Token | quickly |
Token ID | 2952 |
Feature activation | -4.72e+00 |
Pos logit contributions | |
ons | +0.154 |
bes | +0.122 |
ages | +0.115 |
on | +0.110 |
con | +0.109 |
Neg logit contributions | |
ksh | -0.099 |
flood | -0.098 |
rant | -0.097 |
dolphin | -0.095 |
whale | -0.093 |
Token | checked |
Token ID | 10667 |
Feature activation | +1.94e+00 |
Pos logit contributions | |
ons | +0.370 |
bes | +0.293 |
ages | +0.277 |
on | +0.263 |
con | +0.260 |
Neg logit contributions | |
ksh | -0.249 |
flood | -0.248 |
rant | -0.248 |
dolphin | -0.239 |
whale | -0.235 |
Token | for |
Token ID | 329 |
Feature activation | +2.25e+00 |
Pos logit contributions | |
rant | +0.099 |
coming | +0.097 |
flood | +0.094 |
dolphin | +0.093 |
whale | +0.093 |
Neg logit contributions | |
ons | -0.150 |
bes | -0.129 |
ages | -0.122 |
seek | -0.114 |
con | -0.111 |
Token | a |
Token ID | 257 |
Feature activation | -8.58e-01 |
Pos logit contributions | |
flood | +0.126 |
coming | +0.122 |
rant | +0.121 |
dolphin | +0.118 |
whale | +0.118 |
Neg logit contributions | |
ons | -0.168 |
bes | -0.136 |
ages | -0.131 |
con | -0.121 |
seek | -0.119 |
Token | fire |
Token ID | 2046 |
Feature activation | +6.25e+00 |
Pos logit contributions | |
ons | +0.063 |
ages | +0.047 |
bes | +0.047 |
on | +0.047 |
con | +0.046 |
Neg logit contributions | |
ksh | -0.051 |
rant | -0.049 |
iet | -0.046 |
dolphin | -0.043 |
flood | -0.043 |
Token | extingu |
Token ID | 24995 |
Feature activation | +1.29e+01 |
Pos logit contributions | |
ksh | +0.408 |
rant | +0.402 |
iet | +0.393 |
ackers | +0.372 |
ust | +0.372 |
Neg logit contributions | |
ons | -0.396 |
ages | -0.316 |
con | -0.290 |
es | -0.289 |
bes | -0.282 |
Token | isher |
Token ID | 4828 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
ksh | +0.684 |
flood | +0.659 |
dolphin | +0.659 |
ending | +0.644 |
rant | +0.643 |
Neg logit contributions | |
ons | -1.038 |
aid | -0.768 |
on | -0.733 |
ages | -0.730 |
bes | -0.727 |
Token | , |
Token ID | 11 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
ksh | +0.012 |
rant | +0.011 |
flood | +0.011 |
dolphin | +0.011 |
whale | +0.011 |
Neg logit contributions | |
ons | -0.014 |
bes | -0.010 |
ages | -0.010 |
on | -0.009 |
es | -0.009 |
Token | but |
Token ID | 475 |
Feature activation | +5.49e+00 |
Pos logit contributions | |
ksh | +0.005 |
rant | +0.004 |
flood | +0.004 |
dolphin | +0.004 |
iet | +0.004 |
Neg logit contributions | |
ons | -0.011 |
bes | -0.009 |
ages | -0.009 |
on | -0.008 |
con | -0.008 |
Token | there |
Token ID | 612 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
flood | +0.334 |
rant | +0.327 |
ksh | +0.324 |
dolphin | +0.322 |
coming | +0.322 |
Neg logit contributions | |
ons | -0.388 |
bes | -0.308 |
ages | -0.285 |
con | -0.266 |
seek | -0.263 |
Token | wasn |
Token ID | 2492 |
Feature activation | +3.75e+00 |
Pos logit contributions | |
ksh | +0.079 |
flood | +0.077 |
rant | +0.077 |
dolphin | +0.075 |
whale | +0.074 |
Neg logit contributions | |
ons | -0.115 |
bes | -0.091 |
ages | -0.087 |
on | -0.082 |
con | -0.081 |
Token | . |
Token ID | 13 |
Feature activation | +3.37e-01 |
Pos logit contributions | |
ksh | +0.059 |
rant | +0.058 |
flood | +0.057 |
whale | +0.055 |
dolphin | +0.055 |
Neg logit contributions | |
ons | -0.068 |
bes | -0.053 |
ages | -0.050 |
es | -0.047 |
on | -0.047 |
Token | She |
Token ID | 1375 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
ksh | +0.018 |
rant | +0.017 |
ending | +0.017 |
flood | +0.017 |
beginning | +0.017 |
Neg logit contributions | |
ons | -0.027 |
bes | -0.021 |
ages | -0.019 |
aid | -0.019 |
on | -0.019 |
Token | grabbed |
Token ID | 13176 |
Feature activation | +8.96e-01 |
Pos logit contributions | |
ons | +0.141 |
bes | +0.112 |
ages | +0.107 |
on | +0.102 |
con | +0.100 |
Neg logit contributions | |
ksh | -0.086 |
rant | -0.085 |
flood | -0.085 |
dolphin | -0.083 |
whale | -0.081 |
Token | the |
Token ID | 262 |
Feature activation | -9.73e-01 |
Pos logit contributions | |
rant | +0.048 |
flood | +0.047 |
ksh | +0.046 |
dolphin | +0.046 |
coming | +0.045 |
Neg logit contributions | |
ons | -0.069 |
bes | -0.056 |
ages | -0.055 |
seek | -0.052 |
con | -0.050 |
Token | fire |
Token ID | 2046 |
Feature activation | +6.04e+00 |
Pos logit contributions | |
ons | +0.085 |
bes | +0.068 |
ages | +0.066 |
on | +0.064 |
con | +0.062 |
Neg logit contributions | |
ksh | -0.044 |
rant | -0.042 |
iet | -0.039 |
dolphin | -0.039 |
flood | -0.038 |
Token | extingu |
Token ID | 24995 |
Feature activation | +1.27e+01 |
Pos logit contributions | |
ksh | +0.324 |
rant | +0.320 |
flood | +0.314 |
dolphin | +0.307 |
whale | +0.298 |
Neg logit contributions | |
ons | -0.467 |
bes | -0.367 |
ages | -0.359 |
on | -0.335 |
es | -0.333 |
Token | isher |
Token ID | 4828 |
Feature activation | -2.37e-02 |
Pos logit contributions | |
ksh | +0.570 |
flood | +0.566 |
rant | +0.558 |
dolphin | +0.545 |
whale | +0.528 |
Neg logit contributions | |
ons | -1.106 |
bes | -0.880 |
ages | -0.853 |
on | -0.813 |
aid | -0.807 |
Token | and |
Token ID | 290 |
Feature activation | -2.81e-02 |
Pos logit contributions | |
ons | +0.002 |
bes | +0.002 |
ages | +0.002 |
aid | +0.001 |
con | +0.001 |
Neg logit contributions | |
flood | -0.001 |
ksh | -0.001 |
rant | -0.001 |
coming | -0.001 |
dolphin | -0.001 |
Token | put |
Token ID | 1234 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
ons | +0.002 |
bes | +0.002 |
seek | +0.002 |
ages | +0.002 |
con | +0.002 |
Neg logit contributions | |
flood | -0.001 |
dolphin | -0.001 |
coming | -0.001 |
ksh | -0.001 |
whale | -0.001 |
Token | out |
Token ID | 503 |
Feature activation | +6.83e+00 |
Pos logit contributions | |
rant | +0.062 |
ksh | +0.061 |
flood | +0.060 |
coming | +0.060 |
dolphin | +0.058 |
Neg logit contributions | |
ons | -0.088 |
bes | -0.071 |
ages | -0.070 |
seek | -0.066 |
con | -0.064 |
Token | the |
Token ID | 262 |
Feature activation | -9.34e-01 |
Pos logit contributions | |
flood | +0.372 |
rant | +0.372 |
coming | +0.366 |
ksh | +0.364 |
dolphin | +0.360 |
Neg logit contributions | |
ons | -0.506 |
bes | -0.410 |
ages | -0.400 |
con | -0.369 |
seek | -0.366 |
Token | dolls |
Token ID | 36062 |
Feature activation | -6.51e+00 |
Pos logit contributions | |
ons | +0.067 |
bes | +0.052 |
ages | +0.050 |
on | +0.047 |
con | +0.047 |
Neg logit contributions | |
ksh | -0.051 |
flood | -0.050 |
rant | -0.050 |
dolphin | -0.049 |
whale | -0.048 |
Token | and |
Token ID | 290 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
ons | +0.534 |
bes | +0.415 |
ages | +0.406 |
con | +0.387 |
aid | +0.379 |
Neg logit contributions | |
rant | -0.324 |
ksh | -0.324 |
flood | -0.319 |
dolphin | -0.310 |
coming | -0.306 |
Token | pretend |
Token ID | 16614 |
Feature activation | +1.89e+00 |
Pos logit contributions | |
flood | +0.010 |
ksh | +0.010 |
rant | +0.010 |
dolphin | +0.009 |
whale | +0.009 |
Neg logit contributions | |
ons | -0.018 |
bes | -0.015 |
ages | -0.014 |
on | -0.014 |
con | -0.013 |
Token | they |
Token ID | 484 |
Feature activation | +3.31e+00 |
Pos logit contributions | |
ksh | +0.126 |
rant | +0.123 |
flood | +0.121 |
dolphin | +0.117 |
whale | +0.116 |
Neg logit contributions | |
ons | -0.128 |
bes | -0.095 |
ages | -0.089 |
on | -0.084 |
es | -0.082 |
Token | were |
Token ID | 547 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
ksh | +0.192 |
rant | +0.179 |
flood | +0.179 |
dolphin | +0.176 |
whale | +0.174 |
Neg logit contributions | |
ons | -0.256 |
bes | -0.198 |
ages | -0.187 |
es | -0.187 |
on | -0.178 |
Token | mer |
Token ID | 4017 |
Feature activation | +1.26e+01 |
Pos logit contributions | |
ksh | +0.073 |
rant | +0.070 |
flood | +0.067 |
whale | +0.064 |
dolphin | +0.064 |
Neg logit contributions | |
ons | -0.096 |
bes | -0.075 |
ages | -0.072 |
es | -0.069 |
on | -0.068 |
Token | maid |
Token ID | 23151 |
Feature activation | +8.06e+00 |
Pos logit contributions | |
ending | +0.533 |
flood | +0.532 |
storm | +0.499 |
coming | +0.464 |
Why | +0.462 |
Neg logit contributions | |
ons | -1.014 |
bes | -0.950 |
es | -0.910 |
ages | -0.895 |
aged | -0.850 |
Token | s |
Token ID | 82 |
Feature activation | -3.01e+00 |
Pos logit contributions | |
flood | +0.569 |
rant | +0.565 |
ksh | +0.551 |
ending | +0.544 |
tide | +0.532 |
Neg logit contributions | |
ons | -0.503 |
ages | -0.358 |
bes | -0.357 |
es | -0.354 |
on | -0.318 |
Token | or |
Token ID | 393 |
Feature activation | +3.72e+00 |
Pos logit contributions | |
ons | +0.199 |
bes | +0.152 |
ages | +0.144 |
es | +0.141 |
on | +0.135 |
Neg logit contributions | |
ksh | -0.196 |
rant | -0.191 |
flood | -0.190 |
dolphin | -0.182 |
whale | -0.180 |
Token | pirates |
Token ID | 27516 |
Feature activation | +4.37e+00 |
Pos logit contributions | |
ksh | +0.202 |
rant | +0.198 |
flood | +0.195 |
dolphin | +0.186 |
whale | +0.184 |
Neg logit contributions | |
ons | -0.290 |
bes | -0.229 |
ages | -0.219 |
on | -0.206 |
con | -0.203 |
Token | . |
Token ID | 13 |
Feature activation | +6.96e-01 |
Pos logit contributions | |
ksh | +0.240 |
rant | +0.236 |
flood | +0.235 |
dolphin | +0.228 |
whale | +0.224 |
Neg logit contributions | |
ons | -0.334 |
bes | -0.262 |
ages | -0.251 |
on | -0.235 |
con | -0.232 |
Token | . |
Token ID | 13 |
Feature activation | +6.09e-01 |
Pos logit contributions | |
ksh | +0.019 |
rant | +0.019 |
flood | +0.019 |
dolphin | +0.018 |
whale | +0.018 |
Neg logit contributions | |
ons | -0.022 |
bes | -0.017 |
ages | -0.017 |
on | -0.016 |
es | -0.015 |
Token | |
Token ID | 220 |
Feature activation | -3.59e+00 |
Pos logit contributions | |
ksh | +0.032 |
dolphin | +0.029 |
rant | +0.029 |
flood | +0.029 |
Why | +0.029 |
Neg logit contributions | |
ons | -0.048 |
bes | -0.039 |
ages | -0.037 |
aid | -0.035 |
igs | -0.034 |
Token |
|
Token ID | 198 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
ons | +0.267 |
bes | +0.213 |
ages | +0.206 |
con | +0.192 |
seek | +0.189 |
Neg logit contributions | |
ksh | -0.191 |
rant | -0.186 |
flood | -0.179 |
coming | -0.179 |
dolphin | -0.176 |
Token | The |
Token ID | 464 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
Why | +0.009 |
dolphin | +0.009 |
rant | +0.009 |
iet | +0.009 |
coming | +0.009 |
Neg logit contributions | |
ons | -0.017 |
ages | -0.015 |
bes | -0.014 |
aces | -0.013 |
igs | -0.013 |
Token | strong |
Token ID | 1913 |
Feature activation | -2.81e+00 |
Pos logit contributions | |
ons | +0.103 |
bes | +0.088 |
on | +0.082 |
aid | +0.081 |
seek | +0.078 |
Neg logit contributions | |
ksh | -0.047 |
extingu | -0.038 |
rant | -0.037 |
OO | -0.033 |
king | -0.032 |
Token | mer |
Token ID | 4017 |
Feature activation | +1.26e+01 |
Pos logit contributions | |
ons | +0.209 |
bes | +0.162 |
on | +0.152 |
es | +0.144 |
exclaimed | +0.144 |
Neg logit contributions | |
ksh | -0.174 |
extingu | -0.147 |
rant | -0.140 |
flood | -0.137 |
ISE | -0.134 |
Token | maid |
Token ID | 23151 |
Feature activation | +8.03e+00 |
Pos logit contributions | |
flood | +0.468 |
ksh | +0.449 |
extingu | +0.386 |
rant | +0.377 |
coming | +0.377 |
Neg logit contributions | |
ons | -1.153 |
bes | -1.020 |
es | -0.971 |
ages | -0.961 |
aged | -0.924 |
Token | thought |
Token ID | 1807 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
ksh | +0.411 |
dolphin | +0.407 |
rant | +0.407 |
flood | +0.407 |
whale | +0.402 |
Neg logit contributions | |
ons | -0.636 |
bes | -0.509 |
ages | -0.480 |
on | -0.456 |
con | -0.453 |
Token | about |
Token ID | 546 |
Feature activation | +2.87e+00 |
Pos logit contributions | |
ons | +0.092 |
bes | +0.073 |
ages | +0.070 |
on | +0.065 |
con | +0.064 |
Neg logit contributions | |
ksh | -0.065 |
rant | -0.064 |
flood | -0.064 |
dolphin | -0.062 |
whale | -0.061 |
Token | the |
Token ID | 262 |
Feature activation | -6.75e-01 |
Pos logit contributions | |
ksh | +0.148 |
flood | +0.147 |
rant | +0.146 |
dolphin | +0.145 |
whale | +0.142 |
Neg logit contributions | |
ons | -0.227 |
bes | -0.181 |
ages | -0.176 |
on | -0.162 |
con | -0.161 |
Token | promise |
Token ID | 6991 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
ons | +0.046 |
skills | +0.037 |
bes | +0.036 |
powers | +0.035 |
friends | +0.035 |
Neg logit contributions | |
ksh | -0.038 |
rant | -0.036 |
extingu | -0.034 |
ount | -0.031 |
OO | -0.031 |
Token | They |
Token ID | 2990 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
dolphin | +0.010 |
Why | +0.010 |
iet | +0.010 |
Can | +0.010 |
whale | +0.009 |
Neg logit contributions | |
ons | -0.016 |
aces | -0.013 |
bes | -0.013 |
con | -0.013 |
ages | -0.012 |
Token | both |
Token ID | 1111 |
Feature activation | +7.03e-01 |
Pos logit contributions | |
ksh | +0.079 |
rant | +0.079 |
flood | +0.078 |
dolphin | +0.076 |
whale | +0.074 |
Neg logit contributions | |
ons | -0.128 |
bes | -0.104 |
ages | -0.098 |
con | -0.093 |
seek | -0.092 |
Token | skipped |
Token ID | 26684 |
Feature activation | -4.13e+00 |
Pos logit contributions | |
ksh | +0.042 |
rant | +0.041 |
flood | +0.041 |
dolphin | +0.039 |
whale | +0.039 |
Neg logit contributions | |
ons | -0.051 |
bes | -0.039 |
ages | -0.038 |
on | -0.035 |
con | -0.035 |
Token | and |
Token ID | 290 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
ons | +0.351 |
bes | +0.281 |
ages | +0.270 |
on | +0.254 |
con | +0.251 |
Neg logit contributions | |
ksh | -0.197 |
flood | -0.195 |
rant | -0.194 |
dolphin | -0.191 |
mate | -0.183 |
Token | jumped |
Token ID | 11687 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
flood | +0.010 |
dolphin | +0.010 |
ksh | +0.010 |
rant | +0.009 |
coming | +0.009 |
Neg logit contributions | |
ons | -0.019 |
bes | -0.016 |
ages | -0.016 |
powers | -0.015 |
aid | -0.015 |
Token | mer |
Token ID | 4017 |
Feature activation | +1.26e+01 |
Pos logit contributions | |
ksh | +0.083 |
rant | +0.081 |
flood | +0.080 |
dolphin | +0.078 |
whale | +0.077 |
Neg logit contributions | |
ons | -0.109 |
bes | -0.085 |
ages | -0.081 |
on | -0.076 |
con | -0.076 |
Token | rily |
Token ID | 28904 |
Feature activation | +2.39e+00 |
Pos logit contributions | |
flood | +0.619 |
ksh | +0.584 |
ending | +0.583 |
rant | +0.573 |
coming | +0.559 |
Neg logit contributions | |
ons | -1.002 |
bes | -0.869 |
ages | -0.822 |
es | -0.796 |
aged | -0.755 |
Token | as |
Token ID | 355 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
ksh | +0.131 |
rant | +0.128 |
flood | +0.128 |
dolphin | +0.124 |
whale | +0.122 |
Neg logit contributions | |
ons | -0.183 |
bes | -0.144 |
ages | -0.138 |
on | -0.130 |
con | -0.128 |
Token | they |
Token ID | 484 |
Feature activation | +3.33e+00 |
Pos logit contributions | |
ons | +0.017 |
bes | +0.014 |
ages | +0.013 |
con | +0.012 |
seek | +0.012 |
Neg logit contributions | |
flood | -0.012 |
whale | -0.012 |
dolphin | -0.012 |
rant | -0.012 |
tide | -0.012 |
Token | made |
Token ID | 925 |
Feature activation | +4.22e+00 |
Pos logit contributions | |
ksh | +0.167 |
rant | +0.164 |
flood | +0.163 |
dolphin | +0.157 |
whale | +0.155 |
Neg logit contributions | |
ons | -0.271 |
bes | -0.217 |
ages | -0.208 |
on | -0.196 |
con | -0.194 |
Token | their |
Token ID | 511 |
Feature activation | +6.56e+00 |
Pos logit contributions | |
ksh | +0.213 |
rant | +0.210 |
flood | +0.206 |
dolphin | +0.197 |
whale | +0.194 |
Neg logit contributions | |
ons | -0.346 |
bes | -0.273 |
ages | -0.264 |
on | -0.250 |
con | -0.246 |
Token | , |
Token ID | 11 |
Feature activation | +8.35e-02 |
Pos logit contributions | |
rant | +0.121 |
ksh | +0.120 |
flood | +0.119 |
coming | +0.116 |
dolphin | +0.115 |
Neg logit contributions | |
ons | -0.190 |
bes | -0.152 |
ages | -0.147 |
aid | -0.140 |
on | -0.135 |
Token | and |
Token ID | 290 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
ksh | +0.004 |
rant | +0.004 |
flood | +0.004 |
iet | +0.004 |
dolphin | +0.004 |
Neg logit contributions | |
ons | -0.007 |
bes | -0.005 |
ages | -0.005 |
on | -0.005 |
con | -0.005 |
Token | she |
Token ID | 673 |
Feature activation | -2.69e+00 |
Pos logit contributions | |
flood | +0.008 |
rant | +0.008 |
ksh | +0.008 |
dolphin | +0.008 |
coming | +0.008 |
Neg logit contributions | |
ons | -0.011 |
bes | -0.008 |
ages | -0.008 |
on | -0.008 |
aid | -0.007 |
Token | wanted |
Token ID | 2227 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
ons | +0.230 |
bes | +0.185 |
ages | +0.178 |
on | +0.169 |
con | +0.167 |
Neg logit contributions | |
ksh | -0.125 |
rant | -0.123 |
flood | -0.122 |
dolphin | -0.118 |
whale | -0.116 |
Token | the |
Token ID | 262 |
Feature activation | -7.58e-01 |
Pos logit contributions | |
ksh | +0.100 |
rant | +0.097 |
flood | +0.097 |
dolphin | +0.091 |
whale | +0.091 |
Neg logit contributions | |
ons | -0.110 |
bes | -0.081 |
ages | -0.078 |
on | -0.074 |
seek | -0.073 |
Token | mer |
Token ID | 4017 |
Feature activation | +1.25e+01 |
Pos logit contributions | |
ons | +0.054 |
skills | +0.043 |
bes | +0.042 |
on | +0.041 |
powers | +0.040 |
Neg logit contributions | |
ksh | -0.042 |
rant | -0.040 |
extingu | -0.036 |
ount | -0.035 |
OO | -0.034 |
Token | maid |
Token ID | 23151 |
Feature activation | +7.99e+00 |
Pos logit contributions | |
flood | +0.460 |
ksh | +0.436 |
extingu | +0.400 |
Why | +0.398 |
coming | +0.373 |
Neg logit contributions | |
ons | -1.143 |
bes | -1.016 |
ages | -0.948 |
es | -0.939 |
seek | -0.896 |
Token | to |
Token ID | 284 |
Feature activation | -2.18e-01 |
Pos logit contributions | |
ksh | +0.414 |
rant | +0.408 |
flood | +0.402 |
dolphin | +0.401 |
whale | +0.390 |
Neg logit contributions | |
ons | -0.632 |
bes | -0.505 |
ages | -0.473 |
on | -0.451 |
con | -0.450 |
Token | come |
Token ID | 1282 |
Feature activation | +7.62e+00 |
Pos logit contributions | |
ons | +0.018 |
bes | +0.015 |
ages | +0.014 |
on | +0.014 |
seek | +0.013 |
Neg logit contributions | |
ksh | -0.011 |
rant | -0.010 |
flood | -0.010 |
whale | -0.009 |
dolphin | -0.009 |
Token | too |
Token ID | 1165 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
ksh | +0.442 |
flood | +0.439 |
rant | +0.424 |
whale | +0.416 |
dolphin | +0.403 |
Neg logit contributions | |
ons | -0.538 |
bes | -0.423 |
ages | -0.403 |
es | -0.401 |
on | -0.387 |
Token | . |
Token ID | 13 |
Feature activation | +6.09e-01 |
Pos logit contributions | |
ksh | +0.019 |
rant | +0.019 |
flood | +0.019 |
dolphin | +0.018 |
whale | +0.018 |
Neg logit contributions | |
ons | -0.022 |
bes | -0.017 |
ages | -0.017 |
on | -0.016 |
es | -0.015 |
Token | she |
Token ID | 673 |
Feature activation | -2.73e+00 |
Pos logit contributions | |
ons | +0.149 |
bes | +0.128 |
ages | +0.127 |
aid | +0.115 |
seek | +0.109 |
Neg logit contributions | |
flood | -0.115 |
ending | -0.115 |
dolphin | -0.114 |
coming | -0.113 |
tide | -0.109 |
Token | was |
Token ID | 373 |
Feature activation | +1.98e-01 |
Pos logit contributions | |
ons | +0.196 |
bes | +0.146 |
on | +0.140 |
es | +0.134 |
ages | +0.133 |
Neg logit contributions | |
ksh | -0.174 |
rant | -0.159 |
flood | -0.159 |
extingu | -0.151 |
mate | -0.150 |
Token | finished |
Token ID | 5201 |
Feature activation | -5.90e-01 |
Pos logit contributions | |
ksh | +0.014 |
rant | +0.013 |
flood | +0.013 |
extingu | +0.013 |
dolphin | +0.013 |
Neg logit contributions | |
ons | -0.012 |
bes | -0.009 |
on | -0.009 |
ages | -0.009 |
es | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | +8.35e-02 |
Pos logit contributions | |
ons | +0.039 |
bes | +0.030 |
ages | +0.029 |
on | +0.026 |
con | +0.026 |
Neg logit contributions | |
ksh | -0.038 |
flood | -0.037 |
rant | -0.037 |
dolphin | -0.037 |
whale | -0.036 |
Token | the |
Token ID | 262 |
Feature activation | -7.74e-01 |
Pos logit contributions | |
dolphin | +0.005 |
coming | +0.005 |
ending | +0.005 |
beginning | +0.005 |
flood | +0.005 |
Neg logit contributions | |
ons | -0.006 |
bes | -0.005 |
ages | -0.005 |
aid | -0.004 |
ittens | -0.004 |
Token | mer |
Token ID | 4017 |
Feature activation | +1.25e+01 |
Pos logit contributions | |
ons | +0.067 |
bes | +0.056 |
ages | +0.052 |
on | +0.051 |
aid | +0.050 |
Neg logit contributions | |
ksh | -0.037 |
rant | -0.032 |
extingu | -0.029 |
OO | -0.027 |
flood | -0.027 |
Token | maid |
Token ID | 23151 |
Feature activation | +7.96e+00 |
Pos logit contributions | |
ksh | +0.482 |
flood | +0.481 |
Why | +0.443 |
coming | +0.432 |
beginning | +0.415 |
Neg logit contributions | |
ons | -1.057 |
bes | -0.948 |
ages | -0.908 |
es | -0.886 |
aged | -0.865 |
Token | thanked |
Token ID | 26280 |
Feature activation | -4.09e+00 |
Pos logit contributions | |
ksh | +0.494 |
rant | +0.486 |
flood | +0.484 |
dolphin | +0.468 |
whale | +0.462 |
Neg logit contributions | |
ons | -0.554 |
bes | -0.422 |
ages | -0.402 |
on | -0.375 |
con | -0.369 |
Token | her |
Token ID | 607 |
Feature activation | -9.38e-01 |
Pos logit contributions | |
ons | +0.300 |
bes | +0.241 |
ages | +0.235 |
con | +0.210 |
aid | +0.209 |
Neg logit contributions | |
rant | -0.228 |
ksh | -0.225 |
dolphin | -0.225 |
flood | -0.221 |
whale | -0.218 |
Token | for |
Token ID | 329 |
Feature activation | +2.27e+00 |
Pos logit contributions | |
ons | +0.081 |
bes | +0.065 |
ages | +0.063 |
on | +0.059 |
seek | +0.058 |
Neg logit contributions | |
ksh | -0.043 |
rant | -0.042 |
flood | -0.042 |
dolphin | -0.042 |
whale | -0.041 |
Token | coming |
Token ID | 2406 |
Feature activation | +1.12e+01 |
Pos logit contributions | |
flood | +0.121 |
dolphin | +0.121 |
coming | +0.121 |
rant | +0.119 |
whale | +0.117 |
Neg logit contributions | |
ons | -0.173 |
bes | -0.139 |
ages | -0.135 |
con | -0.125 |
seek | -0.122 |
Token | They |
Token ID | 1119 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
rant | +0.030 |
flood | +0.030 |
ksh | +0.029 |
dolphin | +0.029 |
whale | +0.028 |
Neg logit contributions | |
ons | -0.038 |
bes | -0.031 |
ages | -0.030 |
con | -0.028 |
seek | -0.028 |
Token | had |
Token ID | 550 |
Feature activation | +7.30e-01 |
Pos logit contributions | |
ksh | +0.114 |
rant | +0.106 |
flood | +0.106 |
extingu | +0.102 |
ending | +0.100 |
Neg logit contributions | |
ons | -0.127 |
bes | -0.095 |
ages | -0.094 |
on | -0.089 |
aid | -0.089 |
Token | never |
Token ID | 1239 |
Feature activation | -4.42e-01 |
Pos logit contributions | |
ksh | +0.045 |
rant | +0.044 |
flood | +0.042 |
dolphin | +0.041 |
whale | +0.040 |
Neg logit contributions | |
ons | -0.053 |
bes | -0.040 |
ages | -0.039 |
seek | -0.036 |
on | -0.035 |
Token | seen |
Token ID | 1775 |
Feature activation | +7.42e-01 |
Pos logit contributions | |
ons | +0.035 |
ages | +0.027 |
bes | +0.027 |
seek | +0.025 |
on | +0.025 |
Neg logit contributions | |
ksh | -0.024 |
rant | -0.023 |
flood | -0.023 |
dolphin | -0.022 |
whale | -0.022 |
Token | a |
Token ID | 257 |
Feature activation | -8.58e-01 |
Pos logit contributions | |
ksh | +0.046 |
flood | +0.045 |
rant | +0.044 |
ending | +0.042 |
iet | +0.042 |
Neg logit contributions | |
ons | -0.054 |
on | -0.039 |
bes | -0.038 |
con | -0.037 |
es | -0.036 |
Token | mer |
Token ID | 4017 |
Feature activation | +1.25e+01 |
Pos logit contributions | |
ons | +0.060 |
on | +0.049 |
con | +0.046 |
bes | +0.045 |
powers | +0.044 |
Neg logit contributions | |
ksh | -0.053 |
rant | -0.044 |
ount | -0.041 |
extingu | -0.041 |
latable | -0.037 |
Token | maid |
Token ID | 23151 |
Feature activation | +7.97e+00 |
Pos logit contributions | |
ksh | +0.552 |
Why | +0.538 |
flood | +0.531 |
ending | +0.490 |
rant | +0.478 |
Neg logit contributions | |
ons | -0.931 |
bes | -0.851 |
es | -0.781 |
ages | -0.764 |
powers | -0.758 |
Token | before |
Token ID | 878 |
Feature activation | +5.32e-01 |
Pos logit contributions | |
ksh | +0.437 |
flood | +0.428 |
rant | +0.421 |
dolphin | +0.396 |
iet | +0.393 |
Neg logit contributions | |
ons | -0.615 |
bes | -0.480 |
ages | -0.464 |
on | -0.454 |
es | -0.444 |
Token | . |
Token ID | 13 |
Feature activation | +5.93e-01 |
Pos logit contributions | |
ksh | +0.033 |
rant | +0.032 |
flood | +0.032 |
dolphin | +0.030 |
mate | +0.030 |
Neg logit contributions | |
ons | -0.038 |
bes | -0.029 |
ages | -0.027 |
on | -0.026 |
es | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
ksh | +0.034 |
rant | +0.034 |
flood | +0.034 |
dolphin | +0.033 |
whale | +0.033 |
Neg logit contributions | |
ons | -0.043 |
bes | -0.034 |
ages | -0.032 |
on | -0.030 |
con | -0.030 |
Token |
|
Token ID | 198 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
dolphin | +0.009 |
whale | +0.008 |
flood | +0.008 |
rant | +0.008 |
ksh | +0.008 |
Neg logit contributions | |
ons | -0.015 |
bes | -0.012 |
ages | -0.012 |
aces | -0.011 |
con | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | +9.89e-02 |
Pos logit contributions | |
ons | +0.294 |
bes | +0.245 |
ages | +0.231 |
seek | +0.220 |
tail | +0.219 |
Neg logit contributions | |
ksh | -0.165 |
rant | -0.157 |
coming | -0.155 |
mate | -0.151 |
flood | -0.149 |
Token |
|
Token ID | 198 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
dolphin | +0.003 |
Why | +0.003 |
rant | +0.003 |
ksh | +0.003 |
coming | +0.003 |
Neg logit contributions | |
ons | -0.009 |
bes | -0.007 |
ages | -0.007 |
aces | -0.007 |
elbow | -0.007 |
Token | S |
Token ID | 50 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
dolphin | +0.006 |
ksh | +0.006 |
flood | +0.006 |
iet | +0.006 |
Why | +0.006 |
Neg logit contributions | |
ons | -0.009 |
bes | -0.007 |
ages | -0.006 |
on | -0.006 |
aces | -0.006 |
Token | ki |
Token ID | 4106 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
ksh | +0.100 |
rant | +0.098 |
flood | +0.098 |
dolphin | +0.095 |
whale | +0.093 |
Neg logit contributions | |
ons | -0.138 |
bes | -0.107 |
ages | -0.103 |
on | -0.097 |
con | -0.096 |
Token | pping |
Token ID | 2105 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
ons | +0.126 |
bes | +0.106 |
ages | +0.105 |
powers | +0.102 |
aid | +0.099 |
Neg logit contributions | |
rant | -0.078 |
iet | -0.071 |
whale | -0.071 |
coming | -0.070 |
dolphin | -0.070 |
Token | mer |
Token ID | 4017 |
Feature activation | +1.25e+01 |
Pos logit contributions | |
coming | +0.069 |
dolphin | +0.068 |
flood | +0.067 |
rant | +0.067 |
whale | +0.064 |
Neg logit contributions | |
ons | -0.122 |
bes | -0.106 |
ages | -0.096 |
seek | -0.093 |
aid | -0.092 |
Token | rily |
Token ID | 28904 |
Feature activation | +2.29e+00 |
Pos logit contributions | |
flood | +0.775 |
ksh | +0.765 |
rant | +0.750 |
dolphin | +0.726 |
ending | +0.726 |
Neg logit contributions | |
ons | -0.872 |
bes | -0.678 |
ages | -0.647 |
es | -0.607 |
on | -0.582 |
Token | along |
Token ID | 1863 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
flood | +0.107 |
dolphin | +0.107 |
rant | +0.106 |
coming | +0.105 |
ending | +0.104 |
Neg logit contributions | |
ons | -0.188 |
bes | -0.148 |
ages | -0.143 |
aid | -0.138 |
on | -0.136 |
Token | , |
Token ID | 11 |
Feature activation | +8.98e-02 |
Pos logit contributions | |
ons | +0.104 |
bes | +0.081 |
ages | +0.080 |
aid | +0.075 |
on | +0.072 |
Neg logit contributions | |
coming | -0.075 |
flood | -0.074 |
rant | -0.073 |
dolphin | -0.072 |
whale | -0.071 |
Token | Hawk |
Token ID | 26698 |
Feature activation | -9.52e-01 |
Pos logit contributions | |
ksh | +0.005 |
rant | +0.005 |
flood | +0.005 |
dolphin | +0.005 |
whale | +0.004 |
Neg logit contributions | |
ons | -0.007 |
bes | -0.005 |
ages | -0.005 |
on | -0.005 |
con | -0.005 |
Token | came |
Token ID | 1625 |
Feature activation | +6.20e+00 |
Pos logit contributions | |
ons | +0.076 |
bes | +0.061 |
ages | +0.058 |
on | +0.054 |
aid | +0.054 |
Neg logit contributions | |
rant | -0.048 |
ksh | -0.048 |
flood | -0.048 |
dolphin | -0.046 |
whale | -0.046 |
Token | castle |
Token ID | 16669 |
Feature activation | +9.44e-01 |
Pos logit contributions | |
ons | +0.105 |
bes | +0.084 |
ages | +0.080 |
on | +0.075 |
con | +0.074 |
Neg logit contributions | |
ksh | -0.074 |
rant | -0.071 |
flood | -0.070 |
dolphin | -0.066 |
whale | -0.065 |
Token | so |
Token ID | 523 |
Feature activation | -4.22e-01 |
Pos logit contributions | |
flood | +0.046 |
rant | +0.045 |
ksh | +0.045 |
coming | +0.044 |
ending | +0.043 |
Neg logit contributions | |
ons | -0.077 |
bes | -0.062 |
ages | -0.060 |
con | -0.056 |
aid | -0.056 |
Token | much |
Token ID | 881 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
ons | +0.030 |
bes | +0.022 |
ages | +0.021 |
on | +0.021 |
es | +0.020 |
Neg logit contributions | |
ksh | -0.026 |
rant | -0.024 |
extingu | -0.023 |
flood | -0.023 |
mate | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | +5.30e-01 |
Pos logit contributions | |
ons | +0.142 |
bes | +0.099 |
con | +0.097 |
ages | +0.094 |
on | +0.092 |
Neg logit contributions | |
ksh | -0.154 |
rant | -0.142 |
flood | -0.138 |
iet | -0.137 |
mate | -0.135 |
Token | The |
Token ID | 383 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
ksh | +0.032 |
rant | +0.031 |
flood | +0.031 |
dolphin | +0.030 |
whale | +0.030 |
Neg logit contributions | |
ons | -0.039 |
bes | -0.030 |
ages | -0.028 |
on | -0.026 |
con | -0.026 |
Token | mer |
Token ID | 4017 |
Feature activation | +1.25e+01 |
Pos logit contributions | |
ons | +0.108 |
bes | +0.092 |
on | +0.087 |
aid | +0.085 |
ages | +0.085 |
Neg logit contributions | |
ksh | -0.036 |
rant | -0.027 |
extingu | -0.027 |
flood | -0.024 |
OO | -0.022 |
Token | maid |
Token ID | 23151 |
Feature activation | +7.95e+00 |
Pos logit contributions | |
flood | +0.436 |
ksh | +0.421 |
coming | +0.353 |
extingu | +0.352 |
Why | +0.349 |
Neg logit contributions | |
ons | -1.188 |
bes | -1.032 |
ages | -0.984 |
es | -0.983 |
aged | -0.946 |
Token | said |
Token ID | 531 |
Feature activation | -2.59e+00 |
Pos logit contributions | |
ksh | +0.540 |
rant | +0.534 |
flood | +0.532 |
dolphin | +0.523 |
whale | +0.516 |
Neg logit contributions | |
ons | -0.501 |
bes | -0.372 |
ages | -0.352 |
on | -0.322 |
con | -0.319 |
Token | , |
Token ID | 11 |
Feature activation | +8.98e-02 |
Pos logit contributions | |
ons | +0.168 |
bes | +0.124 |
ages | +0.118 |
on | +0.110 |
es | +0.107 |
Neg logit contributions | |
ksh | -0.176 |
rant | -0.173 |
flood | -0.172 |
dolphin | -0.166 |
whale | -0.164 |
Token | " |
Token ID | 366 |
Feature activation | +3.05e+00 |
Pos logit contributions | |
flood | +0.004 |
ksh | +0.004 |
rant | +0.004 |
dolphin | +0.004 |
iet | +0.004 |
Neg logit contributions | |
ons | -0.007 |
bes | -0.006 |
ages | -0.006 |
on | -0.006 |
es | -0.005 |
Token | I |
Token ID | 40 |
Feature activation | +3.70e+00 |
Pos logit contributions | |
flood | +0.165 |
rant | +0.159 |
ksh | +0.158 |
ending | +0.152 |
mate | +0.150 |
Neg logit contributions | |
ons | -0.246 |
bes | -0.189 |
ages | -0.182 |
on | -0.181 |
aid | -0.172 |
Token | " |
Token ID | 1 |
Feature activation | -2.37e+00 |
Pos logit contributions | |
ksh | +0.005 |
flood | +0.005 |
dolphin | +0.005 |
rant | +0.005 |
mate | +0.005 |
Neg logit contributions | |
ons | -0.006 |
bes | -0.005 |
con | -0.005 |
ages | -0.005 |
seek | -0.004 |
Token | And |
Token ID | 1870 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
ons | +0.182 |
bes | +0.151 |
aid | +0.145 |
ages | +0.141 |
on | +0.138 |
Neg logit contributions | |
rant | -0.126 |
flood | -0.118 |
ksh | -0.117 |
ending | -0.112 |
iet | -0.111 |
Token | I |
Token ID | 314 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
flood | +0.082 |
dolphin | +0.082 |
ksh | +0.082 |
whale | +0.081 |
rant | +0.081 |
Neg logit contributions | |
ons | -0.119 |
bes | -0.094 |
ages | -0.089 |
con | -0.084 |
seek | -0.083 |
Token | 'm |
Token ID | 1101 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
ksh | +0.082 |
rant | +0.078 |
whale | +0.075 |
dolphin | +0.074 |
flood | +0.073 |
Neg logit contributions | |
ons | -0.085 |
bes | -0.066 |
ages | -0.066 |
on | -0.062 |
aid | -0.060 |
Token | a |
Token ID | 257 |
Feature activation | -8.97e-01 |
Pos logit contributions | |
ksh | +0.017 |
rant | +0.017 |
flood | +0.017 |
dolphin | +0.016 |
whale | +0.016 |
Neg logit contributions | |
ons | -0.017 |
bes | -0.013 |
ages | -0.012 |
on | -0.012 |
es | -0.011 |
Token | mer |
Token ID | 4017 |
Feature activation | +1.25e+01 |
Pos logit contributions | |
ons | +0.040 |
on | +0.032 |
bes | +0.031 |
ages | +0.030 |
powers | +0.029 |
Neg logit contributions | |
rant | -0.067 |
ksh | -0.065 |
ount | -0.064 |
appearance | -0.060 |
ility | -0.060 |
Token | maid |
Token ID | 23151 |
Feature activation | +7.93e+00 |
Pos logit contributions | |
flood | +0.524 |
ksh | +0.519 |
Why | +0.510 |
rant | +0.509 |
beginning | +0.506 |
Neg logit contributions | |
ons | -0.916 |
bes | -0.874 |
es | -0.830 |
ages | -0.820 |
aps | -0.812 |
Token | !" |
Token ID | 2474 |
Feature activation | -3.37e+00 |
Pos logit contributions | |
ksh | +0.447 |
mate | +0.444 |
dolphin | +0.443 |
whale | +0.437 |
rant | +0.429 |
Neg logit contributions | |
ons | -0.586 |
bes | -0.470 |
ages | -0.421 |
seek | -0.396 |
aid | -0.394 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
ons | +0.340 |
bes | +0.290 |
ages | +0.278 |
seek | +0.266 |
con | +0.265 |
Neg logit contributions | |
flood | -0.113 |
dolphin | -0.107 |
ending | -0.103 |
mate | -0.102 |
whale | -0.099 |
Token | said |
Token ID | 531 |
Feature activation | -2.58e+00 |
Pos logit contributions | |
ksh | +0.067 |
dolphin | +0.065 |
whale | +0.064 |
rant | +0.063 |
mate | +0.062 |
Neg logit contributions | |
ons | -0.074 |
bes | -0.055 |
ages | -0.051 |
igs | -0.050 |
con | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | +5.69e-01 |
Pos logit contributions | |
ons | +0.224 |
bes | +0.180 |
ages | +0.168 |
con | +0.166 |
rob | +0.162 |
Neg logit contributions | |
ksh | -0.119 |
coming | -0.114 |
rant | -0.113 |
dolphin | -0.111 |
flood | -0.110 |
Token | passport |
Token ID | 17981 |
Feature activation | +2.72e-01 |
Pos logit contributions | |
ons | +0.041 |
bes | +0.033 |
ages | +0.032 |
on | +0.030 |
con | +0.030 |
Neg logit contributions | |
ksh | -0.024 |
rant | -0.023 |
flood | -0.022 |
dolphin | -0.021 |
whale | -0.021 |
Token | belonged |
Token ID | 19611 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
flood | +0.012 |
coming | +0.011 |
whale | +0.011 |
ending | +0.011 |
wake | +0.011 |
Neg logit contributions | |
ons | -0.022 |
bes | -0.019 |
ages | -0.019 |
ittens | -0.018 |
seek | -0.018 |
Token | to |
Token ID | 284 |
Feature activation | -8.13e-02 |
Pos logit contributions | |
rant | +0.127 |
ksh | +0.126 |
mate | +0.125 |
coming | +0.124 |
flood | +0.124 |
Neg logit contributions | |
ons | -0.217 |
bes | -0.177 |
ages | -0.168 |
aid | -0.159 |
con | -0.157 |
Token | a |
Token ID | 257 |
Feature activation | -7.87e-01 |
Pos logit contributions | |
ons | +0.006 |
bes | +0.004 |
ages | +0.004 |
on | +0.004 |
es | +0.004 |
Neg logit contributions | |
ksh | -0.005 |
rant | -0.005 |
flood | -0.005 |
dolphin | -0.005 |
whale | -0.004 |
Token | magical |
Token ID | 10883 |
Feature activation | +4.63e+00 |
Pos logit contributions | |
ons | +0.058 |
bes | +0.045 |
ages | +0.044 |
con | +0.042 |
on | +0.042 |
Neg logit contributions | |
ksh | -0.045 |
rant | -0.045 |
iet | -0.039 |
extingu | -0.038 |
flood | -0.037 |
Token | mer |
Token ID | 4017 |
Feature activation | +1.24e+01 |
Pos logit contributions | |
ksh | +0.224 |
rant | +0.212 |
flood | +0.202 |
dolphin | +0.190 |
iet | +0.186 |
Neg logit contributions | |
ons | -0.394 |
bes | -0.316 |
ages | -0.301 |
on | -0.291 |
con | -0.290 |
Token | maid |
Token ID | 23151 |
Feature activation | +7.87e+00 |
Pos logit contributions | |
flood | +0.460 |
ksh | +0.451 |
Why | +0.408 |
rant | +0.406 |
ending | +0.399 |
Neg logit contributions | |
ons | -1.121 |
bes | -0.975 |
ages | -0.932 |
es | -0.912 |
powers | -0.871 |
Token | who |
Token ID | 508 |
Feature activation | +3.98e+00 |
Pos logit contributions | |
rant | +0.382 |
ksh | +0.380 |
dolphin | +0.378 |
flood | +0.373 |
whale | +0.370 |
Neg logit contributions | |
ons | -0.637 |
bes | -0.515 |
ages | -0.489 |
aid | -0.462 |
con | -0.458 |
Token | had |
Token ID | 550 |
Feature activation | +5.78e-01 |
Pos logit contributions | |
rant | +0.181 |
flood | +0.178 |
dolphin | +0.177 |
whale | +0.175 |
ksh | +0.170 |
Neg logit contributions | |
ons | -0.331 |
bes | -0.276 |
ages | -0.266 |
con | -0.249 |
seek | -0.246 |
Token | been |
Token ID | 587 |
Feature activation | +9.56e-01 |
Pos logit contributions | |
rant | +0.027 |
flood | +0.027 |
ksh | +0.027 |
dolphin | +0.027 |
whale | +0.026 |
Neg logit contributions | |
ons | -0.048 |
bes | -0.039 |
ages | -0.037 |
on | -0.035 |
con | -0.035 |
Token | visiting |
Token ID | 10013 |
Feature activation | +2.87e-01 |
Pos logit contributions | |
ksh | +0.049 |
rant | +0.048 |
flood | +0.048 |
dolphin | +0.046 |
whale | +0.045 |
Neg logit contributions | |
ons | -0.077 |
bes | -0.061 |
ages | -0.059 |
on | -0.055 |
con | -0.055 |
Token | . |
Token ID | 13 |
Feature activation | +9.60e-01 |
Pos logit contributions | |
ons | +0.157 |
bes | +0.121 |
ages | +0.116 |
on | +0.107 |
con | +0.106 |
Neg logit contributions | |
ksh | -0.133 |
flood | -0.131 |
rant | -0.131 |
dolphin | -0.128 |
whale | -0.126 |
Token | The |
Token ID | 383 |
Feature activation | -7.84e-01 |
Pos logit contributions | |
ksh | +0.047 |
mate | +0.044 |
rant | +0.043 |
flood | +0.043 |
dolphin | +0.043 |
Neg logit contributions | |
ons | -0.081 |
bes | -0.065 |
ages | -0.063 |
con | -0.059 |
seek | -0.058 |
Token | more |
Token ID | 517 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
ons | +0.066 |
bes | +0.053 |
ages | +0.050 |
on | +0.048 |
con | +0.048 |
Neg logit contributions | |
ksh | -0.038 |
rant | -0.037 |
flood | -0.035 |
dolphin | -0.033 |
whale | -0.032 |
Token | , |
Token ID | 11 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
ons | +0.099 |
bes | +0.080 |
ages | +0.077 |
seek | +0.074 |
aid | +0.072 |
Neg logit contributions | |
flood | -0.075 |
whale | -0.073 |
dolphin | -0.072 |
mate | -0.072 |
ending | -0.070 |
Token | the |
Token ID | 262 |
Feature activation | -7.55e-01 |
Pos logit contributions | |
ksh | +0.009 |
rant | +0.009 |
flood | +0.009 |
dolphin | +0.009 |
whale | +0.008 |
Neg logit contributions | |
ons | -0.013 |
bes | -0.010 |
ages | -0.009 |
on | -0.009 |
con | -0.009 |
Token | mer |
Token ID | 4017 |
Feature activation | +1.24e+01 |
Pos logit contributions | |
ons | +0.062 |
bes | +0.049 |
ages | +0.047 |
on | +0.045 |
con | +0.045 |
Neg logit contributions | |
ksh | -0.038 |
rant | -0.037 |
flood | -0.035 |
dolphin | -0.034 |
whale | -0.033 |
Token | rier |
Token ID | 5277 |
Feature activation | +3.28e+00 |
Pos logit contributions | |
flood | +0.387 |
ksh | +0.384 |
rant | +0.341 |
Why | +0.330 |
coming | +0.326 |
Neg logit contributions | |
ons | -1.220 |
bes | -1.046 |
ages | -1.018 |
es | -0.990 |
con | -0.948 |
Token | . |
Token ID | 13 |
Feature activation | +4.29e-01 |
Pos logit contributions | |
ksh | +0.181 |
rant | +0.179 |
flood | +0.178 |
dolphin | +0.174 |
whale | +0.172 |
Neg logit contributions | |
ons | -0.248 |
bes | -0.195 |
ages | -0.185 |
on | -0.174 |
con | -0.173 |
Token | Come |
Token ID | 7911 |
Feature activation | +3.64e+00 |
Pos logit contributions | |
ksh | +0.023 |
rant | +0.023 |
flood | +0.023 |
dolphin | +0.022 |
whale | +0.022 |
Neg logit contributions | |
ons | -0.033 |
bes | -0.026 |
ages | -0.025 |
on | -0.023 |
con | -0.023 |
Token | to |
Token ID | 284 |
Feature activation | -4.48e-01 |
Pos logit contributions | |
ksh | +0.242 |
rant | +0.238 |
flood | +0.237 |
dolphin | +0.231 |
whale | +0.228 |
Neg logit contributions | |
ons | -0.238 |
bes | -0.178 |
ages | -0.169 |
on | -0.157 |
con | -0.153 |
Token | my |
Token ID | 616 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
ons | +0.033 |
bes | +0.025 |
ages | +0.024 |
on | +0.023 |
con | +0.023 |
Neg logit contributions | |
ksh | -0.026 |
rant | -0.026 |
flood | -0.026 |
dolphin | -0.025 |
whale | -0.024 |
Token | it |
Token ID | 340 |
Feature activation | -6.36e-01 |
Pos logit contributions | |
ons | +0.028 |
bes | +0.021 |
ages | +0.019 |
on | +0.019 |
con | +0.018 |
Neg logit contributions | |
ksh | -0.029 |
rant | -0.028 |
flood | -0.027 |
dolphin | -0.027 |
whale | -0.026 |
Token | was |
Token ID | 373 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
ons | +0.037 |
bes | +0.027 |
es | +0.026 |
ages | +0.025 |
ves | +0.024 |
Neg logit contributions | |
ksh | -0.047 |
extingu | -0.044 |
whale | -0.043 |
rant | -0.043 |
dolphin | -0.043 |
Token | time |
Token ID | 640 |
Feature activation | +5.38e+00 |
Pos logit contributions | |
ksh | +0.007 |
dolphin | +0.006 |
whale | +0.006 |
extingu | +0.006 |
rant | +0.006 |
Neg logit contributions | |
ons | -0.010 |
bes | -0.008 |
on | -0.008 |
igs | -0.007 |
onwards | -0.007 |
Token | for |
Token ID | 329 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
ksh | +0.322 |
flood | +0.317 |
rant | +0.308 |
dolphin | +0.304 |
whale | +0.300 |
Neg logit contributions | |
ons | -0.387 |
ages | -0.290 |
bes | -0.290 |
on | -0.271 |
con | -0.260 |
Token | the |
Token ID | 262 |
Feature activation | -8.46e-01 |
Pos logit contributions | |
ksh | +0.121 |
flood | +0.121 |
rant | +0.120 |
dolphin | +0.118 |
whale | +0.115 |
Neg logit contributions | |
ons | -0.163 |
bes | -0.128 |
ages | -0.124 |
on | -0.115 |
con | -0.114 |
Token | mer |
Token ID | 4017 |
Feature activation | +1.24e+01 |
Pos logit contributions | |
ons | +0.061 |
bes | +0.047 |
on | +0.044 |
con | +0.044 |
ages | +0.043 |
Neg logit contributions | |
ksh | -0.051 |
rant | -0.048 |
flood | -0.043 |
extingu | -0.043 |
iet | -0.042 |
Token | maid |
Token ID | 23151 |
Feature activation | +7.89e+00 |
Pos logit contributions | |
flood | +0.521 |
Why | +0.473 |
ksh | +0.461 |
coming | +0.429 |
beginning | +0.427 |
Neg logit contributions | |
ons | -1.044 |
bes | -0.959 |
ages | -0.892 |
es | -0.867 |
powers | -0.860 |
Token | to |
Token ID | 284 |
Feature activation | -2.87e-01 |
Pos logit contributions | |
ksh | +0.437 |
rant | +0.429 |
flood | +0.427 |
dolphin | +0.414 |
whale | +0.407 |
Neg logit contributions | |
ons | -0.601 |
bes | -0.471 |
ages | -0.450 |
on | -0.423 |
con | -0.418 |
Token | return |
Token ID | 1441 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
ons | +0.027 |
bes | +0.022 |
on | +0.021 |
ages | +0.021 |
aid | +0.020 |
Neg logit contributions | |
ksh | -0.012 |
rant | -0.010 |
flood | -0.009 |
whale | -0.009 |
dolphin | -0.009 |
Token | home |
Token ID | 1363 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
ksh | +0.059 |
rant | +0.058 |
flood | +0.058 |
dolphin | +0.056 |
whale | +0.055 |
Neg logit contributions | |
ons | -0.081 |
bes | -0.064 |
ages | -0.061 |
on | -0.057 |
con | -0.056 |
Token | . |
Token ID | 13 |
Feature activation | +5.16e-01 |
Pos logit contributions | |
ons | +0.163 |
bes | +0.130 |
ages | +0.124 |
con | +0.122 |
aid | +0.122 |
Neg logit contributions | |
coming | -0.083 |
rant | -0.083 |
flood | -0.082 |
ksh | -0.081 |
dolphin | -0.081 |
Token |
|
Token ID | 198 |
Feature activation | +3.73e-02 |
Pos logit contributions | |
ksh | +0.014 |
rant | +0.013 |
flood | +0.013 |
dolphin | +0.013 |
whale | +0.012 |
Neg logit contributions | |
ons | -0.042 |
bes | -0.036 |
ages | -0.034 |
aid | -0.033 |
con | -0.032 |
Token |
|
Token ID | 198 |
Feature activation | +4.03e-02 |
Pos logit contributions | |
dolphin | +0.002 |
rant | +0.002 |
whale | +0.002 |
iet | +0.002 |
ksh | +0.002 |
Neg logit contributions | |
ons | -0.003 |
bes | -0.003 |
ages | -0.002 |
tail | -0.002 |
aces | -0.002 |
Token | Jack |
Token ID | 14295 |
Feature activation | +3.03e+00 |
Pos logit contributions | |
ksh | +0.002 |
rant | +0.002 |
flood | +0.002 |
dolphin | +0.002 |
whale | +0.002 |
Neg logit contributions | |
ons | -0.003 |
bes | -0.003 |
ages | -0.002 |
con | -0.002 |
es | -0.002 |
Token | asked |
Token ID | 1965 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
ksh | +0.155 |
rant | +0.154 |
flood | +0.151 |
dolphin | +0.150 |
whale | +0.147 |
Neg logit contributions | |
ons | -0.242 |
bes | -0.193 |
ages | -0.185 |
con | -0.173 |
on | -0.173 |
Token | the |
Token ID | 262 |
Feature activation | -8.70e-01 |
Pos logit contributions | |
ons | +0.127 |
bes | +0.102 |
ages | +0.097 |
con | +0.088 |
on | +0.087 |
Neg logit contributions | |
rant | -0.103 |
flood | -0.103 |
ksh | -0.102 |
dolphin | -0.101 |
whale | -0.100 |
Token | mer |
Token ID | 4017 |
Feature activation | +1.24e+01 |
Pos logit contributions | |
ons | +0.077 |
bes | +0.066 |
exclaimed | +0.063 |
on | +0.062 |
aid | +0.062 |
Neg logit contributions | |
ksh | -0.038 |
rant | -0.029 |
extingu | -0.028 |
OO | -0.027 |
flood | -0.026 |
Token | maid |
Token ID | 23151 |
Feature activation | +7.86e+00 |
Pos logit contributions | |
flood | +0.539 |
ksh | +0.478 |
ending | +0.447 |
storm | +0.445 |
Why | +0.443 |
Neg logit contributions | |
ons | -1.069 |
bes | -0.925 |
es | -0.916 |
ages | -0.882 |
aged | -0.840 |
Token | why |
Token ID | 1521 |
Feature activation | +8.54e+00 |
Pos logit contributions | |
ksh | +0.441 |
flood | +0.431 |
rant | +0.422 |
dolphin | +0.401 |
whale | +0.394 |
Neg logit contributions | |
ons | -0.606 |
bes | -0.460 |
ages | -0.450 |
es | -0.434 |
on | -0.433 |
Token | the |
Token ID | 262 |
Feature activation | -8.47e-01 |
Pos logit contributions | |
ksh | +0.645 |
rant | +0.628 |
flood | +0.621 |
dolphin | +0.604 |
whale | +0.598 |
Neg logit contributions | |
ons | -0.496 |
bes | -0.347 |
ages | -0.323 |
on | -0.303 |
es | -0.294 |
Token | waves |
Token ID | 9813 |
Feature activation | +4.07e+00 |
Pos logit contributions | |
ons | +0.079 |
bes | +0.067 |
powers | +0.063 |
ages | +0.062 |
on | +0.061 |
Neg logit contributions | |
ksh | -0.034 |
rant | -0.028 |
extingu | -0.025 |
latable | -0.022 |
flood | -0.022 |
Token | were |
Token ID | 547 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
ksh | +0.229 |
rant | +0.212 |
flood | +0.205 |
mate | +0.200 |
dolphin | +0.197 |
Neg logit contributions | |
ons | -0.323 |
bes | -0.243 |
ages | -0.235 |
on | -0.233 |
es | -0.225 |
Token | . |
Token ID | 13 |
Feature activation | +4.04e-01 |
Pos logit contributions | |
ons | +0.193 |
bes | +0.149 |
ages | +0.144 |
on | +0.136 |
es | +0.135 |
Neg logit contributions | |
ksh | -0.166 |
rant | -0.165 |
flood | -0.164 |
dolphin | -0.157 |
iet | -0.157 |
Token | " |
Token ID | 366 |
Feature activation | +2.89e+00 |
Pos logit contributions | |
ksh | +0.018 |
rant | +0.018 |
flood | +0.017 |
dolphin | +0.017 |
whale | +0.016 |
Neg logit contributions | |
ons | -0.035 |
bes | -0.029 |
ages | -0.028 |
on | -0.026 |
con | -0.026 |
Token | I |
Token ID | 40 |
Feature activation | +3.54e+00 |
Pos logit contributions | |
flood | +0.136 |
rant | +0.125 |
ksh | +0.125 |
ending | +0.120 |
dolphin | +0.119 |
Neg logit contributions | |
ons | -0.246 |
bes | -0.204 |
on | -0.193 |
ages | -0.192 |
aid | -0.190 |
Token | am |
Token ID | 716 |
Feature activation | -2.61e+00 |
Pos logit contributions | |
ksh | +0.219 |
rant | +0.196 |
flood | +0.193 |
iet | +0.191 |
dolphin | +0.189 |
Neg logit contributions | |
ons | -0.251 |
bes | -0.192 |
ages | -0.187 |
on | -0.185 |
aid | -0.181 |
Token | a |
Token ID | 257 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
ons | +0.175 |
on | +0.128 |
bes | +0.128 |
ages | +0.122 |
es | +0.115 |
Neg logit contributions | |
ksh | -0.171 |
rant | -0.160 |
iet | -0.156 |
flood | -0.154 |
ending | -0.151 |
Token | mer |
Token ID | 4017 |
Feature activation | +1.24e+01 |
Pos logit contributions | |
ons | +0.055 |
on | +0.047 |
powers | +0.045 |
bes | +0.044 |
ages | +0.044 |
Neg logit contributions | |
ksh | -0.066 |
rant | -0.065 |
ount | -0.059 |
iet | -0.056 |
ains | -0.056 |
Token | maid |
Token ID | 23151 |
Feature activation | +7.82e+00 |
Pos logit contributions | |
ksh | +0.531 |
Why | +0.526 |
flood | +0.518 |
rant | +0.497 |
beginning | +0.479 |
Neg logit contributions | |
ons | -0.900 |
bes | -0.843 |
ages | -0.801 |
aged | -0.790 |
es | -0.787 |
Token | . |
Token ID | 13 |
Feature activation | +4.45e-01 |
Pos logit contributions | |
ksh | +0.429 |
dolphin | +0.427 |
rant | +0.421 |
whale | +0.421 |
mate | +0.420 |
Neg logit contributions | |
ons | -0.596 |
bes | -0.479 |
ages | -0.437 |
con | -0.413 |
seek | -0.411 |
Token | Do |
Token ID | 2141 |
Feature activation | +4.19e-01 |
Pos logit contributions | |
flood | +0.024 |
ksh | +0.024 |
rant | +0.023 |
iet | +0.022 |
dolphin | +0.022 |
Neg logit contributions | |
ons | -0.033 |
bes | -0.026 |
ages | -0.025 |
on | -0.025 |
es | -0.024 |
Token | you |
Token ID | 345 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
dolphin | +0.026 |
mate | +0.026 |
flood | +0.025 |
ksh | +0.025 |
whale | +0.025 |
Neg logit contributions | |
ons | -0.028 |
bes | -0.022 |
ages | -0.021 |
con | -0.019 |
aid | -0.019 |
Token | want |
Token ID | 765 |
Feature activation | +3.18e+00 |
Pos logit contributions | |
ons | +0.190 |
bes | +0.152 |
on | +0.149 |
ages | +0.149 |
aid | +0.145 |
Neg logit contributions | |
ksh | -0.077 |
flood | -0.067 |
rant | -0.066 |
dolphin | -0.064 |
extingu | -0.064 |
Token | so |
Token ID | 523 |
Feature activation | -5.43e-01 |
Pos logit contributions | |
ksh | +0.003 |
rant | +0.003 |
flood | +0.003 |
dolphin | +0.003 |
whale | +0.003 |
Neg logit contributions | |
ons | -0.005 |
bes | -0.004 |
ages | -0.004 |
con | -0.003 |
on | -0.003 |
Token | happy |
Token ID | 3772 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
ons | +0.044 |
bes | +0.035 |
con | +0.032 |
ages | +0.032 |
seek | +0.031 |
Neg logit contributions | |
flood | -0.028 |
rant | -0.027 |
dolphin | -0.027 |
whale | -0.027 |
ksh | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | +3.39e-05 |
Pos logit contributions | |
ons | +0.084 |
bes | +0.065 |
ages | +0.062 |
on | +0.058 |
con | +0.057 |
Neg logit contributions | |
ksh | -0.067 |
rant | -0.066 |
flood | -0.065 |
dolphin | -0.064 |
whale | -0.063 |
Token | thanked |
Token ID | 26280 |
Feature activation | -4.24e+00 |
Pos logit contributions | |
dolphin | +0.000 |
mer | +0.000 |
whale | +0.000 |
coming | +0.000 |
flood | +0.000 |
Neg logit contributions | |
ons | -0.000 |
bes | -0.000 |
orns | -0.000 |
aid | -0.000 |
rob | -0.000 |
Token | the |
Token ID | 262 |
Feature activation | -9.06e-01 |
Pos logit contributions | |
ons | +0.307 |
bes | +0.240 |
ages | +0.232 |
on | +0.211 |
con | +0.208 |
Neg logit contributions | |
ksh | -0.249 |
rant | -0.247 |
flood | -0.246 |
dolphin | -0.242 |
whale | -0.239 |
Token | mer |
Token ID | 4017 |
Feature activation | +1.24e+01 |
Pos logit contributions | |
ons | +0.074 |
bes | +0.061 |
aid | +0.059 |
on | +0.059 |
ages | +0.057 |
Neg logit contributions | |
ksh | -0.048 |
rant | -0.041 |
extingu | -0.036 |
flood | -0.035 |
iet | -0.034 |
Token | maid |
Token ID | 23151 |
Feature activation | +7.84e+00 |
Pos logit contributions | |
flood | +0.472 |
ksh | +0.467 |
coming | +0.410 |
beginning | +0.401 |
rant | +0.400 |
Neg logit contributions | |
ons | -1.095 |
bes | -0.951 |
ages | -0.932 |
es | -0.901 |
aged | -0.882 |
Token | many |
Token ID | 867 |
Feature activation | -7.28e+00 |
Pos logit contributions | |
ksh | +0.333 |
rant | +0.323 |
mate | +0.323 |
dolphin | +0.322 |
whale | +0.316 |
Neg logit contributions | |
ons | -0.693 |
bes | -0.585 |
aid | -0.523 |
ages | -0.520 |
rob | -0.517 |
Token | times |
Token ID | 1661 |
Feature activation | -4.64e+00 |
Pos logit contributions | |
ons | +0.532 |
bes | +0.417 |
ages | +0.399 |
on | +0.371 |
con | +0.365 |
Neg logit contributions | |
ksh | -0.423 |
rant | -0.422 |
flood | -0.419 |
dolphin | -0.414 |
whale | -0.409 |
Token | . |
Token ID | 13 |
Feature activation | +4.57e-01 |
Pos logit contributions | |
ons | +0.388 |
bes | +0.321 |
ages | +0.304 |
aid | +0.287 |
con | +0.283 |
Neg logit contributions | |
rant | -0.211 |
coming | -0.210 |
ksh | -0.207 |
mate | -0.205 |
dolphin | -0.205 |
Token | With |
Token ID | 2080 |
Feature activation | -4.80e+00 |
Pos logit contributions | |
ksh | +0.026 |
rant | +0.025 |
dolphin | +0.025 |
whale | +0.024 |
flood | +0.024 |
Neg logit contributions | |
ons | -0.035 |
bes | -0.027 |
ages | -0.026 |
aid | -0.025 |
on | -0.024 |
Token | bed |
Token ID | 3996 |
Feature activation | -5.91e+00 |
Pos logit contributions | |
ons | +0.033 |
ages | +0.027 |
bes | +0.026 |
aid | +0.024 |
aces | +0.024 |
Neg logit contributions | |
flood | -0.024 |
wake | -0.024 |
coming | -0.023 |
dolphin | -0.022 |
ending | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | +3.95e-01 |
Pos logit contributions | |
ons | +0.495 |
bes | +0.363 |
ages | +0.361 |
aid | +0.361 |
on | +0.358 |
Neg logit contributions | |
coming | -0.294 |
wake | -0.269 |
rant | -0.267 |
flood | -0.267 |
ending | -0.264 |
Token | She |
Token ID | 1375 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
ksh | +0.020 |
rant | +0.019 |
Why | +0.019 |
ending | +0.019 |
wake | +0.019 |
Neg logit contributions | |
ons | -0.032 |
bes | -0.025 |
ages | -0.024 |
on | -0.023 |
con | -0.023 |
Token | dreamed |
Token ID | 27947 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
ons | +0.150 |
bes | +0.120 |
ages | +0.118 |
con | +0.114 |
on | +0.112 |
Neg logit contributions | |
rant | -0.066 |
ksh | -0.066 |
dolphin | -0.066 |
flood | -0.065 |
whale | -0.064 |
Token | of |
Token ID | 286 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
ons | +0.085 |
bes | +0.072 |
ages | +0.068 |
con | +0.060 |
tail | +0.060 |
Neg logit contributions | |
rant | -0.061 |
coming | -0.061 |
flood | -0.060 |
ending | -0.060 |
dolphin | -0.059 |
Token | mer |
Token ID | 4017 |
Feature activation | +1.23e+01 |
Pos logit contributions | |
ons | +0.103 |
bes | +0.081 |
ages | +0.081 |
seek | +0.073 |
on | +0.073 |
Neg logit contributions | |
flood | -0.072 |
coming | -0.070 |
dolphin | -0.070 |
whale | -0.069 |
rant | -0.068 |
Token | maid |
Token ID | 23151 |
Feature activation | +7.81e+00 |
Pos logit contributions | |
flood | +0.542 |
ksh | +0.486 |
ending | +0.470 |
storm | +0.455 |
rant | +0.452 |
Neg logit contributions | |
ons | -1.084 |
bes | -0.952 |
es | -0.912 |
ages | -0.886 |
aged | -0.847 |
Token | s |
Token ID | 82 |
Feature activation | -3.24e+00 |
Pos logit contributions | |
ksh | +0.524 |
rant | +0.517 |
flood | +0.514 |
dolphin | +0.506 |
whale | +0.498 |
Neg logit contributions | |
ons | -0.501 |
bes | -0.374 |
ages | -0.352 |
on | -0.326 |
con | -0.320 |
Token | and |
Token ID | 290 |
Feature activation | +8.40e-03 |
Pos logit contributions | |
ons | +0.261 |
bes | +0.207 |
ages | +0.202 |
con | +0.190 |
on | +0.188 |
Neg logit contributions | |
flood | -0.159 |
rant | -0.158 |
dolphin | -0.156 |
ksh | -0.154 |
coming | -0.154 |
Token | pirates |
Token ID | 27516 |
Feature activation | +4.13e+00 |
Pos logit contributions | |
dolphin | +0.000 |
flood | +0.000 |
ending | +0.000 |
whale | +0.000 |
mate | +0.000 |
Neg logit contributions | |
ons | -0.001 |
bes | -0.001 |
ages | -0.001 |
on | -0.000 |
con | -0.000 |
Token | , |
Token ID | 11 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
rant | +0.198 |
dolphin | +0.196 |
flood | +0.194 |
ksh | +0.194 |
coming | +0.192 |
Neg logit contributions | |
ons | -0.333 |
bes | -0.266 |
ages | -0.257 |
on | -0.242 |
con | -0.241 |
Token | Let |
Token ID | 5756 |
Feature activation | +2.99e+00 |
Pos logit contributions | |
flood | +0.172 |
mate | +0.162 |
rant | +0.161 |
arter | +0.159 |
ending | +0.159 |
Neg logit contributions | |
ons | -0.196 |
bes | -0.163 |
ages | -0.156 |
seek | -0.155 |
tail | -0.150 |
Token | 's |
Token ID | 338 |
Feature activation | +3.41e+00 |
Pos logit contributions | |
rant | +0.219 |
ksh | +0.216 |
flood | +0.206 |
extingu | +0.205 |
mate | +0.204 |
Neg logit contributions | |
ons | -0.194 |
bes | -0.127 |
ages | -0.126 |
on | -0.123 |
exclaimed | -0.115 |
Token | pretend |
Token ID | 16614 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
ksh | +0.213 |
rant | +0.192 |
ending | +0.186 |
king | +0.173 |
whale | +0.172 |
Neg logit contributions | |
ons | -0.245 |
bes | -0.192 |
on | -0.180 |
igs | -0.170 |
tail | -0.170 |
Token | we |
Token ID | 356 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
ksh | +0.108 |
rant | +0.107 |
flood | +0.100 |
iet | +0.098 |
extingu | +0.097 |
Neg logit contributions | |
ons | -0.114 |
bes | -0.083 |
ages | -0.080 |
on | -0.077 |
es | -0.074 |
Token | are |
Token ID | 389 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
ksh | +0.108 |
rant | +0.103 |
ending | +0.101 |
iet | +0.100 |
dolphin | +0.099 |
Neg logit contributions | |
ons | -0.161 |
on | -0.118 |
bes | -0.116 |
es | -0.115 |
ages | -0.114 |
Token | mer |
Token ID | 4017 |
Feature activation | +1.23e+01 |
Pos logit contributions | |
ksh | +0.062 |
rant | +0.061 |
iet | +0.057 |
ending | +0.056 |
flood | +0.056 |
Neg logit contributions | |
ons | -0.079 |
bes | -0.060 |
ages | -0.059 |
on | -0.058 |
es | -0.057 |
Token | maid |
Token ID | 23151 |
Feature activation | +7.81e+00 |
Pos logit contributions | |
flood | +0.507 |
ending | +0.503 |
storm | +0.470 |
rant | +0.443 |
Why | +0.441 |
Neg logit contributions | |
ons | -1.014 |
bes | -0.957 |
es | -0.913 |
ages | -0.911 |
aged | -0.848 |
Token | s |
Token ID | 82 |
Feature activation | -3.24e+00 |
Pos logit contributions | |
rant | +0.568 |
flood | +0.568 |
ksh | +0.567 |
ending | +0.540 |
dolphin | +0.538 |
Neg logit contributions | |
ons | -0.461 |
bes | -0.323 |
ages | -0.317 |
es | -0.291 |
on | -0.286 |
Token | and |
Token ID | 290 |
Feature activation | +8.40e-03 |
Pos logit contributions | |
ons | +0.177 |
bes | +0.125 |
ages | +0.120 |
es | +0.113 |
on | +0.109 |
Neg logit contributions | |
ksh | -0.243 |
rant | -0.239 |
flood | -0.237 |
dolphin | -0.231 |
iet | -0.228 |
Token | swim |
Token ID | 9422 |
Feature activation | +5.59e+00 |
Pos logit contributions | |
flood | +0.000 |
ksh | +0.000 |
dolphin | +0.000 |
rant | +0.000 |
whale | +0.000 |
Neg logit contributions | |
ons | -0.001 |
bes | -0.001 |
ages | -0.001 |
on | -0.000 |
con | -0.000 |
Token | in |
Token ID | 287 |
Feature activation | +1.89e+00 |
Pos logit contributions | |
ksh | +0.355 |
flood | +0.354 |
rant | +0.350 |
ending | +0.349 |
extingu | +0.334 |
Neg logit contributions | |
ons | -0.352 |
bes | -0.272 |
ages | -0.268 |
es | -0.256 |
seek | -0.234 |
Token | many |
Token ID | 867 |
Feature activation | -7.09e+00 |
Pos logit contributions | |
ons | +0.232 |
bes | +0.186 |
ages | +0.177 |
con | +0.165 |
on | +0.164 |
Neg logit contributions | |
ksh | -0.164 |
rant | -0.162 |
flood | -0.161 |
dolphin | -0.158 |
coming | -0.156 |
Token | more |
Token ID | 517 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
ons | +0.546 |
ages | +0.473 |
bes | +0.472 |
seek | +0.433 |
on | +0.426 |
Neg logit contributions | |
coming | -0.342 |
flood | -0.333 |
whale | -0.328 |
dolphin | -0.327 |
rant | -0.321 |
Token | adventures |
Token ID | 17545 |
Feature activation | -2.88e+00 |
Pos logit contributions | |
ons | +0.119 |
bes | +0.098 |
ages | +0.094 |
es | +0.088 |
seek | +0.088 |
Neg logit contributions | |
flood | -0.073 |
dolphin | -0.071 |
coming | -0.071 |
whale | -0.071 |
rant | -0.070 |
Token | with |
Token ID | 351 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
ons | +0.231 |
bes | +0.192 |
ages | +0.183 |
aid | +0.176 |
con | +0.170 |
Neg logit contributions | |
coming | -0.133 |
rant | -0.126 |
ksh | -0.125 |
flood | -0.125 |
dolphin | -0.121 |
Token | the |
Token ID | 262 |
Feature activation | -9.08e-01 |
Pos logit contributions | |
ons | +0.093 |
bes | +0.076 |
ages | +0.075 |
seek | +0.070 |
aid | +0.069 |
Neg logit contributions | |
flood | -0.064 |
coming | -0.063 |
rant | -0.063 |
dolphin | -0.063 |
whale | -0.062 |
Token | mer |
Token ID | 4017 |
Feature activation | +1.23e+01 |
Pos logit contributions | |
ons | +0.071 |
bes | +0.056 |
ages | +0.053 |
con | +0.051 |
on | +0.051 |
Neg logit contributions | |
ksh | -0.049 |
rant | -0.048 |
flood | -0.046 |
dolphin | -0.044 |
whale | -0.043 |
Token | maid |
Token ID | 23151 |
Feature activation | +7.75e+00 |
Pos logit contributions | |
flood | +0.417 |
ksh | +0.410 |
Why | +0.383 |
coming | +0.363 |
ending | +0.360 |
Neg logit contributions | |
ons | -1.173 |
bes | -1.009 |
ages | -0.971 |
es | -0.935 |
powers | -0.914 |
Token | . |
Token ID | 13 |
Feature activation | +3.23e-01 |
Pos logit contributions | |
dolphin | +0.409 |
ksh | +0.402 |
rant | +0.401 |
whale | +0.399 |
flood | +0.392 |
Neg logit contributions | |
ons | -0.593 |
bes | -0.479 |
ages | -0.442 |
aid | -0.421 |
seek | -0.421 |
Token |
|
Token ID | 198 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
ksh | +0.019 |
dolphin | +0.018 |
rant | +0.018 |
flood | +0.018 |
mate | +0.018 |
Neg logit contributions | |
ons | -0.024 |
bes | -0.018 |
ages | -0.018 |
aid | -0.017 |
igs | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
ons | +0.008 |
ages | +0.007 |
bes | +0.007 |
aces | +0.007 |
seek | +0.006 |
Neg logit contributions | |
dolphin | -0.005 |
whale | -0.005 |
iet | -0.005 |
rant | -0.004 |
flood | -0.004 |
Token | The |
Token ID | 464 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
ons | +0.010 |
bes | +0.007 |
ages | +0.007 |
con | +0.007 |
aces | +0.007 |
Neg logit contributions | |
dolphin | -0.008 |
ksh | -0.007 |
iet | -0.007 |
flood | -0.007 |
mate | -0.007 |
Token | her |
Token ID | 607 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
rant | +0.045 |
ksh | +0.044 |
dolphin | +0.043 |
coming | +0.043 |
flood | +0.043 |
Neg logit contributions | |
ons | -0.057 |
bes | -0.046 |
ages | -0.045 |
seek | -0.044 |
powers | -0.041 |
Token | hair |
Token ID | 4190 |
Feature activation | -5.82e+00 |
Pos logit contributions | |
ons | +0.114 |
bes | +0.089 |
ages | +0.086 |
seek | +0.082 |
on | +0.080 |
Neg logit contributions | |
ksh | -0.081 |
rant | -0.080 |
dolphin | -0.078 |
flood | -0.078 |
whale | -0.077 |
Token | in |
Token ID | 287 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
ons | +0.436 |
bes | +0.338 |
ages | +0.323 |
con | +0.305 |
seek | +0.301 |
Neg logit contributions | |
ksh | -0.327 |
rant | -0.327 |
flood | -0.314 |
dolphin | -0.306 |
coming | -0.303 |
Token | a |
Token ID | 257 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
flood | +0.083 |
ksh | +0.082 |
rant | +0.082 |
dolphin | +0.080 |
coming | +0.079 |
Neg logit contributions | |
ons | -0.107 |
bes | -0.083 |
ages | -0.083 |
con | -0.076 |
seek | -0.075 |
Token | pony |
Token ID | 26902 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
ons | +0.109 |
bes | +0.085 |
ages | +0.081 |
on | +0.077 |
con | +0.076 |
Neg logit contributions | |
ksh | -0.078 |
rant | -0.077 |
flood | -0.075 |
dolphin | -0.073 |
whale | -0.072 |
Token | tail |
Token ID | 13199 |
Feature activation | -1.57e+01 |
Pos logit contributions | |
ons | +0.113 |
bes | +0.091 |
ages | +0.088 |
aid | +0.085 |
seek | +0.081 |
Neg logit contributions | |
ksh | -0.069 |
coming | -0.069 |
rant | -0.069 |
dolphin | -0.068 |
flood | -0.067 |
Token | so |
Token ID | 523 |
Feature activation | -9.20e-01 |
Pos logit contributions | |
ons | +1.261 |
bes | +1.004 |
ages | +0.977 |
aid | +0.909 |
con | +0.908 |
Neg logit contributions | |
rant | -0.771 |
flood | -0.768 |
ksh | -0.754 |
dolphin | -0.739 |
mate | -0.735 |
Token | she |
Token ID | 673 |
Feature activation | -3.24e+00 |
Pos logit contributions | |
ons | +0.059 |
bes | +0.047 |
ages | +0.043 |
rob | +0.041 |
seek | +0.041 |
Neg logit contributions | |
dolphin | -0.058 |
flood | -0.058 |
ending | -0.058 |
rant | -0.057 |
ksh | -0.056 |
Token | could |
Token ID | 714 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
ons | +0.271 |
bes | +0.224 |
ages | +0.211 |
seek | +0.205 |
con | +0.200 |
Neg logit contributions | |
ksh | -0.150 |
rant | -0.149 |
dolphin | -0.146 |
flood | -0.143 |
whale | -0.142 |
Token | look |
Token ID | 804 |
Feature activation | +3.79e+00 |
Pos logit contributions | |
ons | +0.199 |
bes | +0.163 |
ages | +0.162 |
rob | +0.152 |
con | +0.152 |
Neg logit contributions | |
wake | -0.087 |
flood | -0.085 |
rant | -0.081 |
dolphin | -0.080 |
coming | -0.079 |
Token | pretty |
Token ID | 2495 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
ksh | +0.182 |
flood | +0.178 |
rant | +0.174 |
dolphin | +0.170 |
whale | +0.167 |
Neg logit contributions | |
ons | -0.317 |
bes | -0.251 |
ages | -0.242 |
on | -0.233 |
es | -0.228 |
Token | , |
Token ID | 11 |
Feature activation | -3.47e-01 |
Pos logit contributions | |
flood | +0.040 |
ksh | +0.040 |
rant | +0.040 |
dolphin | +0.039 |
ending | +0.039 |
Neg logit contributions | |
ons | -0.047 |
bes | -0.036 |
ages | -0.035 |
on | -0.032 |
con | -0.032 |
Token | Emma |
Token ID | 18966 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
ons | +0.030 |
bes | +0.024 |
ages | +0.024 |
con | +0.022 |
aid | +0.022 |
Neg logit contributions | |
ksh | -0.016 |
flood | -0.016 |
rant | -0.015 |
dolphin | -0.015 |
ending | -0.015 |
Token | decided |
Token ID | 3066 |
Feature activation | -4.70e-01 |
Pos logit contributions | |
ksh | +0.067 |
rant | +0.066 |
flood | +0.065 |
dolphin | +0.063 |
whale | +0.062 |
Neg logit contributions | |
ons | -0.106 |
bes | -0.084 |
ages | -0.080 |
on | -0.076 |
con | -0.076 |
Token | a |
Token ID | 257 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
ons | +0.036 |
bes | +0.029 |
ages | +0.027 |
on | +0.026 |
con | +0.025 |
Neg logit contributions | |
ksh | -0.025 |
flood | -0.025 |
rant | -0.025 |
dolphin | -0.024 |
coming | -0.024 |
Token | pony |
Token ID | 26902 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
ons | +0.103 |
bes | +0.081 |
ages | +0.079 |
on | +0.075 |
con | +0.075 |
Neg logit contributions | |
ksh | -0.066 |
rant | -0.066 |
flood | -0.061 |
dolphin | -0.059 |
mate | -0.058 |
Token | tail |
Token ID | 13199 |
Feature activation | -1.56e+01 |
Pos logit contributions | |
ons | +0.106 |
bes | +0.087 |
ages | +0.082 |
aid | +0.076 |
on | +0.076 |
Neg logit contributions | |
flood | -0.063 |
rant | -0.062 |
ksh | -0.062 |
dolphin | -0.061 |
coming | -0.060 |
Token | wasn |
Token ID | 2492 |
Feature activation | +3.30e+00 |
Pos logit contributions | |
ons | +1.338 |
bes | +1.059 |
ages | +1.022 |
on | +0.989 |
es | +0.974 |
Neg logit contributions | |
ksh | -0.772 |
rant | -0.722 |
flood | -0.698 |
dolphin | -0.678 |
whale | -0.662 |
Token | 't |
Token ID | 470 |
Feature activation | -4.49e+00 |
Pos logit contributions | |
flood | +0.230 |
rant | +0.225 |
ksh | +0.224 |
coming | +0.224 |
dolphin | +0.218 |
Neg logit contributions | |
ons | -0.200 |
bes | -0.152 |
ages | -0.147 |
con | -0.131 |
es | -0.126 |
Token | so |
Token ID | 523 |
Feature activation | -7.80e-01 |
Pos logit contributions | |
ons | +0.360 |
bes | +0.287 |
ages | +0.274 |
on | +0.258 |
con | +0.256 |
Neg logit contributions | |
ksh | -0.231 |
flood | -0.228 |
rant | -0.227 |
dolphin | -0.219 |
whale | -0.215 |
Token | important |
Token ID | 1593 |
Feature activation | -9.11e-01 |
Pos logit contributions | |
ons | +0.066 |
bes | +0.054 |
ages | +0.051 |
on | +0.048 |
con | +0.048 |
Neg logit contributions | |
flood | -0.036 |
ksh | -0.036 |
rant | -0.036 |
dolphin | -0.035 |
whale | -0.034 |
Token | after |
Token ID | 706 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
ons | +0.077 |
bes | +0.062 |
ages | +0.059 |
on | +0.057 |
aid | +0.057 |
Neg logit contributions | |
ksh | -0.042 |
coming | -0.041 |
rant | -0.041 |
flood | -0.040 |
dolphin | -0.040 |
Token | with |
Token ID | 351 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
ksh | +0.086 |
rant | +0.084 |
flood | +0.083 |
dolphin | +0.081 |
whale | +0.079 |
Neg logit contributions | |
ons | -0.099 |
bes | -0.076 |
ages | -0.072 |
on | -0.069 |
es | -0.067 |
Token | glasses |
Token ID | 15232 |
Feature activation | -3.94e+00 |
Pos logit contributions | |
ons | +0.086 |
bes | +0.068 |
ages | +0.065 |
on | +0.060 |
con | +0.060 |
Neg logit contributions | |
ksh | -0.063 |
flood | -0.062 |
rant | -0.062 |
dolphin | -0.061 |
whale | -0.060 |
Token | and |
Token ID | 290 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
ons | +0.298 |
bes | +0.235 |
ages | +0.225 |
con | +0.208 |
on | +0.207 |
Neg logit contributions | |
ksh | -0.217 |
rant | -0.215 |
flood | -0.213 |
dolphin | -0.208 |
whale | -0.204 |
Token | a |
Token ID | 257 |
Feature activation | -8.34e-01 |
Pos logit contributions | |
dolphin | +0.008 |
flood | +0.008 |
whale | +0.008 |
ksh | +0.008 |
mate | +0.008 |
Neg logit contributions | |
ons | -0.013 |
bes | -0.010 |
ages | -0.010 |
on | -0.009 |
seek | -0.009 |
Token | pony |
Token ID | 26902 |
Feature activation | -8.59e-01 |
Pos logit contributions | |
ons | +0.067 |
bes | +0.053 |
ages | +0.051 |
on | +0.048 |
con | +0.047 |
Neg logit contributions | |
ksh | -0.043 |
flood | -0.042 |
rant | -0.042 |
dolphin | -0.041 |
whale | -0.040 |
Token | tail |
Token ID | 13199 |
Feature activation | -1.51e+01 |
Pos logit contributions | |
ons | +0.066 |
bes | +0.052 |
ages | +0.050 |
aid | +0.047 |
on | +0.046 |
Neg logit contributions | |
ksh | -0.046 |
flood | -0.045 |
rant | -0.045 |
dolphin | -0.044 |
coming | -0.044 |
Token | . |
Token ID | 13 |
Feature activation | +6.00e-01 |
Pos logit contributions | |
ons | +1.127 |
bes | +0.882 |
ages | +0.840 |
on | +0.788 |
es | +0.779 |
Neg logit contributions | |
ksh | -0.875 |
rant | -0.851 |
flood | -0.847 |
dolphin | -0.824 |
whale | -0.810 |
Token | She |
Token ID | 1375 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
rant | +0.038 |
ksh | +0.038 |
flood | +0.038 |
whale | +0.036 |
mate | +0.036 |
Neg logit contributions | |
ons | -0.040 |
bes | -0.030 |
ages | -0.029 |
on | -0.027 |
con | -0.027 |
Token | is |
Token ID | 318 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
ons | +0.103 |
bes | +0.079 |
ages | +0.074 |
on | +0.070 |
es | +0.070 |
Neg logit contributions | |
ksh | -0.090 |
flood | -0.088 |
rant | -0.088 |
dolphin | -0.084 |
whale | -0.083 |
Token | reading |
Token ID | 3555 |
Feature activation | -8.83e+00 |
Pos logit contributions | |
ons | +0.008 |
bes | +0.006 |
ages | +0.006 |
on | +0.005 |
es | +0.005 |
Neg logit contributions | |
ksh | -0.007 |
flood | -0.007 |
rant | -0.007 |
dolphin | -0.007 |
ending | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | -7.99e-01 |
Pos logit contributions | |
ons | +0.641 |
bes | +0.496 |
ages | +0.476 |
on | +0.442 |
con | +0.437 |
Neg logit contributions | |
ksh | -0.519 |
rant | -0.511 |
flood | -0.507 |
dolphin | -0.495 |
whale | -0.486 |
Token | cars |
Token ID | 5006 |
Feature activation | -6.50e+00 |
Pos logit contributions | |
ons | +0.096 |
bes | +0.073 |
ages | +0.070 |
con | +0.067 |
on | +0.067 |
Neg logit contributions | |
ksh | -0.078 |
rant | -0.076 |
flood | -0.072 |
iet | -0.070 |
dolphin | -0.070 |
Token | and |
Token ID | 290 |
Feature activation | -3.96e-01 |
Pos logit contributions | |
ons | +0.539 |
bes | +0.443 |
ages | +0.421 |
con | +0.395 |
on | +0.391 |
Neg logit contributions | |
flood | -0.303 |
rant | -0.303 |
ksh | -0.302 |
dolphin | -0.287 |
mate | -0.287 |
Token | the |
Token ID | 262 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
ons | +0.030 |
bes | +0.025 |
ages | +0.024 |
seek | +0.022 |
on | +0.022 |
Neg logit contributions | |
flood | -0.021 |
coming | -0.020 |
dolphin | -0.020 |
mate | -0.020 |
ksh | -0.020 |
Token | cr |
Token ID | 1067 |
Feature activation | -3.50e+00 |
Pos logit contributions | |
ons | +0.092 |
bes | +0.071 |
ages | +0.067 |
con | +0.063 |
on | +0.063 |
Neg logit contributions | |
ksh | -0.080 |
rant | -0.080 |
flood | -0.075 |
dolphin | -0.075 |
iet | -0.074 |
Token | ay |
Token ID | 323 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
ons | +0.269 |
bes | +0.221 |
ages | +0.210 |
con | +0.200 |
seek | +0.196 |
Neg logit contributions | |
ksh | -0.181 |
rant | -0.180 |
flood | -0.171 |
iet | -0.167 |
mate | -0.165 |
Token | ons |
Token ID | 684 |
Feature activation | -1.44e+01 |
Pos logit contributions | |
bes | +0.005 |
ons | +0.003 |
seek | -0.002 |
ages | -0.003 |
con | -0.005 |
Neg logit contributions | |
coming | -0.190 |
whale | -0.185 |
rant | -0.184 |
iet | -0.181 |
ksh | -0.180 |
Token | and |
Token ID | 290 |
Feature activation | -4.01e-01 |
Pos logit contributions | |
ons | +1.153 |
bes | +0.929 |
ages | +0.885 |
con | +0.834 |
seek | +0.823 |
Neg logit contributions | |
rant | -0.724 |
ksh | -0.724 |
flood | -0.714 |
dolphin | -0.692 |
mate | -0.681 |
Token | make |
Token ID | 787 |
Feature activation | +4.13e-01 |
Pos logit contributions | |
ons | +0.031 |
bes | +0.025 |
ages | +0.024 |
seek | +0.023 |
on | +0.022 |
Neg logit contributions | |
flood | -0.021 |
coming | -0.020 |
dolphin | -0.020 |
ksh | -0.020 |
mate | -0.020 |
Token | something |
Token ID | 1223 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
ksh | +0.019 |
flood | +0.018 |
rant | +0.018 |
dolphin | +0.018 |
whale | +0.018 |
Neg logit contributions | |
ons | -0.035 |
bes | -0.029 |
ages | -0.028 |
seek | -0.026 |
on | -0.026 |
Token | nice |
Token ID | 3621 |
Feature activation | +3.25e+00 |
Pos logit contributions | |
ksh | +0.099 |
rant | +0.098 |
flood | +0.097 |
dolphin | +0.094 |
whale | +0.092 |
Neg logit contributions | |
ons | -0.140 |
bes | -0.110 |
ages | -0.105 |
on | -0.099 |
con | -0.098 |
Token | ." |
Token ID | 526 |
Feature activation | -4.54e+00 |
Pos logit contributions | |
ksh | +0.185 |
rant | +0.181 |
flood | +0.181 |
dolphin | +0.175 |
whale | +0.172 |
Neg logit contributions | |
ons | -0.243 |
bes | -0.189 |
ages | -0.182 |
on | -0.171 |
es | -0.168 |
Token | . |
Token ID | 13 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
ksh | +0.049 |
rant | +0.048 |
flood | +0.047 |
dolphin | +0.046 |
whale | +0.045 |
Neg logit contributions | |
ons | -0.068 |
bes | -0.053 |
ages | -0.051 |
on | -0.048 |
con | -0.047 |
Token | The |
Token ID | 383 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
rant | +0.002 |
ksh | +0.002 |
flood | +0.002 |
dolphin | +0.002 |
mate | +0.002 |
Neg logit contributions | |
ons | -0.002 |
bes | -0.001 |
ages | -0.001 |
on | -0.001 |
con | -0.001 |
Token | melted |
Token ID | 24178 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
ons | +0.120 |
bes | +0.091 |
ages | +0.088 |
on | +0.084 |
con | +0.080 |
Neg logit contributions | |
ksh | -0.095 |
rant | -0.089 |
mate | -0.085 |
ust | -0.084 |
iet | -0.084 |
Token | cr |
Token ID | 1067 |
Feature activation | -3.50e+00 |
Pos logit contributions | |
ksh | +0.157 |
rant | +0.155 |
flood | +0.155 |
dolphin | +0.150 |
whale | +0.148 |
Neg logit contributions | |
ons | -0.189 |
bes | -0.146 |
ages | -0.139 |
on | -0.129 |
con | -0.128 |
Token | ay |
Token ID | 323 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
ons | +0.265 |
bes | +0.216 |
ages | +0.205 |
con | +0.196 |
seek | +0.192 |
Neg logit contributions | |
rant | -0.186 |
ksh | -0.184 |
flood | -0.175 |
iet | -0.171 |
coming | -0.168 |
Token | ons |
Token ID | 684 |
Feature activation | -1.44e+01 |
Pos logit contributions | |
bes | +0.010 |
ons | +0.010 |
seek | +0.009 |
powers | +0.002 |
ages | +0.001 |
Neg logit contributions | |
rant | -0.182 |
coming | -0.180 |
whale | -0.177 |
iet | -0.176 |
dolphin | -0.173 |
Token | have |
Token ID | 423 |
Feature activation | -3.75e+00 |
Pos logit contributions | |
ons | +1.143 |
bes | +0.924 |
ages | +0.885 |
con | +0.830 |
seek | +0.827 |
Neg logit contributions | |
flood | -0.732 |
rant | -0.731 |
ksh | -0.720 |
dolphin | -0.699 |
coming | -0.693 |
Token | made |
Token ID | 925 |
Feature activation | +3.56e+00 |
Pos logit contributions | |
ons | +0.279 |
bes | +0.220 |
ages | +0.210 |
on | +0.196 |
con | +0.194 |
Neg logit contributions | |
ksh | -0.211 |
rant | -0.209 |
flood | -0.207 |
dolphin | -0.200 |
whale | -0.198 |
Token | a |
Token ID | 257 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
ksh | +0.200 |
rant | +0.196 |
flood | +0.196 |
dolphin | +0.190 |
whale | +0.186 |
Neg logit contributions | |
ons | -0.268 |
bes | -0.210 |
ages | -0.201 |
on | -0.188 |
con | -0.186 |
Token | beautiful |
Token ID | 4950 |
Feature activation | +3.49e+00 |
Pos logit contributions | |
ons | +0.096 |
bes | +0.071 |
ages | +0.070 |
on | +0.068 |
con | +0.066 |
Neg logit contributions | |
rant | -0.089 |
ksh | -0.088 |
iet | -0.079 |
mate | -0.079 |
dolphin | -0.077 |
Token | pattern |
Token ID | 3912 |
Feature activation | -8.87e-01 |
Pos logit contributions | |
ksh | +0.218 |
rant | +0.212 |
flood | +0.205 |
dolphin | +0.201 |
mate | +0.198 |
Neg logit contributions | |
ons | -0.249 |
bes | -0.190 |
ages | -0.178 |
on | -0.173 |
con | -0.166 |
Token | day |
Token ID | 1110 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
ksh | +0.097 |
rant | +0.093 |
flood | +0.091 |
dolphin | +0.087 |
mate | +0.086 |
Neg logit contributions | |
ons | -0.122 |
bes | -0.091 |
ages | -0.089 |
on | -0.088 |
es | -0.081 |
Token | , |
Token ID | 11 |
Feature activation | -4.53e-01 |
Pos logit contributions | |
ons | +0.102 |
bes | +0.083 |
ages | +0.077 |
con | +0.072 |
aid | +0.072 |
Neg logit contributions | |
flood | -0.084 |
dolphin | -0.083 |
whale | -0.082 |
ksh | -0.082 |
rant | -0.082 |
Token | the |
Token ID | 262 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
ons | +0.037 |
bes | +0.030 |
ages | +0.029 |
seek | +0.027 |
con | +0.027 |
Neg logit contributions | |
flood | -0.023 |
dolphin | -0.022 |
ksh | -0.022 |
whale | -0.022 |
rant | -0.022 |
Token | cr |
Token ID | 1067 |
Feature activation | -3.50e+00 |
Pos logit contributions | |
ons | +0.105 |
bes | +0.082 |
ages | +0.078 |
on | +0.077 |
con | +0.073 |
Neg logit contributions | |
ksh | -0.070 |
rant | -0.069 |
flood | -0.067 |
dolphin | -0.063 |
iet | -0.062 |
Token | ay |
Token ID | 323 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
ons | +0.261 |
bes | +0.219 |
ages | +0.206 |
con | +0.196 |
seek | +0.195 |
Neg logit contributions | |
rant | -0.187 |
ksh | -0.181 |
flood | -0.175 |
iet | -0.172 |
whale | -0.168 |
Token | ons |
Token ID | 684 |
Feature activation | -1.44e+01 |
Pos logit contributions | |
ons | +0.013 |
bes | +0.012 |
seek | +0.009 |
ages | +0.008 |
powers | +0.003 |
Neg logit contributions | |
rant | -0.184 |
whale | -0.184 |
dolphin | -0.177 |
coming | -0.177 |
dot | -0.176 |
Token | wanted |
Token ID | 2227 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
ons | +1.214 |
bes | +0.978 |
ages | +0.942 |
on | +0.889 |
con | +0.882 |
Neg logit contributions | |
ksh | -0.680 |
rant | -0.668 |
flood | -0.666 |
dolphin | -0.640 |
whale | -0.628 |
Token | to |
Token ID | 284 |
Feature activation | -7.96e-01 |
Pos logit contributions | |
ksh | +0.056 |
rant | +0.055 |
flood | +0.055 |
dolphin | +0.054 |
whale | +0.053 |
Neg logit contributions | |
ons | -0.076 |
bes | -0.060 |
ages | -0.057 |
on | -0.053 |
con | -0.053 |
Token | know |
Token ID | 760 |
Feature activation | -5.74e+00 |
Pos logit contributions | |
ons | +0.071 |
bes | +0.060 |
ages | +0.059 |
es | +0.056 |
con | +0.055 |
Neg logit contributions | |
flood | -0.034 |
wake | -0.031 |
dolphin | -0.031 |
whale | -0.031 |
rant | -0.030 |
Token | which |
Token ID | 543 |
Feature activation | +2.75e+00 |
Pos logit contributions | |
ons | +0.472 |
bes | +0.370 |
ages | +0.356 |
on | +0.342 |
es | +0.336 |
Neg logit contributions | |
ksh | -0.291 |
rant | -0.282 |
flood | -0.282 |
dolphin | -0.269 |
whale | -0.263 |
Token | one |
Token ID | 530 |
Feature activation | +4.22e+00 |
Pos logit contributions | |
ksh | +0.118 |
rant | +0.115 |
flood | +0.115 |
dolphin | +0.110 |
whale | +0.108 |
Neg logit contributions | |
ons | -0.245 |
bes | -0.200 |
ages | -0.193 |
on | -0.182 |
con | -0.181 |
Token | . |
Token ID | 13 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
ons | +0.222 |
bes | +0.180 |
ages | +0.170 |
aid | +0.160 |
on | +0.159 |
Neg logit contributions | |
coming | -0.175 |
rant | -0.170 |
ksh | -0.169 |
mate | -0.168 |
dolphin | -0.167 |
Token | We |
Token ID | 775 |
Feature activation | +2.40e+00 |
Pos logit contributions | |
ksh | +0.001 |
rant | +0.001 |
flood | +0.001 |
dolphin | +0.001 |
mate | +0.001 |
Neg logit contributions | |
ons | -0.002 |
bes | -0.002 |
ages | -0.002 |
on | -0.002 |
con | -0.002 |
Token | used |
Token ID | 973 |
Feature activation | -4.69e+00 |
Pos logit contributions | |
ksh | +0.108 |
rant | +0.104 |
flood | +0.104 |
dolphin | +0.100 |
whale | +0.098 |
Neg logit contributions | |
ons | -0.210 |
bes | -0.169 |
ages | -0.164 |
on | -0.156 |
es | -0.153 |
Token | cr |
Token ID | 1067 |
Feature activation | -3.50e+00 |
Pos logit contributions | |
ons | +0.348 |
bes | +0.291 |
ages | +0.279 |
seek | +0.264 |
es | +0.254 |
Neg logit contributions | |
rant | -0.246 |
flood | -0.244 |
ksh | -0.242 |
dolphin | -0.240 |
whale | -0.238 |
Token | ay |
Token ID | 323 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
ons | +0.270 |
bes | +0.237 |
ages | +0.219 |
seek | +0.212 |
con | +0.209 |
Neg logit contributions | |
rant | -0.165 |
ksh | -0.164 |
iet | -0.154 |
mate | -0.147 |
dolphin | -0.143 |
Token | ons |
Token ID | 684 |
Feature activation | -1.44e+01 |
Pos logit contributions | |
bes | +0.002 |
ons | +0.001 |
seek | -0.005 |
tail | -0.008 |
ages | -0.011 |
Neg logit contributions | |
coming | -0.188 |
iet | -0.188 |
whale | -0.188 |
rant | -0.187 |
dolphin | -0.185 |
Token | and |
Token ID | 290 |
Feature activation | -4.01e-01 |
Pos logit contributions | |
ons | +1.210 |
bes | +1.038 |
seek | +0.935 |
ages | +0.928 |
rob | +0.915 |
Neg logit contributions | |
mate | -0.611 |
rant | -0.611 |
ksh | -0.603 |
dolphin | -0.586 |
whale | -0.579 |
Token | markers |
Token ID | 19736 |
Feature activation | -5.15e+00 |
Pos logit contributions | |
ons | +0.029 |
bes | +0.024 |
ages | +0.022 |
seek | +0.021 |
on | +0.021 |
Neg logit contributions | |
rant | -0.022 |
ksh | -0.021 |
flood | -0.021 |
whale | -0.021 |
dolphin | -0.021 |
Token | to |
Token ID | 284 |
Feature activation | -7.93e-01 |
Pos logit contributions | |
ons | +0.420 |
bes | +0.354 |
ages | +0.322 |
seek | +0.318 |
on | +0.313 |
Neg logit contributions | |
rant | -0.233 |
ksh | -0.228 |
mate | -0.225 |
coming | -0.212 |
dolphin | -0.211 |
Token | draw |
Token ID | 3197 |
Feature activation | -2.16e+00 |
Pos logit contributions | |
ons | +0.067 |
bes | +0.055 |
ages | +0.054 |
seek | +0.050 |
con | +0.050 |
Neg logit contributions | |
rant | -0.036 |
flood | -0.036 |
ksh | -0.035 |
dolphin | -0.034 |
whale | -0.034 |
Token | pictures |
Token ID | 5986 |
Feature activation | -7.02e+00 |
Pos logit contributions | |
ons | +0.143 |
bes | +0.114 |
ages | +0.111 |
seek | +0.105 |
aid | +0.103 |
Neg logit contributions | |
ksh | -0.130 |
rant | -0.130 |
coming | -0.125 |
dolphin | -0.123 |
flood | -0.122 |
Token | his |
Token ID | 465 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
ons | +0.108 |
bes | +0.074 |
ages | +0.073 |
on | +0.072 |
con | +0.068 |
Neg logit contributions | |
ksh | -0.109 |
rant | -0.103 |
flood | -0.100 |
iet | -0.097 |
dolphin | -0.095 |
Token | paper |
Token ID | 3348 |
Feature activation | -3.70e+00 |
Pos logit contributions | |
ons | +0.131 |
bes | +0.100 |
ages | +0.095 |
on | +0.089 |
con | +0.087 |
Neg logit contributions | |
ksh | -0.115 |
rant | -0.113 |
flood | -0.113 |
dolphin | -0.109 |
whale | -0.108 |
Token | and |
Token ID | 290 |
Feature activation | -4.00e-01 |
Pos logit contributions | |
ons | +0.300 |
bes | +0.238 |
ages | +0.229 |
on | +0.218 |
es | +0.214 |
Neg logit contributions | |
ksh | -0.188 |
rant | -0.184 |
flood | -0.183 |
dolphin | -0.175 |
whale | -0.172 |
Token | cr |
Token ID | 1067 |
Feature activation | -3.49e+00 |
Pos logit contributions | |
ons | +0.029 |
bes | +0.024 |
ages | +0.023 |
seek | +0.023 |
ops | +0.022 |
Neg logit contributions | |
dolphin | -0.022 |
mate | -0.022 |
whale | -0.022 |
flood | -0.021 |
wake | -0.021 |
Token | ay |
Token ID | 323 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
ons | +0.271 |
bes | +0.226 |
ages | +0.213 |
con | +0.206 |
seek | +0.201 |
Neg logit contributions | |
rant | -0.177 |
ksh | -0.177 |
iet | -0.164 |
flood | -0.162 |
mate | -0.160 |
Token | ons |
Token ID | 684 |
Feature activation | -1.44e+01 |
Pos logit contributions | |
bes | +0.001 |
ons | +0.001 |
seek | -0.001 |
ages | -0.007 |
con | -0.008 |
Neg logit contributions | |
rant | -0.191 |
coming | -0.189 |
whale | -0.189 |
iet | -0.186 |
ksh | -0.185 |
Token | back |
Token ID | 736 |
Feature activation | +4.86e+00 |
Pos logit contributions | |
ons | +1.134 |
bes | +0.903 |
ages | +0.860 |
con | +0.804 |
on | +0.802 |
Neg logit contributions | |
ksh | -0.762 |
rant | -0.751 |
flood | -0.745 |
dolphin | -0.723 |
whale | -0.708 |
Token | . |
Token ID | 13 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
rant | +0.214 |
coming | +0.208 |
ksh | +0.206 |
flood | +0.205 |
mate | +0.199 |
Neg logit contributions | |
ons | -0.420 |
bes | -0.345 |
ages | -0.334 |
aid | -0.313 |
con | -0.313 |
Token | He |
Token ID | 679 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
ksh | +0.001 |
flood | +0.001 |
rant | +0.001 |
dolphin | +0.001 |
whale | +0.001 |
Neg logit contributions | |
ons | -0.001 |
bes | -0.001 |
ages | -0.001 |
on | -0.001 |
con | -0.001 |
Token | wants |
Token ID | 3382 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
ons | +0.118 |
bes | +0.091 |
ages | +0.087 |
on | +0.082 |
con | +0.080 |
Neg logit contributions | |
ksh | -0.095 |
rant | -0.093 |
flood | -0.093 |
dolphin | -0.090 |
whale | -0.089 |
Token | his |
Token ID | 465 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
ons | +0.106 |
bes | +0.074 |
ages | +0.071 |
on | +0.071 |
aid | +0.066 |
Neg logit contributions | |
ksh | -0.109 |
rant | -0.104 |
flood | -0.103 |
dolphin | -0.098 |
iet | -0.098 |
Token | are |
Token ID | 389 |
Feature activation | +6.86e-01 |
Pos logit contributions | |
flood | +0.182 |
ending | +0.171 |
whale | +0.167 |
dolphin | +0.165 |
coming | +0.164 |
Neg logit contributions | |
ons | -0.314 |
bes | -0.280 |
ages | -0.267 |
aid | -0.257 |
con | -0.253 |
Token | papers |
Token ID | 9473 |
Feature activation | -8.98e+00 |
Pos logit contributions | |
flood | +0.039 |
ksh | +0.038 |
rant | +0.037 |
whale | +0.037 |
dolphin | +0.037 |
Neg logit contributions | |
ons | -0.051 |
bes | -0.041 |
ages | -0.039 |
con | -0.037 |
seek | -0.036 |
Token | and |
Token ID | 290 |
Feature activation | -4.00e-01 |
Pos logit contributions | |
ons | +0.703 |
bes | +0.559 |
ages | +0.541 |
con | +0.506 |
seek | +0.501 |
Neg logit contributions | |
flood | -0.462 |
rant | -0.458 |
ksh | -0.454 |
dolphin | -0.438 |
coming | -0.437 |
Token | cr |
Token ID | 1067 |
Feature activation | -3.49e+00 |
Pos logit contributions | |
ons | +0.025 |
bes | +0.019 |
ages | +0.018 |
seek | +0.017 |
con | +0.016 |
Neg logit contributions | |
flood | -0.027 |
ksh | -0.026 |
dolphin | -0.026 |
whale | -0.026 |
rant | -0.026 |
Token | ay |
Token ID | 323 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
ons | +0.260 |
bes | +0.208 |
ages | +0.197 |
con | +0.186 |
seek | +0.183 |
Neg logit contributions | |
ksh | -0.194 |
rant | -0.193 |
flood | -0.186 |
dolphin | -0.178 |
iet | -0.177 |
Token | ons |
Token ID | 684 |
Feature activation | -1.44e+01 |
Pos logit contributions | |
ons | +0.001 |
bes | -0.004 |
seek | -0.009 |
tail | -0.010 |
ages | -0.013 |
Neg logit contributions | |
coming | -0.195 |
rant | -0.194 |
ksh | -0.192 |
whale | -0.188 |
iet | -0.186 |
Token | and |
Token ID | 290 |
Feature activation | -4.01e-01 |
Pos logit contributions | |
ons | +1.131 |
bes | +0.909 |
ages | +0.867 |
con | +0.815 |
seek | +0.812 |
Neg logit contributions | |
rant | -0.737 |
ksh | -0.735 |
flood | -0.731 |
dolphin | -0.701 |
coming | -0.698 |
Token | stickers |
Token ID | 28568 |
Feature activation | -7.57e+00 |
Pos logit contributions | |
ons | +0.025 |
bes | +0.019 |
ages | +0.018 |
seek | +0.017 |
on | +0.017 |
Neg logit contributions | |
flood | -0.026 |
ksh | -0.026 |
dolphin | -0.025 |
whale | -0.025 |
rant | -0.025 |
Token | and |
Token ID | 290 |
Feature activation | -3.90e-01 |
Pos logit contributions | |
ons | +0.595 |
bes | +0.481 |
ages | +0.456 |
con | +0.428 |
seek | +0.426 |
Neg logit contributions | |
rant | -0.388 |
ksh | -0.385 |
flood | -0.376 |
dolphin | -0.364 |
coming | -0.363 |
Token | glue |
Token ID | 22749 |
Feature activation | -4.19e+00 |
Pos logit contributions | |
ons | +0.026 |
bes | +0.020 |
ages | +0.019 |
on | +0.018 |
seek | +0.018 |
Neg logit contributions | |
ksh | -0.024 |
flood | -0.023 |
mate | -0.023 |
dolphin | -0.023 |
rant | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | +2.66e-02 |
Pos logit contributions | |
ons | +0.321 |
bes | +0.253 |
ages | +0.242 |
on | +0.227 |
con | +0.225 |
Neg logit contributions | |
ksh | -0.228 |
rant | -0.226 |
flood | -0.223 |
dolphin | -0.216 |
whale | -0.213 |
Token | line |
Token ID | 1627 |
Feature activation | -3.00e+00 |
Pos logit contributions | |
ons | +0.256 |
bes | +0.207 |
ages | +0.203 |
aid | +0.193 |
seek | +0.192 |
Neg logit contributions | |
flood | -0.136 |
ksh | -0.134 |
rant | -0.133 |
whale | -0.129 |
dolphin | -0.129 |
Token | with |
Token ID | 351 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
ons | +0.226 |
bes | +0.185 |
ages | +0.176 |
aid | +0.164 |
con | +0.164 |
Neg logit contributions | |
rant | -0.156 |
coming | -0.156 |
flood | -0.154 |
ksh | -0.151 |
mate | -0.146 |
Token | the |
Token ID | 262 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
ons | +0.119 |
bes | +0.089 |
ages | +0.085 |
on | +0.080 |
con | +0.078 |
Neg logit contributions | |
ksh | -0.107 |
rant | -0.104 |
flood | -0.104 |
dolphin | -0.101 |
whale | -0.099 |
Token | cr |
Token ID | 1067 |
Feature activation | -3.49e+00 |
Pos logit contributions | |
ons | +0.095 |
bes | +0.072 |
ages | +0.069 |
on | +0.065 |
con | +0.064 |
Neg logit contributions | |
ksh | -0.081 |
rant | -0.079 |
flood | -0.078 |
dolphin | -0.076 |
whale | -0.074 |
Token | ay |
Token ID | 323 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
ons | +0.265 |
bes | +0.215 |
ages | +0.204 |
con | +0.191 |
seek | +0.190 |
Neg logit contributions | |
ksh | -0.187 |
rant | -0.186 |
flood | -0.178 |
dolphin | -0.173 |
iet | -0.171 |
Token | ons |
Token ID | 684 |
Feature activation | -1.44e+01 |
Pos logit contributions | |
ons | +0.007 |
bes | -0.002 |
seek | -0.008 |
ages | -0.008 |
tail | -0.013 |
Neg logit contributions | |
rant | -0.196 |
whale | -0.193 |
coming | -0.193 |
ksh | -0.191 |
iet | -0.190 |
Token | . |
Token ID | 13 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
ons | +1.062 |
bes | +0.815 |
ages | +0.786 |
on | +0.745 |
es | +0.731 |
Neg logit contributions | |
ksh | -0.848 |
rant | -0.824 |
flood | -0.823 |
dolphin | -0.795 |
whale | -0.781 |
Token | Then |
Token ID | 3244 |
Feature activation | -5.91e+00 |
Pos logit contributions | |
ksh | +0.001 |
rant | +0.001 |
flood | +0.001 |
dolphin | +0.001 |
whale | +0.001 |
Neg logit contributions | |
ons | -0.001 |
bes | -0.001 |
ages | -0.001 |
on | -0.001 |
aid | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | -4.48e-01 |
Pos logit contributions | |
ons | +0.472 |
bes | +0.380 |
ages | +0.362 |
on | +0.342 |
con | +0.341 |
Neg logit contributions | |
flood | -0.304 |
ksh | -0.294 |
dolphin | -0.290 |
rant | -0.289 |
ending | -0.285 |
Token | they |
Token ID | 484 |
Feature activation | +2.69e+00 |
Pos logit contributions | |
ons | +0.036 |
bes | +0.028 |
ages | +0.027 |
on | +0.025 |
con | +0.025 |
Neg logit contributions | |
ksh | -0.023 |
flood | -0.023 |
rant | -0.023 |
dolphin | -0.022 |
whale | -0.022 |
Token | drew |
Token ID | 9859 |
Feature activation | -2.39e+00 |
Pos logit contributions | |
ksh | +0.135 |
rant | +0.135 |
flood | +0.133 |
dolphin | +0.129 |
whale | +0.127 |
Neg logit contributions | |
ons | -0.216 |
bes | -0.176 |
ages | -0.168 |
con | -0.157 |
seek | -0.156 |
Token | her |
Token ID | 607 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
flood | +0.010 |
coming | +0.010 |
rant | +0.009 |
ksh | +0.009 |
dolphin | +0.009 |
Neg logit contributions | |
ons | -0.018 |
bes | -0.016 |
ages | -0.015 |
con | -0.015 |
seek | -0.014 |
Token | diary |
Token ID | 26339 |
Feature activation | -6.33e+00 |
Pos logit contributions | |
ons | +0.112 |
ages | +0.092 |
seek | +0.091 |
bes | +0.090 |
rob | +0.086 |
Neg logit contributions | |
dolphin | -0.077 |
flood | -0.075 |
coming | -0.074 |
whale | -0.074 |
harbor | -0.072 |
Token | and |
Token ID | 290 |
Feature activation | -3.98e-01 |
Pos logit contributions | |
ons | +0.547 |
bes | +0.431 |
ages | +0.417 |
rob | +0.403 |
con | +0.402 |
Neg logit contributions | |
coming | -0.274 |
flood | -0.272 |
rant | -0.271 |
ksh | -0.269 |
mate | -0.263 |
Token | cr |
Token ID | 1067 |
Feature activation | -3.51e+00 |
Pos logit contributions | |
ons | +0.033 |
rob | +0.029 |
ages | +0.029 |
seek | +0.029 |
bes | +0.028 |
Neg logit contributions | |
coming | -0.016 |
wake | -0.015 |
flood | -0.015 |
dolphin | -0.015 |
spr | -0.014 |
Token | ay |
Token ID | 323 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
ons | +0.274 |
bes | +0.229 |
ages | +0.217 |
con | +0.207 |
seek | +0.203 |
Neg logit contributions | |
rant | -0.174 |
ksh | -0.173 |
iet | -0.161 |
flood | -0.158 |
mate | -0.157 |
Token | ons |
Token ID | 684 |
Feature activation | -1.44e+01 |
Pos logit contributions | |
seek | +0.011 |
bes | +0.011 |
ages | +0.004 |
ons | +0.001 |
tail | +0.000 |
Neg logit contributions | |
coming | -0.179 |
them | -0.176 |
rant | -0.174 |
us | -0.170 |
iet | -0.169 |
Token | . |
Token ID | 13 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
ons | +1.140 |
bes | +0.913 |
ages | +0.870 |
con | +0.811 |
seek | +0.804 |
Neg logit contributions | |
ksh | -0.745 |
rant | -0.739 |
flood | -0.727 |
dolphin | -0.711 |
coming | -0.694 |
Token | She |
Token ID | 1375 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
ksh | +0.001 |
rant | +0.001 |
dolphin | +0.001 |
flood | +0.001 |
coming | +0.001 |
Neg logit contributions | |
ons | -0.001 |
ages | -0.001 |
bes | -0.001 |
aid | -0.001 |
on | -0.001 |
Token | put |
Token ID | 1234 |
Feature activation | +7.97e-01 |
Pos logit contributions | |
ons | +0.177 |
ages | +0.143 |
bes | +0.143 |
seek | +0.139 |
con | +0.139 |
Neg logit contributions | |
rant | -0.080 |
dolphin | -0.080 |
ksh | -0.078 |
whale | -0.078 |
coming | -0.076 |
Token | them |
Token ID | 606 |
Feature activation | +8.46e+00 |
Pos logit contributions | |
rant | +0.041 |
ksh | +0.040 |
flood | +0.039 |
dolphin | +0.039 |
whale | +0.038 |
Neg logit contributions | |
ons | -0.062 |
bes | -0.051 |
ages | -0.050 |
seek | -0.047 |
con | -0.045 |
Token | in |
Token ID | 287 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
rant | +0.446 |
ksh | +0.441 |
flood | +0.435 |
coming | +0.434 |
dolphin | +0.426 |
Neg logit contributions | |
ons | -0.636 |
bes | -0.511 |
ages | -0.498 |
con | -0.484 |
seek | -0.479 |
Token | took |
Token ID | 1718 |
Feature activation | +9.77e-01 |
Pos logit contributions | |
ons | +0.173 |
bes | +0.148 |
ages | +0.145 |
seek | +0.140 |
con | +0.139 |
Neg logit contributions | |
rant | -0.083 |
whale | -0.080 |
wake | -0.079 |
rum | -0.079 |
coming | -0.078 |
Token | out |
Token ID | 503 |
Feature activation | +6.45e+00 |
Pos logit contributions | |
coming | +0.038 |
rant | +0.037 |
wake | +0.035 |
like | +0.034 |
them | +0.034 |
Neg logit contributions | |
ons | -0.081 |
bes | -0.077 |
ages | -0.073 |
rob | -0.071 |
seek | -0.069 |
Token | some |
Token ID | 617 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
coming | +0.293 |
dolphin | +0.293 |
iet | +0.279 |
mate | +0.279 |
whale | +0.278 |
Neg logit contributions | |
ons | -0.498 |
bes | -0.455 |
ages | -0.438 |
rob | -0.421 |
seek | -0.417 |
Token | cr |
Token ID | 1067 |
Feature activation | -3.51e+00 |
Pos logit contributions | |
ons | +0.145 |
bes | +0.136 |
ages | +0.129 |
seek | +0.129 |
rob | +0.116 |
Neg logit contributions | |
coming | -0.106 |
dolphin | -0.105 |
whale | -0.104 |
rant | -0.102 |
flood | -0.100 |
Token | ay |
Token ID | 323 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
ons | +0.269 |
bes | +0.229 |
ages | +0.214 |
con | +0.206 |
seek | +0.204 |
Neg logit contributions | |
rant | -0.175 |
ksh | -0.174 |
iet | -0.162 |
dolphin | -0.155 |
mate | -0.155 |
Token | ons |
Token ID | 684 |
Feature activation | -1.44e+01 |
Pos logit contributions | |
bes | +0.011 |
seek | +0.007 |
ons | +0.001 |
tail | +0.000 |
ages | -0.001 |
Neg logit contributions | |
coming | -0.181 |
rant | -0.180 |
iet | -0.177 |
whale | -0.177 |
dolphin | -0.176 |
Token | and |
Token ID | 290 |
Feature activation | -4.01e-01 |
Pos logit contributions | |
ons | +1.178 |
bes | +0.966 |
ages | +0.910 |
con | +0.856 |
seek | +0.854 |
Neg logit contributions | |
ksh | -0.686 |
rant | -0.684 |
dolphin | -0.667 |
flood | -0.664 |
whale | -0.649 |
Token | paper |
Token ID | 3348 |
Feature activation | -3.69e+00 |
Pos logit contributions | |
ons | +0.031 |
bes | +0.028 |
seek | +0.026 |
ages | +0.026 |
ittens | +0.026 |
Neg logit contributions | |
coming | -0.017 |
rant | -0.017 |
dolphin | -0.017 |
whale | -0.017 |
wake | -0.016 |
Token | , |
Token ID | 11 |
Feature activation | -4.58e-01 |
Pos logit contributions | |
ons | +0.285 |
bes | +0.229 |
ages | +0.218 |
seek | +0.206 |
aid | +0.205 |
Neg logit contributions | |
rant | -0.189 |
ksh | -0.187 |
dolphin | -0.183 |
flood | -0.181 |
coming | -0.180 |
Token | and |
Token ID | 290 |
Feature activation | -4.02e-01 |
Pos logit contributions | |
ons | +0.037 |
bes | +0.031 |
ages | +0.030 |
seek | +0.028 |
con | +0.027 |
Neg logit contributions | |
rant | -0.021 |
coming | -0.021 |
ksh | -0.021 |
dolphin | -0.020 |
flood | -0.020 |
Token | she |
Token ID | 673 |
Feature activation | -3.25e+00 |
Pos logit contributions | |
ons | +0.032 |
bes | +0.027 |
ages | +0.025 |
seek | +0.024 |
con | +0.023 |
Neg logit contributions | |
rant | -0.019 |
coming | -0.019 |
dolphin | -0.019 |
flood | -0.019 |
whale | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
rant | +0.212 |
ksh | +0.208 |
coming | +0.206 |
flood | +0.202 |
dolphin | +0.201 |
Neg logit contributions | |
ons | -0.319 |
bes | -0.263 |
ages | -0.251 |
con | -0.234 |
aid | -0.233 |
Token | She |
Token ID | 1375 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
ksh | +0.001 |
rant | +0.001 |
flood | +0.001 |
dolphin | +0.001 |
whale | +0.001 |
Neg logit contributions | |
ons | -0.002 |
bes | -0.002 |
ages | -0.001 |
on | -0.001 |
con | -0.001 |
Token | used |
Token ID | 973 |
Feature activation | -4.69e+00 |
Pos logit contributions | |
ons | +0.171 |
bes | +0.143 |
ages | +0.139 |
con | +0.131 |
seek | +0.129 |
Neg logit contributions | |
whale | -0.091 |
rant | -0.091 |
dolphin | -0.090 |
ksh | -0.086 |
flood | -0.086 |
Token | cr |
Token ID | 1067 |
Feature activation | -3.51e+00 |
Pos logit contributions | |
ons | +0.346 |
bes | +0.296 |
ages | +0.287 |
seek | +0.268 |
rob | +0.258 |
Neg logit contributions | |
rant | -0.242 |
coming | -0.239 |
dolphin | -0.237 |
mate | -0.234 |
flood | -0.234 |
Token | ay |
Token ID | 323 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
ons | +0.269 |
bes | +0.238 |
ages | +0.220 |
seek | +0.209 |
con | +0.209 |
Neg logit contributions | |
rant | -0.169 |
ksh | -0.162 |
iet | -0.154 |
mate | -0.149 |
coming | -0.146 |
Token | ons |
Token ID | 684 |
Feature activation | -1.44e+01 |
Pos logit contributions | |
bes | +0.005 |
ons | +0.000 |
seek | -0.003 |
tail | -0.004 |
ages | -0.007 |
Neg logit contributions | |
coming | -0.186 |
rant | -0.185 |
iet | -0.185 |
whale | -0.180 |
mate | -0.178 |
Token | of |
Token ID | 286 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
ons | +1.128 |
bes | +0.927 |
ages | +0.866 |
con | +0.810 |
seek | +0.808 |
Neg logit contributions | |
rant | -0.735 |
ksh | -0.729 |
flood | -0.707 |
dolphin | -0.704 |
coming | -0.690 |
Token | different |
Token ID | 1180 |
Feature activation | +3.84e+00 |
Pos logit contributions | |
ons | +0.122 |
bes | +0.103 |
ages | +0.099 |
seek | +0.093 |
aid | +0.088 |
Neg logit contributions | |
rant | -0.096 |
dolphin | -0.094 |
whale | -0.093 |
ksh | -0.092 |
flood | -0.092 |
Token | colors |
Token ID | 7577 |
Feature activation | -8.88e-01 |
Pos logit contributions | |
rant | +0.199 |
ksh | +0.190 |
coming | +0.188 |
appearance | +0.183 |
mate | +0.182 |
Neg logit contributions | |
ons | -0.274 |
bes | -0.240 |
ages | -0.222 |
aid | -0.217 |
seek | -0.211 |
Token | . |
Token ID | 13 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
ons | +0.067 |
bes | +0.054 |
ages | +0.052 |
seek | +0.048 |
con | +0.048 |
Neg logit contributions | |
rant | -0.048 |
ksh | -0.047 |
flood | -0.046 |
dolphin | -0.045 |
whale | -0.045 |
Token | She |
Token ID | 1375 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
ksh | +0.001 |
rant | +0.001 |
flood | +0.001 |
dolphin | +0.001 |
whale | +0.001 |
Neg logit contributions | |
ons | -0.001 |
bes | -0.001 |
ages | -0.001 |
on | -0.001 |
con | -0.001 |
Token | would |
Token ID | 561 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
ksh | +0.132 |
flood | +0.132 |
rant | +0.131 |
dolphin | +0.127 |
whale | +0.125 |
Neg logit contributions | |
ons | -0.220 |
bes | -0.178 |
ages | -0.169 |
on | -0.159 |
con | -0.158 |
Token | spend |
Token ID | 4341 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
ksh | +0.072 |
rant | +0.068 |
flood | +0.063 |
dolphin | +0.062 |
whale | +0.062 |
Neg logit contributions | |
ons | -0.099 |
bes | -0.079 |
ages | -0.075 |
on | -0.072 |
seek | -0.071 |
Token | their |
Token ID | 511 |
Feature activation | +5.89e+00 |
Pos logit contributions | |
ons | +0.174 |
bes | +0.141 |
ages | +0.138 |
on | +0.132 |
con | +0.130 |
Neg logit contributions | |
flood | -0.094 |
rant | -0.091 |
ksh | -0.091 |
dolphin | -0.089 |
coming | -0.089 |
Token | after |
Token ID | 706 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
ksh | +0.318 |
flood | +0.316 |
rant | +0.308 |
dolphin | +0.304 |
whale | +0.299 |
Neg logit contributions | |
ons | -0.454 |
bes | -0.356 |
ages | -0.339 |
on | -0.324 |
es | -0.319 |
Token | no |
Token ID | 3919 |
Feature activation | -5.29e+00 |
Pos logit contributions | |
coming | +0.058 |
flood | +0.056 |
dolphin | +0.054 |
whale | +0.052 |
harbor | +0.049 |
Neg logit contributions | |
ons | -0.088 |
bes | -0.076 |
ages | -0.072 |
aid | -0.071 |
aces | -0.068 |
Token | ons |
Token ID | 684 |
Feature activation | -1.44e+01 |
Pos logit contributions | |
bes | +0.206 |
ons | +0.199 |
con | +0.185 |
ages | +0.162 |
seek | +0.153 |
Neg logit contributions | |
coming | -0.409 |
flood | -0.384 |
ending | -0.379 |
storm | -0.371 |
whale | -0.363 |
Token | in |
Token ID | 287 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
ons | +1.079 |
bes | +0.864 |
ages | +0.802 |
on | +0.760 |
con | +0.752 |
Neg logit contributions | |
ksh | -0.777 |
flood | -0.775 |
dolphin | -0.752 |
coming | -0.750 |
rant | -0.750 |
Token | the |
Token ID | 262 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
ksh | +0.086 |
flood | +0.085 |
rant | +0.085 |
dolphin | +0.082 |
whale | +0.081 |
Neg logit contributions | |
ons | -0.107 |
bes | -0.083 |
ages | -0.080 |
on | -0.074 |
con | -0.073 |
Token | shade |
Token ID | 17979 |
Feature activation | -4.94e-01 |
Pos logit contributions | |
ons | +0.089 |
bes | +0.064 |
ages | +0.064 |
con | +0.062 |
on | +0.061 |
Neg logit contributions | |
rant | -0.085 |
ksh | -0.082 |
mate | -0.076 |
extingu | -0.075 |
dolphin | -0.073 |
Token | together |
Token ID | 1978 |
Feature activation | -4.15e+00 |
Pos logit contributions | |
ons | +0.039 |
bes | +0.033 |
ages | +0.030 |
aid | +0.029 |
con | +0.029 |
Neg logit contributions | |
flood | -0.025 |
rant | -0.025 |
ksh | -0.024 |
coming | -0.023 |
whale | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | +3.43e-02 |
Pos logit contributions | |
ons | +0.292 |
bes | +0.219 |
ages | +0.209 |
es | +0.200 |
on | +0.199 |
Neg logit contributions | |
ksh | -0.261 |
rant | -0.253 |
flood | -0.252 |
dolphin | -0.243 |
whale | -0.240 |
Token | car |
Token ID | 1097 |
Feature activation | -3.33e+00 |
Pos logit contributions | |
ons | +0.157 |
bes | +0.123 |
ages | +0.120 |
es | +0.115 |
on | +0.113 |
Neg logit contributions | |
flood | -0.093 |
ksh | -0.088 |
dolphin | -0.088 |
whale | -0.087 |
ending | -0.086 |
Token | with |
Token ID | 351 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
ons | +0.270 |
bes | +0.222 |
ages | +0.211 |
aid | +0.193 |
con | +0.193 |
Neg logit contributions | |
flood | -0.167 |
coming | -0.166 |
rant | -0.160 |
ksh | -0.157 |
ending | -0.157 |
Token | his |
Token ID | 465 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
ons | +0.112 |
bes | +0.088 |
ages | +0.085 |
con | +0.077 |
es | +0.075 |
Neg logit contributions | |
flood | -0.108 |
rant | -0.108 |
ksh | -0.106 |
coming | -0.104 |
dolphin | -0.104 |
Token | cr |
Token ID | 1067 |
Feature activation | -3.50e+00 |
Pos logit contributions | |
ons | +0.161 |
bes | +0.130 |
ages | +0.126 |
seek | +0.123 |
es | +0.119 |
Neg logit contributions | |
flood | -0.083 |
dolphin | -0.081 |
ksh | -0.080 |
whale | -0.080 |
mate | -0.079 |
Token | ay |
Token ID | 323 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
ons | +0.274 |
bes | +0.230 |
ages | +0.218 |
con | +0.208 |
seek | +0.203 |
Neg logit contributions | |
ksh | -0.171 |
rant | -0.169 |
flood | -0.159 |
iet | -0.158 |
mate | -0.157 |
Token | ons |
Token ID | 684 |
Feature activation | -1.44e+01 |
Pos logit contributions | |
bes | +0.006 |
seek | +0.005 |
ons | +0.004 |
ages | +0.003 |
tail | +0.003 |
Neg logit contributions | |
coming | -0.185 |
rant | -0.181 |
iet | -0.178 |
mate | -0.177 |
whale | -0.177 |
Token | , |
Token ID | 11 |
Feature activation | -4.48e-01 |
Pos logit contributions | |
ons | +1.049 |
bes | +0.810 |
ages | +0.775 |
on | +0.725 |
con | +0.714 |
Neg logit contributions | |
ksh | -0.851 |
rant | -0.836 |
flood | -0.834 |
dolphin | -0.808 |
whale | -0.797 |
Token | but |
Token ID | 475 |
Feature activation | +4.92e+00 |
Pos logit contributions | |
ons | +0.044 |
bes | +0.037 |
ages | +0.036 |
on | +0.034 |
con | +0.034 |
Neg logit contributions | |
ksh | -0.015 |
rant | -0.015 |
flood | -0.014 |
dolphin | -0.014 |
whale | -0.013 |
Token | he |
Token ID | 339 |
Feature activation | -2.44e+00 |
Pos logit contributions | |
flood | +0.328 |
dolphin | +0.321 |
coming | +0.319 |
mate | +0.318 |
ending | +0.317 |
Neg logit contributions | |
ons | -0.314 |
bes | -0.243 |
ages | -0.232 |
con | -0.213 |
seek | -0.211 |
Token | didn |
Token ID | 1422 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
ons | +0.209 |
bes | +0.169 |
ages | +0.162 |
con | +0.153 |
on | +0.152 |
Neg logit contributions | |
ksh | -0.112 |
rant | -0.110 |
flood | -0.110 |
dolphin | -0.107 |
whale | -0.104 |
Token | 't |
Token ID | 470 |
Feature activation | -4.62e+00 |
Pos logit contributions | |
ons | +0.084 |
ages | +0.069 |
bes | +0.068 |
aid | +0.063 |
con | +0.062 |
Neg logit contributions | |
flood | -0.073 |
coming | -0.072 |
rant | -0.069 |
wake | -0.069 |
ksh | -0.067 |
Token | many |
Token ID | 867 |
Feature activation | -7.71e+00 |
Pos logit contributions | |
ons | +0.138 |
bes | +0.110 |
ages | +0.105 |
con | +0.098 |
on | +0.098 |
Neg logit contributions | |
ksh | -0.101 |
rant | -0.099 |
flood | -0.097 |
dolphin | -0.097 |
coming | -0.096 |
Token | colors |
Token ID | 7577 |
Feature activation | -8.86e-01 |
Pos logit contributions | |
ons | +0.534 |
bes | +0.460 |
seek | +0.451 |
ages | +0.448 |
aid | +0.425 |
Neg logit contributions | |
dolphin | -0.395 |
coming | -0.393 |
iet | -0.389 |
whale | -0.385 |
rant | -0.378 |
Token | of |
Token ID | 286 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
ons | +0.068 |
bes | +0.054 |
ages | +0.052 |
con | +0.048 |
on | +0.048 |
Neg logit contributions | |
rant | -0.048 |
ksh | -0.048 |
flood | -0.047 |
dolphin | -0.046 |
whale | -0.045 |
Token | cr |
Token ID | 1067 |
Feature activation | -3.50e+00 |
Pos logit contributions | |
ons | +0.129 |
bes | +0.101 |
ages | +0.098 |
on | +0.090 |
seek | +0.089 |
Neg logit contributions | |
ksh | -0.097 |
rant | -0.096 |
flood | -0.096 |
dolphin | -0.094 |
whale | -0.092 |
Token | ay |
Token ID | 323 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
ons | +0.267 |
bes | +0.232 |
ages | +0.219 |
con | +0.207 |
seek | +0.207 |
Neg logit contributions | |
rant | -0.169 |
ksh | -0.164 |
iet | -0.159 |
coming | -0.148 |
mate | -0.147 |
Token | ons |
Token ID | 684 |
Feature activation | -1.44e+01 |
Pos logit contributions | |
bes | +0.003 |
seek | +0.001 |
ons | +0.000 |
ages | -0.003 |
tail | -0.004 |
Neg logit contributions | |
coming | -0.185 |
rant | -0.183 |
iet | -0.182 |
whale | -0.177 |
them | -0.176 |
Token | . |
Token ID | 13 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
ons | +1.108 |
bes | +0.889 |
ages | +0.844 |
con | +0.786 |
seek | +0.781 |
Neg logit contributions | |
rant | -0.765 |
ksh | -0.755 |
flood | -0.745 |
dolphin | -0.731 |
coming | -0.722 |
Token | She |
Token ID | 1375 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
ksh | +0.002 |
flood | +0.002 |
rant | +0.002 |
dolphin | +0.002 |
whale | +0.001 |
Neg logit contributions | |
ons | -0.002 |
bes | -0.001 |
ages | -0.001 |
on | -0.001 |
con | -0.001 |
Token | likes |
Token ID | 7832 |
Feature activation | -6.01e-01 |
Pos logit contributions | |
ons | +0.159 |
bes | +0.125 |
ages | +0.121 |
on | +0.112 |
con | +0.112 |
Neg logit contributions | |
ksh | -0.110 |
rant | -0.109 |
flood | -0.107 |
dolphin | -0.105 |
whale | -0.103 |
Token | red |
Token ID | 2266 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
ons | +0.046 |
bes | +0.036 |
ages | +0.035 |
on | +0.032 |
con | +0.032 |
Neg logit contributions | |
ksh | -0.033 |
rant | -0.033 |
flood | -0.032 |
dolphin | -0.032 |
whale | -0.031 |
Token | and |
Token ID | 290 |
Feature activation | -3.82e-01 |
Pos logit contributions | |
ksh | +0.097 |
rant | +0.095 |
dolphin | +0.094 |
coming | +0.093 |
dot | +0.091 |
Neg logit contributions | |
ons | -0.143 |
bes | -0.123 |
ages | -0.119 |
aid | -0.114 |
seek | -0.112 |
Token | a |
Token ID | 257 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
ksh | +0.156 |
rant | +0.151 |
flood | +0.150 |
dolphin | +0.143 |
whale | +0.141 |
Neg logit contributions | |
ons | -0.167 |
bes | -0.123 |
ages | -0.119 |
on | -0.113 |
es | -0.111 |
Token | box |
Token ID | 3091 |
Feature activation | -2.65e+00 |
Pos logit contributions | |
ons | +0.099 |
bes | +0.076 |
ages | +0.071 |
on | +0.070 |
con | +0.069 |
Neg logit contributions | |
ksh | -0.090 |
rant | -0.087 |
flood | -0.080 |
iet | -0.077 |
ending | -0.075 |
Token | of |
Token ID | 286 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
ons | +0.185 |
bes | +0.139 |
ages | +0.132 |
on | +0.127 |
es | +0.122 |
Neg logit contributions | |
ksh | -0.170 |
flood | -0.163 |
rant | -0.162 |
dolphin | -0.158 |
whale | -0.155 |
Token | cr |
Token ID | 1067 |
Feature activation | -3.49e+00 |
Pos logit contributions | |
ons | +0.116 |
bes | +0.092 |
ages | +0.089 |
seek | +0.083 |
es | +0.080 |
Neg logit contributions | |
flood | -0.107 |
dolphin | -0.106 |
rant | -0.104 |
whale | -0.104 |
ksh | -0.104 |
Token | ay |
Token ID | 323 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
ons | +0.263 |
bes | +0.216 |
ages | +0.206 |
con | +0.193 |
seek | +0.191 |
Neg logit contributions | |
rant | -0.185 |
ksh | -0.182 |
flood | -0.173 |
iet | -0.170 |
dolphin | -0.168 |
Token | ons |
Token ID | 684 |
Feature activation | -1.44e+01 |
Pos logit contributions | |
ons | +0.001 |
bes | -0.003 |
seek | -0.006 |
ages | -0.008 |
tail | -0.012 |
Neg logit contributions | |
coming | -0.193 |
rant | -0.193 |
whale | -0.189 |
iet | -0.187 |
dolphin | -0.186 |
Token | and |
Token ID | 290 |
Feature activation | -4.00e-01 |
Pos logit contributions | |
ons | +1.137 |
bes | +0.905 |
ages | +0.868 |
on | +0.809 |
con | +0.806 |
Neg logit contributions | |
ksh | -0.744 |
rant | -0.740 |
flood | -0.732 |
dolphin | -0.713 |
whale | -0.701 |
Token | a |
Token ID | 257 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
ons | +0.030 |
bes | +0.025 |
ages | +0.023 |
seek | +0.022 |
aid | +0.022 |
Neg logit contributions | |
dolphin | -0.021 |
whale | -0.021 |
mate | -0.020 |
rant | -0.020 |
coming | -0.020 |
Token | sketch |
Token ID | 17548 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
ons | +0.104 |
bes | +0.081 |
ages | +0.077 |
on | +0.073 |
con | +0.072 |
Neg logit contributions | |
ksh | -0.081 |
rant | -0.080 |
flood | -0.079 |
dolphin | -0.077 |
whale | -0.076 |
Token | book |
Token ID | 2070 |
Feature activation | -5.16e+00 |
Pos logit contributions | |
ksh | +0.087 |
flood | +0.086 |
rant | +0.086 |
dolphin | +0.084 |
whale | +0.083 |
Neg logit contributions | |
ons | -0.065 |
bes | -0.046 |
ages | -0.043 |
on | -0.039 |
con | -0.038 |
Token | . |
Token ID | 13 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
ons | +0.376 |
bes | +0.292 |
ages | +0.279 |
on | +0.263 |
es | +0.258 |
Neg logit contributions | |
ksh | -0.309 |
flood | -0.301 |
rant | -0.299 |
dolphin | -0.289 |
whale | -0.284 |
Token | job |
Token ID | 1693 |
Feature activation | -6.59e+00 |
Pos logit contributions | |
flood | +0.160 |
ksh | +0.159 |
dolphin | +0.154 |
rant | +0.154 |
ending | +0.154 |
Neg logit contributions | |
ons | -0.160 |
bes | -0.120 |
ages | -0.113 |
aid | -0.105 |
es | -0.105 |
Token | organizing |
Token ID | 16924 |
Feature activation | +4.11e-01 |
Pos logit contributions | |
ons | +0.470 |
bes | +0.367 |
ages | +0.348 |
aid | +0.328 |
con | +0.326 |
Neg logit contributions | |
flood | -0.385 |
ksh | -0.383 |
rant | -0.378 |
coming | -0.371 |
dolphin | -0.368 |
Token | your |
Token ID | 534 |
Feature activation | -3.44e+00 |
Pos logit contributions | |
rant | +0.023 |
ksh | +0.023 |
flood | +0.022 |
dolphin | +0.022 |
coming | +0.022 |
Neg logit contributions | |
ons | -0.029 |
ages | -0.023 |
bes | -0.023 |
aid | -0.022 |
seek | -0.022 |
Token | cr |
Token ID | 1067 |
Feature activation | -3.50e+00 |
Pos logit contributions | |
ons | +0.246 |
bes | +0.192 |
ages | +0.188 |
aid | +0.181 |
seek | +0.178 |
Neg logit contributions | |
flood | -0.194 |
ksh | -0.194 |
whale | -0.187 |
rant | -0.186 |
ending | -0.185 |
Token | ay |
Token ID | 323 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
ons | +0.286 |
bes | +0.258 |
ages | +0.243 |
rob | +0.239 |
aid | +0.238 |
Neg logit contributions | |
ksh | -0.132 |
rant | -0.125 |
iet | -0.125 |
mate | -0.111 |
dot | -0.109 |
Token | ons |
Token ID | 684 |
Feature activation | -1.44e+01 |
Pos logit contributions | |
seek | +0.013 |
bes | +0.012 |
aid | +0.008 |
tail | +0.008 |
con | +0.008 |
Neg logit contributions | |
coming | -0.167 |
iet | -0.164 |
rant | -0.159 |
whale | -0.159 |
y | -0.159 |
Token | by |
Token ID | 416 |
Feature activation | -3.48e-01 |
Pos logit contributions | |
ons | +1.078 |
bes | +0.894 |
ages | +0.822 |
aid | +0.817 |
con | +0.804 |
Neg logit contributions | |
ksh | -0.714 |
rant | -0.708 |
coming | -0.708 |
flood | -0.707 |
like | -0.701 |
Token | color |
Token ID | 3124 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
ons | +0.028 |
bes | +0.022 |
ages | +0.022 |
aid | +0.021 |
con | +0.020 |
Neg logit contributions | |
flood | -0.017 |
rant | -0.017 |
ksh | -0.017 |
dolphin | -0.017 |
whale | -0.017 |
Token | ," |
Token ID | 553 |
Feature activation | -7.87e+00 |
Pos logit contributions | |
ksh | +0.082 |
rant | +0.081 |
flood | +0.079 |
coming | +0.078 |
dolphin | +0.077 |
Neg logit contributions | |
ons | -0.097 |
bes | -0.082 |
seek | -0.076 |
ages | -0.075 |
aid | -0.074 |
Token | she |
Token ID | 673 |
Feature activation | -3.24e+00 |
Pos logit contributions | |
ons | +0.589 |
bes | +0.503 |
seek | +0.498 |
ages | +0.487 |
powers | +0.470 |
Neg logit contributions | |
dolphin | -0.400 |
Why | -0.393 |
coming | -0.388 |
why | -0.388 |
whale | -0.383 |
Token | said |
Token ID | 531 |
Feature activation | -3.13e+00 |
Pos logit contributions | |
ons | +0.259 |
orns | +0.226 |
bes | +0.223 |
igs | +0.210 |
seek | +0.206 |
Neg logit contributions | |
dolphin | -0.151 |
ksh | -0.145 |
mer | -0.144 |
whale | -0.143 |
rum | -0.139 |
Token | a |
Token ID | 257 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
ksh | +0.132 |
rant | +0.129 |
dolphin | +0.127 |
flood | +0.126 |
whale | +0.124 |
Neg logit contributions | |
ons | -0.180 |
bes | -0.146 |
ages | -0.135 |
con | -0.128 |
on | -0.126 |
Token | box |
Token ID | 3091 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
ons | +0.103 |
bes | +0.076 |
on | +0.074 |
ages | +0.074 |
con | +0.071 |
Neg logit contributions | |
ksh | -0.085 |
rant | -0.083 |
flood | -0.078 |
iet | -0.075 |
dolphin | -0.074 |
Token | of |
Token ID | 286 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
ons | +0.188 |
bes | +0.145 |
ages | +0.137 |
on | +0.128 |
con | +0.127 |
Neg logit contributions | |
ksh | -0.160 |
rant | -0.157 |
flood | -0.157 |
dolphin | -0.152 |
whale | -0.150 |
Token | cr |
Token ID | 1067 |
Feature activation | -3.50e+00 |
Pos logit contributions | |
ons | +0.120 |
bes | +0.105 |
ages | +0.105 |
seek | +0.101 |
es | +0.095 |
Neg logit contributions | |
flood | -0.099 |
dolphin | -0.097 |
whale | -0.095 |
coming | -0.091 |
tide | -0.091 |
Token | ay |
Token ID | 323 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
ons | +0.270 |
bes | +0.232 |
ages | +0.219 |
con | +0.212 |
seek | +0.206 |
Neg logit contributions | |
rant | -0.168 |
ksh | -0.167 |
iet | -0.157 |
flood | -0.151 |
mate | -0.150 |
Token | ons |
Token ID | 684 |
Feature activation | -1.44e+01 |
Pos logit contributions | |
bes | +0.018 |
seek | +0.016 |
ages | +0.007 |
tail | +0.006 |
powers | +0.006 |
Neg logit contributions | |
coming | -0.176 |
iet | -0.166 |
rant | -0.166 |
whale | -0.166 |
dolphin | -0.163 |
Token | and |
Token ID | 290 |
Feature activation | -4.06e-01 |
Pos logit contributions | |
ons | +1.199 |
bes | +1.008 |
ages | +0.921 |
seek | +0.891 |
con | +0.889 |
Neg logit contributions | |
rant | -0.656 |
ksh | -0.652 |
dolphin | -0.632 |
coming | -0.630 |
flood | -0.618 |
Token | some |
Token ID | 617 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
ons | +0.030 |
bes | +0.028 |
seek | +0.025 |
ages | +0.024 |
tail | +0.023 |
Neg logit contributions | |
dolphin | -0.020 |
whale | -0.020 |
coming | -0.019 |
flood | -0.019 |
mate | -0.019 |
Token | paper |
Token ID | 3348 |
Feature activation | -3.70e+00 |
Pos logit contributions | |
ons | +0.147 |
bes | +0.120 |
ages | +0.112 |
seek | +0.107 |
es | +0.103 |
Neg logit contributions | |
flood | -0.118 |
ksh | -0.117 |
rant | -0.117 |
dolphin | -0.116 |
whale | -0.114 |
Token | . |
Token ID | 13 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
ons | +0.289 |
bes | +0.232 |
ages | +0.219 |
seek | +0.208 |
on | +0.205 |
Neg logit contributions | |
ksh | -0.189 |
rant | -0.188 |
flood | -0.184 |
dolphin | -0.183 |
whale | -0.180 |
Token | Tim |
Token ID | 5045 |
Feature activation | +1.94e+00 |
Pos logit contributions | |
ksh | +0.001 |
flood | +0.001 |
rant | +0.001 |
dolphin | +0.001 |
coming | +0.001 |
Neg logit contributions | |
ons | -0.002 |
bes | -0.002 |
ages | -0.002 |
es | -0.002 |
aid | -0.002 |
Token | new |
Token ID | 649 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
ons | +0.101 |
bes | +0.074 |
on | +0.072 |
ages | +0.071 |
aid | +0.069 |
Neg logit contributions | |
rant | -0.089 |
ksh | -0.088 |
flood | -0.078 |
mate | -0.077 |
iet | -0.076 |
Token | box |
Token ID | 3091 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
ons | +0.008 |
bes | +0.007 |
ages | +0.006 |
on | +0.006 |
con | +0.006 |
Neg logit contributions | |
ksh | -0.006 |
rant | -0.006 |
flood | -0.005 |
dolphin | -0.005 |
mate | -0.005 |
Token | of |
Token ID | 286 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
ons | +0.200 |
bes | +0.158 |
ages | +0.151 |
con | +0.141 |
on | +0.140 |
Neg logit contributions | |
ksh | -0.145 |
rant | -0.144 |
flood | -0.142 |
dolphin | -0.138 |
whale | -0.136 |
Token | cr |
Token ID | 1067 |
Feature activation | -3.50e+00 |
Pos logit contributions | |
ons | +0.118 |
ages | +0.099 |
bes | +0.099 |
seek | +0.093 |
es | +0.089 |
Neg logit contributions | |
flood | -0.103 |
dolphin | -0.103 |
whale | -0.100 |
coming | -0.097 |
rant | -0.097 |
Token | ay |
Token ID | 323 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
ons | +0.273 |
bes | +0.236 |
ages | +0.226 |
con | +0.218 |
seek | +0.209 |
Neg logit contributions | |
rant | -0.161 |
ksh | -0.160 |
iet | -0.153 |
mate | -0.143 |
coming | -0.143 |
Token | ons |
Token ID | 684 |
Feature activation | -1.44e+01 |
Pos logit contributions | |
seek | +0.022 |
bes | +0.019 |
powers | +0.012 |
ages | +0.010 |
tail | +0.008 |
Neg logit contributions | |
coming | -0.172 |
rant | -0.161 |
whale | -0.161 |
iet | -0.160 |
them | -0.160 |
Token | . |
Token ID | 13 |
Feature activation | +1.00e-02 |
Pos logit contributions | |
ons | +1.194 |
bes | +0.993 |
ages | +0.924 |
con | +0.885 |
aid | +0.877 |
Neg logit contributions | |
rant | -0.664 |
ksh | -0.661 |
coming | -0.642 |
dolphin | -0.633 |
flood | -0.630 |
Token | She |
Token ID | 1375 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
rant | +0.001 |
ksh | +0.001 |
dolphin | +0.001 |
whale | +0.001 |
flood | +0.001 |
Neg logit contributions | |
ons | -0.001 |
bes | -0.001 |
ages | -0.000 |
on | -0.000 |
seek | -0.000 |
Token | was |
Token ID | 373 |
Feature activation | -3.28e-01 |
Pos logit contributions | |
ons | +0.173 |
bes | +0.147 |
ages | +0.142 |
seek | +0.138 |
con | +0.136 |
Neg logit contributions | |
dolphin | -0.085 |
whale | -0.085 |
rant | -0.082 |
flood | -0.080 |
ksh | -0.080 |
Token | so |
Token ID | 523 |
Feature activation | -9.21e-01 |
Pos logit contributions | |
ons | +0.027 |
bes | +0.022 |
ages | +0.020 |
con | +0.020 |
seek | +0.019 |
Neg logit contributions | |
flood | -0.016 |
ksh | -0.016 |
coming | -0.016 |
rant | -0.016 |
dolphin | -0.016 |
Token | excited |
Token ID | 6568 |
Feature activation | -2.37e+00 |
Pos logit contributions | |
ons | +0.074 |
bes | +0.061 |
ages | +0.056 |
con | +0.056 |
seek | +0.056 |
Neg logit contributions | |
flood | -0.046 |
ksh | -0.045 |
rant | -0.045 |
dolphin | -0.045 |
whale | -0.044 |
Token | bar |
Token ID | 2318 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
ons | +0.027 |
bes | +0.021 |
ages | +0.020 |
con | +0.019 |
on | +0.019 |
Neg logit contributions | |
ksh | -0.020 |
rant | -0.019 |
flood | -0.019 |
dolphin | -0.018 |
ending | -0.018 |
Token | ks |
Token ID | 591 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
ons | +0.106 |
bes | +0.077 |
ages | +0.076 |
es | +0.072 |
on | +0.071 |
Neg logit contributions | |
ksh | -0.106 |
flood | -0.102 |
rant | -0.101 |
dolphin | -0.098 |
ending | -0.096 |
Token | . |
Token ID | 13 |
Feature activation | +7.47e-02 |
Pos logit contributions | |
ons | +0.118 |
bes | +0.088 |
es | +0.086 |
ages | +0.084 |
on | +0.081 |
Neg logit contributions | |
ksh | -0.115 |
rant | -0.113 |
flood | -0.112 |
whale | -0.105 |
dolphin | -0.104 |
Token | He |
Token ID | 679 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
ksh | +0.005 |
rant | +0.005 |
flood | +0.005 |
dolphin | +0.005 |
whale | +0.005 |
Neg logit contributions | |
ons | -0.005 |
bes | -0.003 |
ages | -0.003 |
on | -0.003 |
con | -0.003 |
Token | scares |
Token ID | 44462 |
Feature activation | +5.20e-01 |
Pos logit contributions | |
ons | +0.119 |
bes | +0.092 |
ages | +0.090 |
on | +0.084 |
aid | +0.082 |
Neg logit contributions | |
ksh | -0.087 |
flood | -0.086 |
rant | -0.085 |
dolphin | -0.084 |
whale | -0.083 |
Token | them |
Token ID | 606 |
Feature activation | +8.53e+00 |
Pos logit contributions | |
ksh | +0.032 |
rant | +0.030 |
ending | +0.029 |
flood | +0.028 |
mate | +0.028 |
Neg logit contributions | |
ons | -0.039 |
on | -0.027 |
bes | -0.027 |
es | -0.027 |
ages | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | +8.20e-02 |
Pos logit contributions | |
coming | +0.377 |
flood | +0.373 |
dolphin | +0.352 |
wake | +0.347 |
mate | +0.346 |
Neg logit contributions | |
ons | -0.744 |
bes | -0.608 |
ages | -0.577 |
con | -0.563 |
rob | -0.553 |
Token |
|
Token ID | 198 |
Feature activation | -3.18e-01 |
Pos logit contributions | |
ksh | +0.005 |
flood | +0.005 |
rant | +0.005 |
dolphin | +0.005 |
whale | +0.005 |
Neg logit contributions | |
ons | -0.006 |
bes | -0.004 |
ages | -0.004 |
on | -0.004 |
con | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -3.17e-01 |
Pos logit contributions | |
ons | +0.025 |
bes | +0.021 |
ages | +0.020 |
aces | +0.019 |
con | +0.019 |
Neg logit contributions | |
flood | -0.015 |
dolphin | -0.015 |
rant | -0.015 |
whale | -0.015 |
coming | -0.014 |
Token | " |
Token ID | 1 |
Feature activation | -2.79e+00 |
Pos logit contributions | |
ons | +0.023 |
bes | +0.018 |
ages | +0.018 |
aces | +0.017 |
con | +0.017 |
Neg logit contributions | |
Why | -0.018 |
flood | -0.017 |
dolphin | -0.017 |
whale | -0.017 |
rant | -0.017 |
Token | Go |
Token ID | 5247 |
Feature activation | +3.62e+00 |
Pos logit contributions | |
ons | +0.219 |
aid | +0.176 |
bes | +0.173 |
ages | +0.160 |
on | +0.158 |
Neg logit contributions | |
rant | -0.155 |
ksh | -0.143 |
ending | -0.140 |
harbor | -0.136 |
mate | -0.136 |
Token | coin |
Token ID | 10752 |
Feature activation | -3.09e+00 |
Pos logit contributions | |
ons | +0.050 |
bes | +0.038 |
ages | +0.036 |
es | +0.034 |
seek | +0.034 |
Neg logit contributions | |
flood | -0.049 |
whale | -0.048 |
dolphin | -0.048 |
mate | -0.047 |
ending | -0.046 |
Token | . |
Token ID | 13 |
Feature activation | +6.09e-01 |
Pos logit contributions | |
ons | +0.229 |
bes | +0.179 |
ages | +0.170 |
aid | +0.157 |
con | +0.157 |
Neg logit contributions | |
rant | -0.174 |
ksh | -0.174 |
flood | -0.174 |
dolphin | -0.169 |
whale | -0.168 |
Token |
|
Token ID | 198 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
ksh | +0.032 |
flood | +0.031 |
rant | +0.031 |
dolphin | +0.031 |
whale | +0.030 |
Neg logit contributions | |
ons | -0.048 |
bes | -0.039 |
ages | -0.037 |
seek | -0.034 |
aid | -0.034 |
Token |
|
Token ID | 198 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
dolphin | +0.010 |
flood | +0.010 |
whale | +0.009 |
rant | +0.009 |
iet | +0.009 |
Neg logit contributions | |
ons | -0.016 |
bes | -0.014 |
ages | -0.013 |
aces | -0.012 |
seek | -0.012 |
Token | " |
Token ID | 1 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
flood | +0.008 |
ksh | +0.008 |
rant | +0.008 |
dolphin | +0.008 |
iet | +0.008 |
Neg logit contributions | |
ons | -0.019 |
bes | -0.016 |
ages | -0.016 |
con | -0.015 |
seek | -0.015 |
Token | Hey |
Token ID | 10814 |
Feature activation | +7.29e+00 |
Pos logit contributions | |
ons | +0.165 |
aid | +0.133 |
bes | +0.132 |
on | +0.125 |
rob | +0.119 |
Neg logit contributions | |
rant | -0.129 |
ksh | -0.120 |
ashing | -0.119 |
flood | -0.118 |
appearance | -0.117 |
Token | , |
Token ID | 11 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
ksh | +0.674 |
rant | +0.663 |
iet | +0.591 |
flood | +0.586 |
extingu | +0.582 |
Neg logit contributions | |
ons | -0.339 |
on | -0.213 |
es | -0.188 |
ages | -0.172 |
con | -0.165 |
Token | that |
Token ID | 326 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
ksh | +0.013 |
rant | +0.012 |
ashing | +0.010 |
iet | +0.010 |
horizon | +0.010 |
Neg logit contributions | |
ons | -0.012 |
on | -0.009 |
aid | -0.008 |
con | -0.008 |
es | -0.008 |
Token | 's |
Token ID | 338 |
Feature activation | +3.66e+00 |
Pos logit contributions | |
ksh | +0.065 |
flood | +0.064 |
rant | +0.063 |
dolphin | +0.062 |
whale | +0.061 |
Neg logit contributions | |
ons | -0.091 |
bes | -0.072 |
ages | -0.069 |
on | -0.064 |
con | -0.064 |
Token | our |
Token ID | 674 |
Feature activation | +3.04e+00 |
Pos logit contributions | |
rant | +0.279 |
ksh | +0.275 |
dolphin | +0.266 |
extingu | +0.264 |
flood | +0.259 |
Neg logit contributions | |
ons | -0.199 |
bes | -0.133 |
ages | -0.126 |
on | -0.125 |
myself | -0.114 |
Token | coin |
Token ID | 10752 |
Feature activation | -3.03e+00 |
Pos logit contributions | |
ksh | +0.214 |
rant | +0.211 |
flood | +0.209 |
dolphin | +0.204 |
whale | +0.201 |
Neg logit contributions | |
ons | -0.188 |
bes | -0.138 |
ages | -0.130 |
on | -0.119 |
con | -0.117 |
Token | and |
Token ID | 290 |
Feature activation | -4.00e-01 |
Pos logit contributions | |
ons | +0.184 |
bes | +0.115 |
ages | +0.112 |
on | +0.111 |
es | +0.107 |
Neg logit contributions | |
ksh | -0.266 |
rant | -0.264 |
flood | -0.256 |
iet | -0.246 |
extingu | -0.235 |
Token | encouraged |
Token ID | 10085 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
ons | +0.034 |
bes | +0.027 |
orns | +0.025 |
con | +0.025 |
ops | +0.025 |
Neg logit contributions | |
dolphin | -0.019 |
flood | -0.019 |
coming | -0.019 |
ending | -0.019 |
wake | -0.019 |
Token | him |
Token ID | 683 |
Feature activation | +2.39e+00 |
Pos logit contributions | |
ons | +0.099 |
bes | +0.081 |
ages | +0.077 |
on | +0.072 |
aid | +0.071 |
Neg logit contributions | |
ksh | -0.059 |
rant | -0.057 |
dolphin | -0.056 |
flood | -0.056 |
coming | -0.054 |
Token | to |
Token ID | 284 |
Feature activation | -7.98e-01 |
Pos logit contributions | |
coming | +0.104 |
mate | +0.104 |
wake | +0.100 |
dolphin | +0.099 |
ending | +0.098 |
Neg logit contributions | |
ons | -0.209 |
bes | -0.168 |
rob | -0.156 |
ages | -0.154 |
aid | -0.153 |
Token | never |
Token ID | 1239 |
Feature activation | -9.84e-01 |
Pos logit contributions | |
ons | +0.074 |
bes | +0.061 |
ages | +0.059 |
con | +0.056 |
on | +0.055 |
Neg logit contributions | |
flood | -0.031 |
rant | -0.030 |
ksh | -0.030 |
dolphin | -0.030 |
whale | -0.029 |
Token | stop |
Token ID | 2245 |
Feature activation | +6.45e+00 |
Pos logit contributions | |
ons | +0.071 |
seek | +0.056 |
tail | +0.053 |
ages | +0.052 |
on | +0.052 |
Neg logit contributions | |
ksh | -0.059 |
rant | -0.055 |
iet | -0.051 |
dolphin | -0.049 |
extingu | -0.049 |
Token | being |
Token ID | 852 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
ksh | +0.404 |
rant | +0.392 |
flood | +0.381 |
iet | +0.369 |
dolphin | +0.358 |
Neg logit contributions | |
ons | -0.453 |
bes | -0.340 |
ages | -0.322 |
on | -0.313 |
es | -0.307 |
Token | creative |
Token ID | 7325 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
ksh | +0.071 |
rant | +0.070 |
flood | +0.069 |
dolphin | +0.066 |
whale | +0.065 |
Neg logit contributions | |
ons | -0.114 |
bes | -0.090 |
ages | -0.087 |
on | -0.082 |
es | -0.081 |
Token | . |
Token ID | 13 |
Feature activation | +2.66e-02 |
Pos logit contributions | |
ons | +0.096 |
bes | +0.075 |
ages | +0.073 |
es | +0.069 |
on | +0.068 |
Neg logit contributions | |
ksh | -0.068 |
flood | -0.067 |
rant | -0.067 |
dolphin | -0.064 |
whale | -0.063 |
Token |
|
Token ID | 198 |
Feature activation | -3.87e-01 |
Pos logit contributions | |
ksh | +0.001 |
dolphin | +0.001 |
whale | +0.001 |
flood | +0.001 |
mate | +0.001 |
Neg logit contributions | |
ons | -0.002 |
bes | -0.002 |
ages | -0.002 |
aches | -0.002 |
aid | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | -3.73e-01 |
Pos logit contributions | |
ons | +0.032 |
ages | +0.027 |
bes | +0.027 |
aces | +0.026 |
seek | +0.025 |
Neg logit contributions | |
dolphin | -0.016 |
iet | -0.015 |
whale | -0.015 |
flood | -0.014 |
coming | -0.014 |
Token | She |
Token ID | 1375 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
ksh | +0.003 |
rant | +0.003 |
flood | +0.003 |
dolphin | +0.003 |
whale | +0.003 |
Neg logit contributions | |
ons | -0.004 |
bes | -0.003 |
ages | -0.003 |
on | -0.003 |
con | -0.003 |
Token | decided |
Token ID | 3066 |
Feature activation | -5.45e-01 |
Pos logit contributions | |
ons | +0.174 |
bes | +0.140 |
ages | +0.134 |
on | +0.129 |
es | +0.127 |
Neg logit contributions | |
ksh | -0.094 |
rant | -0.091 |
flood | -0.090 |
dolphin | -0.085 |
whale | -0.084 |
Token | to |
Token ID | 284 |
Feature activation | -7.45e-01 |
Pos logit contributions | |
ons | +0.040 |
bes | +0.031 |
ages | +0.030 |
on | +0.028 |
con | +0.027 |
Neg logit contributions | |
ksh | -0.032 |
rant | -0.031 |
flood | -0.031 |
dolphin | -0.030 |
whale | -0.030 |
Token | pull |
Token ID | 2834 |
Feature activation | -7.00e-02 |
Pos logit contributions | |
ons | +0.067 |
bes | +0.054 |
ages | +0.053 |
con | +0.050 |
on | +0.050 |
Neg logit contributions | |
flood | -0.031 |
ksh | -0.030 |
rant | -0.030 |
dolphin | -0.029 |
whale | -0.029 |
Token | it |
Token ID | 340 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
ons | +0.005 |
bes | +0.004 |
ages | +0.004 |
seek | +0.004 |
con | +0.004 |
Neg logit contributions | |
rant | -0.004 |
ksh | -0.004 |
flood | -0.004 |
dolphin | -0.004 |
whale | -0.004 |
Token | out |
Token ID | 503 |
Feature activation | +6.51e+00 |
Pos logit contributions | |
ons | +0.096 |
bes | +0.080 |
ages | +0.077 |
on | +0.072 |
con | +0.072 |
Neg logit contributions | |
flood | -0.038 |
rant | -0.038 |
ksh | -0.038 |
coming | -0.035 |
dolphin | -0.035 |
Token | of |
Token ID | 286 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
coming | +0.327 |
rant | +0.325 |
flood | +0.315 |
dolphin | +0.306 |
ksh | +0.302 |
Neg logit contributions | |
ons | -0.508 |
bes | -0.426 |
ages | -0.415 |
seek | -0.381 |
rob | -0.379 |
Token | the |
Token ID | 262 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
ons | +0.119 |
bes | +0.091 |
ages | +0.090 |
es | +0.081 |
con | +0.081 |
Neg logit contributions | |
flood | -0.101 |
rant | -0.100 |
ksh | -0.099 |
dolphin | -0.097 |
whale | -0.095 |
Token | grass |
Token ID | 8701 |
Feature activation | -2.59e+00 |
Pos logit contributions | |
ons | +0.097 |
bes | +0.076 |
ages | +0.072 |
on | +0.068 |
con | +0.068 |
Neg logit contributions | |
ksh | -0.070 |
rant | -0.068 |
flood | -0.068 |
dolphin | -0.066 |
whale | -0.064 |
Token | and |
Token ID | 290 |
Feature activation | -3.34e-01 |
Pos logit contributions | |
ons | +0.194 |
bes | +0.150 |
ages | +0.144 |
on | +0.137 |
es | +0.136 |
Neg logit contributions | |
ksh | -0.148 |
flood | -0.143 |
rant | -0.143 |
dolphin | -0.139 |
whale | -0.136 |
Token | it |
Token ID | 340 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
ons | +0.029 |
bes | +0.023 |
ages | +0.023 |
on | +0.021 |
seek | +0.021 |
Neg logit contributions | |
flood | -0.015 |
rant | -0.014 |
dolphin | -0.014 |
ksh | -0.014 |
coming | -0.014 |
Token | ! |
Token ID | 0 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
ons | +0.152 |
bes | +0.121 |
ages | +0.114 |
con | +0.110 |
on | +0.107 |
Neg logit contributions | |
ksh | -0.097 |
rant | -0.095 |
flood | -0.093 |
dolphin | -0.091 |
coming | -0.089 |
Token | Can |
Token ID | 1680 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
ksh | +0.087 |
rant | +0.086 |
flood | +0.085 |
dolphin | +0.082 |
whale | +0.081 |
Neg logit contributions | |
ons | -0.126 |
bes | -0.100 |
ages | -0.096 |
on | -0.090 |
con | -0.089 |
Token | 't |
Token ID | 470 |
Feature activation | -4.55e+00 |
Pos logit contributions | |
flood | +0.069 |
ksh | +0.069 |
dolphin | +0.068 |
mate | +0.067 |
Why | +0.067 |
Neg logit contributions | |
ons | -0.150 |
ages | -0.126 |
bes | -0.123 |
con | -0.116 |
seek | -0.114 |
Token | we |
Token ID | 356 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
ons | +0.333 |
bes | +0.253 |
ages | +0.238 |
on | +0.236 |
es | +0.222 |
Neg logit contributions | |
ksh | -0.275 |
rant | -0.270 |
flood | -0.256 |
dolphin | -0.248 |
whale | -0.248 |
Token | keep |
Token ID | 1394 |
Feature activation | +1.88e-01 |
Pos logit contributions | |
ksh | +0.088 |
rant | +0.083 |
flood | +0.081 |
dolphin | +0.079 |
whale | +0.079 |
Neg logit contributions | |
ons | -0.142 |
bes | -0.112 |
ages | -0.107 |
on | -0.105 |
aid | -0.102 |
Token | them |
Token ID | 606 |
Feature activation | +8.51e+00 |
Pos logit contributions | |
ksh | +0.006 |
rant | +0.005 |
flood | +0.005 |
dolphin | +0.005 |
whale | +0.005 |
Neg logit contributions | |
ons | -0.019 |
bes | -0.016 |
ages | -0.016 |
on | -0.015 |
con | -0.015 |
Token | all |
Token ID | 477 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
ksh | +0.439 |
flood | +0.434 |
rant | +0.432 |
dolphin | +0.418 |
whale | +0.407 |
Neg logit contributions | |
ons | -0.678 |
bes | -0.539 |
ages | -0.518 |
con | -0.488 |
on | -0.485 |
Token | ?" |
Token ID | 1701 |
Feature activation | -3.02e+00 |
Pos logit contributions | |
ons | +0.081 |
bes | +0.065 |
ages | +0.063 |
es | +0.058 |
con | +0.058 |
Neg logit contributions | |
flood | -0.056 |
rant | -0.056 |
ksh | -0.055 |
coming | -0.053 |
dolphin | -0.053 |
Token |
|
Token ID | 198 |
Feature activation | -3.50e-01 |
Pos logit contributions | |
ons | +0.231 |
bes | +0.187 |
ages | +0.185 |
seek | +0.168 |
igs | +0.166 |
Neg logit contributions | |
ksh | -0.154 |
dolphin | -0.151 |
Why | -0.147 |
coming | -0.145 |
whale | -0.144 |
Token |
|
Token ID | 198 |
Feature activation | -3.56e-01 |
Pos logit contributions | |
ons | +0.028 |
ages | +0.023 |
bes | +0.023 |
aces | +0.021 |
seek | +0.021 |
Neg logit contributions | |
dolphin | -0.016 |
ksh | -0.015 |
iet | -0.015 |
whale | -0.015 |
Why | -0.015 |
Token | Her |
Token ID | 9360 |
Feature activation | -2.87e+00 |
Pos logit contributions | |
ons | +0.028 |
ages | +0.022 |
con | +0.022 |
bes | +0.022 |
aces | +0.021 |
Neg logit contributions | |
dolphin | -0.018 |
iet | -0.017 |
ksh | -0.017 |
Why | -0.017 |
mate | -0.016 |
Token | to |
Token ID | 284 |
Feature activation | -2.84e-01 |
Pos logit contributions | |
coming | +0.048 |
flood | +0.047 |
dolphin | +0.047 |
whale | +0.046 |
rant | +0.045 |
Neg logit contributions | |
ons | -0.100 |
bes | -0.085 |
ages | -0.083 |
aid | -0.076 |
aces | -0.075 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
ons | +0.021 |
bes | +0.016 |
on | +0.015 |
ages | +0.015 |
es | +0.014 |
Neg logit contributions | |
ksh | -0.017 |
rant | -0.016 |
flood | -0.016 |
whale | -0.015 |
dolphin | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
coming | +0.040 |
ksh | +0.037 |
wake | +0.036 |
dolphin | +0.036 |
rant | +0.036 |
Neg logit contributions | |
ons | -0.093 |
bes | -0.077 |
ages | -0.075 |
aid | -0.074 |
rob | -0.072 |
Token | asked |
Token ID | 1965 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
dolphin | +0.007 |
flood | +0.007 |
ksh | +0.007 |
whale | +0.006 |
rant | +0.006 |
Neg logit contributions | |
ons | -0.008 |
bes | -0.006 |
ages | -0.006 |
aces | -0.006 |
on | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | +6.48e-02 |
Pos logit contributions | |
ons | +0.122 |
bes | +0.095 |
ages | +0.091 |
on | +0.084 |
con | +0.083 |
Neg logit contributions | |
ksh | -0.099 |
rant | -0.098 |
flood | -0.098 |
dolphin | -0.096 |
whale | -0.094 |
Token | " |
Token ID | 366 |
Feature activation | +3.04e+00 |
Pos logit contributions | |
rant | +0.004 |
ksh | +0.004 |
iet | +0.004 |
flood | +0.004 |
appearance | +0.004 |
Neg logit contributions | |
ons | -0.004 |
on | -0.003 |
bes | -0.003 |
ner | -0.003 |
es | -0.003 |
Token | What |
Token ID | 2061 |
Feature activation | +7.20e+00 |
Pos logit contributions | |
flood | +0.151 |
ksh | +0.149 |
sing | +0.148 |
rant | +0.146 |
ending | +0.142 |
Neg logit contributions | |
ons | -0.254 |
on | -0.198 |
bes | -0.192 |
aid | -0.189 |
rob | -0.182 |
Token | 's |
Token ID | 338 |
Feature activation | +3.58e+00 |
Pos logit contributions | |
ksh | +0.358 |
flood | +0.354 |
rant | +0.349 |
dolphin | +0.348 |
whale | +0.342 |
Neg logit contributions | |
ons | -0.574 |
bes | -0.469 |
ages | -0.444 |
con | -0.420 |
aid | -0.411 |
Token | wrong |
Token ID | 2642 |
Feature activation | +3.79e-01 |
Pos logit contributions | |
ksh | +0.227 |
rant | +0.220 |
flood | +0.215 |
dolphin | +0.208 |
whale | +0.207 |
Neg logit contributions | |
ons | -0.253 |
bes | -0.191 |
ages | -0.181 |
on | -0.173 |
con | -0.169 |
Token | , |
Token ID | 11 |
Feature activation | +9.50e-02 |
Pos logit contributions | |
ksh | +0.031 |
rant | +0.031 |
flood | +0.027 |
beginning | +0.027 |
whale | +0.026 |
Neg logit contributions | |
ons | -0.019 |
bes | -0.013 |
ages | -0.013 |
es | -0.012 |
on | -0.011 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
ksh | +0.006 |
rant | +0.006 |
flood | +0.005 |
extingu | +0.005 |
beginning | +0.005 |
Neg logit contributions | |
ons | -0.007 |
bes | -0.005 |
on | -0.005 |
aid | -0.005 |
con | -0.004 |
Token | played |
Token ID | 2826 |
Feature activation | -3.49e+00 |
Pos logit contributions | |
ons | +0.018 |
bes | +0.015 |
ages | +0.014 |
aid | +0.013 |
con | +0.013 |
Neg logit contributions | |
flood | -0.010 |
rant | -0.010 |
ksh | -0.009 |
dolphin | -0.009 |
whale | -0.009 |
Token | together |
Token ID | 1978 |
Feature activation | -3.95e+00 |
Pos logit contributions | |
ons | +0.250 |
ages | +0.185 |
bes | +0.184 |
con | +0.171 |
aid | +0.170 |
Neg logit contributions | |
flood | -0.208 |
ksh | -0.203 |
rant | -0.201 |
dolphin | -0.200 |
coming | -0.199 |
Token | all |
Token ID | 477 |
Feature activation | -8.83e-01 |
Pos logit contributions | |
ons | +0.291 |
bes | +0.226 |
ages | +0.215 |
on | +0.203 |
es | +0.199 |
Neg logit contributions | |
ksh | -0.233 |
rant | -0.228 |
flood | -0.227 |
dolphin | -0.220 |
whale | -0.216 |
Token | day |
Token ID | 1110 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
ons | +0.072 |
bes | +0.057 |
ages | +0.055 |
con | +0.052 |
on | +0.052 |
Neg logit contributions | |
ksh | -0.045 |
rant | -0.045 |
flood | -0.044 |
dolphin | -0.043 |
whale | -0.042 |
Token | long |
Token ID | 890 |
Feature activation | -2.80e+00 |
Pos logit contributions | |
ons | +0.102 |
aid | +0.080 |
bes | +0.079 |
ages | +0.076 |
con | +0.072 |
Neg logit contributions | |
flood | -0.060 |
ksh | -0.060 |
coming | -0.059 |
rant | -0.058 |
dolphin | -0.058 |
Token | . |
Token ID | 13 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
ons | +0.217 |
bes | +0.171 |
ages | +0.165 |
on | +0.154 |
aid | +0.154 |
Neg logit contributions | |
ksh | -0.149 |
rant | -0.148 |
flood | -0.147 |
dolphin | -0.142 |
whale | -0.140 |
Token | And |
Token ID | 843 |
Feature activation | -2.82e+00 |
Pos logit contributions | |
ksh | +0.015 |
flood | +0.015 |
rant | +0.015 |
dolphin | +0.014 |
whale | +0.014 |
Neg logit contributions | |
ons | -0.016 |
bes | -0.012 |
ages | -0.011 |
on | -0.010 |
con | -0.010 |
Token | they |
Token ID | 484 |
Feature activation | +2.91e+00 |
Pos logit contributions | |
ons | +0.232 |
bes | +0.188 |
ages | +0.178 |
on | +0.167 |
con | +0.166 |
Neg logit contributions | |
ksh | -0.139 |
flood | -0.137 |
rant | -0.135 |
dolphin | -0.135 |
whale | -0.131 |
Token | never |
Token ID | 1239 |
Feature activation | -7.39e-01 |
Pos logit contributions | |
flood | +0.142 |
rant | +0.141 |
ksh | +0.138 |
dolphin | +0.136 |
whale | +0.134 |
Neg logit contributions | |
ons | -0.239 |
bes | -0.196 |
ages | -0.188 |
con | -0.178 |
seek | -0.174 |
Token | had |
Token ID | 550 |
Feature activation | +4.28e-01 |
Pos logit contributions | |
ons | +0.054 |
ages | +0.041 |
seek | +0.039 |
bes | +0.039 |
con | +0.038 |
Neg logit contributions | |
ksh | -0.042 |
rant | -0.042 |
dolphin | -0.041 |
iet | -0.040 |
whale | -0.039 |
Token | to |
Token ID | 284 |
Feature activation | -5.48e-01 |
Pos logit contributions | |
ksh | +0.024 |
flood | +0.024 |
rant | +0.024 |
dolphin | +0.023 |
whale | +0.023 |
Neg logit contributions | |
ons | -0.032 |
bes | -0.025 |
ages | -0.024 |
on | -0.023 |
con | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
ons | +0.148 |
bes | +0.119 |
ages | +0.111 |
con | +0.104 |
seek | +0.100 |
Neg logit contributions | |
flood | -0.111 |
mate | -0.110 |
dolphin | -0.110 |
ksh | -0.110 |
rant | -0.110 |
Token | " |
Token ID | 366 |
Feature activation | +3.13e+00 |
Pos logit contributions | |
ksh | +0.008 |
rant | +0.008 |
flood | +0.008 |
iet | +0.008 |
dolphin | +0.007 |
Neg logit contributions | |
ons | -0.012 |
bes | -0.010 |
ages | -0.009 |
on | -0.009 |
es | -0.009 |
Token | Sure |
Token ID | 19457 |
Feature activation | +3.91e-01 |
Pos logit contributions | |
ksh | +0.154 |
rant | +0.147 |
flood | +0.147 |
dolphin | +0.145 |
whale | +0.141 |
Neg logit contributions | |
ons | -0.264 |
bes | -0.211 |
ages | -0.201 |
on | -0.191 |
es | -0.188 |
Token | , |
Token ID | 11 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
ksh | +0.026 |
rant | +0.026 |
flood | +0.026 |
dolphin | +0.025 |
whale | +0.025 |
Neg logit contributions | |
ons | -0.026 |
bes | -0.019 |
ages | -0.018 |
on | -0.017 |
con | -0.016 |
Token | Anna |
Token ID | 11735 |
Feature activation | +6.42e-01 |
Pos logit contributions | |
ksh | +0.012 |
rant | +0.012 |
flood | +0.012 |
iet | +0.011 |
extingu | +0.011 |
Neg logit contributions | |
ons | -0.011 |
bes | -0.007 |
on | -0.007 |
es | -0.007 |
aid | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +6.62e-01 |
Pos logit contributions | |
ksh | +0.031 |
rant | +0.031 |
flood | +0.031 |
dolphin | +0.030 |
whale | +0.030 |
Neg logit contributions | |
ons | -0.052 |
bes | -0.043 |
ages | -0.040 |
con | -0.038 |
aid | -0.038 |
Token | He |
Token ID | 679 |
Feature activation | -9.83e-01 |
Pos logit contributions | |
ksh | +0.035 |
rant | +0.034 |
flood | +0.033 |
dolphin | +0.032 |
whale | +0.032 |
Neg logit contributions | |
ons | -0.053 |
bes | -0.042 |
ages | -0.040 |
on | -0.038 |
con | -0.037 |
Token | likes |
Token ID | 7832 |
Feature activation | +4.97e-02 |
Pos logit contributions | |
ons | +0.078 |
bes | +0.062 |
ages | +0.058 |
on | +0.056 |
seek | +0.055 |
Neg logit contributions | |
ksh | -0.050 |
rant | -0.050 |
flood | -0.049 |
dolphin | -0.048 |
whale | -0.047 |
Token | that |
Token ID | 326 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
dolphin | +0.003 |
whale | +0.002 |
coming | +0.002 |
flood | +0.002 |
ksh | +0.002 |
Neg logit contributions | |
ons | -0.004 |
bes | -0.003 |
ages | -0.003 |
on | -0.003 |
es | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +6.96e-01 |
Pos logit contributions | |
flood | +0.062 |
dolphin | +0.061 |
ksh | +0.061 |
whale | +0.059 |
ending | +0.058 |
Neg logit contributions | |
ons | -0.101 |
bes | -0.080 |
ages | -0.077 |
aid | -0.073 |
on | -0.072 |
Token | His |
Token ID | 2399 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
flood | +0.044 |
ksh | +0.044 |
rant | +0.044 |
dolphin | +0.042 |
whale | +0.042 |
Neg logit contributions | |
ons | -0.047 |
bes | -0.035 |
ages | -0.034 |
on | -0.032 |
con | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | +5.24e-01 |
Pos logit contributions | |
ksh | +0.124 |
rant | +0.124 |
flood | +0.123 |
dolphin | +0.120 |
whale | +0.118 |
Neg logit contributions | |
ons | -0.184 |
bes | -0.146 |
ages | -0.141 |
on | -0.131 |
aid | -0.130 |
Token | They |
Token ID | 1119 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
ksh | +0.030 |
rant | +0.029 |
flood | +0.029 |
dolphin | +0.029 |
whale | +0.028 |
Neg logit contributions | |
ons | -0.039 |
bes | -0.030 |
ages | -0.029 |
seek | -0.027 |
aid | -0.027 |
Token | rushed |
Token ID | 13999 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
flood | +0.086 |
rant | +0.086 |
dolphin | +0.085 |
whale | +0.085 |
mate | +0.084 |
Neg logit contributions | |
ons | -0.148 |
bes | -0.124 |
ages | -0.117 |
con | -0.110 |
seek | -0.109 |
Token | off |
Token ID | 572 |
Feature activation | -3.82e+00 |
Pos logit contributions | |
ons | +0.107 |
bes | +0.084 |
ages | +0.080 |
on | +0.077 |
es | +0.076 |
Neg logit contributions | |
ksh | -0.073 |
flood | -0.070 |
rant | -0.070 |
dolphin | -0.067 |
whale | -0.067 |
Token | , |
Token ID | 11 |
Feature activation | +6.48e-02 |
Pos logit contributions | |
ons | +0.326 |
bes | +0.270 |
ages | +0.268 |
aid | +0.252 |
con | +0.240 |
Neg logit contributions | |
coming | -0.176 |
flood | -0.173 |
dolphin | -0.164 |
rant | -0.163 |
mer | -0.161 |
Token | eager |
Token ID | 11069 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
ksh | +0.004 |
rant | +0.003 |
flood | +0.003 |
dolphin | +0.003 |
whale | +0.003 |
Neg logit contributions | |
ons | -0.005 |
bes | -0.004 |
ages | -0.004 |
on | -0.004 |
con | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
ksh | +0.015 |
rant | +0.014 |
flood | +0.014 |
iet | +0.014 |
dolphin | +0.013 |
Neg logit contributions | |
ons | -0.013 |
ages | -0.009 |
bes | -0.009 |
seek | -0.008 |
con | -0.008 |
Token | find |
Token ID | 1064 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
ons | +0.023 |
bes | +0.019 |
ages | +0.018 |
con | +0.017 |
aid | +0.017 |
Neg logit contributions | |
flood | -0.011 |
rant | -0.010 |
dolphin | -0.010 |
ksh | -0.010 |
whale | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | -7.68e-01 |
Pos logit contributions | |
ksh | +0.061 |
flood | +0.056 |
rant | +0.056 |
ending | +0.052 |
dolphin | +0.052 |
Neg logit contributions | |
ons | -0.077 |
bes | -0.057 |
ages | -0.055 |
es | -0.054 |
on | -0.053 |
Token | treasure |
Token ID | 14068 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
ons | +0.052 |
bes | +0.040 |
con | +0.038 |
ages | +0.037 |
puzzles | +0.036 |
Neg logit contributions | |
ksh | -0.050 |
rant | -0.046 |
flood | -0.043 |
iet | -0.042 |
extingu | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | +5.80e-01 |
Pos logit contributions | |
rant | +0.126 |
flood | +0.126 |
ksh | +0.125 |
dolphin | +0.122 |
whale | +0.120 |
Neg logit contributions | |
ons | -0.189 |
bes | -0.150 |
ages | -0.145 |
aid | -0.135 |
on | -0.134 |
Token | look |
Token ID | 804 |
Feature activation | +4.23e+00 |
Pos logit contributions | |
ksh | +0.011 |
rant | +0.011 |
flood | +0.011 |
dolphin | +0.011 |
whale | +0.010 |
Neg logit contributions | |
ons | -0.020 |
bes | -0.016 |
ages | -0.015 |
on | -0.015 |
con | -0.015 |
Token | for |
Token ID | 329 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
ksh | +0.206 |
flood | +0.206 |
rant | +0.198 |
dolphin | +0.198 |
whale | +0.196 |
Neg logit contributions | |
ons | -0.342 |
bes | -0.267 |
ages | -0.259 |
on | -0.250 |
es | -0.246 |
Token | you |
Token ID | 345 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
ksh | +0.116 |
rant | +0.114 |
flood | +0.114 |
dolphin | +0.112 |
whale | +0.110 |
Neg logit contributions | |
ons | -0.164 |
bes | -0.131 |
ages | -0.127 |
on | -0.116 |
con | -0.116 |
Token | ." |
Token ID | 526 |
Feature activation | -4.06e+00 |
Pos logit contributions | |
ons | +0.154 |
bes | +0.117 |
ages | +0.112 |
con | +0.110 |
on | +0.108 |
Neg logit contributions | |
ksh | -0.108 |
rant | -0.108 |
mate | -0.104 |
coming | -0.101 |
flood | -0.100 |
Token | |
Token ID | 220 |
Feature activation | -3.72e+00 |
Pos logit contributions | |
ons | +0.344 |
bes | +0.289 |
ages | +0.283 |
seek | +0.267 |
aid | +0.266 |
Neg logit contributions | |
ksh | -0.174 |
flood | -0.165 |
rant | -0.158 |
coming | -0.157 |
extingu | -0.150 |
Token |
|
Token ID | 198 |
Feature activation | +8.45e-02 |
Pos logit contributions | |
ons | +0.281 |
bes | +0.228 |
ages | +0.217 |
seek | +0.207 |
con | +0.204 |
Neg logit contributions | |
ksh | -0.194 |
rant | -0.185 |
flood | -0.181 |
coming | -0.179 |
mate | -0.173 |
Token |
|
Token ID | 198 |
Feature activation | +8.91e-02 |
Pos logit contributions | |
ksh | +0.004 |
rant | +0.004 |
coming | +0.004 |
Why | +0.003 |
flood | +0.003 |
Neg logit contributions | |
ons | -0.007 |
bes | -0.006 |
ages | -0.006 |
aces | -0.005 |
tail | -0.005 |
Token | Tim |
Token ID | 14967 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
ksh | +0.005 |
flood | +0.005 |
rant | +0.005 |
dolphin | +0.005 |
whale | +0.005 |
Neg logit contributions | |
ons | -0.007 |
bes | -0.005 |
ages | -0.005 |
con | -0.005 |
on | -0.005 |
Token | my |
Token ID | 1820 |
Feature activation | -3.90e-01 |
Pos logit contributions | |
ons | +0.020 |
bes | +0.018 |
ages | +0.017 |
con | +0.017 |
aid | +0.016 |
Neg logit contributions | |
rant | -0.006 |
ksh | -0.005 |
iet | -0.005 |
y | -0.005 |
wake | -0.005 |
Token | ran |
Token ID | 4966 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
ons | +0.036 |
bes | +0.028 |
aid | +0.028 |
con | +0.027 |
rob | +0.027 |
Neg logit contributions | |
iet | -0.014 |
dolphin | -0.014 |
ksh | -0.014 |
wake | -0.014 |
mate | -0.014 |
Token | to |
Token ID | 284 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
ons | +0.161 |
bes | +0.128 |
ages | +0.125 |
on | +0.119 |
aid | +0.116 |
Neg logit contributions | |
rant | -0.076 |
flood | -0.075 |
ksh | -0.073 |
coming | -0.072 |
mate | -0.070 |
Token | again |
Token ID | 757 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
ons | +0.232 |
bes | +0.185 |
ages | +0.178 |
on | +0.165 |
con | +0.163 |
Neg logit contributions | |
rant | -0.156 |
ksh | -0.155 |
flood | -0.154 |
dolphin | -0.151 |
coming | -0.146 |
Token | , |
Token ID | 11 |
Feature activation | -4.02e-01 |
Pos logit contributions | |
flood | +0.055 |
rant | +0.055 |
ksh | +0.054 |
dolphin | +0.054 |
coming | +0.053 |
Neg logit contributions | |
ons | -0.078 |
bes | -0.061 |
ages | -0.058 |
aid | -0.055 |
on | -0.055 |
Token | you |
Token ID | 345 |
Feature activation | -2.37e+00 |
Pos logit contributions | |
ons | +0.028 |
bes | +0.022 |
ages | +0.020 |
on | +0.019 |
seek | +0.019 |
Neg logit contributions | |
flood | -0.025 |
ksh | -0.024 |
dolphin | -0.024 |
rant | -0.024 |
whale | -0.023 |
Token | can |
Token ID | 460 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
ons | +0.230 |
bes | +0.195 |
ages | +0.183 |
con | +0.176 |
seek | +0.176 |
Neg logit contributions | |
flood | -0.083 |
rant | -0.079 |
ksh | -0.077 |
dolphin | -0.077 |
mate | -0.076 |
Token | have |
Token ID | 423 |
Feature activation | -3.68e+00 |
Pos logit contributions | |
flood | +0.066 |
dolphin | +0.063 |
rant | +0.062 |
ksh | +0.061 |
whale | +0.061 |
Neg logit contributions | |
ons | -0.112 |
bes | -0.090 |
ages | -0.087 |
con | -0.081 |
aid | -0.080 |
Token | some |
Token ID | 617 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
ons | +0.298 |
bes | +0.253 |
ages | +0.236 |
es | +0.229 |
on | +0.222 |
Neg logit contributions | |
coming | -0.168 |
ksh | -0.167 |
flood | -0.166 |
dolphin | -0.165 |
whale | -0.160 |
Token | toast |
Token ID | 27805 |
Feature activation | -5.02e+00 |
Pos logit contributions | |
ons | +0.153 |
bes | +0.131 |
ages | +0.118 |
seek | +0.116 |
es | +0.114 |
Neg logit contributions | |
dolphin | -0.100 |
whale | -0.099 |
flood | -0.099 |
coming | -0.098 |
ksh | -0.098 |
Token | , |
Token ID | 11 |
Feature activation | -3.84e-01 |
Pos logit contributions | |
ons | +0.398 |
bes | +0.315 |
ages | +0.301 |
on | +0.284 |
con | +0.282 |
Neg logit contributions | |
ksh | -0.260 |
rant | -0.256 |
flood | -0.253 |
dolphin | -0.246 |
whale | -0.241 |
Token | â |
Token ID | 22940 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
ons | +0.030 |
bes | +0.023 |
ages | +0.022 |
on | +0.021 |
con | +0.021 |
Neg logit contributions | |
ksh | -0.021 |
rant | -0.021 |
flood | -0.020 |
dolphin | -0.020 |
whale | -0.020 |
Token | € |
Token ID | 26391 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
ksh | +0.057 |
rant | +0.057 |
flood | +0.055 |
dolphin | +0.053 |
whale | +0.052 |
Neg logit contributions | |
ons | -0.094 |
bes | -0.075 |
ages | -0.072 |
es | -0.068 |
on | -0.068 |
Token | said |
Token ID | 531 |
Feature activation | -3.04e+00 |
Pos logit contributions | |
ons | +0.083 |
bes | +0.063 |
aid | +0.062 |
ages | +0.059 |
on | +0.058 |
Neg logit contributions | |
ksh | -0.058 |
flood | -0.054 |
dolphin | -0.054 |
whale | -0.052 |
mate | -0.052 |
Token | a |
Token ID | 257 |
Feature activation | -7.87e-01 |
Pos logit contributions | |
ksh | +0.140 |
rant | +0.136 |
flood | +0.135 |
dolphin | +0.134 |
whale | +0.129 |
Neg logit contributions | |
ons | -0.219 |
bes | -0.178 |
ages | -0.172 |
seek | -0.161 |
on | -0.157 |
Token | ball |
Token ID | 2613 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
ons | +0.059 |
bes | +0.046 |
ages | +0.044 |
on | +0.042 |
con | +0.041 |
Neg logit contributions | |
ksh | -0.045 |
rant | -0.044 |
flood | -0.043 |
dolphin | -0.042 |
whale | -0.041 |
Token | for |
Token ID | 329 |
Feature activation | +2.20e+00 |
Pos logit contributions | |
ons | +0.158 |
bes | +0.126 |
ages | +0.120 |
on | +0.112 |
con | +0.112 |
Neg logit contributions | |
ksh | -0.104 |
rant | -0.103 |
flood | -0.103 |
dolphin | -0.099 |
whale | -0.098 |
Token | him |
Token ID | 683 |
Feature activation | +2.85e+00 |
Pos logit contributions | |
coming | +0.100 |
flood | +0.099 |
dolphin | +0.096 |
mate | +0.092 |
whale | +0.090 |
Neg logit contributions | |
ons | -0.180 |
ages | -0.150 |
bes | -0.149 |
es | -0.135 |
con | -0.134 |
Token | . |
Token ID | 13 |
Feature activation | +4.29e-01 |
Pos logit contributions | |
ksh | +0.129 |
flood | +0.128 |
coming | +0.127 |
rant | +0.126 |
mate | +0.125 |
Neg logit contributions | |
ons | -0.237 |
bes | -0.191 |
ages | -0.178 |
aid | -0.173 |
on | -0.172 |
Token | Her |
Token ID | 2332 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
ksh | +0.022 |
flood | +0.021 |
rant | +0.021 |
dolphin | +0.021 |
whale | +0.020 |
Neg logit contributions | |
ons | -0.035 |
bes | -0.027 |
ages | -0.027 |
aid | -0.025 |
on | -0.025 |
Token | dog |
Token ID | 3290 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
ons | +0.092 |
bes | +0.073 |
ages | +0.070 |
on | +0.067 |
es | +0.066 |
Neg logit contributions | |
ksh | -0.048 |
flood | -0.048 |
dolphin | -0.047 |
rant | -0.046 |
whale | -0.046 |
Token | was |
Token ID | 373 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
ksh | +0.011 |
rant | +0.011 |
flood | +0.011 |
dolphin | +0.011 |
whale | +0.010 |
Neg logit contributions | |
ons | -0.014 |
bes | -0.011 |
ages | -0.011 |
on | -0.010 |
es | -0.010 |
Token | very |
Token ID | 845 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
ons | +0.001 |
bes | +0.001 |
ages | +0.001 |
on | +0.001 |
con | +0.001 |
Neg logit contributions | |
ksh | -0.001 |
rant | -0.001 |
flood | -0.001 |
dolphin | -0.001 |
whale | -0.001 |
Token | fast |
Token ID | 3049 |
Feature activation | -5.65e+00 |
Pos logit contributions | |
ons | +0.132 |
bes | +0.105 |
ages | +0.101 |
on | +0.094 |
con | +0.093 |
Neg logit contributions | |
ksh | -0.090 |
rant | -0.088 |
flood | -0.087 |
dolphin | -0.084 |
whale | -0.083 |
Token | and |
Token ID | 290 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
ons | +0.451 |
bes | +0.357 |
ages | +0.342 |
on | +0.323 |
con | +0.318 |
Neg logit contributions | |
ksh | -0.293 |
flood | -0.288 |
rant | -0.288 |
dolphin | -0.279 |
whale | -0.274 |
Token | back |
Token ID | 736 |
Feature activation | +4.89e+00 |
Pos logit contributions | |
ksh | +0.027 |
rant | +0.026 |
flood | +0.026 |
dolphin | +0.025 |
whale | +0.025 |
Neg logit contributions | |
ons | -0.041 |
bes | -0.033 |
ages | -0.031 |
on | -0.029 |
con | -0.029 |
Token | to |
Token ID | 284 |
Feature activation | -7.65e-01 |
Pos logit contributions | |
flood | +0.234 |
rant | +0.228 |
ksh | +0.224 |
coming | +0.223 |
dolphin | +0.219 |
Neg logit contributions | |
ons | -0.407 |
bes | -0.334 |
ages | -0.320 |
on | -0.300 |
aid | -0.297 |
Token | his |
Token ID | 465 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
ons | +0.053 |
bes | +0.040 |
ages | +0.038 |
on | +0.036 |
con | +0.036 |
Neg logit contributions | |
ksh | -0.048 |
rant | -0.047 |
flood | -0.046 |
dolphin | -0.045 |
whale | -0.045 |
Token | dad |
Token ID | 9955 |
Feature activation | +3.99e+00 |
Pos logit contributions | |
ons | +0.151 |
bes | +0.125 |
ages | +0.119 |
con | +0.112 |
on | +0.111 |
Neg logit contributions | |
ksh | -0.088 |
rant | -0.087 |
iet | -0.077 |
flood | -0.077 |
dolphin | -0.076 |
Token | . |
Token ID | 13 |
Feature activation | +5.61e-02 |
Pos logit contributions | |
ksh | +0.225 |
rant | +0.221 |
flood | +0.220 |
dolphin | +0.213 |
whale | +0.210 |
Neg logit contributions | |
ons | -0.301 |
bes | -0.235 |
ages | -0.225 |
on | -0.212 |
con | -0.208 |
Token | He |
Token ID | 679 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
ksh | +0.004 |
flood | +0.003 |
rant | +0.003 |
dolphin | +0.003 |
whale | +0.003 |
Neg logit contributions | |
ons | -0.004 |
bes | -0.003 |
ages | -0.003 |
aid | -0.003 |
on | -0.003 |
Token | started |
Token ID | 2067 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
ons | +0.119 |
bes | +0.093 |
ages | +0.090 |
on | +0.084 |
es | +0.082 |
Neg logit contributions | |
ksh | -0.090 |
rant | -0.088 |
flood | -0.088 |
dolphin | -0.084 |
whale | -0.083 |
Token | to |
Token ID | 284 |
Feature activation | -7.45e-01 |
Pos logit contributions | |
ons | +0.191 |
bes | +0.150 |
ages | +0.143 |
on | +0.135 |
con | +0.134 |
Neg logit contributions | |
ksh | -0.130 |
flood | -0.130 |
rant | -0.127 |
dolphin | -0.125 |
coming | -0.123 |
Token | cry |
Token ID | 3960 |
Feature activation | -2.74e+00 |
Pos logit contributions | |
ons | +0.059 |
bes | +0.047 |
ages | +0.045 |
on | +0.042 |
con | +0.042 |
Neg logit contributions | |
ksh | -0.040 |
rant | -0.039 |
flood | -0.037 |
dolphin | -0.037 |
whale | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | +7.13e-02 |
Pos logit contributions | |
ons | +0.154 |
ages | +0.124 |
es | +0.121 |
bes | +0.118 |
on | +0.106 |
Neg logit contributions | |
rant | -0.192 |
ksh | -0.187 |
flood | -0.185 |
whale | -0.184 |
iet | -0.176 |
Token |
|
Token ID | 198 |
Feature activation | -3.33e-01 |
Pos logit contributions | |
ksh | +0.004 |
flood | +0.004 |
rant | +0.004 |
dolphin | +0.004 |
whale | +0.004 |
Neg logit contributions | |
ons | -0.005 |
bes | -0.004 |
ages | -0.004 |
aid | -0.003 |
es | -0.003 |
Token | Lily |
Token ID | 20037 |
Feature activation | +6.35e-01 |
Pos logit contributions | |
dolphin | +0.032 |
flood | +0.031 |
ksh | +0.030 |
whale | +0.030 |
coming | +0.030 |
Neg logit contributions | |
ons | -0.064 |
bes | -0.054 |
ages | -0.052 |
seek | -0.050 |
con | -0.048 |
Token | and |
Token ID | 290 |
Feature activation | -2.75e-01 |
Pos logit contributions | |
coming | +0.021 |
ksh | +0.019 |
rant | +0.018 |
wake | +0.018 |
extingu | +0.017 |
Neg logit contributions | |
ons | -0.061 |
bes | -0.051 |
con | -0.050 |
seek | -0.049 |
ages | -0.049 |
Token | hugs |
Token ID | 40657 |
Feature activation | -4.60e+00 |
Pos logit contributions | |
ons | +0.018 |
bes | +0.013 |
ages | +0.013 |
on | +0.011 |
seek | +0.011 |
Neg logit contributions | |
ksh | -0.019 |
flood | -0.018 |
rant | -0.018 |
dolphin | -0.018 |
whale | -0.018 |
Token | her |
Token ID | 607 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
ons | +0.298 |
bes | +0.207 |
es | +0.202 |
aid | +0.202 |
on | +0.197 |
Neg logit contributions | |
ksh | -0.324 |
flood | -0.311 |
rant | -0.300 |
whale | -0.291 |
mate | -0.290 |
Token | . |
Token ID | 13 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
ons | +0.127 |
bes | +0.097 |
con | +0.097 |
seek | +0.095 |
ages | +0.094 |
Neg logit contributions | |
ksh | -0.058 |
rant | -0.052 |
dolphin | -0.052 |
coming | -0.052 |
extingu | -0.051 |
Token | She |
Token ID | 1375 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
ksh | +0.011 |
rant | +0.010 |
flood | +0.010 |
dolphin | +0.010 |
whale | +0.010 |
Neg logit contributions | |
ons | -0.011 |
bes | -0.008 |
ages | -0.008 |
on | -0.007 |
con | -0.007 |
Token | is |
Token ID | 318 |
Feature activation | -5.63e-01 |
Pos logit contributions | |
ons | +0.128 |
bes | +0.098 |
ages | +0.096 |
on | +0.089 |
seek | +0.088 |
Neg logit contributions | |
flood | -0.119 |
dolphin | -0.118 |
ksh | -0.118 |
rant | -0.117 |
whale | -0.117 |
Token | bleeding |
Token ID | 16832 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
ons | +0.045 |
bes | +0.036 |
ages | +0.034 |
con | +0.033 |
ops | +0.032 |
Neg logit contributions | |
coming | -0.029 |
flood | -0.029 |
ksh | -0.028 |
rant | -0.028 |
dolphin | -0.028 |
Token | and |
Token ID | 290 |
Feature activation | -2.58e-01 |
Pos logit contributions | |
ons | +0.192 |
bes | +0.155 |
ages | +0.144 |
con | +0.141 |
on | +0.139 |
Neg logit contributions | |
ksh | -0.102 |
rant | -0.100 |
flood | -0.098 |
coming | -0.098 |
dolphin | -0.095 |
Token | hurt |
Token ID | 5938 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
ons | +0.021 |
bes | +0.017 |
ages | +0.017 |
seek | +0.016 |
ops | +0.016 |
Neg logit contributions | |
coming | -0.013 |
flood | -0.013 |
Why | -0.012 |
wake | -0.012 |
dolphin | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
ons | +0.197 |
bes | +0.157 |
ages | +0.149 |
con | +0.141 |
on | +0.140 |
Neg logit contributions | |
ksh | -0.131 |
rant | -0.130 |
flood | -0.129 |
dolphin | -0.124 |
coming | -0.124 |
Token | wanted |
Token ID | 2227 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
ons | +0.013 |
bes | +0.010 |
ages | +0.010 |
on | +0.009 |
con | +0.009 |
Neg logit contributions | |
flood | -0.008 |
dolphin | -0.008 |
coming | -0.008 |
whale | -0.008 |
ksh | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | -5.48e-01 |
Pos logit contributions | |
flood | +0.056 |
coming | +0.056 |
dolphin | +0.056 |
rant | +0.056 |
ksh | +0.055 |
Neg logit contributions | |
ons | -0.103 |
bes | -0.087 |
ages | -0.082 |
es | -0.076 |
on | -0.075 |
Token | make |
Token ID | 787 |
Feature activation | +6.66e-01 |
Pos logit contributions | |
ons | +0.049 |
bes | +0.042 |
ages | +0.041 |
es | +0.040 |
rob | +0.039 |
Neg logit contributions | |
wake | -0.024 |
flood | -0.023 |
stop | -0.021 |
dolphin | -0.019 |
coming | -0.019 |
Token | her |
Token ID | 607 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
flood | +0.034 |
ksh | +0.034 |
dolphin | +0.034 |
coming | +0.033 |
whale | +0.032 |
Neg logit contributions | |
ons | -0.051 |
bes | -0.042 |
ages | -0.039 |
on | -0.037 |
seek | -0.036 |
Token | feel |
Token ID | 1254 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
ons | +0.111 |
bes | +0.087 |
ages | +0.083 |
con | +0.081 |
rob | +0.080 |
Neg logit contributions | |
flood | -0.057 |
wake | -0.055 |
ending | -0.054 |
dolphin | -0.053 |
ksh | -0.052 |
Token | better |
Token ID | 1365 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
ons | +0.151 |
bes | +0.118 |
ages | +0.113 |
con | +0.105 |
on | +0.105 |
Neg logit contributions | |
ksh | -0.114 |
flood | -0.113 |
rant | -0.113 |
dolphin | -0.109 |
whale | -0.107 |
Token | . |
Token ID | 13 |
Feature activation | +2.89e-01 |
Pos logit contributions | |
ons | +0.081 |
bes | +0.063 |
ages | +0.061 |
on | +0.057 |
es | +0.056 |
Neg logit contributions | |
ksh | -0.063 |
rant | -0.062 |
flood | -0.062 |
dolphin | -0.060 |
whale | -0.059 |
Token | So |
Token ID | 1406 |
Feature activation | -6.46e+00 |
Pos logit contributions | |
ksh | +0.013 |
Why | +0.011 |
flood | +0.011 |
ending | +0.010 |
ashing | +0.010 |
Neg logit contributions | |
ons | -0.026 |
bes | -0.020 |
on | -0.020 |
odic | -0.019 |
igs | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
ons | +0.527 |
bes | +0.435 |
ages | +0.398 |
on | +0.380 |
con | +0.372 |
Neg logit contributions | |
ending | -0.336 |
flood | -0.332 |
coming | -0.315 |
mate | -0.311 |
dolphin | -0.309 |
Token | he |
Token ID | 339 |
Feature activation | -2.16e+00 |
Pos logit contributions | |
ons | +0.013 |
bes | +0.011 |
ages | +0.010 |
on | +0.009 |
igs | +0.009 |
Neg logit contributions | |
flood | -0.009 |
Why | -0.009 |
coming | -0.008 |
dolphin | -0.008 |
ending | -0.008 |
Token | brought |
Token ID | 3181 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
ons | +0.182 |
bes | +0.150 |
ages | +0.140 |
seek | +0.135 |
con | +0.132 |
Neg logit contributions | |
ksh | -0.100 |
dolphin | -0.100 |
rant | -0.099 |
flood | -0.097 |
whale | -0.097 |
Token | on |
Token ID | 319 |
Feature activation | -4.24e+00 |
Pos logit contributions | |
ons | +0.253 |
bes | +0.204 |
ages | +0.192 |
on | +0.179 |
con | +0.177 |
Neg logit contributions | |
ksh | -0.167 |
rant | -0.164 |
flood | -0.164 |
coming | -0.160 |
dolphin | -0.153 |
Token | the |
Token ID | 262 |
Feature activation | -7.69e-01 |
Pos logit contributions | |
ons | +0.280 |
bes | +0.208 |
ages | +0.189 |
on | +0.187 |
es | +0.180 |
Neg logit contributions | |
ksh | -0.285 |
rant | -0.275 |
flood | -0.272 |
dolphin | -0.266 |
whale | -0.264 |
Token | floor |
Token ID | 4314 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
ons | +0.065 |
bes | +0.051 |
ages | +0.049 |
aid | +0.047 |
es | +0.046 |
Neg logit contributions | |
ksh | -0.037 |
rant | -0.036 |
flood | -0.036 |
dolphin | -0.035 |
whale | -0.035 |
Token | and |
Token ID | 290 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
ons | +0.117 |
bes | +0.092 |
ages | +0.088 |
on | +0.083 |
con | +0.082 |
Neg logit contributions | |
ksh | -0.084 |
rant | -0.082 |
flood | -0.082 |
dolphin | -0.079 |
whale | -0.078 |
Token | the |
Token ID | 262 |
Feature activation | -7.44e-01 |
Pos logit contributions | |
flood | +0.008 |
coming | +0.008 |
rum | +0.007 |
rant | +0.007 |
Why | +0.007 |
Neg logit contributions | |
ons | -0.013 |
bes | -0.011 |
ages | -0.011 |
seek | -0.011 |
con | -0.010 |
Token | pages |
Token ID | 5468 |
Feature activation | -8.42e+00 |
Pos logit contributions | |
ons | +0.052 |
bes | +0.040 |
ages | +0.039 |
igs | +0.036 |
aid | +0.036 |
Neg logit contributions | |
flood | -0.044 |
ending | -0.044 |
ksh | -0.044 |
dolphin | -0.044 |
coming | -0.044 |
Token | got |
Token ID | 1392 |
Feature activation | +7.88e-01 |
Pos logit contributions | |
ons | +0.591 |
bes | +0.463 |
ages | +0.444 |
con | +0.413 |
seek | +0.406 |
Neg logit contributions | |
flood | -0.503 |
rant | -0.497 |
ksh | -0.492 |
dolphin | -0.478 |
coming | -0.472 |
Token | torn |
Token ID | 12445 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
flood | +0.042 |
ksh | +0.042 |
rant | +0.041 |
coming | +0.040 |
dolphin | +0.039 |
Neg logit contributions | |
ons | -0.062 |
bes | -0.050 |
ages | -0.048 |
seek | -0.044 |
con | -0.044 |
Token | . |
Token ID | 13 |
Feature activation | +6.09e-01 |
Pos logit contributions | |
ksh | +0.012 |
flood | +0.012 |
rant | +0.012 |
dolphin | +0.012 |
whale | +0.011 |
Neg logit contributions | |
ons | -0.014 |
bes | -0.011 |
ages | -0.010 |
es | -0.010 |
on | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
ksh | +0.021 |
flood | +0.020 |
rant | +0.018 |
dolphin | +0.018 |
Why | +0.018 |
Neg logit contributions | |
ons | -0.058 |
bes | -0.048 |
ages | -0.048 |
aid | -0.045 |
seek | -0.045 |
Token |
|
Token ID | 198 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
Why | +0.008 |
rant | +0.008 |
flood | +0.007 |
iet | +0.007 |
coming | +0.007 |
Neg logit contributions | |
ons | -0.018 |
bes | -0.015 |
ages | -0.015 |
aces | -0.015 |
con | -0.014 |
Token | can |
Token ID | 460 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
flood | +0.154 |
rant | +0.151 |
whale | +0.150 |
dolphin | +0.149 |
coming | +0.143 |
Neg logit contributions | |
ons | -0.301 |
bes | -0.263 |
ages | -0.243 |
seek | -0.235 |
con | -0.232 |
Token | reveal |
Token ID | 7766 |
Feature activation | +5.18e+00 |
Pos logit contributions | |
ksh | +0.070 |
rant | +0.068 |
flood | +0.068 |
dolphin | +0.066 |
whale | +0.065 |
Neg logit contributions | |
ons | -0.100 |
bes | -0.079 |
ages | -0.076 |
on | -0.071 |
con | -0.070 |
Token | a |
Token ID | 257 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
ksh | +0.265 |
rant | +0.259 |
flood | +0.258 |
dolphin | +0.249 |
whale | +0.245 |
Neg logit contributions | |
ons | -0.417 |
bes | -0.331 |
ages | -0.318 |
on | -0.300 |
con | -0.297 |
Token | solution |
Token ID | 4610 |
Feature activation | +2.80e+00 |
Pos logit contributions | |
ons | +0.113 |
bes | +0.088 |
ages | +0.086 |
on | +0.081 |
con | +0.080 |
Neg logit contributions | |
ksh | -0.076 |
rant | -0.074 |
flood | -0.072 |
dolphin | -0.069 |
iet | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | +1.22e-02 |
Pos logit contributions | |
rant | +0.145 |
ksh | +0.145 |
flood | +0.143 |
dolphin | +0.141 |
whale | +0.138 |
Neg logit contributions | |
ons | -0.218 |
bes | -0.174 |
ages | -0.166 |
on | -0.156 |
con | -0.155 |
Token | From |
Token ID | 3574 |
Feature activation | -8.35e+00 |
Pos logit contributions | |
ksh | +0.001 |
dolphin | +0.001 |
mate | +0.001 |
rant | +0.001 |
flood | +0.001 |
Neg logit contributions | |
ons | -0.001 |
bes | -0.001 |
ages | -0.001 |
seek | -0.001 |
aches | -0.001 |
Token | that |
Token ID | 326 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
ons | +0.630 |
bes | +0.493 |
ages | +0.475 |
on | +0.441 |
es | +0.434 |
Neg logit contributions | |
rant | -0.471 |
ksh | -0.467 |
flood | -0.467 |
dolphin | -0.459 |
whale | -0.451 |
Token | day |
Token ID | 1110 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
flood | +0.026 |
rant | +0.026 |
ksh | +0.025 |
ending | +0.025 |
dolphin | +0.025 |
Neg logit contributions | |
ons | -0.048 |
bes | -0.040 |
ages | -0.038 |
aid | -0.036 |
con | -0.036 |
Token | on |
Token ID | 319 |
Feature activation | -4.80e+00 |
Pos logit contributions | |
ons | +0.099 |
bes | +0.085 |
con | +0.076 |
ages | +0.075 |
seek | +0.074 |
Neg logit contributions | |
coming | -0.088 |
rant | -0.083 |
beginning | -0.083 |
ending | -0.083 |
flood | -0.083 |
Token | , |
Token ID | 11 |
Feature activation | -4.57e-01 |
Pos logit contributions | |
ons | +0.375 |
aid | +0.324 |
ages | +0.323 |
bes | +0.310 |
seek | +0.307 |
Neg logit contributions | |
flood | -0.228 |
dolphin | -0.226 |
ending | -0.221 |
coming | -0.214 |
whale | -0.208 |
Token | Benny |
Token ID | 44275 |
Feature activation | +6.63e-01 |
Pos logit contributions | |
ons | +0.040 |
bes | +0.033 |
ages | +0.032 |
seek | +0.030 |
aid | +0.030 |
Neg logit contributions | |
flood | -0.020 |
dolphin | -0.020 |
ksh | -0.019 |
ending | -0.019 |
mer | -0.019 |
Token | flips |
Token ID | 45971 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
ons | +0.165 |
bes | +0.132 |
ages | +0.130 |
aid | +0.125 |
on | +0.123 |
Neg logit contributions | |
coming | -0.111 |
flood | -0.109 |
dolphin | -0.107 |
mate | -0.103 |
ending | -0.103 |
Token | . |
Token ID | 13 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
ons | +0.201 |
bes | +0.158 |
ages | +0.152 |
on | +0.143 |
con | +0.141 |
Neg logit contributions | |
ksh | -0.137 |
rant | -0.135 |
flood | -0.134 |
dolphin | -0.130 |
whale | -0.128 |
Token | Every |
Token ID | 3887 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
rant | +0.010 |
ksh | +0.010 |
flood | +0.010 |
dolphin | +0.010 |
whale | +0.010 |
Neg logit contributions | |
ons | -0.012 |
bes | -0.009 |
ages | -0.009 |
on | -0.008 |
seek | -0.008 |
Token | day |
Token ID | 1110 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
ons | +0.106 |
bes | +0.085 |
ages | +0.082 |
on | +0.077 |
con | +0.076 |
Neg logit contributions | |
ksh | -0.065 |
flood | -0.064 |
rant | -0.063 |
dolphin | -0.062 |
whale | -0.060 |
Token | he |
Token ID | 339 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
ons | +0.069 |
ages | +0.045 |
bes | +0.044 |
on | +0.041 |
es | +0.039 |
Neg logit contributions | |
ksh | -0.103 |
rant | -0.100 |
flood | -0.098 |
iet | -0.096 |
dolphin | -0.096 |
Token | practiced |
Token ID | 19893 |
Feature activation | -9.22e+00 |
Pos logit contributions | |
ons | +0.214 |
bes | +0.175 |
ages | +0.170 |
on | +0.162 |
con | +0.161 |
Neg logit contributions | |
ksh | -0.088 |
rant | -0.085 |
flood | -0.085 |
dolphin | -0.080 |
whale | -0.078 |
Token | his |
Token ID | 465 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
ons | +0.699 |
ages | +0.551 |
bes | +0.549 |
aid | +0.520 |
con | +0.492 |
Neg logit contributions | |
coming | -0.509 |
dolphin | -0.471 |
rant | -0.470 |
flood | -0.466 |
crashing | -0.458 |
Token | moves |
Token ID | 6100 |
Feature activation | -2.25e+00 |
Pos logit contributions | |
ons | +0.132 |
bes | +0.098 |
ages | +0.095 |
aid | +0.095 |
es | +0.093 |
Neg logit contributions | |
dolphin | -0.092 |
flood | -0.090 |
ending | -0.089 |
whale | -0.088 |
tide | -0.087 |
Token | . |
Token ID | 13 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
ons | +0.186 |
bes | +0.150 |
ages | +0.145 |
aid | +0.134 |
con | +0.134 |
Neg logit contributions | |
rant | -0.104 |
ksh | -0.104 |
dolphin | -0.103 |
coming | -0.102 |
flood | -0.102 |
Token | One |
Token ID | 1881 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
ksh | +0.013 |
rant | +0.013 |
flood | +0.013 |
dolphin | +0.012 |
whale | +0.012 |
Neg logit contributions | |
ons | -0.014 |
bes | -0.011 |
ages | -0.010 |
on | -0.009 |
con | -0.009 |
Token | day |
Token ID | 1110 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
flood | +0.080 |
dolphin | +0.078 |
ending | +0.077 |
whale | +0.076 |
ksh | +0.075 |
Neg logit contributions | |
ons | -0.156 |
bes | -0.131 |
ages | -0.124 |
con | -0.116 |
es | -0.116 |
Token | of |
Token ID | 286 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
ons | +0.257 |
bes | +0.203 |
ages | +0.190 |
con | +0.179 |
aid | +0.175 |
Neg logit contributions | |
rant | -0.210 |
flood | -0.209 |
ksh | -0.206 |
coming | -0.203 |
dolphin | -0.203 |
Token | old |
Token ID | 1468 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
ons | +0.100 |
bes | +0.081 |
ages | +0.076 |
seek | +0.075 |
es | +0.072 |
Neg logit contributions | |
dolphin | -0.083 |
flood | -0.083 |
whale | -0.082 |
ksh | -0.081 |
rant | -0.080 |
Token | papers |
Token ID | 9473 |
Feature activation | -8.65e+00 |
Pos logit contributions | |
ons | +0.106 |
bes | +0.082 |
on | +0.076 |
ages | +0.076 |
aid | +0.076 |
Neg logit contributions | |
flood | -0.093 |
whale | -0.092 |
dolphin | -0.092 |
mate | -0.090 |
ksh | -0.089 |
Token | and |
Token ID | 290 |
Feature activation | -7.19e-02 |
Pos logit contributions | |
ons | +0.669 |
bes | +0.528 |
ages | +0.507 |
con | +0.471 |
on | +0.471 |
Neg logit contributions | |
rant | -0.457 |
ksh | -0.454 |
flood | -0.451 |
dolphin | -0.441 |
mate | -0.434 |
Token | envelop |
Token ID | 16441 |
Feature activation | -3.05e+00 |
Pos logit contributions | |
ons | +0.005 |
bes | +0.004 |
seek | +0.004 |
on | +0.004 |
ages | +0.004 |
Neg logit contributions | |
mate | -0.004 |
dolphin | -0.004 |
whale | -0.004 |
coming | -0.004 |
flood | -0.004 |
Token | es |
Token ID | 274 |
Feature activation | -1.00e+01 |
Pos logit contributions | |
ons | +0.088 |
bes | +0.046 |
ages | +0.035 |
con | +0.029 |
aid | +0.025 |
Neg logit contributions | |
rant | -0.304 |
ksh | -0.304 |
mate | -0.301 |
dolphin | -0.300 |
whale | -0.296 |
Token | . |
Token ID | 13 |
Feature activation | +3.58e-01 |
Pos logit contributions | |
ons | +0.750 |
bes | +0.580 |
ages | +0.558 |
on | +0.523 |
es | +0.519 |
Neg logit contributions | |
ksh | -0.576 |
flood | -0.564 |
rant | -0.561 |
dolphin | -0.541 |
whale | -0.533 |
Token | They |
Token ID | 1119 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
rant | +0.026 |
flood | +0.026 |
ksh | +0.025 |
ending | +0.025 |
whale | +0.025 |
Neg logit contributions | |
ons | -0.020 |
bes | -0.015 |
ages | -0.014 |
on | -0.013 |
con | -0.012 |
Token | take |
Token ID | 1011 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
ksh | +0.077 |
rant | +0.077 |
flood | +0.077 |
dolphin | +0.074 |
whale | +0.072 |
Neg logit contributions | |
ons | -0.141 |
bes | -0.114 |
ages | -0.110 |
con | -0.103 |
on | -0.103 |
Token | turns |
Token ID | 4962 |
Feature activation | -3.02e+00 |
Pos logit contributions | |
ksh | +0.068 |
rant | +0.067 |
flood | +0.066 |
dolphin | +0.064 |
whale | +0.063 |
Neg logit contributions | |
ons | -0.094 |
bes | -0.075 |
ages | -0.071 |
on | -0.067 |
con | -0.066 |
Token | to |
Token ID | 284 |
Feature activation | -4.29e-01 |
Pos logit contributions | |
ons | +0.229 |
bes | +0.189 |
ages | +0.175 |
aid | +0.172 |
seek | +0.166 |
Neg logit contributions | |
coming | -0.157 |
rant | -0.151 |
flood | -0.149 |
ksh | -0.148 |
whale | -0.147 |
Token | She |
Token ID | 1375 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
ksh | +0.036 |
rant | +0.036 |
flood | +0.036 |
dolphin | +0.035 |
whale | +0.034 |
Neg logit contributions | |
ons | -0.038 |
bes | -0.029 |
ages | -0.028 |
on | -0.026 |
con | -0.025 |
Token | loved |
Token ID | 6151 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
ons | +0.138 |
bes | +0.114 |
con | +0.112 |
ages | +0.110 |
aid | +0.107 |
Neg logit contributions | |
wake | -0.052 |
coming | -0.051 |
dolphin | -0.051 |
ending | -0.050 |
flood | -0.050 |
Token | to |
Token ID | 284 |
Feature activation | -2.29e-01 |
Pos logit contributions | |
ons | +0.012 |
bes | +0.011 |
ages | +0.010 |
es | +0.010 |
aid | +0.009 |
Neg logit contributions | |
coming | -0.007 |
ksh | -0.007 |
flood | -0.007 |
dolphin | -0.007 |
whale | -0.007 |
Token | play |
Token ID | 711 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
ons | +0.021 |
bes | +0.017 |
ages | +0.017 |
es | +0.017 |
con | +0.016 |
Neg logit contributions | |
wake | -0.009 |
flood | -0.009 |
dolphin | -0.009 |
whale | -0.008 |
tide | -0.008 |
Token | with |
Token ID | 351 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
ons | +0.165 |
bes | +0.130 |
ages | +0.128 |
con | +0.119 |
on | +0.118 |
Neg logit contributions | |
ksh | -0.103 |
flood | -0.103 |
dolphin | -0.101 |
rant | -0.101 |
whale | -0.098 |
Token | puzzles |
Token ID | 24367 |
Feature activation | -1.29e+01 |
Pos logit contributions | |
ons | +0.082 |
ages | +0.075 |
seek | +0.073 |
bes | +0.071 |
con | +0.068 |
Neg logit contributions | |
dolphin | -0.052 |
coming | -0.052 |
flood | -0.051 |
whale | -0.050 |
iet | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | +5.90e-01 |
Pos logit contributions | |
ons | +1.154 |
ages | +0.966 |
bes | +0.954 |
con | +0.893 |
aid | +0.890 |
Neg logit contributions | |
coming | -0.513 |
rant | -0.504 |
like | -0.499 |
flood | -0.484 |
mate | -0.482 |
Token | One |
Token ID | 1881 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
flood | +0.034 |
dolphin | +0.034 |
whale | +0.034 |
ksh | +0.034 |
rant | +0.033 |
Neg logit contributions | |
ons | -0.043 |
bes | -0.033 |
on | -0.031 |
ages | -0.030 |
con | -0.030 |
Token | day |
Token ID | 1110 |
Feature activation | -8.62e-01 |
Pos logit contributions | |
ksh | +0.130 |
rant | +0.128 |
flood | +0.124 |
dolphin | +0.120 |
mate | +0.118 |
Neg logit contributions | |
ons | -0.161 |
bes | -0.121 |
ages | -0.119 |
on | -0.116 |
aid | -0.108 |
Token | , |
Token ID | 11 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
ons | +0.062 |
bes | +0.051 |
ages | +0.047 |
con | +0.045 |
aid | +0.044 |
Neg logit contributions | |
flood | -0.048 |
dolphin | -0.047 |
rant | -0.047 |
ksh | -0.047 |
coming | -0.047 |
Token | she |
Token ID | 673 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
ksh | +0.010 |
flood | +0.010 |
rant | +0.010 |
dolphin | +0.010 |
whale | +0.010 |
Neg logit contributions | |
ons | -0.009 |
bes | -0.006 |
ages | -0.006 |
on | -0.005 |
con | -0.005 |