Token | danced |
Token ID | 39480 |
Feature activation | -5.42e-01 |
Pos logit contributions | |
yes | +0.015 |
goals | +0.014 |
best | +0.014 |
mm | +0.014 |
narrative | +0.013 |
Neg logit contributions | |
over | -0.011 |
late | -0.010 |
To | -0.010 |
yelled | -0.010 |
aback | -0.010 |
Token | around |
Token ID | 1088 |
Feature activation | +5.05e-01 |
Pos logit contributions | |
over | +0.012 |
late | +0.011 |
To | +0.011 |
aback | +0.011 |
temperature | +0.011 |
Neg logit contributions | |
yes | -0.014 |
goals | -0.014 |
best | -0.013 |
mm | -0.013 |
narrative | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
mm | +0.021 |
yes | +0.020 |
narrative | +0.019 |
goals | +0.018 |
best | +0.018 |
Neg logit contributions | |
over | -0.006 |
into | -0.005 |
towards | -0.004 |
To | -0.004 |
late | -0.004 |
Token | z |
Token ID | 1976 |
Feature activation | +3.23e+00 |
Pos logit contributions | |
yes | +0.050 |
goals | +0.048 |
best | +0.047 |
mm | +0.044 |
narrative | +0.043 |
Neg logit contributions | |
over | -0.048 |
late | -0.046 |
To | -0.045 |
temperature | -0.045 |
aback | -0.045 |
Token | ig |
Token ID | 328 |
Feature activation | +2.72e+00 |
Pos logit contributions | |
yes | +0.051 |
goals | +0.042 |
narrative | +0.041 |
best | +0.040 |
mm | +0.036 |
Neg logit contributions | |
ray | -0.101 |
into | -0.100 |
Start | -0.100 |
over | -0.099 |
To | -0.098 |
Token | zag |
Token ID | 50183 |
Feature activation | +4.39e+00 |
Pos logit contributions | |
goals | +0.075 |
narrative | +0.075 |
yes | +0.073 |
mm | +0.071 |
nations | +0.066 |
Neg logit contributions | |
over | -0.062 |
into | -0.057 |
any | -0.052 |
towards | -0.051 |
To | -0.050 |
Token | together |
Token ID | 1978 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
mm | +0.124 |
yes | +0.122 |
narrative | +0.113 |
owers | +0.106 |
Deal | +0.104 |
Neg logit contributions | |
over | -0.101 |
into | -0.095 |
towards | -0.087 |
yelled | -0.080 |
around | -0.080 |
Token | ! |
Token ID | 0 |
Feature activation | +2.91e-01 |
Pos logit contributions | |
mm | +0.035 |
yes | +0.033 |
narrative | +0.033 |
goals | +0.031 |
Deal | +0.031 |
Neg logit contributions | |
over | -0.019 |
into | -0.017 |
towards | -0.015 |
To | -0.014 |
aback | -0.014 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
yes | +0.010 |
goals | +0.009 |
best | +0.009 |
mm | +0.009 |
narrative | +0.009 |
Neg logit contributions | |
temperature | -0.004 |
over | -0.004 |
person | -0.004 |
aback | -0.004 |
ray | -0.004 |
Token | John |
Token ID | 7554 |
Feature activation | +4.02e-01 |
Pos logit contributions | |
over | +0.026 |
temperature | +0.025 |
person | +0.025 |
late | +0.024 |
aback | +0.024 |
Neg logit contributions | |
yes | -0.062 |
goals | -0.060 |
best | -0.059 |
mm | -0.057 |
narrative | -0.056 |
Token | and |
Token ID | 290 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
yes | +0.011 |
goals | +0.010 |
best | +0.010 |
mm | +0.009 |
ho | +0.009 |
Neg logit contributions | |
To | -0.008 |
Start | -0.008 |
late | -0.008 |
person | -0.008 |
aback | -0.008 |
Token | back |
Token ID | 736 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
temperature | +0.009 |
ray | +0.009 |
To | +0.009 |
person | +0.009 |
Grand | +0.009 |
Neg logit contributions | |
yes | -0.024 |
goals | -0.023 |
best | -0.022 |
ho | -0.021 |
mm | -0.021 |
Token | down |
Token ID | 866 |
Feature activation | +6.89e-02 |
Pos logit contributions | |
To | +0.002 |
person | +0.002 |
late | +0.002 |
over | +0.002 |
Grand | +0.002 |
Neg logit contributions | |
yes | -0.007 |
goals | -0.006 |
best | -0.006 |
mm | -0.006 |
ho | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
yes | +0.002 |
goals | +0.002 |
best | +0.002 |
mm | +0.002 |
narrative | +0.002 |
Neg logit contributions | |
over | -0.001 |
To | -0.001 |
temperature | -0.001 |
person | -0.001 |
ray | -0.001 |
Token | z |
Token ID | 1976 |
Feature activation | +2.95e+00 |
Pos logit contributions | |
goals | +0.019 |
best | +0.019 |
yes | +0.018 |
ig | +0.015 |
ho | +0.015 |
Neg logit contributions | |
aback | -0.034 |
person | -0.033 |
To | -0.033 |
over | -0.032 |
To | -0.032 |
Token | ig |
Token ID | 328 |
Feature activation | +2.30e+00 |
Pos logit contributions | |
yes | +0.034 |
goals | +0.027 |
best | +0.026 |
narrative | +0.023 |
mm | +0.023 |
Neg logit contributions | |
over | -0.105 |
ray | -0.104 |
temperature | -0.104 |
To | -0.104 |
into | -0.103 |
Token | zag |
Token ID | 50183 |
Feature activation | +4.30e+00 |
Pos logit contributions | |
goals | +0.056 |
yes | +0.055 |
narrative | +0.053 |
mm | +0.052 |
best | +0.050 |
Neg logit contributions | |
over | -0.058 |
into | -0.053 |
To | -0.051 |
To | -0.050 |
towards | -0.050 |
Token | branch |
Token ID | 8478 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
yes | +0.122 |
mm | +0.119 |
narrative | +0.114 |
goals | +0.112 |
best | +0.108 |
Neg logit contributions | |
over | -0.090 |
into | -0.083 |
towards | -0.079 |
To | -0.075 |
To | -0.075 |
Token | and |
Token ID | 290 |
Feature activation | +3.78e-01 |
Pos logit contributions | |
yes | +0.044 |
goals | +0.043 |
mm | +0.042 |
narrative | +0.041 |
best | +0.041 |
Neg logit contributions | |
over | -0.024 |
into | -0.021 |
late | -0.021 |
aback | -0.020 |
To | -0.020 |
Token | lay |
Token ID | 3830 |
Feature activation | -6.67e-01 |
Pos logit contributions | |
yes | +0.010 |
best | +0.010 |
goals | +0.009 |
ho | +0.009 |
mm | +0.008 |
Neg logit contributions | |
temperature | -0.008 |
late | -0.007 |
person | -0.007 |
To | -0.007 |
ray | -0.007 |
Token | in |
Token ID | 287 |
Feature activation | +8.60e-01 |
Pos logit contributions | |
over | +0.009 |
late | +0.008 |
into | +0.008 |
temperature | +0.008 |
aback | +0.008 |
Neg logit contributions | |
yes | -0.023 |
goals | -0.023 |
best | -0.022 |
mm | -0.022 |
narrative | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | +9.90e-01 |
Pos logit contributions | |
yes | +0.028 |
goals | +0.027 |
best | +0.026 |
mm | +0.026 |
narrative | +0.025 |
Neg logit contributions | |
over | -0.013 |
late | -0.012 |
To | -0.012 |
temperature | -0.012 |
aback | -0.012 |
Token | Jenny |
Token ID | 28437 |
Feature activation | -4.82e-01 |
Pos logit contributions | |
over | +0.026 |
temperature | +0.024 |
aback | +0.024 |
To | +0.024 |
late | +0.024 |
Neg logit contributions | |
yes | -0.075 |
goals | -0.071 |
best | -0.070 |
mm | -0.068 |
narrative | -0.066 |
Token | raced |
Token ID | 32897 |
Feature activation | +8.13e-01 |
Pos logit contributions | |
late | +0.009 |
aback | +0.009 |
To | +0.009 |
over | +0.009 |
temperature | +0.009 |
Neg logit contributions | |
yes | -0.014 |
goals | -0.013 |
best | -0.013 |
ig | -0.012 |
mm | -0.012 |
Token | around |
Token ID | 1088 |
Feature activation | +4.82e-01 |
Pos logit contributions | |
yes | +0.025 |
best | +0.024 |
goals | +0.023 |
course | +0.020 |
ig | +0.019 |
Neg logit contributions | |
ray | -0.016 |
Tooth | -0.015 |
temperature | -0.015 |
To | -0.014 |
comm | -0.014 |
Token | on |
Token ID | 319 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
yes | +0.017 |
mm | +0.016 |
goals | +0.016 |
narrative | +0.016 |
best | +0.016 |
Neg logit contributions | |
over | -0.007 |
into | -0.006 |
late | -0.006 |
To | -0.006 |
To | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
yes | +0.046 |
mm | +0.043 |
goals | +0.043 |
best | +0.043 |
narrative | +0.042 |
Neg logit contributions | |
over | -0.021 |
into | -0.018 |
late | -0.018 |
To | -0.017 |
To | -0.017 |
Token | sc |
Token ID | 629 |
Feature activation | +4.19e+00 |
Pos logit contributions | |
yes | +0.055 |
goals | +0.052 |
best | +0.051 |
mm | +0.047 |
narrative | +0.046 |
Neg logit contributions | |
over | -0.055 |
late | -0.055 |
To | -0.054 |
aback | -0.053 |
temperature | -0.053 |
Token | ooter |
Token ID | 25141 |
Feature activation | +2.30e+00 |
Pos logit contributions | |
yes | +0.108 |
goals | +0.101 |
best | +0.099 |
mm | +0.095 |
narrative | +0.090 |
Neg logit contributions | |
over | -0.093 |
late | -0.090 |
temperature | -0.089 |
To | -0.088 |
aback | -0.086 |
Token | . |
Token ID | 13 |
Feature activation | -8.46e-01 |
Pos logit contributions | |
yes | +0.075 |
goals | +0.072 |
mm | +0.070 |
best | +0.070 |
narrative | +0.069 |
Neg logit contributions | |
over | -0.036 |
into | -0.031 |
To | -0.031 |
late | -0.031 |
To | -0.031 |
Token | She |
Token ID | 1375 |
Feature activation | -5.77e-02 |
Pos logit contributions | |
comm | +0.019 |
temperature | +0.019 |
yelled | +0.018 |
plug | +0.018 |
ray | +0.018 |
Neg logit contributions | |
yes | -0.021 |
goals | -0.020 |
best | -0.019 |
mm | -0.018 |
owers | -0.017 |
Token | had |
Token ID | 550 |
Feature activation | -4.97e-01 |
Pos logit contributions | |
over | +0.001 |
late | +0.001 |
To | +0.001 |
temperature | +0.001 |
aback | +0.001 |
Neg logit contributions | |
yes | -0.002 |
goals | -0.002 |
best | -0.002 |
mm | -0.002 |
narrative | -0.002 |
Token | so |
Token ID | 523 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
person | +0.010 |
ray | +0.010 |
temperature | +0.010 |
late | +0.010 |
To | +0.010 |
Neg logit contributions | |
yes | -0.014 |
best | -0.013 |
goals | -0.013 |
course | -0.011 |
mm | -0.011 |
Token | Bob |
Token ID | 5811 |
Feature activation | +3.61e-01 |
Pos logit contributions | |
yes | +0.010 |
best | +0.010 |
goals | +0.009 |
ir | +0.008 |
ho | +0.008 |
Neg logit contributions | |
ray | -0.011 |
plug | -0.010 |
Grand | -0.010 |
any | -0.009 |
To | -0.009 |
Token | onto |
Token ID | 4291 |
Feature activation | +7.45e-02 |
Pos logit contributions | |
yes | +0.012 |
best | +0.011 |
goals | +0.011 |
ho | +0.010 |
mm | +0.010 |
Neg logit contributions | |
late | -0.005 |
To | -0.005 |
over | -0.005 |
Start | -0.005 |
person | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
yes | +0.002 |
goals | +0.002 |
best | +0.002 |
mm | +0.002 |
course | +0.002 |
Neg logit contributions | |
ray | -0.001 |
To | -0.001 |
over | -0.001 |
person | -0.001 |
Grand | -0.001 |
Token | z |
Token ID | 1976 |
Feature activation | +2.84e+00 |
Pos logit contributions | |
yes | +0.030 |
goals | +0.028 |
best | +0.027 |
mm | +0.027 |
narrative | +0.026 |
Neg logit contributions | |
over | -0.023 |
late | -0.022 |
To | -0.022 |
temperature | -0.022 |
aback | -0.021 |
Token | ig |
Token ID | 328 |
Feature activation | +2.45e+00 |
Pos logit contributions | |
yes | +0.024 |
goals | +0.020 |
best | +0.018 |
mm | +0.016 |
narrative | +0.015 |
Neg logit contributions | |
over | -0.110 |
temperature | -0.108 |
ray | -0.108 |
To | -0.108 |
To | -0.107 |
Token | zag |
Token ID | 50183 |
Feature activation | +4.12e+00 |
Pos logit contributions | |
yes | +0.068 |
goals | +0.066 |
mm | +0.063 |
narrative | +0.062 |
best | +0.062 |
Neg logit contributions | |
over | -0.050 |
into | -0.046 |
late | -0.045 |
To | -0.045 |
To | -0.045 |
Token | path |
Token ID | 3108 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
yes | +0.120 |
mm | +0.114 |
goals | +0.112 |
best | +0.109 |
narrative | +0.108 |
Neg logit contributions | |
over | -0.080 |
into | -0.073 |
To | -0.071 |
temperature | -0.070 |
late | -0.070 |
Token | . |
Token ID | 13 |
Feature activation | +4.77e-01 |
Pos logit contributions | |
yes | +0.044 |
mm | +0.043 |
goals | +0.043 |
narrative | +0.042 |
best | +0.041 |
Neg logit contributions | |
over | -0.017 |
into | -0.015 |
late | -0.014 |
aback | -0.014 |
towards | -0.014 |
Token | Together |
Token ID | 17083 |
Feature activation | +4.54e-01 |
Pos logit contributions | |
yes | +0.014 |
goals | +0.013 |
best | +0.013 |
mm | +0.012 |
owers | +0.011 |
Neg logit contributions | |
temperature | -0.009 |
over | -0.008 |
comm | -0.008 |
yelled | -0.008 |
Grand | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | +7.53e-02 |
Pos logit contributions | |
yes | +0.011 |
goals | +0.010 |
mm | +0.010 |
best | +0.010 |
narrative | +0.010 |
Neg logit contributions | |
over | -0.011 |
late | -0.010 |
into | -0.010 |
temperature | -0.010 |
To | -0.010 |
Token | they |
Token ID | 484 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
yes | +0.002 |
goals | +0.002 |
mm | +0.002 |
best | +0.002 |
narrative | +0.002 |
Neg logit contributions | |
over | -0.002 |
late | -0.002 |
into | -0.002 |
To | -0.002 |
To | -0.002 |
Token | wiped |
Token ID | 21122 |
Feature activation | -7.46e-01 |
Pos logit contributions | |
yes | +0.004 |
goals | +0.004 |
best | +0.004 |
mm | +0.003 |
narrative | +0.003 |
Neg logit contributions | |
over | -0.002 |
late | -0.002 |
To | -0.002 |
aback | -0.002 |
temperature | -0.002 |
Token | away |
Token ID | 1497 |
Feature activation | -7.51e-01 |
Pos logit contributions | |
person | +0.017 |
To | +0.017 |
ray | +0.016 |
Grand | +0.016 |
late | +0.016 |
Neg logit contributions | |
goals | -0.019 |
best | -0.019 |
yes | -0.017 |
course | -0.017 |
ig | -0.016 |
Token | his |
Token ID | 465 |
Feature activation | +6.98e-01 |
Pos logit contributions | |
ray | +0.015 |
To | +0.015 |
person | +0.015 |
late | +0.015 |
Grand | +0.015 |
Neg logit contributions | |
yes | -0.021 |
goals | -0.020 |
best | -0.019 |
nations | -0.016 |
mm | -0.016 |
Token | z |
Token ID | 1976 |
Feature activation | +3.14e+00 |
Pos logit contributions | |
yes | +0.020 |
goals | +0.019 |
best | +0.018 |
mm | +0.018 |
narrative | +0.018 |
Neg logit contributions | |
over | -0.013 |
temperature | -0.013 |
late | -0.012 |
To | -0.012 |
aback | -0.012 |
Token | ig |
Token ID | 328 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
yes | +0.037 |
goals | +0.033 |
narrative | +0.030 |
best | +0.029 |
nations | +0.028 |
Neg logit contributions | |
over | -0.106 |
ray | -0.105 |
temperature | -0.105 |
into | -0.104 |
Start | -0.103 |
Token | zag |
Token ID | 50183 |
Feature activation | +4.09e+00 |
Pos logit contributions | |
yes | +0.064 |
goals | +0.062 |
best | +0.058 |
mm | +0.057 |
narrative | +0.057 |
Neg logit contributions | |
over | -0.063 |
To | -0.059 |
late | -0.058 |
temperature | -0.058 |
into | -0.058 |
Token | and |
Token ID | 290 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
yes | +0.117 |
goals | +0.111 |
mm | +0.108 |
best | +0.107 |
narrative | +0.104 |
Neg logit contributions | |
over | -0.080 |
To | -0.073 |
into | -0.073 |
temperature | -0.073 |
late | -0.073 |
Token | went |
Token ID | 1816 |
Feature activation | -5.82e-01 |
Pos logit contributions | |
over | +0.000 |
To | +0.000 |
late | +0.000 |
temperature | +0.000 |
aback | +0.000 |
Neg logit contributions | |
yes | -0.000 |
goals | -0.000 |
best | -0.000 |
mm | -0.000 |
narrative | -0.000 |
Token | on |
Token ID | 319 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
Grand | +0.013 |
comm | +0.013 |
person | +0.013 |
Start | +0.012 |
ray | +0.012 |
Neg logit contributions | |
best | -0.014 |
yes | -0.014 |
goals | -0.013 |
ig | -0.013 |
soaring | -0.011 |
Token | with |
Token ID | 351 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
yes | +0.005 |
goals | +0.005 |
best | +0.005 |
mm | +0.004 |
narrative | +0.004 |
Neg logit contributions | |
ray | -0.003 |
person | -0.003 |
To | -0.003 |
over | -0.003 |
aback | -0.003 |
Token | his |
Token ID | 465 |
Feature activation | -5.98e-02 |
Pos logit contributions | |
yes | +0.004 |
goals | +0.003 |
mm | +0.003 |
narrative | +0.003 |
best | +0.003 |
Neg logit contributions | |
over | -0.002 |
late | -0.001 |
aback | -0.001 |
into | -0.001 |
yelled | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | -3.83e-01 |
Pos logit contributions | |
over | +0.009 |
into | +0.008 |
temperature | +0.008 |
yelled | +0.007 |
towards | +0.007 |
Neg logit contributions | |
yes | -0.019 |
mm | -0.019 |
goals | -0.019 |
narrative | -0.018 |
Deal | -0.018 |
Token | pushed |
Token ID | 7121 |
Feature activation | +5.12e-01 |
Pos logit contributions | |
ray | +0.007 |
person | +0.007 |
To | +0.007 |
temperature | +0.007 |
aback | +0.007 |
Neg logit contributions | |
yes | -0.011 |
best | -0.010 |
goals | -0.010 |
ho | -0.010 |
course | -0.009 |
Token | her |
Token ID | 607 |
Feature activation | -2.64e-01 |
Pos logit contributions | |
yes | +0.013 |
goals | +0.012 |
best | +0.012 |
mm | +0.011 |
narrative | +0.011 |
Neg logit contributions | |
over | -0.011 |
To | -0.011 |
late | -0.011 |
ray | -0.011 |
temperature | -0.011 |
Token | off |
Token ID | 572 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
over | +0.004 |
into | +0.004 |
temperature | +0.004 |
aback | +0.004 |
yelled | +0.004 |
Neg logit contributions | |
yes | -0.009 |
mm | -0.008 |
goals | -0.008 |
narrative | -0.008 |
best | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
yes | +0.006 |
goals | +0.006 |
best | +0.006 |
mm | +0.006 |
narrative | +0.006 |
Neg logit contributions | |
over | -0.002 |
late | -0.002 |
To | -0.002 |
temperature | -0.002 |
into | -0.002 |
Token | sc |
Token ID | 629 |
Feature activation | +4.05e+00 |
Pos logit contributions | |
yes | +0.030 |
goals | +0.028 |
best | +0.028 |
mm | +0.026 |
narrative | +0.026 |
Neg logit contributions | |
over | -0.030 |
late | -0.029 |
To | -0.029 |
aback | -0.029 |
To | -0.028 |
Token | ooter |
Token ID | 25141 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
yes | +0.109 |
goals | +0.102 |
best | +0.100 |
mm | +0.097 |
narrative | +0.093 |
Neg logit contributions | |
over | -0.086 |
late | -0.081 |
To | -0.080 |
temperature | -0.080 |
aback | -0.079 |
Token | . |
Token ID | 13 |
Feature activation | -3.36e-01 |
Pos logit contributions | |
yes | +0.084 |
goals | +0.080 |
best | +0.079 |
mm | +0.077 |
narrative | +0.076 |
Neg logit contributions | |
over | -0.027 |
late | -0.024 |
To | -0.024 |
temperature | -0.023 |
aback | -0.023 |
Token | Anna |
Token ID | 11735 |
Feature activation | -8.48e-01 |
Pos logit contributions | |
any | +0.007 |
temperature | +0.007 |
ray | +0.007 |
person | +0.007 |
ilda | +0.007 |
Neg logit contributions | |
goals | -0.009 |
yes | -0.009 |
mm | -0.008 |
best | -0.008 |
narrative | -0.007 |
Token | fell |
Token ID | 3214 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
over | +0.016 |
late | +0.015 |
To | +0.015 |
temperature | +0.015 |
person | +0.015 |
Neg logit contributions | |
yes | -0.025 |
goals | -0.023 |
best | -0.023 |
mm | -0.022 |
narrative | -0.021 |
Token | on |
Token ID | 319 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
person | +0.003 |
ray | +0.003 |
To | +0.003 |
Grand | +0.003 |
To | +0.003 |
Neg logit contributions | |
yes | -0.006 |
best | -0.005 |
goals | -0.005 |
ho | -0.005 |
ig | -0.005 |
Token | z |
Token ID | 1976 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
yes | +0.020 |
best | +0.019 |
goals | +0.018 |
ig | +0.014 |
mm | +0.014 |
Neg logit contributions | |
person | -0.028 |
To | -0.028 |
ray | -0.028 |
aback | -0.028 |
temperature | -0.027 |
Token | ooming |
Token ID | 30602 |
Feature activation | +4.51e-01 |
Pos logit contributions | |
goals | +0.020 |
yes | +0.018 |
best | +0.017 |
narrative | +0.017 |
mm | +0.017 |
Neg logit contributions | |
over | -0.037 |
into | -0.036 |
ray | -0.036 |
aback | -0.035 |
Start | -0.035 |
Token | around |
Token ID | 1088 |
Feature activation | +2.72e-01 |
Pos logit contributions | |
best | +0.012 |
goals | +0.012 |
yes | +0.011 |
ho | +0.009 |
course | +0.009 |
Neg logit contributions | |
person | -0.010 |
Tooth | -0.010 |
ray | -0.010 |
any | -0.010 |
Em | -0.009 |
Token | on |
Token ID | 319 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
yes | +0.008 |
goals | +0.008 |
best | +0.008 |
mm | +0.007 |
narrative | +0.007 |
Neg logit contributions | |
over | -0.004 |
To | -0.004 |
person | -0.004 |
ray | -0.004 |
aback | -0.004 |
Token | his |
Token ID | 465 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
yes | +0.035 |
goals | +0.034 |
best | +0.033 |
mm | +0.032 |
narrative | +0.031 |
Neg logit contributions | |
over | -0.017 |
To | -0.016 |
late | -0.016 |
aback | -0.015 |
person | -0.015 |
Token | sc |
Token ID | 629 |
Feature activation | +4.05e+00 |
Pos logit contributions | |
yes | +0.048 |
goals | +0.046 |
best | +0.044 |
mm | +0.043 |
narrative | +0.041 |
Neg logit contributions | |
over | -0.046 |
late | -0.044 |
To | -0.043 |
aback | -0.043 |
temperature | -0.043 |
Token | ooter |
Token ID | 25141 |
Feature activation | +2.16e+00 |
Pos logit contributions | |
yes | +0.107 |
goals | +0.101 |
best | +0.099 |
mm | +0.094 |
narrative | +0.090 |
Neg logit contributions | |
over | -0.087 |
late | -0.084 |
temperature | -0.083 |
To | -0.082 |
aback | -0.082 |
Token | when |
Token ID | 618 |
Feature activation | +5.99e-02 |
Pos logit contributions | |
yes | +0.069 |
goals | +0.066 |
best | +0.066 |
mm | +0.060 |
narrative | +0.058 |
Neg logit contributions | |
aback | -0.033 |
late | -0.033 |
over | -0.032 |
To | -0.032 |
Start | -0.032 |
Token | he |
Token ID | 339 |
Feature activation | +6.50e-02 |
Pos logit contributions | |
yes | +0.002 |
goals | +0.002 |
mm | +0.002 |
narrative | +0.002 |
owers | +0.002 |
Neg logit contributions | |
over | -0.001 |
late | -0.001 |
into | -0.001 |
yelled | -0.001 |
To | -0.001 |
Token | stopped |
Token ID | 5025 |
Feature activation | +9.52e-01 |
Pos logit contributions | |
yes | +0.002 |
best | +0.002 |
goals | +0.002 |
ig | +0.002 |
mm | +0.002 |
Neg logit contributions | |
person | -0.001 |
aback | -0.001 |
over | -0.001 |
To | -0.001 |
late | -0.001 |
Token | at |
Token ID | 379 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
yes | +0.028 |
goals | +0.026 |
mm | +0.026 |
best | +0.025 |
narrative | +0.025 |
Neg logit contributions | |
over | -0.019 |
into | -0.017 |
late | -0.017 |
temperature | -0.016 |
yelled | -0.016 |
Token | out |
Token ID | 503 |
Feature activation | -7.43e-01 |
Pos logit contributions | |
yes | +0.004 |
best | +0.003 |
ho | +0.003 |
goals | +0.003 |
course | +0.003 |
Neg logit contributions | |
ray | -0.003 |
Em | -0.003 |
comm | -0.003 |
person | -0.003 |
shave | -0.003 |
Token | to |
Token ID | 284 |
Feature activation | +7.89e-01 |
Pos logit contributions | |
person | +0.013 |
ray | +0.013 |
To | +0.012 |
temperature | +0.012 |
Start | +0.012 |
Neg logit contributions | |
yes | -0.023 |
goals | -0.021 |
best | -0.021 |
narrative | -0.019 |
ho | -0.019 |
Token | see |
Token ID | 766 |
Feature activation | +9.41e-01 |
Pos logit contributions | |
yes | +0.024 |
best | +0.021 |
goals | +0.021 |
course | +0.020 |
tweet | +0.020 |
Neg logit contributions | |
late | -0.013 |
To | -0.013 |
person | -0.013 |
Grand | -0.013 |
temperature | -0.013 |
Token | what |
Token ID | 644 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
yes | +0.028 |
goals | +0.028 |
mm | +0.027 |
narrative | +0.026 |
best | +0.026 |
Neg logit contributions | |
over | -0.018 |
aback | -0.016 |
To | -0.016 |
into | -0.015 |
late | -0.015 |
Token | it |
Token ID | 340 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
yes | +0.039 |
goals | +0.037 |
best | +0.036 |
mm | +0.035 |
narrative | +0.034 |
Neg logit contributions | |
over | -0.025 |
To | -0.023 |
late | -0.023 |
temperature | -0.023 |
aback | -0.023 |
Token | was |
Token ID | 373 |
Feature activation | +4.04e+00 |
Pos logit contributions | |
yes | +0.051 |
goals | +0.048 |
best | +0.047 |
mm | +0.045 |
narrative | +0.044 |
Neg logit contributions | |
over | -0.034 |
late | -0.033 |
To | -0.032 |
temperature | -0.032 |
person | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | +7.31e-02 |
Pos logit contributions | |
yes | +0.137 |
best | +0.131 |
goals | +0.128 |
mm | +0.118 |
narrative | +0.118 |
Neg logit contributions | |
person | -0.055 |
To | -0.055 |
Start | -0.054 |
late | -0.053 |
over | -0.053 |
Token | Sammy |
Token ID | 44067 |
Feature activation | +3.59e-01 |
Pos logit contributions | |
yes | +0.002 |
goals | +0.002 |
best | +0.002 |
owers | +0.002 |
ho | +0.001 |
Neg logit contributions | |
temperature | -0.002 |
shave | -0.002 |
Grand | -0.002 |
bo | -0.002 |
plug | -0.002 |
Token | was |
Token ID | 373 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
yes | +0.010 |
best | +0.009 |
goals | +0.009 |
ho | +0.008 |
course | +0.008 |
Neg logit contributions | |
late | -0.007 |
person | -0.007 |
To | -0.007 |
temperature | -0.007 |
ray | -0.007 |
Token | so |
Token ID | 523 |
Feature activation | +3.90e-01 |
Pos logit contributions | |
yes | +0.028 |
best | +0.027 |
goals | +0.024 |
narrative | +0.020 |
tweet | +0.020 |
Neg logit contributions | |
ray | -0.028 |
person | -0.027 |
To | -0.027 |
temperature | -0.027 |
olyn | -0.026 |
Token | relieved |
Token ID | 22598 |
Feature activation | +6.81e-01 |
Pos logit contributions | |
yes | +0.010 |
goals | +0.009 |
best | +0.009 |
mm | +0.008 |
narrative | +0.008 |
Neg logit contributions | |
over | -0.009 |
temperature | -0.009 |
person | -0.009 |
ray | -0.009 |
To | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
temperature | +0.011 |
person | +0.010 |
ray | +0.010 |
comm | +0.010 |
plug | +0.009 |
Neg logit contributions | |
yes | -0.010 |
best | -0.010 |
goals | -0.009 |
mm | -0.009 |
course | -0.008 |
Token | Jimmy |
Token ID | 40335 |
Feature activation | -6.14e-01 |
Pos logit contributions | |
temperature | +0.004 |
aback | +0.004 |
ray | +0.003 |
into | +0.003 |
source | +0.003 |
Neg logit contributions | |
yes | -0.004 |
goals | -0.003 |
best | -0.003 |
narrative | -0.003 |
mm | -0.003 |
Token | quickly |
Token ID | 2952 |
Feature activation | +9.59e-01 |
Pos logit contributions | |
late | +0.012 |
aback | +0.012 |
To | +0.012 |
person | +0.011 |
temperature | +0.011 |
Neg logit contributions | |
yes | -0.017 |
best | -0.016 |
goals | -0.016 |
ho | -0.015 |
ig | -0.014 |
Token | cleaned |
Token ID | 20750 |
Feature activation | -7.99e-01 |
Pos logit contributions | |
yes | +0.027 |
goals | +0.024 |
best | +0.023 |
ig | +0.023 |
ho | +0.022 |
Neg logit contributions | |
late | -0.020 |
temperature | -0.019 |
person | -0.019 |
aback | -0.018 |
To | -0.018 |
Token | his |
Token ID | 465 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
To | +0.016 |
ray | +0.015 |
person | +0.014 |
Grand | +0.014 |
Start | +0.014 |
Neg logit contributions | |
best | -0.023 |
goals | -0.022 |
yes | -0.022 |
mm | -0.020 |
course | -0.019 |
Token | sc |
Token ID | 629 |
Feature activation | +4.02e+00 |
Pos logit contributions | |
yes | +0.039 |
goals | +0.036 |
mm | +0.034 |
best | +0.034 |
narrative | +0.033 |
Neg logit contributions | |
over | -0.034 |
temperature | -0.032 |
late | -0.031 |
To | -0.031 |
To | -0.031 |
Token | ooter |
Token ID | 25141 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
yes | +0.112 |
goals | +0.106 |
best | +0.103 |
mm | +0.100 |
narrative | +0.098 |
Neg logit contributions | |
over | -0.081 |
late | -0.076 |
To | -0.076 |
temperature | -0.075 |
aback | -0.074 |
Token | and |
Token ID | 290 |
Feature activation | +2.38e-01 |
Pos logit contributions | |
yes | +0.065 |
goals | +0.062 |
best | +0.061 |
mm | +0.059 |
narrative | +0.056 |
Neg logit contributions | |
over | -0.035 |
late | -0.034 |
To | -0.034 |
temperature | -0.033 |
aback | -0.033 |
Token | then |
Token ID | 788 |
Feature activation | +3.96e-01 |
Pos logit contributions | |
yes | +0.007 |
best | +0.006 |
goals | +0.006 |
mm | +0.006 |
narrative | +0.006 |
Neg logit contributions | |
To | -0.004 |
aback | -0.004 |
temperature | -0.004 |
over | -0.004 |
late | -0.004 |
Token | had |
Token ID | 550 |
Feature activation | -5.93e-01 |
Pos logit contributions | |
yes | +0.011 |
best | +0.011 |
goals | +0.010 |
mm | +0.009 |
narrative | +0.009 |
Neg logit contributions | |
To | -0.008 |
aback | -0.008 |
temperature | -0.008 |
late | -0.008 |
Start | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | -2.82e-01 |
Pos logit contributions | |
person | +0.015 |
ray | +0.015 |
temperature | +0.015 |
any | +0.014 |
net | +0.014 |
Neg logit contributions | |
yes | -0.014 |
best | -0.014 |
goals | -0.013 |
course | -0.011 |
mm | -0.011 |
Token | He |
Token ID | 679 |
Feature activation | +7.23e-01 |
Pos logit contributions | |
yes | +0.008 |
goals | +0.008 |
best | +0.007 |
mm | +0.007 |
ho | +0.007 |
Neg logit contributions | |
yelled | -0.007 |
comm | -0.007 |
temperature | -0.006 |
To | -0.006 |
Grand | -0.006 |
Token | loved |
Token ID | 6151 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
yes | +0.022 |
goals | +0.020 |
best | +0.020 |
mm | +0.018 |
ho | +0.018 |
Neg logit contributions | |
late | -0.013 |
aback | -0.013 |
person | -0.013 |
temperature | -0.013 |
over | -0.013 |
Token | how |
Token ID | 703 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
yes | +0.043 |
mm | +0.041 |
narrative | +0.041 |
goals | +0.041 |
best | +0.040 |
Neg logit contributions | |
over | -0.019 |
into | -0.016 |
To | -0.016 |
yelled | -0.016 |
late | -0.016 |
Token | beautiful |
Token ID | 4950 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
yes | +0.061 |
mm | +0.059 |
narrative | +0.059 |
goals | +0.057 |
owers | +0.056 |
Neg logit contributions | |
over | -0.029 |
late | -0.026 |
To | -0.024 |
into | -0.024 |
yelled | -0.023 |
Token | it |
Token ID | 340 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
yes | +0.039 |
goals | +0.037 |
best | +0.036 |
narrative | +0.035 |
mm | +0.035 |
Neg logit contributions | |
over | -0.019 |
into | -0.016 |
late | -0.016 |
To | -0.016 |
yelled | -0.016 |
Token | was |
Token ID | 373 |
Feature activation | +4.00e+00 |
Pos logit contributions | |
yes | +0.038 |
goals | +0.036 |
best | +0.035 |
mm | +0.034 |
narrative | +0.033 |
Neg logit contributions | |
over | -0.025 |
late | -0.023 |
To | -0.023 |
temperature | -0.023 |
aback | -0.023 |
Token | ! |
Token ID | 0 |
Feature activation | +3.95e-01 |
Pos logit contributions | |
yes | +0.121 |
best | +0.117 |
goals | +0.109 |
ho | +0.100 |
mm | +0.099 |
Neg logit contributions | |
Start | -0.072 |
person | -0.070 |
aback | -0.068 |
To | -0.067 |
Grand | -0.067 |
Token | He |
Token ID | 679 |
Feature activation | +6.43e-01 |
Pos logit contributions | |
goals | +0.011 |
yes | +0.011 |
best | +0.010 |
mm | +0.010 |
Deal | +0.009 |
Neg logit contributions | |
yelled | -0.008 |
temperature | -0.008 |
Start | -0.007 |
To | -0.007 |
Grand | -0.007 |
Token | asked |
Token ID | 1965 |
Feature activation | +7.89e-01 |
Pos logit contributions | |
yes | +0.019 |
goals | +0.018 |
best | +0.017 |
mm | +0.016 |
ho | +0.016 |
Neg logit contributions | |
late | -0.012 |
aback | -0.012 |
over | -0.012 |
temperature | -0.012 |
person | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | -8.27e-01 |
Pos logit contributions | |
yes | +0.024 |
goals | +0.022 |
best | +0.022 |
mm | +0.021 |
narrative | +0.021 |
Neg logit contributions | |
over | -0.014 |
late | -0.014 |
To | -0.014 |
temperature | -0.013 |
aback | -0.013 |
Token | " |
Token ID | 366 |
Feature activation | +9.85e-02 |
Pos logit contributions | |
over | +0.016 |
aback | +0.016 |
temperature | +0.016 |
ray | +0.016 |
To | +0.015 |
Neg logit contributions | |
yes | -0.024 |
best | -0.022 |
goals | -0.021 |
mm | -0.020 |
narrative | -0.019 |
Token | to |
Token ID | 284 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
yes | +0.008 |
best | +0.007 |
goals | +0.007 |
ho | +0.007 |
tweet | +0.006 |
Neg logit contributions | |
person | -0.003 |
ray | -0.003 |
To | -0.003 |
aback | -0.003 |
Grand | -0.003 |
Token | see |
Token ID | 766 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
yes | +0.037 |
goals | +0.034 |
best | +0.034 |
mm | +0.032 |
narrative | +0.031 |
Neg logit contributions | |
over | -0.026 |
To | -0.025 |
late | -0.025 |
person | -0.024 |
aback | -0.024 |
Token | how |
Token ID | 703 |
Feature activation | +1.94e+00 |
Pos logit contributions | |
mm | +0.036 |
yes | +0.035 |
narrative | +0.035 |
goals | +0.034 |
owers | +0.034 |
Neg logit contributions | |
over | -0.018 |
into | -0.014 |
late | -0.014 |
aback | -0.014 |
To | -0.013 |
Token | beautiful |
Token ID | 4950 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
yes | +0.051 |
goals | +0.048 |
best | +0.047 |
mm | +0.045 |
narrative | +0.044 |
Neg logit contributions | |
over | -0.043 |
late | -0.040 |
To | -0.040 |
temperature | -0.040 |
aback | -0.039 |
Token | they |
Token ID | 484 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
yes | +0.034 |
goals | +0.033 |
best | +0.032 |
narrative | +0.031 |
mm | +0.031 |
Neg logit contributions | |
over | -0.019 |
into | -0.017 |
late | -0.017 |
To | -0.016 |
yelled | -0.016 |
Token | looked |
Token ID | 3114 |
Feature activation | +3.99e+00 |
Pos logit contributions | |
yes | +0.005 |
mm | +0.005 |
goals | +0.005 |
narrative | +0.004 |
best | +0.004 |
Neg logit contributions | |
over | -0.009 |
yelled | -0.008 |
asked | -0.008 |
into | -0.008 |
To | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
yes | +0.150 |
mm | +0.149 |
goals | +0.146 |
narrative | +0.144 |
best | +0.139 |
Neg logit contributions | |
over | -0.045 |
into | -0.037 |
late | -0.036 |
towards | -0.032 |
temperature | -0.032 |
Token |
|
Token ID | 198 |
Feature activation | -7.26e-01 |
Pos logit contributions | |
temperature | +0.007 |
plug | +0.007 |
comm | +0.006 |
person | +0.006 |
over | +0.006 |
Neg logit contributions | |
yes | -0.007 |
goals | -0.006 |
best | -0.006 |
mm | -0.006 |
owers | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
temperature | +0.012 |
person | +0.011 |
over | +0.011 |
comm | +0.011 |
aback | +0.011 |
Neg logit contributions | |
yes | -0.022 |
best | -0.021 |
goals | -0.020 |
mm | -0.019 |
course | -0.019 |
Token | He |
Token ID | 1544 |
Feature activation | +4.55e-01 |
Pos logit contributions | |
yes | +0.003 |
goals | +0.003 |
best | +0.003 |
course | +0.003 |
mm | +0.003 |
Neg logit contributions | |
temperature | -0.004 |
aback | -0.003 |
into | -0.003 |
comm | -0.003 |
over | -0.003 |
Token | noticed |
Token ID | 6810 |
Feature activation | -6.69e-02 |
Pos logit contributions | |
yes | +0.014 |
best | +0.013 |
goals | +0.013 |
ho | +0.011 |
ig | +0.011 |
Neg logit contributions | |
person | -0.008 |
aback | -0.008 |
late | -0.008 |
To | -0.008 |
temperature | -0.008 |
Token | hop |
Token ID | 1725 |
Feature activation | -2.64e-01 |
Pos logit contributions | |
yes | +0.008 |
goals | +0.007 |
best | +0.007 |
ho | +0.006 |
tweet | +0.006 |
Neg logit contributions | |
late | -0.008 |
temperature | -0.008 |
person | -0.008 |
Grand | -0.007 |
To | -0.007 |
Token | around |
Token ID | 1088 |
Feature activation | +4.54e-01 |
Pos logit contributions | |
over | +0.003 |
temperature | +0.003 |
late | +0.003 |
To | +0.003 |
person | +0.003 |
Neg logit contributions | |
yes | -0.009 |
goals | -0.009 |
best | -0.009 |
mm | -0.008 |
narrative | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | +2.26e+00 |
Pos logit contributions | |
yes | +0.016 |
mm | +0.016 |
goals | +0.016 |
narrative | +0.015 |
best | +0.015 |
Neg logit contributions | |
over | -0.006 |
into | -0.005 |
late | -0.005 |
To | -0.005 |
towards | -0.005 |
Token | z |
Token ID | 1976 |
Feature activation | +3.26e+00 |
Pos logit contributions | |
goals | +0.049 |
yes | +0.049 |
best | +0.048 |
mm | +0.042 |
narrative | +0.041 |
Neg logit contributions | |
over | -0.057 |
late | -0.056 |
person | -0.056 |
To | -0.056 |
aback | -0.055 |
Token | ig |
Token ID | 328 |
Feature activation | +2.30e+00 |
Pos logit contributions | |
yes | +0.042 |
goals | +0.033 |
narrative | +0.033 |
best | +0.032 |
mm | +0.029 |
Neg logit contributions | |
over | -0.113 |
into | -0.111 |
ray | -0.111 |
Start | -0.111 |
To | -0.110 |
Token | zag |
Token ID | 50183 |
Feature activation | +3.98e+00 |
Pos logit contributions | |
goals | +0.057 |
narrative | +0.057 |
yes | +0.057 |
mm | +0.054 |
nations | +0.050 |
Neg logit contributions | |
over | -0.055 |
into | -0.051 |
towards | -0.048 |
To | -0.048 |
To | -0.047 |
Token | too |
Token ID | 1165 |
Feature activation | +3.95e-01 |
Pos logit contributions | |
yes | +0.117 |
mm | +0.114 |
narrative | +0.108 |
goals | +0.103 |
best | +0.101 |
Neg logit contributions | |
over | -0.080 |
into | -0.075 |
towards | -0.069 |
To | -0.067 |
yelled | -0.067 |
Token | ! |
Token ID | 0 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
yes | +0.015 |
goals | +0.014 |
mm | +0.014 |
narrative | +0.014 |
best | +0.014 |
Neg logit contributions | |
over | -0.004 |
late | -0.004 |
To | -0.003 |
into | -0.003 |
aback | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -2.64e-01 |
Pos logit contributions | |
yes | +0.006 |
goals | +0.006 |
best | +0.006 |
mm | +0.005 |
course | +0.005 |
Neg logit contributions | |
temperature | -0.003 |
Grand | -0.003 |
over | -0.003 |
ray | -0.003 |
comm | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +3.36e-01 |
Pos logit contributions | |
over | +0.003 |
temperature | +0.003 |
late | +0.003 |
person | +0.003 |
aback | +0.003 |
Neg logit contributions | |
yes | -0.009 |
goals | -0.009 |
best | -0.009 |
mm | -0.008 |
narrative | -0.008 |
Token | The |
Token ID | 464 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
yes | +0.010 |
goals | +0.010 |
best | +0.009 |
mm | +0.009 |
narrative | +0.009 |
Neg logit contributions | |
over | -0.006 |
To | -0.006 |
late | -0.006 |
Start | -0.006 |
aback | -0.005 |
Token | wanted |
Token ID | 2227 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
goals | +0.027 |
yes | +0.027 |
mm | +0.026 |
narrative | +0.025 |
best | +0.025 |
Neg logit contributions | |
over | -0.016 |
yelled | -0.014 |
late | -0.014 |
asked | -0.014 |
To | -0.014 |
Token | to |
Token ID | 284 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
yes | +0.004 |
best | +0.004 |
goals | +0.004 |
course | +0.003 |
ho | +0.003 |
Neg logit contributions | |
person | -0.002 |
net | -0.002 |
ray | -0.001 |
Start | -0.001 |
aron | -0.001 |
Token | see |
Token ID | 766 |
Feature activation | +4.96e-01 |
Pos logit contributions | |
yes | +0.003 |
best | +0.003 |
are | +0.003 |
goals | +0.003 |
tweet | +0.003 |
Neg logit contributions | |
late | -0.004 |
ray | -0.003 |
Grand | -0.003 |
aron | -0.003 |
Mrs | -0.003 |
Token | what |
Token ID | 644 |
Feature activation | +6.45e-01 |
Pos logit contributions | |
yes | +0.014 |
goals | +0.013 |
mm | +0.013 |
best | +0.013 |
narrative | +0.013 |
Neg logit contributions | |
over | -0.010 |
late | -0.009 |
aback | -0.009 |
into | -0.009 |
To | -0.009 |
Token | it |
Token ID | 340 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
yes | +0.018 |
goals | +0.017 |
best | +0.017 |
mm | +0.016 |
narrative | +0.015 |
Neg logit contributions | |
over | -0.013 |
late | -0.012 |
temperature | -0.012 |
aback | -0.012 |
person | -0.012 |
Token | was |
Token ID | 373 |
Feature activation | +3.96e+00 |
Pos logit contributions | |
yes | +0.038 |
goals | +0.035 |
best | +0.035 |
mm | +0.033 |
narrative | +0.033 |
Neg logit contributions | |
over | -0.026 |
late | -0.025 |
To | -0.025 |
person | -0.025 |
temperature | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | -2.30e-01 |
Pos logit contributions | |
yes | +0.132 |
best | +0.127 |
goals | +0.123 |
mm | +0.113 |
narrative | +0.112 |
Neg logit contributions | |
person | -0.057 |
To | -0.056 |
Start | -0.056 |
ray | -0.054 |
aback | -0.054 |
Token | |
Token ID | 220 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
temperature | +0.005 |
shave | +0.005 |
late | +0.005 |
plug | +0.005 |
Tooth | +0.005 |
Neg logit contributions | |
yes | -0.006 |
best | -0.005 |
goals | -0.005 |
Deal | -0.004 |
mm | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
yes | +0.039 |
best | +0.038 |
goals | +0.037 |
mm | +0.035 |
owers | +0.033 |
Neg logit contributions | |
temperature | -0.021 |
person | -0.020 |
comm | -0.019 |
over | -0.019 |
late | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | +6.49e-01 |
Pos logit contributions | |
yes | +0.003 |
best | +0.003 |
goals | +0.003 |
mm | +0.003 |
course | +0.003 |
Neg logit contributions | |
temperature | -0.002 |
person | -0.002 |
comm | -0.002 |
shave | -0.002 |
ray | -0.002 |
Token | She |
Token ID | 3347 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
yes | +0.017 |
goals | +0.015 |
best | +0.015 |
Deal | +0.014 |
mm | +0.014 |
Neg logit contributions | |
temperature | -0.015 |
olyn | -0.014 |
over | -0.013 |
aback | -0.013 |
shave | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
yes | +0.027 |
goals | +0.025 |
best | +0.025 |
mm | +0.024 |
narrative | +0.023 |
Neg logit contributions | |
over | -0.016 |
To | -0.015 |
late | -0.015 |
temperature | -0.015 |
person | -0.015 |
Token | said |
Token ID | 531 |
Feature activation | +6.18e-01 |
Pos logit contributions | |
yes | +0.051 |
best | +0.044 |
mm | +0.041 |
tweet | +0.039 |
ho | +0.038 |
Neg logit contributions | |
temperature | -0.031 |
person | -0.029 |
To | -0.029 |
late | -0.029 |
nob | -0.029 |
Token | , |
Token ID | 11 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
goals | +0.019 |
narrative | +0.019 |
mm | +0.017 |
nations | +0.017 |
best | +0.017 |
Neg logit contributions | |
over | -0.011 |
into | -0.010 |
To | -0.010 |
to | -0.010 |
aback | -0.010 |
Token | " |
Token ID | 366 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
yes | +0.004 |
best | +0.004 |
goals | +0.004 |
mm | +0.004 |
ho | +0.004 |
Neg logit contributions | |
aback | -0.003 |
over | -0.003 |
temperature | -0.003 |
late | -0.003 |
ray | -0.003 |
Token | Of |
Token ID | 5189 |
Feature activation | +2.26e+00 |
Pos logit contributions | |
yes | +0.046 |
goals | +0.043 |
best | +0.043 |
mm | +0.041 |
narrative | +0.040 |
Neg logit contributions | |
over | -0.027 |
late | -0.025 |
temperature | -0.025 |
To | -0.025 |
aback | -0.025 |
Token | course |
Token ID | 1781 |
Feature activation | +3.93e+00 |
Pos logit contributions | |
yes | +0.021 |
goals | +0.021 |
best | +0.015 |
owers | +0.012 |
narrative | +0.011 |
Neg logit contributions | |
over | -0.087 |
aback | -0.086 |
Start | -0.085 |
To | -0.085 |
yelled | -0.085 |
Token | you |
Token ID | 345 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
yes | +0.094 |
goals | +0.086 |
best | +0.085 |
mm | +0.082 |
narrative | +0.077 |
Neg logit contributions | |
over | -0.096 |
late | -0.092 |
To | -0.091 |
temperature | -0.091 |
aback | -0.090 |
Token | can |
Token ID | 460 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
yes | +0.024 |
goals | +0.022 |
mm | +0.022 |
best | +0.020 |
narrative | +0.020 |
Neg logit contributions | |
over | -0.043 |
To | -0.040 |
ray | -0.040 |
aback | -0.039 |
temperature | -0.039 |
Token | , |
Token ID | 11 |
Feature activation | +6.16e-01 |
Pos logit contributions | |
yes | +0.047 |
best | +0.043 |
goals | +0.043 |
mm | +0.040 |
ig | +0.038 |
Neg logit contributions | |
late | -0.035 |
person | -0.033 |
over | -0.033 |
temperature | -0.033 |
ray | -0.033 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
goals | +0.023 |
narrative | +0.022 |
yes | +0.022 |
owers | +0.022 |
best | +0.021 |
Neg logit contributions | |
over | -0.007 |
late | -0.006 |
To | -0.005 |
To | -0.005 |
into | -0.005 |
Token | ! |
Token ID | 0 |
Feature activation | +2.00e+00 |
Pos logit contributions | |
yes | +0.004 |
best | +0.004 |
goals | +0.004 |
mm | +0.004 |
ig | +0.003 |
Neg logit contributions | |
Start | -0.002 |
To | -0.002 |
To | -0.002 |
aback | -0.002 |
person | -0.002 |
Token | two |
Token ID | 734 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
mm | +0.031 |
owers | +0.030 |
yes | +0.030 |
goals | +0.027 |
rots | +0.026 |
Neg logit contributions | |
over | -0.024 |
temperature | -0.022 |
asked | -0.022 |
comm | -0.022 |
late | -0.021 |
Token | friends |
Token ID | 2460 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
mm | +0.076 |
owers | +0.071 |
Deal | +0.070 |
yes | +0.070 |
narrative | +0.069 |
Neg logit contributions | |
yelled | -0.043 |
asked | -0.042 |
over | -0.041 |
shouted | -0.041 |
into | -0.040 |
Token | laughed |
Token ID | 13818 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
yes | +0.013 |
goals | +0.013 |
best | +0.012 |
mm | +0.012 |
narrative | +0.011 |
Neg logit contributions | |
over | -0.014 |
To | -0.013 |
late | -0.013 |
yelled | -0.013 |
aback | -0.013 |
Token | at |
Token ID | 379 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
yes | +0.006 |
goals | +0.006 |
mm | +0.006 |
narrative | +0.006 |
best | +0.006 |
Neg logit contributions | |
over | -0.004 |
late | -0.003 |
into | -0.003 |
aback | -0.003 |
To | -0.003 |
Token | their |
Token ID | 511 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
mm | +0.046 |
yes | +0.046 |
narrative | +0.045 |
owers | +0.044 |
goals | +0.043 |
Neg logit contributions | |
over | -0.013 |
into | -0.009 |
aback | -0.008 |
towards | -0.008 |
asked | -0.008 |
Token | cle |
Token ID | 1190 |
Feature activation | +3.92e+00 |
Pos logit contributions | |
mm | +0.046 |
yes | +0.041 |
owers | +0.041 |
narrative | +0.040 |
otion | +0.039 |
Neg logit contributions | |
over | -0.032 |
temperature | -0.032 |
comm | -0.030 |
into | -0.030 |
towards | -0.028 |
Token | vern |
Token ID | 933 |
Feature activation | +5.73e-01 |
Pos logit contributions | |
yes | +0.064 |
goals | +0.064 |
mm | +0.060 |
narrative | +0.057 |
best | +0.055 |
Neg logit contributions | |
person | -0.114 |
Start | -0.114 |
over | -0.113 |
yelled | -0.113 |
ray | -0.111 |
Token | ess |
Token ID | 408 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
yes | +0.015 |
goals | +0.014 |
best | +0.014 |
ho | +0.014 |
mm | +0.013 |
Neg logit contributions | |
over | -0.012 |
To | -0.012 |
Start | -0.012 |
late | -0.012 |
To | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | +8.09e-01 |
Pos logit contributions | |
yes | +0.045 |
goals | +0.044 |
mm | +0.044 |
narrative | +0.043 |
best | +0.041 |
Neg logit contributions | |
over | -0.025 |
into | -0.022 |
temperature | -0.020 |
aback | -0.020 |
ray | -0.020 |
Token | went |
Token ID | 1816 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
yes | +0.024 |
goals | +0.022 |
best | +0.022 |
mm | +0.021 |
narrative | +0.021 |
Neg logit contributions | |
over | -0.015 |
late | -0.014 |
To | -0.014 |
temperature | -0.014 |
aback | -0.014 |
Token | off |
Token ID | 572 |
Feature activation | -3.06e-01 |
Pos logit contributions | |
yes | +0.000 |
best | +0.000 |
goals | +0.000 |
ig | +0.000 |
ho | +0.000 |
Neg logit contributions | |
Grand | -0.000 |
ray | -0.000 |
temperature | -0.000 |
To | -0.000 |
comm | -0.000 |
Token | , |
Token ID | 11 |
Feature activation | +5.73e-01 |
Pos logit contributions | |
yes | +0.050 |
goals | +0.048 |
mm | +0.047 |
narrative | +0.046 |
best | +0.045 |
Neg logit contributions | |
over | -0.029 |
into | -0.024 |
yelled | -0.023 |
late | -0.022 |
towards | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | +3.37e-01 |
Pos logit contributions | |
goals | +0.020 |
mm | +0.019 |
yes | +0.019 |
narrative | +0.019 |
ig | +0.019 |
Neg logit contributions | |
over | -0.009 |
late | -0.007 |
into | -0.007 |
yelled | -0.007 |
asked | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
yes | +0.010 |
goals | +0.010 |
mm | +0.010 |
best | +0.010 |
narrative | +0.010 |
Neg logit contributions | |
over | -0.006 |
yelled | -0.005 |
late | -0.005 |
asked | -0.005 |
To | -0.005 |
Token | merry |
Token ID | 44570 |
Feature activation | +2.73e+00 |
Pos logit contributions | |
yes | +0.036 |
mm | +0.035 |
goals | +0.035 |
narrative | +0.032 |
owers | +0.031 |
Neg logit contributions | |
over | -0.028 |
yelled | -0.027 |
late | -0.027 |
To | -0.027 |
Grand | -0.027 |
Token | - |
Token ID | 12 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
yes | +0.084 |
goals | +0.080 |
best | +0.078 |
mm | +0.076 |
narrative | +0.074 |
Neg logit contributions | |
over | -0.047 |
late | -0.044 |
To | -0.044 |
temperature | -0.043 |
aback | -0.042 |
Token | go |
Token ID | 2188 |
Feature activation | +3.91e+00 |
Pos logit contributions | |
goals | +0.087 |
yes | +0.086 |
best | +0.082 |
narrative | +0.080 |
mm | +0.080 |
Neg logit contributions | |
over | -0.041 |
To | -0.038 |
late | -0.037 |
person | -0.037 |
ray | -0.037 |
Token | - |
Token ID | 12 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
yes | +0.138 |
goals | +0.134 |
narrative | +0.130 |
mm | +0.127 |
best | +0.126 |
Neg logit contributions | |
over | -0.054 |
yelled | -0.047 |
into | -0.046 |
To | -0.046 |
towards | -0.044 |
Token | round |
Token ID | 744 |
Feature activation | +8.04e-01 |
Pos logit contributions | |
yes | +0.062 |
goals | +0.058 |
best | +0.057 |
mm | +0.057 |
ho | +0.055 |
Neg logit contributions | |
temperature | -0.023 |
over | -0.023 |
aback | -0.022 |
late | -0.022 |
To | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
yes | +0.030 |
goals | +0.029 |
best | +0.028 |
mm | +0.028 |
narrative | +0.027 |
Neg logit contributions | |
over | -0.009 |
late | -0.008 |
To | -0.008 |
temperature | -0.008 |
into | -0.007 |
Token | Lily |
Token ID | 20037 |
Feature activation | -5.80e-01 |
Pos logit contributions | |
yes | +0.006 |
goals | +0.005 |
best | +0.005 |
mm | +0.005 |
narrative | +0.005 |
Neg logit contributions | |
person | -0.004 |
aback | -0.004 |
ray | -0.004 |
temperature | -0.004 |
plug | -0.004 |
Token | wanted |
Token ID | 2227 |
Feature activation | +7.45e-01 |
Pos logit contributions | |
aback | +0.013 |
late | +0.013 |
temperature | +0.013 |
person | +0.013 |
To | +0.012 |
Neg logit contributions | |
yes | -0.015 |
best | -0.014 |
goals | -0.014 |
mm | -0.012 |
ig | -0.012 |
Token | went |
Token ID | 1816 |
Feature activation | -7.80e-01 |
Pos logit contributions | |
aback | +0.003 |
person | +0.003 |
net | +0.003 |
late | +0.003 |
To | +0.003 |
Neg logit contributions | |
yes | -0.004 |
best | -0.003 |
goals | -0.003 |
are | -0.003 |
ho | -0.003 |
Token | to |
Token ID | 284 |
Feature activation | -3.83e-01 |
Pos logit contributions | |
person | +0.016 |
net | +0.016 |
Start | +0.016 |
ZE | +0.016 |
Grand | +0.015 |
Neg logit contributions | |
yes | -0.020 |
best | -0.019 |
ig | -0.016 |
angels | -0.016 |
hanging | -0.015 |
Token | see |
Token ID | 766 |
Feature activation | +8.92e-01 |
Pos logit contributions | |
aron | +0.011 |
ave | +0.011 |
person | +0.011 |
ol | +0.010 |
lic | +0.010 |
Neg logit contributions | |
yes | -0.008 |
are | -0.006 |
best | -0.006 |
tweet | -0.006 |
goals | -0.006 |
Token | what |
Token ID | 644 |
Feature activation | +6.43e-01 |
Pos logit contributions | |
yes | +0.023 |
goals | +0.021 |
best | +0.021 |
mm | +0.020 |
narrative | +0.019 |
Neg logit contributions | |
over | -0.020 |
To | -0.019 |
late | -0.019 |
temperature | -0.019 |
aback | -0.019 |
Token | it |
Token ID | 340 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
yes | +0.018 |
goals | +0.017 |
best | +0.017 |
mm | +0.016 |
narrative | +0.016 |
Neg logit contributions | |
over | -0.013 |
late | -0.012 |
temperature | -0.012 |
aback | -0.012 |
To | -0.012 |
Token | was |
Token ID | 373 |
Feature activation | +3.90e+00 |
Pos logit contributions | |
yes | +0.032 |
goals | +0.030 |
best | +0.030 |
mm | +0.029 |
narrative | +0.028 |
Neg logit contributions | |
over | -0.022 |
late | -0.021 |
To | -0.021 |
person | -0.021 |
aback | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | +6.28e-02 |
Pos logit contributions | |
yes | +0.124 |
best | +0.121 |
goals | +0.114 |
mm | +0.107 |
narrative | +0.104 |
Neg logit contributions | |
person | -0.063 |
Start | -0.062 |
To | -0.060 |
aback | -0.059 |
Grand | -0.057 |
Token | It |
Token ID | 632 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
yes | +0.001 |
mm | +0.001 |
best | +0.001 |
goals | +0.001 |
Deal | +0.001 |
Neg logit contributions | |
temperature | -0.002 |
olyn | -0.002 |
ilda | -0.002 |
any | -0.002 |
bo | -0.002 |
Token | was |
Token ID | 373 |
Feature activation | +5.37e-01 |
Pos logit contributions | |
yes | +0.003 |
best | +0.002 |
goals | +0.002 |
ho | +0.002 |
narrative | +0.002 |
Neg logit contributions | |
person | -0.003 |
aback | -0.003 |
To | -0.002 |
To | -0.002 |
ray | -0.002 |
Token | just |
Token ID | 655 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
yes | +0.014 |
best | +0.014 |
goals | +0.013 |
mm | +0.011 |
narrative | +0.011 |
Neg logit contributions | |
ray | -0.012 |
person | -0.012 |
To | -0.012 |
Start | -0.012 |
aback | -0.012 |
Token | his |
Token ID | 465 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
person | +0.028 |
To | +0.028 |
ray | +0.027 |
Start | +0.027 |
aback | +0.026 |
Neg logit contributions | |
yes | -0.034 |
best | -0.033 |
goals | -0.033 |
mm | -0.029 |
narrative | -0.028 |
Token | girl |
Token ID | 2576 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
yes | +0.089 |
mm | +0.088 |
goals | +0.088 |
narrative | +0.082 |
best | +0.081 |
Neg logit contributions | |
over | -0.045 |
aback | -0.040 |
asked | -0.040 |
yelled | -0.039 |
temperature | -0.039 |
Token | said |
Token ID | 531 |
Feature activation | +7.04e-01 |
Pos logit contributions | |
yes | +0.048 |
best | +0.043 |
goals | +0.042 |
mm | +0.040 |
ho | +0.040 |
Neg logit contributions | |
late | -0.025 |
To | -0.023 |
over | -0.023 |
temperature | -0.023 |
aback | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | +2.39e-01 |
Pos logit contributions | |
narrative | +0.024 |
goals | +0.023 |
nations | +0.023 |
owers | +0.022 |
mm | +0.022 |
Neg logit contributions | |
over | -0.013 |
into | -0.011 |
to | -0.011 |
To | -0.010 |
towards | -0.010 |
Token | " |
Token ID | 366 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
yes | +0.007 |
goals | +0.007 |
best | +0.007 |
mm | +0.006 |
narrative | +0.006 |
Neg logit contributions | |
over | -0.004 |
late | -0.004 |
To | -0.004 |
temperature | -0.004 |
aback | -0.004 |
Token | Of |
Token ID | 5189 |
Feature activation | +2.45e+00 |
Pos logit contributions | |
yes | +0.041 |
ho | +0.034 |
Deal | +0.034 |
best | +0.033 |
goals | +0.033 |
Neg logit contributions | |
temperature | -0.034 |
over | -0.033 |
into | -0.031 |
late | -0.031 |
ray | -0.031 |
Token | course |
Token ID | 1781 |
Feature activation | +3.88e+00 |
Pos logit contributions | |
goals | +0.021 |
yes | +0.021 |
best | +0.014 |
mm | +0.012 |
owers | +0.012 |
Neg logit contributions | |
over | -0.097 |
aback | -0.095 |
temperature | -0.093 |
To | -0.093 |
yelled | -0.093 |
Token | , |
Token ID | 11 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
yes | +0.105 |
best | +0.096 |
goals | +0.094 |
mm | +0.091 |
ho | +0.085 |
Neg logit contributions | |
over | -0.083 |
To | -0.081 |
late | -0.081 |
temperature | -0.080 |
ray | -0.079 |
Token | Max |
Token ID | 5436 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
yes | +0.045 |
goals | +0.044 |
best | +0.043 |
mm | +0.042 |
narrative | +0.042 |
Neg logit contributions | |
over | -0.021 |
late | -0.019 |
To | -0.019 |
To | -0.018 |
aback | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
yes | +0.060 |
best | +0.058 |
goals | +0.056 |
mm | +0.052 |
ho | +0.052 |
Neg logit contributions | |
To | -0.030 |
Grand | -0.030 |
over | -0.030 |
late | -0.030 |
person | -0.030 |
Token | I |
Token ID | 314 |
Feature activation | +8.14e-01 |
Pos logit contributions | |
yes | +0.058 |
goals | +0.054 |
best | +0.053 |
mm | +0.052 |
narrative | +0.049 |
Neg logit contributions | |
over | -0.034 |
late | -0.033 |
temperature | -0.032 |
To | -0.032 |
yelled | -0.032 |
Token | love |
Token ID | 1842 |
Feature activation | +7.25e-01 |
Pos logit contributions | |
yes | +0.017 |
goals | +0.017 |
mm | +0.016 |
best | +0.015 |
narrative | +0.015 |
Neg logit contributions | |
over | -0.022 |
aback | -0.021 |
To | -0.021 |
late | -0.021 |
temperature | -0.020 |
Token | Mickey |
Token ID | 27232 |
Feature activation | -8.58e-01 |
Pos logit contributions | |
yes | +0.010 |
goals | +0.009 |
best | +0.009 |
mm | +0.009 |
narrative | +0.009 |
Neg logit contributions | |
over | -0.005 |
late | -0.005 |
To | -0.005 |
temperature | -0.004 |
aback | -0.004 |
Token | what |
Token ID | 644 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
over | +0.018 |
late | +0.016 |
To | +0.016 |
aback | +0.016 |
temperature | +0.016 |
Neg logit contributions | |
yes | -0.024 |
goals | -0.023 |
mm | -0.023 |
best | -0.022 |
narrative | -0.022 |
Token | his |
Token ID | 465 |
Feature activation | +8.10e-01 |
Pos logit contributions | |
yes | +0.033 |
goals | +0.031 |
best | +0.030 |
mm | +0.029 |
narrative | +0.028 |
Neg logit contributions | |
over | -0.023 |
late | -0.021 |
To | -0.021 |
temperature | -0.021 |
aback | -0.021 |
Token | favorite |
Token ID | 4004 |
Feature activation | +5.32e-01 |
Pos logit contributions | |
yes | +0.021 |
goals | +0.020 |
mm | +0.019 |
narrative | +0.019 |
course | +0.018 |
Neg logit contributions | |
over | -0.018 |
temperature | -0.017 |
To | -0.016 |
late | -0.016 |
into | -0.016 |
Token | food |
Token ID | 2057 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
yes | +0.014 |
goals | +0.013 |
mm | +0.013 |
narrative | +0.013 |
best | +0.011 |
Neg logit contributions | |
over | -0.013 |
asked | -0.011 |
temperature | -0.011 |
into | -0.011 |
To | -0.011 |
Token | was |
Token ID | 373 |
Feature activation | +3.88e+00 |
Pos logit contributions | |
yes | +0.007 |
goals | +0.007 |
best | +0.006 |
mm | +0.006 |
narrative | +0.006 |
Neg logit contributions | |
over | -0.004 |
temperature | -0.004 |
To | -0.004 |
late | -0.004 |
yelled | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -8.28e-01 |
Pos logit contributions | |
yes | +0.140 |
goals | +0.134 |
best | +0.130 |
mm | +0.129 |
narrative | +0.126 |
Neg logit contributions | |
over | -0.048 |
late | -0.042 |
To | -0.041 |
temperature | -0.041 |
aback | -0.040 |
Token | Mickey |
Token ID | 27232 |
Feature activation | -7.09e-01 |
Pos logit contributions | |
temperature | +0.017 |
To | +0.016 |
over | +0.016 |
plug | +0.016 |
Start | +0.016 |
Neg logit contributions | |
yes | -0.023 |
best | -0.020 |
goals | -0.020 |
mm | -0.020 |
owers | -0.018 |
Token | said |
Token ID | 531 |
Feature activation | +4.60e-03 |
Pos logit contributions | |
over | +0.010 |
late | +0.010 |
To | +0.010 |
temperature | +0.010 |
person | +0.009 |
Neg logit contributions | |
yes | -0.024 |
best | -0.022 |
goals | -0.022 |
mm | -0.021 |
narrative | -0.021 |
Token | he |
Token ID | 339 |
Feature activation | -3.03e-01 |
Pos logit contributions | |
goals | +0.000 |
narrative | +0.000 |
owers | +0.000 |
nations | +0.000 |
ig | +0.000 |
Neg logit contributions | |
over | -0.000 |
asked | -0.000 |
to | -0.000 |
late | -0.000 |
To | -0.000 |
Token | loved |
Token ID | 6151 |
Feature activation | +8.76e-01 |
Pos logit contributions | |
over | +0.006 |
ray | +0.005 |
yelled | +0.005 |
temperature | +0.005 |
asked | +0.005 |
Neg logit contributions | |
yes | -0.009 |
goals | -0.008 |
mm | -0.008 |
best | -0.008 |
narrative | -0.008 |
Token | closer |
Token ID | 5699 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
person | +0.002 |
Grand | +0.002 |
comm | +0.002 |
any | +0.002 |
To | +0.002 |
Neg logit contributions | |
yes | -0.005 |
goals | -0.005 |
best | -0.005 |
ho | -0.004 |
narrative | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | +3.54e-01 |
Pos logit contributions | |
yes | +0.050 |
goals | +0.048 |
mm | +0.047 |
best | +0.047 |
narrative | +0.047 |
Neg logit contributions | |
over | -0.012 |
into | -0.010 |
late | -0.010 |
To | -0.009 |
temperature | -0.009 |
Token | see |
Token ID | 766 |
Feature activation | +7.52e-01 |
Pos logit contributions | |
yes | +0.011 |
goals | +0.009 |
best | +0.009 |
course | +0.009 |
tweet | +0.009 |
Neg logit contributions | |
person | -0.007 |
Grand | -0.007 |
late | -0.007 |
temperature | -0.006 |
To | -0.006 |
Token | what |
Token ID | 644 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
yes | +0.021 |
goals | +0.021 |
mm | +0.020 |
narrative | +0.020 |
best | +0.020 |
Neg logit contributions | |
over | -0.016 |
ray | -0.013 |
into | -0.013 |
aback | -0.013 |
To | -0.013 |
Token | it |
Token ID | 340 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
yes | +0.035 |
goals | +0.033 |
best | +0.032 |
mm | +0.032 |
narrative | +0.031 |
Neg logit contributions | |
over | -0.021 |
To | -0.019 |
late | -0.019 |
temperature | -0.019 |
aback | -0.019 |
Token | was |
Token ID | 373 |
Feature activation | +3.88e+00 |
Pos logit contributions | |
yes | +0.051 |
goals | +0.049 |
best | +0.047 |
mm | +0.047 |
narrative | +0.045 |
Neg logit contributions | |
over | -0.030 |
To | -0.027 |
late | -0.027 |
temperature | -0.027 |
yelled | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
yes | +0.121 |
best | +0.120 |
goals | +0.114 |
mm | +0.100 |
are | +0.099 |
Neg logit contributions | |
person | -0.068 |
Start | -0.065 |
Grand | -0.064 |
To | -0.063 |
aback | -0.060 |
Token | It |
Token ID | 632 |
Feature activation | +5.92e-01 |
Pos logit contributions | |
Grand | +0.005 |
ilda | +0.005 |
temperature | +0.005 |
comm | +0.005 |
plug | +0.005 |
Neg logit contributions | |
yes | -0.005 |
goals | -0.004 |
best | -0.004 |
ho | -0.004 |
mm | -0.003 |
Token | turned |
Token ID | 2900 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
yes | +0.016 |
best | +0.015 |
goals | +0.015 |
ho | +0.013 |
mm | +0.013 |
Neg logit contributions | |
person | -0.012 |
late | -0.012 |
over | -0.012 |
To | -0.012 |
temperature | -0.012 |
Token | out |
Token ID | 503 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
person | +0.005 |
any | +0.004 |
Grand | +0.004 |
ilda | +0.004 |
ray | +0.004 |
Neg logit contributions | |
yes | -0.008 |
goals | -0.008 |
best | -0.008 |
ho | -0.007 |
tweet | -0.006 |
Token | to |
Token ID | 284 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
person | +0.004 |
temperature | +0.004 |
ray | +0.004 |
any | +0.004 |
To | +0.004 |
Neg logit contributions | |
yes | -0.009 |
goals | -0.008 |
best | -0.008 |
ho | -0.008 |
mm | -0.007 |
Token | did |
Token ID | 750 |
Feature activation | -8.48e-01 |
Pos logit contributions | |
yes | +0.039 |
goals | +0.037 |
best | +0.034 |
mm | +0.033 |
narrative | +0.033 |
Neg logit contributions | |
over | -0.062 |
temperature | -0.060 |
yelled | -0.059 |
To | -0.059 |
late | -0.058 |
Token | Mom |
Token ID | 11254 |
Feature activation | -2.25e+00 |
Pos logit contributions | |
over | +0.018 |
late | +0.017 |
To | +0.017 |
yelled | +0.017 |
temperature | +0.017 |
Neg logit contributions | |
yes | -0.022 |
goals | -0.021 |
best | -0.020 |
narrative | -0.020 |
mm | -0.019 |
Token | and |
Token ID | 290 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
person | +0.041 |
To | +0.041 |
Start | +0.040 |
aback | +0.040 |
To | +0.040 |
Neg logit contributions | |
yes | -0.064 |
best | -0.060 |
mm | -0.059 |
goals | -0.058 |
ho | -0.055 |
Token | Dad |
Token ID | 17415 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
To | +0.034 |
ray | +0.032 |
over | +0.032 |
aback | +0.032 |
person | +0.032 |
Neg logit contributions | |
yes | -0.059 |
goals | -0.051 |
best | -0.051 |
mm | -0.050 |
ho | -0.049 |
Token | put |
Token ID | 1234 |
Feature activation | -2.68e+00 |
Pos logit contributions | |
person | +0.027 |
To | +0.027 |
over | +0.026 |
aback | +0.026 |
Start | +0.026 |
Neg logit contributions | |
yes | -0.048 |
best | -0.045 |
goals | -0.044 |
mm | -0.043 |
ho | -0.041 |
Token | soap |
Token ID | 19533 |
Feature activation | -4.78e+00 |
Pos logit contributions | |
To | +0.066 |
person | +0.066 |
over | +0.065 |
Start | +0.065 |
aback | +0.064 |
Neg logit contributions | |
yes | -0.059 |
best | -0.055 |
goals | -0.055 |
mm | -0.051 |
narrative | -0.048 |
Token | in |
Token ID | 287 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
Start | +0.093 |
person | +0.091 |
To | +0.091 |
comm | +0.091 |
aback | +0.089 |
Neg logit contributions | |
yes | -0.134 |
best | -0.124 |
goals | -0.122 |
mm | -0.116 |
narrative | -0.113 |
Token | the |
Token ID | 262 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
over | +0.022 |
To | +0.021 |
person | +0.021 |
ray | +0.021 |
aback | +0.021 |
Neg logit contributions | |
yes | -0.031 |
goals | -0.029 |
best | -0.029 |
mm | -0.028 |
narrative | -0.026 |
Token | jar |
Token ID | 17379 |
Feature activation | -2.14e+00 |
Pos logit contributions | |
over | +0.005 |
late | +0.004 |
To | +0.004 |
aback | +0.004 |
person | +0.004 |
Neg logit contributions | |
yes | -0.005 |
goals | -0.004 |
best | -0.004 |
mm | -0.004 |
narrative | -0.004 |
Token | ?" |
Token ID | 1701 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
person | +0.039 |
Start | +0.039 |
comm | +0.038 |
aback | +0.037 |
To | +0.035 |
Neg logit contributions | |
yes | -0.063 |
best | -0.060 |
goals | -0.058 |
mm | -0.055 |
narrative | -0.054 |
Token |
|
Token ID | 198 |
Feature activation | +9.39e-01 |
Pos logit contributions | |
over | +0.002 |
late | +0.002 |
To | +0.002 |
To | +0.002 |
Start | +0.002 |
Neg logit contributions | |
yes | -0.009 |
goals | -0.009 |
best | -0.008 |
mm | -0.008 |
narrative | -0.008 |
Token | who |
Token ID | 508 |
Feature activation | +2.69e-01 |
Pos logit contributions | |
over | +0.007 |
late | +0.007 |
To | +0.007 |
temperature | +0.006 |
aback | +0.006 |
Neg logit contributions | |
yes | -0.016 |
goals | -0.015 |
best | -0.015 |
mm | -0.015 |
narrative | -0.014 |
Token | don |
Token ID | 836 |
Feature activation | +1.72e-02 |
Pos logit contributions | |
yes | +0.007 |
goals | +0.007 |
mm | +0.007 |
best | +0.007 |
narrative | +0.006 |
Neg logit contributions | |
over | -0.006 |
yelled | -0.005 |
asked | -0.005 |
To | -0.005 |
ray | -0.005 |
Token | 't |
Token ID | 470 |
Feature activation | -3.13e-01 |
Pos logit contributions | |
yes | +0.000 |
goals | +0.000 |
best | +0.000 |
mm | +0.000 |
narrative | +0.000 |
Neg logit contributions | |
over | -0.000 |
aback | -0.000 |
temperature | -0.000 |
ray | -0.000 |
late | -0.000 |
Token | have |
Token ID | 423 |
Feature activation | -2.23e+00 |
Pos logit contributions | |
comm | +0.008 |
Start | +0.007 |
plug | +0.007 |
aback | +0.007 |
temperature | +0.007 |
Neg logit contributions | |
yes | -0.008 |
best | -0.007 |
are | -0.006 |
goals | -0.006 |
mm | -0.006 |
Token | a |
Token ID | 257 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
person | +0.058 |
Grand | +0.055 |
To | +0.054 |
aback | +0.053 |
ray | +0.053 |
Neg logit contributions | |
yes | -0.051 |
goals | -0.048 |
best | -0.048 |
course | -0.038 |
mm | -0.038 |
Token | comfortable |
Token ID | 6792 |
Feature activation | -4.54e+00 |
Pos logit contributions | |
over | +0.054 |
To | +0.053 |
person | +0.053 |
aback | +0.052 |
late | +0.052 |
Neg logit contributions | |
yes | -0.051 |
best | -0.047 |
goals | -0.046 |
ig | -0.038 |
narrative | -0.037 |
Token | bed |
Token ID | 3996 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
To | +0.098 |
over | +0.097 |
Start | +0.097 |
ray | +0.097 |
aback | +0.095 |
Neg logit contributions | |
yes | -0.121 |
best | -0.112 |
goals | -0.110 |
ig | -0.098 |
mm | -0.092 |
Token | . |
Token ID | 13 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
over | +0.029 |
late | +0.027 |
To | +0.027 |
temperature | +0.026 |
aback | +0.026 |
Neg logit contributions | |
yes | -0.076 |
goals | -0.072 |
best | -0.071 |
mm | -0.069 |
narrative | -0.068 |
Token | Her |
Token ID | 2332 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
temperature | +0.029 |
late | +0.029 |
over | +0.029 |
person | +0.028 |
comm | +0.028 |
Neg logit contributions | |
yes | -0.043 |
goals | -0.041 |
best | -0.040 |
mm | -0.037 |
narrative | -0.037 |
Token | mom |
Token ID | 1995 |
Feature activation | +2.70e-01 |
Pos logit contributions | |
over | +0.022 |
temperature | +0.022 |
late | +0.021 |
To | +0.020 |
To | +0.020 |
Neg logit contributions | |
yes | -0.038 |
goals | -0.036 |
mm | -0.035 |
narrative | -0.034 |
best | -0.034 |
Token | was |
Token ID | 373 |
Feature activation | +3.90e-01 |
Pos logit contributions | |
yes | +0.010 |
best | +0.009 |
goals | +0.009 |
mm | +0.009 |
narrative | +0.008 |
Neg logit contributions | |
over | -0.003 |
late | -0.003 |
To | -0.003 |
temperature | -0.003 |
Start | -0.003 |
Token | fighting |
Token ID | 4330 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
over | +0.039 |
To | +0.038 |
late | +0.038 |
aback | +0.038 |
temperature | +0.037 |
Neg logit contributions | |
yes | -0.027 |
goals | -0.025 |
best | -0.024 |
mm | -0.023 |
narrative | -0.022 |
Token | ? |
Token ID | 30 |
Feature activation | -8.34e-01 |
Pos logit contributions | |
person | +0.024 |
Start | +0.022 |
aback | +0.021 |
comm | +0.020 |
Tooth | +0.019 |
Neg logit contributions | |
best | -0.035 |
are | -0.032 |
goals | -0.031 |
yes | -0.029 |
nations | -0.029 |
Token | Why |
Token ID | 4162 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
temperature | +0.019 |
over | +0.018 |
comm | +0.018 |
aback | +0.018 |
source | +0.018 |
Neg logit contributions | |
yes | -0.020 |
mm | -0.019 |
goals | -0.019 |
best | -0.018 |
narrative | -0.018 |
Token | did |
Token ID | 750 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
yes | +0.014 |
goals | +0.012 |
best | +0.011 |
mm | +0.010 |
narrative | +0.009 |
Neg logit contributions | |
over | -0.047 |
late | -0.045 |
To | -0.045 |
temperature | -0.045 |
aback | -0.044 |
Token | you |
Token ID | 345 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
over | +0.030 |
late | +0.029 |
To | +0.028 |
temperature | +0.028 |
aback | +0.028 |
Neg logit contributions | |
yes | -0.033 |
goals | -0.031 |
best | -0.030 |
mm | -0.029 |
narrative | -0.028 |
Token | waste |
Token ID | 7030 |
Feature activation | -4.40e+00 |
Pos logit contributions | |
person | +0.031 |
aback | +0.030 |
ray | +0.030 |
temperature | +0.030 |
To | +0.030 |
Neg logit contributions | |
yes | -0.032 |
are | -0.027 |
best | -0.027 |
goals | -0.026 |
course | -0.025 |
Token | the |
Token ID | 262 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
person | +0.103 |
any | +0.096 |
To | +0.095 |
ray | +0.092 |
Grand | +0.092 |
Neg logit contributions | |
best | -0.110 |
goals | -0.108 |
yes | -0.108 |
mm | -0.100 |
course | -0.098 |
Token | food |
Token ID | 2057 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
over | +0.023 |
late | +0.022 |
temperature | +0.022 |
To | +0.021 |
into | +0.021 |
Neg logit contributions | |
yes | -0.025 |
goals | -0.024 |
mm | -0.023 |
best | -0.022 |
narrative | -0.022 |
Token | ?" |
Token ID | 1701 |
Feature activation | -7.96e-01 |
Pos logit contributions | |
person | +0.045 |
Start | +0.045 |
aback | +0.044 |
To | +0.043 |
To | +0.043 |
Neg logit contributions | |
yes | -0.070 |
best | -0.069 |
goals | -0.067 |
mm | -0.064 |
narrative | -0.062 |
Token |
|
Token ID | 198 |
Feature activation | +3.10e-01 |
Pos logit contributions | |
temperature | +0.014 |
ray | +0.014 |
over | +0.013 |
person | +0.013 |
any | +0.013 |
Neg logit contributions | |
yes | -0.024 |
goals | -0.023 |
mm | -0.023 |
best | -0.022 |
narrative | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | +6.51e-01 |
Pos logit contributions | |
yes | +0.010 |
goals | +0.009 |
best | +0.009 |
mm | +0.009 |
narrative | +0.009 |
Neg logit contributions | |
temperature | -0.005 |
over | -0.005 |
person | -0.005 |
aback | -0.005 |
any | -0.005 |
Token | ay |
Token ID | 323 |
Feature activation | -9.99e-01 |
Pos logit contributions | |
comm | +0.063 |
Start | +0.062 |
yelled | +0.062 |
aback | +0.062 |
late | +0.061 |
Neg logit contributions | |
mm | -0.039 |
yes | -0.038 |
goals | -0.037 |
best | -0.037 |
owers | -0.034 |
Token | ons |
Token ID | 684 |
Feature activation | -2.68e+00 |
Pos logit contributions | |
over | +0.029 |
late | +0.028 |
aback | +0.028 |
To | +0.028 |
Start | +0.028 |
Neg logit contributions | |
yes | -0.019 |
best | -0.017 |
goals | -0.017 |
mm | -0.017 |
narrative | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
person | +0.054 |
Start | +0.053 |
To | +0.051 |
To | +0.051 |
comm | +0.050 |
Neg logit contributions | |
yes | -0.073 |
best | -0.072 |
goals | -0.067 |
mm | -0.065 |
narrative | -0.063 |
Token | making |
Token ID | 1642 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
over | +0.045 |
late | +0.042 |
To | +0.042 |
temperature | +0.041 |
aback | +0.041 |
Neg logit contributions | |
yes | -0.060 |
goals | -0.057 |
best | -0.055 |
mm | -0.054 |
narrative | -0.052 |
Token | a |
Token ID | 257 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
person | +0.039 |
any | +0.038 |
To | +0.037 |
aback | +0.037 |
ray | +0.036 |
Neg logit contributions | |
best | -0.041 |
goals | -0.041 |
yes | -0.039 |
mm | -0.035 |
course | -0.035 |
Token | mess |
Token ID | 2085 |
Feature activation | -4.26e+00 |
Pos logit contributions | |
over | +0.019 |
aback | +0.019 |
person | +0.018 |
any | +0.018 |
temperature | +0.018 |
Neg logit contributions | |
yes | -0.034 |
best | -0.033 |
goals | -0.033 |
mm | -0.031 |
narrative | -0.031 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
To | +0.077 |
Start | +0.077 |
To | +0.077 |
aback | +0.076 |
person | +0.076 |
Neg logit contributions | |
yes | -0.122 |
best | -0.118 |
goals | -0.113 |
mm | -0.110 |
narrative | -0.107 |
Token |
|
Token ID | 198 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
temperature | +0.015 |
over | +0.015 |
person | +0.014 |
ray | +0.014 |
any | +0.013 |
Neg logit contributions | |
yes | -0.033 |
goals | -0.031 |
best | -0.031 |
mm | -0.030 |
narrative | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | +3.78e-01 |
Pos logit contributions | |
temperature | +0.000 |
person | +0.000 |
any | +0.000 |
over | +0.000 |
wood | +0.000 |
Neg logit contributions | |
yes | -0.000 |
goals | -0.000 |
best | -0.000 |
mm | -0.000 |
narrative | -0.000 |
Token | The |
Token ID | 464 |
Feature activation | -4.16e-01 |
Pos logit contributions | |
yes | +0.010 |
goals | +0.010 |
narrative | +0.009 |
best | +0.009 |
mm | +0.009 |
Neg logit contributions | |
temperature | -0.008 |
over | -0.008 |
person | -0.007 |
aback | -0.007 |
source | -0.007 |
Token | boy |
Token ID | 2933 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
over | +0.007 |
late | +0.007 |
temperature | +0.006 |
ray | +0.006 |
Start | +0.006 |
Neg logit contributions | |
yes | -0.013 |
mm | -0.012 |
goals | -0.012 |
owers | -0.012 |
best | -0.011 |
Token | Where |
Token ID | 6350 |
Feature activation | +7.76e-01 |
Pos logit contributions | |
temperature | +0.012 |
over | +0.012 |
shave | +0.012 |
up | +0.011 |
Grand | +0.011 |
Neg logit contributions | |
goals | -0.014 |
yes | -0.014 |
mm | -0.013 |
best | -0.013 |
itt | -0.012 |
Token | had |
Token ID | 550 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
goals | +0.022 |
yes | +0.020 |
mm | +0.020 |
owers | +0.020 |
narrative | +0.019 |
Neg logit contributions | |
over | -0.017 |
yelled | -0.015 |
asked | -0.015 |
ray | -0.015 |
To | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | -8.36e-01 |
Pos logit contributions | |
over | +0.041 |
person | +0.040 |
ray | +0.040 |
aback | +0.040 |
To | +0.039 |
Neg logit contributions | |
yes | -0.056 |
best | -0.052 |
goals | -0.051 |
mm | -0.050 |
course | -0.047 |
Token | salt |
Token ID | 8268 |
Feature activation | -3.42e+00 |
Pos logit contributions | |
over | +0.021 |
person | +0.020 |
aback | +0.020 |
To | +0.020 |
To | +0.019 |
Neg logit contributions | |
yes | -0.020 |
best | -0.019 |
goals | -0.018 |
mm | -0.017 |
narrative | -0.016 |
Token | sh |
Token ID | 427 |
Feature activation | -9.50e-01 |
Pos logit contributions | |
over | +0.062 |
To | +0.060 |
late | +0.060 |
aback | +0.060 |
Start | +0.060 |
Neg logit contributions | |
yes | -0.100 |
best | -0.093 |
goals | -0.092 |
mm | -0.089 |
narrative | -0.085 |
Token | aker |
Token ID | 3110 |
Feature activation | -4.19e+00 |
Pos logit contributions | |
over | +0.018 |
late | +0.016 |
To | +0.016 |
temperature | +0.016 |
aback | +0.016 |
Neg logit contributions | |
yes | -0.028 |
goals | -0.027 |
best | -0.026 |
mm | -0.026 |
narrative | -0.025 |
Token | gone |
Token ID | 3750 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
Grand | +0.077 |
person | +0.076 |
aback | +0.075 |
Start | +0.074 |
To | +0.074 |
Neg logit contributions | |
yes | -0.121 |
best | -0.111 |
mm | -0.101 |
goals | -0.101 |
course | -0.100 |
Token | ? |
Token ID | 30 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
person | +0.036 |
Start | +0.036 |
To | +0.036 |
Grand | +0.035 |
over | +0.035 |
Neg logit contributions | |
yes | -0.067 |
best | -0.064 |
goals | -0.062 |
mm | -0.059 |
narrative | -0.057 |
Token | Now |
Token ID | 2735 |
Feature activation | -3.14e-01 |
Pos logit contributions | |
over | +0.020 |
temperature | +0.020 |
comm | +0.018 |
into | +0.017 |
any | +0.017 |
Neg logit contributions | |
yes | -0.035 |
best | -0.031 |
mm | -0.031 |
goals | -0.031 |
itt | -0.029 |
Token | he |
Token ID | 339 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
late | +0.007 |
Grand | +0.007 |
To | +0.007 |
over | +0.007 |
person | +0.007 |
Neg logit contributions | |
yes | -0.008 |
best | -0.007 |
goals | -0.007 |
mm | -0.007 |
course | -0.006 |
Token | would |
Token ID | 561 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
over | +0.026 |
late | +0.025 |
To | +0.025 |
aback | +0.025 |
temperature | +0.025 |
Neg logit contributions | |
yes | -0.029 |
goals | -0.028 |
best | -0.027 |
mm | -0.026 |
narrative | -0.025 |
Token | said |
Token ID | 531 |
Feature activation | +5.27e-01 |
Pos logit contributions | |
yes | +0.034 |
best | +0.031 |
goals | +0.031 |
mm | +0.028 |
ho | +0.027 |
Neg logit contributions | |
over | -0.018 |
late | -0.018 |
person | -0.018 |
aback | -0.018 |
To | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +8.75e-01 |
Pos logit contributions | |
yes | +0.019 |
goals | +0.018 |
best | +0.018 |
mm | +0.017 |
narrative | +0.016 |
Neg logit contributions | |
over | -0.006 |
To | -0.006 |
late | -0.006 |
temperature | -0.006 |
aback | -0.006 |
Token | " |
Token ID | 366 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
yes | +0.022 |
best | +0.018 |
goals | +0.018 |
mm | +0.017 |
ho | +0.015 |
Neg logit contributions | |
temperature | -0.021 |
any | -0.020 |
comm | -0.020 |
over | -0.020 |
shave | -0.020 |
Token | Where |
Token ID | 8496 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
yes | +0.035 |
best | +0.034 |
goals | +0.033 |
mm | +0.032 |
narrative | +0.031 |
Neg logit contributions | |
To | -0.017 |
late | -0.017 |
over | -0.017 |
Start | -0.016 |
Grand | -0.016 |
Token | is |
Token ID | 318 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
owers | +0.035 |
ir | +0.033 |
ho | +0.032 |
goals | +0.032 |
ement | +0.032 |
Neg logit contributions | |
ray | -0.027 |
over | -0.027 |
to | -0.025 |
asked | -0.024 |
yelled | -0.024 |
Token | dust |
Token ID | 8977 |
Feature activation | -4.18e+00 |
Pos logit contributions | |
over | +0.021 |
late | +0.020 |
To | +0.020 |
temperature | +0.020 |
aback | +0.019 |
Neg logit contributions | |
yes | -0.033 |
goals | -0.031 |
best | -0.030 |
mm | -0.030 |
narrative | -0.029 |
Token | ?" |
Token ID | 1701 |
Feature activation | -4.55e-01 |
Pos logit contributions | |
To | +0.064 |
over | +0.064 |
Start | +0.064 |
person | +0.064 |
aback | +0.063 |
Neg logit contributions | |
yes | -0.132 |
best | -0.124 |
goals | -0.124 |
mm | -0.119 |
narrative | -0.114 |
Token |
|
Token ID | 198 |
Feature activation | +7.55e-01 |
Pos logit contributions | |
temperature | +0.008 |
over | +0.007 |
person | +0.007 |
ray | +0.007 |
any | +0.007 |
Neg logit contributions | |
yes | -0.015 |
best | -0.013 |
goals | -0.013 |
mm | -0.013 |
narrative | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
yes | +0.023 |
goals | +0.022 |
best | +0.022 |
mm | +0.021 |
narrative | +0.020 |
Neg logit contributions | |
temperature | -0.013 |
person | -0.012 |
aback | -0.012 |
over | -0.012 |
any | -0.011 |
Token | L |
Token ID | 43 |
Feature activation | +3.03e-01 |
Pos logit contributions | |
yes | +0.032 |
goals | +0.030 |
best | +0.029 |
mm | +0.029 |
narrative | +0.028 |
Neg logit contributions | |
temperature | -0.021 |
over | -0.021 |
aback | -0.020 |
person | -0.019 |
into | -0.019 |
Token | ily |
Token ID | 813 |
Feature activation | -7.32e-01 |
Pos logit contributions | |
yes | +0.007 |
goals | +0.007 |
best | +0.006 |
mm | +0.006 |
ho | +0.006 |
Neg logit contributions | |
over | -0.008 |
late | -0.007 |
To | -0.007 |
ray | -0.007 |
Start | -0.007 |
Token | as |
Token ID | 355 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
person | +0.018 |
any | +0.018 |
Grand | +0.017 |
aback | +0.017 |
gent | +0.017 |
Neg logit contributions | |
best | -0.025 |
yes | -0.025 |
goals | -0.024 |
mm | -0.021 |
course | -0.020 |
Token | solid |
Token ID | 4735 |
Feature activation | -3.10e+00 |
Pos logit contributions | |
over | +0.034 |
late | +0.032 |
into | +0.031 |
To | +0.031 |
To | +0.030 |
Neg logit contributions | |
yes | -0.051 |
goals | -0.049 |
mm | -0.048 |
narrative | -0.048 |
ho | -0.047 |
Token | as |
Token ID | 355 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
Grand | +0.060 |
ray | +0.059 |
Mrs | +0.057 |
Start | +0.056 |
To | +0.054 |
Neg logit contributions | |
yes | -0.092 |
goals | -0.085 |
best | -0.083 |
mm | -0.075 |
course | -0.075 |
Token | a |
Token ID | 257 |
Feature activation | -5.44e-01 |
Pos logit contributions | |
over | +0.025 |
late | +0.023 |
To | +0.023 |
temperature | +0.023 |
aback | +0.022 |
Neg logit contributions | |
yes | -0.037 |
goals | -0.036 |
best | -0.034 |
mm | -0.034 |
narrative | -0.033 |
Token | table |
Token ID | 3084 |
Feature activation | -3.41e+00 |
Pos logit contributions | |
over | +0.013 |
aback | +0.012 |
late | +0.012 |
To | +0.012 |
ray | +0.012 |
Neg logit contributions | |
yes | -0.014 |
best | -0.012 |
goals | -0.012 |
mm | -0.011 |
ho | -0.011 |
Token | cloth |
Token ID | 44905 |
Feature activation | -4.17e+00 |
Pos logit contributions | |
Start | +0.058 |
aback | +0.057 |
ray | +0.055 |
comm | +0.055 |
To | +0.055 |
Neg logit contributions | |
yes | -0.106 |
best | -0.098 |
goals | -0.096 |
mm | -0.088 |
narrative | -0.088 |
Token | ! |
Token ID | 0 |
Feature activation | -5.53e-01 |
Pos logit contributions | |
Start | +0.079 |
Grand | +0.079 |
plug | +0.077 |
person | +0.076 |
ray | +0.075 |
Neg logit contributions | |
yes | -0.124 |
best | -0.110 |
ho | -0.103 |
are | -0.101 |
course | -0.101 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
temperature | +0.010 |
olyn | +0.010 |
yelled | +0.010 |
over | +0.010 |
ZE | +0.010 |
Neg logit contributions | |
yes | -0.015 |
goals | -0.014 |
best | -0.014 |
mm | -0.014 |
course | -0.012 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
person | +0.036 |
temperature | +0.035 |
over | +0.035 |
aback | +0.034 |
comm | +0.034 |
Neg logit contributions | |
yes | -0.069 |
best | -0.067 |
goals | -0.066 |
mm | -0.063 |
owers | -0.061 |
Token | upon |
Token ID | 2402 |
Feature activation | +7.06e-02 |
Pos logit contributions | |
yes | +0.043 |
goals | +0.043 |
mm | +0.041 |
narrative | +0.040 |
Deal | +0.039 |
Neg logit contributions | |
over | -0.035 |
late | -0.032 |
into | -0.031 |
asked | -0.031 |
To | -0.030 |
Token | a |
Token ID | 257 |
Feature activation | -5.75e-02 |
Pos logit contributions | |
ho | +0.003 |
goals | +0.003 |
narrative | +0.003 |
mm | +0.003 |
ig | +0.002 |
Neg logit contributions | |
late | -0.001 |
over | -0.001 |
To | -0.001 |
Start | -0.001 |
To | -0.001 |
Token | was |
Token ID | 373 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
late | +0.007 |
To | +0.006 |
over | +0.006 |
aback | +0.006 |
person | +0.006 |
Neg logit contributions | |
yes | -0.010 |
best | -0.009 |
goals | -0.009 |
mm | -0.008 |
ig | -0.008 |
Token | disappointed |
Token ID | 11679 |
Feature activation | -6.60e-01 |
Pos logit contributions | |
yes | +0.002 |
best | +0.002 |
goals | +0.002 |
narrative | +0.001 |
mm | +0.001 |
Neg logit contributions | |
ray | -0.003 |
To | -0.003 |
person | -0.003 |
net | -0.003 |
aback | -0.003 |
Token | but |
Token ID | 475 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
person | +0.017 |
plug | +0.016 |
aback | +0.016 |
net | +0.016 |
ZE | +0.016 |
Neg logit contributions | |
best | -0.016 |
yes | -0.015 |
goals | -0.015 |
mm | -0.013 |
narrative | -0.012 |
Token | she |
Token ID | 673 |
Feature activation | -9.04e-01 |
Pos logit contributions | |
over | +0.027 |
late | +0.026 |
To | +0.025 |
temperature | +0.025 |
yelled | +0.025 |
Neg logit contributions | |
yes | -0.026 |
goals | -0.024 |
best | -0.023 |
mm | -0.023 |
narrative | -0.022 |
Token | knew |
Token ID | 2993 |
Feature activation | -9.84e-01 |
Pos logit contributions | |
over | +0.020 |
late | +0.019 |
aback | +0.019 |
To | +0.019 |
To | +0.018 |
Neg logit contributions | |
yes | -0.024 |
goals | -0.022 |
best | -0.022 |
mm | -0.020 |
narrative | -0.020 |
Token | her |
Token ID | 607 |
Feature activation | -4.17e+00 |
Pos logit contributions | |
over | +0.020 |
late | +0.018 |
temperature | +0.018 |
To | +0.018 |
To | +0.018 |
Neg logit contributions | |
yes | -0.027 |
goals | -0.026 |
mm | -0.025 |
best | -0.025 |
narrative | -0.025 |
Token | daddy |
Token ID | 35822 |
Feature activation | -9.12e-01 |
Pos logit contributions | |
over | +0.097 |
aback | +0.096 |
To | +0.095 |
Start | +0.093 |
person | +0.091 |
Neg logit contributions | |
yes | -0.106 |
best | -0.105 |
goals | -0.101 |
narrative | -0.089 |
mm | -0.087 |
Token | was |
Token ID | 373 |
Feature activation | +3.83e-02 |
Pos logit contributions | |
over | +0.020 |
temperature | +0.018 |
yelled | +0.017 |
asked | +0.017 |
ray | +0.017 |
Neg logit contributions | |
goals | -0.024 |
yes | -0.024 |
mm | -0.023 |
narrative | -0.023 |
best | -0.022 |
Token | right |
Token ID | 826 |
Feature activation | +5.88e-02 |
Pos logit contributions | |
yes | +0.001 |
best | +0.001 |
goals | +0.001 |
narrative | +0.001 |
course | +0.001 |
Neg logit contributions | |
ray | -0.001 |
To | -0.001 |
person | -0.001 |
net | -0.001 |
To | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -8.71e-01 |
Pos logit contributions | |
yes | +0.002 |
goals | +0.002 |
mm | +0.002 |
best | +0.002 |
narrative | +0.002 |
Neg logit contributions | |
over | -0.001 |
temperature | -0.001 |
late | -0.001 |
into | -0.000 |
aback | -0.000 |
Token | She |
Token ID | 1375 |
Feature activation | -5.76e-01 |
Pos logit contributions | |
plug | +0.023 |
yelled | +0.023 |
shave | +0.023 |
comm | +0.023 |
rid | +0.023 |
Neg logit contributions | |
goals | -0.017 |
yes | -0.017 |
best | -0.015 |
mm | -0.014 |
ho | -0.012 |
Token | you |
Token ID | 345 |
Feature activation | -5.37e-01 |
Pos logit contributions | |
goals | +0.083 |
yes | +0.080 |
owers | +0.080 |
ement | +0.079 |
narrative | +0.078 |
Neg logit contributions | |
me | -0.033 |
To | -0.032 |
over | -0.031 |
yelled | -0.031 |
To | -0.030 |
Token | want |
Token ID | 765 |
Feature activation | -4.69e-01 |
Pos logit contributions | |
over | +0.011 |
late | +0.010 |
yelled | +0.010 |
To | +0.010 |
aback | +0.010 |
Neg logit contributions | |
yes | -0.015 |
goals | -0.014 |
best | -0.014 |
mm | -0.014 |
narrative | -0.013 |
Token | to |
Token ID | 284 |
Feature activation | -3.29e-01 |
Pos logit contributions | |
person | +0.008 |
aron | +0.008 |
any | +0.008 |
aback | +0.007 |
net | +0.007 |
Neg logit contributions | |
yes | -0.015 |
best | -0.014 |
course | -0.014 |
goals | -0.014 |
mm | -0.013 |
Token | buy |
Token ID | 2822 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
late | +0.007 |
aback | +0.007 |
ray | +0.007 |
person | +0.007 |
temperature | +0.007 |
Neg logit contributions | |
yes | -0.009 |
best | -0.008 |
goals | -0.008 |
mm | -0.008 |
course | -0.007 |
Token | some |
Token ID | 617 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
over | +0.036 |
To | +0.036 |
late | +0.035 |
person | +0.035 |
temperature | +0.035 |
Neg logit contributions | |
yes | -0.049 |
best | -0.046 |
goals | -0.046 |
mm | -0.044 |
course | -0.041 |
Token | laundry |
Token ID | 25724 |
Feature activation | -4.08e+00 |
Pos logit contributions | |
aback | +0.035 |
over | +0.035 |
person | +0.034 |
Start | +0.034 |
late | +0.034 |
Neg logit contributions | |
yes | -0.035 |
best | -0.033 |
goals | -0.033 |
mm | -0.031 |
narrative | -0.029 |
Token | ? |
Token ID | 30 |
Feature activation | -4.62e-01 |
Pos logit contributions | |
Start | +0.086 |
comm | +0.085 |
person | +0.080 |
aback | +0.079 |
yelled | +0.078 |
Neg logit contributions | |
yes | -0.121 |
best | -0.115 |
goals | -0.103 |
mm | -0.102 |
course | -0.094 |
Token | I |
Token ID | 314 |
Feature activation | -7.12e-01 |
Pos logit contributions | |
over | +0.009 |
temperature | +0.009 |
comm | +0.009 |
aback | +0.009 |
yelled | +0.009 |
Neg logit contributions | |
yes | -0.013 |
goals | -0.012 |
mm | -0.011 |
best | -0.011 |
narrative | -0.010 |
Token | have |
Token ID | 423 |
Feature activation | -2.48e+00 |
Pos logit contributions | |
over | +0.017 |
late | +0.016 |
To | +0.016 |
temperature | +0.016 |
aback | +0.016 |
Neg logit contributions | |
yes | -0.018 |
goals | -0.016 |
best | -0.016 |
mm | -0.015 |
narrative | -0.015 |
Token | a |
Token ID | 257 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
over | +0.054 |
person | +0.052 |
aback | +0.051 |
To | +0.051 |
late | +0.051 |
Neg logit contributions | |
yes | -0.066 |
goals | -0.061 |
best | -0.061 |
mm | -0.056 |
narrative | -0.054 |
Token | shirt |
Token ID | 10147 |
Feature activation | -8.95e-01 |
Pos logit contributions | |
over | +0.039 |
late | +0.037 |
To | +0.037 |
aback | +0.036 |
temperature | +0.036 |
Neg logit contributions | |
yes | -0.039 |
goals | -0.036 |
best | -0.036 |
mm | -0.033 |
narrative | -0.032 |
Token | Why |
Token ID | 5195 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
yes | +0.024 |
goals | +0.023 |
best | +0.023 |
mm | +0.022 |
ig | +0.022 |
Neg logit contributions | |
To | -0.012 |
over | -0.012 |
late | -0.012 |
Grand | -0.012 |
Start | -0.011 |
Token | did |
Token ID | 750 |
Feature activation | -6.53e-01 |
Pos logit contributions | |
yes | +0.050 |
goals | +0.048 |
best | +0.044 |
mm | +0.043 |
narrative | +0.043 |
Neg logit contributions | |
over | -0.065 |
temperature | -0.061 |
late | -0.061 |
yelled | -0.060 |
To | -0.060 |
Token | you |
Token ID | 345 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
over | +0.016 |
aback | +0.015 |
late | +0.015 |
temperature | +0.015 |
To | +0.015 |
Neg logit contributions | |
yes | -0.016 |
goals | -0.015 |
best | -0.014 |
mm | -0.014 |
narrative | -0.013 |
Token | take |
Token ID | 1011 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
person | +0.025 |
aback | +0.025 |
ray | +0.025 |
To | +0.024 |
late | +0.024 |
Neg logit contributions | |
yes | -0.028 |
best | -0.023 |
are | -0.023 |
goals | -0.023 |
ir | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
person | +0.051 |
any | +0.049 |
To | +0.048 |
Grand | +0.047 |
rev | +0.045 |
Neg logit contributions | |
goals | -0.049 |
best | -0.049 |
yes | -0.046 |
course | -0.042 |
ig | -0.038 |
Token | soap |
Token ID | 19533 |
Feature activation | -4.03e+00 |
Pos logit contributions | |
person | +0.044 |
To | +0.042 |
Grand | +0.042 |
aback | +0.042 |
ray | +0.042 |
Neg logit contributions | |
best | -0.041 |
yes | -0.041 |
goals | -0.039 |
narrative | -0.033 |
mm | -0.032 |
Token | without |
Token ID | 1231 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
Start | +0.075 |
person | +0.075 |
comm | +0.073 |
To | +0.071 |
To | +0.070 |
Neg logit contributions | |
yes | -0.117 |
best | -0.111 |
goals | -0.104 |
course | -0.099 |
ho | -0.099 |
Token | asking |
Token ID | 4737 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
person | +0.041 |
To | +0.039 |
aback | +0.039 |
ray | +0.039 |
Start | +0.037 |
Neg logit contributions | |
best | -0.041 |
yes | -0.041 |
goals | -0.038 |
mm | -0.034 |
course | -0.034 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
Start | +0.034 |
comm | +0.032 |
To | +0.032 |
person | +0.031 |
aback | +0.031 |
Neg logit contributions | |
yes | -0.045 |
best | -0.042 |
goals | -0.038 |
mm | -0.037 |
course | -0.035 |
Token | Daddy |
Token ID | 26926 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
over | +0.020 |
To | +0.018 |
late | +0.018 |
aback | +0.018 |
temperature | +0.018 |
Neg logit contributions | |
yes | -0.051 |
goals | -0.049 |
best | -0.048 |
mm | -0.047 |
narrative | -0.045 |
Token | asked |
Token ID | 1965 |
Feature activation | +3.68e-01 |
Pos logit contributions | |
To | +0.022 |
To | +0.021 |
aback | +0.020 |
late | +0.019 |
icol | +0.019 |
Neg logit contributions | |
yes | -0.052 |
best | -0.045 |
ement | -0.043 |
course | -0.041 |
ho | -0.041 |
Token | peel |
Token ID | 31738 |
Feature activation | -2.82e+00 |
Pos logit contributions | |
over | +0.045 |
into | +0.040 |
late | +0.040 |
temperature | +0.039 |
To | +0.038 |
Neg logit contributions | |
yes | -0.088 |
goals | -0.086 |
narrative | -0.081 |
best | -0.081 |
mm | -0.080 |
Token | ings |
Token ID | 654 |
Feature activation | -2.82e+00 |
Pos logit contributions | |
over | +0.046 |
person | +0.046 |
To | +0.045 |
Start | +0.045 |
aback | +0.045 |
Neg logit contributions | |
yes | -0.087 |
goals | -0.081 |
best | -0.081 |
mm | -0.078 |
narrative | -0.074 |
Token | and |
Token ID | 290 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
comm | +0.048 |
person | +0.048 |
Start | +0.047 |
aback | +0.047 |
any | +0.046 |
Neg logit contributions | |
yes | -0.088 |
best | -0.078 |
mm | -0.078 |
goals | -0.075 |
course | -0.073 |
Token | grow |
Token ID | 1663 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
aback | +0.022 |
over | +0.021 |
person | +0.021 |
Start | +0.021 |
late | +0.021 |
Neg logit contributions | |
yes | -0.036 |
goals | -0.032 |
best | -0.032 |
mm | -0.031 |
ho | -0.029 |
Token | more |
Token ID | 517 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
over | +0.030 |
late | +0.027 |
To | +0.027 |
temperature | +0.027 |
aback | +0.026 |
Neg logit contributions | |
yes | -0.056 |
goals | -0.053 |
best | -0.052 |
mm | -0.051 |
narrative | -0.050 |
Token | potatoes |
Token ID | 18821 |
Feature activation | -3.99e+00 |
Pos logit contributions | |
person | +0.045 |
aback | +0.045 |
Start | +0.045 |
To | +0.044 |
ray | +0.044 |
Neg logit contributions | |
yes | -0.038 |
best | -0.038 |
goals | -0.037 |
mm | -0.036 |
narrative | -0.030 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
person | +0.079 |
Start | +0.079 |
any | +0.077 |
aback | +0.076 |
ray | +0.075 |
Neg logit contributions | |
yes | -0.118 |
best | -0.111 |
goals | -0.104 |
mm | -0.101 |
narrative | -0.094 |
Token | they |
Token ID | 484 |
Feature activation | -6.97e-01 |
Pos logit contributions | |
over | +0.002 |
late | +0.002 |
asked | +0.002 |
yelled | +0.002 |
To | +0.002 |
Neg logit contributions | |
yes | -0.004 |
goals | -0.004 |
best | -0.004 |
mm | -0.004 |
narrative | -0.004 |
Token | asked |
Token ID | 1965 |
Feature activation | +9.45e-01 |
Pos logit contributions | |
ray | +0.006 |
To | +0.006 |
over | +0.006 |
aback | +0.005 |
person | +0.005 |
Neg logit contributions | |
yes | -0.029 |
best | -0.026 |
goals | -0.026 |
narrative | -0.025 |
mm | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +9.30e-01 |
Pos logit contributions | |
mm | +0.039 |
goals | +0.039 |
narrative | +0.038 |
yes | +0.038 |
nations | +0.037 |
Neg logit contributions | |
over | -0.007 |
into | -0.005 |
asked | -0.004 |
late | -0.004 |
temperature | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | +9.61e-01 |
Pos logit contributions | |
yes | +0.034 |
goals | +0.032 |
best | +0.032 |
mm | +0.031 |
narrative | +0.030 |
Neg logit contributions | |
over | -0.011 |
temperature | -0.010 |
ray | -0.010 |
aback | -0.010 |
late | -0.010 |
Token | healthy |
Token ID | 5448 |
Feature activation | -7.64e-01 |
Pos logit contributions | |
over | +0.012 |
late | +0.011 |
To | +0.011 |
temperature | +0.011 |
yelled | +0.011 |
Neg logit contributions | |
yes | -0.010 |
goals | -0.010 |
best | -0.009 |
mm | -0.009 |
narrative | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +3.51e-01 |
Pos logit contributions | |
Start | +0.013 |
To | +0.012 |
To | +0.012 |
person | +0.012 |
over | +0.012 |
Neg logit contributions | |
yes | -0.024 |
best | -0.023 |
goals | -0.023 |
mm | -0.021 |
narrative | -0.021 |
Token | Do |
Token ID | 2141 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
yes | +0.010 |
goals | +0.009 |
mm | +0.009 |
best | +0.009 |
narrative | +0.008 |
Neg logit contributions | |
over | -0.007 |
temperature | -0.007 |
aback | -0.007 |
late | -0.007 |
yelled | -0.007 |
Token | you |
Token ID | 345 |
Feature activation | -9.58e-01 |
Pos logit contributions | |
goals | +0.067 |
ement | +0.066 |
yes | +0.063 |
ig | +0.062 |
ilot | +0.062 |
Neg logit contributions | |
me | -0.025 |
To | -0.024 |
To | -0.023 |
Grand | -0.022 |
asked | -0.021 |
Token | like |
Token ID | 588 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
over | +0.022 |
To | +0.021 |
late | +0.021 |
temperature | +0.021 |
person | +0.020 |
Neg logit contributions | |
yes | -0.025 |
goals | -0.023 |
best | -0.022 |
mm | -0.021 |
narrative | -0.021 |
Token | rice |
Token ID | 11464 |
Feature activation | -3.99e+00 |
Pos logit contributions | |
aback | +0.031 |
person | +0.031 |
ray | +0.030 |
Start | +0.030 |
any | +0.030 |
Neg logit contributions | |
yes | -0.034 |
goals | -0.032 |
best | -0.031 |
mm | -0.030 |
course | -0.028 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
Start | +0.060 |
aback | +0.059 |
person | +0.059 |
over | +0.059 |
To | +0.058 |
Neg logit contributions | |
yes | -0.131 |
best | -0.124 |
goals | -0.121 |
mm | -0.119 |
narrative | -0.113 |
Token | Grand |
Token ID | 5675 |
Feature activation | -6.56e-01 |
Pos logit contributions | |
ray | +0.024 |
temperature | +0.024 |
aback | +0.024 |
over | +0.024 |
person | +0.023 |
Neg logit contributions | |
yes | -0.026 |
best | -0.022 |
mm | -0.022 |
goals | -0.022 |
narrative | -0.020 |
Token | ma |
Token ID | 2611 |
Feature activation | -6.92e-01 |
Pos logit contributions | |
over | +0.018 |
late | +0.018 |
To | +0.018 |
aback | +0.017 |
into | +0.017 |
Neg logit contributions | |
yes | -0.013 |
mm | -0.012 |
goals | -0.012 |
best | -0.012 |
narrative | -0.011 |
Token | asks |
Token ID | 7893 |
Feature activation | +5.71e-02 |
Pos logit contributions | |
To | +0.016 |
late | +0.015 |
person | +0.015 |
aback | +0.015 |
temperature | +0.015 |
Neg logit contributions | |
yes | -0.016 |
goals | -0.015 |
best | -0.014 |
tur | -0.014 |
ho | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -4.48e-01 |
Pos logit contributions | |
yes | +0.002 |
best | +0.002 |
goals | +0.002 |
mm | +0.002 |
ho | +0.002 |
Neg logit contributions | |
To | -0.001 |
over | -0.001 |
Start | -0.001 |
aback | -0.001 |
ray | -0.001 |
Token | Why |
Token ID | 4162 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
temperature | +0.010 |
over | +0.010 |
aback | +0.010 |
comm | +0.010 |
To | +0.010 |
Neg logit contributions | |
yes | -0.013 |
goals | -0.012 |
mm | -0.012 |
best | -0.012 |
narrative | -0.012 |
Token | are |
Token ID | 389 |
Feature activation | -3.30e-01 |
Pos logit contributions | |
yes | +0.020 |
goals | +0.018 |
best | +0.016 |
mm | +0.015 |
narrative | +0.014 |
Neg logit contributions | |
over | -0.052 |
late | -0.049 |
To | -0.049 |
temperature | -0.049 |
ray | -0.049 |
Token | you |
Token ID | 345 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
over | +0.007 |
late | +0.007 |
yelled | +0.006 |
To | +0.006 |
into | +0.006 |
Neg logit contributions | |
yes | -0.009 |
goals | -0.008 |
best | -0.008 |
mm | -0.008 |
narrative | -0.008 |
Token | making |
Token ID | 1642 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
over | +0.046 |
To | +0.046 |
aback | +0.045 |
person | +0.045 |
late | +0.045 |
Neg logit contributions | |
yes | -0.030 |
goals | -0.027 |
best | -0.027 |
mm | -0.024 |
narrative | -0.023 |
Token | a |
Token ID | 257 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
person | +0.041 |
To | +0.040 |
any | +0.040 |
ray | +0.039 |
aback | +0.039 |
Neg logit contributions | |
best | -0.040 |
goals | -0.039 |
yes | -0.037 |
mm | -0.033 |
course | -0.032 |
Token | mess |
Token ID | 2085 |
Feature activation | -3.95e+00 |
Pos logit contributions | |
aback | +0.019 |
person | +0.018 |
any | +0.018 |
over | +0.017 |
ray | +0.017 |
Neg logit contributions | |
best | -0.032 |
goals | -0.031 |
yes | -0.030 |
narrative | -0.029 |
nations | -0.028 |
Token | ?" |
Token ID | 1701 |
Feature activation | -4.94e-01 |
Pos logit contributions | |
To | +0.070 |
Start | +0.070 |
To | +0.069 |
aback | +0.069 |
person | +0.068 |
Neg logit contributions | |
yes | -0.117 |
best | -0.113 |
goals | -0.109 |
mm | -0.106 |
narrative | -0.102 |
Token |
|
Token ID | 198 |
Feature activation | +6.50e-01 |
Pos logit contributions | |
temperature | +0.008 |
over | +0.007 |
person | +0.007 |
aback | +0.007 |
ray | +0.007 |
Neg logit contributions | |
yes | -0.016 |
goals | -0.015 |
best | -0.015 |
mm | -0.015 |
narrative | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
yes | +0.019 |
goals | +0.019 |
mm | +0.018 |
best | +0.018 |
narrative | +0.017 |
Neg logit contributions | |
temperature | -0.011 |
person | -0.011 |
aback | -0.011 |
any | -0.011 |
over | -0.011 |
Token | Tom |
Token ID | 13787 |
Feature activation | -6.44e-01 |
Pos logit contributions | |
yes | +0.028 |
goals | +0.027 |
mm | +0.026 |
best | +0.026 |
narrative | +0.025 |
Neg logit contributions | |
over | -0.022 |
temperature | -0.022 |
aback | -0.022 |
into | -0.021 |
person | -0.021 |
Token | and |
Token ID | 290 |
Feature activation | +4.82e-01 |
Pos logit contributions | |
person | +0.014 |
To | +0.014 |
late | +0.013 |
aback | +0.013 |
To | +0.013 |
Neg logit contributions | |
goals | -0.016 |
yes | -0.016 |
best | -0.015 |
narrative | -0.014 |
mm | -0.014 |
Token | are |
Token ID | 389 |
Feature activation | -8.92e-02 |
Pos logit contributions | |
yes | +0.027 |
goals | +0.025 |
best | +0.021 |
mm | +0.020 |
narrative | +0.020 |
Neg logit contributions | |
over | -0.073 |
yelled | -0.069 |
To | -0.069 |
ray | -0.069 |
temperature | -0.069 |
Token | you |
Token ID | 345 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
over | +0.002 |
late | +0.002 |
yelled | +0.002 |
into | +0.002 |
To | +0.002 |
Neg logit contributions | |
yes | -0.002 |
goals | -0.002 |
best | -0.002 |
narrative | -0.002 |
mm | -0.002 |
Token | making |
Token ID | 1642 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
over | +0.031 |
To | +0.031 |
aback | +0.030 |
person | +0.030 |
ray | +0.030 |
Neg logit contributions | |
yes | -0.023 |
best | -0.021 |
goals | -0.021 |
mm | -0.019 |
narrative | -0.018 |
Token | such |
Token ID | 884 |
Feature activation | -5.63e-01 |
Pos logit contributions | |
person | +0.036 |
To | +0.036 |
any | +0.035 |
aback | +0.034 |
ray | +0.034 |
Neg logit contributions | |
best | -0.037 |
yes | -0.036 |
goals | -0.035 |
mm | -0.032 |
course | -0.031 |
Token | a |
Token ID | 257 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
over | +0.013 |
To | +0.012 |
person | +0.012 |
ray | +0.012 |
To | +0.012 |
Neg logit contributions | |
yes | -0.014 |
goals | -0.014 |
best | -0.014 |
mm | -0.012 |
narrative | -0.012 |
Token | mess |
Token ID | 2085 |
Feature activation | -3.93e+00 |
Pos logit contributions | |
over | +0.018 |
aback | +0.018 |
To | +0.017 |
person | +0.016 |
To | +0.016 |
Neg logit contributions | |
best | -0.037 |
yes | -0.036 |
goals | -0.035 |
mm | -0.033 |
narrative | -0.033 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
To | +0.076 |
To | +0.076 |
Start | +0.075 |
aback | +0.073 |
person | +0.073 |
Neg logit contributions | |
yes | -0.111 |
best | -0.108 |
mm | -0.102 |
goals | -0.100 |
narrative | -0.097 |
Token | she |
Token ID | 673 |
Feature activation | -2.74e-01 |
Pos logit contributions | |
over | +0.022 |
late | +0.021 |
To | +0.021 |
temperature | +0.020 |
aback | +0.020 |
Neg logit contributions | |
yes | -0.035 |
goals | -0.033 |
best | -0.032 |
mm | -0.031 |
narrative | -0.031 |
Token | asks |
Token ID | 7893 |
Feature activation | +5.05e-01 |
Pos logit contributions | |
person | +0.005 |
To | +0.005 |
aback | +0.005 |
ray | +0.005 |
To | +0.005 |
Neg logit contributions | |
goals | -0.008 |
yes | -0.008 |
best | -0.007 |
narrative | -0.007 |
mm | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | -5.29e-01 |
Pos logit contributions | |
yes | +0.020 |
goals | +0.020 |
mm | +0.020 |
best | +0.019 |
narrative | +0.019 |
Neg logit contributions | |
over | -0.005 |
into | -0.004 |
late | -0.004 |
temperature | -0.003 |
yelled | -0.003 |
Token | looks |
Token ID | 3073 |
Feature activation | +7.83e-01 |
Pos logit contributions | |
To | +0.010 |
person | +0.009 |
temperature | +0.009 |
over | +0.009 |
late | +0.009 |
Neg logit contributions | |
yes | -0.016 |
goals | -0.014 |
best | -0.014 |
mm | -0.013 |
narrative | -0.013 |
Token | Tim |
Token ID | 5045 |
Feature activation | +5.21e-01 |
Pos logit contributions | |
over | +0.014 |
late | +0.011 |
Grand | +0.011 |
Grand | +0.011 |
asked | +0.010 |
Neg logit contributions | |
owers | -0.032 |
goals | -0.031 |
ho | -0.031 |
yes | -0.031 |
mm | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | -4.11e-02 |
Pos logit contributions | |
yes | +0.016 |
best | +0.016 |
goals | +0.016 |
mm | +0.015 |
narrative | +0.014 |
Neg logit contributions | |
To | -0.008 |
Start | -0.008 |
To | -0.008 |
over | -0.008 |
person | -0.008 |
Token | You |
Token ID | 921 |
Feature activation | -2.30e-01 |
Pos logit contributions | |
over | +0.001 |
temperature | +0.001 |
yelled | +0.001 |
aback | +0.001 |
To | +0.001 |
Neg logit contributions | |
yes | -0.001 |
goals | -0.001 |
best | -0.001 |
mm | -0.001 |
narrative | -0.001 |
Token | are |
Token ID | 389 |
Feature activation | -6.53e-02 |
Pos logit contributions | |
over | +0.008 |
late | +0.008 |
To | +0.008 |
temperature | +0.007 |
aback | +0.007 |
Neg logit contributions | |
yes | -0.003 |
goals | -0.003 |
best | -0.003 |
mm | -0.003 |
narrative | -0.002 |
Token | a |
Token ID | 257 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
To | +0.001 |
over | +0.001 |
temperature | +0.001 |
person | +0.001 |
aback | +0.001 |
Neg logit contributions | |
yes | -0.002 |
best | -0.002 |
goals | -0.002 |
mm | -0.001 |
narrative | -0.001 |
Token | smart |
Token ID | 4451 |
Feature activation | -3.92e+00 |
Pos logit contributions | |
over | +0.043 |
late | +0.042 |
To | +0.041 |
Start | +0.040 |
Grand | +0.040 |
Neg logit contributions | |
yes | -0.035 |
goals | -0.034 |
mm | -0.033 |
owers | -0.031 |
ho | -0.030 |
Token | and |
Token ID | 290 |
Feature activation | -7.35e-01 |
Pos logit contributions | |
over | +0.082 |
late | +0.078 |
To | +0.077 |
temperature | +0.076 |
aback | +0.076 |
Neg logit contributions | |
yes | -0.106 |
goals | -0.100 |
best | -0.097 |
mm | -0.094 |
narrative | -0.092 |
Token | brave |
Token ID | 14802 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
over | +0.017 |
late | +0.017 |
To | +0.016 |
aback | +0.016 |
Start | +0.016 |
Neg logit contributions | |
yes | -0.018 |
goals | -0.017 |
mm | -0.016 |
best | -0.016 |
narrative | -0.016 |
Token | boy |
Token ID | 2933 |
Feature activation | +6.92e-01 |
Pos logit contributions | |
over | +0.054 |
late | +0.051 |
To | +0.051 |
temperature | +0.050 |
aback | +0.050 |
Neg logit contributions | |
yes | -0.064 |
goals | -0.060 |
best | -0.058 |
mm | -0.057 |
narrative | -0.055 |
Token | ." |
Token ID | 526 |
Feature activation | -7.86e-01 |
Pos logit contributions | |
yes | +0.026 |
goals | +0.025 |
best | +0.024 |
mm | +0.024 |
narrative | +0.023 |
Neg logit contributions | |
over | -0.008 |
late | -0.007 |
To | -0.007 |
temperature | -0.006 |
aback | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
over | +0.008 |
late | +0.007 |
To | +0.007 |
temperature | +0.007 |
aback | +0.007 |
Neg logit contributions | |
yes | -0.030 |
goals | -0.029 |
best | -0.028 |
mm | -0.028 |
narrative | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | +7.01e-01 |
Pos logit contributions | |
yes | +0.031 |
best | +0.030 |
goals | +0.029 |
mm | +0.029 |
ho | +0.027 |
Neg logit contributions | |
To | -0.013 |
over | -0.013 |
Start | -0.013 |
aback | -0.013 |
late | -0.012 |
Token | Can |
Token ID | 1680 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
yes | +0.021 |
goals | +0.020 |
best | +0.020 |
mm | +0.019 |
narrative | +0.019 |
Neg logit contributions | |
over | -0.012 |
late | -0.012 |
To | -0.012 |
temperature | -0.011 |
aback | -0.011 |
Token | I |
Token ID | 314 |
Feature activation | -3.43e-01 |
Pos logit contributions | |
goals | +0.072 |
best | +0.066 |
yes | +0.065 |
narrative | +0.064 |
nations | +0.063 |
Neg logit contributions | |
over | -0.038 |
To | -0.036 |
To | -0.035 |
asked | -0.034 |
late | -0.034 |
Token | have |
Token ID | 423 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
over | +0.007 |
late | +0.007 |
aback | +0.007 |
person | +0.007 |
Start | +0.007 |
Neg logit contributions | |
yes | -0.009 |
best | -0.009 |
goals | -0.008 |
mm | -0.008 |
narrative | -0.008 |
Token | some |
Token ID | 617 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
person | +0.050 |
any | +0.050 |
aback | +0.048 |
over | +0.048 |
net | +0.047 |
Neg logit contributions | |
yes | -0.050 |
best | -0.045 |
mm | -0.044 |
goals | -0.042 |
course | -0.041 |
Token | butter |
Token ID | 9215 |
Feature activation | -3.91e+00 |
Pos logit contributions | |
person | +0.043 |
aback | +0.043 |
over | +0.043 |
Start | +0.043 |
any | +0.043 |
Neg logit contributions | |
yes | -0.049 |
best | -0.046 |
mm | -0.044 |
goals | -0.044 |
course | -0.040 |
Token | on |
Token ID | 319 |
Feature activation | -6.68e-01 |
Pos logit contributions | |
over | +0.060 |
Start | +0.058 |
To | +0.058 |
aback | +0.058 |
person | +0.057 |
Neg logit contributions | |
yes | -0.126 |
best | -0.118 |
goals | -0.117 |
mm | -0.115 |
narrative | -0.109 |
Token | it |
Token ID | 340 |
Feature activation | -7.38e-01 |
Pos logit contributions | |
over | +0.012 |
late | +0.011 |
To | +0.011 |
temperature | +0.011 |
aback | +0.011 |
Neg logit contributions | |
yes | -0.020 |
goals | -0.019 |
best | -0.019 |
mm | -0.018 |
narrative | -0.018 |
Token | ?" |
Token ID | 1701 |
Feature activation | -3.46e-01 |
Pos logit contributions | |
person | +0.012 |
Start | +0.012 |
aback | +0.011 |
comm | +0.011 |
To | +0.011 |
Neg logit contributions | |
yes | -0.024 |
best | -0.022 |
goals | -0.022 |
mm | -0.021 |
narrative | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | +3.46e-01 |
Pos logit contributions | |
over | +0.004 |
late | +0.003 |
To | +0.003 |
temperature | +0.003 |
aback | +0.003 |
Neg logit contributions | |
yes | -0.013 |
goals | -0.012 |
best | -0.012 |
mm | -0.012 |
narrative | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | +6.00e-01 |
Pos logit contributions | |
yes | +0.011 |
goals | +0.010 |
best | +0.010 |
mm | +0.010 |
ho | +0.009 |
Neg logit contributions | |
temperature | -0.005 |
aback | -0.005 |
over | -0.005 |
person | -0.005 |
ray | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +3.14e-01 |
Pos logit contributions | |
yes | +0.005 |
goals | +0.005 |
best | +0.005 |
mm | +0.004 |
narrative | +0.004 |
Neg logit contributions | |
over | -0.003 |
late | -0.003 |
To | -0.003 |
temperature | -0.003 |
aback | -0.003 |
Token | " |
Token ID | 366 |
Feature activation | +3.91e-01 |
Pos logit contributions | |
yes | +0.007 |
goals | +0.006 |
best | +0.006 |
ho | +0.005 |
mm | +0.005 |
Neg logit contributions | |
ray | -0.008 |
temperature | -0.008 |
aback | -0.008 |
any | -0.008 |
ZE | -0.008 |
Token | How |
Token ID | 2437 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
yes | +0.011 |
goals | +0.011 |
best | +0.011 |
mm | +0.010 |
narrative | +0.010 |
Neg logit contributions | |
over | -0.007 |
To | -0.007 |
late | -0.007 |
Grand | -0.007 |
Start | -0.007 |
Token | did |
Token ID | 750 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
owers | +0.037 |
ho | +0.036 |
goals | +0.036 |
narrative | +0.036 |
yes | +0.035 |
Neg logit contributions | |
late | -0.028 |
over | -0.027 |
To | -0.026 |
ray | -0.026 |
To | -0.025 |
Token | you |
Token ID | 345 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
over | +0.039 |
late | +0.037 |
aback | +0.037 |
To | +0.037 |
temperature | +0.036 |
Neg logit contributions | |
yes | -0.042 |
goals | -0.039 |
best | -0.038 |
mm | -0.037 |
narrative | -0.035 |
Token | sleep |
Token ID | 3993 |
Feature activation | -3.89e+00 |
Pos logit contributions | |
over | +0.030 |
late | +0.029 |
person | +0.029 |
To | +0.029 |
aback | +0.029 |
Neg logit contributions | |
yes | -0.044 |
goals | -0.040 |
best | -0.040 |
mm | -0.038 |
narrative | -0.036 |
Token | ?" |
Token ID | 1701 |
Feature activation | -8.41e-01 |
Pos logit contributions | |
over | +0.068 |
late | +0.063 |
To | +0.063 |
temperature | +0.062 |
aback | +0.062 |
Neg logit contributions | |
yes | -0.119 |
goals | -0.113 |
best | -0.111 |
mm | -0.107 |
narrative | -0.104 |
Token |
|
Token ID | 198 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
ray | +0.012 |
temperature | +0.012 |
aback | +0.012 |
over | +0.012 |
person | +0.012 |
Neg logit contributions | |
yes | -0.028 |
goals | -0.026 |
best | -0.026 |
mm | -0.025 |
narrative | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | +4.35e-01 |
Pos logit contributions | |
yes | +0.008 |
goals | +0.008 |
best | +0.008 |
mm | +0.008 |
narrative | +0.007 |
Neg logit contributions | |
temperature | -0.005 |
aback | -0.005 |
person | -0.005 |
over | -0.004 |
any | -0.004 |
Token | " |
Token ID | 1 |
Feature activation | -4.44e-01 |
Pos logit contributions | |
yes | +0.011 |
goals | +0.010 |
best | +0.010 |
mm | +0.009 |
narrative | +0.009 |
Neg logit contributions | |
over | -0.010 |
temperature | -0.010 |
aback | -0.010 |
comm | -0.009 |
person | -0.009 |
Token | Very |
Token ID | 16371 |
Feature activation | +7.88e-01 |
Pos logit contributions | |
over | +0.007 |
To | +0.007 |
late | +0.007 |
Grand | +0.007 |
Start | +0.007 |
Neg logit contributions | |
yes | -0.014 |
goals | -0.014 |
best | -0.013 |
mm | -0.013 |
narrative | -0.012 |
Token | girl |
Token ID | 2576 |
Feature activation | -2.63e-01 |
Pos logit contributions | |
over | +0.010 |
late | +0.010 |
temperature | +0.010 |
aback | +0.009 |
into | +0.009 |
Neg logit contributions | |
yes | -0.020 |
goals | -0.019 |
mm | -0.019 |
best | -0.018 |
ho | -0.018 |
Token | waited |
Token ID | 13488 |
Feature activation | +3.99e-01 |
Pos logit contributions | |
over | +0.005 |
late | +0.005 |
To | +0.005 |
aback | +0.005 |
person | +0.005 |
Neg logit contributions | |
yes | -0.007 |
best | -0.007 |
goals | -0.007 |
mm | -0.006 |
ho | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | -6.98e-01 |
Pos logit contributions | |
yes | +0.012 |
goals | +0.011 |
mm | +0.011 |
best | +0.011 |
narrative | +0.011 |
Neg logit contributions | |
over | -0.007 |
late | -0.007 |
into | -0.007 |
temperature | -0.007 |
To | -0.007 |
Token | waited |
Token ID | 13488 |
Feature activation | +5.07e-01 |
Pos logit contributions | |
over | +0.011 |
person | +0.011 |
To | +0.011 |
temperature | +0.011 |
aback | +0.010 |
Neg logit contributions | |
yes | -0.022 |
goals | -0.021 |
best | -0.020 |
mm | -0.020 |
narrative | -0.019 |
Token | but |
Token ID | 475 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
yes | +0.014 |
goals | +0.013 |
best | +0.013 |
mm | +0.013 |
narrative | +0.012 |
Neg logit contributions | |
over | -0.011 |
late | -0.010 |
temperature | -0.010 |
To | -0.010 |
into | -0.010 |
Token | her |
Token ID | 607 |
Feature activation | -3.88e+00 |
Pos logit contributions | |
over | +0.026 |
late | +0.024 |
into | +0.024 |
temperature | +0.023 |
To | +0.023 |
Neg logit contributions | |
goals | -0.038 |
yes | -0.038 |
mm | -0.037 |
best | -0.036 |
narrative | -0.035 |
Token | dad |
Token ID | 9955 |
Feature activation | -6.34e-02 |
Pos logit contributions | |
over | +0.079 |
To | +0.076 |
aback | +0.076 |
Start | +0.075 |
late | +0.075 |
Neg logit contributions | |
yes | -0.107 |
goals | -0.103 |
best | -0.102 |
narrative | -0.093 |
mm | -0.093 |
Token | never |
Token ID | 1239 |
Feature activation | +4.15e-01 |
Pos logit contributions | |
over | +0.001 |
yelled | +0.001 |
asked | +0.001 |
temperature | +0.001 |
into | +0.001 |
Neg logit contributions | |
goals | -0.002 |
mm | -0.002 |
yes | -0.002 |
narrative | -0.002 |
best | -0.002 |
Token | came |
Token ID | 1625 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
yes | +0.011 |
best | +0.010 |
tweet | +0.009 |
ho | +0.009 |
are | +0.009 |
Neg logit contributions | |
person | -0.009 |
net | -0.009 |
To | -0.009 |
Start | -0.009 |
To | -0.009 |
Token | back |
Token ID | 736 |
Feature activation | -5.81e-01 |
Pos logit contributions | |
ray | +0.020 |
person | +0.020 |
Start | +0.020 |
To | +0.020 |
aback | +0.019 |
Neg logit contributions | |
yes | -0.043 |
best | -0.041 |
goals | -0.040 |
ho | -0.036 |
narrative | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
over | +0.006 |
into | +0.005 |
late | +0.005 |
temperature | +0.005 |
towards | +0.005 |
Neg logit contributions | |
yes | -0.022 |
goals | -0.022 |
mm | -0.022 |
narrative | -0.021 |
best | -0.021 |
Token | you |
Token ID | 345 |
Feature activation | -9.14e-01 |
Pos logit contributions | |
ilot | +0.073 |
goals | +0.072 |
narrative | +0.072 |
ement | +0.072 |
yes | +0.072 |
Neg logit contributions | |
me | -0.028 |
To | -0.026 |
To | -0.024 |
over | -0.024 |
Grand | -0.023 |
Token | like |
Token ID | 588 |
Feature activation | -5.80e-01 |
Pos logit contributions | |
over | +0.018 |
late | +0.017 |
To | +0.017 |
temperature | +0.017 |
aback | +0.017 |
Neg logit contributions | |
yes | -0.026 |
goals | -0.024 |
best | -0.024 |
mm | -0.023 |
narrative | -0.023 |
Token | d |
Token ID | 288 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
over | +0.011 |
late | +0.011 |
To | +0.011 |
temperature | +0.011 |
aback | +0.011 |
Neg logit contributions | |
yes | -0.016 |
goals | -0.015 |
best | -0.015 |
mm | -0.015 |
narrative | -0.014 |
Token | aff |
Token ID | 2001 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
late | +0.048 |
over | +0.048 |
To | +0.048 |
person | +0.048 |
temperature | +0.048 |
Neg logit contributions | |
yes | -0.028 |
goals | -0.027 |
best | -0.025 |
mm | -0.025 |
ho | -0.024 |
Token | od |
Token ID | 375 |
Feature activation | -7.19e-01 |
Pos logit contributions | |
over | +0.064 |
into | +0.059 |
asked | +0.059 |
yelled | +0.058 |
To | +0.057 |
Neg logit contributions | |
yes | -0.052 |
goals | -0.048 |
narrative | -0.045 |
best | -0.044 |
Deal | -0.043 |
Token | ils |
Token ID | 4487 |
Feature activation | -3.87e+00 |
Pos logit contributions | |
over | +0.018 |
yelled | +0.017 |
ray | +0.017 |
into | +0.017 |
asked | +0.017 |
Neg logit contributions | |
yes | -0.017 |
goals | -0.015 |
narrative | -0.014 |
best | -0.014 |
mm | -0.013 |
Token | ?" |
Token ID | 1701 |
Feature activation | -8.27e-01 |
Pos logit contributions | |
over | +0.058 |
aback | +0.057 |
person | +0.057 |
To | +0.057 |
late | +0.057 |
Neg logit contributions | |
yes | -0.126 |
goals | -0.118 |
best | -0.116 |
mm | -0.112 |
narrative | -0.108 |
Token | Ben |
Token ID | 3932 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
over | +0.016 |
To | +0.015 |
temperature | +0.015 |
aback | +0.014 |
late | +0.014 |
Neg logit contributions | |
yes | -0.024 |
goals | -0.023 |
best | -0.023 |
mm | -0.022 |
narrative | -0.021 |
Token | asks |
Token ID | 7893 |
Feature activation | +4.29e-01 |
Pos logit contributions | |
goals | +0.003 |
yes | +0.003 |
best | +0.003 |
mm | +0.003 |
narrative | +0.003 |
Neg logit contributions | |
person | -0.002 |
To | -0.002 |
late | -0.002 |
temperature | -0.002 |
ray | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | +5.13e-01 |
Pos logit contributions | |
yes | +0.015 |
goals | +0.015 |
best | +0.014 |
mm | +0.014 |
narrative | +0.014 |
Neg logit contributions | |
over | -0.005 |
To | -0.005 |
late | -0.005 |
aback | -0.005 |
temperature | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +6.41e-01 |
Pos logit contributions | |
yes | +0.016 |
goals | +0.015 |
best | +0.014 |
mm | +0.014 |
ho | +0.013 |
Neg logit contributions | |
any | -0.009 |
temperature | -0.009 |
person | -0.009 |
ilda | -0.009 |
ray | -0.009 |
Token | " |
Token ID | 1 |
Feature activation | -6.49e-03 |
Pos logit contributions | |
yes | +0.005 |
goals | +0.004 |
best | +0.004 |
mm | +0.004 |
narrative | +0.004 |
Neg logit contributions | |
temperature | -0.005 |
aback | -0.005 |
person | -0.005 |
over | -0.005 |
comm | -0.004 |
Token | Can |
Token ID | 6090 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
over | +0.000 |
To | +0.000 |
late | +0.000 |
temperature | +0.000 |
Start | +0.000 |
Neg logit contributions | |
yes | -0.000 |
goals | -0.000 |
best | -0.000 |
mm | -0.000 |
narrative | -0.000 |
Token | we |
Token ID | 356 |
Feature activation | -7.37e-01 |
Pos logit contributions | |
goals | +0.051 |
yes | +0.049 |
best | +0.048 |
narrative | +0.046 |
owers | +0.045 |
Neg logit contributions | |
over | -0.041 |
To | -0.038 |
To | -0.038 |
late | -0.037 |
into | -0.037 |
Token | use |
Token ID | 779 |
Feature activation | -2.62e+00 |
Pos logit contributions | |
person | +0.016 |
aback | +0.016 |
comm | +0.015 |
late | +0.015 |
Start | +0.015 |
Neg logit contributions | |
yes | -0.021 |
best | -0.019 |
goals | -0.018 |
mm | -0.017 |
are | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | -7.77e-01 |
Pos logit contributions | |
person | +0.050 |
To | +0.048 |
over | +0.047 |
Start | +0.047 |
ray | +0.046 |
Neg logit contributions | |
yes | -0.076 |
goals | -0.072 |
best | -0.071 |
mm | -0.067 |
course | -0.065 |
Token | soap |
Token ID | 19533 |
Feature activation | -3.87e+00 |
Pos logit contributions | |
over | +0.017 |
person | +0.017 |
late | +0.017 |
To | +0.017 |
aback | +0.017 |
Neg logit contributions | |
yes | -0.020 |
best | -0.019 |
goals | -0.019 |
mm | -0.017 |
narrative | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
Start | +0.065 |
To | +0.064 |
person | +0.064 |
Grand | +0.063 |
comm | +0.062 |
Neg logit contributions | |
yes | -0.119 |
best | -0.112 |
goals | -0.109 |
mm | -0.104 |
ho | -0.102 |
Token | Lily |
Token ID | 20037 |
Feature activation | -2.30e+00 |
Pos logit contributions | |
over | +0.026 |
late | +0.024 |
temperature | +0.023 |
into | +0.022 |
To | +0.022 |
Neg logit contributions | |
yes | -0.048 |
goals | -0.048 |
best | -0.046 |
mm | -0.045 |
narrative | -0.045 |
Token | ?" |
Token ID | 1701 |
Feature activation | -9.66e-01 |
Pos logit contributions | |
To | +0.035 |
Start | +0.035 |
over | +0.035 |
aback | +0.035 |
To | +0.035 |
Neg logit contributions | |
yes | -0.074 |
best | -0.070 |
goals | -0.069 |
mm | -0.066 |
narrative | -0.064 |
Token | Ben |
Token ID | 3932 |
Feature activation | -5.65e-01 |
Pos logit contributions | |
over | +0.017 |
To | +0.016 |
late | +0.016 |
temperature | +0.016 |
aback | +0.016 |
Neg logit contributions | |
yes | -0.029 |
goals | -0.028 |
best | -0.027 |
mm | -0.026 |
narrative | -0.026 |
Token | asks |
Token ID | 7893 |
Feature activation | +2.91e-01 |
Pos logit contributions | |
person | +0.012 |
To | +0.012 |
ray | +0.011 |
har | +0.011 |
To | +0.011 |
Neg logit contributions | |
goals | -0.016 |
yes | -0.015 |
best | -0.014 |
narrative | -0.013 |
nations | -0.012 |
Token | " |
Token ID | 366 |
Feature activation | +8.70e-01 |
Pos logit contributions | |
yes | +0.025 |
best | +0.022 |
goals | +0.022 |
ho | +0.021 |
mm | +0.021 |
Neg logit contributions | |
temperature | -0.021 |
ray | -0.021 |
over | -0.020 |
late | -0.020 |
aback | -0.020 |
Token | We |
Token ID | 1135 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
yes | +0.028 |
goals | +0.028 |
best | +0.027 |
mm | +0.026 |
narrative | +0.025 |
Neg logit contributions | |
over | -0.014 |
To | -0.014 |
late | -0.014 |
Grand | -0.013 |
Start | -0.013 |
Token | are |
Token ID | 389 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
yes | +0.003 |
goals | +0.003 |
mm | +0.003 |
best | +0.003 |
narrative | +0.003 |
Neg logit contributions | |
over | -0.005 |
yelled | -0.005 |
aback | -0.005 |
asked | -0.005 |
To | -0.005 |
Token | Jen |
Token ID | 13364 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
yes | +0.025 |
goals | +0.024 |
mm | +0.022 |
best | +0.022 |
narrative | +0.022 |
Neg logit contributions | |
over | -0.024 |
late | -0.022 |
temperature | -0.022 |
into | -0.022 |
yelled | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
yes | +0.046 |
goals | +0.046 |
narrative | +0.044 |
mm | +0.044 |
best | +0.042 |
Neg logit contributions | |
over | -0.018 |
late | -0.016 |
into | -0.015 |
temperature | -0.015 |
yelled | -0.015 |
Token | Sam |
Token ID | 3409 |
Feature activation | +2.32e+00 |
Pos logit contributions | |
goals | +0.005 |
narrative | +0.005 |
yes | +0.005 |
mm | +0.005 |
best | +0.004 |
Neg logit contributions | |
over | -0.002 |
yelled | -0.002 |
asked | -0.002 |
into | -0.002 |
To | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | +9.18e-01 |
Pos logit contributions | |
yes | +0.083 |
goals | +0.080 |
best | +0.077 |
mm | +0.077 |
narrative | +0.076 |
Neg logit contributions | |
over | -0.028 |
late | -0.025 |
To | -0.024 |
temperature | -0.024 |
aback | -0.024 |
Token | We |
Token ID | 775 |
Feature activation | -9.91e-02 |
Pos logit contributions | |
yes | +0.025 |
goals | +0.024 |
best | +0.023 |
mm | +0.023 |
narrative | +0.022 |
Neg logit contributions | |
over | -0.018 |
temperature | -0.018 |
late | -0.018 |
To | -0.017 |
aback | -0.017 |
Token | like |
Token ID | 588 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
over | +0.003 |
yelled | +0.003 |
To | +0.003 |
aback | +0.003 |
asked | +0.003 |
Neg logit contributions | |
yes | -0.002 |
goals | -0.002 |
mm | -0.002 |
best | -0.002 |
narrative | -0.002 |
Token | you |
Token ID | 345 |
Feature activation | +7.96e-01 |
Pos logit contributions | |
yes | +0.007 |
goals | +0.007 |
best | +0.007 |
mm | +0.007 |
narrative | +0.007 |
Neg logit contributions | |
over | -0.003 |
late | -0.003 |
To | -0.003 |
temperature | -0.002 |
aback | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | +7.43e-01 |
Pos logit contributions | |
yes | +0.029 |
goals | +0.028 |
mm | +0.028 |
narrative | +0.028 |
best | +0.027 |
Neg logit contributions | |
over | -0.009 |
late | -0.008 |
into | -0.008 |
yelled | -0.007 |
temperature | -0.007 |
Token | put |
Token ID | 1234 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
yes | +0.006 |
best | +0.005 |
goals | +0.005 |
course | +0.005 |
ho | +0.005 |
Neg logit contributions | |
person | -0.003 |
late | -0.003 |
net | -0.003 |
any | -0.003 |
temperature | -0.003 |
Token | it |
Token ID | 340 |
Feature activation | +8.43e-01 |
Pos logit contributions | |
over | +0.023 |
late | +0.020 |
into | +0.020 |
aback | +0.020 |
temperature | +0.019 |
Neg logit contributions | |
yes | -0.042 |
goals | -0.040 |
mm | -0.039 |
best | -0.039 |
narrative | -0.038 |
Token | back |
Token ID | 736 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
mm | +0.029 |
yes | +0.028 |
goals | +0.028 |
narrative | +0.027 |
best | +0.026 |
Neg logit contributions | |
over | -0.013 |
into | -0.012 |
late | -0.011 |
towards | -0.010 |
To | -0.010 |
Token | on |
Token ID | 319 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
yes | +0.040 |
goals | +0.039 |
mm | +0.038 |
narrative | +0.037 |
best | +0.036 |
Neg logit contributions | |
over | -0.018 |
into | -0.015 |
late | -0.013 |
towards | -0.013 |
ray | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
yes | +0.054 |
goals | +0.051 |
best | +0.050 |
mm | +0.049 |
narrative | +0.047 |
Neg logit contributions | |
over | -0.025 |
late | -0.023 |
To | -0.023 |
temperature | -0.023 |
aback | -0.023 |
Token | flower |
Token ID | 15061 |
Feature activation | +2.27e+00 |
Pos logit contributions | |
yes | +0.029 |
goals | +0.027 |
best | +0.027 |
mm | +0.026 |
narrative | +0.025 |
Neg logit contributions | |
over | -0.024 |
late | -0.022 |
To | -0.022 |
aback | -0.022 |
temperature | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
yes | +0.090 |
mm | +0.086 |
goals | +0.085 |
best | +0.084 |
narrative | +0.083 |
Neg logit contributions | |
over | -0.022 |
into | -0.019 |
To | -0.017 |
temperature | -0.017 |
late | -0.016 |
Token | She |
Token ID | 1375 |
Feature activation | -8.85e-02 |
Pos logit contributions | |
yes | +0.004 |
goals | +0.004 |
best | +0.004 |
mm | +0.004 |
narrative | +0.003 |
Neg logit contributions | |
over | -0.002 |
temperature | -0.002 |
person | -0.002 |
ray | -0.002 |
aback | -0.002 |
Token | hoped |
Token ID | 10719 |
Feature activation | -2.87e-01 |
Pos logit contributions | |
over | +0.001 |
late | +0.001 |
To | +0.001 |
temperature | +0.001 |
person | +0.001 |
Neg logit contributions | |
yes | -0.003 |
goals | -0.003 |
best | -0.003 |
mm | -0.003 |
narrative | -0.002 |
Token | it |
Token ID | 340 |
Feature activation | -6.06e-01 |
Pos logit contributions | |
over | +0.005 |
late | +0.004 |
into | +0.004 |
To | +0.004 |
temperature | +0.004 |
Neg logit contributions | |
yes | -0.009 |
goals | -0.009 |
best | -0.009 |
mm | -0.009 |
narrative | -0.009 |
Token | was |
Token ID | 373 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
over | +0.014 |
into | +0.011 |
asked | +0.011 |
To | +0.011 |
bo | +0.010 |
Neg logit contributions | |
goals | -0.019 |
mm | -0.019 |
nations | -0.018 |
ocol | -0.018 |
narrative | -0.018 |
Token | for |
Token ID | 329 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
over | +0.020 |
To | +0.020 |
person | +0.019 |
Start | +0.019 |
aback | +0.019 |
Neg logit contributions | |
yes | -0.057 |
best | -0.054 |
goals | -0.053 |
mm | -0.050 |
ho | -0.050 |
Token | dinner |
Token ID | 8073 |
Feature activation | -2.36e+00 |
Pos logit contributions | |
over | +0.029 |
ray | +0.028 |
aback | +0.027 |
To | +0.027 |
person | +0.027 |
Neg logit contributions | |
yes | -0.043 |
goals | -0.040 |
best | -0.040 |
mm | -0.038 |
narrative | -0.036 |
Token | , |
Token ID | 11 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
ray | +0.068 |
Start | +0.068 |
person | +0.067 |
plug | +0.066 |
over | +0.066 |
Neg logit contributions | |
yes | -0.047 |
best | -0.043 |
mm | -0.037 |
goals | -0.037 |
ho | -0.033 |
Token | Ruby |
Token ID | 10888 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
over | +0.024 |
into | +0.023 |
late | +0.021 |
To | +0.021 |
To | +0.021 |
Neg logit contributions | |
mm | -0.040 |
yes | -0.040 |
goals | -0.040 |
narrative | -0.039 |
ho | -0.038 |
Token | would |
Token ID | 561 |
Feature activation | +4.67e-01 |
Pos logit contributions | |
over | +0.029 |
late | +0.028 |
aback | +0.028 |
Start | +0.028 |
To | +0.028 |
Neg logit contributions | |
yes | -0.024 |
goals | -0.022 |
best | -0.022 |
narrative | -0.019 |
mm | -0.019 |
Token | look |
Token ID | 804 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
yes | +0.013 |
goals | +0.012 |
best | +0.012 |
mm | +0.011 |
narrative | +0.011 |
Neg logit contributions | |
over | -0.009 |
late | -0.009 |
aback | -0.009 |
ray | -0.009 |
To | -0.009 |
Token | at |
Token ID | 379 |
Feature activation | -4.27e-01 |
Pos logit contributions | |
yes | +0.061 |
best | +0.057 |
goals | +0.057 |
mm | +0.049 |
narrative | +0.048 |
Neg logit contributions | |
person | -0.042 |
To | -0.041 |
Start | -0.040 |
net | -0.040 |
Grand | -0.039 |
Token | the |
Token ID | 262 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
over | +0.006 |
into | +0.006 |
temperature | +0.006 |
late | +0.005 |
aback | +0.005 |
Neg logit contributions | |
yes | -0.015 |
narrative | -0.014 |
mm | -0.014 |
goals | -0.014 |
ho | -0.014 |
Token | vegetables |
Token ID | 13701 |
Feature activation | -8.02e-01 |
Pos logit contributions | |
over | +0.038 |
person | +0.037 |
Start | +0.037 |
aback | +0.037 |
yelled | +0.037 |
Neg logit contributions | |
yes | -0.033 |
goals | -0.032 |
best | -0.031 |
narrative | -0.028 |
mm | -0.028 |
Token | on |
Token ID | 319 |
Feature activation | +5.86e-01 |
Pos logit contributions | |
over | +0.011 |
into | +0.011 |
late | +0.008 |
towards | +0.008 |
temperature | +0.008 |
Neg logit contributions | |
mm | -0.029 |
goals | -0.029 |
yes | -0.028 |
narrative | -0.028 |
tweet | -0.027 |
Token | her |
Token ID | 607 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
yes | +0.021 |
goals | +0.020 |
best | +0.020 |
mm | +0.020 |
narrative | +0.020 |
Neg logit contributions | |
over | -0.008 |
late | -0.007 |
into | -0.007 |
temperature | -0.007 |
To | -0.006 |
Token | very |
Token ID | 845 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
over | +0.004 |
To | +0.004 |
temperature | +0.004 |
ray | +0.004 |
late | +0.004 |
Neg logit contributions | |
yes | -0.004 |
goals | -0.003 |
best | -0.003 |
mm | -0.003 |
narrative | -0.003 |
Token | happy |
Token ID | 3772 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
over | +0.039 |
To | +0.038 |
late | +0.037 |
temperature | +0.037 |
aback | +0.037 |
Neg logit contributions | |
yes | -0.028 |
goals | -0.026 |
best | -0.025 |
mm | -0.024 |
narrative | -0.022 |
Token | with |
Token ID | 351 |
Feature activation | +4.42e-01 |
Pos logit contributions | |
ray | +0.003 |
aback | +0.003 |
Start | +0.003 |
person | +0.003 |
To | +0.003 |
Neg logit contributions | |
yes | -0.004 |
best | -0.004 |
goals | -0.004 |
mm | -0.003 |
course | -0.003 |
Token | how |
Token ID | 703 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
yes | +0.014 |
owers | +0.014 |
ho | +0.013 |
narrative | +0.013 |
mm | +0.013 |
Neg logit contributions | |
over | -0.008 |
into | -0.006 |
To | -0.006 |
late | -0.006 |
aback | -0.006 |
Token | it |
Token ID | 340 |
Feature activation | -6.55e-02 |
Pos logit contributions | |
yes | +0.033 |
goals | +0.031 |
mm | +0.031 |
best | +0.030 |
narrative | +0.030 |
Neg logit contributions | |
over | -0.029 |
late | -0.028 |
To | -0.027 |
Start | -0.026 |
into | -0.026 |
Token | looked |
Token ID | 3114 |
Feature activation | +2.39e+00 |
Pos logit contributions | |
person | +0.001 |
late | +0.001 |
over | +0.001 |
To | +0.001 |
temperature | +0.001 |
Neg logit contributions | |
yes | -0.002 |
best | -0.002 |
goals | -0.002 |
mm | -0.002 |
narrative | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
yes | +0.072 |
best | +0.067 |
goals | +0.067 |
mm | +0.062 |
narrative | +0.061 |
Neg logit contributions | |
over | -0.041 |
person | -0.040 |
To | -0.040 |
temperature | -0.040 |
Start | -0.040 |
Token | couldn |
Token ID | 3521 |
Feature activation | +8.28e-01 |
Pos logit contributions | |
over | +0.004 |
late | +0.003 |
yelled | +0.003 |
into | +0.003 |
asked | +0.003 |
Neg logit contributions | |
yes | -0.007 |
goals | -0.007 |
mm | -0.006 |
best | -0.006 |
narrative | -0.006 |
Token | 't |
Token ID | 470 |
Feature activation | -5.76e-01 |
Pos logit contributions | |
yes | +0.024 |
goals | +0.023 |
best | +0.022 |
mm | +0.021 |
narrative | +0.021 |
Neg logit contributions | |
over | -0.016 |
late | -0.015 |
To | -0.015 |
To | -0.015 |
Start | -0.015 |
Token | wait |
Token ID | 4043 |
Feature activation | -2.55e-01 |
Pos logit contributions | |
aback | +0.015 |
over | +0.014 |
late | +0.014 |
temperature | +0.014 |
ray | +0.014 |
Neg logit contributions | |
yes | -0.014 |
goals | -0.011 |
best | -0.011 |
mm | -0.011 |
narrative | -0.010 |
Token | to |
Token ID | 284 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
over | +0.002 |
late | +0.001 |
To | +0.001 |
To | +0.001 |
into | +0.001 |
Neg logit contributions | |
yes | -0.011 |
goals | -0.010 |
best | -0.010 |
mm | -0.010 |
narrative | -0.010 |
Token | The |
Token ID | 383 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
temperature | +0.015 |
plug | +0.015 |
rid | +0.015 |
shave | +0.014 |
heads | +0.014 |
Neg logit contributions | |
yes | -0.011 |
goals | -0.008 |
tweet | -0.008 |
are | -0.007 |
course | -0.007 |
Token | man |
Token ID | 582 |
Feature activation | -8.45e-02 |
Pos logit contributions | |
yes | +0.007 |
goals | +0.007 |
best | +0.007 |
mm | +0.007 |
narrative | +0.006 |
Neg logit contributions | |
over | -0.003 |
late | -0.003 |
To | -0.003 |
temperature | -0.003 |
aback | -0.003 |
Token | said |
Token ID | 531 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
late | +0.001 |
over | +0.001 |
temperature | +0.001 |
To | +0.001 |
aback | +0.001 |
Neg logit contributions | |
yes | -0.003 |
best | -0.002 |
goals | -0.002 |
mm | -0.002 |
ho | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -6.22e-01 |
Pos logit contributions | |
over | +0.006 |
To | +0.006 |
into | +0.006 |
aback | +0.006 |
late | +0.006 |
Neg logit contributions | |
goals | -0.009 |
narrative | -0.009 |
mm | -0.008 |
best | -0.008 |
yes | -0.008 |
Token | " |
Token ID | 366 |
Feature activation | +3.84e-01 |
Pos logit contributions | |
temperature | +0.014 |
aback | +0.014 |
To | +0.013 |
ray | +0.013 |
late | +0.013 |
Neg logit contributions | |
yes | -0.017 |
best | -0.014 |
goals | -0.013 |
ho | -0.013 |
tweet | -0.013 |
Token | Do |
Token ID | 5211 |
Feature activation | +2.47e+00 |
Pos logit contributions | |
yes | +0.010 |
course | +0.009 |
ho | +0.009 |
tweet | +0.008 |
Deal | +0.008 |
Neg logit contributions | |
temperature | -0.008 |
over | -0.008 |
into | -0.008 |
To | -0.008 |
comm | -0.008 |
Token | not |
Token ID | 407 |
Feature activation | -8.15e-01 |
Pos logit contributions | |
goals | +0.084 |
ement | +0.081 |
owers | +0.080 |
ig | +0.078 |
narrative | +0.078 |
Neg logit contributions | |
To | -0.030 |
me | -0.030 |
To | -0.028 |
Start | -0.026 |
yelled | -0.025 |
Token | be |
Token ID | 307 |
Feature activation | -5.81e-01 |
Pos logit contributions | |
person | +0.025 |
ilda | +0.023 |
rev | +0.023 |
plug | +0.022 |
Grand | +0.022 |
Neg logit contributions | |
yes | -0.016 |
best | -0.015 |
tweet | -0.015 |
goals | -0.014 |
are | -0.013 |
Token | worried |
Token ID | 7960 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
person | +0.018 |
To | +0.016 |
To | +0.016 |
ilda | +0.016 |
net | +0.016 |
Neg logit contributions | |
best | -0.012 |
goals | -0.011 |
yes | -0.011 |
inches | -0.009 |
tweet | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
person | +0.038 |
To | +0.038 |
Start | +0.037 |
plug | +0.037 |
aback | +0.037 |
Neg logit contributions | |
yes | -0.037 |
goals | -0.035 |
best | -0.035 |
mm | -0.030 |
nations | -0.029 |
Token | little |
Token ID | 1310 |
Feature activation | -4.91e-01 |
Pos logit contributions | |
over | +0.024 |
late | +0.022 |
To | +0.022 |
temperature | +0.022 |
aback | +0.022 |
Neg logit contributions | |
yes | -0.039 |
goals | -0.037 |
best | -0.036 |
mm | -0.036 |
narrative | -0.035 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
person | +0.031 |
temperature | +0.030 |
over | +0.030 |
comm | +0.030 |
aback | +0.029 |
Neg logit contributions | |
yes | -0.059 |
best | -0.058 |
goals | -0.057 |
mm | -0.054 |
narrative | -0.053 |
Token | upon |
Token ID | 2402 |
Feature activation | +2.62e-01 |
Pos logit contributions | |
yes | +0.031 |
goals | +0.030 |
mm | +0.029 |
narrative | +0.028 |
best | +0.028 |
Neg logit contributions | |
over | -0.028 |
late | -0.026 |
into | -0.025 |
To | -0.025 |
asked | -0.025 |
Token | a |
Token ID | 257 |
Feature activation | +9.74e-02 |
Pos logit contributions | |
ho | +0.009 |
narrative | +0.009 |
mm | +0.009 |
goals | +0.009 |
ig | +0.009 |
Neg logit contributions | |
late | -0.003 |
over | -0.003 |
To | -0.003 |
Start | -0.003 |
To | -0.003 |
Token | time |
Token ID | 640 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
yes | +0.003 |
goals | +0.003 |
mm | +0.003 |
narrative | +0.003 |
best | +0.003 |
Neg logit contributions | |
over | -0.002 |
late | -0.002 |
To | -0.002 |
Start | -0.002 |
Grand | -0.002 |
Token | there |
Token ID | 612 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
goals | +0.006 |
narrative | +0.006 |
Deal | +0.006 |
owers | +0.006 |
mm | +0.005 |
Neg logit contributions | |
over | -0.002 |
to | -0.002 |
To | -0.002 |
into | -0.002 |
| -0.001 |
Token | was |
Token ID | 373 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
yes | +0.008 |
goals | +0.008 |
narrative | +0.008 |
mm | +0.007 |
nations | +0.007 |
Neg logit contributions | |
over | -0.003 |
asked | -0.002 |
yelled | -0.002 |
To | -0.002 |
To | -0.002 |
Token | a |
Token ID | 257 |
Feature activation | -6.76e-01 |
Pos logit contributions | |
late | +0.001 |
over | +0.001 |
Out | +0.000 |
outside | +0.000 |
into | +0.000 |
Neg logit contributions | |
ho | -0.006 |
narrative | -0.006 |
ig | -0.006 |
goals | -0.006 |
Deal | -0.005 |
Token | k |
Token ID | 479 |
Feature activation | +2.41e-01 |
Pos logit contributions | |
over | +0.011 |
late | +0.011 |
Grand | +0.010 |
To | +0.010 |
temperature | +0.010 |
Neg logit contributions | |
yes | -0.021 |
goals | -0.020 |
mm | -0.019 |
narrative | -0.019 |
best | -0.019 |
Token | itty |
Token ID | 9760 |
Feature activation | -4.33e-01 |
Pos logit contributions | |
yes | +0.005 |
goals | +0.005 |
mm | +0.004 |
best | +0.004 |
narrative | +0.004 |
Neg logit contributions | |
over | -0.006 |
late | -0.006 |
ray | -0.006 |
To | -0.006 |
yelled | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | +8.27e-02 |
Pos logit contributions | |
aback | +0.011 |
over | +0.011 |
late | +0.010 |
temperature | +0.010 |
To | +0.010 |
Neg logit contributions | |
yes | -0.010 |
best | -0.009 |
goals | -0.009 |
mm | -0.008 |
ho | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
narrative | +0.003 |
goals | +0.003 |
ement | +0.003 |
owers | +0.003 |
nations | +0.003 |
Neg logit contributions | |
asked | -0.001 |
yelled | -0.001 |
over | -0.001 |
shouted | -0.001 |
late | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
over | +0.007 |
temperature | +0.007 |
plug | +0.007 |
shave | +0.007 |
into | +0.007 |
Neg logit contributions | |
yes | -0.010 |
goals | -0.009 |
best | -0.009 |
mm | -0.009 |
ho | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | +6.05e-02 |
Pos logit contributions | |
over | +0.005 |
temperature | +0.005 |
into | +0.005 |
aback | +0.004 |
person | +0.004 |
Neg logit contributions | |
yes | -0.011 |
goals | -0.011 |
best | -0.011 |
mm | -0.010 |
narrative | -0.010 |
Token | She |
Token ID | 3347 |
Feature activation | -3.65e-01 |
Pos logit contributions | |
yes | +0.002 |
goals | +0.002 |
best | +0.001 |
mm | +0.001 |
narrative | +0.001 |
Neg logit contributions | |
over | -0.001 |
temperature | -0.001 |
into | -0.001 |
aback | -0.001 |
late | -0.001 |
Token | asked |
Token ID | 1965 |
Feature activation | +5.79e-01 |
Pos logit contributions | |
over | +0.006 |
late | +0.006 |
aback | +0.006 |
person | +0.006 |
To | +0.006 |
Neg logit contributions | |
yes | -0.011 |
goals | -0.011 |
best | -0.010 |
mm | -0.010 |
ho | -0.010 |
Token | her |
Token ID | 607 |
Feature activation | -9.40e-01 |
Pos logit contributions | |
yes | +0.016 |
goals | +0.016 |
mm | +0.015 |
best | +0.015 |
narrative | +0.015 |
Neg logit contributions | |
over | -0.012 |
temperature | -0.011 |
To | -0.011 |
late | -0.011 |
into | -0.011 |
Token | mom |
Token ID | 1995 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
over | +0.019 |
late | +0.018 |
To | +0.017 |
temperature | +0.017 |
aback | +0.017 |
Neg logit contributions | |
yes | -0.027 |
goals | -0.025 |
best | -0.025 |
mm | -0.024 |
narrative | -0.023 |
Token | but |
Token ID | 475 |
Feature activation | -6.14e-02 |
Pos logit contributions | |
goals | +0.005 |
mm | +0.005 |
yes | +0.004 |
narrative | +0.004 |
owers | +0.004 |
Neg logit contributions | |
over | -0.003 |
into | -0.002 |
To | -0.002 |
temperature | -0.002 |
ray | -0.002 |
Token | she |
Token ID | 673 |
Feature activation | -6.62e-01 |
Pos logit contributions | |
over | +0.001 |
asked | +0.001 |
into | +0.001 |
temperature | +0.001 |
Grand | +0.001 |
Neg logit contributions | |
goals | -0.002 |
mm | -0.002 |
owers | -0.002 |
narrative | -0.002 |
nations | -0.002 |
Token | said |
Token ID | 531 |
Feature activation | -3.17e-01 |
Pos logit contributions | |
over | +0.012 |
To | +0.011 |
yelled | +0.011 |
late | +0.011 |
temperature | +0.011 |
Neg logit contributions | |
yes | -0.019 |
goals | -0.019 |
best | -0.018 |
mm | -0.018 |
narrative | -0.017 |
Token | no |
Token ID | 645 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
over | +0.007 |
into | +0.006 |
asked | +0.006 |
To | +0.006 |
to | +0.006 |
Neg logit contributions | |
goals | -0.008 |
narrative | -0.008 |
nations | -0.007 |
mm | -0.007 |
Deal | -0.007 |
Token | because |
Token ID | 780 |
Feature activation | -4.60e-01 |
Pos logit contributions | |
yes | +0.054 |
goals | +0.052 |
best | +0.051 |
mm | +0.049 |
narrative | +0.047 |
Neg logit contributions | |
over | -0.022 |
To | -0.021 |
aback | -0.021 |
late | -0.020 |
person | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -5.08e-01 |
Pos logit contributions | |
mm | +0.024 |
yes | +0.024 |
goals | +0.023 |
narrative | +0.023 |
best | +0.022 |
Neg logit contributions | |
over | -0.009 |
into | -0.007 |
late | -0.007 |
towards | -0.006 |
To | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | -7.76e-01 |
Pos logit contributions | |
over | +0.008 |
temperature | +0.007 |
To | +0.007 |
aback | +0.007 |
late | +0.007 |
Neg logit contributions | |
yes | -0.017 |
goals | -0.016 |
best | -0.015 |
mm | -0.015 |
narrative | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | -1.58e-01 |
Pos logit contributions | |
over | +0.010 |
temperature | +0.010 |
aback | +0.009 |
ray | +0.009 |
into | +0.009 |
Neg logit contributions | |
yes | -0.027 |
goals | -0.026 |
best | -0.025 |
mm | -0.024 |
narrative | -0.024 |
Token | Bob |
Token ID | 18861 |
Feature activation | -8.66e-01 |
Pos logit contributions | |
over | +0.003 |
temperature | +0.003 |
aback | +0.003 |
late | +0.003 |
ray | +0.003 |
Neg logit contributions | |
yes | -0.004 |
goals | -0.004 |
best | -0.004 |
mm | -0.004 |
narrative | -0.004 |
Token | 's |
Token ID | 338 |
Feature activation | -4.09e-01 |
Pos logit contributions | |
over | +0.016 |
late | +0.015 |
To | +0.015 |
temperature | +0.015 |
aback | +0.015 |
Neg logit contributions | |
yes | -0.026 |
goals | -0.024 |
best | -0.024 |
mm | -0.023 |
narrative | -0.022 |
Token | favourite |
Token ID | 12507 |
Feature activation | +3.51e-01 |
Pos logit contributions | |
over | +0.007 |
temperature | +0.007 |
To | +0.006 |
person | +0.006 |
into | +0.006 |
Neg logit contributions | |
yes | -0.012 |
mm | -0.012 |
goals | -0.012 |
narrative | -0.012 |
owers | -0.011 |
Token | place |
Token ID | 1295 |
Feature activation | -2.77e-02 |
Pos logit contributions | |
yes | +0.009 |
goals | +0.008 |
mm | +0.008 |
best | +0.008 |
narrative | +0.008 |
Neg logit contributions | |
over | -0.008 |
temperature | -0.007 |
To | -0.007 |
late | -0.007 |
To | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | +5.23e-01 |
Pos logit contributions | |
over | +0.000 |
late | +0.000 |
To | +0.000 |
temperature | +0.000 |
aback | +0.000 |
Neg logit contributions | |
yes | -0.001 |
goals | -0.001 |
best | -0.001 |
mm | -0.001 |
narrative | -0.001 |
Token | explore |
Token ID | 7301 |
Feature activation | +3.04e-01 |
Pos logit contributions | |
yes | +0.017 |
goals | +0.016 |
mm | +0.016 |
best | +0.016 |
narrative | +0.016 |
Neg logit contributions | |
over | -0.009 |
To | -0.008 |
To | -0.008 |
late | -0.007 |
temperature | -0.007 |
Token | was |
Token ID | 373 |
Feature activation | +7.65e-01 |
Pos logit contributions | |
mm | +0.011 |
yes | +0.011 |
narrative | +0.010 |
goals | +0.010 |
ho | +0.010 |
Neg logit contributions | |
over | -0.004 |
into | -0.004 |
late | -0.004 |
To | -0.003 |
To | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
yes | +0.022 |
goals | +0.021 |
best | +0.021 |
mm | +0.019 |
narrative | +0.019 |
Neg logit contributions | |
over | -0.014 |
person | -0.014 |
To | -0.014 |
temperature | -0.014 |
ray | -0.014 |
Token | " |
Token ID | 366 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
temperature | +0.007 |
ray | +0.007 |
aback | +0.007 |
over | +0.007 |
To | +0.007 |
Neg logit contributions | |
yes | -0.009 |
best | -0.007 |
goals | -0.007 |
ho | -0.007 |
mm | -0.007 |
Token | I |
Token ID | 40 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
yes | +0.031 |
goals | +0.028 |
best | +0.028 |
mm | +0.027 |
narrative | +0.027 |
Neg logit contributions | |
over | -0.022 |
temperature | -0.022 |
into | -0.021 |
late | -0.020 |
To | -0.020 |
Token | have |
Token ID | 423 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
yes | +0.003 |
best | +0.002 |
goals | +0.002 |
mm | +0.002 |
narrative | +0.002 |
Neg logit contributions | |
over | -0.003 |
late | -0.003 |
temperature | -0.002 |
person | -0.002 |
To | -0.002 |
Token | some |
Token ID | 617 |
Feature activation | -3.59e-01 |
Pos logit contributions | |
person | +0.021 |
over | +0.020 |
temperature | +0.019 |
ray | +0.019 |
aback | +0.019 |
Neg logit contributions | |
yes | -0.038 |
goals | -0.036 |
best | -0.036 |
mm | -0.032 |
narrative | -0.031 |
Token | special |
Token ID | 2041 |
Feature activation | -4.06e-01 |
Pos logit contributions | |
over | +0.008 |
late | +0.008 |
To | +0.008 |
temperature | +0.008 |
aback | +0.008 |
Neg logit contributions | |
yes | -0.009 |
goals | -0.008 |
best | -0.008 |
mm | -0.008 |
narrative | -0.008 |
Token | magic |
Token ID | 5536 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
over | +0.009 |
late | +0.009 |
To | +0.009 |
temperature | +0.009 |
aback | +0.009 |
Neg logit contributions | |
yes | -0.010 |
goals | -0.010 |
best | -0.009 |
mm | -0.009 |
narrative | -0.009 |
Token | just |
Token ID | 655 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
over | +0.003 |
late | +0.003 |
To | +0.002 |
aback | +0.002 |
Start | +0.002 |
Neg logit contributions | |
yes | -0.005 |
goals | -0.005 |
best | -0.005 |
mm | -0.004 |
narrative | -0.004 |
Token | for |
Token ID | 329 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
over | +0.003 |
late | +0.002 |
To | +0.002 |
temperature | +0.002 |
aback | +0.002 |
Neg logit contributions | |
yes | -0.004 |
goals | -0.004 |
best | -0.004 |
mm | -0.004 |
narrative | -0.004 |
Token | you |
Token ID | 345 |
Feature activation | +5.94e-01 |
Pos logit contributions | |
over | +0.027 |
To | +0.026 |
late | +0.026 |
aback | +0.025 |
temperature | +0.025 |
Neg logit contributions | |
yes | -0.035 |
goals | -0.033 |
best | -0.032 |
mm | -0.031 |
narrative | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | +8.02e-01 |
Pos logit contributions | |
yes | +0.021 |
goals | +0.020 |
best | +0.020 |
mm | +0.019 |
narrative | +0.019 |
Neg logit contributions | |
over | -0.008 |
late | -0.007 |
To | -0.007 |
temperature | -0.007 |
aback | -0.007 |
Token | Would |
Token ID | 10928 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
yes | +0.023 |
goals | +0.021 |
mm | +0.021 |
best | +0.021 |
narrative | +0.019 |
Neg logit contributions | |
over | -0.015 |
temperature | -0.015 |
late | -0.014 |
yelled | -0.014 |
comm | -0.014 |
Token | ," |
Token ID | 553 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
person | +0.003 |
aback | +0.003 |
Start | +0.002 |
To | +0.002 |
temperature | +0.002 |
Neg logit contributions | |
yes | -0.005 |
goals | -0.005 |
best | -0.005 |
mm | -0.005 |
narrative | -0.004 |
Token | Tom |
Token ID | 4186 |
Feature activation | +7.98e-01 |
Pos logit contributions | |
yes | +0.007 |
goals | +0.007 |
best | +0.007 |
narrative | +0.007 |
ig | +0.007 |
Neg logit contributions | |
over | -0.003 |
yelled | -0.003 |
late | -0.003 |
asked | -0.003 |
To | -0.003 |
Token | says |
Token ID | 1139 |
Feature activation | +8.96e-01 |
Pos logit contributions | |
yes | +0.019 |
goals | +0.019 |
best | +0.018 |
nations | +0.016 |
are | +0.015 |
Neg logit contributions | |
person | -0.019 |
To | -0.019 |
late | -0.018 |
temperature | -0.018 |
aback | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +9.62e-01 |
Pos logit contributions | |
yes | +0.032 |
goals | +0.030 |
best | +0.030 |
mm | +0.028 |
narrative | +0.028 |
Neg logit contributions | |
over | -0.011 |
To | -0.011 |
late | -0.010 |
temperature | -0.010 |
aback | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | +9.33e-01 |
Pos logit contributions | |
yes | +0.028 |
goals | +0.025 |
best | +0.025 |
mm | +0.023 |
ho | +0.022 |
Neg logit contributions | |
temperature | -0.018 |
any | -0.018 |
ray | -0.018 |
ilda | -0.018 |
aback | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
yes | +0.027 |
goals | +0.026 |
best | +0.025 |
mm | +0.024 |
narrative | +0.022 |
Neg logit contributions | |
temperature | -0.017 |
person | -0.017 |
aback | -0.017 |
comm | -0.016 |
any | -0.016 |
Token | " |
Token ID | 1 |
Feature activation | +6.52e-01 |
Pos logit contributions | |
yes | +0.038 |
goals | +0.036 |
best | +0.035 |
mm | +0.034 |
narrative | +0.032 |
Neg logit contributions | |
over | -0.027 |
aback | -0.027 |
temperature | -0.027 |
comm | -0.026 |
To | -0.026 |
Token | Yes |
Token ID | 5297 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
goals | +0.022 |
yes | +0.021 |
best | +0.021 |
ig | +0.020 |
mm | +0.020 |
Neg logit contributions | |
To | -0.010 |
Grand | -0.009 |
late | -0.009 |
over | -0.009 |
Start | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | +8.08e-01 |
Pos logit contributions | |
yes | +0.049 |
goals | +0.048 |
best | +0.046 |
narrative | +0.045 |
mm | +0.044 |
Neg logit contributions | |
over | -0.044 |
late | -0.041 |
temperature | -0.040 |
into | -0.040 |
To | -0.040 |
Token | Tom |
Token ID | 4186 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
owers | +0.029 |
goals | +0.029 |
narrative | +0.028 |
ho | +0.027 |
best | +0.027 |
Neg logit contributions | |
over | -0.010 |
late | -0.009 |
Grand | -0.008 |
yelled | -0.008 |
Grand | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +8.29e-01 |
Pos logit contributions | |
yes | +0.034 |
goals | +0.032 |
best | +0.032 |
mm | +0.030 |
narrative | +0.028 |
Neg logit contributions | |
over | -0.026 |
To | -0.025 |
late | -0.025 |
person | -0.025 |
Start | -0.024 |
Token | We |
Token ID | 775 |
Feature activation | -3.54e-01 |
Pos logit contributions | |
yes | +0.015 |
goals | +0.014 |
best | +0.014 |
mm | +0.014 |
narrative | +0.013 |
Neg logit contributions | |
over | -0.012 |
temperature | -0.011 |
late | -0.011 |
To | -0.011 |
yelled | -0.011 |
Token | are |
Token ID | 389 |
Feature activation | -5.76e-02 |
Pos logit contributions | |
over | +0.010 |
To | +0.010 |
temperature | +0.010 |
late | +0.010 |
ray | +0.010 |
Neg logit contributions | |
yes | -0.007 |
goals | -0.006 |
mm | -0.006 |
best | -0.006 |
narrative | -0.005 |
Token | friends |
Token ID | 2460 |
Feature activation | -5.02e-01 |
Pos logit contributions | |
over | +0.002 |
late | +0.002 |
into | +0.001 |
temperature | +0.001 |
To | +0.001 |
Neg logit contributions | |
yes | -0.001 |
goals | -0.001 |
mm | -0.001 |
ho | -0.001 |
narrative | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +4.08e-01 |
Pos logit contributions | |
over | +0.005 |
late | +0.005 |
To | +0.004 |
into | +0.004 |
aback | +0.004 |
Neg logit contributions | |
yes | -0.019 |
goals | -0.018 |
mm | -0.018 |
best | -0.018 |
narrative | -0.017 |
Token | We |
Token ID | 775 |
Feature activation | -4.07e-01 |
Pos logit contributions | |
yes | +0.011 |
goals | +0.010 |
best | +0.010 |
mm | +0.010 |
narrative | +0.009 |
Neg logit contributions | |
over | -0.009 |
temperature | -0.008 |
yelled | -0.008 |
Start | -0.008 |
aback | -0.008 |
Token | will |
Token ID | 481 |
Feature activation | -4.47e-01 |
Pos logit contributions | |
over | +0.012 |
To | +0.011 |
temperature | +0.011 |
ray | +0.011 |
late | +0.011 |
Neg logit contributions | |
yes | -0.008 |
goals | -0.007 |
mm | -0.007 |
best | -0.007 |
narrative | -0.006 |
Token | play |
Token ID | 711 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
aback | +0.010 |
over | +0.009 |
late | +0.009 |
ray | +0.009 |
Start | +0.009 |
Neg logit contributions | |
yes | -0.012 |
best | -0.011 |
goals | -0.011 |
mm | -0.011 |
narrative | -0.010 |
Token | with |
Token ID | 351 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
over | +0.004 |
late | +0.004 |
into | +0.003 |
To | +0.003 |
To | +0.003 |
Neg logit contributions | |
yes | -0.007 |
ho | -0.007 |
mm | -0.007 |
goals | -0.007 |
narrative | -0.007 |
Token | our |
Token ID | 674 |
Feature activation | -6.55e-01 |
Pos logit contributions | |
over | +0.016 |
into | +0.015 |
late | +0.014 |
temperature | +0.014 |
ray | +0.013 |
Neg logit contributions | |
yes | -0.034 |
narrative | -0.031 |
mm | -0.031 |
ho | -0.031 |
goals | -0.031 |
Token | models |
Token ID | 4981 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
over | +0.016 |
temperature | +0.015 |
into | +0.015 |
late | +0.015 |
Grand | +0.014 |
Neg logit contributions | |
yes | -0.017 |
mm | -0.015 |
Deal | -0.013 |
ho | -0.013 |
owers | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | -6.28e-01 |
Pos logit contributions | |
yes | +0.037 |
goals | +0.035 |
mm | +0.035 |
best | +0.034 |
narrative | +0.034 |
Neg logit contributions | |
over | -0.014 |
late | -0.012 |
into | -0.012 |
To | -0.012 |
temperature | -0.011 |
Token | lived |
Token ID | 5615 |
Feature activation | +2.81e-04 |
Pos logit contributions | |
yes | +0.006 |
goals | +0.005 |
mm | +0.005 |
best | +0.005 |
narrative | +0.005 |
Neg logit contributions | |
over | -0.006 |
yelled | -0.005 |
asked | -0.005 |
To | -0.005 |
late | -0.005 |
Token | in |
Token ID | 287 |
Feature activation | -8.82e-01 |
Pos logit contributions | |
yes | +0.000 |
mm | +0.000 |
goals | +0.000 |
narrative | +0.000 |
best | +0.000 |
Neg logit contributions | |
over | -0.000 |
late | -0.000 |
into | -0.000 |
To | -0.000 |
yelled | -0.000 |
Token | the |
Token ID | 262 |
Feature activation | -3.41e-01 |
Pos logit contributions | |
late | +0.012 |
over | +0.012 |
To | +0.011 |
Start | +0.011 |
To | +0.010 |
Neg logit contributions | |
yes | -0.031 |
goals | -0.029 |
narrative | -0.029 |
mm | -0.029 |
best | -0.028 |
Token | sea |
Token ID | 5417 |
Feature activation | +8.61e-01 |
Pos logit contributions | |
over | +0.008 |
late | +0.008 |
To | +0.008 |
into | +0.007 |
temperature | +0.007 |
Neg logit contributions | |
yes | -0.009 |
goals | -0.008 |
mm | -0.008 |
best | -0.007 |
narrative | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +9.28e-02 |
Pos logit contributions | |
yes | +0.033 |
goals | +0.031 |
mm | +0.031 |
best | +0.030 |
narrative | +0.030 |
Neg logit contributions | |
over | -0.009 |
late | -0.008 |
To | -0.007 |
into | -0.007 |
yelled | -0.007 |
Token | The |
Token ID | 383 |
Feature activation | -1.99e-01 |
Pos logit contributions | |
yes | +0.003 |
goals | +0.003 |
mm | +0.003 |
best | +0.003 |
course | +0.002 |
Neg logit contributions | |
temperature | -0.002 |
person | -0.002 |
aback | -0.002 |
ray | -0.002 |
any | -0.002 |
Token | crab |
Token ID | 32202 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
over | +0.005 |
late | +0.005 |
into | +0.005 |
temperature | +0.005 |
To | +0.004 |
Neg logit contributions | |
yes | -0.005 |
mm | -0.004 |
goals | -0.004 |
owers | -0.004 |
Deal | -0.004 |
Token | had |
Token ID | 550 |
Feature activation | -9.03e-01 |
Pos logit contributions | |
yes | +0.006 |
goals | +0.006 |
mm | +0.006 |
narrative | +0.005 |
best | +0.005 |
Neg logit contributions | |
over | -0.005 |
asked | -0.005 |
yelled | -0.004 |
To | -0.004 |
into | -0.004 |
Token | many |
Token ID | 867 |
Feature activation | -4.18e-01 |
Pos logit contributions | |
over | +0.017 |
late | +0.016 |
To | +0.016 |
into | +0.016 |
yelled | +0.016 |
Neg logit contributions | |
yes | -0.027 |
mm | -0.025 |
goals | -0.025 |
narrative | -0.024 |
owers | -0.023 |
Token | friends |
Token ID | 2460 |
Feature activation | -5.29e-01 |
Pos logit contributions | |
asked | +0.010 |
yelled | +0.010 |
into | +0.009 |
shouted | +0.009 |
over | +0.009 |
Neg logit contributions | |
yes | -0.011 |
mm | -0.010 |
ho | -0.009 |
otion | -0.009 |
narrative | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
over | +0.009 |
into | +0.007 |
asked | +0.007 |
towards | +0.007 |
To | +0.007 |
Neg logit contributions | |
mm | -0.018 |
yes | -0.018 |
goals | -0.017 |
narrative | -0.017 |
best | -0.016 |
Token | opinions |
Token ID | 9317 |
Feature activation | +3.79e-01 |
Pos logit contributions | |
yes | +0.006 |
mm | +0.006 |
goals | +0.006 |
owers | +0.006 |
narrative | +0.006 |
Neg logit contributions | |
over | -0.004 |
temperature | -0.004 |
late | -0.004 |
yelled | -0.004 |
comm | -0.004 |
Token | had |
Token ID | 550 |
Feature activation | +2.28e-02 |
Pos logit contributions | |
yes | +0.007 |
goals | +0.006 |
best | +0.006 |
mm | +0.006 |
narrative | +0.005 |
Neg logit contributions | |
over | -0.011 |
late | -0.011 |
To | -0.011 |
temperature | -0.011 |
aback | -0.011 |
Token | been |
Token ID | 587 |
Feature activation | +2.53e-01 |
Pos logit contributions | |
yes | +0.001 |
best | +0.001 |
goals | +0.001 |
mm | +0.001 |
narrative | +0.001 |
Neg logit contributions | |
person | -0.000 |
ray | -0.000 |
Grand | -0.000 |
late | -0.000 |
over | -0.000 |
Token | useful |
Token ID | 4465 |
Feature activation | -8.05e-01 |
Pos logit contributions | |
yes | +0.007 |
best | +0.006 |
goals | +0.006 |
mm | +0.005 |
narrative | +0.005 |
Neg logit contributions | |
person | -0.005 |
ray | -0.005 |
To | -0.005 |
To | -0.005 |
temperature | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | -8.71e-01 |
Pos logit contributions | |
over | +0.012 |
late | +0.011 |
temperature | +0.010 |
To | +0.010 |
aback | +0.010 |
Neg logit contributions | |
yes | -0.027 |
goals | -0.026 |
best | -0.025 |
mm | -0.025 |
narrative | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | -5.99e-01 |
Pos logit contributions | |
temperature | +0.021 |
ZE | +0.019 |
comm | +0.019 |
shave | +0.019 |
into | +0.019 |
Neg logit contributions | |
best | -0.019 |
yes | -0.019 |
goals | -0.019 |
mm | -0.019 |
course | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
temperature | +0.012 |
aback | +0.011 |
person | +0.011 |
over | +0.010 |
into | +0.010 |
Neg logit contributions | |
yes | -0.017 |
goals | -0.016 |
best | -0.016 |
mm | -0.015 |
course | -0.014 |
Token | The |
Token ID | 464 |
Feature activation | +3.58e-01 |
Pos logit contributions | |
over | +0.005 |
temperature | +0.005 |
aback | +0.005 |
into | +0.004 |
person | +0.004 |
Neg logit contributions | |
yes | -0.007 |
goals | -0.006 |
best | -0.006 |
mm | -0.006 |
narrative | -0.006 |
Token | mom |
Token ID | 1995 |
Feature activation | -9.72e-01 |
Pos logit contributions | |
yes | +0.010 |
goals | +0.010 |
mm | +0.010 |
best | +0.009 |
ho | +0.009 |
Neg logit contributions | |
over | -0.007 |
late | -0.007 |
temperature | -0.006 |
comm | -0.006 |
yelled | -0.006 |
Token | my |
Token ID | 1820 |
Feature activation | -4.39e-01 |
Pos logit contributions | |
over | +0.016 |
To | +0.016 |
late | +0.016 |
temperature | +0.015 |
aback | +0.015 |
Neg logit contributions | |
yes | -0.030 |
goals | -0.028 |
best | -0.028 |
mm | -0.026 |
narrative | -0.026 |
Token | saw |
Token ID | 2497 |
Feature activation | -2.55e-01 |
Pos logit contributions | |
late | +0.007 |
over | +0.007 |
To | +0.007 |
temperature | +0.007 |
To | +0.007 |
Neg logit contributions | |
yes | -0.013 |
best | -0.012 |
goals | -0.012 |
mm | -0.012 |
narrative | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | +2.79e-01 |
Pos logit contributions | |
over | +0.002 |
late | +0.002 |
To | +0.002 |
temperature | +0.002 |
aback | +0.002 |
Neg logit contributions | |
yes | -0.004 |
goals | -0.004 |
best | -0.004 |
mm | -0.004 |
narrative | -0.004 |
Token | slippery |
Token ID | 32911 |
Feature activation | +7.33e-01 |
Pos logit contributions | |
yes | +0.009 |
goals | +0.008 |
best | +0.008 |
mm | +0.008 |
narrative | +0.008 |
Neg logit contributions | |
over | -0.005 |
late | -0.004 |
temperature | -0.004 |
into | -0.004 |
To | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -5.81e-01 |
Pos logit contributions | |
yes | +0.027 |
goals | +0.025 |
mm | +0.025 |
narrative | +0.024 |
best | +0.024 |
Neg logit contributions | |
over | -0.010 |
into | -0.008 |
late | -0.008 |
temperature | -0.008 |
aback | -0.007 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
temperature | +0.011 |
over | +0.011 |
ray | +0.010 |
person | +0.010 |
plug | +0.010 |
Neg logit contributions | |
yes | -0.017 |
goals | -0.016 |
mm | -0.015 |
best | -0.015 |
course | -0.014 |
Token | didn |
Token ID | 1422 |
Feature activation | -7.06e-01 |
Pos logit contributions | |
late | +0.021 |
over | +0.021 |
To | +0.021 |
person | +0.021 |
aback | +0.021 |
Neg logit contributions | |
yes | -0.031 |
goals | -0.028 |
best | -0.028 |
mm | -0.026 |
ho | -0.026 |
Token | 't |
Token ID | 470 |
Feature activation | -5.95e-01 |
Pos logit contributions | |
over | +0.015 |
temperature | +0.015 |
person | +0.014 |
aback | +0.014 |
ray | +0.014 |
Neg logit contributions | |
yes | -0.018 |
goals | -0.017 |
best | -0.017 |
mm | -0.016 |
narrative | -0.016 |
Token | want |
Token ID | 765 |
Feature activation | -7.09e-01 |
Pos logit contributions | |
plug | +0.017 |
person | +0.017 |
Grand | +0.017 |
Mrs | +0.017 |
To | +0.017 |
Neg logit contributions | |
yes | -0.011 |
best | -0.010 |
are | -0.010 |
goals | -0.010 |
tweet | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | -6.12e-01 |
Pos logit contributions | |
person | +0.010 |
ray | +0.009 |
Grand | +0.009 |
net | +0.009 |
Start | +0.009 |
Neg logit contributions | |
yes | -0.025 |
best | -0.024 |
goals | -0.023 |
course | -0.021 |
mm | -0.021 |
Token | da |
Token ID | 12379 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
over | +0.013 |
late | +0.012 |
To | +0.012 |
aback | +0.012 |
temperature | +0.012 |
Neg logit contributions | |
yes | -0.017 |
goals | -0.016 |
best | -0.015 |
mm | -0.015 |
narrative | -0.014 |
Token | isy |
Token ID | 13560 |
Feature activation | -5.97e-01 |
Pos logit contributions | |
over | +0.003 |
To | +0.003 |
into | +0.003 |
temperature | +0.003 |
late | +0.003 |
Neg logit contributions | |
goals | -0.003 |
yes | -0.003 |
narrative | -0.003 |
best | -0.003 |
nations | -0.003 |
Token | to |
Token ID | 284 |
Feature activation | +6.50e-01 |
Pos logit contributions | |
over | +0.005 |
into | +0.004 |
temperature | +0.004 |
towards | +0.003 |
late | +0.003 |
Neg logit contributions | |
narrative | -0.024 |
yes | -0.024 |
goals | -0.024 |
mm | -0.024 |
nations | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | -4.07e-01 |
Pos logit contributions | |
any | +0.019 |
person | +0.019 |
ray | +0.019 |
over | +0.018 |
ilda | +0.018 |
Neg logit contributions | |
goals | -0.017 |
mm | -0.017 |
yes | -0.016 |
best | -0.015 |
narrative | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
person | +0.008 |
temperature | +0.008 |
any | +0.008 |
ray | +0.008 |
aback | +0.008 |
Neg logit contributions | |
yes | -0.011 |
goals | -0.011 |
mm | -0.010 |
best | -0.010 |
are | -0.010 |
Token | S |
Token ID | 50 |
Feature activation | -9.93e-01 |
Pos logit contributions | |
yes | +0.004 |
goals | +0.004 |
mm | +0.004 |
best | +0.004 |
narrative | +0.004 |
Neg logit contributions | |
person | -0.004 |
over | -0.004 |
temperature | -0.004 |
aback | -0.004 |
ray | -0.004 |
Token | ara |
Token ID | 3301 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
over | +0.022 |
late | +0.021 |
To | +0.020 |
temperature | +0.020 |
aback | +0.020 |
Neg logit contributions | |
yes | -0.026 |
goals | -0.025 |
best | -0.024 |
mm | -0.023 |
narrative | -0.022 |
Token | was |
Token ID | 373 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
over | +0.032 |
late | +0.032 |
aback | +0.032 |
person | +0.032 |
To | +0.031 |
Neg logit contributions | |
yes | -0.037 |
goals | -0.033 |
best | -0.032 |
mm | -0.030 |
ho | -0.030 |
Token | sad |
Token ID | 6507 |
Feature activation | -4.92e-01 |
Pos logit contributions | |
yes | +0.004 |
best | +0.004 |
goals | +0.003 |
narrative | +0.003 |
mm | +0.002 |
Neg logit contributions | |
person | -0.007 |
ray | -0.007 |
To | -0.007 |
aback | -0.006 |
olyn | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | -7.98e-01 |
Pos logit contributions | |
person | +0.012 |
ZE | +0.012 |
net | +0.011 |
any | +0.011 |
aback | +0.011 |
Neg logit contributions | |
yes | -0.013 |
best | -0.012 |
goals | -0.012 |
mm | -0.011 |
are | -0.010 |
Token | angry |
Token ID | 7954 |
Feature activation | +5.85e-02 |
Pos logit contributions | |
person | +0.019 |
To | +0.018 |
aback | +0.018 |
ray | +0.018 |
Grand | +0.018 |
Neg logit contributions | |
yes | -0.020 |
goals | -0.018 |
best | -0.018 |
narrative | -0.016 |
ho | -0.016 |
Token | again |
Token ID | 757 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
yes | +0.002 |
goals | +0.002 |
best | +0.002 |
mm | +0.002 |
narrative | +0.001 |
Neg logit contributions | |
aback | -0.001 |
person | -0.001 |
any | -0.001 |
To | -0.001 |
plug | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
yes | +0.038 |
goals | +0.037 |
best | +0.036 |
mm | +0.033 |
narrative | +0.033 |
Neg logit contributions | |
aback | -0.020 |
person | -0.020 |
ray | -0.020 |
over | -0.019 |
Start | -0.019 |
Token | She |
Token ID | 1375 |
Feature activation | -5.44e-01 |
Pos logit contributions | |
over | +0.004 |
person | +0.004 |
ray | +0.004 |
yelled | +0.004 |
plug | +0.004 |
Neg logit contributions | |
goals | -0.004 |
yes | -0.004 |
mm | -0.004 |
best | -0.004 |
narrative | -0.003 |
Token | She |
Token ID | 1375 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
temperature | +0.025 |
ray | +0.025 |
over | +0.025 |
aback | +0.025 |
olyn | +0.024 |
Neg logit contributions | |
yes | -0.033 |
best | -0.029 |
mm | -0.029 |
goals | -0.028 |
Deal | -0.025 |
Token | explained |
Token ID | 4893 |
Feature activation | -4.51e-01 |
Pos logit contributions | |
late | +0.018 |
aback | +0.018 |
To | +0.017 |
over | +0.017 |
person | +0.017 |
Neg logit contributions | |
yes | -0.038 |
best | -0.034 |
goals | -0.032 |
narrative | -0.031 |
mm | -0.031 |
Token | that |
Token ID | 326 |
Feature activation | -6.36e-01 |
Pos logit contributions | |
over | +0.007 |
To | +0.007 |
late | +0.007 |
temperature | +0.007 |
aback | +0.007 |
Neg logit contributions | |
yes | -0.014 |
goals | -0.013 |
best | -0.013 |
mm | -0.013 |
narrative | -0.012 |
Token | she |
Token ID | 673 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
over | +0.012 |
late | +0.011 |
temperature | +0.011 |
To | +0.010 |
into | +0.010 |
Neg logit contributions | |
yes | -0.019 |
goals | -0.018 |
mm | -0.017 |
best | -0.017 |
narrative | -0.017 |
Token | had |
Token ID | 550 |
Feature activation | -9.84e-01 |
Pos logit contributions | |
over | +0.039 |
late | +0.037 |
To | +0.036 |
temperature | +0.036 |
aback | +0.036 |
Neg logit contributions | |
yes | -0.054 |
goals | -0.051 |
best | -0.050 |
mm | -0.049 |
narrative | -0.047 |
Token | spilled |
Token ID | 31896 |
Feature activation | -2.90e+00 |
Pos logit contributions | |
person | +0.017 |
over | +0.016 |
aback | +0.016 |
ray | +0.015 |
Start | +0.015 |
Neg logit contributions | |
yes | -0.031 |
best | -0.029 |
goals | -0.028 |
mm | -0.025 |
narrative | -0.025 |
Token | the |
Token ID | 262 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
person | +0.066 |
ray | +0.064 |
To | +0.064 |
Start | +0.063 |
Grand | +0.063 |
Neg logit contributions | |
yes | -0.072 |
best | -0.070 |
goals | -0.066 |
mm | -0.062 |
course | -0.060 |
Token | pan |
Token ID | 3425 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
over | +0.064 |
person | +0.063 |
To | +0.063 |
aback | +0.062 |
To | +0.062 |
Neg logit contributions | |
yes | -0.054 |
best | -0.051 |
goals | -0.047 |
mm | -0.044 |
narrative | -0.042 |
Token | when |
Token ID | 618 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
Start | +0.019 |
To | +0.019 |
over | +0.019 |
person | +0.019 |
aback | +0.019 |
Neg logit contributions | |
yes | -0.040 |
best | -0.038 |
goals | -0.037 |
mm | -0.035 |
narrative | -0.034 |
Token | she |
Token ID | 673 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
over | +0.026 |
late | +0.025 |
temperature | +0.024 |
into | +0.024 |
To | +0.023 |
Neg logit contributions | |
yes | -0.040 |
goals | -0.039 |
mm | -0.037 |
narrative | -0.036 |
best | -0.036 |
Token | was |
Token ID | 373 |
Feature activation | -7.49e-01 |
Pos logit contributions | |
over | +0.029 |
aback | +0.028 |
late | +0.028 |
To | +0.028 |
person | +0.027 |
Neg logit contributions | |
yes | -0.042 |
goals | -0.039 |
best | -0.039 |
mm | -0.037 |
narrative | -0.036 |
Token | it |
Token ID | 340 |
Feature activation | -6.16e-01 |
Pos logit contributions | |
person | +0.029 |
any | +0.029 |
Grand | +0.029 |
rev | +0.029 |
comm | +0.029 |
Neg logit contributions | |
yes | -0.027 |
best | -0.026 |
goals | -0.025 |
course | -0.023 |
are | -0.021 |
Token | to |
Token ID | 284 |
Feature activation | -2.88e-02 |
Pos logit contributions | |
person | +0.010 |
comm | +0.009 |
Grand | +0.009 |
rev | +0.009 |
Start | +0.009 |
Neg logit contributions | |
yes | -0.021 |
best | -0.020 |
goals | -0.018 |
ig | -0.016 |
ho | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | +6.10e-03 |
Pos logit contributions | |
over | +0.001 |
To | +0.001 |
aback | +0.001 |
late | +0.001 |
person | +0.001 |
Neg logit contributions | |
yes | -0.001 |
goals | -0.001 |
best | -0.001 |
mm | -0.001 |
narrative | -0.001 |
Token | kitchen |
Token ID | 9592 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
yes | +0.000 |
goals | +0.000 |
best | +0.000 |
mm | +0.000 |
narrative | +0.000 |
Neg logit contributions | |
over | -0.000 |
aback | -0.000 |
late | -0.000 |
To | -0.000 |
Start | -0.000 |
Token | to |
Token ID | 284 |
Feature activation | -7.14e-03 |
Pos logit contributions | |
over | +0.015 |
Start | +0.014 |
aback | +0.014 |
To | +0.014 |
person | +0.014 |
Neg logit contributions | |
yes | -0.036 |
goals | -0.034 |
best | -0.034 |
mm | -0.032 |
narrative | -0.031 |
Token | wash |
Token ID | 13502 |
Feature activation | -2.55e+00 |
Pos logit contributions | |
person | +0.000 |
ray | +0.000 |
aback | +0.000 |
over | +0.000 |
Start | +0.000 |
Neg logit contributions | |
yes | -0.000 |
best | -0.000 |
mm | -0.000 |
goals | -0.000 |
course | -0.000 |
Token | it |
Token ID | 340 |
Feature activation | +3.36e-01 |
Pos logit contributions | |
person | +0.056 |
To | +0.056 |
Grand | +0.055 |
Start | +0.054 |
comm | +0.054 |
Neg logit contributions | |
yes | -0.063 |
best | -0.063 |
goals | -0.061 |
mm | -0.056 |
ig | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
yes | +0.011 |
best | +0.011 |
goals | +0.011 |
mm | +0.010 |
narrative | +0.009 |
Neg logit contributions | |
person | -0.005 |
Start | -0.005 |
To | -0.005 |
aback | -0.005 |
Grand | -0.005 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
shave | +0.024 |
ray | +0.024 |
bo | +0.024 |
temperature | +0.023 |
person | +0.023 |
Neg logit contributions | |
yes | -0.024 |
goals | -0.020 |
mm | -0.020 |
best | -0.020 |
owers | -0.018 |
Token | put |
Token ID | 1234 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
over | +0.034 |
late | +0.033 |
To | +0.032 |
person | +0.032 |
To | +0.032 |
Neg logit contributions | |
yes | -0.043 |
goals | -0.039 |
best | -0.039 |
mm | -0.036 |
narrative | -0.035 |
Token | the |
Token ID | 262 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
To | +0.029 |
person | +0.029 |
over | +0.029 |
ray | +0.028 |
late | +0.028 |
Neg logit contributions | |
yes | -0.046 |
best | -0.043 |
goals | -0.043 |
mm | -0.041 |
narrative | -0.039 |
Token | bird |
Token ID | 6512 |
Feature activation | +2.00e+00 |
Pos logit contributions | |
mm | +0.014 |
yes | +0.013 |
narrative | +0.013 |
goals | +0.013 |
otion | +0.012 |
Neg logit contributions | |
source | -0.009 |
over | -0.008 |
temperature | -0.008 |
comm | -0.007 |
sound | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
mm | +0.073 |
narrative | +0.072 |
yes | +0.070 |
goals | +0.069 |
Deal | +0.066 |
Neg logit contributions | |
over | -0.029 |
into | -0.025 |
towards | -0.021 |
asked | -0.020 |
yelled | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -9.04e-01 |
Pos logit contributions | |
temperature | +0.020 |
ray | +0.019 |
yelled | +0.019 |
comm | +0.019 |
plug | +0.018 |
Neg logit contributions | |
yes | -0.032 |
goals | -0.031 |
mm | -0.029 |
best | -0.029 |
owers | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | -5.52e-01 |
Pos logit contributions | |
temperature | +0.015 |
aback | +0.014 |
ray | +0.014 |
person | +0.014 |
over | +0.014 |
Neg logit contributions | |
yes | -0.028 |
goals | -0.027 |
best | -0.026 |
mm | -0.026 |
narrative | -0.024 |
Token | When |
Token ID | 2215 |
Feature activation | -5.25e-01 |
Pos logit contributions | |
temperature | +0.011 |
over | +0.011 |
aback | +0.011 |
ray | +0.011 |
person | +0.010 |
Neg logit contributions | |
yes | -0.016 |
goals | -0.015 |
best | -0.014 |
mm | -0.014 |
narrative | -0.013 |
Token | evening |
Token ID | 6180 |
Feature activation | -3.08e+00 |
Pos logit contributions | |
over | +0.012 |
ray | +0.012 |
aback | +0.011 |
person | +0.011 |
temperature | +0.011 |
Neg logit contributions | |
yes | -0.014 |
best | -0.013 |
goals | -0.012 |
mm | -0.012 |
narrative | -0.011 |
Token | came |
Token ID | 1625 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
ray | +0.092 |
plug | +0.090 |
any | +0.087 |
Grand | +0.086 |
shave | +0.085 |
Neg logit contributions | |
yes | -0.060 |
best | -0.054 |
mm | -0.048 |
are | -0.047 |
course | -0.044 |
Token | , |
Token ID | 11 |
Feature activation | -2.48e+00 |
Pos logit contributions | |
over | +0.025 |
late | +0.023 |
into | +0.022 |
To | +0.022 |
To | +0.022 |
Neg logit contributions | |
yes | -0.042 |
goals | -0.040 |
mm | -0.038 |
best | -0.038 |
narrative | -0.037 |
Token | Clara |
Token ID | 27443 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
ray | +0.054 |
temperature | +0.051 |
person | +0.051 |
over | +0.050 |
aback | +0.050 |
Neg logit contributions | |
yes | -0.069 |
best | -0.063 |
goals | -0.060 |
mm | -0.059 |
narrative | -0.055 |
Token | climbed |
Token ID | 19952 |
Feature activation | -3.64e-01 |
Pos logit contributions | |
late | +0.024 |
over | +0.024 |
To | +0.024 |
aback | +0.024 |
person | +0.023 |
Neg logit contributions | |
yes | -0.038 |
goals | -0.035 |
best | -0.035 |
mm | -0.033 |
ho | -0.032 |
Token | down |
Token ID | 866 |
Feature activation | -6.92e-01 |
Pos logit contributions | |
over | +0.005 |
into | +0.004 |
late | +0.004 |
aback | +0.004 |
temperature | +0.004 |
Neg logit contributions | |
yes | -0.013 |
goals | -0.013 |
mm | -0.012 |
best | -0.012 |
narrative | -0.012 |
Token | , |
Token ID | 11 |
Feature activation | -6.38e-01 |
Pos logit contributions | |
over | +0.004 |
To | +0.003 |
into | +0.003 |
aback | +0.003 |
late | +0.003 |
Neg logit contributions | |
narrative | -0.005 |
goals | -0.005 |
yes | -0.005 |
mm | -0.005 |
owers | -0.005 |
Token | " |
Token ID | 366 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
aback | +0.013 |
temperature | +0.013 |
over | +0.013 |
ray | +0.013 |
late | +0.012 |
Neg logit contributions | |
yes | -0.018 |
goals | -0.016 |
best | -0.016 |
mm | -0.015 |
ho | -0.015 |
Token | You |
Token ID | 1639 |
Feature activation | +4.21e-01 |
Pos logit contributions | |
yes | +0.029 |
goals | +0.027 |
best | +0.027 |
mm | +0.026 |
narrative | +0.025 |
Neg logit contributions | |
over | -0.020 |
temperature | -0.020 |
aback | -0.019 |
into | -0.019 |
late | -0.019 |
Token | are |
Token ID | 389 |
Feature activation | +9.85e-01 |
Pos logit contributions | |
yes | +0.006 |
best | +0.005 |
goals | +0.005 |
mm | +0.004 |
narrative | +0.004 |
Neg logit contributions | |
over | -0.014 |
late | -0.014 |
temperature | -0.014 |
person | -0.014 |
To | -0.014 |
Token | a |
Token ID | 257 |
Feature activation | -9.73e-01 |
Pos logit contributions | |
best | +0.025 |
yes | +0.024 |
goals | +0.022 |
mm | +0.020 |
course | +0.019 |
Neg logit contributions | |
ray | -0.024 |
person | -0.024 |
temperature | -0.024 |
To | -0.023 |
any | -0.022 |
Token | restless |
Token ID | 42537 |
Feature activation | -2.69e+00 |
Pos logit contributions | |
over | +0.027 |
late | +0.026 |
To | +0.026 |
Start | +0.025 |
temperature | +0.025 |
Neg logit contributions | |
yes | -0.020 |
goals | -0.018 |
mm | -0.017 |
best | -0.017 |
narrative | -0.016 |
Token | little |
Token ID | 1310 |
Feature activation | -7.76e-01 |
Pos logit contributions | |
over | +0.055 |
To | +0.053 |
To | +0.052 |
aback | +0.051 |
ray | +0.051 |
Neg logit contributions | |
yes | -0.074 |
goals | -0.070 |
best | -0.070 |
narrative | -0.063 |
mm | -0.063 |
Token | musician |
Token ID | 21623 |
Feature activation | -9.92e-03 |
Pos logit contributions | |
over | +0.017 |
late | +0.016 |
To | +0.016 |
temperature | +0.016 |
aback | +0.016 |
Neg logit contributions | |
yes | -0.021 |
goals | -0.019 |
best | -0.019 |
mm | -0.018 |
narrative | -0.018 |
Token | !" |
Token ID | 2474 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
over | +0.000 |
To | +0.000 |
late | +0.000 |
temperature | +0.000 |
aback | +0.000 |
Neg logit contributions | |
yes | -0.000 |
goals | -0.000 |
best | -0.000 |
mm | -0.000 |
narrative | -0.000 |
Token | Tim |
Token ID | 5045 |
Feature activation | +8.56e-01 |
Pos logit contributions | |
temperature | +0.000 |
person | +0.000 |
aback | +0.000 |
Grand | +0.000 |
shave | +0.000 |
Neg logit contributions | |
yes | -0.000 |
mm | -0.000 |
best | -0.000 |
course | -0.000 |
goals | -0.000 |
Token | my |
Token ID | 1820 |
Feature activation | +6.33e-01 |
Pos logit contributions | |
yes | +0.023 |
best | +0.022 |
goals | +0.021 |
mm | +0.020 |
ig | +0.020 |
Neg logit contributions | |
To | -0.018 |
late | -0.017 |
To | -0.017 |
Start | -0.017 |
person | -0.017 |
Token | that |
Token ID | 326 |
Feature activation | +6.18e-01 |
Pos logit contributions | |
over | +0.002 |
late | +0.002 |
into | +0.002 |
yelled | +0.002 |
temperature | +0.002 |
Neg logit contributions | |
yes | -0.005 |
goals | -0.005 |
narrative | -0.005 |
mm | -0.005 |
best | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | -8.28e-01 |
Pos logit contributions | |
yes | +0.019 |
goals | +0.018 |
mm | +0.017 |
best | +0.017 |
narrative | +0.017 |
Neg logit contributions | |
over | -0.011 |
late | -0.010 |
temperature | -0.010 |
into | -0.010 |
To | -0.010 |
Token | c |
Token ID | 269 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
over | +0.016 |
temperature | +0.016 |
late | +0.015 |
comm | +0.015 |
yelled | +0.015 |
Neg logit contributions | |
yes | -0.024 |
goals | -0.022 |
mm | -0.021 |
best | -0.020 |
narrative | -0.020 |
Token | actus |
Token ID | 34144 |
Feature activation | -3.12e-01 |
Pos logit contributions | |
over | +0.026 |
aback | +0.025 |
late | +0.024 |
To | +0.024 |
yelled | +0.024 |
Neg logit contributions | |
yes | -0.028 |
goals | -0.027 |
best | -0.026 |
mm | -0.026 |
ho | -0.024 |
Token | was |
Token ID | 373 |
Feature activation | +5.36e-01 |
Pos logit contributions | |
over | +0.005 |
into | +0.004 |
temperature | +0.004 |
asked | +0.004 |
late | +0.004 |
Neg logit contributions | |
yes | -0.010 |
goals | -0.010 |
mm | -0.010 |
narrative | -0.010 |
best | -0.010 |
Token | dro |
Token ID | 3102 |
Feature activation | -2.65e+00 |
Pos logit contributions | |
yes | +0.014 |
goals | +0.013 |
best | +0.013 |
mm | +0.012 |
narrative | +0.012 |
Neg logit contributions | |
over | -0.012 |
To | -0.011 |
late | -0.011 |
temperature | -0.011 |
aback | -0.011 |
Token | opy |
Token ID | 11081 |
Feature activation | -8.53e-01 |
Pos logit contributions | |
over | +0.070 |
To | +0.067 |
To | +0.067 |
temperature | +0.067 |
aback | +0.066 |
Neg logit contributions | |
yes | -0.056 |
goals | -0.051 |
best | -0.050 |
mm | -0.049 |
narrative | -0.046 |
Token | and |
Token ID | 290 |
Feature activation | -7.76e-01 |
Pos logit contributions | |
over | +0.011 |
into | +0.010 |
temperature | +0.009 |
late | +0.009 |
towards | +0.008 |
Neg logit contributions | |
yes | -0.032 |
mm | -0.032 |
narrative | -0.030 |
Deal | -0.029 |
goals | -0.029 |
Token | brown |
Token ID | 7586 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
over | +0.015 |
late | +0.014 |
into | +0.014 |
yelled | +0.014 |
temperature | +0.014 |
Neg logit contributions | |
yes | -0.022 |
goals | -0.021 |
mm | -0.021 |
best | -0.021 |
narrative | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -2.69e-01 |
Pos logit contributions | |
yes | +0.005 |
mm | +0.005 |
narrative | +0.005 |
goals | +0.005 |
Deal | +0.005 |
Neg logit contributions | |
over | -0.001 |
into | -0.001 |
temperature | -0.001 |
towards | -0.001 |
aback | -0.001 |
Token | She |
Token ID | 1375 |
Feature activation | -4.71e-01 |
Pos logit contributions | |
over | +0.005 |
temperature | +0.005 |
late | +0.005 |
To | +0.005 |
aback | +0.005 |
Neg logit contributions | |
yes | -0.008 |
goals | -0.007 |
best | -0.007 |
mm | -0.007 |
narrative | -0.007 |