Token | They |
Token ID | 1119 |
Feature activation | -4.33e-01 |
Pos logit contributions | |
cks | +0.026 |
lair | +0.026 |
pads | +0.025 |
necks | +0.024 |
floor | +0.024 |
Neg logit contributions | |
reward | -0.039 |
plunge | -0.038 |
help | -0.037 |
ar | -0.037 |
less | -0.036 |
Token | are |
Token ID | 389 |
Feature activation | +2.58e-01 |
Pos logit contributions | |
help | +0.028 |
reward | +0.023 |
plunge | +0.023 |
ar | +0.021 |
f | +0.020 |
Neg logit contributions | |
lair | -0.045 |
llers | -0.042 |
cks | -0.042 |
pads | -0.041 |
floor | -0.040 |
Token | different |
Token ID | 1180 |
Feature activation | +5.03e-01 |
Pos logit contributions | |
lair | +0.021 |
opl | +0.021 |
llers | +0.021 |
pads | +0.020 |
floor | +0.020 |
Neg logit contributions | |
help | -0.019 |
plunge | -0.016 |
Bub | -0.016 |
reward | -0.016 |
less | -0.015 |
Token | colors |
Token ID | 7577 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
lair | +0.044 |
cks | +0.043 |
pads | +0.041 |
mess | +0.041 |
floor | +0.040 |
Neg logit contributions | |
help | -0.034 |
reward | -0.033 |
plunge | -0.032 |
ar | -0.032 |
less | -0.031 |
Token | and |
Token ID | 290 |
Feature activation | +4.57e-01 |
Pos logit contributions | |
lair | +0.141 |
cks | +0.132 |
llers | +0.130 |
mess | +0.130 |
necks | +0.128 |
Neg logit contributions | |
help | -0.068 |
less | -0.065 |
iest | -0.060 |
reward | -0.060 |
of | -0.060 |
Token | sizes |
Token ID | 10620 |
Feature activation | +5.10e+00 |
Pos logit contributions | |
lair | +0.034 |
cks | +0.030 |
llers | +0.029 |
floor | +0.029 |
opl | +0.029 |
Neg logit contributions | |
help | -0.042 |
reward | -0.038 |
plunge | -0.035 |
ar | -0.035 |
f | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | +5.22e-01 |
Pos logit contributions | |
lair | +0.520 |
cks | +0.484 |
llers | +0.478 |
mess | +0.472 |
pads | +0.458 |
Neg logit contributions | |
help | -0.301 |
reward | -0.274 |
ar | -0.260 |
less | -0.256 |
plunge | -0.248 |
Token | Some |
Token ID | 2773 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
cks | +0.034 |
lair | +0.034 |
pads | +0.033 |
necks | +0.032 |
floor | +0.032 |
Neg logit contributions | |
reward | -0.046 |
plunge | -0.044 |
help | -0.044 |
ar | -0.043 |
less | -0.042 |
Token | are |
Token ID | 389 |
Feature activation | +3.40e-01 |
Pos logit contributions | |
help | +0.070 |
reward | +0.069 |
ar | +0.067 |
plunge | +0.066 |
bye | +0.063 |
Neg logit contributions | |
lair | -0.098 |
cks | -0.097 |
mess | -0.094 |
pads | -0.092 |
floor | -0.091 |
Token | red |
Token ID | 2266 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
lair | +0.028 |
cks | +0.026 |
opl | +0.026 |
llers | +0.026 |
floor | +0.025 |
Neg logit contributions | |
help | -0.026 |
reward | -0.024 |
less | -0.023 |
plunge | -0.023 |
ar | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | +3.09e-02 |
Pos logit contributions | |
cks | +0.092 |
lair | +0.091 |
mess | +0.090 |
pads | +0.090 |
necks | +0.088 |
Neg logit contributions | |
reward | -0.066 |
help | -0.064 |
ar | -0.063 |
plunge | -0.062 |
Sure | -0.062 |
Token | trucks |
Token ID | 13960 |
Feature activation | +9.10e-01 |
Pos logit contributions | |
lair | +0.036 |
cks | +0.036 |
mess | +0.034 |
pads | +0.033 |
floor | +0.032 |
Neg logit contributions | |
help | -0.047 |
reward | -0.046 |
ar | -0.044 |
plunge | -0.044 |
bye | -0.042 |
Token | of |
Token ID | 286 |
Feature activation | -7.15e-01 |
Pos logit contributions | |
lair | +0.074 |
cks | +0.074 |
mess | +0.071 |
pads | +0.069 |
floor | +0.069 |
Neg logit contributions | |
help | -0.066 |
reward | -0.065 |
ar | -0.063 |
plunge | -0.061 |
bye | -0.060 |
Token | different |
Token ID | 1180 |
Feature activation | +4.79e-01 |
Pos logit contributions | |
ar | +0.049 |
bye | +0.048 |
reward | +0.048 |
plunge | +0.047 |
less | +0.044 |
Neg logit contributions | |
mess | -0.061 |
lair | -0.061 |
cks | -0.060 |
floor | -0.059 |
sizes | -0.058 |
Token | colors |
Token ID | 7577 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
lair | +0.030 |
cks | +0.030 |
mess | +0.028 |
pads | +0.028 |
floor | +0.027 |
Neg logit contributions | |
help | -0.043 |
reward | -0.043 |
ar | -0.042 |
plunge | -0.041 |
bye | -0.041 |
Token | and |
Token ID | 290 |
Feature activation | +5.13e-01 |
Pos logit contributions | |
lair | +0.112 |
cks | +0.110 |
mess | +0.105 |
necks | +0.102 |
web | +0.100 |
Neg logit contributions | |
help | -0.080 |
reward | -0.075 |
ar | -0.074 |
less | -0.073 |
ted | -0.072 |
Token | sizes |
Token ID | 10620 |
Feature activation | +5.00e+00 |
Pos logit contributions | |
lair | +0.024 |
cks | +0.024 |
mess | +0.020 |
pads | +0.019 |
llers | +0.019 |
Neg logit contributions | |
help | -0.058 |
reward | -0.055 |
ar | -0.052 |
plunge | -0.051 |
bye | -0.051 |
Token | . |
Token ID | 13 |
Feature activation | +5.50e-01 |
Pos logit contributions | |
lair | +0.462 |
cks | +0.460 |
mess | +0.434 |
llers | +0.426 |
necks | +0.421 |
Neg logit contributions | |
help | -0.321 |
reward | -0.305 |
ar | -0.293 |
less | -0.283 |
plunge | -0.279 |
Token | He |
Token ID | 679 |
Feature activation | +3.78e-01 |
Pos logit contributions | |
lair | +0.042 |
cks | +0.042 |
mess | +0.040 |
pads | +0.039 |
floor | +0.039 |
Neg logit contributions | |
help | -0.043 |
reward | -0.042 |
ar | -0.041 |
plunge | -0.040 |
Sure | -0.040 |
Token | liked |
Token ID | 8288 |
Feature activation | +3.75e-01 |
Pos logit contributions | |
cks | +0.035 |
lair | +0.033 |
opl | +0.031 |
mess | +0.030 |
web | +0.030 |
Neg logit contributions | |
help | -0.026 |
reward | -0.024 |
ar | -0.024 |
Sure | -0.023 |
f | -0.023 |
Token | to |
Token ID | 284 |
Feature activation | +3.13e-01 |
Pos logit contributions | |
cks | +0.041 |
lair | +0.040 |
pads | +0.039 |
mess | +0.038 |
necks | +0.038 |
Neg logit contributions | |
help | -0.019 |
reward | -0.016 |
ted | -0.015 |
ar | -0.014 |
Bub | -0.014 |
Token | make |
Token ID | 787 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
lair | +0.031 |
cks | +0.030 |
pads | +0.029 |
mess | +0.028 |
floor | +0.027 |
Neg logit contributions | |
help | -0.020 |
reward | -0.018 |
ar | -0.017 |
bye | -0.016 |
plunge | -0.016 |
Token | are |
Token ID | 389 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
help | +0.036 |
reward | +0.029 |
plunge | +0.025 |
f | +0.024 |
Sure | +0.023 |
Neg logit contributions | |
lair | -0.065 |
llers | -0.061 |
opl | -0.060 |
pads | -0.059 |
cks | -0.059 |
Token | in |
Token ID | 287 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
opl | +0.017 |
lair | +0.017 |
llers | +0.016 |
storms | +0.016 |
pads | +0.015 |
Neg logit contributions | |
help | -0.012 |
reward | -0.011 |
Bub | -0.009 |
souven | -0.009 |
less | -0.009 |
Token | different |
Token ID | 1180 |
Feature activation | +4.38e-01 |
Pos logit contributions | |
lair | +0.016 |
cks | +0.015 |
mess | +0.015 |
pads | +0.015 |
floor | +0.015 |
Neg logit contributions | |
help | -0.009 |
reward | -0.009 |
ar | -0.009 |
plunge | -0.009 |
bye | -0.008 |
Token | shapes |
Token ID | 15268 |
Feature activation | +2.71e+00 |
Pos logit contributions | |
cks | +0.027 |
lair | +0.027 |
llers | +0.025 |
pads | +0.025 |
opl | +0.024 |
Neg logit contributions | |
help | -0.040 |
reward | -0.040 |
plunge | -0.037 |
ar | -0.037 |
less | -0.036 |
Token | and |
Token ID | 290 |
Feature activation | +3.99e-01 |
Pos logit contributions | |
lair | +0.292 |
llers | +0.289 |
cks | +0.277 |
die | +0.273 |
opl | +0.272 |
Neg logit contributions | |
help | -0.128 |
reward | -0.108 |
less | -0.108 |
of | -0.104 |
ar | -0.103 |
Token | sizes |
Token ID | 10620 |
Feature activation | +5.00e+00 |
Pos logit contributions | |
lair | +0.024 |
cks | +0.022 |
opl | +0.021 |
llers | +0.020 |
pads | +0.019 |
Neg logit contributions | |
help | -0.040 |
reward | -0.038 |
plunge | -0.035 |
souven | -0.034 |
ar | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | +3.15e-01 |
Pos logit contributions | |
lair | +0.534 |
cks | +0.500 |
llers | +0.497 |
opl | +0.472 |
pads | +0.471 |
Neg logit contributions | |
help | -0.266 |
reward | -0.237 |
less | -0.230 |
ar | -0.218 |
f | -0.211 |
Token | They |
Token ID | 1119 |
Feature activation | -5.28e-01 |
Pos logit contributions | |
pads | +0.018 |
lair | +0.018 |
cks | +0.018 |
floor | +0.017 |
necks | +0.017 |
Neg logit contributions | |
plunge | -0.030 |
reward | -0.029 |
ar | -0.029 |
help | -0.029 |
iest | -0.028 |
Token | are |
Token ID | 389 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
help | +0.034 |
reward | +0.028 |
f | +0.023 |
plunge | +0.022 |
Sure | +0.022 |
Neg logit contributions | |
lair | -0.059 |
llers | -0.055 |
opl | -0.055 |
cks | -0.054 |
pads | -0.054 |
Token | in |
Token ID | 287 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
opl | +0.020 |
lair | +0.020 |
llers | +0.019 |
storms | +0.019 |
cks | +0.018 |
Neg logit contributions | |
help | -0.016 |
reward | -0.014 |
Bub | -0.012 |
souven | -0.012 |
less | -0.012 |
Token | a |
Token ID | 257 |
Feature activation | -3.00e-01 |
Pos logit contributions | |
lair | +0.016 |
cks | +0.015 |
pads | +0.015 |
mess | +0.015 |
floor | +0.014 |
Neg logit contributions | |
help | -0.011 |
reward | -0.011 |
ar | -0.010 |
plunge | -0.010 |
Sure | -0.010 |
Token | it |
Token ID | 340 |
Feature activation | -8.30e-01 |
Pos logit contributions | |
lair | +0.135 |
cks | +0.130 |
pads | +0.129 |
necks | +0.126 |
mess | +0.126 |
Neg logit contributions | |
help | -0.059 |
reward | -0.056 |
plunge | -0.052 |
ar | -0.050 |
bye | -0.046 |
Token | with |
Token ID | 351 |
Feature activation | +6.92e-01 |
Pos logit contributions | |
help | +0.055 |
reward | +0.052 |
plunge | +0.046 |
ar | +0.044 |
bye | +0.044 |
Neg logit contributions | |
cks | -0.077 |
lair | -0.077 |
necks | -0.074 |
pads | -0.073 |
llers | -0.072 |
Token | holes |
Token ID | 10421 |
Feature activation | +2.27e+00 |
Pos logit contributions | |
lair | +0.064 |
cks | +0.063 |
mess | +0.062 |
pads | +0.060 |
floor | +0.060 |
Neg logit contributions | |
reward | -0.042 |
help | -0.042 |
ar | -0.041 |
plunge | -0.040 |
bye | -0.038 |
Token | of |
Token ID | 286 |
Feature activation | -7.80e-01 |
Pos logit contributions | |
lair | +0.185 |
cks | +0.181 |
mess | +0.174 |
pads | +0.170 |
llers | +0.168 |
Neg logit contributions | |
help | -0.166 |
reward | -0.162 |
ar | -0.158 |
plunge | -0.155 |
less | -0.151 |
Token | different |
Token ID | 1180 |
Feature activation | +3.88e-01 |
Pos logit contributions | |
help | +0.047 |
reward | +0.044 |
plunge | +0.041 |
ar | +0.038 |
ted | +0.036 |
Neg logit contributions | |
lair | -0.076 |
cks | -0.074 |
pads | -0.074 |
llers | -0.070 |
mess | -0.070 |
Token | sizes |
Token ID | 10620 |
Feature activation | +4.96e+00 |
Pos logit contributions | |
lair | +0.027 |
cks | +0.025 |
mess | +0.024 |
pads | +0.023 |
llers | +0.023 |
Neg logit contributions | |
help | -0.034 |
reward | -0.033 |
plunge | -0.032 |
ar | -0.032 |
bye | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | +3.04e-01 |
Pos logit contributions | |
lair | +0.457 |
cks | +0.434 |
mess | +0.414 |
llers | +0.411 |
frown | +0.403 |
Neg logit contributions | |
help | -0.333 |
reward | -0.318 |
ar | -0.304 |
plunge | -0.295 |
less | -0.293 |
Token | I |
Token ID | 314 |
Feature activation | +3.23e-01 |
Pos logit contributions | |
lair | +0.024 |
cks | +0.024 |
pads | +0.024 |
floor | +0.023 |
mess | +0.023 |
Neg logit contributions | |
reward | -0.022 |
help | -0.021 |
ar | -0.021 |
plunge | -0.021 |
less | -0.020 |
Token | will |
Token ID | 481 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
lair | +0.029 |
cks | +0.029 |
mess | +0.028 |
pads | +0.027 |
floor | +0.027 |
Neg logit contributions | |
help | -0.020 |
reward | -0.020 |
ar | -0.020 |
plunge | -0.020 |
bye | -0.019 |
Token | make |
Token ID | 787 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
help | +0.070 |
reward | +0.065 |
ar | +0.055 |
plunge | +0.055 |
bye | +0.053 |
Neg logit contributions | |
lair | -0.104 |
cks | -0.102 |
pads | -0.101 |
necks | -0.095 |
mess | -0.094 |
Token | it |
Token ID | 340 |
Feature activation | -8.23e-01 |
Pos logit contributions | |
lair | +0.130 |
cks | +0.126 |
pads | +0.125 |
necks | +0.122 |
mess | +0.121 |
Neg logit contributions | |
help | -0.062 |
reward | -0.058 |
plunge | -0.054 |
ar | -0.053 |
bye | -0.048 |
Token | shapes |
Token ID | 15268 |
Feature activation | +2.70e+00 |
Pos logit contributions | |
lair | +0.031 |
cks | +0.030 |
pads | +0.029 |
mess | +0.028 |
llers | +0.028 |
Neg logit contributions | |
help | -0.039 |
reward | -0.038 |
plunge | -0.037 |
ar | -0.036 |
Sure | -0.035 |
Token | by |
Token ID | 416 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
lair | +0.263 |
cks | +0.261 |
llers | +0.250 |
mess | +0.248 |
pads | +0.247 |
Neg logit contributions | |
help | -0.158 |
reward | -0.152 |
ar | -0.145 |
less | -0.141 |
plunge | -0.140 |
Token | their |
Token ID | 511 |
Feature activation | -5.12e-02 |
Pos logit contributions | |
help | +0.085 |
reward | +0.082 |
ar | +0.078 |
plunge | +0.077 |
less | +0.074 |
Neg logit contributions | |
lair | -0.115 |
cks | -0.109 |
pads | -0.108 |
mess | -0.107 |
llers | -0.103 |
Token | colors |
Token ID | 7577 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
help | +0.005 |
reward | +0.005 |
ar | +0.004 |
plunge | +0.004 |
bye | +0.004 |
Neg logit contributions | |
lair | -0.003 |
cks | -0.003 |
mess | -0.003 |
pads | -0.003 |
floor | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
lair | +0.101 |
cks | +0.099 |
mess | +0.093 |
pads | +0.091 |
llers | +0.090 |
Neg logit contributions | |
help | -0.085 |
reward | -0.084 |
ar | -0.081 |
plunge | -0.079 |
less | -0.079 |
Token | sizes |
Token ID | 10620 |
Feature activation | +4.96e+00 |
Pos logit contributions | |
lair | +0.021 |
cks | +0.021 |
mess | +0.019 |
opl | +0.019 |
pads | +0.019 |
Neg logit contributions | |
help | -0.020 |
reward | -0.020 |
ar | -0.018 |
plunge | -0.018 |
bye | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | +2.79e-01 |
Pos logit contributions | |
lair | +0.455 |
cks | +0.445 |
mess | +0.422 |
pads | +0.421 |
llers | +0.415 |
Neg logit contributions | |
help | -0.322 |
reward | -0.312 |
ar | -0.298 |
plunge | -0.293 |
less | -0.291 |
Token | made |
Token ID | 925 |
Feature activation | +3.87e-01 |
Pos logit contributions | |
cks | +0.023 |
lair | +0.023 |
mess | +0.021 |
pads | +0.020 |
llers | +0.020 |
Neg logit contributions | |
help | -0.021 |
reward | -0.021 |
ar | -0.019 |
plunge | -0.019 |
bye | -0.018 |
Token | their |
Token ID | 511 |
Feature activation | -5.65e-02 |
Pos logit contributions | |
lair | +0.035 |
cks | +0.035 |
mess | +0.034 |
pads | +0.034 |
floor | +0.033 |
Neg logit contributions | |
help | -0.024 |
reward | -0.024 |
ar | -0.023 |
plunge | -0.022 |
Sure | -0.021 |
Token | own |
Token ID | 898 |
Feature activation | -3.84e-01 |
Pos logit contributions | |
help | +0.005 |
reward | +0.005 |
plunge | +0.004 |
f | +0.004 |
bye | +0.004 |
Neg logit contributions | |
cks | -0.004 |
lair | -0.004 |
llers | -0.004 |
pads | -0.004 |
opl | -0.004 |
Token | puzzles |
Token ID | 24367 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
help | +0.031 |
reward | +0.031 |
plunge | +0.028 |
ar | +0.027 |
less | +0.027 |
Neg logit contributions | |
cks | -0.029 |
lair | -0.028 |
pads | -0.028 |
llers | -0.027 |
necks | -0.026 |
Token | clay |
Token ID | 21558 |
Feature activation | +9.63e-01 |
Pos logit contributions | |
lair | +0.150 |
cks | +0.148 |
mess | +0.143 |
pads | +0.142 |
floor | +0.140 |
Neg logit contributions | |
help | -0.091 |
reward | -0.090 |
ar | -0.087 |
plunge | -0.086 |
bye | -0.082 |
Token | of |
Token ID | 286 |
Feature activation | -7.44e-01 |
Pos logit contributions | |
lair | +0.085 |
cks | +0.083 |
pads | +0.080 |
mess | +0.079 |
necks | +0.078 |
Neg logit contributions | |
help | -0.065 |
reward | -0.063 |
ar | -0.060 |
plunge | -0.059 |
ted | -0.057 |
Token | all |
Token ID | 477 |
Feature activation | -6.82e-02 |
Pos logit contributions | |
help | +0.041 |
reward | +0.041 |
ar | +0.039 |
plunge | +0.038 |
bye | +0.037 |
Neg logit contributions | |
lair | -0.073 |
cks | -0.073 |
mess | -0.071 |
pads | -0.069 |
floor | -0.069 |
Token | kinds |
Token ID | 6982 |
Feature activation | +7.52e-01 |
Pos logit contributions | |
help | +0.003 |
reward | +0.002 |
plunge | +0.002 |
iest | +0.002 |
f | +0.002 |
Neg logit contributions | |
lair | -0.008 |
pads | -0.008 |
necks | -0.007 |
cks | -0.007 |
llers | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | +4.24e-01 |
Pos logit contributions | |
lair | +0.098 |
cks | +0.096 |
pads | +0.093 |
mess | +0.092 |
llers | +0.092 |
Neg logit contributions | |
help | -0.020 |
reward | -0.017 |
ar | -0.016 |
less | -0.015 |
f | -0.013 |
Token | sizes |
Token ID | 10620 |
Feature activation | +4.95e+00 |
Pos logit contributions | |
lair | +0.024 |
cks | +0.022 |
pads | +0.020 |
opl | +0.019 |
llers | +0.019 |
Neg logit contributions | |
help | -0.045 |
reward | -0.043 |
plunge | -0.040 |
f | -0.039 |
ar | -0.039 |
Token | and |
Token ID | 290 |
Feature activation | +4.08e-01 |
Pos logit contributions | |
lair | +0.490 |
cks | +0.465 |
llers | +0.448 |
mess | +0.439 |
necks | +0.439 |
Neg logit contributions | |
help | -0.301 |
reward | -0.279 |
less | -0.265 |
ar | -0.264 |
plunge | -0.251 |
Token | shapes |
Token ID | 15268 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
lair | +0.026 |
cks | +0.025 |
opl | +0.022 |
pads | +0.022 |
llers | +0.022 |
Neg logit contributions | |
help | -0.040 |
reward | -0.038 |
plunge | -0.034 |
ar | -0.034 |
f | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | +3.73e-01 |
Pos logit contributions | |
lair | +0.268 |
cks | +0.258 |
llers | +0.256 |
necks | +0.251 |
mess | +0.248 |
Neg logit contributions | |
help | -0.155 |
reward | -0.142 |
less | -0.135 |
ar | -0.132 |
iest | -0.123 |
Token | They |
Token ID | 1119 |
Feature activation | -4.24e-01 |
Pos logit contributions | |
lair | +0.027 |
cks | +0.026 |
pads | +0.025 |
mess | +0.025 |
floor | +0.025 |
Neg logit contributions | |
reward | -0.031 |
help | -0.030 |
plunge | -0.030 |
ar | -0.030 |
less | -0.028 |
Token | are |
Token ID | 389 |
Feature activation | +2.20e-01 |
Pos logit contributions | |
help | +0.027 |
reward | +0.019 |
plunge | +0.017 |
Sure | +0.016 |
f | +0.016 |
Neg logit contributions | |
lair | -0.049 |
pads | -0.046 |
llers | -0.045 |
cks | -0.045 |
opl | -0.044 |
Token | They |
Token ID | 1119 |
Feature activation | -4.61e-01 |
Pos logit contributions | |
lair | +0.033 |
floor | +0.032 |
cks | +0.032 |
pads | +0.031 |
mess | +0.031 |
Neg logit contributions | |
help | -0.040 |
reward | -0.040 |
plunge | -0.040 |
ar | -0.039 |
f | -0.038 |
Token | were |
Token ID | 547 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
help | +0.032 |
reward | +0.029 |
Sure | +0.028 |
ar | +0.027 |
plunge | +0.027 |
Neg logit contributions | |
lair | -0.043 |
cks | -0.041 |
mess | -0.039 |
opl | -0.039 |
llers | -0.038 |
Token | different |
Token ID | 1180 |
Feature activation | +3.62e-01 |
Pos logit contributions | |
lair | +0.018 |
opl | +0.017 |
urious | +0.017 |
llers | +0.016 |
pads | +0.016 |
Neg logit contributions | |
help | -0.015 |
reward | -0.012 |
plunge | -0.012 |
less | -0.012 |
Bub | -0.011 |
Token | colors |
Token ID | 7577 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
cks | +0.028 |
lair | +0.028 |
mess | +0.026 |
pads | +0.026 |
floor | +0.025 |
Neg logit contributions | |
help | -0.028 |
reward | -0.027 |
ar | -0.026 |
plunge | -0.026 |
less | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | +3.74e-01 |
Pos logit contributions | |
lair | +0.114 |
cks | +0.108 |
mess | +0.100 |
necks | +0.100 |
llers | +0.100 |
Neg logit contributions | |
help | -0.070 |
less | -0.068 |
reward | -0.064 |
iest | -0.063 |
ar | -0.061 |
Token | sizes |
Token ID | 10620 |
Feature activation | +4.94e+00 |
Pos logit contributions | |
lair | +0.020 |
cks | +0.018 |
mess | +0.016 |
llers | +0.015 |
pads | +0.015 |
Neg logit contributions | |
help | -0.041 |
reward | -0.039 |
ar | -0.037 |
plunge | -0.037 |
less | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | +4.69e-01 |
Pos logit contributions | |
lair | +0.477 |
cks | +0.460 |
llers | +0.433 |
mess | +0.433 |
pads | +0.424 |
Neg logit contributions | |
help | -0.313 |
reward | -0.292 |
ar | -0.276 |
less | -0.275 |
plunge | -0.268 |
Token | Lily |
Token ID | 20037 |
Feature activation | +7.98e-01 |
Pos logit contributions | |
lair | +0.030 |
floor | +0.030 |
cks | +0.029 |
pads | +0.029 |
mess | +0.028 |
Neg logit contributions | |
reward | -0.041 |
help | -0.041 |
plunge | -0.041 |
ar | -0.040 |
bye | -0.039 |
Token | picked |
Token ID | 6497 |
Feature activation | -6.15e-01 |
Pos logit contributions | |
cks | +0.093 |
lair | +0.090 |
opl | +0.089 |
ric | +0.083 |
fault | +0.080 |
Neg logit contributions | |
ed | -0.035 |
help | -0.033 |
ar | -0.031 |
ted | -0.031 |
f | -0.029 |
Token | a |
Token ID | 257 |
Feature activation | -3.07e-01 |
Pos logit contributions | |
reward | +0.041 |
bye | +0.040 |
ar | +0.039 |
help | +0.039 |
plunge | +0.037 |
Neg logit contributions | |
cks | -0.053 |
pads | -0.053 |
neck | -0.050 |
mess | -0.050 |
goals | -0.049 |
Token | pink |
Token ID | 11398 |
Feature activation | +9.87e-01 |
Pos logit contributions | |
help | +0.025 |
reward | +0.025 |
ar | +0.025 |
plunge | +0.024 |
Sure | +0.024 |
Neg logit contributions | |
lair | -0.022 |
cks | -0.022 |
mess | -0.020 |
pads | -0.020 |
floor | -0.020 |
Token | cars |
Token ID | 5006 |
Feature activation | -4.20e-01 |
Pos logit contributions | |
lair | +0.076 |
cks | +0.075 |
pads | +0.069 |
mess | +0.069 |
llers | +0.068 |
Neg logit contributions | |
help | -0.085 |
reward | -0.081 |
ar | -0.078 |
plunge | -0.077 |
ted | -0.074 |
Token | of |
Token ID | 286 |
Feature activation | -9.00e-01 |
Pos logit contributions | |
reward | +0.027 |
help | +0.026 |
plunge | +0.026 |
ar | +0.026 |
Sure | +0.025 |
Neg logit contributions | |
lair | -0.037 |
cks | -0.037 |
mess | -0.036 |
pads | -0.036 |
floor | -0.035 |
Token | different |
Token ID | 1180 |
Feature activation | +3.14e-01 |
Pos logit contributions | |
ar | +0.071 |
reward | +0.070 |
bye | +0.069 |
plunge | +0.069 |
less | +0.067 |
Neg logit contributions | |
cks | -0.067 |
mess | -0.066 |
lair | -0.066 |
floor | -0.065 |
sizes | -0.064 |
Token | colors |
Token ID | 7577 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
lair | +0.019 |
cks | +0.019 |
mess | +0.017 |
pads | +0.017 |
llers | +0.016 |
Neg logit contributions | |
help | -0.029 |
reward | -0.029 |
ar | -0.028 |
plunge | -0.028 |
bye | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | +3.74e-01 |
Pos logit contributions | |
lair | +0.106 |
cks | +0.099 |
mess | +0.096 |
necks | +0.093 |
web | +0.093 |
Neg logit contributions | |
help | -0.065 |
ar | -0.058 |
iest | -0.057 |
reward | -0.057 |
less | -0.056 |
Token | sizes |
Token ID | 10620 |
Feature activation | +4.92e+00 |
Pos logit contributions | |
lair | +0.021 |
cks | +0.019 |
mess | +0.016 |
irling | +0.016 |
llers | +0.016 |
Neg logit contributions | |
help | -0.041 |
reward | -0.038 |
ar | -0.035 |
Bub | -0.035 |
f | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | +4.21e-01 |
Pos logit contributions | |
lair | +0.475 |
cks | +0.452 |
mess | +0.439 |
llers | +0.431 |
necks | +0.423 |
Neg logit contributions | |
help | -0.309 |
reward | -0.286 |
ar | -0.279 |
f | -0.262 |
less | -0.261 |
Token | He |
Token ID | 679 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
lair | +0.031 |
cks | +0.030 |
mess | +0.029 |
pads | +0.029 |
floor | +0.028 |
Neg logit contributions | |
help | -0.034 |
reward | -0.034 |
ar | -0.033 |
plunge | -0.033 |
bye | -0.032 |
Token | likes |
Token ID | 7832 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
lair | +0.016 |
cks | +0.016 |
llers | +0.015 |
ivering | +0.015 |
cking | +0.015 |
Neg logit contributions | |
help | -0.019 |
reward | -0.017 |
ar | -0.016 |
Bub | -0.016 |
f | -0.015 |
Token | to |
Token ID | 284 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
cks | +0.121 |
lair | +0.120 |
pads | +0.113 |
mess | +0.113 |
necks | +0.111 |
Neg logit contributions | |
help | -0.062 |
reward | -0.055 |
ar | -0.051 |
ted | -0.050 |
plunge | -0.050 |
Token | make |
Token ID | 787 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
lair | +0.015 |
cks | +0.014 |
pads | +0.014 |
mess | +0.014 |
floor | +0.013 |
Neg logit contributions | |
help | -0.011 |
reward | -0.011 |
ar | -0.010 |
plunge | -0.010 |
Sure | -0.010 |
Token | cars |
Token ID | 5006 |
Feature activation | -3.93e-01 |
Pos logit contributions | |
lair | +0.078 |
cks | +0.077 |
mess | +0.072 |
pads | +0.072 |
floor | +0.070 |
Neg logit contributions | |
help | -0.090 |
reward | -0.088 |
ar | -0.085 |
plunge | -0.084 |
Sure | -0.081 |
Token | of |
Token ID | 286 |
Feature activation | -9.03e-01 |
Pos logit contributions | |
plunge | +0.026 |
reward | +0.025 |
ar | +0.024 |
Sure | +0.024 |
help | +0.023 |
Neg logit contributions | |
pads | -0.034 |
mess | -0.034 |
cks | -0.034 |
lair | -0.033 |
floor | -0.032 |
Token | different |
Token ID | 1180 |
Feature activation | +3.60e-01 |
Pos logit contributions | |
ar | +0.070 |
reward | +0.069 |
plunge | +0.068 |
bye | +0.068 |
less | +0.065 |
Neg logit contributions | |
mess | -0.069 |
cks | -0.069 |
lair | -0.068 |
floor | -0.066 |
sizes | -0.066 |
Token | colors |
Token ID | 7577 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
lair | +0.022 |
cks | +0.021 |
mess | +0.020 |
pads | +0.019 |
floor | +0.019 |
Neg logit contributions | |
help | -0.034 |
reward | -0.033 |
ar | -0.033 |
plunge | -0.032 |
bye | -0.031 |
Token | and |
Token ID | 290 |
Feature activation | +3.92e-01 |
Pos logit contributions | |
lair | +0.101 |
cks | +0.095 |
mess | +0.090 |
necks | +0.088 |
pads | +0.087 |
Neg logit contributions | |
help | -0.070 |
reward | -0.065 |
ar | -0.064 |
less | -0.063 |
iest | -0.061 |
Token | sizes |
Token ID | 10620 |
Feature activation | +4.92e+00 |
Pos logit contributions | |
lair | +0.018 |
cks | +0.016 |
mess | +0.013 |
llers | +0.013 |
opl | +0.012 |
Neg logit contributions | |
help | -0.047 |
reward | -0.044 |
ar | -0.041 |
f | -0.041 |
plunge | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | +5.16e-01 |
Pos logit contributions | |
lair | +0.455 |
cks | +0.436 |
mess | +0.420 |
llers | +0.408 |
pads | +0.407 |
Neg logit contributions | |
help | -0.323 |
reward | -0.308 |
ar | -0.300 |
plunge | -0.285 |
less | -0.283 |
Token | They |
Token ID | 1119 |
Feature activation | -4.19e-01 |
Pos logit contributions | |
lair | +0.042 |
cks | +0.041 |
mess | +0.040 |
pads | +0.039 |
floor | +0.039 |
Neg logit contributions | |
help | -0.038 |
reward | -0.037 |
ar | -0.036 |
plunge | -0.036 |
bye | -0.035 |
Token | like |
Token ID | 588 |
Feature activation | -3.24e-01 |
Pos logit contributions | |
help | +0.026 |
reward | +0.022 |
f | +0.019 |
Sure | +0.019 |
ar | +0.018 |
Neg logit contributions | |
lair | -0.045 |
cks | -0.044 |
llers | -0.042 |
pads | -0.041 |
opl | -0.040 |
Token | to |
Token ID | 284 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
help | +0.018 |
reward | +0.016 |
less | +0.014 |
plunge | +0.014 |
Sure | +0.014 |
Neg logit contributions | |
cks | -0.034 |
lair | -0.034 |
pads | -0.032 |
mess | -0.031 |
necks | -0.031 |
Token | drive |
Token ID | 3708 |
Feature activation | -5.88e-01 |
Pos logit contributions | |
lair | +0.023 |
cks | +0.022 |
pads | +0.021 |
mess | +0.021 |
floor | +0.021 |
Neg logit contributions | |
help | -0.016 |
reward | -0.016 |
ar | -0.015 |
plunge | -0.015 |
bye | -0.014 |
Token | with |
Token ID | 351 |
Feature activation | +6.92e-01 |
Pos logit contributions | |
lair | +0.254 |
cks | +0.246 |
llers | +0.239 |
pads | +0.237 |
necks | +0.236 |
Neg logit contributions | |
help | -0.163 |
reward | -0.151 |
ar | -0.143 |
less | -0.142 |
plunge | -0.133 |
Token | the |
Token ID | 262 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
lair | +0.061 |
cks | +0.061 |
mess | +0.059 |
pads | +0.058 |
floor | +0.058 |
Neg logit contributions | |
reward | -0.045 |
help | -0.044 |
ar | -0.044 |
plunge | -0.043 |
bye | -0.041 |
Token | different |
Token ID | 1180 |
Feature activation | +3.96e-01 |
Pos logit contributions | |
cks | +0.025 |
lair | +0.025 |
llers | +0.023 |
pads | +0.023 |
mess | +0.023 |
Neg logit contributions | |
help | -0.022 |
reward | -0.021 |
plunge | -0.021 |
ar | -0.020 |
less | -0.019 |
Token | colors |
Token ID | 7577 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
lair | +0.028 |
cks | +0.028 |
pads | +0.025 |
llers | +0.025 |
mess | +0.025 |
Neg logit contributions | |
help | -0.034 |
reward | -0.033 |
plunge | -0.032 |
ar | -0.031 |
bye | -0.031 |
Token | and |
Token ID | 290 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
lair | +0.116 |
cks | +0.108 |
mess | +0.104 |
llers | +0.103 |
necks | +0.103 |
Neg logit contributions | |
help | -0.067 |
reward | -0.064 |
less | -0.061 |
ar | -0.060 |
plunge | -0.058 |
Token | sizes |
Token ID | 10620 |
Feature activation | +4.92e+00 |
Pos logit contributions | |
lair | +0.015 |
cks | +0.013 |
opl | +0.013 |
llers | +0.012 |
pads | +0.012 |
Neg logit contributions | |
help | -0.024 |
reward | -0.023 |
plunge | -0.021 |
souven | -0.021 |
less | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
lair | +0.473 |
cks | +0.449 |
mess | +0.429 |
llers | +0.428 |
pads | +0.426 |
Neg logit contributions | |
help | -0.309 |
reward | -0.295 |
ar | -0.279 |
less | -0.276 |
plunge | -0.271 |
Token | They |
Token ID | 1119 |
Feature activation | -5.93e-01 |
Pos logit contributions | |
lair | +0.012 |
cks | +0.012 |
pads | +0.012 |
mess | +0.011 |
floor | +0.011 |
Neg logit contributions | |
reward | -0.015 |
help | -0.015 |
plunge | -0.014 |
ar | -0.014 |
bye | -0.014 |
Token | are |
Token ID | 389 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
help | +0.038 |
reward | +0.031 |
f | +0.027 |
plunge | +0.025 |
Sure | +0.024 |
Neg logit contributions | |
lair | -0.066 |
llers | -0.062 |
cks | -0.062 |
pads | -0.061 |
opl | -0.059 |
Token | very |
Token ID | 845 |
Feature activation | +8.75e-02 |
Pos logit contributions | |
lair | +0.021 |
opl | +0.020 |
pads | +0.020 |
cks | +0.019 |
urious | +0.019 |
Neg logit contributions | |
help | -0.014 |
reward | -0.011 |
less | -0.010 |
Bub | -0.009 |
souven | -0.009 |
Token | proud |
Token ID | 6613 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
lair | +0.007 |
cks | +0.007 |
pads | +0.007 |
necks | +0.006 |
mess | +0.006 |
Neg logit contributions | |
help | -0.007 |
reward | -0.007 |
ar | -0.006 |
less | -0.006 |
plunge | -0.006 |
Token | were |
Token ID | 547 |
Feature activation | +9.30e-02 |
Pos logit contributions | |
help | +0.026 |
reward | +0.022 |
plunge | +0.022 |
f | +0.021 |
ar | +0.021 |
Neg logit contributions | |
lair | -0.055 |
cks | -0.051 |
pads | -0.049 |
opl | -0.049 |
mess | -0.049 |
Token | all |
Token ID | 477 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
pads | +0.008 |
lair | +0.008 |
cks | +0.007 |
necks | +0.007 |
opl | +0.007 |
Neg logit contributions | |
help | -0.006 |
plunge | -0.005 |
less | -0.005 |
reward | -0.005 |
Bub | -0.005 |
Token | different |
Token ID | 1180 |
Feature activation | +2.63e-01 |
Pos logit contributions | |
help | +0.020 |
less | +0.018 |
plunge | +0.017 |
reward | +0.017 |
iest | +0.016 |
Neg logit contributions | |
lair | -0.022 |
necks | -0.021 |
pads | -0.021 |
cks | -0.021 |
llers | -0.020 |
Token | colors |
Token ID | 7577 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
lair | +0.017 |
cks | +0.017 |
pads | +0.016 |
mess | +0.016 |
floor | +0.015 |
Neg logit contributions | |
help | -0.023 |
reward | -0.023 |
plunge | -0.022 |
ar | -0.022 |
less | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
lair | +0.109 |
cks | +0.104 |
necks | +0.098 |
mess | +0.097 |
llers | +0.095 |
Neg logit contributions | |
help | -0.063 |
less | -0.061 |
reward | -0.058 |
iest | -0.057 |
plunge | -0.056 |
Token | sizes |
Token ID | 10620 |
Feature activation | +4.91e+00 |
Pos logit contributions | |
lair | +0.014 |
cks | +0.013 |
opl | +0.011 |
mess | +0.011 |
pads | +0.011 |
Neg logit contributions | |
help | -0.022 |
reward | -0.021 |
plunge | -0.020 |
less | -0.020 |
ar | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | +3.33e-01 |
Pos logit contributions | |
lair | +0.480 |
cks | +0.459 |
mess | +0.433 |
llers | +0.430 |
necks | +0.427 |
Neg logit contributions | |
help | -0.307 |
reward | -0.287 |
less | -0.273 |
ar | -0.272 |
plunge | -0.268 |
Token | She |
Token ID | 1375 |
Feature activation | +9.32e-01 |
Pos logit contributions | |
cks | +0.027 |
lair | +0.027 |
mess | +0.025 |
floor | +0.024 |
necks | +0.024 |
Neg logit contributions | |
help | -0.025 |
reward | -0.024 |
Sure | -0.024 |
ar | -0.024 |
plunge | -0.023 |
Token | wondered |
Token ID | 14028 |
Feature activation | +2.20e+00 |
Pos logit contributions | |
cks | +0.095 |
lair | +0.092 |
opl | +0.090 |
fault | +0.083 |
mess | +0.081 |
Neg logit contributions | |
help | -0.054 |
reward | -0.051 |
Sure | -0.050 |
ar | -0.050 |
ted | -0.048 |
Token | how |
Token ID | 703 |
Feature activation | +4.11e-01 |
Pos logit contributions | |
lair | +0.187 |
cks | +0.185 |
mess | +0.179 |
pads | +0.176 |
floor | +0.174 |
Neg logit contributions | |
reward | -0.149 |
help | -0.149 |
ar | -0.145 |
plunge | -0.142 |
bye | -0.138 |
Token | many |
Token ID | 867 |
Feature activation | +9.85e-01 |
Pos logit contributions | |
lair | +0.038 |
cks | +0.038 |
mess | +0.037 |
pads | +0.036 |
floor | +0.036 |
Neg logit contributions | |
reward | -0.024 |
help | -0.024 |
ar | -0.024 |
plunge | -0.023 |
bye | -0.022 |
Token | cars |
Token ID | 5006 |
Feature activation | -4.30e-01 |
Pos logit contributions | |
lair | +0.081 |
cks | +0.078 |
pads | +0.072 |
llers | +0.071 |
mess | +0.070 |
Neg logit contributions | |
help | -0.086 |
reward | -0.081 |
ar | -0.077 |
plunge | -0.075 |
Sure | -0.074 |
Token | of |
Token ID | 286 |
Feature activation | -9.28e-01 |
Pos logit contributions | |
reward | +0.026 |
plunge | +0.026 |
help | +0.025 |
ar | +0.025 |
Sure | +0.024 |
Neg logit contributions | |
cks | -0.039 |
lair | -0.039 |
mess | -0.038 |
pads | -0.038 |
floor | -0.037 |
Token | different |
Token ID | 1180 |
Feature activation | +3.22e-01 |
Pos logit contributions | |
reward | +0.060 |
ar | +0.060 |
bye | +0.058 |
plunge | +0.058 |
help | +0.056 |
Neg logit contributions | |
cks | -0.082 |
mess | -0.081 |
lair | -0.081 |
floor | -0.080 |
sizes | -0.078 |
Token | colors |
Token ID | 7577 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
lair | +0.020 |
cks | +0.019 |
mess | +0.017 |
pads | +0.017 |
llers | +0.016 |
Neg logit contributions | |
help | -0.030 |
reward | -0.029 |
ar | -0.029 |
plunge | -0.028 |
bye | -0.028 |
Token | and |
Token ID | 290 |
Feature activation | +3.56e-01 |
Pos logit contributions | |
lair | +0.107 |
cks | +0.099 |
mess | +0.095 |
necks | +0.095 |
web | +0.094 |
Neg logit contributions | |
help | -0.063 |
iest | -0.057 |
reward | -0.057 |
ar | -0.057 |
less | -0.055 |
Token | sizes |
Token ID | 10620 |
Feature activation | +4.90e+00 |
Pos logit contributions | |
lair | +0.017 |
cks | +0.015 |
irling | +0.012 |
iver | +0.012 |
ivering | +0.012 |
Neg logit contributions | |
help | -0.042 |
reward | -0.039 |
Bub | -0.036 |
ar | -0.036 |
f | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | +3.94e-01 |
Pos logit contributions | |
lair | +0.476 |
cks | +0.452 |
mess | +0.438 |
llers | +0.431 |
necks | +0.424 |
Neg logit contributions | |
help | -0.305 |
reward | -0.283 |
ar | -0.275 |
f | -0.260 |
less | -0.259 |
Token | He |
Token ID | 679 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
lair | +0.032 |
cks | +0.032 |
mess | +0.031 |
pads | +0.030 |
floor | +0.029 |
Neg logit contributions | |
help | -0.029 |
reward | -0.028 |
ar | -0.027 |
Sure | -0.027 |
plunge | -0.026 |
Token | likes |
Token ID | 7832 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
lair | +0.018 |
cks | +0.018 |
llers | +0.017 |
ivering | +0.017 |
iver | +0.016 |
Neg logit contributions | |
help | -0.020 |
reward | -0.017 |
ar | -0.017 |
Bub | -0.016 |
f | -0.016 |
Token | to |
Token ID | 284 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
lair | +0.121 |
cks | +0.121 |
pads | +0.114 |
mess | +0.114 |
necks | +0.112 |
Neg logit contributions | |
help | -0.064 |
reward | -0.056 |
ted | -0.052 |
ar | -0.052 |
Bub | -0.052 |
Token | make |
Token ID | 787 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
lair | +0.017 |
cks | +0.017 |
pads | +0.016 |
mess | +0.016 |
floor | +0.016 |
Neg logit contributions | |
help | -0.013 |
reward | -0.013 |
ar | -0.012 |
plunge | -0.012 |
Sure | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | +2.43e-01 |
Pos logit contributions | |
lair | +0.158 |
cks | +0.155 |
mess | +0.151 |
pads | +0.150 |
floor | +0.147 |
Neg logit contributions | |
help | -0.096 |
reward | -0.094 |
ar | -0.090 |
plunge | -0.088 |
bye | -0.084 |
Token | blocks |
Token ID | 7021 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
cks | +0.018 |
lair | +0.016 |
llers | +0.016 |
pads | +0.016 |
necks | +0.015 |
Neg logit contributions | |
help | -0.021 |
reward | -0.021 |
plunge | -0.019 |
ar | -0.018 |
f | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
lair | +0.098 |
cks | +0.097 |
mess | +0.094 |
pads | +0.092 |
floor | +0.091 |
Neg logit contributions | |
help | -0.065 |
reward | -0.064 |
ar | -0.062 |
plunge | -0.060 |
bye | -0.058 |
Token | compare |
Token ID | 8996 |
Feature activation | +7.98e-01 |
Pos logit contributions | |
help | +0.015 |
reward | +0.014 |
ar | +0.014 |
plunge | +0.014 |
Sure | +0.013 |
Neg logit contributions | |
lair | -0.016 |
cks | -0.016 |
mess | -0.015 |
pads | -0.015 |
floor | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | +2.54e-01 |
Pos logit contributions | |
cks | +0.072 |
lair | +0.072 |
mess | +0.070 |
pads | +0.069 |
floor | +0.068 |
Neg logit contributions | |
reward | -0.050 |
help | -0.050 |
ar | -0.048 |
plunge | -0.047 |
bye | -0.046 |
Token | sizes |
Token ID | 10620 |
Feature activation | +4.90e+00 |
Pos logit contributions | |
cks | +0.019 |
lair | +0.018 |
llers | +0.017 |
pads | +0.017 |
necks | +0.016 |
Neg logit contributions | |
help | -0.022 |
reward | -0.021 |
plunge | -0.019 |
f | -0.018 |
ar | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
lair | +0.488 |
cks | +0.481 |
mess | +0.467 |
pads | +0.461 |
floor | +0.455 |
Neg logit contributions | |
help | -0.266 |
reward | -0.263 |
ar | -0.252 |
plunge | -0.247 |
bye | -0.236 |
Token | and |
Token ID | 290 |
Feature activation | +2.01e-01 |
Pos logit contributions | |
help | +0.018 |
reward | +0.018 |
ar | +0.017 |
plunge | +0.017 |
bye | +0.016 |
Neg logit contributions | |
lair | -0.017 |
cks | -0.016 |
mess | -0.015 |
pads | -0.015 |
floor | -0.015 |
Token | sort |
Token ID | 3297 |
Feature activation | +9.20e-01 |
Pos logit contributions | |
lair | +0.015 |
cks | +0.014 |
pads | +0.013 |
mess | +0.013 |
opl | +0.013 |
Neg logit contributions | |
help | -0.017 |
reward | -0.016 |
plunge | -0.015 |
ar | -0.014 |
f | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | +2.70e-01 |
Pos logit contributions | |
cks | +0.079 |
mess | +0.078 |
sizes | +0.076 |
floor | +0.076 |
lair | +0.075 |
Neg logit contributions | |
reward | -0.061 |
ar | -0.060 |
plunge | -0.059 |
help | -0.058 |
Sure | -0.057 |
Token | colors |
Token ID | 7577 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
cks | +0.019 |
lair | +0.019 |
llers | +0.018 |
pads | +0.017 |
irling | +0.017 |
Neg logit contributions | |
help | -0.023 |
reward | -0.022 |
plunge | -0.021 |
f | -0.020 |
ar | -0.020 |
Token | were |
Token ID | 547 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
help | +0.042 |
reward | +0.040 |
plunge | +0.038 |
ar | +0.038 |
ted | +0.038 |
Neg logit contributions | |
lair | -0.053 |
cks | -0.050 |
pads | -0.048 |
mess | -0.048 |
opl | -0.048 |
Token | of |
Token ID | 286 |
Feature activation | -8.54e-01 |
Pos logit contributions | |
lair | +0.014 |
opl | +0.014 |
pads | +0.013 |
llers | +0.013 |
urious | +0.012 |
Neg logit contributions | |
help | -0.014 |
reward | -0.012 |
plunge | -0.012 |
less | -0.012 |
souven | -0.011 |
Token | different |
Token ID | 1180 |
Feature activation | +3.25e-01 |
Pos logit contributions | |
ar | +0.064 |
reward | +0.064 |
help | +0.063 |
plunge | +0.061 |
bye | +0.061 |
Neg logit contributions | |
lair | -0.067 |
cks | -0.066 |
mess | -0.065 |
floor | -0.062 |
pads | -0.062 |
Token | colors |
Token ID | 7577 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
lair | +0.020 |
cks | +0.020 |
pads | +0.019 |
mess | +0.018 |
llers | +0.018 |
Neg logit contributions | |
help | -0.030 |
reward | -0.029 |
plunge | -0.028 |
ar | -0.028 |
less | -0.028 |
Token | and |
Token ID | 290 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
lair | +0.109 |
cks | +0.104 |
necks | +0.098 |
mess | +0.097 |
llers | +0.096 |
Neg logit contributions | |
help | -0.065 |
less | -0.063 |
reward | -0.060 |
iest | -0.059 |
ar | -0.059 |
Token | sizes |
Token ID | 10620 |
Feature activation | +4.90e+00 |
Pos logit contributions | |
lair | +0.013 |
cks | +0.012 |
mess | +0.011 |
pads | +0.011 |
floor | +0.011 |
Neg logit contributions | |
help | -0.020 |
reward | -0.019 |
ar | -0.018 |
plunge | -0.018 |
bye | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
lair | +0.493 |
cks | +0.474 |
llers | +0.451 |
mess | +0.447 |
necks | +0.443 |
Neg logit contributions | |
help | -0.291 |
reward | -0.270 |
less | -0.254 |
ar | -0.253 |
plunge | -0.244 |
Token | They |
Token ID | 1119 |
Feature activation | -5.66e-01 |
Pos logit contributions | |
lair | +0.017 |
pads | +0.017 |
cks | +0.017 |
floor | +0.017 |
mess | +0.016 |
Neg logit contributions | |
reward | -0.017 |
help | -0.017 |
plunge | -0.017 |
bye | -0.016 |
ar | -0.016 |
Token | filled |
Token ID | 5901 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
help | +0.040 |
reward | +0.038 |
ar | +0.036 |
plunge | +0.035 |
f | +0.035 |
Neg logit contributions | |
lair | -0.050 |
cks | -0.048 |
opl | -0.046 |
pads | -0.045 |
mess | -0.045 |
Token | the |
Token ID | 262 |
Feature activation | +3.37e-01 |
Pos logit contributions | |
cks | +0.023 |
mess | +0.022 |
lair | +0.022 |
floor | +0.021 |
necks | +0.021 |
Neg logit contributions | |
reward | -0.016 |
help | -0.016 |
ar | -0.015 |
plunge | -0.015 |
bye | -0.015 |
Token | air |
Token ID | 1633 |
Feature activation | -3.43e-01 |
Pos logit contributions | |
lair | +0.022 |
cks | +0.022 |
pads | +0.020 |
mess | +0.020 |
llers | +0.020 |
Neg logit contributions | |
help | -0.030 |
reward | -0.030 |
ar | -0.029 |
plunge | -0.029 |
bye | -0.028 |
Token | they |
Token ID | 484 |
Feature activation | -5.68e-01 |
Pos logit contributions | |
ar | +0.032 |
reward | +0.031 |
plunge | +0.031 |
help | +0.030 |
bye | +0.029 |
Neg logit contributions | |
lair | -0.051 |
cks | -0.051 |
mess | -0.051 |
floor | -0.050 |
pads | -0.049 |
Token | 're |
Token ID | 821 |
Feature activation | -9.93e-01 |
Pos logit contributions | |
help | +0.036 |
reward | +0.034 |
ar | +0.032 |
plunge | +0.032 |
f | +0.031 |
Neg logit contributions | |
lair | -0.055 |
cks | -0.053 |
pads | -0.051 |
mess | -0.050 |
floor | -0.049 |
Token | different |
Token ID | 1180 |
Feature activation | +3.24e-01 |
Pos logit contributions | |
help | +0.075 |
reward | +0.070 |
plunge | +0.062 |
less | +0.061 |
ar | +0.061 |
Neg logit contributions | |
lair | -0.084 |
pads | -0.080 |
cks | -0.079 |
necks | -0.077 |
llers | -0.074 |
Token | colors |
Token ID | 7577 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
cks | +0.028 |
lair | +0.028 |
pads | +0.026 |
mess | +0.026 |
necks | +0.026 |
Neg logit contributions | |
help | -0.022 |
reward | -0.021 |
ar | -0.020 |
plunge | -0.020 |
less | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
lair | +0.110 |
cks | +0.106 |
necks | +0.100 |
mess | +0.098 |
frown | +0.098 |
Neg logit contributions | |
help | -0.071 |
reward | -0.066 |
less | -0.066 |
iest | -0.064 |
ar | -0.062 |
Token | sizes |
Token ID | 10620 |
Feature activation | +4.89e+00 |
Pos logit contributions | |
lair | +0.008 |
cks | +0.007 |
mess | +0.007 |
pads | +0.007 |
floor | +0.007 |
Neg logit contributions | |
help | -0.010 |
reward | -0.010 |
ar | -0.009 |
plunge | -0.009 |
bye | -0.009 |
Token | ." |
Token ID | 526 |
Feature activation | +4.72e-01 |
Pos logit contributions | |
lair | +0.431 |
cks | +0.417 |
mess | +0.394 |
llers | +0.387 |
pads | +0.387 |
Neg logit contributions | |
help | -0.343 |
reward | -0.330 |
ar | -0.314 |
plunge | -0.308 |
less | -0.305 |
Token |
|
Token ID | 198 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
lair | +0.045 |
cks | +0.044 |
mess | +0.042 |
floor | +0.041 |
pads | +0.041 |
Neg logit contributions | |
help | -0.029 |
reward | -0.028 |
ar | -0.027 |
Sure | -0.027 |
plunge | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
lair | +0.017 |
cks | +0.017 |
mess | +0.016 |
pads | +0.016 |
floor | +0.016 |
Neg logit contributions | |
help | -0.013 |
reward | -0.013 |
ar | -0.012 |
plunge | -0.012 |
Sure | -0.011 |
Token | Max |
Token ID | 11518 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
lair | +0.023 |
cks | +0.022 |
mess | +0.021 |
sizes | +0.021 |
fault | +0.021 |
Neg logit contributions | |
Sure | -0.010 |
help | -0.009 |
reward | -0.009 |
ar | -0.009 |
f | -0.008 |
Token | started |
Token ID | 2067 |
Feature activation | -3.50e-01 |
Pos logit contributions | |
help | +0.012 |
reward | +0.012 |
ar | +0.012 |
Sure | +0.011 |
bye | +0.011 |
Neg logit contributions | |
cks | -0.017 |
lair | -0.017 |
mess | -0.015 |
pads | -0.015 |
opl | -0.015 |
Token | fish |
Token ID | 5916 |
Feature activation | -6.84e-01 |
Pos logit contributions | |
lair | +0.100 |
cks | +0.096 |
pads | +0.094 |
llers | +0.092 |
mess | +0.091 |
Neg logit contributions | |
help | -0.065 |
reward | -0.063 |
plunge | -0.061 |
ar | -0.059 |
ted | -0.055 |
Token | of |
Token ID | 286 |
Feature activation | -8.67e-01 |
Pos logit contributions | |
help | +0.044 |
reward | +0.041 |
ar | +0.041 |
plunge | +0.040 |
less | +0.038 |
Neg logit contributions | |
lair | -0.062 |
llers | -0.060 |
cks | -0.059 |
mess | -0.059 |
floor | -0.058 |
Token | different |
Token ID | 1180 |
Feature activation | +3.20e-01 |
Pos logit contributions | |
reward | +0.053 |
help | +0.053 |
ar | +0.053 |
plunge | +0.051 |
bye | +0.050 |
Neg logit contributions | |
lair | -0.079 |
cks | -0.079 |
mess | -0.077 |
floor | -0.075 |
pads | -0.074 |
Token | colors |
Token ID | 7577 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
lair | +0.020 |
cks | +0.019 |
pads | +0.018 |
mess | +0.018 |
llers | +0.017 |
Neg logit contributions | |
help | -0.029 |
reward | -0.029 |
plunge | -0.029 |
ar | -0.028 |
ted | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | +3.60e-01 |
Pos logit contributions | |
lair | +0.120 |
cks | +0.106 |
llers | +0.104 |
mess | +0.104 |
necks | +0.104 |
Neg logit contributions | |
help | -0.061 |
iest | -0.055 |
reward | -0.055 |
less | -0.054 |
plunge | -0.052 |
Token | sizes |
Token ID | 10620 |
Feature activation | +4.89e+00 |
Pos logit contributions | |
lair | +0.023 |
opl | +0.019 |
cks | +0.019 |
ivering | +0.019 |
iver | +0.018 |
Neg logit contributions | |
help | -0.037 |
reward | -0.034 |
plunge | -0.032 |
Bub | -0.031 |
ar | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | +8.27e-02 |
Pos logit contributions | |
lair | +0.525 |
cks | +0.477 |
llers | +0.476 |
necks | +0.460 |
pads | +0.458 |
Neg logit contributions | |
help | -0.269 |
reward | -0.243 |
less | -0.228 |
plunge | -0.221 |
ar | -0.218 |
Token | They |
Token ID | 1119 |
Feature activation | -5.16e-01 |
Pos logit contributions | |
lair | +0.006 |
cks | +0.006 |
mess | +0.006 |
pads | +0.006 |
floor | +0.006 |
Neg logit contributions | |
help | -0.006 |
reward | -0.006 |
ar | -0.006 |
plunge | -0.006 |
Sure | -0.006 |
Token | swim |
Token ID | 9422 |
Feature activation | -2.78e-01 |
Pos logit contributions | |
help | +0.036 |
plunge | +0.028 |
reward | +0.027 |
Sure | +0.025 |
f | +0.024 |
Neg logit contributions | |
lair | -0.057 |
llers | -0.051 |
opl | -0.051 |
cks | -0.051 |
pads | -0.050 |
Token | and |
Token ID | 290 |
Feature activation | +3.09e-01 |
Pos logit contributions | |
reward | +0.022 |
help | +0.022 |
plunge | +0.021 |
ar | +0.021 |
bye | +0.020 |
Neg logit contributions | |
cks | -0.020 |
lair | -0.020 |
pads | -0.020 |
mess | -0.019 |
necks | -0.019 |
Token | splash |
Token ID | 22870 |
Feature activation | +7.26e-01 |
Pos logit contributions | |
lair | +0.029 |
cks | +0.027 |
llers | +0.027 |
mess | +0.026 |
opl | +0.026 |
Neg logit contributions | |
help | -0.022 |
reward | -0.020 |
plunge | -0.019 |
ar | -0.018 |
f | -0.018 |
Token | truck |
Token ID | 7779 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
cks | +0.013 |
lair | +0.013 |
llers | +0.012 |
mess | +0.012 |
pads | +0.012 |
Neg logit contributions | |
help | -0.020 |
reward | -0.019 |
ar | -0.019 |
plunge | -0.018 |
bye | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +3.68e-01 |
Pos logit contributions | |
help | +0.082 |
reward | +0.078 |
ar | +0.078 |
bye | +0.075 |
plunge | +0.074 |
Neg logit contributions | |
lair | -0.113 |
cks | -0.112 |
mess | -0.109 |
pads | -0.104 |
llers | -0.104 |
Token | They |
Token ID | 1119 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
lair | +0.031 |
cks | +0.031 |
mess | +0.030 |
pads | +0.029 |
floor | +0.028 |
Neg logit contributions | |
help | -0.025 |
reward | -0.025 |
ar | -0.025 |
plunge | -0.024 |
Sure | -0.024 |
Token | were |
Token ID | 547 |
Feature activation | +3.37e-01 |
Pos logit contributions | |
help | +0.025 |
reward | +0.023 |
ar | +0.022 |
plunge | +0.021 |
f | +0.021 |
Neg logit contributions | |
lair | -0.031 |
cks | -0.030 |
mess | -0.029 |
pads | -0.029 |
llers | -0.029 |
Token | different |
Token ID | 1180 |
Feature activation | +4.87e-01 |
Pos logit contributions | |
lair | +0.029 |
cks | +0.027 |
pads | +0.027 |
mess | +0.027 |
floor | +0.027 |
Neg logit contributions | |
help | -0.024 |
reward | -0.022 |
plunge | -0.021 |
ar | -0.021 |
bye | -0.020 |
Token | sizes |
Token ID | 10620 |
Feature activation | +4.88e+00 |
Pos logit contributions | |
cks | +0.039 |
lair | +0.038 |
pads | +0.037 |
mess | +0.036 |
floor | +0.035 |
Neg logit contributions | |
help | -0.037 |
reward | -0.036 |
ar | -0.034 |
less | -0.033 |
plunge | -0.033 |
Token | and |
Token ID | 290 |
Feature activation | +3.90e-01 |
Pos logit contributions | |
lair | +0.482 |
cks | +0.473 |
mess | +0.445 |
llers | +0.439 |
necks | +0.436 |
Neg logit contributions | |
help | -0.291 |
reward | -0.272 |
ar | -0.263 |
less | -0.259 |
plunge | -0.248 |
Token | colors |
Token ID | 7577 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
lair | +0.011 |
cks | +0.010 |
mess | +0.009 |
pads | +0.009 |
floor | +0.008 |
Neg logit contributions | |
help | -0.049 |
reward | -0.049 |
ar | -0.048 |
plunge | -0.048 |
bye | -0.047 |
Token | . |
Token ID | 13 |
Feature activation | +5.86e-01 |
Pos logit contributions | |
lair | +0.116 |
cks | +0.112 |
mess | +0.109 |
necks | +0.106 |
pads | +0.105 |
Neg logit contributions | |
help | -0.077 |
reward | -0.073 |
ar | -0.072 |
less | -0.070 |
plunge | -0.069 |
Token | Some |
Token ID | 2773 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
lair | +0.045 |
cks | +0.044 |
pads | +0.043 |
mess | +0.042 |
floor | +0.042 |
Neg logit contributions | |
reward | -0.045 |
help | -0.045 |
plunge | -0.044 |
ar | -0.043 |
bye | -0.041 |
Token | had |
Token ID | 550 |
Feature activation | +5.62e-01 |
Pos logit contributions | |
reward | +0.071 |
help | +0.069 |
ar | +0.069 |
plunge | +0.069 |
Sure | +0.066 |
Neg logit contributions | |
cks | -0.100 |
lair | -0.100 |
mess | -0.098 |
pads | -0.097 |
floor | -0.096 |
Token | three |
Token ID | 1115 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
lair | +0.021 |
cks | +0.021 |
pads | +0.020 |
mess | +0.020 |
necks | +0.019 |
Neg logit contributions | |
help | -0.021 |
reward | -0.021 |
ar | -0.019 |
plunge | -0.019 |
less | -0.019 |
Token | shapes |
Token ID | 15268 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
help | +0.099 |
ar | +0.096 |
reward | +0.096 |
f | +0.090 |
plunge | +0.089 |
Neg logit contributions | |
lair | -0.076 |
cks | -0.072 |
pads | -0.068 |
llers | -0.066 |
floor | -0.063 |
Token | , |
Token ID | 11 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
lair | +0.273 |
cks | +0.271 |
necks | +0.256 |
mess | +0.256 |
pads | +0.256 |
Neg logit contributions | |
help | -0.131 |
reward | -0.122 |
less | -0.114 |
ar | -0.114 |
plunge | -0.113 |
Token | all |
Token ID | 477 |
Feature activation | -2.84e-01 |
Pos logit contributions | |
help | +0.013 |
reward | +0.012 |
plunge | +0.012 |
ar | +0.012 |
ted | +0.011 |
Neg logit contributions | |
lair | -0.014 |
cks | -0.013 |
mess | -0.012 |
pads | -0.012 |
llers | -0.012 |
Token | different |
Token ID | 1180 |
Feature activation | +2.90e-01 |
Pos logit contributions | |
help | +0.013 |
reward | +0.011 |
less | +0.011 |
plunge | +0.010 |
iest | +0.010 |
Neg logit contributions | |
lair | -0.032 |
cks | -0.031 |
necks | -0.031 |
pads | -0.030 |
llers | -0.030 |
Token | sizes |
Token ID | 10620 |
Feature activation | +4.88e+00 |
Pos logit contributions | |
lair | +0.017 |
cks | +0.017 |
mess | +0.016 |
pads | +0.016 |
floor | +0.015 |
Neg logit contributions | |
help | -0.027 |
reward | -0.027 |
ar | -0.026 |
plunge | -0.026 |
less | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
lair | +0.474 |
cks | +0.464 |
mess | +0.444 |
pads | +0.436 |
floor | +0.432 |
Neg logit contributions | |
help | -0.289 |
reward | -0.280 |
ar | -0.269 |
plunge | -0.263 |
less | -0.257 |
Token |
|
Token ID | 198 |
Feature activation | +2.71e-01 |
Pos logit contributions | |
lair | +0.020 |
cks | +0.019 |
mess | +0.019 |
pads | +0.019 |
floor | +0.018 |
Neg logit contributions | |
help | -0.015 |
reward | -0.015 |
plunge | -0.015 |
ar | -0.015 |
bye | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | +3.17e-01 |
Pos logit contributions | |
lair | +0.024 |
cks | +0.024 |
mess | +0.023 |
pads | +0.023 |
floor | +0.022 |
Neg logit contributions | |
help | -0.018 |
reward | -0.017 |
ar | -0.017 |
plunge | -0.017 |
Sure | -0.016 |
Token | The |
Token ID | 464 |
Feature activation | -5.61e-01 |
Pos logit contributions | |
lair | +0.026 |
cks | +0.026 |
mess | +0.025 |
pads | +0.024 |
floor | +0.024 |
Neg logit contributions | |
help | -0.023 |
reward | -0.023 |
ar | -0.022 |
plunge | -0.022 |
Sure | -0.022 |
Token | shape |
Token ID | 5485 |
Feature activation | +1.85e+00 |
Pos logit contributions | |
ar | +0.042 |
ted | +0.040 |
plunge | +0.040 |
bye | +0.038 |
less | +0.038 |
Neg logit contributions | |
mess | -0.045 |
floor | -0.045 |
lair | -0.044 |
frown | -0.042 |
neck | -0.041 |
Token | prints |
Token ID | 20842 |
Feature activation | -2.79e-01 |
Pos logit contributions | |
reward | +0.027 |
help | +0.027 |
ar | +0.027 |
plunge | +0.026 |
bye | +0.026 |
Neg logit contributions | |
cks | -0.022 |
lair | -0.022 |
mess | -0.021 |
pads | -0.021 |
floor | -0.020 |
Token | in |
Token ID | 287 |
Feature activation | -5.59e-02 |
Pos logit contributions | |
reward | +0.019 |
plunge | +0.019 |
help | +0.019 |
Sure | +0.018 |
ar | +0.018 |
Neg logit contributions | |
lair | -0.023 |
pads | -0.023 |
cks | -0.022 |
mess | -0.022 |
floor | -0.022 |
Token | different |
Token ID | 1180 |
Feature activation | +2.97e-01 |
Pos logit contributions | |
ar | +0.003 |
reward | +0.003 |
plunge | +0.003 |
help | +0.003 |
bye | +0.003 |
Neg logit contributions | |
lair | -0.005 |
cks | -0.005 |
mess | -0.005 |
floor | -0.005 |
necks | -0.005 |
Token | colors |
Token ID | 7577 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
lair | +0.016 |
cks | +0.016 |
mess | +0.015 |
pads | +0.015 |
floor | +0.015 |
Neg logit contributions | |
help | -0.029 |
reward | -0.029 |
ar | -0.028 |
plunge | -0.028 |
bye | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
lair | +0.094 |
cks | +0.092 |
mess | +0.087 |
pads | +0.084 |
necks | +0.084 |
Neg logit contributions | |
help | -0.078 |
reward | -0.077 |
ar | -0.075 |
less | -0.073 |
plunge | -0.072 |
Token | sizes |
Token ID | 10620 |
Feature activation | +4.88e+00 |
Pos logit contributions | |
lair | +0.010 |
cks | +0.010 |
mess | +0.010 |
pads | +0.010 |
floor | +0.010 |
Neg logit contributions | |
reward | -0.018 |
ar | -0.017 |
help | -0.017 |
plunge | -0.017 |
Sure | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
lair | +0.434 |
cks | +0.425 |
mess | +0.407 |
pads | +0.400 |
floor | +0.395 |
Neg logit contributions | |
help | -0.325 |
reward | -0.318 |
ar | -0.307 |
plunge | -0.301 |
less | -0.294 |
Token | |
Token ID | 220 |
Feature activation | -3.73e-01 |
Pos logit contributions | |
cks | +0.018 |
lair | +0.018 |
pads | +0.018 |
floor | +0.017 |
mess | +0.017 |
Neg logit contributions | |
reward | -0.018 |
help | -0.018 |
ar | -0.018 |
plunge | -0.018 |
bye | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | +3.07e-01 |
Pos logit contributions | |
help | +0.025 |
reward | +0.025 |
ar | +0.024 |
plunge | +0.024 |
bye | +0.023 |
Neg logit contributions | |
lair | -0.032 |
cks | -0.032 |
mess | -0.031 |
pads | -0.030 |
floor | -0.030 |
Token |
|
Token ID | 198 |
Feature activation | +3.38e-01 |
Pos logit contributions | |
lair | +0.026 |
cks | +0.026 |
mess | +0.025 |
pads | +0.024 |
floor | +0.024 |
Neg logit contributions | |
reward | -0.021 |
help | -0.021 |
ar | -0.021 |
plunge | -0.020 |
bye | -0.019 |
Token | The |
Token ID | 464 |
Feature activation | -5.32e-01 |
Pos logit contributions | |
lair | +0.026 |
cks | +0.026 |
mess | +0.025 |
pads | +0.025 |
floor | +0.024 |
Neg logit contributions | |
help | -0.026 |
reward | -0.025 |
ar | -0.025 |
plunge | -0.024 |
bye | -0.024 |
Token | cars |
Token ID | 5006 |
Feature activation | -4.67e-01 |
Pos logit contributions | |
lair | +0.073 |
cks | +0.072 |
mess | +0.068 |
pads | +0.067 |
floor | +0.066 |
Neg logit contributions | |
help | -0.091 |
reward | -0.089 |
ar | -0.087 |
plunge | -0.085 |
Sure | -0.082 |
Token | of |
Token ID | 286 |
Feature activation | -9.45e-01 |
Pos logit contributions | |
plunge | +0.031 |
reward | +0.030 |
help | +0.028 |
ar | +0.028 |
Sure | +0.027 |
Neg logit contributions | |
cks | -0.041 |
mess | -0.041 |
pads | -0.040 |
lair | -0.040 |
floor | -0.039 |
Token | different |
Token ID | 1180 |
Feature activation | +3.33e-01 |
Pos logit contributions | |
ar | +0.070 |
reward | +0.070 |
plunge | +0.068 |
bye | +0.068 |
help | +0.066 |
Neg logit contributions | |
cks | -0.074 |
lair | -0.074 |
mess | -0.074 |
floor | -0.071 |
pads | -0.070 |
Token | colors |
Token ID | 7577 |
Feature activation | +9.98e-01 |
Pos logit contributions | |
lair | +0.021 |
cks | +0.021 |
mess | +0.019 |
pads | +0.019 |
floor | +0.018 |
Neg logit contributions | |
help | -0.031 |
reward | -0.030 |
ar | -0.029 |
plunge | -0.029 |
bye | -0.028 |
Token | and |
Token ID | 290 |
Feature activation | +3.62e-01 |
Pos logit contributions | |
lair | +0.093 |
cks | +0.088 |
mess | +0.084 |
necks | +0.082 |
llers | +0.081 |
Neg logit contributions | |
help | -0.064 |
ar | -0.059 |
reward | -0.059 |
less | -0.058 |
iest | -0.056 |
Token | sizes |
Token ID | 10620 |
Feature activation | +4.88e+00 |
Pos logit contributions | |
lair | +0.016 |
cks | +0.013 |
mess | +0.011 |
llers | +0.011 |
opl | +0.010 |
Neg logit contributions | |
help | -0.044 |
reward | -0.041 |
ar | -0.039 |
f | -0.039 |
plunge | -0.038 |
Token | . |
Token ID | 13 |
Feature activation | +5.17e-01 |
Pos logit contributions | |
lair | +0.456 |
cks | +0.434 |
mess | +0.419 |
llers | +0.409 |
pads | +0.405 |
Neg logit contributions | |
help | -0.317 |
reward | -0.300 |
ar | -0.293 |
plunge | -0.276 |
less | -0.276 |
Token | They |
Token ID | 1119 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
lair | +0.042 |
cks | +0.042 |
mess | +0.040 |
pads | +0.040 |
floor | +0.039 |
Neg logit contributions | |
help | -0.037 |
reward | -0.037 |
ar | -0.036 |
plunge | -0.035 |
bye | -0.034 |
Token | make |
Token ID | 787 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
help | +0.026 |
reward | +0.022 |
f | +0.019 |
ar | +0.019 |
plunge | +0.018 |
Neg logit contributions | |
lair | -0.046 |
cks | -0.044 |
llers | -0.043 |
pads | -0.041 |
opl | -0.040 |
Token | a |
Token ID | 257 |
Feature activation | -2.61e-01 |
Pos logit contributions | |
lair | +0.080 |
cks | +0.077 |
pads | +0.075 |
necks | +0.072 |
mess | +0.071 |
Neg logit contributions | |
help | -0.084 |
reward | -0.079 |
plunge | -0.077 |
ar | -0.076 |
f | -0.074 |
Token | line |
Token ID | 1627 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
help | +0.019 |
reward | +0.019 |
ar | +0.017 |
Sure | +0.017 |
plunge | +0.017 |
Neg logit contributions | |
cks | -0.022 |
lair | -0.021 |
pads | -0.020 |
llers | -0.020 |
necks | -0.019 |
Token | there |
Token ID | 612 |
Feature activation | -7.78e-01 |
Pos logit contributions | |
lair | +0.096 |
cks | +0.095 |
necks | +0.091 |
pads | +0.091 |
mess | +0.090 |
Neg logit contributions | |
help | -0.049 |
reward | -0.047 |
ar | -0.046 |
less | -0.045 |
plunge | -0.045 |
Token | was |
Token ID | 373 |
Feature activation | -8.46e-01 |
Pos logit contributions | |
reward | +0.052 |
help | +0.051 |
ar | +0.050 |
plunge | +0.049 |
bye | +0.048 |
Neg logit contributions | |
lair | -0.066 |
cks | -0.066 |
mess | -0.064 |
pads | -0.063 |
floor | -0.063 |
Token | a |
Token ID | 257 |
Feature activation | -9.23e-01 |
Pos logit contributions | |
ar | +0.048 |
reward | +0.048 |
help | +0.045 |
plunge | +0.044 |
Sure | +0.043 |
Neg logit contributions | |
cks | -0.081 |
lair | -0.080 |
pads | -0.079 |
mess | -0.078 |
floor | -0.078 |
Token | three |
Token ID | 1115 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
help | +0.058 |
reward | +0.058 |
ar | +0.058 |
plunge | +0.057 |
ted | +0.055 |
Neg logit contributions | |
lair | -0.081 |
mess | -0.081 |
cks | -0.079 |
floor | -0.078 |
pads | -0.076 |
Token | - |
Token ID | 12 |
Feature activation | -3.91e+00 |
Pos logit contributions | |
help | +0.078 |
reward | +0.076 |
ar | +0.074 |
plunge | +0.069 |
bye | +0.068 |
Neg logit contributions | |
lair | -0.165 |
cks | -0.163 |
pads | -0.158 |
mess | -0.156 |
llers | -0.154 |
Token | year |
Token ID | 1941 |
Feature activation | -7.30e+00 |
Pos logit contributions | |
reward | +0.154 |
help | +0.149 |
ar | +0.143 |
plunge | +0.141 |
bye | +0.127 |
Neg logit contributions | |
lair | -0.449 |
cks | -0.447 |
mess | -0.430 |
pads | -0.428 |
floor | -0.424 |
Token | - |
Token ID | 12 |
Feature activation | -3.86e+00 |
Pos logit contributions | |
help | +0.277 |
reward | +0.272 |
ar | +0.258 |
plunge | +0.249 |
bye | +0.238 |
Neg logit contributions | |
lair | -0.843 |
cks | -0.839 |
mess | -0.816 |
pads | -0.808 |
floor | -0.798 |
Token | old |
Token ID | 727 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
reward | +0.278 |
help | +0.275 |
ar | +0.268 |
plunge | +0.266 |
bye | +0.251 |
Neg logit contributions | |
lair | -0.317 |
cks | -0.313 |
mess | -0.298 |
pads | -0.296 |
floor | -0.292 |
Token | who |
Token ID | 508 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
reward | +0.111 |
plunge | +0.111 |
ar | +0.108 |
help | +0.108 |
less | +0.103 |
Neg logit contributions | |
lair | -0.097 |
cks | -0.092 |
pads | -0.091 |
mess | -0.091 |
neck | -0.090 |
Token | found |
Token ID | 1043 |
Feature activation | -9.24e-01 |
Pos logit contributions | |
help | +0.134 |
reward | +0.133 |
ar | +0.130 |
plunge | +0.127 |
bye | +0.124 |
Neg logit contributions | |
lair | -0.126 |
cks | -0.125 |
mess | -0.120 |
pads | -0.118 |
floor | -0.116 |
Token | a |
Token ID | 257 |
Feature activation | -6.90e-01 |
Pos logit contributions | |
reward | +0.055 |
ar | +0.054 |
plunge | +0.051 |
bye | +0.051 |
help | +0.051 |
Neg logit contributions | |
lair | -0.089 |
cks | -0.087 |
floor | -0.084 |
mess | -0.083 |
neck | -0.082 |
Token | was |
Token ID | 373 |
Feature activation | -4.99e-01 |
Pos logit contributions | |
lair | +0.050 |
cks | +0.047 |
sizes | +0.043 |
web | +0.043 |
pads | +0.043 |
Neg logit contributions | |
help | -0.039 |
bye | -0.037 |
reward | -0.037 |
ted | -0.036 |
ar | -0.036 |
Token | an |
Token ID | 281 |
Feature activation | +9.80e-02 |
Pos logit contributions | |
help | +0.031 |
plunge | +0.028 |
reward | +0.028 |
ar | +0.027 |
ted | +0.026 |
Neg logit contributions | |
lair | -0.045 |
sizes | -0.045 |
cks | -0.044 |
necks | -0.043 |
llers | -0.043 |
Token | average |
Token ID | 2811 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
lair | +0.007 |
cks | +0.007 |
pads | +0.007 |
mess | +0.007 |
floor | +0.006 |
Neg logit contributions | |
help | -0.008 |
reward | -0.008 |
ar | -0.008 |
plunge | -0.008 |
bye | -0.008 |
Token | three |
Token ID | 1115 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
plunge | +0.087 |
ar | +0.086 |
reward | +0.086 |
help | +0.086 |
ted | +0.084 |
Neg logit contributions | |
lair | -0.115 |
floor | -0.111 |
mess | -0.111 |
neck | -0.111 |
pads | -0.107 |
Token | - |
Token ID | 12 |
Feature activation | -3.93e+00 |
Pos logit contributions | |
help | +0.078 |
reward | +0.073 |
ar | +0.073 |
f | +0.068 |
bye | +0.066 |
Neg logit contributions | |
lair | -0.159 |
cks | -0.157 |
pads | -0.150 |
llers | -0.148 |
mess | -0.147 |
Token | year |
Token ID | 1941 |
Feature activation | -7.18e+00 |
Pos logit contributions | |
reward | +0.154 |
help | +0.140 |
ar | +0.136 |
plunge | +0.135 |
bye | +0.116 |
Neg logit contributions | |
lair | -0.458 |
cks | -0.455 |
pads | -0.439 |
mess | -0.435 |
floor | -0.431 |
Token | - |
Token ID | 12 |
Feature activation | -3.89e+00 |
Pos logit contributions | |
help | +0.254 |
reward | +0.246 |
ar | +0.234 |
plunge | +0.219 |
bye | +0.218 |
Neg logit contributions | |
cks | -0.852 |
lair | -0.850 |
mess | -0.823 |
pads | -0.818 |
floor | -0.802 |
Token | old |
Token ID | 727 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
reward | +0.295 |
help | +0.286 |
plunge | +0.281 |
ar | +0.280 |
ted | +0.263 |
Neg logit contributions | |
lair | -0.308 |
cks | -0.301 |
pads | -0.288 |
mess | -0.286 |
floor | -0.281 |
Token | boy |
Token ID | 2933 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
reward | +0.109 |
help | +0.107 |
plunge | +0.107 |
ar | +0.106 |
Sure | +0.101 |
Neg logit contributions | |
lair | -0.104 |
cks | -0.099 |
pads | -0.098 |
mess | -0.097 |
floor | -0.096 |
Token | . |
Token ID | 13 |
Feature activation | -3.00e-01 |
Pos logit contributions | |
help | +0.001 |
reward | +0.001 |
ar | +0.001 |
plunge | +0.001 |
bye | +0.001 |
Neg logit contributions | |
lair | -0.002 |
cks | -0.002 |
mess | -0.001 |
floor | -0.001 |
pads | -0.001 |
Token | He |
Token ID | 679 |
Feature activation | -3.72e-01 |
Pos logit contributions | |
help | +0.021 |
bye | +0.020 |
ar | +0.020 |
reward | +0.020 |
Sure | +0.019 |
Neg logit contributions | |
cks | -0.025 |
mess | -0.025 |
lair | -0.025 |
necks | -0.023 |
web | -0.023 |
Token | saw |
Token ID | 2497 |
Feature activation | -5.60e-01 |
Pos logit contributions | |
cks | +0.003 |
lair | +0.003 |
llers | +0.002 |
sizes | +0.002 |
mess | +0.002 |
Neg logit contributions | |
help | -0.002 |
reward | -0.002 |
ar | -0.002 |
bye | -0.002 |
plunge | -0.002 |
Token | a |
Token ID | 257 |
Feature activation | -8.12e-01 |
Pos logit contributions | |
reward | +0.027 |
help | +0.027 |
ar | +0.026 |
plunge | +0.026 |
bye | +0.024 |
Neg logit contributions | |
lair | -0.058 |
cks | -0.058 |
mess | -0.056 |
pads | -0.055 |
floor | -0.055 |
Token | little |
Token ID | 1310 |
Feature activation | -6.75e-01 |
Pos logit contributions | |
help | +0.061 |
reward | +0.061 |
ar | +0.059 |
plunge | +0.058 |
bye | +0.056 |
Neg logit contributions | |
lair | -0.063 |
cks | -0.063 |
mess | -0.060 |
pads | -0.059 |
floor | -0.058 |
Token | 3 |
Token ID | 513 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
help | +0.042 |
reward | +0.042 |
ar | +0.042 |
plunge | +0.040 |
bye | +0.039 |
Neg logit contributions | |
lair | -0.061 |
mess | -0.059 |
cks | -0.058 |
floor | -0.057 |
pads | -0.056 |
Token | - |
Token ID | 12 |
Feature activation | -3.94e+00 |
Pos logit contributions | |
help | +0.085 |
ar | +0.082 |
reward | +0.079 |
f | +0.076 |
bye | +0.072 |
Neg logit contributions | |
lair | -0.194 |
pads | -0.192 |
cks | -0.190 |
llers | -0.185 |
sizes | -0.180 |
Token | year |
Token ID | 1941 |
Feature activation | -7.14e+00 |
Pos logit contributions | |
reward | +0.165 |
ar | +0.147 |
plunge | +0.141 |
help | +0.138 |
bye | +0.121 |
Neg logit contributions | |
lair | -0.451 |
cks | -0.449 |
mess | -0.418 |
pads | -0.414 |
floor | -0.409 |
Token | - |
Token ID | 12 |
Feature activation | -3.87e+00 |
Pos logit contributions | |
help | +0.263 |
reward | +0.257 |
ar | +0.245 |
plunge | +0.234 |
bye | +0.225 |
Neg logit contributions | |
lair | -0.834 |
cks | -0.830 |
mess | -0.807 |
pads | -0.802 |
floor | -0.791 |
Token | old |
Token ID | 727 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
reward | +0.305 |
help | +0.291 |
plunge | +0.289 |
ar | +0.288 |
less | +0.265 |
Neg logit contributions | |
lair | -0.299 |
cks | -0.291 |
mess | -0.272 |
pads | -0.269 |
floor | -0.264 |
Token | boy |
Token ID | 2933 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
reward | +0.089 |
ar | +0.088 |
plunge | +0.087 |
help | +0.085 |
Sure | +0.083 |
Neg logit contributions | |
lair | -0.108 |
mess | -0.102 |
pads | -0.100 |
cks | -0.100 |
floor | -0.098 |
Token | . |
Token ID | 13 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
lair | +0.011 |
cks | +0.011 |
mess | +0.010 |
pads | +0.010 |
floor | +0.010 |
Neg logit contributions | |
reward | -0.013 |
help | -0.013 |
ar | -0.013 |
plunge | -0.013 |
bye | -0.013 |
Token | He |
Token ID | 679 |
Feature activation | -8.77e-02 |
Pos logit contributions | |
help | +0.008 |
reward | +0.008 |
ar | +0.007 |
plunge | +0.007 |
Sure | +0.007 |
Neg logit contributions | |
cks | -0.010 |
lair | -0.010 |
mess | -0.009 |
pads | -0.009 |
floor | -0.009 |
Token | olly |
Token ID | 5098 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
help | +0.073 |
reward | +0.072 |
plunge | +0.072 |
ar | +0.071 |
ted | +0.068 |
Neg logit contributions | |
lair | -0.091 |
cks | -0.089 |
mess | -0.088 |
floor | -0.086 |
frown | -0.084 |
Token | was |
Token ID | 373 |
Feature activation | -3.88e-01 |
Pos logit contributions | |
help | +0.112 |
reward | +0.110 |
ar | +0.110 |
bye | +0.105 |
ted | +0.104 |
Neg logit contributions | |
lair | -0.130 |
cks | -0.127 |
mess | -0.119 |
pads | -0.117 |
sizes | -0.116 |
Token | a |
Token ID | 257 |
Feature activation | -6.48e-01 |
Pos logit contributions | |
help | +0.024 |
reward | +0.023 |
plunge | +0.022 |
ar | +0.021 |
less | +0.021 |
Neg logit contributions | |
lair | -0.036 |
cks | -0.035 |
sizes | -0.034 |
mess | -0.033 |
necks | -0.033 |
Token | three |
Token ID | 1115 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
help | +0.047 |
reward | +0.046 |
ar | +0.045 |
plunge | +0.045 |
bye | +0.043 |
Neg logit contributions | |
lair | -0.053 |
cks | -0.052 |
mess | -0.050 |
pads | -0.049 |
floor | -0.049 |
Token | - |
Token ID | 12 |
Feature activation | -3.78e+00 |
Pos logit contributions | |
help | +0.065 |
reward | +0.061 |
ar | +0.060 |
f | +0.056 |
bye | +0.054 |
Neg logit contributions | |
lair | -0.143 |
cks | -0.141 |
pads | -0.137 |
llers | -0.135 |
mess | -0.132 |
Token | year |
Token ID | 1941 |
Feature activation | -7.05e+00 |
Pos logit contributions | |
reward | +0.144 |
help | +0.131 |
ar | +0.129 |
plunge | +0.126 |
bye | +0.110 |
Neg logit contributions | |
lair | -0.443 |
cks | -0.443 |
mess | -0.421 |
pads | -0.420 |
floor | -0.416 |
Token | old |
Token ID | 1468 |
Feature activation | -3.80e-01 |
Pos logit contributions | |
help | +0.260 |
reward | +0.256 |
ar | +0.242 |
plunge | +0.232 |
bye | +0.221 |
Neg logit contributions | |
lair | -0.823 |
cks | -0.819 |
mess | -0.796 |
pads | -0.789 |
floor | -0.780 |
Token | girl |
Token ID | 2576 |
Feature activation | -5.66e-01 |
Pos logit contributions | |
help | +0.022 |
reward | +0.021 |
ar | +0.019 |
plunge | +0.019 |
bye | +0.019 |
Neg logit contributions | |
cks | -0.038 |
lair | -0.037 |
pads | -0.036 |
mess | -0.035 |
llers | -0.035 |
Token | who |
Token ID | 508 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
help | +0.038 |
reward | +0.036 |
ar | +0.036 |
plunge | +0.033 |
bye | +0.033 |
Neg logit contributions | |
lair | -0.050 |
cks | -0.050 |
mess | -0.048 |
pads | -0.046 |
llers | -0.046 |
Token | loved |
Token ID | 6151 |
Feature activation | -7.42e-01 |
Pos logit contributions | |
help | +0.118 |
reward | +0.113 |
ar | +0.112 |
bye | +0.106 |
f | +0.106 |
Neg logit contributions | |
lair | -0.133 |
cks | -0.130 |
pads | -0.123 |
mess | -0.123 |
necks | -0.120 |
Token | to |
Token ID | 284 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
help | +0.042 |
reward | +0.038 |
plunge | +0.035 |
ted | +0.034 |
less | +0.033 |
Neg logit contributions | |
cks | -0.078 |
lair | -0.075 |
pads | -0.072 |
mess | -0.070 |
necks | -0.070 |
Token | Joe |
Token ID | 5689 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
plunge | +0.029 |
reward | +0.028 |
iest | +0.027 |
bye | +0.026 |
help | +0.026 |
Neg logit contributions | |
frown | -0.030 |
neck | -0.028 |
lair | -0.028 |
cks | -0.028 |
mess | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | -6.06e-01 |
Pos logit contributions | |
lair | +0.014 |
cks | +0.014 |
mess | +0.014 |
pads | +0.014 |
floor | +0.013 |
Neg logit contributions | |
help | -0.012 |
reward | -0.012 |
ar | -0.012 |
plunge | -0.012 |
bye | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | -7.13e-01 |
Pos logit contributions | |
help | +0.046 |
reward | +0.044 |
ar | +0.042 |
plunge | +0.042 |
Sure | +0.042 |
Neg logit contributions | |
lair | -0.051 |
cks | -0.049 |
mess | -0.046 |
pads | -0.046 |
floor | -0.045 |
Token | 3 |
Token ID | 513 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
help | +0.052 |
ar | +0.052 |
reward | +0.052 |
plunge | +0.051 |
ted | +0.051 |
Neg logit contributions | |
lair | -0.055 |
mess | -0.054 |
cks | -0.052 |
floor | -0.052 |
pads | -0.050 |
Token | - |
Token ID | 12 |
Feature activation | -3.80e+00 |
Pos logit contributions | |
help | +0.090 |
reward | +0.087 |
ar | +0.086 |
plunge | +0.081 |
bye | +0.079 |
Neg logit contributions | |
lair | -0.162 |
cks | -0.158 |
pads | -0.155 |
mess | -0.153 |
llers | -0.150 |
Token | year |
Token ID | 1941 |
Feature activation | -7.02e+00 |
Pos logit contributions | |
reward | +0.182 |
plunge | +0.153 |
ar | +0.150 |
help | +0.149 |
less | +0.128 |
Neg logit contributions | |
cks | -0.410 |
lair | -0.390 |
pads | -0.384 |
opl | -0.366 |
cking | -0.364 |
Token | - |
Token ID | 12 |
Feature activation | -3.79e+00 |
Pos logit contributions | |
help | +0.268 |
reward | +0.266 |
ar | +0.250 |
plunge | +0.247 |
Sure | +0.228 |
Neg logit contributions | |
lair | -0.808 |
cks | -0.801 |
mess | -0.781 |
pads | -0.770 |
floor | -0.766 |
Token | old |
Token ID | 727 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
reward | +0.327 |
plunge | +0.310 |
help | +0.304 |
ar | +0.298 |
ted | +0.285 |
Neg logit contributions | |
cks | -0.253 |
lair | -0.250 |
pads | -0.236 |
opl | -0.226 |
necks | -0.219 |
Token | boy |
Token ID | 2933 |
Feature activation | +2.38e-01 |
Pos logit contributions | |
reward | +0.083 |
plunge | +0.083 |
ar | +0.082 |
help | +0.079 |
Sure | +0.078 |
Neg logit contributions | |
lair | -0.097 |
mess | -0.093 |
cks | -0.091 |
pads | -0.090 |
neck | -0.090 |
Token | . |
Token ID | 13 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
lair | +0.019 |
cks | +0.019 |
mess | +0.018 |
pads | +0.018 |
floor | +0.017 |
Neg logit contributions | |
reward | -0.018 |
help | -0.017 |
plunge | -0.017 |
ar | -0.017 |
bye | -0.016 |
Token | He |
Token ID | 679 |
Feature activation | -3.03e-01 |
Pos logit contributions | |
help | +0.018 |
reward | +0.018 |
ar | +0.017 |
plunge | +0.017 |
bye | +0.016 |
Neg logit contributions | |
lair | -0.021 |
cks | -0.021 |
mess | -0.020 |
pads | -0.020 |
floor | -0.019 |
Token | Pat |
Token ID | 12130 |
Feature activation | -5.72e-01 |
Pos logit contributions | |
cks | +0.023 |
lair | +0.023 |
pads | +0.023 |
floor | +0.022 |
mess | +0.022 |
Neg logit contributions | |
reward | -0.020 |
help | -0.020 |
plunge | -0.019 |
ar | -0.019 |
bye | -0.018 |
Token | was |
Token ID | 373 |
Feature activation | -3.04e-01 |
Pos logit contributions | |
help | +0.035 |
ar | +0.035 |
reward | +0.034 |
plunge | +0.033 |
ted | +0.033 |
Neg logit contributions | |
lair | -0.053 |
cks | -0.052 |
mess | -0.050 |
floor | -0.049 |
pads | -0.048 |
Token | a |
Token ID | 257 |
Feature activation | -6.56e-01 |
Pos logit contributions | |
help | +0.020 |
reward | +0.018 |
plunge | +0.018 |
bye | +0.017 |
Sure | +0.017 |
Neg logit contributions | |
lair | -0.027 |
cks | -0.027 |
sizes | -0.026 |
llers | -0.026 |
necks | -0.026 |
Token | 3 |
Token ID | 513 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
help | +0.045 |
reward | +0.045 |
ar | +0.044 |
plunge | +0.043 |
bye | +0.042 |
Neg logit contributions | |
lair | -0.055 |
cks | -0.055 |
mess | -0.053 |
pads | -0.052 |
floor | -0.051 |
Token | - |
Token ID | 12 |
Feature activation | -3.74e+00 |
Pos logit contributions | |
help | +0.088 |
ar | +0.083 |
reward | +0.083 |
f | +0.077 |
bye | +0.077 |
Neg logit contributions | |
lair | -0.173 |
cks | -0.170 |
pads | -0.166 |
mess | -0.163 |
llers | -0.162 |
Token | year |
Token ID | 1941 |
Feature activation | -7.01e+00 |
Pos logit contributions | |
reward | +0.176 |
ar | +0.160 |
help | +0.157 |
plunge | +0.156 |
bye | +0.138 |
Neg logit contributions | |
cks | -0.406 |
lair | -0.405 |
pads | -0.383 |
mess | -0.380 |
floor | -0.374 |
Token | - |
Token ID | 12 |
Feature activation | -3.74e+00 |
Pos logit contributions | |
help | +0.267 |
reward | +0.264 |
ar | +0.249 |
plunge | +0.242 |
bye | +0.226 |
Neg logit contributions | |
lair | -0.809 |
cks | -0.803 |
mess | -0.782 |
pads | -0.774 |
floor | -0.766 |
Token | old |
Token ID | 727 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
reward | +0.290 |
help | +0.279 |
plunge | +0.277 |
ar | +0.277 |
ted | +0.258 |
Neg logit contributions | |
lair | -0.288 |
cks | -0.285 |
pads | -0.268 |
mess | -0.266 |
floor | -0.259 |
Token | boy |
Token ID | 2933 |
Feature activation | +8.12e-02 |
Pos logit contributions | |
reward | +0.105 |
plunge | +0.103 |
ar | +0.102 |
help | +0.102 |
Sure | +0.096 |
Neg logit contributions | |
lair | -0.101 |
cks | -0.095 |
pads | -0.095 |
mess | -0.095 |
floor | -0.094 |
Token | . |
Token ID | 13 |
Feature activation | -3.48e-01 |
Pos logit contributions | |
lair | +0.006 |
cks | +0.006 |
mess | +0.006 |
pads | +0.006 |
floor | +0.005 |
Neg logit contributions | |
help | -0.006 |
reward | -0.006 |
ar | -0.006 |
plunge | -0.006 |
bye | -0.006 |
Token | He |
Token ID | 679 |
Feature activation | -3.91e-01 |
Pos logit contributions | |
help | +0.023 |
reward | +0.022 |
bye | +0.021 |
ar | +0.021 |
Sure | +0.021 |
Neg logit contributions | |
cks | -0.032 |
mess | -0.032 |
lair | -0.031 |
irling | -0.029 |
cking | -0.029 |
Token | a |
Token ID | 257 |
Feature activation | -6.96e-01 |
Pos logit contributions | |
help | +0.018 |
reward | +0.017 |
plunge | +0.017 |
ar | +0.016 |
bye | +0.016 |
Neg logit contributions | |
lair | -0.028 |
cks | -0.028 |
mess | -0.027 |
pads | -0.026 |
necks | -0.026 |
Token | very |
Token ID | 845 |
Feature activation | -5.03e-01 |
Pos logit contributions | |
help | +0.051 |
reward | +0.051 |
ar | +0.050 |
plunge | +0.049 |
ted | +0.047 |
Neg logit contributions | |
lair | -0.055 |
cks | -0.054 |
mess | -0.053 |
pads | -0.051 |
floor | -0.051 |
Token | adventurous |
Token ID | 39770 |
Feature activation | -7.94e-01 |
Pos logit contributions | |
reward | +0.046 |
ar | +0.046 |
plunge | +0.046 |
help | +0.046 |
ted | +0.045 |
Neg logit contributions | |
lair | -0.029 |
mess | -0.029 |
cks | -0.028 |
floor | -0.028 |
frown | -0.027 |
Token | three |
Token ID | 1115 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
plunge | +0.054 |
ar | +0.053 |
help | +0.053 |
reward | +0.053 |
ted | +0.052 |
Neg logit contributions | |
lair | -0.064 |
mess | -0.064 |
floor | -0.063 |
neck | -0.060 |
frown | -0.059 |
Token | - |
Token ID | 12 |
Feature activation | -3.73e+00 |
Pos logit contributions | |
help | +0.085 |
reward | +0.083 |
ar | +0.081 |
plunge | +0.076 |
bye | +0.075 |
Neg logit contributions | |
lair | -0.134 |
cks | -0.132 |
pads | -0.126 |
mess | -0.126 |
llers | -0.123 |
Token | year |
Token ID | 1941 |
Feature activation | -6.96e+00 |
Pos logit contributions | |
reward | +0.223 |
help | +0.200 |
plunge | +0.199 |
ar | +0.192 |
ted | +0.180 |
Neg logit contributions | |
lair | -0.361 |
cks | -0.358 |
pads | -0.343 |
opl | -0.331 |
floor | -0.326 |
Token | - |
Token ID | 12 |
Feature activation | -3.74e+00 |
Pos logit contributions | |
help | +0.276 |
reward | +0.274 |
ar | +0.258 |
plunge | +0.257 |
Sure | +0.239 |
Neg logit contributions | |
lair | -0.792 |
cks | -0.781 |
mess | -0.764 |
pads | -0.755 |
floor | -0.751 |
Token | old |
Token ID | 727 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
reward | +0.310 |
help | +0.296 |
plunge | +0.295 |
ar | +0.288 |
ted | +0.277 |
Neg logit contributions | |
lair | -0.271 |
cks | -0.266 |
pads | -0.254 |
mess | -0.242 |
floor | -0.241 |
Token | . |
Token ID | 13 |
Feature activation | -4.33e-01 |
Pos logit contributions | |
reward | +0.102 |
plunge | +0.102 |
help | +0.100 |
ar | +0.099 |
Sure | +0.095 |
Neg logit contributions | |
lair | -0.095 |
pads | -0.091 |
cks | -0.088 |
mess | -0.087 |
neck | -0.087 |
Token | She |
Token ID | 1375 |
Feature activation | +2.79e-01 |
Pos logit contributions | |
bye | +0.027 |
help | +0.027 |
reward | +0.026 |
Enjoy | +0.025 |
Sure | +0.025 |
Neg logit contributions | |
cks | -0.043 |
mess | -0.040 |
cking | -0.039 |
lair | -0.038 |
irling | -0.037 |
Token | was |
Token ID | 373 |
Feature activation | -2.17e-01 |
Pos logit contributions | |
cks | +0.026 |
lair | +0.026 |
llers | +0.024 |
mess | +0.024 |
necks | +0.023 |
Neg logit contributions | |
help | -0.018 |
ar | -0.017 |
reward | -0.016 |
bye | -0.016 |
f | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -3.17e-01 |
Pos logit contributions | |
bye | +0.004 |
help | +0.004 |
ar | +0.004 |
reward | +0.003 |
less | +0.003 |
Neg logit contributions | |
lair | -0.008 |
cks | -0.008 |
mess | -0.008 |
frown | -0.008 |
llers | -0.008 |
Token | Two |
Token ID | 4930 |
Feature activation | -2.98e-01 |
Pos logit contributions | |
help | +0.023 |
reward | +0.023 |
ar | +0.023 |
plunge | +0.023 |
less | +0.022 |
Neg logit contributions | |
lair | -0.025 |
cks | -0.024 |
floor | -0.023 |
pads | -0.023 |
llers | -0.023 |
Token | curious |
Token ID | 11040 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
help | +0.019 |
reward | +0.019 |
ar | +0.018 |
plunge | +0.018 |
bye | +0.017 |
Neg logit contributions | |
lair | -0.027 |
cks | -0.027 |
mess | -0.026 |
pads | -0.025 |
floor | -0.025 |
Token | three |
Token ID | 1115 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
lair | +0.072 |
floor | +0.068 |
mess | +0.067 |
pads | +0.065 |
cks | +0.065 |
Neg logit contributions | |
ar | -0.089 |
reward | -0.088 |
help | -0.087 |
plunge | -0.086 |
Sure | -0.086 |
Token | - |
Token ID | 12 |
Feature activation | -3.70e+00 |
Pos logit contributions | |
help | +0.064 |
ar | +0.062 |
reward | +0.059 |
f | +0.056 |
bye | +0.053 |
Neg logit contributions | |
lair | -0.137 |
pads | -0.130 |
cks | -0.129 |
mess | -0.126 |
floor | -0.124 |
Token | year |
Token ID | 1941 |
Feature activation | -6.95e+00 |
Pos logit contributions | |
reward | +0.235 |
plunge | +0.218 |
help | +0.210 |
ar | +0.205 |
less | +0.196 |
Neg logit contributions | |
cks | -0.347 |
lair | -0.342 |
pads | -0.328 |
sizes | -0.319 |
opl | -0.314 |
Token | - |
Token ID | 12 |
Feature activation | -3.71e+00 |
Pos logit contributions | |
help | +0.287 |
reward | +0.286 |
ar | +0.269 |
plunge | +0.268 |
Sure | +0.252 |
Neg logit contributions | |
lair | -0.780 |
cks | -0.772 |
mess | -0.750 |
pads | -0.742 |
floor | -0.737 |
Token | old |
Token ID | 727 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
reward | +0.302 |
plunge | +0.288 |
help | +0.288 |
ar | +0.281 |
less | +0.270 |
Neg logit contributions | |
lair | -0.275 |
cks | -0.273 |
pads | -0.258 |
llers | -0.248 |
mess | -0.247 |
Token | children |
Token ID | 1751 |
Feature activation | -8.83e-01 |
Pos logit contributions | |
reward | +0.100 |
plunge | +0.100 |
help | +0.095 |
ar | +0.095 |
Sure | +0.093 |
Neg logit contributions | |
lair | -0.093 |
pads | -0.092 |
cks | -0.088 |
sizes | -0.088 |
necks | -0.087 |
Token | , |
Token ID | 11 |
Feature activation | -6.16e-01 |
Pos logit contributions | |
reward | +0.050 |
help | +0.048 |
ar | +0.047 |
plunge | +0.046 |
Sure | +0.045 |
Neg logit contributions | |
cks | -0.086 |
lair | -0.086 |
mess | -0.082 |
pads | -0.082 |
floor | -0.080 |
Token | Bob |
Token ID | 5811 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
help | +0.040 |
reward | +0.039 |
ar | +0.038 |
plunge | +0.037 |
bye | +0.036 |
Neg logit contributions | |
lair | -0.055 |
cks | -0.054 |
mess | -0.053 |
pads | -0.052 |
floor | -0.051 |
Token | there |
Token ID | 612 |
Feature activation | -4.20e-01 |
Pos logit contributions | |
lair | +0.134 |
cks | +0.133 |
mess | +0.128 |
pads | +0.127 |
floor | +0.125 |
Neg logit contributions | |
help | -0.086 |
reward | -0.084 |
ar | -0.082 |
plunge | -0.081 |
less | -0.077 |
Token | was |
Token ID | 373 |
Feature activation | -4.43e-01 |
Pos logit contributions | |
reward | +0.033 |
help | +0.033 |
ar | +0.032 |
plunge | +0.032 |
bye | +0.031 |
Neg logit contributions | |
lair | -0.029 |
cks | -0.029 |
mess | -0.029 |
floor | -0.028 |
frown | -0.028 |
Token | a |
Token ID | 257 |
Feature activation | -7.14e-01 |
Pos logit contributions | |
reward | +0.023 |
help | +0.023 |
ar | +0.022 |
plunge | +0.021 |
Sure | +0.020 |
Neg logit contributions | |
lair | -0.045 |
cks | -0.045 |
mess | -0.043 |
pads | -0.043 |
floor | -0.043 |
Token | 4 |
Token ID | 604 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
help | +0.043 |
reward | +0.042 |
ar | +0.041 |
plunge | +0.041 |
bye | +0.039 |
Neg logit contributions | |
lair | -0.066 |
cks | -0.065 |
mess | -0.064 |
pads | -0.062 |
floor | -0.062 |
Token | - |
Token ID | 12 |
Feature activation | -3.65e+00 |
Pos logit contributions | |
reward | +0.093 |
help | +0.093 |
plunge | +0.093 |
ar | +0.087 |
Sure | +0.086 |
Neg logit contributions | |
cks | -0.172 |
lair | -0.171 |
mess | -0.168 |
floor | -0.167 |
necks | -0.164 |
Token | year |
Token ID | 1941 |
Feature activation | -6.95e+00 |
Pos logit contributions | |
reward | +0.271 |
help | +0.261 |
plunge | +0.258 |
ar | +0.255 |
ted | +0.243 |
Neg logit contributions | |
cks | -0.289 |
lair | -0.286 |
pads | -0.273 |
mess | -0.267 |
floor | -0.267 |
Token | old |
Token ID | 1468 |
Feature activation | -3.55e-01 |
Pos logit contributions | |
help | +0.281 |
reward | +0.278 |
ar | +0.263 |
plunge | +0.261 |
Sure | +0.244 |
Neg logit contributions | |
lair | -0.781 |
cks | -0.772 |
mess | -0.756 |
pads | -0.744 |
floor | -0.742 |
Token | girl |
Token ID | 2576 |
Feature activation | -5.83e-01 |
Pos logit contributions | |
help | +0.025 |
reward | +0.025 |
ar | +0.024 |
plunge | +0.023 |
bye | +0.023 |
Neg logit contributions | |
cks | -0.030 |
lair | -0.030 |
pads | -0.028 |
mess | -0.028 |
llers | -0.027 |
Token | named |
Token ID | 3706 |
Feature activation | -2.93e+00 |
Pos logit contributions | |
help | +0.031 |
reward | +0.030 |
ar | +0.029 |
plunge | +0.028 |
bye | +0.028 |
Neg logit contributions | |
lair | -0.059 |
cks | -0.058 |
mess | -0.057 |
pads | -0.056 |
floor | -0.055 |
Token | Sus |
Token ID | 8932 |
Feature activation | -6.82e-01 |
Pos logit contributions | |
reward | +0.184 |
help | +0.184 |
ar | +0.178 |
plunge | +0.175 |
bye | +0.168 |
Neg logit contributions | |
lair | -0.265 |
cks | -0.263 |
mess | -0.256 |
pads | -0.252 |
floor | -0.250 |
Token | ie |
Token ID | 494 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
help | +0.027 |
f | +0.025 |
less | +0.025 |
ted | +0.024 |
reward | +0.024 |
Neg logit contributions | |
lair | -0.078 |
cks | -0.075 |
pads | -0.072 |
llers | -0.072 |
necks | -0.071 |
Token | buried |
Token ID | 11694 |
Feature activation | -9.62e-01 |
Pos logit contributions | |
help | +0.005 |
reward | +0.005 |
plunge | +0.005 |
ar | +0.005 |
less | +0.005 |
Neg logit contributions | |
lair | -0.006 |
cks | -0.006 |
necks | -0.006 |
pads | -0.006 |
mess | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | -3.53e-01 |
Pos logit contributions | |
reward | +0.063 |
help | +0.061 |
bye | +0.060 |
Sure | +0.059 |
plunge | +0.059 |
Neg logit contributions | |
mess | -0.084 |
cks | -0.083 |
lair | -0.083 |
floor | -0.082 |
pads | -0.079 |
Token | the |
Token ID | 262 |
Feature activation | +4.05e-02 |
Pos logit contributions | |
ar | +0.022 |
help | +0.022 |
reward | +0.022 |
bye | +0.021 |
plunge | +0.021 |
Neg logit contributions | |
lair | -0.031 |
mess | -0.030 |
floor | -0.030 |
neck | -0.029 |
cks | -0.029 |
Token | 3 |
Token ID | 513 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
lair | +0.003 |
cks | +0.003 |
mess | +0.003 |
pads | +0.003 |
floor | +0.003 |
Neg logit contributions | |
help | -0.003 |
reward | -0.003 |
ar | -0.003 |
plunge | -0.003 |
bye | -0.003 |
Token | - |
Token ID | 12 |
Feature activation | -3.80e+00 |
Pos logit contributions | |
help | +0.080 |
ar | +0.078 |
reward | +0.078 |
plunge | +0.072 |
bye | +0.072 |
Neg logit contributions | |
lair | -0.158 |
cks | -0.156 |
pads | -0.151 |
mess | -0.149 |
llers | -0.146 |
Token | year |
Token ID | 1941 |
Feature activation | -6.94e+00 |
Pos logit contributions | |
reward | +0.205 |
help | +0.193 |
plunge | +0.186 |
ar | +0.178 |
ted | +0.159 |
Neg logit contributions | |
cks | -0.381 |
lair | -0.380 |
pads | -0.367 |
mess | -0.353 |
floor | -0.352 |
Token | - |
Token ID | 12 |
Feature activation | -3.79e+00 |
Pos logit contributions | |
help | +0.263 |
reward | +0.259 |
ar | +0.247 |
plunge | +0.237 |
bye | +0.227 |
Neg logit contributions | |
lair | -0.803 |
cks | -0.799 |
mess | -0.776 |
pads | -0.769 |
floor | -0.759 |
Token | old |
Token ID | 727 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
reward | +0.308 |
help | +0.298 |
plunge | +0.293 |
ar | +0.283 |
ted | +0.271 |
Neg logit contributions | |
lair | -0.278 |
cks | -0.273 |
pads | -0.262 |
mess | -0.251 |
llers | -0.249 |
Token | and |
Token ID | 290 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
reward | +0.089 |
help | +0.089 |
plunge | +0.089 |
ar | +0.086 |
less | +0.083 |
Neg logit contributions | |
lair | -0.097 |
mess | -0.093 |
pads | -0.092 |
floor | -0.092 |
cks | -0.091 |
Token | the |
Token ID | 262 |
Feature activation | +3.74e-02 |
Pos logit contributions | |
help | +0.007 |
reward | +0.006 |
ar | +0.006 |
plunge | +0.006 |
bye | +0.006 |
Neg logit contributions | |
lair | -0.009 |
cks | -0.009 |
mess | -0.009 |
pads | -0.009 |
floor | -0.009 |
Token | parent |
Token ID | 2560 |
Feature activation | -3.72e-01 |
Pos logit contributions | |
lair | +0.003 |
mess | +0.003 |
floor | +0.003 |
pads | +0.003 |
cks | +0.003 |
Neg logit contributions | |
ar | -0.003 |
reward | -0.003 |
help | -0.003 |
plunge | -0.002 |
bye | -0.002 |
Token | was |
Token ID | 373 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
cks | +0.044 |
lair | +0.044 |
mess | +0.040 |
pads | +0.039 |
opl | +0.039 |
Neg logit contributions | |
help | -0.031 |
ar | -0.030 |
reward | -0.030 |
bye | -0.029 |
Sure | -0.029 |
Token | an |
Token ID | 281 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
help | +0.010 |
reward | +0.010 |
plunge | +0.010 |
ar | +0.009 |
Sure | +0.009 |
Neg logit contributions | |
lair | -0.015 |
cks | -0.015 |
mess | -0.014 |
pads | -0.014 |
necks | -0.014 |
Token | attractive |
Token ID | 10966 |
Feature activation | -7.11e-01 |
Pos logit contributions | |
lair | +0.010 |
cks | +0.010 |
mess | +0.010 |
pads | +0.009 |
floor | +0.009 |
Neg logit contributions | |
help | -0.013 |
reward | -0.013 |
ar | -0.013 |
plunge | -0.013 |
bye | -0.012 |
Token | three |
Token ID | 1115 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
ar | +0.052 |
help | +0.051 |
plunge | +0.051 |
reward | +0.050 |
ted | +0.050 |
Neg logit contributions | |
lair | -0.056 |
mess | -0.055 |
floor | -0.053 |
neck | -0.051 |
frown | -0.051 |
Token | - |
Token ID | 12 |
Feature activation | -3.79e+00 |
Pos logit contributions | |
help | +0.054 |
reward | +0.050 |
ar | +0.046 |
f | +0.045 |
iest | +0.044 |
Neg logit contributions | |
lair | -0.150 |
cks | -0.149 |
llers | -0.144 |
pads | -0.143 |
ric | -0.136 |
Token | year |
Token ID | 1941 |
Feature activation | -6.94e+00 |
Pos logit contributions | |
reward | +0.155 |
plunge | +0.145 |
help | +0.145 |
ar | +0.137 |
ted | +0.119 |
Neg logit contributions | |
cks | -0.432 |
lair | -0.428 |
pads | -0.413 |
mess | -0.407 |
floor | -0.403 |
Token | - |
Token ID | 12 |
Feature activation | -3.78e+00 |
Pos logit contributions | |
help | +0.230 |
reward | +0.224 |
ar | +0.211 |
bye | +0.201 |
plunge | +0.190 |
Neg logit contributions | |
cks | -0.842 |
lair | -0.835 |
mess | -0.809 |
pads | -0.809 |
llers | -0.791 |
Token | old |
Token ID | 727 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
reward | +0.295 |
help | +0.288 |
plunge | +0.288 |
ar | +0.280 |
ted | +0.265 |
Neg logit contributions | |
lair | -0.285 |
cks | -0.284 |
pads | -0.271 |
mess | -0.265 |
floor | -0.261 |
Token | who |
Token ID | 508 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
reward | +0.078 |
help | +0.077 |
ar | +0.076 |
plunge | +0.076 |
Sure | +0.071 |
Neg logit contributions | |
lair | -0.103 |
cks | -0.100 |
mess | -0.099 |
pads | -0.098 |
floor | -0.097 |
Token | loved |
Token ID | 6151 |
Feature activation | -6.38e-01 |
Pos logit contributions | |
reward | +0.122 |
help | +0.122 |
ar | +0.118 |
plunge | +0.118 |
bye | +0.113 |
Neg logit contributions | |
lair | -0.103 |
cks | -0.102 |
mess | -0.098 |
pads | -0.096 |
floor | -0.095 |
Token | reading |
Token ID | 3555 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
help | +0.032 |
reward | +0.029 |
less | +0.026 |
plunge | +0.026 |
ted | +0.026 |
Neg logit contributions | |
cks | -0.070 |
lair | -0.067 |
pads | -0.065 |
mess | -0.063 |
necks | -0.063 |
Token | there |
Token ID | 612 |
Feature activation | -4.10e-01 |
Pos logit contributions | |
reward | +0.059 |
help | +0.056 |
ar | +0.055 |
plunge | +0.054 |
bye | +0.054 |
Neg logit contributions | |
cks | -0.072 |
lair | -0.069 |
mess | -0.068 |
pads | -0.067 |
floor | -0.066 |
Token | was |
Token ID | 373 |
Feature activation | -4.26e-01 |
Pos logit contributions | |
help | +0.028 |
reward | +0.028 |
ar | +0.027 |
plunge | +0.027 |
bye | +0.026 |
Neg logit contributions | |
lair | -0.035 |
cks | -0.034 |
mess | -0.033 |
pads | -0.032 |
floor | -0.032 |
Token | a |
Token ID | 257 |
Feature activation | -8.35e-01 |
Pos logit contributions | |
ar | +0.029 |
reward | +0.028 |
plunge | +0.027 |
help | +0.026 |
less | +0.025 |
Neg logit contributions | |
lair | -0.036 |
cks | -0.036 |
floor | -0.036 |
pads | -0.036 |
mess | -0.036 |
Token | three |
Token ID | 1115 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
plunge | +0.060 |
ar | +0.059 |
ted | +0.058 |
help | +0.058 |
reward | +0.058 |
Neg logit contributions | |
mess | -0.068 |
lair | -0.066 |
floor | -0.064 |
frown | -0.063 |
neck | -0.063 |
Token | - |
Token ID | 12 |
Feature activation | -3.65e+00 |
Pos logit contributions | |
help | +0.068 |
reward | +0.066 |
ar | +0.064 |
plunge | +0.058 |
bye | +0.058 |
Neg logit contributions | |
lair | -0.147 |
cks | -0.145 |
pads | -0.140 |
mess | -0.138 |
llers | -0.137 |
Token | year |
Token ID | 1941 |
Feature activation | -6.94e+00 |
Pos logit contributions | |
reward | +0.145 |
plunge | +0.129 |
ar | +0.128 |
help | +0.125 |
less | +0.108 |
Neg logit contributions | |
lair | -0.425 |
cks | -0.423 |
pads | -0.400 |
mess | -0.400 |
floor | -0.398 |
Token | - |
Token ID | 12 |
Feature activation | -3.65e+00 |
Pos logit contributions | |
help | +0.268 |
reward | +0.265 |
ar | +0.250 |
plunge | +0.243 |
bye | +0.228 |
Neg logit contributions | |
lair | -0.797 |
cks | -0.791 |
mess | -0.770 |
pads | -0.762 |
floor | -0.754 |
Token | old |
Token ID | 727 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
reward | +0.275 |
help | +0.265 |
plunge | +0.264 |
ar | +0.260 |
ted | +0.245 |
Neg logit contributions | |
lair | -0.292 |
cks | -0.288 |
pads | -0.272 |
mess | -0.272 |
floor | -0.267 |
Token | boy |
Token ID | 2933 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
plunge | +0.105 |
reward | +0.099 |
ar | +0.097 |
less | +0.094 |
help | +0.093 |
Neg logit contributions | |
lair | -0.104 |
neck | -0.101 |
mess | -0.098 |
pads | -0.097 |
cks | -0.096 |
Token | called |
Token ID | 1444 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
lair | +0.015 |
cks | +0.015 |
mess | +0.015 |
pads | +0.014 |
floor | +0.014 |
Neg logit contributions | |
reward | -0.010 |
plunge | -0.009 |
help | -0.009 |
ar | -0.009 |
bye | -0.009 |
Token | Tommy |
Token ID | 19919 |
Feature activation | -6.19e-01 |
Pos logit contributions | |
ar | +0.098 |
ted | +0.097 |
Sure | +0.097 |
bye | +0.095 |
reward | +0.094 |
Neg logit contributions | |
lair | -0.082 |
floor | -0.077 |
frown | -0.074 |
cks | -0.073 |
neck | -0.072 |
Token | and |
Token ID | 290 |
Feature activation | -2.95e-01 |
Pos logit contributions | |
lair | +0.082 |
cks | +0.082 |
mess | +0.079 |
pads | +0.078 |
floor | +0.077 |
Neg logit contributions | |
reward | -0.054 |
help | -0.054 |
ar | -0.052 |
plunge | -0.052 |
Sure | -0.049 |
Token | their |
Token ID | 511 |
Feature activation | -5.90e-01 |
Pos logit contributions | |
ar | +0.020 |
plunge | +0.019 |
reward | +0.019 |
less | +0.018 |
bye | +0.018 |
Neg logit contributions | |
cks | -0.025 |
lair | -0.025 |
mess | -0.024 |
frown | -0.024 |
neck | -0.024 |
Token | little |
Token ID | 1310 |
Feature activation | -7.11e-01 |
Pos logit contributions | |
ar | +0.056 |
ted | +0.048 |
f | +0.046 |
plunge | +0.046 |
less | +0.046 |
Neg logit contributions | |
mess | -0.041 |
floor | -0.040 |
lair | -0.040 |
frown | -0.038 |
neck | -0.036 |
Token | three |
Token ID | 1115 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
ar | +0.061 |
reward | +0.057 |
plunge | +0.055 |
ted | +0.054 |
Sure | +0.054 |
Neg logit contributions | |
mess | -0.049 |
lair | -0.049 |
floor | -0.047 |
frown | -0.047 |
neck | -0.047 |
Token | - |
Token ID | 12 |
Feature activation | -3.61e+00 |
Pos logit contributions | |
help | +0.068 |
reward | +0.066 |
ar | +0.064 |
plunge | +0.060 |
bye | +0.059 |
Neg logit contributions | |
lair | -0.141 |
cks | -0.138 |
mess | -0.133 |
pads | -0.133 |
floor | -0.130 |
Token | year |
Token ID | 1941 |
Feature activation | -6.94e+00 |
Pos logit contributions | |
reward | +0.198 |
ar | +0.176 |
plunge | +0.174 |
help | +0.151 |
less | +0.136 |
Neg logit contributions | |
cks | -0.376 |
lair | -0.365 |
pads | -0.348 |
sizes | -0.345 |
floor | -0.341 |
Token | - |
Token ID | 12 |
Feature activation | -3.62e+00 |
Pos logit contributions | |
reward | +0.305 |
help | +0.300 |
plunge | +0.295 |
ar | +0.289 |
Sure | +0.282 |
Neg logit contributions | |
lair | -0.757 |
cks | -0.745 |
mess | -0.731 |
floor | -0.720 |
pads | -0.718 |
Token | old |
Token ID | 727 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
reward | +0.311 |
plunge | +0.298 |
ar | +0.291 |
help | +0.281 |
less | +0.267 |
Neg logit contributions | |
cks | -0.254 |
lair | -0.253 |
pads | -0.239 |
sizes | -0.230 |
mess | -0.229 |
Token | daughter |
Token ID | 4957 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
plunge | +0.112 |
reward | +0.109 |
ar | +0.108 |
Sure | +0.104 |
help | +0.101 |
Neg logit contributions | |
lair | -0.082 |
pads | -0.079 |
mess | -0.079 |
neck | -0.078 |
necks | -0.078 |
Token | . |
Token ID | 13 |
Feature activation | -4.12e-01 |
Pos logit contributions | |
plunge | +0.135 |
reward | +0.133 |
ar | +0.130 |
Sure | +0.126 |
help | +0.125 |
Neg logit contributions | |
lair | -0.173 |
cks | -0.169 |
pads | -0.166 |
mess | -0.166 |
floor | -0.163 |
Token | They |
Token ID | 1119 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
help | +0.026 |
reward | +0.024 |
Bub | +0.024 |
Sure | +0.024 |
Enjoy | +0.023 |
Neg logit contributions | |
cks | -0.039 |
lair | -0.038 |
mess | -0.038 |
pads | -0.036 |
necks | -0.035 |
Token | ley |
Token ID | 1636 |
Feature activation | -2.91e+00 |
Pos logit contributions | |
help | +0.071 |
reward | +0.070 |
plunge | +0.069 |
ar | +0.066 |
Sure | +0.065 |
Neg logit contributions | |
lair | -0.065 |
cks | -0.063 |
pads | -0.060 |
mess | -0.060 |
necks | -0.058 |
Token | was |
Token ID | 373 |
Feature activation | -3.02e-01 |
Pos logit contributions | |
help | +0.198 |
ar | +0.190 |
reward | +0.188 |
ted | +0.179 |
bye | +0.179 |
Neg logit contributions | |
cks | -0.255 |
lair | -0.252 |
mess | -0.235 |
pads | -0.233 |
sizes | -0.232 |
Token | a |
Token ID | 257 |
Feature activation | -6.88e-01 |
Pos logit contributions | |
help | +0.019 |
reward | +0.017 |
plunge | +0.016 |
ar | +0.016 |
Sure | +0.016 |
Neg logit contributions | |
lair | -0.028 |
cks | -0.028 |
sizes | -0.028 |
necks | -0.027 |
mess | -0.026 |
Token | three |
Token ID | 1115 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
help | +0.047 |
reward | +0.047 |
ar | +0.046 |
plunge | +0.045 |
bye | +0.043 |
Neg logit contributions | |
lair | -0.058 |
cks | -0.058 |
mess | -0.055 |
pads | -0.055 |
floor | -0.054 |
Token | - |
Token ID | 12 |
Feature activation | -3.64e+00 |
Pos logit contributions | |
help | +0.071 |
reward | +0.068 |
ar | +0.067 |
f | +0.062 |
bye | +0.060 |
Neg logit contributions | |
lair | -0.148 |
cks | -0.145 |
pads | -0.140 |
llers | -0.138 |
mess | -0.136 |
Token | year |
Token ID | 1941 |
Feature activation | -6.93e+00 |
Pos logit contributions | |
reward | +0.178 |
plunge | +0.162 |
ar | +0.161 |
help | +0.157 |
bye | +0.137 |
Neg logit contributions | |
lair | -0.389 |
cks | -0.382 |
pads | -0.364 |
mess | -0.357 |
necks | -0.351 |
Token | - |
Token ID | 12 |
Feature activation | -3.64e+00 |
Pos logit contributions | |
help | +0.258 |
reward | +0.255 |
ar | +0.241 |
plunge | +0.233 |
bye | +0.218 |
Neg logit contributions | |
lair | -0.806 |
cks | -0.801 |
mess | -0.780 |
pads | -0.772 |
floor | -0.764 |
Token | old |
Token ID | 727 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
reward | +0.292 |
plunge | +0.284 |
help | +0.281 |
ar | +0.278 |
ted | +0.262 |
Neg logit contributions | |
lair | -0.272 |
cks | -0.263 |
pads | -0.251 |
mess | -0.247 |
necks | -0.240 |
Token | who |
Token ID | 508 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
reward | +0.089 |
plunge | +0.089 |
ar | +0.087 |
help | +0.087 |
Sure | +0.082 |
Neg logit contributions | |
lair | -0.101 |
pads | -0.096 |
mess | -0.095 |
cks | -0.094 |
neck | -0.093 |
Token | loved |
Token ID | 6151 |
Feature activation | -6.95e-01 |
Pos logit contributions | |
help | +0.114 |
reward | +0.110 |
ar | +0.109 |
bye | +0.104 |
f | +0.102 |
Neg logit contributions | |
lair | -0.125 |
cks | -0.125 |
mess | -0.118 |
pads | -0.116 |
necks | -0.114 |
Token | to |
Token ID | 284 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
help | +0.041 |
reward | +0.038 |
plunge | +0.035 |
bye | +0.034 |
ted | +0.033 |
Neg logit contributions | |
cks | -0.071 |
lair | -0.068 |
pads | -0.065 |
mess | -0.063 |
necks | -0.063 |
Token |
|
Token ID | 198 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
help | +0.015 |
reward | +0.015 |
ar | +0.014 |
plunge | +0.014 |
bye | +0.013 |
Neg logit contributions | |
lair | -0.017 |
cks | -0.017 |
pads | -0.016 |
mess | -0.016 |
floor | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
lair | +0.002 |
cks | +0.002 |
mess | +0.002 |
pads | +0.002 |
floor | +0.002 |
Neg logit contributions | |
help | -0.002 |
reward | -0.002 |
ar | -0.001 |
plunge | -0.001 |
bye | -0.001 |
Token | The |
Token ID | 464 |
Feature activation | -9.02e-01 |
Pos logit contributions | |
lair | +0.009 |
cks | +0.009 |
mess | +0.009 |
pads | +0.009 |
floor | +0.008 |
Neg logit contributions | |
help | -0.009 |
reward | -0.009 |
ar | -0.009 |
plunge | -0.009 |
bye | -0.009 |
Token | three |
Token ID | 1115 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
ar | +0.070 |
ted | +0.067 |
less | +0.067 |
help | +0.064 |
plunge | +0.064 |
Neg logit contributions | |
floor | -0.067 |
lair | -0.067 |
mess | -0.066 |
neck | -0.064 |
frown | -0.062 |
Token | - |
Token ID | 12 |
Feature activation | -3.77e+00 |
Pos logit contributions | |
help | +0.074 |
reward | +0.073 |
ar | +0.071 |
plunge | +0.069 |
bye | +0.066 |
Neg logit contributions | |
lair | -0.123 |
cks | -0.121 |
mess | -0.117 |
pads | -0.116 |
floor | -0.114 |
Token | year |
Token ID | 1941 |
Feature activation | -6.93e+00 |
Pos logit contributions | |
reward | +0.165 |
help | +0.149 |
plunge | +0.145 |
ar | +0.141 |
less | +0.123 |
Neg logit contributions | |
cks | -0.422 |
lair | -0.420 |
pads | -0.402 |
necks | -0.390 |
floor | -0.388 |
Token | - |
Token ID | 12 |
Feature activation | -3.76e+00 |
Pos logit contributions | |
help | +0.292 |
reward | +0.289 |
ar | +0.273 |
plunge | +0.269 |
Sure | +0.251 |
Neg logit contributions | |
lair | -0.773 |
cks | -0.765 |
mess | -0.745 |
pads | -0.736 |
floor | -0.732 |
Token | old |
Token ID | 727 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
reward | +0.303 |
help | +0.292 |
plunge | +0.288 |
ar | +0.281 |
ted | +0.268 |
Neg logit contributions | |
lair | -0.280 |
cks | -0.275 |
pads | -0.259 |
llers | -0.251 |
necks | -0.250 |
Token | girl |
Token ID | 2576 |
Feature activation | -4.71e-01 |
Pos logit contributions | |
reward | +0.087 |
plunge | +0.085 |
help | +0.084 |
ar | +0.082 |
less | +0.079 |
Neg logit contributions | |
lair | -0.087 |
pads | -0.083 |
mess | -0.083 |
floor | -0.082 |
cks | -0.082 |
Token | was |
Token ID | 373 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
help | +0.034 |
reward | +0.033 |
ar | +0.033 |
plunge | +0.032 |
bye | +0.031 |
Neg logit contributions | |
lair | -0.039 |
cks | -0.039 |
mess | -0.037 |
pads | -0.036 |
floor | -0.036 |
Token | fascinated |
Token ID | 30103 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
help | +0.013 |
reward | +0.013 |
ar | +0.012 |
plunge | +0.012 |
less | +0.012 |
Neg logit contributions | |
cks | -0.015 |
lair | -0.015 |
pads | -0.014 |
mess | -0.014 |
necks | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | -3.19e-01 |
Pos logit contributions | |
reward | +0.027 |
help | +0.027 |
ar | +0.026 |
plunge | +0.026 |
bye | +0.024 |
Neg logit contributions | |
lair | -0.033 |
cks | -0.032 |
floor | -0.031 |
mess | -0.031 |
pads | -0.031 |
Token |
|
Token ID | 198 |
Feature activation | -2.54e-01 |
Pos logit contributions | |
help | +0.022 |
reward | +0.022 |
ar | +0.021 |
plunge | +0.021 |
bye | +0.020 |
Neg logit contributions | |
lair | -0.027 |
cks | -0.027 |
mess | -0.026 |
pads | -0.025 |
floor | -0.025 |
Token | A |
Token ID | 32 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
help | +0.017 |
reward | +0.017 |
ar | +0.016 |
plunge | +0.016 |
Sure | +0.015 |
Neg logit contributions | |
lair | -0.023 |
cks | -0.022 |
mess | -0.022 |
pads | -0.021 |
floor | -0.021 |
Token | 3 |
Token ID | 513 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
ted | +0.242 |
plunge | +0.226 |
ed | +0.210 |
ar | +0.210 |
Enjoy | +0.205 |
Neg logit contributions | |
mess | -0.167 |
floor | -0.151 |
frown | -0.148 |
web | -0.141 |
lair | -0.134 |
Token | - |
Token ID | 12 |
Feature activation | -3.66e+00 |
Pos logit contributions | |
reward | +0.087 |
help | +0.087 |
plunge | +0.083 |
ar | +0.083 |
Sure | +0.078 |
Neg logit contributions | |
lair | -0.143 |
cks | -0.143 |
mess | -0.138 |
floor | -0.135 |
pads | -0.135 |
Token | year |
Token ID | 1941 |
Feature activation | -6.93e+00 |
Pos logit contributions | |
reward | +0.158 |
plunge | +0.142 |
help | +0.137 |
ar | +0.137 |
less | +0.124 |
Neg logit contributions | |
cks | -0.417 |
lair | -0.413 |
pads | -0.387 |
mess | -0.384 |
floor | -0.380 |
Token | - |
Token ID | 12 |
Feature activation | -3.66e+00 |
Pos logit contributions | |
help | +0.285 |
reward | +0.283 |
ar | +0.268 |
plunge | +0.266 |
Sure | +0.247 |
Neg logit contributions | |
lair | -0.777 |
cks | -0.769 |
mess | -0.750 |
pads | -0.739 |
floor | -0.736 |
Token | old |
Token ID | 727 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
reward | +0.285 |
help | +0.276 |
plunge | +0.274 |
ar | +0.268 |
less | +0.257 |
Neg logit contributions | |
lair | -0.284 |
cks | -0.281 |
pads | -0.262 |
mess | -0.261 |
floor | -0.255 |
Token | toddler |
Token ID | 30773 |
Feature activation | -4.10e-01 |
Pos logit contributions | |
plunge | +0.105 |
reward | +0.101 |
less | +0.097 |
ar | +0.095 |
Sure | +0.094 |
Neg logit contributions | |
lair | -0.087 |
mess | -0.080 |
frown | -0.079 |
neck | -0.079 |
wondered | -0.077 |
Token | walked |
Token ID | 6807 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
reward | +0.029 |
plunge | +0.029 |
bye | +0.027 |
Sure | +0.027 |
less | +0.026 |
Neg logit contributions | |
lair | -0.033 |
cks | -0.031 |
frown | -0.031 |
neck | -0.031 |
mess | -0.031 |
Token | up |
Token ID | 510 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
reward | +0.101 |
ar | +0.094 |
help | +0.093 |
plunge | +0.091 |
bye | +0.087 |
Neg logit contributions | |
lair | -0.095 |
mess | -0.090 |
cks | -0.090 |
pads | -0.089 |
floor | -0.088 |
Token | mum |
Token ID | 25682 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
cks | +0.106 |
lair | +0.100 |
necks | +0.098 |
mess | +0.098 |
llers | +0.096 |
Neg logit contributions | |
help | -0.033 |
reward | -0.026 |
plunge | -0.023 |
less | -0.022 |
bye | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | -6.30e-02 |
Pos logit contributions | |
lair | +0.014 |
cks | +0.014 |
mess | +0.013 |
pads | +0.013 |
floor | +0.013 |
Neg logit contributions | |
help | -0.012 |
reward | -0.012 |
ar | -0.012 |
plunge | -0.011 |
bye | -0.011 |
Token | her |
Token ID | 607 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
reward | +0.004 |
help | +0.004 |
ar | +0.004 |
plunge | +0.004 |
bye | +0.004 |
Neg logit contributions | |
lair | -0.005 |
cks | -0.005 |
mess | -0.005 |
pads | -0.005 |
floor | -0.005 |
Token | three |
Token ID | 1115 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
lair | +0.107 |
cks | +0.105 |
mess | +0.103 |
pads | +0.102 |
floor | +0.101 |
Neg logit contributions | |
help | -0.076 |
reward | -0.076 |
ar | -0.075 |
plunge | -0.073 |
bye | -0.071 |
Token | - |
Token ID | 12 |
Feature activation | -3.70e+00 |
Pos logit contributions | |
help | +0.082 |
reward | +0.080 |
ar | +0.079 |
plunge | +0.075 |
f | +0.074 |
Neg logit contributions | |
lair | -0.100 |
cks | -0.097 |
mess | -0.093 |
pads | -0.092 |
llers | -0.091 |
Token | year |
Token ID | 1941 |
Feature activation | -6.93e+00 |
Pos logit contributions | |
reward | +0.216 |
plunge | +0.198 |
ar | +0.195 |
help | +0.192 |
bye | +0.175 |
Neg logit contributions | |
cks | -0.354 |
lair | -0.347 |
pads | -0.342 |
necks | -0.329 |
sizes | -0.328 |
Token | - |
Token ID | 12 |
Feature activation | -3.64e+00 |
Pos logit contributions | |
help | +0.279 |
reward | +0.278 |
plunge | +0.262 |
ar | +0.262 |
Sure | +0.244 |
Neg logit contributions | |
lair | -0.783 |
cks | -0.774 |
mess | -0.757 |
pads | -0.746 |
floor | -0.743 |
Token | old |
Token ID | 727 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
reward | +0.309 |
plunge | +0.293 |
ar | +0.290 |
help | +0.289 |
Sure | +0.269 |
Neg logit contributions | |
cks | -0.245 |
lair | -0.245 |
pads | -0.237 |
sizes | -0.230 |
llers | -0.228 |
Token | daughter |
Token ID | 4957 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
reward | +0.069 |
ar | +0.068 |
plunge | +0.068 |
help | +0.065 |
Sure | +0.065 |
Neg logit contributions | |
lair | -0.082 |
pads | -0.080 |
mess | -0.079 |
cks | -0.078 |
floor | -0.078 |
Token | went |
Token ID | 1816 |
Feature activation | -6.24e-01 |
Pos logit contributions | |
help | +0.121 |
reward | +0.121 |
ar | +0.118 |
plunge | +0.116 |
bye | +0.112 |
Neg logit contributions | |
lair | -0.134 |
cks | -0.132 |
mess | -0.127 |
pads | -0.126 |
floor | -0.124 |
Token | to |
Token ID | 284 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
reward | +0.041 |
bye | +0.041 |
ar | +0.040 |
plunge | +0.039 |
Sure | +0.038 |
Neg logit contributions | |
lair | -0.051 |
mess | -0.051 |
floor | -0.049 |
pads | -0.049 |
cks | -0.049 |
Token | She |
Token ID | 1375 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
help | +0.024 |
reward | +0.024 |
ar | +0.023 |
plunge | +0.023 |
bye | +0.022 |
Neg logit contributions | |
lair | -0.032 |
cks | -0.032 |
mess | -0.031 |
pads | -0.031 |
floor | -0.030 |
Token | told |
Token ID | 1297 |
Feature activation | -7.92e-01 |
Pos logit contributions | |
lair | +0.019 |
cks | +0.019 |
mess | +0.019 |
floor | +0.018 |
pads | +0.018 |
Neg logit contributions | |
help | -0.021 |
reward | -0.021 |
plunge | -0.020 |
ar | -0.020 |
bye | -0.019 |
Token | her |
Token ID | 607 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
ar | +0.062 |
reward | +0.059 |
Sure | +0.058 |
plunge | +0.057 |
less | +0.057 |
Neg logit contributions | |
mess | -0.061 |
floor | -0.059 |
frown | -0.058 |
cks | -0.057 |
lair | -0.056 |
Token | three |
Token ID | 1115 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
mess | +0.084 |
lair | +0.084 |
floor | +0.081 |
pads | +0.079 |
frown | +0.079 |
Neg logit contributions | |
ar | -0.079 |
plunge | -0.075 |
ted | -0.073 |
bye | -0.072 |
reward | -0.072 |
Token | - |
Token ID | 12 |
Feature activation | -3.74e+00 |
Pos logit contributions | |
help | +0.073 |
reward | +0.073 |
ar | +0.071 |
plunge | +0.066 |
bye | +0.064 |
Neg logit contributions | |
lair | -0.140 |
cks | -0.139 |
pads | -0.133 |
mess | -0.132 |
floor | -0.129 |
Token | year |
Token ID | 1941 |
Feature activation | -6.92e+00 |
Pos logit contributions | |
reward | +0.239 |
plunge | +0.232 |
help | +0.207 |
ar | +0.202 |
less | +0.177 |
Neg logit contributions | |
pads | -0.325 |
cks | -0.320 |
lair | -0.317 |
ents | -0.310 |
vice | -0.300 |
Token | - |
Token ID | 12 |
Feature activation | -3.74e+00 |
Pos logit contributions | |
help | +0.298 |
reward | +0.296 |
plunge | +0.287 |
ar | +0.279 |
Sure | +0.269 |
Neg logit contributions | |
lair | -0.761 |
cks | -0.747 |
mess | -0.734 |
floor | -0.725 |
pads | -0.725 |
Token | old |
Token ID | 727 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
plunge | +0.361 |
reward | +0.360 |
help | +0.335 |
ar | +0.334 |
ted | +0.323 |
Neg logit contributions | |
pads | -0.201 |
lair | -0.200 |
cks | -0.192 |
opl | -0.180 |
necks | -0.179 |
Token | daughter |
Token ID | 4957 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
plunge | +0.084 |
ar | +0.083 |
reward | +0.083 |
help | +0.077 |
less | +0.075 |
Neg logit contributions | |
lair | -0.099 |
pads | -0.096 |
mess | -0.096 |
floor | -0.094 |
neck | -0.093 |
Token | to |
Token ID | 284 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
reward | +0.112 |
ar | +0.111 |
plunge | +0.110 |
help | +0.107 |
bye | +0.104 |
Neg logit contributions | |
lair | -0.192 |
pads | -0.186 |
cks | -0.185 |
mess | -0.185 |
floor | -0.181 |
Token | attach |
Token ID | 10199 |
Feature activation | -9.15e-01 |
Pos logit contributions | |
ar | +0.013 |
ted | +0.012 |
less | +0.012 |
iest | +0.011 |
ed | +0.011 |
Neg logit contributions | |
mess | -0.012 |
frown | -0.012 |
floor | -0.011 |
neck | -0.011 |
lair | -0.011 |
Token | on |
Token ID | 319 |
Feature activation | +3.53e-01 |
Pos logit contributions | |
pads | +0.028 |
cks | +0.026 |
goals | +0.026 |
neck | +0.025 |
beating | +0.024 |
Neg logit contributions | |
plunge | -0.030 |
reward | -0.027 |
bye | -0.026 |
Sure | -0.026 |
help | -0.026 |
Token | , |
Token ID | 11 |
Feature activation | -5.91e-02 |
Pos logit contributions | |
cks | +0.034 |
lair | +0.034 |
mess | +0.033 |
floor | +0.033 |
pads | +0.033 |
Neg logit contributions | |
reward | -0.019 |
help | -0.019 |
plunge | -0.018 |
bye | -0.018 |
ar | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
ar | +0.004 |
plunge | +0.004 |
reward | +0.004 |
help | +0.004 |
less | +0.004 |
Neg logit contributions | |
lair | -0.005 |
pads | -0.005 |
mess | -0.005 |
cks | -0.005 |
sizes | -0.005 |
Token | three |
Token ID | 1115 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
mess | +0.014 |
floor | +0.014 |
lair | +0.013 |
frown | +0.013 |
neck | +0.013 |
Neg logit contributions | |
ar | -0.015 |
ted | -0.014 |
plunge | -0.014 |
Sure | -0.014 |
less | -0.013 |
Token | - |
Token ID | 12 |
Feature activation | -3.68e+00 |
Pos logit contributions | |
help | +0.070 |
reward | +0.070 |
plunge | +0.067 |
ar | +0.067 |
bye | +0.064 |
Neg logit contributions | |
lair | -0.119 |
cks | -0.118 |
mess | -0.115 |
pads | -0.113 |
floor | -0.112 |
Token | year |
Token ID | 1941 |
Feature activation | -6.92e+00 |
Pos logit contributions | |
reward | +0.199 |
plunge | +0.192 |
help | +0.187 |
ar | +0.172 |
ted | +0.156 |
Neg logit contributions | |
cks | -0.360 |
pads | -0.354 |
lair | -0.352 |
llers | -0.337 |
opl | -0.337 |
Token | - |
Token ID | 12 |
Feature activation | -3.69e+00 |
Pos logit contributions | |
help | +0.302 |
reward | +0.301 |
plunge | +0.293 |
Sure | +0.279 |
ar | +0.278 |
Neg logit contributions | |
lair | -0.750 |
cks | -0.737 |
mess | -0.720 |
floor | -0.714 |
pads | -0.713 |
Token | old |
Token ID | 727 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
reward | +0.307 |
plunge | +0.301 |
help | +0.293 |
ar | +0.280 |
ted | +0.272 |
Neg logit contributions | |
cks | -0.250 |
lair | -0.247 |
pads | -0.245 |
opl | -0.231 |
llers | -0.231 |
Token | started |
Token ID | 2067 |
Feature activation | -6.97e-01 |
Pos logit contributions | |
plunge | +0.091 |
reward | +0.091 |
help | +0.090 |
ar | +0.087 |
Sure | +0.086 |
Neg logit contributions | |
lair | -0.077 |
pads | -0.075 |
mess | -0.074 |
floor | -0.073 |
cks | -0.073 |
Token | testing |
Token ID | 4856 |
Feature activation | +2.54e-01 |
Pos logit contributions | |
help | +0.037 |
reward | +0.033 |
less | +0.031 |
ar | +0.031 |
plunge | +0.031 |
Neg logit contributions | |
cks | -0.074 |
lair | -0.072 |
mess | -0.070 |
pads | -0.069 |
floor | -0.068 |
Token | and |
Token ID | 290 |
Feature activation | -9.83e-03 |
Pos logit contributions | |
lair | +0.023 |
cks | +0.022 |
mess | +0.022 |
pads | +0.021 |
floor | +0.021 |
Neg logit contributions | |
help | -0.016 |
reward | -0.016 |
ar | -0.016 |
plunge | -0.016 |
bye | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | -4.97e-01 |
Pos logit contributions | |
plunge | +0.105 |
bye | +0.101 |
Sure | +0.098 |
reward | +0.098 |
ar | +0.097 |
Neg logit contributions | |
frown | -0.080 |
mess | -0.080 |
floor | -0.079 |
neck | -0.074 |
reading | -0.074 |
Token | a |
Token ID | 257 |
Feature activation | -6.74e-01 |
Pos logit contributions | |
ar | +0.044 |
plunge | +0.042 |
bye | +0.042 |
or | +0.041 |
less | +0.039 |
Neg logit contributions | |
floor | -0.032 |
frown | -0.031 |
neck | -0.031 |
mess | -0.031 |
uneven | -0.029 |
Token | little |
Token ID | 1310 |
Feature activation | -7.73e-01 |
Pos logit contributions | |
ted | +0.062 |
plunge | +0.059 |
ar | +0.057 |
bye | +0.056 |
f | +0.055 |
Neg logit contributions | |
mess | -0.045 |
frown | -0.043 |
floor | -0.041 |
lair | -0.040 |
web | -0.039 |
Token | three |
Token ID | 1115 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
reward | +0.051 |
ar | +0.050 |
plunge | +0.049 |
help | +0.048 |
f | +0.046 |
Neg logit contributions | |
mess | -0.066 |
lair | -0.065 |
floor | -0.063 |
frown | -0.063 |
neck | -0.061 |
Token | - |
Token ID | 12 |
Feature activation | -3.61e+00 |
Pos logit contributions | |
reward | +0.062 |
help | +0.061 |
plunge | +0.059 |
ar | +0.058 |
bye | +0.055 |
Neg logit contributions | |
lair | -0.133 |
cks | -0.133 |
mess | -0.130 |
floor | -0.127 |
pads | -0.127 |
Token | year |
Token ID | 1941 |
Feature activation | -6.91e+00 |
Pos logit contributions | |
reward | +0.185 |
plunge | +0.168 |
ar | +0.159 |
help | +0.152 |
less | +0.141 |
Neg logit contributions | |
cks | -0.382 |
lair | -0.375 |
pads | -0.356 |
floor | -0.344 |
mess | -0.344 |
Token | - |
Token ID | 12 |
Feature activation | -3.61e+00 |
Pos logit contributions | |
help | +0.270 |
reward | +0.269 |
ar | +0.253 |
plunge | +0.252 |
Sure | +0.234 |
Neg logit contributions | |
lair | -0.790 |
cks | -0.780 |
mess | -0.762 |
pads | -0.752 |
floor | -0.748 |
Token | old |
Token ID | 727 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
reward | +0.313 |
plunge | +0.303 |
help | +0.294 |
ar | +0.290 |
less | +0.278 |
Neg logit contributions | |
lair | -0.249 |
cks | -0.247 |
pads | -0.230 |
mess | -0.222 |
floor | -0.218 |
Token | boy |
Token ID | 2933 |
Feature activation | +2.78e-01 |
Pos logit contributions | |
plunge | +0.099 |
reward | +0.096 |
ar | +0.091 |
less | +0.090 |
Sure | +0.089 |
Neg logit contributions | |
lair | -0.094 |
mess | -0.088 |
pads | -0.086 |
neck | -0.085 |
frown | -0.085 |
Token | named |
Token ID | 3706 |
Feature activation | -2.60e+00 |
Pos logit contributions | |
lair | +0.021 |
cks | +0.021 |
pads | +0.020 |
mess | +0.020 |
frown | +0.020 |
Neg logit contributions | |
reward | -0.021 |
plunge | -0.021 |
Sure | -0.020 |
less | -0.019 |
bye | -0.019 |
Token | John |
Token ID | 1757 |
Feature activation | +7.77e-01 |
Pos logit contributions | |
reward | +0.173 |
ar | +0.168 |
help | +0.164 |
bye | +0.159 |
plunge | +0.159 |
Neg logit contributions | |
lair | -0.229 |
cks | -0.222 |
mess | -0.220 |
floor | -0.216 |
pads | -0.211 |
Token | , |
Token ID | 11 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
mess | +0.062 |
lair | +0.061 |
cks | +0.061 |
floor | +0.061 |
pads | +0.059 |
Neg logit contributions | |
reward | -0.039 |
ar | -0.038 |
help | -0.038 |
plunge | -0.037 |
bye | -0.036 |
Token | she |
Token ID | 673 |
Feature activation | +7.39e-01 |
Pos logit contributions | |
ar | +0.017 |
bye | +0.016 |
reward | +0.016 |
plunge | +0.016 |
less | +0.016 |
Neg logit contributions | |
mess | -0.013 |
floor | -0.013 |
frown | -0.012 |
lair | -0.012 |
neck | -0.012 |
Token | looked |
Token ID | 3114 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
lair | +0.077 |
cks | +0.077 |
opl | +0.075 |
pads | +0.069 |
sizes | +0.069 |
Neg logit contributions | |
help | -0.043 |
reward | -0.038 |
f | -0.037 |
Sure | -0.036 |
ar | -0.035 |
Token | at |
Token ID | 379 |
Feature activation | +7.75e-02 |
Pos logit contributions | |
lair | +0.009 |
cks | +0.009 |
pads | +0.009 |
mess | +0.009 |
floor | +0.008 |
Neg logit contributions | |
help | -0.007 |
reward | -0.006 |
ar | -0.006 |
plunge | -0.006 |
Sure | -0.006 |
Token | her |
Token ID | 607 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
lair | +0.008 |
cks | +0.008 |
pads | +0.007 |
mess | +0.007 |
necks | +0.007 |
Neg logit contributions | |
help | -0.004 |
reward | -0.004 |
plunge | -0.004 |
ar | -0.004 |
ted | -0.004 |
Token | wall |
Token ID | 3355 |
Feature activation | +2.55e+00 |
Pos logit contributions | |
cks | +0.103 |
lair | +0.101 |
pads | +0.098 |
sizes | +0.095 |
necks | +0.090 |
Neg logit contributions | |
help | -0.070 |
reward | -0.070 |
plunge | -0.060 |
ar | -0.059 |
f | -0.059 |
Token | and |
Token ID | 290 |
Feature activation | -5.87e-02 |
Pos logit contributions | |
cks | +0.268 |
lair | +0.262 |
die | +0.254 |
pads | +0.248 |
opl | +0.242 |
Neg logit contributions | |
help | -0.136 |
less | -0.126 |
f | -0.121 |
reward | -0.119 |
ar | -0.116 |
Token | smiled |
Token ID | 13541 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
help | +0.004 |
ted | +0.003 |
reward | +0.003 |
Sure | +0.003 |
ar | +0.003 |
Neg logit contributions | |
lair | -0.006 |
cks | -0.005 |
opl | -0.005 |
pads | -0.005 |
fault | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | -9.31e-03 |
Pos logit contributions | |
help | +0.065 |
reward | +0.063 |
less | +0.059 |
bye | +0.058 |
ar | +0.058 |
Neg logit contributions | |
cks | -0.099 |
lair | -0.097 |
pads | -0.092 |
necks | -0.091 |
mess | -0.091 |
Token | It |
Token ID | 632 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
help | +0.001 |
reward | +0.001 |
ar | +0.001 |
Sure | +0.001 |
plunge | +0.001 |
Neg logit contributions | |
cks | -0.001 |
lair | -0.001 |
mess | -0.001 |
pads | -0.001 |
necks | -0.001 |
Token | was |
Token ID | 373 |
Feature activation | -8.63e-04 |
Pos logit contributions | |
help | +0.086 |
reward | +0.074 |
Sure | +0.071 |
ar | +0.071 |
f | +0.067 |
Neg logit contributions | |
lair | -0.217 |
cks | -0.210 |
necks | -0.200 |
pads | -0.197 |
goals | -0.193 |
Token | will |
Token ID | 481 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
help | +0.036 |
reward | +0.035 |
ar | +0.033 |
plunge | +0.032 |
bye | +0.031 |
Neg logit contributions | |
lair | -0.050 |
cks | -0.050 |
pads | -0.047 |
mess | -0.047 |
floor | -0.047 |
Token | be |
Token ID | 307 |
Feature activation | +4.06e-01 |
Pos logit contributions | |
help | +0.079 |
reward | +0.075 |
plunge | +0.070 |
ar | +0.070 |
bye | +0.068 |
Neg logit contributions | |
lair | -0.084 |
cks | -0.083 |
pads | -0.080 |
mess | -0.076 |
necks | -0.076 |
Token | in |
Token ID | 287 |
Feature activation | +9.47e-02 |
Pos logit contributions | |
lair | +0.036 |
llers | +0.035 |
sizes | +0.035 |
pads | +0.035 |
floor | +0.035 |
Neg logit contributions | |
help | -0.027 |
reward | -0.025 |
ar | -0.024 |
plunge | -0.023 |
less | -0.022 |
Token | trouble |
Token ID | 5876 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
cks | +0.009 |
pads | +0.009 |
lair | +0.009 |
llers | +0.008 |
necks | +0.008 |
Neg logit contributions | |
help | -0.006 |
reward | -0.006 |
plunge | -0.005 |
ar | -0.005 |
less | -0.005 |
Token | !" |
Token ID | 2474 |
Feature activation | +9.29e-01 |
Pos logit contributions | |
lair | +0.160 |
cks | +0.158 |
pads | +0.152 |
mess | +0.151 |
floor | +0.148 |
Neg logit contributions | |
help | -0.122 |
reward | -0.119 |
ar | -0.116 |
plunge | -0.114 |
bye | -0.110 |
Token | Anna |
Token ID | 11735 |
Feature activation | +2.05e+00 |
Pos logit contributions | |
lair | +0.083 |
cks | +0.082 |
pads | +0.081 |
llers | +0.078 |
sizes | +0.077 |
Neg logit contributions | |
help | -0.066 |
reward | -0.062 |
ar | -0.058 |
Sure | -0.057 |
plunge | -0.055 |
Token | said |
Token ID | 531 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
lair | +0.208 |
cks | +0.204 |
opl | +0.197 |
mess | +0.188 |
pads | +0.188 |
Neg logit contributions | |
help | -0.120 |
ar | -0.112 |
reward | -0.110 |
f | -0.105 |
Sure | -0.105 |
Token | . |
Token ID | 13 |
Feature activation | +3.37e-01 |
Pos logit contributions | |
reward | +0.011 |
ar | +0.011 |
plunge | +0.011 |
help | +0.011 |
bye | +0.011 |
Neg logit contributions | |
cks | -0.010 |
lair | -0.010 |
mess | -0.010 |
pads | -0.009 |
floor | -0.009 |
Token | " |
Token ID | 366 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
lair | +0.035 |
cks | +0.035 |
mess | +0.034 |
opl | +0.034 |
sizes | +0.034 |
Neg logit contributions | |
help | -0.019 |
Sure | -0.017 |
reward | -0.016 |
ted | -0.016 |
ar | -0.015 |
Token | We |
Token ID | 1135 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
lair | +0.016 |
mess | +0.015 |
cks | +0.015 |
pads | +0.015 |
opl | +0.015 |
Neg logit contributions | |
Sure | -0.009 |
help | -0.009 |
bye | -0.008 |
f | -0.008 |
reward | -0.008 |
Token | were |
Token ID | 547 |
Feature activation | +2.56e-01 |
Pos logit contributions | |
reward | +0.110 |
help | +0.108 |
ar | +0.107 |
plunge | +0.106 |
bye | +0.103 |
Neg logit contributions | |
lair | -0.141 |
cks | -0.140 |
mess | -0.137 |
pads | -0.134 |
floor | -0.131 |
Token | 's |
Token ID | 338 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
cks | +0.059 |
lair | +0.059 |
mess | +0.055 |
pads | +0.052 |
llers | +0.052 |
Neg logit contributions | |
help | -0.107 |
reward | -0.105 |
ar | -0.102 |
plunge | -0.101 |
less | -0.098 |
Token | shoulder |
Token ID | 8163 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
lair | +0.137 |
cks | +0.136 |
mess | +0.129 |
pads | +0.127 |
floor | +0.124 |
Neg logit contributions | |
help | -0.186 |
reward | -0.184 |
ar | -0.180 |
plunge | -0.178 |
bye | -0.173 |
Token | . |
Token ID | 13 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
cks | +0.009 |
lair | +0.009 |
mess | +0.008 |
pads | +0.008 |
floor | +0.008 |
Neg logit contributions | |
help | -0.008 |
reward | -0.008 |
ar | -0.008 |
plunge | -0.008 |
bye | -0.008 |
Token | " |
Token ID | 366 |
Feature activation | +6.79e-02 |
Pos logit contributions | |
ar | +0.001 |
plunge | +0.001 |
reward | +0.001 |
help | +0.001 |
bye | +0.001 |
Neg logit contributions | |
lair | -0.001 |
floor | -0.001 |
llers | -0.001 |
pads | -0.001 |
mess | -0.001 |
Token | That |
Token ID | 2504 |
Feature activation | -6.47e-01 |
Pos logit contributions | |
lair | +0.005 |
cks | +0.005 |
pads | +0.005 |
mess | +0.005 |
floor | +0.005 |
Neg logit contributions | |
reward | -0.005 |
help | -0.005 |
ar | -0.005 |
plunge | -0.005 |
bye | -0.005 |
Token | 's |
Token ID | 338 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
help | +0.061 |
reward | +0.061 |
plunge | +0.059 |
bye | +0.058 |
ar | +0.057 |
Neg logit contributions | |
cks | -0.045 |
lair | -0.042 |
necks | -0.036 |
pads | -0.036 |
llers | -0.035 |
Token | very |
Token ID | 845 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
cks | +0.156 |
mess | +0.152 |
lair | +0.149 |
floor | +0.145 |
frown | +0.144 |
Neg logit contributions | |
ar | -0.158 |
plunge | -0.158 |
reward | -0.155 |
ted | -0.155 |
help | -0.154 |
Token | kind |
Token ID | 1611 |
Feature activation | -9.25e-01 |
Pos logit contributions | |
ted | +0.010 |
plunge | +0.010 |
reward | +0.010 |
ar | +0.010 |
help | +0.010 |
Neg logit contributions | |
mess | -0.007 |
cks | -0.007 |
lair | -0.007 |
floor | -0.006 |
frown | -0.006 |
Token | of |
Token ID | 286 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
ar | +0.034 |
reward | +0.034 |
plunge | +0.033 |
help | +0.033 |
Sure | +0.031 |
Neg logit contributions | |
lair | -0.105 |
mess | -0.103 |
pads | -0.103 |
floor | -0.102 |
sizes | -0.101 |
Token | you |
Token ID | 345 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
ar | +0.063 |
Sure | +0.061 |
reward | +0.060 |
plunge | +0.059 |
less | +0.058 |
Neg logit contributions | |
cks | -0.091 |
mess | -0.090 |
lair | -0.088 |
floor | -0.087 |
sizes | -0.086 |
Token | to |
Token ID | 284 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
reward | +0.093 |
ar | +0.091 |
plunge | +0.089 |
Sure | +0.089 |
help | +0.089 |
Neg logit contributions | |
lair | -0.115 |
mess | -0.114 |
cks | -0.112 |
pads | -0.111 |
floor | -0.111 |
Token | and |
Token ID | 290 |
Feature activation | +2.76e-01 |
Pos logit contributions | |
cks | +0.076 |
lair | +0.075 |
pads | +0.071 |
llers | +0.071 |
necks | +0.070 |
Neg logit contributions | |
help | -0.041 |
reward | -0.034 |
less | -0.032 |
ar | -0.032 |
Sure | -0.029 |
Token | angry |
Token ID | 7954 |
Feature activation | +6.57e-02 |
Pos logit contributions | |
lair | +0.019 |
cks | +0.018 |
sizes | +0.017 |
pads | +0.017 |
llers | +0.017 |
Neg logit contributions | |
help | -0.025 |
reward | -0.023 |
ted | -0.023 |
bye | -0.023 |
plunge | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
lair | +0.006 |
cks | +0.006 |
pads | +0.006 |
mess | +0.006 |
necks | +0.006 |
Neg logit contributions | |
help | -0.004 |
reward | -0.004 |
ar | -0.004 |
bye | -0.004 |
ted | -0.004 |
Token | He |
Token ID | 679 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
cks | +0.013 |
mess | +0.013 |
pads | +0.013 |
lair | +0.013 |
cking | +0.013 |
Neg logit contributions | |
help | -0.008 |
Bub | -0.007 |
Sure | -0.007 |
ar | -0.007 |
Enjoy | -0.007 |
Token | does |
Token ID | 857 |
Feature activation | +6.30e-02 |
Pos logit contributions | |
lair | +0.015 |
llers | +0.015 |
sizes | +0.015 |
cks | +0.015 |
opl | +0.014 |
Neg logit contributions | |
help | -0.018 |
ted | -0.017 |
reward | -0.016 |
ar | -0.016 |
Enjoy | -0.015 |
Token | not |
Token ID | 407 |
Feature activation | +3.17e+00 |
Pos logit contributions | |
pads | +0.004 |
sizes | +0.004 |
lair | +0.004 |
necks | +0.004 |
cks | +0.004 |
Neg logit contributions | |
help | -0.006 |
reward | -0.006 |
ar | -0.005 |
Sure | -0.005 |
f | -0.005 |
Token | like |
Token ID | 588 |
Feature activation | -3.40e-01 |
Pos logit contributions | |
lair | +0.282 |
cks | +0.274 |
llers | +0.265 |
necks | +0.260 |
floor | +0.258 |
Neg logit contributions | |
help | -0.235 |
reward | -0.221 |
ar | -0.203 |
bye | -0.200 |
f | -0.198 |
Token | Mia |
Token ID | 32189 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
help | +0.020 |
reward | +0.017 |
ted | +0.016 |
Bub | +0.015 |
plunge | +0.015 |
Neg logit contributions | |
lair | -0.034 |
cks | -0.033 |
pads | -0.033 |
necks | -0.032 |
llers | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
lair | +0.132 |
cks | +0.128 |
llers | +0.128 |
sizes | +0.117 |
necks | +0.117 |
Neg logit contributions | |
help | -0.122 |
reward | -0.109 |
f | -0.107 |
ar | -0.105 |
less | -0.104 |
Token | He |
Token ID | 679 |
Feature activation | +2.31e-01 |
Pos logit contributions | |
lair | +0.013 |
mess | +0.013 |
cks | +0.013 |
pads | +0.013 |
llers | +0.012 |
Neg logit contributions | |
help | -0.009 |
ar | -0.008 |
Sure | -0.008 |
Bub | -0.008 |
reward | -0.008 |
Token | does |
Token ID | 857 |
Feature activation | +7.30e-02 |
Pos logit contributions | |
lair | +0.017 |
llers | +0.016 |
cks | +0.015 |
sizes | +0.015 |
pads | +0.015 |
Neg logit contributions | |
help | -0.020 |
ar | -0.019 |
ted | -0.018 |
reward | -0.018 |
f | -0.017 |
Token | 's |
Token ID | 338 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
mess | +0.014 |
cks | +0.013 |
pads | +0.013 |
lair | +0.012 |
floor | +0.012 |
Neg logit contributions | |
reward | -0.060 |
ar | -0.060 |
help | -0.059 |
plunge | -0.059 |
less | -0.059 |
Token | go |
Token ID | 467 |
Feature activation | +3.52e-01 |
Pos logit contributions | |
lair | +0.241 |
cks | +0.239 |
mess | +0.229 |
pads | +0.229 |
floor | +0.225 |
Neg logit contributions | |
help | -0.125 |
reward | -0.121 |
ar | -0.113 |
plunge | -0.113 |
bye | -0.107 |
Token | play |
Token ID | 711 |
Feature activation | -9.57e-02 |
Pos logit contributions | |
mess | +0.021 |
lair | +0.021 |
pads | +0.021 |
cks | +0.020 |
floor | +0.019 |
Neg logit contributions | |
reward | -0.029 |
ted | -0.028 |
less | -0.028 |
plunge | -0.028 |
ar | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | +2.60e-01 |
Pos logit contributions | |
reward | +0.007 |
plunge | +0.007 |
less | +0.007 |
bye | +0.007 |
ted | +0.007 |
Neg logit contributions | |
lair | -0.007 |
mess | -0.007 |
cks | -0.007 |
web | -0.007 |
floor | -0.007 |
Token | Let |
Token ID | 3914 |
Feature activation | +4.55e-01 |
Pos logit contributions | |
lair | +0.018 |
pads | +0.018 |
cks | +0.018 |
necks | +0.017 |
sizes | +0.017 |
Neg logit contributions | |
plunge | -0.021 |
reward | -0.021 |
ar | -0.020 |
less | -0.020 |
help | -0.020 |
Token | 's |
Token ID | 338 |
Feature activation | +2.32e+00 |
Pos logit contributions | |
mess | +0.012 |
cks | +0.011 |
pads | +0.011 |
lair | +0.010 |
floor | +0.010 |
Neg logit contributions | |
reward | -0.057 |
help | -0.057 |
ar | -0.057 |
plunge | -0.056 |
less | -0.056 |
Token | make |
Token ID | 787 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
lair | +0.237 |
cks | +0.235 |
mess | +0.225 |
pads | +0.224 |
floor | +0.221 |
Neg logit contributions | |
help | -0.125 |
reward | -0.122 |
ar | -0.114 |
plunge | -0.114 |
bye | -0.108 |
Token | a |
Token ID | 257 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
lair | +0.126 |
cks | +0.125 |
mess | +0.122 |
pads | +0.120 |
floor | +0.119 |
Neg logit contributions | |
help | -0.065 |
reward | -0.065 |
ar | -0.062 |
plunge | -0.061 |
bye | -0.058 |
Token | snow |
Token ID | 6729 |
Feature activation | +5.95e-01 |
Pos logit contributions | |
help | +0.022 |
ar | +0.022 |
reward | +0.022 |
plunge | +0.021 |
ted | +0.021 |
Neg logit contributions | |
lair | -0.015 |
mess | -0.015 |
cks | -0.015 |
floor | -0.014 |
pads | -0.014 |
Token | man |
Token ID | 805 |
Feature activation | +4.57e-01 |
Pos logit contributions | |
lair | +0.046 |
cks | +0.045 |
mess | +0.044 |
pads | +0.043 |
floor | +0.042 |
Neg logit contributions | |
help | -0.045 |
reward | -0.045 |
ar | -0.044 |
plunge | -0.044 |
Sure | -0.042 |
Token | friend |
Token ID | 1545 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
lair | +0.040 |
cks | +0.040 |
mess | +0.038 |
pads | +0.038 |
floor | +0.037 |
Neg logit contributions | |
help | -0.031 |
reward | -0.030 |
ar | -0.029 |
plunge | -0.028 |
bye | -0.028 |
Token | for |
Token ID | 329 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
lair | +0.052 |
cks | +0.051 |
mess | +0.051 |
pads | +0.050 |
floor | +0.049 |
Neg logit contributions | |
reward | -0.030 |
help | -0.029 |
ar | -0.029 |
plunge | -0.028 |
Sure | -0.027 |
Token | the |
Token ID | 262 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
Sure | +0.114 |
ar | +0.111 |
less | +0.108 |
f | +0.108 |
ted | +0.103 |
Neg logit contributions | |
mess | -0.137 |
lair | -0.133 |
floor | -0.132 |
cks | -0.127 |
frown | -0.120 |
Token | surprise |
Token ID | 5975 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
mess | +0.011 |
floor | +0.011 |
lair | +0.010 |
pads | +0.010 |
frown | +0.009 |
Neg logit contributions | |
ar | -0.010 |
ted | -0.009 |
Sure | -0.009 |
less | -0.009 |
f | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -2.80e-02 |
Pos logit contributions | |
lair | +0.144 |
cks | +0.141 |
mess | +0.138 |
pads | +0.136 |
floor | +0.134 |
Neg logit contributions | |
help | -0.116 |
reward | -0.115 |
ar | -0.113 |
plunge | -0.111 |
Sure | -0.107 |
Token | Come |
Token ID | 7911 |
Feature activation | -8.98e-01 |
Pos logit contributions | |
iest | +0.002 |
plunge | +0.002 |
ar | +0.002 |
less | +0.002 |
reward | +0.002 |
Neg logit contributions | |
lair | -0.002 |
pads | -0.002 |
llers | -0.002 |
floor | -0.002 |
cks | -0.002 |
Token | to |
Token ID | 284 |
Feature activation | +8.36e-02 |
Pos logit contributions | |
iest | +0.067 |
bye | +0.065 |
plunge | +0.065 |
reward | +0.063 |
ted | +0.062 |
Neg logit contributions | |
mess | -0.062 |
lair | -0.061 |
frown | -0.061 |
neck | -0.060 |
cks | -0.059 |
Token | the |
Token ID | 262 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
lair | +0.007 |
mess | +0.007 |
cks | +0.007 |
floor | +0.007 |
frown | +0.007 |
Neg logit contributions | |
ar | -0.006 |
less | -0.005 |
f | -0.005 |
plunge | -0.005 |
iest | -0.005 |
Token | centre |
Token ID | 7372 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
floor | +0.010 |
mess | +0.010 |
lair | +0.009 |
frown | +0.008 |
pads | +0.008 |
Neg logit contributions | |
ar | -0.010 |
f | -0.009 |
ted | -0.009 |
less | -0.009 |
help | -0.009 |
Token | of |
Token ID | 286 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
help | +0.057 |
reward | +0.054 |
ar | +0.049 |
plunge | +0.048 |
bye | +0.047 |
Neg logit contributions | |
cks | -0.159 |
lair | -0.157 |
pads | -0.151 |
necks | -0.149 |
llers | -0.149 |
Token | the |
Token ID | 262 |
Feature activation | +8.41e-02 |
Pos logit contributions | |
ar | +0.066 |
reward | +0.064 |
Sure | +0.063 |
plunge | +0.062 |
help | +0.062 |
Neg logit contributions | |
lair | -0.088 |
mess | -0.087 |
cks | -0.086 |
floor | -0.083 |
pads | -0.079 |
Token | room |
Token ID | 2119 |
Feature activation | +1.86e+00 |
Pos logit contributions | |
mess | +0.007 |
lair | +0.007 |
floor | +0.007 |
pads | +0.006 |
goals | +0.006 |
Neg logit contributions | |
ar | -0.006 |
help | -0.006 |
ted | -0.006 |
plunge | -0.006 |
less | -0.006 |
Token | The |
Token ID | 383 |
Feature activation | -8.51e-01 |
Pos logit contributions | |
cks | +0.049 |
lair | +0.048 |
mess | +0.048 |
pads | +0.046 |
floor | +0.046 |
Neg logit contributions | |
help | -0.032 |
reward | -0.032 |
ar | -0.031 |
Sure | -0.030 |
bye | -0.029 |
Token | moral |
Token ID | 6573 |
Feature activation | -9.36e-01 |
Pos logit contributions | |
ar | +0.064 |
plunge | +0.063 |
ted | +0.061 |
less | +0.059 |
bye | +0.058 |
Neg logit contributions | |
lair | -0.069 |
mess | -0.069 |
floor | -0.066 |
frown | -0.064 |
web | -0.062 |
Token | of |
Token ID | 286 |
Feature activation | -5.80e-01 |
Pos logit contributions | |
plunge | +0.044 |
reward | +0.042 |
ar | +0.041 |
f | +0.039 |
ted | +0.039 |
Neg logit contributions | |
lair | -0.096 |
floor | -0.093 |
mess | -0.093 |
goals | -0.090 |
pads | -0.090 |
Token | the |
Token ID | 262 |
Feature activation | +4.86e-01 |
Pos logit contributions | |
reward | +0.033 |
ar | +0.033 |
help | +0.032 |
plunge | +0.032 |
Sure | +0.030 |
Neg logit contributions | |
cks | -0.056 |
lair | -0.055 |
mess | -0.053 |
floor | -0.052 |
pads | -0.051 |
Token | story |
Token ID | 1621 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
mess | +0.025 |
lair | +0.024 |
floor | +0.023 |
frown | +0.022 |
web | +0.021 |
Neg logit contributions | |
ar | -0.049 |
ted | -0.049 |
less | -0.048 |
plunge | -0.048 |
bye | -0.047 |
Token | is |
Token ID | 318 |
Feature activation | +8.74e-01 |
Pos logit contributions | |
lair | +0.128 |
cks | +0.127 |
pads | +0.119 |
mess | +0.118 |
necks | +0.116 |
Neg logit contributions | |
help | -0.148 |
reward | -0.147 |
ar | -0.143 |
plunge | -0.139 |
bye | -0.138 |
Token | that |
Token ID | 326 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
lair | +0.076 |
cks | +0.073 |
mess | +0.072 |
floor | +0.070 |
llers | +0.069 |
Neg logit contributions | |
reward | -0.056 |
plunge | -0.055 |
ar | -0.055 |
ted | -0.054 |
help | -0.053 |
Token | it |
Token ID | 340 |
Feature activation | -7.86e-01 |
Pos logit contributions | |
help | +0.010 |
reward | +0.009 |
plunge | +0.009 |
bye | +0.008 |
ar | +0.008 |
Neg logit contributions | |
cks | -0.016 |
lair | -0.016 |
pads | -0.016 |
necks | -0.015 |
mess | -0.015 |
Token | 's |
Token ID | 338 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
help | +0.077 |
reward | +0.072 |
ar | +0.069 |
plunge | +0.068 |
f | +0.065 |
Neg logit contributions | |
lair | -0.052 |
cks | -0.050 |
necks | -0.047 |
pads | -0.046 |
neck | -0.045 |
Token | always |
Token ID | 1464 |
Feature activation | +8.82e-01 |
Pos logit contributions | |
lair | +0.131 |
cks | +0.129 |
mess | +0.123 |
pads | +0.120 |
floor | +0.118 |
Neg logit contributions | |
help | -0.182 |
reward | -0.181 |
ar | -0.178 |
plunge | -0.176 |
bye | -0.171 |
Token | best |
Token ID | 1266 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
lair | +0.079 |
cks | +0.078 |
mess | +0.074 |
pads | +0.074 |
floor | +0.073 |
Neg logit contributions | |
help | -0.059 |
reward | -0.058 |
ar | -0.055 |
plunge | -0.054 |
bye | -0.052 |
Token | until |
Token ID | 1566 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
help | +0.020 |
ar | +0.018 |
reward | +0.018 |
plunge | +0.017 |
less | +0.017 |
Neg logit contributions | |
lair | -0.034 |
cks | -0.033 |
pads | -0.032 |
mess | -0.032 |
floor | -0.031 |
Token | , |
Token ID | 11 |
Feature activation | -6.10e-02 |
Pos logit contributions | |
mess | +0.016 |
floor | +0.016 |
frown | +0.015 |
lair | +0.014 |
llers | +0.014 |
Neg logit contributions | |
ar | -0.016 |
plunge | -0.016 |
bye | -0.016 |
Sure | -0.015 |
f | -0.015 |
Token | eventually |
Token ID | 4191 |
Feature activation | +9.91e-02 |
Pos logit contributions | |
ar | +0.004 |
reward | +0.004 |
help | +0.004 |
plunge | +0.004 |
bye | +0.004 |
Neg logit contributions | |
lair | -0.005 |
mess | -0.005 |
cks | -0.005 |
floor | -0.005 |
pads | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | -7.52e-02 |
Pos logit contributions | |
mess | +0.009 |
lair | +0.009 |
floor | +0.008 |
frown | +0.008 |
cks | +0.008 |
Neg logit contributions | |
Sure | -0.006 |
plunge | -0.006 |
help | -0.006 |
bye | -0.006 |
reward | -0.006 |
Token | a |
Token ID | 257 |
Feature activation | -2.69e-01 |
Pos logit contributions | |
ar | +0.004 |
help | +0.004 |
reward | +0.004 |
plunge | +0.004 |
bye | +0.004 |
Neg logit contributions | |
lair | -0.007 |
mess | -0.007 |
cks | -0.007 |
floor | -0.007 |
pads | -0.007 |
Token | flock |
Token ID | 29556 |
Feature activation | -4.13e-01 |
Pos logit contributions | |
ted | +0.022 |
ar | +0.022 |
plunge | +0.021 |
help | +0.021 |
reward | +0.021 |
Neg logit contributions | |
mess | -0.019 |
lair | -0.019 |
floor | -0.018 |
frown | -0.018 |
web | -0.017 |
Token | of |
Token ID | 286 |
Feature activation | -7.24e-01 |
Pos logit contributions | |
reward | +0.019 |
help | +0.017 |
ar | +0.016 |
plunge | +0.016 |
Sure | +0.016 |
Neg logit contributions | |
cks | -0.043 |
lair | -0.042 |
floor | -0.042 |
mess | -0.041 |
pads | -0.041 |
Token | birds |
Token ID | 10087 |
Feature activation | +7.35e-01 |
Pos logit contributions | |
ar | +0.062 |
reward | +0.061 |
less | +0.057 |
help | +0.057 |
Sure | +0.057 |
Neg logit contributions | |
lair | -0.050 |
mess | -0.049 |
cks | -0.048 |
floor | -0.047 |
pads | -0.047 |
Token | flew |
Token ID | 13112 |
Feature activation | +3.59e-01 |
Pos logit contributions | |
lair | +0.057 |
cks | +0.055 |
mess | +0.053 |
floor | +0.052 |
pads | +0.052 |
Neg logit contributions | |
help | -0.057 |
reward | -0.055 |
ar | -0.055 |
plunge | -0.055 |
ted | -0.053 |
Token | out |
Token ID | 503 |
Feature activation | -9.66e-01 |
Pos logit contributions | |
pads | +0.028 |
cks | +0.028 |
lair | +0.028 |
mess | +0.027 |
necks | +0.026 |
Neg logit contributions | |
reward | -0.027 |
help | -0.025 |
ar | -0.025 |
plunge | -0.024 |
Enjoy | -0.023 |
Token | of |
Token ID | 286 |
Feature activation | -7.98e-01 |
Pos logit contributions | |
help | +0.043 |
reward | +0.043 |
ar | +0.041 |
plunge | +0.040 |
bye | +0.037 |
Neg logit contributions | |
lair | -0.105 |
cks | -0.104 |
mess | -0.102 |
pads | -0.100 |
floor | -0.100 |
Token | could |
Token ID | 714 |
Feature activation | -3.64e-01 |
Pos logit contributions | |
reward | +0.045 |
help | +0.045 |
ar | +0.044 |
plunge | +0.044 |
bye | +0.042 |
Neg logit contributions | |
lair | -0.043 |
cks | -0.043 |
mess | -0.041 |
pads | -0.041 |
floor | -0.040 |
Token | have |
Token ID | 423 |
Feature activation | +2.26e-01 |
Pos logit contributions | |
ar | +0.023 |
reward | +0.022 |
plunge | +0.022 |
ted | +0.022 |
help | +0.022 |
Neg logit contributions | |
lair | -0.031 |
cks | -0.031 |
mess | -0.030 |
pads | -0.030 |
floor | -0.029 |
Token | a |
Token ID | 257 |
Feature activation | -4.26e-01 |
Pos logit contributions | |
cks | +0.020 |
lair | +0.020 |
mess | +0.020 |
pads | +0.020 |
necks | +0.019 |
Neg logit contributions | |
reward | -0.013 |
ar | -0.013 |
help | -0.013 |
plunge | -0.013 |
bye | -0.013 |
Token | turn |
Token ID | 1210 |
Feature activation | +5.38e-01 |
Pos logit contributions | |
ar | +0.031 |
help | +0.031 |
reward | +0.030 |
plunge | +0.030 |
ted | +0.030 |
Neg logit contributions | |
lair | -0.034 |
mess | -0.033 |
cks | -0.032 |
pads | -0.032 |
floor | -0.031 |
Token | and |
Token ID | 290 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
lair | +0.048 |
cks | +0.047 |
mess | +0.046 |
pads | +0.046 |
floor | +0.045 |
Neg logit contributions | |
reward | -0.034 |
help | -0.033 |
ar | -0.032 |
plunge | -0.032 |
bye | -0.031 |
Token | Jane |
Token ID | 12091 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
lair | +0.016 |
mess | +0.016 |
cks | +0.016 |
pads | +0.016 |
floor | +0.015 |
Neg logit contributions | |
ar | -0.011 |
reward | -0.011 |
plunge | -0.011 |
less | -0.010 |
help | -0.010 |
Token | said |
Token ID | 531 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
lair | +0.136 |
cks | +0.134 |
mess | +0.128 |
floor | +0.125 |
pads | +0.123 |
Neg logit contributions | |
help | -0.103 |
reward | -0.101 |
ar | -0.099 |
Sure | -0.093 |
f | -0.092 |
Token | yes |
Token ID | 3763 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
plunge | +0.007 |
reward | +0.006 |
ar | +0.006 |
bye | +0.006 |
ted | +0.006 |
Neg logit contributions | |
neck | -0.015 |
pads | -0.015 |
frown | -0.015 |
goals | -0.015 |
lair | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
help | +0.099 |
reward | +0.098 |
ar | +0.095 |
plunge | +0.095 |
bye | +0.090 |
Neg logit contributions | |
lair | -0.126 |
cks | -0.124 |
mess | -0.121 |
pads | -0.119 |
floor | -0.118 |
Token | |
Token ID | 220 |
Feature activation | -4.47e-01 |
Pos logit contributions | |
lair | +0.012 |
cks | +0.012 |
mess | +0.012 |
pads | +0.012 |
floor | +0.012 |
Neg logit contributions | |
help | -0.011 |
reward | -0.011 |
ar | -0.011 |
plunge | -0.010 |
bye | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
help | +0.030 |
reward | +0.029 |
ar | +0.029 |
plunge | +0.028 |
bye | +0.027 |
Neg logit contributions | |
lair | -0.039 |
cks | -0.038 |
mess | -0.037 |
pads | -0.037 |
floor | -0.036 |
Token | it |
Token ID | 340 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
lair | +0.060 |
floor | +0.057 |
mess | +0.057 |
cks | +0.057 |
pads | +0.056 |
Neg logit contributions | |
reward | -0.049 |
ar | -0.049 |
help | -0.049 |
plunge | -0.048 |
less | -0.047 |
Token | was |
Token ID | 373 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
help | +0.062 |
reward | +0.061 |
ar | +0.061 |
plunge | +0.058 |
bye | +0.055 |
Neg logit contributions | |
cks | -0.104 |
lair | -0.102 |
pads | -0.097 |
necks | -0.097 |
llers | -0.094 |
Token | too |
Token ID | 1165 |
Feature activation | -3.38e-01 |
Pos logit contributions | |
help | +0.008 |
reward | +0.007 |
less | +0.006 |
plunge | +0.006 |
ar | +0.006 |
Neg logit contributions | |
cks | -0.017 |
necks | -0.016 |
lair | -0.016 |
wrists | -0.016 |
pads | -0.016 |
Token | wind |
Token ID | 2344 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
help | +0.026 |
reward | +0.025 |
ar | +0.025 |
plunge | +0.024 |
bye | +0.024 |
Neg logit contributions | |
lair | -0.026 |
cks | -0.026 |
mess | -0.025 |
pads | -0.025 |
floor | -0.024 |
Token | y |
Token ID | 88 |
Feature activation | +6.04e-01 |
Pos logit contributions | |
lair | +0.010 |
cks | +0.009 |
mess | +0.009 |
necks | +0.009 |
floor | +0.009 |
Neg logit contributions | |
help | -0.010 |
less | -0.010 |
ar | -0.010 |
reward | -0.010 |
plunge | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
lair | +0.057 |
cks | +0.057 |
mess | +0.054 |
necks | +0.053 |
floor | +0.053 |
Neg logit contributions | |
help | -0.036 |
reward | -0.035 |
ar | -0.034 |
plunge | -0.033 |
less | -0.032 |
Token | the |
Token ID | 262 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
help | +0.010 |
reward | +0.009 |
plunge | +0.008 |
ted | +0.008 |
ar | +0.008 |
Neg logit contributions | |
lair | -0.013 |
cks | -0.013 |
necks | -0.012 |
pads | -0.012 |
opl | -0.011 |
Token | waves |
Token ID | 9813 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
ar | +0.002 |
reward | +0.002 |
Sure | +0.002 |
ted | +0.002 |
less | +0.002 |
Neg logit contributions | |
lair | -0.002 |
floor | -0.002 |
mess | -0.002 |
neck | -0.002 |
frown | -0.002 |
Token | were |
Token ID | 547 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
reward | +0.025 |
help | +0.024 |
plunge | +0.024 |
ar | +0.024 |
Sure | +0.024 |
Neg logit contributions | |
cks | -0.027 |
mess | -0.027 |
lair | -0.027 |
pads | -0.027 |
floor | -0.026 |
Token | too |
Token ID | 1165 |
Feature activation | -3.32e-01 |
Pos logit contributions | |
help | +0.017 |
less | +0.015 |
reward | +0.015 |
plunge | +0.014 |
ted | +0.013 |
Neg logit contributions | |
lair | -0.021 |
cks | -0.020 |
necks | -0.020 |
pads | -0.019 |
wrists | -0.019 |
Token | big |
Token ID | 1263 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
help | +0.026 |
reward | +0.025 |
ar | +0.025 |
plunge | +0.024 |
bye | +0.024 |
Neg logit contributions | |
lair | -0.025 |
cks | -0.025 |
mess | -0.024 |
pads | -0.024 |
floor | -0.023 |
Token | mom |
Token ID | 1995 |
Feature activation | -4.76e-03 |
Pos logit contributions | |
mess | +0.071 |
floor | +0.068 |
lair | +0.068 |
frown | +0.067 |
web | +0.064 |
Neg logit contributions | |
ar | -0.083 |
plunge | -0.077 |
ted | -0.074 |
Sure | -0.073 |
f | -0.073 |
Token | my |
Token ID | 1820 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
help | +0.000 |
reward | +0.000 |
ar | +0.000 |
bye | +0.000 |
Sure | +0.000 |
Neg logit contributions | |
lair | -0.000 |
cks | -0.000 |
necks | -0.000 |
pads | -0.000 |
goals | -0.000 |
Token | asked |
Token ID | 1965 |
Feature activation | +6.99e-01 |
Pos logit contributions | |
help | +0.018 |
reward | +0.017 |
ar | +0.017 |
plunge | +0.017 |
bye | +0.016 |
Neg logit contributions | |
lair | -0.020 |
cks | -0.020 |
mess | -0.019 |
pads | -0.019 |
floor | -0.018 |
Token | her |
Token ID | 607 |
Feature activation | +9.67e-01 |
Pos logit contributions | |
mess | +0.043 |
floor | +0.038 |
frown | +0.037 |
pads | +0.032 |
cks | +0.031 |
Neg logit contributions | |
plunge | -0.071 |
ar | -0.068 |
reward | -0.067 |
iest | -0.064 |
bye | -0.063 |
Token | to |
Token ID | 284 |
Feature activation | -8.72e-02 |
Pos logit contributions | |
mess | +0.073 |
lair | +0.071 |
frown | +0.069 |
floor | +0.067 |
neck | +0.064 |
Neg logit contributions | |
ar | -0.076 |
plunge | -0.074 |
ted | -0.073 |
bye | -0.070 |
reward | -0.065 |
Token | help |
Token ID | 1037 |
Feature activation | -2.10e+00 |
Pos logit contributions | |
ted | +0.007 |
ar | +0.007 |
ed | +0.006 |
less | +0.006 |
rav | +0.006 |
Neg logit contributions | |
floor | -0.007 |
mess | -0.006 |
frown | -0.006 |
neck | -0.006 |
lair | -0.006 |
Token | with |
Token ID | 351 |
Feature activation | +5.22e-01 |
Pos logit contributions | |
ar | +0.186 |
plunge | +0.182 |
bye | +0.181 |
reward | +0.176 |
ted | +0.174 |
Neg logit contributions | |
mess | -0.142 |
floor | -0.140 |
cks | -0.129 |
frown | -0.125 |
neck | -0.121 |
Token | the |
Token ID | 262 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
mess | +0.039 |
floor | +0.036 |
frown | +0.035 |
cks | +0.033 |
uneven | +0.033 |
Neg logit contributions | |
ar | -0.047 |
bye | -0.045 |
plunge | -0.043 |
Sure | -0.041 |
rav | -0.040 |
Token | grocery |
Token ID | 16918 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
ar | +0.015 |
bye | +0.014 |
reward | +0.014 |
ted | +0.014 |
Sure | +0.014 |
Neg logit contributions | |
mess | -0.015 |
lair | -0.015 |
floor | -0.015 |
frown | -0.014 |
cks | -0.014 |
Token | shopping |
Token ID | 9735 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
ar | +0.011 |
plunge | +0.011 |
bye | +0.010 |
ted | +0.010 |
reward | +0.010 |
Neg logit contributions | |
mess | -0.012 |
floor | -0.012 |
lair | -0.011 |
pads | -0.011 |
cks | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -3.06e-01 |
Pos logit contributions | |
plunge | +0.146 |
ar | +0.146 |
ted | +0.141 |
bye | +0.137 |
Bub | +0.136 |
Neg logit contributions | |
floor | -0.112 |
mess | -0.109 |
lair | -0.095 |
pads | -0.093 |
sizes | -0.091 |
Token | " |
Token ID | 1 |
Feature activation | -9.04e-01 |
Pos logit contributions | |
cks | +0.105 |
lair | +0.102 |
mess | +0.099 |
llers | +0.097 |
pads | +0.097 |
Neg logit contributions | |
help | -0.076 |
reward | -0.073 |
ar | -0.072 |
plunge | -0.070 |
bye | -0.069 |
Token | |
Token ID | 220 |
Feature activation | -7.05e-01 |
Pos logit contributions | |
reward | +0.065 |
help | +0.064 |
ar | +0.063 |
plunge | +0.063 |
Sure | +0.058 |
Neg logit contributions | |
lair | -0.073 |
pads | -0.072 |
cks | -0.071 |
mess | -0.069 |
floor | -0.069 |
Token |
|
Token ID | 198 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
help | +0.048 |
reward | +0.048 |
ar | +0.046 |
plunge | +0.046 |
bye | +0.044 |
Neg logit contributions | |
lair | -0.060 |
cks | -0.059 |
mess | -0.057 |
pads | -0.057 |
floor | -0.056 |
Token |
|
Token ID | 198 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
reward | +0.013 |
help | +0.013 |
plunge | +0.013 |
less | +0.013 |
ar | +0.012 |
Neg logit contributions | |
lair | -0.011 |
cks | -0.011 |
floor | -0.010 |
pads | -0.010 |
llers | -0.010 |
Token | Just |
Token ID | 5703 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
help | +0.009 |
reward | +0.009 |
ar | +0.009 |
plunge | +0.009 |
Sure | +0.009 |
Neg logit contributions | |
lair | -0.008 |
cks | -0.007 |
mess | -0.007 |
pads | -0.007 |
floor | -0.007 |
Token | then |
Token ID | 788 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
reward | +0.094 |
ar | +0.090 |
less | +0.090 |
plunge | +0.089 |
Sure | +0.089 |
Neg logit contributions | |
lair | -0.106 |
cks | -0.106 |
floor | -0.102 |
pads | -0.102 |
neck | -0.101 |
Token | , |
Token ID | 11 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
reward | +0.121 |
plunge | +0.120 |
bye | +0.118 |
Sure | +0.116 |
less | +0.112 |
Neg logit contributions | |
lair | -0.139 |
frown | -0.136 |
neck | -0.128 |
cks | -0.123 |
mess | -0.123 |
Token | the |
Token ID | 262 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
rav | +0.018 |
ar | +0.018 |
bye | +0.018 |
less | +0.017 |
ay | +0.017 |
Neg logit contributions | |
frown | -0.009 |
neck | -0.008 |
not | -0.007 |
also | -0.007 |
mess | -0.007 |
Token | rabbit |
Token ID | 22746 |
Feature activation | -8.41e-01 |
Pos logit contributions | |
lair | +0.008 |
floor | +0.008 |
mess | +0.008 |
pads | +0.008 |
neck | +0.007 |
Neg logit contributions | |
ar | -0.008 |
reward | -0.008 |
f | -0.007 |
less | -0.007 |
help | -0.007 |
Token | heard |
Token ID | 2982 |
Feature activation | -8.69e-03 |
Pos logit contributions | |
help | +0.061 |
ar | +0.060 |
reward | +0.059 |
plunge | +0.056 |
ted | +0.055 |
Neg logit contributions | |
cks | -0.071 |
lair | -0.069 |
mess | -0.066 |
floor | -0.064 |
sizes | -0.064 |
Token | a |
Token ID | 257 |
Feature activation | -5.52e-01 |
Pos logit contributions | |
reward | +0.001 |
plunge | +0.001 |
ar | +0.001 |
Sure | +0.001 |
bye | +0.001 |
Neg logit contributions | |
lair | -0.001 |
frown | -0.001 |
floor | -0.001 |
mess | -0.001 |
cks | -0.001 |
Token | it |
Token ID | 340 |
Feature activation | -8.98e-01 |
Pos logit contributions | |
bye | +0.026 |
ar | +0.025 |
plunge | +0.024 |
less | +0.024 |
ival | +0.023 |
Neg logit contributions | |
frown | -0.020 |
cks | -0.020 |
neck | -0.019 |
wrists | -0.018 |
teeth | -0.018 |
Token | with |
Token ID | 351 |
Feature activation | +6.23e-01 |
Pos logit contributions | |
reward | +0.065 |
ar | +0.064 |
help | +0.064 |
plunge | +0.062 |
bye | +0.060 |
Neg logit contributions | |
lair | -0.071 |
mess | -0.069 |
cks | -0.069 |
floor | -0.069 |
pads | -0.068 |
Token | both |
Token ID | 1111 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
cks | +0.047 |
mess | +0.046 |
frown | +0.045 |
lair | +0.045 |
pads | +0.045 |
Neg logit contributions | |
ar | -0.049 |
reward | -0.046 |
plunge | -0.046 |
bye | -0.045 |
Sure | -0.044 |
Token | hands |
Token ID | 2832 |
Feature activation | -6.49e-01 |
Pos logit contributions | |
help | +0.010 |
reward | +0.009 |
plunge | +0.008 |
less | +0.008 |
Sure | +0.008 |
Neg logit contributions | |
lair | -0.013 |
cks | -0.012 |
llers | -0.011 |
pads | -0.011 |
opl | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | +2.70e-02 |
Pos logit contributions | |
help | +0.040 |
reward | +0.039 |
ar | +0.037 |
plunge | +0.037 |
less | +0.036 |
Neg logit contributions | |
lair | -0.061 |
cks | -0.060 |
mess | -0.058 |
pads | -0.057 |
floor | -0.056 |
Token | It |
Token ID | 632 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
lair | +0.002 |
cks | +0.002 |
mess | +0.002 |
pads | +0.002 |
floor | +0.002 |
Neg logit contributions | |
help | -0.002 |
reward | -0.002 |
ar | -0.002 |
plunge | -0.002 |
Sure | -0.002 |
Token | was |
Token ID | 373 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
help | +0.108 |
reward | +0.100 |
ar | +0.094 |
Sure | +0.092 |
ted | +0.091 |
Neg logit contributions | |
lair | -0.183 |
cks | -0.180 |
opl | -0.170 |
pads | -0.164 |
goals | -0.161 |
Token | too |
Token ID | 1165 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
lair | +0.018 |
cks | +0.017 |
pads | +0.016 |
necks | +0.016 |
llers | +0.016 |
Neg logit contributions | |
help | -0.013 |
reward | -0.013 |
less | -0.011 |
plunge | -0.011 |
ar | -0.011 |
Token | heavy |
Token ID | 4334 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
help | +0.008 |
reward | +0.007 |
less | +0.007 |
ar | +0.007 |
bye | +0.007 |
Neg logit contributions | |
lair | -0.010 |
cks | -0.010 |
pads | -0.009 |
necks | -0.009 |
llers | -0.009 |
Token | for |
Token ID | 329 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
lair | +0.030 |
cks | +0.030 |
mess | +0.029 |
pads | +0.029 |
floor | +0.029 |
Neg logit contributions | |
reward | -0.013 |
help | -0.013 |
ar | -0.013 |
plunge | -0.013 |
Sure | -0.012 |
Token | her |
Token ID | 607 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
ar | +0.144 |
bye | +0.141 |
Sure | +0.139 |
iest | +0.134 |
plunge | +0.134 |
Neg logit contributions | |
mess | -0.103 |
floor | -0.099 |
cks | -0.094 |
neck | -0.093 |
frown | -0.091 |
Token | pulled |
Token ID | 5954 |
Feature activation | -6.58e-01 |
Pos logit contributions | |
lair | +0.017 |
cks | +0.017 |
mess | +0.016 |
pads | +0.016 |
floor | +0.016 |
Neg logit contributions | |
help | -0.017 |
reward | -0.016 |
ar | -0.016 |
plunge | -0.016 |
bye | -0.015 |
Token | a |
Token ID | 257 |
Feature activation | -6.18e-01 |
Pos logit contributions | |
reward | +0.041 |
bye | +0.041 |
ar | +0.040 |
Sure | +0.040 |
help | +0.038 |
Neg logit contributions | |
mess | -0.059 |
lair | -0.058 |
cks | -0.058 |
floor | -0.057 |
pads | -0.056 |
Token | giant |
Token ID | 6175 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
help | +0.048 |
ar | +0.048 |
reward | +0.047 |
bye | +0.047 |
ted | +0.047 |
Neg logit contributions | |
mess | -0.045 |
lair | -0.045 |
floor | -0.043 |
cks | -0.043 |
pads | -0.041 |
Token | carrot |
Token ID | 40562 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
cks | +0.155 |
lair | +0.152 |
pads | +0.146 |
llers | +0.145 |
necks | +0.143 |
Neg logit contributions | |
help | -0.153 |
reward | -0.150 |
ar | -0.141 |
plunge | -0.140 |
less | -0.132 |
Token | . |
Token ID | 13 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
help | +0.008 |
reward | +0.008 |
ar | +0.008 |
plunge | +0.008 |
less | +0.007 |
Neg logit contributions | |
cks | -0.012 |
lair | -0.012 |
mess | -0.011 |
llers | -0.011 |
pads | -0.011 |
Token | It |
Token ID | 632 |
Feature activation | -2.10e+00 |
Pos logit contributions | |
reward | +0.019 |
help | +0.018 |
ar | +0.018 |
plunge | +0.018 |
less | +0.018 |
Neg logit contributions | |
lair | -0.016 |
cks | -0.016 |
floor | -0.015 |
pads | -0.015 |
llers | -0.015 |
Token | was |
Token ID | 373 |
Feature activation | -3.27e-02 |
Pos logit contributions | |
help | +0.129 |
reward | +0.124 |
ar | +0.123 |
plunge | +0.115 |
ted | +0.113 |
Neg logit contributions | |
lair | -0.200 |
cks | -0.196 |
pads | -0.186 |
mess | -0.186 |
necks | -0.185 |
Token | bigger |
Token ID | 5749 |
Feature activation | +3.90e-01 |
Pos logit contributions | |
help | +0.002 |
reward | +0.002 |
plunge | +0.002 |
ar | +0.002 |
less | +0.002 |
Neg logit contributions | |
cks | -0.003 |
lair | -0.003 |
pads | -0.003 |
llers | -0.003 |
mess | -0.003 |
Token | than |
Token ID | 621 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
cks | +0.031 |
llers | +0.028 |
lair | +0.027 |
opl | +0.026 |
frown | +0.026 |
Neg logit contributions | |
help | -0.030 |
ar | -0.029 |
reward | -0.029 |
less | -0.028 |
ted | -0.025 |
Token | the |
Token ID | 262 |
Feature activation | +6.87e-02 |
Pos logit contributions | |
lair | +0.202 |
cks | +0.194 |
floor | +0.194 |
sizes | +0.191 |
mess | +0.189 |
Neg logit contributions | |
Sure | -0.168 |
ar | -0.168 |
bye | -0.164 |
reward | -0.163 |
help | -0.154 |
Token | other |
Token ID | 584 |
Feature activation | -5.30e-01 |
Pos logit contributions | |
cks | +0.006 |
lair | +0.005 |
llers | +0.005 |
pads | +0.005 |
mess | +0.005 |
Neg logit contributions | |
help | -0.005 |
reward | -0.005 |
plunge | -0.005 |
ar | -0.005 |
bye | -0.004 |
Token | She |
Token ID | 1375 |
Feature activation | +3.03e-01 |
Pos logit contributions | |
help | +0.020 |
reward | +0.020 |
ar | +0.019 |
plunge | +0.019 |
Sure | +0.018 |
Neg logit contributions | |
lair | -0.020 |
cks | -0.020 |
mess | -0.019 |
pads | -0.019 |
necks | -0.018 |
Token | saw |
Token ID | 2497 |
Feature activation | -9.74e-02 |
Pos logit contributions | |
lair | +0.024 |
cks | +0.024 |
mess | +0.022 |
pads | +0.022 |
llers | +0.021 |
Neg logit contributions | |
help | -0.024 |
reward | -0.023 |
ar | -0.023 |
bye | -0.022 |
Sure | -0.022 |
Token | a |
Token ID | 257 |
Feature activation | -6.27e-01 |
Pos logit contributions | |
reward | +0.005 |
ar | +0.005 |
help | +0.005 |
plunge | +0.005 |
bye | +0.005 |
Neg logit contributions | |
lair | -0.010 |
cks | -0.010 |
mess | -0.010 |
floor | -0.009 |
pads | -0.009 |
Token | plane |
Token ID | 6614 |
Feature activation | -2.14e-01 |
Pos logit contributions | |
ar | +0.050 |
help | +0.050 |
reward | +0.049 |
plunge | +0.049 |
ted | +0.048 |
Neg logit contributions | |
mess | -0.045 |
lair | -0.045 |
floor | -0.043 |
cks | -0.042 |
frown | -0.041 |
Token | pass |
Token ID | 1208 |
Feature activation | +2.38e-01 |
Pos logit contributions | |
help | +0.011 |
reward | +0.010 |
ar | +0.010 |
bye | +0.010 |
plunge | +0.010 |
Neg logit contributions | |
lair | -0.022 |
pads | -0.021 |
cks | -0.021 |
necks | -0.021 |
mess | -0.021 |
Token | by |
Token ID | 416 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
cks | +0.020 |
pads | +0.019 |
lair | +0.019 |
floor | +0.019 |
mess | +0.019 |
Neg logit contributions | |
reward | -0.017 |
help | -0.017 |
ar | -0.016 |
plunge | -0.016 |
Sure | -0.015 |
Token | in |
Token ID | 287 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
help | +0.116 |
reward | +0.114 |
ar | +0.110 |
plunge | +0.109 |
bye | +0.105 |
Neg logit contributions | |
lair | -0.140 |
cks | -0.138 |
mess | -0.133 |
pads | -0.131 |
floor | -0.129 |
Token | the |
Token ID | 262 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
reward | +0.010 |
help | +0.010 |
ar | +0.010 |
plunge | +0.010 |
bye | +0.009 |
Neg logit contributions | |
lair | -0.016 |
cks | -0.016 |
mess | -0.016 |
floor | -0.016 |
pads | -0.016 |
Token | sky |
Token ID | 6766 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
ar | +0.011 |
reward | +0.011 |
help | +0.011 |
plunge | +0.011 |
less | +0.011 |
Neg logit contributions | |
lair | -0.008 |
mess | -0.008 |
floor | -0.008 |
cks | -0.008 |
pads | -0.008 |
Token | one |
Token ID | 530 |
Feature activation | -7.97e-01 |
Pos logit contributions | |
lair | +0.017 |
cks | +0.017 |
mess | +0.016 |
llers | +0.016 |
pads | +0.015 |
Neg logit contributions | |
help | -0.008 |
bye | -0.008 |
reward | -0.008 |
less | -0.007 |
ar | -0.007 |
Token | day |
Token ID | 1110 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
ar | +0.044 |
help | +0.044 |
reward | +0.044 |
plunge | +0.043 |
ted | +0.041 |
Neg logit contributions | |
cks | -0.076 |
lair | -0.076 |
mess | -0.074 |
pads | -0.073 |
floor | -0.073 |
Token | ball |
Token ID | 2613 |
Feature activation | +6.79e-01 |
Pos logit contributions | |
help | +0.185 |
reward | +0.185 |
ar | +0.179 |
plunge | +0.177 |
bye | +0.171 |
Neg logit contributions | |
lair | -0.181 |
cks | -0.180 |
mess | -0.170 |
pads | -0.168 |
floor | -0.165 |
Token | . |
Token ID | 13 |
Feature activation | +7.81e-02 |
Pos logit contributions | |
cks | +0.063 |
lair | +0.063 |
mess | +0.060 |
pads | +0.059 |
necks | +0.058 |
Neg logit contributions | |
help | -0.042 |
reward | -0.041 |
ar | -0.040 |
plunge | -0.039 |
bye | -0.038 |
Token | The |
Token ID | 383 |
Feature activation | -9.87e-01 |
Pos logit contributions | |
lair | +0.007 |
cks | +0.007 |
mess | +0.006 |
pads | +0.006 |
floor | +0.006 |
Neg logit contributions | |
reward | -0.005 |
help | -0.005 |
ar | -0.005 |
plunge | -0.005 |
bye | -0.005 |
Token | 3 |
Token ID | 513 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
ar | +0.072 |
ted | +0.071 |
plunge | +0.071 |
help | +0.068 |
less | +0.068 |
Neg logit contributions | |
mess | -0.078 |
lair | -0.078 |
floor | -0.077 |
frown | -0.073 |
neck | -0.073 |
Token | - |
Token ID | 12 |
Feature activation | -3.48e+00 |
Pos logit contributions | |
help | +0.060 |
reward | +0.060 |
ar | +0.057 |
plunge | +0.056 |
bye | +0.053 |
Neg logit contributions | |
lair | -0.113 |
cks | -0.112 |
mess | -0.109 |
pads | -0.108 |
floor | -0.107 |
Token | year |
Token ID | 1941 |
Feature activation | -6.70e+00 |
Pos logit contributions | |
plunge | +0.201 |
reward | +0.199 |
help | +0.189 |
ar | +0.161 |
Bub | +0.143 |
Neg logit contributions | |
cks | -0.324 |
ents | -0.308 |
pads | -0.304 |
aron | -0.291 |
thia | -0.286 |
Token | - |
Token ID | 12 |
Feature activation | -3.47e+00 |
Pos logit contributions | |
help | +0.288 |
reward | +0.282 |
plunge | +0.277 |
ar | +0.264 |
Sure | +0.258 |
Neg logit contributions | |
lair | -0.737 |
cks | -0.727 |
mess | -0.713 |
floor | -0.701 |
pads | -0.699 |
Token | old |
Token ID | 727 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
plunge | +0.331 |
reward | +0.320 |
help | +0.315 |
ar | +0.289 |
Bub | +0.283 |
Neg logit contributions | |
cks | -0.196 |
pads | -0.190 |
opl | -0.178 |
ents | -0.174 |
lair | -0.173 |
Token | was |
Token ID | 373 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
plunge | +0.079 |
help | +0.079 |
reward | +0.078 |
ar | +0.076 |
less | +0.074 |
Neg logit contributions | |
lair | -0.072 |
mess | -0.069 |
pads | -0.068 |
cks | -0.068 |
floor | -0.067 |
Token | so |
Token ID | 523 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
lair | +0.022 |
cks | +0.022 |
mess | +0.021 |
necks | +0.021 |
floor | +0.020 |
Neg logit contributions | |
help | -0.016 |
reward | -0.015 |
less | -0.014 |
plunge | -0.014 |
ar | -0.014 |
Token | excited |
Token ID | 6568 |
Feature activation | +5.51e-01 |
Pos logit contributions | |
lair | +0.073 |
cks | +0.071 |
mess | +0.067 |
pads | +0.067 |
floor | +0.066 |
Neg logit contributions | |
help | -0.086 |
reward | -0.085 |
ar | -0.082 |
plunge | -0.082 |
bye | -0.080 |
Token | . |
Token ID | 13 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
help | +0.031 |
reward | +0.030 |
ar | +0.029 |
plunge | +0.029 |
bye | +0.028 |
Neg logit contributions | |
lair | -0.040 |
cks | -0.039 |
mess | -0.038 |
pads | -0.037 |
floor | -0.037 |
Token | She |
Token ID | 1375 |
Feature activation | +4.45e-01 |
Pos logit contributions | |
help | +0.017 |
reward | +0.017 |
ar | +0.016 |
plunge | +0.016 |
bye | +0.016 |
Neg logit contributions | |
cks | -0.017 |
lair | -0.017 |
mess | -0.016 |
pads | -0.016 |
necks | -0.015 |
Token | tried |
Token ID | 3088 |
Feature activation | -6.84e-01 |
Pos logit contributions | |
cks | +0.038 |
lair | +0.037 |
pads | +0.035 |
mess | +0.035 |
opl | +0.034 |
Neg logit contributions | |
help | -0.032 |
reward | -0.032 |
ar | -0.031 |
Sure | -0.029 |
f | -0.029 |
Token | to |
Token ID | 284 |
Feature activation | +4.95e-03 |
Pos logit contributions | |
help | +0.040 |
reward | +0.036 |
plunge | +0.033 |
ar | +0.033 |
less | +0.032 |
Neg logit contributions | |
cks | -0.072 |
lair | -0.070 |
pads | -0.067 |
sizes | -0.064 |
necks | -0.064 |
Token | sh |
Token ID | 427 |
Feature activation | +5.28e-01 |
Pos logit contributions | |
lair | +0.000 |
cks | +0.000 |
pads | +0.000 |
mess | +0.000 |
llers | +0.000 |
Neg logit contributions | |
help | -0.000 |
reward | -0.000 |
plunge | -0.000 |
ar | -0.000 |
Sure | -0.000 |
Token | oo |
Token ID | 2238 |
Feature activation | -4.94e+00 |
Pos logit contributions | |
cks | +0.049 |
lair | +0.048 |
mess | +0.046 |
llers | +0.045 |
pads | +0.044 |
Neg logit contributions | |
reward | -0.033 |
help | -0.033 |
plunge | -0.031 |
ar | -0.030 |
bye | -0.030 |
Token | him |
Token ID | 683 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
reward | +0.320 |
bye | +0.299 |
plunge | +0.294 |
help | +0.291 |
Sure | +0.289 |
Neg logit contributions | |
cks | -0.429 |
mess | -0.423 |
floor | -0.413 |
llers | -0.409 |
pads | -0.404 |
Token | away |
Token ID | 1497 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
help | +0.085 |
reward | +0.084 |
ar | +0.081 |
plunge | +0.080 |
bye | +0.075 |
Neg logit contributions | |
lair | -0.150 |
cks | -0.149 |
mess | -0.144 |
pads | -0.142 |
floor | -0.141 |
Token | , |
Token ID | 11 |
Feature activation | -4.37e-01 |
Pos logit contributions | |
help | +0.091 |
reward | +0.089 |
ar | +0.083 |
plunge | +0.080 |
less | +0.076 |
Neg logit contributions | |
cks | -0.221 |
lair | -0.220 |
pads | -0.211 |
mess | -0.207 |
necks | -0.206 |
Token | but |
Token ID | 475 |
Feature activation | +7.93e-01 |
Pos logit contributions | |
help | +0.035 |
reward | +0.032 |
Sure | +0.029 |
plunge | +0.029 |
ar | +0.029 |
Neg logit contributions | |
cks | -0.037 |
lair | -0.037 |
pads | -0.034 |
llers | -0.033 |
opl | -0.033 |
Token | it |
Token ID | 340 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
lair | +0.064 |
mess | +0.064 |
floor | +0.063 |
frown | +0.062 |
cks | +0.061 |
Neg logit contributions | |
ar | -0.054 |
plunge | -0.053 |
reward | -0.053 |
bye | -0.052 |
help | -0.051 |
Token | ate |
Token ID | 15063 |
Feature activation | -1.93e-01 |
Pos logit contributions | |
cks | +0.019 |
lair | +0.019 |
necks | +0.018 |
pads | +0.018 |
goals | +0.017 |
Neg logit contributions | |
help | -0.014 |
reward | -0.013 |
ar | -0.013 |
bye | -0.012 |
f | -0.012 |
Token | too |
Token ID | 1165 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
ar | +0.013 |
reward | +0.012 |
help | +0.012 |
plunge | +0.012 |
bye | +0.012 |
Neg logit contributions | |
mess | -0.017 |
cks | -0.017 |
floor | -0.016 |
lair | -0.016 |
pads | -0.016 |
Token | much |
Token ID | 881 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
plunge | +0.015 |
ar | +0.015 |
reward | +0.015 |
help | +0.014 |
ted | +0.014 |
Neg logit contributions | |
mess | -0.011 |
lair | -0.011 |
cks | -0.011 |
floor | -0.011 |
pads | -0.010 |
Token | ice |
Token ID | 4771 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
ar | +0.008 |
plunge | +0.008 |
reward | +0.008 |
help | +0.008 |
bye | +0.008 |
Neg logit contributions | |
mess | -0.009 |
cks | -0.009 |
floor | -0.009 |
pads | -0.008 |
lair | -0.008 |
Token | - |
Token ID | 12 |
Feature activation | -3.52e+00 |
Pos logit contributions | |
plunge | +0.112 |
help | +0.111 |
reward | +0.110 |
ar | +0.103 |
Sure | +0.103 |
Neg logit contributions | |
cks | -0.175 |
mess | -0.174 |
pads | -0.170 |
floor | -0.170 |
lair | -0.168 |
Token | cream |
Token ID | 36277 |
Feature activation | -4.57e+00 |
Pos logit contributions | |
reward | +0.264 |
help | +0.256 |
plunge | +0.255 |
mate | +0.235 |
ar | +0.230 |
Neg logit contributions | |
cks | -0.278 |
llers | -0.275 |
pads | -0.265 |
mess | -0.254 |
necks | -0.253 |
Token | already |
Token ID | 1541 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
plunge | +0.362 |
help | +0.354 |
reward | +0.349 |
ar | +0.348 |
Sure | +0.335 |
Neg logit contributions | |
floor | -0.327 |
mess | -0.325 |
pads | -0.317 |
lair | -0.309 |
sizes | -0.303 |
Token | . |
Token ID | 13 |
Feature activation | +8.49e-03 |
Pos logit contributions | |
reward | +0.124 |
help | +0.123 |
ar | +0.121 |
plunge | +0.121 |
Sure | +0.119 |
Neg logit contributions | |
mess | -0.125 |
lair | -0.123 |
floor | -0.122 |
cks | -0.121 |
pads | -0.120 |
Token | Tim |
Token ID | 5045 |
Feature activation | -2.79e-01 |
Pos logit contributions | |
lair | +0.001 |
cks | +0.001 |
mess | +0.001 |
pads | +0.001 |
floor | +0.001 |
Neg logit contributions | |
help | -0.000 |
reward | -0.000 |
ar | -0.000 |
plunge | -0.000 |
Sure | -0.000 |
Token | my |
Token ID | 1820 |
Feature activation | -2.50e-02 |
Pos logit contributions | |
help | +0.015 |
Sure | +0.014 |
bye | +0.014 |
reward | +0.014 |
less | +0.013 |
Neg logit contributions | |
cks | -0.028 |
lair | -0.027 |
mesmer | -0.025 |
web | -0.025 |
wrists | -0.025 |
Token | got |
Token ID | 1392 |
Feature activation | +1.41e-03 |
Pos logit contributions | |
help | +0.001 |
Sure | +0.001 |
reward | +0.001 |
bye | +0.001 |
ar | +0.001 |
Neg logit contributions | |
cks | -0.003 |
lair | -0.003 |
opl | -0.003 |
sizes | -0.003 |
pads | -0.003 |
Token | kinds |
Token ID | 6982 |
Feature activation | +4.72e-01 |
Pos logit contributions | |
help | +0.007 |
reward | +0.006 |
ted | +0.006 |
plunge | +0.006 |
bye | +0.006 |
Neg logit contributions | |
lair | -0.020 |
cks | -0.020 |
necks | -0.019 |
pads | -0.019 |
mess | -0.019 |
Token | of |
Token ID | 286 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
lair | +0.055 |
cks | +0.054 |
mess | +0.054 |
floor | +0.053 |
pads | +0.053 |
Neg logit contributions | |
reward | -0.017 |
help | -0.016 |
plunge | -0.016 |
ar | -0.016 |
bye | -0.014 |
Token | delicious |
Token ID | 12625 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
help | +0.073 |
reward | +0.072 |
ar | +0.070 |
plunge | +0.069 |
bye | +0.066 |
Neg logit contributions | |
lair | -0.083 |
cks | -0.082 |
mess | -0.079 |
pads | -0.078 |
floor | -0.077 |
Token | ice |
Token ID | 4771 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
help | +0.011 |
reward | +0.011 |
ar | +0.010 |
plunge | +0.010 |
bye | +0.010 |
Neg logit contributions | |
lair | -0.015 |
cks | -0.015 |
mess | -0.015 |
pads | -0.014 |
floor | -0.014 |
Token | - |
Token ID | 12 |
Feature activation | -3.59e+00 |
Pos logit contributions | |
help | +0.093 |
reward | +0.092 |
plunge | +0.089 |
ar | +0.086 |
Sure | +0.083 |
Neg logit contributions | |
cks | -0.191 |
lair | -0.190 |
mess | -0.189 |
pads | -0.187 |
floor | -0.185 |
Token | cream |
Token ID | 36277 |
Feature activation | -4.56e+00 |
Pos logit contributions | |
reward | +0.267 |
help | +0.259 |
plunge | +0.257 |
mate | +0.236 |
ar | +0.229 |
Neg logit contributions | |
cks | -0.286 |
pads | -0.282 |
llers | -0.282 |
necks | -0.269 |
sizes | -0.268 |
Token | flavors |
Token ID | 17361 |
Feature activation | +7.75e-01 |
Pos logit contributions | |
help | +0.347 |
plunge | +0.347 |
reward | +0.344 |
ar | +0.337 |
Sure | +0.329 |
Neg logit contributions | |
floor | -0.337 |
mess | -0.337 |
pads | -0.334 |
lair | -0.334 |
cks | -0.329 |
Token | . |
Token ID | 13 |
Feature activation | +1.88e-01 |
Pos logit contributions | |
cks | +0.066 |
lair | +0.066 |
pads | +0.065 |
mess | +0.065 |
floor | +0.064 |
Neg logit contributions | |
help | -0.051 |
reward | -0.051 |
plunge | -0.050 |
ar | -0.049 |
Sure | -0.047 |
Token | " |
Token ID | 366 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
cks | +0.015 |
lair | +0.015 |
mess | +0.014 |
pads | +0.014 |
necks | +0.014 |
Neg logit contributions | |
help | -0.014 |
reward | -0.014 |
ar | -0.013 |
Sure | -0.013 |
plunge | -0.013 |
Token | Can |
Token ID | 6090 |
Feature activation | -4.93e-01 |
Pos logit contributions | |
lair | +0.011 |
cks | +0.011 |
mess | +0.010 |
pads | +0.010 |
floor | +0.010 |
Neg logit contributions | |
help | -0.008 |
reward | -0.008 |
ar | -0.008 |
plunge | -0.008 |
bye | -0.007 |
Token | I |
Token ID | 314 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
less | +0.033 |
ar | +0.033 |
reward | +0.033 |
plunge | +0.033 |
help | +0.031 |
Neg logit contributions | |
llers | -0.038 |
floor | -0.037 |
mess | -0.037 |
lair | -0.037 |
pads | -0.037 |
Token | Alex |
Token ID | 4422 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
bye | +0.007 |
reward | +0.007 |
less | +0.007 |
ar | +0.007 |
plunge | +0.007 |
Neg logit contributions | |
cks | -0.006 |
floor | -0.005 |
mess | -0.005 |
lair | -0.005 |
frown | -0.005 |
Token | for |
Token ID | 329 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
lair | +0.177 |
cks | +0.174 |
mess | +0.167 |
pads | +0.163 |
floor | +0.162 |
Neg logit contributions | |
help | -0.180 |
reward | -0.178 |
ar | -0.173 |
plunge | -0.171 |
bye | -0.165 |
Token | some |
Token ID | 617 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
ar | +0.105 |
bye | +0.104 |
ted | +0.097 |
reward | +0.096 |
plunge | +0.095 |
Neg logit contributions | |
mess | -0.131 |
cks | -0.127 |
lair | -0.127 |
floor | -0.127 |
neck | -0.124 |
Token | ice |
Token ID | 4771 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
ar | +0.016 |
bye | +0.015 |
reward | +0.015 |
Sure | +0.015 |
plunge | +0.015 |
Neg logit contributions | |
mess | -0.019 |
lair | -0.019 |
floor | -0.018 |
cks | -0.018 |
pads | -0.018 |
Token | - |
Token ID | 12 |
Feature activation | -3.61e+00 |
Pos logit contributions | |
reward | +0.125 |
plunge | +0.124 |
help | +0.122 |
ar | +0.117 |
Sure | +0.117 |
Neg logit contributions | |
cks | -0.149 |
pads | -0.149 |
mess | -0.148 |
lair | -0.147 |
floor | -0.146 |
Token | cream |
Token ID | 36277 |
Feature activation | -4.49e+00 |
Pos logit contributions | |
reward | +0.288 |
plunge | +0.270 |
help | +0.263 |
mate | +0.246 |
awaited | +0.231 |
Neg logit contributions | |
pads | -0.263 |
cks | -0.253 |
llers | -0.251 |
necks | -0.247 |
opl | -0.222 |
Token | . |
Token ID | 13 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
reward | +0.331 |
help | +0.328 |
ar | +0.326 |
plunge | +0.324 |
Sure | +0.309 |
Neg logit contributions | |
floor | -0.338 |
pads | -0.323 |
mess | -0.321 |
lair | -0.321 |
sizes | -0.302 |
Token | He |
Token ID | 679 |
Feature activation | +8.24e-02 |
Pos logit contributions | |
cks | +0.016 |
lair | +0.016 |
mess | +0.015 |
necks | +0.015 |
pads | +0.015 |
Neg logit contributions | |
help | -0.012 |
reward | -0.012 |
Sure | -0.012 |
plunge | -0.011 |
ar | -0.011 |
Token | smiled |
Token ID | 13541 |
Feature activation | -8.15e-01 |
Pos logit contributions | |
cks | +0.007 |
lair | +0.007 |
mess | +0.007 |
pads | +0.006 |
floor | +0.006 |
Neg logit contributions | |
help | -0.006 |
reward | -0.006 |
ar | -0.006 |
Sure | -0.005 |
plunge | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
help | +0.054 |
reward | +0.053 |
less | +0.050 |
ar | +0.050 |
bye | +0.049 |
Neg logit contributions | |
cks | -0.076 |
lair | -0.072 |
mess | -0.069 |
floor | -0.068 |
pads | -0.067 |
Token | was |
Token ID | 373 |
Feature activation | +3.56e-01 |
Pos logit contributions | |
lair | +0.015 |
cks | +0.015 |
mess | +0.015 |
pads | +0.014 |
floor | +0.014 |
Neg logit contributions | |
help | -0.014 |
reward | -0.014 |
ar | -0.014 |
plunge | -0.014 |
bye | -0.013 |