Token | mom |
Token ID | 1995 |
Feature activation | +4.62e+00 |
Pos logit contributions | |
he | +0.071 |
answer | +0.064 |
note | +0.064 |
ad | +0.060 |
fairy | +0.060 |
Neg logit contributions | |
ashed | -0.093 |
tanks | -0.092 |
ksh | -0.092 |
rods | -0.087 |
mouths | -0.085 |
Token | said |
Token ID | 531 |
Feature activation | +6.68e+00 |
Pos logit contributions | |
he | +0.221 |
note | +0.197 |
answer | +0.192 |
guardian | +0.192 |
fairy | +0.187 |
Neg logit contributions | |
tanks | -0.315 |
ashed | -0.315 |
ksh | -0.304 |
rods | -0.297 |
trucks | -0.290 |
Token | with |
Token ID | 351 |
Feature activation | +4.93e+00 |
Pos logit contributions | |
he | +0.331 |
answer | +0.305 |
note | +0.302 |
fairy | +0.294 |
voice | +0.287 |
Neg logit contributions | |
tanks | -0.451 |
ashed | -0.438 |
ksh | -0.436 |
rods | -0.430 |
trucks | -0.415 |
Token | a |
Token ID | 257 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
he | +0.240 |
note | +0.213 |
answer | +0.210 |
guardian | +0.207 |
fairy | +0.203 |
Neg logit contributions | |
tanks | -0.335 |
ashed | -0.332 |
ksh | -0.323 |
rods | -0.317 |
trucks | -0.307 |
Token | tw |
Token ID | 665 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
he | +0.110 |
ad | +0.092 |
ibrarian | +0.090 |
guardian | +0.086 |
answer | +0.085 |
Neg logit contributions | |
tanks | -0.146 |
ashed | -0.140 |
rods | -0.139 |
mouths | -0.138 |
pic | -0.136 |
Token | inkle |
Token ID | 19894 |
Feature activation | +1.48e+01 |
Pos logit contributions | |
he | +0.007 |
note | +0.007 |
answer | +0.006 |
ibrarian | +0.006 |
reply | +0.006 |
Neg logit contributions | |
ashed | -0.005 |
pic | -0.004 |
sizes | -0.004 |
tails | -0.004 |
tanks | -0.004 |
Token | in |
Token ID | 287 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
he | +0.764 |
note | +0.683 |
answer | +0.671 |
guardian | +0.664 |
ibrarian | +0.644 |
Neg logit contributions | |
ashed | -0.951 |
tanks | -0.940 |
ksh | -0.907 |
rods | -0.885 |
mouths | -0.863 |
Token | her |
Token ID | 607 |
Feature activation | +4.93e+00 |
Pos logit contributions | |
he | +0.055 |
ibrarian | +0.051 |
iver | +0.049 |
guardian | +0.047 |
orn | +0.047 |
Neg logit contributions | |
pic | -0.058 |
tanks | -0.056 |
mouths | -0.055 |
rods | -0.054 |
tails | -0.054 |
Token | eye |
Token ID | 4151 |
Feature activation | +9.81e+00 |
Pos logit contributions | |
he | +0.266 |
note | +0.216 |
ibrarian | +0.216 |
ad | +0.213 |
answer | +0.211 |
Neg logit contributions | |
mouths | -0.309 |
tanks | -0.304 |
ashed | -0.302 |
tails | -0.292 |
ksh | -0.291 |
Token | . |
Token ID | 13 |
Feature activation | +4.08e-01 |
Pos logit contributions | |
he | +0.512 |
note | +0.452 |
guardian | +0.449 |
answer | +0.449 |
ad | +0.431 |
Neg logit contributions | |
ashed | -0.614 |
tanks | -0.609 |
ksh | -0.584 |
rods | -0.574 |
mouths | -0.560 |
Token |
|
Token ID | 198 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
he | +0.018 |
answer | +0.015 |
fairy | +0.015 |
note | +0.015 |
ibrarian | +0.015 |
Neg logit contributions | |
tanks | -0.030 |
ashed | -0.029 |
rods | -0.029 |
ksh | -0.028 |
trucks | -0.028 |
Token | a |
Token ID | 257 |
Feature activation | +2.60e+00 |
Pos logit contributions | |
he | +0.095 |
note | +0.084 |
answer | +0.083 |
guardian | +0.081 |
fairy | +0.080 |
Neg logit contributions | |
tanks | -0.144 |
ashed | -0.144 |
ksh | -0.140 |
rods | -0.137 |
trucks | -0.133 |
Token | little |
Token ID | 1310 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
he | +0.129 |
answer | +0.112 |
note | +0.112 |
guardian | +0.109 |
ad | +0.107 |
Neg logit contributions | |
tanks | -0.173 |
ashed | -0.171 |
ksh | -0.166 |
rods | -0.164 |
trucks | -0.159 |
Token | star |
Token ID | 3491 |
Feature activation | +8.35e+00 |
Pos logit contributions | |
tanks | +0.130 |
ashed | +0.129 |
ksh | +0.126 |
rods | +0.123 |
trucks | +0.119 |
Neg logit contributions | |
he | -0.098 |
note | -0.087 |
answer | -0.086 |
guardian | -0.084 |
fairy | -0.082 |
Token | named |
Token ID | 3706 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
he | +0.426 |
answer | +0.383 |
note | +0.375 |
guardian | +0.368 |
ibrarian | +0.361 |
Neg logit contributions | |
tanks | -0.535 |
ashed | -0.529 |
rods | -0.498 |
ksh | -0.496 |
trucks | -0.490 |
Token | Tw |
Token ID | 1815 |
Feature activation | +1.85e+00 |
Pos logit contributions | |
he | +0.098 |
note | +0.091 |
answer | +0.090 |
guardian | +0.089 |
fairy | +0.088 |
Neg logit contributions | |
tanks | -0.056 |
ashed | -0.056 |
ksh | -0.053 |
rods | -0.051 |
trucks | -0.049 |
Token | inkle |
Token ID | 19894 |
Feature activation | +1.45e+01 |
Pos logit contributions | |
he | +0.053 |
note | +0.045 |
answer | +0.043 |
guardian | +0.041 |
owl | +0.037 |
Neg logit contributions | |
ashed | -0.156 |
tanks | -0.149 |
rods | -0.146 |
ksh | -0.146 |
pic | -0.145 |
Token | . |
Token ID | 13 |
Feature activation | +5.07e-01 |
Pos logit contributions | |
he | +0.775 |
note | +0.700 |
answer | +0.699 |
guardian | +0.681 |
ibrarian | +0.661 |
Neg logit contributions | |
ashed | -0.900 |
tanks | -0.891 |
ksh | -0.855 |
rods | -0.831 |
mouths | -0.810 |
Token | Tw |
Token ID | 1815 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
he | +0.030 |
answer | +0.027 |
note | +0.027 |
guardian | +0.026 |
fairy | +0.026 |
Neg logit contributions | |
ashed | -0.029 |
tanks | -0.029 |
ksh | -0.028 |
rods | -0.028 |
trucks | -0.026 |
Token | inkle |
Token ID | 19894 |
Feature activation | +1.42e+01 |
Pos logit contributions | |
he | +0.039 |
note | +0.032 |
answer | +0.029 |
guardian | +0.029 |
reply | +0.026 |
Neg logit contributions | |
ashed | -0.132 |
tanks | -0.127 |
rods | -0.125 |
ksh | -0.124 |
pic | -0.123 |
Token | was |
Token ID | 373 |
Feature activation | -2.79e-01 |
Pos logit contributions | |
he | +0.842 |
answer | +0.780 |
note | +0.771 |
guardian | +0.757 |
ibrarian | +0.747 |
Neg logit contributions | |
ashed | -0.800 |
tanks | -0.772 |
rods | -0.718 |
ksh | -0.717 |
sizes | -0.707 |
Token | a |
Token ID | 257 |
Feature activation | +9.10e-01 |
Pos logit contributions | |
tanks | +0.015 |
ashed | +0.015 |
sizes | +0.015 |
ksh | +0.014 |
tails | +0.014 |
Neg logit contributions | |
he | -0.017 |
note | -0.015 |
answer | -0.014 |
ibrarian | -0.014 |
figure | -0.014 |
Token | man |
Token ID | 582 |
Feature activation | +7.99e+00 |
Pos logit contributions | |
tanks | +0.013 |
trucks | +0.013 |
rods | +0.013 |
helmets | +0.012 |
mats | +0.012 |
Neg logit contributions | |
he | -0.009 |
iver | -0.009 |
orn | -0.008 |
guardian | -0.008 |
inkle | -0.008 |
Token | played |
Token ID | 2826 |
Feature activation | -7.83e+00 |
Pos logit contributions | |
he | +0.462 |
answer | +0.415 |
figure | +0.403 |
note | +0.398 |
fairy | +0.397 |
Neg logit contributions | |
ksh | -0.486 |
tanks | -0.470 |
rods | -0.444 |
sizes | -0.437 |
pic | -0.437 |
Token | Tw |
Token ID | 1815 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
ashed | +0.478 |
tanks | +0.477 |
ksh | +0.465 |
rods | +0.447 |
mouths | +0.433 |
Neg logit contributions | |
he | -0.442 |
note | -0.397 |
answer | -0.394 |
fairy | -0.380 |
guardian | -0.377 |
Token | inkle |
Token ID | 19894 |
Feature activation | +1.45e+01 |
Pos logit contributions | |
he | +0.049 |
note | +0.043 |
guardian | +0.042 |
answer | +0.040 |
fairy | +0.040 |
Neg logit contributions | |
rods | -0.142 |
tanks | -0.141 |
ashed | -0.140 |
ksh | -0.135 |
mouths | -0.133 |
Token | Tw |
Token ID | 1815 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
he | +0.603 |
fairy | +0.537 |
answer | +0.518 |
note | +0.496 |
voice | +0.491 |
Neg logit contributions | |
tanks | -1.114 |
ashed | -1.085 |
ksh | -1.082 |
rods | -1.063 |
pic | -1.039 |
Token | inkle |
Token ID | 19894 |
Feature activation | +1.45e+01 |
Pos logit contributions | |
he | +0.046 |
note | +0.037 |
guardian | +0.036 |
answer | +0.035 |
fairy | +0.034 |
Neg logit contributions | |
tanks | -0.150 |
rods | -0.150 |
ashed | -0.149 |
ksh | -0.145 |
mouths | -0.144 |
Token | Little |
Token ID | 7703 |
Feature activation | -7.72e+00 |
Pos logit contributions | |
he | +0.597 |
fairy | +0.538 |
answer | +0.531 |
voice | +0.510 |
note | +0.505 |
Neg logit contributions | |
tanks | -1.111 |
ashed | -1.071 |
ksh | -1.069 |
rods | -1.065 |
pic | -1.045 |
Token | Star |
Token ID | 2907 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
ashed | +0.493 |
tanks | +0.486 |
rods | +0.456 |
ksh | +0.453 |
sizes | +0.451 |
Neg logit contributions | |
he | -0.412 |
answer | -0.390 |
fairy | -0.376 |
ad | -0.372 |
voice | -0.371 |
Token | together |
Token ID | 1978 |
Feature activation | -8.36e+00 |
Pos logit contributions | |
he | +0.021 |
fairy | +0.018 |
note | +0.017 |
answer | +0.017 |
bird | +0.016 |
Neg logit contributions | |
tanks | -0.045 |
ashed | -0.045 |
ksh | -0.044 |
rods | -0.043 |
pic | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | +6.62e-01 |
Pos logit contributions | |
tanks | +0.640 |
ksh | +0.634 |
ashed | +0.631 |
rods | +0.610 |
trucks | +0.594 |
Neg logit contributions | |
he | -0.354 |
answer | -0.310 |
note | -0.304 |
fairy | -0.293 |
voice | -0.278 |
Token | Lily |
Token ID | 20037 |
Feature activation | -3.92e+00 |
Pos logit contributions | |
he | +0.050 |
answer | +0.045 |
fairy | +0.045 |
voice | +0.043 |
ad | +0.043 |
Neg logit contributions | |
ashed | -0.028 |
rods | -0.026 |
tanks | -0.025 |
mons | -0.023 |
trucks | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
he | +0.253 |
note | +0.216 |
answer | +0.215 |
fairy | +0.210 |
ibrarian | +0.208 |
Neg logit contributions | |
tanks | -0.435 |
ksh | -0.429 |
rods | -0.414 |
ashed | -0.413 |
mouths | -0.403 |
Token | the |
Token ID | 262 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
he | +0.087 |
answer | +0.081 |
fairy | +0.080 |
note | +0.078 |
ad | +0.077 |
Neg logit contributions | |
ashed | -0.083 |
tanks | -0.082 |
ksh | -0.082 |
rods | -0.079 |
sizes | -0.077 |
Token | man |
Token ID | 582 |
Feature activation | +7.99e+00 |
Pos logit contributions | |
tanks | +0.013 |
trucks | +0.013 |
rods | +0.013 |
helmets | +0.012 |
mats | +0.012 |
Neg logit contributions | |
he | -0.009 |
iver | -0.009 |
orn | -0.008 |
guardian | -0.008 |
inkle | -0.008 |
Token | played |
Token ID | 2826 |
Feature activation | -7.83e+00 |
Pos logit contributions | |
he | +0.462 |
answer | +0.415 |
figure | +0.403 |
note | +0.398 |
fairy | +0.397 |
Neg logit contributions | |
ksh | -0.486 |
tanks | -0.470 |
rods | -0.444 |
sizes | -0.437 |
pic | -0.437 |
Token | Tw |
Token ID | 1815 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
ashed | +0.478 |
tanks | +0.477 |
ksh | +0.465 |
rods | +0.447 |
mouths | +0.433 |
Neg logit contributions | |
he | -0.442 |
note | -0.397 |
answer | -0.394 |
fairy | -0.380 |
guardian | -0.377 |
Token | inkle |
Token ID | 19894 |
Feature activation | +1.45e+01 |
Pos logit contributions | |
he | +0.049 |
note | +0.043 |
guardian | +0.042 |
answer | +0.040 |
fairy | +0.040 |
Neg logit contributions | |
rods | -0.142 |
tanks | -0.141 |
ashed | -0.140 |
ksh | -0.135 |
mouths | -0.133 |
Token | Tw |
Token ID | 1815 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
he | +0.603 |
fairy | +0.537 |
answer | +0.518 |
note | +0.496 |
voice | +0.491 |
Neg logit contributions | |
tanks | -1.114 |
ashed | -1.085 |
ksh | -1.082 |
rods | -1.063 |
pic | -1.039 |
Token | inkle |
Token ID | 19894 |
Feature activation | +1.45e+01 |
Pos logit contributions | |
he | +0.046 |
note | +0.037 |
guardian | +0.036 |
answer | +0.035 |
fairy | +0.034 |
Neg logit contributions | |
tanks | -0.150 |
rods | -0.150 |
ashed | -0.149 |
ksh | -0.145 |
mouths | -0.144 |
Token | Little |
Token ID | 7703 |
Feature activation | -7.72e+00 |
Pos logit contributions | |
he | +0.597 |
fairy | +0.538 |
answer | +0.531 |
voice | +0.510 |
note | +0.505 |
Neg logit contributions | |
tanks | -1.111 |
ashed | -1.071 |
ksh | -1.069 |
rods | -1.065 |
pic | -1.045 |
Token | Star |
Token ID | 2907 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
ashed | +0.493 |
tanks | +0.486 |
rods | +0.456 |
ksh | +0.453 |
sizes | +0.451 |
Neg logit contributions | |
he | -0.412 |
answer | -0.390 |
fairy | -0.376 |
ad | -0.372 |
voice | -0.371 |
Token | together |
Token ID | 1978 |
Feature activation | -8.36e+00 |
Pos logit contributions | |
he | +0.021 |
fairy | +0.018 |
note | +0.017 |
answer | +0.017 |
bird | +0.016 |
Neg logit contributions | |
tanks | -0.045 |
ashed | -0.045 |
ksh | -0.044 |
rods | -0.043 |
pic | -0.043 |
Token | and |
Token ID | 290 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
he | +0.123 |
note | +0.108 |
ibrarian | +0.108 |
answer | +0.102 |
guardian | +0.102 |
Neg logit contributions | |
ashed | -0.130 |
ksh | -0.125 |
tanks | -0.124 |
mouths | -0.114 |
rods | -0.113 |
Token | said |
Token ID | 531 |
Feature activation | +4.66e+00 |
Pos logit contributions | |
he | +0.059 |
note | +0.052 |
answer | +0.052 |
fairy | +0.051 |
guardian | +0.050 |
Neg logit contributions | |
tanks | -0.111 |
ashed | -0.109 |
ksh | -0.108 |
rods | -0.105 |
trucks | -0.102 |
Token | with |
Token ID | 351 |
Feature activation | +3.62e+00 |
Pos logit contributions | |
he | +0.219 |
note | +0.207 |
answer | +0.206 |
fairy | +0.206 |
guardian | +0.195 |
Neg logit contributions | |
tanks | -0.332 |
ksh | -0.322 |
ashed | -0.317 |
rods | -0.312 |
pic | -0.306 |
Token | a |
Token ID | 257 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
he | +0.215 |
guardian | +0.208 |
wizard | +0.203 |
note | +0.202 |
iver | +0.202 |
Neg logit contributions | |
tanks | -0.171 |
pic | -0.168 |
mouths | -0.162 |
tails | -0.162 |
ashed | -0.162 |
Token | tw |
Token ID | 665 |
Feature activation | -4.48e-01 |
Pos logit contributions | |
he | +0.078 |
ibrarian | +0.068 |
ad | +0.067 |
handed | +0.067 |
iver | +0.064 |
Neg logit contributions | |
pic | -0.062 |
tanks | -0.062 |
they | -0.062 |
mouths | -0.061 |
while | -0.060 |
Token | inkle |
Token ID | 19894 |
Feature activation | +1.45e+01 |
Pos logit contributions | |
ashed | +0.026 |
igs | +0.021 |
rods | +0.021 |
suits | +0.021 |
pic | +0.021 |
Neg logit contributions | |
he | -0.024 |
note | -0.020 |
ibrarian | -0.020 |
answer | -0.020 |
angel | -0.020 |
Token | in |
Token ID | 287 |
Feature activation | +8.84e-01 |
Pos logit contributions | |
he | +0.796 |
note | +0.712 |
guardian | +0.687 |
answer | +0.686 |
ibrarian | +0.684 |
Neg logit contributions | |
ashed | -0.873 |
tanks | -0.851 |
ksh | -0.805 |
rods | -0.791 |
mouths | -0.779 |
Token | her |
Token ID | 607 |
Feature activation | +4.93e+00 |
Pos logit contributions | |
he | +0.056 |
ibrarian | +0.053 |
iver | +0.052 |
ble | +0.048 |
orn | +0.047 |
Neg logit contributions | |
pic | -0.043 |
tails | -0.042 |
mouths | -0.040 |
they | -0.039 |
tanks | -0.038 |
Token | eye |
Token ID | 4151 |
Feature activation | +9.94e+00 |
Pos logit contributions | |
he | +0.251 |
ibrarian | +0.205 |
note | +0.177 |
ad | +0.175 |
iver | +0.175 |
Neg logit contributions | |
mouths | -0.346 |
tails | -0.326 |
ashed | -0.324 |
tanks | -0.321 |
ksh | -0.312 |
Token | , |
Token ID | 11 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
he | +0.562 |
guardian | +0.499 |
answer | +0.487 |
note | +0.485 |
ad | +0.481 |
Neg logit contributions | |
ashed | -0.558 |
tanks | -0.547 |
ksh | -0.520 |
mouths | -0.515 |
rods | -0.511 |
Token | " |
Token ID | 366 |
Feature activation | -5.53e+00 |
Pos logit contributions | |
tanks | +0.059 |
ashed | +0.059 |
ksh | +0.057 |
rods | +0.055 |
trucks | +0.053 |
Neg logit contributions | |
he | -0.074 |
note | -0.067 |
answer | -0.067 |
guardian | -0.066 |
fairy | -0.065 |
Token | window |
Token ID | 4324 |
Feature activation | +5.66e+00 |
Pos logit contributions | |
tanks | +0.014 |
ksh | +0.014 |
trucks | +0.014 |
rods | +0.013 |
mouths | +0.013 |
Neg logit contributions | |
he | -0.012 |
ibrarian | -0.011 |
ad | -0.011 |
note | -0.011 |
guardian | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
he | +0.288 |
note | +0.256 |
answer | +0.254 |
guardian | +0.248 |
fairy | +0.245 |
Neg logit contributions | |
tanks | -0.370 |
ashed | -0.369 |
ksh | -0.359 |
rods | -0.350 |
trucks | -0.338 |
Token | watch |
Token ID | 2342 |
Feature activation | +2.16e+00 |
Pos logit contributions | |
he | +0.072 |
note | +0.062 |
ibrarian | +0.062 |
guardian | +0.060 |
ad | +0.058 |
Neg logit contributions | |
tanks | -0.096 |
mouths | -0.091 |
ashed | -0.091 |
trucks | -0.091 |
ksh | -0.090 |
Token | them |
Token ID | 606 |
Feature activation | -4.32e-01 |
Pos logit contributions | |
he | +0.120 |
note | +0.108 |
answer | +0.105 |
guardian | +0.105 |
fairy | +0.103 |
Neg logit contributions | |
tanks | -0.133 |
ashed | -0.130 |
ksh | -0.127 |
rods | -0.124 |
trucks | -0.121 |
Token | tw |
Token ID | 665 |
Feature activation | -4.73e-01 |
Pos logit contributions | |
tanks | +0.029 |
ashed | +0.028 |
ksh | +0.027 |
trucks | +0.027 |
mouths | +0.027 |
Neg logit contributions | |
he | -0.020 |
ibrarian | -0.019 |
guardian | -0.018 |
note | -0.018 |
ocket | -0.017 |
Token | inkle |
Token ID | 19894 |
Feature activation | +1.45e+01 |
Pos logit contributions | |
ashed | +0.033 |
tanks | +0.032 |
ksh | +0.032 |
rods | +0.031 |
mouths | +0.030 |
Neg logit contributions | |
he | -0.021 |
note | -0.021 |
answer | -0.020 |
guardian | -0.020 |
fairy | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | +6.06e-01 |
Pos logit contributions | |
he | +0.489 |
answer | +0.429 |
fairy | +0.403 |
note | +0.374 |
voice | +0.365 |
Neg logit contributions | |
ashed | -1.230 |
tanks | -1.183 |
rods | -1.165 |
ksh | -1.144 |
pic | -1.131 |
Token | One |
Token ID | 1881 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
he | +0.034 |
answer | +0.030 |
ad | +0.029 |
voice | +0.026 |
mirror | +0.025 |
Neg logit contributions | |
ashed | -0.039 |
ksh | -0.036 |
tanks | -0.036 |
suits | -0.035 |
rods | -0.035 |
Token | day |
Token ID | 1110 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
tanks | +0.057 |
ashed | +0.057 |
ksh | +0.053 |
rods | +0.052 |
trucks | +0.051 |
Neg logit contributions | |
he | -0.083 |
note | -0.077 |
guardian | -0.075 |
answer | -0.075 |
ibrarian | -0.075 |
Token | , |
Token ID | 11 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
ksh | +0.175 |
girls | +0.155 |
sizes | +0.151 |
pic | +0.150 |
habitats | +0.147 |
Neg logit contributions | |
he | -0.033 |
answer | -0.029 |
on | -0.026 |
ad | -0.023 |
voice | -0.022 |
Token | Tim |
Token ID | 5045 |
Feature activation | -6.46e+00 |
Pos logit contributions | |
ksh | +0.075 |
suits | +0.064 |
ashed | +0.064 |
appet | +0.063 |
girls | +0.063 |
Neg logit contributions | |
he | -0.075 |
answer | -0.066 |
on | -0.061 |
figure | -0.061 |
ad | -0.061 |
Token | her |
Token ID | 607 |
Feature activation | +4.77e+00 |
Pos logit contributions | |
tanks | +0.123 |
ashed | +0.117 |
rods | +0.112 |
ksh | +0.110 |
trucks | +0.109 |
Neg logit contributions | |
he | -0.174 |
note | -0.155 |
guardian | -0.155 |
answer | -0.152 |
fairy | -0.150 |
Token | name |
Token ID | 1438 |
Feature activation | +9.89e+00 |
Pos logit contributions | |
he | +0.257 |
note | +0.231 |
answer | +0.229 |
guardian | +0.224 |
fairy | +0.221 |
Neg logit contributions | |
tanks | -0.300 |
ashed | -0.299 |
ksh | -0.290 |
rods | -0.283 |
trucks | -0.273 |
Token | to |
Token ID | 284 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
he | +0.509 |
note | +0.453 |
answer | +0.448 |
guardian | +0.442 |
fairy | +0.433 |
Neg logit contributions | |
tanks | -0.637 |
ashed | -0.636 |
ksh | -0.616 |
rods | -0.601 |
trucks | -0.584 |
Token | be |
Token ID | 307 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
tanks | +0.138 |
trucks | +0.133 |
sizes | +0.126 |
pic | +0.126 |
mouths | +0.125 |
Neg logit contributions | |
he | -0.105 |
note | -0.093 |
guardian | -0.092 |
ibrarian | -0.087 |
orn | -0.085 |
Token | Tw |
Token ID | 1815 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
he | +0.064 |
note | +0.055 |
answer | +0.054 |
guardian | +0.052 |
ibrarian | +0.052 |
Neg logit contributions | |
tanks | -0.087 |
ashed | -0.086 |
ksh | -0.083 |
rods | -0.081 |
pic | -0.080 |
Token | inkle |
Token ID | 19894 |
Feature activation | +1.44e+01 |
Pos logit contributions | |
he | +0.079 |
note | +0.072 |
answer | +0.070 |
guardian | +0.068 |
fairy | +0.068 |
Neg logit contributions | |
ashed | -0.108 |
tanks | -0.105 |
ksh | -0.104 |
rods | -0.101 |
mouths | -0.097 |
Token | ! |
Token ID | 0 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
he | +0.700 |
answer | +0.638 |
fairy | +0.635 |
note | +0.618 |
guardian | +0.611 |
Neg logit contributions | |
tanks | -0.977 |
ashed | -0.970 |
ksh | -0.949 |
rods | -0.948 |
pic | -0.908 |
Token | Her |
Token ID | 2332 |
Feature activation | +3.52e+00 |
Pos logit contributions | |
he | +0.125 |
answer | +0.113 |
voice | +0.110 |
ibrarian | +0.110 |
mirror | +0.109 |
Neg logit contributions | |
ashed | -0.178 |
tanks | -0.173 |
rods | -0.170 |
ksh | -0.164 |
sizes | -0.163 |
Token | mom |
Token ID | 1995 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
he | +0.218 |
note | +0.181 |
ibrarian | +0.180 |
answer | +0.180 |
ad | +0.180 |
Neg logit contributions | |
tanks | -0.201 |
ashed | -0.196 |
mouths | -0.188 |
ksh | -0.185 |
rods | -0.185 |
Token | my |
Token ID | 1820 |
Feature activation | +5.66e+00 |
Pos logit contributions | |
he | +0.070 |
note | +0.062 |
ocket | +0.062 |
guardian | +0.061 |
answer | +0.061 |
Neg logit contributions | |
ashed | -0.092 |
tanks | -0.089 |
trucks | -0.088 |
sizes | -0.086 |
ksh | -0.082 |
Token | and |
Token ID | 290 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
he | +0.306 |
note | +0.269 |
answer | +0.266 |
guardian | +0.259 |
ad | +0.251 |
Neg logit contributions | |
ashed | -0.355 |
tanks | -0.351 |
ksh | -0.328 |
trucks | -0.323 |
rods | -0.322 |
Token | it |
Token ID | 340 |
Feature activation | +3.12e+00 |
Pos logit contributions | |
he | +0.070 |
fairy | +0.066 |
note | +0.066 |
answer | +0.065 |
guardian | +0.065 |
Neg logit contributions | |
ksh | -0.085 |
ashed | -0.081 |
tanks | -0.081 |
rods | -0.078 |
suits | -0.077 |
Token | suddenly |
Token ID | 6451 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
he | +0.178 |
note | +0.160 |
answer | +0.159 |
guardian | +0.156 |
fairy | +0.153 |
Neg logit contributions | |
tanks | -0.185 |
ashed | -0.185 |
ksh | -0.178 |
rods | -0.173 |
trucks | -0.168 |
Token | seemed |
Token ID | 3947 |
Feature activation | -7.16e-01 |
Pos logit contributions | |
tanks | +0.121 |
ashed | +0.121 |
ksh | +0.119 |
rods | +0.114 |
trucks | +0.110 |
Neg logit contributions | |
he | -0.104 |
note | -0.095 |
answer | -0.094 |
guardian | -0.092 |
fairy | -0.091 |
Token | to |
Token ID | 284 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
tanks | +0.047 |
ashed | +0.047 |
ksh | +0.046 |
rods | +0.045 |
trucks | +0.044 |
Neg logit contributions | |
he | -0.035 |
answer | -0.033 |
note | -0.032 |
guardian | -0.032 |
fairy | -0.032 |
Token | tw |
Token ID | 665 |
Feature activation | -5.69e-01 |
Pos logit contributions | |
tanks | +0.113 |
they | +0.109 |
ashed | +0.108 |
sizes | +0.107 |
Nem | +0.106 |
Neg logit contributions | |
ibrarian | -0.126 |
ocket | -0.112 |
he | -0.109 |
guardian | -0.106 |
note | -0.106 |
Token | inkle |
Token ID | 19894 |
Feature activation | +1.44e+01 |
Pos logit contributions | |
ashed | +0.039 |
tanks | +0.036 |
ksh | +0.035 |
rods | +0.035 |
tails | +0.034 |
Neg logit contributions | |
answer | -0.027 |
note | -0.026 |
he | -0.025 |
fairy | -0.024 |
angel | -0.024 |
Token | brighter |
Token ID | 26083 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
he | +0.552 |
answer | +0.490 |
fairy | +0.486 |
note | +0.456 |
guardian | +0.443 |
Neg logit contributions | |
ashed | -1.113 |
tanks | -1.083 |
ksh | -1.080 |
rods | -1.078 |
pic | -1.056 |
Token | . |
Token ID | 13 |
Feature activation | +5.06e-01 |
Pos logit contributions | |
ashed | +0.123 |
tanks | +0.119 |
ksh | +0.118 |
rods | +0.113 |
pic | +0.109 |
Neg logit contributions | |
he | -0.065 |
answer | -0.060 |
note | -0.059 |
fairy | -0.057 |
guardian | -0.057 |
Token | It |
Token ID | 632 |
Feature activation | +3.58e+00 |
Pos logit contributions | |
he | +0.022 |
answer | +0.019 |
ad | +0.018 |
note | +0.018 |
guardian | +0.018 |
Neg logit contributions | |
ashed | -0.037 |
ksh | -0.036 |
tanks | -0.036 |
rods | -0.034 |
suits | -0.034 |
Token | suddenly |
Token ID | 6451 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
he | +0.207 |
note | +0.186 |
answer | +0.183 |
guardian | +0.181 |
ibrarian | +0.178 |
Neg logit contributions | |
ashed | -0.211 |
tanks | -0.210 |
ksh | -0.200 |
rods | -0.193 |
sizes | -0.190 |
Token | reminded |
Token ID | 14516 |
Feature activation | +3.94e+00 |
Pos logit contributions | |
tanks | +0.112 |
ashed | +0.112 |
ksh | +0.109 |
rods | +0.106 |
trucks | +0.102 |
Neg logit contributions | |
he | -0.103 |
note | -0.093 |
answer | -0.093 |
guardian | -0.091 |
fairy | -0.090 |
Token | forgot |
Token ID | 16453 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
he | +0.289 |
answer | +0.267 |
note | +0.264 |
guardian | +0.260 |
fairy | +0.259 |
Neg logit contributions | |
tanks | -0.254 |
ksh | -0.246 |
ashed | -0.244 |
rods | -0.243 |
trucks | -0.230 |
Token | her |
Token ID | 607 |
Feature activation | +4.76e+00 |
Pos logit contributions | |
ashed | +0.169 |
ksh | +0.169 |
tanks | +0.164 |
rods | +0.161 |
suits | +0.159 |
Neg logit contributions | |
he | -0.099 |
answer | -0.092 |
fairy | -0.088 |
note | -0.087 |
guardian | -0.086 |
Token | adventure |
Token ID | 8855 |
Feature activation | -2.46e+00 |
Pos logit contributions | |
he | +0.262 |
note | +0.233 |
answer | +0.230 |
guardian | +0.226 |
fairy | +0.223 |
Neg logit contributions | |
tanks | -0.296 |
ashed | -0.294 |
ksh | -0.284 |
rods | -0.279 |
trucks | -0.270 |
Token | with |
Token ID | 351 |
Feature activation | +3.51e+00 |
Pos logit contributions | |
tanks | +0.165 |
ashed | +0.165 |
ksh | +0.161 |
rods | +0.157 |
trucks | +0.151 |
Neg logit contributions | |
he | -0.123 |
note | -0.109 |
answer | -0.108 |
guardian | -0.105 |
fairy | -0.104 |
Token | Cr |
Token ID | 3864 |
Feature activation | -2.79e+00 |
Pos logit contributions | |
he | +0.137 |
answer | +0.125 |
note | +0.124 |
fairy | +0.120 |
guardian | +0.118 |
Neg logit contributions | |
ashed | -0.271 |
ksh | -0.270 |
tanks | -0.265 |
rods | -0.254 |
suits | -0.249 |
Token | inkle |
Token ID | 19894 |
Feature activation | +1.43e+01 |
Pos logit contributions | |
ashed | +0.130 |
ksh | +0.123 |
tanks | +0.123 |
rods | +0.114 |
sizes | +0.109 |
Neg logit contributions | |
he | -0.196 |
note | -0.185 |
answer | -0.181 |
guardian | -0.181 |
fairy | -0.179 |
Token | . |
Token ID | 13 |
Feature activation | +4.94e-01 |
Pos logit contributions | |
he | +0.603 |
fairy | +0.553 |
answer | +0.541 |
note | +0.530 |
guardian | +0.509 |
Neg logit contributions | |
ashed | -1.060 |
ksh | -1.047 |
tanks | -1.046 |
rods | -1.024 |
trucks | -0.984 |
Token | She |
Token ID | 1375 |
Feature activation | +4.83e+00 |
Pos logit contributions | |
he | +0.020 |
answer | +0.017 |
note | +0.016 |
guardian | +0.016 |
fairy | +0.016 |
Neg logit contributions | |
tanks | -0.039 |
ashed | -0.038 |
ksh | -0.037 |
rods | -0.037 |
trucks | -0.036 |
Token | kept |
Token ID | 4030 |
Feature activation | -4.05e+00 |
Pos logit contributions | |
he | +0.262 |
note | +0.236 |
answer | +0.233 |
guardian | +0.229 |
fairy | +0.226 |
Neg logit contributions | |
tanks | -0.300 |
ashed | -0.299 |
ksh | -0.289 |
rods | -0.282 |
trucks | -0.273 |
Token | the |
Token ID | 262 |
Feature activation | -2.62e-01 |
Pos logit contributions | |
tanks | +0.240 |
ashed | +0.231 |
pic | +0.226 |
rods | +0.224 |
trucks | +0.223 |
Neg logit contributions | |
he | -0.240 |
ibrarian | -0.204 |
note | -0.203 |
answer | -0.200 |
guardian | -0.196 |
Token | dictionary |
Token ID | 22155 |
Feature activation | +4.12e+00 |
Pos logit contributions | |
trucks | +0.015 |
tanks | +0.015 |
sizes | +0.014 |
rods | +0.014 |
mouths | +0.014 |
Neg logit contributions | |
he | -0.014 |
ibrarian | -0.014 |
ad | -0.013 |
iver | -0.012 |
ocket | -0.012 |
Token | to |
Token ID | 284 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
tanks | +0.186 |
ashed | +0.185 |
ksh | +0.176 |
rods | +0.173 |
trucks | +0.165 |
Neg logit contributions | |
he | -0.252 |
note | -0.227 |
answer | -0.225 |
guardian | -0.222 |
ibrarian | -0.220 |
Token | spark |
Token ID | 9009 |
Feature activation | +7.18e+00 |
Pos logit contributions | |
tanks | +0.135 |
sizes | +0.130 |
trucks | +0.128 |
ksh | +0.122 |
rods | +0.119 |
Neg logit contributions | |
ibrarian | -0.109 |
he | -0.105 |
ocket | -0.098 |
orn | -0.094 |
guardian | -0.093 |
Token | le |
Token ID | 293 |
Feature activation | +9.78e+00 |
Pos logit contributions | |
he | +0.253 |
answer | +0.213 |
note | +0.210 |
fairy | +0.203 |
guardian | +0.193 |
Neg logit contributions | |
ashed | -0.585 |
tanks | -0.583 |
ksh | -0.568 |
rods | -0.562 |
pic | -0.548 |
Token | and |
Token ID | 290 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
he | +0.436 |
note | +0.392 |
answer | +0.389 |
fairy | +0.380 |
ad | +0.369 |
Neg logit contributions | |
ashed | -0.695 |
ksh | -0.685 |
tanks | -0.684 |
rods | -0.654 |
pic | -0.639 |
Token | tw |
Token ID | 665 |
Feature activation | -6.02e-01 |
Pos logit contributions | |
he | +0.053 |
note | +0.049 |
answer | +0.049 |
fairy | +0.048 |
guardian | +0.047 |
Neg logit contributions | |
ashed | -0.102 |
ksh | -0.100 |
tanks | -0.100 |
rods | -0.096 |
pic | -0.093 |
Token | inkle |
Token ID | 19894 |
Feature activation | +1.43e+01 |
Pos logit contributions | |
ashed | +0.040 |
tanks | +0.038 |
rods | +0.038 |
ksh | +0.037 |
mouths | +0.036 |
Neg logit contributions | |
he | -0.030 |
answer | -0.028 |
note | -0.028 |
guardian | -0.026 |
fairy | -0.026 |
Token | like |
Token ID | 588 |
Feature activation | +3.40e+00 |
Pos logit contributions | |
he | +0.559 |
fairy | +0.505 |
answer | +0.494 |
note | +0.474 |
voice | +0.461 |
Neg logit contributions | |
ashed | -1.098 |
tanks | -1.082 |
rods | -1.063 |
ksh | -1.063 |
pic | -1.060 |
Token | stars |
Token ID | 5788 |
Feature activation | +2.25e+00 |
Pos logit contributions | |
fairy | +0.128 |
he | +0.128 |
answer | +0.127 |
ad | +0.126 |
note | +0.125 |
Neg logit contributions | |
ashed | -0.265 |
ksh | -0.261 |
suits | -0.248 |
tanks | -0.247 |
rods | -0.238 |
Token | in |
Token ID | 287 |
Feature activation | +7.48e-01 |
Pos logit contributions | |
he | +0.104 |
note | +0.092 |
answer | +0.091 |
guardian | +0.089 |
fairy | +0.087 |
Neg logit contributions | |
tanks | -0.157 |
ashed | -0.157 |
ksh | -0.153 |
rods | -0.149 |
trucks | -0.145 |
Token | the |
Token ID | 262 |
Feature activation | -2.72e-01 |
Pos logit contributions | |
he | +0.040 |
note | +0.036 |
answer | +0.036 |
guardian | +0.035 |
ibrarian | +0.035 |
Neg logit contributions | |
tanks | -0.047 |
ashed | -0.046 |
ksh | -0.044 |
rods | -0.044 |
trucks | -0.043 |
Token | night |
Token ID | 1755 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
tanks | +0.017 |
ashed | +0.016 |
ksh | +0.016 |
rods | +0.016 |
trucks | +0.016 |
Neg logit contributions | |
he | -0.015 |
ibrarian | -0.013 |
note | -0.013 |
answer | -0.013 |
guardian | -0.012 |
Token | But |
Token ID | 887 |
Feature activation | -7.99e-01 |
Pos logit contributions | |
he | +0.025 |
answer | +0.023 |
note | +0.023 |
voice | +0.022 |
guardian | +0.022 |
Neg logit contributions | |
tanks | -0.025 |
ashed | -0.025 |
ksh | -0.024 |
rods | -0.024 |
trucks | -0.023 |
Token | then |
Token ID | 788 |
Feature activation | -4.68e+00 |
Pos logit contributions | |
tanks | +0.048 |
ksh | +0.047 |
ashed | +0.047 |
rods | +0.046 |
sizes | +0.044 |
Neg logit contributions | |
he | -0.047 |
answer | -0.043 |
note | -0.042 |
fairy | -0.042 |
guardian | -0.042 |
Token | , |
Token ID | 11 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
ashed | +0.263 |
tanks | +0.263 |
ksh | +0.247 |
rods | +0.244 |
trucks | +0.237 |
Neg logit contributions | |
he | -0.283 |
note | -0.252 |
answer | -0.251 |
guardian | -0.247 |
ibrarian | -0.240 |
Token | a |
Token ID | 257 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
ashed | +0.086 |
tanks | +0.085 |
ksh | +0.080 |
rods | +0.079 |
trucks | +0.077 |
Neg logit contributions | |
he | -0.069 |
note | -0.061 |
guardian | -0.060 |
answer | -0.060 |
ibrarian | -0.059 |
Token | tw |
Token ID | 665 |
Feature activation | -6.02e-01 |
Pos logit contributions | |
he | +0.059 |
ibrarian | +0.048 |
iver | +0.048 |
ad | +0.047 |
ocket | +0.047 |
Neg logit contributions | |
tanks | -0.075 |
ashed | -0.070 |
rods | -0.068 |
pic | -0.067 |
ksh | -0.067 |
Token | inkle |
Token ID | 19894 |
Feature activation | +1.43e+01 |
Pos logit contributions | |
ashed | +0.033 |
ashing | +0.026 |
iked | +0.026 |
igs | +0.025 |
ug | +0.024 |
Neg logit contributions | |
answer | -0.032 |
he | -0.029 |
note | -0.028 |
reply | -0.028 |
source | -0.028 |
Token | of |
Token ID | 286 |
Feature activation | +3.87e+00 |
Pos logit contributions | |
he | +0.749 |
note | +0.703 |
guardian | +0.694 |
answer | +0.694 |
ibrarian | +0.667 |
Neg logit contributions | |
ashed | -0.892 |
tanks | -0.862 |
ksh | -0.806 |
rods | -0.797 |
trucks | -0.776 |
Token | light |
Token ID | 1657 |
Feature activation | +5.47e+00 |
Pos logit contributions | |
he | +0.223 |
ibrarian | +0.198 |
note | +0.196 |
iver | +0.192 |
guardian | +0.190 |
Neg logit contributions | |
tanks | -0.222 |
ashed | -0.207 |
trucks | -0.205 |
mouths | -0.204 |
tails | -0.202 |
Token | appeared |
Token ID | 4120 |
Feature activation | +5.90e+00 |
Pos logit contributions | |
he | +0.291 |
answer | +0.259 |
note | +0.259 |
guardian | +0.259 |
ibrarian | +0.255 |
Neg logit contributions | |
ashed | -0.339 |
tanks | -0.336 |
ksh | -0.316 |
rods | -0.313 |
trucks | -0.307 |
Token | and |
Token ID | 290 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
he | +0.277 |
note | +0.245 |
answer | +0.242 |
guardian | +0.237 |
fairy | +0.232 |
Neg logit contributions | |
tanks | -0.408 |
ashed | -0.407 |
ksh | -0.396 |
rods | -0.386 |
trucks | -0.375 |
Token | grew |
Token ID | 6348 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
he | +0.068 |
note | +0.061 |
answer | +0.060 |
guardian | +0.059 |
fairy | +0.058 |
Neg logit contributions | |
tanks | -0.095 |
ashed | -0.094 |
ksh | -0.092 |
rods | -0.090 |
trucks | -0.087 |
Token | little |
Token ID | 1310 |
Feature activation | -3.02e+00 |
Pos logit contributions | |
tanks | +0.064 |
ashed | +0.060 |
they | +0.058 |
ksh | +0.058 |
Finn | +0.057 |
Neg logit contributions | |
he | -0.092 |
iver | -0.087 |
ad | -0.081 |
handed | -0.080 |
note | -0.078 |
Token | girl |
Token ID | 2576 |
Feature activation | +2.20e+00 |
Pos logit contributions | |
tanks | +0.236 |
ashed | +0.235 |
ksh | +0.229 |
rods | +0.225 |
trucks | +0.219 |
Neg logit contributions | |
he | -0.117 |
note | -0.094 |
answer | -0.092 |
ibrarian | -0.090 |
guardian | -0.090 |
Token | ," |
Token ID | 553 |
Feature activation | +2.27e+00 |
Pos logit contributions | |
he | +0.115 |
answer | +0.101 |
note | +0.101 |
guardian | +0.097 |
ibrarian | +0.095 |
Neg logit contributions | |
tanks | -0.141 |
ashed | -0.139 |
ksh | -0.135 |
rods | -0.131 |
trucks | -0.130 |
Token | said |
Token ID | 531 |
Feature activation | +4.53e+00 |
Pos logit contributions | |
he | +0.120 |
guardian | +0.107 |
note | +0.102 |
answer | +0.102 |
iver | +0.101 |
Neg logit contributions | |
tanks | -0.143 |
ashed | -0.137 |
ksh | -0.134 |
rods | -0.133 |
trucks | -0.132 |
Token | Tw |
Token ID | 1815 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
he | +0.280 |
note | +0.253 |
answer | +0.250 |
guardian | +0.248 |
fairy | +0.241 |
Neg logit contributions | |
tanks | -0.246 |
ashed | -0.245 |
ksh | -0.237 |
rods | -0.229 |
trucks | -0.221 |
Token | inkle |
Token ID | 19894 |
Feature activation | +1.43e+01 |
Pos logit contributions | |
he | +0.055 |
answer | +0.046 |
note | +0.046 |
guardian | +0.046 |
reply | +0.041 |
Neg logit contributions | |
ashed | -0.121 |
ksh | -0.113 |
tanks | -0.111 |
pic | -0.110 |
rods | -0.109 |
Token | . |
Token ID | 13 |
Feature activation | +4.72e-01 |
Pos logit contributions | |
he | +0.791 |
note | +0.715 |
answer | +0.711 |
guardian | +0.704 |
ibrarian | +0.677 |
Neg logit contributions | |
ashed | -0.863 |
tanks | -0.856 |
ksh | -0.831 |
rods | -0.790 |
trucks | -0.774 |
Token | " |
Token ID | 366 |
Feature activation | -5.72e+00 |
Pos logit contributions | |
he | +0.032 |
answer | +0.029 |
note | +0.029 |
fairy | +0.029 |
guardian | +0.029 |
Neg logit contributions | |
tanks | -0.023 |
ashed | -0.023 |
ksh | -0.022 |
rods | -0.022 |
trucks | -0.021 |
Token | Do |
Token ID | 5211 |
Feature activation | -4.92e+00 |
Pos logit contributions | |
ksh | +0.278 |
ashed | +0.268 |
tanks | +0.257 |
sizes | +0.243 |
pic | +0.233 |
Neg logit contributions | |
ibrarian | -0.348 |
he | -0.347 |
answer | -0.345 |
note | -0.344 |
inkle | -0.341 |
Token | you |
Token ID | 345 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
tanks | +0.289 |
ashed | +0.287 |
ksh | +0.277 |
rods | +0.272 |
trucks | +0.264 |
Neg logit contributions | |
he | -0.284 |
note | -0.253 |
answer | -0.252 |
guardian | -0.249 |
fairy | -0.244 |
Token | want |
Token ID | 765 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
ashed | +0.065 |
tanks | +0.064 |
sizes | +0.057 |
tails | +0.056 |
ksh | +0.055 |
Neg logit contributions | |
he | -0.091 |
note | -0.085 |
ibrarian | -0.083 |
guardian | -0.081 |
oma | -0.079 |
Token | Toby |
Token ID | 37578 |
Feature activation | -2.30e+00 |
Pos logit contributions | |
ashed | +0.027 |
ksh | +0.026 |
tanks | +0.026 |
rods | +0.025 |
mouths | +0.024 |
Neg logit contributions | |
he | -0.029 |
answer | -0.026 |
note | -0.026 |
ad | -0.025 |
fairy | -0.025 |
Token | with |
Token ID | 351 |
Feature activation | +3.42e+00 |
Pos logit contributions | |
tanks | +0.155 |
ashed | +0.152 |
ksh | +0.149 |
rods | +0.146 |
trucks | +0.142 |
Neg logit contributions | |
he | -0.112 |
note | -0.099 |
guardian | -0.097 |
answer | -0.097 |
fairy | -0.096 |
Token | a |
Token ID | 257 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
he | +0.186 |
note | +0.166 |
ibrarian | +0.163 |
fairy | +0.162 |
guardian | +0.160 |
Neg logit contributions | |
tanks | -0.214 |
ashed | -0.210 |
rods | -0.203 |
mouths | -0.198 |
ksh | -0.197 |
Token | naughty |
Token ID | 45128 |
Feature activation | -9.22e-01 |
Pos logit contributions | |
he | +0.055 |
ibrarian | +0.049 |
ad | +0.047 |
ocket | +0.047 |
mother | +0.046 |
Neg logit contributions | |
ashed | -0.073 |
tanks | -0.072 |
rods | -0.070 |
mouths | -0.068 |
ksh | -0.067 |
Token | tw |
Token ID | 665 |
Feature activation | -6.31e-01 |
Pos logit contributions | |
tanks | +0.069 |
ashed | +0.069 |
ksh | +0.065 |
rods | +0.065 |
trucks | +0.064 |
Neg logit contributions | |
he | -0.040 |
ibrarian | -0.032 |
note | -0.032 |
guardian | -0.031 |
fairy | -0.031 |
Token | inkle |
Token ID | 19894 |
Feature activation | +1.43e+01 |
Pos logit contributions | |
ashed | +0.031 |
rods | +0.027 |
tanks | +0.025 |
mouths | +0.025 |
suits | +0.024 |
Neg logit contributions | |
he | -0.042 |
fairy | -0.040 |
answer | -0.040 |
note | -0.040 |
guardian | -0.039 |
Token | in |
Token ID | 287 |
Feature activation | +6.97e-01 |
Pos logit contributions | |
he | +0.628 |
answer | +0.552 |
note | +0.535 |
fairy | +0.510 |
ad | +0.508 |
Neg logit contributions | |
tanks | -1.039 |
ksh | -1.033 |
ashed | -1.025 |
pic | -0.992 |
rods | -0.988 |
Token | his |
Token ID | 465 |
Feature activation | +4.46e-01 |
Pos logit contributions | |
he | +0.038 |
note | +0.034 |
answer | +0.034 |
guardian | +0.033 |
fairy | +0.033 |
Neg logit contributions | |
tanks | -0.044 |
ashed | -0.043 |
ksh | -0.042 |
rods | -0.041 |
trucks | -0.040 |
Token | eye |
Token ID | 4151 |
Feature activation | +9.75e+00 |
Pos logit contributions | |
he | +0.025 |
ibrarian | +0.021 |
note | +0.021 |
fairy | +0.021 |
guardian | +0.020 |
Neg logit contributions | |
ashed | -0.028 |
tanks | -0.028 |
mouths | -0.028 |
rods | -0.027 |
ksh | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
he | +0.445 |
note | +0.393 |
answer | +0.389 |
guardian | +0.376 |
fairy | +0.372 |
Neg logit contributions | |
tanks | -0.689 |
ashed | -0.688 |
ksh | -0.669 |
rods | -0.653 |
trucks | -0.633 |
Token | said |
Token ID | 531 |
Feature activation | +4.56e+00 |
Pos logit contributions | |
he | +0.054 |
answer | +0.050 |
note | +0.049 |
fairy | +0.048 |
figure | +0.048 |
Neg logit contributions | |
ksh | -0.100 |
tanks | -0.100 |
ashed | -0.099 |
rods | -0.096 |
pic | -0.095 |
Token | they |
Token ID | 484 |
Feature activation | -6.78e+00 |
Pos logit contributions | |
ksh | +0.511 |
tanks | +0.495 |
ashed | +0.473 |
rods | +0.466 |
trucks | +0.460 |
Neg logit contributions | |
he | -0.528 |
answer | -0.471 |
note | -0.457 |
fairy | -0.452 |
ad | -0.445 |
Token | watched |
Token ID | 7342 |
Feature activation | -5.53e+00 |
Pos logit contributions | |
tanks | +0.375 |
ksh | +0.365 |
ashed | +0.361 |
rods | +0.355 |
trucks | +0.335 |
Neg logit contributions | |
he | -0.429 |
answer | -0.380 |
note | -0.380 |
fairy | -0.373 |
guardian | -0.371 |
Token | the |
Token ID | 262 |
Feature activation | -3.98e-01 |
Pos logit contributions | |
ashed | +0.411 |
ksh | +0.389 |
tanks | +0.386 |
suits | +0.377 |
rods | +0.373 |
Neg logit contributions | |
he | -0.251 |
answer | -0.232 |
fairy | -0.214 |
ad | -0.214 |
note | -0.213 |
Token | stars |
Token ID | 5788 |
Feature activation | +2.16e+00 |
Pos logit contributions | |
tanks | +0.023 |
ashed | +0.022 |
ksh | +0.022 |
trucks | +0.022 |
sizes | +0.021 |
Neg logit contributions | |
he | -0.023 |
note | -0.020 |
guardian | -0.020 |
ibrarian | -0.020 |
answer | -0.020 |
Token | tw |
Token ID | 665 |
Feature activation | -6.65e-01 |
Pos logit contributions | |
he | +0.109 |
note | +0.097 |
answer | +0.096 |
guardian | +0.094 |
fairy | +0.093 |
Neg logit contributions | |
tanks | -0.142 |
ashed | -0.141 |
ksh | -0.137 |
rods | -0.134 |
trucks | -0.130 |
Token | inkle |
Token ID | 19894 |
Feature activation | +1.43e+01 |
Pos logit contributions | |
ashed | +0.054 |
tanks | +0.053 |
ksh | +0.051 |
rods | +0.051 |
pic | +0.050 |
Neg logit contributions | |
he | -0.023 |
note | -0.022 |
answer | -0.021 |
guardian | -0.020 |
fairy | -0.019 |
Token | in |
Token ID | 287 |
Feature activation | +6.83e-01 |
Pos logit contributions | |
he | +0.531 |
fairy | +0.452 |
answer | +0.443 |
note | +0.411 |
voice | +0.401 |
Neg logit contributions | |
ashed | -1.135 |
tanks | -1.119 |
rods | -1.092 |
ksh | -1.077 |
trucks | -1.056 |
Token | the |
Token ID | 262 |
Feature activation | -3.53e-01 |
Pos logit contributions | |
he | +0.037 |
note | +0.033 |
answer | +0.033 |
guardian | +0.032 |
fairy | +0.031 |
Neg logit contributions | |
ashed | -0.043 |
tanks | -0.043 |
ksh | -0.042 |
rods | -0.041 |
trucks | -0.039 |
Token | night |
Token ID | 1755 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
tanks | +0.020 |
ashed | +0.020 |
ksh | +0.019 |
mouths | +0.019 |
trucks | +0.018 |
Neg logit contributions | |
he | -0.020 |
note | -0.018 |
ibrarian | -0.018 |
guardian | -0.018 |
answer | -0.017 |
Token | sky |
Token ID | 6766 |
Feature activation | +5.60e+00 |
Pos logit contributions | |
he | +0.069 |
note | +0.062 |
answer | +0.061 |
guardian | +0.060 |
fairy | +0.058 |
Neg logit contributions | |
tanks | -0.108 |
ashed | -0.107 |
ksh | -0.105 |
rods | -0.102 |
trucks | -0.099 |
Token | . |
Token ID | 13 |
Feature activation | +4.54e-01 |
Pos logit contributions | |
he | +0.215 |
answer | +0.181 |
note | +0.176 |
fairy | +0.174 |
guardian | +0.166 |
Neg logit contributions | |
ashed | -0.447 |
tanks | -0.447 |
ksh | -0.426 |
rods | -0.425 |
trucks | -0.414 |
Token | en |
Token ID | 551 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
tanks | +0.176 |
rods | +0.169 |
mouths | +0.166 |
ashed | +0.166 |
they | +0.163 |
Neg logit contributions | |
he | -0.186 |
mother | -0.162 |
iver | -0.161 |
ibrarian | -0.160 |
mirror | -0.156 |
Token | vious |
Token ID | 1442 |
Feature activation | +2.87e+00 |
Pos logit contributions | |
guardian | +0.110 |
note | +0.109 |
gen | +0.107 |
he | +0.106 |
ibrarian | +0.105 |
Neg logit contributions | |
ashed | -0.174 |
ksh | -0.160 |
tanks | -0.157 |
ashing | -0.145 |
rods | -0.144 |
Token | of |
Token ID | 286 |
Feature activation | +3.97e+00 |
Pos logit contributions | |
he | +0.096 |
answer | +0.076 |
fairy | +0.075 |
note | +0.074 |
guardian | +0.072 |
Neg logit contributions | |
ksh | -0.243 |
ashed | -0.242 |
tanks | -0.242 |
rods | -0.234 |
mouths | -0.229 |
Token | what |
Token ID | 644 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
he | +0.184 |
answer | +0.169 |
note | +0.162 |
fairy | +0.162 |
ad | +0.161 |
Neg logit contributions | |
ashed | -0.290 |
ksh | -0.278 |
tanks | -0.276 |
suits | -0.266 |
rods | -0.264 |
Token | the |
Token ID | 262 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
he | +0.089 |
note | +0.088 |
answer | +0.087 |
fairy | +0.086 |
wizard | +0.080 |
Neg logit contributions | |
ksh | -0.137 |
ashed | -0.137 |
tanks | -0.129 |
suits | -0.125 |
rods | -0.124 |
Token | gen |
Token ID | 2429 |
Feature activation | +1.43e+01 |
Pos logit contributions | |
tanks | +0.011 |
mouths | +0.010 |
trucks | +0.010 |
rods | +0.010 |
sizes | +0.010 |
Neg logit contributions | |
he | -0.009 |
ibrarian | -0.008 |
iver | -0.008 |
inkle | -0.007 |
alla | -0.006 |
Token | ie |
Token ID | 494 |
Feature activation | +5.01e+00 |
Pos logit contributions | |
he | +0.758 |
guardian | +0.729 |
answer | +0.716 |
note | +0.697 |
voice | +0.685 |
Neg logit contributions | |
ashed | -0.860 |
tanks | -0.860 |
ksh | -0.796 |
sizes | -0.793 |
trucks | -0.782 |
Token | could |
Token ID | 714 |
Feature activation | -7.71e+00 |
Pos logit contributions | |
he | +0.311 |
note | +0.281 |
guardian | +0.277 |
answer | +0.276 |
fairy | +0.266 |
Neg logit contributions | |
tanks | -0.272 |
ashed | -0.268 |
ksh | -0.251 |
rods | -0.250 |
trucks | -0.243 |
Token | do |
Token ID | 466 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
they | +0.374 |
tanks | +0.320 |
trucks | +0.315 |
rods | +0.314 |
sizes | +0.313 |
Neg logit contributions | |
ibrarian | -0.446 |
mirror | -0.414 |
guardian | -0.409 |
ocket | -0.406 |
orn | -0.401 |
Token | that |
Token ID | 326 |
Feature activation | +6.51e+00 |
Pos logit contributions | |
he | +0.067 |
note | +0.059 |
answer | +0.058 |
guardian | +0.056 |
fairy | +0.056 |
Neg logit contributions | |
tanks | -0.098 |
ashed | -0.098 |
ksh | -0.096 |
rods | -0.093 |
mouths | -0.091 |
Token | it |
Token ID | 340 |
Feature activation | +3.33e+00 |
Pos logit contributions | |
he | +0.319 |
note | +0.280 |
guardian | +0.277 |
answer | +0.275 |
ibrarian | +0.269 |
Neg logit contributions | |
tanks | -0.436 |
ashed | -0.428 |
ksh | -0.419 |
rods | -0.410 |
trucks | -0.402 |
Token | Hello |
Token ID | 15496 |
Feature activation | +2.99e+00 |
Pos logit contributions | |
ksh | +0.322 |
ashed | +0.315 |
tanks | +0.306 |
rods | +0.291 |
sizes | +0.285 |
Neg logit contributions | |
he | -0.320 |
answer | -0.302 |
ibrarian | -0.296 |
note | -0.295 |
guardian | -0.295 |
Token | , |
Token ID | 11 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
fairy | +0.112 |
he | +0.110 |
answer | +0.108 |
note | +0.106 |
voice | +0.101 |
Neg logit contributions | |
ksh | -0.233 |
ashed | -0.233 |
rods | -0.223 |
tanks | -0.223 |
mouths | -0.221 |
Token | I |
Token ID | 314 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
tanks | +0.064 |
ashed | +0.062 |
ksh | +0.060 |
rods | +0.058 |
trucks | +0.057 |
Neg logit contributions | |
he | -0.078 |
note | -0.071 |
answer | -0.070 |
guardian | -0.069 |
fairy | -0.068 |
Token | am |
Token ID | 716 |
Feature activation | +3.76e+00 |
Pos logit contributions | |
ashed | +0.103 |
tanks | +0.092 |
pic | +0.090 |
suits | +0.087 |
planned | +0.084 |
Neg logit contributions | |
angel | -0.104 |
he | -0.103 |
ibrarian | -0.097 |
note | -0.095 |
owl | -0.095 |
Token | the |
Token ID | 262 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
he | +0.225 |
note | +0.205 |
answer | +0.203 |
guardian | +0.200 |
fairy | +0.198 |
Neg logit contributions | |
tanks | -0.213 |
ashed | -0.213 |
ksh | -0.205 |
rods | -0.199 |
trucks | -0.192 |
Token | gen |
Token ID | 2429 |
Feature activation | +1.43e+01 |
Pos logit contributions | |
tanks | +0.012 |
trucks | +0.012 |
rods | +0.011 |
sizes | +0.011 |
mouths | +0.011 |
Neg logit contributions | |
he | -0.008 |
ibrarian | -0.006 |
iver | -0.006 |
inkle | -0.006 |
ocket | -0.005 |
Token | ie |
Token ID | 494 |
Feature activation | +5.01e+00 |
Pos logit contributions | |
he | +0.681 |
answer | +0.608 |
note | +0.604 |
guardian | +0.601 |
ad | +0.567 |
Neg logit contributions | |
tanks | -0.971 |
ashed | -0.959 |
ksh | -0.937 |
rods | -0.906 |
trucks | -0.893 |
Token | of |
Token ID | 286 |
Feature activation | +4.01e+00 |
Pos logit contributions | |
he | +0.232 |
note | +0.204 |
answer | +0.204 |
fairy | +0.201 |
guardian | +0.198 |
Neg logit contributions | |
ashed | -0.356 |
tanks | -0.351 |
ksh | -0.341 |
rods | -0.338 |
mouths | -0.327 |
Token | the |
Token ID | 262 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
he | +0.210 |
note | +0.189 |
answer | +0.187 |
guardian | +0.183 |
fairy | +0.181 |
Neg logit contributions | |
ashed | -0.258 |
tanks | -0.257 |
ksh | -0.250 |
rods | -0.243 |
trucks | -0.235 |
Token | bottle |
Token ID | 9294 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
tanks | +0.010 |
rods | +0.009 |
mouths | +0.009 |
trucks | +0.009 |
ashed | +0.009 |
Neg logit contributions | |
he | -0.007 |
ibrarian | -0.006 |
iver | -0.006 |
ad | -0.006 |
answer | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +6.69e-01 |
Pos logit contributions | |
he | +0.094 |
note | +0.085 |
guardian | +0.084 |
answer | +0.084 |
ad | +0.081 |
Neg logit contributions | |
tanks | -0.117 |
ashed | -0.114 |
ksh | -0.110 |
rods | -0.108 |
trucks | -0.107 |
Token | â |
Token ID | 22940 |
Feature activation | +8.08e-01 |
Pos logit contributions | |
he | +0.125 |
note | +0.112 |
answer | +0.111 |
guardian | +0.109 |
fairy | +0.107 |
Neg logit contributions | |
tanks | -0.154 |
ashed | -0.153 |
ksh | -0.150 |
rods | -0.145 |
trucks | -0.140 |
Token | € |
Token ID | 26391 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
he | +0.043 |
guardian | +0.042 |
note | +0.042 |
fairy | +0.042 |
answer | +0.041 |
Neg logit contributions | |
ashed | -0.046 |
ksh | -0.045 |
tanks | -0.042 |
suits | -0.040 |
appet | -0.040 |
Token | With |
Token ID | 2080 |
Feature activation | +7.38e-01 |
Pos logit contributions | |
tanks | +0.068 |
ashed | +0.068 |
ksh | +0.066 |
rods | +0.065 |
mouths | +0.062 |
Neg logit contributions | |
he | -0.058 |
note | -0.052 |
answer | -0.050 |
guardian | -0.050 |
fairy | -0.049 |
Token | a |
Token ID | 257 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
he | +0.038 |
note | +0.034 |
answer | +0.033 |
guardian | +0.033 |
fairy | +0.032 |
Neg logit contributions | |
tanks | -0.048 |
ashed | -0.048 |
ksh | -0.046 |
rods | -0.045 |
mouths | -0.044 |
Token | tw |
Token ID | 665 |
Feature activation | -6.63e-01 |
Pos logit contributions | |
he | +0.056 |
ad | +0.053 |
ibrarian | +0.051 |
iver | +0.051 |
mother | +0.048 |
Neg logit contributions | |
ashed | -0.065 |
tanks | -0.061 |
rods | -0.060 |
suits | -0.059 |
pic | -0.058 |
Token | inkle |
Token ID | 19894 |
Feature activation | +1.43e+01 |
Pos logit contributions | |
ashed | +0.047 |
rods | +0.040 |
appet | +0.040 |
pic | +0.039 |
mouths | +0.038 |
Neg logit contributions | |
he | -0.029 |
owl | -0.025 |
answer | -0.025 |
angel | -0.025 |
guardian | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
he | +0.721 |
note | +0.643 |
answer | +0.637 |
guardian | +0.623 |
fairy | +0.615 |
Neg logit contributions | |
tanks | -0.938 |
ashed | -0.935 |
ksh | -0.909 |
rods | -0.887 |
trucks | -0.858 |
Token | the |
Token ID | 262 |
Feature activation | -3.48e-01 |
Pos logit contributions | |
tanks | +0.086 |
ashed | +0.085 |
ksh | +0.083 |
rods | +0.081 |
pic | +0.079 |
Neg logit contributions | |
he | -0.074 |
note | -0.066 |
answer | -0.065 |
guardian | -0.064 |
ibrarian | -0.063 |
Token | fairy |
Token ID | 25607 |
Feature activation | +1.26e+01 |
Pos logit contributions | |
tanks | +0.024 |
ashed | +0.023 |
sizes | +0.022 |
mouths | +0.022 |
rods | +0.022 |
Neg logit contributions | |
he | -0.016 |
ibrarian | -0.013 |
iver | -0.013 |
ad | -0.013 |
guardian | -0.011 |
Token | disappeared |
Token ID | 12120 |
Feature activation | +2.55e+00 |
Pos logit contributions | |
he | +0.694 |
note | +0.613 |
answer | +0.608 |
guardian | +0.590 |
ad | +0.578 |
Neg logit contributions | |
ashed | -0.767 |
tanks | -0.763 |
ksh | -0.725 |
rods | -0.713 |
mouths | -0.694 |
Token | . |
Token ID | 13 |
Feature activation | +4.43e-01 |
Pos logit contributions | |
he | +0.126 |
note | +0.111 |
answer | +0.110 |
guardian | +0.108 |
fairy | +0.106 |
Neg logit contributions | |
tanks | -0.170 |
ashed | -0.170 |
ksh | -0.165 |
rods | -0.161 |
trucks | -0.156 |
Token | family |
Token ID | 1641 |
Feature activation | +9.71e-01 |
Pos logit contributions | |
he | +0.022 |
note | +0.019 |
answer | +0.019 |
guardian | +0.019 |
ibrarian | +0.018 |
Neg logit contributions | |
tanks | -0.027 |
ashed | -0.027 |
ksh | -0.026 |
rods | -0.026 |
trucks | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | +5.45e-01 |
Pos logit contributions | |
he | +0.037 |
answer | +0.034 |
note | +0.034 |
fairy | +0.032 |
guardian | +0.031 |
Neg logit contributions | |
ashed | -0.076 |
ksh | -0.076 |
tanks | -0.075 |
rods | -0.073 |
mouths | -0.071 |
Token |
|
Token ID | 198 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
he | +0.021 |
ad | +0.016 |
answer | +0.015 |
note | +0.015 |
voice | +0.015 |
Neg logit contributions | |
ksh | -0.042 |
ashed | -0.042 |
tanks | -0.042 |
sizes | -0.040 |
pic | -0.040 |
Token |
|
Token ID | 198 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
guardian | +0.147 |
angel | +0.141 |
fairy | +0.140 |
wizard | +0.135 |
note | +0.134 |
Neg logit contributions | |
rods | -0.123 |
tanks | -0.122 |
ashed | -0.119 |
pic | -0.115 |
they | -0.112 |
Token | The |
Token ID | 464 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
he | +0.129 |
note | +0.111 |
answer | +0.108 |
ad | +0.105 |
ibrarian | +0.104 |
Neg logit contributions | |
tanks | -0.150 |
ashed | -0.149 |
ksh | -0.147 |
mouths | -0.139 |
rods | -0.138 |
Token | gen |
Token ID | 2429 |
Feature activation | +1.42e+01 |
Pos logit contributions | |
he | +0.061 |
iver | +0.044 |
ibrarian | +0.039 |
orn | +0.037 |
inkle | +0.037 |
Neg logit contributions | |
tanks | -0.145 |
trucks | -0.139 |
rods | -0.137 |
mouths | -0.137 |
ashed | -0.133 |
Token | ie |
Token ID | 494 |
Feature activation | +4.96e+00 |
Pos logit contributions | |
he | +0.668 |
answer | +0.600 |
guardian | +0.595 |
note | +0.595 |
fairy | +0.562 |
Neg logit contributions | |
ashed | -0.981 |
tanks | -0.971 |
ksh | -0.929 |
rods | -0.915 |
trucks | -0.895 |
Token | nodded |
Token ID | 14464 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
he | +0.240 |
fairy | +0.224 |
note | +0.224 |
answer | +0.222 |
guardian | +0.219 |
Neg logit contributions | |
tanks | -0.324 |
ksh | -0.323 |
ashed | -0.320 |
rods | -0.307 |
pic | -0.300 |
Token | and |
Token ID | 290 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
tanks | +0.075 |
ashed | +0.075 |
ksh | +0.073 |
rods | +0.071 |
trucks | +0.069 |
Neg logit contributions | |
he | -0.045 |
note | -0.040 |
answer | -0.040 |
fairy | -0.038 |
guardian | -0.038 |
Token | said |
Token ID | 531 |
Feature activation | +4.70e+00 |
Pos logit contributions | |
he | +0.062 |
fairy | +0.062 |
note | +0.060 |
answer | +0.060 |
wizard | +0.057 |
Neg logit contributions | |
ksh | -0.110 |
ashed | -0.106 |
tanks | -0.105 |
sizes | -0.100 |
suits | -0.099 |
Token | he |
Token ID | 339 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
fairy | +0.231 |
answer | +0.224 |
note | +0.223 |
wizard | +0.216 |
ad | +0.213 |
Neg logit contributions | |
ksh | -0.339 |
ashed | -0.329 |
suits | -0.316 |
tanks | -0.311 |
ashing | -0.295 |
Token | kind |
Token ID | 1611 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
he | +0.050 |
ad | +0.042 |
ibrarian | +0.042 |
answer | +0.042 |
note | +0.040 |
Neg logit contributions | |
tanks | -0.065 |
ashed | -0.063 |
rods | -0.061 |
trucks | -0.060 |
ksh | -0.060 |
Token | heart |
Token ID | 2612 |
Feature activation | +6.99e+00 |
Pos logit contributions | |
he | +0.125 |
ibrarian | +0.103 |
note | +0.102 |
iver | +0.098 |
guardian | +0.098 |
Neg logit contributions | |
tanks | -0.178 |
ashed | -0.172 |
rods | -0.163 |
trucks | -0.160 |
sizes | -0.157 |
Token | and |
Token ID | 290 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
he | +0.349 |
note | +0.308 |
answer | +0.307 |
guardian | +0.302 |
fairy | +0.294 |
Neg logit contributions | |
tanks | -0.464 |
ashed | -0.462 |
ksh | -0.441 |
rods | -0.437 |
trucks | -0.425 |
Token | a |
Token ID | 257 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
note | +0.054 |
answer | +0.053 |
fairy | +0.053 |
he | +0.051 |
voice | +0.051 |
Neg logit contributions | |
ksh | -0.088 |
ashed | -0.083 |
tanks | -0.080 |
rods | -0.078 |
suits | -0.077 |
Token | tw |
Token ID | 665 |
Feature activation | -7.22e-01 |
Pos logit contributions | |
he | +0.053 |
ad | +0.044 |
ibrarian | +0.044 |
answer | +0.043 |
note | +0.042 |
Neg logit contributions | |
tanks | -0.067 |
ashed | -0.065 |
rods | -0.063 |
ksh | -0.062 |
mouths | -0.062 |
Token | inkle |
Token ID | 19894 |
Feature activation | +1.42e+01 |
Pos logit contributions | |
ashed | +0.037 |
tanks | +0.030 |
tails | +0.030 |
rods | +0.029 |
suits | +0.029 |
Neg logit contributions | |
he | -0.043 |
answer | -0.043 |
note | -0.041 |
reply | -0.041 |
mirror | -0.040 |
Token | in |
Token ID | 287 |
Feature activation | +5.99e-01 |
Pos logit contributions | |
he | +0.644 |
answer | +0.556 |
note | +0.554 |
fairy | +0.541 |
guardian | +0.534 |
Neg logit contributions | |
tanks | -1.017 |
ashed | -1.011 |
ksh | -1.005 |
rods | -0.984 |
pic | -0.957 |
Token | his |
Token ID | 465 |
Feature activation | +3.47e-01 |
Pos logit contributions | |
he | +0.036 |
ibrarian | +0.032 |
note | +0.032 |
guardian | +0.032 |
answer | +0.031 |
Neg logit contributions | |
tanks | -0.035 |
ashed | -0.032 |
rods | -0.032 |
trucks | -0.031 |
pic | -0.031 |
Token | eye |
Token ID | 4151 |
Feature activation | +9.65e+00 |
Pos logit contributions | |
he | +0.017 |
ibrarian | +0.014 |
answer | +0.013 |
note | +0.013 |
ad | +0.013 |
Neg logit contributions | |
tanks | -0.024 |
mouths | -0.024 |
ashed | -0.023 |
tails | -0.023 |
rods | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | +3.75e-01 |
Pos logit contributions | |
he | +0.417 |
note | +0.370 |
answer | +0.364 |
fairy | +0.351 |
guardian | +0.349 |
Neg logit contributions | |
ashed | -0.704 |
tanks | -0.702 |
ksh | -0.690 |
rods | -0.670 |
pic | -0.650 |
Token | He |
Token ID | 679 |
Feature activation | +3.35e-01 |
Pos logit contributions | |
he | +0.021 |
answer | +0.019 |
ad | +0.018 |
note | +0.017 |
guardian | +0.017 |
Neg logit contributions | |
ashed | -0.024 |
ksh | -0.022 |
tanks | -0.022 |
rods | -0.022 |
mouths | -0.021 |
Token | |
Token ID | 220 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
guardian | +0.134 |
wizard | +0.132 |
note | +0.130 |
fairy | +0.130 |
ocket | +0.130 |
Neg logit contributions | |
ashed | -0.111 |
tanks | -0.111 |
they | -0.107 |
rods | -0.104 |
pic | -0.104 |
Token |
|
Token ID | 198 |
Feature activation | +2.29e+00 |
Pos logit contributions | |
he | +0.050 |
note | +0.049 |
guardian | +0.048 |
fairy | +0.047 |
answer | +0.046 |
Neg logit contributions | |
ashed | -0.069 |
tanks | -0.067 |
ksh | -0.066 |
rods | -0.064 |
pic | -0.062 |
Token | Suddenly |
Token ID | 38582 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
he | +0.136 |
answer | +0.126 |
note | +0.125 |
guardian | +0.123 |
fairy | +0.122 |
Neg logit contributions | |
ashed | -0.127 |
tanks | -0.123 |
ksh | -0.120 |
rods | -0.114 |
pic | -0.112 |
Token | , |
Token ID | 11 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
tanks | +0.132 |
ashed | +0.131 |
ksh | +0.129 |
rods | +0.124 |
sizes | +0.119 |
Neg logit contributions | |
he | -0.119 |
note | -0.108 |
answer | -0.108 |
fairy | -0.106 |
guardian | -0.105 |
Token | a |
Token ID | 257 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
tanks | +0.088 |
ashed | +0.087 |
ksh | +0.086 |
rods | +0.084 |
sizes | +0.080 |
Neg logit contributions | |
he | -0.059 |
answer | -0.051 |
note | -0.051 |
fairy | -0.050 |
guardian | -0.049 |
Token | gen |
Token ID | 2429 |
Feature activation | +1.42e+01 |
Pos logit contributions | |
he | +0.062 |
ibrarian | +0.052 |
ad | +0.051 |
iver | +0.051 |
ocket | +0.051 |
Neg logit contributions | |
tanks | -0.085 |
ashed | -0.078 |
rods | -0.077 |
trucks | -0.076 |
ksh | -0.074 |
Token | ie |
Token ID | 494 |
Feature activation | +4.95e+00 |
Pos logit contributions | |
he | +0.805 |
guardian | +0.752 |
note | +0.740 |
answer | +0.738 |
ibrarian | +0.689 |
Neg logit contributions | |
ashed | -0.835 |
tanks | -0.798 |
ksh | -0.734 |
trucks | -0.728 |
rods | -0.709 |
Token | appeared |
Token ID | 4120 |
Feature activation | +5.98e+00 |
Pos logit contributions | |
he | +0.322 |
answer | +0.265 |
iver | +0.265 |
note | +0.261 |
ibrarian | +0.250 |
Neg logit contributions | |
tanks | -0.273 |
ashed | -0.258 |
trucks | -0.250 |
rods | -0.238 |
ksh | -0.230 |
Token | in |
Token ID | 287 |
Feature activation | +8.36e-01 |
Pos logit contributions | |
he | +0.282 |
note | +0.249 |
answer | +0.246 |
guardian | +0.241 |
fairy | +0.237 |
Neg logit contributions | |
tanks | -0.415 |
ashed | -0.414 |
ksh | -0.403 |
rods | -0.394 |
trucks | -0.382 |
Token | her |
Token ID | 607 |
Feature activation | +4.89e+00 |
Pos logit contributions | |
ibrarian | +0.054 |
ocket | +0.052 |
guardian | +0.052 |
he | +0.051 |
iver | +0.051 |
Neg logit contributions | |
tanks | -0.036 |
pic | -0.036 |
trucks | -0.033 |
rods | -0.031 |
they | -0.031 |
Token | bedroom |
Token ID | 14043 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
he | +0.263 |
answer | +0.224 |
note | +0.222 |
ad | +0.218 |
guardian | +0.214 |
Neg logit contributions | |
tanks | -0.319 |
mouths | -0.304 |
ashed | -0.303 |
trucks | -0.298 |
rods | -0.296 |
Token | fun |
Token ID | 1257 |
Feature activation | -7.97e+00 |
Pos logit contributions | |
ksh | +0.130 |
tanks | +0.129 |
ashed | +0.122 |
rods | +0.115 |
trucks | +0.113 |
Neg logit contributions | |
he | -0.211 |
answer | -0.196 |
note | -0.193 |
fairy | -0.184 |
ad | -0.182 |
Token | . |
Token ID | 13 |
Feature activation | -3.35e-01 |
Pos logit contributions | |
ksh | +0.684 |
tanks | +0.671 |
ashed | +0.656 |
suits | +0.633 |
rods | +0.628 |
Neg logit contributions | |
he | -0.275 |
note | -0.215 |
answer | -0.197 |
fairy | -0.186 |
figure | -0.183 |
Token | Lily |
Token ID | 20037 |
Feature activation | -4.96e+00 |
Pos logit contributions | |
rods | +0.020 |
suits | +0.019 |
tanks | +0.019 |
trucks | +0.018 |
girls | +0.018 |
Neg logit contributions | |
he | -0.017 |
answer | -0.014 |
mirror | -0.014 |
voice | -0.014 |
fairy | -0.014 |
Token | realized |
Token ID | 6939 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
tanks | +0.328 |
ashed | +0.324 |
ksh | +0.321 |
rods | +0.313 |
trucks | +0.300 |
Neg logit contributions | |
he | -0.249 |
note | -0.222 |
answer | -0.219 |
fairy | -0.215 |
guardian | -0.213 |
Token | that |
Token ID | 326 |
Feature activation | +5.52e+00 |
Pos logit contributions | |
he | +0.074 |
note | +0.064 |
answer | +0.064 |
guardian | +0.062 |
fairy | +0.061 |
Neg logit contributions | |
tanks | -0.120 |
ashed | -0.120 |
ksh | -0.116 |
rods | -0.114 |
trucks | -0.111 |
Token | pic |
Token ID | 8301 |
Feature activation | -1.76e+01 |
Pos logit contributions | |
he | +0.308 |
note | +0.285 |
answer | +0.281 |
fairy | +0.274 |
guardian | +0.268 |
Neg logit contributions | |
ashed | -0.339 |
ksh | -0.330 |
tanks | -0.329 |
rods | -0.315 |
suits | -0.310 |
Token | n |
Token ID | 77 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
ashed | +1.085 |
tanks | +1.083 |
ksh | +1.049 |
rods | +1.025 |
pic | +0.987 |
Neg logit contributions | |
he | -0.949 |
note | -0.858 |
answer | -0.850 |
guardian | -0.846 |
fairy | -0.830 |
Token | ics |
Token ID | 873 |
Feature activation | +9.41e-01 |
Pos logit contributions | |
ashed | +0.108 |
tanks | +0.098 |
rods | +0.094 |
ksh | +0.093 |
they | +0.092 |
Neg logit contributions | |
guardian | -0.048 |
wizard | -0.047 |
note | -0.047 |
answer | -0.046 |
fairy | -0.045 |
Token | were |
Token ID | 547 |
Feature activation | -4.38e+00 |
Pos logit contributions | |
he | +0.047 |
note | +0.043 |
answer | +0.042 |
fairy | +0.041 |
guardian | +0.039 |
Neg logit contributions | |
ksh | -0.064 |
tanks | -0.063 |
ashed | -0.063 |
rods | -0.060 |
trucks | -0.058 |
Token | not |
Token ID | 407 |
Feature activation | +4.66e+00 |
Pos logit contributions | |
ksh | +0.317 |
ashed | +0.315 |
tanks | +0.304 |
suits | +0.301 |
rods | +0.300 |
Neg logit contributions | |
answer | -0.189 |
he | -0.184 |
note | -0.180 |
fairy | -0.179 |
guardian | -0.174 |
Token | so |
Token ID | 523 |
Feature activation | -4.17e+00 |
Pos logit contributions | |
answer | +0.204 |
fairy | +0.181 |
he | +0.181 |
note | +0.178 |
owl | +0.172 |
Neg logit contributions | |
ksh | -0.336 |
ashed | -0.332 |
rods | -0.331 |
suits | -0.327 |
iked | -0.314 |
Token | family |
Token ID | 1641 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
he | +0.208 |
note | +0.191 |
answer | +0.190 |
fairy | +0.187 |
guardian | +0.186 |
Neg logit contributions | |
ashed | -0.259 |
tanks | -0.258 |
ksh | -0.258 |
rods | -0.244 |
sizes | -0.237 |
Token | were |
Token ID | 547 |
Feature activation | -4.32e+00 |
Pos logit contributions | |
he | +0.011 |
note | +0.010 |
answer | +0.010 |
fairy | +0.009 |
guardian | +0.009 |
Neg logit contributions | |
ksh | -0.009 |
tanks | -0.009 |
ashed | -0.008 |
rods | -0.008 |
sizes | -0.008 |
Token | ready |
Token ID | 3492 |
Feature activation | -5.54e+00 |
Pos logit contributions | |
ashed | +0.261 |
tanks | +0.259 |
ksh | +0.246 |
rods | +0.241 |
pic | +0.238 |
Neg logit contributions | |
he | -0.236 |
note | -0.216 |
answer | -0.216 |
guardian | -0.210 |
fairy | -0.209 |
Token | to |
Token ID | 284 |
Feature activation | -2.93e+00 |
Pos logit contributions | |
tanks | +0.319 |
ashed | +0.317 |
ksh | +0.308 |
rods | +0.296 |
trucks | +0.288 |
Neg logit contributions | |
he | -0.331 |
answer | -0.305 |
note | -0.303 |
guardian | -0.290 |
fairy | -0.288 |
Token | go |
Token ID | 467 |
Feature activation | -4.39e+00 |
Pos logit contributions | |
tanks | +0.163 |
pic | +0.159 |
rods | +0.152 |
ashed | +0.151 |
mouths | +0.147 |
Neg logit contributions | |
he | -0.168 |
note | -0.155 |
guardian | -0.154 |
ibrarian | -0.153 |
fairy | -0.148 |
Token | pic |
Token ID | 8301 |
Feature activation | -1.76e+01 |
Pos logit contributions | |
tanks | +0.265 |
ashed | +0.261 |
rods | +0.248 |
pic | +0.244 |
ksh | +0.241 |
Neg logit contributions | |
he | -0.240 |
note | -0.225 |
ibrarian | -0.222 |
guardian | -0.222 |
answer | -0.215 |
Token | n |
Token ID | 77 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
tanks | +0.995 |
ashed | +0.979 |
rods | +0.947 |
ksh | +0.937 |
pic | +0.917 |
Neg logit contributions | |
he | -1.018 |
guardian | -0.917 |
note | -0.916 |
answer | -0.909 |
fairy | -0.909 |
Token | icking |
Token ID | 7958 |
Feature activation | -2.54e+00 |
Pos logit contributions | |
ashed | +0.077 |
tanks | +0.077 |
ksh | +0.073 |
rods | +0.072 |
pic | +0.072 |
Neg logit contributions | |
he | -0.069 |
fairy | -0.068 |
guardian | -0.068 |
note | -0.067 |
answer | -0.067 |
Token | again |
Token ID | 757 |
Feature activation | -2.98e+00 |
Pos logit contributions | |
ashed | +0.188 |
tanks | +0.184 |
ksh | +0.183 |
rods | +0.176 |
trucks | +0.171 |
Neg logit contributions | |
he | -0.110 |
answer | -0.097 |
note | -0.096 |
fairy | -0.091 |
voice | -0.088 |
Token | . |
Token ID | 13 |
Feature activation | -3.50e-01 |
Pos logit contributions | |
ashed | +0.236 |
tanks | +0.230 |
ksh | +0.227 |
rods | +0.217 |
mouths | +0.215 |
Neg logit contributions | |
he | -0.118 |
answer | -0.106 |
note | -0.102 |
fairy | -0.099 |
guardian | -0.099 |
Token | But |
Token ID | 887 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
rods | +0.019 |
tanks | +0.019 |
ashed | +0.018 |
trucks | +0.017 |
rect | +0.017 |
Neg logit contributions | |
he | -0.022 |
voice | -0.019 |
answer | -0.019 |
bird | -0.018 |
guardian | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
tanks | +0.241 |
ashed | +0.234 |
sizes | +0.227 |
trucks | +0.224 |
ksh | +0.219 |
Neg logit contributions | |
he | -0.298 |
note | -0.290 |
guardian | -0.289 |
ibrarian | -0.288 |
wizard | -0.275 |
Token | She |
Token ID | 1375 |
Feature activation | +4.01e+00 |
Pos logit contributions | |
rods | +0.025 |
tanks | +0.025 |
ashed | +0.024 |
ksh | +0.024 |
trucks | +0.023 |
Neg logit contributions | |
he | -0.011 |
answer | -0.008 |
fairy | -0.008 |
mirror | -0.007 |
note | -0.007 |
Token | loved |
Token ID | 6151 |
Feature activation | -7.62e-01 |
Pos logit contributions | |
he | +0.272 |
note | +0.251 |
answer | +0.251 |
guardian | +0.249 |
fairy | +0.240 |
Neg logit contributions | |
ashed | -0.197 |
tanks | -0.194 |
ksh | -0.181 |
rods | -0.174 |
trucks | -0.172 |
Token | to |
Token ID | 284 |
Feature activation | -2.93e+00 |
Pos logit contributions | |
ashed | +0.051 |
ksh | +0.051 |
suits | +0.049 |
tanks | +0.047 |
rods | +0.046 |
Neg logit contributions | |
answer | -0.035 |
note | -0.035 |
he | -0.035 |
owl | -0.034 |
mirror | -0.034 |
Token | have |
Token ID | 423 |
Feature activation | +4.32e+00 |
Pos logit contributions | |
tanks | +0.172 |
ashed | +0.170 |
ksh | +0.162 |
rods | +0.161 |
trucks | +0.157 |
Neg logit contributions | |
he | -0.170 |
note | -0.153 |
guardian | -0.151 |
answer | -0.149 |
fairy | -0.147 |
Token | pic |
Token ID | 8301 |
Feature activation | -1.76e+01 |
Pos logit contributions | |
he | +0.280 |
answer | +0.257 |
mirror | +0.255 |
note | +0.255 |
veil | +0.254 |
Neg logit contributions | |
ashed | -0.220 |
ksh | -0.210 |
tanks | -0.207 |
rods | -0.195 |
mouths | -0.191 |
Token | n |
Token ID | 77 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
tanks | +0.978 |
ashed | +0.962 |
rods | +0.933 |
ksh | +0.912 |
pic | +0.909 |
Neg logit contributions | |
he | -1.035 |
guardian | -0.945 |
note | -0.938 |
fairy | -0.931 |
answer | -0.927 |
Token | ics |
Token ID | 873 |
Feature activation | +9.48e-01 |
Pos logit contributions | |
ashed | +0.059 |
tanks | +0.056 |
trucks | +0.054 |
they | +0.054 |
rods | +0.052 |
Neg logit contributions | |
guardian | -0.089 |
fairy | -0.088 |
ocket | -0.086 |
elf | -0.086 |
wizard | -0.085 |
Token | with |
Token ID | 351 |
Feature activation | +2.69e+00 |
Pos logit contributions | |
he | +0.047 |
note | +0.042 |
answer | +0.042 |
guardian | +0.041 |
fairy | +0.040 |
Neg logit contributions | |
tanks | -0.063 |
ashed | -0.062 |
ksh | -0.060 |
rods | -0.059 |
trucks | -0.057 |
Token | her |
Token ID | 607 |
Feature activation | +3.95e+00 |
Pos logit contributions | |
he | +0.059 |
answer | +0.052 |
note | +0.050 |
fairy | +0.045 |
mirror | +0.041 |
Neg logit contributions | |
ashed | -0.258 |
ksh | -0.254 |
tanks | -0.246 |
suits | -0.240 |
rods | -0.240 |
Token | family |
Token ID | 1641 |
Feature activation | +9.83e-02 |
Pos logit contributions | |
he | +0.164 |
note | +0.162 |
fairy | +0.159 |
answer | +0.147 |
voice | +0.146 |
Neg logit contributions | |
ksh | -0.296 |
ashed | -0.295 |
tanks | -0.279 |
rods | -0.269 |
sizes | -0.266 |
Token | a |
Token ID | 257 |
Feature activation | +3.67e-01 |
Pos logit contributions | |
tanks | +0.064 |
ashed | +0.064 |
ksh | +0.063 |
rods | +0.061 |
mouths | +0.060 |
Neg logit contributions | |
he | -0.039 |
note | -0.033 |
answer | -0.033 |
guardian | -0.032 |
fairy | -0.032 |
Token | family |
Token ID | 1641 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
he | +0.018 |
note | +0.016 |
answer | +0.016 |
guardian | +0.016 |
fairy | +0.016 |
Neg logit contributions | |
ashed | -0.025 |
tanks | -0.025 |
ksh | -0.024 |
rods | -0.023 |
mouths | -0.023 |
Token | who |
Token ID | 508 |
Feature activation | -9.09e-02 |
Pos logit contributions | |
he | +0.007 |
note | +0.006 |
answer | +0.006 |
ibrarian | +0.006 |
guardian | +0.006 |
Neg logit contributions | |
tanks | -0.009 |
ashed | -0.008 |
rods | -0.008 |
trucks | -0.008 |
ksh | -0.008 |
Token | really |
Token ID | 1107 |
Feature activation | +7.52e-01 |
Pos logit contributions | |
ksh | +0.005 |
tanks | +0.005 |
ashed | +0.005 |
rods | +0.005 |
mouths | +0.005 |
Neg logit contributions | |
he | -0.005 |
answer | -0.005 |
note | -0.005 |
fairy | -0.005 |
ad | -0.004 |
Token | enjoyed |
Token ID | 8359 |
Feature activation | -5.26e+00 |
Pos logit contributions | |
he | +0.044 |
answer | +0.040 |
note | +0.039 |
fairy | +0.038 |
guardian | +0.038 |
Neg logit contributions | |
tanks | -0.044 |
ksh | -0.043 |
rods | -0.042 |
ashed | -0.042 |
trucks | -0.040 |
Token | pic |
Token ID | 8301 |
Feature activation | -1.76e+01 |
Pos logit contributions | |
ashed | +0.385 |
ksh | +0.373 |
suits | +0.343 |
sizes | +0.336 |
tanks | +0.328 |
Neg logit contributions | |
note | -0.234 |
he | -0.232 |
answer | -0.227 |
voice | -0.217 |
fairy | -0.210 |
Token | n |
Token ID | 77 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
tanks | +0.963 |
ashed | +0.942 |
rods | +0.913 |
ksh | +0.905 |
pic | +0.888 |
Neg logit contributions | |
he | -1.029 |
guardian | -0.951 |
note | -0.949 |
answer | -0.936 |
fairy | -0.936 |
Token | ics |
Token ID | 873 |
Feature activation | +9.92e-01 |
Pos logit contributions | |
tanks | +0.069 |
ashed | +0.068 |
rods | +0.065 |
sizes | +0.064 |
trucks | +0.064 |
Neg logit contributions | |
guardian | -0.077 |
fairy | -0.076 |
note | -0.074 |
mirror | -0.073 |
wizard | -0.073 |
Token | . |
Token ID | 13 |
Feature activation | -3.09e-01 |
Pos logit contributions | |
he | +0.039 |
note | +0.033 |
answer | +0.032 |
fairy | +0.030 |
guardian | +0.029 |
Neg logit contributions | |
ashed | -0.079 |
ksh | -0.078 |
tanks | -0.077 |
rods | -0.074 |
mouths | -0.074 |
Token | They |
Token ID | 1119 |
Feature activation | -6.84e+00 |
Pos logit contributions | |
ashed | +0.016 |
ksh | +0.014 |
balls | +0.013 |
tanks | +0.013 |
rods | +0.013 |
Neg logit contributions | |
he | -0.021 |
answer | -0.019 |
ad | -0.018 |
voice | -0.018 |
whisper | -0.017 |
Token | loved |
Token ID | 6151 |
Feature activation | -7.50e-01 |
Pos logit contributions | |
ksh | +0.448 |
rods | +0.422 |
tanks | +0.411 |
ashed | +0.410 |
mouths | +0.408 |
Neg logit contributions | |
he | -0.386 |
answer | -0.338 |
fairy | -0.321 |
note | -0.308 |
ad | -0.305 |
Token | favourite |
Token ID | 12507 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
ashed | +0.388 |
ksh | +0.382 |
tanks | +0.377 |
rods | +0.357 |
trucks | +0.346 |
Neg logit contributions | |
he | -0.297 |
answer | -0.289 |
note | -0.287 |
guardian | -0.284 |
fairy | -0.278 |
Token | place |
Token ID | 1295 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
ashed | +0.153 |
ksh | +0.151 |
tanks | +0.148 |
rods | +0.141 |
mouths | +0.136 |
Neg logit contributions | |
he | -0.114 |
answer | -0.110 |
note | -0.110 |
fairy | -0.107 |
guardian | -0.104 |
Token | to |
Token ID | 284 |
Feature activation | -2.94e+00 |
Pos logit contributions | |
ashed | +0.079 |
tanks | +0.078 |
ksh | +0.078 |
rods | +0.074 |
trucks | +0.072 |
Neg logit contributions | |
he | -0.054 |
note | -0.049 |
answer | -0.048 |
guardian | -0.046 |
fairy | -0.046 |
Token | go |
Token ID | 467 |
Feature activation | -4.39e+00 |
Pos logit contributions | |
tanks | +0.181 |
ashed | +0.176 |
ksh | +0.171 |
rods | +0.169 |
pic | +0.168 |
Neg logit contributions | |
he | -0.154 |
note | -0.141 |
guardian | -0.137 |
ibrarian | -0.135 |
answer | -0.134 |
Token | for |
Token ID | 329 |
Feature activation | +4.97e-01 |
Pos logit contributions | |
ashed | +0.305 |
tanks | +0.304 |
ksh | +0.297 |
rods | +0.290 |
mouths | +0.281 |
Neg logit contributions | |
he | -0.209 |
note | -0.183 |
answer | -0.182 |
guardian | -0.176 |
fairy | -0.175 |
Token | pic |
Token ID | 8301 |
Feature activation | -1.76e+01 |
Pos logit contributions | |
he | +0.024 |
answer | +0.022 |
note | +0.022 |
ad | +0.021 |
fairy | +0.021 |
Neg logit contributions | |
ashed | -0.035 |
ksh | -0.034 |
tanks | -0.033 |
rods | -0.032 |
suits | -0.031 |
Token | n |
Token ID | 77 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
tanks | +1.032 |
ashed | +1.032 |
ksh | +0.996 |
rods | +0.974 |
pic | +0.944 |
Neg logit contributions | |
he | -0.990 |
note | -0.896 |
answer | -0.887 |
guardian | -0.885 |
fairy | -0.874 |
Token | ics |
Token ID | 873 |
Feature activation | +9.64e-01 |
Pos logit contributions | |
ashed | +0.080 |
tanks | +0.079 |
ksh | +0.076 |
rods | +0.074 |
pic | +0.073 |
Neg logit contributions | |
he | -0.063 |
guardian | -0.062 |
fairy | -0.062 |
note | -0.062 |
answer | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | -3.14e-01 |
Pos logit contributions | |
he | +0.041 |
note | +0.035 |
answer | +0.035 |
fairy | +0.033 |
guardian | +0.033 |
Neg logit contributions | |
ashed | -0.073 |
tanks | -0.072 |
ksh | -0.072 |
rods | -0.069 |
mouths | -0.068 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +5.46e+00 |
Pos logit contributions | |
ashed | +0.025 |
tanks | +0.024 |
ksh | +0.024 |
trucks | +0.023 |
rods | +0.023 |
Neg logit contributions | |
he | -0.013 |
answer | -0.009 |
note | -0.009 |
guardian | -0.009 |
ad | -0.009 |
Token | Once |
Token ID | 7454 |
Feature activation | +7.81e-01 |
Pos logit contributions | |
he | +0.258 |
note | +0.240 |
answer | +0.237 |
guardian | +0.237 |
fairy | +0.227 |
Neg logit contributions | |
tanks | -0.365 |
ashed | -0.362 |
ksh | -0.350 |
rods | -0.345 |
pic | -0.335 |
Token | see |
Token ID | 766 |
Feature activation | +6.10e+00 |
Pos logit contributions | |
tanks | +0.183 |
ashed | +0.181 |
ksh | +0.175 |
rods | +0.172 |
pic | +0.168 |
Neg logit contributions | |
he | -0.154 |
note | -0.140 |
answer | -0.136 |
guardian | -0.135 |
ibrarian | -0.133 |
Token | the |
Token ID | 262 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
he | +0.289 |
answer | +0.266 |
fairy | +0.257 |
note | +0.254 |
guardian | +0.250 |
Neg logit contributions | |
ashed | -0.440 |
ksh | -0.423 |
tanks | -0.417 |
rods | -0.402 |
suits | -0.397 |
Token | bird |
Token ID | 6512 |
Feature activation | +5.71e+00 |
Pos logit contributions | |
tanks | +0.066 |
pic | +0.064 |
ksh | +0.063 |
ashed | +0.063 |
trucks | +0.063 |
Neg logit contributions | |
he | -0.054 |
ibrarian | -0.052 |
ad | -0.048 |
ocket | -0.047 |
note | -0.045 |
Token | and |
Token ID | 290 |
Feature activation | +6.23e-01 |
Pos logit contributions | |
he | +0.302 |
ibrarian | +0.273 |
note | +0.273 |
answer | +0.266 |
guardian | +0.266 |
Neg logit contributions | |
tanks | -0.359 |
ashed | -0.351 |
ksh | -0.343 |
mouths | -0.328 |
rods | -0.328 |
Token | have |
Token ID | 423 |
Feature activation | +4.38e+00 |
Pos logit contributions | |
he | +0.032 |
fairy | +0.031 |
answer | +0.031 |
note | +0.030 |
guardian | +0.030 |
Neg logit contributions | |
ksh | -0.041 |
ashed | -0.039 |
tanks | -0.038 |
rods | -0.037 |
suits | -0.036 |
Token | pic |
Token ID | 8301 |
Feature activation | -1.75e+01 |
Pos logit contributions | |
note | +0.190 |
answer | +0.184 |
fairy | +0.182 |
ad | +0.181 |
he | +0.177 |
Neg logit contributions | |
ksh | -0.294 |
tanks | -0.285 |
ashed | -0.282 |
suits | -0.271 |
trucks | -0.264 |
Token | n |
Token ID | 77 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
ashed | +1.027 |
tanks | +1.016 |
ksh | +0.983 |
rods | +0.966 |
pic | +0.936 |
Neg logit contributions | |
he | -0.991 |
note | -0.899 |
guardian | -0.891 |
answer | -0.883 |
fairy | -0.876 |
Token | ics |
Token ID | 873 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
ashed | +0.072 |
tanks | +0.067 |
ksh | +0.064 |
rods | +0.064 |
trucks | +0.062 |
Neg logit contributions | |
guardian | -0.068 |
answer | -0.068 |
he | -0.067 |
fairy | -0.067 |
note | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
he | +0.059 |
note | +0.053 |
answer | +0.051 |
fairy | +0.051 |
guardian | +0.049 |
Neg logit contributions | |
tanks | -0.066 |
ksh | -0.066 |
ashed | -0.065 |
rods | -0.063 |
mouths | -0.062 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +5.52e+00 |
Pos logit contributions | |
tanks | +0.020 |
ashed | +0.020 |
rods | +0.020 |
ksh | +0.019 |
trucks | +0.019 |
Neg logit contributions | |
he | -0.009 |
fairy | -0.007 |
answer | -0.007 |
note | -0.006 |
guardian | -0.006 |
Token | Once |
Token ID | 7454 |
Feature activation | +8.56e-01 |
Pos logit contributions | |
guardian | +0.276 |
he | +0.265 |
note | +0.265 |
answer | +0.262 |
fairy | +0.257 |
Neg logit contributions | |
tanks | -0.343 |
ashed | -0.333 |
rods | -0.324 |
pic | -0.323 |
ksh | -0.321 |
Token | They |
Token ID | 2990 |
Feature activation | -5.47e+00 |
Pos logit contributions | |
he | +0.089 |
answer | +0.084 |
note | +0.083 |
guardian | +0.082 |
fairy | +0.081 |
Neg logit contributions | |
tanks | -0.089 |
ashed | -0.088 |
rods | -0.085 |
ksh | -0.085 |
pic | -0.082 |
Token | decided |
Token ID | 3066 |
Feature activation | -5.84e+00 |
Pos logit contributions | |
tanks | +0.287 |
ashed | +0.285 |
ksh | +0.279 |
rods | +0.269 |
trucks | +0.258 |
Neg logit contributions | |
he | -0.351 |
note | -0.320 |
answer | -0.318 |
guardian | -0.311 |
fairy | -0.310 |
Token | to |
Token ID | 284 |
Feature activation | -2.82e+00 |
Pos logit contributions | |
ksh | +0.376 |
ashed | +0.362 |
tanks | +0.362 |
rods | +0.338 |
sizes | +0.335 |
Neg logit contributions | |
he | -0.311 |
note | -0.303 |
answer | -0.297 |
guardian | -0.278 |
fairy | -0.276 |
Token | make |
Token ID | 787 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
tanks | +0.173 |
ashed | +0.169 |
rods | +0.163 |
ksh | +0.162 |
pic | +0.159 |
Neg logit contributions | |
he | -0.151 |
note | -0.135 |
guardian | -0.132 |
answer | -0.131 |
ibrarian | -0.130 |
Token | more |
Token ID | 517 |
Feature activation | -3.43e+00 |
Pos logit contributions | |
he | +0.110 |
note | +0.100 |
answer | +0.099 |
fairy | +0.095 |
guardian | +0.095 |
Neg logit contributions | |
ashed | -0.132 |
tanks | -0.129 |
ksh | -0.127 |
rods | -0.121 |
trucks | -0.118 |
Token | pic |
Token ID | 8301 |
Feature activation | -1.74e+01 |
Pos logit contributions | |
ashed | +0.172 |
ksh | +0.168 |
tanks | +0.163 |
rods | +0.149 |
trucks | +0.143 |
Neg logit contributions | |
he | -0.230 |
answer | -0.219 |
note | -0.218 |
fairy | -0.208 |
guardian | -0.206 |
Token | n |
Token ID | 77 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
tanks | +1.018 |
ashed | +1.017 |
rods | +0.974 |
ksh | +0.970 |
pic | +0.925 |
Neg logit contributions | |
he | -0.981 |
guardian | -0.882 |
note | -0.879 |
answer | -0.873 |
fairy | -0.865 |
Token | ics |
Token ID | 873 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
ashed | +0.068 |
tanks | +0.068 |
rods | +0.064 |
ksh | +0.063 |
trucks | +0.063 |
Neg logit contributions | |
he | -0.061 |
guardian | -0.061 |
fairy | -0.061 |
note | -0.060 |
answer | -0.059 |
Token | and |
Token ID | 290 |
Feature activation | +7.15e-01 |
Pos logit contributions | |
he | +0.063 |
note | +0.057 |
answer | +0.056 |
guardian | +0.055 |
fairy | +0.055 |
Neg logit contributions | |
ashed | -0.069 |
tanks | -0.069 |
ksh | -0.068 |
rods | -0.065 |
trucks | -0.063 |
Token | looked |
Token ID | 3114 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
note | +0.034 |
he | +0.034 |
answer | +0.034 |
fairy | +0.032 |
figure | +0.031 |
Neg logit contributions | |
ksh | -0.049 |
ashed | -0.049 |
tanks | -0.047 |
suits | -0.045 |
rods | -0.044 |
Token | in |
Token ID | 287 |
Feature activation | +8.95e-02 |
Pos logit contributions | |
he | +0.060 |
answer | +0.056 |
fairy | +0.052 |
note | +0.051 |
reply | +0.050 |
Neg logit contributions | |
ksh | -0.096 |
ashed | -0.094 |
rods | -0.093 |
tanks | -0.093 |
trucks | -0.090 |
Token | even |
Token ID | 772 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
ashed | +0.488 |
pic | +0.483 |
trucks | +0.480 |
tanks | +0.468 |
mouths | +0.467 |
Neg logit contributions | |
note | -0.436 |
he | -0.423 |
fairy | -0.416 |
wizard | -0.413 |
ocket | -0.412 |
Token | see |
Token ID | 766 |
Feature activation | +6.18e+00 |
Pos logit contributions | |
tanks | +0.010 |
ashed | +0.010 |
ksh | +0.009 |
rods | +0.009 |
trucks | +0.009 |
Neg logit contributions | |
he | -0.007 |
note | -0.006 |
guardian | -0.006 |
ibrarian | -0.006 |
fairy | -0.006 |
Token | other |
Token ID | 584 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
he | +0.299 |
answer | +0.267 |
note | +0.261 |
fairy | +0.258 |
guardian | +0.253 |
Neg logit contributions | |
ashed | -0.435 |
ksh | -0.425 |
tanks | -0.419 |
rods | -0.397 |
suits | -0.384 |
Token | people |
Token ID | 661 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
ashed | +0.121 |
ksh | +0.117 |
tanks | +0.117 |
rods | +0.110 |
suits | +0.104 |
Neg logit contributions | |
he | -0.107 |
answer | -0.099 |
note | -0.096 |
guardian | -0.093 |
fairy | -0.092 |
Token | having |
Token ID | 1719 |
Feature activation | -2.82e+00 |
Pos logit contributions | |
he | +0.009 |
note | +0.008 |
ibrarian | +0.008 |
guardian | +0.008 |
answer | +0.008 |
Neg logit contributions | |
tanks | -0.011 |
ashed | -0.010 |
ksh | -0.010 |
trucks | -0.010 |
rods | -0.010 |
Token | pic |
Token ID | 8301 |
Feature activation | -1.74e+01 |
Pos logit contributions | |
ashed | +0.155 |
ksh | +0.151 |
tanks | +0.145 |
rods | +0.132 |
suits | +0.129 |
Neg logit contributions | |
he | -0.180 |
answer | -0.169 |
note | -0.167 |
ad | -0.165 |
voice | -0.158 |
Token | n |
Token ID | 77 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
tanks | +1.104 |
ashed | +1.098 |
ksh | +1.068 |
rods | +1.038 |
trucks | +1.005 |
Neg logit contributions | |
he | -0.931 |
note | -0.833 |
answer | -0.828 |
guardian | -0.804 |
fairy | -0.794 |
Token | ics |
Token ID | 873 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
ashed | +0.068 |
tanks | +0.066 |
ksh | +0.063 |
rods | +0.063 |
trucks | +0.062 |
Neg logit contributions | |
guardian | -0.061 |
fairy | -0.061 |
note | -0.061 |
he | -0.060 |
answer | -0.059 |
Token | . |
Token ID | 13 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
he | +0.047 |
note | +0.040 |
answer | +0.039 |
fairy | +0.037 |
guardian | +0.036 |
Neg logit contributions | |
ksh | -0.088 |
ashed | -0.087 |
tanks | -0.086 |
mouths | -0.083 |
rods | -0.082 |
Token |
|
Token ID | 198 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
tanks | +0.009 |
trucks | +0.009 |
ashed | +0.008 |
ksh | +0.008 |
rods | +0.008 |
Neg logit contributions | |
he | -0.010 |
answer | -0.009 |
ad | -0.008 |
note | -0.008 |
ibrarian | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
guardian | +0.095 |
angel | +0.095 |
note | +0.091 |
answer | +0.090 |
ring | +0.089 |
Neg logit contributions | |
tanks | -0.089 |
rods | -0.088 |
ashed | -0.088 |
pic | -0.085 |
they | -0.083 |
Token | and |
Token ID | 290 |
Feature activation | +7.06e-01 |
Pos logit contributions | |
ksh | +0.664 |
ashed | +0.664 |
tanks | +0.656 |
rods | +0.625 |
suits | +0.604 |
Neg logit contributions | |
he | -0.420 |
answer | -0.381 |
note | -0.373 |
fairy | -0.349 |
guardian | -0.346 |
Token | it |
Token ID | 340 |
Feature activation | +2.48e+00 |
Pos logit contributions | |
answer | +0.038 |
figure | +0.037 |
he | +0.036 |
note | +0.035 |
ad | +0.035 |
Neg logit contributions | |
ksh | -0.044 |
ashed | -0.040 |
suits | -0.039 |
tanks | -0.038 |
rods | -0.038 |
Token | was |
Token ID | 373 |
Feature activation | -7.07e-01 |
Pos logit contributions | |
he | +0.124 |
answer | +0.115 |
fairy | +0.113 |
note | +0.113 |
figure | +0.108 |
Neg logit contributions | |
ksh | -0.163 |
tanks | -0.161 |
ashed | -0.160 |
rods | -0.158 |
suits | -0.151 |
Token | going |
Token ID | 1016 |
Feature activation | -4.70e+00 |
Pos logit contributions | |
ashed | +0.056 |
ksh | +0.055 |
rods | +0.053 |
tanks | +0.052 |
suits | +0.052 |
Neg logit contributions | |
answer | -0.026 |
he | -0.024 |
fairy | -0.023 |
note | -0.023 |
guardian | -0.023 |
Token | on |
Token ID | 319 |
Feature activation | +4.07e-02 |
Pos logit contributions | |
ksh | +0.352 |
mouths | +0.340 |
ashed | +0.337 |
pic | +0.327 |
rods | +0.322 |
Neg logit contributions | |
he | -0.180 |
answer | -0.180 |
note | -0.171 |
reply | -0.148 |
ad | -0.145 |
Token | pic |
Token ID | 8301 |
Feature activation | -1.74e+01 |
Pos logit contributions | |
he | +0.001 |
answer | +0.001 |
note | +0.001 |
ad | +0.001 |
voice | +0.001 |
Neg logit contributions | |
ashed | -0.004 |
ksh | -0.003 |
mouths | -0.003 |
tanks | -0.003 |
suits | -0.003 |
Token | n |
Token ID | 77 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
tanks | +1.123 |
ashed | +1.122 |
ksh | +1.092 |
rods | +1.057 |
mouths | +1.029 |
Neg logit contributions | |
he | -0.910 |
note | -0.817 |
answer | -0.811 |
guardian | -0.784 |
fairy | -0.773 |
Token | ics |
Token ID | 873 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
ashed | +0.075 |
tanks | +0.074 |
ksh | +0.071 |
rods | +0.071 |
trucks | +0.069 |
Neg logit contributions | |
he | -0.053 |
note | -0.051 |
guardian | -0.050 |
fairy | -0.050 |
answer | -0.049 |
Token | ! |
Token ID | 0 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
he | +0.047 |
note | +0.040 |
answer | +0.039 |
fairy | +0.035 |
figure | +0.034 |
Neg logit contributions | |
ksh | -0.091 |
ashed | -0.090 |
tanks | -0.087 |
mouths | -0.085 |
rods | -0.083 |
Token |
|
Token ID | 198 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
he | +0.078 |
answer | +0.064 |
note | +0.063 |
ad | +0.060 |
ibrarian | +0.059 |
Neg logit contributions | |
ashed | -0.145 |
ksh | -0.142 |
tanks | -0.142 |
trucks | -0.134 |
sizes | -0.134 |
Token |
|
Token ID | 198 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
guardian | +0.100 |
angel | +0.096 |
note | +0.094 |
fairy | +0.093 |
answer | +0.092 |
Neg logit contributions | |
tanks | -0.091 |
rods | -0.089 |
ashed | -0.087 |
pic | -0.084 |
they | -0.081 |
Token | They |
Token ID | 1119 |
Feature activation | -6.70e+00 |
Pos logit contributions | |
ashed | +0.012 |
tanks | +0.012 |
ksh | +0.011 |
rods | +0.011 |
trucks | +0.011 |
Neg logit contributions | |
he | -0.009 |
note | -0.008 |
answer | -0.008 |
guardian | -0.008 |
fairy | -0.008 |
Token | missed |
Token ID | 6825 |
Feature activation | +6.42e-01 |
Pos logit contributions | |
ashed | +0.379 |
tanks | +0.376 |
ksh | +0.353 |
rods | +0.344 |
mouths | +0.337 |
Neg logit contributions | |
he | -0.396 |
note | -0.365 |
answer | -0.358 |
guardian | -0.352 |
ibrarian | -0.344 |
Token | their |
Token ID | 511 |
Feature activation | -5.72e+00 |
Pos logit contributions | |
answer | +0.030 |
he | +0.028 |
note | +0.028 |
fairy | +0.028 |
voice | +0.027 |
Neg logit contributions | |
ashed | -0.045 |
ksh | -0.045 |
tanks | -0.042 |
suits | -0.041 |
rods | -0.041 |
Token | walks |
Token ID | 11114 |
Feature activation | -3.78e+00 |
Pos logit contributions | |
ksh | +0.385 |
ashed | +0.383 |
tanks | +0.378 |
rods | +0.358 |
sizes | +0.347 |
Neg logit contributions | |
he | -0.276 |
note | -0.271 |
answer | -0.269 |
guardian | -0.265 |
fairy | -0.261 |
Token | and |
Token ID | 290 |
Feature activation | +7.49e-01 |
Pos logit contributions | |
ksh | +0.273 |
tanks | +0.264 |
ashed | +0.259 |
mouths | +0.256 |
rods | +0.256 |
Neg logit contributions | |
he | -0.186 |
note | -0.155 |
fairy | -0.150 |
guardian | -0.150 |
answer | -0.146 |
Token | pic |
Token ID | 8301 |
Feature activation | -1.74e+01 |
Pos logit contributions | |
he | +0.039 |
fairy | +0.038 |
answer | +0.038 |
note | +0.038 |
figure | +0.037 |
Neg logit contributions | |
ksh | -0.049 |
ashed | -0.045 |
rods | -0.044 |
tanks | -0.044 |
suits | -0.043 |
Token | n |
Token ID | 77 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
tanks | +1.100 |
ashed | +1.093 |
ksh | +1.064 |
rods | +1.036 |
trucks | +1.003 |
Neg logit contributions | |
he | -0.936 |
note | -0.838 |
answer | -0.832 |
guardian | -0.812 |
fairy | -0.801 |
Token | ics |
Token ID | 873 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
ashed | +0.069 |
tanks | +0.065 |
rods | +0.061 |
ksh | +0.061 |
trucks | +0.060 |
Neg logit contributions | |
he | -0.060 |
answer | -0.059 |
note | -0.059 |
guardian | -0.058 |
fairy | -0.058 |
Token | near |
Token ID | 1474 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
he | +0.063 |
note | +0.057 |
answer | +0.055 |
fairy | +0.055 |
guardian | +0.053 |
Neg logit contributions | |
ksh | -0.076 |
tanks | -0.075 |
ashed | -0.071 |
rods | -0.071 |
trucks | -0.070 |
Token | the |
Token ID | 262 |
Feature activation | -8.95e-01 |
Pos logit contributions | |
he | +0.020 |
answer | +0.018 |
note | +0.017 |
fairy | +0.017 |
figure | +0.017 |
Neg logit contributions | |
ashed | -0.030 |
ksh | -0.029 |
tanks | -0.029 |
rods | -0.027 |
suits | -0.027 |
Token | cliff |
Token ID | 19516 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
tanks | +0.051 |
rods | +0.048 |
ashed | +0.047 |
pic | +0.046 |
ksh | +0.045 |
Neg logit contributions | |
he | -0.052 |
ibrarian | -0.049 |
note | -0.047 |
ad | -0.047 |
ocket | -0.046 |
Token | and |
Token ID | 290 |
Feature activation | +7.11e-01 |
Pos logit contributions | |
he | +0.065 |
note | +0.057 |
answer | +0.057 |
guardian | +0.055 |
fairy | +0.054 |
Neg logit contributions | |
tanks | -0.100 |
ashed | -0.100 |
ksh | -0.097 |
rods | -0.095 |
trucks | -0.092 |
Token | nodded |
Token ID | 14464 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
he | +0.019 |
voice | +0.018 |
fairy | +0.018 |
answer | +0.018 |
note | +0.018 |
Neg logit contributions | |
ksh | -0.061 |
tanks | -0.060 |
rods | -0.058 |
sizes | -0.057 |
ashed | -0.057 |
Token | . |
Token ID | 13 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
ashed | +0.139 |
tanks | +0.138 |
ksh | +0.135 |
rods | +0.133 |
pic | +0.130 |
Neg logit contributions | |
he | -0.068 |
answer | -0.066 |
note | -0.062 |
voice | -0.061 |
fairy | -0.060 |
Token | He |
Token ID | 679 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
ashed | +0.013 |
girls | +0.013 |
pic | +0.013 |
ksh | +0.013 |
tanks | +0.013 |
Neg logit contributions | |
answer | -0.006 |
voice | -0.006 |
he | -0.006 |
ad | -0.005 |
figure | -0.005 |
Token | loved |
Token ID | 6151 |
Feature activation | -6.24e-01 |
Pos logit contributions | |
tanks | +0.014 |
ksh | +0.013 |
ashed | +0.013 |
rods | +0.013 |
trucks | +0.012 |
Neg logit contributions | |
he | -0.010 |
answer | -0.009 |
note | -0.009 |
fairy | -0.009 |
guardian | -0.009 |
Token | pic |
Token ID | 8301 |
Feature activation | -1.74e+01 |
Pos logit contributions | |
ashed | +0.050 |
ksh | +0.049 |
suits | +0.046 |
tanks | +0.042 |
appet | +0.042 |
Neg logit contributions | |
answer | -0.024 |
owl | -0.023 |
note | -0.023 |
ad | -0.022 |
fairy | -0.022 |
Token | n |
Token ID | 77 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
ashed | +0.838 |
tanks | +0.833 |
rods | +0.805 |
ksh | +0.803 |
pic | +0.765 |
Neg logit contributions | |
he | -1.052 |
guardian | -1.018 |
fairy | -1.005 |
note | -0.997 |
mirror | -0.989 |
Token | ics |
Token ID | 873 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
ashed | +0.060 |
tanks | +0.058 |
trucks | +0.056 |
they | +0.055 |
rods | +0.054 |
Neg logit contributions | |
fairy | -0.065 |
guardian | -0.065 |
ocket | -0.064 |
note | -0.063 |
mirror | -0.063 |
Token | and |
Token ID | 290 |
Feature activation | +7.57e-01 |
Pos logit contributions | |
he | +0.050 |
note | +0.044 |
answer | +0.043 |
guardian | +0.041 |
fairy | +0.040 |
Neg logit contributions | |
ashed | -0.088 |
ksh | -0.086 |
tanks | -0.085 |
rods | -0.081 |
mouths | -0.081 |
Token | couldn |
Token ID | 3521 |
Feature activation | -7.24e+00 |
Pos logit contributions | |
answer | +0.034 |
whisper | +0.033 |
fairy | +0.033 |
figure | +0.033 |
he | +0.032 |
Neg logit contributions | |
ksh | -0.052 |
rods | -0.045 |
ashed | -0.045 |
suits | -0.044 |
tanks | -0.044 |
Token | 't |
Token ID | 470 |
Feature activation | +8.70e+00 |
Pos logit contributions | |
tanks | +0.573 |
ashed | +0.566 |
ksh | +0.561 |
rods | +0.550 |
pic | +0.542 |
Neg logit contributions | |
he | -0.225 |
note | -0.218 |
guardian | -0.213 |
fairy | -0.205 |
answer | -0.204 |
Token | saw |
Token ID | 2497 |
Feature activation | +4.47e+00 |
Pos logit contributions | |
tanks | +0.012 |
ksh | +0.012 |
ashed | +0.012 |
rods | +0.012 |
trucks | +0.011 |
Neg logit contributions | |
he | -0.009 |
note | -0.008 |
answer | -0.008 |
guardian | -0.008 |
fairy | -0.008 |
Token | lots |
Token ID | 6041 |
Feature activation | -4.51e+00 |
Pos logit contributions | |
he | +0.215 |
note | +0.192 |
answer | +0.190 |
guardian | +0.185 |
fairy | +0.184 |
Neg logit contributions | |
ashed | -0.306 |
tanks | -0.305 |
ksh | -0.300 |
rods | -0.290 |
mouths | -0.280 |
Token | of |
Token ID | 286 |
Feature activation | +3.25e+00 |
Pos logit contributions | |
ashed | +0.373 |
tanks | +0.372 |
ksh | +0.369 |
rods | +0.357 |
mouths | +0.348 |
Neg logit contributions | |
he | -0.156 |
answer | -0.128 |
note | -0.128 |
fairy | -0.121 |
guardian | -0.121 |
Token | people |
Token ID | 661 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
he | +0.176 |
answer | +0.159 |
note | +0.158 |
fairy | +0.152 |
guardian | +0.152 |
Neg logit contributions | |
ashed | -0.204 |
tanks | -0.200 |
ksh | -0.200 |
rods | -0.190 |
mouths | -0.184 |
Token | having |
Token ID | 1719 |
Feature activation | -2.76e+00 |
Pos logit contributions | |
he | +0.013 |
note | +0.012 |
answer | +0.012 |
guardian | +0.012 |
fairy | +0.012 |
Neg logit contributions | |
tanks | -0.014 |
ashed | -0.014 |
ksh | -0.013 |
rods | -0.013 |
trucks | -0.012 |
Token | pic |
Token ID | 8301 |
Feature activation | -1.74e+01 |
Pos logit contributions | |
ashed | +0.149 |
ksh | +0.145 |
tanks | +0.136 |
rods | +0.125 |
mouths | +0.123 |
Neg logit contributions | |
he | -0.175 |
answer | -0.172 |
note | -0.167 |
voice | -0.162 |
ad | -0.162 |
Token | n |
Token ID | 77 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
tanks | +1.143 |
ashed | +1.132 |
ksh | +1.110 |
rods | +1.070 |
mouths | +1.049 |
Neg logit contributions | |
he | -0.898 |
answer | -0.799 |
note | -0.794 |
ad | -0.762 |
guardian | -0.756 |
Token | ics |
Token ID | 873 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
ashed | +0.067 |
tanks | +0.064 |
ksh | +0.061 |
rods | +0.061 |
trucks | +0.059 |
Neg logit contributions | |
fairy | -0.058 |
guardian | -0.058 |
note | -0.058 |
he | -0.058 |
answer | -0.056 |
Token | and |
Token ID | 290 |
Feature activation | +8.05e-01 |
Pos logit contributions | |
he | +0.043 |
note | +0.036 |
answer | +0.036 |
fairy | +0.033 |
guardian | +0.031 |
Neg logit contributions | |
ksh | -0.101 |
ashed | -0.098 |
tanks | -0.096 |
mouths | -0.094 |
rods | -0.092 |
Token | walking |
Token ID | 6155 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
answer | +0.036 |
fairy | +0.036 |
note | +0.035 |
figure | +0.034 |
voice | +0.034 |
Neg logit contributions | |
ksh | -0.058 |
ashed | -0.052 |
suits | -0.049 |
rods | -0.047 |
appet | -0.047 |
Token | their |
Token ID | 511 |
Feature activation | -5.64e+00 |
Pos logit contributions | |
ksh | +0.113 |
mouths | +0.110 |
tanks | +0.108 |
ashed | +0.107 |
pic | +0.106 |
Neg logit contributions | |
he | -0.057 |
answer | -0.050 |
note | -0.048 |
voice | -0.044 |
reply | -0.044 |
Token | be |
Token ID | 307 |
Feature activation | +6.81e-01 |
Pos logit contributions | |
he | +0.066 |
note | +0.058 |
ibrarian | +0.056 |
answer | +0.056 |
guardian | +0.055 |
Neg logit contributions | |
tanks | -0.084 |
ashed | -0.083 |
ksh | -0.081 |
rods | -0.079 |
pic | -0.078 |
Token | there |
Token ID | 612 |
Feature activation | -5.13e-01 |
Pos logit contributions | |
he | +0.030 |
answer | +0.029 |
guardian | +0.029 |
fairy | +0.029 |
note | +0.028 |
Neg logit contributions | |
tanks | -0.048 |
ashed | -0.047 |
rods | -0.047 |
ksh | -0.046 |
trucks | -0.045 |
Token | to |
Token ID | 284 |
Feature activation | -2.69e+00 |
Pos logit contributions | |
ashed | +0.033 |
ksh | +0.033 |
tanks | +0.033 |
rods | +0.031 |
trucks | +0.030 |
Neg logit contributions | |
he | -0.028 |
answer | -0.026 |
note | -0.025 |
guardian | -0.025 |
fairy | -0.025 |
Token | support |
Token ID | 1104 |
Feature activation | +7.55e-02 |
Pos logit contributions | |
tanks | +0.155 |
pic | +0.148 |
rods | +0.144 |
ashed | +0.142 |
trucks | +0.142 |
Neg logit contributions | |
he | -0.144 |
ibrarian | -0.134 |
note | -0.131 |
guardian | -0.126 |
ocket | -0.123 |
Token | her |
Token ID | 607 |
Feature activation | +4.22e+00 |
Pos logit contributions | |
he | +0.003 |
answer | +0.003 |
note | +0.003 |
fairy | +0.003 |
guardian | +0.003 |
Neg logit contributions | |
ashed | -0.006 |
tanks | -0.005 |
ksh | -0.005 |
rods | -0.005 |
trucks | -0.005 |
Token | pic |
Token ID | 8301 |
Feature activation | -1.73e+01 |
Pos logit contributions | |
he | +0.205 |
note | +0.191 |
answer | +0.191 |
guardian | +0.185 |
fairy | +0.184 |
Neg logit contributions | |
ashed | -0.294 |
ksh | -0.284 |
tanks | -0.281 |
rods | -0.271 |
trucks | -0.259 |
Token | n |
Token ID | 77 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
tanks | +1.024 |
ashed | +1.016 |
ksh | +0.983 |
rods | +0.966 |
pic | +0.930 |
Neg logit contributions | |
he | -0.982 |
note | -0.880 |
answer | -0.867 |
guardian | -0.864 |
fairy | -0.852 |
Token | ics |
Token ID | 873 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
ashed | +0.063 |
tanks | +0.063 |
ksh | +0.059 |
rods | +0.059 |
trucks | +0.058 |
Neg logit contributions | |
he | -0.053 |
fairy | -0.052 |
guardian | -0.052 |
note | -0.052 |
answer | -0.051 |
Token | -- |
Token ID | 438 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
he | +0.065 |
note | +0.058 |
answer | +0.058 |
guardian | +0.056 |
fairy | +0.056 |
Neg logit contributions | |
ashed | -0.080 |
tanks | -0.079 |
ksh | -0.078 |
rods | -0.075 |
mouths | -0.073 |
Token | no |
Token ID | 3919 |
Feature activation | +6.13e-01 |
Pos logit contributions | |
he | +0.089 |
note | +0.081 |
answer | +0.081 |
guardian | +0.079 |
fairy | +0.078 |
Neg logit contributions | |
ashed | -0.115 |
tanks | -0.115 |
ksh | -0.112 |
rods | -0.109 |
pic | -0.106 |
Token | matter |
Token ID | 2300 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
he | +0.028 |
note | +0.025 |
answer | +0.025 |
fairy | +0.024 |
ad | +0.024 |
Neg logit contributions | |
ashed | -0.045 |
ksh | -0.044 |
tanks | -0.044 |
rods | -0.042 |
suits | -0.041 |
Token | so |
Token ID | 523 |
Feature activation | -3.86e+00 |
Pos logit contributions | |
ashed | +0.035 |
tanks | +0.035 |
ksh | +0.033 |
rods | +0.031 |
trucks | +0.031 |
Neg logit contributions | |
he | -0.036 |
note | -0.033 |
answer | -0.032 |
guardian | -0.032 |
fairy | -0.032 |
Token | excited |
Token ID | 6568 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
ashed | +0.248 |
tanks | +0.247 |
ksh | +0.239 |
rods | +0.233 |
trucks | +0.226 |
Neg logit contributions | |
he | -0.202 |
note | -0.182 |
answer | -0.180 |
guardian | -0.177 |
fairy | -0.174 |
Token | ! |
Token ID | 0 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
tanks | +0.084 |
ashed | +0.083 |
ksh | +0.081 |
rods | +0.079 |
mouths | +0.076 |
Neg logit contributions | |
he | -0.075 |
note | -0.068 |
answer | -0.067 |
guardian | -0.065 |
fairy | -0.065 |
Token | She |
Token ID | 1375 |
Feature activation | +4.27e+00 |
Pos logit contributions | |
he | +0.079 |
answer | +0.065 |
note | +0.065 |
guardian | +0.061 |
fairy | +0.061 |
Neg logit contributions | |
tanks | -0.159 |
ashed | -0.156 |
ksh | -0.154 |
rods | -0.152 |
trucks | -0.147 |
Token | loved |
Token ID | 6151 |
Feature activation | -4.99e-01 |
Pos logit contributions | |
he | +0.188 |
voice | +0.181 |
figure | +0.180 |
answer | +0.180 |
note | +0.179 |
Neg logit contributions | |
ksh | -0.291 |
tanks | -0.281 |
rods | -0.275 |
sizes | -0.273 |
ashed | -0.266 |
Token | pic |
Token ID | 8301 |
Feature activation | -1.73e+01 |
Pos logit contributions | |
ashed | +0.033 |
ksh | +0.032 |
suits | +0.030 |
tanks | +0.027 |
ashing | +0.027 |
Neg logit contributions | |
note | -0.025 |
answer | -0.025 |
veil | -0.024 |
he | -0.023 |
owl | -0.023 |
Token | n |
Token ID | 77 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
ashed | +0.884 |
tanks | +0.882 |
ksh | +0.848 |
rods | +0.846 |
pic | +0.831 |
Neg logit contributions | |
he | -1.023 |
guardian | -0.986 |
fairy | -0.967 |
note | -0.963 |
answer | -0.961 |
Token | ics |
Token ID | 873 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
ashed | +0.051 |
tanks | +0.049 |
trucks | +0.047 |
they | +0.047 |
ksh | +0.046 |
Neg logit contributions | |
guardian | -0.061 |
fairy | -0.061 |
ocket | -0.058 |
wizard | -0.058 |
answer | -0.058 |
Token | and |
Token ID | 290 |
Feature activation | +8.84e-01 |
Pos logit contributions | |
he | +0.058 |
note | +0.051 |
answer | +0.050 |
guardian | +0.049 |
fairy | +0.049 |
Neg logit contributions | |
ashed | -0.093 |
tanks | -0.092 |
ksh | -0.091 |
rods | -0.088 |
mouths | -0.086 |
Token | spending |
Token ID | 4581 |
Feature activation | -6.44e+00 |
Pos logit contributions | |
answer | +0.037 |
fairy | +0.036 |
whisper | +0.036 |
figure | +0.036 |
mirror | +0.034 |
Neg logit contributions | |
ksh | -0.063 |
rods | -0.056 |
ashed | -0.055 |
tanks | -0.053 |
suits | -0.053 |
Token | time |
Token ID | 640 |
Feature activation | -2.67e+00 |
Pos logit contributions | |
ksh | +0.444 |
tanks | +0.437 |
ashed | +0.429 |
rods | +0.410 |
trucks | +0.402 |
Neg logit contributions | |
he | -0.320 |
note | -0.280 |
answer | -0.275 |
ad | -0.260 |
voice | -0.254 |
Token | liked |
Token ID | 8288 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
ksh | +0.369 |
tanks | +0.357 |
rods | +0.356 |
pic | +0.348 |
mouths | +0.330 |
Neg logit contributions | |
he | -0.407 |
answer | -0.374 |
fairy | -0.354 |
ad | -0.349 |
figure | -0.346 |
Token | to |
Token ID | 284 |
Feature activation | -2.61e+00 |
Pos logit contributions | |
ksh | +0.200 |
suits | +0.192 |
ashed | +0.191 |
appet | +0.178 |
sizes | +0.173 |
Neg logit contributions | |
answer | -0.073 |
owl | -0.068 |
note | -0.062 |
story | -0.060 |
crying | -0.058 |
Token | play |
Token ID | 711 |
Feature activation | -3.35e+00 |
Pos logit contributions | |
ksh | +0.166 |
ashed | +0.165 |
tanks | +0.163 |
rods | +0.154 |
mouths | +0.148 |
Neg logit contributions | |
he | -0.139 |
answer | -0.133 |
note | -0.125 |
figure | -0.124 |
ad | -0.123 |
Token | and |
Token ID | 290 |
Feature activation | +8.97e-01 |
Pos logit contributions | |
ksh | +0.175 |
ashed | +0.164 |
sizes | +0.163 |
tanks | +0.157 |
mouths | +0.156 |
Neg logit contributions | |
he | -0.234 |
note | -0.206 |
figure | -0.205 |
answer | -0.204 |
fairy | -0.202 |
Token | have |
Token ID | 423 |
Feature activation | +4.66e+00 |
Pos logit contributions | |
he | +0.033 |
answer | +0.033 |
figure | +0.031 |
note | +0.029 |
fairy | +0.028 |
Neg logit contributions | |
ksh | -0.071 |
ashed | -0.070 |
tanks | -0.067 |
rods | -0.066 |
suits | -0.064 |
Token | pic |
Token ID | 8301 |
Feature activation | -1.72e+01 |
Pos logit contributions | |
answer | +0.217 |
note | +0.205 |
he | +0.201 |
story | +0.197 |
ad | +0.190 |
Neg logit contributions | |
ksh | -0.309 |
ashed | -0.295 |
iked | -0.273 |
tanks | -0.269 |
mouths | -0.266 |
Token | n |
Token ID | 77 |
Feature activation | -9.55e-01 |
Pos logit contributions | |
tanks | +1.032 |
ashed | +1.030 |
ksh | +0.988 |
rods | +0.978 |
trucks | +0.936 |
Neg logit contributions | |
he | -0.961 |
note | -0.868 |
answer | -0.858 |
guardian | -0.855 |
fairy | -0.844 |
Token | ics |
Token ID | 873 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
ashed | +0.057 |
tanks | +0.056 |
rods | +0.053 |
ksh | +0.053 |
trucks | +0.053 |
Neg logit contributions | |
he | -0.050 |
guardian | -0.049 |
note | -0.049 |
fairy | -0.049 |
answer | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | +4.58e-02 |
Pos logit contributions | |
he | +0.073 |
note | +0.065 |
answer | +0.063 |
fairy | +0.059 |
guardian | +0.056 |
Neg logit contributions | |
ksh | -0.090 |
tanks | -0.086 |
ashed | -0.085 |
mouths | -0.083 |
rods | -0.080 |
Token | One |
Token ID | 1881 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
he | +0.002 |
answer | +0.002 |
ad | +0.002 |
note | +0.002 |
mirror | +0.002 |
Neg logit contributions | |
ashed | -0.003 |
ksh | -0.003 |
tanks | -0.003 |
trucks | -0.003 |
rods | -0.003 |
Token | day |
Token ID | 1110 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
tanks | +0.100 |
ashed | +0.098 |
ksh | +0.097 |
rods | +0.093 |
trucks | +0.090 |
Neg logit contributions | |
he | -0.110 |
answer | -0.100 |
note | -0.100 |
fairy | -0.096 |
guardian | -0.096 |
Token | children |
Token ID | 1751 |
Feature activation | -8.12e-01 |
Pos logit contributions | |
ksh | +0.241 |
ashed | +0.236 |
tanks | +0.221 |
rods | +0.209 |
suits | +0.208 |
Neg logit contributions | |
he | -0.221 |
answer | -0.213 |
note | -0.209 |
guardian | -0.205 |
voice | -0.203 |
Token | running |
Token ID | 2491 |
Feature activation | -4.26e+00 |
Pos logit contributions | |
tanks | +0.047 |
ashed | +0.046 |
ksh | +0.045 |
rods | +0.044 |
trucks | +0.042 |
Neg logit contributions | |
he | -0.048 |
note | -0.043 |
answer | -0.043 |
guardian | -0.042 |
fairy | -0.042 |
Token | around |
Token ID | 1088 |
Feature activation | +4.69e+00 |
Pos logit contributions | |
ksh | +0.351 |
mouths | +0.341 |
ashed | +0.332 |
rods | +0.331 |
tanks | +0.329 |
Neg logit contributions | |
he | -0.151 |
answer | -0.123 |
note | -0.120 |
voice | -0.116 |
fairy | -0.113 |
Token | , |
Token ID | 11 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
he | +0.182 |
answer | +0.171 |
fairy | +0.158 |
note | +0.157 |
figure | +0.156 |
Neg logit contributions | |
ksh | -0.375 |
ashed | -0.374 |
tanks | -0.344 |
mouths | -0.343 |
rods | -0.336 |
Token | families |
Token ID | 4172 |
Feature activation | -7.27e+00 |
Pos logit contributions | |
ksh | +0.115 |
ashed | +0.114 |
tanks | +0.111 |
rods | +0.110 |
mouths | +0.105 |
Neg logit contributions | |
he | -0.085 |
answer | -0.079 |
guardian | -0.078 |
note | -0.077 |
fairy | -0.076 |
Token | pic |
Token ID | 8301 |
Feature activation | -1.72e+01 |
Pos logit contributions | |
ashed | +0.477 |
ksh | +0.474 |
tanks | +0.472 |
rods | +0.453 |
mouths | +0.438 |
Neg logit contributions | |
he | -0.373 |
answer | -0.330 |
note | -0.329 |
guardian | -0.320 |
fairy | -0.318 |
Token | n |
Token ID | 77 |
Feature activation | -9.39e-01 |
Pos logit contributions | |
tanks | +1.057 |
ashed | +1.053 |
ksh | +1.020 |
rods | +0.997 |
trucks | +0.962 |
Neg logit contributions | |
he | -0.941 |
note | -0.847 |
answer | -0.839 |
guardian | -0.825 |
fairy | -0.815 |
Token | icking |
Token ID | 7958 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
ashed | +0.057 |
tanks | +0.054 |
ksh | +0.051 |
rods | +0.051 |
trucks | +0.050 |
Neg logit contributions | |
he | -0.051 |
note | -0.049 |
answer | -0.049 |
fairy | -0.048 |
guardian | -0.048 |
Token | , |
Token ID | 11 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
ksh | +0.161 |
ashed | +0.158 |
tanks | +0.149 |
rods | +0.145 |
suits | +0.140 |
Neg logit contributions | |
he | -0.086 |
answer | -0.082 |
note | -0.078 |
voice | -0.075 |
fairy | -0.074 |
Token | couples |
Token ID | 11886 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
ksh | +0.123 |
ashed | +0.122 |
tanks | +0.117 |
rods | +0.115 |
mouths | +0.111 |
Neg logit contributions | |
he | -0.074 |
answer | -0.069 |
fairy | -0.068 |
note | -0.067 |
guardian | -0.067 |
Token | st |
Token ID | 336 |
Feature activation | -4.81e+00 |
Pos logit contributions | |
he | +0.079 |
answer | +0.070 |
note | +0.069 |
fairy | +0.068 |
figure | +0.066 |
Neg logit contributions | |
ksh | -0.109 |
ashed | -0.106 |
tanks | -0.100 |
rods | -0.098 |
mouths | -0.097 |
Token | to |
Token ID | 284 |
Feature activation | -2.53e+00 |
Pos logit contributions | |
he | +0.072 |
note | +0.064 |
answer | +0.063 |
guardian | +0.061 |
fairy | +0.060 |
Neg logit contributions | |
ashed | -0.116 |
tanks | -0.116 |
ksh | -0.114 |
rods | -0.110 |
trucks | -0.107 |
Token | be |
Token ID | 307 |
Feature activation | +8.81e-01 |
Pos logit contributions | |
tanks | +0.144 |
ashed | +0.140 |
rods | +0.135 |
trucks | +0.133 |
ksh | +0.133 |
Neg logit contributions | |
he | -0.144 |
ibrarian | -0.129 |
note | -0.128 |
guardian | -0.127 |
fairy | -0.123 |
Token | careful |
Token ID | 8161 |
Feature activation | -5.50e+00 |
Pos logit contributions | |
he | +0.038 |
answer | +0.038 |
note | +0.038 |
guardian | +0.037 |
fairy | +0.036 |
Neg logit contributions | |
ksh | -0.062 |
tanks | -0.061 |
ashed | -0.060 |
rods | -0.058 |
suits | -0.057 |
Token | when |
Token ID | 618 |
Feature activation | -6.26e+00 |
Pos logit contributions | |
tanks | +0.379 |
ashed | +0.378 |
ksh | +0.367 |
rods | +0.359 |
trucks | +0.349 |
Neg logit contributions | |
he | -0.261 |
note | -0.230 |
answer | -0.228 |
guardian | -0.223 |
fairy | -0.220 |
Token | having |
Token ID | 1719 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
ksh | +0.397 |
ashed | +0.367 |
suits | +0.339 |
tanks | +0.302 |
sizes | +0.291 |
Neg logit contributions | |
answer | -0.382 |
note | -0.372 |
fairy | -0.371 |
figure | -0.364 |
voice | -0.362 |
Token | pic |
Token ID | 8301 |
Feature activation | -1.71e+01 |
Pos logit contributions | |
ashed | +0.172 |
ksh | +0.171 |
suits | +0.166 |
tanks | +0.159 |
rods | +0.150 |
Neg logit contributions | |
answer | -0.121 |
note | -0.118 |
he | -0.114 |
voice | -0.113 |
ad | -0.109 |
Token | n |
Token ID | 77 |
Feature activation | -8.53e-01 |
Pos logit contributions | |
tanks | +1.081 |
ashed | +1.077 |
ksh | +1.046 |
rods | +1.019 |
trucks | +0.986 |
Neg logit contributions | |
he | -0.917 |
note | -0.822 |
answer | -0.816 |
guardian | -0.796 |
fairy | -0.786 |
Token | ics |
Token ID | 873 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
tanks | +0.055 |
ashed | +0.055 |
ksh | +0.053 |
rods | +0.052 |
trucks | +0.050 |
Neg logit contributions | |
he | -0.044 |
note | -0.039 |
answer | -0.039 |
guardian | -0.038 |
fairy | -0.038 |
Token | and |
Token ID | 290 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
he | +0.055 |
note | +0.048 |
answer | +0.048 |
fairy | +0.043 |
figure | +0.042 |
Neg logit contributions | |
ksh | -0.120 |
ashed | -0.116 |
tanks | -0.112 |
rods | -0.107 |
mouths | -0.106 |
Token | not |
Token ID | 407 |
Feature activation | +5.18e+00 |
Pos logit contributions | |
note | +0.049 |
answer | +0.049 |
he | +0.047 |
fairy | +0.046 |
guardian | +0.045 |
Neg logit contributions | |
ksh | -0.072 |
ashed | -0.069 |
suits | -0.067 |
tanks | -0.066 |
rods | -0.064 |
Token | to |
Token ID | 284 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
answer | +0.224 |
note | +0.217 |
he | +0.214 |
ring | +0.197 |
fairy | +0.196 |
Neg logit contributions | |
ksh | -0.374 |
ashed | -0.363 |
tanks | -0.360 |
rods | -0.353 |
suits | -0.351 |
Token | Tim |
Token ID | 5045 |
Feature activation | -6.97e+00 |
Pos logit contributions | |
he | +0.006 |
ad | +0.005 |
answer | +0.005 |
on | +0.005 |
voice | +0.005 |
Neg logit contributions | |
ashed | -0.004 |
ksh | -0.004 |
suits | -0.004 |
rods | -0.004 |
tanks | -0.004 |
Token | my |
Token ID | 1820 |
Feature activation | +5.22e+00 |
Pos logit contributions | |
tanks | +0.474 |
ashed | +0.471 |
trucks | +0.457 |
sizes | +0.440 |
ksh | +0.438 |
Neg logit contributions | |
guardian | -0.314 |
note | -0.303 |
ibrarian | -0.292 |
he | -0.292 |
fairy | -0.290 |
Token | loved |
Token ID | 6151 |
Feature activation | -3.42e-01 |
Pos logit contributions | |
he | +0.332 |
note | +0.304 |
answer | +0.302 |
guardian | +0.297 |
fairy | +0.293 |
Neg logit contributions | |
tanks | -0.277 |
ashed | -0.275 |
ksh | -0.263 |
rods | -0.257 |
trucks | -0.248 |
Token | going |
Token ID | 1016 |
Feature activation | -4.40e+00 |
Pos logit contributions | |
ashed | +0.025 |
ksh | +0.024 |
suits | +0.024 |
sizes | +0.022 |
appet | +0.022 |
Neg logit contributions | |
answer | -0.014 |
owl | -0.014 |
note | -0.014 |
mirror | -0.014 |
he | -0.013 |
Token | on |
Token ID | 319 |
Feature activation | +3.42e-01 |
Pos logit contributions | |
ashed | +0.284 |
ksh | +0.284 |
tanks | +0.277 |
mouths | +0.268 |
rods | +0.265 |
Neg logit contributions | |
he | -0.232 |
answer | -0.213 |
note | -0.207 |
guardian | -0.200 |
fairy | -0.194 |
Token | pic |
Token ID | 8301 |
Feature activation | -1.71e+01 |
Pos logit contributions | |
he | +0.021 |
note | +0.020 |
answer | +0.019 |
ad | +0.019 |
voice | +0.018 |
Neg logit contributions | |
ashed | -0.020 |
ksh | -0.019 |
tanks | -0.018 |
suits | -0.017 |
sizes | -0.017 |
Token | n |
Token ID | 77 |
Feature activation | -8.35e-01 |
Pos logit contributions | |
tanks | +0.973 |
ashed | +0.944 |
rods | +0.921 |
ksh | +0.904 |
pic | +0.888 |
Neg logit contributions | |
he | -0.990 |
note | -0.902 |
guardian | -0.900 |
fairy | -0.896 |
answer | -0.893 |
Token | ics |
Token ID | 873 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
tanks | +0.041 |
ashed | +0.040 |
trucks | +0.040 |
they | +0.039 |
rods | +0.038 |
Neg logit contributions | |
fairy | -0.051 |
guardian | -0.051 |
ocket | -0.050 |
mirror | -0.048 |
wizard | -0.048 |
Token | with |
Token ID | 351 |
Feature activation | +3.20e+00 |
Pos logit contributions | |
he | +0.066 |
note | +0.057 |
answer | +0.056 |
guardian | +0.052 |
fairy | +0.052 |
Neg logit contributions | |
ashed | -0.107 |
tanks | -0.106 |
ksh | -0.105 |
rods | -0.100 |
mouths | -0.099 |
Token | his |
Token ID | 465 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
he | +0.106 |
answer | +0.097 |
ad | +0.096 |
note | +0.091 |
mirror | +0.089 |
Neg logit contributions | |
ashed | -0.269 |
ksh | -0.259 |
suits | -0.253 |
appet | -0.243 |
ashing | -0.240 |
Token | family |
Token ID | 1641 |
Feature activation | +6.42e-01 |
Pos logit contributions | |
note | +0.006 |
he | +0.006 |
voice | +0.006 |
fairy | +0.006 |
answer | +0.005 |
Neg logit contributions | |
ksh | -0.014 |
ashed | -0.013 |
sizes | -0.013 |
tanks | -0.012 |
rect | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | -6.35e-01 |
Pos logit contributions | |
he | +0.019 |
answer | +0.016 |
note | +0.016 |
fairy | +0.016 |
guardian | +0.015 |
Neg logit contributions | |
ashed | -0.029 |
ksh | -0.028 |
tanks | -0.027 |
rods | -0.026 |
mouths | -0.026 |
Token | park |
Token ID | 3952 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
ashed | +0.040 |
ksh | +0.039 |
tanks | +0.038 |
sizes | +0.036 |
rods | +0.036 |
Neg logit contributions | |
he | -0.034 |
answer | -0.031 |
guardian | -0.030 |
fairy | -0.030 |
mirror | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
he | +0.061 |
answer | +0.050 |
note | +0.049 |
figure | +0.047 |
fairy | +0.046 |
Neg logit contributions | |
ksh | -0.138 |
ashed | -0.135 |
tanks | -0.130 |
mouths | -0.127 |
rods | -0.127 |
Token | and |
Token ID | 290 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
ksh | +0.117 |
ashed | +0.116 |
tanks | +0.112 |
rods | +0.109 |
mouths | +0.107 |
Neg logit contributions | |
he | -0.071 |
answer | -0.065 |
guardian | -0.062 |
note | -0.062 |
fairy | -0.061 |
Token | have |
Token ID | 423 |
Feature activation | +4.82e+00 |
Pos logit contributions | |
he | +0.057 |
answer | +0.057 |
figure | +0.056 |
fairy | +0.054 |
note | +0.053 |
Neg logit contributions | |
ksh | -0.066 |
ashed | -0.063 |
suits | -0.059 |
tanks | -0.058 |
rods | -0.057 |
Token | pic |
Token ID | 8301 |
Feature activation | -1.71e+01 |
Pos logit contributions | |
ad | +0.212 |
answer | +0.200 |
note | +0.193 |
fairy | +0.191 |
story | +0.183 |
Neg logit contributions | |
ksh | -0.324 |
ashed | -0.307 |
tanks | -0.291 |
suits | -0.289 |
mons | -0.287 |
Token | n |
Token ID | 77 |
Feature activation | -7.94e-01 |
Pos logit contributions | |
tanks | +1.081 |
ashed | +1.074 |
ksh | +1.045 |
rods | +1.016 |
trucks | +0.985 |
Neg logit contributions | |
he | -0.913 |
note | -0.817 |
answer | -0.812 |
guardian | -0.789 |
fairy | -0.779 |
Token | ics |
Token ID | 873 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
ashed | +0.049 |
tanks | +0.047 |
ksh | +0.045 |
rods | +0.045 |
trucks | +0.044 |
Neg logit contributions | |
he | -0.041 |
note | -0.040 |
guardian | -0.039 |
answer | -0.039 |
fairy | -0.039 |
Token | outside |
Token ID | 2354 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
he | +0.072 |
note | +0.063 |
answer | +0.061 |
fairy | +0.060 |
guardian | +0.058 |
Neg logit contributions | |
ksh | -0.107 |
tanks | -0.105 |
ashed | -0.104 |
mouths | -0.101 |
rods | -0.100 |
Token | . |
Token ID | 13 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
he | +0.085 |
answer | +0.068 |
note | +0.065 |
fairy | +0.063 |
guardian | +0.061 |
Neg logit contributions | |
ashed | -0.169 |
ksh | -0.162 |
tanks | -0.159 |
rods | -0.158 |
mouths | -0.155 |
Token | |
Token ID | 220 |
Feature activation | +7.77e-01 |
Pos logit contributions | |
he | +0.011 |
ad | +0.008 |
bird | +0.008 |
answer | +0.008 |
story | +0.008 |
Neg logit contributions | |
suits | -0.014 |
ashed | -0.014 |
trucks | -0.014 |
rods | -0.014 |
tanks | -0.014 |
Token | went |
Token ID | 1816 |
Feature activation | -6.61e+00 |
Pos logit contributions | |
ksh | +0.347 |
tanks | +0.327 |
ashed | +0.323 |
rods | +0.316 |
mouths | +0.299 |
Neg logit contributions | |
he | -0.411 |
fairy | -0.372 |
answer | -0.366 |
note | -0.365 |
figure | -0.360 |
Token | on |
Token ID | 319 |
Feature activation | +4.01e-01 |
Pos logit contributions | |
ksh | +0.449 |
ashed | +0.417 |
tanks | +0.414 |
mouths | +0.405 |
rods | +0.400 |
Neg logit contributions | |
he | -0.338 |
answer | -0.298 |
fairy | -0.296 |
note | -0.295 |
guardian | -0.280 |
Token | adventures |
Token ID | 17545 |
Feature activation | -6.45e+00 |
Pos logit contributions | |
he | +0.020 |
note | +0.020 |
answer | +0.019 |
story | +0.018 |
fairy | +0.018 |
Neg logit contributions | |
ksh | -0.027 |
ashed | -0.026 |
sizes | -0.024 |
tanks | -0.023 |
mons | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
ksh | +0.500 |
ashed | +0.467 |
tanks | +0.448 |
mouths | +0.447 |
rods | +0.429 |
Neg logit contributions | |
he | -0.309 |
note | -0.251 |
answer | -0.243 |
fairy | -0.240 |
figure | -0.230 |
Token | had |
Token ID | 550 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
ksh | +0.116 |
tanks | +0.114 |
ashed | +0.112 |
rods | +0.111 |
mouths | +0.109 |
Neg logit contributions | |
he | -0.069 |
answer | -0.059 |
fairy | -0.058 |
note | -0.057 |
figure | -0.056 |
Token | pic |
Token ID | 8301 |
Feature activation | -1.70e+01 |
Pos logit contributions | |
ad | +0.005 |
he | +0.005 |
note | +0.005 |
fairy | +0.005 |
answer | +0.004 |
Neg logit contributions | |
ksh | -0.008 |
ashed | -0.008 |
tanks | -0.008 |
suits | -0.008 |
trucks | -0.007 |
Token | n |
Token ID | 77 |
Feature activation | -7.56e-01 |
Pos logit contributions | |
ashed | +1.033 |
tanks | +1.028 |
ksh | +0.993 |
rods | +0.974 |
pic | +0.934 |
Neg logit contributions | |
he | -0.935 |
note | -0.846 |
answer | -0.837 |
guardian | -0.831 |
fairy | -0.820 |
Token | ics |
Token ID | 873 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
ashed | +0.047 |
tanks | +0.045 |
ksh | +0.043 |
rods | +0.043 |
trucks | +0.042 |
Neg logit contributions | |
he | -0.039 |
answer | -0.038 |
note | -0.038 |
guardian | -0.038 |
fairy | -0.037 |
Token | , |
Token ID | 11 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
he | +0.073 |
note | +0.064 |
fairy | +0.060 |
answer | +0.060 |
guardian | +0.058 |
Neg logit contributions | |
ksh | -0.114 |
tanks | -0.108 |
ashed | -0.107 |
mouths | -0.103 |
rods | -0.102 |
Token | and |
Token ID | 290 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
ksh | +0.118 |
ashed | +0.115 |
tanks | +0.115 |
rods | +0.111 |
mouths | +0.109 |
Neg logit contributions | |
he | -0.061 |
fairy | -0.053 |
answer | -0.053 |
note | -0.052 |
guardian | -0.051 |
Token | even |
Token ID | 772 |
Feature activation | +3.31e-01 |
Pos logit contributions | |
fairy | +0.058 |
owl | +0.055 |
answer | +0.055 |
ad | +0.054 |
he | +0.054 |
Neg logit contributions | |
ksh | -0.078 |
suits | -0.073 |
rods | -0.069 |
ashed | -0.067 |
mouths | -0.066 |
Token | " |
Token ID | 366 |
Feature activation | -6.46e+00 |
Pos logit contributions | |
ashed | +0.097 |
tanks | +0.096 |
ksh | +0.090 |
rods | +0.090 |
pic | +0.086 |
Neg logit contributions | |
he | -0.141 |
answer | -0.133 |
note | -0.132 |
guardian | -0.129 |
fairy | -0.128 |
Token | No |
Token ID | 2949 |
Feature activation | +3.63e-01 |
Pos logit contributions | |
ksh | +0.385 |
ashed | +0.376 |
tanks | +0.357 |
sizes | +0.351 |
rods | +0.349 |
Neg logit contributions | |
he | -0.345 |
answer | -0.326 |
note | -0.326 |
ibrarian | -0.322 |
guardian | -0.315 |
Token | , |
Token ID | 11 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
answer | +0.017 |
note | +0.016 |
he | +0.016 |
fairy | +0.015 |
voice | +0.015 |
Neg logit contributions | |
ashed | -0.027 |
ksh | -0.025 |
tanks | -0.025 |
suits | -0.024 |
rods | -0.024 |
Token | I |
Token ID | 314 |
Feature activation | -2.66e+00 |
Pos logit contributions | |
ashed | +0.134 |
tanks | +0.130 |
ksh | +0.128 |
rods | +0.123 |
suits | +0.121 |
Neg logit contributions | |
he | -0.105 |
answer | -0.100 |
note | -0.099 |
fairy | -0.097 |
guardian | -0.096 |
Token | didn |
Token ID | 1422 |
Feature activation | -5.43e+00 |
Pos logit contributions | |
ashed | +0.127 |
tanks | +0.108 |
ksh | +0.104 |
pic | +0.104 |
suits | +0.101 |
Neg logit contributions | |
he | -0.174 |
note | -0.168 |
angel | -0.164 |
guardian | -0.162 |
ibrarian | -0.160 |
Token | 't |
Token ID | 470 |
Feature activation | +8.58e+00 |
Pos logit contributions | |
they | +0.278 |
tanks | +0.259 |
First | +0.248 |
Finn | +0.247 |
balls | +0.245 |
Neg logit contributions | |
guardian | -0.285 |
ocket | -0.279 |
fairy | -0.270 |
wizard | -0.269 |
mirror | -0.268 |
Token | feel |
Token ID | 1254 |
Feature activation | +6.70e+00 |
Pos logit contributions | |
he | +0.449 |
ibrarian | +0.438 |
mother | +0.433 |
ocket | +0.422 |
note | +0.417 |
Neg logit contributions | |
sizes | -0.413 |
ashed | -0.410 |
while | -0.398 |
tanks | -0.383 |
ksh | -0.382 |
Token | any |
Token ID | 597 |
Feature activation | -3.60e+00 |
Pos logit contributions | |
he | +0.376 |
note | +0.332 |
ibrarian | +0.330 |
guardian | +0.328 |
ad | +0.328 |
Neg logit contributions | |
tanks | -0.388 |
ashed | -0.380 |
ksh | -0.370 |
rods | -0.357 |
sizes | -0.356 |
Token | pain |
Token ID | 2356 |
Feature activation | +3.68e+00 |
Pos logit contributions | |
tanks | +0.225 |
ashed | +0.221 |
sizes | +0.215 |
ksh | +0.214 |
mouths | +0.213 |
Neg logit contributions | |
he | -0.183 |
ibrarian | -0.158 |
note | -0.154 |
ad | -0.154 |
guardian | -0.150 |
Token | ". |
Token ID | 1911 |
Feature activation | -2.39e+00 |
Pos logit contributions | |
he | +0.183 |
answer | +0.161 |
note | +0.160 |
fairy | +0.154 |
guardian | +0.153 |
Neg logit contributions | |
ashed | -0.250 |
tanks | -0.248 |
ksh | -0.242 |
rods | -0.238 |
pic | -0.230 |
Token | He |
Token ID | 679 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
ashed | +0.141 |
tanks | +0.141 |
ksh | +0.136 |
rods | +0.132 |
trucks | +0.129 |
Neg logit contributions | |
he | -0.139 |
note | -0.124 |
answer | -0.123 |
guardian | -0.120 |
fairy | -0.119 |
Token | and |
Token ID | 290 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
ashed | +0.072 |
tanks | +0.072 |
mouths | +0.070 |
trucks | +0.069 |
ksh | +0.067 |
Neg logit contributions | |
he | -0.082 |
guardian | -0.077 |
ocket | -0.072 |
note | -0.070 |
wizard | -0.069 |
Token | Lily |
Token ID | 20037 |
Feature activation | -4.47e+00 |
Pos logit contributions | |
he | +0.071 |
answer | +0.069 |
fairy | +0.067 |
note | +0.066 |
figure | +0.065 |
Neg logit contributions | |
ksh | -0.046 |
ashed | -0.046 |
tanks | -0.045 |
rods | -0.043 |
suits | -0.041 |
Token | hug |
Token ID | 16225 |
Feature activation | +6.97e+00 |
Pos logit contributions | |
ashed | +0.269 |
mouths | +0.256 |
tanks | +0.251 |
they | +0.238 |
trucks | +0.238 |
Neg logit contributions | |
guardian | -0.234 |
ibrarian | -0.215 |
wizard | -0.214 |
orn | -0.213 |
he | -0.213 |
Token | . |
Token ID | 13 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
he | +0.342 |
answer | +0.315 |
ad | +0.296 |
note | +0.294 |
voice | +0.294 |
Neg logit contributions | |
ashed | -0.474 |
tanks | -0.465 |
ksh | -0.456 |
rods | -0.442 |
pic | -0.440 |
Token | They |
Token ID | 1119 |
Feature activation | -6.37e+00 |
Pos logit contributions | |
voice | +0.009 |
answer | +0.009 |
fairy | +0.008 |
whisper | +0.008 |
say | +0.008 |
Neg logit contributions | |
pic | -0.008 |
ksh | -0.008 |
ashed | -0.007 |
iked | -0.007 |
tanks | -0.007 |
Token | say |
Token ID | 910 |
Feature activation | +9.64e+00 |
Pos logit contributions | |
ashed | +0.395 |
tanks | +0.387 |
mouths | +0.371 |
trucks | +0.367 |
girls | +0.356 |
Neg logit contributions | |
guardian | -0.302 |
note | -0.286 |
wizard | -0.285 |
he | -0.283 |
ibrarian | -0.279 |
Token | , |
Token ID | 11 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
note | +0.353 |
fairy | +0.331 |
voice | +0.331 |
answer | +0.323 |
he | +0.298 |
Neg logit contributions | |
ksh | -0.829 |
tanks | -0.767 |
sizes | -0.766 |
ashed | -0.759 |
pic | -0.748 |
Token | " |
Token ID | 366 |
Feature activation | -6.02e+00 |
Pos logit contributions | |
tanks | +0.089 |
ashed | +0.089 |
ksh | +0.087 |
rods | +0.086 |
sizes | +0.084 |
Neg logit contributions | |
he | -0.096 |
note | -0.090 |
answer | -0.089 |
fairy | -0.089 |
voice | -0.088 |
Token | We |
Token ID | 1135 |
Feature activation | -3.12e+00 |
Pos logit contributions | |
tanks | +0.373 |
ashed | +0.367 |
ksh | +0.353 |
rods | +0.348 |
trucks | +0.336 |
Neg logit contributions | |
he | -0.340 |
note | -0.306 |
answer | -0.297 |
fairy | -0.295 |
guardian | -0.291 |
Token | are |
Token ID | 389 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
ashed | +0.190 |
tanks | +0.182 |
ksh | +0.172 |
rods | +0.171 |
mouths | +0.170 |
Neg logit contributions | |
he | -0.176 |
guardian | -0.161 |
note | -0.156 |
wizard | -0.156 |
fairy | -0.154 |
Token | sorry |
Token ID | 7926 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
he | +0.064 |
note | +0.057 |
answer | +0.057 |
guardian | +0.056 |
ad | +0.055 |
Neg logit contributions | |
tanks | -0.070 |
ashed | -0.069 |
ksh | -0.066 |
trucks | -0.065 |
rods | -0.065 |
Token | down |
Token ID | 866 |
Feature activation | -2.93e+00 |
Pos logit contributions | |
he | +0.039 |
answer | +0.035 |
note | +0.034 |
fairy | +0.034 |
guardian | +0.034 |
Neg logit contributions | |
tanks | -0.046 |
ashed | -0.045 |
ksh | -0.045 |
rods | -0.044 |
pic | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | -2.30e-01 |
Pos logit contributions | |
tanks | +0.190 |
ashed | +0.190 |
ksh | +0.183 |
rods | +0.179 |
trucks | +0.175 |
Neg logit contributions | |
he | -0.145 |
note | -0.130 |
guardian | -0.128 |
answer | -0.127 |
ibrarian | -0.126 |
Token | She |
Token ID | 1375 |
Feature activation | +4.07e+00 |
Pos logit contributions | |
ashed | +0.015 |
tanks | +0.015 |
ksh | +0.014 |
rods | +0.014 |
mouths | +0.013 |
Neg logit contributions | |
he | -0.012 |
note | -0.011 |
answer | -0.011 |
guardian | -0.011 |
fairy | -0.010 |
Token | started |
Token ID | 2067 |
Feature activation | -4.91e+00 |
Pos logit contributions | |
he | +0.233 |
note | +0.211 |
answer | +0.211 |
guardian | +0.208 |
fairy | +0.202 |
Neg logit contributions | |
ashed | -0.241 |
tanks | -0.240 |
ksh | -0.228 |
rods | -0.224 |
mouths | -0.219 |
Token | to |
Token ID | 284 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
tanks | +0.255 |
ashed | +0.254 |
ksh | +0.243 |
rods | +0.237 |
trucks | +0.228 |
Neg logit contributions | |
he | -0.318 |
note | -0.290 |
answer | -0.287 |
guardian | -0.284 |
fairy | -0.281 |
Token | cry |
Token ID | 3960 |
Feature activation | +8.31e+00 |
Pos logit contributions | |
tanks | +0.246 |
ashed | +0.242 |
ksh | +0.237 |
trucks | +0.234 |
rods | +0.231 |
Neg logit contributions | |
he | -0.089 |
ibrarian | -0.075 |
note | -0.074 |
guardian | -0.071 |
ocket | -0.068 |
Token | and |
Token ID | 290 |
Feature activation | +5.83e-01 |
Pos logit contributions | |
he | +0.384 |
answer | +0.341 |
note | +0.338 |
fairy | +0.330 |
guardian | +0.326 |
Neg logit contributions | |
tanks | -0.589 |
ashed | -0.583 |
rods | -0.569 |
ksh | -0.565 |
pic | -0.547 |
Token | her |
Token ID | 607 |
Feature activation | +4.00e+00 |
Pos logit contributions | |
he | +0.028 |
note | +0.025 |
answer | +0.025 |
guardian | +0.024 |
fairy | +0.024 |
Neg logit contributions | |
tanks | -0.040 |
ashed | -0.040 |
ksh | -0.038 |
rods | -0.038 |
trucks | -0.037 |
Token | friends |
Token ID | 2460 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
he | +0.206 |
note | +0.191 |
answer | +0.188 |
guardian | +0.184 |
fairy | +0.182 |
Neg logit contributions | |
tanks | -0.258 |
ashed | -0.257 |
ksh | -0.253 |
rods | -0.243 |
sizes | -0.233 |
Token | saw |
Token ID | 2497 |
Feature activation | +4.28e+00 |
Pos logit contributions | |
tanks | +0.134 |
ashed | +0.133 |
ksh | +0.129 |
rods | +0.126 |
trucks | +0.121 |
Neg logit contributions | |
he | -0.131 |
note | -0.118 |
answer | -0.117 |
guardian | -0.115 |
fairy | -0.114 |
Token | that |
Token ID | 326 |
Feature activation | +5.53e+00 |
Pos logit contributions | |
he | +0.235 |
note | +0.214 |
guardian | +0.211 |
answer | +0.205 |
iver | +0.204 |
Neg logit contributions | |
tanks | -0.251 |
ashed | -0.250 |
trucks | -0.240 |
rods | -0.239 |
ksh | -0.233 |
Token | The |
Token ID | 383 |
Feature activation | +5.31e-01 |
Pos logit contributions | |
he | +0.012 |
answer | +0.011 |
voice | +0.011 |
fairy | +0.011 |
guardian | +0.010 |
Neg logit contributions | |
tanks | -0.020 |
ashed | -0.019 |
rods | -0.019 |
ksh | -0.019 |
trucks | -0.019 |
Token | little |
Token ID | 1310 |
Feature activation | -3.15e+00 |
Pos logit contributions | |
he | +0.023 |
ad | +0.020 |
orn | +0.020 |
iver | +0.019 |
inkle | +0.018 |
Neg logit contributions | |
ashed | -0.038 |
tanks | -0.036 |
suits | -0.036 |
ksh | -0.035 |
trucks | -0.035 |
Token | girl |
Token ID | 2576 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
ashed | +0.179 |
tanks | +0.171 |
ksh | +0.165 |
suits | +0.162 |
trucks | +0.161 |
Neg logit contributions | |
he | -0.184 |
iver | -0.162 |
ad | -0.162 |
ibrarian | -0.162 |
note | -0.158 |
Token | thought |
Token ID | 1807 |
Feature activation | +3.86e+00 |
Pos logit contributions | |
he | +0.108 |
note | +0.096 |
answer | +0.096 |
guardian | +0.095 |
fairy | +0.094 |
Neg logit contributions | |
tanks | -0.135 |
ksh | -0.133 |
ashed | -0.132 |
rods | -0.129 |
pic | -0.124 |
Token | the |
Token ID | 262 |
Feature activation | -4.45e-01 |
Pos logit contributions | |
he | +0.187 |
answer | +0.167 |
note | +0.167 |
fairy | +0.160 |
guardian | +0.159 |
Neg logit contributions | |
tanks | -0.265 |
ashed | -0.265 |
ksh | -0.261 |
rods | -0.251 |
mouths | -0.242 |
Token | woman |
Token ID | 2415 |
Feature activation | +9.76e+00 |
Pos logit contributions | |
trucks | +0.027 |
suits | +0.027 |
they | +0.026 |
mouths | +0.026 |
games | +0.026 |
Neg logit contributions | |
orn | -0.020 |
he | -0.019 |
ad | -0.019 |
iver | -0.019 |
on | -0.018 |
Token | was |
Token ID | 373 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
he | +0.543 |
note | +0.493 |
answer | +0.492 |
fairy | +0.479 |
guardian | +0.478 |
Neg logit contributions | |
tanks | -0.592 |
ashed | -0.587 |
ksh | -0.575 |
rods | -0.565 |
pic | -0.536 |
Token | so |
Token ID | 523 |
Feature activation | -3.44e+00 |
Pos logit contributions | |
tanks | +0.012 |
ashed | +0.011 |
ksh | +0.011 |
sizes | +0.011 |
trucks | +0.011 |
Neg logit contributions | |
he | -0.011 |
note | -0.010 |
ibrarian | -0.010 |
orn | -0.009 |
answer | -0.009 |
Token | silly |
Token ID | 14397 |
Feature activation | -6.71e-01 |
Pos logit contributions | |
ashed | +0.193 |
tanks | +0.186 |
sizes | +0.181 |
they | +0.178 |
pic | +0.175 |
Neg logit contributions | |
he | -0.188 |
ocket | -0.180 |
note | -0.179 |
orn | -0.176 |
ibrarian | -0.175 |
Token | . |
Token ID | 13 |
Feature activation | +3.75e-01 |
Pos logit contributions | |
ksh | +0.054 |
tanks | +0.054 |
ashed | +0.054 |
rods | +0.053 |
mouths | +0.051 |
Neg logit contributions | |
he | -0.024 |
answer | -0.022 |
note | -0.020 |
fairy | -0.020 |
voice | -0.019 |
Token | She |
Token ID | 1375 |
Feature activation | +4.70e+00 |
Pos logit contributions | |
he | +0.016 |
answer | +0.013 |
note | +0.013 |
guardian | +0.013 |
voice | +0.012 |
Neg logit contributions | |
tanks | -0.029 |
ashed | -0.028 |
ksh | -0.027 |
rods | -0.027 |
trucks | -0.026 |
Token | so |
Token ID | 523 |
Feature activation | -4.15e+00 |
Pos logit contributions | |
ashed | +0.046 |
tanks | +0.044 |
pic | +0.040 |
tails | +0.040 |
ksh | +0.040 |
Neg logit contributions | |
he | -0.058 |
note | -0.053 |
answer | -0.052 |
ibrarian | -0.052 |
orn | -0.051 |
Token | scared |
Token ID | 12008 |
Feature activation | +5.46e+00 |
Pos logit contributions | |
ashed | +0.244 |
tanks | +0.226 |
they | +0.208 |
ksh | +0.206 |
sizes | +0.203 |
Neg logit contributions | |
he | -0.222 |
ibrarian | -0.214 |
answer | -0.214 |
wizard | -0.208 |
note | -0.207 |
Token | and |
Token ID | 290 |
Feature activation | +5.37e-01 |
Pos logit contributions | |
he | +0.242 |
answer | +0.210 |
fairy | +0.205 |
note | +0.200 |
guardian | +0.199 |
Neg logit contributions | |
ashed | -0.401 |
ksh | -0.400 |
tanks | -0.399 |
rods | -0.386 |
suits | -0.375 |
Token | trying |
Token ID | 2111 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
he | +0.028 |
note | +0.025 |
answer | +0.025 |
fairy | +0.025 |
guardian | +0.024 |
Neg logit contributions | |
tanks | -0.035 |
ksh | -0.035 |
ashed | -0.034 |
rods | -0.034 |
suits | -0.032 |
Token | to |
Token ID | 284 |
Feature activation | -2.95e+00 |
Pos logit contributions | |
ksh | +0.001 |
tanks | +0.001 |
ashed | +0.001 |
rods | +0.001 |
pic | +0.001 |
Neg logit contributions | |
he | -0.001 |
answer | -0.001 |
note | -0.001 |
guardian | -0.001 |
fairy | -0.001 |
Token | blink |
Token ID | 21019 |
Feature activation | +8.50e+00 |
Pos logit contributions | |
tanks | +0.149 |
pic | +0.144 |
Finn | +0.142 |
tails | +0.140 |
trucks | +0.139 |
Neg logit contributions | |
he | -0.174 |
ibrarian | -0.166 |
note | -0.162 |
ocket | -0.161 |
mirror | -0.158 |
Token | away |
Token ID | 1497 |
Feature activation | +2.64e+00 |
Pos logit contributions | |
he | +0.318 |
answer | +0.317 |
fairy | +0.308 |
voice | +0.298 |
note | +0.295 |
Neg logit contributions | |
ksh | -0.641 |
ashed | -0.638 |
tanks | -0.636 |
pic | -0.615 |
rods | -0.611 |
Token | the |
Token ID | 262 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
answer | +0.097 |
he | +0.094 |
note | +0.092 |
voice | +0.092 |
fairy | +0.091 |
Neg logit contributions | |
ashed | -0.214 |
ksh | -0.208 |
tanks | -0.202 |
rods | -0.195 |
pic | -0.195 |
Token | tears |
Token ID | 10953 |
Feature activation | +7.93e+00 |
Pos logit contributions | |
tanks | +0.076 |
ashed | +0.074 |
ksh | +0.073 |
rods | +0.071 |
trucks | +0.070 |
Neg logit contributions | |
he | -0.055 |
note | -0.047 |
ibrarian | -0.046 |
answer | -0.046 |
guardian | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | -3.31e-01 |
Pos logit contributions | |
he | +0.325 |
answer | +0.284 |
note | +0.277 |
fairy | +0.267 |
guardian | +0.261 |
Neg logit contributions | |
ashed | -0.604 |
tanks | -0.602 |
ksh | -0.594 |
rods | -0.581 |
pic | -0.562 |
Token | Then |
Token ID | 3244 |
Feature activation | -9.63e+00 |
Pos logit contributions | |
ashed | +0.021 |
tanks | +0.021 |
ksh | +0.020 |
trucks | +0.019 |
rods | +0.019 |
Neg logit contributions | |
he | -0.018 |
answer | -0.016 |
voice | -0.015 |
ad | -0.015 |
guardian | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
he | +0.329 |
note | +0.302 |
iver | +0.299 |
inkle | +0.290 |
guardian | +0.289 |
Neg logit contributions | |
tanks | -0.299 |
trucks | -0.283 |
rods | -0.280 |
ashed | -0.269 |
pic | -0.258 |
Token | bus |
Token ID | 1323 |
Feature activation | -5.71e-01 |
Pos logit contributions | |
tanks | +0.011 |
ashed | +0.010 |
trucks | +0.010 |
rods | +0.010 |
ksh | +0.010 |
Neg logit contributions | |
he | -0.012 |
note | -0.011 |
ibrarian | -0.010 |
ad | -0.010 |
fairy | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +6.06e-01 |
Pos logit contributions | |
tanks | +0.035 |
ashed | +0.034 |
ksh | +0.033 |
trucks | +0.033 |
rods | +0.032 |
Neg logit contributions | |
he | -0.029 |
note | -0.026 |
answer | -0.026 |
fairy | -0.026 |
guardian | -0.026 |
Token | |
Token ID | 220 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
he | +0.033 |
answer | +0.028 |
note | +0.028 |
guardian | +0.027 |
fairy | +0.027 |
Neg logit contributions | |
ashed | -0.038 |
tanks | -0.038 |
ksh | -0.038 |
rods | -0.036 |
mouths | -0.035 |
Token |
|
Token ID | 198 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
note | +0.059 |
fairy | +0.059 |
guardian | +0.059 |
he | +0.058 |
answer | +0.058 |
Neg logit contributions | |
tanks | -0.069 |
ashed | -0.067 |
ksh | -0.065 |
rods | -0.065 |
trucks | -0.062 |
Token |
|
Token ID | 198 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
guardian | +0.148 |
angel | +0.146 |
note | +0.142 |
fairy | +0.141 |
answer | +0.138 |
Neg logit contributions | |
tanks | -0.123 |
ashed | -0.120 |
rods | -0.118 |
suits | -0.114 |
pic | -0.113 |
Token | Jack |
Token ID | 14295 |
Feature activation | +5.29e-02 |
Pos logit contributions | |
he | +0.136 |
note | +0.125 |
answer | +0.125 |
guardian | +0.123 |
fairy | +0.121 |
Neg logit contributions | |
tanks | -0.143 |
ashed | -0.143 |
ksh | -0.136 |
rods | -0.134 |
trucks | -0.129 |
Token | was |
Token ID | 373 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
he | +0.003 |
answer | +0.002 |
note | +0.002 |
guardian | +0.002 |
fairy | +0.002 |
Neg logit contributions | |
ksh | -0.003 |
tanks | -0.003 |
ashed | -0.003 |
rods | -0.003 |
mouths | -0.003 |
Token | so |
Token ID | 523 |
Feature activation | -3.12e+00 |
Pos logit contributions | |
he | +0.006 |
note | +0.005 |
fairy | +0.005 |
answer | +0.005 |
ibrarian | +0.005 |
Neg logit contributions | |
ashed | -0.006 |
tanks | -0.006 |
ksh | -0.005 |
rods | -0.005 |
trucks | -0.005 |
Token | excited |
Token ID | 6568 |
Feature activation | -6.32e-01 |
Pos logit contributions | |
ashed | +0.191 |
tanks | +0.178 |
rods | +0.168 |
they | +0.166 |
pic | +0.163 |
Neg logit contributions | |
he | -0.161 |
note | -0.155 |
fairy | -0.153 |
answer | -0.151 |
ibrarian | -0.151 |
Token | he |
Token ID | 339 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
tanks | +0.039 |
ashed | +0.039 |
ksh | +0.038 |
rods | +0.037 |
mouths | +0.036 |
Neg logit contributions | |
he | -0.035 |
note | -0.031 |
answer | -0.031 |
guardian | -0.030 |
fairy | -0.030 |
Token | so |
Token ID | 523 |
Feature activation | -3.31e+00 |
Pos logit contributions | |
tanks | +0.068 |
ashed | +0.066 |
ksh | +0.064 |
rods | +0.063 |
trucks | +0.061 |
Neg logit contributions | |
he | -0.086 |
note | -0.078 |
answer | -0.077 |
guardian | -0.077 |
fairy | -0.075 |
Token | he |
Token ID | 339 |
Feature activation | +8.70e-01 |
Pos logit contributions | |
ksh | +0.205 |
ashed | +0.203 |
tanks | +0.203 |
rods | +0.192 |
sizes | +0.185 |
Neg logit contributions | |
he | -0.181 |
note | -0.167 |
answer | -0.166 |
fairy | -0.163 |
guardian | -0.161 |
Token | hopped |
Token ID | 45230 |
Feature activation | -4.30e+00 |
Pos logit contributions | |
he | +0.044 |
fairy | +0.043 |
answer | +0.042 |
note | +0.042 |
guardian | +0.042 |
Neg logit contributions | |
tanks | -0.053 |
ksh | -0.053 |
ashed | -0.052 |
rods | -0.051 |
sizes | -0.051 |
Token | closer |
Token ID | 5699 |
Feature activation | +3.53e+00 |
Pos logit contributions | |
ashed | +0.241 |
tanks | +0.230 |
suits | +0.214 |
they | +0.212 |
ksh | +0.212 |
Neg logit contributions | |
he | -0.240 |
guardian | -0.229 |
gen | -0.228 |
note | -0.228 |
inkle | -0.227 |
Token | to |
Token ID | 284 |
Feature activation | -2.10e+00 |
Pos logit contributions | |
he | +0.215 |
note | +0.194 |
guardian | +0.192 |
fairy | +0.187 |
ibrarian | +0.187 |
Neg logit contributions | |
tanks | -0.193 |
ashed | -0.193 |
rods | -0.185 |
ksh | -0.184 |
trucks | -0.178 |
Token | investigate |
Token ID | 9161 |
Feature activation | +5.82e+00 |
Pos logit contributions | |
tanks | +0.117 |
trucks | +0.108 |
rods | +0.105 |
Nem | +0.104 |
Finn | +0.103 |
Neg logit contributions | |
he | -0.133 |
ibrarian | -0.122 |
note | -0.115 |
iver | -0.111 |
rot | -0.110 |
Token | . |
Token ID | 13 |
Feature activation | +5.45e-01 |
Pos logit contributions | |
he | +0.305 |
note | +0.271 |
guardian | +0.270 |
ibrarian | +0.261 |
answer | +0.260 |
Neg logit contributions | |
tanks | -0.372 |
ashed | -0.357 |
rods | -0.349 |
ksh | -0.346 |
trucks | -0.341 |
Token | As |
Token ID | 1081 |
Feature activation | -5.24e+00 |
Pos logit contributions | |
he | +0.027 |
note | +0.024 |
answer | +0.024 |
guardian | +0.023 |
fairy | +0.023 |
Neg logit contributions | |
tanks | -0.037 |
ashed | -0.036 |
ksh | -0.035 |
rods | -0.035 |
trucks | -0.033 |
Token | the |
Token ID | 262 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
ksh | +0.366 |
ashed | +0.359 |
tanks | +0.329 |
mouths | +0.326 |
rods | +0.321 |
Neg logit contributions | |
he | -0.272 |
fairy | -0.253 |
wizard | -0.245 |
answer | -0.243 |
figure | -0.238 |
Token | young |
Token ID | 1862 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
tanks | +0.014 |
ashed | +0.013 |
ksh | +0.013 |
rods | +0.013 |
trucks | +0.012 |
Neg logit contributions | |
he | -0.013 |
note | -0.011 |
answer | -0.011 |
ibrarian | -0.011 |
guardian | -0.011 |
Token | rabbit |
Token ID | 22746 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
ksh | +0.091 |
ashed | +0.088 |
tanks | +0.085 |
rods | +0.084 |
mouths | +0.084 |
Neg logit contributions | |
fairy | -0.046 |
he | -0.044 |
owl | -0.044 |
answer | -0.043 |
guardian | -0.043 |
Token | went |
Token ID | 1816 |
Feature activation | -6.34e+00 |
Pos logit contributions | |
he | +0.067 |
fairy | +0.061 |
answer | +0.060 |
note | +0.060 |
wizard | +0.056 |
Neg logit contributions | |
ksh | -0.091 |
tanks | -0.087 |
ashed | -0.086 |
rods | -0.086 |
suits | -0.084 |
Token | to |
Token ID | 284 |
Feature activation | -2.14e+00 |
Pos logit contributions | |
ashed | +0.405 |
tanks | +0.402 |
ksh | +0.377 |
rods | +0.372 |
pic | +0.364 |
Neg logit contributions | |
he | -0.332 |
note | -0.300 |
guardian | -0.295 |
answer | -0.295 |
ibrarian | -0.290 |
Token | bed |
Token ID | 3996 |
Feature activation | +6.40e-01 |
Pos logit contributions | |
tanks | +0.120 |
pic | +0.113 |
trucks | +0.109 |
ashed | +0.106 |
rods | +0.106 |
Neg logit contributions | |
he | -0.127 |
ibrarian | -0.118 |
note | -0.117 |
guardian | -0.115 |
orn | -0.109 |
Token | with |
Token ID | 351 |
Feature activation | +3.52e+00 |
Pos logit contributions | |
he | +0.032 |
note | +0.028 |
answer | +0.028 |
guardian | +0.027 |
fairy | +0.027 |
Neg logit contributions | |
tanks | -0.043 |
ashed | -0.043 |
ksh | -0.042 |
rods | -0.041 |
trucks | -0.040 |
Token | a |
Token ID | 257 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
he | +0.192 |
note | +0.172 |
guardian | +0.169 |
answer | +0.167 |
ibrarian | +0.164 |
Neg logit contributions | |
tanks | -0.222 |
ashed | -0.217 |
ksh | -0.208 |
rods | -0.206 |
mouths | -0.203 |
Token | big |
Token ID | 1263 |
Feature activation | +3.36e+00 |
Pos logit contributions | |
he | +0.060 |
ad | +0.057 |
ibrarian | +0.057 |
mother | +0.050 |
guardian | +0.050 |
Neg logit contributions | |
tanks | -0.075 |
ashed | -0.073 |
pic | -0.072 |
mouths | -0.071 |
rods | -0.069 |
Token | smile |
Token ID | 8212 |
Feature activation | +7.85e+00 |
Pos logit contributions | |
he | +0.134 |
ibrarian | +0.115 |
guardian | +0.110 |
answer | +0.105 |
note | +0.105 |
Neg logit contributions | |
tanks | -0.255 |
ashed | -0.251 |
mouths | -0.241 |
ksh | -0.239 |
rods | -0.238 |
Token | on |
Token ID | 319 |
Feature activation | +7.18e-01 |
Pos logit contributions | |
he | +0.383 |
note | +0.341 |
answer | +0.337 |
guardian | +0.335 |
fairy | +0.324 |
Neg logit contributions | |
ashed | -0.528 |
tanks | -0.523 |
ksh | -0.505 |
rods | -0.492 |
trucks | -0.481 |
Token | his |
Token ID | 465 |
Feature activation | +5.38e-01 |
Pos logit contributions | |
he | +0.046 |
guardian | +0.044 |
note | +0.042 |
fairy | +0.042 |
ibrarian | +0.042 |
Neg logit contributions | |
tanks | -0.036 |
mouths | -0.035 |
ashed | -0.035 |
rods | -0.034 |
ksh | -0.034 |
Token | face |
Token ID | 1986 |
Feature activation | +5.84e+00 |
Pos logit contributions | |
he | +0.026 |
guardian | +0.022 |
ibrarian | +0.022 |
wizard | +0.021 |
fairy | +0.021 |
Neg logit contributions | |
ksh | -0.036 |
ashed | -0.036 |
mouths | -0.036 |
tanks | -0.036 |
tails | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | +5.54e-01 |
Pos logit contributions | |
he | +0.254 |
note | +0.222 |
answer | +0.220 |
guardian | +0.212 |
fairy | +0.210 |
Neg logit contributions | |
tanks | -0.429 |
ashed | -0.427 |
ksh | -0.417 |
rods | -0.409 |
trucks | -0.396 |
Token | floor |
Token ID | 4314 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
tanks | +0.017 |
ashed | +0.017 |
ksh | +0.016 |
rods | +0.016 |
trucks | +0.016 |
Neg logit contributions | |
he | -0.013 |
guardian | -0.012 |
note | -0.011 |
answer | -0.011 |
ad | -0.011 |
Token | while |
Token ID | 981 |
Feature activation | -8.91e+00 |
Pos logit contributions | |
tanks | +0.016 |
ksh | +0.016 |
ashed | +0.016 |
rods | +0.015 |
pic | +0.015 |
Neg logit contributions | |
he | -0.012 |
answer | -0.010 |
note | -0.010 |
guardian | -0.010 |
fairy | -0.010 |
Token | looking |
Token ID | 2045 |
Feature activation | +4.99e+00 |
Pos logit contributions | |
ksh | +0.656 |
ashed | +0.636 |
suits | +0.609 |
appet | +0.576 |
sizes | +0.569 |
Neg logit contributions | |
he | -0.428 |
answer | -0.394 |
wizard | -0.394 |
fairy | -0.389 |
figure | -0.387 |
Token | at |
Token ID | 379 |
Feature activation | -2.90e-01 |
Pos logit contributions | |
he | +0.249 |
answer | +0.239 |
fairy | +0.220 |
note | +0.220 |
voice | +0.213 |
Neg logit contributions | |
ksh | -0.324 |
ashed | -0.319 |
tanks | -0.310 |
rods | -0.309 |
pic | -0.302 |
Token | his |
Token ID | 465 |
Feature activation | +5.69e-01 |
Pos logit contributions | |
tanks | +0.018 |
ashed | +0.017 |
ksh | +0.017 |
rods | +0.016 |
trucks | +0.016 |
Neg logit contributions | |
he | -0.016 |
note | -0.015 |
answer | -0.015 |
guardian | -0.014 |
fairy | -0.014 |
Token | calendar |
Token ID | 11845 |
Feature activation | +6.15e+00 |
Pos logit contributions | |
he | +0.024 |
note | +0.021 |
answer | +0.021 |
guardian | +0.020 |
fairy | +0.020 |
Neg logit contributions | |
tanks | -0.042 |
ashed | -0.042 |
ksh | -0.041 |
rods | -0.040 |
trucks | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | +5.93e-01 |
Pos logit contributions | |
he | +0.251 |
note | +0.215 |
answer | +0.211 |
fairy | +0.200 |
figure | +0.190 |
Neg logit contributions | |
ashed | -0.493 |
ksh | -0.484 |
tanks | -0.471 |
rods | -0.456 |
suits | -0.440 |
Token |
|
Token ID | 198 |
Feature activation | +2.39e+00 |
Pos logit contributions | |
he | +0.024 |
answer | +0.020 |
note | +0.018 |
ad | +0.018 |
voice | +0.017 |
Neg logit contributions | |
ashed | -0.046 |
tanks | -0.044 |
ksh | -0.044 |
rods | -0.043 |
suits | -0.042 |
Token |
|
Token ID | 198 |
Feature activation | +2.40e+00 |
Pos logit contributions | |
guardian | +0.152 |
ocket | +0.143 |
angel | +0.142 |
fairy | +0.142 |
note | +0.141 |
Neg logit contributions | |
tanks | -0.119 |
ashed | -0.113 |
rods | -0.111 |
they | -0.111 |
pic | -0.110 |
Token | One |
Token ID | 3198 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
he | +0.105 |
note | +0.086 |
answer | +0.086 |
ad | +0.082 |
guardian | +0.079 |
Neg logit contributions | |
ashed | -0.178 |
ksh | -0.175 |
tanks | -0.174 |
mouths | -0.170 |
rods | -0.167 |
Token | day |
Token ID | 1110 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
he | +0.015 |
note | +0.014 |
answer | +0.014 |
guardian | +0.014 |
ibrarian | +0.014 |
Neg logit contributions | |
ashed | -0.011 |
tanks | -0.011 |
ksh | -0.010 |
rods | -0.010 |
trucks | -0.009 |
Token | what |
Token ID | 644 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
he | +0.030 |
note | +0.027 |
answer | +0.027 |
fairy | +0.026 |
guardian | +0.026 |
Neg logit contributions | |
ashed | -0.043 |
tanks | -0.042 |
ksh | -0.042 |
rods | -0.040 |
mouths | -0.039 |
Token | is |
Token ID | 318 |
Feature activation | +2.48e+00 |
Pos logit contributions | |
note | +0.057 |
answer | +0.056 |
fairy | +0.054 |
he | +0.052 |
voice | +0.051 |
Neg logit contributions | |
ksh | -0.076 |
ashed | -0.073 |
tanks | -0.066 |
suits | -0.066 |
rods | -0.063 |
Token | right |
Token ID | 826 |
Feature activation | +3.51e+00 |
Pos logit contributions | |
answer | +0.118 |
guardian | +0.098 |
note | +0.096 |
fairy | +0.095 |
owl | +0.092 |
Neg logit contributions | |
ksh | -0.191 |
ashed | -0.190 |
suits | -0.175 |
tanks | -0.173 |
mouths | -0.172 |
Token | can |
Token ID | 460 |
Feature activation | -6.39e+00 |
Pos logit contributions | |
answer | +0.147 |
he | +0.144 |
note | +0.139 |
fairy | +0.136 |
guardian | +0.133 |
Neg logit contributions | |
ksh | -0.268 |
tanks | -0.256 |
ashed | -0.254 |
pic | -0.247 |
rods | -0.243 |
Token | be |
Token ID | 307 |
Feature activation | +5.10e-01 |
Pos logit contributions | |
tanks | +0.374 |
sizes | +0.358 |
trucks | +0.354 |
rods | +0.349 |
they | +0.343 |
Neg logit contributions | |
he | -0.352 |
ibrarian | -0.335 |
ocket | -0.323 |
note | -0.308 |
guardian | -0.306 |
Token | scary |
Token ID | 14343 |
Feature activation | +9.02e-01 |
Pos logit contributions | |
answer | +0.022 |
note | +0.020 |
fairy | +0.019 |
he | +0.019 |
guardian | +0.019 |
Neg logit contributions | |
ksh | -0.039 |
ashed | -0.037 |
tanks | -0.035 |
rods | -0.035 |
mouths | -0.035 |
Token | , |
Token ID | 11 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
answer | +0.034 |
he | +0.029 |
fairy | +0.029 |
voice | +0.028 |
figure | +0.028 |
Neg logit contributions | |
ksh | -0.074 |
ashed | -0.070 |
rods | -0.068 |
mouths | -0.067 |
tanks | -0.066 |
Token | but |
Token ID | 475 |
Feature activation | -2.37e+00 |
Pos logit contributions | |
ashed | +0.139 |
ksh | +0.134 |
tanks | +0.130 |
suits | +0.127 |
rods | +0.126 |
Neg logit contributions | |
answer | -0.103 |
he | -0.101 |
fairy | -0.101 |
note | -0.100 |
figure | -0.097 |
Token | it |
Token ID | 340 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
tanks | +0.139 |
ashed | +0.137 |
ksh | +0.131 |
rods | +0.131 |
trucks | +0.127 |
Neg logit contributions | |
he | -0.137 |
note | -0.122 |
answer | -0.120 |
guardian | -0.119 |
ibrarian | -0.117 |
Token | can |
Token ID | 460 |
Feature activation | -6.45e+00 |
Pos logit contributions | |
he | +0.107 |
note | +0.095 |
answer | +0.094 |
guardian | +0.092 |
fairy | +0.091 |
Neg logit contributions | |
ashed | -0.164 |
tanks | -0.163 |
ksh | -0.159 |
rods | -0.156 |
trucks | -0.150 |
Token | also |
Token ID | 635 |
Feature activation | +5.51e+00 |
Pos logit contributions | |
tanks | +0.335 |
they | +0.328 |
trucks | +0.314 |
rods | +0.299 |
sizes | +0.296 |
Neg logit contributions | |
ocket | -0.375 |
ibrarian | -0.368 |
he | -0.365 |
note | -0.341 |
orn | -0.323 |
Token | cat |
Token ID | 3797 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
tanks | +0.029 |
ashed | +0.029 |
ksh | +0.028 |
rods | +0.027 |
trucks | +0.026 |
Neg logit contributions | |
he | -0.023 |
note | -0.021 |
answer | -0.021 |
guardian | -0.020 |
fairy | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
he | +0.095 |
note | +0.083 |
answer | +0.082 |
fairy | +0.082 |
guardian | +0.079 |
Neg logit contributions | |
ashed | -0.147 |
tanks | -0.147 |
ksh | -0.145 |
rods | -0.142 |
mouths | -0.135 |
Token | Tim |
Token ID | 5045 |
Feature activation | -6.96e+00 |
Pos logit contributions | |
he | +0.011 |
answer | +0.009 |
fairy | +0.009 |
ibrarian | +0.009 |
ad | +0.009 |
Neg logit contributions | |
ashed | -0.013 |
tanks | -0.013 |
rods | -0.013 |
ksh | -0.013 |
sizes | -0.012 |
Token | my |
Token ID | 1820 |
Feature activation | +5.20e+00 |
Pos logit contributions | |
ashed | +0.507 |
tanks | +0.501 |
ksh | +0.484 |
trucks | +0.468 |
rods | +0.466 |
Neg logit contributions | |
he | -0.301 |
note | -0.274 |
answer | -0.265 |
guardian | -0.263 |
fairy | -0.252 |
Token | was |
Token ID | 373 |
Feature activation | -4.90e-01 |
Pos logit contributions | |
he | +0.263 |
note | +0.235 |
answer | +0.232 |
fairy | +0.227 |
guardian | +0.226 |
Neg logit contributions | |
tanks | -0.341 |
ashed | -0.336 |
ksh | -0.335 |
rods | -0.325 |
trucks | -0.311 |
Token | very |
Token ID | 845 |
Feature activation | -3.42e-01 |
Pos logit contributions | |
ashed | +0.027 |
tanks | +0.027 |
ksh | +0.025 |
sizes | +0.024 |
tails | +0.024 |
Neg logit contributions | |
he | -0.030 |
note | -0.027 |
answer | -0.026 |
fairy | -0.026 |
orn | -0.025 |
Token | happy |
Token ID | 3772 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
ashed | +0.023 |
tanks | +0.022 |
ksh | +0.021 |
rods | +0.021 |
mouths | +0.020 |
Neg logit contributions | |
he | -0.017 |
note | -0.015 |
answer | -0.015 |
ad | -0.015 |
fairy | -0.014 |
Token | that |
Token ID | 326 |
Feature activation | +5.81e+00 |
Pos logit contributions | |
he | +0.006 |
note | +0.006 |
answer | +0.006 |
fairy | +0.006 |
voice | +0.005 |
Neg logit contributions | |
ksh | -0.012 |
ashed | -0.012 |
tanks | -0.012 |
rods | -0.011 |
suits | -0.011 |
Token | he |
Token ID | 339 |
Feature activation | +2.87e-01 |
Pos logit contributions | |
he | +0.285 |
answer | +0.272 |
fairy | +0.263 |
note | +0.261 |
figure | +0.250 |
Neg logit contributions | |
ksh | -0.418 |
ashed | -0.415 |
suits | -0.394 |
rods | -0.381 |
tanks | -0.378 |
Token | got |
Token ID | 1392 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
he | +0.017 |
fairy | +0.015 |
answer | +0.015 |
note | +0.015 |
guardian | +0.015 |
Neg logit contributions | |
tanks | -0.017 |
rods | -0.016 |
ksh | -0.016 |
ashed | -0.016 |
mouths | -0.016 |
Token | to |
Token ID | 284 |
Feature activation | -2.73e+00 |
Pos logit contributions | |
tanks | +0.142 |
ashed | +0.142 |
ksh | +0.135 |
rods | +0.132 |
trucks | +0.127 |
Neg logit contributions | |
he | -0.166 |
note | -0.151 |
answer | -0.148 |
guardian | -0.147 |
fairy | -0.145 |
Token | could |
Token ID | 714 |
Feature activation | -8.64e+00 |
Pos logit contributions | |
tanks | +0.405 |
ksh | +0.399 |
rods | +0.393 |
ashed | +0.392 |
pic | +0.367 |
Neg logit contributions | |
he | -0.471 |
answer | -0.425 |
note | -0.414 |
fairy | -0.414 |
guardian | -0.410 |
Token | not |
Token ID | 407 |
Feature activation | +4.71e+00 |
Pos logit contributions | |
tanks | +0.503 |
ashed | +0.492 |
pic | +0.489 |
they | +0.477 |
trucks | +0.476 |
Neg logit contributions | |
he | -0.454 |
orn | -0.433 |
ibrarian | -0.427 |
ocket | -0.421 |
note | -0.421 |
Token | go |
Token ID | 467 |
Feature activation | -4.39e+00 |
Pos logit contributions | |
ibrarian | +0.244 |
he | +0.242 |
note | +0.237 |
orn | +0.233 |
guardian | +0.228 |
Neg logit contributions | |
tanks | -0.268 |
ashed | -0.257 |
trucks | -0.246 |
ksh | -0.242 |
they | -0.241 |
Token | outside |
Token ID | 2354 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
ashed | +0.314 |
tanks | +0.310 |
ksh | +0.310 |
rods | +0.300 |
mouths | +0.292 |
Neg logit contributions | |
he | -0.200 |
answer | -0.178 |
note | -0.175 |
guardian | -0.168 |
fairy | -0.167 |
Token | . |
Token ID | 13 |
Feature activation | -3.37e-01 |
Pos logit contributions | |
he | +0.056 |
answer | +0.048 |
note | +0.043 |
voice | +0.041 |
fairy | +0.041 |
Neg logit contributions | |
ashed | -0.133 |
ksh | -0.129 |
tanks | -0.126 |
rods | -0.125 |
mouths | -0.121 |
Token | They |
Token ID | 1119 |
Feature activation | -6.87e+00 |
Pos logit contributions | |
trucks | +0.017 |
ashed | +0.016 |
tanks | +0.016 |
ksh | +0.016 |
sizes | +0.016 |
Neg logit contributions | |
he | -0.021 |
voice | -0.020 |
answer | -0.020 |
mirror | -0.019 |
note | -0.018 |
Token | looked |
Token ID | 3114 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
tanks | +0.385 |
ksh | +0.378 |
ashed | +0.375 |
rods | +0.368 |
pic | +0.347 |
Neg logit contributions | |
he | -0.421 |
answer | -0.385 |
note | -0.374 |
fairy | -0.367 |
guardian | -0.366 |
Token | at |
Token ID | 379 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
he | +0.048 |
answer | +0.045 |
note | +0.040 |
fairy | +0.040 |
guardian | +0.040 |
Neg logit contributions | |
ashed | -0.086 |
ksh | -0.086 |
rods | -0.085 |
tanks | -0.085 |
sizes | -0.080 |
Token | their |
Token ID | 511 |
Feature activation | -5.91e+00 |
Pos logit contributions | |
ashed | +0.073 |
ksh | +0.072 |
tanks | +0.070 |
rods | +0.066 |
suits | +0.065 |
Neg logit contributions | |
he | -0.067 |
answer | -0.062 |
note | -0.060 |
fairy | -0.059 |
voice | -0.058 |
Token | toys |
Token ID | 14958 |
Feature activation | -6.57e+00 |
Pos logit contributions | |
ashed | +0.376 |
ksh | +0.374 |
tanks | +0.369 |
rods | +0.345 |
sizes | +0.341 |
Neg logit contributions | |
he | -0.305 |
note | -0.292 |
answer | -0.284 |
voice | -0.275 |
guardian | -0.274 |
Token | , |
Token ID | 11 |
Feature activation | -2.14e+00 |
Pos logit contributions | |
ksh | +0.522 |
ashed | +0.520 |
tanks | +0.503 |
rods | +0.479 |
pic | +0.474 |
Neg logit contributions | |
he | -0.272 |
answer | -0.236 |
note | -0.235 |
figure | -0.213 |
fairy | -0.208 |
Token | snow |
Token ID | 6729 |
Feature activation | -6.65e+00 |
Pos logit contributions | |
ksh | +0.212 |
ashed | +0.202 |
tanks | +0.184 |
First | +0.180 |
rods | +0.177 |
Neg logit contributions | |
fairy | -0.129 |
voice | -0.128 |
ad | -0.127 |
answer | -0.124 |
mirror | -0.123 |
Token | fl |
Token ID | 2704 |
Feature activation | -4.09e+00 |
Pos logit contributions | |
ksh | +0.500 |
ashed | +0.495 |
tanks | +0.479 |
First | +0.461 |
pic | +0.454 |
Neg logit contributions | |
he | -0.264 |
note | -0.235 |
answer | -0.228 |
fairy | -0.227 |
ad | -0.220 |
Token | akes |
Token ID | 1124 |
Feature activation | -4.21e+00 |
Pos logit contributions | |
tanks | +0.223 |
ashed | +0.207 |
ksh | +0.205 |
rods | +0.199 |
trucks | +0.194 |
Neg logit contributions | |
he | -0.279 |
ad | -0.230 |
orn | -0.218 |
answer | -0.218 |
iver | -0.216 |
Token | and |
Token ID | 290 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
ksh | +0.315 |
ashed | +0.311 |
tanks | +0.304 |
rods | +0.291 |
mouths | +0.283 |
Neg logit contributions | |
he | -0.189 |
answer | -0.161 |
note | -0.159 |
fairy | -0.154 |
figure | -0.145 |
Token | colorful |
Token ID | 20239 |
Feature activation | +3.59e-01 |
Pos logit contributions | |
answer | +0.060 |
fairy | +0.058 |
note | +0.057 |
ad | +0.055 |
voice | +0.055 |
Neg logit contributions | |
ashed | -0.076 |
ksh | -0.075 |
tanks | -0.069 |
rods | -0.066 |
suits | -0.066 |
Token | fish |
Token ID | 5916 |
Feature activation | -2.73e+00 |
Pos logit contributions | |
answer | +0.019 |
fairy | +0.019 |
ad | +0.018 |
mirror | +0.018 |
note | +0.018 |
Neg logit contributions | |
ksh | -0.022 |
ashed | -0.021 |
tanks | -0.019 |
First | -0.018 |
rods | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +3.33e-01 |
Pos logit contributions | |
ksh | +0.214 |
ashed | +0.209 |
tanks | +0.206 |
rods | +0.199 |
mouths | +0.194 |
Neg logit contributions | |
he | -0.100 |
note | -0.092 |
fairy | -0.089 |
answer | -0.089 |
guardian | -0.084 |
Token |
|
Token ID | 198 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
he | +0.015 |
note | +0.013 |
answer | +0.013 |
guardian | +0.013 |
fairy | +0.013 |
Neg logit contributions | |
tanks | -0.024 |
ashed | -0.024 |
ksh | -0.023 |
rods | -0.023 |
trucks | -0.022 |
Token |
|
Token ID | 198 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
guardian | +0.129 |
angel | +0.124 |
wizard | +0.117 |
answer | +0.116 |
note | +0.115 |
Neg logit contributions | |
ashed | -0.111 |
they | -0.110 |
tanks | -0.110 |
rods | -0.107 |
pic | -0.106 |
Token | Soon |
Token ID | 28093 |
Feature activation | -4.41e+00 |
Pos logit contributions | |
he | +0.130 |
answer | +0.124 |
guardian | +0.122 |
note | +0.120 |
angel | +0.118 |
Neg logit contributions | |
ashed | -0.115 |
tanks | -0.111 |
rods | -0.106 |
ksh | -0.103 |
pic | -0.101 |
Token | , |
Token ID | 11 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
ksh | +0.380 |
tanks | +0.349 |
sizes | +0.336 |
trucks | +0.336 |
suits | +0.329 |
Neg logit contributions | |
fairy | -0.151 |
note | -0.143 |
he | -0.142 |
answer | -0.138 |
ad | -0.137 |
Token | exercise |
Token ID | 5517 |
Feature activation | -1.10e+01 |
Pos logit contributions | |
ksh | +0.392 |
ashed | +0.391 |
tanks | +0.378 |
suits | +0.346 |
sizes | +0.343 |
Neg logit contributions | |
answer | -0.306 |
note | -0.305 |
he | -0.294 |
guardian | -0.286 |
voice | -0.281 |
Token | . |
Token ID | 13 |
Feature activation | -3.66e-01 |
Pos logit contributions | |
ksh | +0.811 |
ashed | +0.780 |
tanks | +0.764 |
mouths | +0.722 |
rods | +0.703 |
Neg logit contributions | |
he | -0.536 |
note | -0.467 |
answer | -0.452 |
figure | -0.425 |
fairy | -0.425 |
Token | And |
Token ID | 843 |
Feature activation | -5.09e+00 |
Pos logit contributions | |
ashed | +0.023 |
ksh | +0.022 |
tanks | +0.021 |
rods | +0.021 |
suits | +0.021 |
Neg logit contributions | |
he | -0.020 |
answer | -0.017 |
guardian | -0.017 |
note | -0.017 |
ad | -0.017 |
Token | from |
Token ID | 422 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
ksh | +0.299 |
ashed | +0.283 |
suits | +0.263 |
tanks | +0.260 |
rods | +0.253 |
Neg logit contributions | |
he | -0.313 |
note | -0.292 |
figure | -0.291 |
answer | -0.290 |
guardian | -0.290 |
Token | that |
Token ID | 326 |
Feature activation | +5.51e+00 |
Pos logit contributions | |
he | +0.066 |
note | +0.058 |
answer | +0.058 |
guardian | +0.056 |
fairy | +0.055 |
Neg logit contributions | |
ashed | -0.107 |
tanks | -0.107 |
ksh | -0.105 |
rods | -0.102 |
mouths | -0.099 |
Token | day |
Token ID | 1110 |
Feature activation | -2.60e+00 |
Pos logit contributions | |
he | +0.356 |
note | +0.326 |
answer | +0.323 |
guardian | +0.318 |
fairy | +0.314 |
Neg logit contributions | |
tanks | -0.289 |
ashed | -0.289 |
ksh | -0.280 |
rods | -0.270 |
trucks | -0.258 |
Token | on |
Token ID | 319 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
ksh | +0.222 |
tanks | +0.212 |
ashed | +0.209 |
sizes | +0.207 |
pic | +0.205 |
Neg logit contributions | |
he | -0.096 |
answer | -0.075 |
note | -0.074 |
figure | -0.069 |
guardian | -0.069 |
Token | , |
Token ID | 11 |
Feature activation | -2.14e+00 |
Pos logit contributions | |
ksh | +0.013 |
ashed | +0.012 |
suits | +0.011 |
tanks | +0.011 |
sizes | +0.011 |
Neg logit contributions | |
he | -0.007 |
fairy | -0.006 |
note | -0.006 |
voice | -0.006 |
answer | -0.006 |
Token | Billy |
Token ID | 15890 |
Feature activation | -2.90e+00 |
Pos logit contributions | |
ksh | +0.142 |
ashed | +0.137 |
suits | +0.134 |
rods | +0.119 |
mouths | +0.116 |
Neg logit contributions | |
he | -0.116 |
figure | -0.106 |
guardian | -0.106 |
answer | -0.105 |
owl | -0.105 |
Token | always |
Token ID | 1464 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
tanks | +0.186 |
ksh | +0.185 |
ashed | +0.184 |
rods | +0.179 |
mouths | +0.171 |
Neg logit contributions | |
he | -0.150 |
answer | -0.133 |
note | -0.133 |
fairy | -0.132 |
guardian | -0.131 |
Token | remembered |
Token ID | 12086 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
he | +0.051 |
answer | +0.050 |
note | +0.049 |
fairy | +0.048 |
guardian | +0.048 |
Neg logit contributions | |
ksh | -0.069 |
tanks | -0.069 |
rods | -0.067 |
ashed | -0.066 |
trucks | -0.063 |
Token | then |
Token ID | 788 |
Feature activation | -5.47e+00 |
Pos logit contributions | |
ashed | +0.032 |
ksh | +0.031 |
tanks | +0.029 |
suits | +0.029 |
sizes | +0.029 |
Neg logit contributions | |
he | -0.023 |
answer | -0.022 |
fairy | -0.022 |
wizard | -0.022 |
note | -0.022 |
Token | another |
Token ID | 1194 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
ksh | +0.396 |
ashed | +0.393 |
tanks | +0.365 |
suits | +0.354 |
sizes | +0.353 |
Neg logit contributions | |
he | -0.242 |
fairy | -0.241 |
note | -0.239 |
answer | -0.239 |
wizard | -0.235 |
Token | man |
Token ID | 582 |
Feature activation | +7.05e+00 |
Pos logit contributions | |
tanks | +0.071 |
ashed | +0.070 |
ksh | +0.068 |
rods | +0.067 |
trucks | +0.065 |
Neg logit contributions | |
he | -0.049 |
note | -0.043 |
answer | -0.042 |
guardian | -0.041 |
fairy | -0.040 |
Token | came |
Token ID | 1625 |
Feature activation | +5.08e-01 |
Pos logit contributions | |
he | +0.364 |
note | +0.325 |
answer | +0.322 |
guardian | +0.315 |
fairy | +0.311 |
Neg logit contributions | |
tanks | -0.456 |
ashed | -0.455 |
ksh | -0.441 |
rods | -0.431 |
trucks | -0.417 |
Token | along |
Token ID | 1863 |
Feature activation | -3.22e+00 |
Pos logit contributions | |
he | +0.029 |
note | +0.025 |
ibrarian | +0.025 |
guardian | +0.024 |
fairy | +0.024 |
Neg logit contributions | |
tanks | -0.031 |
ashed | -0.030 |
ksh | -0.030 |
rods | -0.029 |
trucks | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | -2.14e+00 |
Pos logit contributions | |
ashed | +0.227 |
tanks | +0.226 |
ksh | +0.221 |
rods | +0.215 |
mouths | +0.209 |
Neg logit contributions | |
he | -0.149 |
note | -0.131 |
answer | -0.130 |
guardian | -0.126 |
fairy | -0.125 |
Token | who |
Token ID | 508 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
tanks | +0.138 |
ashed | +0.138 |
ksh | +0.134 |
rods | +0.131 |
trucks | +0.126 |
Neg logit contributions | |
he | -0.111 |
note | -0.100 |
answer | -0.099 |
guardian | -0.097 |
fairy | -0.096 |
Token | saw |
Token ID | 2497 |
Feature activation | +4.24e+00 |
Pos logit contributions | |
tanks | +0.008 |
ashed | +0.007 |
rods | +0.007 |
ksh | +0.007 |
trucks | +0.007 |
Neg logit contributions | |
he | -0.008 |
answer | -0.007 |
note | -0.007 |
guardian | -0.007 |
iver | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
he | +0.207 |
note | +0.185 |
answer | +0.185 |
fairy | +0.179 |
ad | +0.178 |
Neg logit contributions | |
ashed | -0.291 |
ksh | -0.287 |
tanks | -0.278 |
rods | -0.264 |
suits | -0.263 |
Token | bottle |
Token ID | 9294 |
Feature activation | +8.31e-01 |
Pos logit contributions | |
tanks | +0.079 |
rods | +0.074 |
ashed | +0.074 |
trucks | +0.071 |
ksh | +0.071 |
Neg logit contributions | |
he | -0.056 |
ad | -0.051 |
ibrarian | -0.049 |
guardian | -0.048 |
iver | -0.046 |
Token | and |
Token ID | 290 |
Feature activation | +5.24e-01 |
Pos logit contributions | |
he | +0.039 |
guardian | +0.036 |
note | +0.035 |
answer | +0.035 |
ibrarian | +0.034 |
Neg logit contributions | |
tanks | -0.055 |
ashed | -0.055 |
rods | -0.052 |
ksh | -0.052 |
trucks | -0.051 |
Token | . |
Token ID | 13 |
Feature activation | -5.19e-02 |
Pos logit contributions | |
tanks | +0.427 |
ksh | +0.427 |
ashed | +0.415 |
rods | +0.402 |
mouths | +0.399 |
Neg logit contributions | |
he | -0.251 |
note | -0.211 |
answer | -0.210 |
fairy | -0.199 |
ibrarian | -0.190 |
Token | And |
Token ID | 843 |
Feature activation | -4.75e+00 |
Pos logit contributions | |
ashed | +0.003 |
tanks | +0.003 |
rods | +0.003 |
trucks | +0.003 |
suits | +0.003 |
Neg logit contributions | |
he | -0.003 |
answer | -0.003 |
fairy | -0.003 |
guardian | -0.003 |
voice | -0.003 |
Token | maybe |
Token ID | 3863 |
Feature activation | -4.85e+00 |
Pos logit contributions | |
ashed | +0.331 |
ksh | +0.312 |
tanks | +0.300 |
suits | +0.287 |
rods | +0.276 |
Neg logit contributions | |
answer | -0.246 |
note | -0.237 |
fairy | -0.229 |
he | -0.224 |
figure | -0.224 |
Token | some |
Token ID | 617 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
ashed | +0.304 |
ksh | +0.284 |
suits | +0.255 |
tanks | +0.253 |
ashing | +0.237 |
Neg logit contributions | |
answer | -0.302 |
note | -0.290 |
fairy | -0.290 |
ring | -0.275 |
figure | -0.269 |
Token | candy |
Token ID | 18550 |
Feature activation | -5.28e+00 |
Pos logit contributions | |
ksh | +0.108 |
ashed | +0.102 |
tanks | +0.095 |
suits | +0.093 |
rods | +0.088 |
Neg logit contributions | |
answer | -0.104 |
note | -0.100 |
he | -0.098 |
fairy | -0.097 |
voice | -0.092 |
Token | later |
Token ID | 1568 |
Feature activation | -1.15e+01 |
Pos logit contributions | |
ksh | +0.376 |
tanks | +0.361 |
ashed | +0.357 |
sizes | +0.338 |
rods | +0.335 |
Neg logit contributions | |
he | -0.261 |
note | -0.235 |
answer | -0.225 |
fairy | -0.217 |
ibrarian | -0.212 |
Token | , |
Token ID | 11 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
ksh | +0.968 |
tanks | +0.946 |
sizes | +0.929 |
ashed | +0.921 |
rods | +0.884 |
Neg logit contributions | |
he | -0.446 |
answer | -0.397 |
note | -0.377 |
voice | -0.337 |
ibrarian | -0.302 |
Token | if |
Token ID | 611 |
Feature activation | -4.64e+00 |
Pos logit contributions | |
ashed | +0.122 |
ksh | +0.114 |
suits | +0.109 |
tanks | +0.108 |
sizes | +0.102 |
Neg logit contributions | |
answer | -0.095 |
fairy | -0.087 |
figure | -0.085 |
note | -0.084 |
he | -0.083 |
Token | you |
Token ID | 345 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
ksh | +0.358 |
ashed | +0.352 |
suits | +0.319 |
ashing | +0.308 |
sizes | +0.308 |
Neg logit contributions | |
note | -0.219 |
fairy | -0.217 |
answer | -0.213 |
voice | -0.193 |
story | -0.190 |
Token | behave |
Token ID | 17438 |
Feature activation | -2.95e+00 |
Pos logit contributions | |
ashed | +0.121 |
tanks | +0.121 |
ksh | +0.117 |
rods | +0.114 |
mouths | +0.110 |
Neg logit contributions | |
he | -0.093 |
note | -0.084 |
guardian | -0.082 |
answer | -0.081 |
fairy | -0.080 |
Token | ." |
Token ID | 526 |
Feature activation | -3.73e+00 |
Pos logit contributions | |
ashed | +0.194 |
tanks | +0.194 |
ksh | +0.189 |
rods | +0.185 |
trucks | +0.180 |
Neg logit contributions | |
he | -0.153 |
note | -0.136 |
answer | -0.135 |
fairy | -0.128 |
guardian | -0.126 |
Token | were |
Token ID | 547 |
Feature activation | -4.35e+00 |
Pos logit contributions | |
ksh | +0.426 |
rods | +0.415 |
tanks | +0.409 |
pic | +0.406 |
mouths | +0.382 |
Neg logit contributions | |
he | -0.395 |
answer | -0.355 |
fairy | -0.334 |
figure | -0.329 |
ad | -0.327 |
Token | playing |
Token ID | 2712 |
Feature activation | -6.15e+00 |
Pos logit contributions | |
ksh | +0.281 |
ashed | +0.279 |
tanks | +0.267 |
rods | +0.267 |
suits | +0.265 |
Neg logit contributions | |
he | -0.223 |
answer | -0.202 |
guardian | -0.193 |
note | -0.191 |
fairy | -0.189 |
Token | in |
Token ID | 287 |
Feature activation | -8.74e-02 |
Pos logit contributions | |
ksh | +0.423 |
tanks | +0.409 |
sizes | +0.405 |
ashed | +0.403 |
mouths | +0.397 |
Neg logit contributions | |
he | -0.319 |
note | -0.270 |
figure | -0.264 |
answer | -0.264 |
fairy | -0.255 |
Token | the |
Token ID | 262 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
ashed | +0.006 |
ksh | +0.006 |
tanks | +0.006 |
rods | +0.006 |
mouths | +0.006 |
Neg logit contributions | |
he | -0.004 |
answer | -0.004 |
note | -0.004 |
fairy | -0.004 |
ad | -0.003 |
Token | park |
Token ID | 3952 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
tanks | +0.081 |
ashed | +0.081 |
ksh | +0.079 |
rods | +0.077 |
trucks | +0.075 |
Neg logit contributions | |
he | -0.049 |
note | -0.043 |
answer | -0.043 |
guardian | -0.042 |
fairy | -0.041 |
Token | together |
Token ID | 1978 |
Feature activation | -9.36e+00 |
Pos logit contributions | |
he | +0.053 |
answer | +0.046 |
note | +0.046 |
figure | +0.043 |
voice | +0.042 |
Neg logit contributions | |
ksh | -0.085 |
tanks | -0.082 |
ashed | -0.079 |
pic | -0.078 |
mouths | -0.076 |
Token | when |
Token ID | 618 |
Feature activation | -6.70e+00 |
Pos logit contributions | |
ksh | +0.774 |
tanks | +0.765 |
ashed | +0.757 |
rods | +0.711 |
trucks | +0.710 |
Neg logit contributions | |
he | -0.351 |
answer | -0.305 |
note | -0.286 |
figure | -0.269 |
fairy | -0.262 |
Token | Max |
Token ID | 5436 |
Feature activation | -4.38e+00 |
Pos logit contributions | |
ksh | +0.361 |
ashed | +0.333 |
sizes | +0.299 |
suits | +0.293 |
appet | +0.288 |
Neg logit contributions | |
he | -0.444 |
fairy | -0.430 |
note | -0.417 |
answer | -0.413 |
reply | -0.396 |
Token | said |
Token ID | 531 |
Feature activation | +3.77e+00 |
Pos logit contributions | |
ksh | +0.342 |
tanks | +0.335 |
rods | +0.329 |
ashed | +0.328 |
sizes | +0.321 |
Neg logit contributions | |
he | -0.161 |
answer | -0.137 |
figure | -0.137 |
fairy | -0.136 |
note | -0.135 |
Token | , |
Token ID | 11 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
he | +0.172 |
note | +0.171 |
answer | +0.168 |
voice | +0.162 |
fairy | +0.160 |
Neg logit contributions | |
ksh | -0.269 |
tanks | -0.254 |
ashed | -0.252 |
pic | -0.240 |
rods | -0.239 |
Token | " |
Token ID | 366 |
Feature activation | -6.52e+00 |
Pos logit contributions | |
ashed | +0.117 |
ksh | +0.112 |
tanks | +0.111 |
rods | +0.108 |
mouths | +0.106 |
Neg logit contributions | |
he | -0.125 |
note | -0.121 |
fairy | -0.121 |
answer | -0.120 |
wizard | -0.119 |
Token | . |
Token ID | 13 |
Feature activation | -3.86e-02 |
Pos logit contributions | |
ashed | +0.215 |
ksh | +0.214 |
tanks | +0.212 |
rods | +0.203 |
mouths | +0.199 |
Neg logit contributions | |
he | -0.116 |
answer | -0.097 |
note | -0.094 |
fairy | -0.088 |
ad | -0.087 |
Token |
|
Token ID | 198 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
ashed | +0.003 |
tanks | +0.003 |
ksh | +0.003 |
rods | +0.003 |
sizes | +0.003 |
Neg logit contributions | |
he | -0.001 |
answer | -0.001 |
note | -0.001 |
voice | -0.001 |
ad | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
guardian | +0.113 |
angel | +0.107 |
fairy | +0.107 |
wizard | +0.105 |
ocket | +0.105 |
Neg logit contributions | |
tanks | -0.086 |
ashed | -0.084 |
they | -0.082 |
rods | -0.081 |
pic | -0.081 |
Token | After |
Token ID | 3260 |
Feature activation | -3.85e+00 |
Pos logit contributions | |
he | +0.111 |
note | +0.096 |
answer | +0.093 |
figure | +0.092 |
guardian | +0.091 |
Neg logit contributions | |
tanks | -0.099 |
ksh | -0.098 |
ashed | -0.098 |
mouths | -0.095 |
rods | -0.093 |
Token | a |
Token ID | 257 |
Feature activation | +6.94e-01 |
Pos logit contributions | |
ksh | +0.322 |
sizes | +0.301 |
suits | +0.297 |
ashed | +0.288 |
appet | +0.276 |
Neg logit contributions | |
on | -0.159 |
he | -0.155 |
ad | -0.149 |
note | -0.136 |
answer | -0.134 |
Token | while |
Token ID | 981 |
Feature activation | -9.47e+00 |
Pos logit contributions | |
he | +0.051 |
note | +0.051 |
answer | +0.049 |
fairy | +0.049 |
voice | +0.048 |
Neg logit contributions | |
ksh | -0.029 |
tanks | -0.028 |
ashed | -0.028 |
mouths | -0.027 |
sizes | -0.026 |
Token | , |
Token ID | 11 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
ksh | +0.788 |
sizes | +0.728 |
ashed | +0.710 |
suits | +0.699 |
tanks | +0.692 |
Neg logit contributions | |
he | -0.390 |
answer | -0.340 |
fairy | -0.340 |
voice | -0.336 |
note | -0.324 |
Token | Tim |
Token ID | 5045 |
Feature activation | -7.07e+00 |
Pos logit contributions | |
ksh | +0.109 |
suits | +0.103 |
sizes | +0.099 |
ashed | +0.096 |
girls | +0.094 |
Neg logit contributions | |
he | -0.112 |
answer | -0.096 |
on | -0.095 |
hurt | -0.094 |
figure | -0.093 |
Token | my |
Token ID | 1820 |
Feature activation | +5.14e+00 |
Pos logit contributions | |
ashed | +0.518 |
tanks | +0.505 |
ksh | +0.485 |
trucks | +0.480 |
rods | +0.463 |
Neg logit contributions | |
he | -0.303 |
guardian | -0.274 |
note | -0.270 |
ibrarian | -0.258 |
fairy | -0.256 |
Token | 's |
Token ID | 338 |
Feature activation | +9.49e-01 |
Pos logit contributions | |
he | +0.219 |
note | +0.194 |
answer | +0.188 |
fairy | +0.186 |
figure | +0.185 |
Neg logit contributions | |
tanks | -0.375 |
ksh | -0.375 |
ashed | -0.362 |
rods | -0.358 |
sizes | -0.358 |
Token | friend |
Token ID | 1545 |
Feature activation | +3.15e+00 |
Pos logit contributions | |
he | +0.048 |
note | +0.043 |
answer | +0.043 |
guardian | +0.042 |
fairy | +0.041 |
Neg logit contributions | |
tanks | -0.063 |
ashed | -0.062 |
ksh | -0.062 |
rods | -0.059 |
sizes | -0.058 |
Token | . |
Token ID | 13 |
Feature activation | -3.39e-01 |
Pos logit contributions | |
he | +0.128 |
note | +0.115 |
answer | +0.114 |
guardian | +0.111 |
fairy | +0.110 |
Neg logit contributions | |
tanks | -0.180 |
ashed | -0.180 |
ksh | -0.177 |
rods | -0.172 |
mouths | -0.166 |
Token |
|
Token ID | 198 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
ashed | +0.022 |
tanks | +0.021 |
ksh | +0.021 |
rods | +0.021 |
trucks | +0.020 |
Neg logit contributions | |
he | -0.018 |
note | -0.016 |
answer | -0.015 |
fairy | -0.015 |
guardian | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
answer | +0.083 |
note | +0.081 |
guardian | +0.081 |
he | +0.079 |
angel | +0.079 |
Neg logit contributions | |
ashed | -0.081 |
tanks | -0.081 |
rods | -0.072 |
pic | -0.071 |
ksh | -0.071 |
Token | B |
Token ID | 33 |
Feature activation | -9.51e+00 |
Pos logit contributions | |
he | +0.108 |
answer | +0.101 |
note | +0.101 |
guardian | +0.099 |
fairy | +0.097 |
Neg logit contributions | |
tanks | -0.061 |
ashed | -0.061 |
ksh | -0.057 |
rods | -0.055 |
trucks | -0.053 |
Token | un |
Token ID | 403 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
tanks | +0.701 |
ashed | +0.700 |
ksh | +0.682 |
rods | +0.667 |
trucks | +0.648 |
Neg logit contributions | |
he | -0.407 |
note | -0.355 |
answer | -0.351 |
guardian | -0.342 |
fairy | -0.336 |
Token | B |
Token ID | 33 |
Feature activation | -9.49e+00 |
Pos logit contributions | |
he | +0.152 |
answer | +0.149 |
note | +0.146 |
guardian | +0.143 |
mirror | +0.140 |
Neg logit contributions | |
ashed | -0.119 |
tanks | -0.114 |
ksh | -0.100 |
trucks | -0.097 |
Nem | -0.093 |
Token | un |
Token ID | 403 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
tanks | +0.707 |
ashed | +0.705 |
ksh | +0.687 |
rods | +0.673 |
trucks | +0.654 |
Neg logit contributions | |
he | -0.398 |
note | -0.346 |
answer | -0.342 |
guardian | -0.333 |
fairy | -0.327 |
Token | was |
Token ID | 373 |
Feature activation | -8.56e-01 |
Pos logit contributions | |
he | +0.148 |
answer | +0.141 |
note | +0.140 |
guardian | +0.138 |
mirror | +0.134 |
Neg logit contributions | |
ashed | -0.125 |
tanks | -0.119 |
ksh | -0.108 |
trucks | -0.104 |
rods | -0.101 |
Token | heart |
Token ID | 2612 |
Feature activation | +6.38e+00 |
Pos logit contributions | |
tanks | +0.049 |
ashed | +0.049 |
tails | +0.043 |
trucks | +0.042 |
mouths | +0.042 |
Neg logit contributions | |
he | -0.049 |
ibrarian | -0.045 |
answer | -0.043 |
note | -0.043 |
wizard | -0.043 |
Token | broken |
Token ID | 25826 |
Feature activation | -8.08e-01 |
Pos logit contributions | |
he | +0.400 |
answer | +0.383 |
ibrarian | +0.383 |
note | +0.369 |
ad | +0.365 |
Neg logit contributions | |
ashed | -0.338 |
tanks | -0.318 |
ksh | -0.316 |
tails | -0.286 |
rods | -0.282 |
Token | and |
Token ID | 290 |
Feature activation | +5.91e-01 |
Pos logit contributions | |
tanks | +0.053 |
ashed | +0.053 |
ksh | +0.051 |
rods | +0.050 |
trucks | +0.049 |
Neg logit contributions | |
he | -0.041 |
note | -0.036 |
answer | -0.036 |
guardian | -0.035 |
fairy | -0.034 |
Token | make |
Token ID | 787 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
tanks | +0.452 |
rods | +0.431 |
trucks | +0.421 |
pic | +0.418 |
mouths | +0.411 |
Neg logit contributions | |
he | -0.476 |
ibrarian | -0.469 |
ocket | -0.446 |
note | -0.446 |
guardian | -0.442 |
Token | a |
Token ID | 257 |
Feature activation | +4.35e-01 |
Pos logit contributions | |
he | +0.105 |
note | +0.094 |
answer | +0.093 |
guardian | +0.092 |
fairy | +0.090 |
Neg logit contributions | |
tanks | -0.129 |
ashed | -0.127 |
ksh | -0.124 |
rods | -0.121 |
trucks | -0.117 |
Token | new |
Token ID | 649 |
Feature activation | -6.48e-01 |
Pos logit contributions | |
he | +0.025 |
ibrarian | +0.022 |
ad | +0.021 |
guardian | +0.020 |
answer | +0.020 |
Neg logit contributions | |
tanks | -0.027 |
rods | -0.025 |
ksh | -0.025 |
pic | -0.025 |
mouths | -0.025 |
Token | v |
Token ID | 410 |
Feature activation | -3.84e+00 |
Pos logit contributions | |
tanks | +0.040 |
rods | +0.038 |
trucks | +0.038 |
mouths | +0.038 |
ashed | +0.037 |
Neg logit contributions | |
he | -0.036 |
ibrarian | -0.032 |
ocket | -0.029 |
ad | -0.028 |
orn | -0.028 |
Token | ase |
Token ID | 589 |
Feature activation | +5.23e+00 |
Pos logit contributions | |
ashed | +0.233 |
ksh | +0.216 |
tanks | +0.210 |
rods | +0.204 |
suits | +0.196 |
Neg logit contributions | |
answer | -0.207 |
guardian | -0.198 |
reply | -0.196 |
wizard | -0.194 |
note | -0.193 |
Token | together |
Token ID | 1978 |
Feature activation | -9.33e+00 |
Pos logit contributions | |
he | +0.307 |
guardian | +0.278 |
note | +0.277 |
answer | +0.274 |
ibrarian | +0.273 |
Neg logit contributions | |
tanks | -0.291 |
ashed | -0.287 |
ksh | -0.277 |
rods | -0.271 |
trucks | -0.266 |
Token | . |
Token ID | 13 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
ashed | +0.732 |
tanks | +0.721 |
ksh | +0.713 |
rods | +0.686 |
mouths | +0.667 |
Neg logit contributions | |
he | -0.378 |
answer | -0.334 |
note | -0.332 |
fairy | -0.311 |
guardian | -0.304 |
Token | Lily |
Token ID | 20037 |
Feature activation | -4.93e+00 |
Pos logit contributions | |
tanks | +0.020 |
ashed | +0.020 |
rods | +0.019 |
ksh | +0.019 |
sizes | +0.019 |
Neg logit contributions | |
he | -0.018 |
note | -0.015 |
fairy | -0.015 |
voice | -0.015 |
answer | -0.015 |
Token | 's |
Token ID | 338 |
Feature activation | +5.75e-01 |
Pos logit contributions | |
tanks | +0.310 |
ashed | +0.309 |
ksh | +0.299 |
rods | +0.292 |
trucks | +0.282 |
Neg logit contributions | |
he | -0.265 |
note | -0.238 |
answer | -0.236 |
guardian | -0.231 |
fairy | -0.228 |
Token | face |
Token ID | 1986 |
Feature activation | +4.95e+00 |
Pos logit contributions | |
he | +0.038 |
guardian | +0.033 |
ibrarian | +0.033 |
note | +0.033 |
answer | +0.033 |
Neg logit contributions | |
tanks | -0.029 |
ashed | -0.028 |
mouths | -0.027 |
ksh | -0.027 |
rods | -0.027 |
Token | lit |
Token ID | 6578 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
he | +0.299 |
note | +0.273 |
answer | +0.271 |
guardian | +0.265 |
fairy | +0.264 |
Neg logit contributions | |
tanks | -0.278 |
ashed | -0.275 |
ksh | -0.267 |
rods | -0.261 |
trucks | -0.250 |