Token | steps |
Token ID | 4831 |
Feature activation | +2.79e-02 |
Pos logit contributions | |
pieces | +0.126 |
Maria | +0.120 |
od | +0.119 |
iece | +0.114 |
lar | +0.110 |
Neg logit contributions | |
cooler | -0.079 |
robe | -0.072 |
arl | -0.070 |
melon | -0.068 |
Cream | -0.064 |
Token | - |
Token ID | 532 |
Feature activation | -2.02e-02 |
Pos logit contributions | |
pieces | +0.046 |
od | +0.041 |
Maria | +0.040 |
one | +0.036 |
enta | +0.036 |
Neg logit contributions | |
robe | -0.050 |
arl | -0.046 |
melon | -0.044 |
cooler | -0.043 |
ugg | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
robe | +0.032 |
cooler | +0.029 |
arl | +0.029 |
melon | +0.029 |
ugg | +0.027 |
Neg logit contributions | |
pieces | -0.036 |
od | -0.033 |
Maria | -0.032 |
inda | -0.030 |
lar | -0.029 |
Token | see |
Token ID | 766 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
pieces | +0.034 |
od | +0.032 |
Maria | +0.031 |
one | +0.029 |
lier | +0.029 |
Neg logit contributions | |
robe | -0.028 |
cooler | -0.026 |
melon | -0.026 |
arl | -0.024 |
ugg | -0.023 |
Token | what |
Token ID | 644 |
Feature activation | -4.91e-02 |
Pos logit contributions | |
pieces | +0.200 |
od | +0.182 |
Maria | +0.181 |
ata | +0.170 |
enta | +0.170 |
Neg logit contributions | |
robe | -0.137 |
cooler | -0.128 |
melon | -0.121 |
arl | -0.119 |
cabin | -0.115 |
Token | other |
Token ID | 584 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
robe | +0.071 |
cooler | +0.064 |
melon | +0.063 |
Cream | +0.059 |
Sleep | +0.058 |
Neg logit contributions | |
pieces | -0.094 |
od | -0.083 |
Maria | -0.081 |
ata | -0.077 |
enta | -0.077 |
Token | y |
Token ID | 331 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
pieces | +0.374 |
od | +0.341 |
Maria | +0.337 |
enta | +0.320 |
ata | +0.319 |
Neg logit contributions | |
robe | -0.268 |
cooler | -0.255 |
melon | -0.236 |
cabin | -0.220 |
Cream | -0.218 |
Token | ummy |
Token ID | 13513 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
pieces | +0.027 |
od | +0.025 |
Maria | +0.024 |
lier | +0.024 |
ill | +0.023 |
Neg logit contributions | |
robe | -0.020 |
arl | -0.017 |
ugg | -0.015 |
cooler | -0.015 |
ranc | -0.015 |
Token | food |
Token ID | 2057 |
Feature activation | +9.78e-03 |
Pos logit contributions | |
pieces | +0.021 |
Maria | +0.019 |
od | +0.019 |
enta | +0.018 |
inda | +0.017 |
Neg logit contributions | |
robe | -0.014 |
cooler | -0.014 |
melon | -0.013 |
Cream | -0.012 |
arl | -0.012 |
Token | they |
Token ID | 484 |
Feature activation | +3.04e-02 |
Pos logit contributions | |
pieces | +0.017 |
od | +0.015 |
Maria | +0.015 |
enta | +0.013 |
one | +0.013 |
Neg logit contributions | |
robe | -0.017 |
melon | -0.015 |
cooler | -0.015 |
arl | -0.015 |
ugg | -0.014 |
Token | could |
Token ID | 714 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
pieces | +0.056 |
Maria | +0.049 |
od | +0.048 |
one | +0.047 |
circles | +0.046 |
Neg logit contributions | |
robe | -0.047 |
melon | -0.043 |
cooler | -0.042 |
arl | -0.040 |
ugg | -0.039 |
Token | y |
Token ID | 88 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
robe | +0.011 |
arl | +0.011 |
melon | +0.010 |
cooler | +0.009 |
ugg | +0.009 |
Neg logit contributions | |
pieces | -0.008 |
od | -0.007 |
Maria | -0.007 |
enta | -0.006 |
one | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | +6.73e-03 |
Pos logit contributions | |
robe | +0.040 |
arl | +0.037 |
cooler | +0.035 |
melon | +0.035 |
ugg | +0.033 |
Neg logit contributions | |
pieces | -0.034 |
Maria | -0.030 |
od | -0.030 |
enta | -0.027 |
one | -0.026 |
Token | scared |
Token ID | 12008 |
Feature activation | -3.10e-03 |
Pos logit contributions | |
pieces | +0.013 |
od | +0.012 |
circles | +0.011 |
Maria | +0.011 |
lier | +0.011 |
Neg logit contributions | |
robe | -0.010 |
arl | -0.009 |
melon | -0.008 |
ugg | -0.008 |
Cream | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
robe | +0.004 |
arl | +0.004 |
cooler | +0.004 |
melon | +0.004 |
sorry | +0.003 |
Neg logit contributions | |
pieces | -0.006 |
Maria | -0.005 |
enta | -0.005 |
od | -0.005 |
lar | -0.005 |
Token | " |
Token ID | 366 |
Feature activation | +1.75e-03 |
Pos logit contributions | |
pieces | +0.028 |
od | +0.025 |
Maria | +0.024 |
lier | +0.023 |
one | +0.023 |
Neg logit contributions | |
robe | -0.021 |
arl | -0.019 |
ugg | -0.017 |
cooler | -0.017 |
melon | -0.017 |
Token | Ah |
Token ID | 10910 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
pieces | +0.003 |
od | +0.003 |
Maria | +0.003 |
lier | +0.003 |
inda | +0.002 |
Neg logit contributions | |
robe | -0.003 |
arl | -0.003 |
ugg | -0.002 |
cooler | -0.002 |
Cream | -0.002 |
Token | h |
Token ID | 71 |
Feature activation | +2.88e-03 |
Pos logit contributions | |
pieces | +0.332 |
Maria | +0.299 |
od | +0.267 |
Rachel | +0.257 |
created | +0.244 |
Neg logit contributions | |
arl | -0.320 |
robe | -0.300 |
cooler | -0.285 |
ugg | -0.284 |
sorry | -0.274 |
Token | ! |
Token ID | 0 |
Feature activation | +3.34e-02 |
Pos logit contributions | |
pieces | +0.004 |
od | +0.004 |
Maria | +0.004 |
lier | +0.003 |
one | +0.003 |
Neg logit contributions | |
robe | -0.006 |
arl | -0.005 |
melon | -0.005 |
cooler | -0.005 |
ugg | -0.005 |
Token | This |
Token ID | 770 |
Feature activation | -1.14e-03 |
Pos logit contributions | |
pieces | +0.061 |
od | +0.055 |
Maria | +0.053 |
lier | +0.049 |
one | +0.049 |
Neg logit contributions | |
robe | -0.054 |
arl | -0.049 |
melon | -0.046 |
cooler | -0.045 |
ugg | -0.045 |
Token | is |
Token ID | 318 |
Feature activation | -7.49e-04 |
Pos logit contributions | |
robe | +0.002 |
cooler | +0.001 |
arl | +0.001 |
melon | +0.001 |
mall | +0.001 |
Neg logit contributions | |
pieces | -0.002 |
Maria | -0.002 |
od | -0.002 |
lar | -0.002 |
inda | -0.002 |
Token | not |
Token ID | 407 |
Feature activation | -8.19e-02 |
Pos logit contributions | |
robe | +0.001 |
arl | +0.001 |
melon | +0.001 |
ugg | +0.001 |
cooler | +0.001 |
Neg logit contributions | |
pieces | -0.002 |
od | -0.001 |
Maria | -0.001 |
one | -0.001 |
lier | -0.001 |
Token | just |
Token ID | 655 |
Feature activation | +5.03e-02 |
Pos logit contributions | |
pieces | +0.064 |
od | +0.056 |
Maria | +0.056 |
lier | +0.050 |
enta | +0.050 |
Neg logit contributions | |
robe | -0.060 |
arl | -0.054 |
cooler | -0.052 |
melon | -0.050 |
ugg | -0.049 |
Token | need |
Token ID | 761 |
Feature activation | +5.34e-02 |
Pos logit contributions | |
pieces | +0.099 |
Maria | +0.090 |
od | +0.089 |
Rachel | +0.081 |
lier | +0.080 |
Neg logit contributions | |
robe | -0.072 |
arl | -0.065 |
melon | -0.063 |
cooler | -0.062 |
Sleep | -0.060 |
Token | to |
Token ID | 284 |
Feature activation | +4.00e-02 |
Pos logit contributions | |
pieces | +0.118 |
od | +0.109 |
Maria | +0.102 |
iece | +0.099 |
ill | +0.096 |
Neg logit contributions | |
robe | -0.064 |
arl | -0.061 |
cooler | -0.059 |
Sleep | -0.059 |
ugg | -0.052 |
Token | help |
Token ID | 1037 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
pieces | +0.067 |
od | +0.060 |
Maria | +0.058 |
one | +0.053 |
lier | +0.053 |
Neg logit contributions | |
robe | -0.070 |
arl | -0.065 |
cooler | -0.061 |
melon | -0.060 |
ugg | -0.058 |
Token | each |
Token ID | 1123 |
Feature activation | -2.49e-02 |
Pos logit contributions | |
pieces | +0.028 |
od | +0.025 |
Maria | +0.024 |
rect | +0.023 |
ez | +0.023 |
Neg logit contributions | |
robe | -0.019 |
arl | -0.017 |
cooler | -0.016 |
Sleep | -0.016 |
Cream | -0.015 |
Token | other |
Token ID | 584 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
robe | +0.032 |
pack | +0.029 |
cooler | +0.029 |
waitress | +0.029 |
festival | +0.028 |
Neg logit contributions | |
pieces | -0.045 |
Maria | -0.044 |
od | -0.044 |
enta | -0.043 |
ill | -0.043 |
Token | ." |
Token ID | 526 |
Feature activation | +2.60e-03 |
Pos logit contributions | |
pieces | +0.318 |
od | +0.282 |
Maria | +0.279 |
enta | +0.250 |
one | +0.247 |
Neg logit contributions | |
robe | -0.349 |
arl | -0.314 |
cooler | -0.301 |
melon | -0.295 |
ugg | -0.282 |
Token |
|
Token ID | 198 |
Feature activation | +3.90e-02 |
Pos logit contributions | |
pieces | +0.006 |
lier | +0.005 |
ets | +0.005 |
ata | +0.005 |
lar | +0.005 |
Neg logit contributions | |
robe | -0.003 |
ugg | -0.003 |
arl | -0.003 |
Cream | -0.003 |
melon | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +3.84e-02 |
Pos logit contributions | |
pieces | +0.084 |
lier | +0.073 |
od | +0.071 |
ets | +0.071 |
asing | +0.069 |
Neg logit contributions | |
robe | -0.051 |
arl | -0.046 |
ugg | -0.043 |
Cream | -0.039 |
melon | -0.038 |
Token | B |
Token ID | 33 |
Feature activation | +6.50e-04 |
Pos logit contributions | |
pieces | +0.062 |
ets | +0.054 |
lier | +0.052 |
od | +0.050 |
pad | +0.050 |
Neg logit contributions | |
robe | -0.065 |
arl | -0.059 |
ugg | -0.058 |
cabin | -0.056 |
ranc | -0.055 |
Token | ert |
Token ID | 861 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
pieces | +0.001 |
od | +0.001 |
Maria | +0.001 |
lier | +0.001 |
minutes | +0.001 |
Neg logit contributions | |
robe | -0.001 |
arl | -0.001 |
ugg | -0.001 |
melon | -0.001 |
Cream | -0.001 |
Token | time |
Token ID | 640 |
Feature activation | +5.02e-02 |
Pos logit contributions | |
pieces | +0.121 |
Maria | +0.110 |
od | +0.108 |
enta | +0.101 |
asing | +0.098 |
Neg logit contributions | |
robe | -0.094 |
cooler | -0.088 |
arl | -0.086 |
melon | -0.081 |
ugg | -0.075 |
Token | , |
Token ID | 11 |
Feature activation | +6.38e-03 |
Pos logit contributions | |
pieces | +0.094 |
Maria | +0.085 |
od | +0.079 |
enta | +0.075 |
Rachel | +0.075 |
Neg logit contributions | |
robe | -0.077 |
arl | -0.069 |
cooler | -0.069 |
melon | -0.068 |
ugg | -0.063 |
Token | saying |
Token ID | 2282 |
Feature activation | +3.56e-02 |
Pos logit contributions | |
pieces | +0.011 |
od | +0.010 |
Maria | +0.009 |
lier | +0.008 |
one | +0.008 |
Neg logit contributions | |
robe | -0.011 |
cooler | -0.010 |
arl | -0.010 |
melon | -0.010 |
ugg | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | +6.94e-03 |
Pos logit contributions | |
pieces | +0.071 |
od | +0.064 |
Maria | +0.062 |
lier | +0.058 |
one | +0.058 |
Neg logit contributions | |
robe | -0.052 |
arl | -0.047 |
melon | -0.043 |
cooler | -0.043 |
ugg | -0.042 |
Token | " |
Token ID | 366 |
Feature activation | +5.56e-03 |
Pos logit contributions | |
pieces | +0.015 |
od | +0.013 |
Maria | +0.013 |
lier | +0.012 |
one | +0.012 |
Neg logit contributions | |
robe | -0.009 |
arl | -0.008 |
melon | -0.008 |
cooler | -0.008 |
ugg | -0.007 |
Token | Ah |
Token ID | 10910 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
pieces | +0.011 |
od | +0.010 |
Maria | +0.009 |
ets | +0.009 |
inda | +0.009 |
Neg logit contributions | |
robe | -0.008 |
arl | -0.007 |
ugg | -0.007 |
Sleep | -0.007 |
Cream | -0.007 |
Token | - |
Token ID | 12 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
pieces | +0.290 |
Maria | +0.264 |
od | +0.239 |
Rachel | +0.225 |
iece | +0.219 |
Neg logit contributions | |
arl | -0.333 |
robe | -0.323 |
ugg | -0.321 |
cooler | -0.307 |
melon | -0.293 |
Token | ch |
Token ID | 354 |
Feature activation | +3.79e-02 |
Pos logit contributions | |
robe | +0.024 |
ugg | +0.023 |
arl | +0.023 |
Cream | +0.021 |
melon | +0.021 |
Neg logit contributions | |
pieces | -0.017 |
od | -0.015 |
Maria | -0.015 |
circles | -0.013 |
enta | -0.013 |
Token | oo |
Token ID | 2238 |
Feature activation | -2.50e-02 |
Pos logit contributions | |
pieces | +0.068 |
od | +0.061 |
Maria | +0.059 |
one | +0.055 |
lier | +0.054 |
Neg logit contributions | |
robe | -0.062 |
arl | -0.056 |
cooler | -0.053 |
melon | -0.052 |
ugg | -0.051 |
Token | !" |
Token ID | 2474 |
Feature activation | +2.22e-02 |
Pos logit contributions | |
robe | +0.055 |
arl | +0.051 |
melon | +0.048 |
cooler | +0.048 |
ugg | +0.048 |
Neg logit contributions | |
pieces | -0.031 |
od | -0.027 |
Maria | -0.025 |
lier | -0.023 |
one | -0.023 |
Token | He |
Token ID | 679 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
pieces | +0.045 |
od | +0.041 |
Maria | +0.040 |
enta | +0.038 |
lar | +0.038 |
Neg logit contributions | |
robe | -0.032 |
arl | -0.028 |
cooler | -0.027 |
ugg | -0.027 |
melon | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | +2.42e-02 |
Pos logit contributions | |
pieces | +0.038 |
od | +0.033 |
Maria | +0.032 |
one | +0.028 |
lier | +0.028 |
Neg logit contributions | |
robe | -0.049 |
arl | -0.046 |
cooler | -0.043 |
melon | -0.043 |
ugg | -0.042 |
Token | You |
Token ID | 921 |
Feature activation | +3.51e-02 |
Pos logit contributions | |
pieces | +0.040 |
od | +0.035 |
Maria | +0.034 |
one | +0.032 |
lier | +0.031 |
Neg logit contributions | |
robe | -0.043 |
arl | -0.040 |
cooler | -0.037 |
melon | -0.037 |
ugg | -0.036 |
Token | can |
Token ID | 460 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
pieces | +0.057 |
od | +0.051 |
Maria | +0.049 |
lier | +0.045 |
one | +0.044 |
Neg logit contributions | |
robe | -0.064 |
arl | -0.059 |
cooler | -0.055 |
melon | -0.055 |
ugg | -0.054 |
Token | help |
Token ID | 1037 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
pieces | +0.034 |
od | +0.031 |
Maria | +0.030 |
lier | +0.027 |
one | +0.027 |
Neg logit contributions | |
robe | -0.032 |
arl | -0.029 |
melon | -0.027 |
cooler | -0.027 |
ugg | -0.027 |
Token | in |
Token ID | 287 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
pieces | +0.029 |
od | +0.025 |
Maria | +0.025 |
ata | +0.024 |
iece | +0.024 |
Neg logit contributions | |
robe | -0.015 |
arl | -0.014 |
cooler | -0.013 |
Cream | -0.013 |
Sleep | -0.013 |
Token | other |
Token ID | 584 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
pieces | +0.036 |
od | +0.034 |
Maria | +0.033 |
ata | +0.031 |
ill | +0.030 |
Neg logit contributions | |
robe | -0.025 |
cooler | -0.023 |
arl | -0.023 |
cabin | -0.021 |
Sleep | -0.021 |
Token | ways |
Token ID | 2842 |
Feature activation | +3.46e-02 |
Pos logit contributions | |
pieces | +0.428 |
ata | +0.398 |
ez | +0.392 |
od | +0.388 |
Maria | +0.376 |
Neg logit contributions | |
cabin | -0.187 |
old | -0.170 |
refrigerator | -0.166 |
parents | -0.160 |
crib | -0.158 |
Token | , |
Token ID | 11 |
Feature activation | +5.36e-03 |
Pos logit contributions | |
pieces | +0.069 |
Maria | +0.060 |
od | +0.060 |
Rachel | +0.055 |
ata | +0.055 |
Neg logit contributions | |
robe | -0.052 |
arl | -0.048 |
melon | -0.043 |
ugg | -0.043 |
Sleep | -0.043 |
Token | like |
Token ID | 588 |
Feature activation | -1.73e-03 |
Pos logit contributions | |
pieces | +0.008 |
od | +0.008 |
Maria | +0.007 |
enta | +0.007 |
one | +0.007 |
Neg logit contributions | |
robe | -0.010 |
arl | -0.009 |
cooler | -0.009 |
melon | -0.009 |
Sleep | -0.008 |
Token | picking |
Token ID | 10868 |
Feature activation | +2.87e-02 |
Pos logit contributions | |
robe | +0.002 |
cooler | +0.002 |
arl | +0.002 |
melon | +0.002 |
Sleep | +0.002 |
Neg logit contributions | |
pieces | -0.003 |
od | -0.003 |
Maria | -0.003 |
ill | -0.003 |
ata | -0.003 |
Token | flowers |
Token ID | 12734 |
Feature activation | +9.59e-02 |
Pos logit contributions | |
pieces | +0.061 |
Maria | +0.056 |
od | +0.054 |
one | +0.051 |
enta | +0.051 |
Neg logit contributions | |
robe | -0.037 |
cooler | -0.034 |
arl | -0.032 |
melon | -0.031 |
Cream | -0.030 |
Token | ! |
Token ID | 0 |
Feature activation | +3.01e-02 |
Pos logit contributions | |
pieces | +0.012 |
od | +0.010 |
Maria | +0.010 |
enta | +0.009 |
ata | +0.009 |
Neg logit contributions | |
robe | -0.011 |
arl | -0.010 |
cooler | -0.010 |
melon | -0.010 |
ugg | -0.009 |
Token | |
Token ID | 220 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
pieces | +0.063 |
od | +0.057 |
Maria | +0.053 |
enta | +0.052 |
asing | +0.052 |
Neg logit contributions | |
robe | -0.042 |
arl | -0.036 |
ugg | -0.034 |
Cream | -0.034 |
melon | -0.034 |
Token |
|
Token ID | 198 |
Feature activation | +3.98e-02 |
Pos logit contributions | |
robe | +0.022 |
arl | +0.020 |
cooler | +0.018 |
ugg | +0.018 |
melon | +0.018 |
Neg logit contributions | |
pieces | -0.026 |
od | -0.023 |
Maria | -0.022 |
lier | -0.021 |
enta | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | +3.95e-02 |
Pos logit contributions | |
pieces | +0.080 |
od | +0.070 |
rect | +0.065 |
ets | +0.065 |
lier | +0.065 |
Neg logit contributions | |
robe | -0.056 |
arl | -0.054 |
ugg | -0.049 |
Cream | -0.048 |
cooler | -0.046 |
Token | The |
Token ID | 464 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
pieces | +0.065 |
od | +0.060 |
ets | +0.052 |
enta | +0.051 |
ples | +0.051 |
Neg logit contributions | |
robe | -0.066 |
arl | -0.063 |
ugg | -0.060 |
cooler | -0.058 |
Cream | -0.058 |
Token | other |
Token ID | 584 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
pieces | +0.033 |
Maria | +0.028 |
circles | +0.028 |
od | +0.028 |
hours | +0.027 |
Neg logit contributions | |
robe | -0.036 |
arl | -0.034 |
ugg | -0.033 |
melon | -0.031 |
Cream | -0.030 |
Token | animals |
Token ID | 4695 |
Feature activation | +4.27e-02 |
Pos logit contributions | |
pieces | +0.363 |
od | +0.326 |
Maria | +0.313 |
enta | +0.288 |
lier | +0.288 |
Neg logit contributions | |
robe | -0.309 |
arl | -0.276 |
cooler | -0.266 |
melon | -0.259 |
ugg | -0.252 |
Token | were |
Token ID | 547 |
Feature activation | +4.39e-02 |
Pos logit contributions | |
pieces | +0.077 |
od | +0.068 |
Maria | +0.066 |
enta | +0.066 |
rect | +0.064 |
Neg logit contributions | |
robe | -0.063 |
melon | -0.060 |
cooler | -0.057 |
ugg | -0.054 |
tired | -0.053 |
Token | sad |
Token ID | 6507 |
Feature activation | -4.38e-02 |
Pos logit contributions | |
pieces | +0.080 |
od | +0.072 |
Maria | +0.070 |
lier | +0.065 |
one | +0.064 |
Neg logit contributions | |
robe | -0.073 |
arl | -0.066 |
ugg | -0.061 |
melon | -0.059 |
cooler | -0.058 |
Token | and |
Token ID | 290 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
robe | +0.072 |
arl | +0.067 |
cooler | +0.065 |
melon | +0.064 |
ugg | +0.060 |
Neg logit contributions | |
pieces | -0.080 |
Maria | -0.068 |
od | -0.068 |
enta | -0.064 |
ata | -0.062 |
Token | didn |
Token ID | 1422 |
Feature activation | -4.95e-02 |
Pos logit contributions | |
pieces | +0.022 |
od | +0.019 |
Maria | +0.019 |
lier | +0.017 |
one | +0.017 |
Neg logit contributions | |
robe | -0.023 |
arl | -0.021 |
ugg | -0.019 |
cooler | -0.019 |
melon | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | +3.89e-03 |
Pos logit contributions | |
pieces | +0.001 |
od | +0.001 |
Maria | +0.001 |
one | +0.001 |
lier | +0.001 |
Neg logit contributions | |
robe | -0.001 |
arl | -0.001 |
cooler | -0.001 |
melon | -0.001 |
ugg | -0.001 |
Token | toys |
Token ID | 14958 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
pieces | +0.007 |
od | +0.007 |
Maria | +0.007 |
one | +0.006 |
lier | +0.006 |
Neg logit contributions | |
robe | -0.006 |
cooler | -0.005 |
arl | -0.005 |
melon | -0.005 |
ugg | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | +1.67e-02 |
Pos logit contributions | |
pieces | +0.028 |
od | +0.024 |
Maria | +0.024 |
lier | +0.021 |
one | +0.021 |
Neg logit contributions | |
robe | -0.031 |
arl | -0.028 |
cooler | -0.027 |
melon | -0.027 |
ugg | -0.026 |
Token | lots |
Token ID | 6041 |
Feature activation | +4.29e-02 |
Pos logit contributions | |
pieces | +0.033 |
od | +0.031 |
Maria | +0.030 |
ill | +0.029 |
asing | +0.029 |
Neg logit contributions | |
robe | -0.023 |
cooler | -0.021 |
melon | -0.018 |
ranc | -0.018 |
mall | -0.018 |
Token | of |
Token ID | 286 |
Feature activation | +2.81e-02 |
Pos logit contributions | |
pieces | +0.065 |
Maria | +0.060 |
od | +0.058 |
Rachel | +0.055 |
enta | +0.054 |
Neg logit contributions | |
robe | -0.077 |
cooler | -0.073 |
arl | -0.069 |
melon | -0.069 |
ugg | -0.066 |
Token | other |
Token ID | 584 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
pieces | +0.058 |
od | +0.055 |
Maria | +0.055 |
enta | +0.053 |
one | +0.053 |
Neg logit contributions | |
robe | -0.033 |
cooler | -0.031 |
mall | -0.028 |
cabin | -0.027 |
ranc | -0.026 |
Token | things |
Token ID | 1243 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
pieces | +0.401 |
Maria | +0.371 |
od | +0.367 |
Rachel | +0.351 |
one | +0.348 |
Neg logit contributions | |
robe | -0.241 |
cooler | -0.226 |
cabin | -0.215 |
ranc | -0.193 |
mall | -0.190 |
Token | . |
Token ID | 13 |
Feature activation | +2.79e-02 |
Pos logit contributions | |
robe | +0.032 |
arl | +0.028 |
cooler | +0.028 |
melon | +0.027 |
ugg | +0.026 |
Neg logit contributions | |
pieces | -0.033 |
od | -0.029 |
Maria | -0.029 |
one | -0.027 |
enta | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | +4.17e-02 |
Pos logit contributions | |
pieces | +0.057 |
od | +0.054 |
asing | +0.049 |
enta | +0.049 |
rect | +0.049 |
Neg logit contributions | |
robe | -0.037 |
arl | -0.032 |
Sleep | -0.030 |
melon | -0.030 |
ugg | -0.030 |
Token |
|
Token ID | 198 |
Feature activation | +4.11e-02 |
Pos logit contributions | |
pieces | +0.091 |
od | +0.084 |
rect | +0.083 |
asing | +0.080 |
ets | +0.080 |
Neg logit contributions | |
robe | -0.046 |
arl | -0.044 |
ugg | -0.041 |
Sleep | -0.038 |
melon | -0.035 |
Token | At |
Token ID | 2953 |
Feature activation | +4.72e-02 |
Pos logit contributions | |
pieces | +0.066 |
od | +0.064 |
ets | +0.058 |
rect | +0.057 |
iece | +0.057 |
Neg logit contributions | |
robe | -0.062 |
arl | -0.060 |
ugg | -0.058 |
Sleep | -0.057 |
melon | -0.054 |
Token | forest |
Token ID | 8222 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
pieces | +0.013 |
od | +0.012 |
Maria | +0.011 |
lier | +0.010 |
one | +0.010 |
Neg logit contributions | |
robe | -0.008 |
arl | -0.007 |
cooler | -0.007 |
melon | -0.007 |
ugg | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +2.64e-02 |
Pos logit contributions | |
robe | +0.017 |
cooler | +0.015 |
melon | +0.015 |
cabin | +0.014 |
arl | +0.013 |
Neg logit contributions | |
pieces | -0.019 |
Maria | -0.017 |
od | -0.017 |
ez | -0.017 |
one | -0.016 |
Token | Who |
Token ID | 5338 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
pieces | +0.055 |
od | +0.052 |
asing | +0.049 |
lar | +0.048 |
ez | +0.048 |
Neg logit contributions | |
robe | -0.033 |
arl | -0.029 |
melon | -0.028 |
Cream | -0.027 |
ugg | -0.027 |
Token | knows |
Token ID | 4206 |
Feature activation | +5.94e-02 |
Pos logit contributions | |
pieces | +0.175 |
Maria | +0.142 |
od | +0.139 |
circles | +0.134 |
enta | +0.133 |
Neg logit contributions | |
robe | -0.166 |
cooler | -0.144 |
arl | -0.137 |
Sleep | -0.137 |
ugg | -0.137 |
Token | what |
Token ID | 644 |
Feature activation | -4.45e-02 |
Pos logit contributions | |
pieces | +0.124 |
od | +0.110 |
Maria | +0.109 |
ata | +0.104 |
ez | +0.104 |
Neg logit contributions | |
robe | -0.077 |
cooler | -0.068 |
arl | -0.065 |
melon | -0.064 |
Sleep | -0.063 |
Token | other |
Token ID | 584 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
robe | +0.057 |
cooler | +0.051 |
Sleep | +0.050 |
cabin | +0.049 |
melon | +0.048 |
Neg logit contributions | |
pieces | -0.091 |
iece | -0.079 |
od | -0.078 |
ata | -0.078 |
ill | -0.078 |
Token | new |
Token ID | 649 |
Feature activation | +9.28e-02 |
Pos logit contributions | |
pieces | +0.370 |
od | +0.332 |
Maria | +0.327 |
ata | +0.313 |
enta | +0.309 |
Neg logit contributions | |
robe | -0.278 |
cooler | -0.252 |
melon | -0.236 |
arl | -0.234 |
cabin | -0.226 |
Token | adventures |
Token ID | 17545 |
Feature activation | -5.18e-02 |
Pos logit contributions | |
pieces | +0.187 |
od | +0.169 |
Maria | +0.164 |
ata | +0.159 |
enta | +0.157 |
Neg logit contributions | |
robe | -0.130 |
cooler | -0.115 |
melon | -0.106 |
arl | -0.106 |
cabin | -0.105 |
Token | they |
Token ID | 484 |
Feature activation | +3.28e-02 |
Pos logit contributions | |
robe | +0.088 |
arl | +0.077 |
cooler | +0.077 |
melon | +0.077 |
ugg | +0.070 |
Neg logit contributions | |
pieces | -0.090 |
od | -0.079 |
Maria | -0.079 |
enta | -0.071 |
one | -0.071 |
Token | would |
Token ID | 561 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
pieces | +0.057 |
Maria | +0.050 |
od | +0.049 |
one | +0.046 |
enta | +0.046 |
Neg logit contributions | |
robe | -0.054 |
cooler | -0.048 |
arl | -0.047 |
melon | -0.047 |
Sleep | -0.044 |
Token | find |
Token ID | 1064 |
Feature activation | +4.90e-02 |
Pos logit contributions | |
pieces | +0.023 |
Maria | +0.020 |
od | +0.020 |
ata | +0.019 |
enta | +0.019 |
Neg logit contributions | |
robe | -0.017 |
arl | -0.016 |
cooler | -0.016 |
melon | -0.015 |
Sleep | -0.014 |
Token | said |
Token ID | 531 |
Feature activation | +3.28e-02 |
Pos logit contributions | |
pieces | +0.043 |
one | +0.036 |
conclusions | +0.036 |
od | +0.034 |
iece | +0.034 |
Neg logit contributions | |
robe | -0.047 |
cooler | -0.042 |
melon | -0.040 |
bar | -0.038 |
Cream | -0.038 |
Token | . |
Token ID | 13 |
Feature activation | +1.91e-02 |
Pos logit contributions | |
pieces | +0.054 |
od | +0.047 |
Maria | +0.046 |
one | +0.042 |
enta | +0.042 |
Neg logit contributions | |
robe | -0.060 |
arl | -0.054 |
melon | -0.051 |
cooler | -0.051 |
ugg | -0.050 |
Token |
|
Token ID | 198 |
Feature activation | +2.96e-02 |
Pos logit contributions | |
pieces | +0.033 |
od | +0.030 |
Maria | +0.029 |
lier | +0.027 |
one | +0.027 |
Neg logit contributions | |
robe | -0.032 |
arl | -0.029 |
melon | -0.027 |
cooler | -0.027 |
ugg | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | +2.89e-02 |
Pos logit contributions | |
pieces | +0.057 |
od | +0.051 |
Maria | +0.048 |
lier | +0.047 |
one | +0.047 |
Neg logit contributions | |
robe | -0.046 |
arl | -0.042 |
ugg | -0.038 |
cooler | -0.037 |
melon | -0.037 |
Token | The |
Token ID | 464 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
pieces | +0.040 |
od | +0.034 |
Maria | +0.033 |
lier | +0.029 |
one | +0.029 |
Neg logit contributions | |
robe | -0.061 |
arl | -0.056 |
melon | -0.053 |
cooler | -0.053 |
ugg | -0.052 |
Token | other |
Token ID | 584 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
pieces | +0.030 |
od | +0.026 |
Maria | +0.025 |
circles | +0.024 |
lier | +0.023 |
Neg logit contributions | |
robe | -0.030 |
arl | -0.029 |
ugg | -0.027 |
melon | -0.026 |
cooler | -0.025 |
Token | bear |
Token ID | 6842 |
Feature activation | -2.58e-02 |
Pos logit contributions | |
pieces | +0.369 |
od | +0.334 |
Maria | +0.327 |
one | +0.300 |
ill | +0.299 |
Neg logit contributions | |
robe | -0.300 |
cooler | -0.253 |
arl | -0.252 |
melon | -0.247 |
ugg | -0.236 |
Token | said |
Token ID | 531 |
Feature activation | +2.98e-02 |
Pos logit contributions | |
robe | +0.040 |
melon | +0.037 |
cooler | +0.036 |
ugg | +0.035 |
Cream | +0.032 |
Neg logit contributions | |
pieces | -0.046 |
one | -0.039 |
conclusions | -0.039 |
Maria | -0.039 |
od | -0.038 |
Token | , |
Token ID | 11 |
Feature activation | +2.94e-03 |
Pos logit contributions | |
pieces | +0.051 |
od | +0.046 |
Maria | +0.044 |
lier | +0.041 |
one | +0.040 |
Neg logit contributions | |
robe | -0.052 |
arl | -0.048 |
cooler | -0.044 |
melon | -0.044 |
ugg | -0.044 |
Token | " |
Token ID | 366 |
Feature activation | -7.11e-03 |
Pos logit contributions | |
pieces | +0.006 |
od | +0.006 |
Maria | +0.005 |
lier | +0.005 |
one | +0.005 |
Neg logit contributions | |
robe | -0.004 |
arl | -0.004 |
cooler | -0.003 |
melon | -0.003 |
ugg | -0.003 |
Token | L |
Token ID | 43 |
Feature activation | +3.56e-02 |
Pos logit contributions | |
robe | +0.010 |
ugg | +0.008 |
Sleep | +0.008 |
Cream | +0.008 |
arl | +0.008 |
Neg logit contributions | |
pieces | -0.015 |
od | -0.013 |
Maria | -0.012 |
tal | -0.012 |
ets | -0.012 |
Token | brave |
Token ID | 14802 |
Feature activation | +2.87e-02 |
Pos logit contributions | |
pieces | +0.026 |
od | +0.024 |
Maria | +0.023 |
lier | +0.021 |
one | +0.021 |
Neg logit contributions | |
robe | -0.024 |
arl | -0.021 |
cooler | -0.020 |
melon | -0.020 |
ugg | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | +3.19e-03 |
Pos logit contributions | |
pieces | +0.053 |
Maria | +0.046 |
od | +0.045 |
enta | +0.041 |
ill | +0.040 |
Neg logit contributions | |
robe | -0.046 |
arl | -0.042 |
cooler | -0.041 |
melon | -0.040 |
ugg | -0.038 |
Token | he |
Token ID | 339 |
Feature activation | +2.53e-02 |
Pos logit contributions | |
pieces | +0.006 |
od | +0.006 |
Maria | +0.006 |
enta | +0.005 |
asing | +0.005 |
Neg logit contributions | |
robe | -0.005 |
cooler | -0.004 |
melon | -0.004 |
arl | -0.004 |
Cream | -0.004 |
Token | even |
Token ID | 772 |
Feature activation | +6.62e-02 |
Pos logit contributions | |
pieces | +0.043 |
od | +0.037 |
Maria | +0.036 |
enta | +0.033 |
asing | +0.033 |
Neg logit contributions | |
robe | -0.044 |
cooler | -0.040 |
melon | -0.038 |
ugg | -0.036 |
ranc | -0.035 |
Token | visited |
Token ID | 8672 |
Feature activation | -5.20e-02 |
Pos logit contributions | |
pieces | +0.113 |
od | +0.103 |
Maria | +0.095 |
enta | +0.090 |
INK | +0.089 |
Neg logit contributions | |
robe | -0.102 |
cooler | -0.099 |
melon | -0.091 |
tired | -0.089 |
chard | -0.086 |
Token | other |
Token ID | 584 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
robe | +0.069 |
cooler | +0.059 |
melon | +0.056 |
arl | +0.054 |
Cream | +0.052 |
Neg logit contributions | |
pieces | -0.108 |
od | -0.099 |
Maria | -0.097 |
enta | -0.092 |
asing | -0.092 |
Token | ponds |
Token ID | 47339 |
Feature activation | +1.53e-03 |
Pos logit contributions | |
pieces | +0.393 |
Maria | +0.379 |
ez | +0.360 |
od | +0.355 |
enta | +0.351 |
Neg logit contributions | |
cooler | -0.217 |
robe | -0.216 |
cabin | -0.199 |
ranc | -0.191 |
mall | -0.180 |
Token | and |
Token ID | 290 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
pieces | +0.003 |
od | +0.002 |
Maria | +0.002 |
enta | +0.002 |
one | +0.002 |
Neg logit contributions | |
robe | -0.003 |
melon | -0.002 |
cooler | -0.002 |
arl | -0.002 |
ugg | -0.002 |
Token | me |
Token ID | 502 |
Feature activation | +3.66e-02 |
Pos logit contributions | |
pieces | +0.033 |
od | +0.031 |
lier | +0.028 |
Maria | +0.028 |
ill | +0.027 |
Neg logit contributions | |
robe | -0.030 |
cooler | -0.027 |
melon | -0.026 |
cabin | -0.025 |
arl | -0.024 |
Token | adows |
Token ID | 9797 |
Feature activation | +5.17e-02 |
Pos logit contributions | |
pieces | +0.059 |
od | +0.052 |
Maria | +0.050 |
rect | +0.049 |
one | +0.045 |
Neg logit contributions | |
robe | -0.068 |
ugg | -0.057 |
arl | -0.057 |
cooler | -0.056 |
cabin | -0.053 |
Token | ! |
Token ID | 0 |
Feature activation | +3.30e-02 |
Pos logit contributions | |
pieces | +0.085 |
od | +0.075 |
Maria | +0.074 |
enta | +0.067 |
one | +0.066 |
Neg logit contributions | |
robe | -0.093 |
arl | -0.083 |
melon | -0.081 |
cooler | -0.080 |
ugg | -0.077 |
Token | , |
Token ID | 11 |
Feature activation | +9.58e-03 |
Pos logit contributions | |
pieces | +0.028 |
od | +0.024 |
Maria | +0.024 |
enta | +0.022 |
rect | +0.022 |
Neg logit contributions | |
robe | -0.023 |
cooler | -0.020 |
arl | -0.019 |
melon | -0.018 |
cabin | -0.018 |
Token | he |
Token ID | 339 |
Feature activation | +3.16e-02 |
Pos logit contributions | |
pieces | +0.018 |
od | +0.017 |
Maria | +0.016 |
lier | +0.015 |
one | +0.015 |
Neg logit contributions | |
robe | -0.015 |
arl | -0.013 |
cooler | -0.012 |
melon | -0.012 |
ugg | -0.012 |
Token | was |
Token ID | 373 |
Feature activation | +3.64e-02 |
Pos logit contributions | |
pieces | +0.056 |
od | +0.047 |
Maria | +0.046 |
enta | +0.043 |
rect | +0.042 |
Neg logit contributions | |
robe | -0.055 |
cooler | -0.050 |
melon | -0.047 |
arl | -0.044 |
ugg | -0.044 |
Token | at |
Token ID | 379 |
Feature activation | +8.77e-03 |
Pos logit contributions | |
pieces | +0.069 |
od | +0.062 |
Maria | +0.060 |
lier | +0.056 |
one | +0.056 |
Neg logit contributions | |
robe | -0.056 |
arl | -0.050 |
cooler | -0.048 |
melon | -0.047 |
ugg | -0.045 |
Token | the |
Token ID | 262 |
Feature activation | +9.91e-03 |
Pos logit contributions | |
pieces | +0.017 |
od | +0.015 |
Maria | +0.015 |
lier | +0.014 |
one | +0.014 |
Neg logit contributions | |
robe | -0.013 |
arl | -0.012 |
cooler | -0.011 |
melon | -0.011 |
ugg | -0.011 |
Token | other |
Token ID | 584 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
pieces | +0.022 |
od | +0.020 |
Maria | +0.020 |
enta | +0.019 |
asing | +0.018 |
Neg logit contributions | |
robe | -0.012 |
cooler | -0.010 |
arl | -0.010 |
melon | -0.010 |
cabin | -0.009 |
Token | side |
Token ID | 1735 |
Feature activation | +3.78e-02 |
Pos logit contributions | |
pieces | +0.434 |
od | +0.406 |
Maria | +0.405 |
enta | +0.386 |
ata | +0.384 |
Neg logit contributions | |
robe | -0.216 |
cooler | -0.198 |
cabin | -0.193 |
mall | -0.181 |
refrigerator | -0.177 |
Token | of |
Token ID | 286 |
Feature activation | +2.71e-02 |
Pos logit contributions | |
pieces | +0.062 |
od | +0.056 |
Maria | +0.056 |
enta | +0.053 |
Rachel | +0.049 |
Neg logit contributions | |
robe | -0.063 |
melon | -0.058 |
cooler | -0.058 |
arl | -0.054 |
ugg | -0.052 |
Token | the |
Token ID | 262 |
Feature activation | +9.93e-03 |
Pos logit contributions | |
pieces | +0.050 |
od | +0.045 |
Maria | +0.043 |
lier | +0.041 |
one | +0.040 |
Neg logit contributions | |
robe | -0.044 |
arl | -0.040 |
cooler | -0.038 |
melon | -0.037 |
ugg | -0.036 |
Token | field |
Token ID | 2214 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
pieces | +0.022 |
od | +0.021 |
Maria | +0.020 |
ill | +0.019 |
asing | +0.019 |
Neg logit contributions | |
robe | -0.012 |
cooler | -0.011 |
arl | -0.011 |
melon | -0.010 |
cabin | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +3.26e-02 |
Pos logit contributions | |
pieces | +0.035 |
od | +0.031 |
Maria | +0.031 |
enta | +0.029 |
asing | +0.027 |
Neg logit contributions | |
robe | -0.035 |
melon | -0.031 |
cooler | -0.031 |
arl | -0.029 |
ugg | -0.028 |
Token | were |
Token ID | 547 |
Feature activation | +4.19e-02 |
Pos logit contributions | |
pieces | +0.068 |
Maria | +0.061 |
od | +0.058 |
Rachel | +0.055 |
ata | +0.054 |
Neg logit contributions | |
robe | -0.060 |
melon | -0.058 |
cooler | -0.053 |
ugg | -0.052 |
cabin | -0.049 |
Token | more |
Token ID | 517 |
Feature activation | +6.31e-02 |
Pos logit contributions | |
pieces | +0.073 |
Maria | +0.066 |
od | +0.066 |
ata | +0.062 |
ill | +0.061 |
Neg logit contributions | |
robe | -0.063 |
cooler | -0.059 |
melon | -0.059 |
arl | -0.054 |
tired | -0.053 |
Token | competitive |
Token ID | 7606 |
Feature activation | +6.88e-02 |
Pos logit contributions | |
pieces | +0.141 |
Maria | +0.126 |
one | +0.124 |
od | +0.123 |
lar | +0.122 |
Neg logit contributions | |
cooler | -0.064 |
robe | -0.062 |
tired | -0.058 |
arl | -0.055 |
thirsty | -0.052 |
Token | with |
Token ID | 351 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
pieces | +0.121 |
Maria | +0.116 |
od | +0.106 |
Jud | +0.103 |
enta | +0.103 |
Neg logit contributions | |
robe | -0.103 |
cooler | -0.096 |
melon | -0.085 |
mall | -0.084 |
cabin | -0.084 |
Token | each |
Token ID | 1123 |
Feature activation | -2.70e-02 |
Pos logit contributions | |
robe | +0.027 |
melon | +0.025 |
cooler | +0.025 |
Cream | +0.023 |
arl | +0.023 |
Neg logit contributions | |
pieces | -0.041 |
od | -0.037 |
Maria | -0.036 |
ata | -0.035 |
inda | -0.035 |
Token | other |
Token ID | 584 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
job | +0.020 |
bar | +0.019 |
well | +0.019 |
tight | +0.018 |
big | +0.018 |
Neg logit contributions | |
Maria | -0.058 |
ill | -0.055 |
ples | -0.054 |
older | -0.053 |
light | -0.053 |
Token | . |
Token ID | 13 |
Feature activation | +2.98e-02 |
Pos logit contributions | |
pieces | +0.321 |
Maria | +0.298 |
od | +0.286 |
enta | +0.269 |
one | +0.262 |
Neg logit contributions | |
robe | -0.335 |
cooler | -0.280 |
melon | -0.275 |
arl | -0.274 |
cabin | -0.263 |
Token | They |
Token ID | 1119 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
pieces | +0.065 |
od | +0.058 |
inda | +0.056 |
Maria | +0.056 |
enta | +0.055 |
Neg logit contributions | |
robe | -0.034 |
Cream | -0.029 |
melon | -0.029 |
arl | -0.029 |
Sleep | -0.028 |
Token | wanted |
Token ID | 2227 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
robe | +0.016 |
melon | +0.015 |
cooler | +0.014 |
ugg | +0.013 |
tired | +0.012 |
Neg logit contributions | |
pieces | -0.020 |
Maria | -0.018 |
od | -0.018 |
one | -0.017 |
lier | -0.017 |
Token | to |
Token ID | 284 |
Feature activation | +4.01e-02 |
Pos logit contributions | |
melon | +0.018 |
cooler | +0.017 |
robe | +0.017 |
Cream | +0.016 |
not | +0.016 |
Neg logit contributions | |
pieces | -0.042 |
ata | -0.037 |
inda | -0.037 |
od | -0.036 |
enta | -0.036 |
Token | prove |
Token ID | 5879 |
Feature activation | +6.73e-02 |
Pos logit contributions | |
pieces | +0.076 |
Maria | +0.069 |
od | +0.068 |
one | +0.065 |
ata | +0.065 |
Neg logit contributions | |
robe | -0.056 |
cooler | -0.050 |
melon | -0.050 |
arl | -0.050 |
name | -0.047 |
Token | together |
Token ID | 1978 |
Feature activation | +4.34e-02 |
Pos logit contributions | |
robe | +0.040 |
cooler | +0.036 |
melon | +0.034 |
arl | +0.033 |
Cream | +0.033 |
Neg logit contributions | |
pieces | -0.048 |
Maria | -0.044 |
od | -0.044 |
asing | -0.043 |
one | -0.042 |
Token | , |
Token ID | 11 |
Feature activation | -5.31e-03 |
Pos logit contributions | |
pieces | +0.079 |
Maria | +0.073 |
od | +0.069 |
enta | +0.068 |
one | +0.067 |
Neg logit contributions | |
robe | -0.067 |
melon | -0.058 |
cooler | -0.057 |
arl | -0.056 |
ugg | -0.052 |
Token | always |
Token ID | 1464 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
robe | +0.008 |
cooler | +0.008 |
melon | +0.007 |
arl | +0.007 |
Sleep | +0.007 |
Neg logit contributions | |
pieces | -0.009 |
od | -0.008 |
Maria | -0.008 |
one | -0.008 |
enta | -0.008 |
Token | understanding |
Token ID | 4547 |
Feature activation | +2.14e-02 |
Pos logit contributions | |
robe | +0.016 |
cooler | +0.014 |
arl | +0.014 |
melon | +0.014 |
Sleep | +0.014 |
Neg logit contributions | |
pieces | -0.021 |
od | -0.020 |
Maria | -0.019 |
ata | -0.018 |
one | -0.018 |
Token | each |
Token ID | 1123 |
Feature activation | -2.92e-02 |
Pos logit contributions | |
pieces | +0.041 |
od | +0.037 |
Maria | +0.037 |
enta | +0.036 |
ata | +0.035 |
Neg logit contributions | |
robe | -0.029 |
arl | -0.028 |
cooler | -0.026 |
melon | -0.026 |
cabin | -0.024 |
Token | other |
Token ID | 584 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
name | +0.034 |
robe | +0.034 |
bar | +0.034 |
waitress | +0.033 |
refrigerator | +0.033 |
Neg logit contributions | |
pieces | -0.053 |
enta | -0.052 |
od | -0.051 |
Maria | -0.051 |
CR | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | +3.10e-02 |
Pos logit contributions | |
pieces | +0.275 |
od | +0.239 |
Maria | +0.234 |
one | +0.208 |
enta | +0.208 |
Neg logit contributions | |
robe | -0.384 |
arl | -0.350 |
melon | -0.333 |
cooler | -0.333 |
ugg | -0.325 |
Token | It |
Token ID | 632 |
Feature activation | +7.47e-03 |
Pos logit contributions | |
pieces | +0.062 |
od | +0.056 |
asing | +0.053 |
enta | +0.053 |
ced | +0.053 |
Neg logit contributions | |
robe | -0.043 |
arl | -0.037 |
Cream | -0.036 |
ugg | -0.036 |
melon | -0.036 |
Token | was |
Token ID | 373 |
Feature activation | +3.12e-02 |
Pos logit contributions | |
pieces | +0.012 |
od | +0.011 |
Maria | +0.010 |
ill | +0.009 |
enta | +0.009 |
Neg logit contributions | |
robe | -0.014 |
cooler | -0.012 |
melon | -0.012 |
arl | -0.011 |
Sleep | -0.011 |
Token | really |
Token ID | 1107 |
Feature activation | +3.47e-02 |
Pos logit contributions | |
pieces | +0.056 |
od | +0.050 |
Maria | +0.049 |
lier | +0.045 |
one | +0.045 |
Neg logit contributions | |
robe | -0.051 |
arl | -0.047 |
ugg | -0.043 |
melon | -0.043 |
cooler | -0.043 |
Token | sweet |
Token ID | 6029 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
pieces | +0.072 |
od | +0.066 |
Maria | +0.064 |
lier | +0.061 |
inda | +0.059 |
Neg logit contributions | |
robe | -0.047 |
arl | -0.042 |
ugg | -0.038 |
melon | -0.038 |
cooler | -0.037 |
Token | 't |
Token ID | 470 |
Feature activation | -5.99e-03 |
Pos logit contributions | |
robe | +0.076 |
arl | +0.066 |
ugg | +0.061 |
melon | +0.061 |
cooler | +0.060 |
Neg logit contributions | |
pieces | -0.117 |
od | -0.108 |
Maria | -0.103 |
one | -0.099 |
lier | -0.097 |
Token | do |
Token ID | 466 |
Feature activation | +3.24e-02 |
Pos logit contributions | |
robe | +0.009 |
arl | +0.008 |
melon | +0.008 |
cooler | +0.008 |
ugg | +0.007 |
Neg logit contributions | |
pieces | -0.011 |
od | -0.010 |
Maria | -0.010 |
enta | -0.009 |
ata | -0.009 |
Token | it |
Token ID | 340 |
Feature activation | +9.22e-03 |
Pos logit contributions | |
pieces | +0.065 |
Maria | +0.059 |
od | +0.059 |
ill | +0.055 |
inda | +0.054 |
Neg logit contributions | |
robe | -0.044 |
arl | -0.040 |
cooler | -0.038 |
melon | -0.037 |
Cream | -0.035 |
Token | to |
Token ID | 284 |
Feature activation | +3.63e-02 |
Pos logit contributions | |
pieces | +0.015 |
Maria | +0.014 |
od | +0.013 |
Rachel | +0.012 |
Jul | +0.012 |
Neg logit contributions | |
robe | -0.017 |
cooler | -0.015 |
arl | -0.014 |
melon | -0.014 |
ugg | -0.013 |
Token | any |
Token ID | 597 |
Feature activation | -3.67e-03 |
Pos logit contributions | |
pieces | +0.060 |
od | +0.053 |
Maria | +0.051 |
one | +0.047 |
lier | +0.047 |
Neg logit contributions | |
robe | -0.065 |
arl | -0.060 |
melon | -0.056 |
cooler | -0.056 |
ugg | -0.055 |
Token | other |
Token ID | 584 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
cooler | +0.006 |
robe | +0.006 |
arl | +0.005 |
crib | +0.005 |
dirty | +0.005 |
Neg logit contributions | |
od | -0.006 |
Maria | -0.006 |
pieces | -0.006 |
enta | -0.005 |
ill | -0.005 |
Token | sm |
Token ID | 895 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
Maria | +0.372 |
pieces | +0.352 |
od | +0.349 |
ris | +0.345 |
enta | +0.343 |
Neg logit contributions | |
robe | -0.244 |
cooler | -0.236 |
mall | -0.198 |
crib | -0.186 |
dirty | -0.184 |
Token | elly |
Token ID | 6148 |
Feature activation | -9.16e-04 |
Pos logit contributions | |
robe | +0.021 |
arl | +0.020 |
melon | +0.019 |
ugg | +0.019 |
cooler | +0.018 |
Neg logit contributions | |
pieces | -0.021 |
od | -0.018 |
Maria | -0.017 |
enta | -0.016 |
one | -0.016 |
Token | skirts |
Token ID | 42370 |
Feature activation | +8.00e-03 |
Pos logit contributions | |
robe | +0.002 |
cooler | +0.001 |
arl | +0.001 |
melon | +0.001 |
mall | +0.001 |
Neg logit contributions | |
pieces | -0.002 |
Maria | -0.002 |
od | -0.002 |
enta | -0.001 |
one | -0.001 |
Token | that |
Token ID | 326 |
Feature activation | -9.18e-02 |
Pos logit contributions | |
pieces | +0.014 |
Maria | +0.013 |
od | +0.012 |
enta | +0.011 |
Rachel | +0.011 |
Neg logit contributions | |
robe | -0.014 |
arl | -0.012 |
melon | -0.012 |
cooler | -0.012 |
ugg | -0.011 |
Token | you |
Token ID | 345 |
Feature activation | +5.79e-02 |
Pos logit contributions | |
robe | +0.157 |
arl | +0.135 |
cooler | +0.135 |
melon | +0.135 |
ugg | +0.126 |
Neg logit contributions | |
pieces | -0.161 |
od | -0.144 |
Maria | -0.142 |
enta | -0.129 |
lier | -0.127 |
Token | and |
Token ID | 290 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
pieces | +0.160 |
od | +0.147 |
Maria | +0.141 |
enta | +0.134 |
ez | +0.133 |
Neg logit contributions | |
robe | -0.133 |
arl | -0.126 |
cooler | -0.115 |
melon | -0.111 |
ugg | -0.107 |
Token | you |
Token ID | 345 |
Feature activation | +5.54e-02 |
Pos logit contributions | |
pieces | +0.026 |
od | +0.023 |
Maria | +0.022 |
lier | +0.021 |
one | +0.021 |
Neg logit contributions | |
robe | -0.025 |
arl | -0.023 |
cooler | -0.022 |
melon | -0.022 |
ugg | -0.021 |
Token | will |
Token ID | 481 |
Feature activation | -2.82e-02 |
Pos logit contributions | |
pieces | +0.090 |
od | +0.080 |
Maria | +0.077 |
lier | +0.071 |
one | +0.071 |
Neg logit contributions | |
robe | -0.100 |
arl | -0.092 |
melon | -0.087 |
cooler | -0.086 |
ugg | -0.085 |
Token | go |
Token ID | 467 |
Feature activation | -2.72e-03 |
Pos logit contributions | |
robe | +0.051 |
arl | +0.046 |
melon | +0.043 |
ugg | +0.043 |
cooler | +0.043 |
Neg logit contributions | |
pieces | -0.046 |
od | -0.041 |
Maria | -0.039 |
one | -0.037 |
lier | -0.037 |
Token | the |
Token ID | 262 |
Feature activation | -5.43e-04 |
Pos logit contributions | |
robe | +0.005 |
arl | +0.004 |
cooler | +0.004 |
melon | +0.004 |
ugg | +0.004 |
Neg logit contributions | |
pieces | -0.005 |
od | -0.004 |
Maria | -0.004 |
lier | -0.004 |
enta | -0.004 |
Token | other |
Token ID | 584 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
robe | +0.001 |
arl | +0.001 |
ugg | +0.001 |
melon | +0.001 |
cooler | +0.001 |
Neg logit contributions | |
pieces | -0.001 |
od | -0.001 |
Maria | -0.001 |
one | -0.001 |
lier | -0.001 |
Token | way |
Token ID | 835 |
Feature activation | +8.49e-02 |
Pos logit contributions | |
pieces | +0.414 |
Maria | +0.402 |
od | +0.395 |
ez | +0.369 |
inda | +0.365 |
Neg logit contributions | |
robe | -0.220 |
cooler | -0.220 |
mall | -0.204 |
cabin | -0.203 |
festival | -0.185 |
Token | . |
Token ID | 13 |
Feature activation | +2.93e-02 |
Pos logit contributions | |
pieces | +0.152 |
od | +0.138 |
Maria | +0.134 |
enta | +0.123 |
ez | +0.122 |
Neg logit contributions | |
robe | -0.141 |
arl | -0.130 |
cooler | -0.122 |
melon | -0.119 |
ugg | -0.115 |
Token | How |
Token ID | 1374 |
Feature activation | -4.67e-02 |
Pos logit contributions | |
pieces | +0.049 |
od | +0.044 |
Maria | +0.041 |
inda | +0.038 |
enta | +0.038 |
Neg logit contributions | |
robe | -0.051 |
arl | -0.048 |
Sleep | -0.044 |
ugg | -0.044 |
cooler | -0.044 |
Token | does |
Token ID | 857 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
robe | +0.070 |
arl | +0.063 |
cooler | +0.061 |
ugg | +0.056 |
Cream | +0.055 |
Neg logit contributions | |
pieces | -0.089 |
Maria | -0.080 |
od | -0.078 |
enta | -0.076 |
ez | -0.074 |
Token | that |
Token ID | 326 |
Feature activation | -9.34e-02 |
Pos logit contributions | |
robe | +0.031 |
arl | +0.029 |
ugg | +0.027 |
melon | +0.027 |
cooler | +0.024 |
Neg logit contributions | |
pieces | -0.026 |
od | -0.022 |
Maria | -0.021 |
one | -0.020 |
lier | -0.020 |
Token | ez |
Token ID | 8471 |
Feature activation | +9.56e-03 |
Pos logit contributions | |
pieces | +0.061 |
od | +0.050 |
rect | +0.046 |
Maria | +0.044 |
whoever | +0.041 |
Neg logit contributions | |
robe | -0.090 |
ugg | -0.090 |
arl | -0.086 |
Sleep | -0.074 |
uggle | -0.074 |
Token | ed |
Token ID | 276 |
Feature activation | -3.55e-02 |
Pos logit contributions | |
pieces | +0.019 |
Maria | +0.016 |
od | +0.016 |
enta | +0.015 |
rect | +0.015 |
Neg logit contributions | |
robe | -0.016 |
ugg | -0.013 |
arl | -0.013 |
cooler | -0.011 |
Cream | -0.011 |
Token | again |
Token ID | 757 |
Feature activation | +3.30e-02 |
Pos logit contributions | |
robe | +0.070 |
arl | +0.063 |
cooler | +0.062 |
melon | +0.061 |
ugg | +0.058 |
Neg logit contributions | |
pieces | -0.054 |
od | -0.046 |
Maria | -0.045 |
enta | -0.040 |
one | -0.040 |
Token | ! |
Token ID | 0 |
Feature activation | +2.60e-02 |
Pos logit contributions | |
pieces | +0.069 |
Maria | +0.061 |
od | +0.060 |
one | +0.058 |
enta | +0.056 |
Neg logit contributions | |
robe | -0.044 |
cooler | -0.042 |
arl | -0.038 |
melon | -0.037 |
tired | -0.036 |
Token | " |
Token ID | 366 |
Feature activation | +1.68e-03 |
Pos logit contributions | |
pieces | +0.050 |
od | +0.045 |
Maria | +0.043 |
lier | +0.041 |
enta | +0.041 |
Neg logit contributions | |
robe | -0.040 |
arl | -0.036 |
cooler | -0.033 |
melon | -0.033 |
ugg | -0.033 |
Token | Ah |
Token ID | 10910 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
pieces | +0.003 |
od | +0.003 |
Maria | +0.003 |
ets | +0.003 |
inda | +0.003 |
Neg logit contributions | |
robe | -0.002 |
arl | -0.002 |
ugg | -0.002 |
Sleep | -0.002 |
Cream | -0.002 |
Token | - |
Token ID | 12 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
pieces | +0.304 |
Maria | +0.275 |
od | +0.248 |
Rachel | +0.236 |
ill | +0.232 |
Neg logit contributions | |
arl | -0.315 |
ugg | -0.305 |
robe | -0.298 |
cooler | -0.287 |
sorry | -0.281 |
Token | ch |
Token ID | 354 |
Feature activation | +3.62e-02 |
Pos logit contributions | |
ugg | +0.026 |
robe | +0.026 |
arl | +0.025 |
melon | +0.024 |
Cream | +0.023 |
Neg logit contributions | |
pieces | -0.020 |
od | -0.018 |
Maria | -0.017 |
circles | -0.015 |
rect | -0.015 |
Token | oo |
Token ID | 2238 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
pieces | +0.066 |
od | +0.060 |
Maria | +0.058 |
one | +0.054 |
lier | +0.054 |
Neg logit contributions | |
robe | -0.058 |
arl | -0.053 |
cooler | -0.049 |
melon | -0.049 |
ugg | -0.048 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.91e-02 |
Pos logit contributions | |
robe | +0.058 |
arl | +0.054 |
melon | +0.051 |
cooler | +0.051 |
ugg | +0.051 |
Neg logit contributions | |
pieces | -0.030 |
od | -0.026 |
Maria | -0.024 |
lier | -0.021 |
one | -0.021 |
Token | He |
Token ID | 679 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
pieces | +0.039 |
od | +0.035 |
Maria | +0.034 |
enta | +0.033 |
one | +0.032 |
Neg logit contributions | |
robe | -0.027 |
arl | -0.023 |
cooler | -0.023 |
melon | -0.022 |
ugg | -0.022 |
Token | you |
Token ID | 345 |
Feature activation | +5.57e-02 |
Pos logit contributions | |
robe | +0.022 |
arl | +0.020 |
cooler | +0.019 |
ugg | +0.019 |
melon | +0.018 |
Neg logit contributions | |
pieces | -0.025 |
od | -0.022 |
Maria | -0.022 |
enta | -0.020 |
ata | -0.020 |
Token | 'll |
Token ID | 1183 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
pieces | +0.093 |
Maria | +0.084 |
od | +0.082 |
circles | +0.073 |
rect | +0.073 |
Neg logit contributions | |
robe | -0.099 |
arl | -0.088 |
ugg | -0.086 |
cooler | -0.086 |
melon | -0.085 |
Token | have |
Token ID | 423 |
Feature activation | +8.63e-02 |
Pos logit contributions | |
robe | +0.022 |
arl | +0.020 |
melon | +0.019 |
cooler | +0.019 |
ugg | +0.018 |
Neg logit contributions | |
pieces | -0.020 |
od | -0.018 |
Maria | -0.017 |
lier | -0.016 |
one | -0.016 |
Token | success |
Token ID | 1943 |
Feature activation | +4.16e-02 |
Pos logit contributions | |
pieces | +0.182 |
od | +0.176 |
Maria | +0.171 |
iece | +0.165 |
ill | +0.161 |
Neg logit contributions | |
cooler | -0.102 |
robe | -0.093 |
Cream | -0.091 |
Sleep | -0.089 |
Rudolph | -0.083 |
Token | in |
Token ID | 287 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
pieces | +0.066 |
Maria | +0.061 |
od | +0.057 |
enta | +0.052 |
one | +0.051 |
Neg logit contributions | |
robe | -0.076 |
arl | -0.067 |
cooler | -0.065 |
Cream | -0.064 |
Sleep | -0.062 |
Token | other |
Token ID | 584 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
pieces | +0.035 |
od | +0.032 |
Maria | +0.032 |
lier | +0.029 |
ata | +0.029 |
Neg logit contributions | |
robe | -0.025 |
arl | -0.024 |
cooler | -0.023 |
melon | -0.022 |
Sleep | -0.022 |
Token | things |
Token ID | 1243 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
od | +0.363 |
Maria | +0.359 |
pieces | +0.357 |
ata | +0.345 |
ill | +0.330 |
Neg logit contributions | |
cooler | -0.244 |
cabin | -0.227 |
robe | -0.221 |
mall | -0.220 |
crib | -0.213 |
Token | you |
Token ID | 345 |
Feature activation | +5.71e-02 |
Pos logit contributions | |
robe | +0.036 |
arl | +0.033 |
cooler | +0.032 |
melon | +0.031 |
ugg | +0.031 |
Neg logit contributions | |
pieces | -0.031 |
Maria | -0.027 |
od | -0.027 |
enta | -0.024 |
inda | -0.023 |
Token | try |
Token ID | 1949 |
Feature activation | +5.51e-02 |
Pos logit contributions | |
pieces | +0.100 |
Maria | +0.093 |
od | +0.090 |
enta | +0.080 |
Jul | +0.080 |
Neg logit contributions | |
robe | -0.093 |
arl | -0.085 |
cooler | -0.082 |
melon | -0.081 |
ugg | -0.079 |
Token | and |
Token ID | 290 |
Feature activation | +2.51e-02 |
Pos logit contributions | |
pieces | +0.096 |
Maria | +0.088 |
od | +0.083 |
lar | +0.080 |
inda | +0.078 |
Neg logit contributions | |
robe | -0.092 |
arl | -0.080 |
cooler | -0.079 |
melon | -0.076 |
Cream | -0.075 |
Token | sell |
Token ID | 3677 |
Feature activation | -3.18e-02 |
Pos logit contributions | |
pieces | +0.045 |
od | +0.042 |
Maria | +0.041 |
inda | +0.037 |
ill | +0.037 |
Neg logit contributions | |
robe | -0.039 |
cooler | -0.035 |
arl | -0.035 |
melon | -0.035 |
ugg | -0.033 |
Token |
|
Token ID | 198 |
Feature activation | +4.71e-02 |
Pos logit contributions | |
pieces | +0.068 |
od | +0.062 |
Maria | +0.059 |
lier | +0.056 |
one | +0.056 |
Neg logit contributions | |
robe | -0.052 |
arl | -0.047 |
melon | -0.043 |
cooler | -0.043 |
ugg | -0.042 |
Token | When |
Token ID | 2215 |
Feature activation | -4.21e-02 |
Pos logit contributions | |
pieces | +0.086 |
od | +0.076 |
lier | +0.069 |
asing | +0.068 |
rect | +0.067 |
Neg logit contributions | |
robe | -0.077 |
arl | -0.073 |
ugg | -0.067 |
cooler | -0.064 |
Cream | -0.063 |
Token | he |
Token ID | 339 |
Feature activation | +3.00e-02 |
Pos logit contributions | |
robe | +0.066 |
arl | +0.059 |
cooler | +0.058 |
melon | +0.057 |
ugg | +0.054 |
Neg logit contributions | |
pieces | -0.079 |
od | -0.071 |
Maria | -0.068 |
lier | -0.064 |
one | -0.063 |
Token | reached |
Token ID | 4251 |
Feature activation | +6.05e-02 |
Pos logit contributions | |
pieces | +0.065 |
Maria | +0.054 |
asing | +0.052 |
od | +0.051 |
lier | +0.049 |
Neg logit contributions | |
robe | -0.040 |
cooler | -0.040 |
melon | -0.036 |
ugg | -0.034 |
bar | -0.034 |
Token | the |
Token ID | 262 |
Feature activation | +1.96e-03 |
Pos logit contributions | |
pieces | +0.117 |
od | +0.104 |
Maria | +0.100 |
lier | +0.095 |
asing | +0.093 |
Neg logit contributions | |
robe | -0.094 |
arl | -0.084 |
cooler | -0.083 |
melon | -0.080 |
ugg | -0.075 |
Token | other |
Token ID | 584 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
pieces | +0.004 |
od | +0.004 |
Maria | +0.004 |
ez | +0.004 |
enta | +0.004 |
Neg logit contributions | |
robe | -0.002 |
cooler | -0.002 |
arl | -0.002 |
melon | -0.002 |
cabin | -0.002 |
Token | side |
Token ID | 1735 |
Feature activation | +3.72e-02 |
Pos logit contributions | |
pieces | +0.473 |
Maria | +0.469 |
od | +0.457 |
ez | +0.457 |
Rachel | +0.434 |
Neg logit contributions | |
cooler | -0.163 |
robe | -0.159 |
cabin | -0.139 |
melon | -0.124 |
mall | -0.117 |
Token | , |
Token ID | 11 |
Feature activation | +1.00e-03 |
Pos logit contributions | |
pieces | +0.075 |
od | +0.067 |
Maria | +0.064 |
ez | +0.064 |
enta | +0.063 |
Neg logit contributions | |
robe | -0.051 |
melon | -0.047 |
cooler | -0.047 |
cabin | -0.041 |
arl | -0.040 |
Token | the |
Token ID | 262 |
Feature activation | +2.21e-03 |
Pos logit contributions | |
pieces | +0.002 |
od | +0.002 |
Maria | +0.002 |
one | +0.001 |
lier | +0.001 |
Neg logit contributions | |
robe | -0.002 |
arl | -0.002 |
ugg | -0.001 |
melon | -0.001 |
cooler | -0.001 |
Token | girl |
Token ID | 2576 |
Feature activation | +8.05e-02 |
Pos logit contributions | |
pieces | +0.004 |
od | +0.004 |
Maria | +0.004 |
one | +0.003 |
circles | +0.003 |
Neg logit contributions | |
robe | -0.003 |
arl | -0.003 |
ugg | -0.003 |
melon | -0.003 |
cooler | -0.003 |
Token | said |
Token ID | 531 |
Feature activation | +3.22e-02 |
Pos logit contributions | |
pieces | +0.151 |
od | +0.126 |
Maria | +0.124 |
conclusions | +0.121 |
asing | +0.119 |
Neg logit contributions | |
cooler | -0.116 |
robe | -0.116 |
tired | -0.107 |
ugg | -0.103 |
melon | -0.101 |
Token | from |
Token ID | 422 |
Feature activation | -6.83e-02 |
Pos logit contributions | |
pieces | +0.029 |
od | +0.026 |
Maria | +0.025 |
enta | +0.023 |
one | +0.023 |
Neg logit contributions | |
robe | -0.026 |
arl | -0.023 |
cooler | -0.023 |
melon | -0.022 |
ugg | -0.021 |
Token | one |
Token ID | 530 |
Feature activation | +7.86e-03 |
Pos logit contributions | |
robe | +0.089 |
cooler | +0.085 |
cabin | +0.077 |
melon | +0.076 |
arl | +0.075 |
Neg logit contributions | |
pieces | -0.141 |
od | -0.134 |
Maria | -0.126 |
asing | -0.122 |
ata | -0.120 |
Token | foot |
Token ID | 2366 |
Feature activation | +6.74e-02 |
Pos logit contributions | |
od | +0.015 |
pieces | +0.015 |
Maria | +0.015 |
enta | +0.015 |
ez | +0.014 |
Neg logit contributions | |
cooler | -0.010 |
cabin | -0.009 |
robe | -0.009 |
crib | -0.009 |
melon | -0.009 |
Token | to |
Token ID | 284 |
Feature activation | +4.19e-02 |
Pos logit contributions | |
pieces | +0.124 |
Maria | +0.114 |
od | +0.112 |
enta | +0.103 |
Rachel | +0.102 |
Neg logit contributions | |
robe | -0.107 |
cooler | -0.094 |
melon | -0.092 |
arl | -0.091 |
ugg | -0.084 |
Token | the |
Token ID | 262 |
Feature activation | +6.98e-03 |
Pos logit contributions | |
pieces | +0.087 |
od | +0.081 |
Maria | +0.078 |
asing | +0.075 |
lier | +0.075 |
Neg logit contributions | |
robe | -0.052 |
cooler | -0.049 |
arl | -0.047 |
melon | -0.046 |
Sleep | -0.042 |
Token | other |
Token ID | 584 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
pieces | +0.014 |
od | +0.013 |
Maria | +0.013 |
enta | +0.012 |
asing | +0.012 |
Neg logit contributions | |
robe | -0.010 |
cooler | -0.009 |
arl | -0.009 |
melon | -0.008 |
cabin | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +2.49e-02 |
Pos logit contributions | |
pieces | +0.425 |
Maria | +0.390 |
od | +0.379 |
enta | +0.375 |
inda | +0.359 |
Neg logit contributions | |
robe | -0.227 |
cooler | -0.219 |
cabin | -0.215 |
mall | -0.199 |
crib | -0.189 |
Token | |
Token ID | 220 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
pieces | +0.052 |
od | +0.047 |
Maria | +0.045 |
enta | +0.043 |
lar | +0.043 |
Neg logit contributions | |
robe | -0.034 |
arl | -0.030 |
melon | -0.028 |
cooler | -0.028 |
Cream | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | +3.88e-02 |
Pos logit contributions | |
robe | +0.026 |
arl | +0.023 |
cooler | +0.021 |
melon | +0.021 |
ugg | +0.021 |
Neg logit contributions | |
pieces | -0.028 |
od | -0.025 |
Maria | -0.024 |
lier | -0.022 |
enta | -0.022 |
Token |
|
Token ID | 198 |
Feature activation | +3.82e-02 |
Pos logit contributions | |
pieces | +0.081 |
od | +0.071 |
asing | +0.066 |
lier | +0.066 |
rect | +0.066 |
Neg logit contributions | |
robe | -0.052 |
arl | -0.049 |
ugg | -0.044 |
Cream | -0.044 |
cooler | -0.042 |
Token | He |
Token ID | 1544 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
pieces | +0.062 |
od | +0.057 |
Maria | +0.049 |
enta | +0.049 |
ets | +0.049 |
Neg logit contributions | |
robe | -0.065 |
arl | -0.062 |
ugg | -0.057 |
Cream | -0.057 |
cooler | -0.056 |
Token | . |
Token ID | 13 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
pieces | +0.063 |
Maria | +0.059 |
od | +0.057 |
one | +0.054 |
enta | +0.054 |
Neg logit contributions | |
robe | -0.058 |
cooler | -0.051 |
melon | -0.049 |
arl | -0.046 |
ugg | -0.045 |
Token | Soon |
Token ID | 15894 |
Feature activation | +9.38e-02 |
Pos logit contributions | |
pieces | +0.044 |
od | +0.039 |
lier | +0.037 |
Maria | +0.037 |
lar | +0.036 |
Neg logit contributions | |
robe | -0.032 |
arl | -0.028 |
melon | -0.027 |
ugg | -0.026 |
Cream | -0.025 |
Token | , |
Token ID | 11 |
Feature activation | +3.57e-03 |
Pos logit contributions | |
pieces | +0.203 |
Maria | +0.179 |
asing | +0.172 |
od | +0.171 |
enta | +0.169 |
Neg logit contributions | |
cooler | -0.108 |
robe | -0.101 |
melon | -0.096 |
tired | -0.091 |
Otto | -0.088 |
Token | lots |
Token ID | 6041 |
Feature activation | +4.22e-02 |
Pos logit contributions | |
pieces | +0.008 |
od | +0.007 |
lier | +0.007 |
Maria | +0.007 |
enta | +0.007 |
Neg logit contributions | |
robe | -0.004 |
cooler | -0.004 |
melon | -0.004 |
tired | -0.004 |
cabin | -0.003 |
Token | of |
Token ID | 286 |
Feature activation | +2.47e-02 |
Pos logit contributions | |
pieces | +0.076 |
Maria | +0.067 |
od | +0.066 |
enta | +0.065 |
Jul | +0.062 |
Neg logit contributions | |
cooler | -0.061 |
robe | -0.059 |
melon | -0.058 |
arl | -0.057 |
ugg | -0.053 |
Token | other |
Token ID | 584 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
pieces | +0.052 |
ill | +0.051 |
od | +0.051 |
one | +0.048 |
Maria | +0.048 |
Neg logit contributions | |
cooler | -0.026 |
robe | -0.023 |
old | -0.022 |
tired | -0.022 |
mall | -0.021 |
Token | animals |
Token ID | 4695 |
Feature activation | +4.14e-02 |
Pos logit contributions | |
pieces | +0.416 |
Maria | +0.397 |
od | +0.385 |
Rachel | +0.382 |
ill | +0.376 |
Neg logit contributions | |
robe | -0.188 |
cooler | -0.182 |
old | -0.171 |
tired | -0.166 |
cabin | -0.165 |
Token | joined |
Token ID | 5399 |
Feature activation | -3.98e-02 |
Pos logit contributions | |
pieces | +0.073 |
Maria | +0.064 |
od | +0.061 |
one | +0.060 |
enta | +0.058 |
Neg logit contributions | |
robe | -0.065 |
melon | -0.062 |
cooler | -0.059 |
arl | -0.056 |
ugg | -0.056 |
Token | in |
Token ID | 287 |
Feature activation | +1.91e-02 |
Pos logit contributions | |
robe | +0.040 |
melon | +0.038 |
arl | +0.036 |
cooler | +0.034 |
name | +0.031 |
Neg logit contributions | |
pieces | -0.081 |
ata | -0.078 |
one | -0.077 |
enta | -0.076 |
mastered | -0.075 |
Token | , |
Token ID | 11 |
Feature activation | -1.80e-03 |
Pos logit contributions | |
pieces | +0.036 |
od | +0.033 |
Maria | +0.032 |
one | +0.032 |
asing | +0.031 |
Neg logit contributions | |
robe | -0.028 |
cooler | -0.025 |
arl | -0.025 |
melon | -0.024 |
cabin | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | +8.85e-03 |
Pos logit contributions | |
robe | +0.003 |
melon | +0.003 |
cooler | +0.003 |
arl | +0.002 |
tired | +0.002 |
Neg logit contributions | |
pieces | -0.003 |
od | -0.003 |
Maria | -0.003 |
one | -0.003 |
lier | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +5.97e-03 |
Pos logit contributions | |
robe | +0.077 |
arl | +0.069 |
melon | +0.063 |
ugg | +0.061 |
cooler | +0.058 |
Neg logit contributions | |
pieces | -0.101 |
od | -0.088 |
lar | -0.087 |
enta | -0.086 |
Maria | -0.086 |
Token |
|
Token ID | 198 |
Feature activation | +7.66e-03 |
Pos logit contributions | |
pieces | +0.012 |
od | +0.011 |
ets | +0.011 |
asing | +0.011 |
rect | +0.011 |
Neg logit contributions | |
robe | -0.008 |
arl | -0.007 |
ugg | -0.006 |
Sleep | -0.006 |
Cream | -0.006 |
Token | As |
Token ID | 1722 |
Feature activation | -8.84e-02 |
Pos logit contributions | |
pieces | +0.012 |
od | +0.011 |
ets | +0.010 |
enta | +0.010 |
asing | +0.010 |
Neg logit contributions | |
robe | -0.013 |
arl | -0.012 |
ugg | -0.011 |
Cream | -0.011 |
Sleep | -0.011 |
Token | they |
Token ID | 484 |
Feature activation | +3.04e-03 |
Pos logit contributions | |
robe | +0.117 |
cooler | +0.106 |
melon | +0.104 |
arl | +0.102 |
ugg | +0.097 |
Neg logit contributions | |
pieces | -0.180 |
od | -0.165 |
asing | -0.155 |
enta | -0.154 |
lier | -0.153 |
Token | sn |
Token ID | 3013 |
Feature activation | -6.10e-02 |
Pos logit contributions | |
pieces | +0.005 |
od | +0.004 |
Maria | +0.004 |
one | +0.004 |
lier | +0.004 |
Neg logit contributions | |
robe | -0.005 |
arl | -0.005 |
cooler | -0.005 |
melon | -0.005 |
ugg | -0.005 |
Token | ugg |
Token ID | 6837 |
Feature activation | -2.23e-01 |
Pos logit contributions | |
robe | +0.049 |
ranc | +0.036 |
arl | +0.033 |
waitress | +0.032 |
cooler | +0.029 |
Neg logit contributions | |
pieces | -0.152 |
od | -0.134 |
Maria | -0.134 |
ill | -0.132 |
lier | -0.131 |
Token | led |
Token ID | 992 |
Feature activation | -4.27e-02 |
Pos logit contributions | |
robe | +0.419 |
arl | +0.396 |
melon | +0.369 |
cooler | +0.367 |
ugg | +0.363 |
Neg logit contributions | |
pieces | -0.338 |
od | -0.300 |
Maria | -0.284 |
lier | -0.264 |
one | -0.255 |
Token | underneath |
Token ID | 14638 |
Feature activation | +6.69e-02 |
Pos logit contributions | |
robe | +0.069 |
arl | +0.063 |
melon | +0.058 |
cooler | +0.058 |
ugg | +0.057 |
Neg logit contributions | |
pieces | -0.078 |
od | -0.070 |
Maria | -0.068 |
one | -0.063 |
lier | -0.063 |
Token | the |
Token ID | 262 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
pieces | +0.120 |
od | +0.108 |
Maria | +0.104 |
one | +0.097 |
lier | +0.096 |
Neg logit contributions | |
robe | -0.112 |
arl | -0.099 |
cooler | -0.096 |
melon | -0.095 |
ugg | -0.091 |
Token | gear |
Token ID | 7733 |
Feature activation | -3.48e-02 |
Pos logit contributions | |
robe | +0.026 |
arl | +0.024 |
melon | +0.021 |
ugg | +0.021 |
cooler | +0.020 |
Neg logit contributions | |
pieces | -0.037 |
od | -0.033 |
Maria | -0.033 |
circles | -0.032 |
lier | -0.031 |
Token | , |
Token ID | 11 |
Feature activation | -1.61e-02 |
Pos logit contributions | |
robe | +0.060 |
arl | +0.051 |
melon | +0.051 |
cooler | +0.050 |
ugg | +0.048 |
Neg logit contributions | |
pieces | -0.063 |
od | -0.055 |
Maria | -0.054 |
enta | -0.050 |
lier | -0.050 |
Token | to |
Token ID | 284 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
robe | +0.067 |
arl | +0.062 |
Sleep | +0.061 |
melon | +0.061 |
Cream | +0.059 |
Neg logit contributions | |
pieces | -0.032 |
circles | -0.027 |
minutes | -0.024 |
Maria | -0.024 |
od | -0.023 |
Token | their |
Token ID | 511 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
pieces | +0.030 |
od | +0.027 |
Maria | +0.026 |
one | +0.025 |
lier | +0.025 |
Neg logit contributions | |
robe | -0.023 |
arl | -0.021 |
cooler | -0.020 |
melon | -0.019 |
ugg | -0.019 |
Token | room |
Token ID | 2119 |
Feature activation | -6.94e-02 |
Pos logit contributions | |
robe | +0.024 |
arl | +0.023 |
ugg | +0.020 |
melon | +0.019 |
Sleep | +0.018 |
Neg logit contributions | |
pieces | -0.030 |
circles | -0.027 |
Maria | -0.026 |
od | -0.026 |
hours | -0.025 |
Token | and |
Token ID | 290 |
Feature activation | -8.25e-03 |
Pos logit contributions | |
robe | +0.157 |
arl | +0.150 |
melon | +0.142 |
cooler | +0.140 |
ugg | +0.138 |
Neg logit contributions | |
pieces | -0.077 |
od | -0.065 |
Maria | -0.060 |
lier | -0.055 |
ill | -0.054 |
Token | sn |
Token ID | 3013 |
Feature activation | -4.19e-02 |
Pos logit contributions | |
robe | +0.015 |
arl | +0.013 |
cooler | +0.013 |
melon | +0.013 |
ugg | +0.012 |
Neg logit contributions | |
pieces | -0.014 |
od | -0.012 |
Maria | -0.012 |
lier | -0.011 |
one | -0.011 |
Token | ugg |
Token ID | 6837 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
robe | +0.034 |
ranc | +0.026 |
melon | +0.024 |
Sleep | +0.022 |
cooler | +0.022 |
Neg logit contributions | |
pieces | -0.104 |
asing | -0.091 |
e | -0.089 |
ets | -0.089 |
ez | -0.089 |
Token | led |
Token ID | 992 |
Feature activation | -4.30e-02 |
Pos logit contributions | |
robe | +0.446 |
arl | +0.419 |
melon | +0.411 |
cooler | +0.398 |
ugg | +0.396 |
Neg logit contributions | |
pieces | -0.286 |
od | -0.248 |
Maria | -0.230 |
lier | -0.224 |
ets | -0.214 |
Token | in |
Token ID | 287 |
Feature activation | +4.73e-04 |
Pos logit contributions | |
robe | +0.080 |
arl | +0.079 |
melon | +0.072 |
ugg | +0.069 |
ranc | +0.068 |
Neg logit contributions | |
pieces | -0.061 |
od | -0.055 |
circles | -0.053 |
Maria | -0.052 |
lier | -0.051 |
Token | their |
Token ID | 511 |
Feature activation | -1.61e-02 |
Pos logit contributions | |
pieces | +0.001 |
od | +0.001 |
Maria | +0.001 |
circles | +0.001 |
lier | +0.001 |
Neg logit contributions | |
robe | -0.001 |
arl | -0.001 |
melon | -0.001 |
ugg | -0.001 |
cooler | -0.001 |
Token | beds |
Token ID | 20237 |
Feature activation | -5.76e-02 |
Pos logit contributions | |
arl | +0.028 |
robe | +0.027 |
melon | +0.025 |
ugg | +0.023 |
chard | +0.022 |
Neg logit contributions | |
pieces | -0.021 |
circles | -0.021 |
minutes | -0.021 |
hours | -0.019 |
's | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -4.94e-03 |
Pos logit contributions | |
robe | +0.118 |
arl | +0.110 |
melon | +0.105 |
cooler | +0.103 |
ugg | +0.103 |
Neg logit contributions | |
pieces | -0.078 |
od | -0.068 |
Maria | -0.064 |
lier | -0.059 |
one | -0.058 |
Token | you |
Token ID | 345 |
Feature activation | +1.61e-03 |
Pos logit contributions | |
pieces | +0.115 |
Maria | +0.102 |
od | +0.099 |
one | +0.092 |
circles | +0.089 |
Neg logit contributions | |
robe | -0.162 |
arl | -0.156 |
ugg | -0.141 |
melon | -0.136 |
cooler | -0.133 |
Token | go |
Token ID | 467 |
Feature activation | -5.31e-02 |
Pos logit contributions | |
pieces | +0.003 |
Maria | +0.003 |
od | +0.003 |
enta | +0.002 |
one | +0.002 |
Neg logit contributions | |
robe | -0.003 |
arl | -0.002 |
cooler | -0.002 |
melon | -0.002 |
ugg | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -3.08e-02 |
Pos logit contributions | |
robe | +0.087 |
arl | +0.079 |
cooler | +0.075 |
melon | +0.074 |
ugg | +0.072 |
Neg logit contributions | |
pieces | -0.096 |
od | -0.087 |
Maria | -0.085 |
lier | -0.078 |
one | -0.077 |
Token | my |
Token ID | 616 |
Feature activation | -4.25e-03 |
Pos logit contributions | |
robe | +0.050 |
arl | +0.045 |
cooler | +0.043 |
melon | +0.042 |
ugg | +0.040 |
Neg logit contributions | |
pieces | -0.056 |
od | -0.051 |
Maria | -0.049 |
enta | -0.046 |
lier | -0.045 |
Token | d |
Token ID | 288 |
Feature activation | -5.51e-03 |
Pos logit contributions | |
robe | +0.006 |
arl | +0.005 |
melon | +0.005 |
cooler | +0.005 |
ugg | +0.004 |
Neg logit contributions | |
pieces | -0.009 |
od | -0.008 |
Maria | -0.008 |
lier | -0.007 |
one | -0.007 |
Token | arl |
Token ID | 7063 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
robe | +0.004 |
arl | +0.004 |
melon | +0.003 |
ugg | +0.003 |
cooler | +0.003 |
Neg logit contributions | |
pieces | -0.015 |
od | -0.014 |
Maria | -0.014 |
enta | -0.013 |
lier | -0.013 |
Token | ings |
Token ID | 654 |
Feature activation | -5.24e-02 |
Pos logit contributions | |
robe | +0.357 |
arl | +0.321 |
cooler | +0.306 |
melon | +0.303 |
ugg | +0.297 |
Neg logit contributions | |
pieces | -0.388 |
od | -0.352 |
Maria | -0.330 |
one | -0.316 |
lier | -0.312 |
Token | . |
Token ID | 13 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
robe | +0.099 |
arl | +0.091 |
melon | +0.086 |
cooler | +0.085 |
ugg | +0.084 |
Neg logit contributions | |
pieces | -0.084 |
od | -0.073 |
Maria | -0.071 |
enta | -0.064 |
one | -0.063 |
Token | Enjoy |
Token ID | 18179 |
Feature activation | -5.09e-02 |
Pos logit contributions | |
pieces | +0.018 |
od | +0.016 |
Maria | +0.015 |
one | +0.014 |
lier | +0.014 |
Neg logit contributions | |
robe | -0.020 |
arl | -0.018 |
melon | -0.018 |
cooler | -0.017 |
ugg | -0.017 |
Token | your |
Token ID | 534 |
Feature activation | +8.21e-03 |
Pos logit contributions | |
robe | +0.094 |
arl | +0.085 |
melon | +0.080 |
cooler | +0.079 |
ugg | +0.079 |
Neg logit contributions | |
pieces | -0.082 |
od | -0.073 |
Maria | -0.069 |
lier | -0.064 |
one | -0.064 |
Token | lunch |
Token ID | 9965 |
Feature activation | -3.35e-02 |
Pos logit contributions | |
pieces | +0.016 |
od | +0.014 |
Maria | +0.014 |
circles | +0.013 |
lier | +0.013 |
Neg logit contributions | |
robe | -0.011 |
arl | -0.010 |
ugg | -0.009 |
melon | -0.009 |
Cream | -0.008 |
Token | ?" |
Token ID | 1701 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
robe | +0.036 |
arl | +0.033 |
cooler | +0.030 |
melon | +0.030 |
ugg | +0.029 |
Neg logit contributions | |
pieces | -0.050 |
od | -0.044 |
Maria | -0.043 |
one | -0.040 |
enta | -0.040 |
Token | Lily |
Token ID | 20037 |
Feature activation | +3.54e-02 |
Pos logit contributions | |
pieces | +0.053 |
od | +0.048 |
Maria | +0.046 |
lier | +0.044 |
one | +0.044 |
Neg logit contributions | |
robe | -0.037 |
arl | -0.033 |
melon | -0.030 |
cooler | -0.030 |
ugg | -0.029 |
Token | asked |
Token ID | 1965 |
Feature activation | +2.04e-04 |
Pos logit contributions | |
pieces | +0.071 |
od | +0.056 |
Maria | +0.054 |
one | +0.053 |
lar | +0.052 |
Neg logit contributions | |
robe | -0.051 |
cooler | -0.047 |
bar | -0.045 |
ranc | -0.044 |
tired | -0.044 |
Token | , |
Token ID | 11 |
Feature activation | -1.65e-02 |
Pos logit contributions | |
pieces | +0.000 |
od | +0.000 |
Maria | +0.000 |
ata | +0.000 |
one | +0.000 |
Neg logit contributions | |
robe | -0.000 |
arl | -0.000 |
cooler | -0.000 |
melon | -0.000 |
ugg | -0.000 |
Token | sn |
Token ID | 3013 |
Feature activation | -4.05e-02 |
Pos logit contributions | |
robe | +0.030 |
arl | +0.027 |
ugg | +0.026 |
melon | +0.025 |
cooler | +0.024 |
Neg logit contributions | |
pieces | -0.027 |
od | -0.024 |
Maria | -0.024 |
circles | -0.022 |
lier | -0.022 |
Token | ugg |
Token ID | 6837 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
robe | +0.062 |
cooler | +0.055 |
ranc | +0.053 |
cabin | +0.052 |
Sleep | +0.052 |
Neg logit contributions | |
pieces | -0.073 |
od | -0.067 |
ill | -0.066 |
Maria | -0.065 |
asing | -0.064 |
Token | ling |
Token ID | 1359 |
Feature activation | -2.50e-02 |
Pos logit contributions | |
robe | +0.455 |
melon | +0.423 |
arl | +0.423 |
ugg | +0.418 |
cooler | +0.418 |
Neg logit contributions | |
pieces | -0.254 |
od | -0.233 |
iece | -0.208 |
Maria | -0.207 |
lier | -0.203 |
Token | with |
Token ID | 351 |
Feature activation | -4.22e-02 |
Pos logit contributions | |
robe | +0.049 |
arl | +0.048 |
ugg | +0.044 |
melon | +0.043 |
Cream | +0.042 |
Neg logit contributions | |
pieces | -0.034 |
od | -0.031 |
Maria | -0.030 |
circles | -0.029 |
lier | -0.029 |
Token | her |
Token ID | 607 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
ugg | +0.099 |
robe | +0.099 |
arl | +0.097 |
melon | +0.089 |
ranc | +0.088 |
Neg logit contributions | |
pieces | -0.041 |
circles | -0.038 |
Maria | -0.031 |
Rachel | -0.029 |
ill | -0.029 |
Token | mom |
Token ID | 1995 |
Feature activation | -5.46e-02 |
Pos logit contributions | |
robe | +0.023 |
arl | +0.021 |
ugg | +0.020 |
melon | +0.020 |
Sleep | +0.018 |
Neg logit contributions | |
pieces | -0.040 |
od | -0.036 |
Maria | -0.035 |
circles | -0.035 |
ill | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
robe | +0.098 |
arl | +0.088 |
cooler | +0.086 |
melon | +0.085 |
ugg | +0.082 |
Neg logit contributions | |
pieces | -0.088 |
od | -0.079 |
Maria | -0.077 |
one | -0.069 |
enta | -0.069 |
Token | ." |
Token ID | 526 |
Feature activation | -2.02e-02 |
Pos logit contributions | |
pieces | +0.023 |
od | +0.020 |
Maria | +0.019 |
one | +0.017 |
lier | +0.017 |
Neg logit contributions | |
robe | -0.028 |
arl | -0.026 |
cooler | -0.025 |
melon | -0.025 |
ugg | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | +8.36e-03 |
Pos logit contributions | |
robe | +0.037 |
arl | +0.033 |
melon | +0.032 |
cooler | +0.031 |
ugg | +0.030 |
Neg logit contributions | |
pieces | -0.031 |
Maria | -0.028 |
od | -0.027 |
circles | -0.025 |
ill | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | +8.51e-03 |
Pos logit contributions | |
pieces | +0.014 |
Maria | +0.013 |
od | +0.013 |
circles | +0.012 |
ill | +0.011 |
Neg logit contributions | |
robe | -0.014 |
arl | -0.013 |
melon | -0.012 |
cooler | -0.012 |
ugg | -0.011 |
Token | They |
Token ID | 2990 |
Feature activation | -7.06e-02 |
Pos logit contributions | |
pieces | +0.012 |
od | +0.010 |
Maria | +0.010 |
lier | +0.008 |
one | +0.008 |
Neg logit contributions | |
robe | -0.018 |
arl | -0.017 |
melon | -0.016 |
cooler | -0.016 |
ugg | -0.015 |
Token | sn |
Token ID | 3013 |
Feature activation | -4.17e-02 |
Pos logit contributions | |
robe | +0.131 |
arl | +0.124 |
melon | +0.115 |
cooler | +0.112 |
ugg | +0.109 |
Neg logit contributions | |
pieces | -0.107 |
od | -0.096 |
Maria | -0.092 |
lier | -0.084 |
ill | -0.084 |
Token | ugg |
Token ID | 6837 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
robe | +0.029 |
ranc | +0.019 |
melon | +0.019 |
arl | +0.018 |
cooler | +0.018 |
Neg logit contributions | |
pieces | -0.111 |
Maria | -0.098 |
e | -0.098 |
lier | -0.098 |
asing | -0.098 |
Token | led |
Token ID | 992 |
Feature activation | -4.03e-02 |
Pos logit contributions | |
robe | +0.412 |
arl | +0.382 |
melon | +0.365 |
cooler | +0.361 |
ugg | +0.351 |
Neg logit contributions | |
pieces | -0.333 |
od | -0.293 |
Maria | -0.279 |
lier | -0.261 |
one | -0.255 |
Token | in |
Token ID | 287 |
Feature activation | +2.56e-04 |
Pos logit contributions | |
robe | +0.075 |
arl | +0.072 |
melon | +0.065 |
ugg | +0.063 |
Cream | +0.062 |
Neg logit contributions | |
pieces | -0.058 |
od | -0.053 |
Maria | -0.051 |
lier | -0.050 |
circles | -0.050 |
Token | their |
Token ID | 511 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
pieces | +0.000 |
od | +0.000 |
Maria | +0.000 |
circles | +0.000 |
hours | +0.000 |
Neg logit contributions | |
robe | -0.000 |
arl | -0.000 |
melon | -0.000 |
ugg | -0.000 |
cooler | -0.000 |
Token | beds |
Token ID | 20237 |
Feature activation | -5.61e-02 |
Pos logit contributions | |
robe | +0.025 |
arl | +0.025 |
melon | +0.022 |
ugg | +0.021 |
Cream | +0.020 |
Neg logit contributions | |
pieces | -0.020 |
circles | -0.020 |
hours | -0.018 |
rect | -0.018 |
minutes | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | -4.83e-03 |
Pos logit contributions | |
robe | +0.118 |
arl | +0.110 |
cooler | +0.104 |
melon | +0.104 |
ugg | +0.102 |
Neg logit contributions | |
pieces | -0.075 |
od | -0.065 |
Maria | -0.062 |
one | -0.055 |
lier | -0.055 |
Token | " |
Token ID | 1 |
Feature activation | -4.44e-02 |
Pos logit contributions | |
pieces | +0.011 |
od | +0.010 |
Maria | +0.009 |
lier | +0.009 |
one | +0.008 |
Neg logit contributions | |
robe | -0.012 |
arl | -0.011 |
cooler | -0.010 |
ugg | -0.010 |
Cream | -0.010 |
Token | Hello |
Token ID | 15496 |
Feature activation | -3.20e-02 |
Pos logit contributions | |
robe | +0.075 |
arl | +0.069 |
ugg | +0.063 |
cooler | +0.063 |
melon | +0.063 |
Neg logit contributions | |
pieces | -0.079 |
od | -0.070 |
Maria | -0.066 |
lier | -0.063 |
enta | -0.061 |
Token | , |
Token ID | 11 |
Feature activation | -2.07e-02 |
Pos logit contributions | |
robe | +0.050 |
arl | +0.046 |
cooler | +0.046 |
Cream | +0.043 |
melon | +0.042 |
Neg logit contributions | |
pieces | -0.059 |
od | -0.052 |
Maria | -0.051 |
asing | -0.047 |
ez | -0.047 |
Token | my |
Token ID | 616 |
Feature activation | -4.82e-03 |
Pos logit contributions | |
robe | +0.032 |
arl | +0.027 |
melon | +0.027 |
ugg | +0.026 |
cooler | +0.025 |
Neg logit contributions | |
pieces | -0.040 |
od | -0.036 |
Maria | -0.035 |
one | -0.033 |
lier | -0.032 |
Token | d |
Token ID | 288 |
Feature activation | -3.91e-03 |
Pos logit contributions | |
robe | +0.007 |
arl | +0.007 |
melon | +0.007 |
cooler | +0.006 |
ugg | +0.006 |
Neg logit contributions | |
pieces | -0.009 |
od | -0.008 |
Maria | -0.008 |
lier | -0.008 |
one | -0.007 |
Token | arl |
Token ID | 7063 |
Feature activation | -2.18e-01 |
Pos logit contributions | |
robe | +0.001 |
cooler | +0.000 |
melon | +0.000 |
arl | +0.000 |
ugg | +0.000 |
Neg logit contributions | |
pieces | -0.012 |
od | -0.011 |
Maria | -0.011 |
one | -0.011 |
ill | -0.011 |
Token | ings |
Token ID | 654 |
Feature activation | -4.99e-02 |
Pos logit contributions | |
robe | +0.344 |
chard | +0.335 |
cooler | +0.334 |
melon | +0.333 |
ranc | +0.322 |
Neg logit contributions | |
pieces | -0.310 |
od | -0.306 |
one | -0.281 |
ata | -0.274 |
ill | -0.272 |
Token | . |
Token ID | 13 |
Feature activation | -5.73e-03 |
Pos logit contributions | |
robe | +0.089 |
arl | +0.085 |
cooler | +0.078 |
Cream | +0.076 |
ugg | +0.075 |
Neg logit contributions | |
pieces | -0.086 |
od | -0.074 |
Maria | -0.072 |
enta | -0.068 |
rect | -0.065 |
Token | I |
Token ID | 314 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
robe | +0.011 |
arl | +0.010 |
melon | +0.009 |
cooler | +0.009 |
ugg | +0.009 |
Neg logit contributions | |
pieces | -0.009 |
od | -0.008 |
Maria | -0.008 |
one | -0.007 |
lier | -0.007 |
Token | 'm |
Token ID | 1101 |
Feature activation | +4.57e-02 |
Pos logit contributions | |
pieces | +0.031 |
od | +0.027 |
Maria | +0.026 |
one | +0.024 |
lier | +0.024 |
Neg logit contributions | |
robe | -0.033 |
arl | -0.030 |
cooler | -0.028 |
ugg | -0.028 |
melon | -0.028 |
Token | so |
Token ID | 523 |
Feature activation | -5.19e-02 |
Pos logit contributions | |
pieces | +0.082 |
od | +0.076 |
Maria | +0.072 |
circles | +0.065 |
enta | +0.065 |
Neg logit contributions | |
robe | -0.077 |
arl | -0.069 |
ugg | -0.066 |
melon | -0.063 |
cooler | -0.059 |
Token |
|
Token ID | 198 |
Feature activation | +5.90e-03 |
Pos logit contributions | |
robe | +0.081 |
arl | +0.074 |
melon | +0.065 |
ugg | +0.064 |
Cream | +0.062 |
Neg logit contributions | |
pieces | -0.109 |
od | -0.095 |
Maria | -0.091 |
lar | -0.090 |
enta | -0.090 |
Token |
|
Token ID | 198 |
Feature activation | +7.35e-03 |
Pos logit contributions | |
pieces | +0.012 |
od | +0.011 |
ets | +0.010 |
enta | +0.010 |
asing | +0.010 |
Neg logit contributions | |
robe | -0.008 |
arl | -0.007 |
ugg | -0.007 |
Cream | -0.007 |
melon | -0.006 |
Token | Don |
Token ID | 3987 |
Feature activation | -9.24e-02 |
Pos logit contributions | |
pieces | +0.012 |
od | +0.011 |
ets | +0.009 |
enta | +0.009 |
one | +0.009 |
Neg logit contributions | |
robe | -0.013 |
arl | -0.012 |
ugg | -0.011 |
Cream | -0.011 |
melon | -0.011 |
Token | ny |
Token ID | 3281 |
Feature activation | -4.06e-02 |
Pos logit contributions | |
robe | +0.114 |
arl | +0.101 |
cooler | +0.089 |
melon | +0.089 |
ugg | +0.088 |
Neg logit contributions | |
pieces | -0.206 |
od | -0.187 |
Maria | -0.182 |
lier | -0.172 |
enta | -0.171 |
Token | sn |
Token ID | 3013 |
Feature activation | -5.83e-02 |
Pos logit contributions | |
robe | +0.071 |
arl | +0.064 |
melon | +0.060 |
ugg | +0.059 |
cooler | +0.059 |
Neg logit contributions | |
pieces | -0.072 |
od | -0.062 |
Maria | -0.062 |
enta | -0.057 |
circles | -0.056 |
Token | ugg |
Token ID | 6837 |
Feature activation | -2.17e-01 |
Pos logit contributions | |
ranc | +0.079 |
robe | +0.078 |
cabin | +0.074 |
waitress | +0.073 |
cooler | +0.071 |
Neg logit contributions | |
pieces | -0.101 |
ill | -0.098 |
od | -0.093 |
Maria | -0.091 |
e | -0.090 |
Token | led |
Token ID | 992 |
Feature activation | -4.03e-02 |
Pos logit contributions | |
robe | +0.457 |
arl | +0.432 |
melon | +0.421 |
cooler | +0.420 |
ugg | +0.418 |
Neg logit contributions | |
pieces | -0.252 |
od | -0.222 |
Maria | -0.201 |
lier | -0.189 |
inda | -0.187 |
Token | into |
Token ID | 656 |
Feature activation | +2.19e-02 |
Pos logit contributions | |
robe | +0.069 |
arl | +0.063 |
cooler | +0.059 |
melon | +0.059 |
ugg | +0.057 |
Neg logit contributions | |
pieces | -0.070 |
od | -0.063 |
Maria | -0.061 |
one | -0.056 |
lier | -0.055 |
Token | his |
Token ID | 465 |
Feature activation | -2.87e-02 |
Pos logit contributions | |
pieces | +0.043 |
od | +0.039 |
Maria | +0.037 |
lier | +0.036 |
one | +0.035 |
Neg logit contributions | |
robe | -0.033 |
arl | -0.030 |
cooler | -0.029 |
melon | -0.027 |
Sleep | -0.026 |
Token | bed |
Token ID | 3996 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
robe | +0.039 |
arl | +0.036 |
melon | +0.032 |
ugg | +0.031 |
Cream | +0.029 |
Neg logit contributions | |
pieces | -0.059 |
od | -0.051 |
Maria | -0.051 |
circles | -0.050 |
lier | -0.048 |
Token | and |
Token ID | 290 |
Feature activation | -2.34e-03 |
Pos logit contributions | |
robe | +0.022 |
arl | +0.021 |
melon | +0.019 |
cooler | +0.019 |
ugg | +0.019 |
Neg logit contributions | |
pieces | -0.019 |
od | -0.017 |
Maria | -0.016 |
lier | -0.015 |
one | -0.014 |
Token | 's |
Token ID | 338 |
Feature activation | -1.07e-03 |
Pos logit contributions | |
robe | +0.193 |
arl | +0.179 |
melon | +0.169 |
cooler | +0.168 |
ugg | +0.165 |
Neg logit contributions | |
pieces | -0.142 |
od | -0.124 |
Maria | -0.119 |
lier | -0.107 |
one | -0.107 |
Token | wonderful |
Token ID | 7932 |
Feature activation | +4.51e-02 |
Pos logit contributions | |
robe | +0.002 |
ugg | +0.002 |
arl | +0.001 |
melon | +0.001 |
ranc | +0.001 |
Neg logit contributions | |
pieces | -0.002 |
od | -0.002 |
Maria | -0.002 |
lier | -0.002 |
one | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
pieces | +0.069 |
od | +0.060 |
Maria | +0.058 |
lier | +0.053 |
one | +0.053 |
Neg logit contributions | |
robe | -0.087 |
arl | -0.081 |
melon | -0.076 |
cooler | -0.075 |
ugg | -0.074 |
Token | my |
Token ID | 616 |
Feature activation | -1.94e-03 |
Pos logit contributions | |
robe | +0.032 |
arl | +0.029 |
cooler | +0.027 |
melon | +0.027 |
ugg | +0.025 |
Neg logit contributions | |
pieces | -0.044 |
od | -0.040 |
Maria | -0.039 |
one | -0.036 |
enta | -0.036 |
Token | d |
Token ID | 288 |
Feature activation | -3.28e-03 |
Pos logit contributions | |
robe | +0.003 |
arl | +0.002 |
melon | +0.002 |
cooler | +0.002 |
ugg | +0.002 |
Neg logit contributions | |
pieces | -0.004 |
od | -0.004 |
Maria | -0.004 |
lier | -0.003 |
one | -0.003 |
Token | arl |
Token ID | 7063 |
Feature activation | -2.17e-01 |
Pos logit contributions | |
robe | +0.003 |
arl | +0.003 |
melon | +0.002 |
cooler | +0.002 |
ugg | +0.002 |
Neg logit contributions | |
pieces | -0.008 |
od | -0.008 |
Maria | -0.008 |
lier | -0.007 |
one | -0.007 |
Token | ings |
Token ID | 654 |
Feature activation | -4.74e-02 |
Pos logit contributions | |
robe | +0.314 |
arl | +0.282 |
cooler | +0.262 |
melon | +0.261 |
ugg | +0.255 |
Neg logit contributions | |
pieces | -0.427 |
od | -0.388 |
Maria | -0.373 |
one | -0.352 |
lier | -0.350 |
Token | !" |
Token ID | 2474 |
Feature activation | +2.86e-02 |
Pos logit contributions | |
robe | +0.095 |
arl | +0.089 |
melon | +0.084 |
cooler | +0.083 |
ugg | +0.082 |
Neg logit contributions | |
pieces | -0.070 |
od | -0.060 |
Maria | -0.058 |
enta | -0.052 |
one | -0.051 |
Token | Mom |
Token ID | 11254 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
pieces | +0.054 |
od | +0.049 |
enta | +0.046 |
Maria | +0.046 |
one | +0.045 |
Neg logit contributions | |
robe | -0.044 |
melon | -0.039 |
arl | -0.038 |
cooler | -0.038 |
Cream | -0.037 |
Token | my |
Token ID | 1820 |
Feature activation | +3.41e-02 |
Pos logit contributions | |
robe | +0.036 |
melon | +0.035 |
arl | +0.034 |
cooler | +0.034 |
ugg | +0.032 |
Neg logit contributions | |
pieces | -0.038 |
od | -0.032 |
Maria | -0.030 |
hours | -0.030 |
rect | -0.030 |
Token | says |
Token ID | 1139 |
Feature activation | +2.84e-02 |
Pos logit contributions | |
pieces | +0.057 |
od | +0.047 |
Maria | +0.046 |
one | +0.046 |
enta | +0.045 |
Neg logit contributions | |
robe | -0.057 |
melon | -0.056 |
cooler | -0.053 |
Cream | -0.050 |
waitress | -0.049 |
Token | was |
Token ID | 373 |
Feature activation | +4.99e-03 |
Pos logit contributions | |
robe | +0.004 |
arl | +0.004 |
cooler | +0.003 |
melon | +0.003 |
ugg | +0.003 |
Neg logit contributions | |
pieces | -0.003 |
od | -0.003 |
Maria | -0.003 |
asing | -0.002 |
inda | -0.002 |
Token | over |
Token ID | 625 |
Feature activation | -3.34e-02 |
Pos logit contributions | |
pieces | +0.008 |
Maria | +0.007 |
od | +0.007 |
lier | +0.007 |
one | +0.007 |
Neg logit contributions | |
robe | -0.009 |
ugg | -0.008 |
arl | -0.008 |
melon | -0.007 |
Cream | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +6.50e-03 |
Pos logit contributions | |
robe | +0.051 |
cooler | +0.046 |
arl | +0.046 |
cabin | +0.043 |
melon | +0.043 |
Neg logit contributions | |
pieces | -0.068 |
od | -0.058 |
Maria | -0.055 |
ata | -0.053 |
asing | -0.053 |
Token | He |
Token ID | 679 |
Feature activation | -3.93e-02 |
Pos logit contributions | |
pieces | +0.014 |
od | +0.012 |
Maria | +0.012 |
lier | +0.011 |
one | +0.011 |
Neg logit contributions | |
robe | -0.009 |
arl | -0.008 |
melon | -0.007 |
cooler | -0.007 |
ugg | -0.007 |
Token | sn |
Token ID | 3013 |
Feature activation | -3.87e-02 |
Pos logit contributions | |
robe | +0.065 |
cooler | +0.059 |
ugg | +0.055 |
ranc | +0.054 |
cabin | +0.054 |
Neg logit contributions | |
pieces | -0.074 |
od | -0.065 |
Maria | -0.061 |
asing | -0.056 |
one | -0.056 |
Token | ugg |
Token ID | 6837 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
robe | +0.028 |
ranc | +0.023 |
melon | +0.022 |
cooler | +0.020 |
waitress | +0.019 |
Neg logit contributions | |
pieces | -0.097 |
Maria | -0.089 |
ill | -0.088 |
ez | -0.088 |
asing | -0.087 |
Token | led |
Token ID | 992 |
Feature activation | -3.99e-02 |
Pos logit contributions | |
robe | +0.414 |
arl | +0.395 |
melon | +0.381 |
cooler | +0.370 |
ugg | +0.365 |
Neg logit contributions | |
pieces | -0.302 |
od | -0.267 |
Maria | -0.254 |
lier | -0.237 |
lar | -0.230 |
Token | with |
Token ID | 351 |
Feature activation | -4.16e-02 |
Pos logit contributions | |
robe | +0.065 |
arl | +0.061 |
melon | +0.054 |
ugg | +0.054 |
Cream | +0.051 |
Neg logit contributions | |
pieces | -0.069 |
od | -0.063 |
Maria | -0.062 |
circles | -0.060 |
lier | -0.060 |
Token | them |
Token ID | 606 |
Feature activation | -4.03e-02 |
Pos logit contributions | |
robe | +0.070 |
arl | +0.064 |
melon | +0.059 |
ugg | +0.059 |
cooler | +0.058 |
Neg logit contributions | |
pieces | -0.074 |
od | -0.066 |
Maria | -0.064 |
one | -0.059 |
lier | -0.059 |
Token | and |
Token ID | 290 |
Feature activation | +1.27e-04 |
Pos logit contributions | |
robe | +0.076 |
arl | +0.071 |
melon | +0.067 |
cooler | +0.065 |
ugg | +0.065 |
Neg logit contributions | |
pieces | -0.061 |
od | -0.054 |
Maria | -0.052 |
lier | -0.048 |
one | -0.047 |
Token | listened |
Token ID | 16399 |
Feature activation | -6.97e-02 |
Pos logit contributions | |
pieces | +0.000 |
od | +0.000 |
Maria | +0.000 |
lier | +0.000 |
one | +0.000 |
Neg logit contributions | |
robe | -0.000 |
arl | -0.000 |
melon | -0.000 |
cooler | -0.000 |
ugg | -0.000 |
Token | the |
Token ID | 262 |
Feature activation | -2.05e-02 |
Pos logit contributions | |
pieces | +0.071 |
od | +0.063 |
Maria | +0.061 |
one | +0.056 |
lier | +0.056 |
Neg logit contributions | |
robe | -0.071 |
arl | -0.064 |
cooler | -0.060 |
melon | -0.060 |
ugg | -0.059 |
Token | cover |
Token ID | 3002 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
robe | +0.025 |
arl | +0.022 |
melon | +0.020 |
cooler | +0.020 |
ugg | +0.019 |
Neg logit contributions | |
pieces | -0.046 |
od | -0.042 |
Maria | -0.041 |
lier | -0.038 |
one | -0.038 |
Token | . |
Token ID | 13 |
Feature activation | -1.63e-03 |
Pos logit contributions | |
pieces | +0.029 |
od | +0.025 |
Maria | +0.025 |
enta | +0.024 |
asing | +0.023 |
Neg logit contributions | |
robe | -0.031 |
arl | -0.026 |
melon | -0.026 |
cooler | -0.026 |
ugg | -0.024 |
Token | She |
Token ID | 1375 |
Feature activation | -2.63e-02 |
Pos logit contributions | |
robe | +0.003 |
arl | +0.002 |
melon | +0.002 |
cooler | +0.002 |
ugg | +0.002 |
Neg logit contributions | |
pieces | -0.003 |
od | -0.003 |
Maria | -0.003 |
lier | -0.002 |
one | -0.002 |
Token | sn |
Token ID | 3013 |
Feature activation | -3.73e-02 |
Pos logit contributions | |
robe | +0.044 |
cooler | +0.039 |
ugg | +0.036 |
melon | +0.036 |
Cream | +0.035 |
Neg logit contributions | |
pieces | -0.049 |
od | -0.040 |
Maria | -0.039 |
one | -0.038 |
asing | -0.037 |
Token | ugg |
Token ID | 6837 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
robe | +0.028 |
ranc | +0.021 |
cabin | +0.020 |
cooler | +0.020 |
melon | +0.017 |
Neg logit contributions | |
pieces | -0.097 |
od | -0.088 |
lier | -0.088 |
ill | -0.087 |
Maria | -0.087 |
Token | led |
Token ID | 992 |
Feature activation | -3.91e-02 |
Pos logit contributions | |
robe | +0.409 |
arl | +0.386 |
melon | +0.367 |
cooler | +0.363 |
ugg | +0.355 |
Neg logit contributions | |
pieces | -0.320 |
od | -0.284 |
Maria | -0.269 |
lier | -0.252 |
one | -0.246 |
Token | with |
Token ID | 351 |
Feature activation | -3.87e-02 |
Pos logit contributions | |
robe | +0.067 |
arl | +0.064 |
ugg | +0.057 |
melon | +0.055 |
ranc | +0.054 |
Neg logit contributions | |
pieces | -0.065 |
od | -0.060 |
Maria | -0.057 |
circles | -0.056 |
lier | -0.055 |
Token | Mom |
Token ID | 11254 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
robe | +0.077 |
arl | +0.073 |
ugg | +0.069 |
melon | +0.067 |
ranc | +0.066 |
Neg logit contributions | |
pieces | -0.056 |
od | -0.049 |
Maria | -0.048 |
circles | -0.046 |
ill | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | -7.95e-03 |
Pos logit contributions | |
robe | +0.058 |
arl | +0.053 |
cooler | +0.051 |
melon | +0.051 |
ugg | +0.050 |
Neg logit contributions | |
pieces | -0.043 |
od | -0.038 |
Maria | -0.036 |
one | -0.032 |
enta | -0.032 |
Token | listened |
Token ID | 16399 |
Feature activation | -6.68e-02 |
Pos logit contributions | |
robe | +0.016 |
arl | +0.015 |
cooler | +0.014 |
melon | +0.014 |
ugg | +0.013 |
Neg logit contributions | |
pieces | -0.011 |
od | -0.010 |
Maria | -0.009 |
enta | -0.008 |
lier | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | -4.84e-02 |
Pos logit contributions | |
pieces | +0.025 |
Maria | +0.021 |
od | +0.020 |
enta | +0.019 |
one | +0.018 |
Neg logit contributions | |
robe | -0.034 |
arl | -0.031 |
cooler | -0.031 |
melon | -0.029 |
cabin | -0.029 |
Token | both |
Token ID | 1111 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
robe | +0.082 |
arl | +0.073 |
cooler | +0.072 |
melon | +0.071 |
ugg | +0.066 |
Neg logit contributions | |
pieces | -0.083 |
od | -0.077 |
Maria | -0.073 |
one | -0.068 |
lier | -0.068 |
Token | of |
Token ID | 286 |
Feature activation | +2.95e-03 |
Pos logit contributions | |
robe | +0.030 |
cooler | +0.027 |
melon | +0.026 |
arl | +0.026 |
ugg | +0.025 |
Neg logit contributions | |
pieces | -0.031 |
od | -0.028 |
Maria | -0.027 |
enta | -0.026 |
lier | -0.025 |
Token | them |
Token ID | 606 |
Feature activation | -5.79e-02 |
Pos logit contributions | |
pieces | +0.005 |
od | +0.005 |
ata | +0.005 |
Maria | +0.005 |
asing | +0.005 |
Neg logit contributions | |
robe | -0.004 |
cooler | -0.004 |
arl | -0.004 |
melon | -0.004 |
cabin | -0.004 |
Token | sn |
Token ID | 3013 |
Feature activation | -4.92e-02 |
Pos logit contributions | |
robe | +0.099 |
cooler | +0.086 |
melon | +0.085 |
arl | +0.083 |
ugg | +0.080 |
Neg logit contributions | |
pieces | -0.099 |
Maria | -0.091 |
od | -0.089 |
one | -0.082 |
enta | -0.081 |
Token | ugg |
Token ID | 6837 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
robe | +0.065 |
arl | +0.054 |
cooler | +0.053 |
melon | +0.052 |
ranc | +0.049 |
Neg logit contributions | |
pieces | -0.103 |
od | -0.092 |
Maria | -0.089 |
lier | -0.088 |
enta | -0.085 |
Token | led |
Token ID | 992 |
Feature activation | -4.21e-02 |
Pos logit contributions | |
robe | +0.376 |
arl | +0.342 |
cooler | +0.318 |
melon | +0.317 |
ugg | +0.311 |
Neg logit contributions | |
pieces | -0.379 |
od | -0.337 |
Maria | -0.331 |
one | -0.305 |
enta | -0.298 |
Token | together |
Token ID | 1978 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
robe | +0.068 |
arl | +0.060 |
cooler | +0.060 |
melon | +0.058 |
ugg | +0.056 |
Neg logit contributions | |
pieces | -0.077 |
od | -0.070 |
Maria | -0.067 |
one | -0.063 |
enta | -0.063 |
Token | under |
Token ID | 739 |
Feature activation | +5.40e-02 |
Pos logit contributions | |
pieces | +0.053 |
od | +0.048 |
Maria | +0.047 |
enta | +0.043 |
one | +0.042 |
Neg logit contributions | |
robe | -0.057 |
arl | -0.050 |
cooler | -0.050 |
melon | -0.049 |
ugg | -0.047 |
Token | a |
Token ID | 257 |
Feature activation | -3.03e-03 |
Pos logit contributions | |
pieces | +0.103 |
od | +0.093 |
Maria | +0.089 |
asing | +0.084 |
one | +0.084 |
Neg logit contributions | |
robe | -0.083 |
arl | -0.075 |
cooler | -0.073 |
melon | -0.072 |
ugg | -0.067 |
Token | cozy |
Token ID | 37438 |
Feature activation | -4.05e-02 |
Pos logit contributions | |
robe | +0.004 |
arl | +0.003 |
cooler | +0.003 |
melon | +0.003 |
ugg | +0.003 |
Neg logit contributions | |
pieces | -0.007 |
od | -0.006 |
Maria | -0.006 |
one | -0.006 |
lier | -0.006 |
Token | she |
Token ID | 673 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
robe | +0.209 |
arl | +0.195 |
melon | +0.176 |
ugg | +0.174 |
cooler | +0.174 |
Neg logit contributions | |
pieces | -0.241 |
od | -0.217 |
Maria | -0.212 |
one | -0.195 |
lier | -0.194 |
Token | pur |
Token ID | 1308 |
Feature activation | -4.43e-02 |
Pos logit contributions | |
pieces | +0.037 |
od | +0.032 |
Maria | +0.031 |
one | +0.029 |
circles | +0.028 |
Neg logit contributions | |
robe | -0.042 |
cooler | -0.036 |
melon | -0.036 |
ugg | -0.035 |
arl | -0.035 |
Token | red |
Token ID | 445 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
robe | +0.093 |
arl | +0.087 |
ugg | +0.083 |
melon | +0.080 |
Sleep | +0.076 |
Neg logit contributions | |
pieces | -0.065 |
od | -0.053 |
Maria | -0.052 |
one | -0.048 |
circles | -0.046 |
Token | and |
Token ID | 290 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
robe | +0.207 |
arl | +0.189 |
melon | +0.179 |
cooler | +0.179 |
ugg | +0.176 |
Neg logit contributions | |
pieces | -0.167 |
od | -0.146 |
Maria | -0.141 |
one | -0.129 |
lier | -0.127 |
Token | sn |
Token ID | 3013 |
Feature activation | -4.90e-02 |
Pos logit contributions | |
robe | +0.028 |
arl | +0.025 |
cooler | +0.025 |
melon | +0.024 |
ugg | +0.024 |
Neg logit contributions | |
pieces | -0.027 |
od | -0.024 |
Maria | -0.023 |
one | -0.022 |
lier | -0.021 |
Token | ugg |
Token ID | 6837 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
robe | +0.038 |
ranc | +0.027 |
cooler | +0.027 |
cabin | +0.027 |
refrigerator | +0.024 |
Neg logit contributions | |
pieces | -0.125 |
ez | -0.111 |
Maria | -0.111 |
lier | -0.110 |
asing | -0.110 |
Token | led |
Token ID | 992 |
Feature activation | -3.83e-02 |
Pos logit contributions | |
robe | +0.424 |
arl | +0.398 |
melon | +0.388 |
cooler | +0.382 |
ugg | +0.366 |
Neg logit contributions | |
pieces | -0.291 |
od | -0.255 |
Maria | -0.240 |
lier | -0.229 |
inda | -0.218 |
Token | up |
Token ID | 510 |
Feature activation | -7.98e-02 |
Pos logit contributions | |
robe | +0.059 |
arl | +0.052 |
cooler | +0.049 |
melon | +0.049 |
ugg | +0.048 |
Neg logit contributions | |
pieces | -0.074 |
od | -0.067 |
Maria | -0.064 |
one | -0.061 |
enta | -0.060 |
Token | to |
Token ID | 284 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
robe | +0.130 |
arl | +0.117 |
cooler | +0.112 |
melon | +0.111 |
ugg | +0.109 |
Neg logit contributions | |
pieces | -0.150 |
od | -0.134 |
Maria | -0.129 |
one | -0.123 |
enta | -0.121 |
Token | her |
Token ID | 607 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
pieces | +0.048 |
od | +0.045 |
Maria | +0.043 |
one | +0.042 |
lier | +0.041 |
Neg logit contributions | |
robe | -0.027 |
arl | -0.023 |
cooler | -0.022 |
Sleep | -0.021 |
melon | -0.021 |
Token | owner |
Token ID | 4870 |
Feature activation | -1.99e-02 |
Pos logit contributions | |
robe | +0.023 |
arl | +0.020 |
melon | +0.018 |
cooler | +0.018 |
ugg | +0.018 |
Neg logit contributions | |
pieces | -0.037 |
od | -0.034 |
Maria | -0.033 |
lier | -0.031 |
one | -0.031 |
Token | name |
Token ID | 1438 |
Feature activation | -7.91e-02 |
Pos logit contributions | |
robe | +0.020 |
arl | +0.020 |
ugg | +0.018 |
melon | +0.017 |
cooler | +0.014 |
Neg logit contributions | |
pieces | -0.051 |
od | -0.047 |
circles | -0.045 |
Maria | -0.044 |
pad | -0.044 |
Token | is |
Token ID | 318 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
robe | +0.135 |
arl | +0.122 |
cooler | +0.114 |
melon | +0.114 |
ugg | +0.112 |
Neg logit contributions | |
pieces | -0.139 |
od | -0.124 |
Maria | -0.120 |
one | -0.110 |
lier | -0.110 |
Token | Mr |
Token ID | 1770 |
Feature activation | -7.38e-02 |
Pos logit contributions | |
robe | +0.026 |
arl | +0.023 |
melon | +0.021 |
ugg | +0.021 |
cooler | +0.021 |
Neg logit contributions | |
pieces | -0.040 |
od | -0.036 |
Maria | -0.035 |
lier | -0.033 |
one | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -8.39e-03 |
Pos logit contributions | |
robe | +0.146 |
arl | +0.135 |
cooler | +0.127 |
melon | +0.127 |
ugg | +0.124 |
Neg logit contributions | |
pieces | -0.110 |
od | -0.096 |
Maria | -0.092 |
one | -0.083 |
lier | -0.083 |
Token | Sn |
Token ID | 5489 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
robe | +0.010 |
arl | +0.010 |
melon | +0.010 |
ugg | +0.009 |
cooler | +0.008 |
Neg logit contributions | |
pieces | -0.016 |
od | -0.015 |
Maria | -0.015 |
circles | -0.015 |
one | -0.015 |
Token | ugg |
Token ID | 6837 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
robe | +0.139 |
arl | +0.118 |
cooler | +0.117 |
melon | +0.113 |
ranc | +0.100 |
Neg logit contributions | |
pieces | -0.386 |
od | -0.363 |
lier | -0.346 |
Maria | -0.342 |
asing | -0.339 |
Token | les |
Token ID | 829 |
Feature activation | +4.57e-02 |
Pos logit contributions | |
robe | +0.547 |
arl | +0.542 |
melon | +0.539 |
ranc | +0.531 |
chard | +0.530 |
Neg logit contributions | |
pieces | -0.113 |
od | -0.093 |
ets | -0.076 |
one | -0.067 |
circles | -0.066 |
Token | ." |
Token ID | 526 |
Feature activation | -2.75e-02 |
Pos logit contributions | |
pieces | +0.067 |
od | +0.058 |
Maria | +0.056 |
lier | +0.051 |
one | +0.051 |
Neg logit contributions | |
robe | -0.090 |
arl | -0.084 |
melon | -0.079 |
cooler | -0.079 |
ugg | -0.078 |
Token |
|
Token ID | 198 |
Feature activation | +3.28e-03 |
Pos logit contributions | |
robe | +0.054 |
arl | +0.051 |
melon | +0.050 |
cooler | +0.047 |
ugg | +0.045 |
Neg logit contributions | |
pieces | -0.037 |
Maria | -0.032 |
od | -0.032 |
circles | -0.029 |
lier | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | +3.84e-03 |
Pos logit contributions | |
pieces | +0.006 |
od | +0.005 |
Maria | +0.005 |
one | +0.005 |
lier | +0.005 |
Neg logit contributions | |
robe | -0.005 |
arl | -0.005 |
ugg | -0.005 |
cooler | -0.005 |
melon | -0.005 |
Token | T |
Token ID | 51 |
Feature activation | -7.39e-02 |
Pos logit contributions | |
pieces | +0.004 |
od | +0.003 |
Maria | +0.003 |
lier | +0.003 |
ata | +0.003 |
Neg logit contributions | |
robe | -0.009 |
arl | -0.008 |
melon | -0.008 |
cooler | -0.008 |
ugg | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
pieces | +0.064 |
od | +0.060 |
Maria | +0.058 |
enta | +0.053 |
ata | +0.053 |
Neg logit contributions | |
robe | -0.058 |
arl | -0.052 |
cooler | -0.051 |
melon | -0.050 |
ugg | -0.047 |
Token | |
Token ID | 220 |
Feature activation | -4.73e-02 |
Pos logit contributions | |
robe | +0.025 |
arl | +0.023 |
melon | +0.021 |
cooler | +0.021 |
ugg | +0.021 |
Neg logit contributions | |
pieces | -0.033 |
od | -0.030 |
Maria | -0.029 |
enta | -0.027 |
asing | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | +9.35e-03 |
Pos logit contributions | |
robe | +0.078 |
arl | +0.071 |
cooler | +0.064 |
melon | +0.064 |
ugg | +0.064 |
Neg logit contributions | |
pieces | -0.086 |
od | -0.077 |
Maria | -0.073 |
lier | -0.069 |
enta | -0.069 |
Token |
|
Token ID | 198 |
Feature activation | +1.00e-02 |
Pos logit contributions | |
pieces | +0.018 |
od | +0.017 |
asing | +0.016 |
lier | +0.016 |
enta | +0.015 |
Neg logit contributions | |
robe | -0.013 |
arl | -0.012 |
ugg | -0.011 |
Cream | -0.011 |
cooler | -0.011 |
Token | Sn |
Token ID | 16501 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
pieces | +0.016 |
od | +0.014 |
enta | +0.012 |
lier | +0.012 |
Maria | +0.012 |
Neg logit contributions | |
robe | -0.018 |
arl | -0.017 |
ugg | -0.016 |
Cream | -0.016 |
cooler | -0.016 |
Token | ugg |
Token ID | 6837 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
robe | +0.180 |
arl | +0.161 |
cooler | +0.155 |
melon | +0.150 |
ranc | +0.145 |
Neg logit contributions | |
pieces | -0.204 |
od | -0.185 |
Maria | -0.175 |
lier | -0.169 |
one | -0.164 |
Token | les |
Token ID | 829 |
Feature activation | +4.78e-02 |
Pos logit contributions | |
robe | +0.483 |
arl | +0.475 |
melon | +0.446 |
cooler | +0.445 |
ugg | +0.434 |
Neg logit contributions | |
pieces | -0.212 |
od | -0.182 |
Maria | -0.154 |
lier | -0.151 |
lar | -0.143 |
Token | was |
Token ID | 373 |
Feature activation | +7.70e-03 |
Pos logit contributions | |
pieces | +0.077 |
od | +0.067 |
Maria | +0.066 |
one | +0.060 |
enta | +0.059 |
Neg logit contributions | |
robe | -0.090 |
cooler | -0.077 |
arl | -0.077 |
melon | -0.077 |
ugg | -0.076 |
Token | so |
Token ID | 523 |
Feature activation | -5.81e-02 |
Pos logit contributions | |
pieces | +0.014 |
od | +0.013 |
Maria | +0.013 |
lier | +0.012 |
circles | +0.011 |
Neg logit contributions | |
robe | -0.012 |
arl | -0.011 |
melon | -0.010 |
ugg | -0.010 |
cooler | -0.010 |
Token | sad |
Token ID | 6507 |
Feature activation | -6.21e-02 |
Pos logit contributions | |
robe | +0.099 |
arl | +0.087 |
melon | +0.081 |
ranc | +0.077 |
cooler | +0.075 |
Neg logit contributions | |
pieces | -0.097 |
Maria | -0.091 |
od | -0.090 |
realised | -0.082 |
lier | -0.081 |
Token | to |
Token ID | 284 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
robe | +0.097 |
arl | +0.086 |
cooler | +0.083 |
melon | +0.080 |
ugg | +0.078 |
Neg logit contributions | |
pieces | -0.121 |
od | -0.106 |
Maria | -0.106 |
enta | -0.099 |
one | -0.095 |
Token | That |
Token ID | 2504 |
Feature activation | -8.01e-02 |
Pos logit contributions | |
pieces | +0.031 |
lier | +0.027 |
od | +0.025 |
pad | +0.025 |
one | +0.025 |
Neg logit contributions | |
robe | -0.034 |
arl | -0.030 |
ugg | -0.029 |
Sleep | -0.029 |
Cream | -0.027 |
Token | night |
Token ID | 1755 |
Feature activation | +3.04e-02 |
Pos logit contributions | |
robe | +0.128 |
arl | +0.111 |
cooler | +0.108 |
melon | +0.107 |
ugg | +0.106 |
Neg logit contributions | |
pieces | -0.149 |
od | -0.133 |
Maria | -0.129 |
lier | -0.120 |
one | -0.119 |
Token | , |
Token ID | 11 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
pieces | +0.056 |
od | +0.049 |
Maria | +0.047 |
lier | +0.045 |
one | +0.045 |
Neg logit contributions | |
robe | -0.050 |
cooler | -0.041 |
arl | -0.041 |
Sleep | -0.040 |
ugg | -0.040 |
Token | Julie |
Token ID | 21946 |
Feature activation | +6.48e-02 |
Pos logit contributions | |
robe | +0.025 |
Sleep | +0.022 |
cooler | +0.021 |
cabin | +0.021 |
tired | +0.020 |
Neg logit contributions | |
pieces | -0.022 |
lier | -0.019 |
od | -0.018 |
one | -0.018 |
ez | -0.018 |
Token | sn |
Token ID | 3013 |
Feature activation | -3.89e-02 |
Pos logit contributions | |
pieces | +0.108 |
od | +0.093 |
Maria | +0.091 |
asing | +0.083 |
one | +0.081 |
Neg logit contributions | |
robe | -0.116 |
cooler | -0.103 |
ugg | -0.103 |
Sleep | -0.101 |
ranc | -0.098 |
Token | ugg |
Token ID | 6837 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
robe | +0.032 |
ranc | +0.028 |
cooler | +0.025 |
melon | +0.025 |
cabin | +0.024 |
Neg logit contributions | |
pieces | -0.091 |
lier | -0.083 |
mastered | -0.082 |
asing | -0.082 |
aw | -0.082 |
Token | led |
Token ID | 992 |
Feature activation | -3.74e-02 |
Pos logit contributions | |
robe | +0.410 |
arl | +0.394 |
melon | +0.375 |
cooler | +0.369 |
ugg | +0.357 |
Neg logit contributions | |
pieces | -0.309 |
od | -0.274 |
Maria | -0.259 |
lier | -0.240 |
inda | -0.234 |
Token | into |
Token ID | 656 |
Feature activation | +2.49e-02 |
Pos logit contributions | |
robe | +0.058 |
arl | +0.056 |
melon | +0.049 |
ugg | +0.048 |
Cream | +0.046 |
Neg logit contributions | |
pieces | -0.068 |
od | -0.062 |
Maria | -0.060 |
lier | -0.057 |
circles | -0.057 |
Token | her |
Token ID | 607 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
pieces | +0.042 |
od | +0.037 |
Maria | +0.036 |
circles | +0.033 |
lier | +0.032 |
Neg logit contributions | |
robe | -0.044 |
arl | -0.041 |
melon | -0.038 |
ugg | -0.038 |
cooler | -0.037 |
Token | own |
Token ID | 898 |
Feature activation | +7.38e-02 |
Pos logit contributions | |
arl | +0.029 |
melon | +0.029 |
robe | +0.025 |
ugg | +0.024 |
chard | +0.023 |
Neg logit contributions | |
minutes | -0.022 |
hours | -0.021 |
circles | -0.021 |
pieces | -0.020 |
's | -0.020 |
Token | soft |
Token ID | 2705 |
Feature activation | -1.15e-03 |
Pos logit contributions | |
pieces | +0.120 |
circles | +0.113 |
hours | +0.108 |
minutes | +0.106 |
's | +0.106 |
Neg logit contributions | |
arl | -0.116 |
melon | -0.115 |
robe | -0.106 |
ugg | -0.100 |
chard | -0.095 |
Token | asleep |
Token ID | 16039 |
Feature activation | -9.81e-02 |
Pos logit contributions | |
pieces | +0.039 |
circles | +0.037 |
od | +0.037 |
Maria | +0.036 |
ill | +0.035 |
Neg logit contributions | |
robe | -0.029 |
melon | -0.028 |
arl | -0.027 |
Cream | -0.026 |
ugg | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
robe | +0.191 |
arl | +0.177 |
cooler | +0.166 |
melon | +0.166 |
ugg | +0.163 |
Neg logit contributions | |
pieces | -0.148 |
od | -0.129 |
Maria | -0.124 |
one | -0.113 |
lier | -0.112 |
Token | Fl |
Token ID | 1610 |
Feature activation | -9.07e-02 |
Pos logit contributions | |
Maria | +0.013 |
pieces | +0.013 |
od | +0.013 |
circles | +0.012 |
ill | +0.012 |
Neg logit contributions | |
robe | -0.022 |
melon | -0.021 |
arl | -0.020 |
cooler | -0.019 |
yacht | -0.018 |
Token | uffy |
Token ID | 15352 |
Feature activation | -5.56e-02 |
Pos logit contributions | |
robe | +0.107 |
melon | +0.092 |
cooler | +0.086 |
arl | +0.086 |
ranc | +0.083 |
Neg logit contributions | |
od | -0.182 |
pieces | -0.177 |
ill | -0.172 |
asing | -0.168 |
lier | -0.167 |
Token | sn |
Token ID | 3013 |
Feature activation | -4.07e-02 |
Pos logit contributions | |
robe | +0.101 |
arl | +0.094 |
melon | +0.088 |
cooler | +0.087 |
ugg | +0.085 |
Neg logit contributions | |
pieces | -0.089 |
od | -0.079 |
Maria | -0.076 |
lier | -0.070 |
one | -0.070 |
Token | ugg |
Token ID | 6837 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
robe | +0.029 |
ranc | +0.025 |
melon | +0.023 |
cooler | +0.022 |
cabin | +0.021 |
Neg logit contributions | |
pieces | -0.098 |
lier | -0.092 |
aw | -0.091 |
e | -0.090 |
ill | -0.090 |
Token | led |
Token ID | 992 |
Feature activation | -3.86e-02 |
Pos logit contributions | |
robe | +0.438 |
arl | +0.420 |
melon | +0.409 |
cooler | +0.396 |
ugg | +0.389 |
Neg logit contributions | |
pieces | -0.261 |
od | -0.230 |
Maria | -0.214 |
lier | -0.209 |
one | -0.198 |
Token | with |
Token ID | 351 |
Feature activation | -3.89e-02 |
Pos logit contributions | |
robe | +0.080 |
arl | +0.078 |
melon | +0.073 |
ugg | +0.070 |
Cream | +0.069 |
Neg logit contributions | |
pieces | -0.046 |
od | -0.042 |
lier | -0.042 |
Maria | -0.040 |
circles | -0.040 |
Token | her |
Token ID | 607 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
robe | +0.080 |
melon | +0.076 |
arl | +0.075 |
ugg | +0.068 |
ranc | +0.068 |
Neg logit contributions | |
pieces | -0.046 |
circles | -0.043 |
Maria | -0.043 |
od | -0.042 |
e | -0.040 |
Token | and |
Token ID | 290 |
Feature activation | -1.42e-03 |
Pos logit contributions | |
robe | +0.035 |
arl | +0.034 |
melon | +0.033 |
ugg | +0.032 |
Cream | +0.031 |
Neg logit contributions | |
pieces | -0.026 |
od | -0.022 |
circles | -0.021 |
Maria | -0.021 |
lier | -0.020 |
Token | pur |
Token ID | 1308 |
Feature activation | -3.61e-02 |
Pos logit contributions | |
robe | +0.002 |
arl | +0.002 |
melon | +0.002 |
ugg | +0.002 |
Cream | +0.002 |
Neg logit contributions | |
pieces | -0.002 |
Maria | -0.002 |
od | -0.002 |
imagined | -0.002 |
circles | -0.002 |
Token | on |
Token ID | 319 |
Feature activation | -1.39e-04 |
Pos logit contributions | |
pieces | +0.053 |
od | +0.046 |
Maria | +0.045 |
one | +0.041 |
lier | +0.040 |
Neg logit contributions | |
robe | -0.070 |
arl | -0.065 |
cooler | -0.061 |
melon | -0.061 |
ugg | -0.060 |
Token | the |
Token ID | 262 |
Feature activation | -5.70e-02 |
Pos logit contributions | |
robe | +0.000 |
cooler | +0.000 |
crib | +0.000 |
arl | +0.000 |
cabin | +0.000 |
Neg logit contributions | |
pieces | -0.000 |
od | -0.000 |
asing | -0.000 |
ata | -0.000 |
lier | -0.000 |
Token | bed |
Token ID | 3996 |
Feature activation | -3.77e-02 |
Pos logit contributions | |
robe | +0.078 |
arl | +0.067 |
cooler | +0.065 |
melon | +0.064 |
ugg | +0.058 |
Neg logit contributions | |
pieces | -0.120 |
od | -0.112 |
Maria | -0.108 |
lier | -0.102 |
one | -0.102 |
Token | and |
Token ID | 290 |
Feature activation | -3.04e-02 |
Pos logit contributions | |
robe | +0.072 |
arl | +0.064 |
cooler | +0.060 |
melon | +0.060 |
ugg | +0.059 |
Neg logit contributions | |
pieces | -0.061 |
od | -0.053 |
Maria | -0.052 |
one | -0.047 |
enta | -0.047 |
Token | sn |
Token ID | 3013 |
Feature activation | -5.44e-02 |
Pos logit contributions | |
robe | +0.053 |
cooler | +0.047 |
arl | +0.047 |
melon | +0.045 |
ugg | +0.044 |
Neg logit contributions | |
pieces | -0.052 |
od | -0.047 |
Maria | -0.045 |
one | -0.042 |
lier | -0.042 |
Token | ugg |
Token ID | 6837 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
robe | +0.055 |
ranc | +0.050 |
cabin | +0.045 |
cooler | +0.043 |
waitress | +0.043 |
Neg logit contributions | |
pieces | -0.121 |
od | -0.108 |
ill | -0.107 |
Maria | -0.105 |
mastered | -0.104 |
Token | led |
Token ID | 992 |
Feature activation | -4.13e-02 |
Pos logit contributions | |
robe | +0.410 |
arl | +0.386 |
melon | +0.368 |
cooler | +0.361 |
ugg | +0.354 |
Neg logit contributions | |
pieces | -0.312 |
od | -0.277 |
Maria | -0.261 |
lier | -0.245 |
one | -0.238 |
Token | together |
Token ID | 1978 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
robe | +0.064 |
arl | +0.057 |
cooler | +0.054 |
melon | +0.053 |
ugg | +0.052 |
Neg logit contributions | |
pieces | -0.079 |
od | -0.072 |
Maria | -0.069 |
one | -0.065 |
enta | -0.064 |
Token | . |
Token ID | 13 |
Feature activation | +4.17e-03 |
Pos logit contributions | |
pieces | +0.047 |
od | +0.041 |
Maria | +0.041 |
enta | +0.038 |
one | +0.037 |
Neg logit contributions | |
robe | -0.048 |
cooler | -0.042 |
arl | -0.041 |
melon | -0.041 |
ugg | -0.039 |
Token | They |
Token ID | 1119 |
Feature activation | -3.20e-02 |
Pos logit contributions | |
pieces | +0.008 |
od | +0.007 |
Maria | +0.007 |
lier | +0.006 |
one | +0.006 |
Neg logit contributions | |
robe | -0.007 |
arl | -0.006 |
cooler | -0.005 |
melon | -0.005 |
ugg | -0.005 |
Token | said |
Token ID | 531 |
Feature activation | +3.02e-03 |
Pos logit contributions | |
robe | +0.059 |
arl | +0.055 |
melon | +0.051 |
cooler | +0.051 |
ugg | +0.050 |
Neg logit contributions | |
pieces | -0.051 |
od | -0.045 |
Maria | -0.043 |
lier | -0.040 |
one | -0.039 |
Token | L |
Token ID | 43 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
robe | +0.013 |
arl | +0.013 |
ugg | +0.011 |
cooler | +0.011 |
Cream | +0.011 |
Neg logit contributions | |
pieces | -0.012 |
od | -0.011 |
lier | -0.010 |
enta | -0.010 |
ets | -0.009 |
Token | ola |
Token ID | 5708 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
robe | +0.019 |
arl | +0.018 |
ugg | +0.015 |
melon | +0.012 |
waitress | +0.012 |
Neg logit contributions | |
pieces | -0.030 |
rect | -0.025 |
enta | -0.025 |
lier | -0.025 |
ples | -0.025 |
Token | smiled |
Token ID | 13541 |
Feature activation | -2.30e-02 |
Pos logit contributions | |
pieces | +0.029 |
Maria | +0.024 |
od | +0.024 |
one | +0.023 |
asing | +0.022 |
Neg logit contributions | |
robe | -0.027 |
ugg | -0.024 |
melon | -0.024 |
cooler | -0.023 |
Cream | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | -3.08e-03 |
Pos logit contributions | |
robe | +0.051 |
arl | +0.048 |
melon | +0.046 |
ugg | +0.045 |
cooler | +0.045 |
Neg logit contributions | |
pieces | -0.027 |
od | -0.024 |
Maria | -0.022 |
lier | -0.021 |
circles | -0.020 |
Token | sn |
Token ID | 3013 |
Feature activation | -4.76e-02 |
Pos logit contributions | |
robe | +0.006 |
cooler | +0.005 |
arl | +0.005 |
ugg | +0.005 |
melon | +0.005 |
Neg logit contributions | |
pieces | -0.005 |
od | -0.004 |
Maria | -0.004 |
one | -0.004 |
lier | -0.004 |
Token | ugg |
Token ID | 6837 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
robe | +0.032 |
cooler | +0.021 |
ranc | +0.021 |
cabin | +0.021 |
arl | +0.020 |
Neg logit contributions | |
pieces | -0.126 |
od | -0.117 |
Maria | -0.115 |
lier | -0.112 |
enta | -0.112 |
Token | led |
Token ID | 992 |
Feature activation | -3.68e-02 |
Pos logit contributions | |
robe | +0.397 |
arl | +0.370 |
melon | +0.349 |
cooler | +0.347 |
ugg | +0.339 |
Neg logit contributions | |
pieces | -0.334 |
od | -0.298 |
Maria | -0.283 |
lier | -0.261 |
one | -0.258 |
Token | in |
Token ID | 287 |
Feature activation | +1.64e-03 |
Pos logit contributions | |
robe | +0.063 |
arl | +0.057 |
cooler | +0.053 |
melon | +0.053 |
ugg | +0.052 |
Neg logit contributions | |
pieces | -0.064 |
od | -0.057 |
Maria | -0.055 |
one | -0.051 |
lier | -0.051 |
Token | the |
Token ID | 262 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
pieces | +0.003 |
od | +0.003 |
Maria | +0.003 |
one | +0.002 |
lier | +0.002 |
Neg logit contributions | |
robe | -0.003 |
arl | -0.002 |
cooler | -0.002 |
melon | -0.002 |
ugg | -0.002 |
Token | warmth |
Token ID | 23125 |
Feature activation | +1.35e-02 |
Pos logit contributions | |
robe | +0.020 |
arl | +0.019 |
melon | +0.017 |
ugg | +0.016 |
ranc | +0.016 |
Neg logit contributions | |
pieces | -0.024 |
Maria | -0.021 |
circles | -0.021 |
od | -0.020 |
lier | -0.019 |
Token | of |
Token ID | 286 |
Feature activation | +1.35e-02 |
Pos logit contributions | |
pieces | +0.022 |
od | +0.019 |
Maria | +0.019 |
enta | +0.017 |
one | +0.016 |
Neg logit contributions | |
robe | -0.026 |
arl | -0.023 |
cooler | -0.023 |
melon | -0.023 |
ugg | -0.022 |
Token | - |
Token ID | 532 |
Feature activation | -4.79e-02 |
Pos logit contributions | |
robe | +0.053 |
arl | +0.049 |
melon | +0.047 |
cooler | +0.046 |
ugg | +0.045 |
Neg logit contributions | |
pieces | -0.042 |
od | -0.036 |
Maria | -0.036 |
enta | -0.032 |
one | -0.032 |
Token | just |
Token ID | 655 |
Feature activation | +2.54e-02 |
Pos logit contributions | |
robe | +0.079 |
arl | +0.074 |
melon | +0.067 |
ugg | +0.066 |
cooler | +0.063 |
Neg logit contributions | |
pieces | -0.084 |
od | -0.075 |
Maria | -0.073 |
circles | -0.069 |
one | -0.069 |
Token | perfect |
Token ID | 2818 |
Feature activation | +6.62e-02 |
Pos logit contributions | |
pieces | +0.047 |
od | +0.042 |
Maria | +0.041 |
lier | +0.037 |
one | +0.037 |
Neg logit contributions | |
robe | -0.041 |
arl | -0.037 |
cooler | -0.036 |
melon | -0.035 |
ugg | -0.034 |
Token | for |
Token ID | 329 |
Feature activation | +3.59e-02 |
Pos logit contributions | |
pieces | +0.110 |
od | +0.098 |
Maria | +0.094 |
lier | +0.088 |
one | +0.088 |
Neg logit contributions | |
robe | -0.115 |
arl | -0.107 |
melon | -0.099 |
ugg | -0.098 |
cooler | -0.098 |
Token | sn |
Token ID | 3013 |
Feature activation | -4.16e-02 |
Pos logit contributions | |
pieces | +0.068 |
od | +0.061 |
Maria | +0.059 |
one | +0.056 |
lier | +0.056 |
Neg logit contributions | |
robe | -0.055 |
arl | -0.050 |
cooler | -0.048 |
melon | -0.047 |
ugg | -0.045 |
Token | ugg |
Token ID | 6837 |
Feature activation | -2.14e-01 |
Pos logit contributions | |
robe | +0.065 |
melon | +0.056 |
cooler | +0.055 |
waitress | +0.054 |
ranc | +0.054 |
Neg logit contributions | |
pieces | -0.074 |
lier | -0.063 |
ets | -0.062 |
od | -0.062 |
inda | -0.061 |
Token | ling |
Token ID | 1359 |
Feature activation | -2.35e-02 |
Pos logit contributions | |
robe | +0.437 |
arl | +0.411 |
melon | +0.390 |
cooler | +0.386 |
ugg | +0.383 |
Neg logit contributions | |
pieces | -0.291 |
od | -0.255 |
Maria | -0.239 |
lier | -0.224 |
one | -0.216 |
Token | . |
Token ID | 13 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
robe | +0.040 |
arl | +0.036 |
cooler | +0.035 |
melon | +0.034 |
ugg | +0.033 |
Neg logit contributions | |
pieces | -0.041 |
od | -0.037 |
Maria | -0.035 |
one | -0.033 |
enta | -0.032 |
Token | The |
Token ID | 383 |
Feature activation | -8.64e-03 |
Pos logit contributions | |
robe | +0.019 |
arl | +0.018 |
melon | +0.017 |
cooler | +0.016 |
ugg | +0.016 |
Neg logit contributions | |
pieces | -0.020 |
od | -0.018 |
Maria | -0.018 |
one | -0.016 |
lier | -0.016 |
Token | baby |
Token ID | 5156 |
Feature activation | +3.58e-02 |
Pos logit contributions | |
robe | +0.015 |
arl | +0.014 |
melon | +0.013 |
ugg | +0.013 |
cooler | +0.011 |
Neg logit contributions | |
pieces | -0.014 |
Maria | -0.012 |
hours | -0.012 |
pad | -0.012 |
circles | -0.012 |
Token | lamb |
Token ID | 19343 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
pieces | +0.063 |
od | +0.056 |
Maria | +0.054 |
one | +0.049 |
lier | +0.049 |
Neg logit contributions | |
robe | -0.061 |
arl | -0.054 |
cooler | -0.052 |
melon | -0.052 |
ugg | -0.051 |
Token | his |
Token ID | 465 |
Feature activation | -3.47e-02 |
Pos logit contributions | |
robe | +0.028 |
arl | +0.028 |
melon | +0.025 |
ugg | +0.024 |
ranc | +0.023 |
Neg logit contributions | |
pieces | -0.024 |
od | -0.021 |
Maria | -0.021 |
circles | -0.020 |
hours | -0.019 |
Token | painting |
Token ID | 12036 |
Feature activation | +6.20e-02 |
Pos logit contributions | |
robe | +0.048 |
arl | +0.045 |
melon | +0.041 |
ugg | +0.040 |
cooler | +0.039 |
Neg logit contributions | |
pieces | -0.070 |
od | -0.062 |
Maria | -0.061 |
circles | -0.057 |
lier | -0.057 |
Token | supplies |
Token ID | 9416 |
Feature activation | -5.43e-03 |
Pos logit contributions | |
pieces | +0.099 |
od | +0.087 |
Maria | +0.084 |
one | +0.076 |
lier | +0.076 |
Neg logit contributions | |
robe | -0.118 |
arl | -0.105 |
cooler | -0.101 |
melon | -0.100 |
ugg | -0.099 |
Token | , |
Token ID | 11 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
robe | +0.012 |
arl | +0.011 |
melon | +0.010 |
cooler | +0.010 |
ugg | +0.010 |
Neg logit contributions | |
pieces | -0.007 |
od | -0.006 |
Maria | -0.005 |
lier | -0.005 |
one | -0.005 |
Token | sn |
Token ID | 3013 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
robe | +0.035 |
arl | +0.031 |
cooler | +0.031 |
melon | +0.030 |
ugg | +0.029 |
Neg logit contributions | |
pieces | -0.030 |
od | -0.027 |
Maria | -0.026 |
lier | -0.024 |
one | -0.023 |
Token | ugg |
Token ID | 6837 |
Feature activation | -2.14e-01 |
Pos logit contributions | |
robe | +0.042 |
ranc | +0.035 |
cooler | +0.032 |
melon | +0.032 |
waitress | +0.032 |
Neg logit contributions | |
pieces | -0.097 |
lier | -0.087 |
od | -0.086 |
oser | -0.086 |
ets | -0.085 |
Token | led |
Token ID | 992 |
Feature activation | -3.92e-02 |
Pos logit contributions | |
robe | +0.428 |
arl | +0.412 |
melon | +0.398 |
ugg | +0.385 |
cooler | +0.382 |
Neg logit contributions | |
pieces | -0.265 |
od | -0.240 |
Maria | -0.220 |
lier | -0.218 |
ets | -0.211 |
Token | up |
Token ID | 510 |
Feature activation | -7.25e-02 |
Pos logit contributions | |
robe | +0.066 |
arl | +0.064 |
melon | +0.056 |
ugg | +0.056 |
Cream | +0.053 |
Neg logit contributions | |
pieces | -0.066 |
od | -0.060 |
Maria | -0.058 |
circles | -0.056 |
lier | -0.056 |
Token | in |
Token ID | 287 |
Feature activation | -1.13e-03 |
Pos logit contributions | |
robe | +0.131 |
arl | +0.123 |
melon | +0.112 |
cooler | +0.111 |
ugg | +0.111 |
Neg logit contributions | |
pieces | -0.116 |
od | -0.104 |
Maria | -0.100 |
lier | -0.091 |
circles | -0.090 |
Token | his |
Token ID | 465 |
Feature activation | -3.48e-02 |
Pos logit contributions | |
robe | +0.002 |
arl | +0.002 |
cooler | +0.002 |
melon | +0.002 |
ugg | +0.001 |
Neg logit contributions | |
pieces | -0.002 |
od | -0.002 |
Maria | -0.002 |
lier | -0.002 |
one | -0.002 |
Token | bed |
Token ID | 3996 |
Feature activation | -1.76e-02 |
Pos logit contributions | |
robe | +0.047 |
arl | +0.046 |
melon | +0.041 |
ugg | +0.038 |
Cream | +0.037 |
Neg logit contributions | |
pieces | -0.069 |
od | -0.060 |
Maria | -0.059 |
circles | -0.059 |
pad | -0.056 |
Token | lex |
Token ID | 2588 |
Feature activation | +3.96e-02 |
Pos logit contributions | |
robe | +0.006 |
arl | +0.005 |
cooler | +0.005 |
melon | +0.005 |
ranc | +0.005 |
Neg logit contributions | |
pieces | -0.005 |
od | -0.004 |
lier | -0.004 |
Maria | -0.004 |
ill | -0.004 |
Token | ible |
Token ID | 856 |
Feature activation | +6.27e-02 |
Pos logit contributions | |
pieces | +0.072 |
Maria | +0.059 |
od | +0.056 |
circles | +0.051 |
rect | +0.050 |
Neg logit contributions | |
robe | -0.072 |
ugg | -0.061 |
melon | -0.061 |
arl | -0.058 |
Cream | -0.058 |
Token | . |
Token ID | 13 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
pieces | +0.106 |
Maria | +0.098 |
od | +0.092 |
Rachel | +0.088 |
enta | +0.085 |
Neg logit contributions | |
robe | -0.110 |
arl | -0.096 |
cooler | -0.096 |
melon | -0.090 |
Cream | -0.089 |
Token |
|
Token ID | 198 |
Feature activation | -3.24e-04 |
Pos logit contributions | |
robe | +0.021 |
arl | +0.019 |
Cream | +0.018 |
ugg | +0.018 |
melon | +0.018 |
Neg logit contributions | |
pieces | -0.030 |
od | -0.027 |
Maria | -0.026 |
inda | -0.025 |
one | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | +6.25e-04 |
Pos logit contributions | |
robe | +0.000 |
arl | +0.000 |
Cream | +0.000 |
ugg | +0.000 |
Sleep | +0.000 |
Neg logit contributions | |
pieces | -0.001 |
od | -0.001 |
asing | -0.001 |
rect | -0.001 |
lier | -0.001 |
Token | But |
Token ID | 1537 |
Feature activation | +8.40e-02 |
Pos logit contributions | |
pieces | +0.001 |
od | +0.001 |
Maria | +0.001 |
ill | +0.001 |
lier | +0.001 |
Neg logit contributions | |
robe | -0.001 |
arl | -0.001 |
cooler | -0.001 |
Cream | -0.001 |
melon | -0.001 |
Token | then |
Token ID | 788 |
Feature activation | -4.99e-03 |
Pos logit contributions | |
pieces | +0.121 |
od | +0.105 |
Maria | +0.101 |
one | +0.091 |
lier | +0.091 |
Neg logit contributions | |
robe | -0.169 |
arl | -0.157 |
melon | -0.148 |
cooler | -0.147 |
ugg | -0.145 |
Token | they |
Token ID | 484 |
Feature activation | +2.77e-03 |
Pos logit contributions | |
robe | +0.008 |
arl | +0.007 |
cooler | +0.007 |
melon | +0.007 |
ugg | +0.007 |
Neg logit contributions | |
pieces | -0.009 |
od | -0.008 |
Maria | -0.008 |
lier | -0.007 |
one | -0.007 |
Token | had |
Token ID | 550 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
pieces | +0.005 |
od | +0.004 |
Maria | +0.004 |
enta | +0.004 |
asing | +0.003 |
Neg logit contributions | |
robe | -0.005 |
cooler | -0.005 |
tired | -0.004 |
melon | -0.004 |
ranc | -0.004 |
Token | an |
Token ID | 281 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
pieces | +0.025 |
od | +0.023 |
Maria | +0.022 |
lier | +0.020 |
one | +0.020 |
Neg logit contributions | |
robe | -0.021 |
arl | -0.019 |
cooler | -0.018 |
melon | -0.017 |
ugg | -0.017 |
Token | idea |
Token ID | 2126 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
pieces | +0.022 |
od | +0.020 |
Maria | +0.019 |
one | +0.018 |
lier | +0.018 |
Neg logit contributions | |
robe | -0.016 |
arl | -0.014 |
ugg | -0.013 |
melon | -0.013 |
cooler | -0.012 |
Token | reward |
Token ID | 6721 |
Feature activation | +2.82e-02 |
Pos logit contributions | |
robe | +0.006 |
arl | +0.006 |
melon | +0.005 |
cooler | +0.005 |
ugg | +0.005 |
Neg logit contributions | |
pieces | -0.011 |
od | -0.011 |
Maria | -0.010 |
lier | -0.010 |
one | -0.010 |
Token | for |
Token ID | 329 |
Feature activation | +4.98e-02 |
Pos logit contributions | |
pieces | +0.047 |
Maria | +0.042 |
od | +0.042 |
enta | +0.038 |
rect | +0.038 |
Neg logit contributions | |
robe | -0.053 |
cooler | -0.044 |
melon | -0.043 |
arl | -0.041 |
ugg | -0.040 |
Token | studying |
Token ID | 11065 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
pieces | +0.096 |
od | +0.087 |
Maria | +0.084 |
lier | +0.081 |
one | +0.081 |
Neg logit contributions | |
robe | -0.072 |
arl | -0.065 |
cooler | -0.064 |
melon | -0.061 |
Sleep | -0.058 |
Token | was |
Token ID | 373 |
Feature activation | +2.62e-02 |
Pos logit contributions | |
pieces | +0.031 |
od | +0.028 |
Maria | +0.028 |
enta | +0.026 |
one | +0.025 |
Neg logit contributions | |
robe | -0.031 |
cooler | -0.026 |
arl | -0.026 |
melon | -0.024 |
Cream | -0.024 |
Token | a |
Token ID | 257 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
pieces | +0.049 |
od | +0.044 |
Maria | +0.043 |
enta | +0.039 |
one | +0.039 |
Neg logit contributions | |
robe | -0.041 |
cooler | -0.036 |
arl | -0.035 |
melon | -0.035 |
Cream | -0.032 |
Token | massive |
Token ID | 4858 |
Feature activation | +8.46e-02 |
Pos logit contributions | |
pieces | +0.040 |
od | +0.038 |
Maria | +0.036 |
asing | +0.034 |
one | +0.034 |
Neg logit contributions | |
robe | -0.025 |
arl | -0.022 |
cooler | -0.021 |
melon | -0.020 |
ugg | -0.019 |
Token | plate |
Token ID | 7480 |
Feature activation | -4.31e-02 |
Pos logit contributions | |
pieces | +0.167 |
od | +0.152 |
Maria | +0.148 |
one | +0.138 |
enta | +0.136 |
Neg logit contributions | |
robe | -0.124 |
arl | -0.111 |
cooler | -0.106 |
melon | -0.104 |
ugg | -0.099 |
Token | of |
Token ID | 286 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
robe | +0.089 |
arl | +0.080 |
melon | +0.078 |
cooler | +0.077 |
ugg | +0.074 |
Neg logit contributions | |
pieces | -0.062 |
od | -0.053 |
Maria | -0.052 |
enta | -0.047 |
one | -0.046 |
Token | her |
Token ID | 607 |
Feature activation | -5.70e-03 |
Pos logit contributions | |
pieces | +0.041 |
od | +0.038 |
Maria | +0.037 |
one | +0.035 |
enta | +0.035 |
Neg logit contributions | |
robe | -0.024 |
cooler | -0.021 |
arl | -0.020 |
melon | -0.020 |
Cream | -0.018 |
Token | favourite |
Token ID | 12507 |
Feature activation | +2.91e-02 |
Pos logit contributions | |
robe | +0.007 |
arl | +0.006 |
cooler | +0.006 |
melon | +0.006 |
ugg | +0.005 |
Neg logit contributions | |
pieces | -0.012 |
od | -0.012 |
Maria | -0.012 |
one | -0.011 |
enta | -0.011 |
Token | pasta |
Token ID | 26296 |
Feature activation | +4.75e-04 |
Pos logit contributions | |
pieces | +0.061 |
od | +0.057 |
Maria | +0.056 |
one | +0.052 |
asing | +0.052 |
Neg logit contributions | |
robe | -0.040 |
cooler | -0.034 |
melon | -0.031 |
arl | -0.031 |
refrigerator | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
pieces | +0.077 |
od | +0.069 |
Maria | +0.065 |
lier | +0.061 |
one | +0.061 |
Neg logit contributions | |
robe | -0.103 |
arl | -0.097 |
melon | -0.090 |
cooler | -0.089 |
ugg | -0.088 |
Token | you |
Token ID | 345 |
Feature activation | +4.59e-02 |
Pos logit contributions | |
robe | +0.030 |
arl | +0.028 |
melon | +0.026 |
ugg | +0.026 |
cooler | +0.025 |
Neg logit contributions | |
pieces | -0.024 |
od | -0.021 |
Maria | -0.020 |
one | -0.019 |
circles | -0.018 |
Token | are |
Token ID | 389 |
Feature activation | +2.73e-02 |
Pos logit contributions | |
pieces | +0.056 |
od | +0.049 |
Maria | +0.044 |
lier | +0.043 |
ata | +0.042 |
Neg logit contributions | |
robe | -0.100 |
arl | -0.094 |
melon | -0.092 |
ugg | -0.088 |
cooler | -0.087 |
Token | brave |
Token ID | 14802 |
Feature activation | +2.43e-02 |
Pos logit contributions | |
pieces | +0.042 |
od | +0.040 |
Maria | +0.037 |
circles | +0.036 |
lier | +0.035 |
Neg logit contributions | |
robe | -0.052 |
arl | -0.049 |
ugg | -0.048 |
melon | -0.044 |
Cream | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | +7.39e-04 |
Pos logit contributions | |
pieces | +0.041 |
od | +0.035 |
Maria | +0.035 |
enta | +0.032 |
one | +0.030 |
Neg logit contributions | |
robe | -0.044 |
arl | -0.041 |
cooler | -0.039 |
melon | -0.036 |
cabin | -0.036 |
Token | But |
Token ID | 887 |
Feature activation | +9.97e-02 |
Pos logit contributions | |
pieces | +0.001 |
od | +0.001 |
Maria | +0.001 |
one | +0.001 |
circles | +0.001 |
Neg logit contributions | |
robe | -0.001 |
arl | -0.001 |
melon | -0.001 |
cooler | -0.001 |
ugg | -0.001 |
Token | you |
Token ID | 345 |
Feature activation | +4.86e-02 |
Pos logit contributions | |
pieces | +0.144 |
od | +0.129 |
Maria | +0.122 |
one | +0.118 |
circles | +0.112 |
Neg logit contributions | |
robe | -0.194 |
arl | -0.178 |
ugg | -0.167 |
melon | -0.165 |
cooler | -0.160 |
Token | should |
Token ID | 815 |
Feature activation | +3.07e-03 |
Pos logit contributions | |
pieces | +0.065 |
od | +0.058 |
Maria | +0.053 |
ata | +0.050 |
lier | +0.049 |
Neg logit contributions | |
robe | -0.102 |
arl | -0.095 |
melon | -0.089 |
ugg | -0.088 |
cooler | -0.087 |
Token | not |
Token ID | 407 |
Feature activation | -8.97e-02 |
Pos logit contributions | |
pieces | +0.005 |
od | +0.005 |
Maria | +0.004 |
e | +0.004 |
one | +0.004 |
Neg logit contributions | |
robe | -0.005 |
arl | -0.004 |
ugg | -0.004 |
melon | -0.004 |
Cream | -0.004 |
Token | climb |
Token ID | 12080 |
Feature activation | +7.36e-02 |
Pos logit contributions | |
robe | +0.139 |
arl | +0.132 |
melon | +0.123 |
cooler | +0.122 |
ugg | +0.118 |
Neg logit contributions | |
pieces | -0.169 |
od | -0.149 |
Maria | -0.147 |
enta | -0.134 |
inda | -0.133 |
Token | the |
Token ID | 262 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
pieces | +0.134 |
od | +0.120 |
Maria | +0.118 |
one | +0.106 |
inda | +0.106 |
Neg logit contributions | |
robe | -0.120 |
arl | -0.110 |
cooler | -0.105 |
melon | -0.102 |
ugg | -0.097 |
Token | three |
Token ID | 1115 |
Feature activation | +3.68e-02 |
Pos logit contributions | |
robe | +0.016 |
cooler | +0.015 |
arl | +0.015 |
Cream | +0.013 |
Sleep | +0.013 |
Neg logit contributions | |
pieces | -0.025 |
Maria | -0.024 |
od | -0.023 |
lier | -0.022 |
lar | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
pieces | +0.066 |
Maria | +0.063 |
Rachel | +0.058 |
od | +0.058 |
iece | +0.057 |
Neg logit contributions | |
cooler | -0.050 |
robe | -0.050 |
arl | -0.049 |
melon | -0.044 |
Cream | -0.044 |
Token | four |
Token ID | 1440 |
Feature activation | +8.36e-02 |
Pos logit contributions | |
robe | +0.016 |
arl | +0.015 |
cooler | +0.015 |
Cream | +0.013 |
melon | +0.013 |
Neg logit contributions | |
pieces | -0.024 |
Maria | -0.023 |
od | -0.022 |
lar | -0.021 |
lier | -0.021 |
Token | ," |
Token ID | 553 |
Feature activation | +7.43e-03 |
Pos logit contributions | |
Maria | +0.157 |
pieces | +0.154 |
Rachel | +0.148 |
od | +0.142 |
Jud | +0.140 |
Neg logit contributions | |
cooler | -0.110 |
robe | -0.106 |
arl | -0.103 |
melon | -0.092 |
ranc | -0.092 |
Token | she |
Token ID | 673 |
Feature activation | +3.28e-02 |
Pos logit contributions | |
pieces | +0.019 |
od | +0.017 |
Maria | +0.017 |
lier | +0.017 |
lar | +0.016 |
Neg logit contributions | |
robe | -0.006 |
arl | -0.005 |
cooler | -0.005 |
melon | -0.005 |
ugg | -0.004 |
Token | counts |
Token ID | 9853 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
pieces | +0.040 |
od | +0.034 |
Maria | +0.032 |
lier | +0.028 |
one | +0.028 |
Neg logit contributions | |
robe | -0.073 |
arl | -0.068 |
cooler | -0.065 |
melon | -0.065 |
ugg | -0.063 |
Token | . |
Token ID | 13 |
Feature activation | +5.01e-03 |
Pos logit contributions | |
pieces | +0.178 |
Maria | +0.155 |
od | +0.155 |
lier | +0.144 |
lar | +0.141 |
Neg logit contributions | |
robe | -0.170 |
arl | -0.160 |
cooler | -0.150 |
melon | -0.148 |
Cream | -0.138 |
Token |
|
Token ID | 198 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
pieces | +0.011 |
od | +0.009 |
lar | +0.009 |
Maria | +0.009 |
lier | +0.009 |
Neg logit contributions | |
robe | -0.007 |
arl | -0.006 |
cooler | -0.006 |
Cream | -0.006 |
melon | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
pieces | +0.040 |
od | +0.034 |
asing | +0.033 |
lar | +0.033 |
lier | +0.033 |
Neg logit contributions | |
arl | -0.018 |
robe | -0.018 |
Cream | -0.015 |
ugg | -0.015 |
Sleep | -0.014 |
Token | " |
Token ID | 1 |
Feature activation | -3.37e-02 |
Pos logit contributions | |
pieces | +0.030 |
od | +0.026 |
Maria | +0.024 |
ez | +0.024 |
lar | +0.023 |
Neg logit contributions | |
robe | -0.024 |
arl | -0.023 |
cooler | -0.021 |
Cream | -0.021 |
melon | -0.021 |
Token | Look |
Token ID | 8567 |
Feature activation | +4.57e-02 |
Pos logit contributions | |
robe | +0.046 |
arl | +0.043 |
Cream | +0.041 |
cooler | +0.037 |
melon | +0.036 |
Neg logit contributions | |
pieces | -0.069 |
ez | -0.061 |
od | -0.059 |
tal | -0.057 |
lar | -0.057 |
Token | ie |
Token ID | 494 |
Feature activation | +4.42e-02 |
Pos logit contributions | |
pieces | +0.060 |
mastered | +0.051 |
Maria | +0.049 |
circles | +0.049 |
rect | +0.048 |
Neg logit contributions | |
robe | -0.044 |
arl | -0.038 |
melon | -0.036 |
ugg | -0.035 |
cabin | -0.033 |
Token | , |
Token ID | 11 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
pieces | +0.057 |
od | +0.050 |
Maria | +0.047 |
lier | +0.042 |
one | +0.042 |
Neg logit contributions | |
robe | -0.095 |
arl | -0.090 |
cooler | -0.084 |
melon | -0.084 |
ugg | -0.082 |
Token | who |
Token ID | 508 |
Feature activation | -1.68e-02 |
Pos logit contributions | |
robe | +0.042 |
arl | +0.039 |
melon | +0.037 |
cooler | +0.037 |
ugg | +0.036 |
Neg logit contributions | |
pieces | -0.028 |
od | -0.024 |
Maria | -0.023 |
lier | -0.021 |
one | -0.021 |
Token | wanted |
Token ID | 2227 |
Feature activation | -3.91e-02 |
Pos logit contributions | |
arl | +0.043 |
rier | +0.037 |
melon | +0.037 |
robe | +0.036 |
cooler | +0.034 |
Neg logit contributions | |
e | -0.017 |
� | -0.016 |
's | -0.015 |
od | -0.015 |
ez | -0.015 |
Token | to |
Token ID | 284 |
Feature activation | +2.42e-02 |
Pos logit contributions | |
robe | +0.064 |
cooler | +0.056 |
arl | +0.056 |
melon | +0.055 |
ugg | +0.053 |
Neg logit contributions | |
pieces | -0.070 |
od | -0.062 |
Maria | -0.060 |
lier | -0.056 |
enta | -0.055 |
Token | reach |
Token ID | 3151 |
Feature activation | +9.31e-02 |
Pos logit contributions | |
pieces | +0.041 |
od | +0.036 |
Maria | +0.035 |
one | +0.032 |
circles | +0.032 |
Neg logit contributions | |
robe | -0.043 |
arl | -0.040 |
melon | -0.037 |
ugg | -0.036 |
cooler | -0.036 |
Token | a |
Token ID | 257 |
Feature activation | -1.94e-03 |
Pos logit contributions | |
pieces | +0.180 |
od | +0.162 |
Maria | +0.160 |
lier | +0.149 |
enta | +0.147 |
Neg logit contributions | |
robe | -0.136 |
cooler | -0.124 |
arl | -0.121 |
melon | -0.116 |
cabin | -0.109 |
Token | jar |
Token ID | 17379 |
Feature activation | -3.22e-02 |
Pos logit contributions | |
robe | +0.003 |
arl | +0.002 |
ugg | +0.002 |
melon | +0.002 |
Sleep | +0.002 |
Neg logit contributions | |
pieces | -0.004 |
od | -0.004 |
Maria | -0.003 |
lier | -0.003 |
one | -0.003 |
Token | of |
Token ID | 286 |
Feature activation | +9.34e-03 |
Pos logit contributions | |
robe | +0.068 |
arl | +0.064 |
melon | +0.060 |
ugg | +0.060 |
cooler | +0.059 |
Neg logit contributions | |
pieces | -0.042 |
od | -0.036 |
Maria | -0.033 |
one | -0.031 |
lier | -0.031 |
Token | cookies |
Token ID | 14746 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
pieces | +0.017 |
od | +0.015 |
Maria | +0.015 |
circles | +0.014 |
one | +0.014 |
Neg logit contributions | |
robe | -0.015 |
arl | -0.014 |
melon | -0.013 |
ugg | -0.013 |
cooler | -0.012 |
Token | that |
Token ID | 326 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
robe | +0.033 |
arl | +0.031 |
melon | +0.029 |
cooler | +0.029 |
ugg | +0.028 |
Neg logit contributions | |
pieces | -0.026 |
od | -0.023 |
Maria | -0.022 |
lier | -0.020 |
one | -0.020 |
Token | mom |
Token ID | 1995 |
Feature activation | -4.89e-02 |
Pos logit contributions | |
pieces | +0.014 |
od | +0.012 |
Maria | +0.012 |
circles | +0.011 |
enta | +0.011 |
Neg logit contributions | |
robe | -0.007 |
arl | -0.006 |
melon | -0.006 |
ugg | -0.005 |
cooler | -0.005 |
Token | smiled |
Token ID | 13541 |
Feature activation | -2.51e-02 |
Pos logit contributions | |
robe | +0.080 |
melon | +0.073 |
cooler | +0.072 |
ugg | +0.072 |
Cream | +0.068 |
Neg logit contributions | |
pieces | -0.083 |
od | -0.070 |
Maria | -0.068 |
one | -0.065 |
conclusions | -0.064 |
Token | and |
Token ID | 290 |
Feature activation | +3.98e-04 |
Pos logit contributions | |
robe | +0.062 |
arl | +0.059 |
melon | +0.057 |
ugg | +0.056 |
cooler | +0.054 |
Neg logit contributions | |
pieces | -0.022 |
od | -0.020 |
Maria | -0.018 |
lier | -0.017 |
ill | -0.016 |
Token | said |
Token ID | 531 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
pieces | +0.001 |
od | +0.001 |
Maria | +0.001 |
one | +0.000 |
lier | +0.000 |
Neg logit contributions | |
robe | -0.001 |
cooler | -0.001 |
arl | -0.001 |
melon | -0.001 |
ugg | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
pieces | +0.026 |
od | +0.023 |
Maria | +0.021 |
lier | +0.020 |
one | +0.019 |
Neg logit contributions | |
robe | -0.033 |
arl | -0.030 |
cooler | -0.029 |
melon | -0.028 |
ugg | -0.028 |
Token | " |
Token ID | 366 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
robe | +0.022 |
arl | +0.020 |
cooler | +0.019 |
melon | +0.018 |
ugg | +0.018 |
Neg logit contributions | |
pieces | -0.032 |
od | -0.029 |
Maria | -0.028 |
one | -0.026 |
lier | -0.026 |
Token | That |
Token ID | 2504 |
Feature activation | -7.91e-02 |
Pos logit contributions | |
robe | +0.028 |
arl | +0.024 |
ugg | +0.024 |
Sleep | +0.024 |
Cream | +0.023 |
Neg logit contributions | |
pieces | -0.046 |
od | -0.042 |
ets | -0.037 |
inations | -0.037 |
ata | -0.037 |
Token | 's |
Token ID | 338 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
robe | +0.176 |
cooler | +0.155 |
melon | +0.153 |
Sleep | +0.153 |
Cream | +0.152 |
Neg logit contributions | |
pieces | -0.100 |
Maria | -0.082 |
od | -0.075 |
ill | -0.072 |
asing | -0.072 |
Token | a |
Token ID | 257 |
Feature activation | +7.02e-03 |
Pos logit contributions | |
pieces | +0.036 |
od | +0.033 |
Maria | +0.032 |
lier | +0.029 |
inda | +0.029 |
Neg logit contributions | |
robe | -0.043 |
arl | -0.040 |
ugg | -0.039 |
melon | -0.038 |
cooler | -0.035 |
Token | beautiful |
Token ID | 4950 |
Feature activation | +3.46e-02 |
Pos logit contributions | |
pieces | +0.013 |
od | +0.012 |
Maria | +0.011 |
lier | +0.011 |
circles | +0.011 |
Neg logit contributions | |
robe | -0.011 |
arl | -0.010 |
ugg | -0.009 |
melon | -0.009 |
cooler | -0.009 |
Token | drawing |
Token ID | 8263 |
Feature activation | +3.30e-02 |
Pos logit contributions | |
pieces | +0.067 |
Maria | +0.059 |
od | +0.058 |
lier | +0.056 |
asing | +0.055 |
Neg logit contributions | |
robe | -0.053 |
cooler | -0.044 |
arl | -0.044 |
melon | -0.044 |
ugg | -0.041 |
Token | smile |
Token ID | 8212 |
Feature activation | -1.03e-03 |
Pos logit contributions | |
robe | +0.005 |
arl | +0.005 |
melon | +0.004 |
cooler | +0.004 |
ugg | +0.004 |
Neg logit contributions | |
pieces | -0.009 |
od | -0.008 |
Maria | -0.008 |
lier | -0.007 |
one | -0.007 |
Token | on |
Token ID | 319 |
Feature activation | +3.87e-02 |
Pos logit contributions | |
robe | +0.002 |
arl | +0.002 |
melon | +0.002 |
cooler | +0.002 |
ugg | +0.002 |
Neg logit contributions | |
pieces | -0.002 |
od | -0.001 |
Maria | -0.001 |
lier | -0.001 |
one | -0.001 |
Token | her |
Token ID | 607 |
Feature activation | -2.51e-02 |
Pos logit contributions | |
pieces | +0.062 |
od | +0.055 |
Maria | +0.052 |
lier | +0.048 |
one | +0.048 |
Neg logit contributions | |
robe | -0.072 |
arl | -0.066 |
melon | -0.062 |
ugg | -0.062 |
cooler | -0.062 |
Token | face |
Token ID | 1986 |
Feature activation | +8.70e-03 |
Pos logit contributions | |
robe | +0.031 |
arl | +0.026 |
cooler | +0.024 |
melon | +0.023 |
ugg | +0.022 |
Neg logit contributions | |
pieces | -0.056 |
od | -0.052 |
Maria | -0.052 |
one | -0.049 |
lier | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | +1.26e-03 |
Pos logit contributions | |
pieces | +0.014 |
Maria | +0.013 |
od | +0.012 |
enta | +0.011 |
one | +0.011 |
Neg logit contributions | |
robe | -0.016 |
arl | -0.014 |
cooler | -0.014 |
melon | -0.014 |
ugg | -0.013 |
Token | Lily |
Token ID | 20037 |
Feature activation | +3.16e-02 |
Pos logit contributions | |
pieces | +0.002 |
od | +0.002 |
Maria | +0.002 |
one | +0.002 |
lier | +0.002 |
Neg logit contributions | |
robe | -0.002 |
arl | -0.002 |
cooler | -0.002 |
melon | -0.002 |
ugg | -0.002 |
Token | loved |
Token ID | 6151 |
Feature activation | -3.91e-02 |
Pos logit contributions | |
pieces | +0.062 |
od | +0.053 |
Maria | +0.053 |
one | +0.049 |
asing | +0.048 |
Neg logit contributions | |
robe | -0.048 |
cooler | -0.040 |
melon | -0.039 |
ugg | -0.039 |
Cream | -0.038 |
Token | to |
Token ID | 284 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
robe | +0.070 |
arl | +0.064 |
melon | +0.060 |
cooler | +0.059 |
ugg | +0.059 |
Neg logit contributions | |
pieces | -0.065 |
od | -0.058 |
Maria | -0.056 |
one | -0.051 |
lier | -0.051 |
Token | comb |
Token ID | 1974 |
Feature activation | -1.90e-03 |
Pos logit contributions | |
pieces | +0.041 |
od | +0.038 |
Maria | +0.038 |
lier | +0.036 |
one | +0.036 |
Neg logit contributions | |
robe | -0.027 |
arl | -0.024 |
cooler | -0.023 |
melon | -0.022 |
ugg | -0.021 |
Token | Anna |
Token ID | 11735 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
robe | +0.003 |
arl | +0.003 |
melon | +0.003 |
cooler | +0.003 |
ugg | +0.003 |
Neg logit contributions | |
pieces | -0.003 |
od | -0.003 |
Maria | -0.003 |
lier | -0.003 |
one | -0.003 |
Token | 's |
Token ID | 338 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
pieces | +0.021 |
Maria | +0.019 |
od | +0.018 |
one | +0.016 |
lar | +0.014 |
Neg logit contributions | |
robe | -0.052 |
cooler | -0.047 |
melon | -0.046 |
arl | -0.046 |
ugg | -0.045 |
Token | He |
Token ID | 679 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
pieces | +0.046 |
od | +0.042 |
Maria | +0.041 |
enta | +0.039 |
lier | +0.039 |
Neg logit contributions | |
robe | -0.024 |
arl | -0.021 |
cooler | -0.018 |
melon | -0.018 |
ugg | -0.018 |
Token | ran |
Token ID | 4966 |
Feature activation | +2.82e-03 |
Pos logit contributions | |
robe | +0.034 |
cooler | +0.032 |
melon | +0.029 |
ranc | +0.029 |
cabin | +0.028 |
Neg logit contributions | |
pieces | -0.043 |
od | -0.036 |
Maria | -0.034 |
iece | -0.033 |
conclusions | -0.033 |
Token | to |
Token ID | 284 |
Feature activation | +3.20e-02 |
Pos logit contributions | |
pieces | +0.004 |
od | +0.004 |
Maria | +0.004 |
lier | +0.003 |
circles | +0.003 |
Neg logit contributions | |
robe | -0.005 |
arl | -0.005 |
melon | -0.005 |
ugg | -0.005 |
cooler | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | -3.74e-03 |
Pos logit contributions | |
pieces | +0.060 |
od | +0.054 |
Maria | +0.052 |
lier | +0.049 |
one | +0.049 |
Neg logit contributions | |
robe | -0.051 |
arl | -0.046 |
cooler | -0.043 |
melon | -0.043 |
ugg | -0.041 |
Token | kitchen |
Token ID | 9592 |
Feature activation | -3.56e-02 |
Pos logit contributions | |
robe | +0.005 |
arl | +0.005 |
ugg | +0.004 |
melon | +0.004 |
cooler | +0.004 |
Neg logit contributions | |
pieces | -0.008 |
od | -0.007 |
Maria | -0.007 |
circles | -0.006 |
one | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | +6.84e-03 |
Pos logit contributions | |
robe | +0.078 |
arl | +0.074 |
ugg | +0.068 |
melon | +0.068 |
cooler | +0.068 |
Neg logit contributions | |
pieces | -0.043 |
od | -0.038 |
Maria | -0.036 |
lier | -0.031 |
ill | -0.031 |
Token | grabbed |
Token ID | 13176 |
Feature activation | +2.65e-02 |
Pos logit contributions | |
pieces | +0.011 |
od | +0.010 |
Maria | +0.009 |
lier | +0.009 |
one | +0.008 |
Neg logit contributions | |
robe | -0.013 |
cooler | -0.011 |
melon | -0.011 |
arl | -0.011 |
ugg | -0.011 |
Token | some |
Token ID | 617 |
Feature activation | -2.95e-02 |
Pos logit contributions | |
pieces | +0.049 |
od | +0.045 |
Maria | +0.043 |
lier | +0.040 |
one | +0.040 |
Neg logit contributions | |
robe | -0.042 |
arl | -0.038 |
cooler | -0.035 |
melon | -0.035 |
ugg | -0.034 |
Token | lot |
Token ID | 1256 |
Feature activation | +6.98e-02 |
Pos logit contributions | |
robe | +0.040 |
cooler | +0.038 |
arl | +0.036 |
melon | +0.034 |
refrigerator | +0.032 |
Neg logit contributions | |
pieces | -0.060 |
Maria | -0.057 |
od | -0.057 |
enta | -0.052 |
Rachel | -0.052 |
Token | ion |
Token ID | 295 |
Feature activation | -9.17e-02 |
Pos logit contributions | |
pieces | +0.121 |
od | +0.105 |
Maria | +0.104 |
enta | +0.093 |
lar | +0.093 |
Neg logit contributions | |
robe | -0.119 |
arl | -0.111 |
melon | -0.107 |
cooler | -0.105 |
ugg | -0.101 |
Token | . |
Token ID | 13 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
robe | +0.181 |
arl | +0.170 |
ugg | +0.155 |
cooler | +0.154 |
melon | +0.152 |
Neg logit contributions | |
pieces | -0.131 |
od | -0.115 |
Maria | -0.105 |
lier | -0.103 |
one | -0.101 |
Token | hear |
Token ID | 3285 |
Feature activation | +7.10e-02 |
Pos logit contributions | |
robe | +0.080 |
arl | +0.075 |
melon | +0.070 |
ugg | +0.068 |
Cream | +0.067 |
Neg logit contributions | |
pieces | -0.056 |
od | -0.048 |
Maria | -0.044 |
lier | -0.043 |
ata | -0.043 |
Token | the |
Token ID | 262 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
pieces | +0.140 |
od | +0.127 |
Maria | +0.123 |
one | +0.115 |
lier | +0.115 |
Neg logit contributions | |
robe | -0.105 |
arl | -0.095 |
cooler | -0.087 |
melon | -0.087 |
ugg | -0.085 |
Token | ball |
Token ID | 2613 |
Feature activation | -8.05e-03 |
Pos logit contributions | |
robe | +0.027 |
arl | +0.025 |
ugg | +0.024 |
melon | +0.024 |
Cream | +0.022 |
Neg logit contributions | |
pieces | -0.033 |
circles | -0.029 |
od | -0.028 |
Maria | -0.028 |
lier | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | -2.91e-03 |
Pos logit contributions | |
robe | +0.013 |
arl | +0.012 |
cooler | +0.012 |
melon | +0.012 |
ugg | +0.011 |
Neg logit contributions | |
pieces | -0.015 |
od | -0.013 |
Maria | -0.013 |
enta | -0.012 |
one | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | -2.05e-02 |
Pos logit contributions | |
robe | +0.005 |
arl | +0.004 |
cooler | +0.004 |
melon | +0.004 |
ugg | +0.004 |
Neg logit contributions | |
pieces | -0.005 |
od | -0.005 |
Maria | -0.005 |
one | -0.004 |
lier | -0.004 |
Token | sticks |
Token ID | 16461 |
Feature activation | +4.60e-02 |
Pos logit contributions | |
robe | +0.030 |
arl | +0.028 |
ugg | +0.026 |
melon | +0.026 |
cooler | +0.025 |
Neg logit contributions | |
pieces | -0.039 |
od | -0.034 |
Maria | -0.033 |
circles | -0.032 |
lier | -0.031 |
Token | outside |
Token ID | 2354 |
Feature activation | +2.84e-03 |
Pos logit contributions | |
pieces | +0.076 |
od | +0.067 |
Maria | +0.067 |
enta | +0.060 |
one | +0.059 |
Neg logit contributions | |
robe | -0.083 |
arl | -0.077 |
cooler | -0.074 |
melon | -0.073 |
ugg | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | +1.78e-03 |
Pos logit contributions | |
pieces | +0.005 |
od | +0.005 |
Maria | +0.004 |
enta | +0.004 |
one | +0.004 |
Neg logit contributions | |
robe | -0.005 |
arl | -0.004 |
cooler | -0.004 |
melon | -0.004 |
cabin | -0.004 |
Token | They |
Token ID | 1119 |
Feature activation | -3.74e-02 |
Pos logit contributions | |
pieces | +0.004 |
od | +0.003 |
Maria | +0.003 |
lier | +0.003 |
one | +0.003 |
Neg logit contributions | |
robe | -0.003 |
arl | -0.002 |
melon | -0.002 |
cooler | -0.002 |
ugg | -0.002 |
Token | wonder |
Token ID | 4240 |
Feature activation | +2.56e-02 |
Pos logit contributions | |
robe | +0.073 |
arl | +0.066 |
melon | +0.062 |
ugg | +0.062 |
Cream | +0.060 |
Neg logit contributions | |
pieces | -0.056 |
od | -0.049 |
Maria | -0.047 |
circles | -0.044 |
ill | -0.044 |
Token | who |
Token ID | 508 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
pieces | +0.057 |
Maria | +0.052 |
od | +0.051 |
enta | +0.050 |
ez | +0.049 |
Neg logit contributions | |
robe | -0.029 |
arl | -0.028 |
cooler | -0.027 |
melon | -0.024 |
cabin | -0.024 |
Token | a |
Token ID | 257 |
Feature activation | +8.69e-04 |
Pos logit contributions | |
pieces | +0.083 |
od | +0.074 |
Maria | +0.072 |
one | +0.066 |
lier | +0.066 |
Neg logit contributions | |
robe | -0.076 |
arl | -0.069 |
melon | -0.064 |
cooler | -0.064 |
ugg | -0.063 |
Token | monkey |
Token ID | 21657 |
Feature activation | +4.03e-02 |
Pos logit contributions | |
pieces | +0.002 |
od | +0.002 |
Maria | +0.002 |
asing | +0.002 |
one | +0.002 |
Neg logit contributions | |
robe | -0.001 |
cooler | -0.001 |
arl | -0.001 |
melon | -0.001 |
cabin | -0.001 |
Token | play |
Token ID | 711 |
Feature activation | -4.82e-02 |
Pos logit contributions | |
pieces | +0.066 |
Maria | +0.061 |
od | +0.058 |
Rachel | +0.053 |
inda | +0.053 |
Neg logit contributions | |
robe | -0.070 |
cooler | -0.064 |
arl | -0.062 |
melon | -0.062 |
cabin | -0.058 |
Token | with |
Token ID | 351 |
Feature activation | -4.39e-02 |
Pos logit contributions | |
robe | +0.077 |
arl | +0.068 |
cooler | +0.065 |
melon | +0.065 |
ugg | +0.062 |
Neg logit contributions | |
pieces | -0.089 |
od | -0.080 |
Maria | -0.079 |
one | -0.073 |
enta | -0.072 |
Token | a |
Token ID | 257 |
Feature activation | +7.72e-05 |
Pos logit contributions | |
robe | +0.062 |
cooler | +0.055 |
arl | +0.052 |
melon | +0.052 |
Cream | +0.047 |
Neg logit contributions | |
pieces | -0.087 |
od | -0.077 |
Maria | -0.077 |
inda | -0.073 |
asing | -0.073 |
Token | ball |
Token ID | 2613 |
Feature activation | -7.46e-03 |
Pos logit contributions | |
pieces | +0.000 |
Maria | +0.000 |
od | +0.000 |
asing | +0.000 |
ill | +0.000 |
Neg logit contributions | |
robe | -0.000 |
cooler | -0.000 |
arl | -0.000 |
cabin | -0.000 |
melon | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | +5.76e-03 |
Pos logit contributions | |
robe | +0.014 |
arl | +0.012 |
melon | +0.012 |
cooler | +0.012 |
ugg | +0.011 |
Neg logit contributions | |
pieces | -0.012 |
Maria | -0.011 |
od | -0.011 |
enta | -0.010 |
Rachel | -0.009 |
Token | She |
Token ID | 1375 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
pieces | +0.012 |
od | +0.011 |
Maria | +0.011 |
one | +0.010 |
enta | +0.010 |
Neg logit contributions | |
robe | -0.008 |
arl | -0.007 |
cooler | -0.006 |
melon | -0.006 |
ugg | -0.006 |
Token | cl |
Token ID | 537 |
Feature activation | +4.01e-02 |
Pos logit contributions | |
robe | +0.030 |
cooler | +0.027 |
melon | +0.026 |
bar | +0.026 |
Cream | +0.025 |
Neg logit contributions | |
pieces | -0.038 |
iece | -0.031 |
od | -0.031 |
Maria | -0.030 |
asing | -0.029 |
Token | apped |
Token ID | 6320 |
Feature activation | -6.92e-02 |
Pos logit contributions | |
pieces | +0.079 |
od | +0.071 |
Maria | +0.069 |
one | +0.064 |
lier | +0.064 |
Neg logit contributions | |
robe | -0.060 |
arl | -0.055 |
melon | -0.050 |
cooler | -0.050 |
ugg | -0.049 |
Token | and |
Token ID | 290 |
Feature activation | -6.38e-04 |
Pos logit contributions | |
robe | +0.121 |
arl | +0.111 |
cooler | +0.104 |
melon | +0.104 |
ugg | +0.101 |
Neg logit contributions | |
pieces | -0.118 |
od | -0.105 |
Maria | -0.101 |
one | -0.093 |
lier | -0.093 |
Token | . |
Token ID | 13 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
pieces | +0.041 |
od | +0.035 |
Maria | +0.035 |
enta | +0.032 |
one | +0.031 |
Neg logit contributions | |
robe | -0.040 |
arl | -0.035 |
cooler | -0.034 |
melon | -0.032 |
ugg | -0.032 |
Token | They |
Token ID | 1119 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
pieces | +0.024 |
od | +0.022 |
Maria | +0.021 |
one | +0.020 |
lier | +0.020 |
Neg logit contributions | |
robe | -0.017 |
arl | -0.015 |
cooler | -0.014 |
melon | -0.014 |
ugg | -0.014 |
Token | do |
Token ID | 466 |
Feature activation | +2.31e-02 |
Pos logit contributions | |
robe | +0.047 |
arl | +0.043 |
melon | +0.040 |
ugg | +0.040 |
Cream | +0.039 |
Neg logit contributions | |
pieces | -0.036 |
od | -0.032 |
Maria | -0.030 |
lier | -0.028 |
ata | -0.028 |
Token | not |
Token ID | 407 |
Feature activation | -8.80e-02 |
Pos logit contributions | |
pieces | +0.054 |
od | +0.051 |
Maria | +0.048 |
ata | +0.047 |
one | +0.046 |
Neg logit contributions | |
robe | -0.026 |
arl | -0.023 |
melon | -0.021 |
ugg | -0.020 |
cooler | -0.020 |
Token | care |
Token ID | 1337 |
Feature activation | -3.52e-02 |
Pos logit contributions | |
robe | +0.160 |
arl | +0.148 |
melon | +0.138 |
cooler | +0.138 |
ugg | +0.136 |
Neg logit contributions | |
pieces | -0.144 |
od | -0.127 |
Maria | -0.122 |
lier | -0.112 |
one | -0.112 |
Token | about |
Token ID | 546 |
Feature activation | -5.26e-02 |
Pos logit contributions | |
robe | +0.054 |
arl | +0.050 |
cooler | +0.042 |
ugg | +0.042 |
melon | +0.041 |
Neg logit contributions | |
pieces | -0.069 |
Maria | -0.060 |
od | -0.059 |
enta | -0.057 |
lier | -0.055 |
Token | Dad |
Token ID | 17415 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
robe | +0.082 |
arl | +0.074 |
cooler | +0.070 |
melon | +0.070 |
ugg | +0.066 |
Neg logit contributions | |
pieces | -0.099 |
od | -0.088 |
Maria | -0.086 |
one | -0.081 |
lier | -0.080 |
Token | . |
Token ID | 13 |
Feature activation | +1.22e-02 |
Pos logit contributions | |
robe | +0.030 |
cooler | +0.028 |
melon | +0.027 |
arl | +0.027 |
Cream | +0.026 |
Neg logit contributions | |
pieces | -0.018 |
Maria | -0.016 |
od | -0.015 |
inda | -0.013 |
enta | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | +2.47e-02 |
Pos logit contributions | |
pieces | +0.023 |
od | +0.021 |
Maria | +0.021 |
lier | +0.019 |
one | +0.019 |
Neg logit contributions | |
robe | -0.019 |
arl | -0.017 |
melon | -0.016 |
cooler | -0.016 |
ugg | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
pieces | +0.041 |
od | +0.036 |
Maria | +0.035 |
one | +0.032 |
lier | +0.032 |
Neg logit contributions | |
robe | -0.044 |
arl | -0.041 |
melon | -0.038 |
cooler | -0.038 |
ugg | -0.037 |
Token | Mom |
Token ID | 29252 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
pieces | +0.038 |
od | +0.034 |
Maria | +0.032 |
lier | +0.030 |
one | +0.029 |
Neg logit contributions | |
robe | -0.045 |
arl | -0.042 |
cooler | -0.039 |
ugg | -0.039 |
melon | -0.038 |
Token | girl |
Token ID | 2576 |
Feature activation | +4.82e-02 |
Pos logit contributions | |
robe | +0.031 |
arl | +0.027 |
melon | +0.025 |
ugg | +0.025 |
Sleep | +0.023 |
Neg logit contributions | |
pieces | -0.031 |
od | -0.028 |
lier | -0.026 |
Maria | -0.025 |
one | -0.024 |
Token | who |
Token ID | 508 |
Feature activation | -3.07e-02 |
Pos logit contributions | |
pieces | +0.085 |
Maria | +0.075 |
od | +0.074 |
one | +0.067 |
ata | +0.066 |
Neg logit contributions | |
robe | -0.079 |
cooler | -0.069 |
ugg | -0.068 |
arl | -0.068 |
melon | -0.068 |
Token | liked |
Token ID | 8288 |
Feature activation | -4.14e-02 |
Pos logit contributions | |
arl | +0.059 |
robe | +0.057 |
melon | +0.053 |
cooler | +0.050 |
ugg | +0.047 |
Neg logit contributions | |
pieces | -0.042 |
od | -0.041 |
Maria | -0.038 |
e | -0.037 |
lier | -0.036 |
Token | to |
Token ID | 284 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
robe | +0.087 |
arl | +0.083 |
ugg | +0.075 |
melon | +0.075 |
cooler | +0.074 |
Neg logit contributions | |
pieces | -0.054 |
od | -0.047 |
Maria | -0.045 |
circles | -0.042 |
lier | -0.039 |
Token | play |
Token ID | 711 |
Feature activation | -5.43e-02 |
Pos logit contributions | |
pieces | +0.024 |
od | +0.021 |
Maria | +0.021 |
lier | +0.020 |
one | +0.020 |
Neg logit contributions | |
robe | -0.018 |
arl | -0.015 |
cooler | -0.015 |
melon | -0.014 |
ugg | -0.014 |
Token | with |
Token ID | 351 |
Feature activation | -4.83e-02 |
Pos logit contributions | |
robe | +0.085 |
arl | +0.077 |
melon | +0.071 |
cooler | +0.071 |
ugg | +0.069 |
Neg logit contributions | |
pieces | -0.103 |
od | -0.093 |
Maria | -0.090 |
one | -0.084 |
lier | -0.084 |
Token | her |
Token ID | 607 |
Feature activation | -3.38e-02 |
Pos logit contributions | |
robe | +0.076 |
arl | +0.068 |
cooler | +0.064 |
melon | +0.062 |
ugg | +0.060 |
Neg logit contributions | |
pieces | -0.092 |
od | -0.081 |
Maria | -0.079 |
one | -0.075 |
lier | -0.075 |
Token | toys |
Token ID | 14958 |
Feature activation | -1.00e-02 |
Pos logit contributions | |
robe | +0.048 |
arl | +0.044 |
melon | +0.040 |
ugg | +0.040 |
cooler | +0.039 |
Neg logit contributions | |
pieces | -0.068 |
od | -0.061 |
Maria | -0.059 |
lier | -0.055 |
one | -0.055 |
Token | . |
Token ID | 13 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
robe | +0.019 |
arl | +0.017 |
cooler | +0.016 |
melon | +0.016 |
ugg | +0.016 |
Neg logit contributions | |
pieces | -0.016 |
od | -0.014 |
Maria | -0.014 |
one | -0.012 |
lier | -0.012 |
Token | She |
Token ID | 1375 |
Feature activation | -3.20e-02 |
Pos logit contributions | |
robe | +0.027 |
arl | +0.026 |
melon | +0.024 |
ugg | +0.023 |
cooler | +0.023 |
Neg logit contributions | |
pieces | -0.026 |
od | -0.023 |
Maria | -0.023 |
lier | -0.022 |
one | -0.021 |
Token | had |
Token ID | 550 |
Feature activation | +6.53e-03 |
Pos logit contributions | |
robe | +0.058 |
cooler | +0.050 |
ugg | +0.048 |
melon | +0.048 |
arl | +0.047 |
Neg logit contributions | |
pieces | -0.056 |
od | -0.047 |
Maria | -0.046 |
lier | -0.043 |
one | -0.042 |
Token | " |
Token ID | 366 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
robe | +0.019 |
arl | +0.016 |
cooler | +0.016 |
ugg | +0.016 |
melon | +0.015 |
Neg logit contributions | |
pieces | -0.020 |
od | -0.019 |
Maria | -0.018 |
enta | -0.017 |
circles | -0.017 |
Token | I |
Token ID | 40 |
Feature activation | +9.91e-03 |
Pos logit contributions | |
robe | +0.022 |
arl | +0.020 |
melon | +0.019 |
cooler | +0.019 |
ugg | +0.018 |
Neg logit contributions | |
pieces | -0.022 |
od | -0.020 |
Maria | -0.019 |
lier | -0.018 |
one | -0.018 |
Token | 'm |
Token ID | 1101 |
Feature activation | +6.52e-02 |
Pos logit contributions | |
pieces | +0.018 |
od | +0.016 |
Maria | +0.016 |
enta | +0.014 |
one | +0.014 |
Neg logit contributions | |
robe | -0.017 |
arl | -0.015 |
ugg | -0.014 |
cooler | -0.014 |
melon | -0.014 |
Token | sorry |
Token ID | 7926 |
Feature activation | -7.68e-02 |
Pos logit contributions | |
pieces | +0.151 |
od | +0.144 |
Maria | +0.140 |
circles | +0.131 |
enta | +0.131 |
Neg logit contributions | |
robe | -0.075 |
ugg | -0.054 |
arl | -0.052 |
melon | -0.051 |
ranc | -0.047 |
Token | , |
Token ID | 11 |
Feature activation | -8.55e-03 |
Pos logit contributions | |
robe | +0.156 |
arl | +0.141 |
melon | +0.136 |
ugg | +0.134 |
cooler | +0.133 |
Neg logit contributions | |
pieces | -0.110 |
od | -0.098 |
Maria | -0.092 |
lier | -0.085 |
one | -0.085 |
Token | too |
Token ID | 1165 |
Feature activation | -8.31e-02 |
Pos logit contributions | |
robe | +0.014 |
arl | +0.013 |
cooler | +0.012 |
melon | +0.012 |
ugg | +0.012 |
Neg logit contributions | |
pieces | -0.015 |
od | -0.014 |
Maria | -0.013 |
lier | -0.012 |
one | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +1.12e-03 |
Pos logit contributions | |
robe | +0.181 |
arl | +0.163 |
melon | +0.158 |
ugg | +0.158 |
cooler | +0.155 |
Neg logit contributions | |
pieces | -0.104 |
od | -0.095 |
Maria | -0.088 |
lier | -0.080 |
one | -0.078 |
Token | I |
Token ID | 314 |
Feature activation | +3.41e-02 |
Pos logit contributions | |
pieces | +0.002 |
Maria | +0.001 |
od | +0.001 |
asing | +0.001 |
circles | +0.001 |
Neg logit contributions | |
robe | -0.002 |
cooler | -0.002 |
arl | -0.002 |
melon | -0.002 |
ranc | -0.002 |
Token | was |
Token ID | 373 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
pieces | +0.058 |
od | +0.050 |
Maria | +0.049 |
one | +0.045 |
enta | +0.044 |
Neg logit contributions | |
robe | -0.061 |
arl | -0.056 |
cooler | -0.053 |
ugg | -0.053 |
melon | -0.052 |
Token | en |
Token ID | 551 |
Feature activation | -5.46e-02 |
Pos logit contributions | |
pieces | +0.050 |
od | +0.046 |
Maria | +0.045 |
one | +0.042 |
lier | +0.042 |
Neg logit contributions | |
robe | -0.021 |
arl | -0.018 |
cooler | -0.017 |
melon | -0.017 |
ugg | -0.015 |
Token | vious |
Token ID | 1442 |
Feature activation | +6.31e-03 |
Pos logit contributions | |
arl | +0.113 |
robe | +0.109 |
cooler | +0.100 |
melon | +0.099 |
ugg | +0.096 |
Neg logit contributions | |
pieces | -0.075 |
Maria | -0.069 |
od | -0.065 |
enta | -0.060 |
Jo | -0.057 |
Token | ?" |
Token ID | 1701 |
Feature activation | +5.07e-03 |
Pos logit contributions | |
pieces | +0.114 |
od | +0.102 |
Maria | +0.100 |
lier | +0.094 |
one | +0.093 |
Neg logit contributions | |
robe | -0.070 |
arl | -0.062 |
melon | -0.058 |
cooler | -0.057 |
ugg | -0.055 |
Token |
|
Token ID | 198 |
Feature activation | +2.66e-02 |
Pos logit contributions | |
pieces | +0.011 |
od | +0.010 |
one | +0.009 |
Maria | +0.009 |
asing | +0.009 |
Neg logit contributions | |
robe | -0.007 |
arl | -0.006 |
ugg | -0.006 |
Cream | -0.005 |
cooler | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +2.65e-02 |
Pos logit contributions | |
pieces | +0.059 |
od | +0.052 |
asing | +0.051 |
rect | +0.050 |
one | +0.050 |
Neg logit contributions | |
robe | -0.033 |
arl | -0.030 |
ugg | -0.028 |
Cream | -0.027 |
cooler | -0.026 |
Token | L |
Token ID | 43 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
pieces | +0.048 |
od | +0.043 |
Maria | +0.039 |
one | +0.039 |
asing | +0.038 |
Neg logit contributions | |
robe | -0.043 |
arl | -0.040 |
ugg | -0.037 |
cooler | -0.036 |
Cream | -0.036 |
Token | ily |
Token ID | 813 |
Feature activation | +5.92e-02 |
Pos logit contributions | |
pieces | +0.044 |
od | +0.037 |
Maria | +0.037 |
rect | +0.035 |
enta | +0.035 |
Neg logit contributions | |
robe | -0.048 |
arl | -0.044 |
ugg | -0.039 |
cooler | -0.039 |
melon | -0.038 |
Token | thought |
Token ID | 1807 |
Feature activation | -4.45e-02 |
Pos logit contributions | |
pieces | +0.098 |
od | +0.085 |
Maria | +0.082 |
one | +0.076 |
lier | +0.075 |
Neg logit contributions | |
robe | -0.107 |
arl | -0.097 |
melon | -0.092 |
cooler | -0.092 |
ugg | -0.091 |
Token | for |
Token ID | 329 |
Feature activation | +4.91e-02 |
Pos logit contributions | |
robe | +0.067 |
arl | +0.061 |
cooler | +0.057 |
melon | +0.055 |
ugg | +0.054 |
Neg logit contributions | |
pieces | -0.084 |
Maria | -0.076 |
od | -0.076 |
lier | -0.071 |
enta | -0.070 |
Token | a |
Token ID | 257 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
pieces | +0.074 |
od | +0.067 |
Maria | +0.067 |
circles | +0.061 |
one | +0.060 |
Neg logit contributions | |
robe | -0.092 |
arl | -0.091 |
melon | -0.086 |
ugg | -0.082 |
cooler | -0.077 |
Token | minute |
Token ID | 5664 |
Feature activation | +8.93e-02 |
Pos logit contributions | |
pieces | +0.026 |
od | +0.024 |
Maria | +0.023 |
asing | +0.022 |
enta | +0.022 |
Neg logit contributions | |
robe | -0.019 |
arl | -0.017 |
cooler | -0.016 |
ranc | -0.015 |
melon | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | -8.19e-03 |
Pos logit contributions | |
pieces | +0.168 |
Maria | +0.151 |
enta | +0.143 |
od | +0.142 |
lar | +0.135 |
Neg logit contributions | |
robe | -0.136 |
arl | -0.117 |
cooler | -0.115 |
melon | -0.113 |
ugg | -0.108 |
Token | and |
Token ID | 290 |
Feature activation | +9.87e-03 |
Pos logit contributions | |
robe | +0.012 |
cooler | +0.011 |
arl | +0.010 |
melon | +0.010 |
Sleep | +0.010 |
Neg logit contributions | |
pieces | -0.016 |
od | -0.014 |
Maria | -0.014 |
enta | -0.013 |
ata | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | -1.49e-02 |
Pos logit contributions | |
robe | +0.037 |
arl | +0.033 |
melon | +0.032 |
cooler | +0.032 |
ugg | +0.031 |
Neg logit contributions | |
pieces | -0.035 |
Maria | -0.031 |
od | -0.031 |
enta | -0.027 |
inda | -0.027 |
Token | Emily |
Token ID | 17608 |
Feature activation | +4.36e-02 |
Pos logit contributions | |
robe | +0.021 |
arl | +0.019 |
cooler | +0.018 |
melon | +0.017 |
ugg | +0.017 |
Neg logit contributions | |
pieces | -0.030 |
od | -0.028 |
Maria | -0.027 |
lier | -0.025 |
one | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | +4.65e-03 |
Pos logit contributions | |
pieces | +0.060 |
od | +0.052 |
Maria | +0.049 |
one | +0.045 |
lier | +0.045 |
Neg logit contributions | |
robe | -0.090 |
arl | -0.085 |
melon | -0.079 |
cooler | -0.079 |
ugg | -0.078 |
Token | Emily |
Token ID | 17608 |
Feature activation | +4.41e-02 |
Pos logit contributions | |
pieces | +0.010 |
od | +0.009 |
Maria | +0.008 |
one | +0.008 |
lier | +0.008 |
Neg logit contributions | |
robe | -0.006 |
arl | -0.006 |
cooler | -0.005 |
melon | -0.005 |
Sleep | -0.005 |
Token | always |
Token ID | 1464 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
pieces | +0.069 |
od | +0.061 |
Maria | +0.059 |
lier | +0.053 |
one | +0.053 |
Neg logit contributions | |
robe | -0.083 |
arl | -0.076 |
melon | -0.072 |
cooler | -0.071 |
ugg | -0.070 |
Token | put |
Token ID | 1234 |
Feature activation | -3.93e-02 |
Pos logit contributions | |
robe | +0.033 |
arl | +0.032 |
melon | +0.030 |
cooler | +0.029 |
ugg | +0.028 |
Neg logit contributions | |
pieces | -0.023 |
od | -0.021 |
Maria | -0.020 |
one | -0.019 |
enta | -0.018 |
Token | a |
Token ID | 257 |
Feature activation | +4.92e-03 |
Pos logit contributions | |
robe | +0.071 |
arl | +0.067 |
melon | +0.064 |
ugg | +0.063 |
cooler | +0.060 |
Neg logit contributions | |
pieces | -0.061 |
od | -0.053 |
Maria | -0.051 |
circles | -0.048 |
lier | -0.048 |
Token | leash |
Token ID | 40529 |
Feature activation | -3.37e-02 |
Pos logit contributions | |
pieces | +0.010 |
od | +0.008 |
pad | +0.008 |
Maria | +0.008 |
circles | +0.008 |
Neg logit contributions | |
robe | -0.007 |
arl | -0.006 |
melon | -0.006 |
Sleep | -0.005 |
ugg | -0.005 |
Token | on |
Token ID | 319 |
Feature activation | +4.70e-02 |
Pos logit contributions | |
arl | +0.065 |
robe | +0.064 |
melon | +0.059 |
cooler | +0.059 |
waitress | +0.057 |
Neg logit contributions | |
pieces | -0.043 |
od | -0.037 |
one | -0.035 |
pad | -0.035 |
ata | -0.034 |
Token | Max |
Token ID | 5436 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
pieces | +0.089 |
od | +0.080 |
Maria | +0.078 |
one | +0.073 |
enta | +0.072 |
Neg logit contributions | |
robe | -0.074 |
arl | -0.066 |
cooler | -0.061 |
melon | -0.061 |
ugg | -0.058 |
Token | before |
Token ID | 878 |
Feature activation | -1.61e-03 |
Pos logit contributions | |
robe | +0.045 |
arl | +0.042 |
cooler | +0.041 |
melon | +0.040 |
ugg | +0.040 |
Neg logit contributions | |
pieces | -0.014 |
od | -0.011 |
Maria | -0.010 |
one | -0.008 |
lier | -0.008 |
Token | She |
Token ID | 1375 |
Feature activation | +3.05e-03 |
Pos logit contributions | |
pieces | +0.050 |
od | +0.045 |
Maria | +0.043 |
lier | +0.041 |
enta | +0.041 |
Neg logit contributions | |
robe | -0.034 |
arl | -0.030 |
melon | -0.028 |
cooler | -0.028 |
ugg | -0.027 |
Token | was |
Token ID | 373 |
Feature activation | +2.62e-02 |
Pos logit contributions | |
pieces | +0.005 |
Maria | +0.004 |
od | +0.004 |
lier | +0.004 |
ill | +0.004 |
Neg logit contributions | |
robe | -0.005 |
cooler | -0.005 |
ugg | -0.005 |
melon | -0.005 |
cabin | -0.004 |
Token | surprised |
Token ID | 6655 |
Feature activation | -9.21e-03 |
Pos logit contributions | |
pieces | +0.048 |
od | +0.043 |
Maria | +0.042 |
lier | +0.039 |
one | +0.039 |
Neg logit contributions | |
robe | -0.042 |
arl | -0.039 |
melon | -0.036 |
cooler | -0.035 |
ugg | -0.035 |
Token | to |
Token ID | 284 |
Feature activation | +3.78e-02 |
Pos logit contributions | |
robe | +0.015 |
arl | +0.014 |
melon | +0.013 |
cooler | +0.013 |
ugg | +0.013 |
Neg logit contributions | |
pieces | -0.017 |
od | -0.014 |
Maria | -0.014 |
enta | -0.013 |
ata | -0.013 |
Token | see |
Token ID | 766 |
Feature activation | +9.89e-02 |
Pos logit contributions | |
pieces | +0.075 |
lier | +0.072 |
rect | +0.070 |
ata | +0.069 |
od | +0.069 |
Neg logit contributions | |
not | -0.044 |
robe | -0.040 |
name | -0.039 |
sorry | -0.039 |
mom | -0.035 |
Token | how |
Token ID | 703 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
pieces | +0.196 |
od | +0.178 |
Maria | +0.171 |
enta | +0.166 |
ata | +0.165 |
Neg logit contributions | |
robe | -0.135 |
arl | -0.124 |
melon | -0.120 |
cooler | -0.119 |
Cream | -0.109 |
Token | much |
Token ID | 881 |
Feature activation | +6.67e-02 |
Pos logit contributions | |
robe | +0.248 |
arl | +0.219 |
melon | +0.200 |
cooler | +0.196 |
ugg | +0.194 |
Neg logit contributions | |
pieces | -0.340 |
od | -0.308 |
Maria | -0.303 |
one | -0.282 |
lier | -0.281 |
Token | fun |
Token ID | 1257 |
Feature activation | +3.05e-02 |
Pos logit contributions | |
pieces | +0.132 |
od | +0.121 |
Maria | +0.115 |
enta | +0.109 |
ata | +0.108 |
Neg logit contributions | |
robe | -0.095 |
cooler | -0.088 |
arl | -0.085 |
melon | -0.083 |
Sleep | -0.076 |
Token | she |
Token ID | 673 |
Feature activation | +5.01e-02 |
Pos logit contributions | |
pieces | +0.062 |
od | +0.056 |
Maria | +0.055 |
ata | +0.052 |
enta | +0.052 |
Neg logit contributions | |
robe | -0.041 |
cooler | -0.038 |
melon | -0.035 |
arl | -0.033 |
cabin | -0.033 |
Token | had |
Token ID | 550 |
Feature activation | +4.50e-02 |
Pos logit contributions | |
pieces | +0.094 |
Maria | +0.082 |
od | +0.081 |
ill | +0.074 |
weet | +0.074 |
Neg logit contributions | |
robe | -0.081 |
cooler | -0.070 |
ugg | -0.064 |
Cream | -0.063 |
cabin | -0.062 |
Token | , |
Token ID | 11 |
Feature activation | +4.93e-03 |
Pos logit contributions | |
pieces | +0.085 |
Maria | +0.072 |
od | +0.070 |
inda | +0.065 |
rect | +0.064 |
Neg logit contributions | |
robe | -0.071 |
cooler | -0.063 |
Sleep | -0.059 |
not | -0.059 |
well | -0.056 |
Token | |
Token ID | 220 |
Feature activation | -3.29e-02 |
Pos logit contributions | |
pieces | +0.036 |
od | +0.033 |
Maria | +0.031 |
enta | +0.030 |
asing | +0.029 |
Neg logit contributions | |
robe | -0.027 |
arl | -0.024 |
cooler | -0.022 |
melon | -0.022 |
ugg | -0.022 |
Token |
|
Token ID | 198 |
Feature activation | +2.53e-02 |
Pos logit contributions | |
robe | +0.054 |
arl | +0.049 |
melon | +0.045 |
cooler | +0.045 |
ugg | +0.044 |
Neg logit contributions | |
pieces | -0.060 |
od | -0.053 |
Maria | -0.051 |
lier | -0.048 |
enta | -0.048 |
Token |
|
Token ID | 198 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
pieces | +0.049 |
od | +0.044 |
rect | +0.040 |
asing | +0.040 |
enta | +0.040 |
Neg logit contributions | |
robe | -0.037 |
arl | -0.035 |
ugg | -0.032 |
cooler | -0.031 |
Cream | -0.031 |
Token | Suddenly |
Token ID | 38582 |
Feature activation | +6.77e-02 |
Pos logit contributions | |
pieces | +0.040 |
od | +0.036 |
enta | +0.031 |
Maria | +0.031 |
ets | +0.031 |
Neg logit contributions | |
robe | -0.047 |
arl | -0.045 |
ugg | -0.042 |
cooler | -0.041 |
Cream | -0.041 |
Token | , |
Token ID | 11 |
Feature activation | -1.59e-02 |
Pos logit contributions | |
pieces | +0.126 |
od | +0.110 |
Maria | +0.106 |
enta | +0.100 |
asing | +0.099 |
Neg logit contributions | |
robe | -0.109 |
cooler | -0.096 |
arl | -0.095 |
melon | -0.092 |
ugg | -0.088 |
Token | Snow |
Token ID | 7967 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
robe | +0.025 |
arl | +0.023 |
cooler | +0.021 |
melon | +0.021 |
ugg | +0.021 |
Neg logit contributions | |
pieces | -0.030 |
od | -0.027 |
Maria | -0.026 |
lier | -0.024 |
one | -0.024 |
Token | y |
Token ID | 88 |
Feature activation | -2.90e-02 |
Pos logit contributions | |
robe | +0.296 |
arl | +0.264 |
melon | +0.249 |
cooler | +0.247 |
ugg | +0.244 |
Neg logit contributions | |
pieces | -0.254 |
od | -0.211 |
Maria | -0.203 |
one | -0.188 |
asing | -0.187 |
Token | said |
Token ID | 531 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
robe | +0.056 |
cooler | +0.050 |
melon | +0.048 |
arl | +0.046 |
ranc | +0.046 |
Neg logit contributions | |
pieces | -0.045 |
od | -0.037 |
Maria | -0.036 |
asing | -0.034 |
enta | -0.034 |
Token | " |
Token ID | 366 |
Feature activation | -1.92e-02 |
Pos logit contributions | |
pieces | +0.026 |
od | +0.024 |
Maria | +0.022 |
lier | +0.021 |
one | +0.020 |
Neg logit contributions | |
robe | -0.032 |
arl | -0.030 |
cooler | -0.027 |
melon | -0.027 |
ugg | -0.027 |
Token | My |
Token ID | 3666 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
robe | +0.029 |
arl | +0.025 |
ugg | +0.024 |
Sleep | +0.023 |
Cream | +0.023 |
Neg logit contributions | |
pieces | -0.038 |
od | -0.034 |
Maria | -0.032 |
enta | -0.031 |
rect | -0.031 |
Token | turn |
Token ID | 1210 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
pieces | +0.085 |
od | +0.077 |
Maria | +0.074 |
one | +0.071 |
lier | +0.070 |
Neg logit contributions | |
robe | -0.052 |
arl | -0.048 |
ugg | -0.043 |
melon | -0.043 |
cooler | -0.041 |
Token | flowers |
Token ID | 12734 |
Feature activation | +6.44e-02 |
Pos logit contributions | |
robe | +0.051 |
arl | +0.049 |
ugg | +0.044 |
melon | +0.043 |
cooler | +0.042 |
Neg logit contributions | |
pieces | -0.051 |
od | -0.044 |
Maria | -0.042 |
lier | -0.039 |
circles | -0.039 |
Token | and |
Token ID | 290 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
pieces | +0.094 |
od | +0.082 |
Maria | +0.078 |
lier | +0.071 |
one | +0.070 |
Neg logit contributions | |
robe | -0.127 |
arl | -0.119 |
melon | -0.110 |
cooler | -0.110 |
ugg | -0.109 |
Token | hugs |
Token ID | 40657 |
Feature activation | -7.77e-02 |
Pos logit contributions | |
robe | +0.029 |
cooler | +0.027 |
melon | +0.025 |
ranc | +0.023 |
arl | +0.023 |
Neg logit contributions | |
pieces | -0.037 |
od | -0.035 |
Maria | -0.033 |
one | -0.033 |
lier | -0.032 |
Token | and |
Token ID | 290 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
robe | +0.138 |
arl | +0.121 |
melon | +0.116 |
cooler | +0.116 |
ugg | +0.113 |
Neg logit contributions | |
pieces | -0.134 |
od | -0.118 |
Maria | -0.116 |
enta | -0.109 |
asing | -0.107 |
Token | said |
Token ID | 531 |
Feature activation | +6.50e-03 |
Pos logit contributions | |
robe | +0.018 |
arl | +0.016 |
cooler | +0.015 |
melon | +0.015 |
ugg | +0.015 |
Neg logit contributions | |
pieces | -0.021 |
od | -0.019 |
Maria | -0.019 |
one | -0.017 |
lier | -0.017 |
Token | how |
Token ID | 703 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
pieces | +0.011 |
od | +0.010 |
Maria | +0.010 |
lier | +0.009 |
one | +0.009 |
Neg logit contributions | |
robe | -0.011 |
arl | -0.010 |
cooler | -0.009 |
melon | -0.009 |
ugg | -0.009 |
Token | wonderful |
Token ID | 7932 |
Feature activation | +7.54e-02 |
Pos logit contributions | |
robe | +0.293 |
arl | +0.266 |
melon | +0.247 |
cooler | +0.245 |
ugg | +0.238 |
Neg logit contributions | |
pieces | -0.331 |
od | -0.299 |
Maria | -0.288 |
lier | -0.266 |
one | -0.264 |
Token | she |
Token ID | 673 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
pieces | +0.149 |
od | +0.133 |
Maria | +0.130 |
enta | +0.123 |
one | +0.122 |
Neg logit contributions | |
robe | -0.113 |
cooler | -0.100 |
melon | -0.093 |
Cream | -0.090 |
arl | -0.090 |
Token | was |
Token ID | 373 |
Feature activation | +1.93e-02 |
Pos logit contributions | |
pieces | +0.062 |
Maria | +0.055 |
od | +0.052 |
circles | +0.049 |
rect | +0.048 |
Neg logit contributions | |
robe | -0.057 |
cooler | -0.049 |
ugg | -0.046 |
Sleep | -0.045 |
melon | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | -7.56e-03 |
Pos logit contributions | |
pieces | +0.036 |
Maria | +0.036 |
Jul | +0.033 |
rect | +0.033 |
Rachel | +0.033 |
Neg logit contributions | |
robe | -0.025 |
melon | -0.023 |
cooler | -0.022 |
tired | -0.021 |
thirsty | -0.020 |
Token | Lily |
Token ID | 20037 |
Feature activation | +4.37e-02 |
Pos logit contributions | |
robe | +0.011 |
arl | +0.010 |
melon | +0.009 |
ugg | +0.009 |
cooler | +0.009 |
Neg logit contributions | |
pieces | -0.015 |
od | -0.013 |
Maria | -0.013 |
one | -0.012 |
lier | -0.012 |
Token | wave |
Token ID | 6769 |
Feature activation | +5.97e-02 |
Pos logit contributions | |
pieces | +0.011 |
od | +0.010 |
Maria | +0.010 |
one | +0.009 |
lier | +0.009 |
Neg logit contributions | |
robe | -0.008 |
cooler | -0.007 |
melon | -0.007 |
arl | -0.007 |
ugg | -0.007 |
Token | his |
Token ID | 465 |
Feature activation | -1.97e-02 |
Pos logit contributions | |
pieces | +0.122 |
Maria | +0.110 |
od | +0.108 |
enta | +0.105 |
lar | +0.103 |
Neg logit contributions | |
robe | -0.083 |
arl | -0.071 |
cooler | -0.070 |
melon | -0.067 |
ugg | -0.062 |
Token | arms |
Token ID | 5101 |
Feature activation | +4.45e-02 |
Pos logit contributions | |
robe | +0.032 |
arl | +0.029 |
cooler | +0.028 |
melon | +0.027 |
ugg | +0.025 |
Neg logit contributions | |
pieces | -0.036 |
od | -0.033 |
Maria | -0.032 |
one | -0.030 |
lier | -0.029 |
Token | to |
Token ID | 284 |
Feature activation | +3.12e-02 |
Pos logit contributions | |
pieces | +0.072 |
od | +0.063 |
Maria | +0.061 |
lier | +0.056 |
one | +0.056 |
Neg logit contributions | |
robe | -0.081 |
arl | -0.075 |
melon | -0.070 |
cooler | -0.070 |
ugg | -0.069 |
Token | show |
Token ID | 905 |
Feature activation | +9.09e-02 |
Pos logit contributions | |
pieces | +0.062 |
Maria | +0.060 |
od | +0.059 |
asing | +0.058 |
ata | +0.057 |
Neg logit contributions | |
robe | -0.034 |
cooler | -0.033 |
melon | -0.032 |
arl | -0.032 |
bar | -0.032 |
Token | how |
Token ID | 703 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
pieces | +0.181 |
Maria | +0.167 |
od | +0.165 |
enta | +0.159 |
ill | +0.156 |
Neg logit contributions | |
robe | -0.124 |
arl | -0.108 |
cooler | -0.105 |
melon | -0.101 |
Sleep | -0.092 |
Token | proud |
Token ID | 6613 |
Feature activation | +4.88e-02 |
Pos logit contributions | |
robe | +0.288 |
arl | +0.250 |
melon | +0.224 |
ugg | +0.218 |
cooler | +0.216 |
Neg logit contributions | |
pieces | -0.331 |
od | -0.299 |
Maria | -0.289 |
one | -0.275 |
lier | -0.273 |
Token | he |
Token ID | 339 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
pieces | +0.104 |
Maria | +0.093 |
od | +0.090 |
enta | +0.089 |
lar | +0.088 |
Neg logit contributions | |
robe | -0.060 |
cooler | -0.057 |
melon | -0.054 |
arl | -0.052 |
Cream | -0.046 |
Token | was |
Token ID | 373 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
pieces | +0.032 |
Maria | +0.028 |
od | +0.025 |
ill | +0.025 |
asing | +0.025 |
Neg logit contributions | |
robe | -0.024 |
cooler | -0.021 |
Sleep | -0.019 |
Cream | -0.018 |
cabin | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +9.51e-03 |
Pos logit contributions | |
pieces | +0.039 |
Maria | +0.038 |
Jul | +0.037 |
rect | +0.036 |
Ah | +0.036 |
Neg logit contributions | |
robe | -0.022 |
cooler | -0.021 |
melon | -0.021 |
tired | -0.020 |
thirsty | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | +2.75e-02 |
Pos logit contributions | |
pieces | +0.016 |
od | +0.014 |
Maria | +0.014 |
lier | +0.013 |
one | +0.013 |
Neg logit contributions | |
robe | -0.017 |
arl | -0.015 |
melon | -0.014 |
cooler | -0.014 |
ugg | -0.014 |
Token | and |
Token ID | 290 |
Feature activation | +4.10e-03 |
Pos logit contributions | |
pieces | +0.087 |
od | +0.076 |
Maria | +0.075 |
lier | +0.068 |
one | +0.068 |
Neg logit contributions | |
robe | -0.098 |
arl | -0.088 |
melon | -0.084 |
cooler | -0.084 |
ugg | -0.081 |
Token | were |
Token ID | 547 |
Feature activation | +3.07e-02 |
Pos logit contributions | |
pieces | +0.008 |
od | +0.007 |
Maria | +0.007 |
ill | +0.006 |
lier | +0.006 |
Neg logit contributions | |
robe | -0.007 |
cooler | -0.006 |
arl | -0.005 |
melon | -0.005 |
ranc | -0.005 |
Token | very |
Token ID | 845 |
Feature activation | +1.56e-03 |
Pos logit contributions | |
pieces | +0.059 |
od | +0.053 |
Maria | +0.051 |
lier | +0.047 |
one | +0.047 |
Neg logit contributions | |
robe | -0.047 |
arl | -0.043 |
melon | -0.040 |
cooler | -0.039 |
ugg | -0.039 |
Token | careful |
Token ID | 8161 |
Feature activation | +2.21e-02 |
Pos logit contributions | |
pieces | +0.003 |
od | +0.003 |
Maria | +0.003 |
enta | +0.002 |
lier | +0.002 |
Neg logit contributions | |
robe | -0.002 |
arl | -0.002 |
melon | -0.002 |
ugg | -0.002 |
Sleep | -0.002 |
Token | with |
Token ID | 351 |
Feature activation | -3.01e-02 |
Pos logit contributions | |
pieces | +0.037 |
od | +0.033 |
Maria | +0.032 |
lier | +0.029 |
one | +0.029 |
Neg logit contributions | |
robe | -0.039 |
arl | -0.036 |
melon | -0.034 |
cooler | -0.034 |
ugg | -0.033 |
Token | how |
Token ID | 703 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
robe | +0.053 |
arl | +0.048 |
melon | +0.045 |
cooler | +0.045 |
ugg | +0.044 |
Neg logit contributions | |
pieces | -0.051 |
od | -0.046 |
Maria | -0.044 |
lier | -0.040 |
one | -0.040 |
Token | they |
Token ID | 484 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
robe | +0.314 |
arl | +0.287 |
melon | +0.269 |
cooler | +0.259 |
ugg | +0.258 |
Neg logit contributions | |
pieces | -0.285 |
od | -0.249 |
Maria | -0.243 |
one | -0.228 |
lier | -0.224 |
Token | wrapped |
Token ID | 12908 |
Feature activation | -8.41e-02 |
Pos logit contributions | |
pieces | +0.035 |
od | +0.031 |
Maria | +0.029 |
lier | +0.027 |
one | +0.027 |
Neg logit contributions | |
robe | -0.045 |
arl | -0.043 |
melon | -0.040 |
cooler | -0.039 |
ugg | -0.039 |
Token | it |
Token ID | 340 |
Feature activation | +1.87e-03 |
Pos logit contributions | |
robe | +0.146 |
arl | +0.128 |
melon | +0.124 |
cooler | +0.123 |
ugg | +0.117 |
Neg logit contributions | |
pieces | -0.146 |
od | -0.131 |
Maria | -0.126 |
one | -0.117 |
enta | -0.117 |
Token | . |
Token ID | 13 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
pieces | +0.003 |
od | +0.003 |
Maria | +0.003 |
enta | +0.003 |
one | +0.003 |
Neg logit contributions | |
robe | -0.003 |
arl | -0.003 |
cooler | -0.003 |
melon | -0.003 |
ugg | -0.002 |
Token | |
Token ID | 220 |
Feature activation | -2.53e-02 |
Pos logit contributions | |
pieces | +0.030 |
od | +0.027 |
Maria | +0.026 |
asing | +0.026 |
lar | +0.026 |
Neg logit contributions | |
robe | -0.018 |
arl | -0.016 |
Cream | -0.015 |
melon | -0.015 |
ugg | -0.015 |