
ez | +0.016 |
pity | +0.015 |
uggle | +0.015 |
Neg logit contributions | |
groups | -0.019 |
bars | -0.019 |
younger | -0.017 |
eight | -0.017 |
towns | -0.017 |
Token | good |
Token ID | 922 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
dd | +0.135 |
night | +0.107 |
ez | +0.099 |
pity | +0.093 |
amb | +0.092 |
Neg logit contributions | |
groups | -0.173 |
bars | -0.172 |
eight | -0.158 |
towns | -0.156 |
younger | -0.156 |
Token | night |
Token ID | 3847 |
Feature activation | +4.14e-01 |
Pos logit contributions | |
dd | +0.140 |
ez | +0.086 |
faced | +0.083 |
ggy | +0.077 |
Here | +0.073 |
Neg logit contributions | |
bars | -0.257 |
older | -0.241 |
department | -0.241 |
oes | -0.238 |
fighters | -0.236 |
Token | to |
Token ID | 284 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
dd | +0.740 |
night | +0.598 |
ez | +0.557 |
pity | +0.539 |
uggle | +0.520 |
Neg logit contributions | |
groups | -0.744 |
bars | -0.733 |
fighters | -0.656 |
eight | -0.655 |
towns | -0.652 |
Token | the |
Token ID | 262 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
groups | +0.022 |
bars | +0.022 |
towns | +0.020 |
younger | +0.020 |
eight | +0.020 |
Neg logit contributions | |
dd | -0.023 |
night | -0.019 |
ez | -0.017 |
pity | -0.016 |
faced | -0.016 |
Token | cat |
Token ID | 3797 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
groups | +0.226 |
bars | +0.221 |
towns | +0.195 |
fighters | +0.194 |
younger | +0.192 |
Neg logit contributions | |
dd | -0.241 |
night | -0.203 |
ez | -0.184 |
pity | -0.177 |
amb | -0.174 |
Token | and |
Token ID | 290 |
Feature activation | +6.87e-02 |
Pos logit contributions | |
dd | +0.225 |
night | +0.191 |
ez | +0.168 |
pity | +0.156 |
amb | +0.154 |
Neg logit contributions | |
groups | -0.287 |
bars | -0.283 |
towns | -0.253 |
fighters | -0.251 |
eight | -0.249 |
Token | went |
Token ID | 1816 |
Feature activation | +5.78e-02 |
Pos logit contributions | |
dd | +0.126 |
night | +0.104 |
ez | +0.096 |
pity | +0.091 |
amb | +0.090 |
Neg logit contributions | |
groups | -0.121 |
bars | -0.120 |
towns | -0.106 |
fighters | -0.106 |
younger | -0.105 |
Token | it |
Token ID | 340 |
Feature activation | +3.18e-01 |
Pos logit contributions | |
dd | +0.044 |
amb | +0.034 |
faced | +0.034 |
night | +0.033 |
ggy | +0.033 |
Neg logit contributions | |
bars | -0.038 |
groups | -0.037 |
older | -0.034 |
forks | -0.034 |
directions | -0.034 |
Token | away |
Token ID | 1497 |
Feature activation | +9.54e-02 |
Pos logit contributions | |
dd | +0.474 |
night | +0.398 |
ez | +0.336 |
pity | +0.315 |
uggle | +0.311 |
Neg logit contributions | |
groups | -0.674 |
bars | -0.660 |
fighters | -0.599 |
towns | -0.598 |
younger | -0.589 |
Token | . |
Token ID | 13 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
dd | +0.140 |
night | +0.127 |
- | +0.103 |
ez | +0.099 |
pity | +0.097 |
Neg logit contributions | |
bars | -0.203 |
groups | -0.201 |
ents | -0.178 |
eight | -0.178 |
fighters | -0.178 |
Token | " |
Token ID | 366 |
Feature activation | -8.37e-02 |
Pos logit contributions | |
groups | +0.019 |
bars | +0.019 |
eight | +0.017 |
younger | +0.017 |
ents | +0.016 |
Neg logit contributions | |
dd | -0.021 |
night | -0.017 |
ez | -0.016 |
pity | -0.015 |
faced | -0.015 |
Token | Good |
Token ID | 10248 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
bars | +0.115 |
groups | +0.104 |
eight | +0.102 |
highlight | +0.098 |
fastest | +0.097 |
Neg logit contributions | |
dd | -0.167 |
night | -0.148 |
- | -0.136 |
pity | -0.135 |
oo | -0.135 |
Token | night |
Token ID | 3847 |
Feature activation | +4.14e-01 |
Pos logit contributions | |
dd | +0.022 |
ez | +0.017 |
to | +0.015 |
faced | +0.015 |
Aid | +0.014 |
Neg logit contributions | |
first | -0.027 |
size | -0.027 |
fighters | -0.027 |
department | -0.027 |
eight | -0.026 |
Token | , |
Token ID | 11 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
dd | +0.752 |
night | +0.608 |
ez | +0.577 |
pity | +0.547 |
faced | +0.536 |
Neg logit contributions | |
groups | -0.739 |
bars | -0.707 |
eight | -0.651 |
fighters | -0.650 |
older | -0.649 |
Token | Brown |
Token ID | 4373 |
Feature activation | +9.66e-02 |
Pos logit contributions | |
dd | +0.040 |
night | +0.032 |
ez | +0.030 |
pity | +0.028 |
amb | +0.028 |
Neg logit contributions | |
groups | -0.043 |
bars | -0.042 |
younger | -0.038 |
older | -0.038 |
eight | -0.038 |
Token | ie |
Token ID | 494 |
Feature activation | +5.89e-03 |
Pos logit contributions | |
dd | +0.144 |
night | +0.122 |
- | +0.103 |
ez | +0.102 |
oo | +0.096 |
Neg logit contributions | |
groups | -0.208 |
bars | -0.207 |
towns | -0.184 |
lights | -0.183 |
bridges | -0.181 |
Token | ," |
Token ID | 553 |
Feature activation | +9.91e-02 |
Pos logit contributions | |
dd | +0.010 |
night | +0.008 |
ez | +0.007 |
pity | +0.007 |
amb | +0.007 |
Neg logit contributions | |
groups | -0.011 |
bars | -0.011 |
fighters | -0.010 |
towns | -0.010 |
younger | -0.010 |
Token | she |
Token ID | 673 |
Feature activation | +4.10e-02 |
Pos logit contributions | |
dd | +0.200 |
night | +0.165 |
ez | +0.152 |
uggle | +0.149 |
faced | +0.148 |
Neg logit contributions | |
bars | -0.159 |
groups | -0.152 |
younger | -0.143 |
older | -0.140 |
ents | -0.137 |
Token | the |
Token ID | 262 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
groups | +0.172 |
bars | +0.169 |
fighters | +0.155 |
towns | +0.154 |
department | +0.153 |
Neg logit contributions | |
dd | -0.125 |
night | -0.101 |
ez | -0.089 |
pity | -0.087 |
oo | -0.085 |
Token | pedals |
Token ID | 42563 |
Feature activation | -8.27e-02 |
Pos logit contributions | |
groups | +0.276 |
bars | +0.248 |
lights | +0.243 |
fighters | +0.239 |
department | +0.236 |
Neg logit contributions | |
dd | -0.265 |
night | -0.231 |
pity | -0.219 |
puppy | -0.201 |
wish | -0.200 |
Token | or |
Token ID | 393 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
groups | +0.182 |
towns | +0.164 |
fighters | +0.163 |
department | +0.161 |
bowls | +0.160 |
Neg logit contributions | |
dd | -0.108 |
- | -0.088 |
it | -0.086 |
night | -0.086 |
oo | -0.077 |
Token | the |
Token ID | 262 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
groups | +0.239 |
bars | +0.225 |
fighters | +0.210 |
lights | +0.207 |
towns | +0.207 |
Neg logit contributions | |
dd | -0.211 |
night | -0.178 |
pity | -0.161 |
ez | -0.159 |
faced | -0.150 |
Token | handle |
Token ID | 5412 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
groups | +0.289 |
bars | +0.263 |
lights | +0.254 |
fighters | +0.251 |
department | +0.250 |
Neg logit contributions | |
dd | -0.240 |
night | -0.204 |
pity | -0.196 |
puppy | -0.177 |
ez | -0.176 |
Token | bars |
Token ID | 34046 |
Feature activation | -5.23e-01 |
Pos logit contributions | |
groups | +0.282 |
towns | +0.253 |
fighters | +0.245 |
department | +0.243 |
highlight | +0.240 |
Neg logit contributions | |
dd | -0.148 |
- | -0.131 |
it | -0.130 |
night | -0.122 |
happened | -0.110 |
Token | . |
Token ID | 13 |
Feature activation | -4.85e-02 |
Pos logit contributions | |
groups | +1.052 |
bars | +0.944 |
towns | +0.936 |
fighters | +0.912 |
eight | +0.911 |
Neg logit contributions | |
dd | -0.852 |
night | -0.700 |
ez | -0.627 |
- | -0.621 |
pity | -0.607 |
Token | She |
Token ID | 1375 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
groups | +0.091 |
bars | +0.089 |
towns | +0.079 |
fighters | +0.079 |
younger | +0.079 |
Neg logit contributions | |
dd | -0.085 |
night | -0.071 |
ez | -0.063 |
pity | -0.060 |
amb | -0.059 |
Token | felt |
Token ID | 2936 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
groups | +0.027 |
bars | +0.027 |
towns | +0.023 |
fighters | +0.023 |
eight | +0.023 |
Neg logit contributions | |
dd | -0.028 |
night | -0.024 |
ez | -0.021 |
pity | -0.020 |
amb | -0.020 |
Token | sad |
Token ID | 6507 |
Feature activation | -8.98e-03 |
Pos logit contributions | |
groups | +0.204 |
bars | +0.199 |
older | +0.177 |
younger | +0.176 |
eight | +0.174 |
Neg logit contributions | |
dd | -0.226 |
night | -0.194 |
amb | -0.172 |
ez | -0.171 |
oof | -0.165 |
Token | . |
Token ID | 13 |
Feature activation | -4.83e-02 |
Pos logit contributions | |
groups | +0.021 |
bars | +0.020 |
towns | +0.019 |
fighters | +0.018 |
eight | +0.018 |
Neg logit contributions | |
dd | -0.011 |
night | -0.009 |
- | -0.008 |
ez | -0.007 |
it | -0.007 |
Token | slowly |
Token ID | 6364 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
dd | +0.081 |
night | +0.068 |
ez | +0.063 |
pity | +0.060 |
uggle | +0.059 |
Neg logit contributions | |
bars | -0.075 |
groups | -0.074 |
towns | -0.065 |
younger | -0.064 |
fighters | -0.064 |
Token | pulled |
Token ID | 5954 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
groups | +0.202 |
bars | +0.184 |
fighters | +0.177 |
department | +0.176 |
lights | +0.173 |
Neg logit contributions | |
dd | -0.169 |
night | -0.143 |
ez | -0.123 |
- | -0.118 |
pity | -0.117 |
Token | on |
Token ID | 319 |
Feature activation | +9.04e-02 |
Pos logit contributions | |
bars | +0.046 |
groups | +0.046 |
younger | +0.040 |
older | +0.040 |
towns | +0.040 |
Neg logit contributions | |
dd | -0.044 |
night | -0.037 |
ez | -0.033 |
amb | -0.031 |
uggle | -0.031 |
Token | the |
Token ID | 262 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
dd | +0.170 |
night | +0.143 |
ez | +0.130 |
pity | +0.125 |
amb | +0.122 |
Neg logit contributions | |
groups | -0.157 |
bars | -0.154 |
towns | -0.136 |
fighters | -0.135 |
younger | -0.134 |
Token | handle |
Token ID | 5412 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
groups | +0.361 |
bars | +0.335 |
fighters | +0.324 |
ents | +0.318 |
department | +0.318 |
Neg logit contributions | |
dd | -0.194 |
night | -0.160 |
pity | -0.133 |
ez | -0.125 |
faced | -0.123 |
Token | bars |
Token ID | 34046 |
Feature activation | -5.22e-01 |
Pos logit contributions | |
groups | +0.189 |
ains | +0.156 |
towns | +0.153 |
eight | +0.152 |
illas | +0.149 |
Neg logit contributions | |
- | -0.226 |
dd | -0.219 |
night | -0.215 |
it | -0.214 |
amb | -0.193 |
Token | . |
Token ID | 13 |
Feature activation | -4.42e-02 |
Pos logit contributions | |
groups | +1.115 |
towns | +0.989 |
eight | +0.980 |
department | +0.970 |
fighters | +0.969 |
Neg logit contributions | |
dd | -0.772 |
night | -0.643 |
- | -0.597 |
amb | -0.567 |
ez | -0.552 |
Token | He |
Token ID | 679 |
Feature activation | -7.48e-02 |
Pos logit contributions | |
groups | +0.077 |
bars | +0.076 |
towns | +0.066 |
younger | +0.066 |
fighters | +0.066 |
Neg logit contributions | |
dd | -0.084 |
night | -0.070 |
ez | -0.064 |
pity | -0.061 |
amb | -0.060 |
Token | slowly |
Token ID | 6364 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
groups | +0.136 |
bars | +0.133 |
towns | +0.118 |
fighters | +0.118 |
younger | +0.117 |
Neg logit contributions | |
dd | -0.135 |
night | -0.113 |
ez | -0.102 |
pity | -0.097 |
amb | -0.096 |
Token | rolled |
Token ID | 11686 |
Feature activation | -4.16e-02 |
Pos logit contributions | |
groups | +0.203 |
bars | +0.185 |
fighters | +0.177 |
lights | +0.176 |
department | +0.175 |
Neg logit contributions | |
dd | -0.171 |
night | -0.145 |
ez | -0.127 |
- | -0.119 |
pity | -0.119 |
Token | down |
Token ID | 866 |
Feature activation | +2.42e-02 |
Pos logit contributions | |
bars | +0.079 |
groups | +0.078 |
towns | +0.069 |
eight | +0.069 |
bridges | +0.069 |
Neg logit contributions | |
dd | -0.073 |
night | -0.061 |
ez | -0.056 |
oof | -0.051 |
pity | -0.051 |
Token | as |
Token ID | 292 |
Feature activation | -3.18e-01 |
Pos logit contributions | |
dd | +0.027 |
night | +0.021 |
ez | +0.017 |
pity | +0.016 |
amb | +0.016 |
Neg logit contributions | |
groups | -0.050 |
bars | -0.049 |
towns | -0.045 |
fighters | -0.045 |
younger | -0.044 |
Token | sels |
Token ID | 14002 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
groups | +0.655 |
bars | +0.622 |
towns | +0.573 |
fighters | +0.573 |
younger | +0.572 |
Neg logit contributions | |
dd | -0.497 |
night | -0.410 |
ez | -0.361 |
amb | -0.335 |
- | -0.334 |
Token | on |
Token ID | 319 |
Feature activation | +9.05e-02 |
Pos logit contributions | |
groups | +0.297 |
bars | +0.276 |
towns | +0.261 |
fighters | +0.259 |
younger | +0.259 |
Neg logit contributions | |
dd | -0.206 |
night | -0.180 |
- | -0.146 |
happened | -0.139 |
it | -0.138 |
Token | the |
Token ID | 262 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
dd | +0.121 |
night | +0.094 |
ez | +0.080 |
pity | +0.076 |
amb | +0.073 |
Neg logit contributions | |
groups | -0.207 |
bars | -0.205 |
fighters | -0.186 |
towns | -0.185 |
younger | -0.184 |
Token | handle |
Token ID | 5412 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
groups | +0.269 |
bars | +0.262 |
fighters | +0.239 |
department | +0.233 |
ents | +0.229 |
Neg logit contributions | |
dd | -0.280 |
night | -0.250 |
pity | -0.213 |
ez | -0.213 |
- | -0.202 |
Token | bars |
Token ID | 34046 |
Feature activation | -5.19e-01 |
Pos logit contributions | |
groups | +0.252 |
towns | +0.218 |
fighters | +0.215 |
eight | +0.215 |
younger | +0.213 |
Neg logit contributions | |
dd | -0.180 |
night | -0.158 |
- | -0.141 |
it | -0.137 |
amb | -0.127 |
Token | and |
Token ID | 290 |
Feature activation | +4.31e-02 |
Pos logit contributions | |
groups | +1.055 |
bars | +0.963 |
towns | +0.926 |
fighters | +0.919 |
eight | +0.913 |
Neg logit contributions | |
dd | -0.836 |
night | -0.702 |
ez | -0.617 |
amb | -0.586 |
- | -0.584 |
Token | a |
Token ID | 257 |
Feature activation | +3.46e-02 |
Pos logit contributions | |
dd | +0.075 |
night | +0.065 |
ez | +0.058 |
pity | +0.057 |
it | +0.054 |
Neg logit contributions | |
groups | -0.077 |
bars | -0.074 |
fighters | -0.067 |
department | -0.067 |
ains | -0.066 |
Token | bell |
Token ID | 8966 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
dd | +0.058 |
night | +0.049 |
ez | +0.043 |
pity | +0.042 |
amb | +0.040 |
Neg logit contributions | |
groups | -0.066 |
bars | -0.065 |
fighters | -0.058 |
towns | -0.058 |
younger | -0.057 |
Token | that |
Token ID | 326 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
dd | +0.181 |
night | +0.155 |
ez | +0.136 |
- | +0.131 |
amb | +0.129 |
Neg logit contributions | |
groups | -0.216 |
bars | -0.196 |
towns | -0.187 |
fighters | -0.183 |
eight | -0.183 |
Token | sounded |
Token ID | 14846 |
Feature activation | -8.95e-02 |
Pos logit contributions | |
dd | +0.191 |
night | +0.161 |
ez | +0.142 |
pity | +0.140 |
uggle | +0.129 |
Neg logit contributions | |
groups | -0.214 |
bars | -0.204 |
ents | -0.185 |
fighters | -0.185 |
eight | -0.184 |
Token | how |
Token ID | 703 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
dd | +0.031 |
night | +0.027 |
ez | +0.023 |
pity | +0.023 |
amb | +0.022 |
Neg logit contributions | |
groups | -0.038 |
bars | -0.037 |
fighters | -0.034 |
towns | -0.033 |
eight | -0.033 |
Token | to |
Token ID | 284 |
Feature activation | -3.51e-02 |
Pos logit contributions | |
groups | +0.287 |
bars | +0.273 |
lights | +0.258 |
fighters | +0.256 |
ains | +0.255 |
Neg logit contributions | |
dd | -0.194 |
night | -0.161 |
- | -0.141 |
it | -0.133 |
pity | -0.130 |
Token | turn |
Token ID | 1210 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
groups | +0.068 |
bars | +0.065 |
department | +0.060 |
lights | +0.060 |
fighters | +0.060 |
Neg logit contributions | |
dd | -0.058 |
night | -0.047 |
ez | -0.042 |
pity | -0.042 |
amb | -0.042 |
Token | the |
Token ID | 262 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
groups | +0.418 |
fighters | +0.391 |
department | +0.391 |
bars | +0.376 |
eight | +0.370 |
Neg logit contributions | |
dd | -0.296 |
night | -0.242 |
it | -0.228 |
- | -0.225 |
pity | -0.215 |
Token | handle |
Token ID | 5412 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
groups | +0.290 |
department | +0.259 |
ill | +0.257 |
lights | +0.255 |
fighters | +0.254 |
Neg logit contributions | |
dd | -0.232 |
night | -0.209 |
pity | -0.197 |
- | -0.185 |
wish | -0.183 |
Token | bars |
Token ID | 34046 |
Feature activation | -5.18e-01 |
Pos logit contributions | |
groups | +0.108 |
bowls | +0.074 |
fighters | +0.071 |
department | +0.070 |
towns | +0.066 |
Neg logit contributions | |
dd | -0.300 |
- | -0.298 |
night | -0.293 |
it | -0.287 |
uggle | -0.267 |
Token | and |
Token ID | 290 |
Feature activation | +4.68e-02 |
Pos logit contributions | |
groups | +1.137 |
department | +1.006 |
eight | +1.005 |
bowls | +1.002 |
fighters | +0.999 |
Neg logit contributions | |
dd | -0.698 |
night | -0.614 |
- | -0.561 |
it | -0.527 |
ez | -0.511 |
Token | keep |
Token ID | 1394 |
Feature activation | -7.51e-03 |
Pos logit contributions | |
dd | +0.071 |
night | +0.062 |
it | +0.057 |
pity | +0.057 |
ez | +0.053 |
Neg logit contributions | |
groups | -0.091 |
department | -0.080 |
ains | -0.079 |
bars | -0.079 |
fighters | -0.079 |
Token | her |
Token ID | 607 |
Feature activation | +9.62e-02 |
Pos logit contributions | |
bars | +0.015 |
groups | +0.014 |
younger | +0.013 |
older | +0.013 |
towns | +0.013 |
Neg logit contributions | |
dd | -0.013 |
night | -0.011 |
ez | -0.009 |
faced | -0.009 |
amb | -0.009 |
Token | balance |
Token ID | 5236 |
Feature activation | -6.93e-02 |
Pos logit contributions | |
dd | +0.153 |
night | +0.136 |
pity | +0.122 |
- | +0.117 |
ez | +0.115 |
Neg logit contributions | |
groups | -0.183 |
department | -0.167 |
bars | -0.167 |
fighters | -0.163 |
ill | -0.159 |
Token | . |
Token ID | 13 |
Feature activation | -4.66e-02 |
Pos logit contributions | |
groups | +0.141 |
department | +0.133 |
fighters | +0.131 |
bars | +0.131 |
eight | +0.130 |
Neg logit contributions | |
dd | -0.106 |
night | -0.085 |
- | -0.080 |
pity | -0.075 |
ez | -0.075 |
Token | blue |
Token ID | 4171 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
dd | +0.115 |
night | +0.093 |
ez | +0.082 |
pity | +0.079 |
amb | +0.076 |
Neg logit contributions | |
groups | -0.157 |
bars | -0.153 |
fighters | -0.140 |
towns | -0.138 |
department | -0.138 |
Token | wheels |
Token ID | 13666 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
bars | +0.293 |
groups | +0.290 |
towns | +0.249 |
eight | +0.247 |
ents | +0.245 |
Neg logit contributions | |
dd | -0.334 |
night | -0.279 |
ez | -0.256 |
pity | -0.245 |
faced | -0.241 |
Token | and |
Token ID | 290 |
Feature activation | +4.74e-02 |
Pos logit contributions | |
groups | +0.313 |
department | +0.283 |
towns | +0.281 |
fighters | +0.279 |
narrower | +0.275 |
Neg logit contributions | |
dd | -0.139 |
- | -0.111 |
night | -0.108 |
amb | -0.101 |
it | -0.097 |
Token | pink |
Token ID | 11398 |
Feature activation | -4.99e-02 |
Pos logit contributions | |
dd | +0.081 |
night | +0.066 |
ez | +0.059 |
pity | +0.058 |
amb | +0.056 |
Neg logit contributions | |
groups | -0.090 |
bars | -0.087 |
fighters | -0.079 |
department | -0.078 |
towns | -0.078 |
Token | handle |
Token ID | 5412 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
groups | +0.082 |
bars | +0.081 |
towns | +0.070 |
fighters | +0.070 |
younger | +0.069 |
Neg logit contributions | |
dd | -0.099 |
night | -0.084 |
ez | -0.077 |
pity | -0.074 |
amb | -0.072 |
Token | bars |
Token ID | 34046 |
Feature activation | -5.18e-01 |
Pos logit contributions | |
groups | +0.193 |
eight | +0.164 |
department | +0.163 |
fighters | +0.163 |
towns | +0.160 |
Neg logit contributions | |
dd | -0.215 |
- | -0.191 |
night | -0.189 |
it | -0.182 |
amb | -0.175 |
Token | . |
Token ID | 13 |
Feature activation | -4.47e-02 |
Pos logit contributions | |
groups | +0.934 |
bars | +0.887 |
towns | +0.809 |
fighters | +0.805 |
eight | +0.800 |
Neg logit contributions | |
dd | -0.940 |
night | -0.795 |
ez | -0.717 |
pity | -0.692 |
amb | -0.680 |
Token | She |
Token ID | 1375 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
groups | +0.078 |
bars | +0.076 |
fighters | +0.070 |
lights | +0.068 |
department | +0.068 |
Neg logit contributions | |
dd | -0.079 |
night | -0.070 |
ez | -0.061 |
amb | -0.060 |
uggle | -0.059 |
Token | liked |
Token ID | 8288 |
Feature activation | +1.29e-02 |
Pos logit contributions | |
groups | +0.024 |
bars | +0.024 |
towns | +0.022 |
fighters | +0.021 |
younger | +0.021 |
Neg logit contributions | |
dd | -0.020 |
night | -0.017 |
ez | -0.015 |
pity | -0.014 |
amb | -0.014 |
Token | to |
Token ID | 284 |
Feature activation | -3.57e-02 |
Pos logit contributions | |
dd | +0.022 |
night | +0.018 |
ez | +0.016 |
pity | +0.015 |
amb | +0.015 |
Neg logit contributions | |
groups | -0.025 |
bars | -0.024 |
fighters | -0.022 |
department | -0.021 |
lights | -0.021 |
Token | go |
Token ID | 467 |
Feature activation | -6.78e-03 |
Pos logit contributions | |
bars | +0.069 |
groups | +0.066 |
towns | +0.060 |
younger | +0.059 |
older | +0.059 |
Neg logit contributions | |
dd | -0.065 |
night | -0.052 |
ez | -0.048 |
amb | -0.043 |
faced | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | -4.78e-02 |
Pos logit contributions | |
dd | +0.062 |
night | +0.055 |
- | +0.052 |
amb | +0.051 |
uggle | +0.048 |
Neg logit contributions | |
groups | -0.092 |
eight | -0.083 |
towns | -0.082 |
lights | -0.081 |
fighters | -0.081 |
Token | He |
Token ID | 679 |
Feature activation | -7.02e-02 |
Pos logit contributions | |
groups | +0.099 |
bars | +0.093 |
lights | +0.086 |
fastest | +0.085 |
fighters | +0.084 |
Neg logit contributions | |
dd | -0.068 |
night | -0.062 |
Now | -0.055 |
amb | -0.051 |
It | -0.051 |
Token | grabbed |
Token ID | 13176 |
Feature activation | -4.06e-03 |
Pos logit contributions | |
groups | +0.132 |
bars | +0.131 |
towns | +0.115 |
fighters | +0.115 |
younger | +0.114 |
Neg logit contributions | |
dd | -0.122 |
night | -0.101 |
ez | -0.091 |
pity | -0.087 |
amb | -0.085 |
Token | the |
Token ID | 262 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
bars | +0.006 |
groups | +0.006 |
eight | +0.006 |
older | +0.005 |
younger | +0.005 |
Neg logit contributions | |
dd | -0.008 |
night | -0.007 |
ez | -0.007 |
uggle | -0.006 |
amb | -0.006 |
Token | handle |
Token ID | 5412 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
groups | +0.333 |
bars | +0.307 |
fighters | +0.299 |
department | +0.296 |
lights | +0.296 |
Neg logit contributions | |
dd | -0.214 |
night | -0.174 |
pity | -0.154 |
ez | -0.146 |
amb | -0.141 |
Token | bars |
Token ID | 34046 |
Feature activation | -5.17e-01 |
Pos logit contributions | |
groups | +0.207 |
ains | +0.173 |
towns | +0.171 |
department | +0.169 |
fighters | +0.168 |
Neg logit contributions | |
dd | -0.209 |
- | -0.200 |
night | -0.197 |
amb | -0.190 |
it | -0.180 |
Token | with |
Token ID | 351 |
Feature activation | +7.37e-02 |
Pos logit contributions | |
groups | +1.209 |
towns | +1.081 |
bars | +1.070 |
fighters | +1.069 |
department | +1.063 |
Neg logit contributions | |
dd | -0.662 |
night | -0.530 |
- | -0.464 |
amb | -0.462 |
ez | -0.439 |
Token | his |
Token ID | 465 |
Feature activation | -9.26e-03 |
Pos logit contributions | |
dd | +0.125 |
night | +0.103 |
ez | +0.093 |
pity | +0.090 |
amb | +0.087 |
Neg logit contributions | |
groups | -0.141 |
bars | -0.138 |
fighters | -0.125 |
towns | -0.123 |
department | -0.123 |
Token | little |
Token ID | 1310 |
Feature activation | -9.03e-02 |
Pos logit contributions | |
groups | +0.018 |
bars | +0.017 |
fighters | +0.016 |
department | +0.016 |
towns | +0.016 |
Neg logit contributions | |
dd | -0.014 |
night | -0.012 |
pity | -0.011 |
ez | -0.011 |
- | -0.010 |
Token | hands |
Token ID | 2832 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
groups | +0.149 |
bars | +0.142 |
fighters | +0.130 |
lights | +0.130 |
ains | +0.128 |
Neg logit contributions | |
dd | -0.167 |
night | -0.149 |
pity | -0.133 |
ez | -0.133 |
amb | -0.128 |
Token | . |
Token ID | 13 |
Feature activation | -4.95e-02 |
Pos logit contributions | |
groups | +0.247 |
towns | +0.226 |
bars | +0.219 |
lights | +0.218 |
ains | +0.218 |
Neg logit contributions | |
dd | -0.132 |
night | -0.106 |
- | -0.094 |
amb | -0.089 |
it | -0.088 |
Token | . |
Token ID | 13 |
Feature activation | -4.90e-02 |
Pos logit contributions | |
groups | +0.142 |
ains | +0.125 |
towns | +0.124 |
bars | +0.124 |
eight | +0.124 |
Neg logit contributions | |
dd | -0.083 |
night | -0.080 |
- | -0.077 |
amb | -0.065 |
faced | -0.065 |
Token | He |
Token ID | 679 |
Feature activation | -7.32e-02 |
Pos logit contributions | |
groups | +0.104 |
bars | +0.102 |
fighters | +0.094 |
lights | +0.093 |
fastest | +0.092 |
Neg logit contributions | |
dd | -0.069 |
night | -0.060 |
ez | -0.050 |
amb | -0.048 |
Now | -0.048 |
Token | grabbed |
Token ID | 13176 |
Feature activation | -5.44e-03 |
Pos logit contributions | |
groups | +0.132 |
bars | +0.130 |
towns | +0.115 |
fighters | +0.114 |
younger | +0.114 |
Neg logit contributions | |
dd | -0.132 |
night | -0.111 |
ez | -0.100 |
pity | -0.096 |
amb | -0.095 |
Token | the |
Token ID | 262 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
bars | +0.008 |
groups | +0.008 |
eight | +0.008 |
older | +0.007 |
younger | +0.007 |
Neg logit contributions | |
dd | -0.011 |
night | -0.010 |
amb | -0.009 |
uggle | -0.009 |
ake | -0.009 |
Token | handle |
Token ID | 5412 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
groups | +0.307 |
bars | +0.304 |
towns | +0.271 |
fighters | +0.270 |
younger | +0.268 |
Neg logit contributions | |
dd | -0.254 |
night | -0.208 |
ez | -0.186 |
pity | -0.176 |
amb | -0.173 |
Token | bars |
Token ID | 34046 |
Feature activation | -5.17e-01 |
Pos logit contributions | |
groups | +0.249 |
ains | +0.219 |
fighters | +0.214 |
towns | +0.214 |
department | +0.212 |
Neg logit contributions | |
dd | -0.165 |
- | -0.155 |
night | -0.151 |
amb | -0.140 |
it | -0.139 |
Token | and |
Token ID | 290 |
Feature activation | +4.24e-02 |
Pos logit contributions | |
groups | +1.223 |
towns | +1.100 |
fighters | +1.092 |
department | +1.088 |
ains | +1.078 |
Neg logit contributions | |
dd | -0.645 |
night | -0.509 |
- | -0.455 |
amb | -0.445 |
ez | -0.423 |
Token | he |
Token ID | 339 |
Feature activation | -5.17e-02 |
Pos logit contributions | |
dd | +0.080 |
night | +0.067 |
ez | +0.062 |
pity | +0.059 |
amb | +0.058 |
Neg logit contributions | |
bars | -0.073 |
groups | -0.073 |
towns | -0.063 |
younger | -0.063 |
eight | -0.063 |
Token | quickly |
Token ID | 2952 |
Feature activation | -3.34e-02 |
Pos logit contributions | |
groups | +0.095 |
bars | +0.092 |
towns | +0.082 |
fighters | +0.082 |
younger | +0.081 |
Neg logit contributions | |
dd | -0.092 |
night | -0.077 |
ez | -0.070 |
pity | -0.067 |
amb | -0.065 |
Token | pushed |
Token ID | 7121 |
Feature activation | -6.56e-02 |
Pos logit contributions | |
groups | +0.064 |
bars | +0.061 |
fighters | +0.056 |
towns | +0.056 |
lights | +0.055 |
Neg logit contributions | |
dd | -0.058 |
night | -0.048 |
ez | -0.043 |
pity | -0.041 |
amb | -0.039 |
Token | on |
Token ID | 319 |
Feature activation | +9.42e-02 |
Pos logit contributions | |
groups | +0.136 |
bars | +0.136 |
eight | +0.122 |
towns | +0.122 |
younger | +0.121 |
Neg logit contributions | |
dd | -0.102 |
night | -0.084 |
ez | -0.072 |
uggle | -0.069 |
amb | -0.068 |
Token | twisted |
Token ID | 19074 |
Feature activation | -2.90e-02 |
Pos logit contributions | |
groups | +0.332 |
bars | +0.326 |
towns | +0.306 |
ains | +0.296 |
ill | +0.288 |
Neg logit contributions | |
dd | -0.307 |
pity | -0.233 |
night | -0.226 |
ez | -0.224 |
it | -0.222 |
Token | , |
Token ID | 11 |
Feature activation | -7.41e-03 |
Pos logit contributions | |
groups | +0.067 |
bars | +0.064 |
fighters | +0.060 |
towns | +0.060 |
eight | +0.060 |
Neg logit contributions | |
dd | -0.037 |
night | -0.029 |
- | -0.025 |
ez | -0.025 |
pity | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | +4.66e-02 |
Pos logit contributions | |
groups | +0.014 |
bars | +0.014 |
towns | +0.012 |
fighters | +0.012 |
younger | +0.012 |
Neg logit contributions | |
dd | -0.012 |
night | -0.010 |
ez | -0.009 |
pity | -0.009 |
it | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
dd | +0.062 |
it | +0.054 |
night | +0.044 |
happened | +0.042 |
pity | +0.042 |
Neg logit contributions | |
bars | -0.105 |
groups | -0.104 |
fastest | -0.092 |
ains | -0.091 |
highlight | -0.091 |
Token | handle |
Token ID | 5412 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
groups | +0.303 |
bars | +0.295 |
fighters | +0.265 |
towns | +0.264 |
eight | +0.262 |
Neg logit contributions | |
dd | -0.249 |
night | -0.205 |
ez | -0.182 |
pity | -0.176 |
amb | -0.168 |
Token | bars |
Token ID | 34046 |
Feature activation | -5.16e-01 |
Pos logit contributions | |
groups | +0.240 |
towns | +0.211 |
bars | +0.210 |
eight | +0.206 |
younger | +0.205 |
Neg logit contributions | |
dd | -0.192 |
night | -0.154 |
ez | -0.142 |
- | -0.139 |
pity | -0.132 |
Token | were |
Token ID | 547 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
groups | +1.032 |
bars | +0.954 |
towns | +0.913 |
eight | +0.893 |
fighters | +0.888 |
Neg logit contributions | |
dd | -0.848 |
night | -0.695 |
ez | -0.636 |
pity | -0.608 |
- | -0.594 |
Token | bent |
Token ID | 17157 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
groups | +0.343 |
towns | +0.320 |
bars | +0.319 |
ains | +0.311 |
oes | +0.295 |
Neg logit contributions | |
dd | -0.296 |
it | -0.227 |
pity | -0.225 |
ez | -0.218 |
- | -0.214 |
Token | . |
Token ID | 13 |
Feature activation | -4.76e-02 |
Pos logit contributions | |
dd | +0.015 |
night | +0.013 |
- | +0.012 |
it | +0.011 |
amb | +0.010 |
Neg logit contributions | |
groups | -0.027 |
bars | -0.025 |
eight | -0.024 |
towns | -0.024 |
bowls | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | +3.37e-02 |
Pos logit contributions | |
groups | +0.096 |
bars | +0.096 |
towns | +0.084 |
fastest | +0.083 |
younger | +0.083 |
Neg logit contributions | |
dd | -0.075 |
night | -0.062 |
ez | -0.055 |
pity | -0.052 |
amb | -0.051 |
Token |
|
Token ID | 198 |
Feature activation | +3.28e-02 |
Pos logit contributions | |
dd | +0.058 |
night | +0.050 |
ez | +0.044 |
pity | +0.042 |
amb | +0.041 |
Neg logit contributions | |
groups | -0.063 |
bars | -0.060 |
eight | -0.056 |
towns | -0.055 |
fighters | -0.055 |
Token | how |
Token ID | 703 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
dd | +0.181 |
night | +0.158 |
ez | +0.142 |
pity | +0.139 |
uggle | +0.134 |
Neg logit contributions | |
groups | -0.169 |
bars | -0.164 |
fighters | -0.147 |
towns | -0.144 |
eight | -0.144 |
Token | to |
Token ID | 284 |
Feature activation | -3.33e-02 |
Pos logit contributions | |
groups | +0.274 |
bars | +0.254 |
fighters | +0.244 |
department | +0.243 |
lights | +0.241 |
Neg logit contributions | |
dd | -0.207 |
night | -0.168 |
- | -0.144 |
it | -0.140 |
pity | -0.139 |
Token | turn |
Token ID | 1210 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
groups | +0.063 |
bars | +0.061 |
fighters | +0.055 |
department | +0.055 |
towns | +0.054 |
Neg logit contributions | |
dd | -0.057 |
night | -0.047 |
ez | -0.042 |
pity | -0.042 |
amb | -0.040 |
Token | the |
Token ID | 262 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
groups | +0.416 |
fighters | +0.390 |
department | +0.389 |
bars | +0.385 |
eight | +0.372 |
Neg logit contributions | |
dd | -0.295 |
night | -0.237 |
it | -0.216 |
- | -0.212 |
ez | -0.210 |
Token | handle |
Token ID | 5412 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
groups | +0.298 |
department | +0.267 |
ill | +0.266 |
fighters | +0.266 |
lights | +0.263 |
Neg logit contributions | |
dd | -0.223 |
night | -0.194 |
pity | -0.182 |
- | -0.165 |
news | -0.165 |
Token | bars |
Token ID | 34046 |
Feature activation | -5.15e-01 |
Pos logit contributions | |
groups | +0.146 |
bowls | +0.111 |
fighters | +0.110 |
department | +0.107 |
towns | +0.107 |
Neg logit contributions | |
dd | -0.255 |
- | -0.249 |
night | -0.243 |
it | -0.241 |
uggle | -0.218 |
Token | . |
Token ID | 13 |
Feature activation | -4.30e-02 |
Pos logit contributions | |
groups | +1.137 |
department | +1.011 |
towns | +1.007 |
eight | +1.006 |
fighters | +1.002 |
Neg logit contributions | |
dd | -0.705 |
night | -0.601 |
- | -0.560 |
it | -0.533 |
ez | -0.501 |
Token | |
Token ID | 220 |
Feature activation | +9.74e-04 |
Pos logit contributions | |
groups | +0.081 |
bars | +0.080 |
towns | +0.071 |
younger | +0.071 |
fighters | +0.070 |
Neg logit contributions | |
dd | -0.077 |
night | -0.062 |
ez | -0.057 |
pity | -0.054 |
amb | -0.053 |
Token |
|
Token ID | 198 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
dd | +0.001 |
night | +0.001 |
ez | +0.001 |
uggle | +0.001 |
pity | +0.001 |
Neg logit contributions | |
groups | -0.002 |
bars | -0.002 |
fighters | -0.002 |
towns | -0.002 |
eight | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | +3.66e-02 |
Pos logit contributions | |
dd | +0.070 |
night | +0.061 |
ez | +0.054 |
pity | +0.052 |
amb | +0.050 |
Neg logit contributions | |
groups | -0.066 |
bars | -0.063 |
towns | -0.057 |
eight | -0.057 |
younger | -0.056 |
Token | Finally |
Token ID | 11158 |
Feature activation | -7.86e-02 |
Pos logit contributions | |
dd | +0.076 |
night | +0.065 |
ez | +0.060 |
pity | +0.058 |
amb | +0.057 |
Neg logit contributions | |
groups | -0.056 |
bars | -0.055 |
fighters | -0.048 |
younger | -0.047 |
towns | -0.047 |
Token | , |
Token ID | 11 |
Feature activation | -7.97e-03 |
Pos logit contributions | |
groups | +0.184 |
fighters | +0.166 |
towns | +0.165 |
department | +0.162 |
ains | +0.162 |
Neg logit contributions | |
dd | -0.104 |
night | -0.086 |
- | -0.077 |
amb | -0.073 |
it | -0.071 |
Token | held |
Token ID | 2714 |
Feature activation | +1.71e-03 |
Pos logit contributions | |
groups | +0.017 |
bars | +0.015 |
fighters | +0.015 |
department | +0.015 |
lights | +0.015 |
Neg logit contributions | |
dd | -0.012 |
night | -0.009 |
ez | -0.009 |
it | -0.008 |
pity | -0.008 |
Token | onto |
Token ID | 4291 |
Feature activation | -3.32e-02 |
Pos logit contributions | |
dd | +0.003 |
night | +0.003 |
ez | +0.002 |
faced | +0.002 |
pity | +0.002 |
Neg logit contributions | |
bars | -0.003 |
groups | -0.003 |
eight | -0.003 |
older | -0.003 |
younger | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
groups | +0.057 |
bars | +0.057 |
eight | +0.050 |
towns | +0.050 |
younger | +0.049 |
Neg logit contributions | |
dd | -0.063 |
night | -0.053 |
ez | -0.048 |
pity | -0.046 |
amb | -0.045 |
Token | handle |
Token ID | 5412 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
groups | +0.276 |
bars | +0.255 |
lights | +0.242 |
fighters | +0.237 |
department | +0.233 |
Neg logit contributions | |
dd | -0.261 |
night | -0.222 |
pity | -0.201 |
ez | -0.200 |
faced | -0.188 |
Token | bars |
Token ID | 34046 |
Feature activation | -5.15e-01 |
Pos logit contributions | |
groups | +0.182 |
fighters | +0.159 |
department | +0.156 |
towns | +0.156 |
ains | +0.155 |
Neg logit contributions | |
- | -0.222 |
dd | -0.208 |
it | -0.205 |
night | -0.201 |
amb | -0.199 |
Token | carefully |
Token ID | 7773 |
Feature activation | -3.20e-02 |
Pos logit contributions | |
groups | +1.139 |
fighters | +1.033 |
towns | +1.029 |
department | +1.027 |
ains | +0.996 |
Neg logit contributions | |
dd | -0.726 |
night | -0.588 |
- | -0.561 |
amb | -0.547 |
it | -0.501 |
Token | and |
Token ID | 290 |
Feature activation | +4.68e-02 |
Pos logit contributions | |
groups | +0.069 |
department | +0.067 |
fighters | +0.067 |
ains | +0.065 |
towns | +0.063 |
Neg logit contributions | |
dd | -0.038 |
- | -0.032 |
night | -0.032 |
it | -0.030 |
ez | -0.027 |
Token | slowly |
Token ID | 6364 |
Feature activation | -9.66e-02 |
Pos logit contributions | |
dd | +0.084 |
night | +0.071 |
ez | +0.064 |
pity | +0.061 |
amb | +0.060 |
Neg logit contributions | |
groups | -0.085 |
bars | -0.083 |
fighters | -0.074 |
towns | -0.074 |
younger | -0.073 |
Token | rolled |
Token ID | 11686 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
groups | +0.192 |
ains | +0.171 |
fighters | +0.171 |
department | +0.170 |
bars | +0.168 |
Neg logit contributions | |
dd | -0.158 |
night | -0.133 |
- | -0.117 |
ez | -0.115 |
pity | -0.109 |
Token | over |
Token ID | 625 |
Feature activation | -6.13e-02 |
Pos logit contributions | |
groups | +0.076 |
bars | +0.075 |
towns | +0.067 |
fighters | +0.066 |
younger | +0.066 |
Neg logit contributions | |
dd | -0.066 |
night | -0.054 |
ez | -0.048 |
pity | -0.046 |
amb | -0.045 |
Token | the |
Token ID | 262 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
dd | +0.255 |
night | +0.210 |
pity | +0.191 |
ez | +0.189 |
faced | +0.179 |
Neg logit contributions | |
bars | -0.297 |
groups | -0.295 |
fighters | -0.261 |
lights | -0.261 |
towns | -0.260 |
Token | red |
Token ID | 2266 |
Feature activation | -5.48e-02 |
Pos logit contributions | |
bars | +0.284 |
ents | +0.251 |
lights | +0.247 |
department | +0.231 |
ups | +0.228 |
Neg logit contributions | |
dd | -0.235 |
night | -0.217 |
pity | -0.211 |
name | -0.209 |
news | -0.203 |
Token | flags |
Token ID | 9701 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
bars | +0.094 |
groups | +0.091 |
ents | +0.084 |
younger | +0.084 |
department | +0.082 |
Neg logit contributions | |
dd | -0.094 |
night | -0.084 |
- | -0.073 |
pity | -0.071 |
ez | -0.070 |
Token | . |
Token ID | 13 |
Feature activation | -4.90e-02 |
Pos logit contributions | |
groups | +0.377 |
bars | +0.374 |
towns | +0.354 |
eight | +0.353 |
younger | +0.342 |
Neg logit contributions | |
dd | -0.247 |
night | -0.203 |
it | -0.185 |
- | -0.177 |
happened | -0.175 |
Token | Both |
Token ID | 5747 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
bars | +0.103 |
groups | +0.098 |
fastest | +0.095 |
highlight | +0.091 |
lights | +0.090 |
Neg logit contributions | |
night | -0.060 |
dd | -0.060 |
Now | -0.057 |
It | -0.054 |
amb | -0.048 |
Token | groups |
Token ID | 2628 |
Feature activation | -5.15e-01 |
Pos logit contributions | |
bars | +0.472 |
ric | +0.439 |
lands | +0.416 |
ifying | +0.410 |
oes | +0.403 |
Neg logit contributions | |
dd | -0.266 |
it | -0.259 |
- | -0.242 |
night | -0.236 |
happened | -0.227 |
Token | were |
Token ID | 547 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
bars | +0.932 |
groups | +0.900 |
lights | +0.832 |
eight | +0.811 |
narrower | +0.809 |
Neg logit contributions | |
dd | -0.861 |
night | -0.713 |
happened | -0.637 |
it | -0.636 |
pity | -0.633 |
Token | very |
Token ID | 845 |
Feature activation | +1.29e-02 |
Pos logit contributions | |
bars | +0.409 |
groups | +0.361 |
lights | +0.356 |
oes | +0.347 |
ains | +0.343 |
Neg logit contributions | |
dd | -0.253 |
pity | -0.181 |
night | -0.179 |
ggy | -0.174 |
- | -0.168 |
Token | happy |
Token ID | 3772 |
Feature activation | +8.34e-02 |
Pos logit contributions | |
dd | +0.016 |
night | +0.011 |
pity | +0.010 |
- | +0.010 |
ggy | +0.010 |
Neg logit contributions | |
bars | -0.031 |
groups | -0.028 |
ains | -0.028 |
lights | -0.027 |
department | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | -5.07e-02 |
Pos logit contributions | |
dd | +0.139 |
night | +0.116 |
ez | +0.102 |
pity | +0.097 |
amb | +0.095 |
Neg logit contributions | |
bars | -0.162 |
groups | -0.160 |
eight | -0.141 |
younger | -0.141 |
towns | -0.141 |
Token | They |
Token ID | 1119 |
Feature activation | -7.08e-02 |
Pos logit contributions | |
bars | +0.106 |
groups | +0.102 |
fastest | +0.096 |
narrower | +0.094 |
lights | +0.094 |
Neg logit contributions | |
dd | -0.068 |
night | -0.062 |
Now | -0.061 |
It | -0.055 |
amb | -0.054 |
Token | She |
Token ID | 1375 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
groups | +0.079 |
bars | +0.078 |
younger | +0.068 |
eight | +0.068 |
towns | +0.068 |
Neg logit contributions | |
dd | -0.095 |
night | -0.077 |
ez | -0.075 |
pity | -0.070 |
ggy | -0.067 |
Token | grabbed |
Token ID | 13176 |
Feature activation | +2.74e-03 |
Pos logit contributions | |
bars | +0.022 |
groups | +0.022 |
towns | +0.019 |
eight | +0.019 |
younger | +0.019 |
Neg logit contributions | |
dd | -0.022 |
night | -0.018 |
ez | -0.017 |
pity | -0.016 |
uggle | -0.016 |
Token | onto |
Token ID | 4291 |
Feature activation | -2.84e-02 |
Pos logit contributions | |
dd | +0.006 |
night | +0.005 |
ager | +0.005 |
ake | +0.005 |
coincidence | +0.005 |
Neg logit contributions | |
bars | -0.004 |
groups | -0.004 |
eight | -0.004 |
older | -0.003 |
first | -0.003 |
Token | its |
Token ID | 663 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
bars | +0.041 |
groups | +0.040 |
eight | +0.037 |
older | +0.036 |
younger | +0.036 |
Neg logit contributions | |
dd | -0.063 |
night | -0.054 |
ez | -0.050 |
amb | -0.049 |
faced | -0.047 |
Token | handle |
Token ID | 5412 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
dd | +0.203 |
night | +0.177 |
amb | +0.152 |
pity | +0.151 |
ez | +0.151 |
Neg logit contributions | |
groups | -0.219 |
bars | -0.209 |
fighters | -0.190 |
towns | -0.190 |
department | -0.187 |
Token | bars |
Token ID | 34046 |
Feature activation | -5.15e-01 |
Pos logit contributions | |
groups | +0.310 |
fighters | +0.282 |
towns | +0.281 |
department | +0.280 |
ains | +0.278 |
Neg logit contributions | |
dd | -0.085 |
- | -0.083 |
night | -0.076 |
amb | -0.074 |
it | -0.062 |
Token | and |
Token ID | 290 |
Feature activation | +4.85e-02 |
Pos logit contributions | |
groups | +1.163 |
bars | +1.072 |
towns | +1.042 |
fighters | +1.041 |
department | +1.027 |
Neg logit contributions | |
dd | -0.700 |
night | -0.564 |
amb | -0.483 |
ez | -0.478 |
- | -0.460 |
Token | put |
Token ID | 1234 |
Feature activation | +1.71e-04 |
Pos logit contributions | |
dd | +0.094 |
night | +0.079 |
ez | +0.074 |
pity | +0.070 |
uggle | +0.069 |
Neg logit contributions | |
bars | -0.082 |
groups | -0.081 |
towns | -0.070 |
younger | -0.069 |
eight | -0.069 |
Token | her |
Token ID | 607 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
dd | +0.000 |
night | +0.000 |
ez | +0.000 |
faced | +0.000 |
ake | +0.000 |
Neg logit contributions | |
bars | -0.000 |
groups | -0.000 |
bridges | -0.000 |
eight | -0.000 |
older | -0.000 |
Token | feet |
Token ID | 3625 |
Feature activation | -2.26e-01 |
Pos logit contributions | |
dd | +0.169 |
night | +0.149 |
amb | +0.129 |
pity | +0.129 |
ez | +0.128 |
Neg logit contributions | |
groups | -0.197 |
bars | -0.175 |
fighters | -0.172 |
department | -0.171 |
towns | -0.169 |
Token | onto |
Token ID | 4291 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
groups | +0.547 |
fighters | +0.500 |
towns | +0.496 |
bars | +0.495 |
department | +0.492 |
Neg logit contributions | |
dd | -0.263 |
night | -0.196 |
- | -0.169 |
amb | -0.161 |
ez | -0.153 |
Token | let |
Token ID | 1309 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
dd | +0.087 |
night | +0.072 |
ez | +0.067 |
amb | +0.064 |
uggle | +0.064 |
Neg logit contributions | |
groups | -0.079 |
bars | -0.079 |
towns | -0.071 |
eight | -0.069 |
younger | -0.069 |
Token | go |
Token ID | 467 |
Feature activation | -6.01e-03 |
Pos logit contributions | |
groups | +0.194 |
bars | +0.183 |
younger | +0.182 |
older | +0.180 |
towns | +0.178 |
Neg logit contributions | |
dd | -0.275 |
ez | -0.226 |
night | -0.221 |
amb | -0.217 |
oof | -0.211 |
Token | of |
Token ID | 286 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
groups | +0.010 |
bars | +0.010 |
towns | +0.009 |
older | +0.009 |
younger | +0.009 |
Neg logit contributions | |
dd | -0.012 |
night | -0.010 |
ez | -0.009 |
pity | -0.009 |
faced | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
bars | +0.298 |
groups | +0.292 |
fighters | +0.260 |
department | +0.258 |
eight | +0.256 |
Neg logit contributions | |
dd | -0.289 |
night | -0.240 |
ez | -0.216 |
pity | -0.216 |
uggle | -0.201 |
Token | handle |
Token ID | 5412 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
groups | +0.250 |
bars | +0.245 |
fighters | +0.214 |
towns | +0.212 |
lights | +0.211 |
Neg logit contributions | |
dd | -0.288 |
night | -0.248 |
ez | -0.220 |
pity | -0.217 |
uggle | -0.211 |
Token | bars |
Token ID | 34046 |
Feature activation | -5.14e-01 |
Pos logit contributions | |
groups | +0.230 |
eight | +0.210 |
towns | +0.204 |
ains | +0.201 |
fighters | +0.198 |
Neg logit contributions | |
dd | -0.148 |
- | -0.147 |
night | -0.139 |
it | -0.138 |
amb | -0.125 |
Token | . |
Token ID | 13 |
Feature activation | -4.48e-02 |
Pos logit contributions | |
groups | +1.053 |
eight | +0.930 |
towns | +0.926 |
department | +0.921 |
bars | +0.920 |
Neg logit contributions | |
dd | -0.809 |
night | -0.679 |
- | -0.611 |
amb | -0.611 |
ez | -0.587 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | +8.76e-02 |
Pos logit contributions | |
groups | +0.089 |
bars | +0.088 |
fighters | +0.078 |
towns | +0.077 |
lights | +0.077 |
Neg logit contributions | |
dd | -0.072 |
night | -0.060 |
ez | -0.052 |
amb | -0.051 |
pity | -0.050 |
Token | , |
Token ID | 11 |
Feature activation | -8.21e-03 |
Pos logit contributions | |
dd | +0.131 |
night | +0.104 |
pity | +0.084 |
ez | +0.083 |
- | +0.083 |
Neg logit contributions | |
bars | -0.189 |
groups | -0.188 |
younger | -0.166 |
fighters | -0.166 |
fastest | -0.164 |
Token | Joey |
Token ID | 26154 |
Feature activation | +8.24e-02 |
Pos logit contributions | |
groups | +0.016 |
bars | +0.016 |
younger | +0.014 |
fighters | +0.014 |
towns | +0.014 |
Neg logit contributions | |
dd | -0.014 |
night | -0.012 |
ez | -0.010 |
pity | -0.010 |
amb | -0.010 |
Token | heard |
Token ID | 2982 |
Feature activation | -4.49e-02 |
Pos logit contributions | |
dd | +0.134 |
night | +0.103 |
ez | +0.097 |
- | +0.092 |
happened | +0.087 |
Neg logit contributions | |
groups | -0.168 |
bars | -0.161 |
lights | -0.153 |
fighters | -0.146 |
sizes | -0.145 |
Token | were |
Token ID | 547 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
bars | +0.135 |
lights | +0.127 |
groups | +0.123 |
fastest | +0.122 |
younger | +0.119 |
Neg logit contributions | |
dd | -0.102 |
it | -0.073 |
night | -0.069 |
happened | -0.068 |
wrote | -0.067 |
Token | sorting |
Token ID | 29407 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
bars | +0.353 |
groups | +0.325 |
lights | +0.301 |
department | +0.294 |
fighters | +0.290 |
Neg logit contributions | |
dd | -0.300 |
night | -0.230 |
pity | -0.217 |
ez | -0.213 |
oo | -0.209 |
Token | themselves |
Token ID | 2405 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
bars | +0.182 |
groups | +0.181 |
fighters | +0.160 |
younger | +0.158 |
towns | +0.158 |
Neg logit contributions | |
dd | -0.181 |
night | -0.150 |
ez | -0.136 |
pity | -0.132 |
amb | -0.126 |
Token | into |
Token ID | 656 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
bars | +0.400 |
groups | +0.357 |
fighters | +0.348 |
younger | +0.345 |
department | +0.345 |
Neg logit contributions | |
dd | -0.209 |
night | -0.155 |
- | -0.153 |
it | -0.143 |
inside | -0.138 |
Token | two |
Token ID | 734 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
bars | +0.415 |
groups | +0.351 |
fighters | +0.348 |
department | +0.347 |
fastest | +0.342 |
Neg logit contributions | |
dd | -0.383 |
pity | -0.307 |
night | -0.302 |
ez | -0.300 |
it | -0.295 |
Token | groups |
Token ID | 2628 |
Feature activation | -5.13e-01 |
Pos logit contributions | |
bars | +0.370 |
oes | +0.322 |
department | +0.304 |
fighters | +0.303 |
lights | +0.292 |
Neg logit contributions | |
dd | -0.373 |
- | -0.314 |
pity | -0.292 |
ez | -0.291 |
puppy | -0.287 |
Token | . |
Token ID | 13 |
Feature activation | -4.57e-02 |
Pos logit contributions | |
bars | +1.042 |
groups | +1.014 |
lights | +0.963 |
younger | +0.957 |
eight | +0.954 |
Neg logit contributions | |
dd | -0.701 |
night | -0.543 |
- | -0.522 |
it | -0.477 |
ez | -0.468 |
Token | He |
Token ID | 679 |
Feature activation | -7.22e-02 |
Pos logit contributions | |
groups | +0.097 |
bars | +0.094 |
department | +0.089 |
fighters | +0.088 |
eight | +0.087 |
Neg logit contributions | |
dd | -0.063 |
night | -0.049 |
ez | -0.042 |
amb | -0.041 |
Now | -0.041 |
Token | was |
Token ID | 373 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
bars | +0.152 |
groups | +0.147 |
lights | +0.135 |
younger | +0.132 |
directions | +0.131 |
Neg logit contributions | |
dd | -0.109 |
night | -0.084 |
ez | -0.075 |
pity | -0.071 |
- | -0.070 |
Token | curious |
Token ID | 11040 |
Feature activation | +5.01e-02 |
Pos logit contributions | |
dd | +0.027 |
night | +0.020 |
pity | +0.020 |
ez | +0.019 |
oo | +0.019 |
Neg logit contributions | |
bars | -0.038 |
lights | -0.034 |
groups | -0.033 |
department | -0.031 |
ains | -0.031 |
Token | and |
Token ID | 290 |
Feature activation | +4.21e-02 |
Pos logit contributions | |
dd | +0.068 |
night | +0.059 |
ez | +0.048 |
- | +0.045 |
pity | +0.045 |
Neg logit contributions | |
groups | -0.107 |
bars | -0.107 |
eight | -0.097 |
fighters | -0.096 |
towns | -0.096 |
Token | toys |
Token ID | 14958 |
Feature activation | -4.12e-02 |
Pos logit contributions | |
bars | +0.016 |
groups | +0.015 |
thirteen | +0.014 |
ents | +0.014 |
ill | +0.014 |
Neg logit contributions | |
dd | -0.012 |
night | -0.010 |
- | -0.009 |
pity | -0.009 |
ez | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | +4.70e-02 |
Pos logit contributions | |
groups | +0.093 |
bars | +0.090 |
lights | +0.086 |
younger | +0.086 |
eight | +0.085 |
Neg logit contributions | |
dd | -0.045 |
it | -0.038 |
- | -0.035 |
night | -0.035 |
inside | -0.032 |
Token | put |
Token ID | 1234 |
Feature activation | -8.44e-03 |
Pos logit contributions | |
dd | +0.084 |
night | +0.072 |
pity | +0.065 |
ez | +0.063 |
amb | +0.060 |
Neg logit contributions | |
groups | -0.081 |
bars | -0.080 |
fighters | -0.070 |
lights | -0.070 |
ains | -0.069 |
Token | them |
Token ID | 606 |
Feature activation | +7.14e-03 |
Pos logit contributions | |
groups | +0.015 |
bars | +0.015 |
towns | +0.013 |
fighters | +0.013 |
younger | +0.013 |
Neg logit contributions | |
dd | -0.015 |
night | -0.013 |
ez | -0.011 |
amb | -0.011 |
pity | -0.011 |
Token | in |
Token ID | 287 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
dd | +0.009 |
night | +0.008 |
ez | +0.006 |
- | +0.006 |
pity | +0.006 |
Neg logit contributions | |
bars | -0.016 |
groups | -0.016 |
fighters | -0.014 |
younger | -0.014 |
towns | -0.014 |
Token | groups |
Token ID | 2628 |
Feature activation | -5.12e-01 |
Pos logit contributions | |
dd | +0.020 |
it | +0.017 |
pity | +0.016 |
ez | +0.015 |
night | +0.015 |
Neg logit contributions | |
bars | -0.037 |
department | -0.033 |
fighters | -0.033 |
ents | -0.033 |
ups | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -4.75e-02 |
Pos logit contributions | |
groups | +1.037 |
bars | +1.001 |
lights | +0.974 |
fighters | +0.973 |
narrower | +0.968 |
Neg logit contributions | |
dd | -0.697 |
night | -0.578 |
- | -0.517 |
ez | -0.492 |
it | -0.483 |
Token | He |
Token ID | 679 |
Feature activation | -6.93e-02 |
Pos logit contributions | |
groups | +0.090 |
bars | +0.088 |
towns | +0.079 |
fighters | +0.078 |
lights | +0.077 |
Neg logit contributions | |
dd | -0.080 |
night | -0.069 |
ez | -0.060 |
amb | -0.058 |
pity | -0.058 |
Token | counted |
Token ID | 14789 |
Feature activation | -2.29e-01 |
Pos logit contributions | |
groups | +0.127 |
bars | +0.126 |
lights | +0.111 |
towns | +0.110 |
fighters | +0.108 |
Neg logit contributions | |
dd | -0.123 |
night | -0.102 |
ez | -0.093 |
pity | -0.087 |
amb | -0.086 |
Token | to |
Token ID | 284 |
Feature activation | -3.58e-02 |
Pos logit contributions | |
groups | +0.488 |
bars | +0.477 |
highlight | +0.464 |
forget | +0.457 |
fastest | +0.449 |
Neg logit contributions | |
it | -0.262 |
dd | -0.229 |
- | -0.225 |
him | -0.182 |
inside | -0.181 |
Token | zero |
Token ID | 6632 |
Feature activation | -4.03e-02 |
Pos logit contributions | |
bars | +0.051 |
lights | +0.046 |
groups | +0.041 |
ains | +0.040 |
department | +0.039 |
Neg logit contributions | |
dd | -0.065 |
it | -0.064 |
pity | -0.063 |
- | -0.060 |
night | -0.057 |
Token | bird |
Token ID | 6512 |
Feature activation | +7.92e-02 |
Pos logit contributions | |
groups | +0.152 |
bars | +0.148 |
towns | +0.133 |
older | +0.132 |
younger | +0.132 |
Neg logit contributions | |
dd | -0.143 |
night | -0.117 |
ez | -0.106 |
amb | -0.100 |
uggle | -0.099 |
Token | sat |
Token ID | 3332 |
Feature activation | +7.15e-02 |
Pos logit contributions | |
dd | +0.128 |
night | +0.103 |
ez | +0.096 |
- | +0.089 |
amb | +0.087 |
Neg logit contributions | |
groups | -0.161 |
bars | -0.155 |
fighters | -0.140 |
department | -0.139 |
lights | -0.139 |
Token | on |
Token ID | 319 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
dd | +0.115 |
night | +0.094 |
ez | +0.084 |
pity | +0.080 |
amb | +0.078 |
Neg logit contributions | |
groups | -0.144 |
bars | -0.142 |
towns | -0.127 |
fighters | -0.127 |
younger | -0.126 |
Token | the |
Token ID | 262 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
dd | +0.207 |
night | +0.176 |
ez | +0.161 |
pity | +0.154 |
amb | +0.154 |
Neg logit contributions | |
groups | -0.168 |
bars | -0.163 |
towns | -0.145 |
eight | -0.144 |
fighters | -0.143 |
Token | handle |
Token ID | 5412 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
groups | +0.293 |
department | +0.254 |
fighters | +0.254 |
ill | +0.252 |
lights | +0.246 |
Neg logit contributions | |
dd | -0.221 |
night | -0.187 |
- | -0.169 |
amb | -0.165 |
puppy | -0.164 |
Token | bars |
Token ID | 34046 |
Feature activation | -5.09e-01 |
Pos logit contributions | |
groups | +0.242 |
bowls | +0.207 |
department | +0.207 |
fighters | +0.205 |
forks | +0.200 |
Neg logit contributions | |
dd | -0.144 |
- | -0.138 |
night | -0.137 |
amb | -0.123 |
it | -0.123 |
Token | and |
Token ID | 290 |
Feature activation | +5.74e-02 |
Pos logit contributions | |
groups | +1.171 |
bars | +1.062 |
towns | +1.041 |
fighters | +1.039 |
department | +1.035 |
Neg logit contributions | |
dd | -0.676 |
night | -0.548 |
ez | -0.471 |
amb | -0.466 |
- | -0.459 |
Token | held |
Token ID | 2714 |
Feature activation | +1.89e-03 |
Pos logit contributions | |
dd | +0.108 |
night | +0.090 |
ez | +0.083 |
pity | +0.079 |
amb | +0.078 |
Neg logit contributions | |
groups | -0.100 |
bars | -0.099 |
towns | -0.087 |
fighters | -0.087 |
younger | -0.086 |
Token | on |
Token ID | 319 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
dd | +0.004 |
night | +0.003 |
uggle | +0.003 |
ake | +0.003 |
faced | +0.003 |
Neg logit contributions | |
bars | -0.003 |
groups | -0.003 |
eight | -0.003 |
younger | -0.003 |
older | -0.003 |
Token | tight |
Token ID | 5381 |
Feature activation | -2.22e-03 |
Pos logit contributions | |
dd | +0.164 |
night | +0.137 |
ez | +0.120 |
amb | +0.116 |
pity | +0.116 |
Neg logit contributions | |
groups | -0.204 |
bars | -0.199 |
fighters | -0.179 |
lights | -0.178 |
towns | -0.177 |
Token | . |
Token ID | 13 |
Feature activation | -3.48e-02 |
Pos logit contributions | |
groups | +0.004 |
bars | +0.004 |
fighters | +0.004 |
department | +0.004 |
ains | +0.004 |
Neg logit contributions | |
dd | -0.004 |
night | -0.003 |
ez | -0.003 |
- | -0.003 |
amb | -0.003 |
Token | they |
Token ID | 484 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
dd | +0.111 |
night | +0.092 |
ez | +0.083 |
pity | +0.080 |
amb | +0.078 |
Neg logit contributions | |
groups | -0.119 |
bars | -0.117 |
towns | -0.103 |
fighters | -0.103 |
younger | -0.102 |
Token | liked |
Token ID | 8288 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
groups | +0.178 |
bars | +0.174 |
fighters | +0.151 |
towns | +0.148 |
eight | +0.146 |
Neg logit contributions | |
dd | -0.253 |
night | -0.220 |
ez | -0.200 |
amb | -0.194 |
pity | -0.191 |
Token | to |
Token ID | 284 |
Feature activation | -3.62e-02 |
Pos logit contributions | |
dd | +0.028 |
night | +0.023 |
ez | +0.021 |
pity | +0.020 |
oof | +0.019 |
Neg logit contributions | |
bars | -0.028 |
groups | -0.028 |
towns | -0.025 |
fighters | -0.025 |
department | -0.024 |
Token | live |
Token ID | 2107 |
Feature activation | +9.71e-02 |
Pos logit contributions | |
groups | +0.066 |
bars | +0.064 |
eight | +0.057 |
towns | +0.057 |
younger | +0.057 |
Neg logit contributions | |
dd | -0.066 |
night | -0.055 |
ez | -0.049 |
pity | -0.046 |
amb | -0.046 |
Token | in |
Token ID | 287 |
Feature activation | +1.47e-02 |
Pos logit contributions | |
dd | +0.172 |
night | +0.143 |
ez | +0.129 |
pity | +0.123 |
amb | +0.121 |
Neg logit contributions | |
groups | -0.181 |
bars | -0.175 |
towns | -0.158 |
fighters | -0.157 |
younger | -0.155 |
Token | groups |
Token ID | 2628 |
Feature activation | -5.08e-01 |
Pos logit contributions | |
dd | +0.026 |
night | +0.022 |
ez | +0.021 |
pity | +0.020 |
it | +0.019 |
Neg logit contributions | |
bars | -0.027 |
groups | -0.024 |
fighters | -0.023 |
department | -0.023 |
ents | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | +4.59e-02 |
Pos logit contributions | |
bars | +1.068 |
groups | +1.008 |
eight | +0.993 |
narrower | +0.993 |
department | +0.974 |
Neg logit contributions | |
dd | -0.650 |
night | -0.548 |
- | -0.502 |
it | -0.479 |
ez | -0.470 |
Token | talk |
Token ID | 1561 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
dd | +0.078 |
night | +0.066 |
pity | +0.061 |
ez | +0.060 |
it | +0.058 |
Neg logit contributions | |
bars | -0.084 |
groups | -0.079 |
lights | -0.071 |
ains | -0.070 |
department | -0.070 |
Token | to |
Token ID | 284 |
Feature activation | -3.65e-02 |
Pos logit contributions | |
dd | +0.043 |
night | +0.039 |
- | +0.032 |
ez | +0.032 |
pity | +0.030 |
Neg logit contributions | |
bars | -0.059 |
groups | -0.059 |
eight | -0.052 |
younger | -0.052 |
towns | -0.052 |
Token | each |
Token ID | 1123 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
bars | +0.064 |
groups | +0.063 |
fighters | +0.055 |
towns | +0.054 |
department | +0.054 |
Neg logit contributions | |
dd | -0.067 |
night | -0.057 |
ez | -0.052 |
pity | -0.050 |
amb | -0.048 |
Token | other |
Token ID | 584 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
bars | +0.181 |
lights | +0.161 |
groups | +0.158 |
department | +0.150 |
fighters | +0.149 |
Neg logit contributions | |
dd | -0.228 |
night | -0.199 |
ez | -0.189 |
- | -0.184 |
pity | -0.184 |
Token | note |
Token ID | 3465 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
dd | +0.108 |
night | +0.090 |
ez | +0.087 |
amb | +0.082 |
oo | +0.078 |
Neg logit contributions | |
groups | -0.082 |
bars | -0.079 |
younger | -0.073 |
older | -0.072 |
eight | -0.072 |
Token | attached |
Token ID | 7223 |
Feature activation | +4.71e-02 |
Pos logit contributions | |
dd | +0.208 |
night | +0.170 |
ez | +0.154 |
pity | +0.146 |
amb | +0.143 |
Neg logit contributions | |
bars | -0.252 |
groups | -0.251 |
fighters | -0.221 |
older | -0.221 |
younger | -0.220 |
Token | to |
Token ID | 284 |
Feature activation | -2.33e-02 |
Pos logit contributions | |
dd | +0.071 |
night | +0.061 |
ez | +0.052 |
- | +0.051 |
pity | +0.049 |
Neg logit contributions | |
groups | -0.098 |
bars | -0.092 |
towns | -0.087 |
lights | -0.085 |
fighters | -0.084 |
Token | the |
Token ID | 262 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
groups | +0.042 |
bars | +0.042 |
fighters | +0.037 |
towns | +0.037 |
lights | +0.037 |
Neg logit contributions | |
dd | -0.041 |
night | -0.035 |
ez | -0.031 |
pity | -0.031 |
faced | -0.029 |
Token | handle |
Token ID | 5412 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
groups | +0.260 |
bars | +0.250 |
fighters | +0.225 |
lights | +0.223 |
towns | +0.223 |
Neg logit contributions | |
dd | -0.254 |
night | -0.218 |
ez | -0.192 |
pity | -0.191 |
faced | -0.183 |
Token | bars |
Token ID | 34046 |
Feature activation | -5.06e-01 |
Pos logit contributions | |
groups | +0.226 |
towns | +0.203 |
narrower | +0.192 |
eight | +0.191 |
bridges | +0.191 |
Neg logit contributions | |
dd | -0.145 |
night | -0.134 |
- | -0.126 |
it | -0.119 |
amb | -0.110 |
Token | . |
Token ID | 13 |
Feature activation | -3.56e-02 |
Pos logit contributions | |
groups | +0.974 |
bars | +0.891 |
towns | +0.857 |
eight | +0.839 |
fighters | +0.832 |
Neg logit contributions | |
dd | -0.861 |
night | -0.733 |
ez | -0.650 |
amb | -0.626 |
pity | -0.619 |
Token | The |
Token ID | 383 |
Feature activation | -6.49e-02 |
Pos logit contributions | |
groups | +0.073 |
bars | +0.072 |
fighters | +0.065 |
towns | +0.065 |
younger | +0.064 |
Neg logit contributions | |
dd | -0.055 |
night | -0.045 |
ez | -0.039 |
pity | -0.038 |
amb | -0.037 |
Token | note |
Token ID | 3465 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
bars | +0.128 |
groups | +0.124 |
eight | +0.116 |
older | +0.114 |
towns | +0.112 |
Neg logit contributions | |
dd | -0.118 |
night | -0.090 |
ez | -0.089 |
amb | -0.083 |
oo | -0.079 |
Token | said |
Token ID | 531 |
Feature activation | +7.47e-02 |
Pos logit contributions | |
dd | +0.215 |
night | +0.177 |
ez | +0.160 |
pity | +0.151 |
amb | +0.150 |
Neg logit contributions | |
groups | -0.242 |
bars | -0.239 |
towns | -0.212 |
fighters | -0.211 |
younger | -0.210 |
Token | , |
Token ID | 11 |
Feature activation | -1.37e-03 |
Pos logit contributions | |
dd | +0.135 |
night | +0.112 |
ez | +0.102 |
pity | +0.097 |
amb | +0.096 |
Neg logit contributions | |
groups | -0.136 |
bars | -0.133 |
towns | -0.119 |
eight | -0.119 |
fighters | -0.118 |
Token | land |
Token ID | 1956 |
Feature activation | -2.29e-03 |
Pos logit contributions | |
dd | +0.062 |
night | +0.052 |
ez | +0.047 |
pity | +0.045 |
amb | +0.044 |
Neg logit contributions | |
groups | -0.065 |
bars | -0.064 |
towns | -0.056 |
fighters | -0.056 |
younger | -0.056 |
Token | , |
Token ID | 11 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
bars | +0.005 |
groups | +0.005 |
fighters | +0.005 |
department | +0.005 |
younger | +0.005 |
Neg logit contributions | |
dd | -0.003 |
night | -0.002 |
- | -0.002 |
oo | -0.002 |
it | -0.002 |
Token | there |
Token ID | 612 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
dd | +0.016 |
night | +0.012 |
it | +0.012 |
pity | +0.010 |
uggle | +0.010 |
Neg logit contributions | |
bars | -0.022 |
groups | -0.021 |
department | -0.020 |
fighters | -0.020 |
lights | -0.019 |
Token | were |
Token ID | 547 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
groups | +0.022 |
bars | +0.022 |
towns | +0.019 |
fighters | +0.019 |
younger | +0.019 |
Neg logit contributions | |
dd | -0.021 |
night | -0.017 |
ez | -0.016 |
pity | -0.015 |
amb | -0.015 |
Token | two |
Token ID | 734 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
bars | +0.358 |
groups | +0.314 |
ge | +0.302 |
lights | +0.299 |
iful | +0.298 |
Neg logit contributions | |
dd | -0.215 |
it | -0.174 |
news | -0.169 |
night | -0.162 |
ggy | -0.146 |
Token | groups |
Token ID | 2628 |
Feature activation | -5.02e-01 |
Pos logit contributions | |
bars | +0.407 |
oes | +0.344 |
groups | +0.329 |
fighters | +0.326 |
department | +0.325 |
Neg logit contributions | |
dd | -0.294 |
- | -0.251 |
night | -0.216 |
puppy | -0.212 |
news | -0.211 |
Token | of |
Token ID | 286 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
bars | +0.886 |
groups | +0.842 |
younger | +0.769 |
eight | +0.766 |
fighters | +0.765 |
Neg logit contributions | |
dd | -0.873 |
night | -0.751 |
- | -0.662 |
happened | -0.652 |
ez | -0.646 |
Token | animals |
Token ID | 4695 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
bars | +0.233 |
groups | +0.221 |
fighters | +0.192 |
lights | +0.191 |
department | +0.189 |
Neg logit contributions | |
dd | -0.326 |
night | -0.277 |
ez | -0.257 |
pity | -0.252 |
amb | -0.242 |
Token | . |
Token ID | 13 |
Feature activation | -3.57e-02 |
Pos logit contributions | |
bars | +0.204 |
groups | +0.202 |
younger | +0.178 |
towns | +0.177 |
fighters | +0.176 |
Neg logit contributions | |
dd | -0.191 |
night | -0.157 |
ez | -0.143 |
pity | -0.135 |
amb | -0.134 |
Token | They |
Token ID | 1119 |
Feature activation | -5.56e-02 |
Pos logit contributions | |
bars | +0.077 |
groups | +0.072 |
fighters | +0.067 |
department | +0.067 |
fastest | +0.067 |
Neg logit contributions | |
dd | -0.048 |
night | -0.040 |
Now | -0.034 |
It | -0.034 |
ez | -0.034 |
Token | were |
Token ID | 547 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
bars | +0.129 |
groups | +0.115 |
lights | +0.114 |
ents | +0.109 |
younger | +0.109 |
Neg logit contributions | |
dd | -0.076 |
night | -0.056 |
pity | -0.049 |
it | -0.049 |
ez | -0.048 |
Token | They |
Token ID | 1119 |
Feature activation | -5.61e-02 |
Pos logit contributions | |
bars | +0.063 |
groups | +0.061 |
fastest | +0.059 |
fighters | +0.058 |
lights | +0.058 |
Neg logit contributions | |
dd | -0.035 |
night | -0.034 |
It | -0.029 |
Now | -0.029 |
amb | -0.027 |
Token | were |
Token ID | 547 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
bars | +0.110 |
lights | +0.103 |
groups | +0.102 |
fastest | +0.097 |
bridges | +0.094 |
Neg logit contributions | |
dd | -0.091 |
night | -0.067 |
it | -0.065 |
pity | -0.064 |
happened | -0.063 |
Token | all |
Token ID | 477 |
Feature activation | -6.77e-02 |
Pos logit contributions | |
bars | +0.316 |
groups | +0.299 |
lights | +0.274 |
department | +0.266 |
fighters | +0.264 |
Neg logit contributions | |
dd | -0.287 |
night | -0.228 |
pity | -0.210 |
ez | -0.208 |
oo | -0.204 |
Token | organizing |
Token ID | 16924 |
Feature activation | -7.24e-02 |
Pos logit contributions | |
bars | +0.136 |
groups | +0.127 |
lights | +0.122 |
ains | +0.120 |
department | +0.118 |
Neg logit contributions | |
dd | -0.100 |
night | -0.082 |
ez | -0.077 |
pity | -0.074 |
faced | -0.072 |
Token | into |
Token ID | 656 |
Feature activation | -2.14e-01 |
Pos logit contributions | |
bars | +0.131 |
groups | +0.129 |
fighters | +0.116 |
eight | +0.115 |
ents | +0.114 |
Neg logit contributions | |
dd | -0.127 |
night | -0.106 |
ez | -0.095 |
pity | -0.094 |
uggle | -0.089 |
Token | groups |
Token ID | 2628 |
Feature activation | -5.02e-01 |
Pos logit contributions | |
bars | +0.375 |
department | +0.339 |
fighters | +0.334 |
oes | +0.330 |
lights | +0.316 |
Neg logit contributions | |
dd | -0.361 |
pity | -0.299 |
it | -0.292 |
- | -0.285 |
night | -0.283 |
Token | and |
Token ID | 290 |
Feature activation | +5.72e-02 |
Pos logit contributions | |
bars | +1.050 |
eight | +1.011 |
narrower | +1.010 |
groups | +1.009 |
department | +0.987 |
Neg logit contributions | |
dd | -0.599 |
- | -0.483 |
night | -0.477 |
it | -0.425 |
uggle | -0.394 |
Token | chatting |
Token ID | 32046 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
dd | +0.095 |
night | +0.079 |
it | +0.071 |
pity | +0.071 |
ez | +0.068 |
Neg logit contributions | |
bars | -0.104 |
groups | -0.099 |
ains | -0.090 |
lights | -0.090 |
eight | -0.089 |
Token | . |
Token ID | 13 |
Feature activation | -3.20e-02 |
Pos logit contributions | |
groups | +0.204 |
bars | +0.199 |
towns | +0.179 |
fighters | +0.177 |
younger | +0.177 |
Neg logit contributions | |
dd | -0.173 |
night | -0.141 |
ez | -0.127 |
pity | -0.120 |
amb | -0.119 |
Token |
|
Token ID | 198 |
Feature activation | +4.40e-02 |
Pos logit contributions | |
bars | +0.065 |
groups | +0.063 |
fighters | +0.059 |
lights | +0.058 |
fastest | +0.058 |
Neg logit contributions | |
dd | -0.046 |
night | -0.040 |
Now | -0.034 |
amb | -0.034 |
ez | -0.033 |
Token |
|
Token ID | 198 |
Feature activation | +4.32e-02 |
Pos logit contributions | |
dd | +0.074 |
night | +0.062 |
ez | +0.055 |
pity | +0.052 |
amb | +0.051 |
Neg logit contributions | |
groups | -0.086 |
bars | -0.083 |
towns | -0.075 |
eight | -0.075 |
younger | -0.074 |
Token | castle |
Token ID | 18676 |
Feature activation | -9.08e-02 |
Pos logit contributions | |
dd | +0.082 |
night | +0.071 |
- | +0.066 |
amb | +0.061 |
it | +0.059 |
Neg logit contributions | |
groups | -0.119 |
bars | -0.119 |
eight | -0.112 |
younger | -0.111 |
towns | -0.110 |
Token | and |
Token ID | 290 |
Feature activation | +8.85e-02 |
Pos logit contributions | |
groups | +0.197 |
bars | +0.183 |
eight | +0.181 |
towns | +0.179 |
department | +0.177 |
Neg logit contributions | |
dd | -0.121 |
night | -0.102 |
- | -0.091 |
it | -0.084 |
amb | -0.084 |
Token | started |
Token ID | 2067 |
Feature activation | -5.90e-02 |
Pos logit contributions | |
dd | +0.157 |
night | +0.130 |
ez | +0.119 |
pity | +0.112 |
amb | +0.112 |
Neg logit contributions | |
groups | -0.164 |
bars | -0.161 |
towns | -0.143 |
fighters | -0.143 |
younger | -0.142 |
Token | to |
Token ID | 284 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
bars | +0.114 |
groups | +0.113 |
department | +0.100 |
fighters | +0.100 |
lights | +0.100 |
Neg logit contributions | |
dd | -0.095 |
night | -0.078 |
ez | -0.070 |
pity | -0.068 |
- | -0.066 |
Token | eat |
Token ID | 4483 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
dd | +0.022 |
night | +0.019 |
ez | +0.017 |
pity | +0.017 |
faced | +0.016 |
Neg logit contributions | |
bars | -0.021 |
groups | -0.020 |
fighters | -0.018 |
lights | -0.018 |
ents | -0.018 |
Token | it |
Token ID | 340 |
Feature activation | +3.26e-01 |
Pos logit contributions | |
dd | +0.284 |
night | +0.237 |
ez | +0.213 |
pity | +0.206 |
amb | +0.200 |
Neg logit contributions | |
groups | -0.287 |
bars | -0.284 |
towns | -0.251 |
fighters | -0.250 |
younger | -0.248 |
Token | . |
Token ID | 13 |
Feature activation | -9.72e-04 |
Pos logit contributions | |
dd | +0.438 |
night | +0.377 |
- | +0.327 |
it | +0.304 |
ggy | +0.296 |
Neg logit contributions | |
bars | -0.709 |
groups | -0.694 |
towns | -0.663 |
eight | -0.642 |
fighters | -0.638 |
Token | He |
Token ID | 679 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
groups | +0.002 |
bars | +0.002 |
eight | +0.002 |
towns | +0.002 |
younger | +0.002 |
Neg logit contributions | |
dd | -0.002 |
night | -0.001 |
ez | -0.001 |
pity | -0.001 |
amb | -0.001 |
Token | ate |
Token ID | 15063 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
groups | +0.058 |
bars | +0.057 |
towns | +0.051 |
fighters | +0.050 |
younger | +0.050 |
Neg logit contributions | |
dd | -0.051 |
night | -0.042 |
ez | -0.038 |
pity | -0.036 |
faced | -0.035 |
Token | the |
Token ID | 262 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
dd | +0.278 |
night | +0.233 |
ez | +0.216 |
amb | +0.208 |
pity | +0.201 |
Neg logit contributions | |
groups | -0.258 |
bars | -0.247 |
bridges | -0.222 |
eight | -0.222 |
towns | -0.220 |
Token | sand |
Token ID | 6450 |
Feature activation | +5.33e-02 |
Pos logit contributions | |
groups | +0.225 |
bars | +0.221 |
towns | +0.197 |
fighters | +0.196 |
younger | +0.195 |
Neg logit contributions | |
dd | -0.207 |
night | -0.172 |
ez | -0.154 |
pity | -0.147 |
amb | -0.144 |
Token | friend |
Token ID | 1545 |
Feature activation | +2.77e-02 |
Pos logit contributions | |
dd | +0.241 |
night | +0.200 |
ez | +0.190 |
amb | +0.182 |
uggle | +0.181 |
Neg logit contributions | |
groups | -0.154 |
bars | -0.135 |
older | -0.135 |
younger | -0.134 |
fighters | -0.127 |
Token | Sam |
Token ID | 3409 |
Feature activation | -2.65e-02 |
Pos logit contributions | |
dd | +0.039 |
night | +0.033 |
ez | +0.029 |
pity | +0.027 |
ggy | +0.027 |
Neg logit contributions | |
bars | -0.059 |
groups | -0.057 |
lights | -0.053 |
towns | -0.052 |
fighters | -0.051 |
Token | . |
Token ID | 13 |
Feature activation | -4.28e-02 |
Pos logit contributions | |
bars | +0.051 |
groups | +0.050 |
lights | +0.046 |
fighters | +0.045 |
towns | +0.045 |
Neg logit contributions | |
dd | -0.043 |
night | -0.037 |
ez | -0.033 |
pity | -0.030 |
- | -0.030 |
Token | She |
Token ID | 1375 |
Feature activation | -6.96e-03 |
Pos logit contributions | |
bars | +0.083 |
groups | +0.082 |
lights | +0.074 |
fighters | +0.073 |
towns | +0.073 |
Neg logit contributions | |
dd | -0.066 |
night | -0.058 |
amb | -0.049 |
faced | -0.049 |
pity | -0.049 |
Token | put |
Token ID | 1234 |
Feature activation | -1.93e-03 |
Pos logit contributions | |
groups | +0.013 |
bars | +0.012 |
fighters | +0.011 |
younger | +0.011 |
eight | +0.011 |
Neg logit contributions | |
dd | -0.012 |
night | -0.011 |
ez | -0.010 |
uggle | -0.009 |
pity | -0.009 |
Token | it |
Token ID | 340 |
Feature activation | +2.90e-01 |
Pos logit contributions | |
groups | +0.003 |
bars | +0.003 |
eight | +0.003 |
bridges | +0.003 |
forks | +0.003 |
Neg logit contributions | |
dd | -0.004 |
night | -0.003 |
amb | -0.003 |
ager | -0.003 |
des | -0.003 |
Token | in |
Token ID | 287 |
Feature activation | +2.31e-02 |
Pos logit contributions | |
dd | +0.387 |
night | +0.309 |
ez | +0.259 |
pity | +0.241 |
amb | +0.238 |
Neg logit contributions | |
groups | -0.655 |
bars | -0.646 |
fighters | -0.590 |
towns | -0.589 |
younger | -0.580 |
Token | her |
Token ID | 607 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
dd | +0.032 |
night | +0.026 |
ez | +0.023 |
pity | +0.022 |
oof | +0.020 |
Neg logit contributions | |
bars | -0.051 |
groups | -0.049 |
fighters | -0.046 |
towns | -0.045 |
department | -0.045 |
Token | backpack |
Token ID | 22526 |
Feature activation | -8.77e-03 |
Pos logit contributions | |
dd | +0.207 |
night | +0.178 |
ez | +0.162 |
pity | +0.157 |
amb | +0.154 |
Neg logit contributions | |
groups | -0.166 |
bars | -0.163 |
towns | -0.142 |
fighters | -0.142 |
younger | -0.139 |
Token | and |
Token ID | 290 |
Feature activation | +4.98e-02 |
Pos logit contributions | |
groups | +0.020 |
bars | +0.019 |
fighters | +0.019 |
towns | +0.019 |
eight | +0.018 |
Neg logit contributions | |
dd | -0.010 |
night | -0.009 |
- | -0.007 |
ez | -0.007 |
amb | -0.006 |
Token | went |
Token ID | 1816 |
Feature activation | +4.48e-02 |
Pos logit contributions | |
dd | +0.092 |
night | +0.077 |
ez | +0.071 |
pity | +0.067 |
amb | +0.066 |
Neg logit contributions | |
groups | -0.089 |
bars | -0.087 |
towns | -0.077 |
fighters | -0.077 |
younger | -0.076 |
Token | , |
Token ID | 11 |
Feature activation | -7.56e-03 |
Pos logit contributions | |
dd | +0.116 |
night | +0.096 |
ez | +0.087 |
pity | +0.083 |
amb | +0.081 |
Neg logit contributions | |
groups | -0.125 |
bars | -0.123 |
towns | -0.109 |
fighters | -0.109 |
younger | -0.108 |
Token | " |
Token ID | 366 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
groups | +0.014 |
bars | +0.014 |
towns | +0.012 |
fastest | +0.012 |
lights | +0.012 |
Neg logit contributions | |
dd | -0.013 |
night | -0.011 |
oo | -0.010 |
it | -0.010 |
faced | -0.010 |
Token | Let |
Token ID | 5756 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
groups | +0.211 |
bars | +0.209 |
towns | +0.185 |
fighters | +0.184 |
eight | +0.183 |
Neg logit contributions | |
dd | -0.194 |
night | -0.162 |
ez | -0.146 |
pity | -0.140 |
amb | -0.136 |
Token | 's |
Token ID | 338 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
dd | +0.028 |
night | +0.024 |
ez | +0.023 |
amb | +0.022 |
pity | +0.022 |
Neg logit contributions | |
groups | -0.012 |
bars | -0.012 |
older | -0.010 |
younger | -0.010 |
towns | -0.010 |
Token | hold |
Token ID | 1745 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
groups | +0.190 |
bars | +0.184 |
towns | +0.164 |
fighters | +0.163 |
younger | +0.162 |
Neg logit contributions | |
dd | -0.202 |
night | -0.169 |
ez | -0.155 |
pity | -0.146 |
amb | -0.144 |
Token | it |
Token ID | 340 |
Feature activation | +2.87e-01 |
Pos logit contributions | |
groups | +0.040 |
bars | +0.039 |
eight | +0.035 |
towns | +0.034 |
older | +0.034 |
Neg logit contributions | |
dd | -0.043 |
night | -0.037 |
ez | -0.033 |
amb | -0.032 |
pity | -0.032 |
Token | together |
Token ID | 1978 |
Feature activation | +2.90e-02 |
Pos logit contributions | |
dd | +0.412 |
night | +0.372 |
- | +0.315 |
amb | +0.300 |
uggle | +0.290 |
Neg logit contributions | |
groups | -0.588 |
bars | -0.571 |
towns | -0.547 |
fighters | -0.525 |
younger | -0.517 |
Token | ." |
Token ID | 526 |
Feature activation | -9.44e-02 |
Pos logit contributions | |
dd | +0.035 |
night | +0.030 |
it | +0.028 |
- | +0.028 |
oo | +0.023 |
Neg logit contributions | |
bars | -0.065 |
groups | -0.063 |
narrower | -0.060 |
towns | -0.059 |
lights | -0.059 |
Token | |
Token ID | 220 |
Feature activation | -4.35e-03 |
Pos logit contributions | |
groups | +0.187 |
bars | +0.184 |
fighters | +0.166 |
towns | +0.165 |
younger | +0.163 |
Neg logit contributions | |
dd | -0.152 |
night | -0.127 |
ez | -0.110 |
amb | -0.106 |
pity | -0.106 |
Token |
|
Token ID | 198 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
groups | +0.009 |
bars | +0.009 |
towns | +0.008 |
fighters | +0.008 |
eight | +0.008 |
Neg logit contributions | |
dd | -0.007 |
night | -0.005 |
ez | -0.005 |
pity | -0.004 |
amb | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | +3.11e-02 |
Pos logit contributions | |
dd | +0.053 |
night | +0.044 |
ez | +0.040 |
pity | +0.038 |
amb | +0.036 |
Neg logit contributions | |
groups | -0.063 |
bars | -0.061 |
towns | -0.055 |
fighters | -0.054 |
eight | -0.054 |
Token | , |
Token ID | 11 |
Feature activation | +6.75e-03 |
Pos logit contributions | |
bars | +0.081 |
groups | +0.081 |
eight | +0.072 |
ents | +0.072 |
ains | +0.070 |
Neg logit contributions | |
dd | -0.056 |
night | -0.050 |
- | -0.043 |
pity | -0.040 |
happened | -0.040 |
Token | they |
Token ID | 484 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
dd | +0.012 |
night | +0.010 |
it | +0.009 |
ez | +0.009 |
pity | +0.009 |
Neg logit contributions | |
bars | -0.013 |
groups | -0.012 |
department | -0.011 |
fighters | -0.011 |
younger | -0.011 |
Token | remembered |
Token ID | 12086 |
Feature activation | -7.97e-02 |
Pos logit contributions | |
bars | +0.208 |
groups | +0.205 |
lights | +0.184 |
younger | +0.182 |
bridges | +0.180 |
Neg logit contributions | |
dd | -0.192 |
night | -0.159 |
ez | -0.139 |
pity | -0.136 |
amb | -0.134 |
Token | how |
Token ID | 703 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
bars | +0.163 |
groups | +0.157 |
fastest | +0.141 |
lights | +0.140 |
fighters | +0.139 |
Neg logit contributions | |
dd | -0.128 |
night | -0.107 |
ez | -0.091 |
pity | -0.090 |
it | -0.087 |
Token | nice |
Token ID | 3621 |
Feature activation | +7.43e-02 |
Pos logit contributions | |
bars | +0.244 |
groups | +0.242 |
fighters | +0.216 |
towns | +0.214 |
department | +0.213 |
Neg logit contributions | |
dd | -0.190 |
night | -0.154 |
ez | -0.132 |
pity | -0.129 |
faced | -0.125 |
Token | it |
Token ID | 340 |
Feature activation | +3.00e-01 |
Pos logit contributions | |
dd | +0.133 |
night | +0.111 |
ez | +0.099 |
pity | +0.096 |
amb | +0.094 |
Neg logit contributions | |
groups | -0.135 |
bars | -0.133 |
towns | -0.117 |
fighters | -0.117 |
younger | -0.117 |
Token | was |
Token ID | 373 |
Feature activation | +3.36e-02 |
Pos logit contributions | |
dd | +0.436 |
night | +0.359 |
happened | +0.316 |
ez | +0.302 |
pity | +0.299 |
Neg logit contributions | |
bars | -0.634 |
groups | -0.630 |
eight | -0.583 |
ents | -0.567 |
towns | -0.558 |
Token | to |
Token ID | 284 |
Feature activation | -2.24e-02 |
Pos logit contributions | |
dd | +0.053 |
night | +0.044 |
ez | +0.037 |
- | +0.037 |
happened | +0.037 |
Neg logit contributions | |
bars | -0.070 |
groups | -0.067 |
lights | -0.061 |
eight | -0.061 |
towns | -0.060 |
Token | share |
Token ID | 2648 |
Feature activation | +4.67e-02 |
Pos logit contributions | |
groups | +0.040 |
bars | +0.038 |
towns | +0.035 |
eight | +0.034 |
younger | +0.034 |
Neg logit contributions | |
dd | -0.042 |
night | -0.035 |
ez | -0.032 |
amb | -0.030 |
pity | -0.030 |
Token | different |
Token ID | 1180 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
dd | +0.088 |
night | +0.073 |
ez | +0.066 |
amb | +0.065 |
pity | +0.062 |
Neg logit contributions | |
groups | -0.087 |
bars | -0.078 |
towns | -0.073 |
bridges | -0.072 |
directions | -0.071 |
Token | things |
Token ID | 1243 |
Feature activation | +6.36e-02 |
Pos logit contributions | |
bars | +0.162 |
groups | +0.158 |
fighters | +0.139 |
department | +0.137 |
eight | +0.136 |
Neg logit contributions | |
dd | -0.205 |
night | -0.178 |
ez | -0.162 |
pity | -0.156 |
amb | -0.153 |
Token | and |
Token ID | 290 |
Feature activation | +4.83e-02 |
Pos logit contributions | |
dd | +0.088 |
- | +0.076 |
night | +0.073 |
it | +0.068 |
oo | +0.062 |
Neg logit contributions | |
bars | -0.153 |
groups | -0.150 |
forks | -0.143 |
bowls | -0.142 |
lights | -0.133 |
Token | covered |
Token ID | 5017 |
Feature activation | -8.84e-02 |
Pos logit contributions | |
dd | +0.077 |
it | +0.066 |
night | +0.065 |
pity | +0.059 |
happened | +0.058 |
Neg logit contributions | |
groups | -0.093 |
bars | -0.090 |
ill | -0.080 |
ains | -0.079 |
towns | -0.078 |
Token | the |
Token ID | 262 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
bars | +0.172 |
groups | +0.167 |
ents | +0.150 |
younger | +0.149 |
eight | +0.146 |
Neg logit contributions | |
dd | -0.150 |
night | -0.124 |
ez | -0.109 |
- | -0.105 |
pity | -0.103 |
Token | city |
Token ID | 1748 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
bars | +0.248 |
groups | +0.226 |
lights | +0.203 |
ents | +0.203 |
ill | +0.199 |
Neg logit contributions | |
dd | -0.255 |
night | -0.244 |
news | -0.214 |
pity | -0.210 |
wish | -0.207 |
Token | . |
Token ID | 13 |
Feature activation | -4.55e-02 |
Pos logit contributions | |
groups | +0.236 |
bars | +0.225 |
eight | +0.219 |
ains | +0.213 |
fighters | +0.212 |
Neg logit contributions | |
dd | -0.135 |
- | -0.128 |
night | -0.122 |
it | -0.117 |
faced | -0.107 |
Token | It |
Token ID | 632 |
Feature activation | +2.70e-01 |
Pos logit contributions | |
bars | +0.100 |
groups | +0.099 |
fastest | +0.090 |
lights | +0.090 |
fighters | +0.090 |
Neg logit contributions | |
dd | -0.057 |
night | -0.048 |
Now | -0.043 |
It | -0.041 |
ez | -0.041 |
Token | was |
Token ID | 373 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
dd | +0.364 |
night | +0.291 |
ez | +0.258 |
happened | +0.243 |
it | +0.239 |
Neg logit contributions | |
bars | -0.610 |
groups | -0.606 |
forks | -0.540 |
ents | -0.529 |
towns | -0.528 |
Token | so |
Token ID | 523 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
dd | +0.035 |
night | +0.029 |
pity | +0.027 |
ez | +0.026 |
it | +0.026 |
Neg logit contributions | |
bars | -0.047 |
groups | -0.044 |
towns | -0.040 |
ains | -0.039 |
oes | -0.039 |
Token | big |
Token ID | 1263 |
Feature activation | -6.53e-02 |
Pos logit contributions | |
bars | +0.215 |
groups | +0.212 |
fighters | +0.188 |
towns | +0.187 |
ents | +0.184 |
Neg logit contributions | |
dd | -0.174 |
night | -0.144 |
pity | -0.123 |
ez | -0.123 |
faced | -0.116 |
Token | that |
Token ID | 326 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
groups | +0.134 |
bars | +0.132 |
towns | +0.119 |
fighters | +0.119 |
younger | +0.117 |
Neg logit contributions | |
dd | -0.101 |
night | -0.082 |
ez | -0.073 |
pity | -0.069 |
amb | -0.068 |
Token | it |
Token ID | 340 |
Feature activation | +2.89e-01 |
Pos logit contributions | |
dd | +0.143 |
night | +0.108 |
it | +0.105 |
pity | +0.087 |
happened | +0.085 |
Neg logit contributions | |
groups | -0.274 |
bars | -0.270 |
ents | -0.251 |
department | -0.241 |
eight | -0.241 |
Token | had |
Token ID | 550 |
Feature activation | -5.81e-02 |
Pos logit contributions | |
dd | +0.011 |
night | +0.009 |
amb | +0.008 |
ez | +0.008 |
pity | +0.008 |
Neg logit contributions | |
groups | -0.011 |
bars | -0.010 |
eight | -0.009 |
fighters | -0.009 |
towns | -0.009 |
Token | st |
Token ID | 336 |
Feature activation | -4.06e-02 |
Pos logit contributions | |
bars | +0.123 |
groups | +0.119 |
lights | +0.111 |
towns | +0.108 |
fighters | +0.108 |
Neg logit contributions | |
dd | -0.091 |
night | -0.072 |
ez | -0.070 |
pity | -0.066 |
happened | -0.061 |
Token | ung |
Token ID | 2150 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
groups | +0.098 |
bars | +0.097 |
towns | +0.090 |
fighters | +0.089 |
younger | +0.088 |
Neg logit contributions | |
dd | -0.051 |
night | -0.037 |
ez | -0.034 |
oo | -0.033 |
amb | -0.032 |
Token | her |
Token ID | 607 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
dd | +0.163 |
night | +0.130 |
ez | +0.111 |
pity | +0.106 |
amb | +0.098 |
Neg logit contributions | |
groups | -0.267 |
bars | -0.262 |
fighters | -0.240 |
towns | -0.238 |
younger | -0.234 |
Token | . |
Token ID | 13 |
Feature activation | -2.08e-02 |
Pos logit contributions | |
dd | +0.211 |
night | +0.183 |
ez | +0.165 |
pity | +0.158 |
amb | +0.155 |
Neg logit contributions | |
groups | -0.219 |
bars | -0.213 |
towns | -0.195 |
fighters | -0.191 |
department | -0.185 |
Token |
|
Token ID | 198 |
Feature activation | +5.81e-02 |
Pos logit contributions | |
groups | +0.044 |
bars | +0.043 |
towns | +0.039 |
fighters | +0.039 |
younger | +0.039 |
Neg logit contributions | |
dd | -0.031 |
night | -0.025 |
ez | -0.022 |
pity | -0.021 |
amb | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | +5.55e-02 |
Pos logit contributions | |
dd | +0.115 |
night | +0.106 |
ez | +0.090 |
amb | +0.088 |
pity | +0.087 |
Neg logit contributions | |
groups | -0.094 |
bars | -0.087 |
eight | -0.083 |
ents | -0.080 |
towns | -0.078 |
Token | Sam |
Token ID | 16305 |
Feature activation | +2.79e-02 |
Pos logit contributions | |
dd | +0.107 |
night | +0.090 |
ez | +0.082 |
pity | +0.079 |
amb | +0.078 |
Neg logit contributions | |
groups | -0.095 |
bars | -0.094 |
fighters | -0.082 |
towns | -0.082 |
younger | -0.081 |
Token | ran |
Token ID | 4966 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
dd | +0.044 |
night | +0.035 |
ez | +0.034 |
pity | +0.030 |
ggy | +0.030 |
Neg logit contributions | |
groups | -0.058 |
bars | -0.058 |
lights | -0.052 |
fighters | -0.051 |
towns | -0.051 |
Token | to |
Token ID | 284 |
Feature activation | -1.75e-02 |
Pos logit contributions | |
groups | +0.020 |
bars | +0.019 |
eight | +0.018 |
older | +0.018 |
department | +0.018 |
Neg logit contributions | |
dd | -0.033 |
ez | -0.027 |
night | -0.027 |
amb | -0.024 |
oof | -0.024 |
Token | her |
Token ID | 607 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
groups | +0.023 |
older | +0.020 |
bars | +0.020 |
younger | +0.020 |
eight | +0.019 |
Neg logit contributions | |
dd | -0.042 |
night | -0.036 |
ez | -0.034 |
amb | -0.033 |
faced | -0.031 |
Token | They |
Token ID | 1119 |
Feature activation | -6.83e-02 |
Pos logit contributions | |
groups | +0.088 |
bars | +0.085 |
towns | +0.076 |
fighters | +0.076 |
eight | +0.076 |
Neg logit contributions | |
dd | -0.099 |
night | -0.082 |
ez | -0.076 |
pity | -0.073 |
amb | -0.071 |
Token | were |
Token ID | 547 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
bars | +0.132 |
groups | +0.132 |
fighters | +0.118 |
lights | +0.118 |
department | +0.116 |
Neg logit contributions | |
dd | -0.114 |
night | -0.094 |
ez | -0.082 |
pity | -0.079 |
amb | -0.077 |
Token | very |
Token ID | 845 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
bars | +0.376 |
groups | +0.364 |
lights | +0.323 |
department | +0.322 |
towns | +0.322 |
Neg logit contributions | |
dd | -0.283 |
night | -0.229 |
ez | -0.203 |
pity | -0.197 |
oo | -0.191 |
Token | happy |
Token ID | 3772 |
Feature activation | +8.44e-02 |
Pos logit contributions | |
dd | +0.018 |
night | +0.014 |
ez | +0.012 |
pity | +0.012 |
amb | +0.011 |
Neg logit contributions | |
groups | -0.029 |
bars | -0.029 |
towns | -0.026 |
fighters | -0.026 |
eight | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | +4.24e-02 |
Pos logit contributions | |
dd | +0.159 |
night | +0.131 |
ez | +0.124 |
pity | +0.120 |
faced | +0.117 |
Neg logit contributions | |
groups | -0.148 |
bars | -0.139 |
fighters | -0.131 |
towns | -0.129 |
department | -0.126 |
Token | danced |
Token ID | 39480 |
Feature activation | +2.49e-02 |
Pos logit contributions | |
dd | +0.077 |
night | +0.062 |
ez | +0.061 |
pity | +0.056 |
amb | +0.055 |
Neg logit contributions | |
groups | -0.075 |
bars | -0.073 |
towns | -0.067 |
younger | -0.067 |
fighters | -0.066 |
Token | together |
Token ID | 1978 |
Feature activation | +2.78e-02 |
Pos logit contributions | |
dd | +0.046 |
night | +0.037 |
ez | +0.036 |
pity | +0.034 |
oof | +0.033 |
Neg logit contributions | |
groups | -0.044 |
bars | -0.040 |
towns | -0.039 |
fighters | -0.038 |
lights | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | -5.23e-02 |
Pos logit contributions | |
dd | +0.045 |
night | +0.038 |
ez | +0.033 |
pity | +0.031 |
- | +0.031 |
Neg logit contributions | |
bars | -0.056 |
groups | -0.054 |
fighters | -0.048 |
department | -0.048 |
towns | -0.048 |
Token |
|
Token ID | 198 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
groups | +0.086 |
bars | +0.082 |
eight | +0.076 |
towns | +0.075 |
fighters | +0.074 |
Neg logit contributions | |
dd | -0.105 |
night | -0.086 |
ez | -0.080 |
pity | -0.078 |
amb | -0.074 |
Token |
|
Token ID | 198 |
Feature activation | +2.39e-02 |
Pos logit contributions | |
dd | +0.051 |
night | +0.046 |
pity | +0.041 |
ez | +0.041 |
amb | +0.040 |
Neg logit contributions | |
groups | -0.035 |
eight | -0.031 |
bars | -0.030 |
ents | -0.030 |
bridges | -0.029 |
Token | But |
Token ID | 1537 |
Feature activation | -3.02e-02 |
Pos logit contributions | |
dd | +0.054 |
night | +0.049 |
ez | +0.045 |
pity | +0.044 |
amb | +0.044 |
Neg logit contributions | |
groups | -0.031 |
bars | -0.028 |
eight | -0.026 |
towns | -0.025 |
bridges | -0.025 |
Token | farm |
Token ID | 5318 |
Feature activation | -2.09e-03 |
Pos logit contributions | |
groups | +0.005 |
bars | +0.005 |
fighters | +0.005 |
towns | +0.004 |
younger | +0.004 |
Neg logit contributions | |
dd | -0.005 |
night | -0.004 |
ez | -0.004 |
pity | -0.004 |
faced | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | -3.89e-03 |
Pos logit contributions | |
bars | +0.004 |
groups | +0.004 |
lights | +0.004 |
eight | +0.004 |
thirteen | +0.004 |
Neg logit contributions | |
dd | -0.003 |
night | -0.003 |
- | -0.002 |
ggy | -0.002 |
ez | -0.002 |
Token | so |
Token ID | 523 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
bars | +0.006 |
groups | +0.006 |
lights | +0.006 |
towns | +0.005 |
fighters | +0.005 |
Neg logit contributions | |
dd | -0.007 |
night | -0.006 |
it | -0.006 |
pity | -0.006 |
ez | -0.006 |
Token | he |
Token ID | 339 |
Feature activation | -4.57e-02 |
Pos logit contributions | |
groups | +0.162 |
bars | +0.161 |
eight | +0.140 |
towns | +0.140 |
fighters | +0.139 |
Neg logit contributions | |
dd | -0.200 |
night | -0.170 |
ez | -0.159 |
pity | -0.151 |
amb | -0.149 |
Token | got |
Token ID | 1392 |
Feature activation | -7.25e-03 |
Pos logit contributions | |
groups | +0.084 |
bars | +0.082 |
towns | +0.073 |
fighters | +0.073 |
younger | +0.072 |
Neg logit contributions | |
dd | -0.081 |
night | -0.068 |
ez | -0.062 |
pity | -0.059 |
amb | -0.058 |
Token | a |
Token ID | 257 |
Feature activation | +4.70e-02 |
Pos logit contributions | |
groups | +0.013 |
bars | +0.012 |
older | +0.011 |
younger | +0.011 |
eight | +0.011 |
Neg logit contributions | |
dd | -0.014 |
night | -0.011 |
ez | -0.010 |
amb | -0.010 |
oof | -0.010 |
Token | camera |
Token ID | 4676 |
Feature activation | +4.33e-02 |
Pos logit contributions | |
dd | +0.098 |
ez | +0.077 |
night | +0.075 |
amb | +0.074 |
Now | +0.067 |
Neg logit contributions | |
groups | -0.080 |
younger | -0.074 |
older | -0.074 |
eight | -0.072 |
bars | -0.071 |
Token | . |
Token ID | 13 |
Feature activation | -4.34e-02 |
Pos logit contributions | |
dd | +0.075 |
night | +0.063 |
ez | +0.056 |
amb | +0.053 |
pity | +0.053 |
Neg logit contributions | |
groups | -0.081 |
bars | -0.079 |
towns | -0.071 |
fighters | -0.070 |
eight | -0.070 |
Token | He |
Token ID | 679 |
Feature activation | -6.70e-02 |
Pos logit contributions | |
groups | +0.080 |
bars | +0.078 |
fighters | +0.070 |
towns | +0.070 |
younger | +0.069 |
Neg logit contributions | |
dd | -0.077 |
night | -0.065 |
ez | -0.058 |
pity | -0.055 |
amb | -0.055 |
Token | put |
Token ID | 1234 |
Feature activation | -1.22e-03 |
Pos logit contributions | |
groups | +0.123 |
bars | +0.120 |
fighters | +0.107 |
towns | +0.106 |
eight | +0.106 |
Neg logit contributions | |
dd | -0.119 |
night | -0.100 |
ez | -0.090 |
pity | -0.087 |
uggle | -0.085 |
Token | the |
Token ID | 262 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
groups | +0.002 |
bars | +0.002 |
eight | +0.002 |
bridges | +0.001 |
older | +0.001 |
Neg logit contributions | |
dd | -0.003 |
night | -0.002 |
ez | -0.002 |
amb | -0.002 |
ggy | -0.002 |
Token | said |
Token ID | 531 |
Feature activation | +6.79e-02 |
Pos logit contributions | |
dd | +0.112 |
night | +0.091 |
ez | +0.080 |
pity | +0.076 |
amb | +0.074 |
Neg logit contributions | |
groups | -0.152 |
bars | -0.149 |
fighters | -0.134 |
towns | -0.134 |
younger | -0.134 |
Token | . |
Token ID | 13 |
Feature activation | -3.96e-02 |
Pos logit contributions | |
dd | +0.120 |
night | +0.103 |
ez | +0.091 |
pity | +0.087 |
amb | +0.085 |
Neg logit contributions | |
groups | -0.126 |
bars | -0.122 |
towns | -0.110 |
fighters | -0.109 |
eight | -0.107 |
Token |
|
Token ID | 198 |
Feature activation | +3.57e-02 |
Pos logit contributions | |
groups | +0.079 |
bars | +0.077 |
towns | +0.069 |
fighters | +0.069 |
younger | +0.069 |
Neg logit contributions | |
dd | -0.065 |
night | -0.053 |
ez | -0.047 |
pity | -0.045 |
amb | -0.044 |
Token |
|
Token ID | 198 |
Feature activation | +3.46e-02 |
Pos logit contributions | |
dd | +0.062 |
night | +0.052 |
ez | +0.047 |
pity | +0.045 |
faced | +0.043 |
Neg logit contributions | |
groups | -0.067 |
bars | -0.065 |
towns | -0.059 |
eight | -0.058 |
younger | -0.058 |
Token | Ben |
Token ID | 11696 |
Feature activation | +4.42e-02 |
Pos logit contributions | |
dd | +0.049 |
night | +0.031 |
ggy | +0.029 |
amb | +0.028 |
faced | +0.027 |
Neg logit contributions | |
bars | -0.081 |
lights | -0.077 |
department | -0.074 |
groups | -0.074 |
fighters | -0.073 |
Token | went |
Token ID | 1816 |
Feature activation | +4.19e-02 |
Pos logit contributions | |
dd | +0.075 |
night | +0.061 |
ez | +0.055 |
pity | +0.053 |
amb | +0.052 |
Neg logit contributions | |
groups | -0.085 |
bars | -0.084 |
towns | -0.074 |
lights | -0.074 |
fighters | -0.074 |
Token | near |
Token ID | 1474 |
Feature activation | -8.79e-02 |
Pos logit contributions | |
dd | +0.080 |
night | +0.067 |
ez | +0.062 |
pity | +0.058 |
amb | +0.058 |
Neg logit contributions | |
groups | -0.073 |
bars | -0.071 |
towns | -0.063 |
fighters | -0.062 |
younger | -0.062 |
Token | the |
Token ID | 262 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
bars | +0.158 |
groups | +0.142 |
fighters | +0.134 |
lights | +0.134 |
department | +0.133 |
Neg logit contributions | |
dd | -0.157 |
night | -0.135 |
it | -0.127 |
pity | -0.126 |
ez | -0.126 |
Token | chicken |
Token ID | 9015 |
Feature activation | +9.35e-02 |
Pos logit contributions | |
bars | +0.319 |
groups | +0.313 |
lights | +0.284 |
towns | +0.280 |
fighters | +0.278 |
Neg logit contributions | |
dd | -0.211 |
night | -0.169 |
ez | -0.145 |
pity | -0.143 |
faced | -0.135 |
Token | and |
Token ID | 290 |
Feature activation | +5.15e-02 |
Pos logit contributions | |
dd | +0.141 |
night | +0.121 |
ez | +0.110 |
uggle | +0.103 |
pity | +0.102 |
Neg logit contributions | |
bars | -0.187 |
groups | -0.182 |
lights | -0.166 |
fighters | -0.165 |
fastest | -0.163 |
Token | reached |
Token ID | 4251 |
Feature activation | -6.87e-02 |
Pos logit contributions | |
dd | +0.088 |
night | +0.073 |
ez | +0.065 |
pity | +0.063 |
faced | +0.061 |
Neg logit contributions | |
groups | -0.098 |
bars | -0.097 |
towns | -0.086 |
fighters | -0.085 |
lights | -0.085 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.98e-01 |
Pos logit contributions | |
groups | +0.083 |
bars | +0.080 |
fighters | +0.073 |
fastest | +0.073 |
older | +0.073 |
Neg logit contributions | |
dd | -0.052 |
night | -0.044 |
amb | -0.038 |
ez | -0.037 |
Now | -0.037 |
Token | S |
Token ID | 50 |
Feature activation | +8.55e-02 |
Pos logit contributions | |
dd | +0.384 |
night | +0.326 |
ez | +0.293 |
pity | +0.287 |
faced | +0.281 |
Neg logit contributions | |
groups | -0.332 |
bars | -0.320 |
towns | -0.288 |
eight | -0.280 |
fighters | -0.280 |
Token | ue |
Token ID | 518 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
dd | +0.145 |
night | +0.122 |
ez | +0.107 |
uggle | +0.104 |
oo | +0.103 |
Neg logit contributions | |
groups | -0.162 |
bars | -0.158 |
fighters | -0.144 |
sizes | -0.143 |
lights | -0.143 |
Token | was |
Token ID | 373 |
Feature activation | +2.57e-02 |
Pos logit contributions | |
groups | +0.239 |
bars | +0.236 |
towns | +0.212 |
eight | +0.210 |
fighters | +0.206 |
Neg logit contributions | |
dd | -0.233 |
night | -0.192 |
pity | -0.171 |
ez | -0.171 |
faced | -0.167 |
Token | very |
Token ID | 845 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
dd | +0.052 |
night | +0.045 |
ez | +0.040 |
amb | +0.040 |
uable | +0.039 |
Neg logit contributions | |
groups | -0.040 |
older | -0.040 |
younger | -0.038 |
eight | -0.037 |
bars | -0.035 |
Token | excited |
Token ID | 6568 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
dd | +0.040 |
night | +0.033 |
ez | +0.030 |
pity | +0.028 |
amb | +0.028 |
Neg logit contributions | |
groups | -0.046 |
bars | -0.045 |
younger | -0.041 |
older | -0.040 |
eight | -0.040 |
Token | ! |
Token ID | 0 |
Feature activation | +2.75e-02 |
Pos logit contributions | |
dd | +0.257 |
night | +0.214 |
ez | +0.197 |
pity | +0.189 |
amb | +0.185 |
Neg logit contributions | |
groups | -0.246 |
bars | -0.241 |
towns | -0.213 |
fighters | -0.212 |
younger | -0.210 |
Token | Today |
Token ID | 6288 |
Feature activation | +4.86e-02 |
Pos logit contributions | |
dd | +0.048 |
night | +0.040 |
ez | +0.036 |
pity | +0.034 |
amb | +0.034 |
Neg logit contributions | |
groups | -0.051 |
bars | -0.051 |
towns | -0.045 |
fighters | -0.045 |
younger | -0.044 |
Token | she |
Token ID | 673 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
dd | +0.081 |
night | +0.067 |
ez | +0.060 |
pity | +0.057 |
amb | +0.056 |
Neg logit contributions | |
groups | -0.094 |
bars | -0.093 |
towns | -0.083 |
fighters | -0.082 |
younger | -0.082 |
Token | was |
Token ID | 373 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
dd | +0.040 |
night | +0.033 |
ez | +0.032 |
pity | +0.031 |
uggle | +0.030 |
Neg logit contributions | |
groups | -0.032 |
bars | -0.031 |
eight | -0.028 |
older | -0.027 |
towns | -0.027 |
Token | going |
Token ID | 1016 |
Feature activation | +1.52e-02 |
Pos logit contributions | |
dd | +0.044 |
night | +0.036 |
ez | +0.033 |
pity | +0.031 |
amb | +0.031 |
Neg logit contributions | |
groups | -0.048 |
bars | -0.047 |
towns | -0.042 |
fighters | -0.041 |
younger | -0.041 |
Token | loved |
Token ID | 6151 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
dd | +0.129 |
night | +0.109 |
ez | +0.099 |
pity | +0.097 |
uggle | +0.091 |
Neg logit contributions | |
groups | -0.132 |
bars | -0.130 |
towns | -0.117 |
fighters | -0.114 |
eight | -0.114 |
Token | to |
Token ID | 284 |
Feature activation | -3.23e-02 |
Pos logit contributions | |
dd | +0.321 |
night | +0.271 |
ez | +0.250 |
pity | +0.238 |
amb | +0.235 |
Neg logit contributions | |
groups | -0.275 |
bars | -0.272 |
towns | -0.238 |
fighters | -0.236 |
younger | -0.236 |
Token | slide |
Token ID | 10649 |
Feature activation | -7.74e-03 |
Pos logit contributions | |
groups | +0.062 |
bars | +0.061 |
towns | +0.055 |
younger | +0.054 |
fighters | +0.054 |
Neg logit contributions | |
dd | -0.055 |
night | -0.045 |
ez | -0.041 |
pity | -0.039 |
amb | -0.038 |
Token | down |
Token ID | 866 |
Feature activation | +3.05e-02 |
Pos logit contributions | |
groups | +0.018 |
department | +0.017 |
fighters | +0.017 |
eight | +0.016 |
bars | +0.016 |
Neg logit contributions | |
dd | -0.009 |
- | -0.007 |
it | -0.007 |
night | -0.007 |
oo | -0.006 |
Token | it |
Token ID | 340 |
Feature activation | +2.88e-01 |
Pos logit contributions | |
dd | +0.043 |
it | +0.042 |
- | +0.036 |
night | +0.034 |
Now | +0.033 |
Neg logit contributions | |
groups | -0.059 |
bars | -0.058 |
department | -0.056 |
fighters | -0.056 |
younger | -0.055 |
Token | every |
Token ID | 790 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
dd | +0.363 |
night | +0.312 |
- | +0.284 |
it | +0.270 |
ez | +0.251 |
Neg logit contributions | |
groups | -0.620 |
fighters | -0.596 |
bars | -0.595 |
department | -0.595 |
towns | -0.574 |
Token | day |
Token ID | 1110 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
bars | +0.270 |
groups | +0.269 |
eight | +0.246 |
older | +0.243 |
younger | +0.243 |
Neg logit contributions | |
dd | -0.175 |
night | -0.130 |
ez | -0.120 |
pity | -0.113 |
amb | -0.111 |
Token | . |
Token ID | 13 |
Feature activation | -4.41e-02 |
Pos logit contributions | |
dd | +0.222 |
night | +0.182 |
ez | +0.157 |
- | +0.154 |
amb | +0.151 |
Neg logit contributions | |
groups | -0.265 |
bars | -0.256 |
fighters | -0.242 |
department | -0.242 |
towns | -0.234 |
Token | But |
Token ID | 887 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
groups | +0.083 |
bars | +0.082 |
towns | +0.072 |
younger | +0.072 |
eight | +0.071 |
Neg logit contributions | |
dd | -0.078 |
night | -0.064 |
ez | -0.058 |
pity | -0.055 |
amb | -0.054 |
Token | one |
Token ID | 530 |
Feature activation | -4.18e-02 |
Pos logit contributions | |
dd | +0.285 |
night | +0.239 |
ez | +0.226 |
pity | +0.217 |
amb | +0.215 |
Neg logit contributions | |
groups | -0.212 |
bars | -0.208 |
eight | -0.180 |
older | -0.180 |
younger | -0.178 |
Token | day |
Token ID | 1110 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
groups | +0.066 |
eight | +0.065 |
bars | +0.065 |
older | +0.062 |
first | +0.061 |
Neg logit contributions | |
dd | -0.078 |
icol | -0.059 |
pity | -0.058 |
amb | -0.057 |
night | -0.057 |
Token | together |
Token ID | 1978 |
Feature activation | +4.21e-02 |
Pos logit contributions | |
dd | +0.229 |
night | +0.191 |
ez | +0.173 |
pity | +0.165 |
amb | +0.163 |
Neg logit contributions | |
groups | -0.228 |
bars | -0.225 |
towns | -0.198 |
fighters | -0.197 |
younger | -0.196 |
Token | . |
Token ID | 13 |
Feature activation | -3.64e-02 |
Pos logit contributions | |
dd | +0.065 |
night | +0.052 |
ez | +0.046 |
- | +0.045 |
it | +0.044 |
Neg logit contributions | |
bars | -0.089 |
groups | -0.086 |
younger | -0.078 |
department | -0.078 |
older | -0.078 |
Token | One |
Token ID | 1881 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
groups | +0.070 |
bars | +0.068 |
towns | +0.061 |
fighters | +0.061 |
younger | +0.060 |
Neg logit contributions | |
dd | -0.063 |
night | -0.052 |
ez | -0.047 |
pity | -0.044 |
amb | -0.044 |
Token | day |
Token ID | 1110 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
groups | +0.283 |
bars | +0.279 |
towns | +0.255 |
younger | +0.254 |
fighters | +0.253 |
Neg logit contributions | |
dd | -0.159 |
night | -0.120 |
ez | -0.105 |
pity | -0.098 |
amb | -0.096 |
Token | , |
Token ID | 11 |
Feature activation | +2.36e-03 |
Pos logit contributions | |
dd | +0.248 |
night | +0.207 |
ez | +0.187 |
pity | +0.179 |
amb | +0.176 |
Neg logit contributions | |
groups | -0.260 |
bars | -0.254 |
towns | -0.227 |
eight | -0.226 |
fighters | -0.226 |
Token | the |
Token ID | 262 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
dd | +0.005 |
night | +0.004 |
ez | +0.003 |
pity | +0.003 |
faced | +0.003 |
Neg logit contributions | |
bars | -0.004 |
groups | -0.004 |
ents | -0.003 |
towns | -0.003 |
fighters | -0.003 |
Token | clown |
Token ID | 25573 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
groups | +0.290 |
bars | +0.283 |
older | +0.273 |
fastest | +0.267 |
towns | +0.266 |
Neg logit contributions | |
dd | -0.240 |
ez | -0.188 |
night | -0.187 |
amb | -0.167 |
oof | -0.159 |
Token | had |
Token ID | 550 |
Feature activation | -7.46e-02 |
Pos logit contributions | |
dd | +0.272 |
night | +0.224 |
ez | +0.199 |
pity | +0.186 |
amb | +0.186 |
Neg logit contributions | |
groups | -0.325 |
bars | -0.323 |
towns | -0.285 |
lights | -0.285 |
younger | -0.283 |
Token | a |
Token ID | 257 |
Feature activation | +4.76e-02 |
Pos logit contributions | |
bars | +0.167 |
groups | +0.165 |
lights | +0.155 |
department | +0.149 |
fighters | +0.148 |
Neg logit contributions | |
dd | -0.101 |
night | -0.081 |
ez | -0.067 |
pity | -0.065 |
it | -0.064 |
Token | big |
Token ID | 1263 |
Feature activation | -6.24e-02 |
Pos logit contributions | |
dd | +0.107 |
ez | +0.087 |
amb | +0.086 |
night | +0.084 |
uggle | +0.077 |
Neg logit contributions | |
groups | -0.075 |
younger | -0.070 |
older | -0.068 |
eight | -0.065 |
fastest | -0.063 |
Token | game |
Token ID | 983 |
Feature activation | +4.68e-03 |
Pos logit contributions | |
groups | +0.107 |
older | +0.097 |
younger | +0.094 |
bars | +0.092 |
eight | +0.087 |
Neg logit contributions | |
dd | -0.134 |
night | -0.101 |
ez | -0.100 |
amb | -0.099 |
to | -0.095 |
Token | . |
Token ID | 13 |
Feature activation | -1.76e-02 |
Pos logit contributions | |
groups | +0.223 |
bars | +0.207 |
towns | +0.198 |
fighters | +0.189 |
eight | +0.188 |
Neg logit contributions | |
dd | -0.175 |
night | -0.152 |
ez | -0.134 |
- | -0.129 |
amb | -0.126 |
Token |
|
Token ID | 198 |
Feature activation | +5.79e-02 |
Pos logit contributions | |
groups | +0.035 |
bars | +0.034 |
towns | +0.031 |
fighters | +0.030 |
younger | +0.030 |
Neg logit contributions | |
dd | -0.029 |
night | -0.024 |
ez | -0.021 |
pity | -0.020 |
amb | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | +5.46e-02 |
Pos logit contributions | |
dd | +0.102 |
night | +0.089 |
ez | +0.077 |
pity | +0.074 |
amb | +0.073 |
Neg logit contributions | |
groups | -0.108 |
bars | -0.105 |
eight | -0.094 |
towns | -0.093 |
ents | -0.092 |
Token | " |
Token ID | 1 |
Feature activation | -3.56e-02 |
Pos logit contributions | |
dd | +0.101 |
night | +0.081 |
ez | +0.075 |
pity | +0.072 |
amb | +0.071 |
Neg logit contributions | |
groups | -0.098 |
bars | -0.098 |
fighters | -0.087 |
younger | -0.086 |
department | -0.086 |
Token | Grab |
Token ID | 48400 |
Feature activation | +7.24e-02 |
Pos logit contributions | |
groups | +0.064 |
bars | +0.063 |
eight | +0.057 |
towns | +0.056 |
fighters | +0.055 |
Neg logit contributions | |
dd | -0.063 |
night | -0.054 |
ez | -0.047 |
amb | -0.047 |
oo | -0.046 |
Token | my |
Token ID | 616 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
dd | +0.149 |
night | +0.132 |
ez | +0.118 |
faced | +0.114 |
amb | +0.114 |
Neg logit contributions | |
groups | -0.108 |
bars | -0.105 |
eight | -0.096 |
towns | -0.091 |
older | -0.091 |
Token | hand |
Token ID | 1021 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
groups | +0.203 |
bars | +0.183 |
towns | +0.179 |
fighters | +0.177 |
lights | +0.174 |
Neg logit contributions | |
dd | -0.185 |
night | -0.175 |
- | -0.158 |
pity | -0.154 |
ez | -0.147 |
Token | , |
Token ID | 11 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
groups | +0.266 |
bars | +0.249 |
towns | +0.245 |
lights | +0.234 |
younger | +0.233 |
Neg logit contributions | |
dd | -0.142 |
night | -0.113 |
- | -0.098 |
amb | -0.095 |
ez | -0.093 |
Token | Ben |
Token ID | 3932 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
dd | +0.026 |
night | +0.022 |
ez | +0.020 |
pity | +0.019 |
amb | +0.019 |
Neg logit contributions | |
groups | -0.024 |
bars | -0.024 |
towns | -0.021 |
fighters | -0.021 |
younger | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | -2.26e-02 |
Pos logit contributions | |
groups | +0.026 |
bars | +0.025 |
fighters | +0.022 |
eight | +0.022 |
towns | +0.022 |
Neg logit contributions | |
dd | -0.024 |
night | -0.020 |
ez | -0.018 |
amb | -0.017 |
pity | -0.017 |
Token | I |
Token ID | 314 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
bars | +0.041 |
groups | +0.041 |
eight | +0.037 |
towns | +0.036 |
fighters | +0.036 |
Neg logit contributions | |
dd | -0.041 |
night | -0.033 |
ez | -0.029 |
pity | -0.028 |
faced | -0.028 |
Token | throw |
Token ID | 3714 |
Feature activation | +6.97e-02 |
Pos logit contributions | |
bars | +0.109 |
groups | +0.108 |
lights | +0.096 |
fighters | +0.096 |
towns | +0.094 |
Neg logit contributions | |
dd | -0.097 |
night | -0.081 |
pity | -0.073 |
ez | -0.072 |
uggle | -0.072 |
Token | their |
Token ID | 511 |
Feature activation | -6.56e-02 |
Pos logit contributions | |
dd | +0.135 |
night | +0.114 |
ez | +0.102 |
amb | +0.100 |
pity | +0.096 |
Neg logit contributions | |
groups | -0.121 |
bars | -0.116 |
towns | -0.103 |
eight | -0.102 |
bridges | -0.102 |
Token | balls |
Token ID | 11333 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
bars | +0.123 |
groups | +0.108 |
fighters | +0.108 |
department | +0.104 |
younger | +0.103 |
Neg logit contributions | |
dd | -0.102 |
night | -0.087 |
- | -0.083 |
ez | -0.082 |
pity | -0.081 |
Token | to |
Token ID | 284 |
Feature activation | -3.36e-02 |
Pos logit contributions | |
groups | +0.245 |
bars | +0.243 |
fighters | +0.222 |
towns | +0.221 |
younger | +0.220 |
Neg logit contributions | |
dd | -0.164 |
night | -0.133 |
ez | -0.114 |
uggle | -0.109 |
amb | -0.107 |
Token | each |
Token ID | 1123 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
bars | +0.056 |
groups | +0.054 |
fighters | +0.048 |
department | +0.047 |
lights | +0.047 |
Neg logit contributions | |
dd | -0.064 |
night | -0.055 |
pity | -0.051 |
ez | -0.050 |
oo | -0.048 |
Token | other |
Token ID | 584 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
bars | +0.144 |
groups | +0.137 |
fighters | +0.120 |
lights | +0.120 |
department | +0.116 |
Neg logit contributions | |
dd | -0.251 |
night | -0.223 |
ez | -0.209 |
pity | -0.205 |
- | -0.201 |
Token | and |
Token ID | 290 |
Feature activation | +5.08e-02 |
Pos logit contributions | |
bars | +0.458 |
groups | +0.431 |
fighters | +0.412 |
eight | +0.411 |
department | +0.405 |
Neg logit contributions | |
dd | -0.251 |
night | -0.223 |
- | -0.192 |
ez | -0.188 |
uggle | -0.174 |
Token | catch |
Token ID | 4929 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
dd | +0.088 |
night | +0.074 |
ez | +0.066 |
pity | +0.064 |
amb | +0.061 |
Neg logit contributions | |
groups | -0.095 |
bars | -0.094 |
fighters | -0.083 |
towns | -0.082 |
younger | -0.081 |
Token | them |
Token ID | 606 |
Feature activation | +1.37e-02 |
Pos logit contributions | |
dd | +0.388 |
night | +0.328 |
ez | +0.295 |
amb | +0.287 |
pity | +0.279 |
Neg logit contributions | |
groups | -0.355 |
bars | -0.336 |
towns | -0.301 |
younger | -0.298 |
older | -0.296 |
Token | . |
Token ID | 13 |
Feature activation | -3.91e-02 |
Pos logit contributions | |
dd | +0.019 |
night | +0.016 |
ez | +0.015 |
- | +0.014 |
it | +0.013 |
Neg logit contributions | |
bars | -0.029 |
groups | -0.028 |
towns | -0.026 |
eight | -0.026 |
lights | -0.025 |
Token | They |
Token ID | 1119 |
Feature activation | -5.93e-02 |
Pos logit contributions | |
groups | +0.064 |
bars | +0.063 |
towns | +0.055 |
fighters | +0.055 |
younger | +0.054 |
Neg logit contributions | |
dd | -0.078 |
night | -0.066 |
ez | -0.061 |
pity | -0.058 |
amb | -0.057 |
Token | 's |
Token ID | 338 |
Feature activation | -9.70e-02 |
Pos logit contributions | |
dd | +0.275 |
night | +0.240 |
ez | +0.211 |
- | +0.205 |
pity | +0.204 |
Neg logit contributions | |
groups | -0.283 |
bars | -0.282 |
lights | -0.249 |
ents | -0.241 |
towns | -0.241 |
Token | getting |
Token ID | 1972 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
bars | +0.209 |
groups | +0.201 |
lights | +0.179 |
towns | +0.179 |
fighters | +0.178 |
Neg logit contributions | |
dd | -0.150 |
night | -0.121 |
pity | -0.105 |
ez | -0.105 |
oo | -0.101 |
Token | closer |
Token ID | 5699 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
bars | +0.341 |
groups | +0.338 |
fighters | +0.300 |
towns | +0.298 |
ains | +0.296 |
Neg logit contributions | |
dd | -0.287 |
night | -0.239 |
pity | -0.209 |
it | -0.206 |
- | -0.193 |
Token | , |
Token ID | 11 |
Feature activation | +9.59e-03 |
Pos logit contributions | |
dd | +0.016 |
night | +0.014 |
- | +0.012 |
it | +0.011 |
ez | +0.010 |
Neg logit contributions | |
bars | -0.028 |
groups | -0.028 |
towns | -0.025 |
younger | -0.025 |
lights | -0.025 |
Token | let |
Token ID | 1309 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
dd | +0.016 |
night | +0.013 |
ez | +0.011 |
amb | +0.011 |
pity | +0.011 |
Neg logit contributions | |
groups | -0.019 |
bars | -0.018 |
towns | -0.016 |
fighters | -0.016 |
younger | -0.016 |
Token | 's |
Token ID | 338 |
Feature activation | -9.59e-02 |
Pos logit contributions | |
groups | +0.155 |
bars | +0.151 |
younger | +0.132 |
eight | +0.132 |
towns | +0.131 |
Neg logit contributions | |
dd | -0.256 |
night | -0.222 |
ez | -0.208 |
pity | -0.202 |
amb | -0.200 |
Token | go |
Token ID | 467 |
Feature activation | +8.54e-03 |
Pos logit contributions | |
groups | +0.167 |
bars | +0.161 |
towns | +0.145 |
younger | +0.144 |
eight | +0.143 |
Neg logit contributions | |
dd | -0.181 |
night | -0.152 |
ez | -0.140 |
pity | -0.131 |
amb | -0.130 |
Token | ." |
Token ID | 526 |
Feature activation | -7.62e-02 |
Pos logit contributions | |
dd | +0.015 |
night | +0.012 |
ez | +0.011 |
pity | +0.011 |
faced | +0.010 |
Neg logit contributions | |
groups | -0.016 |
bars | -0.016 |
towns | -0.014 |
fighters | -0.014 |
younger | -0.014 |
Token | Rob |
Token ID | 3851 |
Feature activation | +2.80e-02 |
Pos logit contributions | |
groups | +0.145 |
bars | +0.143 |
towns | +0.127 |
fighters | +0.126 |
eight | +0.125 |
Neg logit contributions | |
dd | -0.131 |
night | -0.109 |
ez | -0.097 |
pity | -0.093 |
amb | -0.091 |
Token | tremb |
Token ID | 29868 |
Feature activation | -5.46e-02 |
Pos logit contributions | |
dd | +0.047 |
night | +0.039 |
ez | +0.034 |
amb | +0.033 |
pity | +0.032 |
Neg logit contributions | |
groups | -0.055 |
bars | -0.054 |
towns | -0.048 |
bridges | -0.048 |
lights | -0.048 |
Token | led |
Token ID | 992 |
Feature activation | +9.02e-02 |
Pos logit contributions | |
groups | +0.112 |
bars | +0.110 |
towns | +0.100 |
lights | +0.099 |
older | +0.098 |
Neg logit contributions | |
dd | -0.083 |
night | -0.068 |
ez | -0.059 |
uggle | -0.058 |
pity | -0.058 |
Token | help |
Token ID | 1037 |
Feature activation | +2.67e-03 |
Pos logit contributions | |
groups | +0.039 |
bars | +0.038 |
towns | +0.034 |
fighters | +0.034 |
younger | +0.033 |
Neg logit contributions | |
dd | -0.037 |
night | -0.031 |
ez | -0.028 |
pity | -0.027 |
amb | -0.026 |
Token | you |
Token ID | 345 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
dd | +0.004 |
night | +0.003 |
it | +0.003 |
- | +0.003 |
ez | +0.003 |
Neg logit contributions | |
bars | -0.006 |
groups | -0.006 |
lights | -0.005 |
department | -0.005 |
fighters | -0.005 |
Token | !" |
Token ID | 2474 |
Feature activation | +6.51e-02 |
Pos logit contributions | |
dd | +0.016 |
night | +0.014 |
uggle | +0.012 |
- | +0.012 |
ez | +0.012 |
Neg logit contributions | |
bars | -0.022 |
groups | -0.021 |
lights | -0.020 |
fighters | -0.019 |
towns | -0.019 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
dd | +0.085 |
night | +0.072 |
ez | +0.062 |
amb | +0.060 |
Now | +0.059 |
Neg logit contributions | |
groups | -0.145 |
bars | -0.142 |
fastest | -0.134 |
department | -0.132 |
fighters | -0.131 |
Token | Once |
Token ID | 7454 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
dd | +0.393 |
night | +0.333 |
ez | +0.301 |
pity | +0.293 |
amb | +0.287 |
Neg logit contributions | |
groups | -0.345 |
bars | -0.335 |
towns | -0.298 |
fighters | -0.293 |
eight | -0.292 |
Token | upon |
Token ID | 2402 |
Feature activation | -3.10e-01 |
Pos logit contributions | |
groups | +0.109 |
bars | +0.103 |
fighters | +0.094 |
towns | +0.092 |
sizes | +0.090 |
Neg logit contributions | |
dd | -0.247 |
night | -0.219 |
it | -0.208 |
happened | -0.201 |
- | -0.197 |
Token | a |
Token ID | 257 |
Feature activation | +5.31e-02 |
Pos logit contributions | |
groups | +0.443 |
bars | +0.384 |
eight | +0.377 |
older | +0.356 |
younger | +0.351 |
Neg logit contributions | |
dd | -0.676 |
night | -0.573 |
amb | -0.558 |
ez | -0.544 |
pity | -0.525 |
Token | time |
Token ID | 640 |
Feature activation | +9.11e-02 |
Pos logit contributions | |
dd | +0.088 |
night | +0.071 |
ez | +0.066 |
amb | +0.065 |
pity | +0.064 |
Neg logit contributions | |
groups | -0.102 |
bars | -0.093 |
eight | -0.090 |
younger | -0.087 |
older | -0.087 |
Token | there |
Token ID | 612 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
dd | +0.148 |
night | +0.120 |
ez | +0.106 |
pity | +0.100 |
amb | +0.099 |
Neg logit contributions | |
groups | -0.181 |
bars | -0.180 |
fighters | -0.161 |
towns | -0.160 |
younger | -0.160 |
Token | was |
Token ID | 373 |
Feature activation | +3.49e-02 |
Pos logit contributions | |
groups | +0.022 |
bars | +0.020 |
ill | +0.018 |
towns | +0.018 |
fighters | +0.018 |
Neg logit contributions | |
dd | -0.030 |
night | -0.026 |
amb | -0.024 |
ez | -0.024 |
pity | -0.023 |
Token | a |
Token ID | 257 |
Feature activation | +5.20e-02 |
Pos logit contributions | |
dd | +0.077 |
pity | +0.074 |
night | +0.072 |
uable | +0.072 |
person | +0.070 |
Neg logit contributions | |
groups | -0.040 |
eight | -0.036 |
older | -0.030 |
first | -0.030 |
younger | -0.028 |
Token | had |
Token ID | 550 |
Feature activation | -7.28e-02 |
Pos logit contributions | |
groups | +0.207 |
bars | +0.202 |
younger | +0.182 |
older | +0.180 |
towns | +0.178 |
Neg logit contributions | |
dd | -0.200 |
night | -0.165 |
ez | -0.147 |
pity | -0.143 |
ggy | -0.139 |
Token | not |
Token ID | 407 |
Feature activation | -1.00e-01 |
Pos logit contributions | |
bars | +0.151 |
groups | +0.147 |
lights | +0.137 |
towns | +0.128 |
department | +0.126 |
Neg logit contributions | |
dd | -0.115 |
night | -0.097 |
pity | -0.085 |
ez | -0.084 |
ggy | -0.083 |
Token | touched |
Token ID | 12615 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
bars | +0.221 |
lights | +0.214 |
groups | +0.212 |
ains | +0.201 |
department | +0.191 |
Neg logit contributions | |
dd | -0.134 |
night | -0.113 |
happened | -0.108 |
ez | -0.105 |
pity | -0.097 |
Token | the |
Token ID | 262 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
bars | +0.042 |
groups | +0.041 |
department | +0.038 |
fighters | +0.037 |
ents | +0.037 |
Neg logit contributions | |
dd | -0.041 |
night | -0.035 |
it | -0.033 |
ez | -0.032 |
pity | -0.032 |
Token | or |
Token ID | 393 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
bars | +0.274 |
groups | +0.266 |
ill | +0.254 |
fighters | +0.248 |
lights | +0.247 |
Neg logit contributions | |
dd | -0.186 |
puppy | -0.171 |
pity | -0.168 |
night | -0.162 |
jewel | -0.161 |
Token | nam |
Token ID | 7402 |
Feature activation | -2.98e-01 |
Pos logit contributions | |
groups | +0.196 |
bars | +0.189 |
lands | +0.172 |
fighters | +0.169 |
lights | +0.168 |
Neg logit contributions | |
dd | -0.185 |
night | -0.167 |
it | -0.162 |
happened | -0.156 |
pity | -0.150 |
Token | ents |
Token ID | 658 |
Feature activation | -4.20e-01 |
Pos logit contributions | |
groups | +0.198 |
department | +0.194 |
fighters | +0.184 |
bars | +0.175 |
narrower | +0.172 |
Neg logit contributions | |
dd | -0.788 |
- | -0.697 |
amb | -0.692 |
ear | -0.684 |
ake | -0.679 |
Token | . |
Token ID | 13 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
groups | +0.865 |
bars | +0.830 |
eight | +0.770 |
towns | +0.770 |
fighters | +0.759 |
Neg logit contributions | |
dd | -0.620 |
night | -0.532 |
- | -0.485 |
it | -0.473 |
ez | -0.470 |
Token | They |
Token ID | 1119 |
Feature activation | -5.20e-02 |
Pos logit contributions | |
groups | +0.058 |
bars | +0.057 |
towns | +0.050 |
fighters | +0.050 |
younger | +0.049 |
Neg logit contributions | |
dd | -0.068 |
night | -0.058 |
ez | -0.053 |
pity | -0.051 |
amb | -0.050 |
Token | hoped |
Token ID | 10719 |
Feature activation | -5.49e-02 |
Pos logit contributions | |
groups | +0.108 |
bars | +0.107 |
fighters | +0.096 |
towns | +0.096 |
younger | +0.095 |
Neg logit contributions | |
dd | -0.079 |
night | -0.064 |
ez | -0.057 |
pity | -0.054 |
amb | -0.052 |
Token | Mom |
Token ID | 11254 |
Feature activation | -3.61e-02 |
Pos logit contributions | |
groups | +0.102 |
bars | +0.099 |
fastest | +0.088 |
eight | +0.086 |
towns | +0.085 |
Neg logit contributions | |
dd | -0.096 |
night | -0.084 |
it | -0.072 |
pity | -0.069 |
ggy | -0.069 |
Token |
|
Token ID | 198 |
Feature activation | +3.43e-02 |
Pos logit contributions | |
dd | +0.062 |
night | +0.052 |
ez | +0.046 |
pity | +0.044 |
amb | +0.043 |
Neg logit contributions | |
groups | -0.064 |
bars | -0.062 |
eight | -0.057 |
towns | -0.057 |
younger | -0.055 |
Token | She |
Token ID | 3347 |
Feature activation | -8.20e-02 |
Pos logit contributions | |
dd | +0.076 |
night | +0.064 |
ez | +0.060 |
pity | +0.058 |
ggy | +0.058 |
Neg logit contributions | |
bars | -0.050 |
groups | -0.049 |
fighters | -0.044 |
lights | -0.042 |
department | -0.042 |
Token | was |
Token ID | 373 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
groups | +0.133 |
bars | +0.126 |
eight | +0.115 |
younger | +0.112 |
ents | +0.112 |
Neg logit contributions | |
dd | -0.162 |
night | -0.139 |
uggle | -0.130 |
ez | -0.126 |
amb | -0.125 |
Token | scared |
Token ID | 12008 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
dd | +0.035 |
night | +0.030 |
ez | +0.027 |
pity | +0.026 |
amb | +0.025 |
Neg logit contributions | |
groups | -0.034 |
bars | -0.033 |
towns | -0.029 |
fighters | -0.029 |
younger | -0.029 |
Token | at |
Token ID | 379 |
Feature activation | +9.80e-02 |
Pos logit contributions | |
dd | +0.016 |
night | +0.013 |
ez | +0.012 |
- | +0.011 |
it | +0.011 |
Neg logit contributions | |
groups | -0.022 |
bars | -0.022 |
fighters | -0.020 |
towns | -0.020 |
ains | -0.019 |
Token | first |
Token ID | 717 |
Feature activation | -3.82e-01 |
Pos logit contributions | |
dd | +0.255 |
night | +0.234 |
oo | +0.218 |
pity | +0.217 |
uggle | +0.216 |
Neg logit contributions | |
groups | -0.083 |
bars | -0.083 |
lights | -0.066 |
fighters | -0.065 |
towns | -0.063 |
Token | , |
Token ID | 11 |
Feature activation | -7.47e-03 |
Pos logit contributions | |
groups | +0.687 |
bars | +0.673 |
towns | +0.596 |
fighters | +0.592 |
younger | +0.588 |
Neg logit contributions | |
dd | -0.698 |
night | -0.583 |
ez | -0.528 |
pity | -0.506 |
amb | -0.498 |
Token | but |
Token ID | 475 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
groups | +0.010 |
bars | +0.010 |
fighters | +0.009 |
towns | +0.009 |
department | +0.008 |
Neg logit contributions | |
dd | -0.017 |
night | -0.014 |
ez | -0.013 |
pity | -0.013 |
amb | -0.013 |
Token | then |
Token ID | 788 |
Feature activation | -4.54e-02 |
Pos logit contributions | |
groups | +0.271 |
bars | +0.270 |
eight | +0.240 |
older | +0.240 |
younger | +0.240 |
Neg logit contributions | |
dd | -0.345 |
night | -0.292 |
ez | -0.274 |
amb | -0.267 |
uggle | -0.258 |
Token | she |
Token ID | 673 |
Feature activation | +1.22e-02 |
Pos logit contributions | |
groups | +0.083 |
bars | +0.081 |
towns | +0.073 |
younger | +0.072 |
eight | +0.072 |
Neg logit contributions | |
dd | -0.081 |
night | -0.067 |
ez | -0.062 |
amb | -0.060 |
uggle | -0.058 |
Token | saw |
Token ID | 2497 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
dd | +0.021 |
night | +0.017 |
ez | +0.015 |
pity | +0.014 |
- | +0.014 |
Neg logit contributions | |
bars | -0.024 |
groups | -0.024 |
towns | -0.021 |
lights | -0.021 |
fighters | -0.021 |
Token | exciting |
Token ID | 7895 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
groups | +0.007 |
bars | +0.007 |
older | +0.007 |
younger | +0.006 |
towns | +0.006 |
Neg logit contributions | |
dd | -0.006 |
night | -0.005 |
ez | -0.004 |
amb | -0.004 |
faced | -0.004 |
Token | adventure |
Token ID | 8855 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
dd | +0.033 |
ez | +0.026 |
night | +0.024 |
amb | +0.024 |
faced | +0.024 |
Neg logit contributions | |
groups | -0.029 |
towns | -0.027 |
directions | -0.026 |
older | -0.026 |
bridges | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -3.60e-02 |
Pos logit contributions | |
dd | +0.179 |
night | +0.149 |
ez | +0.131 |
- | +0.126 |
ggy | +0.124 |
Neg logit contributions | |
groups | -0.196 |
bars | -0.189 |
eight | -0.173 |
towns | -0.170 |
fighters | -0.169 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.98e-01 |
Pos logit contributions | |
groups | +0.057 |
bars | +0.057 |
eight | +0.052 |
ents | +0.051 |
fighters | +0.050 |
Neg logit contributions | |
dd | -0.071 |
night | -0.059 |
ez | -0.054 |
pity | -0.053 |
faced | -0.050 |
Token | Once |
Token ID | 7454 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
dd | +0.374 |
night | +0.316 |
ez | +0.281 |
pity | +0.277 |
amb | +0.269 |
Neg logit contributions | |
groups | -0.347 |
bars | -0.330 |
towns | -0.301 |
eight | -0.299 |
fighters | -0.291 |
Token | upon |
Token ID | 2402 |
Feature activation | -3.19e-01 |
Pos logit contributions | |
groups | +0.128 |
bars | +0.123 |
fighters | +0.108 |
towns | +0.108 |
department | +0.102 |
Neg logit contributions | |
dd | -0.267 |
night | -0.234 |
oo | -0.212 |
ez | -0.211 |
- | -0.210 |
Token | a |
Token ID | 257 |
Feature activation | +4.45e-02 |
Pos logit contributions | |
groups | +0.409 |
eight | +0.360 |
bars | +0.338 |
first | +0.322 |
older | +0.322 |
Neg logit contributions | |
dd | -0.737 |
night | -0.619 |
amb | -0.617 |
ez | -0.581 |
pity | -0.569 |
Token | time |
Token ID | 640 |
Feature activation | +8.22e-02 |
Pos logit contributions | |
dd | +0.078 |
night | +0.065 |
ez | +0.061 |
pity | +0.060 |
amb | +0.059 |
Neg logit contributions | |
groups | -0.079 |
bars | -0.069 |
eight | -0.069 |
younger | -0.065 |
older | -0.065 |
Token | , |
Token ID | 11 |
Feature activation | -3.28e-03 |
Pos logit contributions | |
dd | +0.129 |
night | +0.104 |
ez | +0.091 |
oo | +0.086 |
amb | +0.086 |
Neg logit contributions | |
bars | -0.167 |
groups | -0.167 |
fighters | -0.149 |
towns | -0.149 |
younger | -0.148 |
Token | there |
Token ID | 612 |
Feature activation | -2.33e-02 |
Pos logit contributions | |
groups | +0.005 |
bars | +0.005 |
eight | +0.005 |
older | +0.004 |
towns | +0.004 |
Neg logit contributions | |
dd | -0.007 |
night | -0.006 |
ez | -0.005 |
amb | -0.005 |
pity | -0.005 |
Token | was |
Token ID | 373 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
groups | +0.035 |
bars | +0.033 |
eight | +0.030 |
towns | +0.029 |
ill | +0.029 |
Neg logit contributions | |
dd | -0.049 |
night | -0.042 |
amb | -0.039 |
ez | -0.038 |
pity | -0.037 |
Token | by |
Token ID | 416 |
Feature activation | -2.47e-02 |
Pos logit contributions | |
groups | +0.136 |
bars | +0.130 |
ents | +0.122 |
eight | +0.121 |
department | +0.121 |
Neg logit contributions | |
dd | -0.087 |
night | -0.077 |
ez | -0.063 |
- | -0.061 |
it | -0.060 |
Token | himself |
Token ID | 2241 |
Feature activation | -5.76e-02 |
Pos logit contributions | |
bars | +0.050 |
groups | +0.048 |
department | +0.046 |
ents | +0.044 |
fighters | +0.044 |
Neg logit contributions | |
dd | -0.039 |
it | -0.029 |
night | -0.029 |
ez | -0.029 |
oo | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -2.86e-02 |
Pos logit contributions | |
bars | +0.131 |
groups | +0.126 |
bridges | +0.123 |
lights | +0.121 |
department | +0.121 |
Neg logit contributions | |
dd | -0.066 |
night | -0.054 |
- | -0.054 |
it | -0.049 |
ggy | -0.042 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
groups | +0.054 |
bars | +0.053 |
towns | +0.047 |
fighters | +0.047 |
eight | +0.047 |
Neg logit contributions | |
dd | -0.050 |
night | -0.041 |
ez | -0.037 |
pity | -0.036 |
amb | -0.034 |
Token | Once |
Token ID | 7454 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
dd | +0.395 |
night | +0.339 |
ez | +0.301 |
pity | +0.298 |
faced | +0.292 |
Neg logit contributions | |
groups | -0.325 |
bars | -0.308 |
towns | -0.282 |
eight | -0.274 |
fighters | -0.271 |
Token | upon |
Token ID | 2402 |
Feature activation | -3.11e-01 |
Pos logit contributions | |
groups | +0.105 |
bars | +0.101 |
fighters | +0.087 |
towns | +0.086 |
department | +0.082 |
Neg logit contributions | |
dd | -0.279 |
night | -0.246 |
it | -0.225 |
ez | -0.224 |
- | -0.223 |
Token | a |
Token ID | 257 |
Feature activation | +5.67e-02 |
Pos logit contributions | |
groups | +0.388 |
eight | +0.341 |
bars | +0.309 |
first | +0.301 |
older | +0.301 |
Neg logit contributions | |
dd | -0.721 |
amb | -0.623 |
night | -0.622 |
ez | -0.574 |
icol | -0.572 |
Token | time |
Token ID | 640 |
Feature activation | +9.22e-02 |
Pos logit contributions | |
dd | +0.101 |
night | +0.084 |
amb | +0.080 |
ez | +0.078 |
pity | +0.077 |
Neg logit contributions | |
groups | -0.099 |
eight | -0.086 |
bars | -0.084 |
younger | -0.082 |
older | -0.081 |
Token | , |
Token ID | 11 |
Feature activation | +2.28e-03 |
Pos logit contributions | |
dd | +0.153 |
night | +0.125 |
ez | +0.112 |
pity | +0.106 |
amb | +0.105 |
Neg logit contributions | |
groups | -0.180 |
bars | -0.178 |
towns | -0.159 |
fighters | -0.158 |
younger | -0.157 |
Token | there |
Token ID | 612 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
dd | +0.005 |
night | +0.004 |
ez | +0.004 |
pity | +0.004 |
amb | +0.003 |
Neg logit contributions | |
groups | -0.004 |
bars | -0.003 |
eight | -0.003 |
towns | -0.003 |
older | -0.003 |
Token | was |
Token ID | 373 |
Feature activation | +3.58e-02 |
Pos logit contributions | |
groups | +0.022 |
bars | +0.021 |
eight | +0.019 |
towns | +0.019 |
ill | +0.018 |
Neg logit contributions | |
dd | -0.028 |
night | -0.025 |
amb | -0.022 |
ez | -0.022 |
pity | -0.021 |