Token | , |
Token ID | 11 |
Feature activation | -3.95e-02 |
Pos logit contributions | |
couple | +0.014 |
rob | +0.012 |
seek | +0.011 |
lady | +0.010 |
thank | +0.010 |
Neg logit contributions | |
aying | -0.025 |
fears | -0.024 |
arse | -0.024 |
Ow | -0.023 |
uning | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
aying | +0.002 |
arse | +0.002 |
fears | +0.002 |
uning | +0.002 |
Ow | +0.002 |
Neg logit contributions | |
couple | -0.001 |
rob | -0.001 |
seek | -0.001 |
lady | -0.001 |
gentleman | -0.001 |
Token | Jill |
Token ID | 22772 |
Feature activation | +4.55e-01 |
Pos logit contributions | |
rob | +0.008 |
couple | +0.007 |
seek | +0.006 |
gentleman | +0.006 |
trainer | +0.006 |
Neg logit contributions | |
aying | -0.009 |
fears | -0.009 |
arse | -0.008 |
omed | -0.008 |
uning | -0.008 |
Token | knew |
Token ID | 2993 |
Feature activation | +3.11e-01 |
Pos logit contributions | |
couple | +0.018 |
rob | +0.017 |
seek | +0.015 |
gentleman | +0.015 |
trainer | +0.014 |
Neg logit contributions | |
aying | -0.024 |
fears | -0.022 |
arse | -0.022 |
uning | -0.022 |
omed | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | +2.74e-01 |
Pos logit contributions | |
couple | +0.010 |
rob | +0.009 |
seek | +0.009 |
lady | +0.009 |
gentleman | +0.008 |
Neg logit contributions | |
aying | -0.018 |
arse | -0.018 |
fears | -0.018 |
Ow | -0.017 |
uning | -0.017 |
Token | elderly |
Token ID | 13830 |
Feature activation | +2.39e+00 |
Pos logit contributions | |
rob | +0.010 |
couple | +0.010 |
seek | +0.009 |
gentleman | +0.008 |
trainer | +0.008 |
Neg logit contributions | |
aying | -0.015 |
fears | -0.014 |
uning | -0.014 |
arse | -0.014 |
Ow | -0.013 |
Token | neighbour |
Token ID | 12250 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
rob | +0.074 |
robat | +0.053 |
seek | +0.052 |
Ar | +0.050 |
seek | +0.044 |
Neg logit contributions | |
aying | -0.140 |
uning | -0.136 |
wipes | -0.134 |
fears | -0.128 |
icking | -0.127 |
Token | was |
Token ID | 373 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
couple | +0.037 |
rob | +0.035 |
seek | +0.032 |
gentleman | +0.031 |
lady | +0.031 |
Neg logit contributions | |
aying | -0.056 |
arse | -0.054 |
fears | -0.053 |
Ow | -0.052 |
uning | -0.052 |
Token | right |
Token ID | 826 |
Feature activation | +2.37e-01 |
Pos logit contributions | |
rob | +0.004 |
couple | +0.004 |
seek | +0.003 |
robat | +0.003 |
gentleman | +0.003 |
Neg logit contributions | |
aying | -0.005 |
fears | -0.005 |
arse | -0.005 |
uning | -0.005 |
omed | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +5.01e-02 |
Pos logit contributions | |
rob | +0.009 |
couple | +0.009 |
seek | +0.008 |
gentleman | +0.007 |
trainer | +0.007 |
Neg logit contributions | |
aying | -0.013 |
fears | -0.012 |
arse | -0.012 |
omed | -0.012 |
uning | -0.012 |
Token | The |
Token ID | 383 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
rob | +0.002 |
couple | +0.002 |
seek | +0.002 |
trainer | +0.002 |
gentleman | +0.002 |
Neg logit contributions | |
aying | -0.002 |
fears | -0.002 |
uning | -0.002 |
arse | -0.002 |
omed | -0.002 |
Token | having |
Token ID | 1719 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
aying | +0.040 |
arse | +0.039 |
fears | +0.038 |
Ow | +0.038 |
omed | +0.037 |
Neg logit contributions | |
couple | -0.025 |
rob | -0.022 |
seek | -0.022 |
thank | -0.020 |
gentleman | -0.020 |
Token | fun |
Token ID | 1257 |
Feature activation | +4.68e-01 |
Pos logit contributions | |
aying | +0.052 |
fears | +0.049 |
arse | +0.049 |
uning | +0.048 |
omed | +0.047 |
Neg logit contributions | |
couple | -0.040 |
rob | -0.039 |
seek | -0.035 |
gentleman | -0.033 |
trainer | -0.033 |
Token | together |
Token ID | 1978 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
couple | +0.018 |
rob | +0.018 |
gentleman | +0.016 |
trainer | +0.015 |
seek | +0.015 |
Neg logit contributions | |
aying | -0.024 |
fears | -0.022 |
omed | -0.022 |
uning | -0.022 |
adoes | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
aying | +0.007 |
fears | +0.007 |
arse | +0.007 |
uning | +0.007 |
omed | +0.007 |
Neg logit contributions | |
rob | -0.005 |
couple | -0.005 |
seek | -0.004 |
gentleman | -0.004 |
trainer | -0.004 |
Token | The |
Token ID | 383 |
Feature activation | +4.35e-01 |
Pos logit contributions | |
couple | +0.006 |
rob | +0.005 |
seek | +0.005 |
gentleman | +0.005 |
trainer | +0.005 |
Neg logit contributions | |
aying | -0.006 |
arse | -0.006 |
fears | -0.006 |
uning | -0.006 |
omed | -0.006 |
Token | elderly |
Token ID | 13830 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
couple | +0.013 |
rob | +0.012 |
seek | +0.010 |
gentleman | +0.010 |
trainer | +0.010 |
Neg logit contributions | |
aying | -0.028 |
fears | -0.026 |
arse | -0.026 |
uning | -0.026 |
omed | -0.026 |
Token | rh |
Token ID | 9529 |
Feature activation | +4.35e-01 |
Pos logit contributions | |
rob | +0.074 |
couple | +0.063 |
seek | +0.063 |
robat | +0.060 |
seek | +0.058 |
Neg logit contributions | |
aying | -0.122 |
adoes | -0.117 |
fears | -0.111 |
uning | -0.110 |
wipes | -0.109 |
Token | in |
Token ID | 259 |
Feature activation | +5.23e-01 |
Pos logit contributions | |
couple | +0.019 |
rob | +0.019 |
seek | +0.017 |
gentleman | +0.017 |
trainer | +0.016 |
Neg logit contributions | |
aying | -0.020 |
fears | -0.019 |
arse | -0.019 |
uning | -0.018 |
omed | -0.018 |
Token | oc |
Token ID | 420 |
Feature activation | +2.60e-01 |
Pos logit contributions | |
rob | +0.012 |
couple | +0.012 |
gentleman | +0.010 |
seek | +0.010 |
businessman | +0.009 |
Neg logit contributions | |
aying | -0.034 |
fears | -0.033 |
ist | -0.031 |
omed | -0.031 |
cleans | -0.031 |
Token | eros |
Token ID | 27498 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
couple | +0.010 |
rob | +0.009 |
seek | +0.009 |
gentleman | +0.008 |
lady | +0.008 |
Neg logit contributions | |
aying | -0.014 |
uning | -0.013 |
fears | -0.013 |
adoes | -0.013 |
arse | -0.013 |
Token | was |
Token ID | 373 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
couple | +0.005 |
rob | +0.004 |
seek | +0.004 |
gentleman | +0.004 |
trainer | +0.004 |
Neg logit contributions | |
aying | -0.006 |
fears | -0.006 |
arse | -0.006 |
uning | -0.006 |
omed | -0.006 |
Token | |
Token ID | 220 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
couple | +0.002 |
rob | +0.002 |
seek | +0.001 |
trainer | +0.001 |
gentleman | +0.001 |
Neg logit contributions | |
aying | -0.002 |
fears | -0.002 |
arse | -0.001 |
omed | -0.001 |
wipes | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -6.60e-01 |
Pos logit contributions | |
couple | +0.001 |
rob | +0.001 |
seek | +0.001 |
gentleman | +0.001 |
lady | +0.001 |
Neg logit contributions | |
aying | -0.001 |
fears | -0.001 |
arse | -0.001 |
uning | -0.001 |
omed | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
aying | +0.032 |
Ow | +0.030 |
wipes | +0.030 |
omed | +0.029 |
arse | +0.029 |
Neg logit contributions | |
couple | -0.028 |
rob | -0.028 |
seek | -0.024 |
gentleman | -0.024 |
trainer | -0.024 |
Token | After |
Token ID | 3260 |
Feature activation | +9.23e-01 |
Pos logit contributions | |
aying | +0.014 |
fears | +0.013 |
arse | +0.013 |
wipes | +0.013 |
omed | +0.012 |
Neg logit contributions | |
rob | -0.010 |
couple | -0.010 |
gentleman | -0.008 |
trainer | -0.008 |
robat | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | +4.64e-01 |
Pos logit contributions | |
couple | +0.034 |
rob | +0.033 |
seek | +0.029 |
gentleman | +0.028 |
trainer | +0.028 |
Neg logit contributions | |
aying | -0.050 |
fears | -0.047 |
arse | -0.047 |
uning | -0.047 |
omed | -0.046 |
Token | haircut |
Token ID | 45548 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
rob | +0.021 |
couple | +0.019 |
seek | +0.018 |
gentleman | +0.018 |
trainer | +0.017 |
Neg logit contributions | |
aying | -0.021 |
cleans | -0.020 |
fears | -0.020 |
arse | -0.020 |
wipes | -0.020 |
Token | , |
Token ID | 11 |
Feature activation | +7.81e-02 |
Pos logit contributions | |
couple | +0.088 |
rob | +0.085 |
seek | +0.076 |
gentleman | +0.073 |
trainer | +0.072 |
Neg logit contributions | |
aying | -0.123 |
fears | -0.116 |
arse | -0.115 |
uning | -0.113 |
omed | -0.112 |
Token | Lily |
Token ID | 20037 |
Feature activation | +5.03e-01 |
Pos logit contributions | |
couple | +0.003 |
rob | +0.003 |
seek | +0.003 |
gentleman | +0.003 |
trainer | +0.003 |
Neg logit contributions | |
aying | -0.004 |
fears | -0.004 |
arse | -0.004 |
uning | -0.004 |
omed | -0.004 |
Token | looked |
Token ID | 3114 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
couple | +0.020 |
rob | +0.020 |
seek | +0.018 |
gentleman | +0.017 |
trainer | +0.017 |
Neg logit contributions | |
aying | -0.026 |
fears | -0.024 |
arse | -0.024 |
uning | -0.023 |
omed | -0.023 |
Token | in |
Token ID | 287 |
Feature activation | -3.71e-01 |
Pos logit contributions | |
couple | +0.008 |
rob | +0.007 |
lady | +0.007 |
gentleman | +0.007 |
man | +0.007 |
Neg logit contributions | |
aying | -0.013 |
arse | -0.012 |
Ow | -0.012 |
uning | -0.012 |
adoes | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | -8.54e-01 |
Pos logit contributions | |
aying | +0.019 |
fears | +0.018 |
wipes | +0.018 |
Ow | +0.018 |
arse | +0.018 |
Neg logit contributions | |
rob | -0.015 |
couple | -0.014 |
seek | -0.013 |
trainer | -0.012 |
robat | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
aying | +0.020 |
fears | +0.019 |
arse | +0.018 |
uning | +0.018 |
omed | +0.018 |
Neg logit contributions | |
couple | -0.014 |
rob | -0.014 |
seek | -0.012 |
gentleman | -0.012 |
trainer | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | -2.14e-03 |
Pos logit contributions | |
couple | +0.010 |
rob | +0.009 |
gentleman | +0.009 |
lady | +0.009 |
man | +0.008 |
Neg logit contributions | |
aying | -0.013 |
arse | -0.012 |
fears | -0.012 |
uning | -0.012 |
omed | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
aying | +0.000 |
fears | +0.000 |
arse | +0.000 |
uning | +0.000 |
omed | +0.000 |
Neg logit contributions | |
couple | -0.000 |
rob | -0.000 |
seek | -0.000 |
gentleman | -0.000 |
lady | -0.000 |
Token | After |
Token ID | 3260 |
Feature activation | +9.41e-01 |
Pos logit contributions | |
aying | +0.001 |
fears | +0.001 |
arse | +0.001 |
uning | +0.001 |
omed | +0.001 |
Neg logit contributions | |
couple | -0.001 |
rob | -0.001 |
seek | -0.001 |
gentleman | -0.001 |
trainer | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | +5.35e-01 |
Pos logit contributions | |
couple | +0.034 |
rob | +0.032 |
seek | +0.029 |
gentleman | +0.027 |
lady | +0.027 |
Neg logit contributions | |
aying | -0.054 |
fears | -0.051 |
arse | -0.050 |
uning | -0.050 |
omed | -0.049 |
Token | haircut |
Token ID | 45548 |
Feature activation | +2.30e+00 |
Pos logit contributions | |
couple | +0.023 |
rob | +0.022 |
seek | +0.020 |
gentleman | +0.019 |
trainer | +0.019 |
Neg logit contributions | |
aying | -0.026 |
fears | -0.025 |
arse | -0.024 |
uning | -0.024 |
omed | -0.024 |
Token | was |
Token ID | 373 |
Feature activation | +5.48e-01 |
Pos logit contributions | |
couple | +0.087 |
rob | +0.083 |
seek | +0.074 |
gentleman | +0.071 |
lady | +0.070 |
Neg logit contributions | |
aying | -0.124 |
fears | -0.117 |
arse | -0.117 |
uning | -0.115 |
Ow | -0.113 |
Token | finished |
Token ID | 5201 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
couple | +0.020 |
seek | +0.017 |
thank | +0.017 |
rob | +0.017 |
lady | +0.016 |
Neg logit contributions | |
aying | -0.030 |
fears | -0.030 |
arse | -0.029 |
Ow | -0.029 |
uning | -0.029 |
Token | , |
Token ID | 11 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
couple | +0.038 |
rob | +0.032 |
seek | +0.031 |
man | +0.029 |
lady | +0.029 |
Neg logit contributions | |
aying | -0.081 |
fears | -0.078 |
Ow | -0.076 |
arse | -0.076 |
uning | -0.075 |
Token | the |
Token ID | 262 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
couple | +0.006 |
rob | +0.005 |
lady | +0.005 |
seek | +0.005 |
gentleman | +0.005 |
Neg logit contributions | |
aying | -0.012 |
fears | -0.011 |
arse | -0.011 |
uning | -0.011 |
omed | -0.011 |
Token | three |
Token ID | 1115 |
Feature activation | +7.18e-01 |
Pos logit contributions | |
aying | +0.010 |
fears | +0.010 |
fists | +0.009 |
wipes | +0.009 |
arse | +0.009 |
Neg logit contributions | |
rob | -0.012 |
seek | -0.010 |
couple | -0.010 |
robat | -0.010 |
trainer | -0.009 |
Token | ." |
Token ID | 526 |
Feature activation | +3.23e-01 |
Pos logit contributions | |
couple | +0.016 |
seek | +0.015 |
lady | +0.014 |
rob | +0.014 |
thank | +0.014 |
Neg logit contributions | |
aying | -0.027 |
arse | -0.026 |
fears | -0.026 |
omed | -0.025 |
uning | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | +6.15e-02 |
Pos logit contributions | |
couple | +0.012 |
rob | +0.011 |
lady | +0.010 |
seek | +0.010 |
gentleman | +0.010 |
Neg logit contributions | |
aying | -0.018 |
fears | -0.017 |
arse | -0.017 |
omed | -0.016 |
uning | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | +5.86e-02 |
Pos logit contributions | |
couple | +0.002 |
rob | +0.002 |
seek | +0.002 |
gentleman | +0.002 |
trainer | +0.002 |
Neg logit contributions | |
aying | -0.003 |
Ow | -0.003 |
arse | -0.003 |
fears | -0.003 |
omed | -0.003 |
Token | After |
Token ID | 3260 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
couple | +0.002 |
rob | +0.002 |
seek | +0.002 |
gentleman | +0.002 |
trainer | +0.002 |
Neg logit contributions | |
aying | -0.003 |
fears | -0.003 |
arse | -0.003 |
uning | -0.003 |
omed | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | +6.75e-01 |
Pos logit contributions | |
couple | +0.040 |
rob | +0.039 |
seek | +0.035 |
gentleman | +0.033 |
trainer | +0.033 |
Neg logit contributions | |
aying | -0.058 |
fears | -0.054 |
arse | -0.054 |
uning | -0.053 |
omed | -0.053 |
Token | haircut |
Token ID | 45548 |
Feature activation | +2.27e+00 |
Pos logit contributions | |
rob | +0.030 |
couple | +0.028 |
seek | +0.026 |
gentleman | +0.025 |
trainer | +0.024 |
Neg logit contributions | |
aying | -0.031 |
fears | -0.030 |
arse | -0.029 |
uning | -0.029 |
wipes | -0.029 |
Token | , |
Token ID | 11 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
couple | +0.090 |
rob | +0.088 |
seek | +0.079 |
gentleman | +0.076 |
trainer | +0.075 |
Neg logit contributions | |
aying | -0.117 |
fears | -0.109 |
uning | -0.108 |
arse | -0.107 |
omed | -0.106 |
Token | Lily |
Token ID | 20037 |
Feature activation | +4.87e-01 |
Pos logit contributions | |
couple | +0.007 |
rob | +0.007 |
seek | +0.006 |
gentleman | +0.006 |
trainer | +0.006 |
Neg logit contributions | |
aying | -0.009 |
fears | -0.008 |
arse | -0.008 |
uning | -0.008 |
omed | -0.008 |
Token | looked |
Token ID | 3114 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
rob | +0.021 |
couple | +0.020 |
seek | +0.018 |
gentleman | +0.017 |
trainer | +0.017 |
Neg logit contributions | |
aying | -0.024 |
fears | -0.022 |
cleans | -0.022 |
arse | -0.022 |
wipes | -0.022 |
Token | in |
Token ID | 287 |
Feature activation | -3.16e-01 |
Pos logit contributions | |
couple | +0.008 |
rob | +0.007 |
lady | +0.007 |
gentleman | +0.007 |
man | +0.007 |
Neg logit contributions | |
aying | -0.012 |
arse | -0.011 |
Ow | -0.011 |
uning | -0.011 |
adoes | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | -4.26e-01 |
Pos logit contributions | |
fears | +0.015 |
aying | +0.015 |
wipes | +0.015 |
fists | +0.014 |
Ow | +0.014 |
Neg logit contributions | |
rob | -0.014 |
couple | -0.013 |
seek | -0.012 |
robat | -0.012 |
trainer | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
aying | +0.008 |
omed | +0.007 |
fears | +0.007 |
uning | +0.007 |
wipes | +0.007 |
Neg logit contributions | |
rob | -0.011 |
couple | -0.010 |
Offic | -0.009 |
trainer | -0.009 |
businessman | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | +3.73e-02 |
Pos logit contributions | |
couple | +0.004 |
rob | +0.004 |
seek | +0.004 |
gentleman | +0.003 |
trainer | +0.003 |
Neg logit contributions | |
aying | -0.007 |
fears | -0.007 |
arse | -0.007 |
uning | -0.007 |
omed | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
couple | +0.001 |
rob | +0.001 |
seek | +0.001 |
gentleman | +0.001 |
trainer | +0.001 |
Neg logit contributions | |
aying | -0.002 |
fears | -0.002 |
Ow | -0.002 |
arse | -0.002 |
omed | -0.002 |
Token | When |
Token ID | 2215 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
rob | +0.005 |
couple | +0.005 |
gentleman | +0.004 |
seek | +0.004 |
trainer | +0.004 |
Neg logit contributions | |
aying | -0.005 |
arse | -0.004 |
fears | -0.004 |
omed | -0.004 |
Ow | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | +5.44e-01 |
Pos logit contributions | |
rob | +0.010 |
couple | +0.009 |
seek | +0.009 |
trainer | +0.009 |
gentleman | +0.009 |
Neg logit contributions | |
aying | -0.012 |
Ow | -0.011 |
omed | -0.011 |
arse | -0.011 |
fears | -0.011 |
Token | elderly |
Token ID | 13830 |
Feature activation | +2.26e+00 |
Pos logit contributions | |
rob | +0.020 |
robat | +0.015 |
seek | +0.013 |
couple | +0.012 |
thank | +0.011 |
Neg logit contributions | |
aying | -0.032 |
fears | -0.031 |
uning | -0.030 |
omed | -0.030 |
cleans | -0.030 |
Token | lady |
Token ID | 10846 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
rob | +0.086 |
robat | +0.068 |
Ar | +0.056 |
seek | +0.055 |
yak | +0.054 |
Neg logit contributions | |
aying | -0.117 |
cleans | -0.114 |
wipes | -0.112 |
uning | -0.111 |
omed | -0.107 |
Token | got |
Token ID | 1392 |
Feature activation | +4.72e-01 |
Pos logit contributions | |
rob | +0.077 |
couple | +0.076 |
seek | +0.067 |
gentleman | +0.064 |
thank | +0.064 |
Neg logit contributions | |
aying | -0.096 |
omed | -0.090 |
fears | -0.089 |
uning | -0.088 |
arse | -0.088 |
Token | home |
Token ID | 1363 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
rob | +0.019 |
couple | +0.017 |
robat | +0.015 |
seek | +0.015 |
trainer | +0.015 |
Neg logit contributions | |
aying | -0.025 |
fears | -0.023 |
uning | -0.023 |
omed | -0.023 |
arse | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | +7.93e-02 |
Pos logit contributions | |
rob | +0.049 |
couple | +0.045 |
robat | +0.041 |
trainer | +0.040 |
seek | +0.040 |
Neg logit contributions | |
aying | -0.049 |
arse | -0.045 |
fears | -0.045 |
uning | -0.045 |
omed | -0.045 |
Token | she |
Token ID | 673 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
couple | +0.002 |
rob | +0.002 |
seek | +0.002 |
gentleman | +0.002 |
trainer | +0.002 |
Neg logit contributions | |
aying | -0.005 |
fears | -0.005 |
arse | -0.005 |
uning | -0.005 |
omed | -0.005 |
Token | on |
Token ID | 319 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
rob | +0.021 |
couple | +0.021 |
seek | +0.019 |
gentleman | +0.018 |
trainer | +0.017 |
Neg logit contributions | |
aying | -0.029 |
arse | -0.027 |
fears | -0.027 |
uning | -0.026 |
omed | -0.026 |
Token | his |
Token ID | 465 |
Feature activation | -3.08e-01 |
Pos logit contributions | |
aying | +0.018 |
fears | +0.017 |
arse | +0.017 |
uning | +0.017 |
Ow | +0.017 |
Neg logit contributions | |
couple | -0.007 |
rob | -0.006 |
seek | -0.005 |
gentleman | -0.005 |
trainer | -0.005 |
Token | face |
Token ID | 1986 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
aying | +0.012 |
arse | +0.011 |
uning | +0.010 |
fears | +0.010 |
omed | +0.010 |
Neg logit contributions | |
couple | -0.017 |
rob | -0.016 |
seek | -0.015 |
gentleman | -0.015 |
trainer | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +9.23e-02 |
Pos logit contributions | |
couple | +0.007 |
rob | +0.007 |
seek | +0.006 |
gentleman | +0.006 |
lady | +0.005 |
Neg logit contributions | |
aying | -0.012 |
fears | -0.011 |
arse | -0.011 |
uning | -0.011 |
Ow | -0.011 |
Token | The |
Token ID | 383 |
Feature activation | +9.51e-02 |
Pos logit contributions | |
couple | +0.003 |
rob | +0.003 |
gentleman | +0.003 |
seek | +0.003 |
lady | +0.002 |
Neg logit contributions | |
aying | -0.005 |
arse | -0.005 |
fears | -0.005 |
uning | -0.005 |
adoes | -0.005 |
Token | elderly |
Token ID | 13830 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
rob | +0.003 |
seek | +0.002 |
couple | +0.002 |
robat | +0.002 |
seek | +0.002 |
Neg logit contributions | |
aying | -0.006 |
fears | -0.006 |
fists | -0.006 |
wipes | -0.006 |
arse | -0.006 |
Token | woman |
Token ID | 2415 |
Feature activation | +6.63e-01 |
Pos logit contributions | |
rob | +0.073 |
seek | +0.065 |
robat | +0.059 |
yak | +0.054 |
Ar | +0.053 |
Neg logit contributions | |
aying | -0.122 |
wipes | -0.119 |
cleans | -0.117 |
adoes | -0.112 |
fears | -0.112 |
Token | thanked |
Token ID | 26280 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
couple | +0.024 |
rob | +0.022 |
seek | +0.020 |
lady | +0.020 |
gentleman | +0.019 |
Neg logit contributions | |
aying | -0.038 |
arse | -0.036 |
fears | -0.035 |
uning | -0.035 |
Ow | -0.035 |
Token | him |
Token ID | 683 |
Feature activation | -2.38e-01 |
Pos logit contributions | |
aying | +0.011 |
fears | +0.011 |
arse | +0.010 |
uning | +0.010 |
omed | +0.010 |
Neg logit contributions | |
couple | -0.009 |
rob | -0.008 |
seek | -0.007 |
gentleman | -0.007 |
lady | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
aying | +0.015 |
arse | +0.015 |
Ow | +0.015 |
fears | +0.014 |
uning | +0.014 |
Neg logit contributions | |
couple | -0.007 |
rob | -0.005 |
lady | -0.005 |
gentleman | -0.005 |
seek | -0.005 |
Token | gave |
Token ID | 2921 |
Feature activation | -6.10e-01 |
Pos logit contributions | |
aying | +0.009 |
fears | +0.008 |
wipes | +0.008 |
arse | +0.008 |
cleans | +0.008 |
Neg logit contributions | |
rob | -0.006 |
couple | -0.006 |
seek | -0.006 |
gentleman | -0.005 |
trainer | -0.005 |
Token | closed |
Token ID | 4838 |
Feature activation | -3.60e-02 |
Pos logit contributions | |
couple | +0.004 |
rob | +0.003 |
seek | +0.003 |
lady | +0.003 |
thank | +0.003 |
Neg logit contributions | |
aying | -0.005 |
fears | -0.005 |
arse | -0.005 |
uning | -0.004 |
Ow | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
aying | +0.002 |
arse | +0.002 |
fears | +0.002 |
uning | +0.002 |
Ow | +0.002 |
Neg logit contributions | |
couple | -0.001 |
rob | -0.001 |
seek | -0.001 |
lady | -0.001 |
thank | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
couple | +0.007 |
rob | +0.006 |
seek | +0.006 |
gentleman | +0.005 |
trainer | +0.005 |
Neg logit contributions | |
aying | -0.009 |
fears | -0.009 |
arse | -0.009 |
uning | -0.009 |
omed | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
couple | +0.005 |
rob | +0.005 |
seek | +0.004 |
thank | +0.004 |
trainer | +0.004 |
Neg logit contributions | |
aying | -0.005 |
omed | -0.005 |
fears | -0.005 |
Ow | -0.005 |
arse | -0.005 |
Token | The |
Token ID | 464 |
Feature activation | +2.56e-01 |
Pos logit contributions | |
rob | +0.009 |
couple | +0.008 |
king | +0.008 |
seek | +0.008 |
trainer | +0.007 |
Neg logit contributions | |
aying | -0.008 |
omed | -0.007 |
arse | -0.007 |
fears | -0.007 |
arding | -0.007 |
Token | elderly |
Token ID | 13830 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
rob | +0.011 |
couple | +0.008 |
amer | +0.008 |
cest | +0.008 |
seek | +0.008 |
Neg logit contributions | |
aying | -0.012 |
fears | -0.012 |
omed | -0.012 |
oken | -0.012 |
wipes | -0.011 |
Token | lady |
Token ID | 10846 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
rob | +0.060 |
Ar | +0.049 |
robat | +0.049 |
seek | +0.047 |
yak | +0.046 |
Neg logit contributions | |
wipes | -0.114 |
aying | -0.113 |
cleans | -0.113 |
omed | -0.108 |
adoes | -0.106 |
Token | was |
Token ID | 373 |
Feature activation | +6.38e-02 |
Pos logit contributions | |
couple | +0.039 |
rob | +0.037 |
seek | +0.034 |
gentleman | +0.032 |
lady | +0.032 |
Neg logit contributions | |
aying | -0.058 |
fears | -0.055 |
arse | -0.055 |
uning | -0.053 |
Ow | -0.053 |
Token | so |
Token ID | 523 |
Feature activation | +3.28e-01 |
Pos logit contributions | |
rob | +0.003 |
couple | +0.002 |
robat | +0.002 |
seek | +0.002 |
trainer | +0.002 |
Neg logit contributions | |
aying | -0.003 |
fears | -0.003 |
omed | -0.003 |
cleans | -0.003 |
arse | -0.003 |
Token | grateful |
Token ID | 14066 |
Feature activation | -7.52e-02 |
Pos logit contributions | |
rob | +0.011 |
couple | +0.010 |
seek | +0.009 |
gentleman | +0.008 |
trainer | +0.008 |
Neg logit contributions | |
aying | -0.020 |
fears | -0.019 |
arse | -0.019 |
omed | -0.018 |
uning | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | -1.49e-04 |
Pos logit contributions | |
aying | +0.004 |
fears | +0.003 |
omed | +0.003 |
wipes | +0.003 |
uning | +0.003 |
Neg logit contributions | |
rob | -0.003 |
couple | -0.003 |
seek | -0.003 |
robat | -0.003 |
trainer | -0.003 |
Token | her |
Token ID | 607 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
aying | +0.002 |
omed | +0.002 |
fears | +0.002 |
uning | +0.002 |
arse | +0.002 |
Neg logit contributions | |
couple | -0.001 |
seek | -0.001 |
rob | -0.001 |
lady | -0.001 |
man | -0.001 |
Token | !" |
Token ID | 2474 |
Feature activation | +5.18e-01 |
Pos logit contributions | |
omed | +0.007 |
aying | +0.007 |
arse | +0.006 |
outed | +0.006 |
ours | +0.006 |
Neg logit contributions | |
couple | -0.004 |
seek | -0.004 |
lady | -0.004 |
man | -0.004 |
thank | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
rob | +0.020 |
couple | +0.019 |
seek | +0.017 |
gentleman | +0.016 |
trainer | +0.016 |
Neg logit contributions | |
aying | -0.028 |
fears | -0.026 |
Ow | -0.026 |
uning | -0.026 |
arse | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
couple | +0.008 |
rob | +0.007 |
seek | +0.006 |
gentleman | +0.006 |
thank | +0.006 |
Neg logit contributions | |
aying | -0.010 |
Ow | -0.010 |
arse | -0.010 |
fears | -0.010 |
omed | -0.009 |
Token | The |
Token ID | 464 |
Feature activation | -6.63e-02 |
Pos logit contributions | |
rob | +0.010 |
couple | +0.010 |
thank | +0.009 |
gentleman | +0.008 |
seek | +0.008 |
Neg logit contributions | |
aying | -0.011 |
fears | -0.010 |
arse | -0.010 |
Ow | -0.010 |
omed | -0.010 |
Token | elderly |
Token ID | 13830 |
Feature activation | +2.21e+00 |
Pos logit contributions | |
aying | +0.004 |
fears | +0.004 |
fists | +0.004 |
arse | +0.004 |
cleans | +0.004 |
Neg logit contributions | |
rob | -0.002 |
seek | -0.002 |
robat | -0.002 |
couple | -0.002 |
thank | -0.001 |
Token | woman |
Token ID | 2415 |
Feature activation | +8.15e-01 |
Pos logit contributions | |
rob | +0.060 |
seek | +0.054 |
robat | +0.046 |
Ar | +0.041 |
seek | +0.040 |
Neg logit contributions | |
aying | -0.133 |
wipes | -0.126 |
cleans | -0.125 |
uning | -0.121 |
fears | -0.121 |
Token | win |
Token ID | 1592 |
Feature activation | +6.60e-01 |
Pos logit contributions | |
couple | +0.030 |
lady | +0.026 |
man | +0.026 |
gentleman | +0.024 |
seek | +0.024 |
Neg logit contributions | |
aying | -0.047 |
arse | -0.046 |
Ow | -0.045 |
uning | -0.044 |
fears | -0.043 |
Token | ced |
Token ID | 771 |
Feature activation | +6.08e-01 |
Pos logit contributions | |
couple | +0.025 |
rob | +0.023 |
seek | +0.022 |
lady | +0.021 |
gentleman | +0.021 |
Neg logit contributions | |
aying | -0.036 |
arse | -0.034 |
fears | -0.034 |
Ow | -0.033 |
uning | -0.033 |
Token | in |
Token ID | 287 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
couple | +0.018 |
rob | +0.016 |
lady | +0.015 |
seek | +0.015 |
man | +0.015 |
Neg logit contributions | |
aying | -0.037 |
ours | -0.035 |
uning | -0.035 |
fears | -0.035 |
arse | -0.035 |
Token | pain |
Token ID | 2356 |
Feature activation | +3.51e-01 |
Pos logit contributions | |
aying | +0.013 |
fears | +0.012 |
arse | +0.012 |
uning | +0.012 |
omed | +0.012 |
Neg logit contributions | |
couple | -0.008 |
rob | -0.008 |
seek | -0.007 |
gentleman | -0.007 |
trainer | -0.007 |
Token | The |
Token ID | 464 |
Feature activation | -2.71e-02 |
Pos logit contributions | |
aying | +0.009 |
arse | +0.008 |
fears | +0.008 |
omed | +0.008 |
Ow | +0.008 |
Neg logit contributions | |
couple | -0.009 |
rob | -0.009 |
gentleman | -0.008 |
seek | -0.008 |
lady | -0.008 |
Token | animals |
Token ID | 4695 |
Feature activation | -3.71e-01 |
Pos logit contributions | |
aying | +0.001 |
fears | +0.001 |
arse | +0.001 |
omed | +0.001 |
uning | +0.001 |
Neg logit contributions | |
rob | -0.001 |
couple | -0.001 |
seek | -0.001 |
gentleman | -0.001 |
thank | -0.001 |
Token | all |
Token ID | 477 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
aying | +0.020 |
arse | +0.020 |
Ow | +0.019 |
fears | +0.019 |
uning | +0.019 |
Neg logit contributions | |
couple | -0.014 |
rob | -0.013 |
seek | -0.012 |
lady | -0.012 |
gentleman | -0.011 |
Token | thanked |
Token ID | 26280 |
Feature activation | +2.67e-01 |
Pos logit contributions | |
aying | +0.007 |
arse | +0.007 |
fears | +0.007 |
uning | +0.007 |
Ow | +0.007 |
Neg logit contributions | |
couple | -0.005 |
rob | -0.005 |
seek | -0.004 |
gentleman | -0.004 |
trainer | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | +8.72e-02 |
Pos logit contributions | |
couple | +0.009 |
rob | +0.009 |
trainer | +0.007 |
seek | +0.007 |
lady | +0.007 |
Neg logit contributions | |
aying | -0.016 |
arse | -0.015 |
fears | -0.015 |
omed | -0.015 |
ours | -0.015 |
Token | elderly |
Token ID | 13830 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
rob | +0.003 |
couple | +0.003 |
seek | +0.002 |
gentleman | +0.002 |
thank | +0.002 |
Neg logit contributions | |
aying | -0.005 |
fears | -0.005 |
arse | -0.005 |
uning | -0.005 |
omed | -0.005 |
Token | rh |
Token ID | 9529 |
Feature activation | +3.80e-01 |
Pos logit contributions | |
rob | +0.067 |
seek | +0.066 |
couple | +0.061 |
robat | +0.057 |
seek | +0.055 |
Neg logit contributions | |
aying | -0.108 |
uning | -0.101 |
adoes | -0.101 |
fears | -0.099 |
cleans | -0.098 |
Token | in |
Token ID | 259 |
Feature activation | +3.97e-01 |
Pos logit contributions | |
couple | +0.017 |
rob | +0.017 |
seek | +0.015 |
gentleman | +0.015 |
trainer | +0.014 |
Neg logit contributions | |
aying | -0.018 |
fears | -0.016 |
arse | -0.016 |
uning | -0.016 |
omed | -0.016 |
Token | oc |
Token ID | 420 |
Feature activation | +2.80e-01 |
Pos logit contributions | |
rob | +0.010 |
couple | +0.009 |
seek | +0.009 |
thank | +0.007 |
businessman | +0.007 |
Neg logit contributions | |
aying | -0.026 |
fears | -0.024 |
uning | -0.024 |
omed | -0.023 |
arse | -0.023 |
Token | eros |
Token ID | 27498 |
Feature activation | +2.45e-01 |
Pos logit contributions | |
couple | +0.010 |
rob | +0.010 |
seek | +0.009 |
gentleman | +0.008 |
lady | +0.008 |
Neg logit contributions | |
aying | -0.015 |
uning | -0.014 |
fears | -0.014 |
arse | -0.014 |
adoes | -0.014 |
Token | for |
Token ID | 329 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
couple | +0.007 |
rob | +0.006 |
seek | +0.006 |
gentleman | +0.005 |
trainer | +0.005 |
Neg logit contributions | |
aying | -0.016 |
arse | -0.015 |
fears | -0.015 |
Ow | -0.015 |
uning | -0.014 |
Token | help |
Token ID | 1037 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
aying | +0.046 |
omed | +0.045 |
arse | +0.044 |
uning | +0.042 |
outed | +0.042 |
Neg logit contributions | |
couple | -0.024 |
rob | -0.021 |
seek | -0.021 |
lady | -0.020 |
man | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | +5.13e-02 |
Pos logit contributions | |
couple | +0.004 |
seek | +0.003 |
rob | +0.003 |
lady | +0.003 |
man | +0.003 |
Neg logit contributions | |
aying | -0.007 |
arse | -0.006 |
omed | -0.006 |
fears | -0.006 |
ours | -0.006 |
Token | Even |
Token ID | 3412 |
Feature activation | +5.51e-02 |
Pos logit contributions | |
couple | +0.002 |
lady | +0.002 |
gentleman | +0.002 |
ranger | +0.002 |
man | +0.002 |
Neg logit contributions | |
arse | -0.003 |
aying | -0.003 |
fears | -0.003 |
arding | -0.003 |
iqu | -0.003 |
Token | though |
Token ID | 996 |
Feature activation | -5.33e-02 |
Pos logit contributions | |
couple | +0.002 |
rob | +0.002 |
seek | +0.002 |
lady | +0.002 |
gentleman | +0.002 |
Neg logit contributions | |
aying | -0.003 |
arse | -0.003 |
fears | -0.003 |
uning | -0.003 |
Ow | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
aying | +0.003 |
arse | +0.003 |
uning | +0.003 |
omed | +0.003 |
fears | +0.003 |
Neg logit contributions | |
couple | -0.002 |
seek | -0.001 |
lady | -0.001 |
rob | -0.001 |
man | -0.001 |
Token | elderly |
Token ID | 13830 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
rob | +0.004 |
couple | +0.003 |
seek | +0.003 |
gentleman | +0.003 |
robat | +0.003 |
Neg logit contributions | |
aying | -0.006 |
fears | -0.005 |
arse | -0.005 |
uning | -0.005 |
fists | -0.005 |
Token | couple |
Token ID | 3155 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
rob | +0.047 |
seek | +0.044 |
robat | +0.035 |
seek | +0.032 |
Ar | +0.025 |
Neg logit contributions | |
aying | -0.139 |
wipes | -0.125 |
ours | -0.124 |
cleans | -0.123 |
uning | -0.122 |
Token | were |
Token ID | 547 |
Feature activation | +3.86e-01 |
Pos logit contributions | |
couple | +0.055 |
rob | +0.053 |
seek | +0.047 |
gentleman | +0.045 |
trainer | +0.045 |
Neg logit contributions | |
aying | -0.072 |
fears | -0.068 |
arse | -0.068 |
uning | -0.067 |
omed | -0.066 |
Token | feeling |
Token ID | 4203 |
Feature activation | +8.35e-01 |
Pos logit contributions | |
couple | +0.013 |
seek | +0.011 |
lady | +0.010 |
trainer | +0.010 |
man | +0.010 |
Neg logit contributions | |
aying | -0.023 |
Ow | -0.023 |
arse | -0.023 |
omed | -0.022 |
ours | -0.022 |
Token | tired |
Token ID | 10032 |
Feature activation | +3.43e-01 |
Pos logit contributions | |
couple | +0.027 |
l | +0.022 |
knowing | +0.022 |
man | +0.021 |
thank | +0.021 |
Neg logit contributions | |
arse | -0.050 |
uning | -0.048 |
Ow | -0.047 |
ains | -0.047 |
aying | -0.047 |
Token | , |
Token ID | 11 |
Feature activation | -6.93e-02 |
Pos logit contributions | |
couple | +0.011 |
rob | +0.008 |
man | +0.008 |
lady | +0.008 |
seek | +0.008 |
Neg logit contributions | |
aying | -0.021 |
arse | -0.021 |
fears | -0.021 |
ours | -0.020 |
Ow | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | +6.73e-02 |
Pos logit contributions | |
couple | +0.024 |
rob | +0.021 |
seek | +0.021 |
lady | +0.020 |
gentleman | +0.019 |
Neg logit contributions | |
aying | -0.040 |
fears | -0.038 |
arse | -0.038 |
uning | -0.037 |
Ow | -0.037 |
Token | |
Token ID | 220 |
Feature activation | +4.47e-01 |
Pos logit contributions | |
couple | +0.003 |
rob | +0.002 |
gentleman | +0.002 |
lady | +0.002 |
seek | +0.002 |
Neg logit contributions | |
aying | -0.004 |
fears | -0.003 |
arse | -0.003 |
uning | -0.003 |
adoes | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -2.76e-01 |
Pos logit contributions | |
couple | +0.006 |
rob | +0.005 |
lady | +0.005 |
gentleman | +0.005 |
man | +0.005 |
Neg logit contributions | |
fears | -0.033 |
arse | -0.032 |
aying | -0.031 |
adoes | -0.031 |
fists | -0.031 |
Token |
|
Token ID | 198 |
Feature activation | -8.34e-02 |
Pos logit contributions | |
aying | +0.016 |
arse | +0.015 |
fears | +0.015 |
omed | +0.014 |
uning | +0.014 |
Neg logit contributions | |
couple | -0.010 |
rob | -0.009 |
seek | -0.008 |
gentleman | -0.008 |
trainer | -0.008 |
Token | The |
Token ID | 464 |
Feature activation | +2.54e-02 |
Pos logit contributions | |
aying | +0.004 |
arse | +0.004 |
omed | +0.004 |
fears | +0.004 |
Ow | +0.004 |
Neg logit contributions | |
rob | -0.004 |
couple | -0.004 |
seek | -0.003 |
trainer | -0.003 |
gentleman | -0.003 |
Token | elderly |
Token ID | 13830 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
rob | +0.001 |
robat | +0.001 |
seek | +0.001 |
couple | +0.001 |
thank | +0.001 |
Neg logit contributions | |
aying | -0.002 |
fears | -0.001 |
wipes | -0.001 |
fists | -0.001 |
omed | -0.001 |
Token | lady |
Token ID | 10846 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
rob | +0.058 |
robat | +0.044 |
seek | +0.043 |
yak | +0.038 |
Ar | +0.036 |
Neg logit contributions | |
aying | -0.130 |
wipes | -0.127 |
cleans | -0.121 |
omed | -0.121 |
outed | -0.117 |
Token | laughed |
Token ID | 13818 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
couple | +0.048 |
rob | +0.043 |
seek | +0.040 |
lady | +0.039 |
gentleman | +0.039 |
Neg logit contributions | |
aying | -0.074 |
arse | -0.070 |
fears | -0.069 |
Ow | -0.068 |
uning | -0.068 |
Token | and |
Token ID | 290 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
couple | +0.006 |
lady | +0.005 |
man | +0.005 |
gentleman | +0.005 |
seek | +0.005 |
Neg logit contributions | |
aying | -0.015 |
fears | -0.015 |
arse | -0.015 |
uning | -0.015 |
ours | -0.015 |
Token | said |
Token ID | 531 |
Feature activation | +3.13e-01 |
Pos logit contributions | |
couple | +0.005 |
rob | +0.005 |
seek | +0.004 |
gentleman | +0.004 |
trainer | +0.004 |
Neg logit contributions | |
aying | -0.010 |
fears | -0.009 |
arse | -0.009 |
uning | -0.009 |
omed | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
couple | +0.009 |
lady | +0.008 |
gentleman | +0.007 |
rob | +0.007 |
man | +0.007 |
Neg logit contributions | |
aying | -0.020 |
fears | -0.020 |
arse | -0.020 |
ains | -0.019 |
oken | -0.019 |
Token | |
Token ID | 220 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
aying | +0.016 |
fears | +0.015 |
arse | +0.015 |
uning | +0.015 |
fists | +0.014 |
Neg logit contributions | |
couple | -0.013 |
rob | -0.013 |
seek | -0.012 |
gentleman | -0.012 |
lady | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | -8.33e-01 |
Pos logit contributions | |
fears | +0.010 |
aying | +0.010 |
arse | +0.010 |
uning | +0.009 |
fists | +0.009 |
Neg logit contributions | |
rob | -0.003 |
couple | -0.003 |
seek | -0.003 |
lady | -0.003 |
gentleman | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -3.88e-01 |
Pos logit contributions | |
aying | +0.041 |
Ow | +0.039 |
omed | +0.038 |
arse | +0.038 |
fears | +0.037 |
Neg logit contributions | |
couple | -0.035 |
rob | -0.033 |
gentleman | -0.030 |
seek | -0.030 |
trainer | -0.029 |
Token | After |
Token ID | 3260 |
Feature activation | +9.43e-01 |
Pos logit contributions | |
aying | +0.017 |
arse | +0.016 |
fears | +0.016 |
omed | +0.016 |
Ow | +0.016 |
Neg logit contributions | |
couple | -0.018 |
rob | -0.018 |
gentleman | -0.016 |
seek | -0.016 |
trainer | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | +5.72e-01 |
Pos logit contributions | |
couple | +0.036 |
rob | +0.035 |
seek | +0.031 |
gentleman | +0.030 |
trainer | +0.029 |
Neg logit contributions | |
aying | -0.050 |
fears | -0.047 |
arse | -0.047 |
uning | -0.046 |
omed | -0.046 |
Token | haircut |
Token ID | 45548 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
rob | +0.020 |
couple | +0.018 |
seek | +0.017 |
thank | +0.016 |
keeping | +0.016 |
Neg logit contributions | |
aying | -0.031 |
fears | -0.030 |
wipes | -0.029 |
fists | -0.029 |
arse | -0.029 |
Token | was |
Token ID | 373 |
Feature activation | +4.92e-01 |
Pos logit contributions | |
couple | +0.088 |
rob | +0.087 |
seek | +0.078 |
gentleman | +0.075 |
trainer | +0.074 |
Neg logit contributions | |
aying | -0.104 |
fears | -0.097 |
uning | -0.096 |
cleans | -0.095 |
wipes | -0.095 |
Token | done |
Token ID | 1760 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
couple | +0.019 |
rob | +0.018 |
seek | +0.017 |
lady | +0.016 |
thank | +0.016 |
Neg logit contributions | |
aying | -0.026 |
fears | -0.025 |
arse | -0.024 |
uning | -0.024 |
Ow | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | +4.90e-03 |
Pos logit contributions | |
couple | +0.009 |
rob | +0.008 |
seek | +0.007 |
gentleman | +0.007 |
lady | +0.007 |
Neg logit contributions | |
aying | -0.015 |
fears | -0.014 |
arse | -0.014 |
uning | -0.014 |
Ow | -0.013 |
Token | Tim |
Token ID | 5045 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
couple | +0.000 |
rob | +0.000 |
lady | +0.000 |
man | +0.000 |
seek | +0.000 |
Neg logit contributions | |
aying | -0.000 |
fears | -0.000 |
arse | -0.000 |
uning | -0.000 |
oken | -0.000 |
Token | my |
Token ID | 1820 |
Feature activation | -4.85e-01 |
Pos logit contributions | |
aying | +0.008 |
fears | +0.008 |
arse | +0.007 |
uning | +0.007 |
omed | +0.007 |
Neg logit contributions | |
couple | -0.010 |
rob | -0.010 |
seek | -0.009 |
gentleman | -0.009 |
trainer | -0.009 |
Token | goodbye |
Token ID | 24829 |
Feature activation | -5.02e-01 |
Pos logit contributions | |
arse | +0.011 |
aying | +0.011 |
fears | +0.010 |
Ow | +0.010 |
ains | +0.010 |
Neg logit contributions | |
couple | -0.005 |
lady | -0.004 |
man | -0.004 |
gentleman | -0.004 |
rob | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +6.90e-02 |
Pos logit contributions | |
aying | +0.027 |
fears | +0.026 |
arse | +0.026 |
uning | +0.025 |
omed | +0.025 |
Neg logit contributions | |
couple | -0.018 |
rob | -0.018 |
seek | -0.016 |
gentleman | -0.015 |
trainer | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
couple | +0.003 |
rob | +0.003 |
gentleman | +0.003 |
lady | +0.003 |
seek | +0.003 |
Neg logit contributions | |
aying | -0.003 |
arse | -0.003 |
fears | -0.003 |
uning | -0.003 |
iqu | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
aying | +0.007 |
arse | +0.007 |
fears | +0.007 |
omed | +0.007 |
Ow | +0.007 |
Neg logit contributions | |
couple | -0.005 |
rob | -0.005 |
seek | -0.004 |
gentleman | -0.004 |
trainer | -0.004 |
Token | The |
Token ID | 464 |
Feature activation | -2.63e-02 |
Pos logit contributions | |
aying | +0.008 |
arse | +0.007 |
fears | +0.007 |
omed | +0.007 |
Ow | +0.007 |
Neg logit contributions | |
couple | -0.007 |
rob | -0.007 |
seek | -0.006 |
gentleman | -0.006 |
trainer | -0.006 |
Token | elderly |
Token ID | 13830 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
aying | +0.001 |
fears | +0.001 |
arse | +0.001 |
uning | +0.001 |
fists | +0.001 |
Neg logit contributions | |
rob | -0.001 |
couple | -0.001 |
seek | -0.001 |
robat | -0.001 |
gentleman | -0.001 |
Token | woman |
Token ID | 2415 |
Feature activation | +5.69e-01 |
Pos logit contributions | |
rob | +0.057 |
seek | +0.049 |
robat | +0.043 |
seek | +0.040 |
Ar | +0.037 |
Neg logit contributions | |
aying | -0.125 |
wipes | -0.120 |
cleans | -0.118 |
fears | -0.116 |
uning | -0.115 |
Token | was |
Token ID | 373 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
couple | +0.021 |
rob | +0.020 |
seek | +0.018 |
gentleman | +0.017 |
trainer | +0.017 |
Neg logit contributions | |
aying | -0.031 |
arse | -0.029 |
fears | -0.029 |
uning | -0.029 |
Ow | -0.029 |
Token | so |
Token ID | 523 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
aying | +0.008 |
fears | +0.007 |
arse | +0.007 |
omed | +0.007 |
cleans | +0.007 |
Neg logit contributions | |
rob | -0.007 |
couple | -0.006 |
robat | -0.005 |
seek | -0.005 |
trainer | -0.005 |
Token | happy |
Token ID | 3772 |
Feature activation | -3.40e-01 |
Pos logit contributions | |
rob | +0.005 |
couple | +0.004 |
seek | +0.004 |
robat | +0.004 |
gentleman | +0.004 |
Neg logit contributions | |
aying | -0.008 |
fears | -0.007 |
arse | -0.007 |
omed | -0.007 |
uning | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
aying | +0.018 |
fears | +0.016 |
uning | +0.016 |
arse | +0.016 |
omed | +0.016 |
Neg logit contributions | |
rob | -0.014 |
couple | -0.013 |
seek | -0.011 |
trainer | -0.011 |
gentleman | -0.011 |
Token | day |
Token ID | 1110 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
aying | +0.024 |
fears | +0.023 |
arse | +0.023 |
uning | +0.023 |
omed | +0.022 |
Neg logit contributions | |
couple | -0.009 |
rob | -0.008 |
seek | -0.007 |
gentleman | -0.006 |
trainer | -0.006 |
Token | ?" |
Token ID | 1701 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
aying | +0.057 |
arse | +0.056 |
fears | +0.056 |
uning | +0.054 |
Ow | +0.053 |
Neg logit contributions | |
couple | -0.046 |
rob | -0.042 |
seek | -0.040 |
lady | -0.038 |
gentleman | -0.037 |
Token |
|
Token ID | 198 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
aying | +0.010 |
fears | +0.009 |
arse | +0.009 |
uning | +0.009 |
omed | +0.009 |
Neg logit contributions | |
couple | -0.010 |
rob | -0.010 |
seek | -0.009 |
gentleman | -0.008 |
lady | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
aying | +0.010 |
Ow | +0.009 |
arse | +0.009 |
cleans | +0.009 |
omed | +0.009 |
Neg logit contributions | |
couple | -0.008 |
rob | -0.008 |
seek | -0.007 |
trainer | -0.007 |
gentleman | -0.007 |
Token | The |
Token ID | 464 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
aying | +0.010 |
arse | +0.009 |
fears | +0.009 |
omed | +0.009 |
Ow | +0.009 |
Neg logit contributions | |
rob | -0.009 |
couple | -0.008 |
seek | -0.007 |
trainer | -0.007 |
robat | -0.007 |
Token | elderly |
Token ID | 13830 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
aying | +0.007 |
fears | +0.007 |
cleans | +0.007 |
ok | +0.007 |
arse | +0.006 |
Neg logit contributions | |
rob | -0.008 |
robat | -0.006 |
seek | -0.006 |
seek | -0.006 |
trainer | -0.006 |
Token | woman |
Token ID | 2415 |
Feature activation | +6.23e-01 |
Pos logit contributions | |
rob | +0.075 |
seek | +0.073 |
robat | +0.063 |
Ar | +0.061 |
seek | +0.059 |
Neg logit contributions | |
aying | -0.107 |
cleans | -0.106 |
wipes | -0.103 |
adoes | -0.095 |
uning | -0.094 |
Token | laughed |
Token ID | 13818 |
Feature activation | -2.65e-02 |
Pos logit contributions | |
couple | +0.018 |
lady | +0.015 |
man | +0.015 |
rob | +0.014 |
seek | +0.014 |
Neg logit contributions | |
aying | -0.040 |
arse | -0.040 |
uning | -0.038 |
Ow | -0.038 |
arding | -0.038 |
Token | and |
Token ID | 290 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
aying | +0.002 |
arse | +0.002 |
fears | +0.002 |
uning | +0.002 |
ours | +0.002 |
Neg logit contributions | |
couple | -0.001 |
rob | -0.001 |
lady | -0.001 |
man | -0.001 |
gentleman | -0.001 |
Token | said |
Token ID | 531 |
Feature activation | -7.87e-02 |
Pos logit contributions | |
aying | +0.012 |
fears | +0.012 |
arse | +0.012 |
wipes | +0.012 |
uning | +0.012 |
Neg logit contributions | |
couple | -0.006 |
rob | -0.006 |
seek | -0.005 |
gentleman | -0.005 |
trainer | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | -3.41e-01 |
Pos logit contributions | |
aying | +0.005 |
arse | +0.005 |
fears | +0.005 |
iqu | +0.005 |
ains | +0.005 |
Neg logit contributions | |
couple | -0.002 |
lady | -0.002 |
man | -0.001 |
rob | -0.001 |
gentleman | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +2.54e-01 |
Pos logit contributions | |
couple | +0.013 |
rob | +0.013 |
seek | +0.011 |
gentleman | +0.011 |
trainer | +0.011 |
Neg logit contributions | |
aying | -0.019 |
fears | -0.018 |
arse | -0.018 |
uning | -0.018 |
omed | -0.017 |
Token | |
Token ID | 220 |
Feature activation | +3.54e-01 |
Pos logit contributions | |
couple | +0.008 |
rob | +0.007 |
gentleman | +0.007 |
lady | +0.007 |
man | +0.007 |
Neg logit contributions | |
fears | -0.015 |
aying | -0.015 |
arse | -0.014 |
uning | -0.014 |
iqu | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | -2.61e-01 |
Pos logit contributions | |
couple | +0.005 |
rob | +0.004 |
lady | +0.003 |
man | +0.003 |
gentleman | +0.003 |
Neg logit contributions | |
fears | -0.026 |
arse | -0.025 |
aying | -0.025 |
adoes | -0.025 |
uning | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | -9.70e-02 |
Pos logit contributions | |
aying | +0.016 |
fears | +0.015 |
arse | +0.015 |
uning | +0.015 |
omed | +0.015 |
Neg logit contributions | |
couple | -0.008 |
rob | -0.007 |
seek | -0.007 |
gentleman | -0.006 |
lady | -0.006 |
Token | The |
Token ID | 464 |
Feature activation | +6.70e-02 |
Pos logit contributions | |
aying | +0.005 |
fears | +0.005 |
arse | +0.005 |
uning | +0.005 |
omed | +0.005 |
Neg logit contributions | |
couple | -0.004 |
rob | -0.004 |
seek | -0.003 |
gentleman | -0.003 |
trainer | -0.003 |
Token | elderly |
Token ID | 13830 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
rob | +0.002 |
couple | +0.002 |
seek | +0.002 |
robat | +0.002 |
thank | +0.001 |
Neg logit contributions | |
aying | -0.004 |
fears | -0.004 |
arse | -0.004 |
fists | -0.004 |
uning | -0.004 |
Token | person |
Token ID | 1048 |
Feature activation | +8.54e-01 |
Pos logit contributions | |
rob | +0.062 |
robat | +0.048 |
seek | +0.045 |
thank | +0.039 |
please | +0.038 |
Neg logit contributions | |
aying | -0.128 |
wipes | -0.119 |
adoes | -0.117 |
cleans | -0.116 |
fears | -0.115 |
Token | said |
Token ID | 531 |
Feature activation | +5.24e-01 |
Pos logit contributions | |
couple | +0.032 |
rob | +0.031 |
seek | +0.028 |
gentleman | +0.026 |
lady | +0.026 |
Neg logit contributions | |
aying | -0.047 |
arse | -0.044 |
fears | -0.044 |
uning | -0.043 |
Ow | -0.043 |
Token | , |
Token ID | 11 |
Feature activation | -2.62e-01 |
Pos logit contributions | |
couple | +0.015 |
lady | +0.012 |
seek | +0.012 |
man | +0.012 |
gentleman | +0.012 |
Neg logit contributions | |
aying | -0.034 |
arse | -0.033 |
fears | -0.033 |
iqu | -0.032 |
ains | -0.032 |
Token | " |
Token ID | 366 |
Feature activation | -2.63e-01 |
Pos logit contributions | |
aying | +0.016 |
fears | +0.016 |
arse | +0.016 |
uning | +0.015 |
iqu | +0.015 |
Neg logit contributions | |
couple | -0.008 |
rob | -0.007 |
lady | -0.007 |
man | -0.007 |
gentleman | -0.006 |
Token | You |
Token ID | 1639 |
Feature activation | -3.32e-02 |
Pos logit contributions | |
aying | +0.013 |
arse | +0.012 |
Ow | +0.011 |
fears | +0.011 |
omed | +0.011 |
Neg logit contributions | |
couple | -0.011 |
rob | -0.011 |
seek | -0.010 |
gentleman | -0.010 |
trainer | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | +6.47e-02 |
Pos logit contributions | |
couple | +0.029 |
rob | +0.027 |
seek | +0.025 |
gentleman | +0.024 |
lady | +0.024 |
Neg logit contributions | |
aying | -0.050 |
fears | -0.046 |
arse | -0.046 |
uning | -0.046 |
Ow | -0.046 |
Token | looked |
Token ID | 3114 |
Feature activation | -6.85e-02 |
Pos logit contributions | |
couple | +0.003 |
rob | +0.003 |
seek | +0.002 |
gentleman | +0.002 |
lady | +0.002 |
Neg logit contributions | |
aying | -0.003 |
fears | -0.003 |
arse | -0.003 |
uning | -0.003 |
Ow | -0.003 |
Token | back |
Token ID | 736 |
Feature activation | -9.65e-03 |
Pos logit contributions | |
aying | +0.004 |
uning | +0.004 |
Ow | +0.004 |
arse | +0.004 |
fears | +0.004 |
Neg logit contributions | |
couple | -0.002 |
rob | -0.002 |
gentleman | -0.002 |
seek | -0.002 |
lady | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | +8.31e-02 |
Pos logit contributions | |
aying | +0.001 |
arse | +0.000 |
fears | +0.000 |
uning | +0.000 |
Ow | +0.000 |
Neg logit contributions | |
couple | -0.000 |
rob | -0.000 |
seek | -0.000 |
gentleman | -0.000 |
lady | -0.000 |
Token | The |
Token ID | 383 |
Feature activation | +9.83e-02 |
Pos logit contributions | |
couple | +0.003 |
gentleman | +0.002 |
seek | +0.002 |
lady | +0.002 |
man | +0.002 |
Neg logit contributions | |
aying | -0.005 |
fears | -0.005 |
arse | -0.005 |
uning | -0.005 |
adoes | -0.005 |
Token | elderly |
Token ID | 13830 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
rob | +0.003 |
seek | +0.003 |
robat | +0.002 |
seek | +0.002 |
couple | +0.002 |
Neg logit contributions | |
fears | -0.006 |
aying | -0.006 |
fists | -0.006 |
arse | -0.006 |
wipes | -0.006 |
Token | woman |
Token ID | 2415 |
Feature activation | +7.72e-01 |
Pos logit contributions | |
rob | +0.078 |
seek | +0.072 |
robat | +0.068 |
yak | +0.061 |
Ar | +0.060 |
Neg logit contributions | |
aying | -0.105 |
wipes | -0.102 |
cleans | -0.101 |
adoes | -0.094 |
omed | -0.092 |
Token | was |
Token ID | 373 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
couple | +0.029 |
rob | +0.026 |
seek | +0.025 |
lady | +0.024 |
gentleman | +0.024 |
Neg logit contributions | |
aying | -0.043 |
arse | -0.041 |
uning | -0.040 |
fears | -0.040 |
Ow | -0.040 |
Token | not |
Token ID | 407 |
Feature activation | -5.28e-03 |
Pos logit contributions | |
rob | +0.001 |
couple | +0.001 |
seek | +0.001 |
robat | +0.001 |
gentleman | +0.001 |
Neg logit contributions | |
aying | -0.001 |
fears | -0.001 |
arse | -0.001 |
omed | -0.001 |
fists | -0.001 |
Token | angry |
Token ID | 7954 |
Feature activation | +6.66e-01 |
Pos logit contributions | |
arse | +0.000 |
aying | +0.000 |
ok | +0.000 |
fears | +0.000 |
streaming | +0.000 |
Neg logit contributions | |
rob | -0.000 |
couple | -0.000 |
robat | -0.000 |
trainer | -0.000 |
gentleman | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
couple | +0.033 |
rob | +0.032 |
seek | +0.029 |
gentleman | +0.028 |
trainer | +0.028 |
Neg logit contributions | |
aying | -0.029 |
fears | -0.026 |
arse | -0.026 |
uning | -0.026 |
omed | -0.025 |
Token | okay |
Token ID | 8788 |
Feature activation | -7.94e-01 |
Pos logit contributions | |
couple | +0.004 |
rob | +0.003 |
seek | +0.003 |
lady | +0.003 |
gentleman | +0.003 |
Neg logit contributions | |
aying | -0.002 |
fears | -0.001 |
arse | -0.001 |
uning | -0.001 |
omed | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -2.73e-01 |
Pos logit contributions | |
aying | +0.049 |
fears | +0.047 |
arse | +0.047 |
uning | +0.046 |
Ow | +0.045 |
Neg logit contributions | |
couple | -0.024 |
rob | -0.021 |
seek | -0.020 |
lady | -0.019 |
gentleman | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | +6.93e-02 |
Pos logit contributions | |
aying | +0.014 |
fears | +0.013 |
arse | +0.013 |
uning | +0.012 |
omed | +0.012 |
Neg logit contributions | |
couple | -0.011 |
rob | -0.011 |
seek | -0.010 |
gentleman | -0.010 |
lady | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | -2.87e-02 |
Pos logit contributions | |
couple | +0.003 |
rob | +0.002 |
seek | +0.002 |
gentleman | +0.002 |
trainer | +0.002 |
Neg logit contributions | |
aying | -0.004 |
arse | -0.004 |
fears | -0.004 |
Ow | -0.004 |
omed | -0.004 |
Token | The |
Token ID | 464 |
Feature activation | +5.53e-02 |
Pos logit contributions | |
aying | +0.002 |
arse | +0.001 |
fears | +0.001 |
omed | +0.001 |
uning | +0.001 |
Neg logit contributions | |
couple | -0.001 |
rob | -0.001 |
seek | -0.001 |
gentleman | -0.001 |
trainer | -0.001 |
Token | elderly |
Token ID | 13830 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
couple | +0.002 |
rob | +0.002 |
seek | +0.001 |
gentleman | +0.001 |
thank | +0.001 |
Neg logit contributions | |
aying | -0.003 |
fears | -0.003 |
arse | -0.003 |
uning | -0.003 |
Ow | -0.003 |
Token | man |
Token ID | 582 |
Feature activation | +6.99e-01 |
Pos logit contributions | |
rob | +0.047 |
seek | +0.042 |
robat | +0.036 |
thank | +0.032 |
seek | +0.032 |
Neg logit contributions | |
aying | -0.130 |
wipes | -0.125 |
uning | -0.123 |
cleans | -0.123 |
ours | -0.120 |
Token | smiled |
Token ID | 13541 |
Feature activation | +5.10e-01 |
Pos logit contributions | |
couple | +0.021 |
rob | +0.018 |
lady | +0.017 |
man | +0.016 |
trainer | +0.016 |
Neg logit contributions | |
aying | -0.045 |
arse | -0.043 |
fears | -0.042 |
Ow | -0.042 |
adoes | -0.042 |
Token | weak |
Token ID | 4939 |
Feature activation | -3.91e-01 |
Pos logit contributions | |
rob | +0.020 |
couple | +0.020 |
seek | +0.017 |
gentleman | +0.016 |
trainer | +0.016 |
Neg logit contributions | |
aying | -0.027 |
fears | -0.026 |
arse | -0.025 |
uning | -0.025 |
omed | -0.025 |
Token | ly |
Token ID | 306 |
Feature activation | -8.00e-01 |
Pos logit contributions | |
aying | +0.021 |
arse | +0.020 |
fears | +0.020 |
Ow | +0.019 |
uning | +0.019 |
Neg logit contributions | |
couple | -0.015 |
rob | -0.014 |
lady | -0.013 |
gentleman | -0.013 |
man | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | -3.12e-01 |
Pos logit contributions | |
aying | +0.048 |
arse | +0.046 |
fears | +0.045 |
uning | +0.045 |
Ow | +0.045 |
Neg logit contributions | |
couple | -0.026 |
rob | -0.023 |
gentleman | -0.021 |
seek | -0.021 |
lady | -0.021 |
Token | feel |
Token ID | 1254 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
couple | +0.004 |
seek | +0.004 |
rob | +0.004 |
lady | +0.003 |
thank | +0.003 |
Neg logit contributions | |
aying | -0.005 |
omed | -0.005 |
uning | -0.005 |
fears | -0.005 |
arse | -0.005 |
Token | excited |
Token ID | 6568 |
Feature activation | +9.74e-02 |
Pos logit contributions | |
couple | +0.001 |
rob | +0.001 |
seek | +0.001 |
lady | +0.000 |
gentleman | +0.000 |
Neg logit contributions | |
aying | -0.001 |
arse | -0.001 |
uning | -0.001 |
Ow | -0.001 |
omed | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -3.87e-01 |
Pos logit contributions | |
couple | +0.003 |
rob | +0.003 |
seek | +0.003 |
gentleman | +0.003 |
lady | +0.003 |
Neg logit contributions | |
aying | -0.006 |
fears | -0.005 |
arse | -0.005 |
uning | -0.005 |
Ow | -0.005 |
Token | When |
Token ID | 1649 |
Feature activation | +2.27e-02 |
Pos logit contributions | |
aying | +0.020 |
fears | +0.019 |
oken | +0.019 |
uning | +0.019 |
arse | +0.019 |
Neg logit contributions | |
couple | -0.015 |
gentleman | -0.012 |
rob | -0.012 |
lady | -0.012 |
seek | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | +3.54e-01 |
Pos logit contributions | |
couple | +0.001 |
seek | +0.001 |
rob | +0.001 |
lady | +0.001 |
man | +0.001 |
Neg logit contributions | |
aying | -0.001 |
fears | -0.001 |
uning | -0.001 |
arse | -0.001 |
outed | -0.001 |
Token | haircut |
Token ID | 45548 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
rob | +0.016 |
couple | +0.014 |
seek | +0.013 |
trainer | +0.013 |
robat | +0.013 |
Neg logit contributions | |
aying | -0.016 |
fears | -0.016 |
wipes | -0.015 |
fists | -0.015 |
arse | -0.015 |
Token | was |
Token ID | 373 |
Feature activation | +4.91e-01 |
Pos logit contributions | |
couple | +0.079 |
rob | +0.076 |
seek | +0.068 |
gentleman | +0.065 |
lady | +0.063 |
Neg logit contributions | |
aying | -0.114 |
fears | -0.107 |
arse | -0.107 |
uning | -0.105 |
omed | -0.103 |
Token | done |
Token ID | 1760 |
Feature activation | +3.79e-01 |
Pos logit contributions | |
couple | +0.016 |
seek | +0.014 |
thank | +0.013 |
lady | +0.013 |
rob | +0.013 |
Neg logit contributions | |
aying | -0.028 |
fears | -0.028 |
Ow | -0.028 |
uning | -0.028 |
arse | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | +9.32e-03 |
Pos logit contributions | |
couple | +0.011 |
seek | +0.009 |
rob | +0.009 |
lady | +0.008 |
gentleman | +0.008 |
Neg logit contributions | |
fears | -0.024 |
aying | -0.024 |
Ow | -0.023 |
arse | -0.023 |
outed | -0.023 |
Token | Little |
Token ID | 7703 |
Feature activation | -7.62e-01 |
Pos logit contributions | |
couple | +0.000 |
seek | +0.000 |
rob | +0.000 |
lady | +0.000 |
man | +0.000 |
Neg logit contributions | |
aying | -0.001 |
arse | -0.000 |
fears | -0.000 |
uning | -0.000 |
omed | -0.000 |
Token | one |
Token ID | 530 |
Feature activation | +3.53e-01 |
Pos logit contributions | |
aying | +0.041 |
fears | +0.039 |
ither | +0.038 |
oken | +0.038 |
arse | +0.037 |
Neg logit contributions | |
couple | -0.031 |
man | -0.028 |
lady | -0.028 |
seek | -0.028 |
rob | -0.027 |
Token | home |
Token ID | 1363 |
Feature activation | +6.34e-01 |
Pos logit contributions | |
aying | +0.006 |
fears | +0.006 |
uning | +0.005 |
arse | +0.005 |
omed | +0.005 |
Neg logit contributions | |
rob | -0.004 |
couple | -0.003 |
seek | -0.003 |
gentleman | -0.003 |
trainer | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
couple | +0.022 |
rob | +0.021 |
seek | +0.018 |
gentleman | +0.017 |
lady | +0.017 |
Neg logit contributions | |
aying | -0.037 |
fears | -0.035 |
arse | -0.034 |
uning | -0.034 |
Ow | -0.033 |
Token |
|
Token ID | 198 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
couple | +0.006 |
rob | +0.006 |
seek | +0.006 |
gentleman | +0.005 |
trainer | +0.005 |
Neg logit contributions | |
aying | -0.008 |
fears | -0.008 |
arse | -0.008 |
uning | -0.008 |
Ow | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
couple | +0.008 |
rob | +0.008 |
seek | +0.007 |
thank | +0.007 |
gentleman | +0.007 |
Neg logit contributions | |
aying | -0.006 |
Ow | -0.006 |
fears | -0.006 |
omed | -0.006 |
arse | -0.006 |
Token | The |
Token ID | 464 |
Feature activation | -6.89e-02 |
Pos logit contributions | |
couple | +0.008 |
rob | +0.008 |
seek | +0.007 |
gentleman | +0.007 |
trainer | +0.007 |
Neg logit contributions | |
aying | -0.008 |
arse | -0.007 |
omed | -0.007 |
fears | -0.007 |
Ow | -0.007 |
Token | elderly |
Token ID | 13830 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
aying | +0.004 |
fears | +0.004 |
arse | +0.004 |
uning | +0.004 |
omed | +0.004 |
Neg logit contributions | |
rob | -0.002 |
couple | -0.002 |
seek | -0.002 |
gentleman | -0.002 |
thank | -0.002 |
Token | lady |
Token ID | 10846 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
rob | +0.048 |
seek | +0.042 |
robat | +0.035 |
couple | +0.032 |
seek | +0.032 |
Neg logit contributions | |
aying | -0.130 |
wipes | -0.124 |
cleans | -0.121 |
adoes | -0.119 |
fears | -0.116 |
Token | was |
Token ID | 373 |
Feature activation | +6.78e-02 |
Pos logit contributions | |
couple | +0.041 |
rob | +0.040 |
seek | +0.035 |
gentleman | +0.034 |
trainer | +0.033 |
Neg logit contributions | |
aying | -0.059 |
fears | -0.056 |
arse | -0.056 |
uning | -0.055 |
omed | -0.054 |
Token | very |
Token ID | 845 |
Feature activation | +7.42e-01 |
Pos logit contributions | |
rob | +0.003 |
couple | +0.003 |
robat | +0.002 |
seek | +0.002 |
gentleman | +0.002 |
Neg logit contributions | |
aying | -0.003 |
fears | -0.003 |
arse | -0.003 |
omed | -0.003 |
ok | -0.003 |
Token | happy |
Token ID | 3772 |
Feature activation | -3.35e-02 |
Pos logit contributions | |
rob | +0.032 |
couple | +0.027 |
robat | +0.026 |
seek | +0.024 |
gentleman | +0.023 |
Neg logit contributions | |
aying | -0.038 |
fears | -0.037 |
outed | -0.035 |
omed | -0.035 |
arse | -0.035 |
Token | to |
Token ID | 284 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
aying | +0.002 |
fears | +0.002 |
uning | +0.002 |
wipes | +0.001 |
omed | +0.001 |
Neg logit contributions | |
rob | -0.001 |
couple | -0.001 |
seek | -0.001 |
robat | -0.001 |
trainer | -0.001 |
Token | She |
Token ID | 1375 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
couple | +0.014 |
rob | +0.013 |
lady | +0.012 |
gentleman | +0.012 |
man | +0.011 |
Neg logit contributions | |
aying | -0.024 |
fears | -0.023 |
arse | -0.023 |
uning | -0.022 |
omed | -0.022 |
Token | takes |
Token ID | 2753 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
couple | +0.005 |
rob | +0.005 |
man | +0.005 |
lady | +0.005 |
trainer | +0.005 |
Neg logit contributions | |
arse | -0.005 |
aying | -0.005 |
Ow | -0.004 |
uning | -0.004 |
arding | -0.004 |
Token | out |
Token ID | 503 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
aying | +0.010 |
fears | +0.009 |
arse | +0.009 |
omed | +0.009 |
uning | +0.009 |
Neg logit contributions | |
couple | -0.008 |
rob | -0.008 |
seek | -0.007 |
lady | -0.007 |
gentleman | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | +7.28e-02 |
Pos logit contributions | |
aying | +0.016 |
arse | +0.016 |
fears | +0.015 |
arding | +0.015 |
uning | +0.015 |
Neg logit contributions | |
couple | -0.007 |
lady | -0.006 |
man | -0.005 |
rob | -0.005 |
seek | -0.005 |
Token | magn |
Token ID | 7842 |
Feature activation | -2.45e-01 |
Pos logit contributions | |
couple | +0.003 |
seek | +0.003 |
rob | +0.003 |
lady | +0.003 |
man | +0.003 |
Neg logit contributions | |
aying | -0.004 |
uning | -0.003 |
fears | -0.003 |
arse | -0.003 |
adoes | -0.003 |
Token | ifying |
Token ID | 4035 |
Feature activation | -2.40e+00 |
Pos logit contributions | |
fears | +0.006 |
cleans | +0.006 |
wipes | +0.006 |
Acc | +0.006 |
adoes | +0.006 |
Neg logit contributions | |
couple | -0.015 |
rob | -0.013 |
trainer | -0.013 |
amer | -0.013 |
seek | -0.013 |
Token | glass |
Token ID | 5405 |
Feature activation | -7.67e-02 |
Pos logit contributions | |
aying | +0.083 |
fears | +0.076 |
Ow | +0.074 |
arse | +0.073 |
uning | +0.073 |
Neg logit contributions | |
couple | -0.136 |
rob | -0.134 |
seek | -0.123 |
gentleman | -0.121 |
trainer | -0.119 |
Token | . |
Token ID | 13 |
Feature activation | +5.92e-01 |
Pos logit contributions | |
aying | +0.004 |
fears | +0.004 |
arse | +0.004 |
uning | +0.004 |
omed | +0.004 |
Neg logit contributions | |
couple | -0.003 |
rob | -0.003 |
seek | -0.003 |
gentleman | -0.003 |
trainer | -0.002 |
Token | She |
Token ID | 1375 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
couple | +0.019 |
lady | +0.018 |
man | +0.016 |
gentleman | +0.015 |
rob | +0.015 |
Neg logit contributions | |
fears | -0.035 |
arse | -0.035 |
aying | -0.034 |
oken | -0.032 |
iqu | -0.032 |
Token | looks |
Token ID | 3073 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
couple | +0.005 |
rob | +0.005 |
seek | +0.004 |
lady | +0.004 |
trainer | +0.004 |
Neg logit contributions | |
arse | -0.004 |
aying | -0.004 |
Ow | -0.004 |
uning | -0.004 |
arding | -0.004 |
Token | at |
Token ID | 379 |
Feature activation | +3.93e-02 |
Pos logit contributions | |
couple | +0.001 |
rob | +0.001 |
seek | +0.001 |
lady | +0.001 |
man | +0.001 |
Neg logit contributions | |
aying | -0.001 |
arse | -0.001 |
uning | -0.001 |
fears | -0.001 |
Ow | -0.001 |
Token | a |
Token ID | 257 |
Feature activation | +3.75e-01 |
Pos logit contributions | |
couple | +0.006 |
rob | +0.006 |
seek | +0.005 |
gentleman | +0.005 |
trainer | +0.005 |
Neg logit contributions | |
aying | -0.007 |
fears | -0.006 |
arse | -0.006 |
Ow | -0.006 |
uning | -0.006 |
Token | flashlight |
Token ID | 38371 |
Feature activation | -4.79e-01 |
Pos logit contributions | |
couple | +0.016 |
rob | +0.015 |
seek | +0.014 |
gentleman | +0.013 |
trainer | +0.013 |
Neg logit contributions | |
aying | -0.019 |
fears | -0.018 |
arse | -0.018 |
uning | -0.017 |
omed | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | +3.40e-02 |
Pos logit contributions | |
aying | +0.028 |
fears | +0.027 |
arse | +0.027 |
Ow | +0.027 |
uning | +0.026 |
Neg logit contributions | |
couple | -0.016 |
rob | -0.015 |
seek | -0.013 |
lady | -0.013 |
trainer | -0.013 |
Token | a |
Token ID | 257 |
Feature activation | -3.08e-01 |
Pos logit contributions | |
couple | +0.001 |
rob | +0.001 |
seek | +0.001 |
gentleman | +0.001 |
trainer | +0.001 |
Neg logit contributions | |
aying | -0.002 |
arse | -0.002 |
fears | -0.002 |
uning | -0.002 |
omed | -0.002 |
Token | magn |
Token ID | 7842 |
Feature activation | -5.86e-01 |
Pos logit contributions | |
aying | +0.015 |
arse | +0.014 |
fears | +0.014 |
uning | +0.014 |
omed | +0.014 |
Neg logit contributions | |
couple | -0.013 |
rob | -0.013 |
seek | -0.012 |
gentleman | -0.011 |
lady | -0.011 |
Token | ifying |
Token ID | 4035 |
Feature activation | -2.38e+00 |
Pos logit contributions | |
fears | +0.015 |
cleans | +0.014 |
wipes | +0.014 |
adoes | +0.013 |
Acc | +0.013 |
Neg logit contributions | |
couple | -0.037 |
rob | -0.036 |
trainer | -0.034 |
robat | -0.033 |
king | -0.033 |
Token | glass |
Token ID | 5405 |
Feature activation | -7.18e-02 |
Pos logit contributions | |
aying | +0.079 |
Ow | +0.073 |
fears | +0.072 |
uning | +0.069 |
arse | +0.068 |
Neg logit contributions | |
couple | -0.138 |
rob | -0.136 |
gentleman | -0.125 |
seek | -0.123 |
trainer | -0.121 |
Token | to |
Token ID | 284 |
Feature activation | +4.28e-01 |
Pos logit contributions | |
aying | +0.004 |
fears | +0.004 |
arse | +0.004 |
uning | +0.004 |
omed | +0.004 |
Neg logit contributions | |
couple | -0.003 |
rob | -0.003 |
seek | -0.002 |
gentleman | -0.002 |
thank | -0.002 |
Token | examine |
Token ID | 10716 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
couple | +0.021 |
rob | +0.021 |
gentleman | +0.018 |
seek | +0.018 |
thank | +0.017 |
Neg logit contributions | |
aying | -0.019 |
fears | -0.017 |
uning | -0.017 |
arse | -0.017 |
wipes | -0.016 |
Token | their |
Token ID | 511 |
Feature activation | +2.59e-01 |
Pos logit contributions | |
couple | +0.007 |
rob | +0.007 |
seek | +0.006 |
lady | +0.006 |
man | +0.006 |
Neg logit contributions | |
arse | -0.013 |
aying | -0.013 |
fears | -0.013 |
uning | -0.013 |
omed | -0.012 |
Token | finds |
Token ID | 7228 |
Feature activation | +8.15e-02 |
Pos logit contributions | |
couple | +0.011 |
rob | +0.010 |
seek | +0.010 |
lady | +0.009 |
trainer | +0.009 |
Neg logit contributions | |
aying | -0.013 |
arse | -0.012 |
omed | -0.012 |
uning | -0.011 |
fears | -0.011 |
Token | He |
Token ID | 679 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
couple | +0.023 |
rob | +0.022 |
seek | +0.021 |
gentleman | +0.020 |
lady | +0.020 |
Neg logit contributions | |
aying | -0.027 |
arse | -0.025 |
fears | -0.025 |
uning | -0.024 |
omed | -0.024 |
Token | took |
Token ID | 1718 |
Feature activation | -8.16e-03 |
Pos logit contributions | |
aying | +0.008 |
fears | +0.007 |
cleans | +0.007 |
wipes | +0.007 |
outed | +0.007 |
Neg logit contributions | |
rob | -0.008 |
couple | -0.008 |
robat | -0.007 |
gentleman | -0.007 |
thank | -0.007 |
Token | out |
Token ID | 503 |
Feature activation | +4.92e-02 |
Pos logit contributions | |
aying | +0.000 |
fears | +0.000 |
arse | +0.000 |
uning | +0.000 |
Ow | +0.000 |
Neg logit contributions | |
couple | -0.000 |
rob | -0.000 |
seek | -0.000 |
gentleman | -0.000 |
trainer | -0.000 |
Token | a |
Token ID | 257 |
Feature activation | +4.32e-01 |
Pos logit contributions | |
couple | +0.001 |
rob | +0.001 |
lady | +0.001 |
man | +0.001 |
seek | +0.001 |
Neg logit contributions | |
aying | -0.003 |
arse | -0.003 |
uning | -0.003 |
arding | -0.003 |
ours | -0.003 |
Token | magn |
Token ID | 7842 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
couple | +0.018 |
rob | +0.018 |
seek | +0.016 |
gentleman | +0.016 |
lady | +0.015 |
Neg logit contributions | |
aying | -0.022 |
fears | -0.020 |
arse | -0.020 |
uning | -0.020 |
Ow | -0.019 |
Token | ifying |
Token ID | 4035 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
fears | +0.004 |
aying | +0.004 |
cleans | +0.004 |
wipes | +0.003 |
adoes | +0.003 |
Neg logit contributions | |
couple | -0.012 |
rob | -0.011 |
trainer | -0.010 |
gentleman | -0.010 |
seek | -0.010 |
Token | glass |
Token ID | 5405 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
aying | +0.074 |
Ow | +0.068 |
fears | +0.068 |
omed | +0.065 |
uning | +0.065 |
Neg logit contributions | |
couple | -0.125 |
rob | -0.125 |
seek | -0.113 |
gentleman | -0.112 |
trainer | -0.110 |
Token | and |
Token ID | 290 |
Feature activation | +3.26e-01 |
Pos logit contributions | |
couple | +0.006 |
rob | +0.006 |
seek | +0.005 |
gentleman | +0.005 |
trainer | +0.005 |
Neg logit contributions | |
aying | -0.008 |
fears | -0.007 |
arse | -0.007 |
uning | -0.007 |
omed | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | -8.26e-02 |
Pos logit contributions | |
rob | +0.014 |
couple | +0.013 |
gentleman | +0.012 |
robat | +0.011 |
trainer | +0.011 |
Neg logit contributions | |
aying | -0.015 |
fears | -0.015 |
wipes | -0.015 |
cleans | -0.015 |
uning | -0.014 |
Token | notebook |
Token ID | 20922 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
aying | +0.004 |
fears | +0.004 |
arse | +0.004 |
uning | +0.004 |
omed | +0.004 |
Neg logit contributions | |
couple | -0.004 |
rob | -0.003 |
seek | -0.003 |
gentleman | -0.003 |
trainer | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | +2.53e-01 |
Pos logit contributions | |
rob | +0.007 |
couple | +0.007 |
gentleman | +0.006 |
seek | +0.006 |
robat | +0.006 |
Neg logit contributions | |
aying | -0.008 |
omed | -0.007 |
ows | -0.007 |
wipes | -0.007 |
uning | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | -4.81e-02 |
Pos logit contributions | |
aying | +0.014 |
arse | +0.014 |
arding | +0.013 |
uning | +0.013 |
fears | +0.013 |
Neg logit contributions | |
couple | -0.004 |
lady | -0.003 |
l | -0.003 |
seek | -0.002 |
man | -0.002 |
Token | necklace |
Token ID | 38987 |
Feature activation | +8.12e-01 |
Pos logit contributions | |
aying | +0.003 |
arse | +0.002 |
fears | +0.002 |
uning | +0.002 |
omed | +0.002 |
Neg logit contributions | |
couple | -0.002 |
rob | -0.002 |
seek | -0.002 |
lady | -0.002 |
gentleman | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
rob | +0.041 |
couple | +0.038 |
gentleman | +0.034 |
seek | +0.034 |
businessman | +0.033 |
Neg logit contributions | |
aying | -0.035 |
ows | -0.032 |
icking | -0.032 |
uning | -0.032 |
adoes | -0.031 |
Token | the |
Token ID | 262 |
Feature activation | -3.92e-01 |
Pos logit contributions | |
aying | +0.005 |
wipes | +0.005 |
fears | +0.005 |
cleans | +0.005 |
uning | +0.005 |
Neg logit contributions | |
rob | -0.005 |
couple | -0.005 |
gentleman | -0.004 |
seek | -0.004 |
robat | -0.004 |
Token | magn |
Token ID | 7842 |
Feature activation | -2.80e-01 |
Pos logit contributions | |
aying | +0.021 |
fears | +0.020 |
arse | +0.019 |
uning | +0.019 |
Ow | +0.019 |
Neg logit contributions | |
couple | -0.015 |
rob | -0.015 |
seek | -0.013 |
gentleman | -0.013 |
trainer | -0.012 |
Token | ifying |
Token ID | 4035 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
aying | +0.007 |
fears | +0.007 |
cleans | +0.006 |
wipes | +0.006 |
adoes | +0.006 |
Neg logit contributions | |
couple | -0.018 |
rob | -0.018 |
seek | -0.016 |
trainer | -0.016 |
gentleman | -0.016 |
Token | glass |
Token ID | 5405 |
Feature activation | +2.47e-03 |
Pos logit contributions | |
Ow | +0.057 |
aying | +0.056 |
wipes | +0.050 |
fears | +0.050 |
cleans | +0.049 |
Neg logit contributions | |
couple | -0.142 |
rob | -0.142 |
gentleman | -0.132 |
seek | -0.130 |
robat | -0.128 |
Token | and |
Token ID | 290 |
Feature activation | -7.37e-03 |
Pos logit contributions | |
couple | +0.000 |
rob | +0.000 |
seek | +0.000 |
gentleman | +0.000 |
trainer | +0.000 |
Neg logit contributions | |
aying | -0.000 |
fears | -0.000 |
arse | -0.000 |
uning | -0.000 |
omed | -0.000 |
Token | started |
Token ID | 2067 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
wipes | +0.000 |
cleans | +0.000 |
aying | +0.000 |
fears | +0.000 |
fists | +0.000 |
Neg logit contributions | |
rob | -0.000 |
couple | -0.000 |
gentleman | -0.000 |
robat | -0.000 |
trainer | -0.000 |
Token | to |
Token ID | 284 |
Feature activation | +3.19e-01 |
Pos logit contributions | |
couple | +0.010 |
rob | +0.010 |
seek | +0.009 |
gentleman | +0.008 |
trainer | +0.008 |
Neg logit contributions | |
aying | -0.014 |
fears | -0.014 |
arse | -0.013 |
uning | -0.013 |
omed | -0.013 |
Token | examine |
Token ID | 10716 |
Feature activation | +2.58e-01 |
Pos logit contributions | |
couple | +0.013 |
seek | +0.013 |
rob | +0.013 |
thank | +0.011 |
lady | +0.011 |
Neg logit contributions | |
aying | -0.016 |
omed | -0.015 |
arse | -0.015 |
fears | -0.015 |
uning | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
arse | +0.013 |
aying | +0.012 |
omed | +0.012 |
fears | +0.012 |
arding | +0.011 |
Neg logit contributions | |
couple | -0.006 |
lady | -0.005 |
rob | -0.005 |
trainer | -0.005 |
l | -0.005 |
Token | necklace |
Token ID | 38987 |
Feature activation | +8.16e-01 |
Pos logit contributions | |
aying | +0.008 |
fears | +0.008 |
arse | +0.008 |
uning | +0.008 |
omed | +0.007 |
Neg logit contributions | |
couple | -0.007 |
rob | -0.007 |
seek | -0.006 |
gentleman | -0.006 |
trainer | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | +9.57e-02 |
Pos logit contributions | |
rob | +0.036 |
couple | +0.035 |
seek | +0.030 |
gentleman | +0.029 |
thank | +0.029 |
Neg logit contributions | |
aying | -0.040 |
fears | -0.036 |
uning | -0.036 |
omed | -0.036 |
cleans | -0.035 |
Token | the |
Token ID | 262 |
Feature activation | -7.03e-01 |
Pos logit contributions | |
rob | +0.004 |
couple | +0.004 |
robat | +0.004 |
thank | +0.003 |
seek | +0.003 |
Neg logit contributions | |
cleans | -0.005 |
wipes | -0.005 |
ours | -0.004 |
fists | -0.004 |
fears | -0.004 |
Token | magn |
Token ID | 7842 |
Feature activation | -4.35e-01 |
Pos logit contributions | |
aying | +0.035 |
fears | +0.033 |
arse | +0.033 |
uning | +0.032 |
Ow | +0.032 |
Neg logit contributions | |
couple | -0.029 |
rob | -0.029 |
seek | -0.025 |
gentleman | -0.025 |
trainer | -0.024 |
Token | ifying |
Token ID | 4035 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
fears | +0.011 |
aying | +0.011 |
wipes | +0.010 |
cleans | +0.010 |
adoes | +0.010 |
Neg logit contributions | |
couple | -0.028 |
rob | -0.027 |
trainer | -0.024 |
seek | -0.024 |
robat | -0.024 |
Token | glass |
Token ID | 5405 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
aying | +0.053 |
Ow | +0.052 |
wipes | +0.047 |
fears | +0.047 |
cleans | +0.047 |
Neg logit contributions | |
rob | -0.146 |
couple | -0.144 |
gentleman | -0.135 |
seek | -0.133 |
robat | -0.132 |
Token | and |
Token ID | 290 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
aying | +0.001 |
fears | +0.001 |
arse | +0.001 |
uning | +0.001 |
omed | +0.001 |
Neg logit contributions | |
couple | -0.001 |
rob | -0.001 |
seek | -0.001 |
gentleman | -0.001 |
trainer | -0.001 |
Token | told |
Token ID | 1297 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
rob | +0.006 |
couple | +0.006 |
robat | +0.005 |
gentleman | +0.005 |
seek | +0.005 |
Neg logit contributions | |
cleans | -0.007 |
wipes | -0.006 |
aying | -0.006 |
fears | -0.006 |
fists | -0.006 |
Token | him |
Token ID | 683 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
aying | +0.009 |
arse | +0.008 |
fears | +0.008 |
uning | +0.008 |
omed | +0.008 |
Neg logit contributions | |
couple | -0.007 |
rob | -0.007 |
seek | -0.006 |
gentleman | -0.006 |
trainer | -0.006 |
Token | that |
Token ID | 326 |
Feature activation | -4.74e-01 |
Pos logit contributions | |
couple | +0.001 |
lady | +0.001 |
ranger | +0.001 |
man | +0.001 |
trainer | +0.001 |
Neg logit contributions | |
arse | -0.002 |
aying | -0.002 |
Ow | -0.001 |
uning | -0.001 |
akes | -0.001 |
Token | She |
Token ID | 1375 |
Feature activation | +2.38e-01 |
Pos logit contributions | |
couple | +0.027 |
rob | +0.025 |
seek | +0.023 |
gentleman | +0.023 |
lady | +0.023 |
Neg logit contributions | |
aying | -0.029 |
arse | -0.027 |
fears | -0.027 |
omed | -0.026 |
Ow | -0.026 |
Token | took |
Token ID | 1718 |
Feature activation | -8.02e-02 |
Pos logit contributions | |
couple | +0.010 |
rob | +0.010 |
gentleman | +0.009 |
seek | +0.009 |
trainer | +0.008 |
Neg logit contributions | |
aying | -0.012 |
fears | -0.011 |
uning | -0.011 |
cleans | -0.011 |
omed | -0.011 |
Token | out |
Token ID | 503 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
aying | +0.004 |
arse | +0.004 |
fears | +0.004 |
uning | +0.004 |
omed | +0.004 |
Neg logit contributions | |
couple | -0.003 |
rob | -0.003 |
seek | -0.003 |
gentleman | -0.003 |
lady | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | +3.31e-01 |
Pos logit contributions | |
aying | +0.013 |
arse | +0.013 |
Ow | +0.012 |
fears | +0.012 |
arding | +0.012 |
Neg logit contributions | |
couple | -0.006 |
lady | -0.004 |
rob | -0.004 |
seek | -0.004 |
l | -0.004 |
Token | magn |
Token ID | 7842 |
Feature activation | -2.55e-01 |
Pos logit contributions | |
couple | +0.013 |
seek | +0.012 |
lady | +0.011 |
man | +0.011 |
rob | +0.011 |
Neg logit contributions | |
aying | -0.019 |
uning | -0.018 |
adoes | -0.017 |
outed | -0.017 |
fears | -0.017 |
Token | ifying |
Token ID | 4035 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
cleans | +0.007 |
fears | +0.007 |
wipes | +0.006 |
ours | +0.006 |
Acc | +0.006 |
Neg logit contributions | |
couple | -0.015 |
rob | -0.014 |
trainer | -0.014 |
amer | -0.013 |
seek | -0.013 |
Token | glass |
Token ID | 5405 |
Feature activation | +2.71e-04 |
Pos logit contributions | |
aying | +0.074 |
fears | +0.067 |
Ow | +0.067 |
uning | +0.065 |
omed | +0.064 |
Neg logit contributions | |
couple | -0.126 |
rob | -0.125 |
seek | -0.113 |
gentleman | -0.113 |
trainer | -0.110 |
Token | and |
Token ID | 290 |
Feature activation | +3.33e-01 |
Pos logit contributions | |
couple | +0.000 |
rob | +0.000 |
seek | +0.000 |
gentleman | +0.000 |
trainer | +0.000 |
Neg logit contributions | |
aying | -0.000 |
fears | -0.000 |
arse | -0.000 |
uning | -0.000 |
omed | -0.000 |
Token | looked |
Token ID | 3114 |
Feature activation | -5.32e-01 |
Pos logit contributions | |
rob | +0.014 |
couple | +0.014 |
gentleman | +0.012 |
seek | +0.012 |
trainer | +0.012 |
Neg logit contributions | |
aying | -0.016 |
fears | -0.015 |
uning | -0.015 |
wipes | -0.015 |
arse | -0.015 |
Token | very |
Token ID | 845 |
Feature activation | +4.71e-02 |
Pos logit contributions | |
aying | +0.030 |
arse | +0.029 |
Ow | +0.028 |
uning | +0.028 |
fears | +0.027 |
Neg logit contributions | |
couple | -0.019 |
rob | -0.016 |
seek | -0.016 |
lady | -0.015 |
gentleman | -0.015 |
Token | carefully |
Token ID | 7773 |
Feature activation | -5.00e-01 |
Pos logit contributions | |
couple | +0.002 |
seek | +0.001 |
man | +0.001 |
lady | +0.001 |
king | +0.001 |
Neg logit contributions | |
aying | -0.003 |
arse | -0.003 |
Ow | -0.003 |
uning | -0.003 |
adoes | -0.003 |
Token | flashlight |
Token ID | 38371 |
Feature activation | -4.96e-01 |
Pos logit contributions | |
aying | +0.014 |
fears | +0.013 |
arse | +0.013 |
uning | +0.013 |
omed | +0.013 |
Neg logit contributions | |
couple | -0.011 |
rob | -0.010 |
seek | -0.010 |
gentleman | -0.009 |
lady | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | -3.77e-02 |
Pos logit contributions | |
aying | +0.029 |
fears | +0.027 |
arse | +0.027 |
Ow | +0.027 |
uning | +0.026 |
Neg logit contributions | |
couple | -0.017 |
rob | -0.016 |
seek | -0.015 |
lady | -0.014 |
gentleman | -0.014 |
Token | and |
Token ID | 290 |
Feature activation | -2.72e-02 |
Pos logit contributions | |
aying | +0.002 |
arse | +0.002 |
fears | +0.002 |
uning | +0.002 |
Ow | +0.002 |
Neg logit contributions | |
couple | -0.001 |
rob | -0.001 |
seek | -0.001 |
lady | -0.001 |
gentleman | -0.001 |
Token | a |
Token ID | 257 |
Feature activation | -4.97e-01 |
Pos logit contributions | |
aying | +0.001 |
fears | +0.001 |
uning | +0.001 |
arse | +0.001 |
wipes | +0.001 |
Neg logit contributions | |
rob | -0.001 |
couple | -0.001 |
seek | -0.001 |
gentleman | -0.001 |
trainer | -0.001 |
Token | magn |
Token ID | 7842 |
Feature activation | -3.47e-01 |
Pos logit contributions | |
aying | +0.026 |
arse | +0.024 |
fears | +0.024 |
uning | +0.024 |
omed | +0.023 |
Neg logit contributions | |
couple | -0.021 |
rob | -0.019 |
seek | -0.018 |
gentleman | -0.017 |
lady | -0.017 |
Token | ifying |
Token ID | 4035 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
fears | +0.010 |
cleans | +0.009 |
wipes | +0.009 |
adoes | +0.009 |
Acc | +0.009 |
Neg logit contributions | |
couple | -0.021 |
rob | -0.020 |
king | -0.018 |
trainer | -0.018 |
gentleman | -0.018 |
Token | glass |
Token ID | 5405 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
aying | +0.083 |
fears | +0.076 |
arse | +0.075 |
uning | +0.074 |
Ow | +0.074 |
Neg logit contributions | |
couple | -0.117 |
rob | -0.114 |
seek | -0.105 |
gentleman | -0.103 |
trainer | -0.102 |
Token | . |
Token ID | 13 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
couple | +0.006 |
rob | +0.006 |
seek | +0.005 |
gentleman | +0.005 |
lady | +0.005 |
Neg logit contributions | |
aying | -0.010 |
fears | -0.009 |
arse | -0.009 |
uning | -0.009 |
Ow | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | +3.87e-01 |
Pos logit contributions | |
couple | +0.009 |
rob | +0.008 |
seek | +0.008 |
lady | +0.008 |
gentleman | +0.008 |
Neg logit contributions | |
aying | -0.011 |
arse | -0.011 |
fears | -0.011 |
uning | -0.010 |
fists | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | +3.12e-01 |
Pos logit contributions | |
couple | +0.014 |
rob | +0.013 |
seek | +0.012 |
gentleman | +0.011 |
trainer | +0.011 |
Neg logit contributions | |
aying | -0.022 |
fears | -0.021 |
arse | -0.020 |
omed | -0.020 |
Ow | -0.020 |
Token | As |
Token ID | 1722 |
Feature activation | +7.58e-01 |
Pos logit contributions | |
couple | +0.011 |
rob | +0.011 |
seek | +0.009 |
gentleman | +0.009 |
trainer | +0.009 |
Neg logit contributions | |
aying | -0.018 |
fears | -0.016 |
arse | -0.016 |
omed | -0.016 |
uning | -0.016 |
Token | When |
Token ID | 1649 |
Feature activation | +3.60e-01 |
Pos logit contributions | |
couple | +0.016 |
rob | +0.014 |
seek | +0.014 |
gentleman | +0.013 |
lady | +0.013 |
Neg logit contributions | |
aying | -0.019 |
arse | -0.018 |
fears | -0.018 |
uning | -0.017 |
omed | -0.017 |
Token | Daisy |
Token ID | 40355 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
couple | +0.011 |
rob | +0.010 |
seek | +0.009 |
lady | +0.009 |
gentleman | +0.009 |
Neg logit contributions | |
aying | -0.022 |
fears | -0.021 |
arse | -0.021 |
uning | -0.021 |
omed | -0.021 |
Token | put |
Token ID | 1234 |
Feature activation | +4.58e-01 |
Pos logit contributions | |
couple | +0.008 |
rob | +0.007 |
seek | +0.006 |
gentleman | +0.006 |
lady | +0.006 |
Neg logit contributions | |
aying | -0.012 |
arse | -0.011 |
fears | -0.011 |
adoes | -0.011 |
uning | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | +3.36e-01 |
Pos logit contributions | |
couple | +0.017 |
rob | +0.016 |
seek | +0.014 |
gentleman | +0.014 |
lady | +0.013 |
Neg logit contributions | |
aying | -0.026 |
fears | -0.024 |
arse | -0.024 |
uning | -0.024 |
omed | -0.023 |
Token | magn |
Token ID | 7842 |
Feature activation | +2.83e-01 |
Pos logit contributions | |
couple | +0.014 |
rob | +0.012 |
lady | +0.012 |
seek | +0.011 |
man | +0.011 |
Neg logit contributions | |
aying | -0.018 |
arse | -0.017 |
fears | -0.017 |
uning | -0.017 |
omed | -0.017 |
Token | ifying |
Token ID | 4035 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
couple | +0.017 |
rob | +0.015 |
seek | +0.015 |
trainer | +0.014 |
lady | +0.014 |
Neg logit contributions | |
fears | -0.009 |
cleans | -0.008 |
wipes | -0.008 |
Acc | -0.008 |
aying | -0.008 |
Token | glass |
Token ID | 5405 |
Feature activation | +2.80e-01 |
Pos logit contributions | |
aying | +0.081 |
Ow | +0.074 |
fears | +0.073 |
uning | +0.071 |
omed | +0.071 |
Neg logit contributions | |
couple | -0.118 |
rob | -0.116 |
seek | -0.106 |
gentleman | -0.105 |
trainer | -0.103 |
Token | down |
Token ID | 866 |
Feature activation | +6.27e-01 |
Pos logit contributions | |
couple | +0.010 |
rob | +0.009 |
seek | +0.009 |
lady | +0.008 |
man | +0.008 |
Neg logit contributions | |
aying | -0.016 |
arse | -0.015 |
fears | -0.015 |
uning | -0.015 |
Ow | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | +6.22e-01 |
Pos logit contributions | |
couple | +0.019 |
seek | +0.017 |
lady | +0.016 |
rob | +0.016 |
man | +0.015 |
Neg logit contributions | |
aying | -0.039 |
arse | -0.038 |
fears | -0.037 |
uning | -0.036 |
iqu | -0.036 |
Token | the |
Token ID | 262 |
Feature activation | +9.07e-02 |
Pos logit contributions | |
couple | +0.014 |
rob | +0.012 |
seek | +0.012 |
lady | +0.010 |
gentleman | +0.010 |
Neg logit contributions | |
aying | -0.044 |
arse | -0.042 |
fears | -0.042 |
uning | -0.041 |
ours | -0.041 |
Token | cater |
Token ID | 20825 |
Feature activation | -3.11e-01 |
Pos logit contributions | |
couple | +0.003 |
seek | +0.003 |
lady | +0.003 |
rob | +0.003 |
man | +0.003 |
Neg logit contributions | |
aying | -0.005 |
arse | -0.005 |
uning | -0.005 |
fears | -0.005 |
omed | -0.005 |
Token | learned |
Token ID | 4499 |
Feature activation | -6.90e-02 |
Pos logit contributions | |
rob | +0.002 |
couple | +0.002 |
gentleman | +0.002 |
trainer | +0.002 |
robat | +0.002 |
Neg logit contributions | |
cleans | -0.003 |
fears | -0.003 |
aying | -0.003 |
wipes | -0.003 |
adoes | -0.003 |
Token | even |
Token ID | 772 |
Feature activation | -4.20e-01 |
Pos logit contributions | |
aying | +0.004 |
arse | +0.004 |
uning | +0.004 |
fears | +0.004 |
omed | +0.004 |
Neg logit contributions | |
couple | -0.002 |
rob | -0.002 |
lady | -0.002 |
seek | -0.002 |
man | -0.002 |
Token | more |
Token ID | 517 |
Feature activation | +8.25e-02 |
Pos logit contributions | |
aying | +0.020 |
fears | +0.020 |
wipes | +0.018 |
uning | +0.018 |
cleans | +0.018 |
Neg logit contributions | |
rob | -0.019 |
couple | -0.018 |
seek | -0.016 |
gentleman | -0.016 |
robat | -0.016 |
Token | about |
Token ID | 546 |
Feature activation | -3.83e-01 |
Pos logit contributions | |
couple | +0.004 |
rob | +0.003 |
seek | +0.003 |
lady | +0.003 |
gentleman | +0.003 |
Neg logit contributions | |
aying | -0.004 |
arse | -0.004 |
Ow | -0.004 |
fears | -0.004 |
uning | -0.004 |
Token | how |
Token ID | 703 |
Feature activation | -9.48e-01 |
Pos logit contributions | |
aying | +0.021 |
fears | +0.020 |
arse | +0.020 |
uning | +0.019 |
omed | +0.019 |
Neg logit contributions | |
couple | -0.014 |
rob | -0.014 |
seek | -0.012 |
gentleman | -0.012 |
trainer | -0.011 |
Token | volcan |
Token ID | 17516 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
aying | +0.056 |
arse | +0.054 |
Ow | +0.052 |
uning | +0.051 |
omed | +0.051 |
Neg logit contributions | |
couple | -0.033 |
seek | -0.027 |
lady | -0.026 |
man | -0.025 |
rob | -0.025 |
Token | oes |
Token ID | 3028 |
Feature activation | -3.57e-01 |
Pos logit contributions | |
wipes | +0.091 |
arse | +0.091 |
fists | +0.088 |
fears | +0.086 |
cleans | +0.085 |
Neg logit contributions | |
rob | -0.092 |
couple | -0.088 |
thank | -0.082 |
king | -0.081 |
trainer | -0.078 |
Token | work |
Token ID | 670 |
Feature activation | +4.31e-01 |
Pos logit contributions | |
arse | +0.019 |
aying | +0.019 |
fears | +0.018 |
omed | +0.018 |
outed | +0.018 |
Neg logit contributions | |
couple | -0.013 |
rob | -0.012 |
seek | -0.012 |
man | -0.011 |
lady | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
couple | +0.012 |
rob | +0.011 |
seek | +0.011 |
lady | +0.010 |
thank | +0.010 |
Neg logit contributions | |
aying | -0.027 |
arse | -0.026 |
Ow | -0.026 |
fears | -0.026 |
uning | -0.025 |
Token | She |
Token ID | 1375 |
Feature activation | +2.04e-02 |
Pos logit contributions | |
couple | +0.002 |
rob | +0.001 |
seek | +0.001 |
gentleman | +0.001 |
trainer | +0.001 |
Neg logit contributions | |
aying | -0.002 |
fears | -0.002 |
arse | -0.002 |
uning | -0.002 |
omed | -0.002 |
Token | thought |
Token ID | 1807 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
couple | +0.001 |
rob | +0.001 |
gentleman | +0.001 |
seek | +0.001 |
robat | +0.001 |
Neg logit contributions | |
aying | -0.001 |
fears | -0.001 |
cleans | -0.001 |
wipes | -0.001 |
uning | -0.001 |
Token | job |
Token ID | 1693 |
Feature activation | +7.73e-01 |
Pos logit contributions | |
aying | +0.004 |
fears | +0.004 |
arse | +0.004 |
uning | +0.004 |
Ow | +0.004 |
Neg logit contributions | |
couple | -0.004 |
rob | -0.004 |
seek | -0.003 |
gentleman | -0.003 |
trainer | -0.003 |
Token | cleaning |
Token ID | 12724 |
Feature activation | +5.48e-01 |
Pos logit contributions | |
couple | +0.027 |
rob | +0.027 |
seek | +0.023 |
gentleman | +0.022 |
trainer | +0.022 |
Neg logit contributions | |
aying | -0.043 |
fears | -0.041 |
arse | -0.040 |
uning | -0.040 |
omed | -0.040 |
Token | the |
Token ID | 262 |
Feature activation | +7.52e-02 |
Pos logit contributions | |
couple | +0.022 |
rob | +0.021 |
seek | +0.019 |
gentleman | +0.018 |
lady | +0.018 |
Neg logit contributions | |
aying | -0.028 |
fears | -0.027 |
arse | -0.027 |
uning | -0.026 |
omed | -0.026 |
Token | garage |
Token ID | 15591 |
Feature activation | +5.26e-01 |
Pos logit contributions | |
couple | +0.004 |
rob | +0.004 |
seek | +0.003 |
gentleman | +0.003 |
trainer | +0.003 |
Neg logit contributions | |
aying | -0.003 |
fears | -0.003 |
arse | -0.003 |
uning | -0.003 |
omed | -0.003 |
Token | ." |
Token ID | 526 |
Feature activation | -5.17e-02 |
Pos logit contributions | |
couple | +0.020 |
rob | +0.020 |
seek | +0.017 |
gentleman | +0.017 |
trainer | +0.017 |
Neg logit contributions | |
aying | -0.028 |
fears | -0.026 |
arse | -0.026 |
uning | -0.025 |
omed | -0.025 |
Token | Little |
Token ID | 7703 |
Feature activation | -2.10e+00 |
Pos logit contributions | |
fears | +0.003 |
aying | +0.003 |
uning | +0.003 |
oken | +0.003 |
arding | +0.003 |
Neg logit contributions | |
couple | -0.002 |
rob | -0.002 |
gentleman | -0.001 |
seek | -0.001 |
lady | -0.001 |
Token | Jimmy |
Token ID | 12963 |
Feature activation | -3.13e-01 |
Pos logit contributions | |
fears | +0.119 |
aying | +0.118 |
uning | +0.118 |
oken | +0.114 |
ither | +0.114 |
Neg logit contributions | |
couple | -0.077 |
man | -0.075 |
rob | -0.074 |
lady | -0.069 |
gentleman | -0.069 |
Token | smiled |
Token ID | 13541 |
Feature activation | +5.14e-01 |
Pos logit contributions | |
aying | +0.017 |
arse | +0.016 |
uning | +0.016 |
fears | +0.016 |
omed | +0.015 |
Neg logit contributions | |
couple | -0.012 |
rob | -0.011 |
seek | -0.010 |
gentleman | -0.010 |
lady | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -8.75e-02 |
Pos logit contributions | |
couple | +0.019 |
rob | +0.018 |
seek | +0.016 |
gentleman | +0.015 |
trainer | +0.015 |
Neg logit contributions | |
aying | -0.029 |
fears | -0.027 |
arse | -0.027 |
uning | -0.026 |
omed | -0.026 |
Token | He |
Token ID | 679 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
aying | +0.005 |
arse | +0.005 |
fears | +0.005 |
uning | +0.005 |
omed | +0.005 |
Neg logit contributions | |
couple | -0.003 |
rob | -0.003 |
gentleman | -0.003 |
seek | -0.003 |
lady | -0.003 |
Token | had |
Token ID | 550 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
aying | +0.007 |
wipes | +0.007 |
fears | +0.007 |
cleans | +0.007 |
omed | +0.007 |
Neg logit contributions | |
rob | -0.008 |
couple | -0.007 |
seek | -0.007 |
gentleman | -0.006 |
trainer | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -5.31e-01 |
Pos logit contributions | |
aying | +0.053 |
fears | +0.050 |
uning | +0.047 |
ours | +0.047 |
arse | +0.047 |
Neg logit contributions | |
rob | -0.044 |
couple | -0.042 |
gentleman | -0.037 |
seek | -0.036 |
robat | -0.035 |
Token | They |
Token ID | 1119 |
Feature activation | +3.23e-01 |
Pos logit contributions | |
arse | +0.022 |
aying | +0.022 |
fears | +0.021 |
iqu | +0.021 |
omed | +0.020 |
Neg logit contributions | |
couple | -0.026 |
rob | -0.024 |
seek | -0.023 |
trainer | -0.023 |
lady | -0.023 |
Token | wear |
Token ID | 5806 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
couple | +0.012 |
rob | +0.011 |
seek | +0.011 |
trainer | +0.011 |
king | +0.010 |
Neg logit contributions | |
aying | -0.017 |
arse | -0.017 |
adoes | -0.017 |
uning | -0.017 |
omed | -0.017 |
Token | masks |
Token ID | 20680 |
Feature activation | -6.72e-01 |
Pos logit contributions | |
aying | +0.011 |
arse | +0.011 |
fears | +0.010 |
uning | +0.010 |
omed | +0.010 |
Neg logit contributions | |
couple | -0.008 |
rob | -0.008 |
seek | -0.007 |
trainer | -0.007 |
par | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
aying | +0.038 |
fears | +0.037 |
arse | +0.037 |
Ow | +0.036 |
uning | +0.036 |
Neg logit contributions | |
couple | -0.024 |
rob | -0.022 |
seek | -0.020 |
lady | -0.019 |
trainer | -0.019 |
Token | fins |
Token ID | 42080 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
couple | +0.005 |
rob | +0.005 |
seek | +0.004 |
gentleman | +0.004 |
trainer | +0.004 |
Neg logit contributions | |
aying | -0.005 |
fears | -0.004 |
arse | -0.004 |
uning | -0.004 |
omed | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | +2.87e-02 |
Pos logit contributions | |
arse | +0.138 |
aying | +0.136 |
fears | +0.132 |
Ow | +0.130 |
uning | +0.130 |
Neg logit contributions | |
couple | -0.054 |
lady | -0.043 |
man | -0.042 |
trainer | -0.042 |
rob | -0.042 |
Token | see |
Token ID | 766 |
Feature activation | -3.30e-01 |
Pos logit contributions | |
couple | +0.001 |
rob | +0.001 |
gentleman | +0.001 |
seek | +0.001 |
thank | +0.001 |
Neg logit contributions | |
aying | -0.001 |
fears | -0.001 |
arse | -0.001 |
wipes | -0.001 |
uning | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | +6.61e-02 |
Pos logit contributions | |
aying | +0.019 |
arse | +0.019 |
omed | +0.018 |
fears | +0.018 |
uning | +0.018 |
Neg logit contributions | |
couple | -0.011 |
rob | -0.010 |
lady | -0.009 |
trainer | -0.009 |
seek | -0.009 |
Token | fish |
Token ID | 5916 |
Feature activation | -8.36e-01 |
Pos logit contributions | |
couple | +0.002 |
rob | +0.002 |
seek | +0.002 |
trainer | +0.002 |
lady | +0.002 |
Neg logit contributions | |
aying | -0.004 |
arse | -0.003 |
fears | -0.003 |
uning | -0.003 |
omed | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
aying | +0.045 |
arse | +0.043 |
fears | +0.043 |
uning | +0.042 |
Ow | +0.042 |
Neg logit contributions | |
couple | -0.031 |
rob | -0.029 |
seek | -0.027 |
trainer | -0.026 |
lady | -0.025 |
Token | said |
Token ID | 531 |
Feature activation | -5.54e-02 |
Pos logit contributions | |
rob | +0.010 |
couple | +0.009 |
robat | +0.008 |
thank | +0.008 |
gentleman | +0.008 |
Neg logit contributions | |
wipes | -0.010 |
outed | -0.010 |
aying | -0.010 |
cleans | -0.010 |
fears | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | +9.06e-01 |
Pos logit contributions | |
aying | +0.003 |
fears | +0.003 |
arse | +0.003 |
Ow | +0.003 |
uning | +0.003 |
Neg logit contributions | |
couple | -0.002 |
rob | -0.001 |
seek | -0.001 |
lady | -0.001 |
gentleman | -0.001 |
Token | holding |
Token ID | 4769 |
Feature activation | -5.97e-02 |
Pos logit contributions | |
couple | +0.034 |
rob | +0.033 |
seek | +0.029 |
gentleman | +0.028 |
trainer | +0.027 |
Neg logit contributions | |
aying | -0.049 |
fears | -0.046 |
arse | -0.046 |
uning | -0.045 |
omed | -0.045 |
Token | the |
Token ID | 262 |
Feature activation | -5.74e-02 |
Pos logit contributions | |
aying | +0.003 |
arse | +0.003 |
fears | +0.003 |
omed | +0.003 |
uning | +0.003 |
Neg logit contributions | |
couple | -0.002 |
rob | -0.002 |
lady | -0.002 |
seek | -0.002 |
man | -0.002 |
Token | magn |
Token ID | 7842 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
aying | +0.003 |
omed | +0.003 |
arse | +0.003 |
fears | +0.003 |
ours | +0.003 |
Neg logit contributions | |
couple | -0.002 |
seek | -0.002 |
rob | -0.002 |
lady | -0.002 |
man | -0.002 |
Token | ifying |
Token ID | 4035 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
couple | +0.015 |
rob | +0.014 |
seek | +0.014 |
trainer | +0.014 |
robat | +0.014 |
Neg logit contributions | |
cleans | -0.006 |
fears | -0.006 |
wipes | -0.006 |
aying | -0.006 |
ours | -0.005 |
Token | glass |
Token ID | 5405 |
Feature activation | +3.52e-01 |
Pos logit contributions | |
aying | +0.060 |
Ow | +0.057 |
fears | +0.054 |
wipes | +0.052 |
uning | +0.052 |
Neg logit contributions | |
rob | -0.129 |
couple | -0.129 |
gentleman | -0.117 |
seek | -0.117 |
trainer | -0.114 |
Token | over |
Token ID | 625 |
Feature activation | +4.40e-01 |
Pos logit contributions | |
couple | +0.012 |
rob | +0.010 |
seek | +0.010 |
lady | +0.010 |
trainer | +0.009 |
Neg logit contributions | |
aying | -0.021 |
arse | -0.020 |
fears | -0.020 |
uning | -0.019 |
Ow | -0.019 |
Token | a |
Token ID | 257 |
Feature activation | +4.31e-01 |
Pos logit contributions | |
couple | +0.016 |
rob | +0.016 |
seek | +0.014 |
gentleman | +0.013 |
trainer | +0.013 |
Neg logit contributions | |
aying | -0.024 |
fears | -0.023 |
arse | -0.023 |
uning | -0.023 |
omed | -0.022 |
Token | black |
Token ID | 2042 |
Feature activation | +3.58e-01 |
Pos logit contributions | |
rob | +0.019 |
couple | +0.018 |
seek | +0.015 |
gentleman | +0.015 |
thank | +0.015 |
Neg logit contributions | |
aying | -0.020 |
fears | -0.020 |
arse | -0.020 |
Ow | -0.019 |
omed | -0.019 |
Token | ant |
Token ID | 1885 |
Feature activation | -4.52e-01 |
Pos logit contributions | |
couple | +0.015 |
rob | +0.014 |
seek | +0.013 |
gentleman | +0.012 |
trainer | +0.012 |
Neg logit contributions | |
aying | -0.018 |
fears | -0.017 |
arse | -0.017 |
uning | -0.017 |
omed | -0.017 |
Token | examine |
Token ID | 10716 |
Feature activation | +3.60e-01 |
Pos logit contributions | |
couple | +0.041 |
rob | +0.039 |
seek | +0.036 |
gentleman | +0.034 |
trainer | +0.034 |
Neg logit contributions | |
aying | -0.045 |
fears | -0.042 |
arse | -0.042 |
uning | -0.041 |
omed | -0.041 |
Token | it |
Token ID | 340 |
Feature activation | -2.25e-01 |
Pos logit contributions | |
couple | +0.013 |
rob | +0.012 |
seek | +0.011 |
lady | +0.010 |
trainer | +0.010 |
Neg logit contributions | |
aying | -0.020 |
arse | -0.020 |
fears | -0.019 |
uning | -0.019 |
omed | -0.019 |
Token | with |
Token ID | 351 |
Feature activation | +5.66e-01 |
Pos logit contributions | |
aying | +0.009 |
arse | +0.008 |
uning | +0.008 |
fears | +0.008 |
Ow | +0.008 |
Neg logit contributions | |
couple | -0.011 |
rob | -0.011 |
gentleman | -0.010 |
seek | -0.010 |
trainer | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | +3.92e-01 |
Pos logit contributions | |
couple | +0.021 |
rob | +0.019 |
seek | +0.018 |
lady | +0.017 |
trainer | +0.017 |
Neg logit contributions | |
aying | -0.032 |
arse | -0.030 |
fears | -0.030 |
uning | -0.029 |
omed | -0.029 |
Token | magn |
Token ID | 7842 |
Feature activation | -6.64e-03 |
Pos logit contributions | |
couple | +0.017 |
rob | +0.016 |
seek | +0.015 |
lady | +0.015 |
trainer | +0.014 |
Neg logit contributions | |
aying | -0.019 |
arse | -0.018 |
fears | -0.018 |
uning | -0.018 |
omed | -0.017 |
Token | ifying |
Token ID | 4035 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
fears | +0.000 |
cleans | +0.000 |
wipes | +0.000 |
aying | +0.000 |
fists | +0.000 |
Neg logit contributions | |
couple | -0.000 |
rob | -0.000 |
robat | -0.000 |
trainer | -0.000 |
seek | -0.000 |
Token | glass |
Token ID | 5405 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
aying | +0.062 |
Ow | +0.055 |
fears | +0.054 |
wipes | +0.053 |
uning | +0.053 |
Neg logit contributions | |
couple | -0.126 |
rob | -0.124 |
seek | -0.114 |
gentleman | -0.114 |
trainer | -0.111 |
Token | ?" |
Token ID | 1701 |
Feature activation | +2.96e-02 |
Pos logit contributions | |
couple | +0.006 |
rob | +0.006 |
seek | +0.006 |
gentleman | +0.006 |
trainer | +0.005 |
Neg logit contributions | |
aying | -0.006 |
fears | -0.005 |
uning | -0.005 |
omed | -0.005 |
arse | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
rob | +0.001 |
couple | +0.001 |
robat | +0.001 |
seek | +0.001 |
thank | +0.001 |
Neg logit contributions | |
aying | -0.001 |
Ow | -0.001 |
wipes | -0.001 |
omed | -0.001 |
fears | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -7.06e-02 |
Pos logit contributions | |
aying | +0.008 |
omed | +0.007 |
fears | +0.007 |
Ow | +0.007 |
arse | +0.007 |
Neg logit contributions | |
couple | -0.005 |
rob | -0.005 |
seek | -0.004 |
gentleman | -0.004 |
trainer | -0.004 |
Token | " |
Token ID | 1 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
aying | +0.004 |
omed | +0.004 |
arse | +0.004 |
fears | +0.004 |
wipes | +0.004 |
Neg logit contributions | |
rob | -0.003 |
couple | -0.003 |
gentleman | -0.002 |
robat | -0.002 |
trainer | -0.002 |
Token | big |
Token ID | 1263 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
couple | +0.002 |
rob | +0.002 |
seek | +0.002 |
lady | +0.002 |
gentleman | +0.002 |
Neg logit contributions | |
aying | -0.003 |
fears | -0.003 |
arse | -0.003 |
uning | -0.003 |
omed | -0.003 |
Token | book |
Token ID | 1492 |
Feature activation | -4.71e-01 |
Pos logit contributions | |
aying | +0.008 |
arse | +0.007 |
fears | +0.007 |
uning | +0.007 |
omed | +0.007 |
Neg logit contributions | |
couple | -0.006 |
rob | -0.006 |
seek | -0.005 |
lady | -0.005 |
gentleman | -0.005 |
Token | with |
Token ID | 351 |
Feature activation | +7.09e-01 |
Pos logit contributions | |
aying | +0.022 |
omed | +0.020 |
uning | +0.019 |
arse | +0.019 |
fears | +0.019 |
Neg logit contributions | |
couple | -0.022 |
rob | -0.021 |
seek | -0.019 |
gentleman | -0.018 |
trainer | -0.018 |
Token | pictures |
Token ID | 5986 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
rob | +0.036 |
couple | +0.034 |
seek | +0.033 |
gentleman | +0.032 |
robat | +0.032 |
Neg logit contributions | |
aying | -0.027 |
fears | -0.027 |
Ow | -0.027 |
fists | -0.026 |
cleans | -0.026 |
Token | of |
Token ID | 286 |
Feature activation | -6.36e-01 |
Pos logit contributions | |
aying | +0.013 |
omed | +0.011 |
uning | +0.011 |
cleans | +0.011 |
ours | +0.011 |
Neg logit contributions | |
couple | -0.010 |
rob | -0.010 |
seek | -0.009 |
gentleman | -0.009 |
trainer | -0.009 |
Token | volcan |
Token ID | 17516 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
aying | +0.032 |
fears | +0.031 |
Ow | +0.030 |
arse | +0.030 |
wipes | +0.029 |
Neg logit contributions | |
couple | -0.025 |
rob | -0.025 |
seek | -0.022 |
gentleman | -0.022 |
thank | -0.021 |
Token | oes |
Token ID | 3028 |
Feature activation | -6.53e-01 |
Pos logit contributions | |
aying | +0.074 |
arse | +0.071 |
fears | +0.071 |
wipes | +0.070 |
fists | +0.067 |
Neg logit contributions | |
rob | -0.107 |
couple | -0.105 |
seek | -0.094 |
trainer | -0.093 |
thank | -0.093 |
Token | . |
Token ID | 13 |
Feature activation | -5.18e-01 |
Pos logit contributions | |
aying | +0.037 |
fears | +0.035 |
arse | +0.035 |
uning | +0.035 |
omed | +0.034 |
Neg logit contributions | |
couple | -0.022 |
rob | -0.022 |
seek | -0.019 |
gentleman | -0.018 |
trainer | -0.018 |
Token | Vol |
Token ID | 4709 |
Feature activation | -5.16e-01 |
Pos logit contributions | |
arse | +0.024 |
aying | +0.024 |
fears | +0.024 |
uning | +0.022 |
iqu | +0.022 |
Neg logit contributions | |
couple | -0.025 |
rob | -0.022 |
lady | -0.022 |
gentleman | -0.021 |
man | -0.021 |
Token | can |
Token ID | 5171 |
Feature activation | -7.65e-01 |
Pos logit contributions | |
fears | +0.027 |
aying | +0.026 |
wipes | +0.025 |
fists | +0.025 |
uning | +0.025 |
Neg logit contributions | |
rob | -0.021 |
couple | -0.019 |
seek | -0.018 |
gentleman | -0.017 |
trainer | -0.017 |
Token | oes |
Token ID | 3028 |
Feature activation | -7.65e-01 |
Pos logit contributions | |
arse | +0.040 |
fears | +0.039 |
aying | +0.039 |
fists | +0.038 |
iqu | +0.037 |
Neg logit contributions | |
couple | -0.027 |
seek | -0.026 |
lady | -0.025 |
rob | -0.025 |
ranger | -0.025 |
Token | a |
Token ID | 257 |
Feature activation | -4.08e-01 |
Pos logit contributions | |
aying | +0.008 |
fears | +0.008 |
arse | +0.008 |
uning | +0.007 |
omed | +0.007 |
Neg logit contributions | |
couple | -0.006 |
rob | -0.005 |
seek | -0.005 |
gentleman | -0.005 |
trainer | -0.004 |
Token | furry |
Token ID | 40857 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
fears | +0.018 |
aying | +0.018 |
arse | +0.017 |
Ow | +0.017 |
wipes | +0.017 |
Neg logit contributions | |
rob | -0.018 |
couple | -0.016 |
robat | -0.015 |
seek | -0.014 |
seek | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
aying | +0.011 |
fears | +0.010 |
uning | +0.010 |
arse | +0.010 |
Ow | +0.010 |
Neg logit contributions | |
rob | -0.007 |
couple | -0.007 |
seek | -0.006 |
gentleman | -0.006 |
thank | -0.006 |
Token | brown |
Token ID | 7586 |
Feature activation | -4.32e-01 |
Pos logit contributions | |
aying | +0.012 |
arse | +0.011 |
uning | +0.011 |
fears | +0.011 |
ours | +0.011 |
Neg logit contributions | |
couple | -0.007 |
seek | -0.007 |
rob | -0.007 |
lady | -0.006 |
man | -0.006 |
Token | bear |
Token ID | 6842 |
Feature activation | -7.11e-01 |
Pos logit contributions | |
aying | +0.024 |
fears | +0.023 |
uning | +0.023 |
arse | +0.023 |
Ow | +0.023 |
Neg logit contributions | |
couple | -0.015 |
rob | -0.015 |
seek | -0.013 |
gentleman | -0.012 |
thank | -0.012 |
Token | cub |
Token ID | 13617 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
aying | +0.030 |
fears | +0.028 |
uning | +0.028 |
ist | +0.027 |
aves | +0.026 |
Neg logit contributions | |
rob | -0.033 |
couple | -0.032 |
seek | -0.029 |
thank | -0.028 |
gentleman | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -3.01e-01 |
Pos logit contributions | |
aying | +0.112 |
arse | +0.106 |
fears | +0.105 |
uning | +0.103 |
omed | +0.102 |
Neg logit contributions | |
couple | -0.076 |
rob | -0.072 |
seek | -0.064 |
trainer | -0.062 |
gentleman | -0.062 |
Token |
|
Token ID | 198 |
Feature activation | +7.54e-02 |
Pos logit contributions | |
aying | +0.018 |
fears | +0.017 |
arse | +0.017 |
uning | +0.016 |
omed | +0.016 |
Neg logit contributions | |
couple | -0.010 |
rob | -0.010 |
seek | -0.009 |
lady | -0.008 |
gentleman | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | +6.22e-02 |
Pos logit contributions | |
couple | +0.003 |
rob | +0.003 |
seek | +0.003 |
thank | +0.003 |
gentleman | +0.003 |
Neg logit contributions | |
aying | -0.004 |
Ow | -0.003 |
omed | -0.003 |
fears | -0.003 |
arse | -0.003 |
Token | One |
Token ID | 3198 |
Feature activation | +6.15e-01 |
Pos logit contributions | |
couple | +0.002 |
rob | +0.002 |
seek | +0.002 |
lady | +0.002 |
gentleman | +0.002 |
Neg logit contributions | |
aying | -0.003 |
fears | -0.003 |
uning | -0.003 |
arse | -0.003 |
fists | -0.003 |
Token | day |
Token ID | 1110 |
Feature activation | -3.02e-01 |
Pos logit contributions | |
rob | +0.018 |
couple | +0.016 |
gentleman | +0.013 |
thank | +0.013 |
seek | +0.013 |
Neg logit contributions | |
aying | -0.039 |
fears | -0.037 |
Ow | -0.036 |
uning | -0.035 |
omed | -0.035 |
Token | fox |
Token ID | 21831 |
Feature activation | -6.32e-01 |
Pos logit contributions | |
aying | +0.044 |
arse | +0.042 |
fears | +0.041 |
uning | +0.041 |
omed | +0.040 |
Neg logit contributions | |
couple | -0.021 |
rob | -0.018 |
lady | -0.017 |
seek | -0.016 |
man | -0.016 |
Token | living |
Token ID | 2877 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
aying | +0.032 |
fears | +0.031 |
uning | +0.030 |
Ow | +0.029 |
omed | +0.029 |
Neg logit contributions | |
couple | -0.025 |
rob | -0.024 |
seek | -0.021 |
gentleman | -0.021 |
thank | -0.020 |
Token | in |
Token ID | 287 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
fears | +0.042 |
aying | +0.042 |
Ow | +0.041 |
adoes | +0.040 |
fists | +0.039 |
Neg logit contributions | |
rob | -0.050 |
couple | -0.050 |
gentleman | -0.045 |
trainer | -0.043 |
seek | -0.042 |
Token | a |
Token ID | 257 |
Feature activation | +6.92e-01 |
Pos logit contributions | |
couple | +0.009 |
rob | +0.009 |
seek | +0.008 |
gentleman | +0.008 |
thank | +0.008 |
Neg logit contributions | |
aying | -0.011 |
fears | -0.011 |
Ow | -0.010 |
arse | -0.010 |
fists | -0.010 |
Token | big |
Token ID | 1263 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
rob | +0.035 |
couple | +0.033 |
thank | +0.031 |
gentleman | +0.030 |
robat | +0.029 |
Neg logit contributions | |
fears | -0.028 |
aying | -0.027 |
arse | -0.027 |
Ow | -0.026 |
wipes | -0.026 |
Token | chim |
Token ID | 18205 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
rob | +0.009 |
couple | +0.009 |
thank | +0.008 |
seek | +0.007 |
gentleman | +0.007 |
Neg logit contributions | |
fears | -0.013 |
aying | -0.013 |
arse | -0.012 |
wipes | -0.012 |
Ow | -0.012 |
Token | ney |
Token ID | 1681 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
aying | +0.071 |
outed | +0.065 |
icker | +0.064 |
fears | +0.064 |
uning | +0.063 |
Neg logit contributions | |
couple | -0.114 |
rob | -0.109 |
gentleman | -0.102 |
seek | -0.099 |
trainer | -0.098 |
Token | . |
Token ID | 13 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
aying | +0.080 |
arse | +0.076 |
fears | +0.076 |
uning | +0.075 |
Ow | +0.074 |
Neg logit contributions | |
couple | -0.043 |
rob | -0.041 |
seek | -0.036 |
gentleman | -0.034 |
lady | -0.033 |
Token | Every |
Token ID | 3887 |
Feature activation | +3.25e-01 |
Pos logit contributions | |
arse | +0.012 |
aying | +0.012 |
fears | +0.011 |
iqu | +0.011 |
arding | +0.011 |
Neg logit contributions | |
couple | -0.009 |
rob | -0.008 |
seek | -0.008 |
lady | -0.008 |
gentleman | -0.008 |
Token | day |
Token ID | 1110 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
couple | +0.007 |
rob | +0.006 |
seek | +0.005 |
gentleman | +0.004 |
trainer | +0.004 |
Neg logit contributions | |
aying | -0.023 |
fears | -0.022 |
arse | -0.022 |
uning | -0.022 |
omed | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | +9.86e-02 |
Pos logit contributions | |
aying | +0.014 |
arse | +0.014 |
fears | +0.014 |
uning | +0.013 |
iqu | +0.013 |
Neg logit contributions | |
couple | -0.007 |
rob | -0.006 |
seek | -0.006 |
lady | -0.006 |
man | -0.006 |
Token | big |
Token ID | 1263 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
couple | +0.003 |
rob | +0.003 |
seek | +0.003 |
lady | +0.002 |
gentleman | +0.002 |
Neg logit contributions | |
aying | -0.004 |
fears | -0.004 |
arse | -0.004 |
uning | -0.004 |
omed | -0.004 |
Token | book |
Token ID | 1492 |
Feature activation | -4.35e-01 |
Pos logit contributions | |
aying | +0.006 |
arse | +0.006 |
fears | +0.006 |
uning | +0.006 |
omed | +0.006 |
Neg logit contributions | |
couple | -0.005 |
rob | -0.005 |
seek | -0.005 |
lady | -0.004 |
gentleman | -0.004 |
Token | with |
Token ID | 351 |
Feature activation | +6.85e-01 |
Pos logit contributions | |
aying | +0.020 |
omed | +0.019 |
uning | +0.018 |
arse | +0.018 |
ours | +0.018 |
Neg logit contributions | |
couple | -0.020 |
rob | -0.020 |
seek | -0.017 |
gentleman | -0.017 |
trainer | -0.016 |
Token | pictures |
Token ID | 5986 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
rob | +0.035 |
couple | +0.034 |
seek | +0.032 |
gentleman | +0.031 |
robat | +0.031 |
Neg logit contributions | |
aying | -0.026 |
fears | -0.026 |
Ow | -0.026 |
fists | -0.025 |
cleans | -0.025 |
Token | of |
Token ID | 286 |
Feature activation | -6.53e-01 |
Pos logit contributions | |
aying | +0.012 |
omed | +0.011 |
uning | +0.010 |
cleans | +0.010 |
ours | +0.010 |
Neg logit contributions | |
couple | -0.010 |
rob | -0.010 |
seek | -0.009 |
gentleman | -0.008 |
trainer | -0.008 |
Token | volcan |
Token ID | 17516 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
aying | +0.033 |
fears | +0.031 |
Ow | +0.030 |
wipes | +0.030 |
arse | +0.030 |
Neg logit contributions | |
couple | -0.026 |
rob | -0.026 |
seek | -0.023 |
gentleman | -0.023 |
thank | -0.022 |
Token | oes |
Token ID | 3028 |
Feature activation | -6.09e-01 |
Pos logit contributions | |
aying | +0.075 |
arse | +0.072 |
fears | +0.072 |
wipes | +0.071 |
fists | +0.069 |
Neg logit contributions | |
rob | -0.105 |
couple | -0.103 |
seek | -0.092 |
trainer | -0.092 |
thank | -0.091 |
Token | . |
Token ID | 13 |
Feature activation | -5.16e-01 |
Pos logit contributions | |
aying | +0.034 |
fears | +0.032 |
arse | +0.032 |
uning | +0.032 |
omed | +0.031 |
Neg logit contributions | |
couple | -0.022 |
rob | -0.021 |
seek | -0.018 |
gentleman | -0.017 |
trainer | -0.017 |
Token | Vol |
Token ID | 4709 |
Feature activation | -4.82e-01 |
Pos logit contributions | |
arse | +0.023 |
aying | +0.023 |
fears | +0.023 |
uning | +0.021 |
iqu | +0.021 |
Neg logit contributions | |
couple | -0.025 |
rob | -0.022 |
lady | -0.022 |
gentleman | -0.021 |
man | -0.021 |
Token | can |
Token ID | 5171 |
Feature activation | -7.48e-01 |
Pos logit contributions | |
fears | +0.025 |
aying | +0.025 |
wipes | +0.024 |
fists | +0.023 |
arse | +0.023 |
Neg logit contributions | |
rob | -0.019 |
couple | -0.018 |
seek | -0.016 |
gentleman | -0.016 |
trainer | -0.015 |
Token | oes |
Token ID | 3028 |
Feature activation | -7.01e-01 |
Pos logit contributions | |
arse | +0.037 |
fears | +0.037 |
aying | +0.036 |
fists | +0.035 |
wipes | +0.035 |
Neg logit contributions | |
couple | -0.028 |
rob | -0.027 |
seek | -0.027 |
lady | -0.026 |
ranger | -0.025 |
Token | bear |
Token ID | 6842 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
aying | +0.030 |
cleans | +0.028 |
ist | +0.028 |
uning | +0.028 |
icking | +0.027 |
Neg logit contributions | |
rob | -0.041 |
couple | -0.039 |
seek | -0.039 |
gentleman | -0.036 |
robat | -0.036 |
Token | who |
Token ID | 508 |
Feature activation | -3.34e-01 |
Pos logit contributions | |
aying | +0.052 |
uning | +0.050 |
fears | +0.048 |
cleans | +0.048 |
ours | +0.048 |
Neg logit contributions | |
rob | -0.039 |
couple | -0.038 |
seek | -0.034 |
gentleman | -0.032 |
robat | -0.032 |
Token | had |
Token ID | 550 |
Feature activation | -5.66e-01 |
Pos logit contributions | |
aying | +0.016 |
fears | +0.015 |
arse | +0.015 |
uning | +0.014 |
Ow | +0.014 |
Neg logit contributions | |
couple | -0.015 |
rob | -0.014 |
seek | -0.013 |
gentleman | -0.013 |
trainer | -0.012 |
Token | a |
Token ID | 257 |
Feature activation | -2.47e-02 |
Pos logit contributions | |
aying | +0.028 |
fears | +0.027 |
arse | +0.026 |
uning | +0.026 |
Ow | +0.026 |
Neg logit contributions | |
couple | -0.023 |
rob | -0.023 |
seek | -0.020 |
gentleman | -0.020 |
trainer | -0.019 |
Token | little |
Token ID | 1310 |
Feature activation | -7.66e-01 |
Pos logit contributions | |
aying | +0.001 |
omed | +0.001 |
arse | +0.001 |
uning | +0.001 |
adoes | +0.001 |
Neg logit contributions | |
couple | -0.001 |
seek | -0.001 |
lady | -0.001 |
rob | -0.001 |
man | -0.001 |
Token | cub |
Token ID | 13617 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
aying | +0.048 |
arse | +0.045 |
omed | +0.045 |
uning | +0.044 |
Ow | +0.043 |
Neg logit contributions | |
couple | -0.025 |
rob | -0.021 |
lady | -0.020 |
seek | -0.020 |
gentleman | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -5.14e-01 |
Pos logit contributions | |
aying | +0.114 |
arse | +0.110 |
fears | +0.107 |
uning | +0.105 |
Ow | +0.105 |
Neg logit contributions | |
couple | -0.074 |
rob | -0.068 |
seek | -0.060 |
trainer | -0.059 |
gentleman | -0.059 |
Token | One |
Token ID | 1881 |
Feature activation | +3.62e-01 |
Pos logit contributions | |
aying | +0.026 |
fears | +0.026 |
arse | +0.025 |
iqu | +0.024 |
fists | +0.024 |
Neg logit contributions | |
couple | -0.023 |
rob | -0.020 |
lady | -0.020 |
seek | -0.019 |
gentleman | -0.019 |
Token | day |
Token ID | 1110 |
Feature activation | -3.03e-01 |
Pos logit contributions | |
rob | +0.010 |
couple | +0.009 |
seek | +0.007 |
gentleman | +0.007 |
trainer | +0.007 |
Neg logit contributions | |
aying | -0.024 |
fears | -0.023 |
Ow | -0.022 |
arse | -0.022 |
uning | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
aying | +0.021 |
fears | +0.020 |
arse | +0.019 |
iqu | +0.019 |
oken | +0.019 |
Neg logit contributions | |
couple | -0.008 |
rob | -0.006 |
seek | -0.006 |
lady | -0.006 |
man | -0.006 |
Token | little |
Token ID | 1310 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
aying | +0.009 |
fears | +0.008 |
arse | +0.008 |
uning | +0.008 |
omed | +0.008 |
Neg logit contributions | |
couple | -0.004 |
rob | -0.003 |
seek | -0.003 |
lady | -0.003 |
man | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
rob | +0.000 |
couple | +0.000 |
gentleman | +0.000 |
trainer | +0.000 |
robat | +0.000 |
Neg logit contributions | |
aying | -0.000 |
Ow | -0.000 |
fears | -0.000 |
cleans | -0.000 |
wipes | -0.000 |
Token | house |
Token ID | 2156 |
Feature activation | -8.36e-01 |
Pos logit contributions | |
rob | +0.009 |
couple | +0.008 |
thank | +0.008 |
gentleman | +0.008 |
robat | +0.008 |
Neg logit contributions | |
fears | -0.006 |
aying | -0.006 |
Ow | -0.006 |
wipes | -0.006 |
arse | -0.006 |
Token | with |
Token ID | 351 |
Feature activation | +6.49e-01 |
Pos logit contributions | |
aying | +0.041 |
ours | +0.037 |
uning | +0.037 |
fears | +0.037 |
wipes | +0.036 |
Neg logit contributions | |
rob | -0.036 |
couple | -0.034 |
robat | -0.030 |
gentleman | -0.030 |
trainer | -0.029 |
Token | a |
Token ID | 257 |
Feature activation | -2.90e-01 |
Pos logit contributions | |
rob | +0.027 |
couple | +0.025 |
gentleman | +0.023 |
robat | +0.022 |
trainer | +0.021 |
Neg logit contributions | |
aying | -0.033 |
Ow | -0.032 |
fears | -0.031 |
cleans | -0.031 |
wipes | -0.031 |
Token | big |
Token ID | 1263 |
Feature activation | -3.78e-01 |
Pos logit contributions | |
aying | +0.016 |
arse | +0.015 |
uning | +0.015 |
fears | +0.015 |
omed | +0.015 |
Neg logit contributions | |
couple | -0.012 |
rob | -0.010 |
lady | -0.010 |
seek | -0.010 |
man | -0.010 |
Token | chim |
Token ID | 18205 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
aying | +0.022 |
arse | +0.020 |
omed | +0.020 |
fears | +0.020 |
uning | +0.020 |
Neg logit contributions | |
couple | -0.014 |
rob | -0.013 |
lady | -0.012 |
seek | -0.012 |
man | -0.012 |
Token | ney |
Token ID | 1681 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
aying | +0.068 |
icker | +0.063 |
uning | +0.061 |
outed | +0.060 |
akes | +0.060 |
Neg logit contributions | |
couple | -0.115 |
rob | -0.113 |
gentleman | -0.103 |
seek | -0.101 |
trainer | -0.101 |
Token | . |
Token ID | 13 |
Feature activation | -4.46e-01 |
Pos logit contributions | |
aying | +0.086 |
fears | +0.082 |
arse | +0.081 |
uning | +0.080 |
omed | +0.079 |
Neg logit contributions | |
couple | -0.052 |
rob | -0.050 |
seek | -0.044 |
gentleman | -0.042 |
trainer | -0.041 |
Token | The |
Token ID | 383 |
Feature activation | -5.59e-01 |
Pos logit contributions | |
aying | +0.022 |
arse | +0.021 |
fears | +0.021 |
uning | +0.020 |
fists | +0.020 |
Neg logit contributions | |
couple | -0.019 |
rob | -0.018 |
seek | -0.017 |
lady | -0.017 |
gentleman | -0.016 |
Token | chim |
Token ID | 18205 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
aying | +0.026 |
arse | +0.025 |
fears | +0.025 |
omed | +0.024 |
uning | +0.024 |
Neg logit contributions | |
couple | -0.025 |
rob | -0.024 |
lady | -0.022 |
seek | -0.022 |
man | -0.021 |
Token | ney |
Token ID | 1681 |
Feature activation | -9.51e-01 |
Pos logit contributions | |
aying | +0.063 |
fears | +0.056 |
uning | +0.056 |
omed | +0.055 |
Ow | +0.055 |
Neg logit contributions | |
couple | -0.074 |
rob | -0.071 |
seek | -0.065 |
gentleman | -0.064 |
trainer | -0.063 |
Token | . |
Token ID | 13 |
Feature activation | -5.44e-01 |
Pos logit contributions | |
aying | +0.025 |
uning | +0.023 |
fears | +0.023 |
omed | +0.022 |
arse | +0.022 |
Neg logit contributions | |
couple | -0.019 |
rob | -0.018 |
seek | -0.016 |
gentleman | -0.016 |
thank | -0.015 |
Token | They |
Token ID | 1119 |
Feature activation | +2.36e-01 |
Pos logit contributions | |
arse | +0.032 |
aying | +0.031 |
fears | +0.031 |
fists | +0.029 |
uning | +0.029 |
Neg logit contributions | |
couple | -0.017 |
rob | -0.015 |
ranger | -0.015 |
lady | -0.015 |
seek | -0.015 |
Token | had |
Token ID | 550 |
Feature activation | -6.11e-03 |
Pos logit contributions | |
couple | +0.009 |
rob | +0.009 |
seek | +0.008 |
gentleman | +0.008 |
trainer | +0.008 |
Neg logit contributions | |
aying | -0.013 |
fears | -0.012 |
arse | -0.012 |
uning | -0.011 |
omed | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
aying | +0.000 |
arse | +0.000 |
fears | +0.000 |
uning | +0.000 |
omed | +0.000 |
Neg logit contributions | |
couple | -0.000 |
rob | -0.000 |
seek | -0.000 |
trainer | -0.000 |
lady | -0.000 |
Token | magn |
Token ID | 7842 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
couple | +0.008 |
rob | +0.008 |
seek | +0.007 |
lady | +0.007 |
trainer | +0.007 |
Neg logit contributions | |
aying | -0.009 |
uning | -0.008 |
arse | -0.008 |
fears | -0.008 |
omed | -0.008 |
Token | ifying |
Token ID | 4035 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
cleans | +0.004 |
fears | +0.004 |
wipes | +0.004 |
Acc | +0.004 |
adoes | +0.004 |
Neg logit contributions | |
couple | -0.009 |
rob | -0.009 |
trainer | -0.008 |
robat | -0.008 |
king | -0.008 |
Token | glass |
Token ID | 5405 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
aying | +0.071 |
fears | +0.064 |
Ow | +0.064 |
uning | +0.061 |
omed | +0.061 |
Neg logit contributions | |
couple | -0.114 |
rob | -0.112 |
gentleman | -0.102 |
seek | -0.101 |
trainer | -0.099 |
Token | and |
Token ID | 290 |
Feature activation | +2.83e-01 |
Pos logit contributions | |
couple | +0.007 |
rob | +0.007 |
seek | +0.006 |
gentleman | +0.006 |
thank | +0.006 |
Neg logit contributions | |
aying | -0.009 |
fears | -0.009 |
uning | -0.009 |
arse | -0.009 |
omed | -0.009 |
Token | a |
Token ID | 257 |
Feature activation | -5.81e-02 |
Pos logit contributions | |
couple | +0.011 |
rob | +0.011 |
seek | +0.010 |
gentleman | +0.009 |
lady | +0.009 |
Neg logit contributions | |
aying | -0.015 |
arse | -0.014 |
fears | -0.014 |
uning | -0.014 |
omed | -0.014 |
Token | small |
Token ID | 1402 |
Feature activation | -3.55e-02 |
Pos logit contributions | |
aying | +0.003 |
arse | +0.003 |
fears | +0.003 |
uning | +0.003 |
omed | +0.003 |
Neg logit contributions | |
couple | -0.002 |
rob | -0.002 |
seek | -0.002 |
gentleman | -0.002 |
lady | -0.002 |
Token | box |
Token ID | 3091 |
Feature activation | -7.82e-01 |
Pos logit contributions | |
aying | +0.002 |
arse | +0.002 |
fears | +0.002 |
uning | +0.002 |
omed | +0.002 |
Neg logit contributions | |
couple | -0.001 |
rob | -0.001 |
seek | -0.001 |
lady | -0.001 |
gentleman | -0.001 |
Token | He |
Token ID | 679 |
Feature activation | +2.20e-01 |
Pos logit contributions | |
couple | +0.005 |
gentleman | +0.004 |
seek | +0.004 |
rob | +0.004 |
ranger | +0.004 |
Neg logit contributions | |
arse | -0.008 |
aying | -0.008 |
fears | -0.008 |
arding | -0.008 |
cleans | -0.008 |
Token | found |
Token ID | 1043 |
Feature activation | -6.84e-02 |
Pos logit contributions | |
couple | +0.008 |
rob | +0.008 |
seek | +0.007 |
gentleman | +0.007 |
lady | +0.006 |
Neg logit contributions | |
aying | -0.012 |
arse | -0.012 |
fears | -0.011 |
uning | -0.011 |
Ow | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | +6.40e-01 |
Pos logit contributions | |
aying | +0.004 |
arse | +0.003 |
fears | +0.003 |
uning | +0.003 |
omed | +0.003 |
Neg logit contributions | |
couple | -0.003 |
rob | -0.003 |
seek | -0.002 |
gentleman | -0.002 |
trainer | -0.002 |
Token | big |
Token ID | 1263 |
Feature activation | +3.82e-01 |
Pos logit contributions | |
couple | +0.025 |
rob | +0.024 |
seek | +0.022 |
gentleman | +0.021 |
lady | +0.021 |
Neg logit contributions | |
aying | -0.034 |
fears | -0.032 |
arse | -0.032 |
uning | -0.031 |
omed | -0.031 |
Token | tree |
Token ID | 5509 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
couple | +0.014 |
rob | +0.013 |
seek | +0.012 |
lady | +0.012 |
gentleman | +0.012 |
Neg logit contributions | |
aying | -0.022 |
arse | -0.020 |
fears | -0.020 |
uning | -0.020 |
Ow | -0.020 |
Token | with |
Token ID | 351 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
aying | +0.016 |
fears | +0.015 |
arse | +0.015 |
uning | +0.015 |
omed | +0.015 |
Neg logit contributions | |
couple | -0.012 |
rob | -0.011 |
seek | -0.010 |
gentleman | -0.009 |
trainer | -0.009 |
Token | an |
Token ID | 281 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
couple | +0.052 |
rob | +0.049 |
seek | +0.044 |
gentleman | +0.043 |
lady | +0.042 |
Neg logit contributions | |
aying | -0.064 |
arse | -0.060 |
fears | -0.060 |
uning | -0.059 |
omed | -0.058 |
Token | old |
Token ID | 1468 |
Feature activation | +7.19e-01 |
Pos logit contributions | |
rob | +0.010 |
couple | +0.009 |
seek | +0.008 |
trainer | +0.008 |
gentleman | +0.008 |
Neg logit contributions | |
aying | -0.011 |
fears | -0.010 |
uning | -0.010 |
arse | -0.009 |
fists | -0.009 |
Token | branch |
Token ID | 8478 |
Feature activation | +3.14e-02 |
Pos logit contributions | |
couple | +0.025 |
rob | +0.024 |
seek | +0.022 |
gentleman | +0.021 |
lady | +0.020 |
Neg logit contributions | |
aying | -0.041 |
fears | -0.038 |
arse | -0.038 |
uning | -0.038 |
omed | -0.037 |
Token | reaching |
Token ID | 8978 |
Feature activation | +3.11e-01 |
Pos logit contributions | |
couple | +0.001 |
rob | +0.001 |
seek | +0.001 |
gentleman | +0.001 |
lady | +0.001 |
Neg logit contributions | |
aying | -0.002 |
fears | -0.002 |
arse | -0.002 |
Ow | -0.002 |
uning | -0.002 |
Token | down |
Token ID | 866 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
couple | +0.009 |
rob | +0.008 |
man | +0.008 |
seek | +0.008 |
lady | +0.008 |
Neg logit contributions | |
aying | -0.018 |
arse | -0.017 |
fears | -0.017 |
wipes | -0.017 |
cleans | -0.017 |
Token | â |
Token ID | 22940 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
aying | +0.005 |
fears | +0.005 |
arse | +0.005 |
uning | +0.005 |
fists | +0.005 |
Neg logit contributions | |
couple | -0.004 |
rob | -0.004 |
seek | -0.004 |
gentleman | -0.004 |
lady | -0.004 |
Token | € |
Token ID | 26391 |
Feature activation | +7.57e-01 |
Pos logit contributions | |
aying | +0.012 |
fears | +0.012 |
arse | +0.012 |
uning | +0.012 |
Ow | +0.012 |
Neg logit contributions | |
rob | -0.009 |
couple | -0.009 |
seek | -0.008 |
gentleman | -0.007 |
thank | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
couple | +0.029 |
rob | +0.028 |
seek | +0.026 |
gentleman | +0.024 |
lady | +0.024 |
Neg logit contributions | |
aying | -0.041 |
arse | -0.039 |
fears | -0.039 |
uning | -0.038 |
fists | -0.037 |
Token |
|
Token ID | 198 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
couple | +0.004 |
rob | +0.004 |
seek | +0.003 |
gentleman | +0.003 |
lady | +0.003 |
Neg logit contributions | |
aying | -0.007 |
fears | -0.006 |
arse | -0.006 |
uning | -0.006 |
adoes | -0.006 |
Token | The |
Token ID | 464 |
Feature activation | +1.67e-02 |
Pos logit contributions | |
couple | +0.009 |
rob | +0.009 |
seek | +0.008 |
gentleman | +0.008 |
trainer | +0.008 |
Neg logit contributions | |
aying | -0.013 |
fears | -0.012 |
arse | -0.012 |
uning | -0.012 |
omed | -0.012 |
Token | old |
Token ID | 1468 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
rob | +0.000 |
couple | +0.000 |
seek | +0.000 |
robat | +0.000 |
gentleman | +0.000 |
Neg logit contributions | |
aying | -0.001 |
fears | -0.001 |
wipes | -0.001 |
uning | -0.001 |
cleans | -0.001 |
Token | lady |
Token ID | 10846 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
rob | +0.034 |
couple | +0.023 |
robat | +0.022 |
amer | +0.020 |
cest | +0.019 |
Neg logit contributions | |
wipes | -0.088 |
cleans | -0.086 |
aying | -0.084 |
ours | -0.081 |
uning | -0.081 |
Token | was |
Token ID | 373 |
Feature activation | -2.49e-02 |
Pos logit contributions | |
couple | +0.039 |
rob | +0.034 |
seek | +0.032 |
lady | +0.032 |
trainer | +0.031 |
Neg logit contributions | |
aying | -0.072 |
arse | -0.070 |
Ow | -0.068 |
fears | -0.068 |
uning | -0.068 |
Token | astonished |
Token ID | 40962 |
Feature activation | +2.66e-01 |
Pos logit contributions | |
aying | +0.001 |
fears | +0.001 |
arse | +0.001 |
omed | +0.001 |
fists | +0.001 |
Neg logit contributions | |
rob | -0.001 |
couple | -0.001 |
seek | -0.001 |
robat | -0.001 |
gentleman | -0.001 |
Token | at |
Token ID | 379 |
Feature activation | -2.25e-01 |
Pos logit contributions | |
couple | +0.009 |
rob | +0.007 |
seek | +0.007 |
lady | +0.007 |
gentleman | +0.007 |
Neg logit contributions | |
aying | -0.016 |
arse | -0.015 |
fears | -0.015 |
Ow | -0.015 |
uning | -0.015 |
Token | Daisy |
Token ID | 40355 |
Feature activation | -5.88e-02 |
Pos logit contributions | |
aying | +0.012 |
arse | +0.011 |
omed | +0.011 |
fears | +0.011 |
uning | +0.010 |
Neg logit contributions | |
couple | -0.009 |
rob | -0.008 |
lady | -0.008 |
seek | -0.008 |
gentleman | -0.008 |
Token | mom |
Token ID | 1995 |
Feature activation | +5.69e-01 |
Pos logit contributions | |
couple | +0.009 |
rob | +0.009 |
seek | +0.008 |
gentleman | +0.008 |
lady | +0.008 |
Neg logit contributions | |
aying | -0.011 |
arse | -0.010 |
fears | -0.010 |
uning | -0.010 |
omed | -0.010 |
Token | holding |
Token ID | 4769 |
Feature activation | -9.41e-02 |
Pos logit contributions | |
couple | +0.019 |
rob | +0.017 |
seek | +0.016 |
lady | +0.015 |
gentleman | +0.015 |
Neg logit contributions | |
aying | -0.034 |
fears | -0.032 |
Ow | -0.032 |
uning | -0.031 |
arse | -0.031 |
Token | a |
Token ID | 257 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
aying | +0.006 |
arse | +0.006 |
uning | +0.005 |
fears | +0.005 |
omed | +0.005 |
Neg logit contributions | |
couple | -0.003 |
rob | -0.003 |
lady | -0.002 |
seek | -0.002 |
gentleman | -0.002 |
Token | big |
Token ID | 1263 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
couple | +0.009 |
rob | +0.008 |
lady | +0.008 |
seek | +0.008 |
man | +0.008 |
Neg logit contributions | |
aying | -0.014 |
uning | -0.013 |
arse | -0.013 |
fears | -0.013 |
Ow | -0.012 |
Token | file |
Token ID | 2393 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
couple | +0.005 |
rob | +0.004 |
seek | +0.004 |
lady | +0.004 |
gentleman | +0.004 |
Neg logit contributions | |
aying | -0.007 |
arse | -0.007 |
fears | -0.007 |
uning | -0.007 |
Ow | -0.007 |
Token | with |
Token ID | 351 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
rob | +0.007 |
couple | +0.006 |
seek | +0.006 |
gentleman | +0.005 |
trainer | +0.005 |
Neg logit contributions | |
aying | -0.008 |
arse | -0.008 |
fears | -0.008 |
omed | -0.008 |
uning | -0.008 |
Token | many |
Token ID | 867 |
Feature activation | +6.10e-01 |
Pos logit contributions | |
couple | +0.047 |
rob | +0.043 |
seek | +0.040 |
lady | +0.038 |
gentleman | +0.038 |
Neg logit contributions | |
aying | -0.065 |
arse | -0.061 |
fears | -0.060 |
uning | -0.059 |
omed | -0.058 |
Token | papers |
Token ID | 9473 |
Feature activation | +7.09e-01 |
Pos logit contributions | |
couple | +0.026 |
rob | +0.024 |
seek | +0.022 |
lady | +0.022 |
gentleman | +0.022 |
Neg logit contributions | |
aying | -0.031 |
arse | -0.029 |
uning | -0.028 |
Ow | -0.028 |
fears | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -9.86e-02 |
Pos logit contributions | |
rob | +0.030 |
couple | +0.028 |
seek | +0.025 |
gentleman | +0.025 |
trainer | +0.024 |
Neg logit contributions | |
aying | -0.036 |
omed | -0.033 |
uning | -0.032 |
arse | -0.032 |
fears | -0.032 |
Token |
|
Token ID | 198 |
Feature activation | +2.78e-01 |
Pos logit contributions | |
aying | +0.006 |
arse | +0.006 |
fears | +0.006 |
uning | +0.005 |
ist | +0.005 |
Neg logit contributions | |
couple | -0.003 |
lady | -0.003 |
rob | -0.003 |
gentleman | -0.003 |
man | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
couple | +0.010 |
rob | +0.010 |
seek | +0.008 |
gentleman | +0.008 |
trainer | +0.008 |
Neg logit contributions | |
aying | -0.016 |
arse | -0.015 |
fears | -0.014 |
omed | -0.014 |
Ow | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | +2.35e-01 |
Pos logit contributions | |
rob | +0.008 |
couple | +0.008 |
seek | +0.007 |
gentleman | +0.007 |
trainer | +0.007 |
Neg logit contributions | |
aying | -0.011 |
fears | -0.011 |
arse | -0.011 |
omed | -0.011 |
uning | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | -4.33e-02 |
Pos logit contributions | |
couple | +0.010 |
rob | +0.009 |
seek | +0.008 |
gentleman | +0.008 |
trainer | +0.008 |
Neg logit contributions | |
aying | -0.012 |
fears | -0.011 |
arse | -0.011 |
uning | -0.011 |
omed | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
aying | +0.002 |
fears | +0.002 |
arse | +0.002 |
Ow | +0.002 |
omed | +0.002 |
Neg logit contributions | |
couple | -0.002 |
rob | -0.002 |
seek | -0.002 |
gentleman | -0.002 |
trainer | -0.002 |
Token | After |
Token ID | 3260 |
Feature activation | +9.26e-01 |
Pos logit contributions | |
rob | +0.001 |
couple | +0.001 |
gentleman | +0.001 |
seek | +0.001 |
trainer | +0.001 |
Neg logit contributions | |
aying | -0.001 |
arse | -0.001 |
fears | -0.001 |
omed | -0.001 |
Ow | -0.001 |
Token | a |
Token ID | 257 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
rob | +0.038 |
couple | +0.037 |
seek | +0.033 |
gentleman | +0.033 |
thank | +0.032 |
Neg logit contributions | |
aying | -0.046 |
fears | -0.043 |
omed | -0.042 |
arse | -0.042 |
uning | -0.042 |
Token | few |
Token ID | 1178 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
rob | +0.055 |
couple | +0.048 |
gentleman | +0.045 |
thank | +0.045 |
robat | +0.044 |
Neg logit contributions | |
arse | -0.052 |
fears | -0.050 |
aying | -0.050 |
omed | -0.050 |
Ow | -0.047 |
Token | hours |
Token ID | 2250 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
couple | +0.047 |
rob | +0.045 |
seek | +0.039 |
gentleman | +0.039 |
thank | +0.037 |
Neg logit contributions | |
aying | -0.086 |
fears | -0.083 |
omed | -0.080 |
arse | -0.080 |
wipes | -0.080 |
Token | , |
Token ID | 11 |
Feature activation | +4.89e-01 |
Pos logit contributions | |
rob | +0.044 |
couple | +0.042 |
seek | +0.040 |
gentleman | +0.038 |
trainer | +0.038 |
Neg logit contributions | |
aying | -0.050 |
omed | -0.047 |
uning | -0.046 |
arse | -0.045 |
cleans | -0.045 |
Token | the |
Token ID | 262 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
couple | +0.017 |
rob | +0.017 |
seek | +0.015 |
gentleman | +0.014 |
trainer | +0.014 |
Neg logit contributions | |
aying | -0.027 |
fears | -0.026 |
arse | -0.025 |
uning | -0.025 |
Ow | -0.025 |
Token | witch |
Token ID | 16365 |
Feature activation | +8.34e-01 |
Pos logit contributions | |
rob | +0.006 |
couple | +0.004 |
robat | +0.004 |
gentleman | +0.004 |
seek | +0.004 |
Neg logit contributions | |
fears | -0.008 |
aying | -0.008 |
wipes | -0.008 |
cleans | -0.008 |
oken | -0.008 |
Token | stirred |
Token ID | 33091 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
rob | +0.036 |
couple | +0.034 |
seek | +0.030 |
gentleman | +0.030 |
robat | +0.029 |
Neg logit contributions | |
aying | -0.040 |
fears | -0.038 |
omed | -0.037 |
cleans | -0.037 |
arse | -0.037 |
Token | there |
Token ID | 612 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
aying | +0.023 |
fears | +0.023 |
arse | +0.022 |
uning | +0.022 |
adoes | +0.022 |
Neg logit contributions | |
couple | -0.010 |
rob | -0.010 |
seek | -0.010 |
lady | -0.008 |
gentleman | -0.008 |
Token | was |
Token ID | 373 |
Feature activation | +4.26e-03 |
Pos logit contributions | |
aying | +0.017 |
fears | +0.016 |
arse | +0.016 |
uning | +0.016 |
omed | +0.016 |
Neg logit contributions | |
couple | -0.010 |
rob | -0.009 |
seek | -0.008 |
gentleman | -0.008 |
lady | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
couple | +0.000 |
rob | +0.000 |
seek | +0.000 |
gentleman | +0.000 |
trainer | +0.000 |
Neg logit contributions | |
aying | -0.000 |
Ow | -0.000 |
fears | -0.000 |
arse | -0.000 |
uning | -0.000 |
Token | city |
Token ID | 1748 |
Feature activation | -8.10e-01 |
Pos logit contributions | |
aying | +0.008 |
fears | +0.008 |
arse | +0.008 |
uning | +0.008 |
Ow | +0.008 |
Neg logit contributions | |
couple | -0.009 |
rob | -0.009 |
seek | -0.008 |
gentleman | -0.008 |
thank | -0.008 |
Token | filled |
Token ID | 5901 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
aying | +0.039 |
omed | +0.036 |
uning | +0.035 |
ours | +0.035 |
fears | +0.035 |
Neg logit contributions | |
rob | -0.033 |
couple | -0.031 |
seek | -0.028 |
robat | -0.028 |
gentleman | -0.027 |
Token | with |
Token ID | 351 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
rob | +0.007 |
couple | +0.007 |
gentleman | +0.006 |
robat | +0.006 |
trainer | +0.006 |
Neg logit contributions | |
aying | -0.006 |
adoes | -0.006 |
Ow | -0.005 |
ours | -0.005 |
icker | -0.005 |
Token | joy |
Token ID | 8716 |
Feature activation | -4.74e-01 |
Pos logit contributions | |
rob | +0.062 |
couple | +0.059 |
seek | +0.055 |
gentleman | +0.053 |
robat | +0.052 |
Neg logit contributions | |
aying | -0.067 |
fears | -0.064 |
Ow | -0.061 |
uning | -0.061 |
arse | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | -3.73e-01 |
Pos logit contributions | |
aying | +0.028 |
fears | +0.027 |
arse | +0.027 |
uning | +0.026 |
Ow | +0.026 |
Neg logit contributions | |
couple | -0.016 |
rob | -0.014 |
seek | -0.013 |
lady | -0.012 |
gentleman | -0.012 |
Token | Everyone |
Token ID | 11075 |
Feature activation | -5.26e-01 |
Pos logit contributions | |
arse | +0.019 |
fears | +0.019 |
aying | +0.018 |
iqu | +0.018 |
icking | +0.017 |
Neg logit contributions | |
couple | -0.017 |
seek | -0.015 |
lady | -0.015 |
rob | -0.014 |
gentleman | -0.014 |
Token | was |
Token ID | 373 |
Feature activation | -6.04e-01 |
Pos logit contributions | |
aying | +0.028 |
arse | +0.027 |
fears | +0.026 |
uning | +0.026 |
omed | +0.026 |
Neg logit contributions | |
couple | -0.021 |
rob | -0.020 |
seek | -0.018 |
gentleman | -0.017 |
trainer | -0.017 |
Token | happy |
Token ID | 3772 |
Feature activation | -5.95e-01 |
Pos logit contributions | |
aying | +0.029 |
fears | +0.026 |
uning | +0.026 |
ours | +0.025 |
ains | +0.025 |
Neg logit contributions | |
rob | -0.028 |
robat | -0.024 |
couple | -0.023 |
seek | -0.022 |
gentleman | -0.022 |
Token | her |
Token ID | 607 |
Feature activation | +8.75e-02 |
Pos logit contributions | |
couple | +0.041 |
rob | +0.039 |
seek | +0.036 |
gentleman | +0.034 |
trainer | +0.034 |
Neg logit contributions | |
aying | -0.047 |
fears | -0.044 |
arse | -0.043 |
uning | -0.043 |
omed | -0.042 |
Token | paper |
Token ID | 3348 |
Feature activation | +5.31e-01 |
Pos logit contributions | |
couple | +0.004 |
rob | +0.004 |
seek | +0.003 |
gentleman | +0.003 |
trainer | +0.003 |
Neg logit contributions | |
aying | -0.004 |
fears | -0.004 |
arse | -0.004 |
uning | -0.004 |
Ow | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
couple | +0.020 |
rob | +0.020 |
seek | +0.017 |
gentleman | +0.017 |
trainer | +0.016 |
Neg logit contributions | |
aying | -0.029 |
fears | -0.027 |
arse | -0.027 |
uning | -0.026 |
omed | -0.026 |
Token | stamped |
Token ID | 28305 |
Feature activation | +9.79e-02 |
Pos logit contributions | |
rob | +0.010 |
couple | +0.009 |
gentleman | +0.009 |
robat | +0.009 |
seek | +0.009 |
Neg logit contributions | |
cleans | -0.007 |
wipes | -0.007 |
fears | -0.007 |
fists | -0.007 |
aying | -0.007 |
Token | them |
Token ID | 606 |
Feature activation | -8.04e-02 |
Pos logit contributions | |
rob | +0.004 |
couple | +0.004 |
seek | +0.004 |
gentleman | +0.003 |
trainer | +0.003 |
Neg logit contributions | |
aying | -0.005 |
fears | -0.005 |
arse | -0.004 |
uning | -0.004 |
Ow | -0.004 |
Token | with |
Token ID | 351 |
Feature activation | +3.78e-01 |
Pos logit contributions | |
aying | +0.005 |
omed | +0.005 |
Ow | +0.005 |
ours | +0.005 |
uning | +0.005 |
Neg logit contributions | |
couple | -0.002 |
seek | -0.002 |
rob | -0.002 |
lady | -0.002 |
man | -0.002 |
Token | the |
Token ID | 262 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
couple | +0.015 |
rob | +0.014 |
seek | +0.013 |
lady | +0.013 |
gentleman | +0.013 |
Neg logit contributions | |
aying | -0.020 |
arse | -0.018 |
fears | -0.018 |
uning | -0.018 |
omed | -0.018 |
Token | heart |
Token ID | 2612 |
Feature activation | +3.20e-01 |
Pos logit contributions | |
aying | +0.008 |
fears | +0.007 |
arse | +0.007 |
uning | +0.007 |
omed | +0.007 |
Neg logit contributions | |
couple | -0.007 |
rob | -0.006 |
seek | -0.006 |
gentleman | -0.006 |
lady | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
couple | +0.010 |
rob | +0.009 |
seek | +0.008 |
lady | +0.008 |
trainer | +0.008 |
Neg logit contributions | |
aying | -0.019 |
arse | -0.019 |
fears | -0.018 |
Ow | -0.018 |
uning | -0.018 |
Token | She |
Token ID | 1375 |
Feature activation | -2.62e-01 |
Pos logit contributions | |
couple | +0.008 |
lady | +0.007 |
rob | +0.007 |
gentleman | +0.007 |
thank | +0.007 |
Neg logit contributions | |
arse | -0.012 |
aying | -0.012 |
fears | -0.011 |
omed | -0.011 |
ist | -0.011 |
Token | wrote |
Token ID | 2630 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
aying | +0.013 |
fears | +0.013 |
wipes | +0.012 |
cleans | +0.012 |
uning | +0.012 |
Neg logit contributions | |
couple | -0.011 |
rob | -0.010 |
seek | -0.009 |
gentleman | -0.009 |
trainer | -0.009 |
Token | He |
Token ID | 679 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
fears | +0.046 |
aying | +0.046 |
arse | +0.046 |
arding | +0.043 |
iqu | +0.043 |
Neg logit contributions | |
rob | -0.027 |
seek | -0.026 |
couple | -0.026 |
ranger | -0.025 |
thank | -0.024 |
Token | loved |
Token ID | 6151 |
Feature activation | -7.76e-02 |
Pos logit contributions | |
couple | +0.007 |
rob | +0.007 |
seek | +0.006 |
gentleman | +0.006 |
trainer | +0.006 |
Neg logit contributions | |
aying | -0.007 |
fears | -0.007 |
arse | -0.007 |
uning | -0.007 |
Ow | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
aying | +0.004 |
fears | +0.004 |
uning | +0.004 |
arse | +0.004 |
Ow | +0.004 |
Neg logit contributions | |
couple | -0.003 |
rob | -0.003 |
seek | -0.003 |
gentleman | -0.003 |
trainer | -0.002 |
Token | go |
Token ID | 467 |
Feature activation | -5.91e-02 |
Pos logit contributions | |
aying | +0.008 |
omed | +0.008 |
arse | +0.008 |
outed | +0.008 |
fears | +0.008 |
Neg logit contributions | |
couple | -0.004 |
seek | -0.004 |
rob | -0.003 |
lady | -0.003 |
par | -0.003 |
Token | for |
Token ID | 329 |
Feature activation | +5.00e-01 |
Pos logit contributions | |
aying | +0.003 |
uning | +0.003 |
arse | +0.003 |
fears | +0.003 |
omed | +0.003 |
Neg logit contributions | |
couple | -0.003 |
rob | -0.002 |
gentleman | -0.002 |
seek | -0.002 |
trainer | -0.002 |
Token | walks |
Token ID | 11114 |
Feature activation | +4.01e-01 |
Pos logit contributions | |
couple | +0.023 |
rob | +0.022 |
seek | +0.021 |
gentleman | +0.020 |
trainer | +0.020 |
Neg logit contributions | |
aying | -0.022 |
fears | -0.021 |
arse | -0.021 |
uning | -0.020 |
Ow | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -3.65e-01 |
Pos logit contributions | |
couple | +0.015 |
rob | +0.015 |
gentleman | +0.013 |
seek | +0.013 |
trainer | +0.012 |
Neg logit contributions | |
aying | -0.021 |
fears | -0.020 |
arse | -0.020 |
uning | -0.020 |
omed | -0.019 |
Token | One |
Token ID | 1881 |
Feature activation | +3.17e-01 |
Pos logit contributions | |
aying | +0.018 |
fears | +0.017 |
arse | +0.017 |
adoes | +0.017 |
omed | +0.017 |
Neg logit contributions | |
couple | -0.015 |
rob | -0.015 |
seek | -0.013 |
gentleman | -0.013 |
lady | -0.013 |
Token | day |
Token ID | 1110 |
Feature activation | -6.17e-01 |
Pos logit contributions | |
rob | +0.009 |
couple | +0.008 |
seek | +0.007 |
trainer | +0.007 |
gentleman | +0.007 |
Neg logit contributions | |
aying | -0.020 |
fears | -0.019 |
Ow | -0.018 |
uning | -0.018 |
arse | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | -3.04e-01 |
Pos logit contributions | |
aying | +0.037 |
fears | +0.036 |
arse | +0.035 |
uning | +0.034 |
omed | +0.034 |
Neg logit contributions | |
couple | -0.020 |
rob | -0.019 |
seek | -0.017 |
gentleman | -0.016 |
lady | -0.016 |
Token | Tim |
Token ID | 5045 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
aying | +0.018 |
fears | +0.017 |
arse | +0.017 |
uning | +0.016 |
omed | +0.016 |
Neg logit contributions | |
couple | -0.010 |
rob | -0.010 |
seek | -0.009 |
gentleman | -0.008 |
lady | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
aying | +0.005 |
Ow | +0.005 |
fears | +0.004 |
omed | +0.004 |
arse | +0.004 |
Neg logit contributions | |
couple | -0.004 |
rob | -0.004 |
gentleman | -0.004 |
seek | -0.004 |
businessman | -0.004 |
Token | After |
Token ID | 3260 |
Feature activation | +9.81e-01 |
Pos logit contributions | |
aying | +0.006 |
fears | +0.005 |
Ow | +0.005 |
arse | +0.005 |
omed | +0.005 |
Neg logit contributions | |
couple | -0.006 |
rob | -0.006 |
gentleman | -0.005 |
trainer | -0.005 |
seek | -0.005 |
Token | playing |
Token ID | 2712 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
couple | +0.043 |
rob | +0.043 |
gentleman | +0.039 |
seek | +0.038 |
thank | +0.038 |
Neg logit contributions | |
aying | -0.044 |
omed | -0.041 |
uning | -0.041 |
Ow | -0.041 |
fears | -0.041 |
Token | for |
Token ID | 329 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
couple | +0.046 |
rob | +0.044 |
gentleman | +0.039 |
seek | +0.038 |
trainer | +0.038 |
Neg logit contributions | |
aying | -0.055 |
fears | -0.051 |
uning | -0.050 |
adoes | -0.050 |
wipes | -0.050 |
Token | a |
Token ID | 257 |
Feature activation | +9.94e-01 |
Pos logit contributions | |
couple | +0.057 |
rob | +0.055 |
gentleman | +0.053 |
trainer | +0.050 |
thank | +0.050 |
Neg logit contributions | |
aying | -0.048 |
Ow | -0.046 |
fears | -0.046 |
cleans | -0.045 |
oken | -0.044 |
Token | while |
Token ID | 981 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
rob | +0.041 |
couple | +0.040 |
seek | +0.035 |
gentleman | +0.035 |
thank | +0.034 |
Neg logit contributions | |
aying | -0.048 |
arse | -0.047 |
fears | -0.046 |
omed | -0.045 |
Ow | -0.044 |
Token | , |
Token ID | 11 |
Feature activation | -3.42e-02 |
Pos logit contributions | |
couple | +0.006 |
rob | +0.006 |
seek | +0.005 |
gentleman | +0.005 |
trainer | +0.005 |
Neg logit contributions | |
aying | -0.007 |
fears | -0.007 |
arse | -0.006 |
uning | -0.006 |
omed | -0.006 |
Token | Tim |
Token ID | 5045 |
Feature activation | -5.19e-02 |
Pos logit contributions | |
aying | +0.002 |
fears | +0.002 |
arse | +0.002 |
omed | +0.002 |
uning | +0.002 |
Neg logit contributions | |
couple | -0.001 |
rob | -0.001 |
seek | -0.001 |
lady | -0.001 |
man | -0.001 |
Token | started |
Token ID | 2067 |
Feature activation | +5.00e-01 |
Pos logit contributions | |
aying | +0.002 |
fears | +0.002 |
arse | +0.002 |
cleans | +0.002 |
omed | +0.002 |
Neg logit contributions | |
couple | -0.002 |
rob | -0.002 |
gentleman | -0.002 |
seek | -0.002 |
robat | -0.002 |
Token | to |
Token ID | 284 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
couple | +0.020 |
rob | +0.020 |
seek | +0.017 |
gentleman | +0.016 |
trainer | +0.016 |
Neg logit contributions | |
aying | -0.026 |
fears | -0.025 |
arse | -0.024 |
uning | -0.024 |
Ow | -0.024 |
Token | feel |
Token ID | 1254 |
Feature activation | +3.50e-01 |
Pos logit contributions | |
couple | +0.010 |
seek | +0.010 |
rob | +0.010 |
king | +0.009 |
lady | +0.009 |
Neg logit contributions | |
aying | -0.010 |
omed | -0.009 |
ours | -0.009 |
uning | -0.009 |
outed | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
aying | +0.017 |
arse | +0.017 |
Ow | +0.016 |
omed | +0.016 |
uning | +0.016 |
Neg logit contributions | |
couple | -0.010 |
seek | -0.009 |
lady | -0.009 |
rob | -0.008 |
man | -0.008 |
Token | The |
Token ID | 383 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
couple | +0.005 |
rob | +0.005 |
seek | +0.004 |
gentleman | +0.004 |
lady | +0.004 |
Neg logit contributions | |
aying | -0.007 |
arse | -0.006 |
fears | -0.006 |
uning | -0.006 |
omed | -0.006 |
Token | end |
Token ID | 886 |
Feature activation | +6.26e-01 |
Pos logit contributions | |
rob | +0.007 |
couple | +0.006 |
seek | +0.006 |
robat | +0.005 |
gentleman | +0.005 |
Neg logit contributions | |
aying | -0.008 |
fears | -0.008 |
cleans | -0.007 |
wipes | -0.007 |
arse | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +2.31e-01 |
Pos logit contributions | |
couple | +0.025 |
rob | +0.024 |
seek | +0.020 |
thank | +0.020 |
gentleman | +0.020 |
Neg logit contributions | |
aying | -0.033 |
fears | -0.031 |
omed | -0.031 |
arse | -0.030 |
uning | -0.030 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.74e-01 |
Pos logit contributions | |
couple | +0.009 |
rob | +0.008 |
seek | +0.008 |
trainer | +0.007 |
gentleman | +0.007 |
Neg logit contributions | |
aying | -0.013 |
fears | -0.012 |
arse | -0.012 |
Ow | -0.012 |
omed | -0.011 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
aying | +0.014 |
Ow | +0.013 |
omed | +0.013 |
arse | +0.013 |
fears | +0.013 |
Neg logit contributions | |
couple | -0.011 |
rob | -0.010 |
seek | -0.009 |
gentleman | -0.009 |
trainer | -0.009 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
couple | +0.021 |
rob | +0.020 |
seek | +0.018 |
gentleman | +0.018 |
lady | +0.018 |
Neg logit contributions | |
aying | -0.017 |
fears | -0.016 |
arse | -0.016 |
uning | -0.016 |
omed | -0.015 |
Token | a |
Token ID | 257 |
Feature activation | +4.42e-01 |
Pos logit contributions | |
aying | +0.005 |
Ow | +0.005 |
fears | +0.005 |
arse | +0.005 |
wipes | +0.005 |
Neg logit contributions | |
rob | -0.005 |
couple | -0.005 |
gentleman | -0.005 |
seek | -0.004 |
trainer | -0.004 |
Token | time |
Token ID | 640 |
Feature activation | -3.93e-01 |
Pos logit contributions | |
couple | +0.019 |
rob | +0.018 |
seek | +0.017 |
lady | +0.016 |
gentleman | +0.016 |
Neg logit contributions | |
uning | -0.021 |
aying | -0.021 |
adoes | -0.020 |
arse | -0.020 |
ours | -0.019 |
Token | there |
Token ID | 612 |
Feature activation | -3.84e-01 |
Pos logit contributions | |
aying | +0.024 |
fears | +0.024 |
arse | +0.023 |
uning | +0.023 |
adoes | +0.022 |
Neg logit contributions | |
couple | -0.011 |
rob | -0.011 |
seek | -0.010 |
lady | -0.009 |
gentleman | -0.009 |
Token | were |
Token ID | 547 |
Feature activation | +3.37e-01 |
Pos logit contributions | |
aying | +0.022 |
fears | +0.021 |
arse | +0.021 |
uning | +0.021 |
omed | +0.020 |
Neg logit contributions | |
couple | -0.013 |
rob | -0.013 |
seek | -0.011 |
gentleman | -0.011 |
lady | -0.010 |
Token | away |
Token ID | 1497 |
Feature activation | +4.18e-02 |
Pos logit contributions | |
aying | +0.002 |
uning | +0.002 |
fears | +0.002 |
omed | +0.002 |
adoes | +0.002 |
Neg logit contributions | |
couple | -0.002 |
rob | -0.002 |
gentleman | -0.002 |
trainer | -0.002 |
seek | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -2.38e-01 |
Pos logit contributions | |
couple | +0.002 |
rob | +0.001 |
seek | +0.001 |
gentleman | +0.001 |
trainer | +0.001 |
Neg logit contributions | |
aying | -0.002 |
fears | -0.002 |
arse | -0.002 |
uning | -0.002 |
omed | -0.002 |
Token | One |
Token ID | 1881 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
iqu | +0.013 |
arse | +0.013 |
fears | +0.012 |
icking | +0.012 |
arding | +0.012 |
Neg logit contributions | |
couple | -0.009 |
rob | -0.008 |
lady | -0.008 |
seek | -0.008 |
thank | -0.008 |
Token | day |
Token ID | 1110 |
Feature activation | -3.03e-01 |
Pos logit contributions | |
rob | +0.004 |
couple | +0.003 |
seek | +0.003 |
gentleman | +0.003 |
trainer | +0.003 |
Neg logit contributions | |
aying | -0.010 |
fears | -0.009 |
uning | -0.009 |
Ow | -0.009 |
arse | -0.009 |
Token | it |
Token ID | 340 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
aying | +0.023 |
fears | +0.022 |
arse | +0.022 |
iqu | +0.022 |
oken | +0.021 |
Neg logit contributions | |
couple | -0.006 |
man | -0.005 |
seek | -0.005 |
lady | -0.005 |
rob | -0.004 |
Token | stored |
Token ID | 8574 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
aying | +0.008 |
cleans | +0.008 |
wipes | +0.007 |
ours | +0.007 |
outed | +0.007 |
Neg logit contributions | |
rob | -0.009 |
couple | -0.008 |
robat | -0.007 |
gentleman | -0.007 |
trainer | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | +3.18e-01 |
Pos logit contributions | |
couple | +0.008 |
rob | +0.008 |
seek | +0.007 |
gentleman | +0.007 |
lady | +0.007 |
Neg logit contributions | |
aying | -0.011 |
arse | -0.010 |
fears | -0.010 |
uning | -0.010 |
omed | -0.010 |
Token | toy |
Token ID | 13373 |
Feature activation | +5.28e-01 |
Pos logit contributions | |
couple | +0.013 |
rob | +0.013 |
seek | +0.012 |
gentleman | +0.011 |
trainer | +0.011 |
Neg logit contributions | |
aying | -0.016 |
fears | -0.015 |
arse | -0.015 |
uning | -0.014 |
omed | -0.014 |
Token | plane |
Token ID | 6614 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
rob | +0.022 |
couple | +0.022 |
seek | +0.020 |
gentleman | +0.019 |
thank | +0.018 |
Neg logit contributions | |
aying | -0.025 |
uning | -0.024 |
fears | -0.024 |
Ow | -0.023 |
adoes | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | +2.78e-01 |
Pos logit contributions | |
aying | +0.010 |
fears | +0.010 |
arse | +0.009 |
uning | +0.009 |
omed | +0.009 |
Neg logit contributions | |
couple | -0.009 |
rob | -0.008 |
seek | -0.007 |
gentleman | -0.007 |
trainer | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
couple | +0.008 |
seek | +0.007 |
man | +0.007 |
lady | +0.007 |
par | +0.006 |
Neg logit contributions | |
aying | -0.018 |
arse | -0.018 |
omed | -0.018 |
fears | -0.018 |
uning | -0.017 |
Token | small |
Token ID | 1402 |
Feature activation | -3.97e-01 |
Pos logit contributions | |
rob | +0.025 |
couple | +0.024 |
thank | +0.022 |
gentleman | +0.022 |
seek | +0.021 |
Neg logit contributions | |
aying | -0.019 |
fears | -0.019 |
arse | -0.018 |
Ow | -0.018 |
wipes | -0.018 |
Token | house |
Token ID | 2156 |
Feature activation | -8.04e-01 |
Pos logit contributions | |
aying | +0.019 |
fears | +0.019 |
arse | +0.018 |
Ow | +0.018 |
wipes | +0.018 |
Neg logit contributions | |
rob | -0.016 |
couple | -0.016 |
seek | -0.014 |
thank | -0.014 |
gentleman | -0.014 |
Token | with |
Token ID | 351 |
Feature activation | +7.56e-01 |
Pos logit contributions | |
aying | +0.045 |
fears | +0.041 |
uning | +0.041 |
arse | +0.041 |
ours | +0.041 |
Neg logit contributions | |
couple | -0.029 |
rob | -0.028 |
gentleman | -0.024 |
seek | -0.023 |
trainer | -0.023 |
Token | his |
Token ID | 465 |
Feature activation | +6.72e-02 |
Pos logit contributions | |
couple | +0.030 |
rob | +0.030 |
gentleman | +0.025 |
seek | +0.025 |
thank | +0.024 |
Neg logit contributions | |
aying | -0.039 |
Ow | -0.038 |
fears | -0.037 |
uning | -0.036 |
wipes | -0.036 |
Token | best |
Token ID | 1266 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
rob | +0.002 |
couple | +0.002 |
seek | +0.002 |
gentleman | +0.002 |
thank | +0.002 |
Neg logit contributions | |
aying | -0.004 |
fears | -0.004 |
uning | -0.003 |
arse | -0.003 |
Ow | -0.003 |
Token | friend |
Token ID | 1545 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
aying | +0.078 |
arse | +0.074 |
omed | +0.073 |
uning | +0.071 |
fears | +0.071 |
Neg logit contributions | |
couple | -0.042 |
rob | -0.034 |
seek | -0.034 |
lady | -0.033 |
trainer | -0.032 |
Token | , |
Token ID | 11 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
arse | +0.017 |
aying | +0.016 |
fears | +0.016 |
iqu | +0.016 |
isel | +0.016 |
Neg logit contributions | |
couple | -0.005 |
rob | -0.005 |
seek | -0.005 |
lady | -0.005 |
king | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | -5.29e-01 |
Pos logit contributions | |
couple | +0.006 |
seek | +0.006 |
rob | +0.006 |
lady | +0.005 |
ranger | +0.005 |
Neg logit contributions | |
aying | -0.012 |
arse | -0.011 |
fears | -0.011 |
uning | -0.011 |
arding | -0.011 |
Token | little |
Token ID | 1310 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
aying | +0.028 |
fears | +0.026 |
arse | +0.026 |
uning | +0.026 |
omed | +0.025 |
Neg logit contributions | |
couple | -0.022 |
rob | -0.020 |
seek | -0.019 |
lady | -0.019 |
gentleman | -0.018 |
Token | boy |
Token ID | 2933 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
aying | +0.107 |
arse | +0.102 |
omed | +0.100 |
fears | +0.098 |
uning | +0.098 |
Neg logit contributions | |
couple | -0.040 |
rob | -0.032 |
lady | -0.032 |
seek | -0.030 |
man | -0.030 |
Token | named |
Token ID | 3706 |
Feature activation | -6.66e-01 |
Pos logit contributions | |
aying | +0.009 |
fears | +0.008 |
uning | +0.008 |
cleans | +0.008 |
Ow | +0.008 |
Neg logit contributions | |
couple | -0.006 |
rob | -0.006 |
gentleman | -0.005 |
seek | -0.005 |
thank | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
couple | +0.004 |
rob | +0.003 |
seek | +0.003 |
lady | +0.003 |
thank | +0.003 |
Neg logit contributions | |
aying | -0.009 |
arse | -0.009 |
fears | -0.009 |
Ow | -0.008 |
uning | -0.008 |
Token | They |
Token ID | 1119 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
couple | +0.002 |
rob | +0.002 |
lady | +0.002 |
seek | +0.002 |
gentleman | +0.002 |
Neg logit contributions | |
arse | -0.002 |
aying | -0.002 |
fears | -0.002 |
uning | -0.002 |
iqu | -0.002 |
Token | mixed |
Token ID | 7668 |
Feature activation | +8.39e-02 |
Pos logit contributions | |
aying | +0.009 |
fears | +0.008 |
arse | +0.008 |
uning | +0.008 |
omed | +0.008 |
Neg logit contributions | |
couple | -0.008 |
rob | -0.008 |
seek | -0.007 |
gentleman | -0.007 |
trainer | -0.006 |
Token | them |
Token ID | 606 |
Feature activation | -6.82e-01 |
Pos logit contributions | |
couple | +0.003 |
rob | +0.003 |
seek | +0.003 |
thank | +0.003 |
lady | +0.003 |
Neg logit contributions | |
aying | -0.004 |
arse | -0.004 |
fears | -0.004 |
uning | -0.004 |
omed | -0.004 |
Token | with |
Token ID | 351 |
Feature activation | +5.44e-01 |
Pos logit contributions | |
aying | +0.041 |
arse | +0.041 |
fears | +0.040 |
omed | +0.038 |
iqu | +0.038 |
Neg logit contributions | |
couple | -0.022 |
seek | -0.019 |
rob | -0.018 |
thank | -0.018 |
l | -0.018 |
Token | a |
Token ID | 257 |
Feature activation | -6.98e-02 |
Pos logit contributions | |
couple | +0.019 |
l | +0.016 |
thank | +0.015 |
rob | +0.015 |
par | +0.015 |
Neg logit contributions | |
aying | -0.032 |
arse | -0.032 |
uning | -0.030 |
fears | -0.030 |
omed | -0.030 |
Token | spoon |
Token ID | 24556 |
Feature activation | -3.25e-01 |
Pos logit contributions | |
aying | +0.004 |
uning | +0.003 |
adoes | +0.003 |
fears | +0.003 |
arse | +0.003 |
Neg logit contributions | |
couple | -0.003 |
rob | -0.003 |
seek | -0.003 |
lady | -0.003 |
man | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +8.83e-02 |
Pos logit contributions | |
aying | +0.018 |
fears | +0.017 |
arse | +0.017 |
uning | +0.016 |
omed | +0.016 |
Neg logit contributions | |
couple | -0.012 |
rob | -0.012 |
seek | -0.010 |
gentleman | -0.010 |
trainer | -0.010 |
Token | They |
Token ID | 1119 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
couple | +0.003 |
rob | +0.003 |
lady | +0.003 |
seek | +0.003 |
ranger | +0.003 |
Neg logit contributions | |
arse | -0.005 |
fears | -0.005 |
aying | -0.005 |
iqu | -0.005 |
fists | -0.005 |
Token | said |
Token ID | 531 |
Feature activation | +2.43e-01 |
Pos logit contributions | |
aying | +0.016 |
fears | +0.015 |
arse | +0.015 |
uning | +0.014 |
omed | +0.014 |
Neg logit contributions | |
couple | -0.013 |
rob | -0.013 |
seek | -0.012 |
gentleman | -0.011 |
trainer | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
couple | +0.007 |
lady | +0.005 |
man | +0.005 |
trainer | +0.005 |
par | +0.005 |
Neg logit contributions | |
arse | -0.017 |
aying | -0.016 |
iqu | -0.016 |
fears | -0.016 |
ist | -0.015 |
Token | Sarah |
Token ID | 10490 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
couple | +0.012 |
rob | +0.012 |
seek | +0.010 |
gentleman | +0.010 |
trainer | +0.010 |
Neg logit contributions | |
aying | -0.019 |
fears | -0.018 |
arse | -0.017 |
uning | -0.017 |
Ow | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | -2.91e-01 |
Pos logit contributions | |
couple | +0.007 |
lady | +0.006 |
man | +0.006 |
gentleman | +0.005 |
ranger | +0.005 |
Neg logit contributions | |
aying | -0.017 |
arse | -0.017 |
fears | -0.016 |
Ow | -0.016 |
oken | -0.016 |
Token | " |
Token ID | 366 |
Feature activation | -3.96e-01 |
Pos logit contributions | |
aying | +0.017 |
fears | +0.017 |
arse | +0.016 |
uning | +0.016 |
Ow | +0.016 |
Neg logit contributions | |
couple | -0.010 |
rob | -0.008 |
lady | -0.008 |
gentleman | -0.008 |
seek | -0.008 |
Token | If |
Token ID | 1532 |
Feature activation | +4.41e-01 |
Pos logit contributions | |
aying | +0.019 |
arse | +0.017 |
fears | +0.017 |
uning | +0.017 |
omed | +0.017 |
Neg logit contributions | |
couple | -0.018 |
rob | -0.017 |
seek | -0.015 |
gentleman | -0.015 |
trainer | -0.015 |
Token | you |
Token ID | 345 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
couple | +0.012 |
seek | +0.011 |
lady | +0.010 |
man | +0.009 |
rob | +0.009 |
Neg logit contributions | |
aying | -0.029 |
arse | -0.028 |
omed | -0.027 |
uning | -0.027 |
fears | -0.027 |
Token | can |
Token ID | 460 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
adoes | +0.005 |
arse | +0.005 |
ours | +0.005 |
aying | +0.005 |
fists | +0.005 |
Neg logit contributions | |
seek | -0.004 |
couple | -0.004 |
trainer | -0.003 |
king | -0.003 |
thank | -0.003 |
Token | guess |
Token ID | 4724 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
aying | +0.010 |
fears | +0.009 |
adoes | +0.009 |
arse | +0.009 |
omed | +0.009 |
Neg logit contributions | |
couple | -0.008 |
seek | -0.007 |
rob | -0.007 |
trainer | -0.006 |
lady | -0.006 |
Token | what |
Token ID | 644 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
couple | +0.005 |
l | +0.005 |
seek | +0.005 |
rob | +0.005 |
trainer | +0.004 |
Neg logit contributions | |
arse | -0.006 |
aying | -0.006 |
outed | -0.006 |
omed | -0.006 |
oken | -0.006 |
Token | 's |
Token ID | 338 |
Feature activation | -4.97e-01 |
Pos logit contributions | |
arse | +0.007 |
aying | +0.007 |
omed | +0.006 |
fears | +0.006 |
uning | +0.006 |
Neg logit contributions | |
couple | -0.005 |
rob | -0.005 |
seek | -0.005 |
lady | -0.004 |
man | -0.004 |
Token | inside |
Token ID | 2641 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
aying | +0.027 |
arse | +0.026 |
omed | +0.026 |
fears | +0.025 |
Ow | +0.025 |
Neg logit contributions | |
couple | -0.018 |
rob | -0.017 |
seek | -0.017 |
lady | -0.016 |
trainer | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
couple | +0.006 |
rob | +0.005 |
seek | +0.005 |
lady | +0.005 |
man | +0.005 |
Neg logit contributions | |
aying | -0.010 |
arse | -0.010 |
fears | -0.010 |
uning | -0.010 |
omed | -0.009 |
Token | time |
Token ID | 640 |
Feature activation | -4.78e-01 |
Pos logit contributions | |
couple | +0.010 |
rob | +0.009 |
seek | +0.009 |
lady | +0.008 |
trainer | +0.008 |
Neg logit contributions | |
aying | -0.010 |
uning | -0.010 |
adoes | -0.010 |
arse | -0.009 |
ours | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
aying | +0.027 |
fears | +0.025 |
arse | +0.025 |
uning | +0.025 |
wipes | +0.024 |
Neg logit contributions | |
couple | -0.017 |
rob | -0.016 |
seek | -0.015 |
gentleman | -0.014 |
lady | -0.014 |
Token | there |
Token ID | 612 |
Feature activation | -7.60e-02 |
Pos logit contributions | |
couple | +0.006 |
rob | +0.006 |
seek | +0.006 |
lady | +0.005 |
man | +0.005 |
Neg logit contributions | |
aying | -0.009 |
fears | -0.008 |
uning | -0.008 |
wipes | -0.008 |
arse | -0.008 |
Token | was |
Token ID | 373 |
Feature activation | -7.22e-02 |
Pos logit contributions | |
aying | +0.004 |
Ow | +0.004 |
fears | +0.003 |
arse | +0.003 |
cleans | +0.003 |
Neg logit contributions | |
couple | -0.003 |
rob | -0.003 |
seek | -0.003 |
gentleman | -0.003 |
trainer | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | -3.02e-01 |
Pos logit contributions | |
aying | +0.003 |
Ow | +0.003 |
fears | +0.003 |
arse | +0.003 |
wipes | +0.003 |
Neg logit contributions | |
rob | -0.003 |
couple | -0.003 |
gentleman | -0.003 |
seek | -0.003 |
trainer | -0.003 |
Token | little |
Token ID | 1310 |
Feature activation | -8.86e-01 |
Pos logit contributions | |
aying | +0.011 |
fears | +0.011 |
arse | +0.011 |
Ow | +0.010 |
wipes | +0.010 |
Neg logit contributions | |
rob | -0.016 |
couple | -0.015 |
seek | -0.014 |
thank | -0.014 |
gentleman | -0.014 |
Token | boy |
Token ID | 2933 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
aying | +0.056 |
fears | +0.053 |
arse | +0.053 |
uning | +0.052 |
Ow | +0.051 |
Neg logit contributions | |
couple | -0.025 |
rob | -0.024 |
seek | -0.021 |
gentleman | -0.019 |
trainer | -0.019 |
Token | named |
Token ID | 3706 |
Feature activation | -3.47e-01 |
Pos logit contributions | |
aying | +0.010 |
fears | +0.009 |
cleans | +0.009 |
wipes | +0.009 |
uning | +0.009 |
Neg logit contributions | |
couple | -0.008 |
rob | -0.008 |
gentleman | -0.007 |
seek | -0.007 |
trainer | -0.006 |
Token | Tim |
Token ID | 5045 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
aying | +0.016 |
fears | +0.015 |
uning | +0.014 |
arse | +0.014 |
omed | +0.014 |
Neg logit contributions | |
couple | -0.016 |
rob | -0.016 |
seek | -0.015 |
lady | -0.014 |
trainer | -0.014 |
Token | my |
Token ID | 1820 |
Feature activation | -2.83e-01 |
Pos logit contributions | |
aying | +0.006 |
fears | +0.006 |
arse | +0.006 |
Ow | +0.005 |
uning | +0.005 |
Neg logit contributions | |
couple | -0.008 |
rob | -0.008 |
seek | -0.007 |
gentleman | -0.007 |
trainer | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -6.20e-01 |
Pos logit contributions | |
aying | +0.013 |
fears | +0.012 |
Ow | +0.012 |
arse | +0.012 |
cleans | +0.012 |
Neg logit contributions | |
couple | -0.013 |
rob | -0.013 |
seek | -0.012 |
gentleman | -0.011 |
trainer | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | +4.37e-01 |
Pos logit contributions | |
couple | +0.008 |
seek | +0.007 |
rob | +0.007 |
lady | +0.007 |
gentleman | +0.006 |
Neg logit contributions | |
aying | -0.015 |
uning | -0.015 |
arse | -0.014 |
fears | -0.014 |
omed | -0.014 |
Token | great |
Token ID | 1049 |
Feature activation | +8.84e-02 |
Pos logit contributions | |
couple | +0.020 |
rob | +0.019 |
seek | +0.018 |
lady | +0.017 |
gentleman | +0.017 |
Neg logit contributions | |
aying | -0.020 |
uning | -0.019 |
arse | -0.019 |
fears | -0.019 |
omed | -0.018 |
Token | job |
Token ID | 1693 |
Feature activation | +8.37e-01 |
Pos logit contributions | |
rob | +0.004 |
couple | +0.004 |
seek | +0.004 |
trainer | +0.003 |
gentleman | +0.003 |
Neg logit contributions | |
aying | -0.004 |
fears | -0.004 |
uning | -0.004 |
Ow | -0.004 |
arse | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | -9.07e-01 |
Pos logit contributions | |
couple | +0.030 |
rob | +0.029 |
seek | +0.026 |
gentleman | +0.024 |
lady | +0.024 |
Neg logit contributions | |
aying | -0.047 |
fears | -0.045 |
arse | -0.044 |
uning | -0.044 |
Ow | -0.043 |
Token | Lily |
Token ID | 20037 |
Feature activation | +4.69e-01 |
Pos logit contributions | |
aying | +0.050 |
arse | +0.048 |
uning | +0.047 |
fears | +0.047 |
omed | +0.046 |
Neg logit contributions | |
couple | -0.034 |
rob | -0.031 |
seek | -0.030 |
lady | -0.029 |
gentleman | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | -3.06e-01 |
Pos logit contributions | |
couple | +0.019 |
rob | +0.019 |
seek | +0.016 |
gentleman | +0.016 |
trainer | +0.016 |
Neg logit contributions | |
aying | -0.024 |
fears | -0.022 |
arse | -0.022 |
cleans | -0.021 |
wipes | -0.021 |
Token | I |
Token ID | 314 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
fears | +0.015 |
aying | +0.015 |
arse | +0.015 |
fists | +0.014 |
omed | +0.014 |
Neg logit contributions | |
couple | -0.012 |
rob | -0.012 |
seek | -0.011 |
gentleman | -0.011 |
thank | -0.010 |
Token | 'm |
Token ID | 1101 |
Feature activation | -9.23e-02 |
Pos logit contributions | |
couple | +0.006 |
rob | +0.006 |
gentleman | +0.005 |
trainer | +0.005 |
seek | +0.005 |
Neg logit contributions | |
aying | -0.005 |
fears | -0.005 |
cleans | -0.005 |
wipes | -0.005 |
omed | -0.005 |
Token | happy |
Token ID | 3772 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
aying | +0.005 |
ok | +0.004 |
ours | +0.004 |
arse | +0.004 |
cleans | +0.004 |
Neg logit contributions | |
rob | -0.004 |
couple | -0.004 |
robat | -0.004 |
seek | -0.004 |
trainer | -0.003 |
Token | you |
Token ID | 345 |
Feature activation | -9.57e-03 |
Pos logit contributions | |
aying | +0.009 |
fears | +0.008 |
arse | +0.008 |
uning | +0.008 |
omed | +0.008 |
Neg logit contributions | |
couple | -0.006 |
rob | -0.006 |
seek | -0.005 |
gentleman | -0.005 |
trainer | -0.005 |
Token | 're |
Token ID | 821 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
aying | +0.000 |
fears | +0.000 |
arse | +0.000 |
omed | +0.000 |
uning | +0.000 |
Neg logit contributions | |
couple | -0.000 |
rob | -0.000 |
gentleman | -0.000 |
seek | -0.000 |
trainer | -0.000 |
Token | it |
Token ID | 340 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
aying | +0.006 |
uning | +0.006 |
arse | +0.006 |
omed | +0.006 |
fears | +0.005 |
Neg logit contributions | |
couple | -0.004 |
rob | -0.003 |
seek | -0.003 |
lady | -0.003 |
trainer | -0.003 |
Token | ! |
Token ID | 0 |
Feature activation | -3.56e-01 |
Pos logit contributions | |
couple | +0.006 |
rob | +0.005 |
gentleman | +0.005 |
seek | +0.004 |
trainer | +0.004 |
Neg logit contributions | |
aying | -0.007 |
arse | -0.006 |
fears | -0.006 |
Ow | -0.006 |
uning | -0.006 |
Token | We |
Token ID | 775 |
Feature activation | +3.19e-01 |
Pos logit contributions | |
arse | +0.017 |
aying | +0.017 |
fears | +0.017 |
omed | +0.016 |
uning | +0.016 |
Neg logit contributions | |
couple | -0.015 |
rob | -0.014 |
seek | -0.013 |
trainer | -0.013 |
gentleman | -0.013 |
Token | have |
Token ID | 423 |
Feature activation | -4.44e-01 |
Pos logit contributions | |
couple | +0.012 |
rob | +0.012 |
seek | +0.012 |
thank | +0.011 |
trainer | +0.011 |
Neg logit contributions | |
aying | -0.016 |
arse | -0.016 |
adoes | -0.016 |
uning | -0.015 |
fears | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | -4.54e-01 |
Pos logit contributions | |
aying | +0.025 |
arse | +0.024 |
fears | +0.023 |
uning | +0.023 |
omed | +0.023 |
Neg logit contributions | |
couple | -0.017 |
rob | -0.015 |
seek | -0.014 |
lady | -0.013 |
thank | -0.013 |
Token | best |
Token ID | 1266 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
aying | +0.025 |
omed | +0.024 |
arse | +0.024 |
ours | +0.023 |
arding | +0.023 |
Neg logit contributions | |
couple | -0.016 |
king | -0.015 |
lady | -0.015 |
rob | -0.014 |
seek | -0.014 |
Token | tower |
Token ID | 10580 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
aying | +0.072 |
fears | +0.067 |
arse | +0.066 |
uning | +0.065 |
omed | +0.064 |
Neg logit contributions | |
couple | -0.071 |
rob | -0.069 |
seek | -0.063 |
gentleman | -0.061 |
trainer | -0.060 |
Token | !" |
Token ID | 2474 |
Feature activation | +3.54e-02 |
Pos logit contributions | |
aying | +0.016 |
fears | +0.015 |
cleans | +0.015 |
adoes | +0.014 |
omed | +0.014 |
Neg logit contributions | |
couple | -0.008 |
rob | -0.008 |
gentleman | -0.007 |
seek | -0.006 |
trainer | -0.006 |
Token | They |
Token ID | 1119 |
Feature activation | -4.55e-01 |
Pos logit contributions | |
couple | +0.001 |
rob | +0.001 |
lady | +0.001 |
man | +0.001 |
trainer | +0.001 |
Neg logit contributions | |
arse | -0.002 |
aying | -0.002 |
uning | -0.002 |
fears | -0.002 |
iqu | -0.002 |
Token | see |
Token ID | 766 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
aying | +0.024 |
adoes | +0.023 |
arse | +0.023 |
uning | +0.023 |
fears | +0.023 |
Neg logit contributions | |
couple | -0.017 |
lady | -0.016 |
seek | -0.016 |
trainer | -0.015 |
rob | -0.015 |
Token | Lily |
Token ID | 20037 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
aying | +0.012 |
arse | +0.012 |
omed | +0.011 |
uning | +0.011 |
fears | +0.011 |
Neg logit contributions | |
couple | -0.006 |
rob | -0.006 |
lady | -0.005 |
man | -0.005 |
trainer | -0.005 |
Token | always |
Token ID | 1464 |
Feature activation | +2.85e-01 |
Pos logit contributions | |
couple | +0.021 |
seek | +0.019 |
lady | +0.018 |
thank | +0.018 |
king | +0.018 |
Neg logit contributions | |
omed | -0.039 |
aying | -0.037 |
fists | -0.037 |
ours | -0.037 |
arse | -0.037 |
Token | be |
Token ID | 307 |
Feature activation | -6.67e-01 |
Pos logit contributions | |
couple | +0.011 |
rob | +0.011 |
seek | +0.010 |
gentleman | +0.010 |
trainer | +0.009 |
Neg logit contributions | |
aying | -0.015 |
arse | -0.014 |
fears | -0.014 |
omed | -0.014 |
uning | -0.014 |
Token | kind |
Token ID | 1611 |
Feature activation | -3.10e-01 |
Pos logit contributions | |
aying | +0.039 |
arse | +0.037 |
fears | +0.036 |
uning | +0.036 |
Ow | +0.036 |
Neg logit contributions | |
couple | -0.023 |
rob | -0.021 |
seek | -0.020 |
lady | -0.019 |
gentleman | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
aying | +0.018 |
arse | +0.018 |
omed | +0.017 |
Ow | +0.017 |
fears | +0.017 |
Neg logit contributions | |
couple | -0.010 |
rob | -0.008 |
seek | -0.008 |
lady | -0.008 |
gentleman | -0.008 |
Token | think |
Token ID | 892 |
Feature activation | -6.00e-01 |
Pos logit contributions | |
aying | +0.012 |
fears | +0.011 |
arse | +0.011 |
uning | +0.011 |
omed | +0.011 |
Neg logit contributions | |
couple | -0.010 |
rob | -0.010 |
seek | -0.009 |
gentleman | -0.009 |
trainer | -0.009 |
Token | of |
Token ID | 286 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
arse | +0.037 |
Ow | +0.036 |
fears | +0.034 |
uning | +0.034 |
aying | +0.034 |
Neg logit contributions | |
couple | -0.018 |
man | -0.015 |
lady | -0.015 |
knowing | -0.014 |
king | -0.014 |
Token | others |
Token ID | 1854 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
omed | +0.079 |
aying | +0.077 |
arse | +0.076 |
outed | +0.074 |
uning | +0.074 |
Neg logit contributions | |
couple | -0.050 |
rob | -0.045 |
man | -0.044 |
lady | -0.043 |
knowing | -0.042 |
Token | before |
Token ID | 878 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
arse | +0.006 |
Ow | +0.006 |
aying | +0.006 |
uning | +0.006 |
omed | +0.006 |
Neg logit contributions | |
couple | -0.003 |
man | -0.003 |
rob | -0.003 |
seek | -0.003 |
lady | -0.003 |
Token | we |
Token ID | 356 |
Feature activation | -6.86e-02 |
Pos logit contributions | |
couple | +0.008 |
seek | +0.006 |
rob | +0.006 |
man | +0.006 |
lady | +0.006 |
Neg logit contributions | |
aying | -0.011 |
uning | -0.011 |
omed | -0.011 |
arse | -0.011 |
fears | -0.011 |
Token | think |
Token ID | 892 |
Feature activation | -6.27e-01 |
Pos logit contributions | |
aying | +0.003 |
arse | +0.003 |
omed | +0.003 |
adoes | +0.003 |
ours | +0.003 |
Neg logit contributions | |
couple | -0.003 |
seek | -0.003 |
trainer | -0.003 |
thank | -0.003 |
rob | -0.003 |
Token | of |
Token ID | 286 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
arse | +0.037 |
Ow | +0.036 |
fears | +0.036 |
aying | +0.036 |
uning | +0.036 |
Neg logit contributions | |
couple | -0.020 |
man | -0.016 |
lady | -0.016 |
rob | -0.016 |
seek | -0.016 |
Token | appetite |
Token ID | 20788 |
Feature activation | -3.74e-01 |
Pos logit contributions | |
aying | +0.004 |
arse | +0.004 |
fears | +0.004 |
omed | +0.004 |
uning | +0.004 |
Neg logit contributions | |
couple | -0.003 |
rob | -0.003 |
seek | -0.003 |
gentleman | -0.003 |
lady | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
aying | +0.019 |
uning | +0.018 |
fears | +0.017 |
ows | +0.017 |
icking | +0.017 |
Neg logit contributions | |
rob | -0.016 |
couple | -0.015 |
seek | -0.013 |
gentleman | -0.013 |
thank | -0.012 |
Token | lived |
Token ID | 5615 |
Feature activation | -5.08e-02 |
Pos logit contributions | |
couple | +0.005 |
rob | +0.005 |
robat | +0.005 |
gentleman | +0.005 |
thank | +0.005 |
Neg logit contributions | |
aying | -0.004 |
fears | -0.004 |
cleans | -0.004 |
wipes | -0.004 |
adoes | -0.004 |
Token | in |
Token ID | 287 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
aying | +0.002 |
fears | +0.002 |
uning | +0.002 |
Ow | +0.002 |
outed | +0.002 |
Neg logit contributions | |
couple | -0.003 |
rob | -0.003 |
gentleman | -0.002 |
robat | -0.002 |
businessman | -0.002 |
Token | a |
Token ID | 257 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
couple | +0.001 |
rob | +0.001 |
gentleman | +0.001 |
trainer | +0.001 |
thank | +0.001 |
Neg logit contributions | |
aying | -0.001 |
fears | -0.001 |
Ow | -0.001 |
fists | -0.001 |
adoes | -0.001 |
Token | swamp |
Token ID | 26837 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
rob | +0.010 |
couple | +0.010 |
thank | +0.009 |
gentleman | +0.009 |
robat | +0.008 |
Neg logit contributions | |
fears | -0.008 |
aying | -0.008 |
arse | -0.007 |
Ow | -0.007 |
wipes | -0.007 |
Token | in |
Token ID | 287 |
Feature activation | -1.69e-04 |
Pos logit contributions | |
aying | +0.081 |
fears | +0.077 |
arse | +0.077 |
uning | +0.075 |
omed | +0.075 |
Neg logit contributions | |
couple | -0.048 |
rob | -0.046 |
seek | -0.041 |
gentleman | -0.039 |
trainer | -0.038 |
Token | the |
Token ID | 262 |
Feature activation | +3.68e-01 |
Pos logit contributions | |
aying | +0.000 |
Ow | +0.000 |
fears | +0.000 |
adoes | +0.000 |
cleans | +0.000 |
Neg logit contributions | |
couple | -0.000 |
rob | -0.000 |
gentleman | -0.000 |
trainer | -0.000 |
thank | -0.000 |
Token | middle |
Token ID | 3504 |
Feature activation | +2.03e-02 |
Pos logit contributions | |
rob | +0.017 |
couple | +0.016 |
gentleman | +0.014 |
robat | +0.014 |
thank | +0.013 |
Neg logit contributions | |
aying | -0.017 |
fears | -0.017 |
Ow | -0.016 |
arse | -0.015 |
wipes | -0.015 |
Token | of |
Token ID | 286 |
Feature activation | +3.62e-01 |
Pos logit contributions | |
rob | +0.001 |
couple | +0.001 |
gentleman | +0.001 |
thank | +0.001 |
trainer | +0.001 |
Neg logit contributions | |
aying | -0.001 |
fears | -0.001 |
omed | -0.001 |
uning | -0.001 |
ours | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | +3.36e-01 |
Pos logit contributions | |
couple | +0.014 |
rob | +0.013 |
seek | +0.012 |
gentleman | +0.012 |
trainer | +0.011 |
Neg logit contributions | |
aying | -0.019 |
fears | -0.018 |
Ow | -0.018 |
arse | -0.018 |
uning | -0.018 |
Token | his |
Token ID | 465 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
aying | +0.016 |
arse | +0.015 |
fears | +0.015 |
uning | +0.015 |
omed | +0.014 |
Neg logit contributions | |
couple | -0.006 |
seek | -0.005 |
rob | -0.005 |
l | -0.004 |
lady | -0.004 |
Token | best |
Token ID | 1266 |
Feature activation | -9.48e-01 |
Pos logit contributions | |
aying | +0.012 |
omed | +0.011 |
arse | +0.011 |
uning | +0.011 |
Ow | +0.011 |
Neg logit contributions | |
couple | -0.008 |
seek | -0.007 |
rob | -0.007 |
man | -0.006 |
lady | -0.006 |
Token | to |
Token ID | 284 |
Feature activation | -4.92e-01 |
Pos logit contributions | |
aying | +0.054 |
arse | +0.051 |
fears | +0.050 |
uning | +0.050 |
omed | +0.049 |
Neg logit contributions | |
couple | -0.034 |
rob | -0.032 |
seek | -0.029 |
gentleman | -0.028 |
lady | -0.027 |
Token | get |
Token ID | 651 |
Feature activation | -5.87e-01 |
Pos logit contributions | |
aying | +0.028 |
omed | +0.027 |
uning | +0.027 |
ains | +0.027 |
ours | +0.027 |
Neg logit contributions | |
seek | -0.018 |
couple | -0.017 |
lady | -0.015 |
thank | -0.014 |
l | -0.013 |
Token | rid |
Token ID | 5755 |
Feature activation | -3.57e-01 |
Pos logit contributions | |
aying | +0.036 |
arse | +0.035 |
fears | +0.034 |
uning | +0.033 |
Ow | +0.033 |
Neg logit contributions | |
couple | -0.018 |
rob | -0.016 |
seek | -0.016 |
lady | -0.015 |
l | -0.015 |
Token | of |
Token ID | 286 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
arse | +0.024 |
fears | +0.024 |
iqu | +0.024 |
ours | +0.024 |
uning | +0.023 |
Neg logit contributions | |
couple | -0.006 |
rob | -0.005 |
seek | -0.005 |
lady | -0.005 |
thank | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | -3.56e-01 |
Pos logit contributions | |
aying | +0.074 |
arse | +0.071 |
uning | +0.069 |
fears | +0.069 |
omed | +0.069 |
Neg logit contributions | |
couple | -0.037 |
rob | -0.034 |
seek | -0.031 |
lady | -0.031 |
gentleman | -0.029 |
Token | big |
Token ID | 1263 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
aying | +0.021 |
arse | +0.020 |
uning | +0.019 |
Ow | +0.019 |
fears | +0.018 |
Neg logit contributions | |
couple | -0.014 |
lady | -0.012 |
man | -0.012 |
seek | -0.012 |
rob | -0.012 |
Token | spot |
Token ID | 4136 |
Feature activation | +4.42e-02 |
Pos logit contributions | |
aying | +0.007 |
arse | +0.006 |
uning | +0.006 |
fears | +0.006 |
Ow | +0.006 |
Neg logit contributions | |
couple | -0.004 |
rob | -0.004 |
seek | -0.004 |
lady | -0.004 |
gentleman | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
couple | +0.001 |
rob | +0.001 |
seek | +0.001 |
lady | +0.001 |
gentleman | +0.001 |
Neg logit contributions | |
aying | -0.003 |
arse | -0.003 |
fears | -0.003 |
Ow | -0.002 |
uning | -0.002 |
Token | Lucy |
Token ID | 22162 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
aying | +0.007 |
arse | +0.007 |
fears | +0.007 |
uning | +0.006 |
arding | +0.006 |
Neg logit contributions | |
couple | -0.006 |
rob | -0.005 |
lady | -0.005 |
seek | -0.005 |
gentleman | -0.005 |
Token | colorful |
Token ID | 20239 |
Feature activation | +5.31e-01 |
Pos logit contributions | |
rob | +0.034 |
couple | +0.034 |
seek | +0.031 |
thank | +0.030 |
gentleman | +0.030 |
Neg logit contributions | |
aying | -0.042 |
fears | -0.041 |
omed | -0.039 |
wipes | -0.039 |
arse | -0.039 |
Token | cand |
Token ID | 2658 |
Feature activation | +9.81e-02 |
Pos logit contributions | |
couple | +0.021 |
rob | +0.021 |
seek | +0.019 |
gentleman | +0.018 |
thank | +0.018 |
Neg logit contributions | |
aying | -0.027 |
fears | -0.025 |
uning | -0.025 |
omed | -0.025 |
arse | -0.025 |
Token | ies |
Token ID | 444 |
Feature activation | -4.86e-01 |
Pos logit contributions | |
couple | +0.005 |
rob | +0.005 |
seek | +0.005 |
gentleman | +0.005 |
trainer | +0.005 |
Neg logit contributions | |
aying | -0.004 |
fears | -0.003 |
arse | -0.003 |
uning | -0.003 |
omed | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | +5.34e-02 |
Pos logit contributions | |
aying | +0.028 |
fears | +0.026 |
arse | +0.026 |
uning | +0.026 |
Ow | +0.026 |
Neg logit contributions | |
couple | -0.017 |
rob | -0.016 |
seek | -0.014 |
gentleman | -0.014 |
trainer | -0.013 |
Token | ch |
Token ID | 442 |
Feature activation | -6.30e-01 |
Pos logit contributions | |
rob | +0.002 |
robat | +0.002 |
couple | +0.002 |
gentleman | +0.002 |
seek | +0.002 |
Neg logit contributions | |
wipes | -0.002 |
fears | -0.002 |
fists | -0.002 |
cleans | -0.002 |
aying | -0.002 |
Token | ocol |
Token ID | 4668 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
arse | +0.031 |
aying | +0.029 |
cleans | +0.028 |
wipes | +0.028 |
Ow | +0.027 |
Neg logit contributions | |
couple | -0.026 |
rob | -0.022 |
trainer | -0.022 |
thank | -0.020 |
par | -0.020 |
Token | ates |
Token ID | 689 |
Feature activation | -8.17e-01 |
Pos logit contributions | |
aying | +0.057 |
uning | +0.054 |
Ow | +0.051 |
omed | +0.051 |
arse | +0.050 |
Neg logit contributions | |
couple | -0.064 |
rob | -0.058 |
seek | -0.058 |
gentleman | -0.057 |
lady | -0.056 |
Token | . |
Token ID | 13 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
aying | +0.049 |
fears | +0.047 |
arse | +0.046 |
Ow | +0.045 |
uning | +0.045 |
Neg logit contributions | |
couple | -0.027 |
rob | -0.024 |
seek | -0.022 |
lady | -0.021 |
trainer | -0.021 |
Token | She |
Token ID | 1375 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
couple | +0.007 |
lady | +0.006 |
gentleman | +0.006 |
rob | +0.005 |
man | +0.005 |
Neg logit contributions | |
aying | -0.010 |
arse | -0.010 |
fears | -0.010 |
omed | -0.009 |
oken | -0.009 |
Token | said |
Token ID | 531 |
Feature activation | +2.99e-01 |
Pos logit contributions | |
aying | +0.009 |
fears | +0.009 |
outed | +0.009 |
wipes | +0.009 |
ours | +0.009 |
Neg logit contributions | |
rob | -0.008 |
couple | -0.008 |
robat | -0.007 |
seek | -0.007 |
gentleman | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | +2.56e-01 |
Pos logit contributions | |
couple | +0.010 |
rob | +0.009 |
seek | +0.008 |
lady | +0.008 |
gentleman | +0.008 |
Neg logit contributions | |
aying | -0.018 |
arse | -0.017 |
fears | -0.017 |
uning | -0.016 |
omed | -0.016 |