Token | want |
Token ID | 765 |
Feature activation | -7.89e-01 |
Pos logit contributions | |
chalk | +0.033 |
ch | +0.031 |
ink | +0.030 |
bar | +0.030 |
wind | +0.027 |
Neg logit contributions | |
Jacob | -0.062 |
Megan | -0.060 |
Jacob | -0.057 |
Pen | -0.055 |
restaurant | -0.055 |
Token | . |
Token ID | 13 |
Feature activation | -2.35e-02 |
Pos logit contributions | |
Jacob | +0.057 |
Megan | +0.056 |
restaurant | +0.056 |
Pen | +0.049 |
Nick | +0.049 |
Neg logit contributions | |
bar | -0.062 |
itted | -0.058 |
bullet | -0.057 |
ch | -0.056 |
gol | -0.056 |
Token | They |
Token ID | 1119 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
Jacob | +0.002 |
restaurant | +0.002 |
Megan | +0.002 |
Jacob | +0.002 |
Pen | +0.002 |
Neg logit contributions | |
chalk | -0.001 |
ink | -0.001 |
bar | -0.001 |
itted | -0.001 |
Bounce | -0.001 |
Token | have |
Token ID | 423 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
Jacob | +0.011 |
Megan | +0.010 |
Jacob | +0.010 |
restaurant | +0.010 |
Pen | +0.010 |
Neg logit contributions | |
chalk | -0.005 |
ink | -0.005 |
gol | -0.004 |
arrow | -0.004 |
sun | -0.004 |
Token | more |
Token ID | 517 |
Feature activation | +9.29e-01 |
Pos logit contributions | |
Jacob | +0.018 |
Megan | +0.017 |
restaurant | +0.017 |
Jacob | +0.017 |
Pen | +0.017 |
Neg logit contributions | |
chalk | -0.009 |
ink | -0.009 |
bar | -0.008 |
cho | -0.008 |
itted | -0.008 |
Token | chalk |
Token ID | 30860 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
bar | +0.078 |
plets | +0.071 |
cho | +0.070 |
ch | +0.067 |
trigger | +0.066 |
Neg logit contributions | |
restaurant | -0.066 |
Jacob | -0.060 |
Megan | -0.057 |
foods | -0.055 |
pic | -0.053 |
Token | and |
Token ID | 290 |
Feature activation | -9.65e-02 |
Pos logit contributions | |
bar | +0.693 |
ink | +0.680 |
chalk | +0.679 |
bullet | +0.663 |
ch | +0.659 |
Neg logit contributions | |
Jacob | -0.816 |
Megan | -0.784 |
restaurant | -0.771 |
Jacob | -0.714 |
Pen | -0.714 |
Token | more |
Token ID | 517 |
Feature activation | +9.27e-01 |
Pos logit contributions | |
Jacob | +0.008 |
Megan | +0.007 |
restaurant | +0.007 |
Pen | +0.007 |
Jacob | +0.007 |
Neg logit contributions | |
bar | -0.007 |
ink | -0.007 |
ch | -0.006 |
chalk | -0.006 |
bullet | -0.006 |
Token | sidewalk |
Token ID | 24058 |
Feature activation | +5.34e+00 |
Pos logit contributions | |
bar | +0.084 |
plets | +0.075 |
ch | +0.074 |
cho | +0.073 |
peac | +0.071 |
Neg logit contributions | |
restaurant | -0.058 |
Jacob | -0.054 |
Megan | -0.050 |
foods | -0.046 |
searches | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | -2.93e-02 |
Pos logit contributions | |
cho | +0.322 |
bar | +0.319 |
chalk | +0.313 |
ink | +0.308 |
itted | +0.296 |
Neg logit contributions | |
Jacob | -0.455 |
Megan | -0.438 |
Jacob | -0.430 |
restaurant | -0.427 |
Pen | -0.424 |
Token | Sara |
Token ID | 24799 |
Feature activation | -2.67e+00 |
Pos logit contributions | |
Jacob | +0.003 |
restaurant | +0.003 |
Megan | +0.003 |
Jacob | +0.002 |
abel | +0.002 |
Neg logit contributions | |
ink | -0.002 |
chalk | -0.002 |
itted | -0.002 |
ch | -0.001 |
sun | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | +2.77e-01 |
Pos logit contributions | |
Jacob | +0.558 |
restaurant | +0.538 |
Megan | +0.535 |
Jacob | +0.516 |
Pen | +0.488 |
Neg logit contributions | |
bar | -0.435 |
ink | -0.435 |
chalk | -0.428 |
cho | -0.426 |
ch | -0.413 |
Token | let |
Token ID | 1309 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
bar | +0.018 |
chalk | +0.018 |
ink | +0.018 |
bullet | +0.017 |
ch | +0.017 |
Neg logit contributions | |
Jacob | -0.023 |
Megan | -0.022 |
restaurant | -0.022 |
Pen | -0.021 |
Jacob | -0.020 |
Token | 's |
Token ID | 338 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
ink | +0.011 |
bar | +0.011 |
chalk | +0.011 |
ch | +0.011 |
cho | +0.011 |
Neg logit contributions | |
Jacob | -0.015 |
Megan | -0.015 |
restaurant | -0.015 |
Jacob | -0.014 |
Pen | -0.013 |
Token | cut |
Token ID | 2005 |
Feature activation | +2.37e+00 |
Pos logit contributions | |
Jacob | +0.084 |
Megan | +0.080 |
restaurant | +0.078 |
Pen | +0.073 |
Jacob | +0.073 |
Neg logit contributions | |
bar | -0.078 |
itted | -0.077 |
ink | -0.076 |
bullet | -0.075 |
ch | -0.075 |
Token | the |
Token ID | 262 |
Feature activation | -6.71e-01 |
Pos logit contributions | |
gol | +0.171 |
bullet | +0.170 |
itted | +0.169 |
peac | +0.167 |
bar | +0.163 |
Neg logit contributions | |
Jacob | -0.184 |
Megan | -0.179 |
restaurant | -0.160 |
Pen | -0.159 |
Nick | -0.156 |
Token | chalk |
Token ID | 30860 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
restaurant | +0.057 |
chamber | +0.053 |
foods | +0.052 |
Jacob | +0.051 |
shelf | +0.051 |
Neg logit contributions | |
cho | -0.041 |
ug | -0.041 |
ch | -0.038 |
plets | -0.036 |
br | -0.035 |
Token | ! |
Token ID | 0 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
bar | +0.718 |
chalk | +0.699 |
ink | +0.695 |
ch | +0.661 |
cho | +0.654 |
Neg logit contributions | |
Jacob | -0.820 |
Megan | -0.793 |
restaurant | -0.772 |
Pen | -0.737 |
Jacob | -0.733 |
Token | It |
Token ID | 632 |
Feature activation | +6.30e-01 |
Pos logit contributions | |
ink | +0.019 |
chalk | +0.019 |
bar | +0.019 |
bullet | +0.018 |
cho | +0.018 |
Neg logit contributions | |
Jacob | -0.026 |
Megan | -0.025 |
Pen | -0.023 |
restaurant | -0.023 |
Jacob | -0.023 |
Token | 'll |
Token ID | 1183 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
bar | +0.047 |
ink | +0.045 |
chalk | +0.045 |
ch | +0.044 |
cho | +0.044 |
Neg logit contributions | |
Jacob | -0.048 |
Megan | -0.046 |
restaurant | -0.046 |
Jacob | -0.043 |
Pen | -0.042 |
Token | be |
Token ID | 307 |
Feature activation | +2.74e+00 |
Pos logit contributions | |
Jacob | +0.123 |
Megan | +0.121 |
restaurant | +0.113 |
Pen | +0.109 |
Nick | +0.106 |
Neg logit contributions | |
itted | -0.131 |
gol | -0.125 |
bullet | -0.123 |
chalk | -0.122 |
bar | -0.121 |
Token | fun |
Token ID | 1257 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
gol | +0.214 |
bullet | +0.211 |
itted | +0.210 |
bar | +0.202 |
peac | +0.201 |
Neg logit contributions | |
Jacob | -0.184 |
Megan | -0.179 |
restaurant | -0.172 |
Frost | -0.158 |
Pen | -0.154 |
Token | She |
Token ID | 1375 |
Feature activation | -6.08e-01 |
Pos logit contributions | |
Jacob | +0.002 |
restaurant | +0.002 |
Megan | +0.002 |
Jacob | +0.002 |
Pen | +0.002 |
Neg logit contributions | |
ink | -0.001 |
chalk | -0.001 |
bar | -0.001 |
Bounce | -0.001 |
itted | -0.001 |
Token | loved |
Token ID | 6151 |
Feature activation | +7.36e-01 |
Pos logit contributions | |
Jacob | +0.047 |
Megan | +0.045 |
restaurant | +0.044 |
Pen | +0.042 |
Jacob | +0.042 |
Neg logit contributions | |
bar | -0.044 |
chalk | -0.043 |
ink | -0.043 |
bullet | -0.041 |
itted | -0.041 |
Token | to |
Token ID | 284 |
Feature activation | +5.78e-01 |
Pos logit contributions | |
bar | +0.046 |
chalk | +0.044 |
ink | +0.044 |
bullet | +0.043 |
itted | +0.042 |
Neg logit contributions | |
Jacob | -0.064 |
Megan | -0.062 |
restaurant | -0.062 |
Pen | -0.058 |
Jacob | -0.057 |
Token | draw |
Token ID | 3197 |
Feature activation | +4.50e+00 |
Pos logit contributions | |
bar | +0.044 |
chalk | +0.044 |
ink | +0.043 |
bullet | +0.041 |
ch | +0.041 |
Neg logit contributions | |
Jacob | -0.043 |
Megan | -0.041 |
restaurant | -0.040 |
Jacob | -0.037 |
Pen | -0.037 |
Token | with |
Token ID | 351 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
bar | +0.352 |
gol | +0.334 |
bullet | +0.334 |
itted | +0.326 |
crashing | +0.319 |
Neg logit contributions | |
Jacob | -0.315 |
restaurant | -0.310 |
Megan | -0.298 |
Jacob | -0.270 |
Pen | -0.267 |
Token | chalk |
Token ID | 30860 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
restaurant | +0.052 |
cereal | +0.046 |
foods | +0.046 |
Jacob | +0.045 |
Megan | +0.042 |
Neg logit contributions | |
plets | -0.186 |
bar | -0.178 |
EEP | -0.176 |
gol | -0.174 |
plet | -0.173 |
Token | on |
Token ID | 319 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
bullet | +0.876 |
crashing | +0.863 |
bar | +0.861 |
gol | +0.853 |
upid | +0.838 |
Neg logit contributions | |
Jacob | -0.588 |
restaurant | -0.581 |
Megan | -0.543 |
abel | -0.488 |
ide | -0.479 |
Token | the |
Token ID | 262 |
Feature activation | -6.73e-01 |
Pos logit contributions | |
bar | +0.118 |
itted | +0.115 |
gol | +0.110 |
bullet | +0.109 |
ch | +0.108 |
Neg logit contributions | |
Jacob | -0.092 |
restaurant | -0.086 |
Megan | -0.086 |
Pen | -0.077 |
Jacob | -0.075 |
Token | sidewalk |
Token ID | 24058 |
Feature activation | +5.34e+00 |
Pos logit contributions | |
restaurant | +0.051 |
Jacob | +0.043 |
chamber | +0.042 |
section | +0.041 |
shelf | +0.040 |
Neg logit contributions | |
itted | -0.052 |
isted | -0.049 |
ged | -0.049 |
bar | -0.049 |
isel | -0.049 |
Token | outside |
Token ID | 2354 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
bar | +0.387 |
chalk | +0.383 |
ink | +0.380 |
itted | +0.356 |
ch | +0.353 |
Neg logit contributions | |
Jacob | -0.413 |
Megan | -0.396 |
restaurant | -0.383 |
Jacob | -0.373 |
Pen | -0.371 |
Token | her |
Token ID | 607 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
Jacob | +0.091 |
Megan | +0.087 |
restaurant | +0.085 |
Jacob | +0.082 |
Pen | +0.081 |
Neg logit contributions | |
chalk | -0.069 |
bar | -0.069 |
ink | -0.069 |
ch | -0.065 |
bullet | -0.064 |
Token | laughed |
Token ID | 13818 |
Feature activation | +9.58e-01 |
Pos logit contributions | |
bar | +0.053 |
chalk | +0.053 |
ink | +0.052 |
ch | +0.049 |
cho | +0.049 |
Neg logit contributions | |
Jacob | -0.057 |
Megan | -0.056 |
restaurant | -0.055 |
Pen | -0.051 |
Jacob | -0.051 |
Token | and |
Token ID | 290 |
Feature activation | -1.00e-01 |
Pos logit contributions | |
bar | +0.077 |
chalk | +0.075 |
ink | +0.075 |
bullet | +0.074 |
ch | +0.073 |
Neg logit contributions | |
Jacob | -0.067 |
Megan | -0.063 |
restaurant | -0.061 |
Pen | -0.057 |
Jacob | -0.057 |
Token | he |
Token ID | 339 |
Feature activation | +2.94e-01 |
Pos logit contributions | |
Jacob | +0.008 |
Megan | +0.007 |
Nick | +0.007 |
Pen | +0.006 |
restaurant | +0.006 |
Neg logit contributions | |
bar | -0.008 |
bullet | -0.008 |
itted | -0.007 |
ink | -0.007 |
ch | -0.007 |
Token | took |
Token ID | 1718 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
bar | +0.019 |
chalk | +0.019 |
ink | +0.019 |
ch | +0.018 |
cho | +0.018 |
Neg logit contributions | |
Jacob | -0.025 |
Megan | -0.024 |
restaurant | -0.023 |
Pen | -0.022 |
Jacob | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | -6.77e-01 |
Pos logit contributions | |
bar | +0.137 |
peac | +0.136 |
cho | +0.128 |
bullet | +0.127 |
itted | +0.125 |
Neg logit contributions | |
Jacob | -0.133 |
Megan | -0.124 |
restaurant | -0.117 |
Nick | -0.112 |
Jacob | -0.110 |
Token | chalk |
Token ID | 30860 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
restaurant | +0.056 |
search | +0.052 |
Jacob | +0.050 |
Megan | +0.050 |
drinks | +0.048 |
Neg logit contributions | |
cho | -0.043 |
ug | -0.042 |
Bree | -0.038 |
bar | -0.038 |
Cle | -0.038 |
Token | from |
Token ID | 422 |
Feature activation | -8.51e-01 |
Pos logit contributions | |
bullet | +0.747 |
ink | +0.747 |
bar | +0.740 |
ch | +0.738 |
cho | +0.738 |
Neg logit contributions | |
Jacob | -0.737 |
Megan | -0.705 |
restaurant | -0.701 |
Pen | -0.629 |
Jacob | -0.623 |
Token | Mum |
Token ID | 38367 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
Jacob | +0.069 |
Megan | +0.066 |
restaurant | +0.065 |
Pen | +0.063 |
Jacob | +0.063 |
Neg logit contributions | |
chalk | -0.058 |
ink | -0.057 |
bar | -0.054 |
bullet | -0.052 |
Bounce | -0.051 |
Token | . |
Token ID | 13 |
Feature activation | -2.64e-02 |
Pos logit contributions | |
Jacob | +0.140 |
Megan | +0.135 |
restaurant | +0.131 |
Jacob | +0.124 |
Pen | +0.124 |
Neg logit contributions | |
bar | -0.137 |
chalk | -0.136 |
ink | -0.136 |
ch | -0.129 |
bullet | -0.129 |
Token | He |
Token ID | 679 |
Feature activation | -5.85e-01 |
Pos logit contributions | |
Jacob | +0.002 |
Megan | +0.002 |
restaurant | +0.002 |
Jacob | +0.002 |
Pen | +0.002 |
Neg logit contributions | |
bar | -0.002 |
ink | -0.002 |
cho | -0.002 |
chalk | -0.002 |
bullet | -0.002 |
Token | started |
Token ID | 2067 |
Feature activation | +2.60e+00 |
Pos logit contributions | |
Jacob | +0.047 |
Megan | +0.046 |
restaurant | +0.045 |
Pen | +0.043 |
Jacob | +0.042 |
Neg logit contributions | |
bar | -0.039 |
chalk | -0.039 |
ink | -0.039 |
bullet | -0.037 |
ch | -0.037 |
Token | about |
Token ID | 546 |
Feature activation | -3.11e+00 |
Pos logit contributions | |
bar | +0.036 |
bullet | +0.034 |
ink | +0.034 |
chalk | +0.034 |
ch | +0.033 |
Neg logit contributions | |
Jacob | -0.033 |
Megan | -0.032 |
restaurant | -0.031 |
Pen | -0.029 |
Jacob | -0.028 |
Token | numbers |
Token ID | 3146 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
Jacob | +0.260 |
Megan | +0.250 |
restaurant | +0.245 |
Pen | +0.231 |
Jacob | +0.228 |
Neg logit contributions | |
bar | -0.204 |
ink | -0.195 |
chalk | -0.194 |
itted | -0.192 |
ch | -0.192 |
Token | with |
Token ID | 351 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
bar | +0.074 |
ink | +0.074 |
chalk | +0.074 |
ch | +0.069 |
cho | +0.069 |
Neg logit contributions | |
Jacob | -0.081 |
Megan | -0.079 |
restaurant | -0.076 |
Jacob | -0.073 |
Pen | -0.073 |
Token | their |
Token ID | 511 |
Feature activation | +9.97e-01 |
Pos logit contributions | |
Jacob | +0.132 |
Megan | +0.127 |
restaurant | +0.124 |
Pen | +0.116 |
Jacob | +0.116 |
Neg logit contributions | |
bar | -0.106 |
ink | -0.100 |
bullet | -0.100 |
chalk | -0.100 |
ch | -0.100 |
Token | new |
Token ID | 649 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
bar | +0.069 |
ink | +0.066 |
chalk | +0.066 |
ch | +0.065 |
bullet | +0.064 |
Neg logit contributions | |
Jacob | -0.080 |
Megan | -0.077 |
restaurant | -0.076 |
Pen | -0.071 |
Jacob | -0.070 |
Token | chalk |
Token ID | 30860 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
bar | +0.105 |
ch | +0.100 |
bullet | +0.099 |
ink | +0.099 |
cho | +0.099 |
Neg logit contributions | |
Jacob | -0.131 |
restaurant | -0.128 |
Megan | -0.125 |
Pen | -0.116 |
Jacob | -0.114 |
Token | . |
Token ID | 13 |
Feature activation | -2.37e-02 |
Pos logit contributions | |
bar | +0.766 |
chalk | +0.759 |
ink | +0.758 |
ch | +0.717 |
bullet | +0.714 |
Neg logit contributions | |
Jacob | -0.757 |
Megan | -0.731 |
restaurant | -0.705 |
Pen | -0.671 |
Jacob | -0.670 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +8.56e-01 |
Pos logit contributions | |
Jacob | +0.002 |
Megan | +0.002 |
restaurant | +0.002 |
Pen | +0.002 |
Jacob | +0.002 |
Neg logit contributions | |
bar | -0.002 |
ink | -0.002 |
chalk | -0.002 |
bullet | -0.002 |
ch | -0.002 |
Token | Once |
Token ID | 7454 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
ch | +0.039 |
chalk | +0.038 |
ink | +0.038 |
Bounce | +0.035 |
sun | +0.034 |
Neg logit contributions | |
Jacob | -0.084 |
Megan | -0.084 |
restaurant | -0.081 |
abel | -0.080 |
Pen | -0.077 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
Jacob | +0.113 |
Megan | +0.109 |
restaurant | +0.105 |
Pen | +0.097 |
Jacob | +0.096 |
Neg logit contributions | |
bar | -0.142 |
ink | -0.141 |
chalk | -0.140 |
bullet | -0.137 |
ch | -0.135 |
Token | a |
Token ID | 257 |
Feature activation | +4.30e-01 |
Pos logit contributions | |
Jacob | +0.017 |
Megan | +0.017 |
restaurant | +0.016 |
Pen | +0.016 |
Jacob | +0.016 |
Neg logit contributions | |
chalk | -0.010 |
ink | -0.010 |
bar | -0.010 |
ch | -0.010 |
bullet | -0.009 |
Token | gave |
Token ID | 2921 |
Feature activation | +2.76e-01 |
Pos logit contributions | |
chalk | +0.073 |
ink | +0.073 |
bar | +0.072 |
cho | +0.069 |
ch | +0.067 |
Neg logit contributions | |
Jacob | -0.101 |
Megan | -0.097 |
restaurant | -0.096 |
Jacob | -0.092 |
Pen | -0.091 |
Token | Ben |
Token ID | 3932 |
Feature activation | +3.12e-01 |
Pos logit contributions | |
bar | +0.021 |
ink | +0.020 |
chalk | +0.020 |
bullet | +0.020 |
itted | +0.020 |
Neg logit contributions | |
Jacob | -0.021 |
Megan | -0.020 |
restaurant | -0.019 |
Pen | -0.018 |
Jacob | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | -3.83e-02 |
Pos logit contributions | |
ink | +0.017 |
chalk | +0.017 |
bar | +0.016 |
ch | +0.016 |
cho | +0.015 |
Neg logit contributions | |
Jacob | -0.029 |
Megan | -0.029 |
restaurant | -0.028 |
Jacob | -0.027 |
Pen | -0.027 |
Token | Lily |
Token ID | 20037 |
Feature activation | +6.52e-01 |
Pos logit contributions | |
Jacob | +0.003 |
Megan | +0.003 |
restaurant | +0.003 |
Pen | +0.003 |
Jacob | +0.002 |
Neg logit contributions | |
bar | -0.003 |
itted | -0.003 |
ch | -0.003 |
chalk | -0.003 |
ink | -0.003 |
Token | their |
Token ID | 511 |
Feature activation | +9.98e-01 |
Pos logit contributions | |
bar | +0.040 |
chalk | +0.040 |
ink | +0.040 |
ch | +0.037 |
bullet | +0.037 |
Neg logit contributions | |
Jacob | -0.057 |
Megan | -0.056 |
restaurant | -0.054 |
Jacob | -0.052 |
Pen | -0.052 |
Token | chalk |
Token ID | 30860 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
bar | +0.064 |
ink | +0.062 |
chalk | +0.062 |
ch | +0.060 |
bullet | +0.060 |
Neg logit contributions | |
Jacob | -0.084 |
Megan | -0.081 |
restaurant | -0.080 |
Pen | -0.075 |
Jacob | -0.074 |
Token | and |
Token ID | 290 |
Feature activation | -9.98e-02 |
Pos logit contributions | |
ch | +0.825 |
Bree | +0.811 |
Bounce | +0.800 |
bullet | +0.797 |
bar | +0.796 |
Neg logit contributions | |
Jacob | -0.664 |
Megan | -0.627 |
restaurant | -0.621 |
Pen | -0.546 |
Jacob | -0.544 |
Token | walked |
Token ID | 6807 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
Jacob | +0.008 |
Megan | +0.007 |
restaurant | +0.007 |
Pen | +0.007 |
Nick | +0.006 |
Neg logit contributions | |
bar | -0.007 |
ch | -0.007 |
itted | -0.007 |
chalk | -0.007 |
crashing | -0.007 |
Token | them |
Token ID | 606 |
Feature activation | +6.19e-01 |
Pos logit contributions | |
Jacob | +0.128 |
Megan | +0.124 |
restaurant | +0.121 |
Pen | +0.115 |
Jacob | +0.114 |
Neg logit contributions | |
bar | -0.090 |
chalk | -0.089 |
ink | -0.089 |
bullet | -0.084 |
ch | -0.084 |
Token | back |
Token ID | 736 |
Feature activation | +8.61e-01 |
Pos logit contributions | |
bar | +0.037 |
chalk | +0.036 |
ink | +0.036 |
itted | +0.034 |
cho | +0.034 |
Neg logit contributions | |
Jacob | -0.055 |
Megan | -0.053 |
restaurant | -0.052 |
Jacob | -0.050 |
Pen | -0.050 |
Token | to |
Token ID | 284 |
Feature activation | +5.73e-01 |
Pos logit contributions | |
bar | +0.043 |
chalk | +0.042 |
ink | +0.040 |
itted | +0.039 |
cho | +0.039 |
Neg logit contributions | |
Jacob | -0.085 |
Megan | -0.082 |
Jacob | -0.080 |
restaurant | -0.080 |
Pen | -0.076 |
Token | with |
Token ID | 351 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
Jacob | +0.070 |
Megan | +0.066 |
restaurant | +0.065 |
Pen | +0.059 |
Jacob | +0.059 |
Neg logit contributions | |
bar | -0.080 |
chalk | -0.080 |
ink | -0.080 |
bullet | -0.079 |
ch | -0.078 |
Token | a |
Token ID | 257 |
Feature activation | +4.46e-01 |
Pos logit contributions | |
Jacob | +0.124 |
Megan | +0.115 |
restaurant | +0.112 |
Pen | +0.103 |
Nick | +0.103 |
Neg logit contributions | |
bar | -0.120 |
EEP | -0.118 |
gol | -0.115 |
ch | -0.115 |
plets | -0.114 |
Token | piece |
Token ID | 3704 |
Feature activation | -5.90e-01 |
Pos logit contributions | |
itted | +0.031 |
bar | +0.031 |
ully | +0.030 |
Bounce | +0.030 |
ug | +0.029 |
Neg logit contributions | |
restaurant | -0.034 |
Jacob | -0.033 |
Megan | -0.030 |
chamber | -0.029 |
stall | -0.029 |
Token | of |
Token ID | 286 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
Jacob | +0.044 |
Megan | +0.042 |
restaurant | +0.042 |
Pen | +0.038 |
abel | +0.037 |
Neg logit contributions | |
gol | -0.040 |
Bounce | -0.039 |
bullet | -0.039 |
bar | -0.038 |
peac | -0.038 |
Token | white |
Token ID | 2330 |
Feature activation | +4.54e-01 |
Pos logit contributions | |
bar | +0.084 |
gol | +0.079 |
itted | +0.079 |
EEP | +0.078 |
cho | +0.075 |
Neg logit contributions | |
Jacob | -0.120 |
restaurant | -0.113 |
Megan | -0.113 |
Pen | -0.104 |
Eva | -0.101 |
Token | chalk |
Token ID | 30860 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
Bounce | +0.027 |
bar | +0.027 |
gol | +0.026 |
cho | +0.026 |
Bree | +0.026 |
Neg logit contributions | |
restaurant | -0.038 |
Jacob | -0.038 |
Megan | -0.035 |
foods | -0.034 |
Pen | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
bar | +0.800 |
bullet | +0.787 |
chalk | +0.785 |
ink | +0.782 |
cho | +0.775 |
Neg logit contributions | |
Jacob | -0.698 |
Megan | -0.656 |
restaurant | -0.649 |
Pen | -0.596 |
Jacob | -0.587 |
Token | Then |
Token ID | 3244 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
Jacob | +0.001 |
Megan | +0.001 |
restaurant | +0.001 |
Pen | +0.001 |
Jacob | +0.001 |
Neg logit contributions | |
bullet | -0.001 |
bar | -0.001 |
ink | -0.001 |
cho | -0.001 |
chalk | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
bar | +0.207 |
itted | +0.198 |
bullet | +0.197 |
chalk | +0.195 |
ink | +0.193 |
Neg logit contributions | |
Jacob | -0.186 |
Megan | -0.180 |
restaurant | -0.167 |
Pen | -0.162 |
Nick | -0.155 |
Token | she |
Token ID | 673 |
Feature activation | +9.39e-01 |
Pos logit contributions | |
bar | +0.021 |
itted | +0.020 |
bullet | +0.020 |
alks | +0.019 |
ink | +0.019 |
Neg logit contributions | |
Jacob | -0.022 |
Megan | -0.022 |
restaurant | -0.020 |
Pen | -0.020 |
Jacob | -0.018 |
Token | turned |
Token ID | 2900 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
bar | +0.055 |
chalk | +0.054 |
ink | +0.054 |
ch | +0.051 |
bullet | +0.051 |
Neg logit contributions | |
Jacob | -0.084 |
Megan | -0.081 |
restaurant | -0.080 |
Jacob | -0.076 |
Pen | -0.076 |
Token | , |
Token ID | 11 |
Feature activation | +2.76e-01 |
Pos logit contributions | |
Jacob | +0.125 |
Megan | +0.122 |
restaurant | +0.115 |
Pen | +0.110 |
Nick | +0.106 |
Neg logit contributions | |
bar | -0.125 |
chalk | -0.120 |
bullet | -0.118 |
ink | -0.117 |
ch | -0.116 |
Token | they |
Token ID | 484 |
Feature activation | +6.18e-01 |
Pos logit contributions | |
bar | +0.019 |
bullet | +0.018 |
itted | +0.018 |
ch | +0.018 |
chalk | +0.017 |
Neg logit contributions | |
Jacob | -0.023 |
Megan | -0.023 |
restaurant | -0.022 |
Pen | -0.021 |
Nick | -0.021 |
Token | looked |
Token ID | 3114 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
bar | +0.050 |
ink | +0.049 |
ch | +0.049 |
bullet | +0.049 |
cho | +0.049 |
Neg logit contributions | |
Jacob | -0.039 |
Megan | -0.037 |
restaurant | -0.037 |
Pen | -0.033 |
Jacob | -0.033 |
Token | for |
Token ID | 329 |
Feature activation | +5.51e-01 |
Pos logit contributions | |
Jacob | +0.117 |
Megan | +0.113 |
restaurant | +0.109 |
Jacob | +0.104 |
Pen | +0.104 |
Neg logit contributions | |
bar | -0.099 |
chalk | -0.099 |
ink | -0.099 |
bullet | -0.094 |
ch | -0.094 |
Token | the |
Token ID | 262 |
Feature activation | -6.84e-01 |
Pos logit contributions | |
bar | +0.045 |
ch | +0.043 |
itted | +0.042 |
cho | +0.041 |
br | +0.040 |
Neg logit contributions | |
Jacob | -0.039 |
Megan | -0.038 |
restaurant | -0.037 |
Nick | -0.033 |
Pen | -0.033 |
Token | chalk |
Token ID | 30860 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
restaurant | +0.060 |
search | +0.055 |
chamber | +0.055 |
Jacob | +0.055 |
foods | +0.052 |
Neg logit contributions | |
cho | -0.042 |
ug | -0.041 |
plets | -0.040 |
ch | -0.040 |
br | -0.037 |
Token | and |
Token ID | 290 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
bar | +0.727 |
chalk | +0.717 |
ink | +0.716 |
bullet | +0.680 |
ch | +0.674 |
Neg logit contributions | |
Jacob | -0.803 |
Megan | -0.777 |
restaurant | -0.752 |
Pen | -0.715 |
Jacob | -0.715 |
Token | finally |
Token ID | 3443 |
Feature activation | -7.79e-01 |
Pos logit contributions | |
Jacob | +0.007 |
Megan | +0.007 |
restaurant | +0.006 |
Nick | +0.006 |
Pen | +0.006 |
Neg logit contributions | |
bar | -0.008 |
itted | -0.008 |
ch | -0.008 |
br | -0.008 |
bullet | -0.008 |
Token | found |
Token ID | 1043 |
Feature activation | -3.50e-01 |
Pos logit contributions | |
Jacob | +0.057 |
Megan | +0.055 |
restaurant | +0.053 |
Pen | +0.049 |
Jacob | +0.049 |
Neg logit contributions | |
bar | -0.059 |
ink | -0.057 |
chalk | -0.057 |
ch | -0.057 |
bullet | -0.056 |
Token | it |
Token ID | 340 |
Feature activation | +3.79e-01 |
Pos logit contributions | |
Jacob | +0.026 |
Megan | +0.025 |
Nick | +0.022 |
restaurant | +0.022 |
Pen | +0.022 |
Neg logit contributions | |
bar | -0.027 |
ch | -0.026 |
itted | -0.026 |
bullet | -0.026 |
gol | -0.025 |
Token | under |
Token ID | 739 |
Feature activation | -3.12e+00 |
Pos logit contributions | |
ink | +0.026 |
bar | +0.026 |
chalk | +0.026 |
Bounce | +0.024 |
cho | +0.024 |
Neg logit contributions | |
Jacob | -0.031 |
Megan | -0.030 |
restaurant | -0.029 |
Jacob | -0.029 |
Pen | -0.028 |
Token | It |
Token ID | 632 |
Feature activation | +6.31e-01 |
Pos logit contributions | |
restaurant | +0.003 |
ide | +0.003 |
Jacob | +0.003 |
abel | +0.003 |
Megan | +0.003 |
Neg logit contributions | |
sun | -0.000 |
ink | -0.000 |
bullet | -0.000 |
ch | -0.000 |
itted | -0.000 |
Token | eats |
Token ID | 25365 |
Feature activation | -4.68e+00 |
Pos logit contributions | |
ch | +0.041 |
cho | +0.038 |
ink | +0.036 |
bullet | +0.036 |
chalk | +0.035 |
Neg logit contributions | |
Jacob | -0.052 |
Megan | -0.050 |
restaurant | -0.049 |
Jacob | -0.048 |
ide | -0.048 |
Token | some |
Token ID | 617 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
Jacob | +0.414 |
Megan | +0.401 |
restaurant | +0.390 |
Jacob | +0.373 |
Pen | +0.373 |
Neg logit contributions | |
bar | -0.282 |
ink | -0.281 |
chalk | -0.281 |
bullet | -0.266 |
ch | -0.265 |
Token | of |
Token ID | 286 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
restaurant | +0.127 |
Jacob | +0.119 |
Megan | +0.118 |
foods | +0.104 |
Nick | +0.103 |
Neg logit contributions | |
bar | -0.116 |
Bree | -0.105 |
itted | -0.104 |
ully | -0.104 |
peac | -0.102 |
Token | the |
Token ID | 262 |
Feature activation | -6.84e-01 |
Pos logit contributions | |
bar | +0.098 |
itted | +0.096 |
ch | +0.092 |
gol | +0.092 |
bullet | +0.091 |
Neg logit contributions | |
Jacob | -0.103 |
Megan | -0.100 |
restaurant | -0.093 |
Pen | -0.091 |
Nick | -0.089 |
Token | chalk |
Token ID | 30860 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
restaurant | +0.072 |
foods | +0.065 |
cafe | +0.062 |
Jacob | +0.062 |
chamber | +0.062 |
Neg logit contributions | |
bar | -0.032 |
cho | -0.031 |
ug | -0.027 |
plets | -0.027 |
itted | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
ink | +0.737 |
bar | +0.734 |
chalk | +0.732 |
bullet | +0.684 |
ch | +0.682 |
Neg logit contributions | |
Jacob | -0.796 |
Megan | -0.774 |
restaurant | -0.738 |
Jacob | -0.718 |
Pen | -0.716 |
Token | It |
Token ID | 632 |
Feature activation | +6.32e-01 |
Pos logit contributions | |
restaurant | +0.003 |
Jacob | +0.003 |
Megan | +0.003 |
ide | +0.002 |
Gor | +0.002 |
Neg logit contributions | |
ink | -0.001 |
ch | -0.001 |
chalk | -0.001 |
sun | -0.001 |
itted | -0.001 |
Token | makes |
Token ID | 1838 |
Feature activation | -4.19e-01 |
Pos logit contributions | |
ch | +0.040 |
cho | +0.038 |
ink | +0.037 |
bullet | +0.036 |
chalk | +0.036 |
Neg logit contributions | |
Jacob | -0.053 |
Megan | -0.050 |
Jacob | -0.049 |
restaurant | -0.048 |
ide | -0.048 |
Token | a |
Token ID | 257 |
Feature activation | +4.42e-01 |
Pos logit contributions | |
Jacob | +0.035 |
Megan | +0.034 |
restaurant | +0.033 |
Pen | +0.031 |
Jacob | +0.030 |
Neg logit contributions | |
bar | -0.028 |
chalk | -0.027 |
ink | -0.026 |
gol | -0.026 |
itted | -0.026 |
Token | face |
Token ID | 1986 |
Feature activation | +3.58e+00 |
Pos logit contributions | |
bar | +0.033 |
itted | +0.033 |
ully | +0.030 |
gol | +0.030 |
chalk | +0.030 |
Neg logit contributions | |
restaurant | -0.034 |
Jacob | -0.032 |
Megan | -0.030 |
Frost | -0.029 |
cabin | -0.028 |
Token | at |
Token ID | 379 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
Jacob | +0.096 |
restaurant | +0.094 |
Megan | +0.093 |
Pen | +0.086 |
Nick | +0.084 |
Neg logit contributions | |
bar | -0.071 |
chalk | -0.067 |
ink | -0.066 |
bullet | -0.066 |
ch | -0.065 |
Token | and |
Token ID | 290 |
Feature activation | -9.94e-02 |
Pos logit contributions | |
Jacob | +0.010 |
Megan | +0.010 |
restaurant | +0.009 |
Jacob | +0.009 |
Pen | +0.009 |
Neg logit contributions | |
ink | -0.008 |
chalk | -0.008 |
bar | -0.007 |
cho | -0.007 |
bullet | -0.007 |
Token | playing |
Token ID | 2712 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
Jacob | +0.008 |
Megan | +0.008 |
restaurant | +0.008 |
Pen | +0.008 |
Jacob | +0.007 |
Neg logit contributions | |
bar | -0.006 |
chalk | -0.006 |
ink | -0.006 |
bullet | -0.006 |
ch | -0.006 |
Token | with |
Token ID | 351 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
ink | +0.127 |
chalk | +0.126 |
bar | +0.115 |
ch | +0.114 |
bullet | +0.112 |
Neg logit contributions | |
Jacob | -0.151 |
Megan | -0.146 |
Jacob | -0.137 |
Pen | -0.136 |
restaurant | -0.136 |
Token | the |
Token ID | 262 |
Feature activation | -6.71e-01 |
Pos logit contributions | |
Jacob | +0.135 |
Megan | +0.131 |
restaurant | +0.127 |
Pen | +0.120 |
Jacob | +0.119 |
Neg logit contributions | |
bar | -0.111 |
chalk | -0.108 |
ink | -0.107 |
bullet | -0.105 |
ch | -0.104 |
Token | chalk |
Token ID | 30860 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
restaurant | +0.051 |
Jacob | +0.045 |
chamber | +0.044 |
Megan | +0.041 |
foods | +0.041 |
Neg logit contributions | |
bar | -0.049 |
ch | -0.047 |
cho | -0.046 |
br | -0.045 |
itted | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
bar | +0.744 |
chalk | +0.734 |
ink | +0.733 |
ch | +0.690 |
cho | +0.681 |
Neg logit contributions | |
Jacob | -0.785 |
Megan | -0.759 |
restaurant | -0.733 |
Jacob | -0.702 |
Pen | -0.698 |
Token | She |
Token ID | 1375 |
Feature activation | -6.13e-01 |
Pos logit contributions | |
Jacob | +0.001 |
Megan | +0.001 |
restaurant | +0.001 |
Pen | +0.001 |
Jacob | +0.001 |
Neg logit contributions | |
bar | -0.001 |
bullet | -0.001 |
ink | -0.001 |
cho | -0.001 |
chalk | -0.001 |
Token | crawled |
Token ID | 45668 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
Jacob | +0.055 |
Megan | +0.053 |
restaurant | +0.052 |
Jacob | +0.050 |
Pen | +0.049 |
Neg logit contributions | |
ink | -0.035 |
chalk | -0.035 |
bar | -0.035 |
cho | -0.032 |
ch | -0.032 |
Token | back |
Token ID | 736 |
Feature activation | +8.71e-01 |
Pos logit contributions | |
Jacob | +0.140 |
Megan | +0.136 |
restaurant | +0.129 |
Pen | +0.123 |
Nick | +0.121 |
Neg logit contributions | |
bar | -0.141 |
ink | -0.135 |
chalk | -0.135 |
bullet | -0.134 |
itted | -0.131 |
Token | home |
Token ID | 1363 |
Feature activation | +5.97e-01 |
Pos logit contributions | |
chalk | +0.046 |
ink | +0.044 |
bar | +0.042 |
itted | +0.039 |
Bree | +0.039 |
Neg logit contributions | |
Jacob | -0.084 |
Megan | -0.082 |
Jacob | -0.080 |
restaurant | -0.079 |
Pen | -0.076 |
Token | give |
Token ID | 1577 |
Feature activation | -7.31e-01 |
Pos logit contributions | |
chalk | +0.040 |
ink | +0.039 |
bar | +0.039 |
bullet | +0.037 |
ch | +0.036 |
Neg logit contributions | |
Jacob | -0.049 |
Megan | -0.047 |
restaurant | -0.046 |
Pen | -0.044 |
Jacob | -0.044 |
Token | Jack |
Token ID | 3619 |
Feature activation | +7.43e-01 |
Pos logit contributions | |
Jacob | +0.056 |
Megan | +0.054 |
restaurant | +0.053 |
Pen | +0.051 |
Jacob | +0.050 |
Neg logit contributions | |
chalk | -0.053 |
ink | -0.052 |
bar | -0.051 |
ch | -0.049 |
bullet | -0.049 |
Token | a |
Token ID | 257 |
Feature activation | +4.40e-01 |
Pos logit contributions | |
chalk | +0.038 |
ink | +0.038 |
bar | +0.036 |
ch | +0.032 |
cho | +0.031 |
Neg logit contributions | |
Jacob | -0.073 |
Megan | -0.072 |
Jacob | -0.068 |
Pen | -0.068 |
restaurant | -0.068 |
Token | box |
Token ID | 3091 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
plets | +0.037 |
itted | +0.035 |
bar | +0.035 |
iece | +0.035 |
ug | +0.034 |
Neg logit contributions | |
restaurant | -0.031 |
Jacob | -0.027 |
Megan | -0.024 |
chamber | -0.023 |
cafe | -0.022 |
Token | of |
Token ID | 286 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
Jacob | +0.103 |
Megan | +0.099 |
restaurant | +0.098 |
Pen | +0.092 |
Jacob | +0.092 |
Neg logit contributions | |
bar | -0.067 |
chalk | -0.067 |
ink | -0.067 |
bullet | -0.064 |
ch | -0.064 |
Token | chalk |
Token ID | 30860 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
itted | +0.055 |
plets | +0.054 |
bar | +0.054 |
cho | +0.053 |
ch | +0.050 |
Neg logit contributions | |
restaurant | -0.143 |
Jacob | -0.141 |
Megan | -0.135 |
foods | -0.128 |
chamber | -0.127 |
Token | , |
Token ID | 11 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
bar | +0.730 |
chalk | +0.719 |
ink | +0.717 |
bullet | +0.667 |
ch | +0.663 |
Neg logit contributions | |
Jacob | -0.811 |
Megan | -0.792 |
restaurant | -0.757 |
Pen | -0.737 |
Jacob | -0.732 |
Token | but |
Token ID | 475 |
Feature activation | +2.98e-02 |
Pos logit contributions | |
bar | +0.022 |
chalk | +0.021 |
ink | +0.021 |
bullet | +0.021 |
ch | +0.021 |
Neg logit contributions | |
Jacob | -0.021 |
Megan | -0.020 |
restaurant | -0.019 |
Pen | -0.018 |
Jacob | -0.018 |
Token | they |
Token ID | 484 |
Feature activation | +6.17e-01 |
Pos logit contributions | |
chalk | +0.002 |
ink | +0.002 |
bar | +0.002 |
Bounce | +0.002 |
bullet | +0.002 |
Neg logit contributions | |
Jacob | -0.003 |
Megan | -0.003 |
restaurant | -0.003 |
Pen | -0.002 |
Jacob | -0.002 |
Token | couldn |
Token ID | 3521 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
bar | +0.036 |
chalk | +0.035 |
ink | +0.034 |
cho | +0.032 |
bullet | +0.032 |
Neg logit contributions | |
Jacob | -0.056 |
Megan | -0.055 |
restaurant | -0.055 |
Jacob | -0.053 |
Pen | -0.052 |
Token | 't |
Token ID | 470 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
bar | +0.055 |
ch | +0.052 |
chalk | +0.049 |
ink | +0.049 |
cho | +0.048 |
Neg logit contributions | |
restaurant | -0.100 |
Jacob | -0.099 |
Megan | -0.099 |
Pen | -0.091 |
Jacob | -0.090 |
Token | ground |
Token ID | 2323 |
Feature activation | +2.37e+00 |
Pos logit contributions | |
Jacob | +0.061 |
restaurant | +0.061 |
Megan | +0.060 |
Nick | +0.056 |
Pen | +0.055 |
Neg logit contributions | |
bar | -0.038 |
itted | -0.038 |
Bree | -0.037 |
Bounce | -0.036 |
ch | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -2.33e-02 |
Pos logit contributions | |
bar | +0.173 |
chalk | +0.171 |
ink | +0.170 |
bullet | +0.163 |
ch | +0.162 |
Neg logit contributions | |
Jacob | -0.179 |
Megan | -0.173 |
restaurant | -0.168 |
Pen | -0.159 |
Jacob | -0.159 |
Token | Mum |
Token ID | 38367 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
Jacob | +0.002 |
Megan | +0.002 |
restaurant | +0.002 |
Jacob | +0.002 |
Pen | +0.002 |
Neg logit contributions | |
bar | -0.002 |
ink | -0.002 |
chalk | -0.002 |
cho | -0.002 |
bullet | -0.002 |
Token | took |
Token ID | 1718 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
Jacob | +0.156 |
Megan | +0.154 |
restaurant | +0.152 |
Pen | +0.142 |
Jacob | +0.142 |
Neg logit contributions | |
bar | -0.119 |
chalk | -0.116 |
ink | -0.115 |
ch | -0.111 |
cho | -0.110 |
Token | some |
Token ID | 617 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
bar | +0.145 |
peac | +0.141 |
bullet | +0.137 |
cho | +0.137 |
itted | +0.137 |
Neg logit contributions | |
Jacob | -0.120 |
Megan | -0.110 |
restaurant | -0.107 |
Jacob | -0.100 |
Nick | -0.099 |
Token | chalk |
Token ID | 30860 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
restaurant | +0.126 |
Jacob | +0.121 |
Megan | +0.120 |
Nick | +0.112 |
Pen | +0.107 |
Neg logit contributions | |
bar | -0.111 |
Bree | -0.106 |
itted | -0.105 |
Bounce | -0.104 |
peac | -0.104 |
Token | from |
Token ID | 422 |
Feature activation | -8.46e-01 |
Pos logit contributions | |
bar | +0.730 |
ink | +0.724 |
chalk | +0.723 |
bullet | +0.713 |
ch | +0.707 |
Neg logit contributions | |
Jacob | -0.776 |
Megan | -0.744 |
restaurant | -0.730 |
Pen | -0.677 |
Jacob | -0.672 |
Token | her |
Token ID | 607 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
Jacob | +0.070 |
Megan | +0.069 |
restaurant | +0.066 |
Pen | +0.065 |
Jacob | +0.064 |
Neg logit contributions | |
chalk | -0.057 |
ink | -0.055 |
bar | -0.052 |
bullet | -0.049 |
gol | -0.049 |
Token | pocket |
Token ID | 10000 |
Feature activation | +4.81e-02 |
Pos logit contributions | |
Jacob | +0.011 |
Megan | +0.011 |
restaurant | +0.010 |
Pen | +0.010 |
Jacob | +0.010 |
Neg logit contributions | |
bar | -0.010 |
ink | -0.010 |
chalk | -0.010 |
ch | -0.010 |
bullet | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | -9.39e-02 |
Pos logit contributions | |
bar | +0.004 |
chalk | +0.004 |
ink | +0.004 |
bullet | +0.003 |
ch | +0.003 |
Neg logit contributions | |
Jacob | -0.003 |
Megan | -0.003 |
restaurant | -0.003 |
Pen | -0.003 |
Jacob | -0.003 |
Token | she |
Token ID | 673 |
Feature activation | +9.39e-01 |
Pos logit contributions | |
Jacob | +0.008 |
Megan | +0.007 |
restaurant | +0.007 |
Pen | +0.007 |
Jacob | +0.007 |
Neg logit contributions | |
bar | -0.006 |
ink | -0.006 |
chalk | -0.006 |
ch | -0.006 |
bullet | -0.006 |
Token | that |
Token ID | 326 |
Feature activation | +4.23e-01 |
Pos logit contributions | |
bar | +0.138 |
chalk | +0.136 |
ink | +0.136 |
bullet | +0.128 |
ch | +0.128 |
Neg logit contributions | |
Jacob | -0.218 |
Megan | -0.211 |
restaurant | -0.206 |
Pen | -0.197 |
Jacob | -0.196 |
Token | she |
Token ID | 673 |
Feature activation | +9.39e-01 |
Pos logit contributions | |
bar | +0.026 |
ink | +0.026 |
chalk | +0.026 |
cho | +0.024 |
ch | +0.024 |
Neg logit contributions | |
Jacob | -0.037 |
Megan | -0.036 |
restaurant | -0.035 |
Jacob | -0.033 |
Pen | -0.033 |
Token | had |
Token ID | 550 |
Feature activation | +7.33e-01 |
Pos logit contributions | |
ink | +0.047 |
chalk | +0.046 |
bar | +0.045 |
ch | +0.044 |
cho | +0.043 |
Neg logit contributions | |
Megan | -0.092 |
Jacob | -0.092 |
restaurant | -0.090 |
Pen | -0.085 |
Jacob | -0.085 |
Token | used |
Token ID | 973 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
bar | +0.056 |
bullet | +0.054 |
ch | +0.054 |
plets | +0.053 |
trigger | +0.053 |
Neg logit contributions | |
Jacob | -0.053 |
Megan | -0.050 |
restaurant | -0.049 |
Pen | -0.045 |
Jacob | -0.044 |
Token | her |
Token ID | 607 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
plets | +0.174 |
bar | +0.170 |
bullet | +0.169 |
ch | +0.168 |
EEP | +0.165 |
Neg logit contributions | |
Jacob | -0.130 |
Megan | -0.126 |
restaurant | -0.115 |
Nick | -0.106 |
Pen | -0.105 |
Token | chalk |
Token ID | 30860 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
restaurant | +0.013 |
Jacob | +0.013 |
search | +0.012 |
searches | +0.012 |
Megan | +0.012 |
Neg logit contributions | |
cho | -0.008 |
ch | -0.007 |
bar | -0.007 |
plets | -0.007 |
crashing | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | +5.94e-01 |
Pos logit contributions | |
bar | +0.720 |
bullet | +0.715 |
ch | +0.711 |
cho | +0.707 |
ink | +0.706 |
Neg logit contributions | |
Jacob | -0.764 |
Megan | -0.728 |
restaurant | -0.723 |
Jacob | -0.658 |
Pen | -0.657 |
Token | make |
Token ID | 787 |
Feature activation | +7.19e-01 |
Pos logit contributions | |
ch | +0.047 |
itted | +0.047 |
cho | +0.046 |
plets | +0.045 |
bar | +0.045 |
Neg logit contributions | |
Jacob | -0.037 |
restaurant | -0.037 |
Megan | -0.036 |
search | -0.032 |
Nick | -0.031 |
Token | a |
Token ID | 257 |
Feature activation | +4.52e-01 |
Pos logit contributions | |
bar | +0.052 |
bullet | +0.050 |
itted | +0.049 |
ch | +0.048 |
chalk | +0.048 |
Neg logit contributions | |
Jacob | -0.056 |
Megan | -0.054 |
restaurant | -0.052 |
Pen | -0.049 |
Jacob | -0.048 |
Token | difference |
Token ID | 3580 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
itted | +0.039 |
bar | +0.036 |
ged | +0.035 |
plets | +0.034 |
ch | +0.033 |
Neg logit contributions | |
restaurant | -0.033 |
Jacob | -0.028 |
Megan | -0.027 |
chamber | -0.026 |
cabin | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
Jacob | +0.108 |
Megan | +0.106 |
restaurant | +0.103 |
Jacob | +0.101 |
Pen | +0.097 |
Neg logit contributions | |
ink | -0.084 |
chalk | -0.083 |
bar | -0.081 |
cho | -0.077 |
gol | -0.076 |
Token | found |
Token ID | 1043 |
Feature activation | -3.42e-01 |
Pos logit contributions | |
ink | +0.014 |
chalk | +0.013 |
printed | +0.012 |
bullet | +0.011 |
ch | +0.010 |
Neg logit contributions | |
Jacob | -0.032 |
Megan | -0.032 |
Pen | -0.030 |
Jacob | -0.030 |
Nick | -0.029 |
Token | a |
Token ID | 257 |
Feature activation | +4.40e-01 |
Pos logit contributions | |
Jacob | +0.022 |
restaurant | +0.020 |
Megan | +0.020 |
Frost | +0.019 |
Nick | +0.018 |
Neg logit contributions | |
itted | -0.029 |
gol | -0.028 |
utes | -0.028 |
EEP | -0.028 |
crashing | -0.028 |
Token | big |
Token ID | 1263 |
Feature activation | +3.35e-01 |
Pos logit contributions | |
itted | +0.031 |
bar | +0.030 |
alks | +0.028 |
plets | +0.028 |
ch | +0.027 |
Neg logit contributions | |
restaurant | -0.037 |
Jacob | -0.034 |
Megan | -0.032 |
Pen | -0.030 |
chamber | -0.030 |
Token | piece |
Token ID | 3704 |
Feature activation | -5.85e-01 |
Pos logit contributions | |
itted | +0.023 |
bar | +0.023 |
ch | +0.022 |
utes | +0.021 |
cho | +0.021 |
Neg logit contributions | |
restaurant | -0.027 |
Jacob | -0.026 |
Megan | -0.024 |
Jacob | -0.022 |
Pen | -0.022 |
Token | of |
Token ID | 286 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
Jacob | +0.050 |
restaurant | +0.048 |
Megan | +0.047 |
Pen | +0.043 |
Jacob | +0.043 |
Neg logit contributions | |
bar | -0.034 |
itted | -0.034 |
chalk | -0.033 |
gol | -0.033 |
Bounce | -0.033 |
Token | chalk |
Token ID | 30860 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
itted | +0.089 |
plets | +0.087 |
ived | +0.081 |
ug | +0.078 |
utes | +0.077 |
Neg logit contributions | |
restaurant | -0.120 |
foods | -0.109 |
cereal | -0.106 |
Jacob | -0.104 |
refrigerator | -0.100 |
Token | and |
Token ID | 290 |
Feature activation | -8.97e-02 |
Pos logit contributions | |
bar | +0.651 |
chalk | +0.643 |
ink | +0.643 |
bullet | +0.605 |
ch | +0.604 |
Neg logit contributions | |
Jacob | -0.876 |
Megan | -0.852 |
restaurant | -0.824 |
Pen | -0.789 |
Jacob | -0.787 |
Token | decided |
Token ID | 3066 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
Jacob | +0.007 |
Megan | +0.007 |
restaurant | +0.006 |
Pen | +0.006 |
Jacob | +0.006 |
Neg logit contributions | |
bar | -0.006 |
itted | -0.006 |
ink | -0.006 |
ch | -0.006 |
chalk | -0.006 |
Token | to |
Token ID | 284 |
Feature activation | +5.95e-01 |
Pos logit contributions | |
bar | +0.082 |
ink | +0.082 |
itted | +0.081 |
ch | +0.080 |
chalk | +0.079 |
Neg logit contributions | |
Jacob | -0.095 |
Megan | -0.091 |
restaurant | -0.089 |
Pen | -0.082 |
Nick | -0.081 |
Token | eat |
Token ID | 4483 |
Feature activation | -3.52e+00 |
Pos logit contributions | |
chalk | +0.041 |
ink | +0.038 |
bar | +0.036 |
bullet | +0.033 |
gol | +0.033 |
Neg logit contributions | |
Jacob | -0.051 |
Megan | -0.050 |
Pen | -0.048 |
restaurant | -0.048 |
Jacob | -0.048 |
Token | it |
Token ID | 340 |
Feature activation | +3.84e-01 |
Pos logit contributions | |
Jacob | +0.249 |
restaurant | +0.236 |
Megan | +0.234 |
Nick | +0.215 |
Frost | +0.211 |
Neg logit contributions | |
itted | -0.280 |
bar | -0.270 |
peac | -0.261 |
bullet | -0.261 |
Bree | -0.259 |
Token | some |
Token ID | 617 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
bar | +0.362 |
gol | +0.345 |
bullet | +0.344 |
EEP | +0.334 |
crashing | +0.334 |
Neg logit contributions | |
Jacob | -0.287 |
restaurant | -0.285 |
Megan | -0.281 |
chamber | -0.238 |
abel | -0.238 |
Token | pictures |
Token ID | 5986 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
restaurant | +0.130 |
foods | +0.112 |
Megan | +0.109 |
Jacob | +0.107 |
chamber | +0.106 |
Neg logit contributions | |
bar | -0.115 |
cho | -0.105 |
plets | -0.103 |
itted | -0.103 |
bullet | -0.101 |
Token | with |
Token ID | 351 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
bar | +0.102 |
ink | +0.100 |
chalk | +0.100 |
bullet | +0.098 |
ch | +0.096 |
Neg logit contributions | |
Jacob | -0.109 |
Megan | -0.105 |
restaurant | -0.102 |
Pen | -0.096 |
Jacob | -0.095 |
Token | her |
Token ID | 607 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
Jacob | +0.117 |
Megan | +0.116 |
restaurant | +0.112 |
Nick | +0.103 |
abel | +0.100 |
Neg logit contributions | |
bar | -0.123 |
EEP | -0.123 |
plets | -0.119 |
bullet | -0.118 |
ch | -0.118 |
Token | special |
Token ID | 2041 |
Feature activation | -4.51e-01 |
Pos logit contributions | |
Jacob | +0.012 |
restaurant | +0.012 |
Megan | +0.012 |
chamber | +0.011 |
Nick | +0.011 |
Neg logit contributions | |
ch | -0.009 |
bar | -0.009 |
cho | -0.009 |
bullet | -0.008 |
crashing | -0.008 |
Token | chalk |
Token ID | 30860 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
restaurant | +0.040 |
foods | +0.039 |
search | +0.037 |
chamber | +0.035 |
searches | +0.033 |
Neg logit contributions | |
cho | -0.026 |
plets | -0.025 |
crashing | -0.024 |
ch | -0.023 |
ug | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -2.00e-02 |
Pos logit contributions | |
bar | +0.849 |
bullet | +0.841 |
ch | +0.827 |
cho | +0.822 |
ink | +0.821 |
Neg logit contributions | |
Jacob | -0.648 |
Megan | -0.615 |
restaurant | -0.608 |
Pen | -0.548 |
Jacob | -0.534 |
Token | She |
Token ID | 1375 |
Feature activation | -6.18e-01 |
Pos logit contributions | |
Jacob | +0.001 |
Megan | +0.001 |
restaurant | +0.001 |
Jacob | +0.001 |
Pen | +0.001 |
Neg logit contributions | |
bar | -0.002 |
bullet | -0.002 |
cho | -0.002 |
peac | -0.002 |
ch | -0.002 |
Token | loved |
Token ID | 6151 |
Feature activation | +7.40e-01 |
Pos logit contributions | |
Jacob | +0.052 |
Megan | +0.051 |
restaurant | +0.050 |
Jacob | +0.048 |
Clive | +0.047 |
Neg logit contributions | |
chalk | -0.039 |
bar | -0.039 |
ink | -0.039 |
cho | -0.036 |
ch | -0.036 |
Token | to |
Token ID | 284 |
Feature activation | +5.84e-01 |
Pos logit contributions | |
plets | +0.065 |
bar | +0.063 |
trigger | +0.061 |
bullet | +0.061 |
EEP | +0.060 |
Neg logit contributions | |
Jacob | -0.047 |
restaurant | -0.046 |
Megan | -0.042 |
Nick | -0.038 |
pic | -0.037 |
Token | draw |
Token ID | 3197 |
Feature activation | +4.50e+00 |
Pos logit contributions | |
bar | +0.045 |
itted | +0.045 |
cho | +0.044 |
ch | +0.044 |
bullet | +0.043 |
Neg logit contributions | |
Jacob | -0.041 |
Megan | -0.039 |
restaurant | -0.039 |
search | -0.035 |
searches | -0.034 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
ink | +0.046 |
chalk | +0.042 |
cho | +0.042 |
ch | +0.040 |
Bree | +0.040 |
Neg logit contributions | |
Megan | -0.053 |
Jacob | -0.053 |
Jacob | -0.048 |
Nick | -0.048 |
Pen | -0.047 |
Token |
|
Token ID | 198 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
Jacob | +0.069 |
Megan | +0.069 |
restaurant | +0.063 |
Pen | +0.062 |
Jacob | +0.058 |
Neg logit contributions | |
bar | -0.087 |
bullet | -0.080 |
gol | -0.079 |
crashing | -0.078 |
cho | -0.078 |
Token |
|
Token ID | 198 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
bar | +0.009 |
chalk | +0.009 |
ink | +0.009 |
bullet | +0.008 |
itted | +0.008 |
Neg logit contributions | |
Jacob | -0.009 |
Megan | -0.008 |
restaurant | -0.008 |
Pen | -0.008 |
Jacob | -0.008 |
Token | " |
Token ID | 1 |
Feature activation | -9.50e-01 |
Pos logit contributions | |
bar | +0.008 |
chalk | +0.007 |
ink | +0.007 |
bullet | +0.007 |
cho | +0.007 |
Neg logit contributions | |
Jacob | -0.010 |
Megan | -0.010 |
restaurant | -0.010 |
Jacob | -0.009 |
Pen | -0.009 |
Token | My |
Token ID | 3666 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
Jacob | +0.070 |
Megan | +0.068 |
restaurant | +0.065 |
Jacob | +0.064 |
Pen | +0.064 |
Neg logit contributions | |
bar | -0.071 |
ink | -0.066 |
chalk | -0.066 |
bullet | -0.066 |
itted | -0.065 |
Token | chalk |
Token ID | 30860 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
ch | +0.134 |
br | +0.128 |
plets | +0.127 |
cho | +0.126 |
crashing | +0.118 |
Neg logit contributions | |
restaurant | -0.139 |
Jacob | -0.135 |
chamber | -0.126 |
Megan | -0.124 |
search | -0.118 |
Token | is |
Token ID | 318 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
bar | +0.966 |
ink | +0.946 |
chalk | +0.944 |
ch | +0.931 |
bullet | +0.928 |
Neg logit contributions | |
Jacob | -0.540 |
Megan | -0.504 |
restaurant | -0.498 |
Jacob | -0.443 |
Pen | -0.439 |
Token | broken |
Token ID | 5445 |
Feature activation | +2.74e+00 |
Pos logit contributions | |
Jacob | +0.094 |
restaurant | +0.086 |
Megan | +0.083 |
chamber | +0.082 |
Frost | +0.079 |
Neg logit contributions | |
itted | -0.147 |
utes | -0.146 |
plets | -0.139 |
vals | -0.137 |
ully | -0.136 |
Token | ! |
Token ID | 0 |
Feature activation | +3.00e-01 |
Pos logit contributions | |
bar | +0.212 |
ink | +0.211 |
chalk | +0.210 |
bullet | +0.201 |
ch | +0.200 |
Neg logit contributions | |
Jacob | -0.196 |
Megan | -0.189 |
restaurant | -0.183 |
Jacob | -0.173 |
Pen | -0.172 |
Token | I |
Token ID | 314 |
Feature activation | +7.44e-01 |
Pos logit contributions | |
bar | +0.021 |
chalk | +0.020 |
ink | +0.020 |
bullet | +0.019 |
itted | +0.019 |
Neg logit contributions | |
Jacob | -0.024 |
Megan | -0.023 |
Pen | -0.022 |
restaurant | -0.022 |
Jacob | -0.021 |
Token | can |
Token ID | 460 |
Feature activation | +7.67e-01 |
Pos logit contributions | |
bar | +0.052 |
chalk | +0.052 |
ink | +0.052 |
bullet | +0.049 |
ch | +0.049 |
Neg logit contributions | |
Jacob | -0.059 |
Megan | -0.057 |
restaurant | -0.055 |
Jacob | -0.052 |
Pen | -0.052 |
Token | a |
Token ID | 257 |
Feature activation | +4.44e-01 |
Pos logit contributions | |
bar | +0.539 |
gol | +0.513 |
alks | +0.505 |
peac | +0.498 |
bullet | +0.490 |
Neg logit contributions | |
Jacob | -0.448 |
restaurant | -0.442 |
Megan | -0.423 |
Jacob | -0.386 |
abel | -0.383 |
Token | big |
Token ID | 1263 |
Feature activation | +3.29e-01 |
Pos logit contributions | |
itted | +0.039 |
bar | +0.037 |
ged | +0.037 |
ug | +0.037 |
ging | +0.035 |
Neg logit contributions | |
restaurant | -0.029 |
chamber | -0.026 |
bar | -0.023 |
Jacob | -0.023 |
shelf | -0.022 |
Token | sun |
Token ID | 4252 |
Feature activation | +4.51e+00 |
Pos logit contributions | |
itted | +0.027 |
utes | +0.025 |
Mat | +0.025 |
amed | +0.024 |
bar | +0.024 |
Neg logit contributions | |
restaurant | -0.023 |
chamber | -0.021 |
Jacob | -0.019 |
bar | -0.018 |
shelf | -0.018 |
Token | with |
Token ID | 351 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
bullet | +0.345 |
Bounce | +0.337 |
chalk | +0.333 |
gol | +0.333 |
itted | +0.332 |
Neg logit contributions | |
Jacob | -0.302 |
restaurant | -0.293 |
Megan | -0.280 |
Pen | -0.267 |
abel | -0.256 |
Token | yellow |
Token ID | 7872 |
Feature activation | +2.39e+00 |
Pos logit contributions | |
Jacob | +0.080 |
Megan | +0.076 |
restaurant | +0.066 |
Nick | +0.064 |
abel | +0.063 |
Neg logit contributions | |
utes | -0.161 |
EEP | -0.157 |
gol | -0.156 |
bar | -0.156 |
wrench | -0.154 |
Token | chalk |
Token ID | 30860 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
gol | +0.174 |
Bounce | +0.172 |
utes | +0.171 |
ug | +0.166 |
Bree | +0.165 |
Neg logit contributions | |
restaurant | -0.136 |
foods | -0.134 |
bar | -0.123 |
cereal | -0.123 |
drinks | -0.117 |
Token | when |
Token ID | 618 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
Bree | +0.816 |
Bounce | +0.805 |
bullet | +0.798 |
bar | +0.794 |
gol | +0.790 |
Neg logit contributions | |
Jacob | -0.642 |
restaurant | -0.615 |
Megan | -0.581 |
abel | -0.556 |
Pen | -0.536 |
Token | her |
Token ID | 607 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
bar | +0.087 |
chalk | +0.082 |
ink | +0.082 |
bullet | +0.081 |
itted | +0.081 |
Neg logit contributions | |
Jacob | -0.115 |
Megan | -0.112 |
restaurant | -0.109 |
Pen | -0.103 |
Jacob | -0.101 |
Token | little |
Token ID | 1310 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
restaurant | +0.006 |
Jacob | +0.006 |
Megan | +0.006 |
chamber | +0.005 |
Nick | +0.005 |
Neg logit contributions | |
bar | -0.015 |
ch | -0.015 |
plets | -0.015 |
br | -0.015 |
cho | -0.014 |
Token | brother |
Token ID | 3956 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
bar | +0.119 |
ula | +0.119 |
Mat | +0.118 |
br | +0.118 |
plets | +0.116 |
Neg logit contributions | |
restaurant | -0.078 |
Jacob | -0.074 |
Megan | -0.068 |
Pen | -0.063 |
abel | -0.061 |
Token | came |
Token ID | 1625 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
bar | +0.141 |
itted | +0.137 |
utes | +0.137 |
alks | +0.131 |
ch | +0.129 |
Neg logit contributions | |
Jacob | -0.149 |
Megan | -0.138 |
restaurant | -0.137 |
Pen | -0.125 |
abel | -0.124 |
Token | saw |
Token ID | 2497 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
Jacob | +0.052 |
Megan | +0.052 |
restaurant | +0.050 |
Pen | +0.048 |
Jacob | +0.048 |
Neg logit contributions | |
chalk | -0.034 |
bar | -0.034 |
ink | -0.033 |
cho | -0.032 |
ch | -0.032 |
Token | Emily |
Token ID | 17608 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
Jacob | +0.016 |
Megan | +0.016 |
restaurant | +0.015 |
Pen | +0.014 |
Jacob | +0.014 |
Neg logit contributions | |
bar | -0.014 |
chalk | -0.013 |
ink | -0.013 |
bullet | -0.013 |
ch | -0.013 |
Token | with |
Token ID | 351 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
ink | +0.077 |
chalk | +0.076 |
bar | +0.076 |
crashing | +0.067 |
laser | +0.066 |
Neg logit contributions | |
Megan | -0.116 |
Jacob | -0.114 |
Jacob | -0.110 |
restaurant | -0.109 |
Pen | -0.105 |
Token | the |
Token ID | 262 |
Feature activation | -6.70e-01 |
Pos logit contributions | |
Jacob | +0.135 |
Megan | +0.128 |
restaurant | +0.125 |
Pen | +0.117 |
Jacob | +0.116 |
Neg logit contributions | |
bar | -0.112 |
ch | -0.107 |
cho | -0.105 |
itted | -0.104 |
ink | -0.104 |
Token | thick |
Token ID | 6546 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
restaurant | +0.058 |
Jacob | +0.055 |
chamber | +0.052 |
search | +0.051 |
Megan | +0.051 |
Neg logit contributions | |
cho | -0.043 |
ug | -0.040 |
ch | -0.040 |
utes | -0.038 |
bar | -0.037 |
Token | chalk |
Token ID | 30860 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
restaurant | +0.002 |
Jacob | +0.002 |
chamber | +0.001 |
Megan | +0.001 |
foods | +0.001 |
Neg logit contributions | |
cho | -0.001 |
ch | -0.001 |
bar | -0.001 |
utes | -0.001 |
crashing | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
bar | +0.735 |
chalk | +0.725 |
ink | +0.725 |
bullet | +0.698 |
ch | +0.697 |
Neg logit contributions | |
Jacob | -0.783 |
Megan | -0.752 |
restaurant | -0.734 |
Pen | -0.690 |
Jacob | -0.688 |
Token | asked |
Token ID | 1965 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
Jacob | +0.008 |
Megan | +0.007 |
restaurant | +0.007 |
Pen | +0.006 |
Nick | +0.006 |
Neg logit contributions | |
bar | -0.008 |
ch | -0.008 |
itted | -0.008 |
ink | -0.007 |
cho | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | +2.81e-01 |
Pos logit contributions | |
Jacob | +0.124 |
Megan | +0.123 |
Jacob | +0.122 |
restaurant | +0.119 |
Pen | +0.115 |
Neg logit contributions | |
ink | -0.071 |
chalk | -0.069 |
Bounce | -0.064 |
Bree | -0.063 |
cho | -0.061 |
Token | " |
Token ID | 366 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
bar | +0.015 |
chalk | +0.015 |
ink | +0.015 |
bullet | +0.014 |
ch | +0.014 |
Neg logit contributions | |
Jacob | -0.027 |
Megan | -0.026 |
restaurant | -0.025 |
Pen | -0.024 |
Jacob | -0.024 |
Token | What |
Token ID | 2061 |
Feature activation | -4.19e-01 |
Pos logit contributions | |
bar | +0.096 |
ink | +0.095 |
chalk | +0.095 |
itted | +0.092 |
bullet | +0.092 |
Neg logit contributions | |
Jacob | -0.095 |
Megan | -0.090 |
restaurant | -0.090 |
Jacob | -0.086 |
Pen | -0.084 |
Token | girl |
Token ID | 2576 |
Feature activation | +7.68e-01 |
Pos logit contributions | |
chalk | +0.066 |
ink | +0.066 |
bar | +0.062 |
gol | +0.060 |
sun | +0.059 |
Neg logit contributions | |
Jacob | -0.140 |
Megan | -0.135 |
Jacob | -0.129 |
restaurant | -0.128 |
Pen | -0.127 |
Token | ran |
Token ID | 4966 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
bar | +0.050 |
chalk | +0.049 |
ink | +0.049 |
bullet | +0.047 |
ch | +0.047 |
Neg logit contributions | |
Jacob | -0.064 |
Megan | -0.062 |
restaurant | -0.060 |
Pen | -0.057 |
Jacob | -0.057 |
Token | off |
Token ID | 572 |
Feature activation | +2.82e+00 |
Pos logit contributions | |
bar | +0.013 |
peac | +0.013 |
ink | +0.013 |
bullet | +0.013 |
chalk | +0.013 |
Neg logit contributions | |
Jacob | -0.017 |
Megan | -0.016 |
restaurant | -0.016 |
Nick | -0.015 |
Pen | -0.014 |
Token | with |
Token ID | 351 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
bar | +0.198 |
ink | +0.192 |
chalk | +0.190 |
bullet | +0.190 |
ch | +0.189 |
Neg logit contributions | |
Jacob | -0.224 |
Megan | -0.217 |
restaurant | -0.211 |
Pen | -0.200 |
Nick | -0.194 |
Token | her |
Token ID | 607 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
Jacob | +0.143 |
Megan | +0.138 |
restaurant | +0.136 |
Jacob | +0.131 |
Pen | +0.129 |
Neg logit contributions | |
chalk | -0.102 |
ink | -0.102 |
bar | -0.097 |
bullet | -0.093 |
cho | -0.093 |
Token | chalk |
Token ID | 30860 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
Jacob | +0.012 |
restaurant | +0.012 |
Megan | +0.012 |
Pen | +0.011 |
Nick | +0.010 |
Neg logit contributions | |
bar | -0.009 |
ch | -0.009 |
bullet | -0.008 |
cho | -0.008 |
itted | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | -9.60e-02 |
Pos logit contributions | |
ink | +0.777 |
chalk | +0.775 |
bar | +0.773 |
ch | +0.759 |
bullet | +0.758 |
Neg logit contributions | |
Jacob | -0.724 |
Megan | -0.690 |
restaurant | -0.677 |
Pen | -0.628 |
Jacob | -0.620 |
Token | her |
Token ID | 607 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
Jacob | +0.008 |
Megan | +0.007 |
restaurant | +0.007 |
Pen | +0.007 |
Nick | +0.007 |
Neg logit contributions | |
bar | -0.007 |
ch | -0.007 |
ink | -0.007 |
itted | -0.006 |
chalk | -0.006 |
Token | picture |
Token ID | 4286 |
Feature activation | +3.64e+00 |
Pos logit contributions | |
Jacob | +0.011 |
restaurant | +0.010 |
Megan | +0.010 |
Pen | +0.009 |
Nick | +0.009 |
Neg logit contributions | |
bar | -0.011 |
ch | -0.010 |
itted | -0.010 |
cho | -0.010 |
bullet | -0.010 |
Token | of |
Token ID | 286 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
bar | +0.273 |
chalk | +0.271 |
ink | +0.270 |
bullet | +0.263 |
gol | +0.259 |
Neg logit contributions | |
Jacob | -0.265 |
Megan | -0.256 |
restaurant | -0.245 |
Pen | -0.234 |
Jacob | -0.228 |
Token | her |
Token ID | 607 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
chalk | +0.089 |
bar | +0.089 |
ink | +0.089 |
bullet | +0.084 |
ch | +0.083 |
Neg logit contributions | |
Jacob | -0.108 |
Megan | -0.104 |
restaurant | -0.102 |
Jacob | -0.096 |
Pen | -0.096 |
Token | on |
Token ID | 319 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
bar | +0.170 |
chalk | +0.167 |
ink | +0.167 |
bullet | +0.160 |
ch | +0.159 |
Neg logit contributions | |
Jacob | -0.204 |
Megan | -0.196 |
restaurant | -0.191 |
Pen | -0.181 |
Jacob | -0.181 |
Token | the |
Token ID | 262 |
Feature activation | -6.76e-01 |
Pos logit contributions | |
ink | +0.083 |
chalk | +0.083 |
bar | +0.079 |
bullet | +0.076 |
ch | +0.076 |
Neg logit contributions | |
Jacob | -0.122 |
Megan | -0.120 |
restaurant | -0.117 |
Pen | -0.112 |
Jacob | -0.112 |
Token | black |
Token ID | 2042 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
Jacob | +0.052 |
restaurant | +0.051 |
Megan | +0.048 |
Pen | +0.045 |
Jacob | +0.044 |
Neg logit contributions | |
bar | -0.052 |
cho | -0.048 |
Bounce | -0.048 |
Bree | -0.047 |
ch | -0.047 |
Token | board |
Token ID | 3526 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
bar | +0.109 |
cho | +0.103 |
ink | +0.102 |
Bounce | +0.102 |
Bree | +0.101 |
Neg logit contributions | |
Jacob | -0.092 |
restaurant | -0.090 |
Megan | -0.086 |
Jacob | -0.079 |
Pen | -0.077 |
Token | with |
Token ID | 351 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
Jacob | +0.073 |
Megan | +0.070 |
restaurant | +0.068 |
Pen | +0.064 |
Jacob | +0.064 |
Neg logit contributions | |
bar | -0.078 |
chalk | -0.077 |
ink | -0.077 |
bullet | -0.074 |
ch | -0.074 |
Token | chalk |
Token ID | 30860 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
Jacob | +0.121 |
Megan | +0.109 |
restaurant | +0.108 |
Pen | +0.098 |
Nick | +0.097 |
Neg logit contributions | |
bar | -0.129 |
gol | -0.120 |
utes | -0.118 |
plets | -0.117 |
ch | -0.117 |
Token | . |
Token ID | 13 |
Feature activation | -1.99e-02 |
Pos logit contributions | |
bar | +0.719 |
chalk | +0.703 |
ink | +0.700 |
bullet | +0.696 |
cho | +0.686 |
Neg logit contributions | |
Jacob | -0.785 |
restaurant | -0.740 |
Megan | -0.739 |
Pen | -0.679 |
Jacob | -0.677 |
Token | Lily |
Token ID | 20037 |
Feature activation | +6.30e-01 |
Pos logit contributions | |
Jacob | +0.002 |
Megan | +0.002 |
restaurant | +0.002 |
Pen | +0.002 |
Jacob | +0.001 |
Neg logit contributions | |
bar | -0.001 |
chalk | -0.001 |
ink | -0.001 |
bullet | -0.001 |
ch | -0.001 |
Token | thought |
Token ID | 1807 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
ink | +0.037 |
chalk | +0.037 |
printed | +0.035 |
bar | +0.034 |
cho | +0.033 |
Neg logit contributions | |
Megan | -0.056 |
Jacob | -0.054 |
restaurant | -0.053 |
Jacob | -0.050 |
Pen | -0.050 |
Token | the |
Token ID | 262 |
Feature activation | -6.77e-01 |
Pos logit contributions | |
Jacob | +0.080 |
Megan | +0.075 |
restaurant | +0.073 |
Pen | +0.069 |
Jacob | +0.068 |
Neg logit contributions | |
bar | -0.078 |
chalk | -0.074 |
ink | -0.074 |
bullet | -0.072 |
itted | -0.072 |
Token | black |
Token ID | 2042 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
Jacob | +0.049 |
restaurant | +0.048 |
Megan | +0.046 |
Jacob | +0.042 |
Pen | +0.042 |
Neg logit contributions | |
bar | -0.055 |
cho | -0.051 |
bullet | -0.049 |
ch | -0.049 |
itted | -0.049 |
Token | ummy |
Token ID | 13513 |
Feature activation | -9.29e-01 |
Pos logit contributions | |
Jacob | +0.369 |
Megan | +0.355 |
restaurant | +0.344 |
Jacob | +0.333 |
Pen | +0.332 |
Neg logit contributions | |
bar | -0.257 |
ink | -0.257 |
chalk | -0.256 |
ch | -0.241 |
itted | -0.239 |
Token | beef |
Token ID | 12023 |
Feature activation | -5.92e+00 |
Pos logit contributions | |
Jacob | +0.075 |
Megan | +0.071 |
Jacob | +0.069 |
Pen | +0.066 |
restaurant | +0.063 |
Neg logit contributions | |
ink | -0.068 |
chalk | -0.066 |
ch | -0.063 |
sun | -0.062 |
br | -0.058 |
Token | at |
Token ID | 379 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
Jacob | +0.522 |
Megan | +0.510 |
Jacob | +0.505 |
restaurant | +0.480 |
Pen | +0.471 |
Neg logit contributions | |
ink | -0.354 |
chalk | -0.345 |
bar | -0.344 |
cho | -0.320 |
sun | -0.301 |
Token | the |
Token ID | 262 |
Feature activation | -7.08e-01 |
Pos logit contributions | |
Jacob | +0.013 |
Megan | +0.013 |
Jacob | +0.012 |
abel | +0.012 |
restaurant | +0.012 |
Neg logit contributions | |
ink | -0.008 |
chalk | -0.008 |
bar | -0.007 |
sun | -0.007 |
cho | -0.007 |
Token | fancy |
Token ID | 14996 |
Feature activation | -7.53e-01 |
Pos logit contributions | |
Jacob | +0.050 |
Megan | +0.049 |
restaurant | +0.044 |
Jacob | +0.044 |
Pen | +0.043 |
Neg logit contributions | |
bar | -0.056 |
ink | -0.056 |
chalk | -0.055 |
ch | -0.053 |
cho | -0.053 |
Token | restaurant |
Token ID | 7072 |
Feature activation | -9.77e+00 |
Pos logit contributions | |
Jacob | +0.061 |
Megan | +0.061 |
Jacob | +0.055 |
Pen | +0.052 |
Nick | +0.051 |
Neg logit contributions | |
ink | -0.055 |
chalk | -0.052 |
bar | -0.052 |
ch | -0.052 |
cho | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | -5.30e-02 |
Pos logit contributions | |
Jacob | +0.835 |
Megan | +0.814 |
Jacob | +0.795 |
Pen | +0.769 |
restaurant | +0.756 |
Neg logit contributions | |
bar | -0.582 |
ink | -0.578 |
chalk | -0.578 |
cho | -0.539 |
ch | -0.507 |
Token | From |
Token ID | 3574 |
Feature activation | +5.97e-01 |
Pos logit contributions | |
Jacob | +0.004 |
Megan | +0.004 |
restaurant | +0.004 |
Pen | +0.004 |
Jacob | +0.004 |
Neg logit contributions | |
bar | -0.004 |
chalk | -0.004 |
ink | -0.004 |
bullet | -0.003 |
ch | -0.003 |
Token | that |
Token ID | 326 |
Feature activation | +3.99e-01 |
Pos logit contributions | |
ink | +0.030 |
chalk | +0.030 |
bar | +0.029 |
cho | +0.028 |
ch | +0.028 |
Neg logit contributions | |
Jacob | -0.059 |
Megan | -0.057 |
restaurant | -0.056 |
Jacob | -0.054 |
Pen | -0.053 |
Token | day |
Token ID | 1110 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
bar | +0.017 |
ink | +0.017 |
chalk | +0.017 |
ch | +0.016 |
cho | +0.016 |
Neg logit contributions | |
Jacob | -0.042 |
Megan | -0.041 |
restaurant | -0.040 |
Jacob | -0.039 |
Pen | -0.038 |
Token | on |
Token ID | 319 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
chalk | +0.128 |
gol | +0.126 |
peac | +0.125 |
ink | +0.125 |
bar | +0.125 |
Neg logit contributions | |
Jacob | -0.103 |
Pen | -0.097 |
Megan | -0.096 |
Jacob | -0.089 |
restaurant | -0.087 |
Token | to |
Token ID | 284 |
Feature activation | +5.51e-01 |
Pos logit contributions | |
Jacob | +0.049 |
Megan | +0.048 |
restaurant | +0.047 |
Pen | +0.044 |
Nick | +0.043 |
Neg logit contributions | |
bar | -0.042 |
bullet | -0.040 |
ink | -0.039 |
chalk | -0.039 |
gol | -0.039 |
Token | come |
Token ID | 1282 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
chalk | +0.039 |
ink | +0.038 |
bar | +0.037 |
bullet | +0.036 |
crashing | +0.036 |
Neg logit contributions | |
Jacob | -0.043 |
Megan | -0.042 |
Jacob | -0.040 |
restaurant | -0.039 |
Pen | -0.038 |
Token | back |
Token ID | 736 |
Feature activation | +8.43e-01 |
Pos logit contributions | |
bar | +0.008 |
chalk | +0.008 |
ink | +0.007 |
ch | +0.007 |
cho | +0.007 |
Neg logit contributions | |
Jacob | -0.019 |
Megan | -0.018 |
Jacob | -0.018 |
restaurant | -0.017 |
Pen | -0.017 |
Token | to |
Token ID | 284 |
Feature activation | +5.54e-01 |
Pos logit contributions | |
chalk | +0.048 |
ink | +0.047 |
ch | +0.045 |
bar | +0.045 |
sun | +0.044 |
Neg logit contributions | |
Jacob | -0.075 |
Megan | -0.073 |
Jacob | -0.073 |
Pen | -0.068 |
restaurant | -0.066 |
Token | the |
Token ID | 262 |
Feature activation | -7.03e-01 |
Pos logit contributions | |
ink | +0.034 |
chalk | +0.033 |
sun | +0.031 |
cho | +0.030 |
ch | +0.030 |
Neg logit contributions | |
Jacob | -0.048 |
Megan | -0.047 |
Jacob | -0.046 |
restaurant | -0.043 |
Pen | -0.042 |
Token | restaurant |
Token ID | 7072 |
Feature activation | -9.77e+00 |
Pos logit contributions | |
Jacob | +0.046 |
Megan | +0.045 |
Jacob | +0.040 |
Pen | +0.040 |
Nick | +0.037 |
Neg logit contributions | |
ink | -0.059 |
chalk | -0.058 |
sun | -0.057 |
cho | -0.056 |
bar | -0.055 |
Token | again |
Token ID | 757 |
Feature activation | -2.37e+00 |
Pos logit contributions | |
Megan | +0.881 |
Jacob | +0.876 |
Jacob | +0.863 |
Pen | +0.839 |
Clive | +0.819 |
Neg logit contributions | |
bar | -0.499 |
ink | -0.440 |
chalk | -0.438 |
cho | -0.429 |
sun | -0.421 |
Token | . |
Token ID | 13 |
Feature activation | -4.60e-02 |
Pos logit contributions | |
Jacob | +0.167 |
Megan | +0.158 |
Jacob | +0.149 |
restaurant | +0.148 |
Pen | +0.142 |
Neg logit contributions | |
ink | -0.186 |
bar | -0.185 |
chalk | -0.184 |
ch | -0.178 |
peac | -0.173 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +8.39e-01 |
Pos logit contributions | |
Jacob | +0.004 |
restaurant | +0.004 |
Megan | +0.004 |
Jacob | +0.004 |
Clive | +0.004 |
Neg logit contributions | |
chalk | -0.002 |
Bree | -0.002 |
ink | -0.002 |
Bounce | -0.002 |
bar | -0.002 |
Token | Once |
Token ID | 7454 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
ch | +0.031 |
chalk | +0.029 |
ink | +0.029 |
sun | +0.027 |
Bounce | +0.027 |
Neg logit contributions | |
restaurant | -0.085 |
Megan | -0.085 |
Jacob | -0.085 |
abel | -0.083 |
icity | -0.082 |
Token | upon |
Token ID | 2402 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
Jacob | +0.124 |
Megan | +0.119 |
restaurant | +0.116 |
Jacob | +0.109 |
Pen | +0.109 |
Neg logit contributions | |
bar | -0.131 |
chalk | -0.130 |
ink | -0.129 |
ch | -0.124 |
bullet | -0.123 |
Token | . |
Token ID | 13 |
Feature activation | -4.86e-02 |
Pos logit contributions | |
Jacob | +0.101 |
Megan | +0.097 |
restaurant | +0.095 |
Pen | +0.090 |
Jacob | +0.090 |
Neg logit contributions | |
bar | -0.082 |
chalk | -0.081 |
ink | -0.081 |
bullet | -0.077 |
ch | -0.077 |
Token | |
Token ID | 220 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
Jacob | +0.004 |
Megan | +0.004 |
restaurant | +0.004 |
Pen | +0.004 |
Jacob | +0.004 |
Neg logit contributions | |
bar | -0.003 |
ink | -0.003 |
chalk | -0.003 |
bullet | -0.003 |
ch | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +8.63e-02 |
Pos logit contributions | |
bar | +0.090 |
chalk | +0.089 |
ink | +0.088 |
ch | +0.085 |
bullet | +0.084 |
Neg logit contributions | |
Jacob | -0.107 |
Megan | -0.104 |
restaurant | -0.102 |
Pen | -0.096 |
Jacob | -0.095 |
Token |
|
Token ID | 198 |
Feature activation | +9.65e-02 |
Pos logit contributions | |
bar | +0.006 |
ink | +0.006 |
chalk | +0.006 |
bullet | +0.006 |
ch | +0.006 |
Neg logit contributions | |
Jacob | -0.007 |
Megan | -0.006 |
restaurant | -0.006 |
Jacob | -0.006 |
Pen | -0.006 |
Token | The |
Token ID | 464 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
ink | +0.006 |
chalk | +0.006 |
bar | +0.006 |
bullet | +0.006 |
itted | +0.006 |
Neg logit contributions | |
Jacob | -0.008 |
Megan | -0.008 |
restaurant | -0.008 |
Jacob | -0.008 |
Pen | -0.007 |
Token | restaurant |
Token ID | 7072 |
Feature activation | -9.76e+00 |
Pos logit contributions | |
bar | +0.068 |
chalk | +0.066 |
ink | +0.065 |
ch | +0.064 |
itted | +0.064 |
Neg logit contributions | |
Jacob | -0.085 |
restaurant | -0.082 |
Megan | -0.082 |
Pen | -0.076 |
Jacob | -0.075 |
Token | provided |
Token ID | 2810 |
Feature activation | -3.31e+00 |
Pos logit contributions | |
Jacob | +0.701 |
Megan | +0.695 |
Pen | +0.661 |
restaurant | +0.649 |
Jacob | +0.635 |
Neg logit contributions | |
bar | -0.711 |
chalk | -0.704 |
ink | -0.701 |
cho | -0.681 |
bullet | -0.670 |
Token | Tommy |
Token ID | 19919 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
Jacob | +0.280 |
Megan | +0.275 |
restaurant | +0.265 |
Pen | +0.258 |
Jacob | +0.258 |
Neg logit contributions | |
ink | -0.204 |
chalk | -0.202 |
bar | -0.200 |
cho | -0.192 |
bullet | -0.188 |
Token | and |
Token ID | 290 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
bar | +0.077 |
chalk | +0.073 |
ink | +0.073 |
itted | +0.068 |
cho | +0.066 |
Neg logit contributions | |
Megan | -0.107 |
Jacob | -0.106 |
Jacob | -0.102 |
restaurant | -0.100 |
Pen | -0.099 |
Token | his |
Token ID | 465 |
Feature activation | +5.22e-02 |
Pos logit contributions | |
Jacob | +0.011 |
Megan | +0.010 |
restaurant | +0.010 |
Pen | +0.010 |
Jacob | +0.009 |
Neg logit contributions | |
bar | -0.009 |
chalk | -0.008 |
ink | -0.008 |
ch | -0.008 |
itted | -0.008 |
Token | family |
Token ID | 1641 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
chalk | +0.003 |
ink | +0.003 |
sun | +0.003 |
bullet | +0.002 |
bar | +0.002 |
Neg logit contributions | |
Jacob | -0.005 |
Megan | -0.005 |
Jacob | -0.005 |
Pen | -0.004 |
ide | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -4.86e-02 |
Pos logit contributions | |
chalk | +0.127 |
ink | +0.123 |
cho | +0.107 |
bar | +0.103 |
sun | +0.101 |
Neg logit contributions | |
Megan | -0.242 |
Jacob | -0.239 |
Pen | -0.238 |
Jacob | -0.233 |
restaurant | -0.226 |
Token | They |
Token ID | 1119 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
Jacob | +0.004 |
Megan | +0.004 |
restaurant | +0.004 |
Pen | +0.003 |
Nick | +0.003 |
Neg logit contributions | |
bar | -0.003 |
ink | -0.003 |
chalk | -0.003 |
cho | -0.003 |
peac | -0.003 |
Token | made |
Token ID | 925 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
Jacob | +0.012 |
Megan | +0.011 |
restaurant | +0.011 |
Jacob | +0.010 |
Pen | +0.010 |
Neg logit contributions | |
ink | -0.009 |
bar | -0.009 |
itted | -0.009 |
chalk | -0.009 |
ch | -0.009 |
Token | a |
Token ID | 257 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
Jacob | +0.015 |
Megan | +0.015 |
restaurant | +0.015 |
Pen | +0.014 |
Jacob | +0.014 |
Neg logit contributions | |
chalk | -0.010 |
ink | -0.010 |
bar | -0.010 |
cho | -0.010 |
ch | -0.010 |
Token | cozy |
Token ID | 37438 |
Feature activation | -4.33e+00 |
Pos logit contributions | |
itted | +0.028 |
bar | +0.028 |
ink | +0.027 |
chalk | +0.027 |
Bree | +0.026 |
Neg logit contributions | |
restaurant | -0.035 |
Jacob | -0.034 |
Megan | -0.032 |
Jacob | -0.030 |
Pen | -0.030 |
Token | restaurant |
Token ID | 7072 |
Feature activation | -9.76e+00 |
Pos logit contributions | |
Jacob | +0.301 |
Megan | +0.296 |
Jacob | +0.274 |
Pen | +0.272 |
abel | +0.255 |
Neg logit contributions | |
ink | -0.344 |
chalk | -0.337 |
bar | -0.333 |
cho | -0.328 |
bullet | -0.327 |
Token | with |
Token ID | 351 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
Megan | +0.878 |
Jacob | +0.875 |
Pen | +0.860 |
Jacob | +0.832 |
Clive | +0.812 |
Neg logit contributions | |
chalk | -0.553 |
ink | -0.538 |
bar | -0.536 |
cho | -0.520 |
br | -0.486 |
Token | chairs |
Token ID | 18791 |
Feature activation | -3.68e+00 |
Pos logit contributions | |
Jacob | +0.149 |
Megan | +0.146 |
restaurant | +0.140 |
Jacob | +0.139 |
Pen | +0.139 |
Neg logit contributions | |
chalk | -0.102 |
ink | -0.101 |
bar | -0.095 |
cho | -0.092 |
bullet | -0.090 |
Token | and |
Token ID | 290 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
Jacob | +0.314 |
Megan | +0.312 |
restaurant | +0.297 |
Pen | +0.296 |
Jacob | +0.295 |
Neg logit contributions | |
bar | -0.237 |
chalk | -0.229 |
ink | -0.229 |
cho | -0.213 |
itted | -0.207 |
Token | tables |
Token ID | 8893 |
Feature activation | -4.63e+00 |
Pos logit contributions | |
Jacob | +0.009 |
Megan | +0.009 |
restaurant | +0.009 |
Pen | +0.008 |
Jacob | +0.008 |
Neg logit contributions | |
chalk | -0.010 |
ink | -0.010 |
bar | -0.010 |
bullet | -0.009 |
ch | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | +2.58e-01 |
Pos logit contributions | |
Jacob | +0.367 |
Megan | +0.357 |
restaurant | +0.345 |
Jacob | +0.334 |
Pen | +0.332 |
Neg logit contributions | |
bar | -0.323 |
chalk | -0.316 |
ink | -0.315 |
cho | -0.302 |
bullet | -0.297 |
Token | her |
Token ID | 607 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
Jacob | +0.280 |
Jacob | +0.275 |
Megan | +0.274 |
restaurant | +0.262 |
Pen | +0.249 |
Neg logit contributions | |
chalk | -0.183 |
ink | -0.182 |
crashing | -0.162 |
cho | -0.162 |
Bounce | -0.158 |
Token | fun |
Token ID | 1257 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
Jacob | +0.015 |
Megan | +0.014 |
Jacob | +0.014 |
restaurant | +0.013 |
Pen | +0.013 |
Neg logit contributions | |
chalk | -0.011 |
ink | -0.011 |
peac | -0.010 |
sun | -0.010 |
bar | -0.010 |
Token | day |
Token ID | 1110 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
Jacob | +0.018 |
Megan | +0.017 |
Jacob | +0.016 |
Pen | +0.016 |
restaurant | +0.016 |
Neg logit contributions | |
chalk | -0.011 |
ink | -0.011 |
bar | -0.011 |
ch | -0.010 |
cho | -0.010 |
Token | at |
Token ID | 379 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
ink | +0.102 |
bar | +0.102 |
chalk | +0.102 |
ch | +0.097 |
itted | +0.096 |
Neg logit contributions | |
Jacob | -0.130 |
Megan | -0.127 |
restaurant | -0.123 |
Jacob | -0.119 |
Pen | -0.117 |
Token | the |
Token ID | 262 |
Feature activation | -7.01e-01 |
Pos logit contributions | |
Jacob | +0.013 |
Megan | +0.013 |
Jacob | +0.013 |
restaurant | +0.012 |
abel | +0.012 |
Neg logit contributions | |
ink | -0.007 |
chalk | -0.007 |
cho | -0.006 |
bar | -0.006 |
sun | -0.006 |
Token | restaurant |
Token ID | 7072 |
Feature activation | -9.76e+00 |
Pos logit contributions | |
Jacob | +0.055 |
Megan | +0.054 |
Jacob | +0.049 |
Pen | +0.048 |
restaurant | +0.047 |
Neg logit contributions | |
ink | -0.049 |
chalk | -0.049 |
bar | -0.048 |
cho | -0.046 |
ch | -0.046 |
Token | . |
Token ID | 13 |
Feature activation | -4.85e-02 |
Pos logit contributions | |
Jacob | +0.831 |
Megan | +0.821 |
Jacob | +0.789 |
Pen | +0.777 |
Frost | +0.765 |
Neg logit contributions | |
ink | -0.565 |
chalk | -0.565 |
bar | -0.560 |
cho | -0.521 |
ch | -0.505 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +8.23e-01 |
Pos logit contributions | |
Jacob | +0.004 |
Megan | +0.004 |
restaurant | +0.003 |
Pen | +0.003 |
Jacob | +0.003 |
Neg logit contributions | |
bar | -0.004 |
ink | -0.003 |
bullet | -0.003 |
chalk | -0.003 |
cho | -0.003 |
Token | Once |
Token ID | 7454 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
ink | +0.043 |
chalk | +0.043 |
ch | +0.042 |
sun | +0.039 |
Jazz | +0.038 |
Neg logit contributions | |
Jacob | -0.073 |
Megan | -0.073 |
restaurant | -0.070 |
abel | -0.068 |
Gor | -0.066 |
Token | there |
Token ID | 612 |
Feature activation | -4.06e+00 |
Pos logit contributions | |
Jacob | +0.124 |
Megan | +0.119 |
restaurant | +0.116 |
Pen | +0.109 |
Jacob | +0.109 |
Neg logit contributions | |
bar | -0.133 |
chalk | -0.132 |
ink | -0.132 |
bullet | -0.126 |
ch | -0.126 |
Token | was |
Token ID | 373 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
restaurant | +0.326 |
Jacob | +0.311 |
Jacob | +0.297 |
Megan | +0.297 |
Pen | +0.291 |
Neg logit contributions | |
chalk | -0.252 |
bar | -0.236 |
ink | -0.231 |
Bounce | -0.226 |
cho | -0.222 |
Token | to |
Token ID | 284 |
Feature activation | +5.59e-01 |
Pos logit contributions | |
Jacob | +0.015 |
restaurant | +0.014 |
Megan | +0.013 |
Pen | +0.012 |
Jacob | +0.012 |
Neg logit contributions | |
bar | -0.016 |
plets | -0.016 |
utes | -0.015 |
bullet | -0.015 |
trigger | -0.015 |
Token | be |
Token ID | 307 |
Feature activation | +2.71e+00 |
Pos logit contributions | |
bar | +0.039 |
chalk | +0.038 |
ink | +0.038 |
bullet | +0.036 |
ch | +0.036 |
Neg logit contributions | |
Jacob | -0.045 |
Megan | -0.043 |
restaurant | -0.042 |
Pen | -0.040 |
Jacob | -0.040 |
Token | careful |
Token ID | 8161 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
ink | +0.205 |
bar | +0.204 |
chalk | +0.204 |
plets | +0.201 |
peac | +0.200 |
Neg logit contributions | |
Jacob | -0.195 |
Megan | -0.188 |
restaurant | -0.181 |
Jacob | -0.172 |
Pen | -0.168 |
Token | , |
Token ID | 11 |
Feature activation | +2.54e-01 |
Pos logit contributions | |
bar | +0.001 |
ink | +0.001 |
chalk | +0.001 |
cho | +0.001 |
ch | +0.001 |
Neg logit contributions | |
Jacob | -0.001 |
Megan | -0.001 |
restaurant | -0.001 |
Pen | -0.001 |
Jacob | -0.001 |
Token | this |
Token ID | 428 |
Feature activation | -9.82e-01 |
Pos logit contributions | |
bar | +0.017 |
chalk | +0.017 |
ink | +0.016 |
cho | +0.016 |
bullet | +0.016 |
Neg logit contributions | |
Jacob | -0.021 |
Megan | -0.020 |
restaurant | -0.020 |
Pen | -0.019 |
Jacob | -0.019 |
Token | restaurant |
Token ID | 7072 |
Feature activation | -9.76e+00 |
Pos logit contributions | |
Jacob | +0.068 |
Megan | +0.065 |
restaurant | +0.062 |
Pen | +0.059 |
Jacob | +0.059 |
Neg logit contributions | |
bar | -0.078 |
ink | -0.078 |
chalk | -0.078 |
ch | -0.075 |
cho | -0.075 |
Token | is |
Token ID | 318 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
Jacob | +0.728 |
Megan | +0.718 |
Pen | +0.669 |
Jacob | +0.655 |
restaurant | +0.643 |
Neg logit contributions | |
ink | -0.708 |
cho | -0.705 |
chalk | -0.699 |
bar | -0.688 |
ch | -0.683 |
Token | not |
Token ID | 407 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
Jacob | +0.133 |
Megan | +0.127 |
restaurant | +0.125 |
Pen | +0.117 |
Jacob | +0.116 |
Neg logit contributions | |
bar | -0.127 |
chalk | -0.126 |
ink | -0.124 |
itted | -0.121 |
bullet | -0.120 |
Token | for |
Token ID | 329 |
Feature activation | +5.25e-01 |
Pos logit contributions | |
Jacob | +0.127 |
Megan | +0.123 |
restaurant | +0.121 |
Pen | +0.114 |
Jacob | +0.113 |
Neg logit contributions | |
bar | -0.088 |
chalk | -0.087 |
ink | -0.087 |
bullet | -0.082 |
itted | -0.082 |
Token | playing |
Token ID | 2712 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
chalk | +0.032 |
ink | +0.032 |
bar | +0.032 |
cho | +0.030 |
bullet | +0.030 |
Neg logit contributions | |
Jacob | -0.045 |
Megan | -0.044 |
restaurant | -0.043 |
Jacob | -0.041 |
Pen | -0.041 |
Token | ". |
Token ID | 1911 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
ink | +0.111 |
chalk | +0.110 |
golf | +0.106 |
cho | +0.106 |
bullet | +0.103 |
Neg logit contributions | |
Megan | -0.160 |
Jacob | -0.154 |
Pen | -0.147 |
abel | -0.146 |
Jacob | -0.143 |
Token | was |
Token ID | 373 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
ink | +0.022 |
chalk | +0.022 |
ch | +0.022 |
bar | +0.022 |
cho | +0.021 |
Neg logit contributions | |
Jacob | -0.030 |
restaurant | -0.029 |
Megan | -0.029 |
Jacob | -0.028 |
Pen | -0.027 |
Token | to |
Token ID | 284 |
Feature activation | +5.67e-01 |
Pos logit contributions | |
Jacob | +0.097 |
Megan | +0.094 |
restaurant | +0.090 |
Jacob | +0.088 |
Pen | +0.087 |
Neg logit contributions | |
ink | -0.086 |
chalk | -0.086 |
bar | -0.085 |
cho | -0.080 |
ch | -0.080 |
Token | go |
Token ID | 467 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
ink | +0.039 |
chalk | +0.039 |
bar | +0.036 |
sun | +0.034 |
bullet | +0.034 |
Neg logit contributions | |
Jacob | -0.045 |
Megan | -0.044 |
Jacob | -0.043 |
Pen | -0.041 |
restaurant | -0.041 |
Token | to |
Token ID | 284 |
Feature activation | +5.61e-01 |
Pos logit contributions | |
bar | +0.109 |
chalk | +0.108 |
ink | +0.108 |
bullet | +0.103 |
ch | +0.103 |
Neg logit contributions | |
Jacob | -0.130 |
Megan | -0.125 |
restaurant | -0.122 |
Pen | -0.115 |
Jacob | -0.115 |
Token | the |
Token ID | 262 |
Feature activation | -7.03e-01 |
Pos logit contributions | |
ink | +0.034 |
chalk | +0.034 |
bar | +0.031 |
bullet | +0.030 |
Bree | +0.029 |
Neg logit contributions | |
Jacob | -0.049 |
Megan | -0.048 |
Jacob | -0.047 |
restaurant | -0.045 |
abel | -0.044 |
Token | restaurant |
Token ID | 7072 |
Feature activation | -9.76e+00 |
Pos logit contributions | |
Jacob | +0.050 |
Megan | +0.048 |
restaurant | +0.045 |
Jacob | +0.044 |
Pen | +0.044 |
Neg logit contributions | |
ink | -0.055 |
bar | -0.055 |
chalk | -0.055 |
bullet | -0.052 |
ch | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | -5.68e-02 |
Pos logit contributions | |
Megan | +0.873 |
Jacob | +0.868 |
Jacob | +0.833 |
Pen | +0.831 |
Clive | +0.797 |
Neg logit contributions | |
bar | -0.556 |
chalk | -0.544 |
ink | -0.537 |
cho | -0.483 |
ch | -0.465 |
Token | Delicious |
Token ID | 39528 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
Jacob | +0.005 |
Megan | +0.005 |
restaurant | +0.005 |
Jacob | +0.004 |
Pen | +0.004 |
Neg logit contributions | |
bar | -0.004 |
chalk | -0.004 |
ink | -0.004 |
bullet | -0.003 |
ch | -0.003 |
Token | food |
Token ID | 2057 |
Feature activation | -4.69e+00 |
Pos logit contributions | |
Megan | +0.104 |
Jacob | +0.103 |
Jacob | +0.097 |
Pen | +0.097 |
| | +0.091 |
Neg logit contributions | |
ink | -0.093 |
chalk | -0.089 |
cho | -0.086 |
ch | -0.085 |
bar | -0.083 |
Token | and |
Token ID | 290 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
Jacob | +0.383 |
Megan | +0.374 |
restaurant | +0.358 |
Jacob | +0.354 |
Pen | +0.349 |
Neg logit contributions | |
ink | -0.314 |
chalk | -0.312 |
bar | -0.305 |
cho | -0.291 |
bullet | -0.289 |
Token | a |
Token ID | 257 |
Feature activation | +4.16e-01 |
Pos logit contributions | |
Jacob | +0.011 |
Megan | +0.010 |
restaurant | +0.010 |
Jacob | +0.010 |
Pen | +0.010 |
Neg logit contributions | |
ink | -0.008 |
chalk | -0.008 |
bar | -0.008 |
cho | -0.008 |
bullet | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
Jacob | +0.449 |
Megan | +0.440 |
Jacob | +0.417 |
restaurant | +0.402 |
Pen | +0.399 |
Neg logit contributions | |
bar | -0.336 |
ink | -0.335 |
chalk | -0.324 |
ch | -0.317 |
cho | -0.316 |
Token | we |
Token ID | 356 |
Feature activation | -3.41e-01 |
Pos logit contributions | |
Jacob | +0.010 |
Megan | +0.010 |
restaurant | +0.009 |
Jacob | +0.009 |
Pen | +0.009 |
Neg logit contributions | |
bar | -0.008 |
chalk | -0.008 |
ink | -0.008 |
bullet | -0.008 |
ch | -0.008 |
Token | can |
Token ID | 460 |
Feature activation | +7.49e-01 |
Pos logit contributions | |
Jacob | +0.031 |
Megan | +0.030 |
Jacob | +0.029 |
restaurant | +0.028 |
Pen | +0.027 |
Neg logit contributions | |
bar | -0.021 |
chalk | -0.021 |
ink | -0.020 |
ch | -0.020 |
cho | -0.019 |
Token | open |
Token ID | 1280 |
Feature activation | -3.61e+00 |
Pos logit contributions | |
bar | +0.046 |
chalk | +0.043 |
ink | +0.042 |
peac | +0.040 |
sun | +0.040 |
Neg logit contributions | |
Jacob | -0.068 |
Megan | -0.066 |
Jacob | -0.063 |
restaurant | -0.063 |
Pen | -0.061 |
Token | the |
Token ID | 262 |
Feature activation | -7.03e-01 |
Pos logit contributions | |
Jacob | +0.328 |
Megan | +0.318 |
restaurant | +0.311 |
Jacob | +0.300 |
Pen | +0.298 |
Neg logit contributions | |
chalk | -0.208 |
ink | -0.206 |
bar | -0.203 |
cho | -0.191 |
ch | -0.190 |
Token | restaurant |
Token ID | 7072 |
Feature activation | -9.76e+00 |
Pos logit contributions | |
Jacob | +0.057 |
Megan | +0.056 |
restaurant | +0.056 |
Pen | +0.052 |
Jacob | +0.051 |
Neg logit contributions | |
bar | -0.047 |
ink | -0.045 |
chalk | -0.045 |
itted | -0.044 |
bullet | -0.044 |
Token | again |
Token ID | 757 |
Feature activation | -2.37e+00 |
Pos logit contributions | |
Jacob | +0.846 |
Megan | +0.828 |
Jacob | +0.812 |
Pen | +0.795 |
Clive | +0.771 |
Neg logit contributions | |
chalk | -0.557 |
ink | -0.550 |
bar | -0.534 |
sun | -0.503 |
cho | -0.500 |
Token | ." |
Token ID | 526 |
Feature activation | -9.96e-01 |
Pos logit contributions | |
Jacob | +0.170 |
Megan | +0.163 |
Jacob | +0.153 |
restaurant | +0.151 |
Pen | +0.146 |
Neg logit contributions | |
ink | -0.184 |
chalk | -0.182 |
bar | -0.178 |
ch | -0.174 |
cho | -0.172 |
Token |
|
Token ID | 198 |
Feature activation | +9.70e-02 |
Pos logit contributions | |
Jacob | +0.083 |
restaurant | +0.080 |
Megan | +0.080 |
Jacob | +0.075 |
Pen | +0.075 |
Neg logit contributions | |
chalk | -0.062 |
ink | -0.061 |
bar | -0.060 |
Bounce | -0.058 |
Bree | -0.058 |
Token |
|
Token ID | 198 |
Feature activation | +8.95e-02 |
Pos logit contributions | |
chalk | +0.006 |
ink | +0.006 |
bar | +0.006 |
ch | +0.005 |
cho | +0.005 |
Neg logit contributions | |
Jacob | -0.008 |
Megan | -0.008 |
restaurant | -0.008 |
Pen | -0.008 |
Jacob | -0.008 |
Token | Her |
Token ID | 9360 |
Feature activation | +2.52e+00 |
Pos logit contributions | |
bar | +0.005 |
chalk | +0.005 |
ink | +0.005 |
ch | +0.005 |
bullet | +0.005 |
Neg logit contributions | |
Jacob | -0.008 |
Megan | -0.008 |
restaurant | -0.008 |
Pen | -0.007 |
Jacob | -0.007 |
Token | were |
Token ID | 547 |
Feature activation | -5.42e-01 |
Pos logit contributions | |
chalk | +0.039 |
ink | +0.038 |
bar | +0.038 |
cho | +0.036 |
bullet | +0.036 |
Neg logit contributions | |
Jacob | -0.053 |
Megan | -0.052 |
restaurant | -0.050 |
Pen | -0.048 |
Jacob | -0.048 |
Token | going |
Token ID | 1016 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
Jacob | +0.054 |
Megan | +0.053 |
Jacob | +0.052 |
Pen | +0.050 |
abel | +0.049 |
Neg logit contributions | |
ink | -0.027 |
chalk | -0.026 |
cho | -0.026 |
sun | -0.024 |
crashing | -0.024 |
Token | to |
Token ID | 284 |
Feature activation | +5.78e-01 |
Pos logit contributions | |
bar | +0.066 |
ink | +0.065 |
chalk | +0.065 |
bullet | +0.063 |
itted | +0.062 |
Neg logit contributions | |
Jacob | -0.091 |
restaurant | -0.089 |
Megan | -0.087 |
Pen | -0.080 |
Jacob | -0.079 |
Token | a |
Token ID | 257 |
Feature activation | +4.30e-01 |
Pos logit contributions | |
chalk | +0.037 |
ink | +0.037 |
bar | +0.035 |
peac | +0.033 |
cho | +0.033 |
Neg logit contributions | |
Jacob | -0.049 |
Megan | -0.048 |
Jacob | -0.046 |
abel | -0.044 |
restaurant | -0.044 |
Token | nice |
Token ID | 3621 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
chalk | +0.034 |
ink | +0.034 |
ch | +0.034 |
bar | +0.034 |
cho | +0.034 |
Neg logit contributions | |
Jacob | -0.030 |
Megan | -0.028 |
Pen | -0.026 |
Jacob | -0.026 |
Nick | -0.025 |
Token | restaurant |
Token ID | 7072 |
Feature activation | -9.76e+00 |
Pos logit contributions | |
ink | +0.012 |
chalk | +0.011 |
sun | +0.011 |
ch | +0.011 |
gol | +0.011 |
Neg logit contributions | |
Jacob | -0.009 |
Megan | -0.008 |
Jacob | -0.008 |
Pen | -0.008 |
Clive | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -4.06e-02 |
Pos logit contributions | |
Jacob | +0.901 |
Megan | +0.898 |
Pen | +0.873 |
Jacob | +0.864 |
Clive | +0.844 |
Neg logit contributions | |
chalk | -0.486 |
ink | -0.476 |
bar | -0.470 |
cho | -0.461 |
ch | -0.435 |
Token | When |
Token ID | 1649 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
Jacob | +0.003 |
Megan | +0.003 |
restaurant | +0.003 |
Pen | +0.003 |
Jacob | +0.003 |
Neg logit contributions | |
bar | -0.003 |
ink | -0.003 |
chalk | -0.003 |
bullet | -0.003 |
ch | -0.003 |
Token | they |
Token ID | 484 |
Feature activation | +5.96e-01 |
Pos logit contributions | |
Jacob | +0.015 |
Megan | +0.014 |
restaurant | +0.014 |
Jacob | +0.014 |
Pen | +0.013 |
Neg logit contributions | |
ink | -0.008 |
chalk | -0.008 |
bar | -0.007 |
cho | -0.007 |
ch | -0.007 |
Token | got |
Token ID | 1392 |
Feature activation | +5.27e-01 |
Pos logit contributions | |
chalk | +0.019 |
bar | +0.018 |
ink | +0.017 |
printed | +0.016 |
ch | +0.015 |
Neg logit contributions | |
Jacob | -0.066 |
Megan | -0.064 |
Pen | -0.064 |
Jacob | -0.063 |
elope | -0.059 |
Token | there |
Token ID | 612 |
Feature activation | -4.06e+00 |
Pos logit contributions | |
ink | +0.024 |
chalk | +0.024 |
bar | +0.023 |
cho | +0.022 |
ch | +0.021 |
Neg logit contributions | |
Jacob | -0.054 |
Megan | -0.052 |
Jacob | -0.052 |
Pen | -0.050 |
abel | -0.049 |
Token | Lily |
Token ID | 20037 |
Feature activation | +6.22e-01 |
Pos logit contributions | |
bar | +0.017 |
chalk | +0.017 |
ink | +0.017 |
bullet | +0.016 |
ch | +0.016 |
Neg logit contributions | |
Jacob | -0.024 |
Megan | -0.023 |
restaurant | -0.022 |
Jacob | -0.021 |
Pen | -0.021 |
Token | 's |
Token ID | 338 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
ink | +0.040 |
bar | +0.040 |
chalk | +0.040 |
ch | +0.037 |
bullet | +0.037 |
Neg logit contributions | |
Jacob | -0.052 |
Megan | -0.051 |
restaurant | -0.049 |
Jacob | -0.047 |
Pen | -0.047 |
Token | family |
Token ID | 1641 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
Jacob | +0.091 |
Megan | +0.088 |
restaurant | +0.085 |
Jacob | +0.082 |
Pen | +0.081 |
Neg logit contributions | |
chalk | -0.074 |
ink | -0.074 |
bar | -0.073 |
bullet | -0.070 |
ch | -0.069 |
Token | left |
Token ID | 1364 |
Feature activation | -2.74e+00 |
Pos logit contributions | |
Jacob | +0.164 |
Megan | +0.160 |
restaurant | +0.153 |
Jacob | +0.150 |
Pen | +0.149 |
Neg logit contributions | |
ink | -0.125 |
chalk | -0.124 |
bar | -0.123 |
cho | -0.118 |
bullet | -0.116 |
Token | the |
Token ID | 262 |
Feature activation | -7.00e-01 |
Pos logit contributions | |
Jacob | +0.226 |
Megan | +0.218 |
restaurant | +0.211 |
Jacob | +0.207 |
Pen | +0.202 |
Neg logit contributions | |
ink | -0.182 |
chalk | -0.181 |
bar | -0.174 |
cho | -0.168 |
bullet | -0.166 |
Token | restaurant |
Token ID | 7072 |
Feature activation | -9.76e+00 |
Pos logit contributions | |
Jacob | +0.053 |
Megan | +0.051 |
Jacob | +0.047 |
Pen | +0.047 |
restaurant | +0.046 |
Neg logit contributions | |
ink | -0.053 |
chalk | -0.053 |
bar | -0.051 |
bullet | -0.049 |
cho | -0.049 |
Token | and |
Token ID | 290 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
Jacob | +0.850 |
Megan | +0.828 |
Pen | +0.793 |
Jacob | +0.790 |
Clive | +0.767 |
Neg logit contributions | |
chalk | -0.598 |
bar | -0.590 |
ink | -0.587 |
cho | -0.537 |
itted | -0.532 |
Token | walked |
Token ID | 6807 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
Jacob | +0.010 |
Megan | +0.010 |
Jacob | +0.010 |
restaurant | +0.009 |
Pen | +0.009 |
Neg logit contributions | |
ink | -0.007 |
chalk | -0.007 |
bar | -0.007 |
cho | -0.007 |
bullet | -0.006 |
Token | home |
Token ID | 1363 |
Feature activation | +5.75e-01 |
Pos logit contributions | |
Jacob | +0.134 |
Megan | +0.130 |
restaurant | +0.127 |
Pen | +0.121 |
Jacob | +0.121 |
Neg logit contributions | |
bar | -0.085 |
chalk | -0.084 |
ink | -0.084 |
ch | -0.080 |
bullet | -0.080 |
Token | . |
Token ID | 13 |
Feature activation | -4.86e-02 |
Pos logit contributions | |
ink | +0.031 |
bar | +0.031 |
chalk | +0.031 |
cho | +0.031 |
itted | +0.030 |
Neg logit contributions | |
Jacob | -0.051 |
Jacob | -0.050 |
Megan | -0.048 |
restaurant | -0.048 |
abel | -0.046 |
Token | On |
Token ID | 1550 |
Feature activation | -4.20e+00 |
Pos logit contributions | |
Jacob | +0.004 |
Megan | +0.004 |
restaurant | +0.004 |
Jacob | +0.004 |
Pen | +0.004 |
Neg logit contributions | |
chalk | -0.003 |
ink | -0.003 |
bar | -0.003 |
ch | -0.003 |
bullet | -0.003 |
Token | her |
Token ID | 607 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
bar | +0.122 |
bullet | +0.117 |
chalk | +0.117 |
ink | +0.115 |
peac | +0.115 |
Neg logit contributions | |
Jacob | -0.134 |
Megan | -0.129 |
restaurant | -0.124 |
Pen | -0.119 |
Jacob | -0.115 |
Token | to |
Token ID | 284 |
Feature activation | +5.61e-01 |
Pos logit contributions | |
Jacob | +0.013 |
restaurant | +0.013 |
Megan | +0.013 |
Pen | +0.012 |
Jacob | +0.012 |
Neg logit contributions | |
bar | -0.010 |
ink | -0.010 |
cho | -0.010 |
bullet | -0.010 |
itted | -0.010 |
Token | a |
Token ID | 257 |
Feature activation | +4.16e-01 |
Pos logit contributions | |
chalk | +0.033 |
ink | +0.032 |
bar | +0.030 |
Bree | +0.029 |
ch | +0.029 |
Neg logit contributions | |
Jacob | -0.050 |
Megan | -0.049 |
restaurant | -0.047 |
Jacob | -0.047 |
abel | -0.045 |
Token | y |
Token ID | 331 |
Feature activation | -4.19e+00 |
Pos logit contributions | |
chalk | +0.028 |
bar | +0.028 |
ink | +0.028 |
ch | +0.027 |
cho | +0.027 |
Neg logit contributions | |
Jacob | -0.034 |
Megan | -0.033 |
Pen | -0.030 |
Jacob | -0.030 |
restaurant | -0.029 |
Token | ummy |
Token ID | 13513 |
Feature activation | -9.29e-01 |
Pos logit contributions | |
Jacob | +0.415 |
Megan | +0.400 |
Pen | +0.389 |
Jacob | +0.379 |
restaurant | +0.360 |
Neg logit contributions | |
bar | -0.228 |
ink | -0.221 |
chalk | -0.219 |
ch | -0.199 |
cho | -0.195 |
Token | restaurant |
Token ID | 7072 |
Feature activation | -9.75e+00 |
Pos logit contributions | |
Jacob | +0.067 |
Megan | +0.066 |
Pen | +0.060 |
Jacob | +0.059 |
Nick | +0.057 |
Neg logit contributions | |
chalk | -0.073 |
ink | -0.072 |
bar | -0.070 |
ch | -0.070 |
cho | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | -4.61e-02 |
Pos logit contributions | |
Jacob | +0.871 |
Megan | +0.861 |
Jacob | +0.819 |
Pen | +0.817 |
Clive | +0.793 |
Neg logit contributions | |
chalk | -0.562 |
bar | -0.559 |
ink | -0.554 |
cho | -0.517 |
ch | -0.500 |
Token | They |
Token ID | 1119 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
Jacob | +0.004 |
Megan | +0.004 |
restaurant | +0.003 |
Pen | +0.003 |
Jacob | +0.003 |
Neg logit contributions | |
bar | -0.003 |
bullet | -0.003 |
ink | -0.003 |
chalk | -0.003 |
cho | -0.003 |
Token | ate |
Token ID | 15063 |
Feature activation | -2.74e+00 |
Pos logit contributions | |
Megan | +0.013 |
Jacob | +0.013 |
Jacob | +0.013 |
Pen | +0.013 |
Clive | +0.012 |
Neg logit contributions | |
chalk | -0.007 |
printed | -0.007 |
bar | -0.007 |
cho | -0.006 |
ink | -0.006 |
Token | spaghetti |
Token ID | 48241 |
Feature activation | -4.27e+00 |
Pos logit contributions | |
Jacob | +0.231 |
Megan | +0.224 |
Jacob | +0.220 |
Pen | +0.217 |
abel | +0.215 |
Neg logit contributions | |
chalk | -0.170 |
ink | -0.169 |
cho | -0.159 |
sun | -0.155 |
ch | -0.144 |
Token | and |
Token ID | 290 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
Megan | +0.379 |
Jacob | +0.372 |
Pen | +0.361 |
Jacob | +0.356 |
restaurant | +0.349 |
Neg logit contributions | |
ink | -0.241 |
bar | -0.238 |
chalk | -0.238 |
cho | -0.227 |
ch | -0.218 |
Token | The |
Token ID | 464 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
bar | +0.006 |
chalk | +0.006 |
ink | +0.006 |
ch | +0.006 |
bullet | +0.006 |
Neg logit contributions | |
Jacob | -0.009 |
Megan | -0.009 |
restaurant | -0.009 |
Pen | -0.008 |
Jacob | -0.008 |
Token | wealthy |
Token ID | 11574 |
Feature activation | -3.00e+00 |
Pos logit contributions | |
ink | +0.069 |
chalk | +0.068 |
bar | +0.066 |
sun | +0.066 |
bullet | +0.065 |
Neg logit contributions | |
Jacob | -0.082 |
Megan | -0.080 |
Pen | -0.075 |
Jacob | -0.073 |
restaurant | -0.071 |
Token | man |
Token ID | 582 |
Feature activation | -7.98e-01 |
Pos logit contributions | |
Jacob | +0.270 |
Megan | +0.269 |
Pen | +0.256 |
| | +0.252 |
Jacob | +0.251 |
Neg logit contributions | |
sun | -0.156 |
ink | -0.152 |
chalk | -0.149 |
bar | -0.146 |
gol | -0.145 |
Token | left |
Token ID | 1364 |
Feature activation | -2.75e+00 |
Pos logit contributions | |
Jacob | +0.066 |
Megan | +0.065 |
restaurant | +0.065 |
Jacob | +0.061 |
Pen | +0.061 |
Neg logit contributions | |
bar | -0.048 |
ink | -0.045 |
chalk | -0.044 |
cho | -0.043 |
printed | -0.042 |
Token | the |
Token ID | 262 |
Feature activation | -7.11e-01 |
Pos logit contributions | |
Jacob | +0.225 |
Megan | +0.219 |
restaurant | +0.212 |
Jacob | +0.207 |
Pen | +0.204 |
Neg logit contributions | |
ink | -0.174 |
bar | -0.173 |
chalk | -0.172 |
cho | -0.166 |
bullet | -0.164 |
Token | restaurant |
Token ID | 7072 |
Feature activation | -9.75e+00 |
Pos logit contributions | |
Jacob | +0.047 |
Megan | +0.046 |
Jacob | +0.041 |
Pen | +0.041 |
Nick | +0.039 |
Neg logit contributions | |
ink | -0.060 |
chalk | -0.059 |
bar | -0.058 |
sun | -0.056 |
bullet | -0.056 |
Token | with |
Token ID | 351 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
Jacob | +0.836 |
Megan | +0.831 |
Pen | +0.791 |
Jacob | +0.786 |
restaurant | +0.750 |
Neg logit contributions | |
bar | -0.593 |
chalk | -0.558 |
ink | -0.555 |
cho | -0.519 |
itted | -0.502 |
Token | a |
Token ID | 257 |
Feature activation | +4.11e-01 |
Pos logit contributions | |
Jacob | +0.140 |
Megan | +0.135 |
restaurant | +0.131 |
Jacob | +0.126 |
Pen | +0.126 |
Neg logit contributions | |
bar | -0.108 |
ink | -0.108 |
chalk | -0.108 |
bullet | -0.102 |
cho | -0.101 |
Token | big |
Token ID | 1263 |
Feature activation | +3.06e-01 |
Pos logit contributions | |
bar | +0.026 |
ink | +0.025 |
chalk | +0.025 |
itted | +0.025 |
ch | +0.024 |
Neg logit contributions | |
Jacob | -0.035 |
restaurant | -0.034 |
Megan | -0.034 |
Pen | -0.031 |
Jacob | -0.031 |
Token | smile |
Token ID | 8212 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
chalk | +0.020 |
ink | +0.020 |
bar | +0.020 |
bullet | +0.019 |
cho | +0.019 |
Neg logit contributions | |
Jacob | -0.025 |
Megan | -0.025 |
Pen | -0.023 |
restaurant | -0.023 |
Jacob | -0.023 |
Token | on |
Token ID | 319 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
chalk | +0.078 |
ink | +0.077 |
bar | +0.077 |
bullet | +0.074 |
Bounce | +0.074 |
Neg logit contributions | |
Jacob | -0.095 |
Megan | -0.090 |
restaurant | -0.089 |
Pen | -0.084 |
Jacob | -0.081 |
Token | song |
Token ID | 3496 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
Jacob | +0.275 |
Megan | +0.268 |
restaurant | +0.256 |
Jacob | +0.253 |
Pen | +0.249 |
Neg logit contributions | |
ink | -0.160 |
chalk | -0.159 |
bar | -0.150 |
ch | -0.148 |
bullet | -0.145 |
Token | and |
Token ID | 290 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
Jacob | +0.010 |
Megan | +0.010 |
restaurant | +0.010 |
Jacob | +0.009 |
Pen | +0.009 |
Neg logit contributions | |
bar | -0.010 |
ink | -0.010 |
chalk | -0.010 |
ch | -0.009 |
cho | -0.009 |
Token | everyone |
Token ID | 2506 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
Jacob | +0.010 |
Megan | +0.010 |
restaurant | +0.009 |
Pen | +0.009 |
Nick | +0.009 |
Neg logit contributions | |
bar | -0.009 |
ink | -0.008 |
chalk | -0.008 |
bullet | -0.008 |
itted | -0.008 |
Token | in |
Token ID | 287 |
Feature activation | -5.35e-01 |
Pos logit contributions | |
chalk | +0.078 |
ink | +0.078 |
bar | +0.077 |
ch | +0.075 |
cho | +0.075 |
Neg logit contributions | |
Jacob | -0.107 |
restaurant | -0.103 |
Megan | -0.102 |
Jacob | -0.097 |
Pen | -0.094 |
Token | the |
Token ID | 262 |
Feature activation | -7.11e-01 |
Pos logit contributions | |
Jacob | +0.047 |
restaurant | +0.046 |
Megan | +0.046 |
Jacob | +0.044 |
Gor | +0.044 |
Neg logit contributions | |
ink | -0.030 |
chalk | -0.030 |
sun | -0.027 |
crashing | -0.026 |
gol | -0.026 |
Token | restaurant |
Token ID | 7072 |
Feature activation | -9.75e+00 |
Pos logit contributions | |
restaurant | +0.051 |
Jacob | +0.051 |
Megan | +0.049 |
Pen | +0.045 |
Jacob | +0.045 |
Neg logit contributions | |
bar | -0.056 |
bullet | -0.052 |
chalk | -0.052 |
itted | -0.052 |
ink | -0.051 |
Token | joined |
Token ID | 5399 |
Feature activation | +6.32e-02 |
Pos logit contributions | |
Jacob | +0.770 |
Megan | +0.743 |
restaurant | +0.712 |
Jacob | +0.709 |
Pen | +0.703 |
Neg logit contributions | |
chalk | -0.650 |
ink | -0.645 |
bar | -0.634 |
cho | -0.615 |
itted | -0.594 |
Token | in |
Token ID | 287 |
Feature activation | -5.37e-01 |
Pos logit contributions | |
chalk | +0.004 |
ink | +0.004 |
bar | +0.004 |
ch | +0.004 |
itted | +0.004 |
Neg logit contributions | |
Jacob | -0.005 |
Megan | -0.005 |
restaurant | -0.005 |
Jacob | -0.005 |
Pen | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | -4.85e-02 |
Pos logit contributions | |
Jacob | +0.043 |
Megan | +0.042 |
restaurant | +0.041 |
Jacob | +0.040 |
Pen | +0.039 |
Neg logit contributions | |
ink | -0.035 |
chalk | -0.035 |
cho | -0.033 |
ch | -0.033 |
bar | -0.033 |
Token | They |
Token ID | 1119 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
Jacob | +0.004 |
Megan | +0.004 |
restaurant | +0.004 |
Pen | +0.004 |
Jacob | +0.004 |
Neg logit contributions | |
bar | -0.003 |
ink | -0.003 |
chalk | -0.003 |
bullet | -0.003 |
cho | -0.003 |
Token | sang |
Token ID | 25889 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
Jacob | +0.013 |
Megan | +0.013 |
restaurant | +0.013 |
Jacob | +0.012 |
Pen | +0.012 |
Neg logit contributions | |
chalk | -0.008 |
ink | -0.008 |
bar | -0.008 |
cho | -0.008 |
ch | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -4.67e-02 |
Pos logit contributions | |
Jacob | +0.360 |
Megan | +0.359 |
Jacob | +0.346 |
Pen | +0.340 |
restaurant | +0.338 |
Neg logit contributions | |
ink | -0.247 |
chalk | -0.242 |
bullet | -0.222 |
cho | -0.219 |
bar | -0.219 |
Token | They |
Token ID | 1119 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
Jacob | +0.004 |
restaurant | +0.004 |
Megan | +0.004 |
abel | +0.004 |
Jacob | +0.004 |
Neg logit contributions | |
chalk | -0.003 |
ink | -0.002 |
Bounce | -0.002 |
bar | -0.002 |
itted | -0.002 |
Token | looked |
Token ID | 3114 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
Jacob | +0.013 |
restaurant | +0.013 |
Megan | +0.013 |
Jacob | +0.012 |
Pen | +0.012 |
Neg logit contributions | |
chalk | -0.007 |
ink | -0.007 |
bar | -0.006 |
bullet | -0.006 |
printed | -0.006 |
Token | around |
Token ID | 1088 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
Jacob | +0.145 |
Megan | +0.141 |
Jacob | +0.136 |
Pen | +0.134 |
restaurant | +0.132 |
Neg logit contributions | |
chalk | -0.076 |
ink | -0.076 |
ch | -0.069 |
cho | -0.068 |
bullet | -0.066 |
Token | the |
Token ID | 262 |
Feature activation | -7.01e-01 |
Pos logit contributions | |
Jacob | +0.178 |
Megan | +0.173 |
Jacob | +0.171 |
Pen | +0.164 |
restaurant | +0.164 |
Neg logit contributions | |
ink | -0.104 |
chalk | -0.103 |
sun | -0.093 |
crashing | -0.091 |
cho | -0.088 |
Token | restaurant |
Token ID | 7072 |
Feature activation | -9.75e+00 |
Pos logit contributions | |
Jacob | +0.051 |
Megan | +0.050 |
restaurant | +0.045 |
Pen | +0.045 |
Jacob | +0.045 |
Neg logit contributions | |
ink | -0.054 |
chalk | -0.054 |
bar | -0.053 |
bullet | -0.051 |
ch | -0.051 |
Token | . |
Token ID | 13 |
Feature activation | -5.35e-02 |
Pos logit contributions | |
Jacob | +0.906 |
Pen | +0.894 |
Megan | +0.893 |
Jacob | +0.891 |
Clive | +0.869 |
Neg logit contributions | |
chalk | -0.474 |
ink | -0.473 |
sun | -0.438 |
bar | -0.426 |
cho | -0.412 |
Token | They |
Token ID | 1119 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
Jacob | +0.005 |
restaurant | +0.005 |
Megan | +0.005 |
abel | +0.005 |
Jacob | +0.005 |
Neg logit contributions | |
chalk | -0.003 |
ink | -0.003 |
Bounce | -0.002 |
itted | -0.002 |
Bree | -0.002 |
Token | saw |
Token ID | 2497 |
Feature activation | -2.30e-01 |
Pos logit contributions | |
Jacob | +0.014 |
Megan | +0.014 |
Pen | +0.014 |
Jacob | +0.014 |
ide | +0.014 |
Neg logit contributions | |
chalk | -0.005 |
printed | -0.004 |
ink | -0.004 |
peac | -0.004 |
bullet | -0.004 |
Token | other |
Token ID | 584 |
Feature activation | +3.62e+00 |
Pos logit contributions | |
Jacob | +0.021 |
Megan | +0.020 |
restaurant | +0.019 |
Jacob | +0.019 |
Pen | +0.019 |
Neg logit contributions | |
chalk | -0.013 |
ink | -0.013 |
bar | -0.013 |
bullet | -0.012 |
cho | -0.012 |
Token | people |
Token ID | 661 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
chalk | +0.167 |
sun | +0.166 |
ink | +0.161 |
ol | +0.134 |
wind | +0.129 |
Neg logit contributions | |
Pen | -0.369 |
Megan | -0.360 |
Jacob | -0.359 |
Jacob | -0.354 |
zzy | -0.350 |
Token |
|
Token ID | 198 |
Feature activation | +9.70e-02 |
Pos logit contributions | |
bar | +0.007 |
chalk | +0.007 |
ink | +0.006 |
ch | +0.006 |
bullet | +0.006 |
Neg logit contributions | |
Jacob | -0.008 |
Megan | -0.007 |
restaurant | -0.007 |
Pen | -0.007 |
Jacob | -0.007 |
Token | The |
Token ID | 464 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
bar | +0.006 |
ink | +0.006 |
chalk | +0.006 |
itted | +0.005 |
bullet | +0.005 |
Neg logit contributions | |
Jacob | -0.009 |
Megan | -0.009 |
restaurant | -0.008 |
Jacob | -0.008 |
Pen | -0.008 |
Token | people |
Token ID | 661 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
ink | +0.068 |
chalk | +0.068 |
bar | +0.066 |
bullet | +0.064 |
ch | +0.063 |
Neg logit contributions | |
Jacob | -0.084 |
Megan | -0.082 |
restaurant | -0.077 |
Pen | -0.076 |
Jacob | -0.075 |
Token | in |
Token ID | 287 |
Feature activation | -5.42e-01 |
Pos logit contributions | |
bar | +0.008 |
ink | +0.008 |
chalk | +0.008 |
bullet | +0.008 |
ch | +0.008 |
Neg logit contributions | |
Jacob | -0.009 |
Megan | -0.009 |
restaurant | -0.009 |
Pen | -0.008 |
Jacob | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | -7.08e-01 |
Pos logit contributions | |
Jacob | +0.048 |
Megan | +0.047 |
restaurant | +0.046 |
Jacob | +0.045 |
Pen | +0.043 |
Neg logit contributions | |
ink | -0.032 |
chalk | -0.032 |
bullet | -0.030 |
cho | -0.029 |
ch | -0.029 |
Token | restaurant |
Token ID | 7072 |
Feature activation | -9.75e+00 |
Pos logit contributions | |
Jacob | +0.043 |
Megan | +0.041 |
restaurant | +0.041 |
Pen | +0.037 |
Jacob | +0.037 |
Neg logit contributions | |
bar | -0.062 |
chalk | -0.062 |
ink | -0.061 |
bullet | -0.060 |
ch | -0.059 |
Token | were |
Token ID | 547 |
Feature activation | -5.55e-01 |
Pos logit contributions | |
Jacob | +0.796 |
Megan | +0.780 |
Pen | +0.742 |
Jacob | +0.737 |
restaurant | +0.729 |
Neg logit contributions | |
chalk | -0.628 |
ink | -0.627 |
bar | -0.608 |
cho | -0.599 |
bullet | -0.586 |
Token | not |
Token ID | 407 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
Jacob | +0.053 |
Megan | +0.051 |
restaurant | +0.050 |
Jacob | +0.048 |
Pen | +0.048 |
Neg logit contributions | |
bar | -0.030 |
ink | -0.030 |
chalk | -0.029 |
cho | -0.028 |
bullet | -0.028 |
Token | happy |
Token ID | 3772 |
Feature activation | -5.41e-01 |
Pos logit contributions | |
Jacob | +0.123 |
Megan | +0.119 |
restaurant | +0.116 |
Pen | +0.110 |
Jacob | +0.109 |
Neg logit contributions | |
bar | -0.091 |
chalk | -0.089 |
ink | -0.089 |
bullet | -0.086 |
ch | -0.085 |
Token | . |
Token ID | 13 |
Feature activation | -4.74e-02 |
Pos logit contributions | |
Jacob | +0.043 |
Megan | +0.042 |
Jacob | +0.042 |
Pen | +0.041 |
restaurant | +0.040 |
Neg logit contributions | |
ink | -0.037 |
chalk | -0.037 |
bar | -0.033 |
cho | -0.033 |
bullet | -0.033 |
Token | They |
Token ID | 1119 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
Jacob | +0.004 |
restaurant | +0.004 |
Megan | +0.004 |
Jacob | +0.004 |
Pen | +0.004 |
Neg logit contributions | |
ink | -0.003 |
chalk | -0.002 |
bar | -0.002 |
itted | -0.002 |
bullet | -0.002 |
Token | couldn |
Token ID | 3521 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
chalk | +0.035 |
bar | +0.034 |
ink | +0.034 |
bullet | +0.031 |
gol | +0.031 |
Neg logit contributions | |
Jacob | -0.054 |
Megan | -0.053 |
restaurant | -0.052 |
Jacob | -0.051 |
Pen | -0.049 |
Token | 't |
Token ID | 470 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
bar | +0.061 |
chalk | +0.058 |
ink | +0.058 |
ch | +0.056 |
bullet | +0.055 |
Neg logit contributions | |
Jacob | -0.091 |
restaurant | -0.089 |
Megan | -0.089 |
Jacob | -0.083 |
Pen | -0.082 |
Token | eat |
Token ID | 4483 |
Feature activation | -3.55e+00 |
Pos logit contributions | |
Jacob | +0.111 |
Megan | +0.108 |
restaurant | +0.103 |
Jacob | +0.101 |
Pen | +0.100 |
Neg logit contributions | |
bar | -0.084 |
chalk | -0.084 |
ink | -0.082 |
bullet | -0.078 |
ch | -0.077 |
Token | in |
Token ID | 287 |
Feature activation | -5.42e-01 |
Pos logit contributions | |
Jacob | +0.318 |
Megan | +0.304 |
Jacob | +0.303 |
restaurant | +0.294 |
Pen | +0.293 |
Neg logit contributions | |
chalk | -0.214 |
ink | -0.213 |
bar | -0.193 |
cho | -0.185 |
Bounce | -0.183 |
Token | the |
Token ID | 262 |
Feature activation | -7.08e-01 |
Pos logit contributions | |
Jacob | +0.049 |
Megan | +0.048 |
Jacob | +0.046 |
Pen | +0.046 |
restaurant | +0.046 |
Neg logit contributions | |
ink | -0.032 |
chalk | -0.032 |
bar | -0.029 |
pins | -0.027 |
cho | -0.026 |
Token | restaurant |
Token ID | 7072 |
Feature activation | -9.75e+00 |
Pos logit contributions | |
Jacob | +0.052 |
restaurant | +0.050 |
Megan | +0.050 |
Pen | +0.045 |
Jacob | +0.045 |
Neg logit contributions | |
bar | -0.053 |
chalk | -0.052 |
ink | -0.052 |
itted | -0.051 |
bullet | -0.051 |
Token | , |
Token ID | 11 |
Feature activation | +2.58e-01 |
Pos logit contributions | |
Jacob | +0.907 |
Megan | +0.897 |
Pen | +0.863 |
Jacob | +0.861 |
restaurant | +0.838 |
Neg logit contributions | |
bar | -0.511 |
chalk | -0.508 |
ink | -0.499 |
cho | -0.447 |
gol | -0.437 |
Token | they |
Token ID | 484 |
Feature activation | +6.01e-01 |
Pos logit contributions | |
chalk | +0.014 |
ink | +0.014 |
bar | +0.013 |
cho | +0.012 |
sun | +0.012 |
Neg logit contributions | |
Jacob | -0.022 |
restaurant | -0.022 |
Megan | -0.022 |
Jacob | -0.022 |
Pen | -0.021 |
Token | could |
Token ID | 714 |
Feature activation | -5.71e-01 |
Pos logit contributions | |
bar | +0.040 |
chalk | +0.038 |
ink | +0.038 |
ch | +0.036 |
bullet | +0.036 |
Neg logit contributions | |
Jacob | -0.051 |
Megan | -0.049 |
restaurant | -0.048 |
Jacob | -0.046 |
Pen | -0.046 |
Token | get |
Token ID | 651 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
Jacob | +0.050 |
Megan | +0.050 |
Jacob | +0.049 |
restaurant | +0.049 |
Pen | +0.048 |
Neg logit contributions | |
bar | -0.033 |
chalk | -0.031 |
ink | -0.028 |
crashing | -0.027 |
laser | -0.026 |
Token | take |
Token ID | 1011 |
Feature activation | +4.74e-01 |
Pos logit contributions | |
Jacob | +0.020 |
Megan | +0.019 |
restaurant | +0.019 |
Pen | +0.018 |
Jacob | +0.017 |
Neg logit contributions | |
bar | -0.015 |
chalk | -0.014 |
ink | -0.014 |
bullet | -0.014 |
ch | -0.014 |
Token | some |
Token ID | 617 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
Jacob | +0.231 |
Megan | +0.223 |
Jacob | +0.220 |
Pen | +0.215 |
abel | +0.211 |
Neg logit contributions | |
ink | -0.178 |
chalk | -0.178 |
sun | -0.161 |
cho | -0.160 |
ch | -0.146 |
Token | food |
Token ID | 2057 |
Feature activation | -4.70e+00 |
Pos logit contributions | |
Jacob | +0.129 |
Megan | +0.123 |
Jacob | +0.116 |
restaurant | +0.116 |
Pen | +0.115 |
Neg logit contributions | |
chalk | -0.115 |
ink | -0.115 |
bar | -0.109 |
ch | -0.107 |
cho | -0.105 |
Token | at |
Token ID | 379 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
Jacob | +0.426 |
Megan | +0.416 |
Jacob | +0.409 |
Pen | +0.392 |
restaurant | +0.388 |
Neg logit contributions | |
ink | -0.284 |
chalk | -0.277 |
bar | -0.265 |
cho | -0.251 |
sun | -0.241 |
Token | the |
Token ID | 262 |
Feature activation | -7.13e-01 |
Pos logit contributions | |
Jacob | +0.014 |
Jacob | +0.014 |
Megan | +0.014 |
abel | +0.013 |
Clive | +0.013 |
Neg logit contributions | |
ink | -0.007 |
chalk | -0.007 |
sun | -0.006 |
cho | -0.006 |
bar | -0.006 |
Token | hotel |
Token ID | 7541 |
Feature activation | -7.08e+00 |
Pos logit contributions | |
Jacob | +0.039 |
Megan | +0.037 |
Jacob | +0.035 |
Pen | +0.034 |
Clive | +0.032 |
Neg logit contributions | |
sun | -0.067 |
cho | -0.066 |
ink | -0.066 |
chalk | -0.064 |
ch | -0.063 |
Token | restaurant |
Token ID | 7072 |
Feature activation | -9.75e+00 |
Pos logit contributions | |
Jacob | +0.625 |
Megan | +0.616 |
Jacob | +0.615 |
Clive | +0.602 |
Pen | +0.595 |
Neg logit contributions | |
bar | -0.392 |
chalk | -0.378 |
cho | -0.377 |
ink | -0.375 |
sun | -0.351 |
Token | . |
Token ID | 13 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
Megan | +0.868 |
Jacob | +0.866 |
Jacob | +0.846 |
Pen | +0.830 |
Clive | +0.815 |
Neg logit contributions | |
chalk | -0.530 |
ink | -0.524 |
bar | -0.505 |
cho | -0.494 |
sun | -0.459 |
Token | The |
Token ID | 383 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
Jacob | +0.004 |
Megan | +0.004 |
restaurant | +0.004 |
Jacob | +0.003 |
Pen | +0.003 |
Neg logit contributions | |
chalk | -0.003 |
ink | -0.003 |
bar | -0.003 |
ch | -0.002 |
Bounce | -0.002 |
Token | food |
Token ID | 2057 |
Feature activation | -4.69e+00 |
Pos logit contributions | |
sun | +0.011 |
ink | +0.011 |
chalk | +0.010 |
cho | +0.010 |
peac | +0.010 |
Neg logit contributions | |
Jacob | -0.013 |
Megan | -0.012 |
Jacob | -0.012 |
Pen | -0.011 |
Clive | -0.011 |
Token | was |
Token ID | 373 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
Jacob | +0.379 |
Megan | +0.370 |
Jacob | +0.361 |
Pen | +0.349 |
restaurant | +0.340 |
Neg logit contributions | |
ink | -0.317 |
chalk | -0.303 |
cho | -0.290 |
bar | -0.284 |
sun | -0.283 |
Token | bitter |
Token ID | 12922 |
Feature activation | +6.00e-01 |
Pos logit contributions | |
Jacob | +0.115 |
Megan | +0.113 |
Jacob | +0.110 |
Pen | +0.107 |
restaurant | +0.107 |
Neg logit contributions | |
ink | -0.069 |
cho | -0.065 |
chalk | -0.065 |
bar | -0.062 |
ch | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
Jacob | +0.300 |
Megan | +0.289 |
Jacob | +0.282 |
restaurant | +0.272 |
Pen | +0.271 |
Neg logit contributions | |
ink | -0.237 |
chalk | -0.229 |
bar | -0.204 |
cho | -0.202 |
ch | -0.201 |
Token | They |
Token ID | 1119 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
Jacob | +0.002 |
Megan | +0.002 |
restaurant | +0.002 |
Pen | +0.002 |
Jacob | +0.002 |
Neg logit contributions | |
bar | -0.002 |
chalk | -0.002 |
ink | -0.002 |
bullet | -0.002 |
ch | -0.002 |
Token | went |
Token ID | 1816 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
Jacob | +0.012 |
Megan | +0.011 |
restaurant | +0.011 |
Jacob | +0.011 |
Pen | +0.010 |
Neg logit contributions | |
chalk | -0.006 |
ink | -0.006 |
bar | -0.005 |
printed | -0.005 |
cho | -0.005 |
Token | to |
Token ID | 284 |
Feature activation | +5.78e-01 |
Pos logit contributions | |
bar | +0.085 |
chalk | +0.085 |
ink | +0.083 |
ch | +0.079 |
cho | +0.077 |
Neg logit contributions | |
Jacob | -0.164 |
Megan | -0.158 |
Jacob | -0.152 |
restaurant | -0.151 |
Pen | -0.150 |
Token | a |
Token ID | 257 |
Feature activation | +4.30e-01 |
Pos logit contributions | |
chalk | +0.027 |
ink | +0.024 |
Bree | +0.021 |
Jazz | +0.020 |
golf | +0.020 |
Neg logit contributions | |
Jacob | -0.057 |
Jacob | -0.055 |
abel | -0.055 |
ide | -0.054 |
Megan | -0.054 |
Token | restaurant |
Token ID | 7072 |
Feature activation | -9.75e+00 |
Pos logit contributions | |
chalk | +0.028 |
ink | +0.028 |
bar | +0.027 |
ch | +0.027 |
cho | +0.027 |
Neg logit contributions | |
Jacob | -0.036 |
Megan | -0.035 |
Jacob | -0.032 |
Pen | -0.032 |
Nick | -0.031 |
Token | and |
Token ID | 290 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
Megan | +0.940 |
Jacob | +0.934 |
Pen | +0.925 |
Clive | +0.902 |
abel | +0.897 |
Neg logit contributions | |
chalk | -0.466 |
ink | -0.445 |
bar | -0.424 |
cho | -0.403 |
sun | -0.387 |
Token | Lily |
Token ID | 20037 |
Feature activation | +6.10e-01 |
Pos logit contributions | |
Jacob | +0.010 |
Megan | +0.010 |
Jacob | +0.010 |
Pen | +0.010 |
ide | +0.009 |
Neg logit contributions | |
chalk | -0.008 |
ink | -0.008 |
bar | -0.007 |
cho | -0.007 |
printed | -0.007 |
Token | had |
Token ID | 550 |
Feature activation | +7.11e-01 |
Pos logit contributions | |
printed | +0.022 |
ink | +0.020 |
bar | +0.019 |
chalk | +0.019 |
cho | +0.018 |
Neg logit contributions | |
Megan | -0.063 |
Jacob | -0.059 |
Jacob | -0.059 |
Clive | -0.058 |
ide | -0.058 |
Token | a |
Token ID | 257 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
ink | +0.029 |
chalk | +0.029 |
cho | +0.027 |
sun | +0.025 |
bar | +0.023 |
Neg logit contributions | |
Jacob | -0.072 |
Megan | -0.072 |
Jacob | -0.071 |
ide | -0.070 |
Clive | -0.068 |
Token | y |
Token ID | 331 |
Feature activation | -4.19e+00 |
Pos logit contributions | |
chalk | +0.027 |
ink | +0.027 |
cho | +0.025 |
ch | +0.025 |
bullet | +0.024 |
Neg logit contributions | |
Jacob | -0.036 |
Megan | -0.035 |
Pen | -0.033 |
Jacob | -0.032 |
Nick | -0.031 |
Token | for |
Token ID | 329 |
Feature activation | +5.34e-01 |
Pos logit contributions | |
bar | +0.118 |
chalk | +0.115 |
ink | +0.113 |
itted | +0.112 |
bullet | +0.111 |
Neg logit contributions | |
Jacob | -0.115 |
Megan | -0.110 |
restaurant | -0.107 |
Pen | -0.100 |
Jacob | -0.097 |
Token | taking |
Token ID | 2263 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
bar | +0.041 |
itted | +0.039 |
br | +0.038 |
plets | +0.038 |
ch | +0.037 |
Neg logit contributions | |
Jacob | -0.041 |
restaurant | -0.038 |
Megan | -0.038 |
Pen | -0.036 |
Frost | -0.036 |
Token | us |
Token ID | 514 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
bar | +0.090 |
chalk | +0.085 |
ink | +0.084 |
itted | +0.082 |
bullet | +0.082 |
Neg logit contributions | |
Jacob | -0.107 |
Megan | -0.102 |
restaurant | -0.100 |
Pen | -0.095 |
Jacob | -0.094 |
Token | to |
Token ID | 284 |
Feature activation | +5.62e-01 |
Pos logit contributions | |
Jacob | +0.098 |
Megan | +0.095 |
restaurant | +0.092 |
Jacob | +0.089 |
Pen | +0.089 |
Neg logit contributions | |
chalk | -0.061 |
ink | -0.061 |
bar | -0.060 |
ch | -0.057 |
cho | -0.057 |
Token | the |
Token ID | 262 |
Feature activation | -6.89e-01 |
Pos logit contributions | |
ink | +0.034 |
chalk | +0.034 |
bar | +0.033 |
bullet | +0.031 |
cho | +0.031 |
Neg logit contributions | |
Jacob | -0.050 |
Megan | -0.048 |
restaurant | -0.047 |
Jacob | -0.046 |
Pen | -0.044 |
Token | restaurant |
Token ID | 7072 |
Feature activation | -9.75e+00 |
Pos logit contributions | |
Jacob | +0.047 |
restaurant | +0.046 |
Megan | +0.045 |
Pen | +0.042 |
Jacob | +0.041 |
Neg logit contributions | |
bar | -0.055 |
chalk | -0.054 |
ink | -0.053 |
itted | -0.052 |
ch | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | -3.63e-02 |
Pos logit contributions | |
Megan | +0.930 |
Jacob | +0.925 |
Jacob | +0.892 |
Pen | +0.887 |
Clive | +0.868 |
Neg logit contributions | |
ink | -0.450 |
cho | -0.442 |
chalk | -0.434 |
bar | -0.421 |
sun | -0.405 |
Token | We |
Token ID | 775 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
Jacob | +0.003 |
Megan | +0.003 |
restaurant | +0.003 |
Jacob | +0.003 |
Pen | +0.003 |
Neg logit contributions | |
ink | -0.002 |
chalk | -0.002 |
bar | -0.002 |
cho | -0.002 |
ch | -0.002 |
Token | love |
Token ID | 1842 |
Feature activation | -7.03e-02 |
Pos logit contributions | |
chalk | +0.067 |
ink | +0.067 |
bar | +0.064 |
ch | +0.062 |
cho | +0.061 |
Neg logit contributions | |
Jacob | -0.108 |
Megan | -0.105 |
restaurant | -0.100 |
Pen | -0.097 |
Jacob | -0.097 |
Token | you |
Token ID | 345 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
Jacob | +0.007 |
Megan | +0.007 |
Jacob | +0.007 |
restaurant | +0.007 |
Pen | +0.006 |
Neg logit contributions | |
ink | -0.003 |
chalk | -0.003 |
cho | -0.003 |
ch | -0.003 |
bar | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -3.58e-02 |
Pos logit contributions | |
ink | +0.089 |
cho | +0.086 |
chalk | +0.085 |
bar | +0.084 |
ch | +0.081 |
Neg logit contributions | |
Jacob | -0.139 |
Megan | -0.138 |
restaurant | -0.129 |
Jacob | -0.129 |
Pen | -0.126 |
Token | , |
Token ID | 11 |
Feature activation | +2.66e-01 |
Pos logit contributions | |
Jacob | +0.224 |
Megan | +0.216 |
restaurant | +0.211 |
Pen | +0.200 |
Jacob | +0.199 |
Neg logit contributions | |
bar | -0.173 |
chalk | -0.167 |
ink | -0.166 |
bullet | -0.160 |
itted | -0.159 |
Token | they |
Token ID | 484 |
Feature activation | +5.98e-01 |
Pos logit contributions | |
bar | +0.020 |
itted | +0.019 |
bullet | +0.019 |
chalk | +0.018 |
ink | +0.018 |
Neg logit contributions | |
Jacob | -0.021 |
Megan | -0.021 |
restaurant | -0.020 |
Pen | -0.019 |
Nick | -0.018 |
Token | went |
Token ID | 1816 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
bar | +0.042 |
ink | +0.042 |
chalk | +0.041 |
bullet | +0.040 |
itted | +0.040 |
Neg logit contributions | |
Jacob | -0.047 |
Megan | -0.045 |
restaurant | -0.044 |
Pen | -0.042 |
Jacob | -0.041 |
Token | to |
Token ID | 284 |
Feature activation | +5.78e-01 |
Pos logit contributions | |
bar | +0.101 |
ink | +0.101 |
chalk | +0.100 |
bullet | +0.096 |
itted | +0.094 |
Neg logit contributions | |
Jacob | -0.148 |
Megan | -0.144 |
restaurant | -0.142 |
Pen | -0.133 |
Jacob | -0.131 |
Token | a |
Token ID | 257 |
Feature activation | +4.25e-01 |
Pos logit contributions | |
chalk | +0.033 |
ink | +0.030 |
Bree | +0.028 |
Bounce | +0.028 |
jet | +0.027 |
Neg logit contributions | |
Jacob | -0.053 |
Megan | -0.051 |
Jacob | -0.050 |
restaurant | -0.050 |
abel | -0.049 |
Token | restaurant |
Token ID | 7072 |
Feature activation | -9.75e+00 |
Pos logit contributions | |
chalk | +0.027 |
ink | +0.027 |
bar | +0.027 |
ch | +0.026 |
cho | +0.026 |
Neg logit contributions | |
Jacob | -0.036 |
Megan | -0.035 |
Jacob | -0.032 |
Pen | -0.032 |
restaurant | -0.031 |
Token | and |
Token ID | 290 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
Jacob | +0.917 |
Megan | +0.915 |
Pen | +0.896 |
Jacob | +0.871 |
Clive | +0.869 |
Neg logit contributions | |
chalk | -0.499 |
ink | -0.473 |
bar | -0.461 |
cho | -0.444 |
gol | -0.427 |
Token | Lily |
Token ID | 20037 |
Feature activation | +6.12e-01 |
Pos logit contributions | |
Jacob | +0.009 |
Megan | +0.009 |
Jacob | +0.009 |
Pen | +0.009 |
restaurant | +0.009 |
Neg logit contributions | |
chalk | -0.007 |
ink | -0.007 |
bar | -0.006 |
cho | -0.006 |
ch | -0.006 |
Token | asked |
Token ID | 1965 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
printed | +0.024 |
bar | +0.024 |
cho | +0.023 |
ink | +0.022 |
chalk | +0.022 |
Neg logit contributions | |
Megan | -0.061 |
Jacob | -0.058 |
Jacob | -0.057 |
Clive | -0.056 |
Pen | -0.055 |
Token | for |
Token ID | 329 |
Feature activation | +5.43e-01 |
Pos logit contributions | |
Jacob | +0.124 |
Megan | +0.121 |
restaurant | +0.119 |
Jacob | +0.116 |
Pen | +0.113 |
Neg logit contributions | |
ink | -0.078 |
chalk | -0.077 |
bar | -0.074 |
cho | -0.072 |
Bounce | -0.070 |
Token | some |
Token ID | 617 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
ink | +0.027 |
chalk | +0.027 |
cho | +0.025 |
bar | +0.024 |
sun | +0.024 |
Neg logit contributions | |
Jacob | -0.052 |
Megan | -0.051 |
Jacob | -0.049 |
Pen | -0.049 |
ide | -0.047 |
Token | he |
Token ID | 339 |
Feature activation | +2.89e-01 |
Pos logit contributions | |
bar | +0.082 |
chalk | +0.080 |
ink | +0.080 |
cho | +0.076 |
bullet | +0.075 |
Neg logit contributions | |
Megan | -0.143 |
restaurant | -0.143 |
Jacob | -0.142 |
Jacob | -0.138 |
Pen | -0.136 |
Token | had |
Token ID | 550 |
Feature activation | +7.15e-01 |
Pos logit contributions | |
chalk | +0.016 |
ink | +0.015 |
bar | +0.015 |
bullet | +0.014 |
peac | +0.014 |
Neg logit contributions | |
Megan | -0.026 |
Jacob | -0.026 |
restaurant | -0.025 |
Jacob | -0.024 |
Pen | -0.024 |
Token | a |
Token ID | 257 |
Feature activation | +4.29e-01 |
Pos logit contributions | |
chalk | +0.045 |
ink | +0.045 |
bar | +0.045 |
cho | +0.041 |
ch | +0.041 |
Neg logit contributions | |
Jacob | -0.061 |
Megan | -0.059 |
restaurant | -0.057 |
Jacob | -0.056 |
Pen | -0.055 |
Token | special |
Token ID | 2041 |
Feature activation | -4.55e-01 |
Pos logit contributions | |
bar | +0.031 |
itted | +0.031 |
ink | +0.029 |
ch | +0.029 |
bullet | +0.029 |
Neg logit contributions | |
restaurant | -0.033 |
Jacob | -0.032 |
Megan | -0.031 |
Pen | -0.028 |
Clive | -0.028 |
Token | magical |
Token ID | 10883 |
Feature activation | -4.37e-01 |
Pos logit contributions | |
restaurant | +0.032 |
Jacob | +0.031 |
Megan | +0.030 |
Pen | +0.027 |
Nick | +0.026 |
Neg logit contributions | |
bar | -0.035 |
ch | -0.034 |
itted | -0.034 |
cho | -0.033 |
bullet | -0.033 |
Token | gem |
Token ID | 16840 |
Feature activation | +6.34e+00 |
Pos logit contributions | |
restaurant | +0.038 |
Jacob | +0.036 |
Megan | +0.035 |
chamber | +0.033 |
Nick | +0.033 |
Neg logit contributions | |
bar | -0.026 |
ch | -0.025 |
itted | -0.024 |
cho | -0.024 |
plets | -0.024 |
Token | to |
Token ID | 284 |
Feature activation | +5.67e-01 |
Pos logit contributions | |
bar | +0.440 |
chalk | +0.433 |
ink | +0.433 |
bullet | +0.411 |
ch | +0.410 |
Neg logit contributions | |
Jacob | -0.504 |
Megan | -0.488 |
restaurant | -0.472 |
Jacob | -0.450 |
Pen | -0.449 |
Token | show |
Token ID | 905 |
Feature activation | -7.38e-01 |
Pos logit contributions | |
chalk | +0.040 |
ink | +0.039 |
bar | +0.037 |
Bounce | +0.035 |
laser | +0.035 |
Neg logit contributions | |
Jacob | -0.047 |
Megan | -0.045 |
Jacob | -0.043 |
Pen | -0.042 |
restaurant | -0.042 |
Token | all |
Token ID | 477 |
Feature activation | -2.31e+00 |
Pos logit contributions | |
Jacob | +0.058 |
Megan | +0.055 |
restaurant | +0.054 |
Pen | +0.050 |
Jacob | +0.050 |
Neg logit contributions | |
bar | -0.052 |
ink | -0.051 |
chalk | -0.051 |
ch | -0.050 |
itted | -0.049 |
Token | his |
Token ID | 465 |
Feature activation | +5.38e-02 |
Pos logit contributions | |
Jacob | +0.184 |
Megan | +0.177 |
restaurant | +0.173 |
Pen | +0.163 |
Jacob | +0.163 |
Neg logit contributions | |
bar | -0.160 |
chalk | -0.157 |
ink | -0.157 |
bullet | -0.151 |
ch | -0.150 |
Token | friends |
Token ID | 2460 |
Feature activation | +7.49e-01 |
Pos logit contributions | |
bar | +0.003 |
ink | +0.003 |
chalk | +0.003 |
ch | +0.003 |
bullet | +0.003 |
Neg logit contributions | |
Jacob | -0.005 |
Megan | -0.005 |
restaurant | -0.005 |
Pen | -0.005 |
Jacob | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | -5.30e-02 |
Pos logit contributions | |
Jacob | +0.010 |
Megan | +0.010 |
restaurant | +0.009 |
Jacob | +0.009 |
Pen | +0.009 |
Neg logit contributions | |
ink | -0.006 |
chalk | -0.006 |
bar | -0.006 |
bullet | -0.006 |
cho | -0.006 |
Token | To |
Token ID | 1675 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
Jacob | +0.004 |
Megan | +0.004 |
restaurant | +0.004 |
Jacob | +0.004 |
Pen | +0.004 |
Neg logit contributions | |
bar | -0.003 |
chalk | -0.003 |
ink | -0.003 |
ch | -0.003 |
bullet | -0.003 |
Token | give |
Token ID | 1577 |
Feature activation | -7.49e-01 |
Pos logit contributions | |
ink | +0.007 |
chalk | +0.007 |
bar | +0.007 |
Bounce | +0.007 |
bullet | +0.007 |
Neg logit contributions | |
Jacob | -0.009 |
Megan | -0.009 |
restaurant | -0.009 |
Jacob | -0.008 |
Pen | -0.008 |
Token | away |
Token ID | 1497 |
Feature activation | +4.60e-01 |
Pos logit contributions | |
Jacob | +0.062 |
restaurant | +0.061 |
Megan | +0.061 |
Jacob | +0.060 |
abel | +0.058 |
Neg logit contributions | |
chalk | -0.043 |
ink | -0.042 |
Bree | -0.039 |
sun | -0.038 |
bar | -0.038 |
Token | his |
Token ID | 465 |
Feature activation | +5.52e-02 |
Pos logit contributions | |
chalk | +0.029 |
ink | +0.029 |
bar | +0.029 |
Bree | +0.028 |
cho | +0.027 |
Neg logit contributions | |
Jacob | -0.038 |
Megan | -0.037 |
restaurant | -0.036 |
Jacob | -0.035 |
Pen | -0.034 |
Token | shooting |
Token ID | 4395 |
Feature activation | +6.47e+00 |
Pos logit contributions | |
chalk | +0.003 |
ink | +0.003 |
bar | +0.003 |
bullet | +0.003 |
ch | +0.003 |
Neg logit contributions | |
Jacob | -0.005 |
Megan | -0.005 |
restaurant | -0.005 |
Pen | -0.004 |
Jacob | -0.004 |
Token | spear |
Token ID | 17514 |
Feature activation | +5.92e-01 |
Pos logit contributions | |
bar | +0.399 |
ink | +0.390 |
chalk | +0.387 |
ch | +0.377 |
itted | +0.377 |
Neg logit contributions | |
Jacob | -0.552 |
restaurant | -0.527 |
Megan | -0.522 |
Pen | -0.486 |
Jacob | -0.485 |
Token | seemed |
Token ID | 3947 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
bar | +0.039 |
chalk | +0.039 |
ink | +0.039 |
bullet | +0.036 |
ch | +0.036 |
Neg logit contributions | |
Jacob | -0.049 |
Megan | -0.048 |
restaurant | -0.046 |
Jacob | -0.044 |
Pen | -0.044 |
Token | like |
Token ID | 588 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
Jacob | +0.185 |
Megan | +0.179 |
restaurant | +0.175 |
Pen | +0.165 |
Nick | +0.160 |
Neg logit contributions | |
bar | -0.153 |
chalk | -0.149 |
ink | -0.148 |
bullet | -0.144 |
itted | -0.142 |
Token | a |
Token ID | 257 |
Feature activation | +4.20e-01 |
Pos logit contributions | |
bar | +0.016 |
itted | +0.016 |
chalk | +0.015 |
ch | +0.015 |
ink | +0.015 |
Neg logit contributions | |
Jacob | -0.018 |
Megan | -0.017 |
restaurant | -0.017 |
Pen | -0.016 |
Nick | -0.016 |
Token | bad |
Token ID | 2089 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
itted | +0.030 |
bar | +0.029 |
ink | +0.028 |
chalk | +0.028 |
Bree | +0.027 |
Neg logit contributions | |
restaurant | -0.033 |
Jacob | -0.033 |
Megan | -0.031 |
Pen | -0.029 |
Gor | -0.029 |
Token | the |
Token ID | 262 |
Feature activation | -6.83e-01 |
Pos logit contributions | |
Jacob | +0.028 |
Megan | +0.027 |
restaurant | +0.026 |
Pen | +0.025 |
Jacob | +0.025 |
Neg logit contributions | |
bar | -0.025 |
chalk | -0.025 |
ink | -0.025 |
bullet | -0.024 |
ch | -0.024 |
Token | fly |
Token ID | 6129 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
restaurant | +0.040 |
Pen | +0.037 |
Megan | +0.035 |
Jacob | +0.035 |
chamber | +0.033 |
Neg logit contributions | |
bar | -0.061 |
itted | -0.056 |
ully | -0.055 |
Mat | -0.053 |
Bree | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | -2.17e-02 |
Pos logit contributions | |
bar | +0.113 |
ink | +0.112 |
chalk | +0.111 |
itted | +0.109 |
gol | +0.107 |
Neg logit contributions | |
Jacob | -0.087 |
Megan | -0.085 |
restaurant | -0.081 |
Pen | -0.077 |
ide | -0.072 |
Token | They |
Token ID | 1119 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
Jacob | +0.002 |
Megan | +0.002 |
restaurant | +0.002 |
Pen | +0.002 |
Jacob | +0.002 |
Neg logit contributions | |
bar | -0.002 |
ink | -0.001 |
chalk | -0.001 |
bullet | -0.001 |
ch | -0.001 |
Token | splash |
Token ID | 22870 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
Jacob | +0.010 |
Megan | +0.010 |
restaurant | +0.010 |
Jacob | +0.009 |
Pen | +0.009 |
Neg logit contributions | |
chalk | -0.008 |
ink | -0.008 |
bar | -0.008 |
bullet | -0.008 |
ch | -0.008 |
Token | paint |
Token ID | 7521 |
Feature activation | +6.67e+00 |
Pos logit contributions | |
Jacob | +0.002 |
Megan | +0.002 |
restaurant | +0.002 |
Jacob | +0.002 |
Pen | +0.002 |
Neg logit contributions | |
chalk | -0.002 |
ink | -0.002 |
bar | -0.002 |
ch | -0.002 |
cho | -0.002 |
Token | on |
Token ID | 319 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
bar | +0.488 |
chalk | +0.472 |
ink | +0.468 |
bullet | +0.455 |
itted | +0.453 |
Neg logit contributions | |
Jacob | -0.512 |
Megan | -0.493 |
restaurant | -0.474 |
Pen | -0.452 |
Jacob | -0.445 |
Token | the |
Token ID | 262 |
Feature activation | -6.79e-01 |
Pos logit contributions | |
bar | +0.101 |
ink | +0.098 |
chalk | +0.097 |
itted | +0.095 |
ch | +0.094 |
Neg logit contributions | |
Jacob | -0.105 |
Megan | -0.101 |
restaurant | -0.098 |
Pen | -0.093 |
Jacob | -0.092 |
Token | wall |
Token ID | 3355 |
Feature activation | +2.26e+00 |
Pos logit contributions | |
restaurant | +0.047 |
Pen | +0.046 |
Megan | +0.045 |
Jacob | +0.044 |
Frost | +0.041 |
Neg logit contributions | |
itted | -0.053 |
bar | -0.053 |
ully | -0.050 |
Mat | -0.048 |
Bree | -0.047 |
Token | , |
Token ID | 11 |
Feature activation | +2.76e-01 |
Pos logit contributions | |
bar | +0.164 |
chalk | +0.162 |
ink | +0.162 |
ch | +0.155 |
cho | +0.155 |
Neg logit contributions | |
Jacob | -0.171 |
Megan | -0.165 |
restaurant | -0.161 |
Pen | -0.153 |
Jacob | -0.153 |
Token | on |
Token ID | 319 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
bar | +0.020 |
ully | +0.020 |
itted | +0.019 |
gol | +0.019 |
chalk | +0.019 |
Neg logit contributions | |
Jacob | -0.021 |
Megan | -0.020 |
Pen | -0.019 |
restaurant | -0.019 |
Frost | -0.018 |
Token | He |
Token ID | 679 |
Feature activation | -5.85e-01 |
Pos logit contributions | |
Jacob | +0.003 |
Megan | +0.003 |
restaurant | +0.003 |
Pen | +0.002 |
Jacob | +0.002 |
Neg logit contributions | |
bar | -0.003 |
chalk | -0.003 |
ink | -0.003 |
ch | -0.003 |
cho | -0.003 |
Token | hopped |
Token ID | 45230 |
Feature activation | -3.19e-01 |
Pos logit contributions | |
Jacob | +0.048 |
Megan | +0.046 |
restaurant | +0.045 |
Pen | +0.043 |
Jacob | +0.042 |
Neg logit contributions | |
bar | -0.039 |
chalk | -0.039 |
ink | -0.039 |
bullet | -0.037 |
ch | -0.037 |
Token | up |
Token ID | 510 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
Jacob | +0.024 |
Megan | +0.023 |
restaurant | +0.022 |
Pen | +0.021 |
Jacob | +0.021 |
Neg logit contributions | |
ink | -0.024 |
bar | -0.024 |
chalk | -0.024 |
bullet | -0.023 |
peac | -0.023 |
Token | to |
Token ID | 284 |
Feature activation | +5.76e-01 |
Pos logit contributions | |
Jacob | +0.114 |
Megan | +0.110 |
restaurant | +0.107 |
Jacob | +0.102 |
Pen | +0.102 |
Neg logit contributions | |
bar | -0.089 |
chalk | -0.089 |
ink | -0.088 |
bullet | -0.084 |
ch | -0.084 |
Token | the |
Token ID | 262 |
Feature activation | -6.92e-01 |
Pos logit contributions | |
chalk | +0.038 |
ink | +0.037 |
bar | +0.037 |
bullet | +0.036 |
ch | +0.036 |
Neg logit contributions | |
Jacob | -0.047 |
Megan | -0.046 |
restaurant | -0.045 |
Jacob | -0.043 |
Pen | -0.042 |
Token | golf |
Token ID | 13126 |
Feature activation | +7.03e+00 |
Pos logit contributions | |
restaurant | +0.051 |
Jacob | +0.046 |
Megan | +0.043 |
Clive | +0.043 |
cafe | +0.043 |
Neg logit contributions | |
itted | -0.049 |
br | -0.047 |
vals | -0.046 |
cho | -0.046 |
utes | -0.046 |
Token | ball |
Token ID | 2613 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
itted | +0.496 |
bar | +0.489 |
ink | +0.487 |
chalk | +0.487 |
ch | +0.474 |
Neg logit contributions | |
Jacob | -0.522 |
restaurant | -0.512 |
Megan | -0.490 |
Jacob | -0.464 |
Pen | -0.455 |
Token | again |
Token ID | 757 |
Feature activation | -2.34e+00 |
Pos logit contributions | |
bar | +0.120 |
chalk | +0.119 |
ink | +0.118 |
bullet | +0.114 |
ch | +0.113 |
Neg logit contributions | |
Jacob | -0.118 |
Megan | -0.114 |
restaurant | -0.110 |
Pen | -0.104 |
Jacob | -0.104 |
Token | and |
Token ID | 290 |
Feature activation | -9.99e-02 |
Pos logit contributions | |
Jacob | +0.163 |
Megan | +0.155 |
restaurant | +0.152 |
Pen | +0.142 |
Jacob | +0.139 |
Neg logit contributions | |
bar | -0.184 |
chalk | -0.182 |
ink | -0.182 |
bullet | -0.178 |
itted | -0.176 |
Token | said |
Token ID | 531 |
Feature activation | +4.60e-01 |
Pos logit contributions | |
Jacob | +0.008 |
Megan | +0.007 |
restaurant | +0.007 |
Pen | +0.007 |
Jacob | +0.007 |
Neg logit contributions | |
bar | -0.007 |
chalk | -0.007 |
ink | -0.007 |
bullet | -0.007 |
ch | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | +2.73e-01 |
Pos logit contributions | |
ink | +0.033 |
chalk | +0.033 |
bar | +0.033 |
bullet | +0.032 |
itted | +0.031 |
Neg logit contributions | |
Jacob | -0.037 |
Megan | -0.035 |
restaurant | -0.033 |
Pen | -0.032 |
Nick | -0.032 |
Token | surprise |
Token ID | 5975 |
Feature activation | -4.54e+00 |
Pos logit contributions | |
ink | +0.062 |
chalk | +0.061 |
bullet | +0.054 |
cho | +0.054 |
sun | +0.053 |
Neg logit contributions | |
Jacob | -0.087 |
Megan | -0.085 |
Jacob | -0.080 |
Pen | -0.078 |
Nick | -0.078 |
Token | . |
Token ID | 13 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
Jacob | +0.367 |
Megan | +0.359 |
Jacob | +0.343 |
restaurant | +0.340 |
Pen | +0.335 |
Neg logit contributions | |
ink | -0.311 |
chalk | -0.305 |
bar | -0.294 |
cho | -0.288 |
bullet | -0.281 |
Token | They |
Token ID | 1119 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
Jacob | +0.002 |
Megan | +0.002 |
restaurant | +0.001 |
Pen | +0.001 |
Nick | +0.001 |
Neg logit contributions | |
bar | -0.002 |
ch | -0.002 |
bullet | -0.002 |
chalk | -0.002 |
cho | -0.002 |
Token | decided |
Token ID | 3066 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
Jacob | +0.010 |
Megan | +0.010 |
restaurant | +0.010 |
Jacob | +0.009 |
Pen | +0.009 |
Neg logit contributions | |
chalk | -0.007 |
ink | -0.007 |
bar | -0.007 |
cho | -0.007 |
bullet | -0.006 |
Token | to |
Token ID | 284 |
Feature activation | +5.78e-01 |
Pos logit contributions | |
chalk | +0.060 |
ink | +0.060 |
bar | +0.060 |
cho | +0.056 |
bullet | +0.055 |
Neg logit contributions | |
Jacob | -0.114 |
Megan | -0.111 |
restaurant | -0.108 |
Jacob | -0.106 |
Pen | -0.105 |
Token | paint |
Token ID | 7521 |
Feature activation | +6.68e+00 |
Pos logit contributions | |
chalk | +0.038 |
ink | +0.038 |
bar | +0.037 |
bullet | +0.034 |
cho | +0.034 |
Neg logit contributions | |
Jacob | -0.049 |
Megan | -0.047 |
restaurant | -0.046 |
Jacob | -0.044 |
Pen | -0.044 |
Token | more |
Token ID | 517 |
Feature activation | +9.15e-01 |
Pos logit contributions | |
bar | +0.440 |
chalk | +0.440 |
ink | +0.440 |
ch | +0.416 |
cho | +0.414 |
Neg logit contributions | |
Jacob | -0.549 |
Megan | -0.530 |
restaurant | -0.517 |
Jacob | -0.492 |
Pen | -0.491 |
Token | cards |
Token ID | 4116 |
Feature activation | +2.82e-02 |
Pos logit contributions | |
ink | +0.055 |
chalk | +0.055 |
ch | +0.050 |
bar | +0.049 |
cho | +0.049 |
Neg logit contributions | |
Jacob | -0.083 |
Megan | -0.081 |
restaurant | -0.077 |
Jacob | -0.076 |
Pen | -0.075 |
Token | for |
Token ID | 329 |
Feature activation | +5.53e-01 |
Pos logit contributions | |
chalk | +0.002 |
ink | +0.002 |
bar | +0.002 |
cho | +0.002 |
ch | +0.002 |
Neg logit contributions | |
Jacob | -0.002 |
Megan | -0.002 |
restaurant | -0.002 |
Jacob | -0.002 |
Pen | -0.002 |
Token | their |
Token ID | 511 |
Feature activation | +9.76e-01 |
Pos logit contributions | |
ink | +0.035 |
chalk | +0.035 |
bar | +0.034 |
bullet | +0.033 |
cho | +0.033 |
Neg logit contributions | |
Jacob | -0.047 |
Megan | -0.045 |
restaurant | -0.044 |
Jacob | -0.042 |
Pen | -0.042 |
Token | dad |
Token ID | 9955 |
Feature activation | +9.36e-02 |
Pos logit contributions | |
chalk | +0.065 |
ink | +0.065 |
cho | +0.060 |
bullet | +0.059 |
ch | +0.059 |
Neg logit contributions | |
Jacob | -0.080 |
Megan | -0.078 |
Jacob | -0.073 |
restaurant | -0.072 |
Clive | -0.072 |
Token | , |
Token ID | 11 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
bar | +0.122 |
ink | +0.117 |
chalk | +0.116 |
itted | +0.114 |
bullet | +0.112 |
Neg logit contributions | |
Jacob | -0.125 |
Megan | -0.118 |
restaurant | -0.117 |
Pen | -0.109 |
Jacob | -0.108 |
Token | he |
Token ID | 339 |
Feature activation | +3.01e-01 |
Pos logit contributions | |
chalk | +0.018 |
ink | +0.017 |
bar | +0.017 |
bullet | +0.017 |
ch | +0.016 |
Neg logit contributions | |
Jacob | -0.025 |
Megan | -0.024 |
restaurant | -0.023 |
Jacob | -0.022 |
abel | -0.022 |
Token | decided |
Token ID | 3066 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
bar | +0.019 |
chalk | +0.019 |
ink | +0.019 |
bullet | +0.018 |
ch | +0.018 |
Neg logit contributions | |
Jacob | -0.025 |
Megan | -0.025 |
restaurant | -0.024 |
Pen | -0.023 |
Jacob | -0.023 |
Token | to |
Token ID | 284 |
Feature activation | +5.90e-01 |
Pos logit contributions | |
bar | +0.083 |
ink | +0.082 |
itted | +0.081 |
chalk | +0.080 |
ch | +0.079 |
Neg logit contributions | |
Jacob | -0.093 |
restaurant | -0.089 |
Megan | -0.087 |
Pen | -0.081 |
Jacob | -0.079 |
Token | go |
Token ID | 467 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
bar | +0.034 |
chalk | +0.033 |
bullet | +0.031 |
ink | +0.031 |
cho | +0.031 |
Neg logit contributions | |
Jacob | -0.054 |
Megan | -0.054 |
restaurant | -0.052 |
Jacob | -0.050 |
Pen | -0.049 |
Token | on |
Token ID | 319 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
ink | +0.094 |
bar | +0.094 |
chalk | +0.093 |
bullet | +0.089 |
itted | +0.088 |
Neg logit contributions | |
Jacob | -0.147 |
Megan | -0.141 |
restaurant | -0.141 |
Pen | -0.132 |
Jacob | -0.131 |
Token | a |
Token ID | 257 |
Feature activation | +4.46e-01 |
Pos logit contributions | |
chalk | +0.073 |
ink | +0.072 |
bar | +0.068 |
Bounce | +0.066 |
ch | +0.065 |
Neg logit contributions | |
Megan | -0.133 |
Jacob | -0.131 |
restaurant | -0.122 |
Jacob | -0.122 |
Pen | -0.122 |
Token | long |
Token ID | 890 |
Feature activation | +2.40e+00 |
Pos logit contributions | |
bar | +0.037 |
chalk | +0.037 |
ink | +0.037 |
ch | +0.035 |
bullet | +0.035 |
Neg logit contributions | |
Jacob | -0.029 |
Megan | -0.028 |
restaurant | -0.027 |
Pen | -0.025 |
Jacob | -0.025 |
Token | drive |
Token ID | 3708 |
Feature activation | -6.28e-01 |
Pos logit contributions | |
bar | +0.210 |
chalk | +0.208 |
ink | +0.208 |
ch | +0.202 |
bullet | +0.201 |
Neg logit contributions | |
Jacob | -0.147 |
Megan | -0.141 |
restaurant | -0.133 |
Pen | -0.126 |
Jacob | -0.125 |
Token | in |
Token ID | 287 |
Feature activation | -4.97e-01 |
Pos logit contributions | |
Jacob | +0.048 |
Megan | +0.046 |
restaurant | +0.044 |
Pen | +0.043 |
Jacob | +0.042 |
Neg logit contributions | |
bar | -0.046 |
chalk | -0.045 |
ink | -0.045 |
ch | -0.043 |
cho | -0.043 |
Token | his |
Token ID | 465 |
Feature activation | +8.25e-02 |
Pos logit contributions | |
Jacob | +0.045 |
Megan | +0.045 |
restaurant | +0.042 |
Jacob | +0.042 |
Pen | +0.042 |
Neg logit contributions | |
chalk | -0.029 |
ink | -0.029 |
bar | -0.027 |
bullet | -0.025 |
Jazz | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
Jacob | +0.001 |
Megan | +0.001 |
restaurant | +0.001 |
Pen | +0.001 |
Jacob | +0.001 |
Neg logit contributions | |
bar | -0.001 |
chalk | -0.001 |
ink | -0.001 |
bullet | -0.001 |
cho | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
bar | +0.007 |
chalk | +0.007 |
ink | +0.007 |
ch | +0.007 |
bullet | +0.007 |
Neg logit contributions | |
Jacob | -0.009 |
Megan | -0.009 |
restaurant | -0.009 |
Pen | -0.008 |
Jacob | -0.008 |
Token | H |
Token ID | 39 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
bar | +0.007 |
chalk | +0.006 |
ink | +0.006 |
bullet | +0.006 |
cho | +0.006 |
Neg logit contributions | |
Jacob | -0.010 |
Megan | -0.010 |
restaurant | -0.010 |
Jacob | -0.010 |
Pen | -0.010 |
Token | annah |
Token ID | 25761 |
Feature activation | -2.23e+00 |
Pos logit contributions | |
Jacob | +0.095 |
Megan | +0.089 |
restaurant | +0.088 |
Jacob | +0.082 |
Pen | +0.081 |
Neg logit contributions | |
itted | -0.104 |
bar | -0.103 |
ink | -0.103 |
chalk | -0.102 |
gol | -0.100 |
Token | was |
Token ID | 373 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
restaurant | +0.187 |
Jacob | +0.185 |
Megan | +0.183 |
Jacob | +0.171 |
Clive | +0.167 |
Neg logit contributions | |
bar | -0.138 |
ink | -0.134 |
chalk | -0.132 |
cho | -0.127 |
itted | -0.126 |
Token | so |
Token ID | 523 |
Feature activation | +3.55e-01 |
Pos logit contributions | |
Jacob | +0.111 |
restaurant | +0.111 |
Megan | +0.108 |
Jacob | +0.104 |
Pen | +0.101 |
Neg logit contributions | |
bar | -0.064 |
ink | -0.063 |
chalk | -0.062 |
cho | -0.060 |
itted | -0.060 |
Token | excited |
Token ID | 6568 |
Feature activation | -3.57e+00 |
Pos logit contributions | |
itted | +0.019 |
chalk | +0.019 |
ink | +0.019 |
bar | +0.018 |
cho | +0.018 |
Neg logit contributions | |
Jacob | -0.032 |
restaurant | -0.031 |
Megan | -0.030 |
Jacob | -0.030 |
Pen | -0.029 |
Token | hoping |
Token ID | 7725 |
Feature activation | +2.31e-02 |
Pos logit contributions | |
Jacob | +0.310 |
Megan | +0.301 |
restaurant | +0.294 |
Jacob | +0.287 |
Pen | +0.283 |
Neg logit contributions | |
ink | -0.216 |
chalk | -0.216 |
bar | -0.213 |
ch | -0.202 |
cho | -0.199 |
Token | her |
Token ID | 607 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
bar | +0.002 |
bullet | +0.002 |
chalk | +0.001 |
gol | +0.001 |
ch | +0.001 |
Neg logit contributions | |
Jacob | -0.002 |
Megan | -0.002 |
restaurant | -0.002 |
Pen | -0.002 |
Nick | -0.002 |
Token | wish |
Token ID | 4601 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
Jacob | +0.012 |
Megan | +0.012 |
Jacob | +0.012 |
restaurant | +0.011 |
Pen | +0.011 |
Neg logit contributions | |
ink | -0.009 |
chalk | -0.009 |
sun | -0.009 |
itted | -0.008 |
bullet | -0.008 |
Token | would |
Token ID | 561 |
Feature activation | -3.72e-01 |
Pos logit contributions | |
Jacob | +0.089 |
Megan | +0.086 |
restaurant | +0.084 |
Jacob | +0.080 |
Pen | +0.079 |
Neg logit contributions | |
bar | -0.079 |
ink | -0.079 |
chalk | -0.079 |
bullet | -0.075 |
cho | -0.075 |
Token | Everybody |
Token ID | 28172 |
Feature activation | +2.40e+00 |
Pos logit contributions | |
chalk | +0.009 |
ink | +0.009 |
bar | +0.008 |
bullet | +0.008 |
gol | +0.008 |
Neg logit contributions | |
Jacob | -0.010 |
Megan | -0.010 |
restaurant | -0.010 |
Jacob | -0.009 |
Pen | -0.009 |
Token | wanted |
Token ID | 2227 |
Feature activation | -6.61e-01 |
Pos logit contributions | |
bar | +0.171 |
chalk | +0.169 |
ink | +0.169 |
bullet | +0.161 |
ch | +0.161 |
Neg logit contributions | |
Jacob | -0.185 |
Megan | -0.178 |
restaurant | -0.173 |
Pen | -0.164 |
Jacob | -0.163 |
Token | a |
Token ID | 257 |
Feature activation | +4.42e-01 |
Pos logit contributions | |
Jacob | +0.058 |
Megan | +0.056 |
restaurant | +0.055 |
Pen | +0.052 |
Jacob | +0.051 |
Neg logit contributions | |
bar | -0.041 |
chalk | -0.040 |
ink | -0.040 |
bullet | -0.039 |
itted | -0.038 |
Token | tw |
Token ID | 665 |
Feature activation | +3.02e+00 |
Pos logit contributions | |
bar | +0.030 |
itted | +0.029 |
bullet | +0.029 |
ink | +0.029 |
chalk | +0.028 |
Neg logit contributions | |
Jacob | -0.035 |
Megan | -0.035 |
restaurant | -0.035 |
Pen | -0.033 |
Nick | -0.032 |
Token | ig |
Token ID | 328 |
Feature activation | +3.54e+00 |
Pos logit contributions | |
bar | +0.139 |
chalk | +0.137 |
ink | +0.137 |
bullet | +0.130 |
ch | +0.126 |
Neg logit contributions | |
Jacob | -0.308 |
Megan | -0.301 |
restaurant | -0.294 |
Pen | -0.283 |
Jacob | -0.281 |
Token | just |
Token ID | 655 |
Feature activation | +4.67e-01 |
Pos logit contributions | |
ink | +0.244 |
bullet | +0.244 |
chalk | +0.241 |
gol | +0.239 |
cho | +0.227 |
Neg logit contributions | |
Jacob | -0.270 |
Megan | -0.261 |
restaurant | -0.256 |
Pen | -0.239 |
Nick | -0.231 |
Token | like |
Token ID | 588 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
bar | +0.036 |
bullet | +0.035 |
ch | +0.034 |
ink | +0.034 |
cho | +0.034 |
Neg logit contributions | |
Jacob | -0.034 |
Megan | -0.033 |
restaurant | -0.032 |
Pen | -0.029 |
Nick | -0.029 |
Token | Bo |
Token ID | 3248 |
Feature activation | -9.84e-01 |
Pos logit contributions | |
bar | +0.018 |
chalk | +0.017 |
ink | +0.017 |
bullet | +0.017 |
itted | +0.017 |
Neg logit contributions | |
Jacob | -0.019 |
Megan | -0.018 |
restaurant | -0.017 |
Pen | -0.017 |
Nick | -0.016 |
Token | 's |
Token ID | 338 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
Jacob | +0.080 |
Megan | +0.077 |
restaurant | +0.076 |
Jacob | +0.073 |
Pen | +0.069 |
Neg logit contributions | |
ch | -0.065 |
ink | -0.064 |
bar | -0.064 |
chalk | -0.063 |
cho | -0.063 |
Token | , |
Token ID | 11 |
Feature activation | +2.84e-01 |
Pos logit contributions | |
Jacob | +0.105 |
Megan | +0.102 |
restaurant | +0.099 |
Pen | +0.096 |
Jacob | +0.094 |
Neg logit contributions | |
bar | -0.058 |
chalk | -0.057 |
ink | -0.056 |
bullet | -0.055 |
ch | -0.054 |
Token | so |
Token ID | 523 |
Feature activation | +3.79e-01 |
Pos logit contributions | |
bar | +0.020 |
chalk | +0.019 |
ink | +0.019 |
bullet | +0.019 |
ch | +0.018 |
Neg logit contributions | |
Jacob | -0.023 |
Megan | -0.022 |
restaurant | -0.021 |
Pen | -0.020 |
Jacob | -0.020 |
Token | source |
Token ID | 2723 |
Feature activation | +6.54e-02 |
Pos logit contributions | |
Jacob | +0.056 |
restaurant | +0.054 |
Megan | +0.052 |
Pen | +0.050 |
Jacob | +0.049 |
Neg logit contributions | |
bar | -0.048 |
ink | -0.044 |
chalk | -0.043 |
cho | -0.043 |
itted | -0.043 |
Token | of |
Token ID | 286 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
bar | +0.003 |
chalk | +0.003 |
ink | +0.003 |
bullet | +0.003 |
ch | +0.003 |
Neg logit contributions | |
Jacob | -0.007 |
Megan | -0.007 |
restaurant | -0.006 |
Pen | -0.006 |
Jacob | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | -6.89e-01 |
Pos logit contributions | |
chalk | +0.082 |
ink | +0.082 |
bar | +0.080 |
bullet | +0.076 |
cho | +0.076 |
Neg logit contributions | |
Jacob | -0.112 |
Megan | -0.108 |
restaurant | -0.106 |
Jacob | -0.101 |
Pen | -0.100 |
Token | language |
Token ID | 3303 |
Feature activation | +2.86e+00 |
Pos logit contributions | |
Jacob | +0.054 |
restaurant | +0.053 |
Megan | +0.050 |
Pen | +0.047 |
Jacob | +0.047 |
Neg logit contributions | |
bar | -0.052 |
ink | -0.045 |
itted | -0.045 |
chalk | -0.045 |
cho | -0.044 |
Token | . |
Token ID | 13 |
Feature activation | -4.02e-02 |
Pos logit contributions | |
bar | +0.211 |
chalk | +0.210 |
ink | +0.209 |
ch | +0.199 |
bullet | +0.198 |
Neg logit contributions | |
Jacob | -0.212 |
Megan | -0.205 |
restaurant | -0.200 |
Jacob | -0.189 |
Pen | -0.188 |
Token | It |
Token ID | 632 |
Feature activation | +6.13e-01 |
Pos logit contributions | |
Jacob | +0.003 |
Megan | +0.003 |
restaurant | +0.003 |
Pen | +0.003 |
Jacob | +0.003 |
Neg logit contributions | |
bar | -0.003 |
ink | -0.003 |
chalk | -0.003 |
bullet | -0.003 |
cho | -0.003 |
Token | was |
Token ID | 373 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
bar | +0.045 |
ch | +0.045 |
cho | +0.044 |
ink | +0.044 |
chalk | +0.044 |
Neg logit contributions | |
Jacob | -0.045 |
restaurant | -0.045 |
Megan | -0.044 |
Jacob | -0.041 |
Pen | -0.040 |
Token | a |
Token ID | 257 |
Feature activation | +4.22e-01 |
Pos logit contributions | |
Jacob | +0.107 |
Megan | +0.103 |
restaurant | +0.101 |
Pen | +0.096 |
Jacob | +0.095 |
Neg logit contributions | |
bar | -0.077 |
chalk | -0.075 |
ink | -0.075 |
bullet | -0.072 |
ch | -0.071 |
Token | group |
Token ID | 1448 |
Feature activation | -7.16e-01 |
Pos logit contributions | |
itted | +0.031 |
bar | +0.030 |
alks | +0.028 |
ink | +0.027 |
chalk | +0.027 |
Neg logit contributions | |
restaurant | -0.034 |
Jacob | -0.032 |
Megan | -0.030 |
Pen | -0.029 |
Jacob | -0.028 |
Token | of |
Token ID | 286 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
Jacob | +0.067 |
Megan | +0.064 |
restaurant | +0.064 |
Pen | +0.060 |
Jacob | +0.059 |
Neg logit contributions | |
bar | -0.039 |
ink | -0.038 |
chalk | -0.037 |
bullet | -0.037 |
itted | -0.037 |
Token | people |
Token ID | 661 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
bar | +0.100 |
itted | +0.089 |
ink | +0.087 |
alks | +0.087 |
vals | +0.085 |
Neg logit contributions | |
Jacob | -0.103 |
restaurant | -0.100 |
Megan | -0.098 |
Pen | -0.091 |
Jacob | -0.089 |
Token | . |
Token ID | 13 |
Feature activation | -5.68e-02 |
Pos logit contributions | |
bar | +0.189 |
ink | +0.188 |
chalk | +0.187 |
itted | +0.182 |
bullet | +0.181 |
Neg logit contributions | |
Jacob | -0.157 |
restaurant | -0.149 |
Megan | -0.148 |
Pen | -0.135 |
Jacob | -0.134 |
Token | The |
Token ID | 383 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
restaurant | +0.006 |
Jacob | +0.006 |
Megan | +0.006 |
Jacob | +0.005 |
abel | +0.005 |
Neg logit contributions | |
chalk | -0.003 |
ink | -0.002 |
bar | -0.002 |
Bounce | -0.002 |
bullet | -0.002 |
Token | car |
Token ID | 1097 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
vals | +0.011 |
itted | +0.011 |
bar | +0.011 |
alks | +0.011 |
Bree | +0.011 |
Neg logit contributions | |
restaurant | -0.011 |
Jacob | -0.010 |
cabin | -0.010 |
Megan | -0.009 |
Frost | -0.009 |
Token | liked |
Token ID | 8288 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
bar | +0.176 |
chalk | +0.174 |
ink | +0.173 |
cho | +0.167 |
bullet | +0.166 |
Neg logit contributions | |
Jacob | -0.175 |
Megan | -0.170 |
restaurant | -0.163 |
Pen | -0.155 |
Jacob | -0.154 |
Token | to |
Token ID | 284 |
Feature activation | +5.59e-01 |
Pos logit contributions | |
Jacob | +0.076 |
restaurant | +0.072 |
Megan | +0.068 |
Frost | +0.066 |
Pen | +0.066 |
Neg logit contributions | |
bar | -0.085 |
plets | -0.083 |
itted | -0.083 |
ink | -0.083 |
vals | -0.080 |
Token | go |
Token ID | 467 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
bar | +0.037 |
ink | +0.037 |
chalk | +0.036 |
itted | +0.035 |
bullet | +0.034 |
Neg logit contributions | |
Jacob | -0.046 |
Megan | -0.044 |
restaurant | -0.043 |
Pen | -0.041 |
Jacob | -0.041 |
Token | fast |
Token ID | 3049 |
Feature activation | -4.44e-01 |
Pos logit contributions | |
itted | +0.118 |
ink | +0.117 |
bar | +0.113 |
chalk | +0.113 |
gol | +0.110 |
Neg logit contributions | |
restaurant | -0.113 |
Jacob | -0.111 |
Megan | -0.103 |
Frost | -0.101 |
cabin | -0.101 |
Token | . |
Token ID | 13 |
Feature activation | -4.60e-02 |
Pos logit contributions | |
Jacob | +0.037 |
Megan | +0.037 |
restaurant | +0.035 |
Jacob | +0.034 |
Pen | +0.033 |
Neg logit contributions | |
ink | -0.028 |
chalk | -0.028 |
bar | -0.028 |
ch | -0.028 |
cho | -0.028 |
Token | One |
Token ID | 1881 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
Jacob | +0.005 |
Megan | +0.005 |
restaurant | +0.005 |
Jacob | +0.004 |
Pen | +0.004 |
Neg logit contributions | |
chalk | -0.002 |
ink | -0.002 |
bar | -0.002 |
ch | -0.002 |
bullet | -0.002 |
Token | day |
Token ID | 1110 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
Jacob | +0.232 |
Megan | +0.225 |
restaurant | +0.222 |
Pen | +0.210 |
Jacob | +0.210 |
Neg logit contributions | |
bar | -0.135 |
ink | -0.131 |
chalk | -0.130 |
bullet | -0.124 |
itted | -0.122 |
Token | , |
Token ID | 11 |
Feature activation | +2.57e-01 |
Pos logit contributions | |
bar | +0.145 |
itted | +0.134 |
ink | +0.134 |
ully | +0.134 |
bullet | +0.133 |
Neg logit contributions | |
Jacob | -0.103 |
Megan | -0.097 |
restaurant | -0.091 |
Pen | -0.089 |
Nick | -0.088 |
Token | , |
Token ID | 11 |
Feature activation | +2.74e-01 |
Pos logit contributions | |
bar | +0.177 |
ully | +0.162 |
bullet | +0.161 |
peac | +0.159 |
gol | +0.158 |
Neg logit contributions | |
Jacob | -0.080 |
Megan | -0.078 |
restaurant | -0.072 |
Pen | -0.067 |
Nick | -0.063 |
Token | Lily |
Token ID | 20037 |
Feature activation | +6.29e-01 |
Pos logit contributions | |
bar | +0.026 |
itted | +0.024 |
alks | +0.023 |
ully | +0.023 |
bullet | +0.023 |
Neg logit contributions | |
Jacob | -0.018 |
Megan | -0.018 |
restaurant | -0.017 |
Pen | -0.016 |
Lucy | -0.015 |
Token | was |
Token ID | 373 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
ink | +0.035 |
chalk | +0.034 |
bar | +0.031 |
bullet | +0.030 |
ch | +0.029 |
Neg logit contributions | |
Megan | -0.063 |
Jacob | -0.063 |
Jacob | -0.057 |
Pen | -0.057 |
Clive | -0.056 |
Token | walking |
Token ID | 6155 |
Feature activation | -4.49e-01 |
Pos logit contributions | |
Jacob | +0.099 |
restaurant | +0.096 |
Megan | +0.095 |
Pen | +0.091 |
Jacob | +0.086 |
Neg logit contributions | |
chalk | -0.081 |
ink | -0.079 |
bar | -0.079 |
bullet | -0.078 |
gol | -0.076 |
Token | up |
Token ID | 510 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
Jacob | +0.038 |
Megan | +0.036 |
restaurant | +0.035 |
Jacob | +0.034 |
Pen | +0.033 |
Neg logit contributions | |
chalk | -0.030 |
bar | -0.029 |
ink | -0.029 |
ch | -0.028 |
bullet | -0.027 |
Token | the |
Token ID | 262 |
Feature activation | -6.75e-01 |
Pos logit contributions | |
Jacob | +0.116 |
Megan | +0.112 |
restaurant | +0.107 |
Jacob | +0.104 |
Pen | +0.102 |
Neg logit contributions | |
ink | -0.086 |
chalk | -0.086 |
bar | -0.082 |
ch | -0.082 |
cho | -0.080 |
Token | steps |
Token ID | 4831 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
restaurant | +0.047 |
Jacob | +0.047 |
Megan | +0.045 |
Pen | +0.043 |
Jacob | +0.041 |
Neg logit contributions | |
bar | -0.052 |
itted | -0.050 |
ink | -0.050 |
chalk | -0.050 |
bullet | -0.049 |
Token | to |
Token ID | 284 |
Feature activation | +5.87e-01 |
Pos logit contributions | |
Jacob | +0.142 |
Megan | +0.134 |
restaurant | +0.132 |
Pen | +0.123 |
abel | +0.119 |
Neg logit contributions | |
bullet | -0.165 |
ink | -0.163 |
chalk | -0.161 |
peac | -0.159 |
Bounce | -0.155 |
Token | her |
Token ID | 607 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
bar | +0.041 |
chalk | +0.041 |
ink | +0.040 |
ch | +0.038 |
bullet | +0.038 |
Neg logit contributions | |
Jacob | -0.047 |
Megan | -0.045 |
restaurant | -0.044 |
Jacob | -0.042 |
Pen | -0.042 |
Token | house |
Token ID | 2156 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
Jacob | +0.013 |
Megan | +0.013 |
Jacob | +0.012 |
Clive | +0.012 |
restaurant | +0.012 |
Neg logit contributions | |
chalk | -0.009 |
ink | -0.008 |
bar | -0.008 |
sun | -0.008 |
laser | -0.008 |
Token | when |
Token ID | 618 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
Jacob | +0.166 |
Megan | +0.160 |
Jacob | +0.152 |
restaurant | +0.151 |
Pen | +0.148 |
Neg logit contributions | |
bar | -0.125 |
chalk | -0.124 |
ink | -0.123 |
ch | -0.114 |
itted | -0.113 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
chalk | +0.097 |
ink | +0.096 |
bar | +0.094 |
ch | +0.093 |
cho | +0.089 |
Neg logit contributions | |
Jacob | -0.098 |
Megan | -0.096 |
Pen | -0.088 |
Jacob | -0.088 |
Gor | -0.087 |
Token |
|
Token ID | 198 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
Jacob | +0.083 |
restaurant | +0.081 |
Megan | +0.080 |
Pen | +0.076 |
abel | +0.075 |
Neg logit contributions | |
chalk | -0.065 |
ink | -0.063 |
bar | -0.061 |
ch | -0.059 |
Bree | -0.059 |
Token |
|
Token ID | 198 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
chalk | +0.008 |
ink | +0.007 |
bar | +0.007 |
ch | +0.007 |
bullet | +0.007 |
Neg logit contributions | |
Jacob | -0.010 |
Megan | -0.009 |
restaurant | -0.009 |
Pen | -0.009 |
Jacob | -0.009 |
Token | M |
Token ID | 44 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
bar | +0.008 |
chalk | +0.008 |
ink | +0.008 |
ch | +0.008 |
bullet | +0.008 |
Neg logit contributions | |
Jacob | -0.010 |
Megan | -0.009 |
restaurant | -0.009 |
Pen | -0.008 |
Jacob | -0.008 |
Token | ia |
Token ID | 544 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
Jacob | +0.128 |
Megan | +0.122 |
restaurant | +0.117 |
Jacob | +0.107 |
Pen | +0.107 |
Neg logit contributions | |
bar | -0.211 |
chalk | -0.208 |
ink | -0.208 |
bullet | -0.202 |
ch | -0.202 |
Token | is |
Token ID | 318 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
restaurant | +0.163 |
Jacob | +0.162 |
Megan | +0.162 |
ide | +0.153 |
Pen | +0.152 |
Neg logit contributions | |
chalk | -0.130 |
ink | -0.126 |
er | -0.112 |
pins | -0.111 |
bullet | -0.109 |
Token | happy |
Token ID | 3772 |
Feature activation | -5.23e-01 |
Pos logit contributions | |
Jacob | +0.160 |
Megan | +0.154 |
restaurant | +0.154 |
Jacob | +0.146 |
Pen | +0.145 |
Neg logit contributions | |
chalk | -0.093 |
ink | -0.093 |
bar | -0.091 |
itted | -0.088 |
cho | -0.086 |
Token | . |
Token ID | 13 |
Feature activation | -2.33e-02 |
Pos logit contributions | |
Jacob | +0.051 |
Jacob | +0.050 |
Megan | +0.050 |
Pen | +0.049 |
Gor | +0.049 |
Neg logit contributions | |
ink | -0.026 |
chalk | -0.025 |
sun | -0.022 |
Jazz | -0.020 |
ch | -0.020 |
Token | She |
Token ID | 1375 |
Feature activation | -6.15e-01 |
Pos logit contributions | |
Jacob | +0.002 |
Megan | +0.002 |
restaurant | +0.002 |
Jacob | +0.002 |
Pen | +0.002 |
Neg logit contributions | |
chalk | -0.001 |
ink | -0.001 |
bar | -0.001 |
bullet | -0.001 |
ch | -0.001 |
Token | says |
Token ID | 1139 |
Feature activation | -8.11e-02 |
Pos logit contributions | |
Jacob | +0.051 |
Megan | +0.049 |
restaurant | +0.047 |
Pen | +0.046 |
Jacob | +0.046 |
Neg logit contributions | |
chalk | -0.042 |
ink | -0.041 |
bullet | -0.038 |
ch | -0.038 |
bar | -0.038 |
Token | , |
Token ID | 11 |
Feature activation | +2.84e-01 |
Pos logit contributions | |
Jacob | +0.006 |
Megan | +0.005 |
Pen | +0.005 |
restaurant | +0.005 |
Nick | +0.005 |
Neg logit contributions | |
bar | -0.007 |
ink | -0.007 |
chalk | -0.007 |
bullet | -0.007 |
itted | -0.007 |
Token | around |
Token ID | 1088 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
bar | +0.175 |
ink | +0.173 |
chalk | +0.171 |
bullet | +0.167 |
itted | +0.163 |
Neg logit contributions | |
Jacob | -0.201 |
Megan | -0.194 |
restaurant | -0.186 |
Pen | -0.177 |
Jacob | -0.177 |
Token | the |
Token ID | 262 |
Feature activation | -7.03e-01 |
Pos logit contributions | |
Jacob | +0.161 |
Megan | +0.157 |
restaurant | +0.155 |
Jacob | +0.149 |
Clive | +0.145 |
Neg logit contributions | |
chalk | -0.117 |
ink | -0.115 |
ch | -0.108 |
bar | -0.107 |
cho | -0.106 |
Token | garden |
Token ID | 11376 |
Feature activation | +4.03e-01 |
Pos logit contributions | |
Jacob | +0.049 |
Megan | +0.047 |
restaurant | +0.046 |
Pen | +0.043 |
Jacob | +0.043 |
Neg logit contributions | |
bar | -0.056 |
ink | -0.055 |
chalk | -0.055 |
bullet | -0.053 |
itted | -0.053 |
Token | , |
Token ID | 11 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
bar | +0.027 |
chalk | +0.027 |
ink | +0.027 |
cho | +0.025 |
ch | +0.025 |
Neg logit contributions | |
Jacob | -0.033 |
Megan | -0.032 |
restaurant | -0.031 |
Jacob | -0.030 |
Pen | -0.029 |
Token | feeling |
Token ID | 4203 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
bar | +0.017 |
ink | +0.017 |
chalk | +0.016 |
bullet | +0.016 |
ch | +0.016 |
Neg logit contributions | |
Jacob | -0.020 |
Megan | -0.019 |
restaurant | -0.018 |
Pen | -0.018 |
Jacob | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | -7.01e-01 |
Pos logit contributions | |
bar | +0.126 |
chalk | +0.122 |
ink | +0.120 |
bullet | +0.118 |
ch | +0.115 |
Neg logit contributions | |
Jacob | -0.137 |
Megan | -0.133 |
restaurant | -0.128 |
Pen | -0.123 |
Jacob | -0.119 |
Token | sunshine |
Token ID | 34488 |
Feature activation | +2.91e+00 |
Pos logit contributions | |
restaurant | +0.069 |
Jacob | +0.067 |
Megan | +0.067 |
Pen | +0.064 |
Nick | +0.061 |
Neg logit contributions | |
bar | -0.035 |
Bree | -0.031 |
itted | -0.031 |
Bounce | -0.030 |
cho | -0.030 |
Token | on |
Token ID | 319 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
bar | +0.189 |
chalk | +0.186 |
ink | +0.183 |
bullet | +0.179 |
gol | +0.176 |
Neg logit contributions | |
Jacob | -0.244 |
Megan | -0.237 |
restaurant | -0.229 |
Pen | -0.221 |
Jacob | -0.215 |
Token | her |
Token ID | 607 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
chalk | +0.078 |
ink | +0.078 |
bar | +0.071 |
bullet | +0.068 |
cho | +0.068 |
Neg logit contributions | |
Jacob | -0.122 |
Megan | -0.118 |
restaurant | -0.117 |
Jacob | -0.113 |
Pen | -0.111 |
Token | face |
Token ID | 1986 |
Feature activation | +3.55e+00 |
Pos logit contributions | |
Jacob | +0.016 |
Megan | +0.016 |
restaurant | +0.016 |
Pen | +0.015 |
Jacob | +0.015 |
Neg logit contributions | |
bar | -0.011 |
ink | -0.010 |
ch | -0.010 |
chalk | -0.010 |
cho | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -4.60e-02 |
Pos logit contributions | |
bar | +0.290 |
bullet | +0.285 |
chalk | +0.285 |
ink | +0.283 |
gol | +0.281 |
Neg logit contributions | |
Jacob | -0.231 |
Megan | -0.222 |
restaurant | -0.214 |
Pen | -0.201 |
Jacob | -0.192 |
Token | says |
Token ID | 1139 |
Feature activation | -8.01e-02 |
Pos logit contributions | |
Jacob | +0.048 |
Megan | +0.047 |
restaurant | +0.045 |
Jacob | +0.044 |
Pen | +0.044 |
Neg logit contributions | |
chalk | -0.039 |
ink | -0.039 |
bar | -0.037 |
ch | -0.037 |
cho | -0.037 |
Token | they |
Token ID | 484 |
Feature activation | +6.09e-01 |
Pos logit contributions | |
Jacob | +0.006 |
Megan | +0.006 |
restaurant | +0.006 |
Pen | +0.006 |
Nick | +0.005 |
Neg logit contributions | |
bar | -0.006 |
itted | -0.006 |
ink | -0.006 |
chalk | -0.006 |
bullet | -0.005 |
Token | are |
Token ID | 389 |
Feature activation | -2.80e-01 |
Pos logit contributions | |
chalk | +0.037 |
ink | +0.036 |
bar | +0.035 |
ch | +0.034 |
bullet | +0.033 |
Neg logit contributions | |
Jacob | -0.056 |
Megan | -0.055 |
restaurant | -0.052 |
Pen | -0.052 |
Jacob | -0.051 |
Token | too |
Token ID | 1165 |
Feature activation | -3.51e+00 |
Pos logit contributions | |
Jacob | +0.024 |
Megan | +0.023 |
restaurant | +0.022 |
Pen | +0.021 |
Jacob | +0.021 |
Neg logit contributions | |
bar | -0.018 |
chalk | -0.018 |
ink | -0.017 |
bullet | -0.017 |
itted | -0.017 |
Token | small |
Token ID | 1402 |
Feature activation | +5.15e-01 |
Pos logit contributions | |
Jacob | +0.334 |
Megan | +0.323 |
restaurant | +0.316 |
Pen | +0.303 |
Jacob | +0.303 |
Neg logit contributions | |
chalk | -0.188 |
ink | -0.186 |
bar | -0.185 |
cho | -0.177 |
ch | -0.176 |
Token | and |
Token ID | 290 |
Feature activation | -1.00e-01 |
Pos logit contributions | |
ink | +0.031 |
chalk | +0.031 |
bar | +0.030 |
cho | +0.029 |
ch | +0.029 |
Neg logit contributions | |
Jacob | -0.046 |
Megan | -0.046 |
Jacob | -0.043 |
Pen | -0.042 |
restaurant | -0.042 |
Token | too |
Token ID | 1165 |
Feature activation | -3.50e+00 |
Pos logit contributions | |
Jacob | +0.009 |
Megan | +0.008 |
restaurant | +0.008 |
Pen | +0.008 |
Jacob | +0.008 |
Neg logit contributions | |
bar | -0.006 |
chalk | -0.006 |
ink | -0.006 |
bullet | -0.006 |
ch | -0.006 |
Token | slow |
Token ID | 3105 |
Feature activation | -6.03e-01 |
Pos logit contributions | |
Jacob | +0.331 |
Megan | +0.324 |
restaurant | +0.311 |
Jacob | +0.302 |
Pen | +0.301 |
Neg logit contributions | |
chalk | -0.189 |
ink | -0.188 |
bar | -0.185 |
ch | -0.180 |
cho | -0.179 |
Token | . |
Token ID | 13 |
Feature activation | -2.62e-02 |
Pos logit contributions | |
Jacob | +0.038 |
Megan | +0.035 |
restaurant | +0.034 |
Pen | +0.032 |
Jacob | +0.031 |
Neg logit contributions | |
chalk | -0.051 |
bar | -0.051 |
ink | -0.051 |
bullet | -0.049 |
peac | -0.049 |
Token | He |
Token ID | 679 |
Feature activation | -5.96e-01 |
Pos logit contributions | |
Jacob | +0.002 |
Megan | +0.002 |
restaurant | +0.002 |
Jacob | +0.002 |
Pen | +0.002 |
Neg logit contributions | |
bar | -0.002 |
chalk | -0.002 |
ink | -0.002 |
cho | -0.002 |
ch | -0.002 |
Token | says |
Token ID | 1139 |
Feature activation | -7.42e-02 |
Pos logit contributions | |
Jacob | +0.047 |
Megan | +0.046 |
restaurant | +0.044 |
Pen | +0.043 |
Jacob | +0.043 |
Neg logit contributions | |
chalk | -0.041 |
ink | -0.041 |
bar | -0.040 |
ch | -0.039 |
cho | -0.039 |
Token | the |
Token ID | 262 |
Feature activation | -6.66e-01 |
Pos logit contributions | |
chalk | +0.021 |
bar | +0.021 |
ink | +0.021 |
cho | +0.020 |
sun | +0.020 |
Neg logit contributions | |
Jacob | -0.038 |
Jacob | -0.037 |
Megan | -0.037 |
restaurant | -0.036 |
abel | -0.036 |
Token | birds |
Token ID | 10087 |
Feature activation | +7.35e-01 |
Pos logit contributions | |
Jacob | +0.065 |
Megan | +0.064 |
restaurant | +0.063 |
Pen | +0.060 |
Jacob | +0.059 |
Neg logit contributions | |
bar | -0.034 |
chalk | -0.033 |
ink | -0.033 |
bullet | -0.031 |
ch | -0.031 |
Token | got |
Token ID | 1392 |
Feature activation | +5.50e-01 |
Pos logit contributions | |
bar | +0.050 |
chalk | +0.050 |
ink | +0.050 |
ch | +0.048 |
cho | +0.047 |
Neg logit contributions | |
Jacob | -0.059 |
Megan | -0.056 |
restaurant | -0.055 |
Jacob | -0.052 |
Pen | -0.052 |
Token | bigger |
Token ID | 5749 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
bar | +0.040 |
chalk | +0.039 |
bullet | +0.038 |
ink | +0.038 |
gol | +0.037 |
Neg logit contributions | |
Jacob | -0.041 |
Megan | -0.040 |
restaurant | -0.039 |
Pen | -0.037 |
Nick | -0.036 |
Token | too |
Token ID | 1165 |
Feature activation | -3.50e+00 |
Pos logit contributions | |
ink | +0.091 |
chalk | +0.089 |
bar | +0.087 |
cho | +0.083 |
itted | +0.083 |
Neg logit contributions | |
Jacob | -0.119 |
Megan | -0.114 |
Jacob | -0.110 |
restaurant | -0.110 |
Pen | -0.106 |
Token | . |
Token ID | 13 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
Jacob | +0.276 |
Megan | +0.266 |
restaurant | +0.260 |
Jacob | +0.249 |
Pen | +0.245 |
Neg logit contributions | |
chalk | -0.242 |
ink | -0.242 |
bar | -0.241 |
cho | -0.229 |
ch | -0.229 |
Token | |
Token ID | 220 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
restaurant | +0.002 |
Jacob | +0.002 |
Megan | +0.001 |
Jacob | +0.001 |
abel | +0.001 |
Neg logit contributions | |
bar | -0.001 |
chalk | -0.001 |
ink | -0.001 |
sun | -0.001 |
ch | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
bar | +0.077 |
chalk | +0.073 |
itted | +0.071 |
ink | +0.071 |
ch | +0.070 |
Neg logit contributions | |
restaurant | -0.120 |
Jacob | -0.120 |
Megan | -0.118 |
Jacob | -0.110 |
Pen | -0.110 |
Token |
|
Token ID | 198 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
bar | +0.007 |
ch | +0.007 |
chalk | +0.006 |
cho | +0.006 |
Bounce | +0.006 |
Neg logit contributions | |
restaurant | -0.011 |
Jacob | -0.011 |
Megan | -0.010 |
abel | -0.010 |
Pen | -0.010 |
Token | But |
Token ID | 1537 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
bar | +0.007 |
ch | +0.007 |
cho | +0.006 |
chalk | +0.006 |
crashing | +0.006 |
Neg logit contributions | |
Jacob | -0.011 |
restaurant | -0.010 |
Megan | -0.010 |
abel | -0.010 |
Pen | -0.009 |
Token | Bob |
Token ID | 5811 |
Feature activation | -5.82e+00 |
Pos logit contributions | |
bar | +0.123 |
chalk | +0.114 |
cho | +0.113 |
sun | +0.110 |
ink | +0.109 |
Neg logit contributions | |
restaurant | -0.171 |
Jacob | -0.170 |
Megan | -0.164 |
Jacob | -0.164 |
abel | -0.160 |
Token | out |
Token ID | 503 |
Feature activation | +3.37e+00 |
Pos logit contributions | |
ch | +0.103 |
ink | +0.102 |
bar | +0.101 |
chalk | +0.100 |
cho | +0.098 |
Neg logit contributions | |
restaurant | -0.143 |
Jacob | -0.142 |
Megan | -0.139 |
Jacob | -0.132 |
Pen | -0.128 |
Token | his |
Token ID | 465 |
Feature activation | +7.07e-02 |
Pos logit contributions | |
ink | +0.167 |
chalk | +0.164 |
ch | +0.153 |
bar | +0.148 |
cho | +0.147 |
Neg logit contributions | |
Jacob | -0.317 |
restaurant | -0.313 |
Megan | -0.311 |
Jacob | -0.300 |
Pen | -0.293 |
Token | phone |
Token ID | 3072 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
bar | +0.005 |
crashing | +0.004 |
cho | +0.004 |
gol | +0.004 |
plets | +0.004 |
Neg logit contributions | |
restaurant | -0.006 |
Jacob | -0.006 |
Megan | -0.005 |
Pen | -0.005 |
chamber | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | -1.00e-01 |
Pos logit contributions | |
bar | +0.109 |
chalk | +0.107 |
ink | +0.107 |
itted | +0.099 |
bullet | +0.097 |
Neg logit contributions | |
Jacob | -0.152 |
Megan | -0.152 |
Jacob | -0.145 |
restaurant | -0.145 |
Pen | -0.140 |
Token | showed |
Token ID | 3751 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
Jacob | +0.008 |
Megan | +0.008 |
restaurant | +0.008 |
Jacob | +0.007 |
Pen | +0.007 |
Neg logit contributions | |
chalk | -0.007 |
ink | -0.007 |
bar | -0.007 |
bullet | -0.006 |
ch | -0.006 |
Token | Anna |
Token ID | 11735 |
Feature activation | -5.45e+00 |
Pos logit contributions | |
ch | +0.085 |
ink | +0.083 |
bar | +0.083 |
cho | +0.083 |
chalk | +0.082 |
Neg logit contributions | |
restaurant | -0.115 |
Jacob | -0.110 |
Megan | -0.108 |
Jacob | -0.104 |
Pen | -0.100 |
Token | a |
Token ID | 257 |
Feature activation | +4.34e-01 |
Pos logit contributions | |
Jacob | +0.505 |
Jacob | +0.501 |
Megan | +0.496 |
restaurant | +0.486 |
Pen | +0.463 |
Neg logit contributions | |
chalk | -0.271 |
ink | -0.267 |
bar | -0.256 |
cho | -0.235 |
itted | -0.223 |
Token | photo |
Token ID | 4590 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
bar | +0.031 |
itted | +0.030 |
gol | +0.030 |
iece | +0.029 |
plets | +0.029 |
Neg logit contributions | |
restaurant | -0.033 |
Jacob | -0.032 |
Megan | -0.029 |
Pen | -0.029 |
chamber | -0.028 |
Token | of |
Token ID | 286 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
bar | +0.172 |
chalk | +0.170 |
ink | +0.170 |
bullet | +0.161 |
ch | +0.161 |
Neg logit contributions | |
Jacob | -0.237 |
Megan | -0.229 |
restaurant | -0.224 |
Pen | -0.213 |
Jacob | -0.212 |
Token | a |
Token ID | 257 |
Feature activation | +4.32e-01 |
Pos logit contributions | |
chalk | +0.090 |
ink | +0.090 |
bar | +0.089 |
ch | +0.085 |
bullet | +0.084 |
Neg logit contributions | |
Jacob | -0.104 |
Megan | -0.100 |
restaurant | -0.098 |
Jacob | -0.093 |
Pen | -0.093 |
Token | fluffy |
Token ID | 39145 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
itted | +0.031 |
bar | +0.030 |
ully | +0.027 |
amed | +0.027 |
iece | +0.027 |
Neg logit contributions | |
restaurant | -0.035 |
Jacob | -0.033 |
Megan | -0.031 |
chamber | -0.030 |
Pen | -0.029 |
Token | and |
Token ID | 290 |
Feature activation | -4.85e-02 |
Pos logit contributions | |
chalk | +0.038 |
bar | +0.038 |
ink | +0.037 |
ch | +0.036 |
bullet | +0.036 |
Neg logit contributions | |
Jacob | -0.039 |
Megan | -0.038 |
restaurant | -0.037 |
Jacob | -0.034 |
Pen | -0.034 |
Token | Tom |
Token ID | 4186 |
Feature activation | -3.41e-01 |
Pos logit contributions | |
Jacob | +0.004 |
Megan | +0.004 |
restaurant | +0.003 |
Pen | +0.003 |
Jacob | +0.003 |
Neg logit contributions | |
bar | -0.004 |
ink | -0.004 |
chalk | -0.004 |
bullet | -0.003 |
itted | -0.003 |
Token | are |
Token ID | 389 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
Jacob | +0.029 |
Megan | +0.028 |
restaurant | +0.027 |
Jacob | +0.026 |
Pen | +0.026 |
Neg logit contributions | |
chalk | -0.022 |
bar | -0.021 |
ink | -0.021 |
ch | -0.020 |
itted | -0.020 |
Token | at |
Token ID | 379 |
Feature activation | -9.16e-02 |
Pos logit contributions | |
Jacob | +0.020 |
restaurant | +0.019 |
Megan | +0.019 |
Nick | +0.017 |
Pen | +0.017 |
Neg logit contributions | |
bullet | -0.019 |
bar | -0.019 |
br | -0.018 |
plets | -0.018 |
ch | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | -6.70e-01 |
Pos logit contributions | |
Jacob | +0.007 |
Megan | +0.007 |
restaurant | +0.007 |
Jacob | +0.007 |
Pen | +0.007 |
Neg logit contributions | |
chalk | -0.006 |
ink | -0.006 |
bar | -0.006 |
bullet | -0.006 |
cho | -0.006 |
Token | shop |
Token ID | 6128 |
Feature activation | -5.11e+00 |
Pos logit contributions | |
Jacob | +0.047 |
Megan | +0.045 |
restaurant | +0.044 |
Pen | +0.041 |
Jacob | +0.041 |
Neg logit contributions | |
bar | -0.052 |
chalk | -0.052 |
ink | -0.052 |
bullet | -0.050 |
itted | -0.050 |
Token | with |
Token ID | 351 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
Jacob | +0.442 |
Megan | +0.426 |
Jacob | +0.422 |
Pen | +0.413 |
Clive | +0.397 |
Neg logit contributions | |
bar | -0.321 |
chalk | -0.297 |
itted | -0.294 |
ink | -0.290 |
cho | -0.288 |
Token | Mom |
Token ID | 11254 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
Jacob | +0.148 |
Megan | +0.144 |
restaurant | +0.141 |
Jacob | +0.141 |
Pen | +0.137 |
Neg logit contributions | |
chalk | -0.099 |
ink | -0.097 |
bar | -0.091 |
cho | -0.088 |
bullet | -0.087 |
Token | . |
Token ID | 13 |
Feature activation | -2.87e-02 |
Pos logit contributions | |
Jacob | +0.140 |
Megan | +0.138 |
Jacob | +0.132 |
Pen | +0.129 |
restaurant | +0.128 |
Neg logit contributions | |
bar | -0.099 |
ink | -0.097 |
chalk | -0.097 |
itted | -0.091 |
ch | -0.088 |
Token | They |
Token ID | 1119 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
Jacob | +0.003 |
restaurant | +0.003 |
Jacob | +0.002 |
Megan | +0.002 |
abel | +0.002 |
Neg logit contributions | |
chalk | -0.002 |
ink | -0.001 |
Bree | -0.001 |
bar | -0.001 |
cho | -0.001 |
Token | want |
Token ID | 765 |
Feature activation | -7.99e-01 |
Pos logit contributions | |
Jacob | +0.012 |
Megan | +0.011 |
restaurant | +0.011 |
Jacob | +0.011 |
Pen | +0.010 |
Neg logit contributions | |
chalk | -0.007 |
ink | -0.006 |
sun | -0.006 |
bar | -0.006 |
wind | -0.006 |
Token | decided |
Token ID | 3066 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
chalk | +0.029 |
ink | +0.027 |
peac | +0.023 |
printed | +0.023 |
Jazz | +0.022 |
Neg logit contributions | |
Jacob | -0.066 |
Megan | -0.063 |
Jacob | -0.063 |
Pen | -0.062 |
Clive | -0.059 |
Token | to |
Token ID | 284 |
Feature activation | +5.75e-01 |
Pos logit contributions | |
chalk | +0.060 |
bar | +0.060 |
ink | +0.060 |
cho | +0.057 |
bullet | +0.056 |
Neg logit contributions | |
Jacob | -0.113 |
Megan | -0.110 |
restaurant | -0.107 |
Jacob | -0.105 |
Pen | -0.103 |
Token | play |
Token ID | 711 |
Feature activation | -8.10e-01 |
Pos logit contributions | |
chalk | +0.032 |
ink | +0.029 |
jet | +0.027 |
paint | +0.026 |
arrow | +0.025 |
Neg logit contributions | |
Jacob | -0.056 |
Jacob | -0.055 |
Megan | -0.053 |
restaurant | -0.052 |
Clive | -0.050 |
Token | hide |
Token ID | 7808 |
Feature activation | -3.83e+00 |
Pos logit contributions | |
Jacob | +0.083 |
Megan | +0.082 |
Gor | +0.079 |
Jacob | +0.078 |
Pen | +0.077 |
Neg logit contributions | |
ink | -0.044 |
chalk | -0.043 |
golf | -0.042 |
pins | -0.036 |
Bounce | -0.036 |
Token | and |
Token ID | 290 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
Jacob | +0.396 |
Jacob | +0.374 |
Megan | +0.363 |
Clive | +0.349 |
Gor | +0.349 |
Neg logit contributions | |
ink | -0.194 |
chalk | -0.182 |
pins | -0.169 |
cho | -0.163 |
jet | -0.150 |
Token | seek |
Token ID | 5380 |
Feature activation | -5.50e+00 |
Pos logit contributions | |
Jacob | +0.006 |
Megan | +0.005 |
Jacob | +0.005 |
restaurant | +0.005 |
abel | +0.005 |
Neg logit contributions | |
chalk | -0.010 |
ink | -0.010 |
bar | -0.010 |
bullet | -0.010 |
Bounce | -0.009 |
Token | again |
Token ID | 757 |
Feature activation | -2.35e+00 |
Pos logit contributions | |
Gor | +0.561 |
Jacob | +0.547 |
Jacob | +0.534 |
Megan | +0.522 |
Pen | +0.522 |
Neg logit contributions | |
ink | -0.235 |
chalk | -0.231 |
pins | -0.191 |
spinning | -0.188 |
golf | -0.181 |
Token | . |
Token ID | 13 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
Jacob | +0.203 |
Megan | +0.194 |
Jacob | +0.191 |
restaurant | +0.189 |
abel | +0.181 |
Neg logit contributions | |
chalk | -0.142 |
ink | -0.141 |
bar | -0.130 |
ch | -0.129 |
Bree | -0.127 |
Token | Sara |
Token ID | 24799 |
Feature activation | -2.68e+00 |
Pos logit contributions | |
Jacob | +0.003 |
restaurant | +0.003 |
Jacob | +0.003 |
Megan | +0.003 |
abel | +0.003 |
Neg logit contributions | |
chalk | -0.002 |
ink | -0.002 |
Bounce | -0.002 |
bar | -0.002 |
Bree | -0.002 |
Token | was |
Token ID | 373 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
Clive | +0.260 |
Jacob | +0.260 |
Megan | +0.260 |
restaurant | +0.250 |
Jacob | +0.250 |
Neg logit contributions | |
ink | -0.111 |
chalk | -0.110 |
printed | -0.099 |
spun | -0.094 |
Jazz | -0.093 |
Token | the |
Token ID | 262 |
Feature activation | -6.78e-01 |
Pos logit contributions | |
Jacob | +0.111 |
restaurant | +0.108 |
Megan | +0.106 |
Jacob | +0.103 |
Pen | +0.100 |
Neg logit contributions | |
chalk | -0.066 |
ink | -0.065 |
bar | -0.064 |
cho | -0.062 |
itted | -0.062 |
Token | inside |
Token ID | 2641 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
Jacob | +0.352 |
Megan | +0.339 |
restaurant | +0.330 |
Pen | +0.312 |
Jacob | +0.311 |
Neg logit contributions | |
bar | -0.326 |
chalk | -0.323 |
ink | -0.322 |
bullet | -0.308 |
ch | -0.307 |
Token | . |
Token ID | 13 |
Feature activation | -4.19e-02 |
Pos logit contributions | |
Megan | +0.107 |
Jacob | +0.107 |
Pen | +0.104 |
Jacob | +0.102 |
restaurant | +0.102 |
Neg logit contributions | |
chalk | -0.074 |
ink | -0.072 |
bar | -0.068 |
cho | -0.062 |
crashing | -0.062 |
Token | It |
Token ID | 632 |
Feature activation | +6.08e-01 |
Pos logit contributions | |
Jacob | +0.003 |
restaurant | +0.003 |
Megan | +0.003 |
Jacob | +0.003 |
Pen | +0.003 |
Neg logit contributions | |
chalk | -0.003 |
ink | -0.003 |
bar | -0.003 |
bullet | -0.003 |
ch | -0.003 |
Token | was |
Token ID | 373 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
ink | +0.037 |
bar | +0.036 |
chalk | +0.036 |
printed | +0.035 |
ch | +0.034 |
Neg logit contributions | |
restaurant | -0.052 |
Jacob | -0.050 |
Megan | -0.050 |
Jacob | -0.048 |
Pen | -0.047 |
Token | a |
Token ID | 257 |
Feature activation | +4.13e-01 |
Pos logit contributions | |
Jacob | +0.109 |
Megan | +0.106 |
restaurant | +0.103 |
Jacob | +0.099 |
Pen | +0.099 |
Neg logit contributions | |
ink | -0.074 |
bar | -0.074 |
chalk | -0.074 |
bullet | -0.069 |
cho | -0.069 |
Token | stuffed |
Token ID | 22259 |
Feature activation | -4.94e+00 |
Pos logit contributions | |
itted | +0.030 |
bar | +0.030 |
Bree | +0.027 |
alks | +0.027 |
ully | +0.027 |
Neg logit contributions | |
restaurant | -0.032 |
Jacob | -0.032 |
Megan | -0.030 |
Pen | -0.028 |
Jacob | -0.028 |
Token | toy |
Token ID | 13373 |
Feature activation | +1.75e+00 |
Pos logit contributions | |
Jacob | +0.453 |
Megan | +0.442 |
Jacob | +0.415 |
restaurant | +0.412 |
Pen | +0.406 |
Neg logit contributions | |
chalk | -0.300 |
ink | -0.297 |
bar | -0.274 |
bullet | -0.273 |
laser | -0.269 |
Token | bear |
Token ID | 6842 |
Feature activation | -2.48e+00 |
Pos logit contributions | |
bar | +0.129 |
chalk | +0.127 |
ink | +0.127 |
itted | +0.125 |
ch | +0.123 |
Neg logit contributions | |
Jacob | -0.130 |
Megan | -0.124 |
restaurant | -0.122 |
Pen | -0.113 |
Jacob | -0.112 |
Token | ! |
Token ID | 0 |
Feature activation | +2.67e-01 |
Pos logit contributions | |
Megan | +0.201 |
Jacob | +0.200 |
restaurant | +0.191 |
Jacob | +0.187 |
Pen | +0.184 |
Neg logit contributions | |
bar | -0.166 |
ink | -0.164 |
chalk | -0.163 |
ch | -0.153 |
cho | -0.153 |
Token | Bob |
Token ID | 5811 |
Feature activation | -5.84e+00 |
Pos logit contributions | |
bar | +0.018 |
chalk | +0.018 |
ink | +0.018 |
bullet | +0.017 |
ch | +0.017 |
Neg logit contributions | |
Jacob | -0.021 |
Megan | -0.021 |
restaurant | -0.020 |
Pen | -0.019 |
Jacob | -0.019 |
Token | gave |
Token ID | 2921 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
restaurant | +0.489 |
Megan | +0.470 |
Jacob | +0.465 |
Jacob | +0.429 |
Pen | +0.422 |
Neg logit contributions | |
bar | -0.369 |
chalk | -0.352 |
ink | -0.346 |
itted | -0.326 |
ch | -0.325 |
Token | I |
Token ID | 40 |
Feature activation | +4.23e-01 |
Pos logit contributions | |
ink | +0.093 |
chalk | +0.091 |
bullet | +0.089 |
bar | +0.089 |
cho | +0.087 |
Neg logit contributions | |
Jacob | -0.100 |
Megan | -0.098 |
Jacob | -0.093 |
restaurant | -0.092 |
Pen | -0.091 |
Token | think |
Token ID | 892 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
ch | +0.024 |
chalk | +0.024 |
peac | +0.023 |
printed | +0.023 |
br | +0.023 |
Neg logit contributions | |
Jacob | -0.039 |
Megan | -0.037 |
restaurant | -0.036 |
Jacob | -0.035 |
Pen | -0.035 |
Token | your |
Token ID | 534 |
Feature activation | +3.66e+00 |
Pos logit contributions | |
Jacob | +0.101 |
Megan | +0.097 |
restaurant | +0.095 |
Pen | +0.090 |
Jacob | +0.090 |
Neg logit contributions | |
bar | -0.074 |
chalk | -0.073 |
ink | -0.073 |
bullet | -0.069 |
ch | -0.069 |
Token | name |
Token ID | 1438 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
bar | +0.294 |
ink | +0.276 |
ch | +0.275 |
chalk | +0.274 |
bullet | +0.273 |
Neg logit contributions | |
Jacob | -0.255 |
Megan | -0.244 |
restaurant | -0.240 |
Pen | -0.228 |
Nick | -0.215 |
Token | is |
Token ID | 318 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
Jacob | +0.014 |
restaurant | +0.014 |
Megan | +0.014 |
Jacob | +0.013 |
Nick | +0.012 |
Neg logit contributions | |
bar | -0.015 |
ch | -0.015 |
cho | -0.014 |
chalk | -0.014 |
br | -0.014 |
Token | Fl |
Token ID | 1610 |
Feature activation | -5.77e+00 |
Pos logit contributions | |
Jacob | +0.135 |
Megan | +0.130 |
restaurant | +0.126 |
Pen | +0.119 |
Jacob | +0.119 |
Neg logit contributions | |
bar | -0.121 |
chalk | -0.119 |
ink | -0.119 |
bullet | -0.114 |
ch | -0.114 |
Token | uffy |
Token ID | 15352 |
Feature activation | -4.18e+00 |
Pos logit contributions | |
Jacob | +0.452 |
Megan | +0.419 |
restaurant | +0.402 |
Jacob | +0.393 |
Pen | +0.374 |
Neg logit contributions | |
ch | -0.433 |
ink | -0.429 |
chalk | -0.422 |
bar | -0.421 |
cho | -0.408 |
Token | ," |
Token ID | 553 |
Feature activation | -3.85e-01 |
Pos logit contributions | |
Jacob | +0.336 |
Megan | +0.329 |
Jacob | +0.321 |
restaurant | +0.318 |
Nick | +0.298 |
Neg logit contributions | |
bar | -0.263 |
cho | -0.257 |
ch | -0.257 |
ink | -0.256 |
chalk | -0.242 |
Token | she |
Token ID | 673 |
Feature activation | +9.44e-01 |
Pos logit contributions | |
restaurant | +0.037 |
Jacob | +0.037 |
Megan | +0.035 |
Jacob | +0.034 |
ide | +0.033 |
Neg logit contributions | |
ink | -0.019 |
chalk | -0.019 |
bar | -0.019 |
ch | -0.018 |
Bree | -0.018 |
Token | said |
Token ID | 531 |
Feature activation | +4.60e-01 |
Pos logit contributions | |
bar | +0.057 |
ink | +0.057 |
chalk | +0.057 |
ch | +0.054 |
cho | +0.053 |
Neg logit contributions | |
Jacob | -0.082 |
Megan | -0.079 |
restaurant | -0.079 |
Jacob | -0.074 |
Pen | -0.074 |
Token | . |
Token ID | 13 |
Feature activation | -2.85e-02 |
Pos logit contributions | |
bar | +0.035 |
chalk | +0.034 |
ink | +0.034 |
bullet | +0.033 |
ch | +0.033 |
Neg logit contributions | |
Jacob | -0.034 |
Megan | -0.032 |
restaurant | -0.032 |
Jacob | -0.030 |
Pen | -0.030 |